Defense of th e

opment of the anti-aircraft g uided missile, the Nike. See page 446.
Courtesy Jr"E 1\l aga=in P. Wes tern Electric Compan y, Incorporated.
 STANLEY S. BALLARD, M.A., Ph.D.
Professor and Chairman, Department of Physics, Tufts College

EDGAR P. SLACK, S.B., M.S.

Professor and Chairman, Undergraduate Physics
Polytechnic Institute of Brooklyn
Director, Training School, National Board of Fire Underwriters

ERICH HAUSMANN, E.E. Sc.D.

Dean Emeritus, Polytechnic Institute of Brooklyn
Consultant, Nussbaumer, Clarke g Velzy, Inc.

Physies
Prineiples

D. Van Nostrand Company, Jue.

TORONTO NEW YORK LONDON
NEW YORK
D. VAN NOSTRAND COMPANY, INC.
250 Fourth Avenue, New York 3

TORONTO
D. VAN NOSTRAND COMPANY (Canada), Ltd.
25 Hollinger Road, Toronto

LONDON
MACMILLAN & COMPANY, Ltd.
St. Martin's Street, London, W.C. 2

Copyright 1954, by D. Van Nostrand Company, Inc.

Published simultaneously in Canada by
D. Van Nostrand Company (Canada), Ltd.

All rights in this book are reserved. Without written authorization from D. Van Nostrand
Company, Inc., 250 FourtJi Ave., New York 3, N. Y., it may not be reproduced in any
form in whole or in part (except for quotations in critical articles or reviews), nor may it
be used for dramatic, motion-talking-picture, radio, television, or any other similar purpose.

Library of Congress Catalog Card No. 54-6280

Printed in the United States of America

 PREFACE
This is an introductory textbook on the principles of Physics, written espe-
cially for students who are majoring in one of the physical sciences or in
engineering. It is designed both for the 3- or 4-semester courses offered in
some schools and for the 2-semester course in others. For the shorter courses
certain sections can be omitted without disturbing the continuity of treat-
ment; a suggested list of sections for omission is given in Appendix III.
It is assumed that the students who use this book will be taking calculus
concurrently, or will already be familiar with its elements. Calculus is intro-
duced in Chapter 3, and is used in later chapters in those places where its
powerful methods are most helpful.
The treatment is based principally on three systems of units: the British
gravitational system because of its extensive use in engineering, the cgs sys-
tem because of its wide acceptance in scientific work, and the mks system
because of its growing use in electrical theory. The relationships between
the mks units and the older electrical and magnetic units are given so that
the student can make the transition to the older units when he encounters
them in the literature.
The authors have striven for that hard-to-define property called "teach-
ability," which despite its elusive nature is so very real both to instructor
and to student. Introductions of new concepts and definitions of units are
clearly stated, and many completely solved problems are given, especially in
the earlier chapters. The many figures are designed to illustrate the basic
concepts, devices, and circuits in the most straightforward manner; line draw-
ings are chiefly used because of their simplicity. More than 800 problems
are included at the ends of the chapters to give the student ample practice
in applying physical principles. Answers to the odd-numbered problems are
given at the back of the book.
The selection of topics and the order of their presentation were carefully
considered. For instance, the concept of mass is def erred until definitely
needed in Chapter 4, rotation is taken up before work and energy so that
kinetic energy of translation and rotation can be covered together, and the
transfer of heat precedes a treatment of the effects of heat to attain a more
logical development. Such important fields as electronics and thermionics,
high-frequency circuits, and semiconductors are given treatments appropriate
lll
lV PREFACE

to their theoretical and practical importance. In the division on light, em-

phasis is given to physical optics and to up-to-date treatments of visual
phenomena, photometry, color specification, and polarization-fields some-
times neglected in introductory courses.
Topics in so-called modern physics are introduced throughout the text
wherever they follow naturally from the "classical" treatment. Additionally,
two chapters at the end of the book cover briefly the major phenomena in
the fields of atomic and nuclear physics. It is urged that this important
material be included in all basic courses, so that students of science and
engineering may be introduced to atomic and sub-atomic physics early in
their college careers.
The authors believe that the book reflects the great strides taken in the
physical sciences in recent years and that it provides a good background of
basic physics for the engineers and scientists of the future.

The entire text has been taught for two years in preprint form, and has
been improved greatly thereby. Sincere appreciation is expressed to our
many colleagues who have offered suggestions for the betterment of the book,
especially to Professors Kathryn A. McCarthy and Charles R. lvfingins of
Tufts College for their criticism of large portions of the manuscript and
preprint and for other assistance. Thanks are expressed to Professor Charles
A. Hachemeister of the Polytechnic Institute of Brooklyn for a critical review
of Chapter 24 and to Professor Leo E. A. Saidla, also of the Polytechnic, for
the editorial reading of the entire manuscript.
February, 1954. STANLEY 8. BALLARD
EDGAR P. SLACK
ERICH HAUSMANN
 CONTE.NTS
MECHANICS

CHAPTER PAGI<J
1 INTRODUCTORY CONCEPTS AND J\1:I<JTHODS 1
2 STATICS 16
3 THE FUNDAMENTALS OF MOTION . 33
4 FORCE AND TRANSLATION 58
5 TORQUE AND ROTA'l'ION 82
6 J\bJCHANICAL ENERGY . 98
7 ELASTICITY AND HARMONIC MOTION . 116
8 LIQUIDS AT RI<JST 135
9 GASES . 154
10 FLUIDS IN MOTION 165
HEAT

11 HEAT AND HEAT TRANSFI<JR 181

12 HEAT QUANTITIES AND CHANGE OF PHASE . 198
13 THERMAL B1<mAVIOR OF GASES 219
14 THERMODYNAMICS 237
ELECTRICITY & MAG1'..~TIS::\1

15 ELECTROSTATICS . 257
16 ELECTRIC CURRENT AND ELECTROMOTIVE FoRCr~ 279
17 RESISTANCE OF CIRCUITS 297
18 ELECTROMAGNI<JTISM . 317
19 MAGNETS AND J\1AGNETIC CIRcurrs 335
20 ELECTROMAGNETIC INDUCTION . 351
21 INDUCTANCI<J AND CAPACI'l'ANCE 363
22 ALTERNATING CURRENTS 380
23 ELECTRONICS . 401
2.J: RADIATION FROM CIRCUITS 428
WAVE MOTION & SOUND

25 WAVE MOTION 449

26 SOUND. 470
VI CONTENTS

OPTICS

CHAPTER PAGE
27 LIGHT AND VISION . 503
28 REFLECTION, REFRACTION, AND DISPERSION 526
29 LENSES AND CURVED 1\1mRORS 548
30 OPTICAL INSTRUMENTS . 574
31 SPI<JCTRA, COLOR, AND ATMOSPHERIC OPTICS 598
32 JNTERFJ<JRI<JNCE AND DIFFRACTION 617
33 POLARIZATION 640

MODERN PHYSICS

34 QUANTUM OPTICS 659

35 NUCLEAR PHYSICS 682
APPENDIX I-PHYSICAL AND MATHEMATICAL TABLES 713
APPENDIX II-RI<JLATION BI<JTWEEN MKS AND OTHER ME'I'RIC
UNITS 721
APPENDIX III-SI<JCTIONS OF TI<JXT SUGGESTED FOR OMISSION IN
SHORTER COURSES 724
ANSWERS TO ODD-NUMBERED PROBLEMS 725
INDEX . 729
 MECHANICS

1
INTRODUCTORY CONCEPTS
AND METHODS
Physics, or "Natural Philosophy," as it used to be called, takes its place
among the physical sciences with astronomy, chemistry, and geology, in
applying the laws of nature to material objects. The areas of knowledge of
these sciences overlap considerably and give rise to such subjects as astro-
physics, physical chemistry, and geophysics. Originally the range of Physics
did not include biology, but it now appears that physical laws, when properly
interpreted, apply to living things, and a science called biophysics has become
recognized. Chiefly, however, Physics deals with inanimate bodies. It is
concerned broadly with matter and energy, together with such related quan-
tities as force and motion. These concepts and the relations among them are
common to all the subdivisions of Physics, which usually include mechanics,
heat, electricity, sound, light, and atomic structure. The scientist or engineer
requires a foundation of physics on which to build his professional career.
1-1. Physical Measurement. Our knowledge of natural phenomena is
based upon the results of a multitude of experiments or tests. If these results
were indefinite or ambiguous, they might be given different interpretations,
and for that reason the conclusions drawn from the experiments would be
open to speculation. To have definite meaning, the evidence gathered must
be quantitative, and evidence of this kind can be gotten only by measurement.
Indeed, measurement is one of the most important operations in all scientific
work. It is because of the precision usually obtainable in physical measure-
ments that physics is called an exact science.
The ordinary way to measure a quantity is to compare it with some other
quantity of the same kind which is used as a basis of reference. Everyone is
familiar with the process of measuring the length of an object by stepping it
1
2 MECHANICS Chap. 1

off with a foot-rule or yardstick. If a pole is 15 feet long, anyone having a

foot-rule can arrive at a correct value of its length by laying off a distance
equal to 15 one-foot distances. The length could also be expressed as 180
inches. This illustration shows that the measurement of a quantity involves
two things: a number and a unit. To say that the pole measures 15 or 180
would be an incomplete statement; it is necessary to say 15 feet or 180 inches.
The unit indicates how large a quantity is used as the basis of reference and
the number how many of these are contained in the quantity being measured.
A physical quantity involves a number and a unit.
Some statements based on physical measurements are given herewith to
indicate the necessity for both number and unit: The average speed of the
S.S. United States in her maiden Atlantic voyage was 35.59 knots. A certain
automobile engine is capable of developing 135 horsepower. Comfortable
room temperature is 68 degrees fahrenheit ... The angular speed of a particu-
lar motor is 1800 revolutions per minute. Atmospheric pressure is about
14.7 pounds per square inch. The speed of radio and light waves is 186,000
miles per second or 300,000,000 meters per second. The wavelength of yel-
low sodium light is 0.0000589 centimeter.
A very large or very small number is often written as a number from 1 to
10 that is multiplied by 10 raised to a suitable power; thus, 300,000,000 is
written 3.00 X 108 and 0.0000589 · is written 5.89 X 10-5 • This practice
makes it easy to deal with such numbers in multiplication and division.
Physical quantities can be expressed in terms of several systems of units,
but this textbook employs only the three most commonly used: the British
gravitational system generally used in engineering, and two metric absolute
systems; the latter are known as the meter-kilogram-second (mks) and the cen-
timeter-gram-second (cgs) system. All three systems are described in Chap. 4.
1-2. Position and Displacement. The position of an object ,vith respect
to any fixed reference point can be specified in either rectangular or polar

----------------- Q

W-t-E
I
s

e
Q'-------v------~~~~~~
3mi 4mi
FIG. 1-l. Position and displacement. Initial position at P; displacement is PQ.
§ 1-3 LJiJNGTJJS A.ND ANGLES ..,
,:)

coordinates. An automobile at point P in Fig. 1 will serve as an example.

Its position is described in rectangular coordinates by saying that it is 3 miles
east and 4 miles north of the reference point 0. In polar coordinates, the
distance OP = V (3) 2 + (4) 2 = fi miles, and () (theta) is the angle of which
the sine is !-, namely, 53. l degrees; therefore the position of the automobile is
described as 5 miles from O along a direction 53.1 degrees north of east.
vVhen an object moves from one position to another, its displacement is
defined as the straight line extending from its initial position to its final posi-
tion and is commonly represented by an arrow directed avrny from the initial
position. Thus, if the automobile in Fig. 1 moves from P to Q, regardless of
the path actually followed in traveling between these points, its displacement
+
is -Y (4) 2 (4) 2 = 5.66 miles along a direction 4.5 degrees north of east .and
is represented by the arrow in the figure. Displacements are specified by
lengths and angles; these quantities will now be considered in more detail.
1-3. Lengths and Angles. The units of length in the metric systems are
based on a standard known as the meter. The meter is defined as· the dis-
tance at zero degrees centigrade(0°C) between the centers of two,lines.traced
on a platinum-iridium bar kept in a subterranean vault of the International
Bureau of Weights and Measures at
Sevres, France. Copies of this stand-
ard meter bar are kept at the National
Bureau of Standards in Washington,
D. C., and at other national labora-
tories; a portion of such a bar is shown
in Fig. 2. Within recent years, the Fm. 1-2. Portion of a standard meter
bar. The defining lines are engraved on
meter has been "calibrated" in terms the center rib.
of a single wavelength in the light from
a glowing vapor. Such a wavelength is constant and reproducible, and hence
the meter could be established again accurately if the standard bar should
be damaged or destroyed. The standard of length in this country is the
U. S. yard. The yard was fixed by an Act of Congress in 1866 as equal to
3600/3937 meter.
The multiples and submultiples of the meter and of the yard in common
use, together with their abbreviations and equivalents, are given below for
reference purposes. Some common prefixes used with metric units include
the following: kilo = 1000, centi = 1/100, and milli = 1/1000.
TABLE I. UNITS OF LENGTH

Metric British

1 kilometer (km) = 1000 meters 1 mile (mi) = 1760 yards

1 meter (m) = 100 centimeters 1 mile (mi) = 5280 feet
1 meter (m) = 1000 millimeters 1 yard (yd) = 3 feet
1 centimeter (cm) 10 millimeters 1 foot (ft) 12 inches
1 millimeter (mm) = 1000 microns 1 inch (in.) = 1000 mils
4 MECHANICS Chap. 1

Expressions of length in one group of units can be converted to correspond-

ing ones in the other group by means of the following relations, which are
sufficiently exact for most purposes. The two italicized conversion factors
should be remembered, since they will be used most frequently.
TABLE II. CONVERSION FACTORS FOR LENGTHS

1 mile = 1.6093 kilometers

1 kilometer = 0.6214 mile
1 meter = 39 .37 inches
1 meter = 3.281 feet
1 foot = 30.48 centimeters
1 inch = 2.540 centimeters

Directions of lines are specified by angles. The entire angle in a plane

around a point has a value of 360 degrees (0 ) ; one-quarter of this angle is a
right angle or 90°. A degree contains 60 minutes of arc ('), and a minute
contains 60 seconds of arc ("). These units are used to express the angle
between any two lines that lie along different directions.
1-4. Force. A force is a push or a pull exerted upon a body. Forces can
be applied through muscular effort or exerted by inanimate objects. Thus, a
cable supporting an elevator exerts an upward pull or force on the elevator;
again, a column supporting a roof exerts an upward push or force on the roof.
There are two particular forces with which everyone is
familiar. One of these is the force of gravity. This is the
pull that is exerted upon an object by the earth; it acts
Spring downward and is called the weight of the object. At high
altitudes the pull of the earth is a little less than at sea
level, but the variations are small. The other force is the
Q
force of friction. This force is encountered when one object
~ Scale moves along another. For example, when a man pushes a
6
heavy box along the floor, the floor exerts a backward fric-
tional drag on the box, because the surfaces sliding upon
each other are rough. Friction oppose5 motion, and the
Fw. 1-3. Sche-
matic diagram force of friction on a body always acts in a direction just op-
of a spring bal- posite to that in which the body slides or tends to slide.
ance. For engineering and for general household purposes the
pound (lb) is used as the unit for force. This unit refers to
a particular block of platinum preserved at the Standards Office in West-
minster, London; the pull exerted by the earth upon this block is the pound
of force. For definiteness, this weight measurement is made at 45 ° latitude
and at sea level.
Weights and other forces are often measured with a spring balance, the
essential parts of which are shown in Fig. 3. The body to be weighed is sus-
pended by a spring; as the body settles downward, the spring stretches and
exerts an increasing upward force upon it, and a balance is reached ,vhen the
§ 1-6 VECTOR ADDITION 5

upward force due to the spring equals the downward pull of gravity on the
body. Such a device is calibrated by hanging bodies of known weights on it
and marking the corresponding extensions of the spring on a scale. The
scale can then be used to furnish a direct reading of any weight or other force
within its range.
1-5. Scalar and Vector Quantities. Many physical quantities are fully
specified by a number and a unit; that is, they are completely determined by
magnitude. Others are directed in space and involve an additional element
--direction. These two types of quantities are distinguished by character-
istic names.
The term scalar refers to quantities that have magnitude only, for example:
the time required for a particular operation, the volume of water in a reservoir,
or the horsepower of a motor; none of these quantities involves the idea of
direction.
The term vector is applied to quantities that have both magnitude and
direction; for completeness of definition it is necessary to add that they shall
be capable of being added by the vector method, as described below. As
examples may be mentioned the force exerted by a guy wire, the displace-
ment of a moving object, or the velocity of an airplane; in order to deal with
these quantities their directions must be considered. Any vector quantity
can be represented by an arrow drawn in the appropriate direction and hav-
ing a length which indicates, to some convenient scale, the numerical value of
the quantity it represents.
There are many occasions when physical quantities of the same kind are to
be added, but the process is quite different for scalars and vectors. Scalar
quantities can be added by simple arithmetic. Thus, the pouring of 100 cm3
of liquid into a vessel which already contains 500 cm3 of it brings the total
liquid volume to 600 cm 3 ; again, a 4-minute interval followed without pause
by an interval of 8 minutes makes a total elapsed time of 12 minutes. Addi-
tions can be carried out in this manner when the quantities do not involve
the idea of direction. The addition of vector quantities is more involved,
but not difficult; several forms of the vector method are described in the fol-
lowing sections.
1-6. Vector Addition. The addition of vector quantities can be illustrated
by an example in which several displacements are to be combined. A man
may be supposed to walk a certain .distance in one direction, then another
distance in a different direction, and so on, and at the end of his walk desire
to know his final position with respect to the starting point. His entire
journey can be mapped out as a series of displacements, one after another,
each represented by an arrow of appropriate length and direction. An arrow
drawn from the beginning to the end of the diagram provides the desired
information. This arrow represents the "sum" of the separate displace-
ments when their directions are taken into account; the vector sum is also
known as the resultant displacement.
6 MECHANICS Chap. 1

In the graphical solution of such a problem the diagram is made carefully

to scale, and the resultant is measured with rule and protractor to determine
its length and direction. In the analytical solution a rough diagram will
suffice as a guide, and the length and direction of the resultant are found
mathematically. Both procedures will be illustrated by a specific problem.
Consider a pedestrian to walk 4 mi due east, then 3 mi in a direction 60° north of
east. How far is he from the starting point, and in what direction is he from that
point?
To solve the problem graphically, lay off the
arrows representing the displacements accurately to
a suitable scale and at the exact inclinations, with
the tail-end of the second arrow placed at the head-
end of the first, thereby forming two sides of a tri-
4 mi angle, as in the plan view of Fig. 4. Draw the
closing side of the triangle and measure its length
Fm. 1-4. Addition of two di~- carefully to the same scale. This gives the numeri-
placements. The resultant is R. cal value of the displacement sought as R = 6.1 mi.
Measure with a protractor the direction that this
closing side of the triangle makes with the east direction; this angle is marked 8 in
the diagram. The result shows that side R is north of east by an angle 8 = 25°. It
will be observed tha't both R and 8 are obtained to only two "figures.

The accuracy of a graphical result depends not only upon laying out the
lengths and directions correctly, but also upon making the drawing of ade-
quate size. If a vector were represented as an arrow 6 in. long, an error of
-h in. in its length would mean an inaccuracy of one part in 6 X 32 or 192-
that is, an error of ! per cent. Graphical solutions like the foregoing can be
depended upon for an accuracy of 1 per cent if care is exercised.
To solve the foregoing problem analytically, make a rough sketch to resemble Fig. 4,
and determine the unknown parts of the triangle by trigonometry. First, note that
the angle between the two known sides of the triangle is 120°, and use the Law of
Cosines to determine the magnitude of the resultant R. Thus,
R2 = (4) 2 + (3) 2 - 2 X 4 X 3 cos 120°
= 16 + 9 - 24( -0.500) = 37.00,
whence R = 6.08 mi. Next, apply the Law of Sines to find the direction of R:
3 mi
- - - 6.08 mi
sin (J sin J.20°'
whence sin () = (3/6.08) sin 120° = 0.493 X 0.866 = 0.427. From trigonometric
tables, the angle having this value for its sine is() = 25.3°.

It is apparent from this illustration that the precision is better in the ana-
lytical solution, since the result does not depend upon the accuracy of the
diagram.
1-7. Vector Addition by the Parallelogram Method. A different procedure
from that just described is often used to find the resultant of two vectors.
§ 1-7 VECTOR ADDITION BY THE PARALLEWGRAM METHOD 7

This procedure is known as the Parallelogram Method; it will be applied to

finding the resultant of the vectors a and b shown in Fig. 5. In this method
the vectors are laid off in the proper directions from a common starting point S,
so as to form two adjacent sides of a parallelogram, as in part I of the figure.
The parallelogram is next completed, as in part II, by drawing the sides a' and
b' parallel to a and b respectively; these intersect at T. The diagonal R
drawn from the starting point, called the concurrent diagonal, gives the result-

a' T
-------1
I
/b'
I
s---------
a s - - -0- - - -
I II m
Fm. 1-5. Adding vectors a and b by the Parallelogram Method; the resultant is R.

ant both in length and direction, as shown in part III of the figure. Since
lengths b and b' are equal, the resultant R is the same as though vector b' had
been added to vector a by the method described in the preceding section.
Consequently, the same mathematical steps used in the solution of the oblique
triangle can be used to obtain an analytical solution by the Parallelogram
Method.
The process of vector addition just described applies also to the reverse
operation of vector subtraction "Yvith but slight modification. Thus, one
vector can be subtracted from another by reversing
its direction and proceeding as in addition. For
example, to subtract vector b from vector a in
Fig. 5, first b is reversed by shifting it through
180° to the position -bas in Fig. 6, and then a a
and - b are added by the Parallelogram Method I
I
to get the difference D, as shown. It should be I
I
noted that rotating a vector through 180° is equiv- I
I
alent to reversing its sign. I
The addition of vectors has its widest applica- I
I
tion in connection with forces. There are innu-
Frn. 1-6. Subtracting vector
merable situations where an object is subjected b from vector a; the difference
to several forces acting in different directions at is the diagonal D.
the same time, and it is necessary to determine
a single force that would be equivalent to all of them acting together-
that is, to determine their resultant.
For example, two forces are exerted through strings by pulling on spring balances
in the directions shown in part I of Fig. 7. The forces, A of 10 lb and B of 7 lb, are
8 MECHANICS Chap. 1

exerted upon a knot k where the strings are tied, and the knot is kept fixed in position
by applying a suitable force along the dotted line C. Find the resultant of the forces
A and B.
To solve the problem graphically, draw the
forces A and B to scale as in part II of the
figure, so as to form two sides of a parallelo-
gram; next, complete the parallelogram; and
finally, draw the concurrent diagonal. This
diagonal R represents the resultant of the
forces A and B, and its length and direction
can be scaled off. It is found to have a mag-
nitude of 5.3 lb and to make an angle of
8 = 42° with A.
I To find the resultant force analytically, a
Il diagram of the general form shown in part
II of the figure is drawn but it need not be to
Fm. 1-7. Addition of two forces. scale. The Law of Cosines applied to the
lower triangle in the diagram shows that
R 2 = A 2 + B 2 - 2 AB cos 30° = (10) 2 + (7) 2 - 2 X 10 X 7 X 0.866 = 27.8; whence
the resultant R = 5.27 lb. Since the upper angle is 8, it follows from the Law of Sines
that B/sin 8 = R/sin 30°; then
. 8 = B sin 30° = 71l> X 0.500 = 0 664
sm R 5.27~ · '
and 8 = 41.6°. The analytical solution verifies the graphical solution and is more
precise.
The method of vector addition just described can be used to combine any
number of vectors, either graphically or analytically, by first finding the
resultant of two of them, then adding another vector to this resultant in
the same way, and continuing the process until all are included. To apply
the Parallelogram Method over and over is laborious, however, and for this
reason the method described in the following section is preferred.
1-8. Polygon Method of Vector Addition. A particularly useful method
for adding a number of vectors consists of constructing a polygon with the
vectors as sides; the closing side of the figure is the resultant. The procedure
in the Polygon Method is as follows: from a chosen starting point lay off one
of the vectors to appropriate scale; from the head-end or terminal of this first
vector lay off the next one similarly; from the terminal of the latter lay off

s
d

I II

Fm. 1-8. The Polygon Method of adding vectors.

§ 1-10 VECTOR ADDITION BY TIIE RESOLUTION METHOD 9

the third, and so on until all are included. Lastly, draw the line from the
starting point of the first vector to the terminal of the last one; this line closes
the polygon and forms the resultant. The method is illustrated in Fig. 8,
wherein the four vectors a, b, c, and d shown in part I are added. In part II
of the figure the order of addition, starting at S, is a, b, c, d; in part III the
order is b, c, d, a; the resultant is the same whatever sequence is followed.
The resultant is Rand is directed away from the starting point.
The Polygon Method is an extension of that used in the problem of § 1-6
in which only two vectors were added; in that case the polygon had the simple
form of a triangle.
1-9. Resolution of Vector Quantities. In the preceding section attention
has been focused upon methods by which vectors can be added to form a
resultant. The reverse process, that of breaking a vec-
tor into its components, is called resolution. It was vi
pointed out that R in Fig. 5 is the resultant of vectors
a and b; in reverse, a and b may be considered the
components of vector R. Similarly, in Fig. 9, x and y
are the components of vector R along the X and Y
axes. Here the components are at right angles to each
other and hence are termed rectangular components.
The components of any vector may be used instead Fm. l-9. Resolution
x x
{ZJ _ _
of a vector into X and
of the vector itself. Y components.
Figure 10 shows pictorially in successive steps I to V
hmv the vector R can be resolved into components along directions 1 and
2; the respective components are found to be P and Q. The same proce-
dure is followed whatever the directions of the initial vector or the direction
lines.

2
I
I
I
1
/
I

/f'
R
'-
'-
',l /

I
I
,-..,
'
I

'
I

1'
I
1
'-
R //',,
I
'

I n m IY v
Fm. 1-10. The process of resolution. I, Showing vector R and direction lines 1 and 2 for
its components. II, Starting a parallelogram. III, Completing the parallelogram. IV,
Designating the components and showing their directions. V, Components P and Q now
replace the original vector R.

1-10. Vector Addition by the Resolution Method. The process of reso-

lution just discussed is extremely useful when applied to the addition of
vectors analytically, particularly where there are more than two to be added.
In adding vectors by the Resolution Method, each one is first resolved into
rectangular components, usually in the horizontal and vertical directions;
10 MECHANICS Chap. 1

next the horizontal and the vertical components are added separately; and
finally these values are combined at right angles to form the resultant. The
diagrams need not be drawn to scale, since the lengths are determined acru-
rately by trigonometric methods.
To illustrate this procedure, suppose that it is desired to find the resultant of four
forces acting on a single body. These are shown in Fig. 11, where A = 50 lb, acting

8=401b
-:z:-
--
400

C=30 lb D=25lb

Frn. 1-11. Forces A, B, C, and D to be added by the Resolution Method.

upward to the right, at an angle of 20° with the horizontal; B = 40 lb, acting directly
upward; C = 30 lb, acting downward to the left at an angle of 40° with the horizontal;
and D = 25 lb, acting toward the right. The horizontal and vertical directions will
be used for the components, since two of the forces already have these directions and,
for that reason, will require no resolution. Thus, A (50 lb) is composed of a horizontal
component 50 cos 20° and a vertical component 50 sin 20°, as shown in Fig. 12;

50 sin 20°= 17.1 30 cos 40°=23.0

37.8 lb

t§?J 50 cos 20°= 47.0

30 sin 40°
=19.3

Frn. 1-12. Forces resolved into rectangular com- Frn. 1-13. Final addition.
ponents.

B (40 lb) is vertical and hence has no horizontal component; C (30 lb) consists of a
horizontal component 30 cos 40° and a vertical component 30 sin 40° as shown; and
D (25 lb) is horizontal. The components are tabulated and added as follows:
Components
Force Up Dovrn Right Left
A 17.1 0 47.0 0
B 40.0 0 0 0
c 0 19.3 0 23.0
D 0 0 25.0 0

Totals 57.1 19.3 72.0 23.0

19.3 23.0

37.8 49.0
The result of the addition is a force of 37.8 lb upward and one of 49.0 lb to the right;
these are combined as shown in Fig. 13 and give the resultant of the four forces
§ 1-12 MOMENT OF A FORCE; TORQUE 11

A, B, C, and D as R = v (37.8) 2 + (49.0) 2 = 61.9 lb, directed upward toward the

right at an angle() = tan- 1 (37.8/49.0) = tan-1 0.772 = 37.7° with the horizontal.
1-11. Vector Addition in Three Dimensions. The methods used in the
preceding sections for determining the resultant of vector quantities are now
extended to more than one plane. In the diagram of Fig. 14, the axes OX,
OY, and OZ represent the intersections of three planes at right angles to one

YI
I

x
z/
/
/

Fm. 1-14. Addition of forces in three

dimensions; the resultant of A, B, and
C is R.

another, with the XZ plane horizontal; and A, B, ood C represent forces

acting along the directions of these rectangular coordinate axes. The result-
ant of any two of the forces is found exactly as before; for example, the
addition of B and C gives their resultant R1 = V B 2 + C2, which lies in the
YZ plane at the angle() = tan-1 ~ with the horizontal XZ plane. The force
R 1 is next combined with the third force A ; the plane which includes these
forces is inclined at the angle () with the horizontal, and in this plane the
final resultant makes an angle </> (phi) with the axis OX such that tan cJ, =
VB2 + C2 /A. The value of the resultant of the three forces A, B, and C is
R = VA 2 + B 2 + C2•
In more involved problems, where the quantities to be added do not lie
along mutually perpendicular axes, they can be resolved into components that
do have these directions, and the solution can then be carried forward in the
manner just explained.
1-12. Moment of a Force; Torque. In many situations, a force applied
to a body tends to turn the body about some axis. The turning effect depends
upon the magnitude of the force, its direction, and its distance from the axis
of rotation. The truth of this statement can be verified readily with a heavy
door. If the hand exerts a force upon the door knob in the various directions
shown from l to 4 in part I of Fig. 15, it ,vill be observed that a smaller force
is needed in direction 2 to open the door than in direction 3. A force along
direction 4, parallel to the door, will not produce any rotational effect, no
matter how large the force may be. The most favorable direction is that
12 MECHANICS Chap. 1

along direction 1, perpendicular to the door, for the least force is required
along this line.
Again, if the door is pushed at various places along its width in a direction
at right angles to the door, as in part II, it can be opened with the least force
applied at position 8 in line with the knob; more and more force must be
exerted in approaching the hinges, positions 7 and 6; and a force applied at
the hinges, position 5, will not turn the door, however great the force may be.

51 6
I

7
8 Least
force

Least force_........,.,.;
1 2 3
I II
Fm. 1-15. The force required to open a door varies with direction and position.

Experience in opening and closing doors, together with a variety of similar

experiences, shows the need of expressing definitely the effectiveness of a
force in turning an object about an axis. The cause that tends to produce
rotation is known as the moment of force, or torque, and is defined as the
product of the force and the perpendicular distance from the axis of rotation to
the line of action of the force. This perpendicular distance is called the lever
arm; consequently,
Torque = force X lever arm.
In symbols, the force is represented by F and the lever arm by L, and the
torque becomes
:J = FL. (1.1)

For example, in the crank shown in Fig. 16, the moment of force F tending
to turn the crank clockwise about an axis O is equal to F X OB for the position

Cranki

AxisO

FIG. 1-17. A cou-

ple acting upon a
body. The torque
FIG. 1-16. A crank subjected to is the product of F
torque. and D.
§ 1-13 CONVERSION OF UNITS 13

shown, where OB is the lever arm. It should be noted that the lever arm L
is not the length l of the crank but is the perpendicular distance from the axis
(at 0) to the line of action of the force (at B). The lever arm can, of course,
he expressed in terms of l by replacing L by its equal, l sine, thus making the
torque J = Fl sin e. Another way to obtain this result is to resolve the
force Finto two components, namely, F cos() along the crank, and F sin() at
right angles to the crank. The former component produces no rotation
(why?), and the latter component multiplied by the length l of the crank
yields the correct value of the torque, F l sin e.
Another example of torque is shown in Fig. 17, which represents a pair of
forces acting upon a body to turn it. Two equal and opposite forces that do
not act along the same line constitute a couple. The torque exerted by a
couple is equal to the product of one of the forces, F, and the perpendicular
distance, D, between their lines of action; Eq. 1.1 applies to this example also.
To prove this statement, a line can be drawn perpendicularly between the
lines of action of the forces, and an axis selected at any point on this line. For
an axis midway between the forces the moment of each force is F D /2, and
since both moments have the same direction (counterclockwise in this exam-
ple) the total moment will be their sum, or FD, as stated. The same result
will be obtained wherever the axis is selected.
The units for torque depend on the units chosen for the force and for the
lever arm; units such as pound· inches (lb· in.) and pound· feet (lb· ft) are com-
monly used.
1-13. Conversion of Units. The surest way to detect errors in calculations
is to carry the units of all physical quantities throughout, in order to verify
the unit of the result. In this process the units may be canceled, multiplied,
or divided as though they were numbers.
For example, it may be desired to convert to centimeters a reading of! in.
taken on a measuring scale, by using the fact that 1 in. = 2.54 cm. The
fraction 2.54 cm/1 in. has a value of unity, since the numerator and denomi-
nator are equal, and the specified reading can be multiplied by this factor
without altering its value; thus,
!. = !. X 2.54 cm = 2.54 cm = 0 318
8 Ill· 8 ~· 1 in..... 8 · cm .

Again, if it is desired to find the number of meters in a mile by using the fact
that 1 m = 39.37 in., three factors are used, each of which has the value of
unity, and the solution is given by

1 . _ 1 . 5280 fx 12 ~ 1m _ 1
m1 - Hl,!, X 1 m.i X 1 ~ X 39 _37 i~. - 609 m.

This procedure may seem unnecessary for such simple conversions, but in
more involved computations, many of which will be met later, it is almost the
only means of being sure that the result is expressed in the correct unit.
14 Af ECIIANICS Chap. 1

1-14. The Solution of Problems. Numerical problems form an essential

part of a course in Physics, because they show the student how well he has
succeeded in grasping the principles of the subject .. Compliance with a few
simple rules will make the problem-work easier and more meaningful.
(a) Read the problem carefully and find out exactly what is wanted. If
the result cannot be attained directly, try to break down the problem into
parts that can be solved separately. (b) Sometimes the data seem insuffi-
cient; in this event assign a symbol to the missing quantity and proceed with
the solution; the symbol that was assigned will either disappear by cancella-
tion or will become a necessary part of the answer. (c) If more data are
available than necessary-this is often true in actual engineering problems-
disregard those that are not needed. (d) Make a practice of carrying the
units through the entire solution, and be sure that the computation gives the
correct unit for the answer.
Arithmetical work should be carried out to a number of places consistent
with the data provided and the nature of the problem. Most problems should
be solved with a standard 10-in. slide rule; this instrument will give results to
three or four figures-that is, to a precision between -lo and i of 1 per cent.
Where data are supplied to many places (as in dealing with the wavelength
of light), or where the effect under observation is very small (as in the expan-
sion caused by heat), the number of figures to be kept in computations should
be correspondingly increased, but the result should never be expressed to more
figures than are justified by the data.
For example, in computing the area of a rectangle that is measured as
2.26 in. long and 1.89 in. wide, it should be noted that these dimensions are
expressed to the nearest 0.01 in. and consequently are precise to about 1 part
in 200, or! of 1 per cent. Since the result cannot be more precise than the
data from which it is obtained, the area should not be expressed as 4.2714 in.2,
although this value is correct arithmetically, but rather as 4.27 in. 2 It is
misleading to state the result with more figures, as these would imply a pre-
cision not obtainable from the data provided.
In stating problems in this textbook, numerical quantities are often given
to only one or two digits, such as 7 ft or 0.12 mm. This is done in order to
lessen mathematical detail and to avoid distraction from the physical prin-
ciples involved. Such data should be regarded as precise to three or four
figures-that is, to the number of figures that can be read on a 10-in. slide rule.

PROBLEMS
1. An acre contains 43,560 ft2. If a square tract of land measures 1 mi on each
edge, how many acres does it contain?
2. A fathom is a length of 6 ft. Express this length in centimeters and in milli-
meters.
3. A tower casts a shadow on level ground and the angular elevation of the tip of
the tower is 43° as measured from the end of its shadow. At a point 50 ft farther away
 PROBLEMS 15

from the tower the angular elevation of the tip is 30°. Compute the height of the
tower.
4. To determine the width of a stream a datum line 100 ft long is laid off along one
bank, and from the ends of this line an object on the opposite bank is sighted. The
lines of sight make angles of 32° and 47° with the datum line; what is the width of
the stream?
5. An airplane in still air travels 200 mi south and then 140 mi southwest. Com-
pute its distance from the starting point and its direction from that point.
6. A boat in still water sails 3 mi east, then 4 mi northeast, and finally 2 mi north.
How far is the boat from the starting point, and along what direction is it from that
point?
7. Find the resultant of two forces, one of 10 lb vertically downward and the other
of 8.66 lb upward toward the right at 30° with the vertical.
8. A is a force of 1000 lb directed horizontally to the right, and B is a force of
800 lb directed upward toward the right at 60° with the horizontal. (a) Add the
forces A and B. (b) Subtract force B from force A.
9. An object is acted upon by two forces: one, 2500 lb vertically downward and the
other, P, acting upward toward the left at 70° with the horizontal. The resultant R
of these forces is known to be horizontal. . Find the value of the force P and that of
the resultant R.
10. The velocity of a body, originally 10 cm/sec directed horizontally toward the
left, is to be changed to one of 10 cm/sec directed toward the left at an angle of 20°
downward from the horizontal. What additional velocity must be given to the body
in order to produce this change?
11. An object is displaced 1800 ft in a direction 40° north of west. Resolve this
displacement into two components: N, directed north, and W, directed west, and
determine their values.
12. A 500-lb force acts vertically downward. Resolve this force into two compo-
nents: A, inclined 30° from the vertical, and B, inclined 60° from the vertical.
13. A sled drawn along level snow is acted upon by four forces: (a) its weight,
40 lb, vertically downward; (b) the applied pull, 10 lb, upward toward the right at
30° with the horizontal; (c) the vertical push of the snow, 35 lb, directed upward; and
(d) the force of friction, 4 lb, horizontally toward the left. Compute the resultant
force acting upon the sled.
14. Four forces act upon a body: (a) 10 lb horizontally toward the left, (b) 20 lb
vertically downward, (c) 30 lb upward toward the right at 30° with the horizontal, and
(d) 40 lb downward toward the right at 45° with the horizontal. Determine the
resultant force acting upon the body, by the resolution method.
15. A ship in still water travels successively 5 mi north, 12 mi east, 8 mi southeast,
and 9 mi in a direction 60° south of west. Find the resultant displacement of the
ship by the resolution method.
16. Find the resultant of the following displacements by the resolution method:
(a) 400 ft north, (b) 200 ft 30° north of east, and (c) 300 ft 45° south of east.
17. When a boy pushes upon the pedal of a bicycle, the pedal crank being 6 in.
long, what torque is exerted about the center of rotation when the crank makes an
angle of 40° with the horizontal, if at that instant the force is 20 lb vertically down-
ward?
18. A yardstick is mounted so that it can rotate in a vertical plane about a fixed
pivot at its midpoint. The stick is placed in a horizontal position and a force is
exerted upon it at the 6-in. mark, acting in the plane of rotation. If this force amounts
to 12 lb and is directed 30° downward from the horizontal, what force applied vertically
downward at the 33-in. mark will prevent rotation about the pivot?
 2
STATICS
The subject matter of Statics has to do with bodies at rest and with the
requirements that must be fulfilled if those bodies are to remain at rest. No
other branch of Physics contains so few principles or so many different appli-
cations. In Statics, the study of vectors is continued, and extensive use is
made of force and torque. Wherever referred to in this chapter, forces are
assumed to act in a single plane.
2-1. Requirements for Equilibrium. A body that continues in a state of
rest is said to be in equilibrium. A stone lying on the ground, a girder sup-
porting a structure, or a picture hanging on the wall: these objects are not in
motion, and since they are standing still they are in equilibrium.

~Axis

I
Fm. 2-1. Distinction between motion of I, translation; and II, rotation.

Broadly speaking, motion is of two kinds: motion of translation and motion

of rotation. In translation, any straight line of the moving body remains
parallel to its original position, as represented in part I of Fig. 1. A very
simple example of translation is the motion of a block sliding along a plane
surface. In rotation, the body turns about some fixed line as an axis, as
represented in part II of the figure. A simple example of rotation is the mo-
tion of a wheel turning in stationary bearings. Often the motion of a body is
a combination of translation and rotation-for example, a wheel rolling along
the ground or a stick thrown through the air.
A body initially at rest would be set into motion either by an unbalanced
force or by an unbalanced torque; an unbalanced force would produce trans-
16
§ 2-2 FIRST CONDITION OF EQUILIBRIUM 17

lation and an unbalanced torque would produce rotation. It follows, there-

fore, that two conditions must be satisfied if a body is to be in equilibrium:
first, the forces acting upon the body must balance since there is to be no
translation; and second, the torques acting upon it must balance since there
is to be no rotation.
In Chap. 4 it is shown that a body that moves along a straight line at con-
stant speed is also in equilibrium, but the present treatment deals only with
bodies that are stancling still.
2-2. First Condition of Equilibrium. The simplest example of equilibrium
is that of a body which stays at rest when acted upon by only two forces.
Here one of the forces must be equal and opposite to the other, and both must
act along the same line; a plumb bob suspended by a ~/~~~~~
cord as in Fig. 2 will serve as an illustration. In the
wctor diagram at the right, 0 represents the bob, the T
force TV is the attraction of the earth pulling it down-
ward, and the force T is the tension in the cord pulling
0
it up,rnrd. Since the bob is in equilibrium, the up-
ward pull of the cord is equal and opposite to the
downward pull of the earth; that is, the forces T and w
lV are equal and opposite. Frn. 2-2. The sus-
When a body is in equilibrium under the action of pended bob is acted
three forces, the resultant of any two must be equal and upon by two equal and
opposite forces.
opposite to the third force. This principle can be used
to find the forces that are needed to support a body. To illustrate, a ball
of weight W is supported from the ceiling by two cords as shown at I in
Fig. 3; the cords a and b make angles of a (alpha) and (3 (beta), respectively,

0 I
w
II
FIG. 2-3. For equilibrium of three forces, any one must be equal and opposite to the
resultant of the other two.

with the ceiling, and are knotted at O to the vertical cord c. The tension in
this cord is equal to the weight of the ball, as in the previous illustration, and
the tensions in the other cords can be determined. The knot is in equilibrium
under the action of three forces, as represented in part II of the figure; these
18 MECHANICS Chap. 2

are the weight W of the ball directly downward, the tensjon in cord a (un-
known) along the direction of that cord, and the tension in cord b (also un-
known) along the direction of that cord. Since the forces must balance, the
tensions in a and b form a resultant W' that will be equal and opposite to W;
hence, W' can be drawn and a parallelogram constructed, giving A as the
tension in cord a and B as the tension in cord b. The values of A and B can
be found graphically by constructing a vector diagram accurately to scale, as
shown, and measuring the corresponding lengths, or they can be found ana-
lytically by the Law of Sines. It should not be thought that the shorter cord
has the smaller tension; in fact, this cord is more nearly vertical and hence
supports the greater part of the weight.
To illustrate the analytical solution, suppose that in the foregoing example the
weight of the ball is W = 10 lb and that the angles which the cords make with the
ceiling are a = 45° and /3 = 60°. For clarity, the parallelogram of part II is redrawn
to larger scale in part III of the figure, and the angles are indicated. The obtuse
angles are seen to be 105° each. From the Law of Sines it follows that

-A- -
sin 30°
W'
sin 105°
.
But W and W' are numerically equal, and therefore the tension in cord a is

A = 10 lb X sin 30° = 5 _18 lb;

cos 15°
similarly the tension in cord bis B = 7.32 lb. Thus, in supporting the 10-lb weight
in the manner described, the cords a and bare subjected to tensions of 5.18 and 7.32 lb,
respectively.

The illustrations point to a general conclusion regarding the equilibrium of

a body under the action of forces that individually tend to give it motion of
translation. Since there is to be no translation, the first condition of equi-
librium requires that the vector sum of all the forces acting on a body must equal
zero.
2-3. Application to Components of Forces. The study of a body that is in
equilibrium can often be simplified by resolving all of the forces which act
upon it into components, and considering the components instead of the forces
themselves. The forces can be resolved along any rectangular axes that may
be desired; in the following treatment the horizontal and vertical directions
are chosen. When a body is in equilibrium, it must, of course, be in equilib-
rium along the horizontal direction, which requires that the horizontal com-
ponents of the forces acting upon it must balance. If components to the
right are regarded as positive and those to the left negative, the algebraic sum
of all the horizontal components must be zero. Similarly, the body must be
in equilibrium along the vertical direction, and therefore the algebraic sum of
all the vertical components must be zero. These facts can be stated in sym-
bols by using ~ (sigma) as a sign of summation, and expressing the sum of
§ 2-4 SECOND CONDITION OF EQCILIBRIUM

the horizontal components as~ Hand the sum of the vertical components as
~ V; then the first condition of equilibrium takes the form:

~H = O; ~ V = 0. (2.1)
To illustrate this procedure the problem of the ball supported by the two cords
shown in Fig. 3 is worked out by resolving the forces exerted by the inclined cords
into horizontal and vertical components, and applying Eqs.
2.1. This resolution, indicated in Fig. 4, shows that the com-
ponents of tension A are A cos 45° horizontally and A sin 45°
vertically, and that those of tension B are B cos 60° horizon-
tally and B sin 60° vertically. All four components act at
point 0, which represents the knot where the weight Wis sup-
ported; the forces A and B are crossed out, since they are
completely replaced by their components. The force W is
vertical and has no horizontal component. Then
~ H = B cos 60° - A cos 45° = 0, Bcos 60°
and ~ V = B sin 60° +A sin 45° - W = 0,
whence 0.500 B = 0.707 A, and 0.866 B + 0.70} A = 10 lb.
These equations, when solved simultaneously, give A = 5.18
lb and B = 7.32 lb. These values for the tensions in the W
cords agree with those previously found. Fm. 2-4. Use of
components in the
The foregoing principles, summarized by Eqs. 2.1, can study of equilibrium.
be extended without change in method to bodies that
are in equilibrium under the action of any number of forces.
2-4. Second Condition of Equilibrium. Even though the forces acting
upon a body are such as to satisfy the "first condition" just described, it does
not necessarily follow that the body will be in equilibrium, for the forces may
be applied in such a way as to produce rotation. Thus, a body may be acted
upon by two equal and opposite forces that form a couple, § 1-12, a situation
which satisfies Eqs. 2.1 that~ H = 0 and~ V = 0, and yet this body will not
be in equilibrium, because the torque or moment of the couple will tend to
make the body rotate. To overcome this tendency, the couple must be
opposed by an equal and opposite couple. Hence, if a body is standing still,
not only must the first condition of equilibrium be satisfied, or the body
would be given motion of translation; but also a second condition must be
met, or it would be set into rotation.
The second condition of equilibrium requires that the torques.or moments of
the forces acting upon a body shall balance, the clockwise torques being equal to
the counterclockwise torqttes. If clockwise torques are taken as positive and
counterclockwise torques as negative, the second condition requires that the
algebraic sum of the torques acting upon the body shall be zero; symbolically,
~ :J = 0. (2.2)
When the forces which maintain equilibrium all act through a common
point-these are called "concurrent" forces-their moments are necessarily
20 MECHANICS . Chap. 2

balanced and the first condition of equilibrium is sufficient, but where they
do not, the second condition must be satisfied also.
To apply this condition of equilibrium to a specific problem, suppose that a bar 10 ft
long supports loads of 20 lb and 30 lb at its ends, and that it is desired to find the
location of a fulcrum upon which the loaded bar will balance, as shown in part I of
Fig. 5. In order to simplify the problem, the weight of the bar will be neglected.
The first condition shows only that the fulcrum must exert an upward force of
20 + 30 = 50 lb upon the bar and does not help directly in the solution of the problem.

_,:- x lOft -J
>I( 10-x--i /~<--x--~c---10-x ---j

Fulcrum
20
30lb 30
50
I Il
Fm. 2-5. Positioning the fulcrum of a bar for equilibrium.

Draw the forces acting upon the bar as in part II of the figure, and show the position
of the fulcrum at a distance x from the end where the 20-lb load is applied. Then
determine the moments of all the forces about the fulcrum and equate their algebraic
sum to zero. Thus, the moment of the 20-lb force is 20 x and that of the 30-lb force
is 30 (10 - x), both in pound· feet. The 50-lb force has no moment about this axis,
since its lever arm is zero. Hence, to satisfy the second condition of equilibrium,
30 (10 - x) - 20 x = 0,
from which x = 6.00 ft. This is the distance from the 20-lb load at which the fulcrum
should be located to have the bar remain in balance.
Although it was natural in the foregoing problem to choose the fulcrum as
the axis about which to take moments, any other point might have been
selected equally well, provided the same point was used throughout the
cal cu la tion.
2-5. Theorem of Three Forces in Equilibrium. A theorem of great value,
based on the second condition of equilibrium, states that, if three nonparallel
forces acting upon a body produce equilibrium, their lines of action must pass
through a common point. This theorem will be demonstrated by taking
moments of the forces about the intersection of two of the lines of action and
showing that the lever arm of the third force must be zero. Figure 6 repre-
sents a body in equilibrium under the action of the forces F1, F2, and F3; for
the purposes of this demonstration the forces are shown incorrectly in the
figure. Since their lines of action are not parallel, any two, if extended, will
meet at some point. Thus, the lines of action of F1 and F2 meet at 0. From
this point a line OA is drawn perpendicular to the third force, F3, meeting its
line of action at A. To satisfy the second condition of equilibrium, ~ 3 = 0,
§ 2-0 RESULTANT OF PARALLEL FORCES 21

the moments of all the forces are taken about 0, and their sum is equated to
zero. The moments of F1 and F2 are both zero since the lines of action of
these forces pass through 0, and thus their lever arms are zero. But the
body, by definition, is in equilibrium, and therefore the moment of the
remaining force, F3 , about O must be zero. Since the force F3 itself is not
zero, its lever arm OA is necessarily zero; that is, the line of action of F3 has
to pass through 0. Hence, the lines of action of these forces must pass
through a common point.
This theorem is often useful in showing the line along which an unknown
force must act in order to produce equilibrium.

A~{_o
I I'
I I
I I
I I
I I
I

Frn. 2-6. For three non- Frn. 2-7. Parallel forces

parallel forces in equi- in equilibrium. The re-
librium, their direction sultant of F1 and F2 is
lines must meet at a equal and opposite to Fa
common point. and acts along the same
line.

2-6. Resultant of Parallel Forces. It is pointed out in §2-2 that when a

body is in equilibrium under the action of three forces, the resultant of any
two of these forces is equal and opposite to the third force. This principle
can be used in finding the resultant of two parallel forces. The body repre-
sented in Fig. 7 is in equilibrium under the action of the three parallel forces,
F1, F2, and F3. From the first condition of equilibrium, F3 = F1 + F2; the
resultant R of the forces F 1 and F 2 is equal and opposite to F 3• From the
second condition of equilibrium, with moments taken about 0, it follows that
F2 X2 - Fi X1 = 0. This equation indicates an important relation between
the two forces and their lever arms, namely,
F2 X1 (2.3)
F1 = X2 •

Hence, the resultant of two parallel forces has the same direction as the forces
and is equal to their sum, and its line of action divides the distance between
them into t,rn parts inversely proportional to the respective forces.
22 MECHANICS Chap. 2

This principle can be extended to find the resultant of any number of paral-
lel forces. First, the force is determined which, included with the forces
given, will produce equilibrium; such a force is called the equilibrant of the
given forces, and the resultant is equal and opposite to the equilibrant and acts
along the same line. .
Under certain conditions a body which is acted upon by several parallel
forces is held in equilibrium by two or more forces instead of by a single
equilibrant. These forces can usually be found by applying the conditions
of equilibrium.
As an illustration, consider a rod 12 ft long supported by two cords and carrying four
loads totaling 30 lb, as shown in Fig. 8. Neglect the weight of the rod and find the
tensions A and B in the supporting cords.

12 ft A
--;. l 10 B

3 3 -6 3 3 6
4
10 9 7
41b lib
lOlb 9lb
I IT
Frn. 2-8. W Pighted rod supported by two rords. The distances are indicated in feet.

From the firnt condition of equilibrium,

~ V = A +B - 10 - 4 - 9 - 7 = 0,
whence A + B = 30. From the second condition of equilibrium, with moments
taken about the left end of the rod,
2": J = 4 X 3 + 9 X 6 + 7 X 12 - A X 1 - B X 11 = 0,
or A + 11 B = 150. In these equations, the forces are in pounds and the lever arms
in feet. The equations, when solved simultaneously, give the tensions in the cords
as A = 18 lb and B = 12 lb.
It is interesting to note that this problem can also be solved by two applications
of the condition 2": J = 0. With moments taken about the left end of the rod,
A +11 B = 150, as before, and with moments taken about the right end,
A X 11 +B X 1 - 10 X 12 - 4 X 9 - 9 X 6 = O,
or 11 A + B = 210. From these equations the tensions in the cords are found to be
A = 18 lb and B = 12 lb, as before.
If the rod of the foregoing problem were rested on props instead of being
suspended by cords, and if these were located at the same places as the cords,
the props would divide the total load between them in the same way; each
prop would be under compression and would exert an upward force on the rod.
§ 2-7 CENTER OF GRAVITY 23

2-7. Center of Gravity. The earth's attraction for a body extends to each
particle of matter that the body contains, and thus the weight of a body may
be regarded as a system of parallel (or very nearly parallel) forces acting upon
the individual particles of which the body is made up. For a given object
there is a particular point about which the object will balance, whatever its
orientation. This point is called the center of gravity of the object and is
based on the expression ~ w x = 0, where w is the weight of an individual
particle, and x is the lever arm extending from the force w to the center of
gravity of the object. The resultant of the individual weight forces, as well
as their equilibrant, passes through the center of gravity. Hence, the weight
of a body, although actually a system of individual forces acting upon its
component parts, can be correctly represented by a single force acting down-
ward at the center of gravity.
As an illustrative problem, consider the forces on a truck having a 160-in. wheel base.
The empty truck weighs 9000 lb, and its center of gravity is 74 in. in front of the rear
axle. The truck carries a load of 11,000 lb that is placed centrally with its center of
gravity 34 in. in front of the rear axle. \Vhat force does each tire exert upon the
ground?

r--160 in.---,
34 40 86 --x
9000 lb
11,000 lb
R1
20,000 lb
II ill
Fm. 2-9. Forces acting on a loaded truck.

The truck with load in place is represented in part I of Fig. 9 and the force diagram
is shown in part II. The weights of the truck and the load are shown acting down-
ward, each at its own center of gravity; the upward thrust of the ground on the two
rear tires is represented as a single force R1; a similar thrust on the front tires is repre-
sented as R2. From the first condition of equilibrium, R1 + R2 - 9000 - 11,000 = 0,
whence R1 = 20,000 - R2. From the second condition of equilibrium, with moments
about the rear axle, 11,000 X 34- + 9000 X 74' - R2 X 160 = 0, from which R2 = 6500.
Finally, R1 = 20,000 - 6500 = 13,500. In all
these expressions, the forces are in
pounds and the lever arms in inches. The forces exerted by the tires on the ground
are equal to the forces with which the ground reacts on the tires; hence, each rear tire
presses against the ground with a force of Ri/2 = 6750 lb and each front tire presses
with a force of R2/2 = 3250 lb.
It may also be of interest in this problem to find the center of gravity of the loaded
truck. This can be done by considering the combined weight of 20,000 lb to act down-
ward at a single point distant x from the rear axle, as shown in part III. With mo-
ments taken about the rear axle, 20,000 x - 6500 X 160 = 0, from which x = 52.
Hence, the center of gravity of the loaded truck is 52 in. in front of the rear axle.
The center of gravity of a symmetrical body is at its geometrical center if
the material of the body is assumed to be of uniform density throughout.
24 MECHANICS Chap. 2

Thus for a rod, tube, or bar, whether of circular or rectangular cross-section,

the center of gravity is located on the axis and midway between the ends; for
a sphere it is at the center; for a cone it is on the axis at a point one-fourth of
the way from the base to the vertex. When a body is in the form of a thin,
uniform sheet, the center of gravity is located at the center of area of the
sheet. For a rectangular sheet, the center of area is at the intersection of the
diagonals, and for a triangle it is one-third of the distance from the middle
point of any side to the opposite vertex. For an unsymmetrical sheet, the
center of gravity can be located by applying the principles of equilibrium, as
in the following problem.
An L-shaped sheet is formed by cutting a square from one corner of a square sheet,
as shown in part I of Fig. 10. The entire square has a length L on each side, and the

I
I
10

l II
Fm. 2-10. Finding the center of gravity of an irregular sheet.

part removed has a corresponding length L/2; their respective centers are at O and C',
and their relative weights are 4 and 1. After the small square has been removed, the
£-shaped remainder will have a weight of 3; suppose this to be centered at C, a distance
x from 0. From the force diagram in part II of the figure, with moments taken
about 0,
p p 3 x - 1 X a= 0.

Hence, x = a/3 = LV2/12, a result which locates

the center of gravity of the £-shaped sheet.
The center of gravity of an irregular solid
can be located experimentally by suspending
the body successively in three different posi-
I II tions. In each test the body will come to rest
Fm. 2-11. Experiment to locate with its center of gravity on a line extending
the center of gravity. The object vertically downward from the point of sus-
pivoted at P rotates until its cen- pension; the intersection of the three lines is
ter of gravity G is directly below
this point. the center of gravity. A plane object can be
pivoted at two points and tested similarly; in
Fig. 11 a sheet cut to represent the cross-section of a retaining wall is under
test in this manner. 'When pivoted at P and displaced from its position of
§ 2-9 THE BALANCE 25

rest, as shown in part I, its weight W sets up a torque which will rotate
the model until its center of gravity G is directly below the pivot, as in part
II. After the vertical line through P is determined, the model is pivoted
about another point and the test is repeated. The intersection of the two
lines locates the center of gravity.
2-8. Conditions of Stability. The way in which a body is supported with
respect to its center of gravity has a great effect upon the stability of equilib-
rium of the body. A cubical box at rest on a level floor, as shown in part I of
Fig. 12, illustrates stable equilibrium; if one edge is raised a little and released,
the box falls back again to its original position. The tilting of the box shifts
the reaction R of the floor to the edge S, and the unbalanced torque set up
by the weight W of the box acting at the center of gravity G produces rota-
tion and returns the box to its initial position. The same effect is produced
in a suspended body hung from a point above its center of gravity, as men-
tioned in the preceding section.

w w
w
I II m
Frn. 2-12. Types of equilibrium: I, stable; II, neutral; and III, unstable.

A ball resting on a level surface, as in part II of the figure, is in neutral

equilibrium. No matter how the ball is moved on the surface, it shows no
tendency either to return to its former position or to go the other way. Its
weight Wand the reaction R of the surface lie along a single line and there-
fore do not produce a couple; the same is true for a cylinder lying on its side.
The third part of Fig. 12 represents a cone (not revolving) in unstable
equilibrium while momentarily balanced upon its vertex. It is evident that
if the cone is given a slight displacement its weight W will produce a torque
about the vertex S which will increase the displacement and cause the cone
to fall over.
2-9. The Balance. The equal-arm analytical balance offers a useful study
in equilibrium. In this instrument, shown diagrammatically in Fig. 13, a
long slender beam of length 2 L supports a scale-pan at each end and is pivoted
at its midpoint. At this point it carries a downwardly projecting pointer, the
lower end of which plays across a fixed scale. The moving system has a
weight W and, being symmetrical, has its center of gravity located at some
point C along the pointer, a distanced below the pivot. This distance is of
26 MECHANICS Chap. 2
'
importance in the operation of the balance and is exaggerated in the figure
for clearness. The dotted lines represent the beam and pointer when not
deflected, and the solid lines show the result of adding a very small weight p
to the right-hand scale-pan. This action sets up a deflecting torque about the
pivot equal to p(L cos e), which becomes smaller as() increases. At the same
time, the center of gravity is displaced; this produces an opposing torque
about the same axis equal to W(d sin 0) which becomes larger as () increases.
Evidently the balance will come to rest at some angle () such that
p (L cos e) = vV (d sin e).

A balance fa sensitive if the pointer is displaced through a large angle when

the weight added to one scale-pan is very small, and for this reason its sensi-
tivity is usually expressed as the ratio of the deflection in scale divisions to the

Beam

I
el
I
wl
Scale-,J//1,j;f/1i1/1_/_i!J I\\\\1\1\~\\\
Frn. 2-1:3. The equal-arm balance.

added weight in milligrams. Physically, the sensitivity of a balance is defined

as the ratio of the deflection () (or its practica.l equivalent, tan e) to the load p;
thus,
. . .
S ens1trnty tan () sin () L
= -- = - --·
p p cos() - w d

The equation sho,vs that, for a balance to have great sensitivity, its arms
should have considerahle length L, the beam should have a small weight 1V
-which usually implies a truss-like design to obtain the necessary rigidity-
and the center of gravity of the moving system should be only a short dis-
tance d below the point of support. The moving system sometimes carries,
near the pivot, an adjustable ,veight by means of which the center of gravity
can be raised or lowered; this adapts the balance to measurements requiring
greater or less sensitivity.
2-10. Applications of the Conditions of Equilibrium. In illustrating the
principles of statics, the problems in this chapter have been solved by follow-
ing a definite pattern. It will be well for the student to use the same pro-
§ 2-10 APPLICATIONS OF THE CONDITIO.V8 OF EQUILIBllIUM 27

cedure, especially in the handling of the more complicated problems. The

method recommended is outlined as follows:
L Make a sketch of the device or structure, mark upon it all the known
data, and assign symbols to the desired quantities.
2. Select some member or point of the body and make a separate force dia-
gram showjng all the forces acting v,pon the portion considered.
3. Apply the conditions of equilibrium to this portion by writing the equa-
tions expressing ~ H = 0, ~ V = 0, and i J = 0. The moments must all
be taken about a single axis. If another equation is required, the moments
may all be taken about some other axis.
4. Solve these equations for the quantities desired.
5. Follow the same procedure for the other points or members involved in
the problem.
The procedure will be illustrated by two additional problems, dealing with
the equilibrium of a ladder and a derrick.
I. A uniform ladder, 18 ft long and weighing 90 lb, leans against a vertical wall at
an angle of 70° with the ground. Assume the wall to be smooth (that is, frictionless),
and find what forces the ladder exerts against the wall and ground.

I h
--p-----
B,
I ,::::
I 0.
I
18ft,90lb I

ladder v

II

Fm. 2-14. Analysis of forces on a ladder leaning against a smooth wall.

The ladder is shown in position in part I of Fig. 14, and the forces acting upon it
are shown in part IL Herein the weight of the ladder is represented as a single force
acting vertically downward through the center of gravity, as discussed in § 2-7.
Force h is the force exerted by the wall upon the ladder and is equal in magnitude,
but opposite in direction, to the push of the ladder against the wall. This reaction
force h is perpendicular to the wall because that surface was assumed frictionless, and
a force along (that is, parallel to) the wall could exist only if there were friction.
Force v is the vertically upward reaction of the ground, and f is the frictional force at
the ground; the latter acts opposite to the direction in which the ladder tends to slide.
,-rrite the equations for equilibrium, with forces in pounds and lever arms in feet.
In the torque equation, take moments about the foot of the ladder, and use the lever
arms indicated in part II of the figure. The equations follow;
28 MEOlfANlCS Chap. 2

~If= f - h = O;

~ -V = v - 90 = O;
~ 3 = 90 (OA) - h (OB) = O;
= 90 (9 cos 70°) - h (18 sin 70°) = 0.
Solve these equations simultaneously and obtain: v = 90 lb; also, f = h = 45/tan 70°
= 16.4 lb. Hence, the ladder pushes against the ground with a force having a hori-
zontal component of 16.4 lb and a vertical component of 90 lb. The resultant push
of the ladder against the ground is therefore V (16.4) 2 + (90) 2 = 91.5 lb, and the
angle that this force makes with the ground is tan-1 1~~4 = 79.7°. Note that this
force is not directed along the ladder, which makes an angle of 70° with the ground.
Although moments could be taken about any point, the foot of the ladder was
selected because two of the forces have lines of action which pass through this point,
and accordingly the moments of these forces are zero.
II. A derrick consists of a uniform boom hinged to a vertical mast and supported
at the outer end by a cable. The arrangement and angles are shown in part I of
Fig. 15. The boom is 24 ft long, weighs 700 lb, and supports a load of 5000 lb at its
outer end. Compute the tension in the cable, and also the horizontal and vertical
thrusts on the hmge-pin.

h
700 0
B

5000lb 5000
%
II m
Fm. 2-15. Diagram for calculating forces acting in a derrick.

Select the boom as the body in equilibrium and make a force diagram as in part II.
The weight of the boom is considered as a single force acting at its center of gravity,
the tension in the cable is designated P, and the horizontal and vertical reactions
on the boom due to the thrust of the hinge-pin are marked h and v, respectively.
Write the equations for equilibrium; in the torque equation, take moments r,,bout the
hinge-pin 0, and refer to part III of the figure for the lever arms of the forces. The
equations follow, ,vith forces m pounds and lever arms in feet:
~ II = P cos 30° - h = 0;
~ V = P sm 30° +v- 5000 - 700 = O;
l: 3 = P (OA) - 5000 (OB) - 700 (OC) = 0,
= P (24 sin 45°) - 5000 (24 sin 75°) - 700 (12 sin 75°) = 0.
From these equations, it is found that P = 7310, h = 6330, and v = 2045. Thus,
the tension in the cable is 7310 lb, and the horizontal and vertical thrusts on the hinge-
pm are respectively 6330 and 204,5 lb. The resultant force acting on the lower end
§ 2-10 APPLICATIONS OF THE CONDITIONS OF EQUILIBRIUM 29

of the boom could be computed if desired; its value is V (6330) 2 + (2045) 2 = 6660 lb,
and it acts at an angle 8 with the vertical such that tan 8 = ~= ;~:~ = 3.10, whence
8 = 72°.
The principles of equilibrium find a wide application in the design of struc-
tures of all kinds. The designer must first determine the forces in the various
members of a proposed structure, so that he may know how large to make
their cross-sections in order to provide sufficient strength.

PROBLE1lJS
1. A 100-lb weight is supported by two ropes: a, 3 ft long and b, 4 ft long; these
ropes are fastened to the ceiling at two points 5 ft apart. Compute the tension in
each rope.
2. In order to suspend a given load, a rope 20 ft long is tied between two walls
16 ft apart, the support at one wall being 6 ft higher than the other. At what point
on the rope should the load be tied in order that both segments of the rope will be
subjected to the same tension?
3. The diagram shows a plan view of a force table having a hole at the center in
which a pin fits loosely. Three cords are tied to the pin: two are shown at a and b,
and the third one c is not shown. If the tension in cord a is 7 lb and that in cord b
is 11 lb, what direction must the cord c have, and what must be the tension in it, to
keep the pin in equilibrium?

270°

180°
w E

b s
PROBLEM 3. PROBLEM 5.

4. A wire is stretched horizontally between fixed supports 40 ft apart. A branch

weighing 150 lb breaks off of a tree and falls on the wire at its midpoint, causing that
point to sag 3 ft. Compute the tension in the wire.
5. The diagram shows a pole to which are attached two cables a and b and a guy-
wire c. The cables extend northeast and southeast and may be regarded as hori-
zontal; the tension in each is 300 lb. The guy-wire is directed to the west at an angle
of 30° with the pole. To what tension should the guy-wire be stretched in order for
it to balance the horizontal pulls on the pole by the cables?
6. A slender pole AB, 10 ft long, rests on two props placed 2 ft from end A and
3 ft from end B. The pole supports a weight of 10 lb at A and 20 lb at B. Neglect
the weight of the pole and compute the reaction forces exerted by the props.
7. A yardstick loaded with 2 lb at the O mark and 5 lb at the 20-in. mark is placed
on a fulcrum at the 12-in. mark and is kept horizontal by applying a vertically upward
force at the 3o-in. mark. Neglect the weight of th~ stick and compute the value of
this upward force.
30 MECHANICS Chap. 2

8. Two tnen carry between them a weight of 81 lb supported from the middle of a
10-ft pole. One end of the pole rests on the rear man's shoulder, and the other end
is pushed 1 ft beyond the front man's shoulder. What part of the weight does each
man carry, if the weight of the pole is negligible?
9. In the diagram, the rectangular frame ABCD is hanging in a vertical plane
with AB and CD horizontal. The frame is acted upon by the forces indicated and
by an additional force, not shown, applied at some
9lb point along the side CD. Find the value of this force
and also its point of application in order to keep
the frame in equilibrium in the position described.
3lb The weight of the frame is to be neglected.
~10 in.- 10. A yardstick is loaded with four weights: 10 oz
ti====~r 2 lb at the 4-in. mark, 12 oz at the 10-in. mark, 20 oz at
A B the 20-in. mark, and 8 oz at the 30-in. mark. At
what mark should the stick be supported in order to
12 in. balance, if its own weight is neglected?

71b D c1 11. A body is acted upon by two vertically down-

ward forces: A, of 10 lb, and B, of 12 lb. Their lines
of action are 3 ft apart. Determine the resultant of
these forces.
41b
12. A thin square sheet measures 20 in. along each
PROBLEM 9. edge. If one-quarter of the sheet is removed by
cutting out a piece along diagonals from the center
to two adjacent corners, how far is the center of gravity of the sheet shifted?
13. A platform scale balances at 1600 lb when the front wheels of an automobile
are run on it, and it balances at 1500 lb when the rear wheels are run on it. If the
automobile has a wheelbase of 120 in., where is the center of gravity of the car with
respect to the rear axle?
14. A yardstick that weighs ! lb carries two loads: one of 3 lb at the 15-in. mark
and one of 1 lb at the 36-in. mark. At what mark should the loaded stick be sup-
ported in order to balance?
15. A uniform ladder 12 ft long weighs 40 lb. Its upper end rests against a smooth
vertical wall, and its lower end rests upon smooth level ice. The ladder is kept from
slipping by a rope 6 ft long that extends horizontally from the foot of the ladder to
the base of the wall. Compute the tension in the TOpe.
16. One end of a uniform ladder 18 ft long and weighing 60 lb rests on level ground,
and the other end leans against a smooth vertical surface. Suppose that the ladder
is inclined at 75° with the ground and that a 150-lb man has climbed up the ladder a
distance of 12 ft. Find (a) the force that the vertical surface exerts upon the ladder,
and (b) the horizontal and vertical components of the force that the ground exerts
upon it.
17. In the diagram AC represents an aluminum ladder used in servicing overhead
street lamps; it is pinned to a truck at A and is braced by a tie-rod BD pinned to its
midpoint B. The ladder AC is 13 ft long and the top C is 12 ft above E. The tie-rod
makes an angle of 45° with the horizontal. Suppose that a 150-lb man stands on the
ladder half-way between B and C, and determine the tension in the tie-rod and
the horizontal and vertical components of the pin reaction at A.. The weight of
the ladder may be neglected.
18. A 10-ft plank rests at one end on level ground, and at a place 2 ft from its other
end leans against a horizontal cylindrical rail at a point 1 ft above the ground. If
the plank is uniform and weighs 80 lb and if the rail is smooth, what force does the
plank exert (a) on the rail, and (b) on the ground?
 PROBLEMS 31

19. The diagram shows a horizontal beam hinged to an upright mast and supported
as shown. The beam BC is uniform, 20 ft long, and weighs 1000 lb. The members
AB, AC, and AD are hinged at the ends and may be regarded as of negligible weight.
Find the force in each of these members, and state for each whElther it is in tension
or compress10n.

c
I
I A
I
I
I
I
I
I
I
I
I
I
D A 1 E
====fC====~----~---
PROBLEM 17. PROBLEM 19.

20. A derrick consists of a boom hinged to the bottom of a vertical mast and sup-
ported at the outer end by a cable that extends to the top of the mast. The boom is
uniform and weighs 500 lb. Both the boom and the cable are 16 ft long and are
inclined at 30° with the horizontal. When a 2-ton load is attached to the boom at a
point 12 ft from the hinge, what is (a) the tension in the cable, and (b) the thrust of
the mast against the boom?

-~-
rI

5 Ift
1

~~=====;::\======i==::9
A ---lO~~t---- \{ ft

~) 24001b

PROBLEM 21. PROBLEM 22.

21. The diagram represents a uniform gate 6 ft wide and 4 ft high that weighs
200 lb. It is supported by two hooks as shown; the hooks engage with rings carried
by the gate and serve as hinges: Compute the force that each hook exerts upon the
gate.
22. The derrick in the diagram has a horizontal boom supported at its outer end
by a cable as shown. The boom is uniform, weighs 600 lb, and carries a 2400-lb load
at a point 2 ft from its outer end. Compute (a) the tension in the cable, and (b) the
horizontal and vertical components of the pin reaction at A.
32 MECHANICS Chap. 2

23. Find the reactions A, B, and Con a roof truss due to the wind and snow loads
shown in the diagram. The truss is fixed at the left support and is free to roll at the
right support.
3000lb
Snow l o a d /
\
3000lb ----r-1
3000 lb 5 1ft

_l 10 ft

PROBLEM 23.
 3
THE FUNDAMENTALS
OF MOTION
The preceding chapter deals with objects at rest, whereas the present one
deals with objects in motion. The concept of motion is important in Physics
and appears in all its branches. The present purpose is to introduce the basic
principles involved and to apply them to motion of simple types.

LINEAR MOTION

3-1. Motion Along a Line. The motion of a body from one place to another
along a line of some sort is of common occurrence. Thus, a crate hoisted by
a rope moves directly upward, and an automobile traveling between two cities
follows the line of the highway. In each of these illustrations the body men-
tioned moves along a line, and therefore its motion is said to be
linear.
8
During any linear motion, the moving body traverses acer-
tain path and undergoes a certain displacement, § 1-2. Length
of path and displacement often have the same numerical value, c
but not always. Thus, a ball may be thrown vertically upward
and then start to fall, as indicated in Fig. 1, and it may be
desired to study the motion between points A and C in the A
figure. The length of path traveled by the ball is AB + BC,
but the displacement of the ball has the value AC and is directed Fm. 3-L
upward. Umgth of path is a scalar quantity and may be either The length
of path is
straight or curved; displacement is a vector quantity and AB + BC;
extends in a straight line from the initial position of the body the dis_
to its final position. placement
In describing the motion of an object, a person might tell how is AC and
fast it was going-say, 25 miles per hour-this would be a state- is directed
ment of its speed. He would give more information if he added upward.
the direction in which the object was moving-say, 25 miles per
hour toward the north-this would be a statement of its velocity. Speed is a
scalar quantity; velocity includes. direction and is a vector quantity. The
magnitude of the velocity is the same numerically as the speed, instant by
33
34 MECHANICS Chap. 3

instant, and the direction of the velocity is necessarily along the direction of
motion.
It frequently happens that a body in motion does not move steadily: it may
speed up, or slow down, or its motion may be quite irregular. Under these
circumstances the speed and velocity change from instant to instant but their
average values can easily be found. For any linear motion, the average speed
of the body is the length of path traversed divided by the time required to traverse
this path, and the average velocity is the displacement that the body undergoes
divided by the time during which this displacement occurs. Hence, speed is
associated with length of path, both of these quantities being scalars, and
velocity is associated with displacement, both of these quantities being vec-
tors. In equation form,
length of path
Average speed = . '
time
. displacement
and Average velocity = t.
1me
Both are expressed in symbols by a single equation. which is true for any linear
motion:
- s (3.1)
v = -t '

where either (1) v stands for speed and s for length of path, or (2) v stands for
velocity and s for displacement. The bar above the v signifies an aYerage
value, and t represents the time of travel.
Measurement of Time. Time measurements are based on the motion of
the earth around the sun. The earth revolves about the sun once a year
(about 365} days). Its orbit is strictly an ellipse with the sun at one focus,
but may be considered as a circle having a radius of 9.296 X 107 mi. Along
this orbit the speed of the earth varies slightly, being greater where the earth
is nearer the sun. The earth also rotates uniformly on its axis once a day.
The time required for an apparent rotation of the earth with respect to the
sun is called a solar day, and the average throughout the year is a mean solar
day. This is divided into 24 hours; 1 hour (hr) = 60 minutes (min); 1 min =
60 seconds (sec).
Units for Speed and Velocity. The same units are used for speed and
velocity, as would be inferred from Eq. 3.1. If the path length (or displace~
ment) is stated in miles and the time in hours, the speed (or velocity) will be
given in miles per hour. Linear speed or velocity can also be stated in feet
per second, meters per second, and centimeters per second. A unit com-
monly used for speed over water is the knot, which is a speed of 1 nautical
mile per hour. The nautical mile is a distance on the earth's surface that
subtends an angle of l min at the center of the earth; the average radius of
§ 3-3 RELATIVE MOTION OF TWO BODIES 35

the earth is 3958 mi, and consequently the nautical mile is equal to 1.152
(land) miles.
A speed expressed in miles per hour can be converted to feet per second by
multiplying it by suitable conversion factors, each of which is equal to unity,
§ 1-13. Thus,
30 mi = 30 mi_ X 5280 ft X 1 m: X 1 irlm- = 30 X 5280 ft = 44 .l!...
hr ht:_ 1 mi 60 :urin._ 60 sec 3600 sec sec
It will be convenient to remember that a speed of 30 mi/hr is exactly equal
to 44 ft/sec in converting from one of these units to the other.
3-2. Uniform Linear Motion. Linear motion can be studied by observing
the moving body carefully for a period of time and measuring how far it
moves during, say, every second. If the motion of the body is uniform, these
distances will be found to be equal. For example, if a car in uniform motion
passes some definite point and then another point 30 ft away one second later,
it will pass a third point 30 ft still farther away at the end of the next second,
and so on, covering the same distance each second. If a shorter time inter-
val is chosen, the distances will be less but will still be equal. In general, a
body in uniform linear motion traverses equal distances during equal inter-
vals of time, however brief these intervals may be.
Often linear motion is not uniform; for example, the above-mentioned car,
after starting from rest, had to acquire the uniform motion referred to, and
in stopping will lose that uniform motion. Other situations will doubtless
suggest themselves in which uniform motion is preceded or followed by non-
uniform motion. Such nonuniform motion is treated later; for the present,
in dealing with uniform motion, it is assumed that the body is traveling at a
steady rate before any consideration of its behavior is begun.
In uniform linear motion the speed or velocity can be computed from the
express10n
8
v= -' (3.2)
t
wheres and t have the same meanings as in Eq. 3.1.
Suppose that a racing car after getting under way crosses the starting line and
travels at constant speed over a 2-mi course in exactly 30 sec. Find the speed of
the car in miles per hour. Since the car travels 2 mi in t min; that is, in 1/120 hr,
it would travel 240 mi in 1 hr if it continued at the same rate. The speed of the car
is therefore 240 mi/hr. The same result is found from Eq. 3.2 as follows:
s 2mi mi
v = - = -- = 240 - ·
t 1 h hr
120 r

3-3. Relative Motion of Two Bodies. It is difficult to imagine absolute

motion and absolute rest, because all that can be observed is relative motion
3G MECHANICS Chap. 3

-that is, the motion of bodies with respect to each other. A passenger in a
train standing at a railroad station and looking out of the window at another
train nearby is often unable to tell whether "his" train or the other train is
in motion when one of them starts; he can observe only that one is moving
relative to the other. In ordinary usage the term "velocity of a body" meam;
its velocity relative to the earth, and the expression "condition of rest" refers
to a state of rest relative to the earth. Thus, one says that a house is sta-
tionary, although the house is carried along with the earth as it rotates on
its axis and moves in its orbit. The idea of relative motion will be illustrated
by some everyday examples.
I. Suppose that a man walks forward through a train at 4 mi/hr while the train is
moving northward at 20 mi/hr. What is the velocity of the man with respect to the
earth? The velocity of the man relative to the train is 4 mi/hr and that of the train
relative to the earth is 20 mi/hr, both northward. Since these velocities have the
same direction, the velocity of the man relative to the earth is 24 mi/hr toward
the north. If, instead, he were to ,rnlk to the rear of the train at the same rate, the
two velocities would have opposite directions and his resultant velocity relative to
the earth would be only 16 mi/hr toward the north.
II. An automobile 17 ft long driven at 45 mi/hr overtakes and passes a 30-ft truck
moving along a straight road at 25 mi/hr. Assume that one car is passing the other
so long as any point on the first is opposite any point on the second, and compute the
time required in passing; also compute the distance traveled by each vehicle during
this time. The velocity of the automobile relative to the truck is (45 - 25) mi/hr
or (66.0 - 36.7) ft/sec; that is, 29.3 ft/sec in the direction of travel. The displace-
ment of the automobile relative to the truck while passing is 17 + 30 = 47 ft, in the
same direction. Hence, the time occupied in passing is t = s/v, the relative displace-
ment divided by the relative velocity; namely, 47 ft/(29.3 ft/sec) or 1.602 sec. During
this time the automobile travels a distances = v t = 66.0 ft/sec X 1.602 sec = 105.8 ft,
and the truck travels 36.7 ft/sec X 1.602 sec = 58.8 ft.

In a more general study of relative motion the individual velocities do not

act along the same line. For definiteness, it will be assumed that an object,
No. 1, is moving with respect to a second object (or medium), No. 2, with a
velocity V1-2, while the second object is moving with respect to a third object
(or medium), No. 3, \Yith a velocity v2- 3. Then, the velocity of object No. 1
with respect to object No. 3 will be the vector sum or resultant of v1_. 2 and
lh_.3. Symbolically,
(3.3)

It must be remembered that this expression is only a shorthand statement of

a vector addition and is not intended in general to represent an arithmetic
addition.
This analysis will be applied to a traveling crane like those used in power
plants and factories for moving heavy machinery. The crane, illustrated in
Fig. 2, extends across the building and rolls on tracks supported along the
side walls near the eaves. A machine can be moved along the crane, cross-
wise of the building, at the same time that the crane is moving lengthwise of
 RELATIVE MOTION OF TWO BODIES 37

the building. The velocity of the machine with respect to the floor will
then be the resultant or vector sum of two separate velocities: that of the
machine with respect to the crane and that of the crane with respect to the
floor. In subscript notation, VM...,,F will represent the velocity of the machine

Fm. 3-2. Overhead traveling crane.

(Courtesy o.f Whiting Corporation).

relative to the floor, VM...,,c that of the machine relative to the crane, and
Vc...,,F that of the crane relative to the floor; then, as a vector addition,

Attention is drawn again to the fact that vector addition takes account of
the directions of the quantities involved. In order to evaluate the resultant
velocity v,i:r_,,F the separate velocities v,i:r__,.c and Vc_,J? must be
taken in their proper directions and then added vectorially
by the procedures described in Chap. 1.
As a numerical example, suppose the machine just mentioned to
be moved along the crane with a velocity VM--,c = 3 ft/sec, while
at the same time the crane is moving at right angles to this di-
rection, its velocity with respect to the floor being vc...,,F = 4
ft/sec. The resultant velocity of the machine relative to the floor
is seen from Fig. 3 to be Fm. 3-3. Addi-
+4 tion of velocities
VM...,,J? = VvM...,,c? + vc__,,p 2 = V3 2 2 = V25 = 5.00 ft/sec,
in the study of
and makes an angle () with the side walls such that relative motion.
() = tan-1 ! = tan-1 0.750 = 36.9°.

The principles of relative velocity apply to a body traveling through a

medium which is itself in motion---for example, an airplane flying in a wind
38 MECHANICS Chap.3

or a ship sailing in moving water. By way of illustration, the following prob-

lem deals with the motion of a ship in a stream.
In what direction should a ship be steered in order to reach a destination 60° east
of north from its initial position, if it steams through the water at 15 mi/hr while the
water is flowing due east at 6 mi/hr? The term "steered" denotes the direction in
which the ship moves relative to the water.
The velocity of the ship with respect to the land, vs..... L, is along the direction 60°
east of north; that of the ship with respect to the water, Vs-w, is 15 mi/hr in a direction
to be determined; that of the water with respect to the land, vw-L, is 6 mi/hr east.
These are related by the vector equation Vs_,,L = Vs-w + Vw._.L.
To solve the problem graphically, lay off to scale the vector vw..... L and the direction
line of Vs....,,L, as in Fig. 4. Then, since vs ..... w = 15 mi/hr, draw an arc with a radius of
15 units about the head of the vector vw..... L as a center. From this center draw Vs .....w
to the point where the arc intersects the direction line of vs ..... L, The angle 8 shows
the direction, reckoned from the east, in which the ship should be steered. The
numerical solution is left as an exercise for the student.

60

-
5i
E
40
80
~ 20
Q

3
Time, sec

FIG. 3-4. Path of a ship in Frn. 3-5. Displacement of a

moving water. body plotted with respect to
time.

MOTION IN A STRAIGHT LINE

3-4. Instantaneous Velocity. The considerations of § 3-1, which apply to

linear motion in general, will now be used to determine the instantaneous
velocity of a body in nonuniform motion. For simplicity, the motion will
be restricted to a straight line.
To find the velocity of a body at a particular instant, it may be supposed
that the body has just completed a displacements in a time t and that it now
undergoes an additional displacement As in a time interval At, the prefix
A (delta) being commonly used to represent an increment of a quantity. The
average velocity over this time interval is then As/ At. If the interval At
is taken shorter and shorter, the displacement As becomes shorter also, and
the instantaneous velocity v at the instant chosen is the limiting value of the
ratio As/ At as t::.t approaches zero; that is,
 JNS7'11NTANEO[/S l'ELOCI'l'Y 39

. . ~8
v = L1m1t - .
At-+0 ~t

In the limit, the ratio 1s represented by ds/ dt; hence, the instantaneous
velocity is
ds (3.4)
v = -.
dt
The quantity ds/dt is known in calculus as the derivative of s with respect to
t. If a graph is made with displacements of a moving body as ordinates and
elapsed times as abscissas, the velocity ds/ dt of the body at any instant is
the slope of the curve at the corresponding point.
Suppose that for a few seconds the displacement of a body along the x axis varies
with time t in accordance with the expression x = 10 t 2 - t3 , where xis the displace-
ment in feet and t is the elapsed time in seconds. This means that at time t = 0
the displacement is x = O; at t = 1 sec, the displacement is x = 10 (1) 2 - (1) 3 = 9 ft;
at 2 sec, x = 32 ft; and at 3 sec, x = 63 ft; as plotted in Fig. 5. The average velocity
during the first second is 9 ft = 9 ft/sec, and the average velocity over the 3-sec period
1 sec
is 53·3 ft = 21 ft/sec. These displacements and velocities are positive and therefore
sec
are directed toward the right, in accordance with the usual convention as to signs.
The velocity equation is found by differentiating the expression for x; thus,
v = ~: = 1 (10 t2 - t3), and since the derivative of t2 is 2 t and the derivative of
t3 is 3 t2 , it follows that v = 20 t - 3 t2 • Hence, the instantaneous velocity when
t = 2 sec is v = 20 X 2 - 3 (2) 2 = 28 ft/sec. The dotted lines in the figure show how
to verify this result graphically by drawing a tangent to the curve at point D, which
corresponds to a time of 2 sec. The slope of this tangent is

BC = (60 - 4) ft = 28 ~'
AC (3 - 1) sec sec
which agrees with the analytical result.

The expression v = ds/ dt when rewritten in the form ds = v dt gives the

infinitesimal displacement ds that a body having an instantaneous velocity v
undergoes in an infinitesimal time dt. The actual displacement in any finite
time is found by summing up a series of such infinitesimal terms. This
process is known as integration and is represented in symbols by:
s = fv dt. (3.5)
To perform the integration, the velocity vis expressed in terms of t, and the
expression is rearranged if necessary to assume a form such that its integral
can be recognized. One of the many forms encountered is given with its solu-
tion as follows:
10 ;t1ECIIL1NIC8 Chap. 3

where u is a variable, n a constant, and C a constant of integration that is to

be evaluated from the conditions of the problem.
In the problem solved above, the velocity of the body was found to be v = 20 t - 3 t2,
hence fv dt = f(20 t - 3 t2 ) dt = 20 ft dt - 3 ft 2 dt, the numerical constants 20
and 3 being set outside the integral signs. Then, by integration, the displacement is
t2 t3
x = 20 2 - 3 3 + C. To evaluate the constant C, note from the first part of this
problem that x = 0 when t = O; if these values are substituted in the equation just
found it will be evident that C = 0. Consequently, the expression for the displace-
ment is x = 10 t2 - t3 , which agrees with that initially given.
3-5. Acceleration. Every car mvner speaks of pick-up as one of the requi-
sites of a good automobile; by this term he means the rapidity with which
the car gains velocity. The scientific term for this concept is acceleration.
If the car, starting from rest, acquires a velocity along a straight highway of

60
B
50 /.,1

~ ~ I
I
~ 40
...... I
I
~ I
:: 30 I
g I
---------'
~ 20
10

5 10 15 20 25 30
Time, sec
FIG. 3-6. The acceleration of an automobile shown by a velocity-time curve.

4 ft/sec by the end of one second, gains an additional velocity of 4 ft/sec

during the next second, and so on, the car is said to have an acceleration of
4 ft/ sec in each second. By the end of three seconds the car will have a
velocity of 12 ft/sec, and so on, as long as it can maintain the same accelera-
tion. During this period of increasing velocity every conceivable value of
velocity, from zero up to the maximum reached by the car, is passed through
at some instant. The instantaneous velocity values fall along a curve such
as that in Fig. 6. The curve is straight from O to A, showing that during
the first five seconds the car gains velocity at a uniform rate; that is, the
acceleration is constant. The curve slopes off beyond A, showing that the
acceleration is reduced, and the curve becomes horizontal at B, which shows
that the acceleration is zero.
The acceleration of a body is defined as the change in its velocity during any
interval of time divided by the duration of that interval. Acceleration is a vector
quantity. For motion along a straight path, the acceleration is in the direc-
§ 3-5 ACCELERATION 41

tion of motion if the speed is increasing, and in the opposite direction if the
speed is decreasing.
The definition of acceleration is based on change of velocity, and not on
amount of displacement. If the motion of a body is observed for some stated
time interval l:!.t and if the velocity changes by an amount Av in this interval,
then the acceleration is
l:!.v
a=-.
l:!.t

If the change in velocity is uniform, the acceleration is constant and this

expression gives its constant value. If the change in velocity is not uniform,
the time interval may be taken shorter and shorter, and in the limit, the
instantaneous acceleration is
dv (3.6)
a= -
dt
and can also be expressed as the second derivative of displacement with
. d 2s
respect to time, or a = ptz · Acceleration denotes the time rate of change
of velocity, and its value at a particular instant is represented graphically by
the slope of the velocity-time curve at the corresponding point.
Any velocity unit divided by any time unit forms an acceleration unit.
The usual acceleration units and their abbreviations are:

feet per second per second, -ft-2,

sec

meters per second per second, -m- ,

sec 2
cm
centimeters per second per second, - -2,
sec
m1
miles per hour per second,
hr·sec

To say that a body has an acceleration of 1 cm per sec per sec means that
during every second the velocity of the body increases by 1 cm/sec. An
acceleration of 2 mi per hr per sec means that during each second the velocity
increases by 2 mi/hr.
The foregoing equations will be applied to the moti.on represented in Fig. 6, which
refers to an automobile starting from rest. During the first 5 sec the acceleration of
20 __!!_ - 0
. d . . l:!.v sec ft
the car is constant, an its va1ue is - = = 4 - .2 During the time
l:!.t 5 sec sec
interval from the fifth to the twenty-fifth second the acceleration varies from instant
to instant and its average value is
42 MECHANICS Chap. 3

l::,.v
60 Jl -20 Jl
sec sec = 9 _i!;_.
l::,.t - 25 sec - 5 sec - sec2
The instantaneous acceleration is found graphically by drawing a tangent to the curve
and determining its slope. For a tangent at point C the slope is
ft
(52.5 - 22.5) - f
______s_e_c = 3 _t_,
(15 - 5) sec sec2
and this is the value of the acceleration at an instant 10 sec after the start.
The instantaneous acceleration can be found analytically if the equation of the
velocity-time curve is known. The curve in Fig. 6, over the region from ,:S to 25 sec,
is given hy the equation v = 5 t - 0.1 t2 - 2.5, with v in feet per second and t in
seconds, reckoned from the instant when the motion starts. The derivative of this
expression gives the value of the acceleration; namely, dv/dt = 5 - 0.1 X 2 t, or
a = 5 - 0.2 t, with a in feet per second per second. vVhen t = 10, a = 5 - 0.2 X 10
= 3.0 ft/sec2 , a result which agrees with the value found graphically. All of these
accelerations are in the direction of motion.
The expression a = dv/dt can be rewritten as dv = a dt, whence, by
integration,
v = fa dt, (3.7)
an expression for velocity which is similar to Eq. 3.5 for displacement.
When the acceleration is constant, its instantaneous values are all alike
and the same as the average acceleration. In the remainder of this chapter
attention is given only to motion in which the acceleration is constant.
3-6. Equations of Motion with Constant Acceleration. The relations which
exist among velocity, displacement, acceleration, and time, in motion along
a straight line with constant acceleration, can be expressed by three equations.
In these, the following symbols will be used:
Vo initial velocity-the velocity of the body at the moment the considera-
tion of its motion begins,
v1 final velocity-the velocity at the moment the consideration of the
motion ends,
time interval between the two moments mentioned,
s displacement-the straight-line distance from the initial to the final
position of the body, and
a acceleration, assumed constant.
I. One of the equations follows from the definition of acceleration in the
preceding section; namely, a = !::,.v/t::,.t. In terms of the symbols just given,
VJ - V0
l::,.v = v1 - v0 and t::,.t = t; therefore a = t , and by rearrangement,

VJ = V0 + at. (3.8)
Suppose that while an automobile is being driven along a straight road its speed-
ometer reads 20 mi/hr at a certain instant and 45 mi/hr 10 sec later; find the ac-
§ 3-6 EQUATIONS OF MOTION WITH CONSTANT ACCELERATION 43

celeration (assumed constant) of the car during this interval. From Eq. 3.8, the
acceleration is
mi mi mi
45- - 20- 25 -
hr hr hr
a = - - - - - - = - - = 2.5 mi/(hr·sec)
10 sec 10 sec
and is in the direction of motion.
II. The second equation of accelerated motion gives the displacement of a
body having constant acceleration. To derive the equation, it is convenient
first to find the average velocity of the moving body during the interval con-
sidered. Since the velocity changes at a steady rate, the average is the mean
of the velocities at the beginning and end of the interval. But these are v 0

and Vt respectively, and therefore the average velocity is

_ Vo+ V[
v= . (3.9)
2
Then the displacement s during a time interval t is the product of the aver-
age velocity and the duration of the interval as indicated in § 3-1, or
_ _ _ Vo+ V[
s - vt - 2 X t. (3.10)

The value of Vt from Eq. 3.8 is now substituted in this expression, and the
displacement is found to be
= Vo + (vo + a l) X t
8 2 '
or (3.11)
This equation can be derived more simply by calculus. Since the velocity
at time t (previously called Vt) is v = Vo + at, and since s = fv dt, it follows
a t2
that 8 = f (vo + at) dt = Vol + 2 + C.
I
If displacements are reckoned from the starting I
at I
position, then s = 0 when t = 0; whence C = 0.

l
I
~
Thus the displacement is s = v t + ! a t , as
I

ij
0
2

above.
The inclined line in Fig. 7 represents graph-
ically the motion of a body in which the velocity
Vo ! V0 Vi

increases uniformly with time; that is, motion

with constant acceleration. From Eq. 3.11 it Time
is evident that the area below the line repre- FIG. 3-7. The area under a
sents the displacement of the body. Also, in velocity-time curve represents
Fig. 6, the portion of the curve from O to A displacement.
represents this type of motion.
The displacements of a body moving with constant acceleration are pic-
tured to scale in Fig. 8 for the first few seconds after starting from rest, to
44 MECHANICS Chap. 3

show how much farther the body moves each second. The figure is drawn
for an acceleration of 2 ft/sec 2 •
To calculate how far the body just mentioned is displaced during the fifth second
after starting from rest, the displacements in 4 and 5 sec are computed; these are,
respectively,

84 1(
= - 2 - ft 2 ) (4 sec) 2 = 16 ft
2 sec
and 85 = -1 ( 2 - ft 2 ) (5 sec) 2
2 Rec
= 25 ft
'

as shown in the figure. Hence the displacement during the fifth second itself 1s
25 - 16 = 9 ft in the direction of motion.

III. The third equation of motion with constant acceleration is derived

from the other two by eliminating the time interval t, which appears in both.

t=O l 2 3 5 sec

,~I l~I l~I I I 111~1 I

s=O l 4 9 16 25 ft
Fm. 3-8. Successive displacements of a block moving with constant acceleration.

This result is obtained by multiplying a from Eq. 3.8 bys from Eq. 3.10, as
follows:

a X S = VJ -
t
Va
X VJ +2 Va t =. 1 (
2 Vt - Va
) (
Vt
+ Vo),
from which vi= v} + 2 as. (3.12)
Suppose that an electron in an x-ray tube is given a velocity of 109 cm/sec in moving
from the cathode to the target, a distance of 4 cm. Assume that the electron starts
from rest, and compute its average acceleration. From Eq. 3.12,

(109)2 cm2 - 0
v/ - v} sec2 = 1.25 X 1011 cm .
a=
28 8 cm sec 2
For ease of reference the three equation,,s for rectilinear motion with con-
stant acceleration are collected herewith, together with the useful "average
velof'.ity" equation:
(3.8) VJ = V0 + at;
(3.11) S = V ot + i a t 2;

(3.12) vi = Va 2+ 2 a s;
(3.10) S = ijf = Vo 1 VJ X f.

The student should understand these equations thoroughly and should gain
experience in using them with consistent units to solve practical problems.
§ 3-8 FALLING BODIES 45

There should be no difficulty in choosing the equation to use in a specific

example, for the first one lacks a displacement term, the second a final velocity,
the third a time interval, and the fourth an acceleration.
3-7. Acceleration Due to Gravity. Bodies fall to the ground because the
earth exerts a pull upon them. This pull, called the force of gravity, causes
a body to accelerate while falling. Moreover, at any given location all bodies
fall with the same acceleration, regardless of their masses, if the resistance
and buoyancy of the air are neglected. This fact, some believe, was first
shown experimentally at the Leaning Tower of Pisa by Galileo Galilei (1564-
1642), Italian natural philosopher and astronomer.
The force of gravity upon a body and hence the acceleration that the force
causes depend upon the distance of the body from the center of the earth,
and decrease as this distance increases. The acceleration of a falling body,
if measured at points above the earth's surface, would be found to diminish
as the elevation became greater. All over the surface of the earth this accel-
eration is nearly the same; it is slightly greater at the poles than at the
equator because the earth is flattened at the poles, its polar radius being 13
miles shorter than the equatorial radius. 'l'he values of the acceleration due
to gravity at a few places are given in Table I.

TABLE I. AccELERATION DuE TO GRAVITY

Ft Cm
-Sec - -Ft Cm
-
2 , Sec 2 Sec 2 Sec 2

Berlin, Germany .......... 32.19 981.2 New Orleans, La ......... 32.13 979.3
Boston, Mass ............. 32.17 980.6 New York, N.Y ......... 32.16 980.2
Equator at sea level ....... 32.09 978.1 North Pole ............. 32.26 983.3
London, England .......... 32.19 981.2 Paris, France ........... 32.18 980.8
Madison, Wis ............. 32.16 980.2 San Francisco, Cal. ...... 32.15 979.9

The acceleration imparted by the earth's pull is generally designated by

the symbol g, and the value 980.665 cm/sec2 has been adopted internationally
as a standard. For many purposes the value is taken as

g = 32 ft/sec2,

or 9.80 m/sec 2,

or 980 cm/ sec 2 •

These approximate values may be used in problems unless other values or

particular places are specified.
3-8. Falling Bodies. The laws of falling bodies are given by the general
equations for motion with constant acceleration, except that the acceleration
g due to gravity is used instead of acceleration a. The equations for falling
bodies are, therefore, simple modifications of Eqs. 3.8, 3.11, and 3.12. In
46 MECHANICS Chap. 3

applying them, either the upward or the downward direction is arbitrarily

taken as positive and the opposite direction as negative, and and - signs +
are given accordingly to the vector quantities displacement, velocity, and
acceleration. A few problems will illustrate the laws of falling bodies. In
these, air friction will be neglected, and the upward direction will be regarded
as positive.
I. A ball is dropped and allowed to fall freely; compute the distance it will travel in
1 sec, 2 sec, . . . . The ball has no initial velocity; that is, v = 0, and its displace-
0

mentB are given by Eq. 3,11 as

s1 = ! (-32 ~2) (1 sec) 2 = -16 ft, s2 = -1 ( -32 - ft 2 ) (2 sec) 2 = -64 ft,

2 sec 2 sec
and so on, as shown in Fig. 9.

Fm. 3-10. The

path of a ball
5- .... - - - -
thrown verti-
Fm. 3-9. Displace- cally upward
ments of a freely fall- from an elevated
ing body. point.

II. A ball is thrown vertically downward with a speed of 10.3 m/sec from a cliff
150 m high. With what velocity will it strike, and how long will it take to reach the
ground? In this problem the initial velocity Vo = -10.3 m/sec, the displacement
s = -150 m, and the acceleration due to gravity is -9.8 m/sec 2 • The velocity at
impact is found from Eq. 3.12 as follows:
2
v/ = v} + 2 gs= (-10.3) 2 - m2
sec2
+ 2 ( -9.8 -sec
m) (-150
2
m) = (106 + 2940) -n1 2
sec
;

whence VJ = ±55.2 m/sec. In this problem the ball is moving downward; therefore,
VJ= -55.2 m/sec. The time of flight, from Eq. 3.8, is

-55.2 ~- (-10.3 ~ )
t= VJ - Vo = sec sec = 4.58 sec.
g ill
-9.8-2
sec
§ 3-9 COMBINATION OF UNIFORM AND ACCELERATED MOTIONS 47

Consequently, the ball will strike the ground in 4.58 sec with a downward velocity of
55.2 m/sec.
III. A ball is thrown vertically upward with a velocity of 80 ft/sec from the bank
of a river and falls into the water 50 ft below the bank, as shown in Fig. 10. (a) Find
the height to which the ball will rise. (b) For what length of time does it continue to
rise? (c) With what velocity does it pass the level of the bank on the way down?
(d) What is the velocity of the ball just before it hits the water?
(a) In order to find how high the ball will rise, refer to the displacement AB, for
which the final velocity is v1 = 0 and the initial velocity Vo = (+) 80 ft/sec. From
Eq. 3.12,
0 - (80)2 ft2.
s=
v/ - Va 2 sec 2
= 100 ft.
2g f
2 ( -32 _t_ )
sec 2
Hence the ball is displaced upward 100 ft in moving from A to B.
(b) The time required to reach the top point Bis found by applying Eq. 3.10 to the
displacement AB as'follows: the average velocity is O ~ SO = 40 ft/sec, and therefore
. . 8 100 ft -
t h e time 1st = - = -- = 2.oO sec.
v 40 l!_
sec
(c) For this part of the problem, first find how long a time the ball falls before it
passes the bank by using Eq. 3.11 for the displacement BC. Since for this displace-
ment the initial velocity is Vo = 0, the equation reduces to s = ! g t2, whence
t = = 2( - 100f ft) = 2.50 sec, sh owmg
. t h at t h e time
. of ascent equa1s t h e time
.
g -32-t
sec2
of fall. Then use Eq. 3.8 for the displacement BC to obtain the velocity of the ball as
it passes the bank; thus, v1 = v0 + gt = 0 + (-32 sec
ft 2) (2.5 sec) = -80 Jl..
sec
This
result shows that the ball has the same speed at this level that it had when thrown,
but is moving in the opposite direction.
(d) To find the velocity of the ball just before it hits the water, apply Eq. 3.12 to
the displacement BD, for which s = -150 ft. There results

vj = v} + 2 gs = 0
, - ft )
+ 2 ( -32 (-150 ft) = 9600 -ft 2 ;
2
2
sec sec
whence v1 = ±98 ft/sec. Consequently, the velocity of the ball just before it hits the
water is -98 ft/sec. This result could also be found by applying Eq. 3.12 to the dis-
placement AD, for which v0 = 80 ft/sec ands = -50 ft. Then

vj = ( 80 -ft ) + 2 ( - 32 - ft 2 )
2
( - 50 ft) = 9600 -ft2 2 ,
~c ~c ooc
which gives the same result as before.

MOTION OF PROJECTILES

3-9. Combination of Uniform and Accelerated Motions. An experiment

commonly performed in the lecture room makes use of two metal balls sup-
48 MECHANICS Chap. 3

ported at the same level near the ceiling, one having a compressed spring
behind it and the other having a trap door beneath it. These parts are
released simultaneously by closing an electric circuit which controls both
spring and door; the spring strikes one ball horizontally and the door opens
to allow the other ball to fall vertically downward. It is found that both
balls strike the floor at the same instant, which shows that the horizontal
motion given to one ball by the spring does not affect its motion in the verti-
cal direction.
The ball that is projected horizontally follows a curved path, moving both
horizontally and vertically at the same time. The horizontal motion of the
ball is uniform, for there is nothing (air resistance being neglected) to change
its speed along this direction, and is unaffected by falling at the same time.
The vertical motion is accelerated, like that of any
0
/eet
10 200 300 falling body, and is neither helped nor hindered
--...... by moving horizontally while falling. Conse-

100
" '\
quently, such motion can be studied by regiwd-
ing it as made up of two parts, one horizontal and
\ the other vertical, and by recognizing these mo-
\ tions as independent of each other.
\ Suppose that a ball is thrown horizontally from the
\ top of a high building with a velocity of 50 ft per sec.
300 The horizontal motion is uniform, and Eq. 3.2 gives
\ the horizontal displacement as s = vt. The vertical
\ motion is accelerated, and Eq. 3.11 gives the vertical
400
displacement ass = ! gt2 • Therefore, during the first
Frn. 3-11. Graph representing second of flight the ball moves horizontally .50 X 1 =
a combination of uniform mo- 50 ft, and falls ! g (1) 2 = 16 ft. During the first two
tion to the right and acceler- seconds of flight it travels horizontally 100 ft and falls
ated motion downward. a total of t g (2) 2 = 64 ft. The horizontal motion
with constant velocity and the vertical motion with
constant acceleration result in the following displacements of the ball during the
first few seconds, air friction being neglected:
Total time, sec . . . . . . . . . . . . 1 2 3 4 5
Horizontal travel, ft . . . . . . . . 50 100 150 200 250
Vertical fall, ft . . . . . . . . . . . . 16 64 144 256 400
These values are coordinated in Fig. 11, and the resulting curve, a parabola, is the
trajectory of the ball. If the building were just 400 ft high, the ball would strike
level ground at a point 250 ft from the base.

3-10. Projectile Motion. In the study of ballistics to determine the range

of projectiles and to trace their paths, it is convenient to resoh·e the actual
motion of the projectile into two independent motions, one vertical and the
other horizontal. Two illustrations will explain the procedure; in both, air
friction is neglected.
I. A gun on level ground fires a projectile at a speed v and directed at an
§ 3-lO f'RO.J ECTILE li,!OTION 49

angle () upward from the horizontal. How far from the gun will the pro-
jectile strike the ground? The first step is to resolve the projection velocity
into a vertical and a horizontal component. The time of flight is deter-
mined by the vertical component and can be found from the expressions for
falling bodies, § 3-8. After this time interval has been found, the horizontal
range of the projectile is obtained from the horizontal component of the
projection velocity by applying the equation for uniform motion.
The vertical component of the projection velocity is v sin 8, and the hori-
zontal component is v cos (), as shown in Fig. 12. To find the time of flight,
only the vertical motion is considered, as if the projectile were not moving
horizontally at all. Some direction, say upward, is selected as positive, and

~
K-------R-------
Fm. 3-12. The trajectory of a projectile.

the time t1 is found in which the projectile reaches the top point of its flight,
where its vertical velocity, for an instant, is zero. For this part of the
flight, Eq. 3.8 becomes O = v sin () + (-g) t1, or v sin () = gt1, from which t1
can be found. During this period t1 the projectile rises to a height given by
Eq. 3.11 as h = (v sin ()) t1 - t gt12, or

The time t2 for the projectile to return to the ground is found similarly from
the same equation. Thus,
-h = t (-g)ti.

Since h in the last two equations represents the same height, it follows that
t1 = t 2 , or the time to reach the maximum elevation is the same as the time
to drop from that level to the datum plane. The total time of flight is there-
fore t = 2 t1.
To find the horizontal range of the projectile, only its horizontal motion is
considered, as if it were not moving vertically at all. The horizontal motion
is uniform, with the velocity v cos () acting for a time t. The horizontal dis-
placement, known as the horizontal range, is therefore

R = (v cos O) t.

If air friction had been taken into account, both the vertical and horizontal
displacements ,vould be noticeably reduced, particularly during the latter
part of the flight.
50 MECHANICS Chap.3

II. A bomber is flying horizontally with constant velocity in the same

vertical plane as its objective shown at O in Fig. 13. At what angle of sight
</> with respect to the vertical should the observer release a bomb so that it
will strike the objective? The bomb, at the instant of release, has the same
velocity as the airplane; consequently, it travels horizontally with constant
velocity all the time it is falling toward the earth; in fact, its trajectory is
exactly like that of the ball thrown horizontally from the building, described
in the preceding section. Hence, the bomb will not strike a point on the
ground which was directly below the airplane at the moment of release, but
will strike beyond that point in the direction of flight by an amount which
depends upon the velocity and height of the airplane. This distance sub-
tends an angle </> at the observer's position and is the angle of sight to be
solved for.
If t represents the time of flight reckoned from the release of the bomb and
x the horizontal distance at that moment between the plane and the objective
0, then the following relations hold:
x = h tan </>; h = ! gt2 ; x = vt.
It is necessary to eliminate both x and t from these expressions, for the angle
of sight </> should involve only the elevation h of the airplane and its hori-
zontal velocity v; the solution is:
12
tan</> = v
'17j,; .
Numerical values substituted in this equation will give data from which
curves can be plotted to show the proper angle of sight </> for any definite
airplane speed v and flight altitude h.

Fw. 3-13. The path of a bomb Fw. 3-14. Diagram used

dropped from an airplane that in defining the radian.
is traveling horizontally.

ANGULAR MOTION

3-11. Rotation and the Radian. A body spinning about a fixed axis, such
as a flywheel rotating in stationary bearings or a chuck turning in a lathe,
§ 3-12 ANGULAR VELOCITY 51

stays where it is; it does not move from one point to another along a line, and
consequently its motion is not linear. The motion of a rotating body can
be described by referring to radii extending outward from its axis of rotation.
As the body rotates, every radius sweeps through an angle-for instance, it
sweeps through 360° in making a complete revolution. For this reason a
rotating body is said to have angular motion.
A unit frequently used for measuring angles, and which is particularly
useful in calculations dealing with rotation, is based upon the relative dimen-
sions of a sector of a circle. Fig. 14 shows a sector bounded by two straight
lines at an angle e with each other. So long as the angle is not changed, the
ratio of the arc s 1 to its radius r1 is the same as that of any other intercepted
arc s2 to its radius r 2 ; this ratio is not affected by the size of the circle but
depends only on the central angle. The angle subtended at the center of a
circle by an arc equal in length to the radius is called a radian. In general,
. . arc length
Angle m radians = d" ,
ra ms
or in symbols,
s
e= -.
r
(3.13)

The angle e is a ratio of one length to another, and these must be expressed
in the same unit in order that the angle may be in radians. In consequence,
an angle stated in radians is merely a number, since the units of s and r
cancel.
Radians can be converted to degrees, or vice versa, by remembering that
the circumference of a circle is 2 1r (that is, 2 pi or 2 X 3.1416) times its
radius; therefore the central angle subtended by the complete circumference
is 2 1r radians. A right angle is one-fourth as large, or 1r /2 radians. Thus,
2 1r radians are equivalent to 360°, and 1 radian= 360/(2 7r) = 57° 17' 45",
or 57.3° approximately:

TABLE II. CONVERSION FACTORS FOR ANGLE!',

1 revolution = 2 1r or 6.283 radians

1 radian = 360/ (2 1r) or 57 .3 degrees
1 degree = 2 1r/360 or 0.0175 radian
1 minute = 2 tr/(360 X 60) or 0.000291 radian

3-12. Angular Velocity. The rotation of the earth and the motion of the
hands of a clock are familiar examples in which the angular motion is uni-
form. The angidar velocity of a body in uniform rotation is defined as the
angular displacement of the body divided by the time occupied in sweeping
through this angle; that is,

A ngu1ar ve1oc1"ty =
angular displacement
t" ,
1me
52 MECHANICS Chap. 3

or, in symbols,
fJ
w = -, (3.14)
t
where w (omega) is the angular velocity, () is the angular displac"ement, and
t is the corresponding time. For uniform rotation, equal angles are swept
through in equal time intervals and the ratio of () to tis a constant; accord-
ingly, a body in uniform angular motion has constant angular speed.
Angular velocity is usually expressed in revolutions per minute (rev per
min) or radians per second. Since 1 rev = 2 1r radians, it follows that
1 rev/min = (2 1r/60) radians/sec.
To find the time required for a wheel to make 10 complete revolutions while turning
at a constant velocity of 80 radians per sec, refer to Eq. 3.14, from which the time is
10 X 2 radians . sec
t = d" = 10 X 2 ~ x 80 -·- ,.:i:--- = 0.785 sec.
80 ra ians ~

sec
The average angular velocity of a rotating body, whether its motion is uni-
form or not, can be found over a given period by dividing the total angular
displacement () by the total elapsed time t: in symbols,
fJ (3.15)
w = -.
t

As the time interval is taken shorter and shorter the average angular velocity
approaches nearer and nearer the instantaneous value, and in the limit the
instantaneous angular velocity is
d()
w =-· (3.16)
dt
The distinction between speed and velocity, which was described in § 3-1
for linear motion, applies equally to angular motion.
3-13. Angular Acceleration. When a rotating machine starts from rest, a
certain period of time elapses before it reaches a steady velocity. If this
time interval is short, the machine is said to have a high angular acceleration.
The close analogy of angular to linear motion makes it possible to shorten
the discussion of angular acceleration by merely following the pattern already
adopted for linear acceleration.
Angular acceleration is defined as the change in angular velocity during any
interval of time divided by the duration of that interval. When a change of
angular velocity ~w occurs in a time ~t, the ratio !; gives the angular accel-
eration, or the average angular acceleration if the velocity does not change
uniformly. The instantaneous angular acceleration is
dw (3.17)
a=-·
dt'
§3-14 RELATION BETWEEN ANGULAR AND LINEAR QUANTITIES 53

that is, the derivative of w with respect to t.

For motion with constant angular acceleration, the equations can be ex-
pressed by analogy with those in § 3-6 as follows:
Wf = Wo + at; (3.18)

8 = Wot + ! ai j2 (3.19)
wJ2 = Wo 2 + 2 a:8. (3.20)

Herein 8 is the angular displacement in time t, a is the angular acceleration,

w 0 is the initial angular Yelocity, and w1 is the final angular velocity. Care
must be exercised in applying these equations to use consistent units; for
example, the same unit of angular measure, either the revolution ~r the
radian, should be used in all terms.

As an illustration, suppose that the flywheel of a metal punch-press, whi.ch normally

runs at 80 rev/min in a clockwise direction, slmvs down ,Yhile punching the metal so
that it takes 0.01 min to turn through i rev during the punching operation. Compute
(a) the angular acceleration, assumed constant, and (b) the angular velocity of the
flywheel at the end of the punching process.
The angular displacement is

e = 3-rev-
-X
2 1r radians
= -321r ra d'ians'
4 1 -re;:;..
the initial angular velocity is

80 ~ 2 1r radians 1 i'1:tm.. 8 1r radians

wo = ~ X 1~ X 60 sec = sec 3
.
both clock wise, an d t h e time
. of t h e operat10n
. .
IS t =
0 .0 1 .um:t.
. X 60ffiiQ
sec_ = 0.6 sec.
1

(a) The angular acceleration is given by Eq. 3.19 as

8 7r radians
2 X 3- 7r ra d'ians 2 x----
2e
a=~-~=~~~~~~

t2
2 Wo
t
2
(0.6) 2 sec 2
3 sec
0.6 sec
-1. 75 radians,
sec2

and (b) the final angular velocity follows from Eq. 3.18 as

w
1
= 8 1r radians
3 sec
+ (-1. 75 radians) (0. 6 sec) = 7 _32 radians,
sec 2 sec

which is equivalent to 70 rev/min, in a clockwise direction.

3-14. Relation Between Angular and Linear Quantities. When a body is

rotating about a fixed axis, all points on the body describe concentric circles
ahout that axis. Hence, while the body as a whole has angular motion the
individual points on it move along lines, and so these points have linear
54 MECHANICS Chap. :J

motion. A disk rotating about a fixed axis, as shown at O in Fig. 15, has
angular motion and the points A and Bon it have linear motion. The linear
speed of point A, located near the rim, is greater than that of point B, near
the axis, since A moves through a larger circle than
Direction does B for each revolution of the disk.
of rotation
A simple relation exists between the angular speed
~
of a rotating body and the linear speed of any point
on it. If the disk in Fig. 15 turns through e radians

(A
m · w = 8 ra d.ians per sec.
· t sec, 1·ts angu1ar speed 1s
In this time any point P located a distance r cm
t
from the axis travels a distance s = Br cm, and con-
sequently the linear speed of the point is v = ~ = ert
t
cm per sec. A comparison of these results shows
ter, whence
Fm. 3-15. Comparing
the angular motion of a that v =
rotating disk and tho
linear motion of two v = w r, (3.21)
points on it.
or Linear speed = angular speed X radius.
In this equation, the angular speed w must be expressed in radians per unit
time, say radians per second, and the unit for the linear speed will depend
upon that used for the radius r. Thus, if r is expressed in centimeters, v is
given in centimeters per second.
As an example, what is the maximum allowable diameter of a cast-iron flywheel
which is to rotate at 90 rev/min, if the speed of the rim is not to exceed 6000 ft/min?
ft
6000-. ft .
v mm mm
From Eq. 3.21. r = - = d" = 6000 - . X - - - - ~. = 10.6 ft,
w 90 X 2 1r ra 1ans mm !.-"-
90 X 2 71"~
mm
whence the flywheel diameter = 2 X 10.6 ft = 21.2 ft.
When a rotating body accelerates, any point on it has a corresponding
acceleration along its circular path. The angular acceleration of the body
can be expressed as a = WJ ~ wo and the linear acceleration of the point as

a = vr ~ Vo, in accordance with Eqs. 3.18 and 3.8. The relation between
these quantities can be found by applying the expression v = wr as given in
Eq. 3.21, in which r is the distance from the axis of rotation to the point in
question. It follows that
a = Wjr - Wor = (Wf - W 0 )
t t r,
or a= a r, (3.22)
§3-14 RELATION BETWEEN ANGULAR AND LINEAR QUANTITIES 55

which states that the linear acceleration of any point on a rotating body along
its circular path is equal to the product of the angular acceleration of the
rotating body and the radius extending to that point.
An examination of Eqs. 3.13, 3.21, and 3.22, which correlate the magni-
tudes of linear and angular quantities, shows that in each instance the linear
quantity s, v, or a, is equal to the corresponding angular quantity (in radian
measure) multiplied by the radius.
PROBLEJl:fS
1. How much longer would it take to travel a distance of 5 mi at an average speed
of 40 mi/hr than it would at an average speed of 60 mi/hr?
2. Two towns are 2 mi apart, and an observer is locate<l between the two ! mi
away from one of them on a straight line joining the t,vo. If town clocks at the two
towns strike the hour of one at the same instant, what time will elapse between the
two sounds heard by the observer? Take the speed of sound as 1100 ft/sec.
3. An airplane is following a straight course at 120 mi/hr, its elevation being
constant at ! mi. A gun on the plane is fired at an instant when the plane is directly
above an observer on the ground. '\Yhere will the observer see the plane at the instant
he hears the report? Take the speed of sound as 1100 ft/sec.
4. A freight train 1400 ft long, traveling at 45 mi/hr, overtakes and passes a man
walking alongside the track at 3 mi/hr. ·what time is occupied in passing?
5. Two trains are running on parallel tracks in opposite directions. One train is
400 ft long and is moving at 60 mi/hr. The other is 300 ft long and is moving at
40 mi/hr. Assume that the trains are passing so long as any point on one train is
opposite some point on the other, and compute the time occupied in passing.
6. A local train leaves Buffalo headed east at an average speed of 40 mi/hr, and
an express train leaves that city 1 hr later headed east at an average speed of 55 mi/hr.
At what distance from Buffalo does the express overtake the local?
7. An airplane moving with an air speed of 200 mi/hr travels 400 mi due east
and returns to its starting point. If there is a wind blowing from west to east at
40 mi/hr, what time is required for the round trip?
8. A river 1 mi wide flows due south at 2 mi_/hr. A ferry boat leaves the east bank
of the river and crosses to a point ! mi downstream on the west bank. If the pilot
finds that he must steer the boat in a direction 10° south of west, how long does the
trip take?
9. For a ball thrown vertically upward with an initial velocity of 96 ft/sec, the
displacement in t sec, reckoned from the initial position, is given in feet by the ex-
pre8sion s = 96 t - 16 t2 • (a) Derive a general expression for the velocity of the ball.
b) Find the displacement and the velocity of the ball at the end of 1 sec, 3 sec, and
."'i sec.
10. Suppose the velocity v of a body (in feet per second) to vary with time t (in
:--econds) as given by the expression v = 1 + 8 t - 6 t2• (a) Let the displacement s
be reckoned from the initial position of the body (that is, s = 0 when t = 0) and
•ieriYe general expressions for the displacement of the body and for its acceleration.
1:,, At what displacement will the acceleration become zero?
11. Compute the acceleration, assumed constant, of a body that moves 20 ft in
4 sec, starting from rest.
12. If a body moving with constant acceleration travels 30 ft during the first two
.aeconds after starting from rest, how far will it travel during the next two seconds?
13. An automobile is being driven at 30 mi/hr along a straight highway when the
,i~iwr suddenly notice~ an obstruction 5 car lengths (80 ft) ahead. His reaction time
56 MECHANICS Chap. 3

before he starts to apply the brakes is f sec, after which he slows the car down at
-20 ft/sec 2 • By how many feet does he miss the obstruction?
14. An outside gallery of the Leaning Tower of Pisa, on which Galileo may have
conducted his demonstration of falling bodies, is 150 ft above the ground. How long
would it take a freely falling body to drop to the ground from this height?
15. An object starts from rest and falls freely toward the ground. What distance
does it fall during the first second, and what is its speed at the end of that time?
16. A ball is thrown vertically upward with a speed of 30 ft/sec from a point 20 ft
above the ground. How far above the ground will it be 2 sec after it is thrown?
17. A ball thrown in a vertical direction from a point 40 ft above the ground reaches
the ground in 3 sec. Determine the velocity with which the ball was thrown.
18. In the diagram, water in an elevated tank flows horizontally through an orifice
A and strikes level ground at C. If AB = 9 ft and BC = 12 ft, what is the speed of
the stream at A?

A
_-_-_-_ A ~
I
I
I
I
c
I
IB c
PROBLEM 18.

19. A projectile is fired horizontally from a gun with a speed of 1000 ft/sec. If the
gun is 20 ft above a level terrain, how far will the projectile travel horizontally before
it strikes the ground?
20. The diagram shows an airplane A and a target C. If the airplane has a steady
speed of 180 mi/hr and a constant elevation of 1600 ft, at what horizontal distance BC
from the target should a bomb be released in order to obtain a hit?
21. A projectile is fired at 1000 ft/sec in a direction 30° upward from a level plain.
Compute (a) the maximum height reached by the projectile, and (b) the horizontal
range of the projectile. Neglect the effects of air friction.
22. Refer to Prob. 21 and determine the location of the projectile at an instant
10 sec after it is fired. Express the result by giving the vertical displacement and the
horizontal displacement of the projectile at the instant selected.
23. Refer to Prob. 21 and determine the velocity of the projectile at an instant 25 sec
after it is fired.
24. A piece of railroad track 200 ft long forms a circular curve of 640-ft radius.
What angle does it subtend at the center (a) in radians, (b) in degrees?
25. The earth has an average radius of 3958 mi, and a quadrant extending from
equator to pole contains 90° of latitude. What distance on such a quadrant subtends
an angle of 1 min at the center of the earth?
26. How far away is the horizon at sea for an observer whose eyes are 60 ft above
the water?
27. Phonograph turntables operate at the old conventional speed of 78 rev/min
and at the newer speeds of 45 and 33t rev /min. Express each ,af these speeds in
radians per second.
28. A motor armature is brought from rest to its normal speed of 1200 rev/min in
30 sec. Find the average angular acceleration of the armature and the number of
revolutions that it makes during the starting period.
 PROBLEMS 57

29. The minute hand of a watch makes a revolution every hour. Compute (a) the
angular speed of the hand in radians per second, and (b) the linear speed of a point
on the hand t in. from the axis.
30. A pulley 10 in. in diameter drives another which is 12 in. in diameter by means
of a belt. If the 10-in. pulley revolves 96 times a minute and if the belt does not slip
on either pulley, what is (a) the linear speed of the belt, and (b) the speed of the 12-in.
pulley in revolutions per minute?
 4
FORCE AND TRANSLATION
In the preceding chapter, accelerated motion is considered "Without stressing
the means by which the acceleration is brought about; the present chapter
and the one to follow deal with the agencies that cause it. Linear acceleration
is always produced by a force, and angular acceleration by a torque. Thus,
to make any change in the velocity of a body requires the action of a force for
motion of translation, or of a torque for motion of rotation. The branch of
Physics that deals ,,·ith the effect of force or torque on motion is called
Dynamics.

BASIC PRINCIPLES

4-1. Effect of Force on Motion. A body at rest tends to remain in that

condition; this tendency illustrates a property of matter called ·inertia. If
the body is to be set into linear motion it must be accelerated, and a force
must be applied to it to cause the acceleration. Again, if the body is already
in motion it tends to maintain constant velocity-another manifestation of
inertia. A force is needed to speed the body up, slow it down, or change its
direction; any change in velocity implies an acceleration, and linear accelera-
tion can be produced only by a force acting on the body.
A force has already been described as a push or a pull. A man pushing a
lawn mower exerts a force on the lawn mower, and a book resting on a table
exerts a force on the table. In each of these illustrations the t,vo bodies
touch each other. The man must be in contact ,vith the lawn mower in
order to exert a force on it; moreover, as the lawn mower moves, he must
move along also and stay in contact with it if he is to coritinue to exert a
force upon it. Again, when a man throws a ball, he exerts a force on it only
so long as it stays in contact with his hand. There is, however, an exception
continually met with in Mechanics to the idea that one body cannot exert a
force on another unless the two are in contact. This exception is gravita-
tion, by means of which a body, say the earth, exerts a force of attraction on
other bodies, whether they are in contact with it or not. This gravitational
attracti6n was mentioned in Chap. 3 and will be considered more fully in
§ 4-5.
'Whenever a body is accelerating a force must be acting upon it, but it does
58
§ 4-2 MASS 59

not follow necessarily that a body will always accelerate when a force is
exerted on it. For example, a man may push on a crate while at the same
time friction or some other agent exerts an equal force on it in the opposite
direction; these two forces balance each other and the crate will not accelerate.
If, however, all of the forces acting on a body are taken into account and if
these do not balance, then the unbalanced force will always cause accelera-
tion. Whenever a body is accelerating there must be an unbalanced force
acting upon it; also, whenever an unbalanced force acts on a body, the body
must accelerate.
Particles and Rigid Bodies. A distinction is frequently made in Mechanics
between particles and rigid bodies because they behave differently under the
action of forces. A particle is a portion of matter, usually small, upon which
all forces may be regarded as acting at a common point; a particle can have
only motion of translation. A rigid body is an aggregation of particles which
maintain their relative positions; it is not deformed by the forces acting on it,
and these forces do not, in general, act upon a common point of the body.
The motion of a rigid body can be either translation or rotation, or both.
The location of the center of gravity, § 2-7, has an important bearing on the
behavior of the body under the action of applied forces. If the various
forces acting on the body produce a resultant that passes through its center
of gravity, and if they do not form a couple, then the motion will be the same
as though the body were concentrated at the center of gravity, with all of the
forces applied to this point with their directions unchanged. Under these
conditions the body may be treated as a particle, and its motion will be pure
translation. If the forces acting upon the body produce only an unbalanced
couple, then the body will rotate about its center of gravity if free to do so.
Often the applied forces are equivalent to an unbalanced force plus a
couple, and the resulting motion is due to both. For example, when a stick
is picked up by one end and thrown, the center of gravity will move as though
the stick were concentrated at that point with the unbalanced force acting
upon it there, while at the same time the stick will be set into rotation about
its center of gravity by the action of the couple. Throughout the present
chapter the resultant force acting upon a body is assumed to pass through
its center of gravity.
4-2. Mass. From the statements at the beginning of this chapter, it is
possible to explain what is meant by the mass of a body. Inertia is a gen-
eral property of matter, and mass is a measure of the inertia of a particular
body. Mass is that property of a body that requires an unbalanced force to
make the body accelerate. The larger the mass of the body, the larger the
force needed to cause a given acceleration.
The primary standard of mass is the kilogram. The kilogram is defined
as the mass of a certain platinum-iridium cylinder preserved at the Interna-
tional Bureau of Weights and Measures and known as the standard kilogram.
If the standard kilogram and some other body were to receive the same
GO MECHANICS Chap. 4

acceleration when acted upon by the same unbalanced force, then the other
body would also have a mass of 1 kilogram.
The pound is sometimes used as a unit of mass, although in a system of
units not very widely used. The pound of mass is the mass of the platinum
block referred to in §1-4; to avoid duplication of standards, however, the U.S.
pound is defined (to five significant figures) as 0.45359 kilogram. The unit
of mass gaining recognition in engineering practice is the slug. To the near-
est whole number, a slug is equal to 32 pounds of mass, and is so used in prob-
lems; the reason for selecting this number will appear in §4-6. Throughout
the remaining sections of Mechanics in this textbook, the pound will be used
as a unit of force only, not mass. The slug is a unit of mass only, and is never
used as a unit of force.
Several units of mass, their abbreviations, and the relations among them,
are listed in Table I.

TABLE I. u NITS OF MASS

Metric British

1 kilogram (kg) = 1000 grams l pound = 453.6 grams

1 gram (gm) = 1000 milligrams (mg) l pound = 0.03108 slug
1 kilogram 2.205 pounds l slug 14.59 kilograms
1 kilogram 0.0685 slug l slug = 32.17 pounds

The mass of a substance per unit volume is known as the density of that
substance. For water the mass of 1 cm 3 is 1 gm; therefore, the density is
1 gm/cm 3 • Again, for water the mass of 1 ft 3 is 1.95 slugs, hence the density
is 1.95 slugs/ft 3 • Density is discussed more fully in § 8-2.
4-3. Newton's Laws of Motion. The relation of force to motion was set
forth by the English philosopher and mathematician Sir Isaac Newton (1642-
1727), one of the most profound scientists of all time. He interpreted and
correlated many diverse observations in Mechanics and combined the results
into three fundamental statements, now known as Newton's Laws of :Motion,
which have vastly simplified the science. In the following paragraphs, each
of these laws is stated in terms of quantities aiready described and is sup-
plemented by a brief discussion.
First Law of Motion. A body at rest remains at rest, and a body in motion
continues to move at constant speed along a straight line, unless the body is acted
upon in either case by an unbalanced force.
The first part of this law is evident from everyday experience; for instance,
a book placed upon a table will remain at rest. In explanation, one might
be inclined to say that the book stays at rest simply because no force acts
upon it. Further thought shows this reasoning to be false, for it is knmvn
that the book is pulled downward by the force of gravity and it can be inferred
that it is pushed upward by the table. The fact is that these two forces on
§ 4-3 NEWTON'S LAWS OF MOTION

the book are equal and opposite, one balancing the other;· hence, there is no
unbalanced force acting on it.
The second part of the law is more difficult to visualize; it states in effect
that if a body is set into motion and left to itself, it will keep on moving with-
out any further force being exerted on it. This statement is correct; experi-
ence shows, however, that a retarding force is always present in the nature of
friction. A block of wood thrown along a rough road slides only a short dis-
tance because the friction is large; along a floor it would slide farther, the
friction being smaller; along a sheet of ice it vmuld slide much farther, since
here the friction is still smaller. From examples like these, it is reasoned
that if friction could be eliminated entirely-a practical impossibility-a
body once set into motion on a level surface would continue to move indefi-
nitely with unchanged velocity. Thus, uniform motion along a straight line
is a natural condition and maintains itself without the action of any unbal-
anced force. Indeed, a body moving with constant velocity satisfies the
conditions of equilibrium just as
truly as if it were at rest. The tend- Lift
ency of a body to remain at rest or Drag
in uniform motion along a straight Thrust
line is known as inertia, as men- c
tioned previously.
An airplane traveling horizontally
at constant speed illustrates the first Weight
law of motion. The forces on the Frn. 4-1. Forces on an airplane moving with
plane are indicated in Fig. 1: the constant velocity.
propeller thrust acts forward along
the propeller axis, the drag of air friction is backward, the weight acts ver-
tically downward, and the "lift" acts upward at right angles to the thrust of
the propeller. When the thrust equals the drag and the lift equals the weight,
there will be no unbalanced force on the plane; consequently, the velocity it
has attained will continue without change as long as the propeller thrust can
be maintained.
It is interesting to note that whether a body is at rest or is moving with
constant speed along a straight line, its acceleration is zero. Hence, the first
law of motion states that a body will not accelerate unless an unbalanced
force acts upon it.
Second Law of Motion. An unbalanced force acting on a body causes it to
accelerate in the direction of the force, and the acceleration produced is directly
proportional to the unbalanced force and inversely proportional to the mass of
the body.
The meaning of this law will be illustrated by supposing that two identical
boxes are being drawn across the floor, and that more force is applied to one
than to the other. Experience shows that the box acted upon by the larger
force will receive the greater acceleration. If one box were full and the other
62 MECHANICS Chap. 4

empty and if both were subjected to the same force, then the empty one
would receive the greater acceleration. In general, the larger the unbalanced
force or the smaller the mass, the greater will be the acceleration. If either
box is started in motion across the floor and the applied force is reduced until
it is equal to the backward force of friction that the floor exerts on the box,
then the box has one force pulling it forward and an equal force pulling it
backward. Therefore, the unbalanced force acting upon the box is zero, and
the box will continue to slide with constant speed, as expressed in the First
Law.
Third Law of Motion. For every action, there is an equal and opposite reac-
tion, and the two act along the same straight line.
In the foregoing statement, the term "action" means the force that one
body exerts upon a second body, and "reaction" means the force that the
second body therefore exerts upon the first. A useful way to express this
law is as follows: If body A exerts a force on body B, then body B must exert
an equal and opposite force on body A.
Some illustrations will clarify the law. If a book presses downward on a
table with a force of 2 lb, then the table presses upward on the book with a
force of 2 lb. Again, if a man pulls on a rope ·with a force of 50 lb, the rope
pulls in the opposite direction on the man with a force of 50 lb. Evidently
the ability to exert a force depends not only on the agent "\Yhich is exerting
the force but also on the agent which is supplying the reaction. A truck
striking a tree can exert only as much force on the tree as the tree is able to
exert against the truck. It will be observed that two bodies are involved in
each of these examples, and that the action and reaction are never exerted on
the same body. Thus, action and reaction, although equal and opposite, can
never balance each other, since in order for two equal and opposite forces to
balance each other they must be exerted on the same body.
The first and second laws of motion deal with forces acting on a single
body; the third law deals with the mutual forces acting between two bodies.
4-4. Relation of Force to Mass and Acceleration. Of the three laws of
motion discussed in the preceding section, only the Second Law deals quan-
titatively with force, mass, and acceleration. Expressed mathematically, the
law states that
F
a o: - ,
m
where a is the acceleration given to a mass m by an unbalanced force F. A
constant of proportionality k can be introduced to form an equation; thus,
ka = F/m, or
F = kma,
in which the value of k depends upon the units used in the expression.
To make use of this relation for numerical calculations, units of force or
mass must be established which will be consistent with the other units already
adopted. Several systems of units are recogni;.,;ed; the most common ones
§ 4-4 RELA.TION OF FORCE TO kIA.SS A.ND A.CCELERA.TION G3

are described in the following paragraphs and will be used throughout this
textbook. In each of them, the units of force and mass are so related that
a unit of force acting on a unit of mass will give it unit acreleration. Then,
in the foregoing equation, k becomes unity, and the equation can be rewritten
in the simpler form
F = ma. (-1.1)
In British gravitational units, used quite widely in America for engineering
purposes, the unit of force is 1 lb, unit acceleration is 1 ft/ sec 2 (both defined
previously), 2J1d the unit of mass is 1 slug. Consequently, the slug is defined
.as the arnovnt of rnass that will receive an acceleration of 1 ft/ sec2 when acted
vpon by an unbalanced force of 1 lb. These units substituted in the foregoing
ft
equation show that 1 lb = 1 slug X 1 - -2 ; hence, the term "pound of force"
sec
. . slug· ft
can a1,vays 1)e rep1ace d 1)Y its eqmva1ent, ? •
sec-
The metric units, universally used in scientific work, belong to two abso-
lute~stems already referred to in §1-1 as the mks system and the cgs system.
In the mks system the unit of mass is 1 kg, unit acceleration is 1 m/sec 2 ,
and the unit of force is named the newton, after the great English scientist.
The newton is defined as the unbalanced force that gives a mass of 1 kg an
acceleration of 1 m/sec2 • The term "newton" can always be replaced by its
kg·m
equivalent, --., The mks system is particularly useful in the study of
sec-
electricity, because it simplifies many of the equations of that subject.
In the cgs system the unit of mass is 1 gm, unit acceleration is 1 cm/sec2 ,
and the unit of force is called the dyne. The dyne is defined as the unbalanced
force that gives a mass of 1 gm an acceleration of 1 cm/sec2• The term "dyne"
'
can always b e rep1ace d b y its
· eqmva· Ien t , gm· cm •
sec2
Table II lists the units for several basic quantities, including force, mass,
and acceleration, in each of the three unit systems mentioned. In any expres-
sion-F = ma, for example-the units will be found consistent if all are
taken from the same system.

TABLE II. UNITS FOR SoME BASIC QuAN'rITIES

British Metric
Quantity
Gravitational Mks Cgs

Length ................. ft m cm
Time ................... sec sec sec
Velocity ................ ft/sec m/sec cm/sec
Acceleration ............. ft/sec 2 m/sec 2 cm/sec2
Force ................... lb newton dyne
Mass ................... slug kg gm
G4 MECHANICS Chap. 4

To illustrate the use of these units, two elementary problems will be consid-
ered in which objects are caused to slide along "smooth" surfaces; that is,
surfaces for which friction can be neglected.
I. Suppose that a body having a mass of 5 slugs is placed on a smooth horizontal
surface and is acted upon by a steady horizontal force of 10 lb. What is the resulting
acceleration?
Three forces act upon the body: its weight acting downward, the upward reaction
of the supporting surface, and the horizontal 10-lb force. The vertical forces balance
each other, and the applied force of 10 lb acts as an unbalanced force on the body.
The acceleration, from Eq. 4.1, is as follows:

10~
a = !!_ = 10 lb = sec 2 = 2 .0 ~ ,
m 5 slugs 5~ sec2

in which the term "lb" is replaced by slug·2ft.

sec
II. What horizontal force must be applied to a 1.4-kg body on a smooth horizontal
plane to give it an acceleration of 3.5 m/sec 2?
The force analysis is similar to that in Prob. I, and the force required is found to be
m kg·m
F = ma = 1.4 kg X 3.5 - = 4.90 - - = 4.90 newtons.
sec 2 sec 2

Several units of force and their conversion factors appear in Table III.

TABLE III. UNITS OF FORCE

Metric British

1 newton = 105 dy:nes 1 ton = 2000 pounds

1 newton = 0.2248 pound 1 pound = 16 ounces (oz)
1 dyne = 3.597 X 10-5 ounce 1 pound = 4.448 newtons

4-5. Law of Universal Gravitation. Newton's investigations in Mechanics

were not limited to the observations incorporated in his three Laws of :Mo-
tion, but extended to the general subject of gravitation. He postulated that
every body in the universe attracts every other body, and showed how this
attraction is influenced by the masses of the bodies concerned and the dis-
tance separating them. The Law of Universal Gravitation follows: Any
two particles of matter attract each other with a force that is directly propor-
tional to the product of their masses and inversely proportional to the square of
the distance between them. Expressed mathematically,
F = Gm1m2, (4.2)
d2
\Vhere m 1 and m2 are the masses of the particles, d is the distance between them,
F is the force with which either particle attracts the other, and G is known as
§ 4-6 PL'LL OF GRA VIT1'; WEIGHT 65

the gravitational constant-not to be confused 'With g, the acceleration due to

gravity. When F is expressed in dynes, m1 and m2 in grams, and din cen-
timeters, G has the value 6.67 X 10-s dyne· cm 2/ gm 2 ; thus, two particles
1 cm apart and each having a mass of 1 gm attract each other with a force of
6.67 X 10-s dyne.
Gravitational attraction can be observed by means of two metal spheres
mounted on the ends of a slender rod that is suspended horizontally by a
thin wire, with the spheres close to a pair of massive lead globes. When the
globes are moved slightly, the spheres tend to follow them. From a measure-
ment of the torque set up in the suspension wire, and a knowledge of the
masses and their separation, the value of the gravitational constant is deter-
mined.
4-6. Pull of Gravity; Weight. The force of attraction which the earth
exerts upon objects near it has been mentioned previously and is the most
familiar illustration of universal gravitation. It is because of this force that
a body accelerates downward when it is allowed to fall. If the body is con-
strained so that it cannot fall, the earth still exerts the same force on it, but
the pull of the earth is balanced by an equal and opposite force exerted by
the restraining agent. Thus, a box resting on a table is pulled downward
by the earth, but it is pushed upward with an equal force by the table, and so
it stays at rest. If the table were flimsy and incapable of pushing upward on
the box with as much force as the earth pulls downward on it, the unbalanced
force on the box would make it accelerate downward, and the table, being in
the way of its motion, ,vould collapse. The force of attraction that the earth
exerts on a body-that is, the pull of gravity on it-is called the weight of
the body and indicates how heavy the body is. The weight of a body is,
therefore, a force and is treated in exactly the same way as any other force.
Its direction is toward the center of the earth.
It was stated in § 3-7 that the acceleration due to gravity varies slightly
from point to point on or near the earth's surface; thus, a body dropped at
the pole would have a slightly greater downward acceleration than if it were
dropped at the equator. This increased acceleration must be caused by an
increased force of gravity; hence, the body must weigh slightly more at the
pole than at the equator. The same result follows from the Law of Universal
Gravitation. Since the earth is flattened at the poles, a body located there
is nearer the center of the earth than if it were at the equator; thus, the dis-
tanced in Eq. 4.2 is less, and consequently the attraction Fis greater.
It is a matter of everyday experience that bodies of different masses also
have different weights. A full barrel not only has a larger mass than an
empty one, but it also has the larger weight. This result also is to be expected
from the Law of Universal Gravitation. If m1 represents the mass of the
earth and m2 the mass of any other object, then, for any particular value of
d, it follows that Fo:. mz. Hence the weights of bodies are directly propor-
tional to their masses, a statement that is strictly true for any given location
66 MECHANICS Chap. 4

and is approximately true for all places on the earth's surface where the
weights might be measured. This proportionality can be expressed as
W = cm, where c is a constant to be determined. It follows that Newton's
Second Law of Motion, Eq. 4.1, can be restated mathematically in the form

F= W a
c '
in which a is the acceleration produced in a body of weight W by an unbal-
anced force F. To evaluate the constant c, it may be observed that if an
object were allowed to fall freely, the unbalanced force acting upon it would
be its weight W, and the resulting acceleration a would necessarily be that
due to gravity, namely, g. Therefore, when F becomes TV, a becomes g,
and the equation reduces to W = W g. Hence c = g, and the original equa-
c
tion becomes
(4.3)

In this expression, a is the acceleration produced by an unbalanced force F

acting upon a body of weight W, and g is the acceleration due to gravity.
The form of Eq. 4.3 shO\vs that if the accelerations a and g are expressed
in the same units of acceleration, then the forces F and TV are to be expressed
in the same units of force. The units commonly used are as follows:
British: forces in pounds, accelerations in feet per second2 ;
mks: forces in newtons, accelerations in meters per second2 ;
cgs: forces in dynes, accelerations in centimeters per second2 •
As an illustration, the unbalanced force needed to give a 100-lb body an acceleration
of 4 ft/sec 2 is
W 100 lb rt-.
F= - a= - X4- = 12.5 lb.
g 'it--- ~
32-
~

The relation between weight and mass, expressed above by the equation
W = cm, can be stated more usefully now that the proportionality constant
c has been shown to equal g, by writing it in the form

W = mg, (4.4)
where Wis the weight of a body, mis its mass, and g is the acceleration due
to gravity. For example, a body of mass m = 1 slug has a weight W = mg
= 1 slug X 32 ft/sec 2 = 32 slug·ft/sec 2 = 32 lb. Similarly, a body of mass
m = 1 kg has a weight W = 1 kg X 9.8 m/sec 2 = 9.8 kg·m/sec 2 = 9.8
newtons. Also, a body of mass 1 gm has a weight W = 1 gm X 980 cm/sec2
= 980 gm· cm/sec 2 = 980 dynes.
4-7. Problems in Force and Translation. Experience has shown that diffi-
culty can be avoided in the solution of problems in dynamics by follO\ving a
§ 4-7 PROBLEMS IN FORCE AND 'TRANSLATION 67

definite procedure, at least until a thorough familiarity with the foregoing

principles has been attained. In any problem involving force and accelera-
tion, the following steps are suggested:
1. E:elect some one body for consideration, usually the body which is in
motion.
2. Construct a force diagram, entirely separate from any picture (or
"space" diagram) that may be used to show the conditions of the problem.
On this force diagram, use a point to represent the body which has been
selected and represent all the forces acting on this body by suitable vectors.
Be careful that no forces are omitted, and also that only forces acting on
the body are used (not forces that the body may be exerting on other things).
If any forces are unknown, represent them also by vectors, but mark them
as unknown quantities.
3. F'rom the force diagram, find the resultant or unbalanced force acting
on the body. When some of the forces are unknown, the expression for the
unbalanced force will involve these unknown quantities.
4. Introduce the unbalanced force in either Eq. 4.1 or Eq. 4.3, and solve
for the acceleration or unknown force, according to the objective of the
problem.
5. Should the problem call for the displacement or the resulting velocity,
then use the equations of accelerated motion in Chap. 3 to find the quantity
required.
Several illustrations, with their analyses, follow:
I. What acceleration will a horizontal force of 25 lb produce in an 80-lb block placed
on a rough horizontal plane, if there is a backward frictional forre of lO lb acting on
the block?
This problem is visualized by constructing the simple diagram indicated in Fig. 2.
The block in part I is represented in
part II by point 0, and the forces which 80lb
act upon it are represented by vectors
drawn to scale and properly directed.
The vertical forces balance each other,
and the 25-lb horizontal force is partly ~ o
balanced by the 10-lb force of friction. ~ ~ - 10 lb 25 lb
The resultant or unbalanced force is,
therefore,. horizontal and amounts to
15 lb. Equation 4.3, with the terms
rearranged, gives the acceleration as
n 80lb
ft
F 15~ X 32 sec2 ft Fm. 4-2. An unbalanced force produces ac-
a=~= =60-· celerated motion. The unbalanced force on
W 80~ · sec 2 the block of Prob. I is 15 lb.
II. A 2-kg block slides from rest down
a smooth plane inclined 20° with the horizontal. Find its velocity after sliding a
distance of 4 m.
A diagram of block and plane is shown in part I of Fig. 3. The block is chosen as
the body for consideration and is represented by point O in the force diagram shown
(\8 MECHANICS Chap. 4

as part II of the figure. Two forces only act on the block-in symbols these are:
W, the downward pull of gravity, and P, the outward push of the plane. Since
the plane is smooth, it can exert no frictional drag on the block along the incline;
hence, the force P must be at right angles to the plane. Resolve the weight W into
components, one perpendicular to the plane and the other along the plane. The
angle indicated in part II of the figure is 20°, since its sides are perpendicular respec-
tively to the incline and the base in part I; consequently, the components of W are

Wsin 20°

Wcos 20°

Fm. 4-3. Forces on a block sliding down a smooth plane.

W cos 20° and W sin 20°, as shown. The first of these, W cos 20°, balances the push P,
and the other, W sin 20°, is the unbalanced force that accelerates the block down the
plane. Since the weight of the block is
m
W = 2 kg X 9.8 - = 19.6 newtons,
sec2
it follows that F = W sin 20° = 19.6 X 0.342 = 6.70 newtons, and the acceleration
of the block, from Eq. 4.1, is
6.70 kg·~
F 6.70 newtons
a= - =
___se_c_ = 3 _35 m .
m 2 kg 2 kg sec2
This value, when substituted in Eq. 3.12, v/ = Vo 2 + 2 as, gives
m m2
v/ = 0 + 2 X 3.35 sec
- 2 X 4 m = 26.8 - 2 ,
sec
from which the velocity of the block along the incline as it passes the specified point
is v1 = 5.18 m/ sec.

6
III. What upward force must be applied to the cable of a
p 10-ton elevator in order that the car may have an upward ac-
Cable celeration of 4 ft/sec 2, if friction is neglected? 1
O The elevator, shown in part I of Fig. 4, is acted upon by two
forces: W, the downward pull of gravity, and P, the unknown
W upward pull of the cable. In part II of the figure, the elevator
Elevator is indicated by point 0, and the forces Wand Pare represented
II by vectors pointing in the appropriate directions. Since the
elevator is to accelerate upward, the upward force P must be
Fm. 4-4. Forces on greater than the downward force W, and the excess, P - W,
an elevator. must be the unbalanced force F acting on the elevator. The
values W = 10 X 2000 = 20,000 lb, g = 32 ft/sec 2, and a = 4
ft/sec 2 are substituted in Eq. 4.3 as follows:
§ 4.:7 PROBLEMS IN FORCE AND TRANSLATION 69

W
F = P - 20 000 lb = - a =
20,000 lb t\
X 4 - - = 2500 lb·
' g ft ~ '
32--
~

hence, the upward force on the elevator is P = 20,000 +

2500 = 22,500 lb.
IV. A device known as the Atwood machine, commonly used in the study of ac-
celerated motion, consists essentially of two unequal masses joined by a cord hung
over a pulley. If one of these masses is 500 gm and the other 400 gm, what is the
tension in the cord and the acceleration of the moving system? Disregard friction
and neglect the mass of the cord and that of the pulley.

T T
500-gm 400-gm
• mass mass
0 O'
490,000 392,000
500 gm 400gm dynes dynes

II III
FIG. 4-5. Diagram for the Atwood machine. I, Two masses suspended over a pulley;
II, the forces acting on one mass; and III, the forces acting on the other mass.

The two masses are represented in part I of Fig. 5 and will be considered separately
in solving the problem.
The 500-gm mass is acted upon by two forces: the downward pull of gravity, which
equals 500 gm X 980 cm2 = 490,000 dynes, and the unknown tension T in the cord
sec
pulling upward, as shown at O in part II of the figure. Since this mass accelerates
downward, the downward force must be greater than the upward force, and the differ-
ence between them, namely 490,000 dynes - T, is the unbalanced force acting on the
500-gm mass. The expression F = ma is applied to this body alone, thus:
490,000 dynes - T = 500 gm X a.
This equation cannot be solved by itself, since it contains two unknown quantities,
T and a.
The 400-gm body is acted upon by a downward force of 400 gm X 980 c~ = 392,000
sec
dynes, and by the upward force T, as indicated at 0' in part III of the figure. The
excess upward force acting upon this body is T - 392,000 dynes, and therefore,
T - 392,000 dynes = 400 gm X a.
This equation also contains the same two unknown quantities.
The tw.o equations when solved simultaneously give the tension T in the cord as
4.36 X 105 dynes and the acceleration a of the moving system as 108.9 cm/sec 2•

In motion where the acceleration varies from instant to instant, the unbal-
anced force that causes the acceleration must vary in a similar manner, for
acceleration and unbalanced force are proportional to each other, as stated
in Newton's Second Law of l\fotion.
70 MECHANICS Chap. 4

FRIC'l'ION

4-8. Force of Friction. The surfaces of solids are seen under magnification
to be uneven, and when two bodies are placed in contact, their surface irregu-
larities intermesh to some extent. Modern experiments indicate, moreover,
that two surfaces in contact, even if they are of polished metal, touch only
at the peaks of these roughnesses, and that when the surfaces are pressed
together the peaks become crushed and yield until their total area is suffi-
cient to support the load. The peaks of one surface may become welded to
those of the other, and when a force produces sliding the peaks are sheared
off and plow up the surfaces. Because of these effects, opposition is encoun-
tered in sliding one surface along another; this opposition is called sliding
.friction. It must not be inferred that friction has no useful aspects; indeed,
friction is commonly needed in order for a person to walk or a wheel to roll
on the ground. A perfectly smooth surface would be one without friction;
such a frictionless surface is never attained in practice but is assumed some-
times, for simplicity.
Many tests have shown that the value of the force of friction depends
upon the materials in contact, the condition of the sliding surfaces, and the
force with which they are pressed together, but is virtually independent of
the speed of sliding over a \Yide range of speeds. For years, it has seemed
that friction is unaffected by the size of the contacting surfaces; it now
appears, however, that friction is proportional to the area of actual contact,
but depends little, if any, upon the area of apparent contact. These state-
ments, based on experimental results, are true only approximately, even for
dry surfaces. The force of friction has the same properties and is expressed
in the same units as any other force.
Much can be learned about friction by merely sliding a book around on a
table top. It will be observed that the force of friction always acts in the
plane of sliding, and that no matter which way the book is moved, the force
of friction always acts on it in the opposite direction. If a person exerts a
horizontal force of t lb on the book toward the right and the book does not
move, then the unbalanced force acting on it must be zero, and therefore the
surface of the table must be exerting a force of friction of l lb on the book
toward the left. It should not be thought that since the book stays at rest,
the force of friction is greater than the !-lb force applied; if this were so, the
book move backward-an absurdity. If the person increases his push on
the book from t lb to ~ IL and finds that the book still does not move, it
means that the force of friction has increased to ~ lb. If he sets the book in
motion and then finds that a horizontal force of 1 lb is just sufficient to keep
it sliding at constant speed, the unbalanced force acting on the book is again
zero and the force of friction acting on the book now amounts to 1 lb.
When a body is starting from rest, the friction experienced when motion
is about to begin is called static .friction; it is somewhat greater than sliding
friction, but for simplicity this difference will be neglected.
§ 4-9 COEFFICIENT OF FRIC'TION 71

It is common experience that a heavy object can be moved much more

easily by rolling it along on two pieces of pipe than by sliding it along the
floor. Rolling friction is much smaller than sliding friction, and this is the
reason that wheels are used instead of runners on wagons, that casters are
provided on heavy furniture, and that for many purposes ball bearings are
preferred to those of the sleeve type. In the sleeve bearing the shaft slides,
whereas in the hall bearing it rolls, and although some sliding takes place,
the force of friction is much reduced. Rolling friction is probably due to
slight deformations of the bodies at the points of contact.
4-9. Coefficient of Friction. The force of friction that one surface exerts
upon another sliding along it is directly proportional to the normal or per-
pendicular force with which the surfaces are pressed together. Thus, if f is
the force of sliding friction between two surfaces, and N is the normal force
pressing them together, then f cc N, or

f = µN,
where the proportionality factor µ (mu) is called the coefficient of sliding
friction. This expression can be transposed to read

µ = 1,
N
(4.5)

from which the coefficient of sliding friction is defined as the ratio of the force
of sliding friction between two surfaces to the normal force pressing them together.
The same, definition would apply to the coefficient of static friction, except
that f would be the force of static friction, \vhich is some\vhat larger than the
force of sliding friction. Since f and Nin Eq. 4.5 are both forces they may
be expressed in any desired force unit, so long as
the same unit is used for both. The coefficient of
friction µ is a numeral and has no unit.
The frictional force f is always in the plane of
-
Motion

sliding, and the normal force N pressing the sur-

faces together is perpendicular to this plane; hence,
f and N are always at right angles to each other. Frn. 4-6. A block being
Fig. 6 shows the forces acting upon a block whieh brought to rest by friction.
has been set in motion toward the right and iE; now
sliding to rest along a level plane. In this example the supporting surface is
horizontal, and so the normal reaction N of the plane is equal and opposite to
the weight W of the block. The force of frictionj is at right angles to N and
opposite to the direction of motion.
The coefficient of sliding friction may be taken as constant for two par-
ticular surfaces; it depends only on the materials involved and the condition
of the contacting surfaces. Some representative values for dry surfaces
appear in Table IV.
72 MEC!IANICS Chap. 4

TABLE IV. CoEFFICIEN'.rs OF SLIDING FmcTION

Wood on wood ......... . 0.25 to 0.5
Metals on wood ....... . 0.2 to 0.6
Metals on metals ....... . 0.15 to 0.2
Leather on oak ......... . 0.27 to 0.38
Leather on metals ....... . 0.56

For surfaces which are carefully machined and thoroughly lubricated, the
coefficient of sliding friction is much smaller, 0.005 being a representative
value. Both rolling friction and sliding friction are treated in the same way
in numerical problems. Some typical values of the coefficients of rolling
friction are given in Table V.
TABLE v. COEFFICIENTS OF ROLLING FRICTION

Cast-iron wheels on rails ......... . 0.004

Roller bearings in rolling contact 0.002 to 0.007
Ball bearings in rolling contact .... 0.001 to 0.003

Suppose that a sled, starting from rest, is drawn on level snow by a constant force
of 10 lb applied at an angle of 30° upward from the horizontal. If the sled with its
load weighs 45 lb, and if the coefficient of sliding friction between the sled and the
snow is 0.05, what is the velocity of the sled at the end of 5 sec, and how far does it
travel in this time?
In accordance with the procedure of § 4-7, the sled, shown in part I of Fig. 7, is
represented at O in the force diagram forming part II of the figure. The forces acting

R
Pull of
lOlb
45-lb sled"' A~o p cos 30°
I J
II w
Frn. 4-7. Analysis of the forces on a sled drawn along level snow.

on the sled are as follows: the downward pull of gravity, TV (45 lb); the pull P (10 lb)
applied in the direction shown; the upward reaction of the snow, R (unknown); and
the backward force of friction f (also unknown). The pull P is resolved into a hori-
zontal component P cos 30° = 10 X 0.866 = 8.66 lb, which is effective in drawing
the sled horizontally, and a vertical component P sin 30° = 10 X 0.500 = 5.00 lb,
which pulls the sled upward, reducing its force against the snow.
The net force which the sled exerts downward on the snow is 45 - 5.0 = 40.0 lb,
and therefore the upward thrust R of the snow on the sled is also 40.0 lb; this is the
normal force between the sliding surfaces. The frictional force, from Eq. 4.5, is
f = µN = 0.05 X 40.0 = 2.00 lb. The unbalanced force acting on the sled is di-
rected horizontally toward the right and amounts to 8.66 - 2.00 = 6.66 lb. Con-
sequently, the acceleration of the sled, from Eq. 4.3, is
§ 4-10 IMPL'LSE AND MOMENTUM

x 32 -ft2
6.66 lb
_ _ _ _ _s_ec_
f
Fg =
4 74 _t__
a= W 45 lb · sec 2
This acceleration value, substituted in Eq. 3.8, gives the final velocity as
ft ft
VJ = V0 + at = 0 + 4.74 sec
- 2 X 5 Sec = 23.7 - .
sec

Then, since displacement = average velocity X time, as stated in Eq. 3.10, the dis-
placement in 5 sec is found to be

s= v t = 0 +223. 7 -sec
ft
X
~
b sec =
~
b9 .3
f
t.

Fluid Friction. In addition to sliding and rolling friction, retardation is

also offered by fluids, either liquid or gaseous, to the motion of an object
through them. Air friction is of particular importance in the propulsion of
automobiles, trains, and aircraft. This retardation increases with velocity,
and at relatively low velocities is often regarded as directly proportional to
the speed. On this assumption, the retarding force can be expressed as
f o:. v or f = Rv, ,vhere vis the velocity of an object and Risa proportionality
factor which may be termed mechanical resistance. 'When this frictional
force is included in Eq. 4.1 for accelerated motion, the total force required to
produce an acceleration a = dv/dt, § 3-5, in a mass mis written as
dv
P = ma + Rv = m dt + Rv.
In starting, v = 0 and the entire unbalanced force produces acceleration;
when the body has accelerated to the maximum velocity permitted by the
frictional drag, v = constant, dv/dt = 0, and the entire force is exerted
against the mechanical resistance in maintaining the speed acquired. The
foregoing equation is useful in studying motion at relatively low velocities;
the behavior at high speeds is complicated by the fact that air resistance is
then more nearly proportional to the square of the speed.

IMPULSIVE FORCES

4-10. Impulse and Momentum. Newton's Second Law of Motion will be

expressed in a different form that introduces two new and useful concepts:
impulse and momentum. The equation F = ma is stated in the form
F = rn dv, then rewritten as F dt = m dv, and integrated, as follows:
dt

l~ l~ dt =
0
dv = rn VJ - rn Vo.

If the manner in which the force F varies with time t is known, it may be
possible to perform the integration; usually, however, this relation is un-
74 MECHANICS Chap. 4

known, and only the average value of the force can be found. Thus, if

F is regarded as constant, then l t

F dt = F l t
dt = F t, and: the equation
becomes
Ft = m VJ - m Vo, (4.6)
in which F represents the constant value of the force, or its average value if
it is variable.
The product of force and time, f F dt or F t, is known as linear impulse and
usually refers to a short time interval, as in a collision. The product of mass
and velocity, m v, is known as linear momentum. Both of these quantities
are vectors; impulse has the direction of the force, and momentum has the
direction of the velocity. Since the equation just stated was derived from
Newton's Second Law, F = ma, the comments in§ 4-4 regarding units apply
equally to both. As typical units, impulse may be stated in pound· seconds
and momentum in slug· feet per second. Equation 4.6 may be stated in words
as follmvs: An ·unbalanced impulse acting ·upon a body changes the momentum
of the body, and the change in momentum equals the unbalanced impulse.
As an example, suppose that a 10-gm bullet, fired with a velocity of 300 m/sec,
passes straight through a hoard 3 cm thick and emerges with a velocity of 200 m/sec.
Assume the deceleration of the bullet to be constant, and determine the force that the
bullet exerts upon the board.
The change in momentum of the bullet ism v1 - m Vo = 10 gm X (20,000 - 30,000)
cm/sec = -105 gm· cm/sec, the direction of motion of the bullet being taken as posi-
tive. Its average speed v while passing through the board is (30,000 20,000)/2 +
= 25,000 cm/sec, and the time required, from Eq. 3.10, is

s 3 cm
t = : = = 1.2 X 10-4 sec.
v 25,000 cm
sec
Therefore, the force exerted on the bullet by the board, from Eq. 4.G, is
_ 105 gm·cm
= m, Vf - m
F -- Vo __ sec
t 1.2 X 10-4 sec - 8.33 X 108 dynes.

The force exerted on the board by the bullet is equal and opposite, namely,
8.33 X 108 dynes, along the direction of motion.

4-11. Collision. The idea of impulse is particularly useful in situations

where the force acts only for a very short time, as in a collision. vVhen two
bodies collide, they exert equal and opposite forces on each other, in accord-
ance with Newton's Third Law of Motion. Furthermore, the time intervals
over which these forces act are also identical, and hence each body exerts the
same impulse on the other. The two bodies are said to form a "system,"
and the equal and opposite forces due to impact are merely internal forces
which one part of the system exerts on the other part; they do not affect the
§ 4-11 COLLISION 75

motion of the system as a whole. If there are no other forces acting to affect
the motion of the colliding bodies, then these bodies will be subjected to
equal unbalanced impulses and will undergo equal changes in momentum.
Thus, if one body gains a certain amount of momentum, the other loses the
same amount, and the total momentum of the system remains unchanged.
This statement illustrates a general principle known as Conservation of
Momentum, which states that, when a system of bodies is not acted upon by
any external forces, the total momentum of the system remains constant; this is
true along any direction.
To apply this principle, two colliding bodies will be considered, and their
motion \vill be resolved along rectangular coordinate axes X, Y, and Z. The
masses of the bodies Y1·ill be taken as m1 and m2, and their respective velocities
along rmy axis will be denoted by v1 and v2 before impact and by vi' and v/
after impact. Then, if no external forces act upon the system, the Law of
Conservation of Momentum states:
Along X axis m1(v1)x + m2(v2)x = m1(vi')x + m2(v2')x}
Y axis m1(v1)v + m2(v2)v = m1(vi')v + m2(v/)v (4.7)
Z axis m1(v1)z + m2(v2)z = m1(vi'),, + m2(v2')z
Velocities are taken positive in one direction and negative in the opposite
direction along any axis. From the form of these equations, it will be clear
that the only requirement regarding units is that they
shall match on both sides of an equation.
Two examples are given to illustrate the conserva-
tion of momentum; in both of them external forces \
\
affecting the motion may be neglected. \
\
\
\ \
I. A bullet of known mass is fired horizontally into a sta- \ \

tionary wood block suspended as shown in Fig. 8, and the \

\
\\
resulting velocity of the block with embedded bullet is de-
termined from the height to which it swings. From these
Bullet~ __
1--.--.
1
I
data: the speed of the bullet can be found by using the prin- ,--~-' I
ciples of momentum. Suppose that a 10-gm bullet moving D
with a velocity v in the manner described becomes embedded
Blo~;7_J
in a 20-kg block of wood and that the velocity of the block FIG. 4-8. A test to
and bullet after impact is 25 cm/sec. The momenturn of measure the velocity of
the system before collision equals that after collision along a bullet.
the direction of motion; therefore, by Eq. 4.7,
10 X v + 20,000 X O = (10 + 20,000) X 25,
the masses being in grams and velocities in centimeters per second. From this
equation the velocity of the impinging bullet is found to be v = 50,025 cm/sec.
II. Suppose an automobile to be hit by a truck at a road intersection, despite at-
tempts to stop both vehicles. Assume that at the time of the collision the 2-ton
automobile (mass = 125 slugs) is moving north at a speed of 50 mi/hr, the 6-ton truck
(mass = 37.5 slugs) is moving east at 30 mi/hr, and that the two become interlocked.
Find the velocity of the wreckage immediately after collision.
76 MECHANICS Chap. 4

A diagram as in Fig. 9 shmvs the conditions of the problem. Let the velocity of
the wreckage after collision have components Vn and v. toward the north and east,
respectively, and apply the Law of Conser-
N vation of Momentum along these direc-
r tions, separately, thus:
north direction
375-slug 125 X 50 +
375 X O = 500 X Vn,
truck
east direction
=500 Ye
30~
hr 375 X 30 + 125 X O = 500 X v.,
50 mi
with masses in slugs and velocities in
hr 125-slug
automobile miles per hour. From these expressions,
Vn = 12.5 mi/hr and v. = 22.5 mi/hr,
whence the resulting velocity of the wreck-
Frn. 4-9. Vector diagram representing age is V (12.5) 2 + (22.5) 2 = 25.7 mi/hr
the collision of an automobile and a directed at an angle tan-1 (12.5/22.5) =
truck. 29.1 ° north of east.

CIRCULAR l\foTION
4-12. Force Involved in Motion Along a Curve. Whenever a body moves
along a curve, an analysis shows that there is always an unbalanced force
acting upon the body and that this force is directed toward the center of the
curve. In accordance with Newton's Second Law of Motion, the unbalanced
force continually accelerates the body toward the center; in fact, it is this
lateral motion combined with the forward motion that gives motion along
the curve.
Many examples show the existence of the lateral force necessary for motion
in a curved path. \Vhen a locomotive encounters a curve, its forward mo-
tion brings the flanges on the wheels into close contact with the outer rail
and the rail presses inwardly against them; the locomotive, under the action
of this inward force, undergoes a change of direction and continues to follow
the curved track. Again, when a stone is whirled around at the end of a
cord, the cord becomes taut and pulls inwardly upon the stone. In other
instances, the cause of the lateral force is not so easily visualized. A little
thought, however, will show why the earth is able to follow its orbit around
the sun, or why an automobile can make a turn on a level street-nnless the
turn is too sharp or the street too slippery.
The central force which acts upon a body while it is moving around a curve
is called the centripetal force. For the locomotive rounding the curve, the
centripetal force is the inward push exerted by the outer rail upon the flanges
on the wheels. Since for every action there is an equal and opposite reac-
tion, the flanges on the wheels push outward against the rail. This reaction
is called the centnfugal force; it does not act upon the moving object and
hence is of subordinate interest. In all motion on curves, the centripetal
§ 4-13 ACCELERATION TOWARD CENTER 77

and centrifugal forces are equal and opposite, and both are exerted in the
plane in which the curve lies. Although equal, these forces cannot balance
each other, because they are not exerted on the same object.
4-13. Acceleration Toward Center. Motion at constant speed around a
circular curve is of special interest; here the centripetal force on the moving
body is constant and is directed exactly toward the center. In studying this
type of motion, emphasis is placed upon the fact that even though the body
moves at constant speed its velocity changes continually, the change taking
place in direction but not in magnitude. Part I of Fig. 10 indicates a circle
of radius r, and a particle P moving around it at constant speed v. It is
desired to compute the acceleration a of the particle and to show that this

0
n m
FIG. 4-10. A body moving uniformly around a circle has an acceleration toward the center.

acceleration is directed toward the center 0. If the particle moves from 11f
to N, shown in part II of the figure, during a time interval t, then MN = v t.
The velocity of the particle at 1\/I and at N can be represented by vectors v0
and VJ, tangent to the circle at these points, respectively, and each having
the same magnitude as v. These velocities differ in direction, and thus some
additional velocity must have been given to the body in moving from kl to
N in order to change its velocity from Vo to VJ. This additional velocity can
be found by drawing v0 and VJ from a common point Q, as in part III of the
figure, and subtracting v from v,. This result is obtained by reversing the
0

direction of Vo and proceeding as in vector addition, § 1-7; the change in

velocity is found to be vd, Since this velocity change occurs in a time interval
t, the average acceleration over this interval is a = va/t. To obtain the
instantaneous acceleration, it will be observed that the two angles marked (J
are equal, and that as the time interval is taken shorter and shorter, approach-
ing zero as a limit, the sector of angle (J in part II becomes more and more
nearly similar to the isosceles triangle of apex angle (Jin part III. Hence, in
the limit,
Va_ MN at vt
or - = -.
VJ - OM

Numerically, VJ = v, and hence the magnitude of the instantaneous arcelera.-

tion a i.s obtai11ecl from this equation as
78 MRCllANICS Chap.4

vz
a (4.8)
r
The direction of this acceleration can be found by observing that, as the
time interval is taken shorter and shorter, the more nearly Va becomes per-
pendicular to Vo and v1 ; in the limit it is perpendieular to both of these vec-
tors. Therefore, the acceleration is truly centripetal; that is to say, it is
directed toward the center of the circle, as previously stated.
The force that must be exerted upon a body to produce the acceleration is
given by Eq. 4.1, F = ma; hence, the centripetal force acting upon a body of
mass m ,vhen moving with speed v around a curve of radius r is given by the
express10n
v2
F=m-, (4.9)
1'
in which consistent units ,vill be suggested by the outline in § 4-4.
Alternately, centripetal force can be expressed in terms of Eq. 4.3 as
follows:
FT = v2/r, (4.10)
Tl g

wherein the centripetal force F and the weight W of the moving body are in
the same force unit, and the accelerations v2/r and g are in the same accelera-
tion unit.
It will be observed that centripetal force varies directly as the square of
the speed and inversely as the radius. Thus, to travel around a given curve
with doubled speed requires four times as much centripetal force; to main-
tain a given speed around a curve of half the radius requires twice as much.
As an illustrative problem, suppose that a 100-gm ball is whirled in a vertical circle
at the end of a cord 50 cm long. If its speed is 400 cm/sec at the top point of the
circle, energy considerations (Chap. 6) show that its speed will be 597 cm/sec at the
bottom point. It is desired to find the tension in the cord as the ball passes each of
these points.
Two forces only act upon the ball; its weight W = mg = 100 gm X 980 cm/sec2
= 98,000 dynes, and the tension 'l\ in the cord. At the top point both of these act
downward and together make up the centripetal force mv2/r, which at this point has
cm2
100 gm X (400) 2 - 2
sec
the value = 320,000 dynes. Hence, the tension in the cord as
50 cm
the ball passes the top point is 'J'i = 320,000 - 98,000 = 222,000 dynes. At the
bottom point the weight W acts downward and the tension 'l\ upward, and T2 must
be larger than W, for the unhalanced force must be directed toward the center of the
circle; at this point the difference T 2 - W constitutes the centripetal force, which has
cm 2
100 gm X (597) 2 - 2
the value sec = 713,000 dynes. Therefore, the tension in the cord
50 cm
as the ball pasRes the bottom point is T2 = 713,000 + 98,000 = 811,000 dynes.
§ 4-14 BANKING OF CURVES 79

As another example, compute the highe8t speed at whirh an automobile (of

weight HT) can round a curve of 2,5-ft radius on level ground. taking the coefficient
of frirtion behveen tires and ground as 0.3. In this example, the centripetal for<'e
Wv 2/g r is provided entirely by the force of friction µ W, whence, µ = v2/g r and the
corresponding speed is v = V µgr =
'\J/0.3 ~-;2 secft ~ 25 f~
2 = 15.5 ft/sec.

4-14. Banking of Curves. When an automobile rounds a curve at constant

speed, it is accelerating toward the center of the eurve; accordingly the
resultant force acting upon it must also have this direction. If the highway
is level, such a foree can come only through friction between tires and road-
way. Under these circumstances, the central force is likely to be small and
uncertain, and the turn must be made slowly to avoid skidding. Curves on
highways are banked by sloping the roadbed upward from the inner to the
outer edge; when the angle of banking is correct, a car can round the curve
at a given speed without relying upon friction.

' \
I \
\

e _____ L __ \
w
I Il
Fm. 4-11. Forces on a car rounding a banked curve.

In Fig. 11, part I represents a car moving horizontally around a curve

banked at an angle e, the roadbed being shmvn in cross-section. Two forces
act upon the car, as shown in part II of the figure: the pull W of gravity
do,Ynward, and the push P exerted by the road. The latter force is taken
perpendicular to the road, since friction is not considered. The forces TV
and P must be so proportioned that their resultant, F, is horizontal, this
being the centripetal force needed for circular motion around a horizontal
curve. Since the two angles marked e are equal, it follows that
Wv 2
F gr vz
tan e= w = w gr '
v2 .
and hence the banking angle is e=tan~ 1 - , where v 1s the speed of the car,
(J r
g the acceleration due to gra,·ity, and r the radius of the curve.
80 MECHANICS Chap. 4

PROBLEMS
1. How much unbalanced force is needed to give a mass of 5 slugs an acceleration
of 3 ft/sec 2 ?
2. Find the acceleration that will be produced in a 10-kg mass when acted upon
by an unbalanced force of 2 newtons.
3. Compute the mass of a body that is given an acceleration of 50 cm/sec2 by an
unbalanced force of 104 dynes.
4. (a) Express a force of 40 newtons in pounds. (b) Express a force of 5 X 106
dynes in pounds.
5. A horizontal force of 165 newtons is applied continuously to a box having a
mass of 50 kg. How far ,vill the box move along a level floor in 3 sec, starting from
rest, if friction exerts a backward drag of 120 newtons upon it?
6. A horizontal force of 700 lb is applied continuously to a truck of 200 slugs mass
along a level road. If the truck moves 100 ft in 10 sec, starting from rest, what is
the backward force of friction acting upon it?
7. The standard bowling ball weighs 16 lb and has a diameter of 8t in. Consider
-the ball to be a uniform sphere and compute (a) its mass in slugs, and (b) the density
of the material of which it is made.
8. Take the density of concrete to be 4.7 slugs/ft3 and compute the mass in slugs
and the weight in pounds of a concrete block that measures 2 X 3 X 4 ft.
9. A steady horizontal force of 40 newtons is applied to a sled weighing 300 newtons
which is resting on level snow. What Rpeed will the sled acquire in 5 sec if friction is
neglected?
10. What steady horizontal force must a locomotive exert on a train weighing 200
tons in order to increase its speed along a level track from 10 mi/hr to 30 mi/hr in
10 sec, if the backward force of friction amounts to 8 lb per ton of weight?
11. What steady force must a car exert on a 4000-lb trailer in order to give it a
speed of 20 mi/hr in moving 200 ft along a level highway, starting from rest? Assume
that friction exerts a constant backward force on the trailer amounting to 20 lb per
ton of weight.
12. The diagram shows a 20-lb block on a plane inclined at 30° with the horizontal;
a cord attached to the block passes to a pulley at the top of the plane and thence
downward, and a steady pull of 15 lb is applied downward to the cord. If friction is
neglected, what is the acceleration of the block?

PROBLEM 12. PROBLEM 13.

13. The diagram shows a 20-lb block on a plane inclined at 30° with the horizontal;
a cord attached to the block passes to a pulley at the top of the plane and thence down-
ward to a 15-lb weight. If friction is neglected, what is the acceleration of the block?
14. What is the coefficient of sliding friction between the sled and the snow in
Prob. 13 of Chap. 1?
15. What steady horizontal force must be applied to a 100-kg block in order to slide
it 20 m along a level surface in 10 sec, starting from rest, if the coefficient of friction
between the block and the supporting surface is 0.25?
16. A 100-lb block is resting on a level surface and a 30-lb pull is applied to it in a
direction 25° upward from the horizontal. V{hat is the smallest value for the co-
 PROBLEMS 81

efficient of friction between the block and the supporting surface in order that the
block shall just start to slide?
17. How long will it take a block to slide from rest a distance of 200 cm down a
plane inclined at 30° with the horizontal, if the coefficient of friction between the block
and plane is 0.2?
18. A uniform plank 12 ft long and weighing 110 lb leans against a smooth vertical
wall. Its lower end stands on a level concrete surface, the coefficient of sliding friction
between the plank and the concrete being 0.3. At what angle with the horizontal
will the plank just start to slip?
19. A 200-gm wood block is sliding along a smooth level surface at 800 cm/sec.
A 10-gm bullet is fired into the block at right angles to its motion, and remains em-
bedded within it. If the impact changes the direction of motion of the block by 30°,
what was the speed of the bullet just before it struck the block?
20. A "hot rod" traveling due east collided at an intersection with a police car
traveling due north. The cars interlocked, and an examination showed that the
tangled wreckage slid along a line 25° north of east. The police car weighed 3000 lb
and was going at 25 mi/hr. The other car weighed 1500 lb; how fast was it going
just before the collision?
21. A boy is sitting on a horizontal turntable at a place 4 ft from the center. The
coefficient of friction between the boy and the turntable is 0.1. If the turntable is set
into rotation and its speed is increased, at \Vhat speed in revolu-
tions per minute will the boy begin to slide? ,,,--~---,I
22. A shaft extends upward through a horizontal frictionless
' ...... __ __ /
I

surface, as in the diagram, and a block that rests on the surface

is fastened to the shaft by a cord, so that when the shaft is rotated
the block describes a circular path. The block weighs 3 lb. If
the cord is 10 in. long, what is the tension in it when the shaft is
driven at 120 rev/min? PROBLEM 22.
23. A stone is being whirled around in a vertical circle at the
end of a cord 2 ft long. What is the lowest speed at which the stone can move through
the top point of its circular path?
24. On a highway being designed for traffic moving at 60 mi/hr it is desired that
the banking angle at curves should not exceed 15°. In order to satisfy this require-
ment, what is the smallest radius that a curve may have?
 5
TORQUE AND ROTATION
It has been pointed out that the motion of a body may consist of transla-
tion, of rotation, or of a combination of the two. The preceding chapter
deals with translation, and shows how motion of this type is affected by the
action of a force. The present chapter deals in a similar way ,vith rotation
and shows how rotary or angular motion is influenced by the action of a
torque.
5-1. Effect of Torque on Rotation. As already shown, the linear accelera-
tion imparted to a hody is directly proportional to the unbalanced force act-
ing upon it and inversely proportional to the mass of the body. This state-
ment, which is Newton's Second Law of l\Iotion, is expressed mathematically
by F = ma, or a = F /m, where a is the acceleration
of the body, F is the unbalanced force upon it, and
mis its mass.
The body is first regarded as a single particle of
F mass m attached to the end of a crank, Fig. 1, and
o the force which acts upon it is taken as a steady force
F so directed that it is always at right angles to the
Fm. 5-1. Effect of · 1e moves, t h e crank mak es
crank . Then as t h e partic
torque in producing an-
gular acceleration. it follow a circular path, and the force gives it an
acceleration a = F /m along the circumference of the
circle. The crank, of length r, is imagined to have negligible mass and to
operate without friction. The torque of the force F about the axis is given
by Eq. 1.1 as 3 = Fr, or
::5 = mar.

Should the moving system consist of two particles of masses m1 and m2, rig-
idly held together at distances r1 and r2 from the axis, then the total torque
needed to accelerate them both is
3 = m1 a1 r1 + m2 a2 r2.
These particles will in general have different linear accelerations a1 and a2,
but since they rotate as a unit their radii will have a common angular accelera-
tion a such that a1 = a r 1 and a 2 = a r2. The total torque acting upon the
rotating system can therefore be expressed as 3 = m1 r1 2a + m2 r2 2a, or
82
§ 5-2 BASIC EQUATION FOR ROTATION 83

J = (m1 ri2 + m2 rl)Cl'..

The torque on an extended body can be found similarly by considering the
body to be made up of a large number of particles; thus, the torque needed
to give a body of n particles an angular acceleration Cl'. is
J = (m1 r12 + m2 r22 + m3 rl + · · · + mn rn )a. 2

If the masses are regarded as infinitesimal and each is represented by the

symbol dm, the entire parenthesis can be replaced by the summation f r 2 dm.
This quantity is usually represented by the single letter I; thus,

I= fr 2 dm, (5.1)

·whence the expression for torque can be rewritten in the simple form

j = !Cl'.. (5.2)

This equation applies to a rigid body rotating about a fixed axis and to
objects, such as a rolling wheel, in which the axis passes through the center
of gravity and keeps the same direction in space. For such purposes it is
not necessary to consider the vector properties of the angular quantities
involved; it will suffice to compute their magnitudes and, if desired, to specify
their directions as clockwise or counterclockwise.
The quantity I in this equation is called the moment of inertia of the rotat-
ing system; it depends not only upon the total mass but also upon the way
the mass is distributed with reference to the axis of rotation.
6-2. Basic Equation for Rotation. The expression J = I a is a funda-
mental equation for rotation. It is identical in form with that expressing
Newton's Second Law for linear acceleration, quoted at the beginning of this
chapter. This likeness is more than one of form-indeed, equation J = I a
is an exact counterpart of equation F = m a, except that it applies to angular
instead of linear motion. An unbalanced torque is necessary to give a body
angular acceleration, just as an unbalanced force is necessary to give it
linear acceleration. The factor that determines how much angular accelera-
tion a g;iYen unbalanced torque will produce is the moment of inertia of the
body acted upon, just as the factor that determines how much linear accel-
eration a given unbalanced force will produce is the mass of the body acted
upon. In symbols, J, I, and a play the same parts, respectively, in angular
motion that F, m, and a play in linear motion.
The moment of inertia of any body whatever, fr 2 dm, involves the same
two quantities-a mass and the square of a length. It follows from Table II
of Chap. 4 that the units for moment of inertia are as follows:

British gravitational system slug· ft 2

Metric {mks system kg·m 2
cgs system gm·cm2
84 MECHANICS Chap. 5

The same table (Table II of Chap. 4) shows how to find the correct unit for
torque in each system. Since torque is the product of a force and a length,
its unit in any system is the force unit multiplied by the length unit of that
system. In British gravitational units, for example, torque is expressed in
pound· feet. Angular acceleration is stated in radians per second per second
in any system of units. From this analysis, the following groups of units are
found to apply to the equation 3 = I a:
3 I a
lb·ft slug·ft2 x radians/ sec 2
newton·m kg·m2 x radians/ sec2
dyne·cm gm·cm2 x radians/ sec 2
Two problems will illustrate the concept of moment of inertia and the use
of the basic equation for angular motion.
I. A motor armature is brought from rest to its normal velocity of 1200 rev /min in
4 sec. The driving torque exerted upon the armature is 53 lb·ft, and the frictional
drag upon it is 3 lb·ft. Determine the moment of inertia of the armature.
The normal velocity of the motor is 1200 rev/min = 20 rev/sec = 40 7r radians/sec,
.
and smce . too k 4 sec to acqmre
it . t h"1s veloe1ty,
. t h e angu lar acce lerat10n
· 1s· 40 1(' = 31.4
4
radians/sec 2 • The unbalanced torque upon the armature is 53 - 3 = 50 lb· ft. Hence,
by Eq. 5.2, the moment of inertia of the armature is ---------

1 slug·ft
50 lb·ft ~ 2
I = d" X 11i'1., = 1.59 slug·ft2 •
31. 4 ra ians Ll"

Sec'.2

II. A water turbine which has a moment of inertia of 250 kg· m 2 is brought to rest
by friction from a velocity of 180 rev/min in 26 min after its gate-valve is closed.
What is the frictional torque?
The angular acceleration of the rotating body is found in radian measure from the
velocities given. The initial angular velocity is
rev 2 7r radians 1 min radians
6 7f' -
Wo = 180 - . X X -- = --
mm 1 rev 60 sec sec

and the final angular velocity, w1 , is zero; therefore, the angular acceleration, from
Eq. 3.18, is
(O _ 6 7r) radians
a = w1 - w0 = _____s_ec_ = -0.01 2 l radians.
t 26 X 60 sec sec2
In this problem the unbalanced torque is that due to friction, which is the only torque
acting; consequently, the frictional torque is

3 = Ia= 250 kg·m2 X ( -0.0121 radians)

2 =
-
-3.03 newton·m.
sec
§ 5-4 CALC[/LATION OF M01vIENT OF INERTIA 85

5-3. Moment of Inertia of a Rigid Body. In the discussion in § 5-1,

moment of inertia is given as I = fr 2 dm. In order to apply this expression
to a particular body, the body is regarded as divided into a large number of
tiny pieces, so small that they are virtually particles of infinitesimal mass.
Then, the moment of inertia of the body is defined as the mass of one particle
multiplied by the square of its distance from the axis of rotation, plus the mass of
another particle multiplied by the square of i"ts distance from the axis, plus . . .
and so on, until all the particles of the body are included.
For some bodies, the moment of inertia can be found by inspection. Thus,
for a particle of mass m located at a distance r from the axis of rotation,
I = m r 2 , since there is but one particle and no need for summation as called
for by the expression fr 2 dm. For a thin ring of radius r, the mass is dis-
tributed all around the axis of rotation, but the various particles are at the
same distance from the axis, making r the same for all; hence, the moment of
inertia of the entire ring is I = r2 fdm = m r2 , where m is the total mass.
These values follow directly from the fore-
going definition, but in general the computa-
tion of moment of inertia requires integration.
It will be clear from the definition that the
moment of inertia of a body depends not only
upon the mass of the body but also upon the
distribution of the many tiny masses of which
it is composed. For example, the two wheels
FIG 5 2 Wheels of the same
in Fig. 2 are disks of the same radius, com- size. an~ ·mass but having differ-
posed in part of metal and in part of wood, ent moments of inertia.
and so designed that each has the same total
mass. The wheel with the metal rim has a larger moment of inertia about
the central axis than the other, because most of its mass is located farther
from the axis.
Moment of inertia is a scalar quantity, and therefore if a body is composed
of several parts its total moment of inertia is simply the arithmetical sum of
the moments of inertia of its parts, all with reference to a single axis. The
moment of inertia of a body is independent of the speed of the body and of
the forces acting; it depends only upon the mass of the body and the dis-
tribution of that mass, and upon the location of the axis.
5-4. Calculation of Moment of Inertia. For a body of simple geometrical
shape, the moment of inertia is easily determined by integration, as shown
in the following paragraphs. Some useful values are given in Table I, for
reference.
Straight Rod. To determine the moment of inertia of a slim rod, the rod
will be regarded as composed of an infinite number of pieces, the moment of
inertia of one infinitesimal piece will be expressed, and the expression will be
integrated to include the entire rod. Part I of Fig. 3 shows a slim rod of
length L vvith a transverse axis through one end, about 1Yhich the moment
8fi MECHANICS Chap. !i

TABLic I. lvimrnN'rs OF IN1mTIA

Form of Body Location of Axis I

- ---
Slim rod of mai-s m and length L At right angles to rod and ! mL 2
through one end
-------~-~-----
Same At right. angles to rod and 1~ mU
through center

Hollow cylinder (thin wall) of I

mass m, radius R, and any axial Axis of symmetry mR2
length
---------
Solid cylinder or disk of mass m, Axis of symmetry ! mlC1-
radius R, and any axial length
---- I
Thin disk of mass m and radius R Along any diameter i mR2
--·

Rectangular bar of mass 1n, At right angles to Land B and m (L2 + fl2)
length L, and width B through center of bar 12
-----·-----------
Solid sphere of mass m and radius Through center I g- rnR 2
R
I
of inertia is to be found. The stick is di,·ided into infinitesimal pieces; one
of these is shown at a distance x from the axis and its length is designated as
dx. If m represents the mass of the rod and A (lambda) the mass per unit

L
II I
r-~ L
2 - - .•

A~x ~I~ ~Rod of Axis x~~

\dx dx
mass m
n
Fw. 5-3. Diagram used in computing the moment of inertia of a rod about a transverse
axis I, located at one end; and II, located at the center.

length (that is, A = m/L), then the mass of the tiny piece is dm = 'A. dx, and
its moment of inertia with respect to the axis is dl = (;\ dx) x 2 = 'A.x 2 dx. For
the entire rod, it is necessary to make a summation of such quantities to
include all the particles comprising the rod. These extend from x = 0 to
x = L, and therefore, by integration, the moment of inertia of the entire
rod is

I = A r:2 dx = [x3JL
A = £3. A
Jo 3 3
0

Since A = m/L, this expression reduces to

I= imV. (5.3)
§ 5-5 PARALLEL-AXIS THEOREM 87

If the axis were to pass through the center of the rod, as indicated in part
II of the figure, the procedure in finding the moment of inertia would be
exactly as before up to the point of imposing the limits of integration. These
limits would then be -L/2 and L/2, and therefore the integration for I
would become
2
I = }. f!L~x
L/z
= A[ ;a]L/-LI;= ~ [(L)3
2 -
( - 2L)3] = }. 1;.
L

Since X = m/L, it follows that the moment of inertia of the slim rod about a
transverse axis through its center is
I = 1\ mL2. (5.4)
Solid Cylinder. The moment of inertia of a solid cylinder about its axis of
symmetry is found by the same procedure.
A cylinder of radius R and length L is
shown in Fig. 4; its mass will be represented
by m and the density of the cylinder mate-
rial by p (rho), whence p = 1r;; L" The
cylinder will be regarded as composed of
a series of concentric tubes having infini- Axis
tesimally thin walls. One such tube of /"'
radius r and thickness dr is shown; its
1r
volume is dV = 2 r L dr, its mass is Frn. 5-4. Diagram used in comput-
ing the moment of inertia of a cylinder
1r
dm = p(2 r L dr), and its moment of about its axis of symmetry.
inertia with respect to the central axis is
dl = p(2 1r r L dr) r 2 or 2 1rp L r 3 dr. For the entire cylinder,

I = 2 1rp L l 0
R
r 3 dr = 2 1rp L -
4 0
[r4JR = 2 1rp L (R4 -
4 2
~-~
0) 1rp LR 4

To obtain the result in terms of the mass m of the cylinder, pis replaced by
its equivalent, 1r;{!; L' ,vhence the moment of inertia of the cylinder about its
axis of symmetry is
I = t mR 2• (5.5)
6-6. Parallel-axis Theorem. Sometimes the moment of inertia of a body
about a particular axis is known, and it is desired to determine its moment of
inertia about another axis parallel to the first. This can be done in a simple
manner when one of the axes passes through the center of gravity of the
body. Analysis shows that, if the moment of inertia of a body of mass m
about an axis through its center of gravity is I G, the moment of inertia about
a parallel axis distant h from the first one is
I= JG+ mh2. (5.6)
Stated in words: The moment of inertia of a body about any axis is equal to its
88 MECHANICS Chap. 5

moment of 1:nertia about a parallel axis through the center of gravity plus the
product of the mass of the entire body and the square of the distance between the
axes.
For example, when a solid cylinder rolls on a flat surface it may be of inter-
est to know the moment of inertia of the cylinder about its line of contact
with the surface. The moment of inertia about its axis of symmetry is
I a = ! mR2 , and when the axis is shifted parallel to itself a distance r to the
rim, the moment of inertia of the cylinder becomes ! mR2 mR 2 = ! mR2 •+
Again, the moment of inertia of a slim rod about a transverse axis through
its center is given by Eq. 5.4 as I a = 1\ mV, and about a parallel axis through
the end of the rod is given by Eq. 5.3 as I = l mV. This result is verified
by the Parallel-Axis Theorem, from which
1
+ m 2~ 1
2
I = Ia + mh 2 = 12 mV = 3 mV.
5-6. Radius of Gyration. The moment of inertia of an object, as previously
stated, is determined by the masses of the particles that make up the object
and the squares of the individual radii that extend to these particles from the
axis. Moment of inertia is often expressed quite differently, however, in
terms of the aggregate mass of the body and a single fictitious radius called
the radius of gyration. A rotating object of any shape can be imagined to
have its entire mass compressed into a thin ring of such size that its moment
of inertia remains unchanged. Then, if K is the radius of the ring, the radius
of gyration of the body itself will also be K, with reference to the axis of
symmetry of the ring. It follows that the moment of inertia of any body of
mass m and radius of gyration K is
I= mK2 • (5.7)
The radius of gyration of a body with respect to any axis is the radius of a thin
ring centered upon the axis mentioned and which has the same mass as the body
and the same moment of inertia about this axis.
The properties of moment of inertia and radius of gyration will be illus-
trated by referring again to a slim rod.
For a meter stick of 120-gm mass, the moment of inertia about a transverse axis
through one end is I= t mL2 = t X 120 gm X 1002 cm2 = 4 X 105 gm·cm2 • The
radius of gyration of the stick with reference to this axis follows from Eq. 5.7 as
4 X 105 gm·cm2
K = = = 57.7 cm. It may be noted that the mass of the
120 gm
stick is not needed in this calculation; since t mL2 = mK2 , the radius of gyration of
. . h . K L 100 cm
t h e stick about a transverse axis through t e end 1s = V3 = 1.732 = 57 .7 cm,
as before.
Manufacturers of turbines, generators, and other machines give the values
of the radius of gyration of the rotating parts so that users can readily com-
pute the moments of inertia of these odd-shaped elements.
 § 5-7 TlTICAL PROBLEMS ON ROTATION 89

5-7. Typical Problems on Rotation. The following problems illustrate the

effects of torque and moment of inertia on rotary motion.
' I. A steel wheel of a gyro-stabilizBr is a solid disk oI diameter 12.5 ft and mass
1
1175 slugs. What torque must a motor exert on the wheel to bring it from rest to its
operating speed of 600 rev /min in 1.5 hr, if friction is neglected?
To solve the problem, compute the moment of inertia of the wheel from its mass
and dimensions, and also the angular acceleration during the starting period. The
torque required of the motor can be calculated from these data by using Eq. 5.2.
The moment of inertia of the disk about its axis is

I= 2 2 T
1 rnR 2 = 1 X 1175 slugs X (125) 2 ft 2 = 2.29 X 104 slug·ft 2 •

The angular velocity of the ·wheel at full speed is ~~O X 2 1r = 20 1r radians/sec, and
20 1r
its angular acceleration during the l.5-hr starting period 1s 3600 X 1. 5 = 0.0116
radian/sec 2 • Therefore,
radian
:J = 2.29 X 104 slug·ft2 X 0.0116 - -2- = 266 lb·ft,
sec
which is the torque exerted by the motor to accelerate the wheel.
II. The wheel and axle combination represented in Fig. 5 has a total mass of
40 kg, and the axle has a diameter of 10 cm. The apparatus is set into rotation by re-
leasing a 2-kg block that is fastened to a cord wrapped
around the axle. The block is observed to fall a dis-
tance of 3 m, starting from rest, in 7.6 sec. Neglect
friction and find the radius of gyration of the moving
system.
In solving this problem, first determine the linear ac-
celeration of the descending block and then the tension
2kg
in the cord. From these data compute the applied torque
and the angular acceleration of the wheel and axle. Fi-
nally, calculate the moment of inertia and the radius of
13-,
gyration.
The block starts from rest and moves a distance of l_J_i
J~ r
3 m with constant acceleration in 7 .G sec, and therefore
FIG. 5-5. A method for
its linear acceleration is
measuring the radius of
2s 2X3m m gyration of a rotating sys-
a = - = = 0.104 - .2 tem.
t2 (7 .6) 2 sec 2 sec
The tension in the cord is found by considering the forces acting upon the 2-kg mass;
these are the force of gravity of 2 X 9.8 = 19.6 newtons acting downward, and the
unknown tension P in the cord pulling upward. The unbalanced force acting on this
mass is 19.6 newtons - P, and produces the observed acceleration of 0.104 m/sec 2 •
From Eq. 4.1,
m
19.6 newtons - P = 2 kg X 0.104 - 2 = 0.208 newton,
sec
from which tho tension is found to be P = 19.39 newtons, and thus the torque exerted
on the moving system is 19.39 X 0.05 = 0.970 newton·m. This is the unbalanced
torque, since friction is neglected. The angular acceleration is found from the linear
acceleration and the radius of the axle; thus,
!)0 MECHANICS Chap. 5

ill
0.104- .
a=~= ___s_e_c2 = 2 .08 radians.
r 0.05 rn sec 2
Therefore the moment of inertia of the wheel and axle, from Eq. 5.2, must be

1 = 0.970 new~on·m = 0.4 66 kg·m 2,

2 .08 radians
sec2
nnd the radius of gyration is

I( = .Jf = ~0.4:~ ~~- rn 2

= 0.108 m.

III. A solid cylinder rolls down a plane which makes an angle of 25° with the hori-
zontal. If the cylinder starts from rest and does not slip, in what time will it travel
200 cm?
From a study of the forces acting on the cylinder, apply Eq. 4.1, F = ma, to its
motion of translation, and Eq. 5.2, J = I a, to its motion of rotation. Solve these
equations simultaneously to get the linear acceleration, and find the time by using
the equations for uniformly accelerated motion. A cylinder of mass m and radius R
is represented in part I of Fig. 6, and the forces acting upon it are shown in part II.

R
Cylinder
of massm

mg sin 25°

25°

mg cos 25°

J]

FIG. 5-G. Analysis of forces on a cylindPr rolling down an inclined plane.

These consist of the weight mg acting downward, the normal reaction N of the plane,
and the backward force of friction fat the contacting surface. The weight is replaced
by its two components: mg sin 25° down the plane and mg cos 25° normal to the
plane, the latter being counteracted by the reaction N. The unbalanced force
F = (mg sin 25° - f) acts down the plane, and from Eq. 4.1 the linear acceleration a
is such that
mg sin 25° - f = ma.

The torque that causes the cylinder to rotate about its center is f X R, and hence
the angular acceleration a is such that
mR 2 a m Ra
f X R =la= - X - = --,
2 R 2
 ANOULAR MOMBNTU11,l 91

from which f can be expressed as ma/2. This value, substituted in the first expression
above, gives mg sin 25° - ma/2 = ma, whence
2 . 2 cm cm
a = - g sm 25 0 = - X 980 - X 0.423 = 276 - .
3 3 sec2 sec 2
1Yith this value for the constant acceleration, and zero initial velocity, Eq. 3.11 gives

t2 = 2 s = 2 X 200 cm = l .4S sec 2•

a 276 cm
sec 2
Hence the time taken by the cylinder to roll down a distance of 200 cm is t = 1.20 sec.
5-8. Angular Momentum. The concept of angular momentum is of inter-
est in mechanics and is of fundamental importance in modern physics. This
concept illustrates further the close analogy between linear and angular mo-
tion that has been repeatedly pointed out.
In order to derive the angular momentum relation, the basic equation for
rotation, 3 = I a, is written with a expressed as dw/dt, the rate of change of
dw
angular velocity, in accordance with Eq. 3.17. Thus 3 = I di' or
3 dt = I dw. (5.8)
This expression is integrated, as follows:

t dt = J,/
J/ ('' dw,
0

and gives as a result, if the torque is constant,

3t = Iw1 - Iwo, (5.9)
in which the product 3 t (or f 3 dt, if the torque is not constant) is called
angular imptllse, and the product Iw is called angular momentum. This
equation states that an unbalanced angular impulse causes a change in the
angular momentum of a body and, moreover, that these two qttantities are
numerically equal. The analogy between this equation and Eq. 4.6 for
linear motion is evident by inspection.
It is instructive to see what happens to a rotating body when the angular
impulse 3 tor f3 dt is zero; clearly, from Eq. 5.9, Iw, will equal Iwo, or
Iw = constant. (5.10)
Hence, a rotating body, when not subjected to any angular impulse that tends
either to accelerate or retard it, maintains a constant angular momentum.
This statement illustrates a general principle known as the Law of Conserva-
tion of Angular Momentum, which states that when a body or system of bodies
is not acted upon by an unbalanced external torque, the angular momentum of
the system about any fixed axis remains -constant. For example, a wheel or
other rigid body turning without friction continues to rotate with undimin-
92 MECIIANICS Chap. !i

ished speed. It is interesting to observe, moreover, that if the moment of

inertia of a rotating body can be altered, its angular velocity will change
oppositely, so as to keep the product Iw constant. This effect can be illus-
trated by standing a man on a low piano stool which has little friction in its
bearings, and spinning him around a few times. If he extends his arms, thus
increasing his moment of inertia, his speed will diminish, and if he draws them
inward, thereby reducing his moment of inertia, his speed of rotation will
mcrease.
As a numerical example, imagine a small metal block set into rotation in a horizontal
circle at the end of a cord which passes down through
a hole at the center of a glass table top, as represented
in Fig. 7; if the system is rotating at 1 rev /sec in a cir-
cle of 20-cm radius, and if the cord is then pulled down
to shorten the radius to 10 cm, what will the speed of
rotation become?
If friction is neglected, the torque on the rotating
system is zero, and hence the initial angular momen-
tum equals the final angular momentum, or I oWo = l 1 w1 •
Note that in this problem the moment of inertia
changes from Io = m r a2 to 11 = m ri when the radius
Fm. 5-7. A test to illus- of the circle changes from ro = 20 cm to r1 = 10 cm,
trate the conservation of the block being considered a particle of mass m.
angular momentum. Therefore, m r a2 Wa = m ri Wt, or
2
Wj = W 0 (r 0 /r1 ) 2 = 1 -rev (20
- cm)
- = 4 rev
-.
sec 10 cm sec
5-9. Motion of Precession. Up to this point the study of rotation has
been confined to a fixed direction of the axis, and for this reason it has
not been necessary to consider the vector properties of angular quantities.
Angular impulse and angular momentum, also torque, angular velocity, and
most other angular quantities (angular displacement is an exception) are
vectors and are subject to the rules for vector addition described in Chap. l.
Because of their angular nature, however, they require a special form of
vector representation. Such a quantity is represented vectorially by an
arrow having a length which indicates the numerical value of the quantity
to some suitable scale, the arrow being parallel to the axis of rotation or, for
torque, parallel to the axis about which the torque would by itself produce
rotation. The direction of this arrow is chosen to be that in which the rota-
tion would advance the usual right-handed screw. In Fig. 8, the angular
momentum of Iw of a rotating wheel is represented in this manner.
When a rotating body is acted upon by a torque that tends to change the

----~-~
Fm. 5-R. Vedor repr0sentation of angular quantities.
 § 5-9 MOTION OF PRECESSION 93

direction of its axis, this action gives rise to an angular motion about a third
axis, if the body is free to move in this way; such motion is called precession.
This type of motion will be explained by Eq. 5.9, rearranged to read
Iwo +3t = Iw1,
and applied vectorially to the wheel shown in Fig. 9. This wheel, mounted
· loosely between collars on a shaft A, is set spinning in the direction a by some
agency not illustrated. The shaft does not rotate with the wheel, and one
end of iii-is set horizontally upon a pivot on the vertical support C, the other
end of the shaft being free. If the wheel were not spinning, the free end of
the shaft would drop and the whole system would fall off the vertical sup-
port. But with the wheel revolving, the tendency for the free end of the
shaft to drop causes the shaft to rotate horizontally about the pivot-that
is, the shaft precesses in a horizontal plane.

'
' \
\
I
I
I

'c

b c
c
Frn. 5-9. Precession of a revolving wheel about one end of its axle.

For a wheel which has a moment of inertia I and which is spinning with
an angular velocity wo, both with reference to its axis of symmetry, the
angular momentum is Iwo; for the wheel in the figure the angular momentum
is represented by the vector drawn parallel to A in the perspective diagram
a.t the right. The pull of the earth upon the wheel, acting for a very short
time interval t, produces an angular impulse ·which tends to turn the entire
moving system about axis Bin the direction shown by arrow b. This angular
impulse is represented in the vector diagram as 3 t. If Iwo and 3 tare added
vectorially, which is possible because these quantities have the same unit,
the vector sum will be Iw1, from the foregoing equation. This resultant lies
in the same horizontal plane as its components and makes an angle () with
Iwo. Accordingly, the shaft will shift its position to point in the direction of
Iw1, turning about the vertical axis C, as shown by arrow c. This motion
of precession occurs, therefore, about an axis perpendicular to both of the
other axes, A and B. The precession continues, for the wheel when displaced
is subjected to the same amount of angular impulse as at first, and the fore-
going actions are repeated. As described, it would appear that the shaft
progresses in discrete steps; actually, the process is one of infinitesimally
small angular shifts that result in a uniform motion of precession.
!)4 ME'CJJANICS Chap. 5

The angular velocity of precession Q ( omega) is easily found, being the

ratio of the angle O to the time t. Since the time interval is very small, the
spin velocity of the wheel may he taken as constant at the value w. Also, 0
may be replaced by tan 0, which, from the vector diagram, is :J t/Iw. Conse-
quently,
:J (5.11)
Q = -,
Iw

where I is the moment of inertia of the wlieel, w its spin velocity, and :J the
torque which tends to change the direction of its axis.
5-10. The Gyroscope. A gyroscope is a wheel and axle supported in
gimbal rings so that it can be set in rotation with its axis in any desired direc-
tion. The revolving wheel maintains its axis in the initial direction, even
though the gyroscope as a whole is moved from one position to another,
because forces exerted on the support

·::~·
B toward are not transmitted through the gim-
bal rings to produce a torque on the
,,/· A toward east wheel.
/// The action of the gyroscope as a
Observer compass is an application of the addi-
tion of two angular velocities in differ-
ent directions. In Fig. 10, the earth
is viewed from a point in space be-
yond the south pole, from which point
it revolves in a clockwise direction.
West The gyroscope is shown in various
positions along the equator; position
Earth G3 is the ultimate one and the other
two are disturbed positions used in
explaining the action. The arrows
Fm. 5-10. A gyroscope used as a com-
on the wheel in positions G1 and G2
pass. The observer is supposed to be indicate the direction in which the
looking northward toward the earth. rim of the wheel nearest the reader is
moving. A weight w placed on the
inner gimbal ring always seeks the position nearest the center of the earth
because of the earth's attraction for it, no matter where the gyroscope is
situated.
If the gyroscope is placed at position G1, with its axis directed arbitrarily
from east to west, then the angular veloeity of the rotating wheel is repre-
sented by a vector directed along its axis toward the east; it is shown as A
in the vector diagram. As the earth revolves and carries the gyroscope to
position G2, its axis continues to poiht eastward, an action ·which brings the
weight w out of line with the earth's radius. This weight sets up a torque
which tends to produce clockwise rotation of the gyroscope about an axis
§ 5-10 THE GYROSCOPE 95

perpendicular to the page. The additional angular veloeity produced by

this torque is represented by a vector directed into the paper; it is shown in
perspective as Bin the vector diagram. The addition of A and B forms the
resultant R, and ll lies in the direction toward which the axis of the wheel
turnR. Consequently, the Rhaft-end 7J will move northward (into the paper)
and the Rhaft-end x will move southward (out of the paper); this precessional
motion occurs about the axis C. As the gyroscope is carried farther by the
earth, the precession continues until the shaft of the gyroscope becomes
parallel to the earth's axis 0, with the shaft-end x pointing toward the
reader, as at Gs. When the gyroscope reaches this position, the earth's pull
on the weight w causes no further torque, and the gimbal ring remains in a
plane with the earth's axis.
Thus, the wheel of a gyro-compass seeks a position such that its axis points
north and south, and thereafter maintains that position. An advantage of
the gyro-compass over the magnetic compass is that the gyroscope indicates
the true geographic north-south direction, whereas the magnetic compass
points along the magnetic meridian for the particular locality in which it is
placed and is, moreover, influenced by variations in the earth's magnetic
field and by masses of iron or other magnetic materials in its neighborhood.

PROBLEMS
1. A solid cylinder of 10-cm radius is mounted in stationary bearings and is set
into rotation by pulling steadily on a cord wound tightly around its rim. The applied
torque is 5 X 105 dyne· cm, of which one-fourth is used in opposing friction. If the
cylinder acquires a speed of 2 rev /sec in 20 sec, what is its moment of inertia?
2. In the preceding problem, when the applied torque ceases to act, with what
angular acceleration will friction bring the cylinder to rest?
3. A motor armature rotating at its rated speed of 1200 rev /min is disconnected
from the supply mains and friction brings it to rest in 3! min. If the moment of
inertia of the armature is 1t kg· m2, what is the value of the retarding torque due to
friction?
4. How much driving torque must be applied to the armature described in Prob. 3
in order to bring it from rest to its rated speed in 25 sec? Assume that the retarding
torque due to friction is constant.
5. What unbalanced torque must be exerted upon a flywheel
which has a moment of inertia of 45 slug· ft2 in order to bring it from "'-B~
rest to a speed of 90 rev /min in 30 sec? I

6. A slender rod of negligible mass carries two concentrated

masses of 0.5 kg each, located 0.6 m apart. Compute the moment
of inertia of the combination about a transverse axis midway between
fL
the masses.
7. The diagram represents a thin book cover of mass m, breadth 1---------J_l
B, and of any length L, with an axis along one of its long edges. -Axis
(a) Show by calculus that the moment of inertia of the cover about
this axis is m B 2/3. Hint: use a symbol to represent mass per unit PROBLEJI 7.
area, thus: {3 = m/BL. (b) A book cover has a mass of 40 gm and
a breadth of 15 cm; find its moment of inertia about the axis shown.
8. Compute the mass of the cylinder described in Prob. 1.
9fi .MRCIIANIC8 Chap. 5

9. For a thick-walled hollow cylinder of mass m, inner radius R 1 , and outer radius
R2, show that the moment of inertia about the geometrical axis is given by the expres-
. rn
s10n I = 2 (R1 + R2
2 2 ).

10. Compute the moment of inertia of the bowling ball described m Prob. 7 of
Chap. 4 about an axis through its center.
11. A yardstick has a mass of 0.01 slug. Compute the moment of inertia of this
slim stick about an axis perpendicular to the stick at the 12-in. mark.
12. The moment of inertia of a slim rod about a transverse axis halfway between
the center and one end is 3.5 kg·m 2 • Find the moment of inertia of the rod about a
transverse axis at one end.
13. Four meter sticks, each of 300-gm mass, are fastened together at the ends to
form a square. Compute the moment of inertia of the combination about an axis
perpendicular to the plane of the square and passing through one corner.
14. A solid cylinder has a radius of 3 cm and a mass of 800 gm. Compute the mo-
ment of inertia of the cylinder (a) about its geometrical axis, and (b) about a parallel
axis tangent to its rim.
15. Determine the radius of gyration of a yardstick that is arranged to rotate about
a transverse axis through one end of the stick.
16. If the flywheel described in Prob. 5 has a mass of 20 slugs, what is its radius of
gyration?
17. A wheel that has a radius of 0.3 m and a mass of 10 kg is mounted in stationary
bearings. A cord is wrapped tightly around the rim of the wheel and is pulled with
a steady force of 15 newtons. This action causes the wheel to make 3 rev in 2.2 sec,
starting from rest. Assume friction to be negligible and compute the radius of gyra-
tion of the wheel.
18. The grindstone in the diagram consists of a solid cylinder 2 ft in diameter and
weighing 100 lb; it is turned by a handle fastened to a radial crank at a point 9 in.
from the axis. If a steady force of 2 lb is applied to the handle, always at right angles
to the crank, what speed in revolutions per minute will the grindstone acquire in
10 sec, if friction is negligible?

PROBLEM 18. PROBLEM 21.

19. An arrangement like the Atwood machine, Fig. 4-5, is used to determine the
moment of inertia of a pulley. The pulley, of 6-cm radius, is mounted in stationary
bearings having negligible friction and is set into motion by hanging over it a cord to
which unequal masses are applied: 500 gm at one end and 510 gm at the other. A test
shows that the masses, when released, move 36.0 cm in 3.0 sec. -What is the moment
of inertia of the pulley?
 PROBLEMS 97

20. A wheel that has a mass of 2t slugs and a diameter of 16 in. is turning at 90
rev /min in stationary bearings and is brought to rest by pressing a brake shoe radially
against its rim with a force of 10 lb. If the radius of gyration of the wheel is 6 in.
and if the coefficient of friction between the Rhoe and the rim has the steady value of
0.20, how long will the wheel continue to rotate?
21. The diagram shows a solid cylinder that is free to turn about its geometric axis
in stationary bearings. The cylinder has a mass of 5 slugs and a radius of 8 in. It
is set into rotation by releasing an 8-lb block attached to a cord that is wound tightly
around the rim of the cylinder. Determine the speed of the block when it is 4 ft
below its starting point.
22. A person standing upright on a rotating stool is spun around at the rate of
24 rev /min with his arms at his sides; in this position his moment of inertia is 0.8
slug· ft 2 • If he extends his arms horizontally, thereby increasing his moment of inertia
by 2.0 slug· ft 2 , at what speed will he then rotate? Assume that friction is negligible.
23. A solid disk 20 cm in diameter and having a mass of 104 gm is mounted on an
axle as shown in Fig. 9, with the center of gravity of the disk 12 cm from the vertical
support. Consider the mass of the axle to be negligible, and determine the speed of
precession of the disk when it revolves at 1000 rev/min. Also, explain how to predict
the direction of precession.
 6
MECHANICAL ENERGY
Work and energy are embodied in every part of Physics and are among the
broadest concepts with which this science deals. The term "·work" is applied
in popular language to any form of labor, physical or mental, for producing
any kind of result. In science and engineering, on the other hand, "work"
has a definite technical meaning which involves two physi<'al quantities that
have been developed in previous chapters--namely, force and displacement.
Although made up of vector constituents, work itself is a scalar quantity,
and amounts of work can be summed up simply by arithmetical addition.

WORK AND ENERGY

6-1. Work Done in Translation. The concept of ,rnrk will be introduced
by some everyday actions, in which a man moves a piano from one place to
another while exerting a steady push on the instrument, or pulls an oar
toward himself in rowing, or lifts a weight from the floor to the top of a
table. In each of these instances, two things should be noted: first, that the
man exerts a force, and second, that he exerts it through a distance in the
direction of that force. Under these conditions, the man is said to do work,
and the amount of work depends upon the two factors: the force that he
exerts, and the distance along the direction of the force through which he
exerts it. The amount of work done is measured by the product of these two
quantities.
Inanimate objects, also, can exert forces, and any agent ·which exerts a
force through a distance does work; hence, a general statement can be made
as follows: Any agent does work on a body when it exerts a force on the body
through a distance along the direction of the force.
More precisely, work should be expressed in terms of displacement rather
than distance along a path, there being some instances ·where displacement
and length of path are not the same, § 3-1. In motion along a straight line
without reversal of direction, the displacement of the point of application of
the force is the same as the distance through which the force is exerted; unless
otherwise stated this type of motion will be assumed, and the two terms will
be used interchangeably.
In general, when an agent exerts a steady force F upon a body, and the
point of application of the force undergoes a displacement .s along the direc-
98
§ f>-1 WORK DONR IN 'l'RANSLA1'10N

tion of the force, the amount of work done by the agent is

E = Fs. (6.1)
The symbol Eis used for work (rather than the initial letter of work) in order
that W may continue to represent weight, as previously, without confusion.
It should be emphasized that F is the force applied by the agent doing the
work, and is not, in general, the unbalanced force acting on the body.
The product of force and distance was encountered in § 1-12, but these
quantities were at right angles to each other to produce torque, instead of
being in the same direction, as at present, to produce work.
Since ,vork is essentially the product of a force and a length, the unit for
work is the product of a force unit and a length unit. In British gravitational
units, the unit of work is the foot·pound. One foot·pound is the amount
of work done on a body when a force of 1 lb is exerted upon it through a distance
of 1 ft in the direction of the force. In the mks system, the unit of work is
called the joule, after the British experimenter James P. Joule (1818-1889).
One joule of work is done on a body when a force of 1 newton acts upon it
through a distance of 1 min the direction of the force. In the cgs system,
the unit of work is called the erg. One erg of work is done on a body when
a force of 1 dyne acts upon it through a distance of 1 cm in the direction of
the fdrce. Since the meter is 102 cm and the newton is 105 dynes, it follows
that 1 newton·m = 107 dyne·cm; in other words, 1 joule= 107 ergs. To
summanze:
1 ft· lb = (1 ft) (1 lb);
1 joule = (1 m) (1 newton);
1 erg = (1 cm) (1 dyne).
Force which Varies with Displacement. Sometimes it is desired to calculate
the work done by a force which is not constant, but which changes when the
displacement changes. In such situations the work done for an infinitesimal
displacement ds is given by
dE = F ds. (6.2)

By integration, the total work is

E=fFds, (6.3)

and the manner in which the force varies with the displacement must be
known in order to make the computation.
The force required to stretch a spring, for example, increases as the spring
is stretched-in fact, the force is directly proportional to the amount of
stretch. Thus, if a spring is stretched by an amount s, the force needed to
hold it in the stretched position is F = Ks, where K is a constant of propor-
tionality known as the spring constant. Then, in order to increase the
amount of stretch from s1 to 82 the work needed is
100 1vIECHANIC8 Chap.fi

E = f F ds = f Ks ds = K J'.; ds = K [ ~I: = ~ (s22 - si2).

Suppose that an unstretched spring is 5 cm long and that a force of 1000 dynes is
needed to hold it when stretched 1 cm. How much work is required to increase its
total length from 5 to 7 cm?
.
S mce t h e sprmg . K = F
. constant Is - =
1000 dynes an d t h e amount of stretc h IS
. to
s 1 cm
be increased from O to 2 cm, the work needed is

1 dvnes
E =? X 1000 - " - (4 cm2 - 0) = 2000 ergs .
.., cm

If the total length of this spring were to be increased from 7 to 8 cm, thereby increasing
the stretch from 2 to 3 cm, the work required would be

1 dynes
E = - X 1000 - - (9 cm2 - 4 cm2 ) = 2500 ergs.
2 cm

Force and Displacement Along Different Directions. It frequently happens,

when work is being done, that the force exerted on the body does not have
the same direction as the distance through which this force acts; that is, F is
not in the same direction as s. Then, in applying the foregoing work equa-
tions the term F should be replaced by that component of the force which is
along the direction of s. Thus, the amount of work E done upon a body by
an agent that exerts a force Fon it through a distances, where the directions
of F ands make an angle e with each other, is given by

E = Fs cos e. (6.4)

For example, a boy may push a sled a distance s along level snow with a
force F which makes an angle () with the horizontal, as shown in Fig. 1.
Here F can be resolved into two components: F cos () along the direction of
s, and F sin () at right angles to this direction. Only the first of these is
effective in doing work on the sled; the other merely increases the pressure
of the sled against the snow and in doing so increases the backward drag of
friction. The work which the boy does on the sled is, therefore, E = Fs cos e.

F cos O

fslnOtgf (Sled
7
Frn. 6-1. Work being done in movin,:!; a Frn. 6-2. Work being done on a whPel to
sled horiwntally. produce rotation.
§ 6-3 ENERGY 101

As an illustrative problem, calculate the work done by a tractor in-plowing a-furrow

200 yd long, if it exerts a steady force of 150 lb on the plow at an angle of 15° with the
ground.
From Eq. 6.4, the work done is E = Fs cos (} = 150 lb X 200 yd X cos 15°
= 150 lb X 600 ft X 0.966 = 86, 940 ft· lb.
6-2. Work Done in Rotation. The amount of work done in rotating a
body can be calculated from the foregoing equations, but usually can be
found more directly in terms of the angular quantities, torque and angle. If
a drum of radius r has a cord fastened to the rim, as represented in Fig. 2,
and if the cord is pulled with a force F for a distance s along its length, an
amount of work E = Fs is done upon the drum. The torque ;} resulting
from the force F is ;} = Fr, whence F = J/r. The angle through which the
drum turns is found by observing that, as the cord is pulled, every point on
the rim traverses an arc of lengths, and therefore the drum rotates through
an angle e such thats = re. When these values are substituted for F ands,
the expression for work becomes E = Fs = 1r r8, or

E = ;1 e. (6.5)
This is a general expression for the work E done upon a rotating body by an
agent which exerts a torque ;} upon the body through an angle of e radians.
It will be clear from a consideration of Eq. 6.5 that if the torque ;} is not
constant, but varies when the angular displacement 8 changes, the work
done is given by the expression :

E = fJ d8. (6.6)
Sometimes bodies move in opposition to the efforts exerted upon them; for
instance, a rotating wheel may continue to turn in a certain direction even
after a brake has been applied which exerts a torque upon it in the opposite
direction, or a heavy weight may slide down an incline in spite of attempts to
pull it upward. In such cases the body does work instead of having work done
upon it; the amount of work done can be found by using Eqs. 6.1 to 6.6.
6-3. Energy. An agent is said to possess energy if it is able to do work.
For example, a horse possesses energy because it can draw a load; steam pos-
sesses energy since it can push the piston within the cylinder of a steam
engine; the mainspring of a watch possesses energy when wound, for it is
able to drive the hands of the timepiece. Moreover, when an agent does
work, its energy is reduced by an amount exactly equal to the work done.
There are many different forms of energy-thus, coal has chemical energy,
a hot substance has internal energy, a stretched spring has mechanical energy,
a charged capacitor has electrical energy, and so on. In the study of Me-
chanics, mechanical energy is naturally the type which is of special interest.
A body or system of bodies may possess mechanical energy from either or
both of two causes. First, a body in motion is able to do work in coming to
102 MECHANICS Chap. 6

rest: a moving body always possesses energy because of its motion; this is
called kinetic energy. A moving hammer has kinetic energy, and this enables
it to do work in driving a nail. Second, a body which has been moved to a
new position is sometimes able to do work because of this fact. For example,
a raised weight is commonly said to possess energy because of its position;
this is called potential energy. Strictly, it is not correct to regard potential
energy as residing in a particular object, but rather as residing in a system
which includes that object. Thus, while a weight is raised it is being sepa-
rated from the earth against their mutual attraction; in the process, work is
done and the potential energy of the system, made up of the weight and the
earth, is increased. Nevertheless, the potential energy is customarily ascribed
to the weight, since this is the tangible part of the system which makes the
energy available. A body is not necessarily given potential energy by dis-
placing it, but only when it is able to do more work in its new position than
it was originally. A weight that has been raised or a spring that has been
wound is thereby given potential energy, but a weight that is merely moved
along a level floor from one position to another does not thereby gain potential
energy.
6-4. Conservation and Transformation of Energy. After a body has done
work its capability of doing further work is lessened, and this means that it
possesses less energy than before. This reduction must not be regarded as
a loss of energy, for in doing work the body has contributed an equal amount
of energy to some other body, which, together with the first, constitutes a
system. Thus, when a man does 20 ft· lb of work in throwing a ball he gives
up this amount of energy, and the ball gains 20 ft· lb of energy, since that
amount of work was done upon it. The energy that is given up by a body is
imparted to others without loss, and thus within the system comprising these
bodies, the total amount of energy remains unchanged. This illustrates a
general law, known as the Law of Conservation of Energy, which states that
energy can neither be created nor destroyed. Expressed differently, the total
amount of energy in the universe remains constant.
This principle is illustrated by a body which has been raised to the top of
an inclined plane and is then allowed to slide down. A certain amount of
work was done on the body in raising it, thereby increasing its potential
energy. As the body slides down the incline, it gives up potential energy
and gains kinetic energy, and if no work is done against friction, the increase
of kinetic energy equals the reduction of potential energy. If, however, part
of the potential energy possessed by the body is expended in doing work
against friction on the way down, then the amount that its kinetic energy
increases will be lessened by exactly this amount. This example shows that,
although energy can be transformed from one kind to another, it is not
destroyed in the process. When energy is expended in work against friction,
it is sometimes wasted-that is, rendered unavailable for useful purposes-
but it is not destroyed, for it is converted to heat, and heat is recognized_as
§ G-5 KINETIC ENERGY 103

a form of energy. When a train is driven at constant speed along a level

track there is no increase in either its kinetic energy or its potential energy,
and hence all of the energy in the fuel burned is used in doing work against
friction. With the eventual exhaustion of natural fuels, the provision of
energy in sufficient quantities to meet the future demands of mankind will
be an engineering problem of utmost importance.
6-5. Kinetic Energy. Kinetic energy in the kind of energy that a body has
because of its motion. An expression for kinetic energy can be derived from
the discussion of the preceding sections, crystallized into the following state-
ment. When mechanical work is done on a body, the amount of work done
is entirely accounted for by one or more of the following effects: (1) increase
in the kinetic energy of the body, (2) increase in its potential energy, or
(3) production of heat due to friction. In particular, the equation will be
derived for motion of translation by calculating the amount of work done in
sliding a body along a surface under such conditions that there is no change
in its potential energy and that no energy is used in opposing friction; then
the work done will appear entirely in the form of kinetic
energy.
It will be clear that if a block of mass m is at rest on a
smooth horizontal plane and if a steady horizontal force F
is exerted on it through a distance s, then an amount
of work Fs is performed on the block. Since friction is
absent, F acts as an unbalanced force and gives the block
a constant acceleration in accordance with the equation
F = ma. From the la-\vs of accelerated motion, § 3-6, the
distance traversed by a body which starts from rest with a Fm. 6-3. Dia-
constant acceleration a and acquires a speed vis found to be gram used in de-
s = v2 /2a. The energy acquired by the block is obtained termining kinetic
energy of rotation.
by substituting for F and s the values found; it becomes
Fs = (ma) (;:). Consequently, the desired expression for kinetic energy is
mv 2
Ek=--. (6.7)
2
It can easily be shown that if the mass m of the body is expressed in slugs
and the speed v is in feet per second, the kinetic energy is in foot· pounds.
Similarly, with the mass in kilograms and the speed in meters per second,
the kinetic energy will be in newton·meters or joules; and with the mass in
grams and the speed in centimeters per second, the kinetic energy will be in
dyne· centimeters or ergs.
In order to find an expression for kinetic energy of rotation, it may be
supposed that a body, initially at rest, is free to turn in frictionless bearings,
and that a force F is applied to a pin A at the end of a radius r, the force
being directed always at right angles to the radius, as shown in Fig. 3. The
104 MECI!ANIC8 Chap. 6

force produces a torque 3 = F'r, causing the body to accelerate in accordance

with Eq. 5.2, a = 3/I, where I is the moment of inertia of the body. In
sweeping through an angle e from rest, with an angular acceleration a, the
body will acquire a speed w such that e = w2/(2a). The work performed upon
the body, from Eq. 6.5, is

3 e = (Ia) ( w2 ) = I w2.
2a 2
But the kinetic energy of the body equals the work done upon it; hence, the
expression for the kinetic energy of a body having a moment of inertia I
and rota ting at an angular speed w is
Jw2
Ek= 2 - (6.8)

There should be no difficulty in showing that with w in radians per second, if

I is in slug· feet2 , Ek is in foot· pounds; if I is in kilogram· meters2, Ek is in
joules; and if I is in gram· centimeters2, Ek is in ergs.
A body that is undergoing both translation and rotation at the same time
possesses an amount of energy mv 2/2 on account of its translation, and an
amount Iw 2 /2 on account of its rotation, where I is the moment of inertia of
the body for an axis through its center of gravity. The total kinetic energy
of such a body is, therefore,
mv2 Iw 2
Ek=-+-·
2 2
6-6. Potential Energy. Potential energy is the kind of energy that a body
has because of its position. ·when a body is raised to a higher level, it is able
to do a certain amount of work in falling
back again, and hence it was given this
amount of potential energy in raising it.
Although elevating a body is not the only
way to give it potential energy, it is per-
haps the most usual way. If the body is
raised in such a manner that its kinetic
energy is not changed and without energy
being used in opposing friction, the work
Frn. 6-4. Diagram used in determin-
ing potential energy due to elevation. done in raising it will all be expended in
giving the body potential energy. Thus,
a body of mass m may be raised from one level to another at constant speed
along a smooth plane, as represented in Fig. 4, the second level being a vertical
distance h above the first. The plane has a length l and makes an angle fJ with
the horizontal. The force F' needed to move the body of weight mg up the
plane at constant speed is rng sin fJ; and, since this force is exerted through
a distance Z, the work done on the body, and hence the potential energy given
§ G-8 WORK DONE AGAINST FRICTION 10:i

to it, is Fs = (mg sin 0) l = ( mg 1) l. Consequently, the expression for

potential energy is
Ep = mgh, (6.9)
which shows that the amount of potential energy given to a body by raising
it from one level to another depends only on the weight mg of the body and
the vertical distance h between the two levels. Since the result is independent
of the value of 0, the path of the body may be inclined at any angle. The
potential energy of the body is not affected by the path over 1vhich it moves
in reaching a fixed elevation.
The units to be used may be those of any standard system. For example,
in the cgs system, with m in grams, g in centimeters per second 2 , and h in
centimeters, EP will be in ergs.
6-7. Kinetic and Potential Energy in Gravitational Units. Kinetic energy
due to translation can be found in gravitational units by exactly the same
method as used in § 6-5. The weight W of the body is used rather than its
mass, the acceleration a is obtained from Eq. 4.3, F /W = a/g, and the expres-
sion for kinetic energy becomes
Wv 2 (6.10)
Ek=-·
2g
With the weight of the body expressed in pounds, its speed in feet per sec-
ond, and the acceleration due to gravity in feet per second 2, the kinetic
energy will be in foot · pounds.
Kinetic energy due to rotation can be found in foot· pounds by Eq. 6.8,
Ek = lw 2 /2, as already described.
The potential energy of a body due to elevation can be expressed in foot·
pounds by the relation
Ep = Wh, (6.11)

where W = mg is the weight of the body in pounds, and h is the vertical

height in feet through which its center of gravity is raised.
6-8. Work Done Against Friction. Whenever one body slides upon another,
some worlt is done against the friction between them. The amount of energy
dispersed in this manner can be found by supposing work to be done upon a
body under such conditions that neither the kinetic energy nor the potential
energy of the body is increased, since all the work done will then be expended
against friction. Consequently, if a body is moved at constant speed along
a rough horizontal surface, a force must be exerted upon it equal and opposite
to the force f of sliding friction, and the work done in exerting this force
through a distances is equal to the product fs. Hence, the energy expended
against friction is
E1 = fs, (6.12)
lO(i MECHANICS Chap. G

an expression that applies also to rolling friction. The method of calculating

f was discussed in § 4-9.
The work done against friction is converted into heat and is frequently
wasted. Often the heating is so slight as to escape notice, but sometimes it
is quite evident, as in the heating produced in an automobile brake, in an
overheated bearing, or in striking a match.
6-9. Application of Energy Principles. The principles of work and energy
provide a method of procedure by means of which many problems can be
solved with great simplicity. To illustrate how these principles are applied,
three typical problems will be considered.
I. A 2-ton weight in a pile driver is raised 12 ft above the top of a pile and then
released. What force (assumed constant) does it exert against the pile upon falling
if it drives the pile ! ft into the ground in one blow?
In this problem, the potential energy given up by the weight is expended in work
done upon the pile; these two quantities can be equated, and from this equation the
desired force can be found. Since the weight finally comes to rest 12! ft below its
initial position, its potential energy is reduced by an amount Wh = 4000 lb X 12! ft
= 5 X 104 ft·lb. The work done on the pile is E = Fs = F X ! ft, where Fis the
force exerted during the blow. Therefore F X t ft = 5 X 104 ft· lb, or the force is
105 lb or 50 tons.

8 30ft c Frn. 6-6. Motion of a cylin-

der rolling down an inclined
FIG. 6-5. A problem in the transfor- plane studied by energy prin-
mation of energy. ciples.

II. A box is placed at the top of a chute 50 ft long, shown in Fig. 5. For the first
20 ft (AB), the chute is inclined downward at 30° with the horizontal, and the rest
(BC) is level. If the coefficient of sliding friction between the box and the chute is
0.2, with what speed will the box leave the chute at C? Neglect the effect of the bend
at B.
In this problem, the potential energy that the box gives up as it moves from A is
partly used in opposing friction from A to B and from B to C, and the remainder
appears as kinetic energy at C. The foregoing sentence will be written as an equation,
and from this the desired speed can be found. The potential energy given up i8
mgh, where m is the mass of the box, and h is its elevation. The energy used in
opposing friction = (force of friction) X (distance) = (coefficient of friction) X (nor-
mal force between surfaces) X (distance). The box leaves the chute at C with the
kinetic energy t mv2, where vis its speed at this point. The energy equation becomes:

+ µ (mg) (30 ft) + 2mv ·

2
mgh = µ (mg cos 30°) (20 ft)
§ 6-10 RATE OF DOING WORK 107

When the common factor mis canceled and the equation solved for v, it is found that
the box leaves the end of the chute with a speed of v = 5.9 ft/sec.
III. Use the principles of work and energy to find the time taken by a solid cylinder,
starting from rest, to roll without sliding down a plane 200 cm long, that is inclined
at 25° with the horizontal.
The conditions of the problem are represented in Fig. 6. The cylinder will possess
the same amount of energy when it reaches the bottom of the plane as it had at the top
if the dispen,ion of energy by rolling friction is considered negligible. At the top,
its energy is entirely in potential form, but as the cylinder rolls down, its potential
energy becomes transformed into kinetic energy, partly of translation and partly of
rotation. If the energy at the top of the incline is equated to that at the bottom, the
speed of the cylinder upon reaching the bottom can be found, and the desired time
follows at once from the laws of accelerated motion.
The reduction in potential energy from top to bottom of the plane is mgh, and
· ·
the gam · ·
m kmetic .
energy upon reaching . mv2
the bottom 1s 2 + Iw
2
2 , These expres-
sions are equated, thus:
mv 2
+ lw
2
mgh = -
2
-,
2
where mis the mass of the cylinder, I is its moment of inertia about its axis of sym-
metry, v and ware its linear and angular speeds respectively upon reaching the bottom
of the incline, his the elevation of the upper end of the plane, and g is the acceleration
due to gravity. Since the moment of inertia of a solid cylinder of radius r about its
· of symmetry 1s
axis · I = mr2 , § 5-4 , an d its
· angu1ar speed 1s
· w = -,v t h e term -I w2
9~ r 2
can be replaced by 1 (
2 2
mr 2
)(v)2
mv
4
2
; = , whence mgh = mv 2
+ mv
2
2

4
3 mv 2
= - - , and
4

v= ~ 4:h, But the height h of the incline is 200 cm X sin 25° = 84.6 cm; conse-
quently, the speed of the cylinder at the bottom of the incline is

cm
4 X 980 - 2 X 84.6 cm
v=
sec = 334 cm.
3 sec
Since the cylinder started from rest, its average speed along the incline 1s
v = (O + 334)/2 = 167 cm/sec, and the time of descent is, therefore
S 200 cm
t=- = = 1.20 sec.
v 167 cm
sec
The student should compare the foregoing solution with the treatment in
§ 5-7, where the same problem is solved independently by the method of
dynamics.

POWER

6-10. Rate of Doing Work. In the doing of work, the time required or
allowed is of importance as well as the amount of work done. For example,
a motor-driven hoist may be needed for raising a certain load; if the load has
108 MECHANICS Chap. 6

to be raised quickly, a more powerful hoist and a larger driving motor must
be provided than if more time is allowed. Usually the size of machinery is
determined, not by the total amount of ,vork to be done, but by the rate at
which it must be done; that is, by the amount of work required per unit of
time. The time rate of doing work is called power.
·when 1 ft· lb of work is done per second, the power is 1 ft· lb per sec. If a
machine working steadily performs 15,000 ft· lb of work in 10 min, it delivers
1500 ft· lb every minute, and its power output is, therefore, 1500 ft· lb /min or
25 ft· lb/sec. In general terms, the power P of an agent which performs an
amount of work E in a time interval tis given by the ratio:
E
p = t' (6.13)

A machine which operates steadily, doing the same amount of work every
second, provides a constant amount of power. But if it works irregularly,
doing more work in some intervals than in others, the power it provides
fluctuates from moment to moment. Under these circumstances, Eq. 6.13
gives the average power throughout the time considered.
In order to find the instantaneous power at a particular moment, it is
necessary to imagine the average power to be found over a short time inter-
val which includes the instant under consideration, and then to take the time
interval shorter and shorter, approaching zero as a limit. The limiting value
of the power for an infinitesimal time interval is the instantaneous power at
the instant selected and is expressed as
p = dE. (6.14)
dt
Since power is the time rate of doing work, the power unit in any system is
found by dividing the work unit in that system by the time unit. Thus, in
the cgs system, power is expressed in ergs per second. In addition to the
units which occur in the standard systems, there are other practical units for
power that are much used in engineering. The horsepower (hp) is the power
provided by an agent working at the rate of 550 ft· lb per sec or 83 ,000 ft· lb per
min. The watt is a rate of doing work equal to 1 joule per sec; this unit is
named after the Scottish engineer, James Watt (1736-1819). The kilowatt
(kw) is a power unit used in rating electrical machines. The relations be-
tween some power units are given in Table I.
TABLE I. UNITS OF PowER

1 watt = 107 ergs/sec = 1 joule/sec

1 hp = 550 ft·lb/sec = 33,000 ft·lb/min
1 hp = 746 watts
1 kw = 1000 watts = 1.34 hp
1 ft·lb/sec = 1.356 watts
The power of mechanical devices and machinery can be expressed in terms
of speed, for speed includes the idea of time. Work done in motion of trans-
§ G-10 RA'PE OF DOING WORK 109

lation is the product of the force F and the displacement s, and the power
involved can be stated as P = E/t = Fs/t, where tis the time in which the
work is done. Numerically, the ratio s/t is the speed v, and therefore
the power is the product of force and speed, or

P = Fv. (6.15)

Similarly, the work done in rotation is the product of the torque :J and the
angular displacement O, from Eq. 6.5; that is, E = 3 O; consequently, the
power is P = E/t = 3 0/t. Since 0/t represents the angular speed w, it fol-
lows that power is the product of torque and angular speed, or
P = 3w. (6.16)
The concept of power will be illustrated by some practical examples.
I. A horse exerts a horizontal pull of 100 lb on a wagon; how fast must he walk in
order to develop exactly 1 hp?
The horse is to do 550 ft·lb of work per second, meantime exerting a force of 100 lb;
hence, from Eq. 6.15, he must trnvel at a speed

550 ft·lb
P sec ft
v= F= 100 lb = 5 ·5 sec.

II. What power is expended against friction in a sleeve bearing supporting a 2-in.
shaft which turns at 460 rev/min, if the shaft presses against the bearing with a force
of 1 ton, and the coefficient of sliding friction between the sliding surfaces is 0.005?
The force of friction, from Eq. 4.5, is 0.005 X 2000 lb = 10 lb. This force is tangent
to the shaft at its rim, and is therefore exerted through a distance (2 1r/12) ft in each
revolution. Hence, the work done per minute against friction, from Eq. 6.12, is
2 1r ft ft·lb . .
E1 = fs = 10 lb X -12 - X 480~ = 2510 ft·lb, and 2510 -.- 1s eqmvalent to
~ mm
2510
33 000 = 0.076 hp.
'As an alternate solution, the torque due to friction is :J = 10 lb X \ ft, and the
1
angular speed is w = 480 X 2 1r radians/min; the power expended, by Eq. 6.16, is
. 10 radians ft· lb
accordmgly P = 3 w = 12 lb· ft X 960 . = 2510 - .-, as before.
mm mm

The relation between work and power is emphasized by transposing Eq.

6.13 to read E = Pt. In this form, the expression shows that if an agent
provides an amount of power P continuously for a period of time t, the agent
does an amount of work equal to the product Pt. This idea of work as the
product of power and time leads to some widely used energy units. The
horsepower-hour (hp· hr) is thus a unit of work, being the amount of work
performed when the rate is 1 hp over a time of 1 hr. Similarly, the kilowatt·
hour (kw·hr), the unit upon which the cost of electrical energy is based,.is
the amount of work performed when the rate is 1 kw over a time of 1 hr. If
an engine delivers 60 hp steadily for 5 hr, it does 60 X 5 = 300 hp· hr of work
110 MECIIANTCS Chap. G

during that interval; again, if a motor takes 12 kw steadily for 4 hr, it receives
12 X 4 = 48 kw·hr of electrical energy.

MACHINES

6-11. Simple Machines. Experience shows that a stone embedded in the

ground can usually be dislodged with a crowbar and that a heavy automobile
can easily be raised by means of a jack. In these illustrations, the crowbar or
jack serves as an intermediate device upon which work is done, and which in
turn does work upon some other object. A device used for this purpose is
technically called a machine. The complex machines in industry are made
up largely of certain elements that may be regarded as simple machines in
themselves. These simple machines are generally considered as comprising
the lever, the wheel and axle, the pulley, the inclined plane, the screw, and
the wedge.
Usually a machine is employed in order to lessen the force required in
doing a certain piece of work. Thus, if a 500-lb weight is to be lifted, a
machine can be used to exert this amount of upward force upon it while the
person operating the machine exerts a smaller force, say only 50 lb. It is
thus possible, and indeed usual, to obtain a larger force from a machine than
that which is exerted upon it. This statement applies to force and not to
energy, for according to the Law of Conservation of Energy more work cannot
be obtained from a machine than the energy supplied to it. It is evident, since
work = force X distance, that when the operator exerts a smaller force than
does the machine, he must exert it through a correspondingly greater distance.
6-12. Efficiency. Because of energy wasted through friction in all moving
machinery, however well designed, the energy given out by a machine is less
than that supplied to it. More definitely, the Principle of Conservation of
Energy shows that
Energy input = energy output + energy wasted,
if no energy is stored up in the machine. This statement is true for any
period of time, and hence for unit time; and since energy per unit time is
power, it can also be said that
Power input = power output + power wasted.
The efficiency of a machine is defined as the ratio of its output to its input, both
output and input being expressed in the same unit of energy or power. This
ratio is always less than unity; it is usually multiplied by 100 and stated in
per cent. Thus,

.
Per cent efficiency =
energy output
. t X 100 }
energy mpu
(6.17)
= power ~utput X lOO.
power mput
§ 6-13 MECHANICAL ADVANTAGE 111

A highly efficient machine is one in which a large part of the power supplied
is given out to the load and only a small part wasted. The efficiency of a
large electrical generator may be as high as 98 per cent. In some simple
machines-a screw jack, for example-considerable friction is necessary to
prevent the load from running down after it has been raised; because of the
energy used in opposing friction the efficiency of a screw jack is less than 50
per cent.
Calculate the power supplied to a 10-hp motor which is delivering its full rated
output at an efficiency of 85 per cent.
From Eq. 6.17,
p . t power output X 100 10 hp X 100 = 11 75 h
ower mpu = t ffi . 85 . p.
per cen e ciency
Machines are rated in terms of output-thus, a 10-hp motor, such as that
referred to in the foregoing problem, is one which is capable of delivering 10
hp without exceeding its design limitations.
6-13. Mechanical Advantage. The utility of a machine, as pointed out
in § 6-11, is chiefly its capacity for exerting a larger force than the person
operating the machine is willing or able to exert himself. The ratio of the
force W exerted by the machine on a load to the force F exerted by an operator
on the machine is called the mechanical advantage of the machine. For exam-
ple, if a hoist has a mechanical advantage of 20, a force of 100 lb exerted upon
it will enable the hoist to raise a load weighing 20 X 100 lb or 1 ton.
The ideal mechanical advantage of a machine is the mechanical advantage
that the machine would have if friction ,vere absent, and can be computed by
geometry. If the machine exerts a force W through a distance h while the
operator exerts his force F through a distances, the energy output is Wh and
the energy input is Fs. On the assumption that friction is absent, the output
and input may be equated-an ideal condition not attainable in practice--
and it follows that the ratio of W to Fis the same as the ratio of s to h. There-
for~, the ideal mechanical advantage can be stated in terms of these distances,
as follows:
s
Ai= h. (6.18)

If, on the other hand, friction is considered in calculating the mechanical

advantage of a machine, or if the mechanical advantage is found by actual
test, in which case friction is present with its wasteful effects, the actual force
ratio W/Fis called the actual mechanical advantage of the machine. Thus,
lV (6.19)
A"= F ·
The actual mechanical advantage cannot be calculated exactly, tor such a
calculation would require the frictional forces to be known, and these are
based on the coefficient of friction, which is not very definite, § 4-9. The
ideal mechanical advantage, on the contrary, can be found easily and is
112 MECHANICS Chap. G

computed in the following section for several of the simple machines men-
tioned.
6-14. Ideal Mechanical Advantage of Some Machines. In order to find
the ideal mechanical advantage of some simple machines, the procedure for
each will be to assume that the machine exerts a force W through a distance
h while the operator exerts his force F through a distance s, and to express
the ratio s/h in terms of the dimensions of the machine. The several arrange-
ments, with dimensions, are shown in Fig. 7.
I. The Pulley. A typical pulley system is shown in part I of the figure,
with the pulleys separated for clearness. In order to raise the load W through
a height h, each of the four ropes (assumed parallel) must be shortened by
this amount, and hence the operator must exert his force F through the dis-
tance s = 4 h, making s/h = 4. Therefore, the ideal mechanical advantage
of the arrangement shown is 4, corresponding to the four ropes used to sup-
port the load.
II. The Lever. In one class of lever, shown in part II, s/h = Zi/Z2, by
similar triangles; hence, the ideal mechanical advantage is

Ai = y;·Z1
III. 1'he Inclined Plane. Part III of the figure represents a smooth plane
inclined at an angle () with the horizontal. In order to elevate the load W by

TI

.
f--,>1

F
( into paperJ

m
Frn. G-7. Some simple machines: I, pulley; II, lever; III, inclined pla.ne; IV, screw jack.
§ 6-H JT)Ji}AL 1v1ECITANICAL ADVAN1'AGE OF SOiHE ivIACJIINBS 113

an amount h, a force F = W sin 8 must be exerted through a distance s = L

along the plane. Consequently, s/h = L/h, and the ideal mechanical advan-
tage is
L
Ai=!/
IV. The Screw Jack. In part IV is shown a screw jack, which may be
looked upon as an inclined plane wrapped around a cylinder. The load W is
raised by an amount p, the pitch of the screw, by exerting a horizontal force
F through a complete circle of radius Z. Therefore, s/h = 2 1rl/p, and the
ideal mechanical advantage of this machine is
A. = 2 -rrZ.
i p

Example. A single problem will serve to coordinate the ideas of efficiency

and mechanical advantage of a machine.
Ruppose an operator to exert a force of 40 lb in raising a load weighing 360 lb through
a vertical height of 1 ft by pushing it up along an inclined plane 12 ft in length. The
actual mechanical advantage of the machine is W /F = 360 lb/40 lb = 9, and the ideal
· 1 a d vantage Is
mec h amca L = 12it-
. h it-- = 12 . I n raismg
. . t h e 1oa d , the energy output .IS
1
360 lb X 1 ft = 360 ft· lb, and the energy input is 40 lb X 12 ft = 480 ft· lb; conse-
.
quen tl y, the effi ciency 1s ~ X 100 = 7~o per cent.
. 360 -ft-.-JJi_
480

Incidentally, it may be stated that the efficiency of a machine is always

equal to the ratio of the ackial mechanical advantage to the ideal mechanical
advantage.

PROBLEMS
1. How much work is needed to raise a 90-lb anchor through a height of 5 fathoms
(1 fathom = 6 ft) at constant speed, if friction and buoyancy are neglected?
2. A 3-kg object is to be moved 2 m along a smooth level surface in 3 sec, starting
from rest. (a) How much work is needed to accomplish this result? (b) What is the
final speed of the object?
3. A 4CO-grn block is drawn along a level surface at a constant speed of 100 cm/sec
by a steady horizontal force applied for 2 sec. The coefficient of friction between the
block and the supporting surface is 0.35. Calculate the work done on the block.
4. A 60-lb sled rests on level snow, the coefficient of friction between sled and snow
being 0.04. How much work does a boy do upon the sled if he exerts a steady hori-
zontal push of 6 lb upon it through a distance of 50 ft? The sled is painted red.
(Some unnecessary data are given in this problem.)
5. A uniform ladder 20 ft long and weighing 40 lb is lying on the ground. How
much work is needed to stand it in a vertical position?
6. A 400-lb log is dragged 80 ft along level ground by a rope attached to a tractor;
the rope makes an angle of 25° with the ground and is under a tension of 100 lb. How
much work is done on the log?
7. How much work is required to bring a flywheel that has a moment of inertia
of 20 slug· ft 2 from rest to a speed of 120 rev /min in 40 sec?
114 MECIIANIC8 Chap. 6

8. A 50-lb weight is hung on a spring balance and causes the spring to elongate
1 in. Then a weight of 100 lb is added, causing a further elongation of 2 in. How
much work is done upon the spring in each of the8e operations?
9. An automobile wheel weighing 40 lb has a diameter of 2.5 ft and a radius of
gyration of 1 ft. What is the kinetic energy of the wheel when it is rolling along the
ground at HO mi/hr?
10. A ball hanging on a thread, when pulled aside and released, vibrates as a
pendulum. If the thread is 40 cm long, through what angle must it be displaced in
order that the ball may sweep through its midposition with a speed of 100 cm/sec?
11. A stone is whirled around in a vertical circle at the end of a cord 2 ft long.
If the stone has a speed of 12 ft/sec at the top of the circle, what is its speed at the
bottom of the circle?
12. A 50-lb box is initially at rest. A man pushes it along a level floor by applying
a horizontal force of 15 lb to it for a distance of 10 ft. He then stops pushing and
the box comes to rest in sliding 1 ft farther. Compute the coefficient of friction
between the box and the floor.
13. A 50-lb sled, starting from rest, is drawn along level snow a distance of 20 ft by
a steady pull of 10 lb applied to it in a direction 30° upward from the horizontal.
The coefficient of friction between the sled and the snow is 0.10. Compute (a) the
total work done on the sled, (b) the energy wasted in opposing friction, and (c) the
final kinetic energy of the sled.
14. How much of the work done on the sled in Prob. 4 was used (a) to oppose
friction, and (b) to give the sled kinetic energy?
15. A 10-lb block sliding along a level surface encounters a plane inclined upward
at 30° from the horizontal and starts up the incline at a speed of 20 ft/sec. The co-
efficient of friction between the block and plane is 0.2. (a) How far along the incline
will the block slide before it comes to rest? (b) After stopping, will the block slide
down again, and if so, with what acceleration?
16. A hammer head that weighs 2 lb is moving at 10 ft/sec at the instant it strikes
a nail. What average force does it exert upon the nail if it drives the nail -a- in. into
the wood?
17. A block slides down a chute like that shown in Fig. 5, except that the portion
AB is 4 m long, inclined at 40° with the horizontal, and the level portion BC is 10 m
long. The block is released at A and slides to rest between B and Cat a point 7 m
from B. Assume the coefficient of friction between the block and the chute to be
constant throughout and determine its value.
18. A 20-lb box is placed on a plane 10 ft long that is inclined at 30° with the hori-
zontal. A man applies a 8teady pull P to the box and moves it up the incline. The
force P is parallel to the plane and its magnitude is such that the box, which starts
from rest at the bottom of the incline, has a speed of 10 ft/sec at the top. The co-
efficient of friction between the box and the plane is 0.3. Determine (a) the potential
energy acquired hy the box, (b) its gain in kinetic energy, (c) the energy used in
opposing friction, (d) the total energy expended by the man, and (e) the value of the
force P.
19. \Vhat average horsepower is required to produce the result specified in Prob. 7?
20. In the punch-press problem solved in § 3-13, suppose the moment of inertia of
the :flywheel to be 30 slug· ft2 and determine how many horsepower are made available
for the punching operation by the slowing down of the :flywheel.
21. A locomotive is coupled to a 400-ton train on a level track. If the friction
force on the train amounts to 8 lb per ton of weight, (a) what average horsepower
must the locomotive develop to bring the train from rest to a speed of 45 mi/hr in a
distance of 500 yd? (b) What horsepower must it develop to maintain this speed?
 PROBLEMS 115

22. A grinding wheel 8 in. in diameter is driven at a constant speed of 3000 rev/min.
A tool bit is sharpened by pressing it perpendicularly against the rim of the revolving
wheel with a steady force of ! lb. If the coefficient of friction between the wheel
and the bit is 0.5, what power is used during the sharpening process?
23. A pulley system lifts a 250-lb load through a height of 1 ft while the operator
exerts a force of 50 lb upon the pulley system through a distance of 8 ft. Compute
(a) the ideal mechanical advantage, (b) the actual mechanical advantage, and (c) the
efficiency, of the pulley system.
24. A man rowing a boat grips the oar at a place 2 ft from the oarlock and pulls
the oar toward himself with a force of 20 lb. If the blade of the oar is 5 ft from the
oarlock, what is (a) the ideal mechanical advantage of the oar, and (b) the force that
the oar exerts on the water?
 7
ELASTICITY AND
HARMONIC MOTION
In the foregoing chapters it was assumed that bodies were rigid and that,
under the action of applied forces, a body would move as a whole, perhaps
with translation, or rotation, or a combination of these two types of motion.
Consideration is given in this chapter to the action of forces in distorting a
body, the particles of the body moving relatively to each other through rather
narrow ranges, and it is understood that the forces are applied in such a man-
ner that the body in its entirety does not move.
It will also be shown that when a body is distorted, provided it is elastic,
those parts which have been displaced will spring back when the distorting
force is removed and will vibrate with harmonic motion about their rest posi-
tions. Such vibrations are commonly observed in machinery and structures.

ELASTICITY

7-1. Hooke's Law of Elasticity. Many observers have noted the bending
of a piece of wood and its return to the original shape upon removal of the
bending force, or the stretching of a metal wire and its return to the original
length upon removal of the stretching force. These effects are evidences of
elasticity; an elastic body is one which, having been strained by the applica-
tion of distorting forces, returns to its original condition upon the removal of
those forces.
When a load is applied to an elastic body, the body becomes distorted; that
is, some portion of it is displaced with respect to some other portion which
has not been disturbed. Because of this displacement, forces are developed
between the molecules of the body which resist the change that the applied
load has brought about, and thereby tend to restore the body to its original
condition. The aggregate of the molecular forces will just balance the
applied load when the body has reached a condition of rest under the action
of these forces. The larger the load, the larger will be the deformation neces-
sary to establish molecular restoring forces that balance the increased load.
Robert Hooke (1635-1703), English experimental physicist, found that the
force developed within an elastic body to withstand an outside load is directly
proportional to the deformation of the body under that load.
116
§ 7-1 HOOKE'S LAW OP ELASTICITY 117

The correctness of this statement can be verified by applying weights to a

helical spring or a flat spring in the manner indicated in Fig. 1. If different
weights W 1 , W2, W3 , etc., are placed
in turn upon either spring, it will be
found that the steady deflections
produced by them, respectively di,
d2, d3, etc., are proportional to the
applied weights, as shown by the
straight line plotted in part I of Fig.
2. The spring balance pictured in
Fig. 1-3 operates on this principle.
For any steady deflection, the restor-
ing force within the spring is equal w
(and opposite) to the applied weight,
and therefore the restoring force is Frn. 7-1. Two types of spring deformed by
proportional to the deflection or de- loading.
formation.
The deformation of an elastic body is usually expressed as the ratio of the
change produced by the applied force to some original dimension of the body;
the term strain is used as the measure of deformation. Thus, if a sheet of
rubber 10 in. square is stretched to 11 in. in both directions, the strain would
appropriately be reckoned as the ratio of change in area to original area, or
(11)2 - (10)2 -
(10)2 - 0.21.
The restoring action set up in an elastic body when strained is commonly
stated as the force per unit area over which the force acts, and the term stress
is used for this purpose. Thus, stress is the restoring force per unit area.
Engineers commonly refer to this quantity as "unit stress."
Expressed in terms of strain and stress, Hooke's Law states that the stress
set up within an elastic body is proportional to the strain to which the body is

Breaking
point

1 2 3 Strain
Extension of spring
I II
Frn. 7-2. Relation between stress and strain. I, Variation of stretch with applied load
within elaRtic- region; II, loading continued to rupture.
118 MECHANICS Chnp. 7

subjected by the applied load. Stated concisely, stress rx :,,train,

or Stress = k X strain;

hence, k = stre~s, (7.1)

st ram
where k is the modulus of elasticity of the m::i,terial of the body.
For each material there is a limiting stress beyond which the proportion-
ality expressed by Hooke's Law does not hold; this is called the proportional
limit. The elastic limit is a limiting stress beyond which the specimen does
not return to its original condition when the distorting force is removed.
Usually the proportional limit and elastic limit have about the same value,
and they are often taken as equal. If the stress is larger than the elastic
limit of the material under test, a permanent distortion will result; at a
higher stress called the yield point, many substances continue to stretch
without increase in load; and at a still higher value, called the breaking stress,
rupture will occur. Some of these quantities are indicated in part II of Fig.
2, which is a typical stress-strain diagram for soft steel, not drawn to scale.
This chapter deals with elastic bodies subjected to forces which do not cause
the proportional or elastic limit to be exceeded.
The deformation of a body may occur in three different ways, depending
upon the manner in which the external forces are applied.
In one of these, the load tends to change the length of the

r
body, in another to change its volume, and in the third to
change its shape. The strain and the stress are evaluated
differently for these cases, but Hooke's Law applies equally
to all three.
7-2. Elasticity of Length; Young's Modulus. In order to
measure elasticity of length, a specimen in the form of a rod
or wire is generally used. If a wire of length L and cross-
sectional area A, as shown in Fig. 3, is subjected to tension
by applying two equal and opposite forces F, F' to its ends,
and if the wire elongates an amount 1:1L, then the strain is
Fm. 7-3. A test
to measure the elongation 1:1L divided by the original length L and the
stress and strain stress is the force F divided by the cro::'ls-sectional area A.
in a wire or rod. Therefore, strain = 1:1L/L and stress = F/A; consequently,
the ratio of stress to strain is
y = stress = F / A . (7.2)
strain 1:1L / L
This constant Y, the stretch modulus of elasticity, is called Young's modulus,
after the English physicist Thomas Young (1773-1829). It should be men-
tioned that as the wire elongates its diameter contracts, but this fact is not
used in computing the modulus.
§ 7-3 ELASTICITY OF VOLUME; BULK MODULUS 119

Young's modulus, like the other elastic moduli, is a property of the material
concerned and basically is not dependent upon the dimensions of the particu-
lar sample used in a test. This statement agrees in general with experiment,
although the modulus is influenced somewhat by the shape, size, and treat-
ment of the specimen. Young's modulus applies equally well to decrease of
length in compression and to increase of length in tension.
In British units, stress is often measured in pounds per square inch and
strain in inches of extension (or contraction) per inch of original length; conse-
quently, the corresponding unit for Young's modulus is Y ~bl)~·
=
2

Ill. lll.
= lb/in. 2
In metric units, stress may be measured in dynes per square centimeter, in
newtons per squarn meter, or in similar units; and strain is usually measured
in centimeters per centimeter. Since strain has no unit, the unit for Young's
modulus is the same as that for stress.
Young's modulus has a very large numerical value as compared with elastic
limit and breaking stress. For example, for soft steel the stretch modulus of
elasticity is 30,000,000 lb/in. 2 , the elastic limit is 35,000 lb/in. 2 , and the break-
ing stress is 60,000 lb/in.~ In designing structures, engineers allow a "unit
stress'' for steel of perhaps 16,000 lb/in. 2 to give an ample factor of safety.
As a numerical example, compute the elongation of a steel tie bar 24 ft long, 4 in.
wide, and ~ in, thick, when subjected to a tensile force of 3 X 104 lb. The stress is
3 X 104 lb/(4 X ! in. 2 ) = 1..50 X 104 lh/in. 2 By Eq. 7.2, the strain is found to be

\J.b
. stress
1.50 x 104 -;-~
1 .2 1
Stram = --- = \ = --.
y 3 00 101 Jb 2000
. x in\2
Since the strain is the elongation b,L divided by the original length L, it follows that
the elongation is /11_, = L X strain = 24 ft X (1/2000) = 0.012 ft = 0.144 in.

The extensions which occur ,Yhen a rod or wire is placed under tensile loads
are so small that they are difficult to measure. In laboratory tests on sus-
pended ·wires an optical lever is of ten used for this purpose. A small inclined
mirror leans against a clamp carried by the wire near its lower end, and a
narrow beam of light played upon the mirror is reflected back to a scale for
observation. As the wire lengthens under load the tilting of the mirror
increases, and the reflected light beam moves along the scale. By this
means, the scale readings furnish a highly magnified reproduction of the
elongations of the wire.
7-3. Elasticity of Volume; Bulk Modulus. A body that is subjected to
normal forces pressing over its entire surface is said to be subjected to a
press1tre. Pressure can be applied by submerging a body in a liquid, for the
force due to the liquid will act perpendicularly on each portion of the surface,
120 MECHANICS Chap. 7

as described in the following chapter. The term "pressure" means the force
acting per unit area.
Figure 4 represents a cube of some material subjected to hydrostatic pres-
sure, with equal forces f acting normally on all its faces. If the pressure on
the cube is increased, its shape remains unchanged but its volume decreases.
The strain is defined as the ratio of the change of volume AV to the original
volume V of the body; thus, strain is AV/ V. The corresponding stress set
up within the material is equal to the increase in pressure Ap. The ratio of
stress to strain in this instance is called the bulk modulus of elasticity and is
expressed as
- -Ap (7.3)
B - AV/V'
the minus sign being introduced because an increase in pressure causes a
decrease in volume-that is, when Ap is positive, AV is inherently negative.
The unit for the modulus B is the same as that for the stress, since the
strain is the ratio of one volume to another and has no unit. This modulus
is applied particularly to fluids; they offer resistance only to change of vol-
ume and so have only a bulk modulus of elasticity.
The reciprocal of the bulk modulus of a substance is known as the com-
pressibility of that material. The bulk modulus of elasticity of steel is
24,000,000 = 24 X 106 lb/in. 2 ; therefore, its compressibility is 1/(24 X 106)
= 4.2 X 10-s in. 2 /lb, meaning that an increase of hydrostatic pressure of
1 lb/in. 2 would decrease the volume of 1 cubic inch of steel by 4.2 X 10-s in. 3

f
Fm. 7-4. A cube subjected to hy- FIG. 7-5. The effect of shearing
drostatic pressure. forces on a block.

7-4. Elasticity of Shape; Shear Modulus. Often a body is subjected to a

pair of equal forces which act in opposite directions but not along the same
line, as in a couple, § 1-12. Such a pair of forces SS is shown acting upon
the upper and lower faces of a rectangular block abed in Fig. 5, and shear it
into the parallelopiped ab' c' d, turning the end faces ab and cd through the
small angle ct,. Upon removal of the shearing forces, the body will resume
§ 7-4 ELASTICITY OF SHAPE; SHEAR MODULUS 121

its original shape provided the elastic limit of the material has not been
exceeded. This type of elasticity is called elasticity of shape, or shear, and is
of great importance in structural design and in power transmission by shafting.
The block shown in Fig. 5 may be regarded as made up of a large number of
horizontal sheets of area A, each of which is forced to slide along slightly
,vith respect to its neighbors under the action of the shearing forces. The
strain is measured by the angle cp through which the ends of the block are
sheared, and is expressed in radians. The area designated as A, which is
equal to the top or bottom area of the block, is the area that resists the
shearing action; hence, the shearing stress is S/ A. The ratio of the stress
to the strain is constant for a particular material, in accordance with Hooke's
Law, and is expressed as

E = S/A. (7.4)
<P

This constant E is called the shear modulus of elasticity and is also known as
the coefficient of rigidity. Since the angle of shear is a number only, the unit
in which the modulus is expressed is the same as that for stress.

Fw. 7-6. The optical lever as used to measure the angle of shear.

Figure 6 indicates how the optical lever can be used to measure the angle
of shear. One end face of the block is polished to serve as a mirror, and a
ray of light from a lamp is reflected by this polished surface upon the scale
shown. As the face of the block turns through an angle ¢, the reflected ray
of light turns through 2 </>, see § 28-4. If the scale is at a distance r from
the mirror, and if the scale deflection due to shear is s, the angle 2 <f> = s/r.
Consequently, measurements of r ands determine the angle of shear <f>.
As a numerical example, assume a cube of alloy having faces 5 X 5 cm to be sub-
jected to a pair of shearing forces applied as shown and each amounting to 1.2 X 1010
dynes. The angle of shear is observed by the lamp-mirror-scale arrangement just
mentioned, and the reflected beam is shifted 8.2 mm along a scale located 3 m from
the mirror surface of the cube. What is the shear modulus of the alloy?

Here, the strain is the angle of shear ¢ = 2\ = 2 X8 ·:000 = 1.37 X 10-3 radian,
122 JfECHANICS Chap. 7

. S 1.2 X 1010 dvnes

and the stress 1s -A = 9 ,. • = 4.80 X 108 dynes/cm 2 ; therefore, the shear
... o cm2
modulus of the alloy is E = (4.80 X 108 )/(1.37 X 10-3 ) = 3.5 X 1011 dynes/cm 2 •

7-5. Summary of Elastic Constants. In the foregoing, the expressions for

the three moduli of elasticity were all derived from Hooke's Law, the stress
in each being measured in force units per unit area, and the strain being a
ratio of two like quantities and therefore a number; consequently, the unit
for the modulus is the same as that for the stress. The expressions for elastic
modulus of length, volume, and shape are summarized here for convenience
of reference:

Y oung ,s mod u1us = longitudinal stress F/ A

= ·
change of length per unit length t:.L/L'
_ increase of pressure _ - t:.p .
Bulk modulus
decrease of volume per unit volume t:. V /V'
shearing stress S/ A
Shear modulus = angle of shear = -;;;-·
Representative values of these elastic constants for several materials arc
listed in Table I.
T ABU, I. ELASTIC CoxsTAi'iTS

Young's l\:Iodulu;; Bulk Modulus Shear Modulus

Material
lb/in. 2 dynes/em2 lb/in. 2 dynes/cm 2 lb/in. 2 dynes/em 2
-----
Solids
Aluminum ... 10 x lQ6 69 x 1010 8.0 x 106 55 x 1010 ;3_9 x 106 27 x 1010
Brass ........ 16 x lQ6 110 x 1010 16 x 10 6 110 x 1010 6.0 x 106 41 x 1010
Copper ...... 15.5 x 106 107 x 10 10 18 x lQ6 124 x 1010 5.7 x 10 6 39 x 1010
Glass (crown) 9.5 x 106 65 x 10 10 5.6 x 106 38 x 1010 3.9 x 10 6 27 x 1010
Iron (cast) . . . 13 x 106 90 x 10 10 8.7 x 106 60 x 1010 5.2 x 10 6 :.36 x 1010
Eiteel (soft) ... 30 x 10 6 210 x 1010 24 x 1()6 165 x 1010 11.6 x 10 6 80 x 1010
Timber ...... 1.4 x 10 6 9.7 x 10 10 ......... . ....
' . . . ........ . ......
Liquids
Alcohol ...... 1G x 104 110 x 108
Ether ....... 8.7 x 1G4 60 x 108
Mercury ..... 400 x ]04 2800 x 108
Water ....... :31 x 104 210 x 108

7-6. Effect of Elasticity on Impact. The behavior of bodies during impact

or collision was described in § 4-11, and the Law of Conservation of Momen-
tum was introduced and illustrated. This law, which states that momentum
cannot be destroyed, is universally true and applies to every collision. The
energy transformations that take place during impact, however, depend upon
the elasticity of the colliding bodies, and vary widely under different condi-
tions.
Impact between bodies occurs in a very short interval of time, but in that
~ 7-6 JtFFJ!JCT OF ELASTICITY ON IMPACT 123

brief period the bodies become deformed and a certain amount of energy is
used to change their shape. The deformation of a golf ball when struck by
a club is clearly evident in Fig. 7. If the colliding
bodies are inelastic, they remain distorted and the
energy used to distort them is not recovered. If,
on the other hand, the colliding bodies are elastic,
they will spring back to their original shapes and
the energy used to cause the deformation is re-
stored.
In an inelastic collision the colliding bodies do
not spring apart but remain in contact after im-
pact. In studying such a collision the only infor-
mation available to work with comes from the
Law of Conservation of Momentum, to which
reference has just been made.
For an elastic collision two independent prin-
ciples can be applied: (1) the conservation of mo- Frn. 7-7. Deformation of
mentum, as before, and (2) the conservation of golf ball when struck by club.
energy, which states that the sum of the kinetic X-ray photograph taken in
one-millionth of a second.
energies of the bodies before impact equals the
(Courtesy of Westinghouse
sum of their kinetic energies after impact, since Electric Corporation).
there is no loss of energy in the system. In mathe-
matical form, if an elastic body having mass m1 and velocity v1 collides with
another having mass m2 and velocity v2, and if the velocities of these bodies
after impact are vi' and v2', respectively, then,
! m1v12 + ! m2vl = ! m1(v1') 2 + ! m2(v2') 2.
Elastic impact can be illustrated by two blocks of steel which slide toward each
other on a smooth horizontal surface and collide head-on. Block 1 has a mass of
1000 gm and a velocity to the right of 20
1000-gm block 250-gm block cm/sec; block 2 has a mass of 250 gm and
moves to the left at 10 cm/sec, as represented

~~~
in Fig. 8. Assume that these bodies are elas-
tic and determine their velocities after colli-
20~ sec
10~sec sion.
Let the desired velocities after collision be v1 '
Frn. 7-8. A problem to illustrate for block l and v2' for block 2, and consider
elastic collision. velocities to the right as positive. Since the
total momentum of the two bodies is the same
before and after impact, as stated in Eq. 4.7,

(1000 gm) (20 cm)

sec
+ (250 gm) (-10 sec
cm) = (1000 gm) vi' + (2.50 gm) v 2 '.

Also, since the bodies are elastic, their total kinetic energy is the same before and after
impact; therefore,

1000 gm ( 20 cm) 2
2
+ 2502gm ( - 10· cm ) 2
= 1000 gm (V1 ') 2
2 + 2502gm (V2 ') •
2
sec sec
124 MEOIIANJCS Chap.7

These expressions when simplified become, respectively,

cm
4 V1 1 + v/ = 70 -sec
,

cm2
and 4 (vi') 2 + (vl) 2 = 1700 - .
sec2
When these equations are solved simultaneously, it is found that v1 ' = 8 cm/sec and
v2' = 38 cm/sec. The results show 'that the small block reverses its direction and
moves to the right with a speed of 38 cm/sec, while the large block follows behind it at
8 cm/sec.

HARMONIC MOTION

7-7. Elasticity and Harmonic Motion. It was pointed out at the begin-
ning of this chapter that when an elastic body is deformed and released, it
returns to its original position. As it does so, its inertia causes it to move
beyond this position, and it vibrates to and fro before finally coming to rest.
Thus, a weight fastened to a coiled spring, as in Fig. 1, when pulled down
from its rest position and released, vibrates up and down about its original
position as a midpoint. In pulling the weight down, if only a small force is
applied, the spring stretches only slightly, and if more force is applied, it
stretches more. The stretching force and the extension of the spring vary
in direct proportion, provided the force is not too large. This behavior is,
indeed, characteristic of elastic substances, as already described.
Whenever the spring is stretched by a force, there is set up within it an
equal force in the opposite direction. It is this restoring force which pulls
the weight up when it is released. Since acceleration is proportional to the
force that causes it, the acceleration of the weight is proportional to the
restoring force, and thus varies directly as the displacement of the weight
from its central position of equilibrium. It is observed, also, that when the
displacement is downward, the restoring force and acceleration are directed
upward, and vice versa. Therefore, in this vibratory motion, the acceleration
of the vibrating body and the restoring force acting upon it are proportional to its
displacement from the midpoint of its path and are directed toward that point.
Motion of this type is known as harmonic motion.
It should be remarked that all to-and-fro motion is not necessarily har-
monic; in fact, it will not be harmonic unless it has the properties just de-
scribed. There are many examples of machine parts in which motion occurs
back and forth repeatedly over the same path in equal time intervals but does
not conform to the foregoing definition. Such motion is periodic (recurring
in equal periods) but not harmonic. Harmonic motion is illustrated by the
oscillation of a weight on a spring, the vibration of elastic bodies in general,
and the motion of the bob of a long pendulum..
7-8. Relation Between Circular and Harmonic Motion. The study of
harmonic motion is much simplified by a relation that it bears to motion in a
circle. If a body moving at constant speed around a circular path is viewed
§ 7-9 TERMS RELATING TO HARMONIC MOTION 125

in the plane of the circle, it is observed to move to and fro along the diameter
of the circle, and this particular form of periodic motion is found to be har-
monic. In brief, the projection of uniform circular motion in the plane of
the circle, as in Fig. 9, is harmonic motion. It can be produced mechanically
by a disk and a bar coupled by a pin-and-slot arrangement as shown in Fig.

Fw. 7-9. Shadowgraph dem- Frn. 7-10. A

onstration of harmonic motion. mechanical
illustration of
harmonic mo-
tion.

10. The projecting pin near the rim of the disk engages the slot in the bar,
the bar itself being supported in stationary guides to constrain its motion to
the vertical direction. When the disk is driven at constant speed, the pin
travels uniformly in a circle, but the bar receives only the vertical projection
of this motion, and consequently slides up
and down the diameter of the circle with D D'
I p'
harmonic motion.
7-9. Terms Relating to Harmonic Mo-
tion. Usually when a body describes har- M
p 1~----- --r~71y~
___ J__:Jl_J_ M'
monic motion, there is no associated body O i-v,2-y2
which actually travels with circular motion I
·I
as just described. It is therefore conveni- I
I
ent to construct a so-called reference circle, c -------=--.JC---C'
and to imagine a body moving uniformly
around it, as in Fig. 11. Here, if a body FIG. 7-11. Diagram showing the
relation between harmonic and cir-
vibrates up and down harmonically along cular motion.
CD, its motion can be considered as the
projection of the motion of another body that is assumed to revolve uniformly
around the reference circle of diameter C' D'. Thus, when the revolving body
is at M', the vibrating body is at the midpoint ~f of its path; as the revolving
126 MECHANICS Chap. 7

body moves to P', the vibrating body moves to P; when the revolving body
reaches D', the vibrating body arrives at D. As the motion continues, the
angle e at the center O increases, sweeping out 2 1r radians in each revolution.
When the body in harmonic motion moves from M to D, then to C, and
finally back to NI, it has completed one vibration. In general terms, a vibra-
tion is a complete to-and-fro movement of the body over its entire path; it
corresponds to a complete revolution of the body in the reference circle.
The displacement of a body in harmonic motion at any instant is its dis-
tance from the midposition at that instant. In Fig. 11, when the body is at
Pits displacement is y = 11f P and is directed upward.
The amplitude of a vibration is the maximum displacement and is repre-
sented in the figure by either MC or MD. The amplitude is equal to the
radius r of the reference circle.

Frn. 7-12. Harmonic motion represented by a sine curve.

The period or periodic time is the time interval during which a vibrating
body completes one vibration. In this time the body in the reference circle
makes one revolution.
The frequency of vibratory motion is the number of vibrations per second
(vib/sec) and is therefore the reciprocal of the period. If the frequency is
denoted by f and the period by T, it follows that f = 1/T.
The equilibrium position of a body in harmonic motion is the midpoint of
its path. In the figure, the equilibrium position is at M.
Another characteristic of harmonic motion will be explafoed by reference
to Fig. 12, in which harmonic motion along the path CD is combined with
uniform motion at right angles to this path, as though the vibrating body
made a trace on a strip of paper while the strip was moving uniformly across
the page. The starting point M" of the curve represents the vibrating body
at an instant when it is at J\,1 and moving upward, the corresponding body in
the reference circle being at M'. As the latter body moves to P', the radius
rotates through the angle e and the curve advances to P", which therefore
represents the vibrating body when at point P. In a similar manner, the
entire vibration can be mapped out, and the position of the vibrating body
§ 7-11 VELOCITY AND ACCELERATION IN llARMONIC MOTION 127

and its direction of motion can be identified at any instant by a correspond-

ing point on the curve. The displacement at a particular instant is repre-
sented on the curve by the distance y, and the amplitude of vibration by the
distance r between the extreme points and the axis. The angle () through
which the radius of the reference circle rotates while the corresponding body
moYes from its assumed datum position 111' to position P' is called the phase
angle. At the instant represented by P, P', and P" in the figure, the phase
of the vibration is 30° or --/2 period. Phase angles can be indicated on the
rurve as shown at the bottom; in one period T the curve advances a complete
eycle, and the phase angle meanwhile increases from 0° to 360°. Phase
angles can also be reckoned from any other position taken as a datum, such
a8 C or D.
7-10. Displacement in Harmonic Motion. For a particle in harmonic
motion, the displacement can be expressed directly in terms of amplitude and
phase angle. From Fig. 12, the ratio of the displacement y to the amplitude
r is the sine of the phase angle e; that is,
y = r sin &.

From the form of this equation, the curve in the figure is called a sine curve.
If time is reckoned from the instant when the phase angle is zero, then
t = 0 when () = 0, and the angular speed of the radius of the reference circle
can be stated as w = Bjt. It follows that 8 = wt, and the foregoing equation
can be rewritten as
y = r sin wt.

The angular speed of the radius is also equal to w = 2 1r / T, where T is the

period, or w = 2 1rf, where f is the frequency of vibration. Therefore, the
displacement can also be expressed as

y = r sin 2 1r f t. (7.5)

7-11. Velocity and Acceleration in Harmonic Motion. In the preceding

section, the displacement of a particle .in harmonic motion was given as
y = r sin wt. The velocity of the particle can be obtained at once by dif-
ferentiating this expression with respect to time; thus,

V = -dy
dt
-d (r sin
= dt . wt ) = rw cos wt = rw cos ()
' .

The term w can be replaced by its equivalent 2 1rf, and the function cos () by
Vr 2 - y2 /r, as will be clear from Fig. 7-11, and therefore the velocity is
(7.6)

The acceleration of the particle is found by differentiating the velocity

equation with respect to time; since v = rw cos (),
128 MECIIANICS Chap. 7

dv d . dO
a = -
dt
= - (rw cos 0)
dt
= -rw sm () -
dt
= -rw2 sin 0.

But r sin O = y; consequently, the acceleration is

a= - w2 Y,

and if w is replaced by 2 1r f, as before, the acceleration becomes

a = -4 1r2 f2 y. (7.7)
In deriving the expressions for displacement, velocity, and acceleration in
harmonic motion, a vertical vibration was used for illustration. The final
equations, however, involve only the frequency, displacement, and amplitude,
and hence apply to vibrations along any direction. It should be noted that
when y is positive, a is negative, and vice versa, showing that the acceleration
is always directed oppositely to the displacement.
7-12. Restoring Force and Frequency of Vibration. A body describing
harmonic motion is continually being accelerated toward its midposition,
and, since an unbalanced force is necessary to produce acceleration, it follows
that such a body must be acted upon continually by an unbalanced restoring
force. This force can be calculated in absolute units from Eqs. 4.1 and 7.7,
thus: F = ma = m X (-4 1r2 f2 y), from which the ratio of the force to the
displacement is found to be
F
-4 1r2 f2 m, (7.8)
y

where m represents the mass of the vibrating body. Since 4 1r2 in this equa-
tion is a numerical constant, and f and mare fixed for a given body vibrating
under particular conditions, the expressions for acceleration and force can be
written: a a: (-y) and Fa: (-y), which show that both acceleration and force
are proportional to the displacement of the body. These facts conform to the
definition in § 7-7 and show that the projection of uniform circular motion
upon a diameter of the circle is harmonic motion.
The frequency of vibration can be expressed in absolute units by rewriting
Eq. 7.8 as follows:

1 t-F (7.9)
f = 2 7r '\lym.
Similarly, in gravitational units,

f =2-1r
-y x JL
1 ~-F
W'
(7.10)

where W is the weight of the vibrating body, and g is the acceleration due to
gravity. In order to evaluate the ratio -F/y (the magnitude of which was
referred to in § 6-1 as the "spring constant"), the procedure is to select any
§ 7-13 THE SIMPLE PENDULUM 129

value for y and, from the conditions of the problem, to find the corresponding
value for F. It is often simplest to let y = r, the full amplitude, and to
determine the value of F for this displacement. Since the restoring force F
and the displacement y are always opposite in direction, the ratio -F/y,
although it looks negative, is inherently positive.
As an illustrative problem, suppose that a 15-lb block, which originally hangs at
the bottom of a long helical spring, is pulled down 1 ft, and that it takes a force of
5 lb to hold it at this point. Neglect the weight of the spring and compute the fre-
quency with which the block will vibrate when released. Determine also its velocity
and acceleration at an instant when it is 3 in. above its original position.
The block upon release sweeps upward through its original position and vibrates
above and below it with an amplitude r = 1 ft, which will remain the same if friction
is negligible. Take some direction, say downward, as positive. Since a force of 5 lb
is needed to produce a displacement of 1 ft, the restoring force is 5 lb upward when
the displacement is 1 ft downward; that is, F = -5 lb when y = +1 ft. Also, the
weight of the vibrating body is W = 15 lb. Therefore, by Eq. 7.10, the frequency is
found to be

39~
5 lb "" sec 2 _ 0 59 "b/
1 ft X 15 lb - · ""' vi sec.

At an instant when the block is 3 in. above its equilibrium position, y = - 3 in.
- t ft, and the velocity, from Eq. 7.6, is
-- ft
v = 21T'fVr2 - y2 = 271' x 0.52
- V ( l ft)
sec
2 - (-tft) 2 = 3.16-·
sec

Similarly, the acceleration of the block at this instant is found from

Eq. 7.7 as follows:

-4 71'2 f2 y = -4 71'2 ( 0.52) X ( --1 ft) = 2.67 ·-,

ft ·
2
a = -
sec 4 sec 2
7-13. The Simple Pendulum. The simple pendulum con-
sists of a concentrated mass supported by a cord of negligible
weight. When displaced from its midposition and released,
it swings to and fro. For displacements which are small as
B
compared with the length of the cord, the vibrations closely
approximate harmonic motion, and the period can be found
from the foregoing expressions.
In Fig. 13 a cord of length l is supported at Sand carries mg
at its lower end a bob of weight mg. When the bob is moved Frn. 7-13. Dia-
from A to B, the cord is displaced through an angle (3. Two gram used in
forces act upon the bob: its weight mg and the pull P of the finding the pe-
riod of a simple
cord; their resultant Fis tangent to the arc AB. It follows
pendulum.
from the force diagram that F /mg = sin (3. For small values
of (3, sin (3 may be taken as equal to tan (3; also, the displacement may be
taken as the horizontal line x instead of the arc AB. '.,Vith these approxima-
130 MECHANICS Chap. 7

tions, the foregoing equation can be written

F . x
- = sm /3 = tan /3 = -.
mg l

When this expression is rearranged and a minus sign introduced to show that
F and x are oppositely directed, it becomes F/x = -mg/l. The resemblance
of this equation to Eq. 7.8, namely, F /y = -4 1r2 .f2 m, shows that the motion
is harmonic, because the restoring force is proportional to the displacement
x or y and opposite in direction. The last two equations may, therefore, be
set equal to each other, whence

mg - -4
l -
7r2 f2 m.

It follows that the frequency of vibration is .f = 2\ ~ , and that the period

of the pendulum is
1'=21r /1.
\ g
(7.11)

7-14. Angular Harmonic Motion; the Torsion Pendulum. A torsion pen-

dulum consists of a mass supported by a wire. w·hen the mass is twisted
about the wire as an axis and released,
angular vibrations are set up which are
similar to the linear vibrations in har-
monic motion. In the torsion pendulum
Wire shown in Fig. 14, the forces FF, acting at
a distance R from the center, produce a
torque. The consequent twisting of the
wire sets up within it an oppositely di-
Disk
rected restoring torque which is propor-
tional to the angular displacement.
F Hence, by analogy with Eq. 7.8 for linear
p harmonic motion,

FIG. 7-14. Forces acting on a torsion

(7.12)
pendulum.
wherein 3 represents the restoring torque,
ct, is the corresponding angular displacement, f is the frequency, and I is
the moment of inertia of the vibrating body about the axis of rotation.
When rearranged, the expression gives the frequency of angular vibration as

f = 2\ ~ Jlct,, and therefore the period, the reciprocal off, is

T = 2 1r ~ ~/1>. (7.13)
§ 7-15 TIIE PHYSICAL PENDULUJ11 131

The ratio of the restoring torque 3 to the corresponding twist </J is a constant
determined by the stiffness of the suspension wire and is essentially negative,
the restoring torque being clockwise when the twist is counterclockwise.
This ratio is commonly designated by - r (tau). Hence, the period of vibra-
tion is

(7.14)

an expression which is analogous to Eq. 7.11 for the rate of linear vibration of
the simple pendulum.
The torsion principle can be used to determine the moment of inertia of a
body; the body is supported as a torsion pendulum and its moment of inertia
is computed from measurements of its period of vibration and of the angle of
twist produced by a measured torque.
7-15. The Physical Pendulum. Any body which vibrates in the manner
of a simple pendulum but in which the mass is distributed and not concen-
trated is called a physical or compound pendulum. A rod suspended at one
end, or a hoop hung on a nail, would vibrate as a physical
pendulum if displaced and released.
A body of mass m vibrating about an axis Sis shown fS
in Fig. 15 at an instant when displaced from its equilib- /1
h 1<1>1
rium position by an angle </J. The weight of the body /~
acts as a downward force mg at the center of gravity G, G I __ JI
IX'
which is distant h from the axis of suspension S. The
force mg produces a restoring torque about point S which
amounts to 3 = -mgx, and, since x = h tan </J, it follows
that 3 = -mgh tan </J. If the angular displacement </J is
mg
small, then tan </J may be taken equal to </>, whence
3 = -mgh <J>. The restoring torque is, therefore, pro- Fm. 7-15. Dia-
portional to the angular displacement; it is also opposite gram used in find-
ing the period of a
in direction. The body consequently undergoes angular
physical pendu-
harmonic vibration like that of the torsion pendulum in lum.
Fig. 14, except that the vibrations occur in a vertical
plane instead of a horizontal plane. The period is found from Eq. 7.13, in
which -3/</J = mgh ¢/</J = mgh; hence

T=21r /I, (7.15)

\Jmgh
where I is the moment of inertia of the pendulum about a transverse axis at
the point of suspension, and h is the distance between the axis of suspension
and the center of gravity.

Suppose it is desired to find the time of vibration of a meter stick pivoted at one end,
as pictured in part I of Fig. 16.
132 MECHANICS Chap. 7

The length of the stick is L, the distance between the axis of suspension S and the
center of gravity G is h = L/2, and the moment of inertia of the stick about an axis
at Sis I = ! mL2, from § 5-4; whence, the period is

ImL2/3 i2L -2-X-100

-- cm- = 1.64 sec.
T = 2 1r \j mgL/Z = 2 1r "\J3g = 2 1r 3 X 980 cm/sec 2

As another example, find the time of vibration of a hoop 1 ft in diameter about an

axis 8 at its circumference, as shown in part II of Fig. 16.

I n
Fm. 7-16. Examples of the physical pendulum: I, A vibrating rod; II, a vibrating hoop.

The moment of inertia of the hoop about a transverse axis at G is I a = mr2 , where
m is its mass and r its radius, and hence about a parallel axis at S its moment of
inertia is
I = Ia +mh2 = mr2 +
mr2 = 2 mr2,

where the distance h hehveen S and G is replaced by r. Therefore, the period of

vibration is
/2 mr2 /2 r 2 X ! ft
T = 2 7r \I mgr = 2 7r \lg = 2 7r 32 ft/sec2 = 1.11 sec.

The time of vibration of the pendulum is seen to be determined by its

dimensions and by the acceleration due to gravity g at the place where the
pendulum is located. By using a pendulum of known dimensions and by
measuring its period of vibration, the value of g can be determined with
accuracy. As the vibration of a pendulum continues, the amplitude gets
smaller and smaller but the period remains practically unchanged. It is this
property that makes the pendulum of value in controlling· the escapements
of timekeepers.

PROBLEMS
1. A hard-drawn copper wire that has a diameter of 0.135 in. hangs vertically with
a load of 300 lb applied to its lower end. The material has an elastic limit of 28,000
lh/in. 2 and a stretch modulus of 15.5 X 106 lb/in. 2 Will the wire return to its original
length when the load is removed? What elongation takes place per foot of length when
the load is applied?
2. A rod of Mone] metal i in. in diameter is subjected to a load of 1000 lb in a
 PROBLEMS 133

testing machine. An 8-in. length of the rod is observed to elongate 0.0075 in. Com-
pute the stress, the strain, and Young's modulus of elasticity for the specimen.
3. Fibers of spun glass can withstand extremely large stresses; a fiber 0.00035 in.
in diameter was found to have a breaking strength of 0.385 oz. What is the breaking
stress of this fiber?
4. Rubber intended for use under tension has a stretch modulus of 240 lb/in. 2 for
moderate elongations. A sheet of this material i- in. thick, 24 in. wide, and 30 in. long
is stretched; what force will increase the length to 31 in.?
5. A small platform is supported from the ceiling by four steel rods, one at each
corner, each rod being 1 in. in diameter and 4 ft long. How much will each rod
elongate when a machine weighing 1 ton is mounted centrally on the platform?
6. A drill used in boring an oil well is made of sections of steel pipe having an
outside diameter of 4 in. and a wall thickness of ! in. If the drill is 2 mi long and gets
stuck at the bottom, how much will it elongate when a pull of 50 tons is exerted upward
upon it?
7. A vertical steel column 14 ft high supports a load of 80 tons. Take Young's
modulus for steel as 30 X 106 lb/in. 2 and compute (a) the cross-sectional area of the
column for an allowable stress of 16,000 lb/in. 2 , and (b) the decrease in length that
the load produces.
8. At a point 6 mi below the surface of the ocean the pressure is about 14,000 lb/in. 2
Assume the bulk modulus of sea water to be the same as for pure water and determine
how much a cubic foot of water will contract when the pressure on it is increased by
this amount.
9. A cubical block of aluminum measuring 2 cm along each edge is subjected to
shearing forces of 5 X 108 dynes on its opposite faces. Compute the angle of shear of
the block.
10. Theory shows that for an isotropic solid, the bulk modulus B, shear modulus E,
and Young's modulus Y are related by the expression B = E Y / (9 E - 3 Y). Com-
pute the value of B for steel from the other values listed in Table I.
11. A 1000-gm steel ball hangs at the bottom of a vertical cord 20 cm long; in this
position it touches the side of a steel block, also of 1000-gm mass, which rests on a
frictionless horizontal surface. The ball is now displaced until the cord stands at
60° with the vertical and then released, whereupon it s,vings as a pendulum and strikes
squarely against the block. Assume elastic impact and compute the velocity (a) of
the ball just before impact, (b) of the ball just after impact, and (c) of the block just
after impact.
12. A 2-kg steel block sliding along a level, frictionless surface at 50 cm/sec over-
takes and collides squarely with an 8-kg steel block that initially is sliding at 20 cm/sec
in the same direction. Consider the collision to be elastic and find the velocity of
each block after impact.
13. A 500-gm mass suspended from a coiled spring is pulled down 2 cm and released,
whereupon it is found to vibrate with a frequency of 3 vib/sec. With what speed does
it move through the midposition of its path?
14. Determine the acceleration of the vibrating mass in the preceding problem at
an instant (a) 1 sec after it is released, and (b) 1.1 sec after it is released.
15. How much additional mass placed on the spring of Prob. 13 would reduce the
frequency to 2 vib/sec?
16. Determine the spring constant of the spring described in Prob. 13.
17. A machine that weighs 900 lb is evenly supported at the corners upon four
identical compression springs. If the machine when so supported has a natural fre-
quency of 7 vib/sec, what is the spring constant of each supporting spring?
18. An instructor wishes to demonstrate the laws of vibratory motion with a
helical spring that elongates 6 cm when subjected to a tension of 3 X \0 5 dynes. What
134 MECHANICS Chap. 7

mass should be attached to the bottom of the spring in order that when pulled down
and released the mass will vibrate exactly twice per second?
19. A slender wooden rod is fixed at one end and extends horizontally. A load of
400 gm tied to the free end of the rod causes that end to sag 2.8 cm. If this load is
set into up-and-down vibration, with what frequency will it oscillate?
20. A helical spring stret.ches 2 in. when a 5-lb weight is suspended from it. What
weight hung on this spring would vibrate four times a second when pulled down and
released?
21. A mass of i slug is hung on a vertical coiled spring and extends it 10 in. The
mass is then decreased to i slug and set into up-and-down vibration with an amplitude
of 6 in. Calculate. (a) the frequency of vibration and (b) the speed of the vibrating
body when it is 1 in. from the top of its path.
22. A "seconds" pendulum is one that sweeps from one side to the other in 1 sec.
(a) What is the period of a seconds pendulum? (b) What is the length of a simple
seconds pendulum?
23. In a test to determine the moment of inertia of an airplane propeller, a torsion
pendulum (Fig. 14) is used. When the wire supports a solid disk having a mass of
0.9 slug and a radius of 8 in., its period of torsional vibration is 1.6 sec. When the
disk is removed and the propeller substituted for it the period of vibration is 4.5 sec.
Compute the moment of inertia of the propeller.
24. A hole is bored through a yardstick at the 12-in. mark and the stick is set into
vibration about a horizontal axis at this point. Compute the period of vibration.
 8
LIQUIDS AT REST
The study of Mechanics up to this point has been confined to solids. It
will now be extended to fluids-that is, to substances which flow; the term
"fluids" includes both liquids and gases. Fluids at rest are taken up first-
liquids in the present chapter and gases in the next-after which fluids in
motion are considered. In some respects the behavior of liquids at rest
appears fantastic, until the basic principles are clearly understood.
8-1. The Liquid Phase. Most substances can be classified readily as to
phase-that is, solid, liquid, or gaseous, but the lines dividing these phases
cannot always be sharply drawn. Tar, for example, if struck with a hammer,
will fracture like a solid, and yet if a can containing some tar is left tilted to
one side, the tar will flow very slowly, like a liquid. Transitions frequently
occur from one phase to another, many solids becoming liquids when suffi-
ciently heated, and liquids becoming solids when sufficiently cooled. Again,
gases can be liquefied by cooling and compressing them sufficiently, and
liquids can be converted to gases by boiling or simply by evaporation.
It is generally agreed that matter is composed of molecules that are in
more or less violent agitation, depending upon the temperature. In a solid
the molecules are generally bound closer together than in a liquid, and their
motion is more restricted, consisting of vibrations over narrow ranges. In a
gas the molecules are relatively far apart and move with comparative free-
dom throughout the entire confining space. The liquid phase is intermedi-
ate between the other two. Upon heating a solid, say a metal, the molecules
become more violently agitated and less closely bound together; if the process
is continued, the metal melts and becomes a liquid.
Liquids at rest, if not completely confined, present a free surface, which is
horizontal except at the edges, § 8-11. Liquids possess elasticity of volume,
but not of length or of shape; they conform to the shape of any container.
An ideal liquid is, by definition, one on which a shearing force cannot be
exerted. A stream of such liquid would experience no frictional drag from
an object floating along its surface. Actual liquids deviate somewhat from
this definition and do withstand small shearing forces. This effect is neglected
for the present, but is considered in Chap. 10.
8-2. Density and Specific Gravity. The behavior of a liquid is, to a large
extent; influenced by its, density. The density of a substance has already
135
13f\ MBCH ANIC8 Chap. 8

been defined in § 4-2 as its mass per unit vohtme; thus,

d. m
= -. (8.1)
v
In metric units, density is ordinarily expressed in grams per cubic centimeter,
but in the mks system it would be stated in kilograms per cubic meter. The
density of water is 1 gm per cm3 , slight variations due to temperature being
neglected.
For engineering purposes, "weight density" is used more commonly than
"mass density," as it might be called. The weight density of a substance is
its weight per unit volume, or
w
w=-, (8.2)
v
and is usually expressed in pounds per cubic foot. For water, the weight
density is 62.4 lb/ft 3 •
The specific gravity of a substance is the ratio of the density of the substance
to that of water, reference being either to mass density or to weight density.
Specific gravity is thus a pure number, and tells how many times a substance
is as "heavy" (dense) as water. If the specific gravity of a substance is 5,
then 1 cm3 of the substance has a mass of 5 X 1 = 5 gm; 1 ft 3 of this substance
weighs 5 X 62.4 lb. It will be noted that in the cgs system of units, the spe-
cific gravity of a substance has the same numerical value as its density, since
the density of water is 1 gm/cm 3 • Density and specific gravity do not de-
pend upon the size of the sample under test, but only upon the substance of
which it is composed. The density of a liquid does not change appreciably

TABLE I. SPECIFIC GRAVITIES

Liquids Metals

Benzene ............... . 0.89 ( 0°C) Aluminum, hard drawn ...... . 2.70

Gasoline ............... . 0.66-0.69 Brass ..................... . 8.25- 8.70
Glycerine .............. . 1.26 ( 0°C) Copper .................... . 8.80- 8.95
Mercury ............... . 13.596 ( 0°C) Iron, gray cast ............. . 7.03- 7.13
Oils, lubricating .. . 0.90-0.93 (20°C) Lead ...................... . 11.34-11.36
Turpentine ............. . 0.873 (l6°C) Platinum .................. . 21.50
Water, pure ............ . 1.000 ( 4°C) Silver ..................... . 10.4 -10.5
\Yater, sea ............. . 1.025 (15°C) Steel ...................... . 7.82- 7.85
Uranium .................. . 18.7
Woods
Miscellaneous
Balsa wood ..... , .............. 0.11-0.13
Cedar. . . . . . . . . . . . . . . . . . . . . . . . . 0.49-0.57 Diamond. . . . . . . . . . . . . . . . . . . . . 3.5 -3.6
Cork. . . . . . . . . . . . . . . . . . . . . . . . . 0.22-0.26 Glass, common. . . . . . . . . . . . . . . . 2.4 -2.8
Lignum-vitae .................. 1.17-1.33 Ice ........................... 0.88-0.91
Oak .......................... 0.60-0.90 Kapok (in pillows) ............. 0.05-0.10
Pine. . . . . . . . . . . . . . . . . . . . . . . . . . 0.35-0.85 Masonry. . . . . . . . . . . . . . . . . . . . . 1.85-2.3
§ g..3 LIQUID PRESSURH 137

with pressure, since liquids cannot be compressed very much. Values of the
specific gravity of some substances are given in Table I.
8-3. Liquid Pressure. A liquid exerts a force against any surface with
which it is in contact. The force exerted per unit area is an important con-
cept, especially in Hydraulics, and is called pressure.
The idea of pressure at a point is often found helpful in
dealing with liquids. This term presupposes a small
plane surface at the point in question and is the limiting
value of the ratio of the force t:.F exerted upon this sur-
face to the area t:.A of the surface, as the surface is taken
smaller and smaller, approaching a point. That is, the
pressure at a point is
Frn. 8-1. Diagram
. . 1)..F dF
p = L 1m1t- - - (8.3) used in evaluating
A->O t:.A - dA. pressure due to a
liquid.
To apply this expression to a particular point, a horizon-
tal surface of area t:.A is indicated at point P in Fig. 1. In this position the
surface supports the column of liquid directly above it. This column of
height h has a weight t:.F = hw t:.A, where w is the weight density of the
liquid. Hence, the ratio t:.F /AA, or in the limit dF / dA, is equal to the
product hw; consequently, the liquid pressure at the point is
p = hw. (8.4)
In British gravitational units, the pressure in pounds per square foot is the
product of the depth in feet and the weight density of the liquid in pounds per
cubic foot. Thus, the liquid pressure at a point 100 ft below the surface of
water is 100 ft X 62.4 lb/ft 3 = 6240 lb/ft 2 •
In metric units, commonly used with the absolute system, the weight of
the liquid per unit volume can be expressed more conveniently as mg/V or
dg, where d = m/V is the density (or mass density), by Eq. 8.1. Hence, the
pressure at a point can be stated as
p = hdg. (8.5)
The pressure in dynes per square centimeter is the product of the depth in
centimeters, the density of the liquid in grams per cubic centimeter, and the
acceleration due to gravity, 980 cm/sec 2 • The liquid pressure at a point
100 cm below the surface of water is 100 cm X (1 gm/cm3) X (980 cm/sec2 )
= 9.8 X 104 dynes}cm 2 •
The pressure at a point due to a liquid is seen from these expressions to be
determined by the depth of the point below the liquid surface and the weight
density of the liquid. This pressure is not affected by the depth to which the
liquid may extend below the point in question, nor by the size or shape of
the body of liquid in which the point is located.
The foregoing expressions give the pressure due to the liquid only; if the
138 MECHANICS Chap. 8

total or absolute pressure is desired, the atmospheric or other pressure on

the surface of the liquid should be added to the value found. Atmospheric
pressure is discussed in Chap. 9; its value at sea level is approximately 14. 7
lb/in. 2, or 1.013 X 106 dynes/cm 2 •
8-4. Direction of Liquid Force. It can be shown that the force exerted by
a liquid at rest is normal to any surface with which the liquid is in contact. To
prove this statement, the force will be assumed to act in some other direction,
as in part I of Fig. 2, where F, the force exerted by the liquid on the wall at P,
is (incorrectly) represented as inclined to the normal. Under these circum-
stances F could be resolved into two components N and T, respectively
normal and tangent to the wall. If the tangential component T could exist,

Frn. 8-2. The foree due to liquid pressure is normal to the surface at all points.

it would tend to move the wall in the direction shown, and hence the wall
would exert a tangential force on the adjacent layer of liquid. But such a
result is contrary to the assumption that a shearing stress cannot be exerted
on a liquid; hence, the component T cannot exist, and the force F must be
normal to the surface. An experimental proof of the statement is indicated
in part II of the figure, where the liquid is shown emerging normally through
small openings in various surfaces with which it is in contact.
It can also be shown that the liquid pressure at a point is independent of
direction. This statement can be verified by imagining a tiny wedge, of some
material having the same density as the liquid and submerged within it, to
stand at rest in the position shown in Fig. 3. The forces A, B, and C act
normally upon its faces, which have areas, respectively, of a, b, and c. The
force C can be resolved into a horizontal component C sin a and a vertical
component C cos a, where a is the angle of the wedge. For equilibrium,
A = C sin a and B = C cos a. But the areas of the wedge surfaces are simi-
A C sin a
larly related, for a = c sin a and b = c cos a. Then - = . and
a c sm a
B C cos a
-b = ; therefore, the pressures A/a, B/b, and C/c are equal. Next,
c cos a
the wedge may be imagined smaller and smaller, eventually dwindling to a
point; then the three forces will pass through a common point, thus completing
the requirements for equilibrium. It is apparent that the same results would
§ 8-5 FORCE AND TORQUE DUE TO LIQUID 139

be obtained whatever the value of the wedge angle a; consequently, the

pressure at the point is independent of direction.

Fw. 8-3. Diagram used in FIG. 8-4. The reirnltant force due to the
showing that liquid presimre is liquid acts at the center of pressure.
independent of direction.

8-5. Force and Torque Due to Liquid. The force due to liquid pressure on
any plane submerged surface can be calculated from the basic definition of
pressure, p = dF/dA, as given in Eq. 8.3. From this expression it follows
that the force due to the liquid is
F = fpdA. (8.6)
For a horizontal surface, the pressure p is constant at all points, and this
equation reduces to
F = pA, (8.7)
in which Fis the force exerted by the liquid on a horizontal surface of area A.
For a vertical surface, the surface is divided into a series of horizontal
strips so narrow that the pressure can be considered constant all over any
one strip; the force due to the liquid upon such a strip is expressed as dF
= p dA; and the total force on all the strips is found by integration, as indi-
cated by Eq. 8.6. Thus, the verti<'al rectangle in Fig. 4 has an altitude h
and width b; liquid of ,veight density w is assumed to be in contact with one
of its faces, with the top edge of the rectangle in the surface 8 of the liquid.
The rectangle is divided into horizontal strips; the one in the figure is at a
depth y below the liquid surface and has an infinitesimal width dy. The
liquid force on this elementary strip is
dF = p dA = (yw) (b dy); (8.8)

hence, the total force upon the entire rectangle is F = bw l h

y dy, or

F = b ~ h2_ (8.9)
 140 MECHANICS Chap. 8

. express10n can b e wn"tten F = hw hb, m

· wh'1ch th e f actor hw 1s
Th1s 2 2 · t he .
liquid pressure p at the middle of the rectangle, in accordance with Eq. 8.4,
and the product hb is the area A of the rectangle. Hence the resultant force
which the liquid exerts on the rectangle can be stated in exactly the form of
Eq. 8.7, namely:
F = pA,
where A represents the area in contact with the liquid and p represents the
liquid pressure at the middle of that area.
Center of Pressure. The resultant force F in the foregoing derivation is
shown in the figure applied normally to the surface at point CP, called the
center of pressure; the depth e of this point below the surface is to be deter-
mined, by taking moments of the liquid force about any convenient axis.
Thus, the torque of the force F about an axis in the surface S is J = F e;
which, from Eq. 8.9, can be written

= -b W2- h
2
J Xe.

The torque J can also be computed by expressing the torque on one of the
elementary strips and integrating this expression. The force on the elemen-
tary strip in Fig. 4 is given by Eq. 8.8 as dF = bwy dy; consequently, the
torque of this force about the axis S is dJ = dF y = bwy2 dy, and the total
torque is
h
l
J = bw y2 dy = --.
0
bwh 3
3
bwh2 bwh3
The two expressions for J can be equated, thus: - 2 - e = - 3 -, from which

e = -i h. (8.10)

The center of pressure is, therefore, located! of the way from the liquid sur-
face to the bottom of the rectangle and, by symmetry, is also located midway
between the sides of the rectangle. These statements are correct for any
rectangle, vertical or inclined, provided its upper edge is not submerged
within the liquid.
In designing and building tanks, dams, and other structures used to con-
fine liquids, it is necessary to know how much force the liquid will exert and
its point of application, in order that each surface may have the proper
strength, weight, and bracing to withstand the applied load. To evaluate
such a force, the pressure is found, usually by the expression p = hw, and
then the force from the equation F = pA.
As an illustrative problem, consider the rectangular tide gate shown edge-on in
Fig. 5; the gate is hinged at the top, and sea water extends to its top edge on one side.
Suppose the height of the gate to be 6 ft and its width (perpendicular to the page) to
§ 8-G TRANSMISSION OF CHANGE OF PRESSURE 141

be 4 ft, and find the torque due to the water tending to rotate the gate about its top
edge. Take the density of sea water as 64 lb/ft3.
The force exerted by the water increases in proportion to the depth, as indicated by
the arrows in part I of the figure, and the resultant must first be found. Part II shows
the midpoint M of the gate, at a depth of 3 ft below the water surface. The liquid
pressure at this point is 3 ft X 64 lb/ft3 = 192 lb/ft2• This pressure multiplied by

II III
FIG. 8-5. Liquid force on a tide gate.
the area in contact with the water gives the magnitude of the resultant force as
(192 lb/ft2) X 24 ft2 = 4610 lb. In part III of the figure, this force is represented
by F; it is perpendicular to the gate at the center of pressure CP, located a distance
e = 4 ft from the top of the gate. The torque of this force about the top edge is
clockwise and amounts to 4610 lb X 4 ft = 18,440 lb· ft.
An interesting paradox appears in Fig. 6, which shows a vertical cross-
section of three tanks with bases of equal area and filled with water to the
same depth. The force exerted on the base of each tank is the same, show-
ing that the force does not depend on the shape of the tank.

=------ - __=

FIG. 8-6. The liquid exerts the same amount of force on the bottom of each tank.

8·6. Transmission of Change of Pressure; Pascal's Principle. A number

of principles in the mechanics of fluids are associated with the name of the
French philosopher Blaise Pascal (1623-1662). One of these states that an
increase in pressure at any point in a confined liquid results in a like increase in
pressure at every other point in the liquid.
This principle is employed in the hydraulic press as used for compressing
materials into bales, for forming the lead sheathing upon electric cables, and
the like. An elementary diagram of this device is shown in Fig. 7, which
represents a confined body of liquid connecting two cylinders of areas a and A,
respectively, each fitted with a piston. When a force f is applied to the small
piston the pressure beneath it is increased by an amount f / a. An equal
142 MECHANICS Chap. s
pressure increase will occur at the large piston, and this piston will therefore
exert a force F = 1
a
A. It follows that F / A = f / a. In this machine, as is
customary, a large force, exerted through a small distance, is obtained by
exerting a small force through a large distance.
The mechanical advantage of the hydraulic press can be computed as
described in § 6-13; the ideal advantage is A/a, and the actual advantage
is smaller than this because of the friction of the pistons in the cylinders.
Hydraulic brakes on automobiles operate by liquid pressure. A master
cylinder near the foot-brake pedal is connected to cylinders at each of the
wheel brakes, and the entire system is filled with nonfreezing liquid. When
foot pressure is applied, the piston of the master cylinder is forced inward,
increasing the pressure on the liquid at that point; this pressure increase is
transmitted to the wheel cylinders and their pistons are forced outward,
tightening the brake bands against the drums.

f F

Frn. 8-7. Schematic diagram of the Frn. 8-8. Diagram

hydraulic press. used in proving Archi-
medes' Principle.

8-7. Buoyancy; Archimedes' Principle. It is well known that a stone can

be lifted more easily in water than in air, that many objects float upon water,
and that a swimmer cannot sink in the Great Salt Lake. These illustrations
show that a liquid exerts an upward force upon a body placed in it. Archi-
medes (c 287-212 B.c.), Greek mathematician and inventor, determined the
amount of this buoyancy. The principle known by his name applies to both
liquids and gases; for a liquid it states that a body submerged entirely or par-
tially in a liquid is buoyed up by a force equal to the weight of the liquid displaced.
Archimedes' Principle can be demonstrated experimentally by weighing a
body in air and again while submerged in a liquid which initially fills the
container to the top. The buoyant force is the apparent loss of weight of
the body and will be found equal to the weight of the liquid that overflows.
Figure 8 will be used in proving Archimedes' Principle theoretically; the
buoyant force acting upon a submerged body will be calculated, then the
weight of the displaced liquid, and it will be shown that the two quantities
are equal. In the figure a block of rectangular cross-section, which has a
height h and top and bottom faces of area a, is shown immersed in a liquid
§ 8-8 MEASUREMENT OF DENSIT}. 143

which is assumed to have a weight density w. On the vertical faces, the

liquid exerts horizontal forces that are balanced on all sides; on the top face
it exerts a downward force of h1wa and on the bottom face an upward force
of h2wa. Since h2 is greater than h1, the liquid exerts a net upward force on
the block which amounts to h2wa - h1wa = hwa. But the volume of the
block, and hence that of the liquid displaced, is ha, and therefore the weight
of the liquid displaced is hwa, and this is identical with the buoyant force due
to the liquid. Bodies of irregular shape may be considered as though they
were made up of a number of blocks as described, and therefore any body
submerged in a liquid is buoyed up by a force equal to the weight of the
liquid displaced.
This principle is employed in the control of submarines. The boat is sub-
merged by admitting sea water into its tanks, thereby increasing its weight.
The boat may be brought to the surface by expelling the water from these
tanks with compressed air or with pumps.
8-8. Measurement of Density. The density of a solid can usually be
found by measuring its mass and its volume separately. The mass is ob-
tained with an equal-arm balance, and the volume is determined either from
the dimensions of the body or by measuring the displacement it produces
when submerged in water. With the mass and volume known, the density
follows from Eq. 8.1.
An indirect method which offers some advantages consists of weighing the
body first in air and again while submerged in water, and using Archimedes'
Principle to calculate the density from the data thus obtained.
Suppose, for example, that a body suspended in air from one end of an equal-arm
balance is held in equilibrium by a mass of 350 gm suspended from the other end;
the weight of the body in air is the same as that of 350 gm. If the body while sub-
merged in water is balanced by 280 gm, the buoyant force is the weight of 350 - 280
= 70 gm. But 1 cm 3 of water has the same weight as 1 gm, and it follows that the
volume of the displaced water, and hence the volume of the body, is 70 cm3. Since
the body has a mass of 350 gm and a volume of 70 cm3, its density is 350 gm/70 cm3,
or 5.0 gm/cm 3•

The density of a liquid can be determined by several methods. In one of

these, a specific gravity bottle of known mass is filled with the liquid and the
combined mass is determined with an equal-arm balance. The volume of the
container, if not known, is determined by another measurement using water,
and the density of the liquid can then be found by a direct proportion. An-
other method for measuring the density of a liquid consists of weighing some
solid body first in air, second while submerged in water, and third while sub-
merged in the liquid under test, and calculating the density from these data
by Archimedes' Principle.
In measuring the density of alcohol by the method just described, suppose that a
body on an equal-arm balance is balanced by 350 gm in air, by 280 gm while sub-
merged in water, and by 295 gm while submerged in alcohol. The displaced alcohol
144 MECHANICS Chap. 8

has a volume of 350 - 280 = 70 cm3, and a weight equal to that of 350 - 295 = 55 gm;
therefore, the mass of the alcohol is 55 gm and its density is 55/70 = 0.79 gm/cm3 •
A third method for finding the density of a liquid is that of balanced col-
umns. In this method, illustrated in Fig. 9, one arm of an inverted U-tube
is dipped into water and the other into the liquid under test; then some of the
air is sucked out from the tube at A, causing the liquids to rise in the tubes.
The heights of the liquid columns hw for water and hx for the test liquid are
then measured. Since the same difference in pressure exists between the
top and bottom of each column, it follows that hwdwg = hxdxg, where dw repre-
sents the density of water and dx that of the liquid under test. From this
equation the value of dx can readily be found.

A -- ~

Frn. 8-9. The balanced-colunm Frn. 8-10. The hydrometer

method for measuring the density for the direct reading of the
of a liquid. specific gravity of a liquid.

The density of a liquid can also be measured with a hydrometer, Fig. 10,
which consists of a hollow glass chamber weighted at the bottom and having
a graduated stem at the top. The hydrometer has a constant mass and,
when floated in a liquid, sinks until it displaces its own weight of liquid (see
following section); the "lighter" the liquid the deeper the hydrometer settles
before it comes to rest. By suitably calibrating the scale of the instrument,
the specific gravity of the liquid can be read directly at the point where the
stem projects through the liquid surface.
8-9. Stability of Floating Bodies. A floating body, as well as a submerged
body, experiences an upward buoyant force equal to the weight of the liquid
displaced, in accordance with Archimedes' Principle. A body floating at
rest, moreover, is in equilibrium, and hence the buoyant force acting upon it
must be equal and opposite to the weight of the body. A body that floats
when placed in a liquid settles until it displaces its own weight of liquid. Thus,
§ 8-10 MOLECULAR MOTION AND MOLECULAR FORCES 145

a ship which displaces 20,000 tons of water also weighs this same amount.
The forces acting upon a ship are indicated in part I of Fig. 11, where W
represents the weight of the ship, acting downward through its center of
gravity G; and B represents the equal buoyant force, which acts upward
through C, the center of gravity of the displaced water. In order for the
ship to be stable, it must tend to right itself when tipped to one side, as
represented in part II of the figure. Although the weight W acts downward
at G, as before, the upward force of buoyancy B acts at the shifted center of
gravity of the displaced water, now at C'. The forces B and W constitute
a couple which tends to restore the ship to an upright position, and the ship
is said to be stable.

G
c

w w
I II
FIG. 8-11. Forces on a ship in stable equilibrium I, when in upright position; and II, when
listed to one side.

The stability of a ship is determined by the position of the metacenter M.

This point is the intersection of two lines: one lies vertically through the cen-
ter of gravity of the displaced water, and the other is a line which extends
vertically through the center of gravity of the ship when in an upright posi-
tion. For stability, the metacenter must be above the center of gravity of
the ship, as in the figure. The metacentric height MG is, nevertheless, made
relatively low, in order that the restoring couple will not be unduly large and
cause the ship to roll too quickly for riding comfort.
8-10. Molecular Motion and Molecular Forces. Experiments show that
the molecules of a body are in a state of continual motion that is entirely
erratic and irregular. The molecules themselves are, of course, too small to
be visible, and indirect methods must be used to study their motion. The
first observations were made in 1827 by the English botanist Robert Brown
(1773-1858), who noted that very fine particles placed in suspension in a
liquid move about in an irregular and lifelike manner. This action he
attributed to uneven bombardment of the particles by the moving molecules
of the liquid, and this hypothesis has been amply borne out by further re-
search. These so-called Brownian movements can be observed with a high-
power microscope.
14G MECHANICS Chap.8

There is also evidence to show that, as two molecules approach each other,
a nearness is reached where their combined potential energy is a minimum.
Work is then needed to move them either nearer together or farther apart,
because of forces which are probably electrical. '\Vhen molecules are ex-
tremely close together, these forces produce a tremendous repulsive effect and
keep the centers of the molecules apart; at greater separations the force
becomes one of attraction. The molecular attractive forces are known to
be very great; for example, they hold a solid body together and enable it not
only to retain its shape but also to support large external loads. The attrac-
tion between molecules of the same substance is called cohesion, whereas the
attraction between molecules of unlike substances is called adhesion.
The attraction between two molecules falls off rapidly as their separation
increases. It is therefore convenient for discussion to imagine a tiny sphere
around each molecule, called its sphere of action, and to regard the molecule
as exerting an attractive force upon other molecules that lie within this
boundary, but not upon those that lie outside of it. This idea of molecular
attraction being limited by a sphere is entirely artificial but is of value in
indicating that the forces of cohesion and adhesion are very small except for
molecules that lie close together. The sphere of action is usually considered
a8 having a radius smaller than a millionth of a centimeter.
8-11. Surface Tension. The theory of molecular attraction explains the
interesting behavior of a liquid surface. In Fig. 12, the line .MN repre:,;ents
a free liquid surface, and A, B, and C
represent molecules of the liquid, each
surrounded by its sphere of action. Mole-
cule A, which is well within the body of
the liquid, is, on the average, attracted
equally in all directions by other mole-
Fm. 8-12. Molecules of a liquid and
their spheres of action. Molecules cules within its sphere of action, and its
near the surface are pulled downward. motion will be unaffected. Molecule B,
which is in the liquid near the surface,
experiences a downward force, since more molecules are in the lower half of
ittl 8phere of action than in the upper half. Similarly, molecule C, ,vhich is
in air just above the free surface, is acted upon by a considerable downward
force. Hence, those molecules which in their motion tend. to rise above the
surface are pulled downward. In consequence, the surface acts like a stretched
membrane, tending to shorten itself as much as possible. Drops of liquid
tend to assume a spherical shape because a sphere has the smallest surface
for a given volume. Figure 13 shows an enlarged view of the splash produced
by a drop of milk falling upon a surface of that liquid and reveals the rebound-
ing droplets just before they become detached spheres.
The surface of a liquid at any point assumes a direction at right angles to
the resultant force acting on the surface molecules. For this reason the edges
of a free liquid surface are usually curved, as in Fig. 14, in which the adhesive
§ 8-11 SURFACE TENSION 147

and cohesive forces acting upon a surface molecule of the liquid are indicated.
The force of adhesion A. exerted by the wall is perpendicular to its surface;
the force of cohesion C, which is due to the surrounding molecules of liquid,
has the general direction shown; and their resultant R is at right angles to
the surface of the liquid. For substances such that the adhesion between

Fm. 8-13. Splash of a drop of milk.

(Photographed in 1/100,000 second by Professor Edgerton
and associates at Massachusetts Institute of Technology).

the liquid and the solid is greater than the cohesion of the liquid-as with
water against glass-the surface has the characteristic shape shown in part I
of the figure. Here liquid molecules near the wall ·which happen to rise
above the free liquid surface are pulled to the wall by the force of adhesion,
causing the liquid to pile up along the edge, and the liquid wets the wall.
For substances such that the adhesion is less than the cohesion-as with

c R C R

TI
Fm. 8-14. Capillary action at edge of liquid surface.

mercury against glass-the liquid surface curves downward, as in part II of

the figure, and the liquid does not wet the wall. The angle of contact at
the edge of a liquid surface is defined as the angle formed by the liquid surface
and a vertical wall at the point of contact, measured within the liquid. The
angle of contact depends upon the substances used; its value is 0° between
water and glass, 90° between water and silver, and 132° between mercury
and glass, when pure liquids and clean surfaces are used.
148 MECHANICS Chap. 8

Any portion of the liquid surface exerts a tension upon adjacent portions
or upon other objects with which it is in contact. This force acts in the plane
of the surface, and its amount per unit of length is known as surface tension.
Surface tension occurs not only at the free surface of a liquid, but also at the
boundaries or interfaces separating two
B B liquids. Its value depends upon the sub-
A A stances and is influenced by the tempera-
ture. By way of illustration, the forces
acting upon a drop of oil floating on hot
FIG. 8-15. Surface-tension forces on water are shown in Fig. 15. The surface
floating oil drop. tension A of the water in contact with air
tends to spread out the drop, whereas the
surface tension B of the oil in contact with air and the surface tension C of
the oil in contact with water tend to make the drop contract. Under the
combined action of these forces, the drop assumes such a shape as to remain
in equilibrium. At lower temperatures surface tension in general increases,
although not equally for all substances. The drop shown in Fig. 15 spreads
out into a thin film at ordinary room temperature because of the relatively
great increase in the surface tension A. Some approximate values of surface
tension, based on measurements at ordinary room temperature, appear in
Table IL

TABLE II. SURFACE TENSIONS OF LIQUIDS

Liquid In Contact with dynes/cm

Benzene. . . . . . . . . . . . air 29
Glycerine. . . . . . . . . . . air 63
Mercury.. . . . . . . . . . air 470
Mercury. . . . . . . . . . . water 392
Olive oil. .. . . . . . . . . . air 35
Olive oil. . . . . . . . . . . . water 19
Water............. air 75

The surface tension of a liquid can be found by measuring the force needed
to pull an inverted U-shaped wire upward through the surface. Part I of
Fig. 16 shows such a wire of horizontal length l; F represents the force in

F F

Cross-section
/ of wire

11
Fm. 8-16. Measuring surface tension of liquid with a U-shapcd frame.
§ 8-12 CURVED SURFACES AND CAPJLLARJ'IT 149

excess of its weight that is needed to pull it through the surface. This force
is required because a film of liquid adheres to the wire and pulls downward
upon it. Since the film has two faces, the force that it exerts on the wire is
2 T l, where T represents the surface tension of the liquid. The forces are
shown in the cross-sectional diagram forming part II of the figure. The force
applied to the wire must be sufficient to balance the downward pull, and
hence F = 2 T l, from which the surface tension is
F (8.11)
T=-·
2l
It can also be shown that work is needed to enlarge the surface of a liquid, a
fact which might be inferred from the tendency of the surface to become as small
as possible. In the experiment just described, the work done in pulling the wire
and film upward a distance dy is F dy, and the additional surface generated
in the two faces of the film in this motion is 2 Zdy. The ratio of the work done
to the surface generated is ;z t, :z'
or which is equal to the surface
tension T of the liquid, from Eq. 8.11. Therefore, surface tension can be
defined either as work per unit of generated area or as force per unit length;
the units erg/cm2 and dyne/cm are equivalent.
For many purposes a low value of surface tension is desirable, because a
liquid with low surface tension can spread out into a thin film. The low
surface tension of oil allows this liquid to penetrate narrow spaces where
lubrication is desired. Tinctures provide better penetration than water solu-
tions because of the low surface tension of alcohol. Certain medicinal prepa-
rations have low surface tension and hence spread out and medicate a large
area. Water, which has a fairly high surface tension, can be supported on a
piece of gauze; the liquid sags down in the openings but does not fall through
them, because the surface tension is great enough to keep the free surfaces
from rupturing. But if some detergent is added to the water, it causes a
marked lowering of surface tension and the liquid pours through the gauze.
Detergents improve the cleaning properties of water partly because they let
the liquid seep into fine pores that otherwise it would not reach.
8-12. Curved Surfaces and Capillarity. Wherever the surface of a liquid
is curved, a difference of pressure exists between the regions inside and out-
side the surface. The spherical shape of a soap bubble is evidence of the
exces8 pressure within it; the value of this pressure can be computed by inves-
tigating the amount of work needed to blow the bubble. Within a bubble of
radius r the pressure exceeds that of the atmosphere, say by an amount p.
The surface of the bubble, including both sides of the film, is S = 8 1rr2. If
the bubble is enlarged, the radius increasing by an amount dr, the increase
in its surface will be dS = (8 1r) 2 r dr = 16 1rr dr. The work needed in this
process can be expressed from the preceding section as T dS or T (16 1rr dr).
The amount of work can also be computed as the product of force and c-IiR-
150 MECHANICS Chap. 8

placement. The pressure p causes a force of magnitude p to act on each unit

area all over the inside of the bubble; as the unit area moves outward a
distance dr, the work done upon it is p dr, and the total work done on the
entire bubble will be p dr (4 1rr2 ) or p (4 1rr2 dr). Therefore,
T (16 1rr dr) = p (4 1rr2 dr),
and from this expression the excess pressure inside the bubble is
4T (8.12)
p=-·
r

It is a little surprising to note from this relation that the smaller the bubble
the greater the pressure inside it. If a little soap bubble and a big one are
placed in communication with each other, the little one will collapse and
make the big one bigger.
The pressure difference that exists between the inside and outside of a
curved liquid surface depends only upon the surface tension T and the
radius r of the surface, and the greater pressure is on the concave side. The
pressure difference is 2 T /r for a sphere with only one surface, like a drop of
liquid or an air bubble in water; it is 4 T/r for a bubble which has two sur-
faces; it is T /r for a cylinder with a single surface; it is 2 T /r for a cylinder
with a double surface.
The pressure difference just described explains the rise of liquid in a small
tube, open at the ends, when lowered vertically into a liquid that wets it.
The liquid inside the tube is initially at the
same level as that outside, but inside the
tube a meniscus forms which is approxi-
A mately spherical and which has its concave
surface on top. Just above this surface
the pressure is that of the atmosphere, and
just below it the pressure must be less.
Such a condition is unstable, and atmos-
h pheric pressure on the surrounding liquid
causes the liquid inside the tube to rise
--=c=---- O until equilibrium is established; this effect
_: : __::_ : _-~_-=-~ ---cc:,:_=_--=-c-e_'=_=_=-"'-~:: is called capillarity.
The equilibrium conditions are repre-
sented in Fig. 17. Liquid of density d and
FIG. 8-17. Rise of liquid in fine-bore surface tension T has risen to a height h in
tube. a capillary tube of radius R. The surface
makes contact with the tube at an angle 8
and forms a meniscus which is assumed to have a spherical surface of radius r.
An evaluation of the absolute pressure at certain points gives the following
values:
§ 8-12 CURVED SURFACES AND CAPILLARITY 151

At A, just above the liquid surface: atmospheric pressure, Pa;

at B, just below the liquid surface: Pa - 2 T /r;
at C, the base of the column: Pa - 2 T /r + hdg;
at D, outside the tube: Pa·
Since the pressures at points C and D are the same, it follows that Pa - 2 T /r
+ hdg = Pa, or h = 2 T/rdg. But R/r = cos 8, and therefore the capillary
nse 1s
2 T cos e (8.13)
h = Rdg '
expressed in terms of the radius R of the tube rather than the radius of curva-
ture of the meniscus.
PROBLEMS
1. A solid cylinder of metal is 5 cm long and has a diameter of 2 cm. If the
cylinder has a mass of 134.5 gm, what is the specific gravity of the metal?
~· A dredge is equipped with a bucket that can lift 10 cubic yards at a time.
Assume the specific gravity of mud to be 1.4 and calculate the weight of mud that the
bucket can lift.
3. At what depth below the surface of fresh water is the absolute pressure 2 atmos-
pheres?
4. Compute the pressure in dynes per square centimeter due to a column of
glycerine 60 cm high.
5. A closed metal drum 2 ft in diameter and 3 ft high stands with its axis vertical,
and from its top a pipe 1 in. in diameter extends 10 ft vertically upward, as shown in
the diagram. When the drum and pipe are full of water, what pres-
sure does the water exert (a) upon the top of the drum, and (b) upon
the bottom of the drum?
6. A pipe 25 ft long extends straight down from the bottom of an
open-top tank 10 ft high. The tank and pipe are full of water and
the bottom of the pipe is closed. What is the liquid pressure at the
bottom of the pipe?
7. A cubical tank measures 4 ft along each edge; it is open at the
top and is filled with water. Calculate (a) the liquid pressure at
the bottom of the tank, and (b) the total force exerted by the water
on the bottom of the tank.
8. An open-top tank has a horizontal base in which there is a
patch measuring 6 X 6 in. The patch can safely withstand a force
of only 300 lb; how deep can the tank be filled with water without
exceeding this limit?
9. Placed alongside the drum and pipe described in Prob. 5 is a
vertical tank 2 ft in diameter and 13 ft high, open at the top. Sup- PROBLEM 5.
pose the drum and pipe, and also the tank, to be full of water and
compute the force exerted by the water (a) on the base of the tank, and (b) on the
base of the drum.
10. What force does the water exert upon one of the vertical side walls of the tank
described in Prob. 7 . Express the result in pounds and in newtons.
11. An open-top tank has a square base measuring 3 ft along each edge. The tank
has vertical sides, and one side cannot withstand a force greater than 1000 lb. To
what depth can the tank be filled with water without exceeding this limit?
152 MECHANICS Chap. 8

12. A square tank is open at the top; its base measures 8 X 8 in. and its height is 6 in.
(a) If the base is horizontal and the tank is full of water, what force does the water
exert upon the base? (b) If the tank is tilted about one edge until half of the water
flows out, what force does the water exert upon the base in the inclined position?
13. A V-shaped trough used for watering animals is formed of two boards each
3 ft long and 1 ft wide; the boards are fastened together at right angles to each other
and the ends of the trough are suitably closed. The trough ir. laid on level ground
and each board makes an angle of 45° with the horizontal. When the trough is filled
with water, what force does the water exert on each board?
14. The diagram represents part of a long wall that has a rectangular cross-section
6 ft high and 4 ft wide. Water stands in contact with one face of the wall up to the
top edge. Consider a 10-ft length of the wall and compute (a) the force that the
water exerts upon it and (b) the torque of this force tending to overturn the wall.

PROBLEM 14. PROBLEM 16.

15. A rectangular tide gate 3 ft wide by 6 ft high is hinged along its top edge.
When sea water stands at a level 3 ft below the hinge line on one side of the gate and
at a level 4 ft below the hinge line on the other side, what is the net torque due to
the water tending to turn the gate on its hinges?
16. The diagram represents a triangle immersed vertically in a liquid with its top
edge in the liquid surface. Determine by calcqlus how deep in the liquid the center
of pressure of the triangle is located.
17. A dam 10 ft long and 14 ft high has a triangular cross-section. Water in con-
tact with the vertical face of the dam extends to a level 2 ft below the top. Compute
(a) the force exerted by the water on the dam, and (b) the moment of this force
tending to overturn the dam.
18. A dam 8 ft long has a triangular cross-section with a 4!-ft base, and its sloping
surface makes an angle of 60° with the horizontal. Water extends to the top of the
dam, in contact with the sloping face. Compute (a) the force exerted by the water
on the dam, and (b) the moment of this force tending to overturn the dam.
19. The two pistons of a hydraulic press have radii of 3 and 12 in., respectively.
What force applied to the small piston will make the large one exert a force of 10,000 lb,
if friction is neglected? What is the ideal mechanical advantage of this machine?
20. A metal sample of irregular shape is balanced by 161.0 gm in air and by 146.8 gm
when immersed in water. What is the volume of the sample?
21. An alloy specimen is balanced by 123.0 gm in air, by 108.0 gm when immersed
in water, and by 111.1 gm when immersed in alcohol. From these data determine
(a) the density of the alloy and (b) the density of alcohol. '
22. A metal block has a volume of 1 cm 3 ; one-fourth of its volume is silver (specific
gravity 10.5) and the rest is gold (specific gravity 19.1). Find how many grams will
balance the block (a) in air, and (b) when immersed in water.
23. An alloy is formed of gold (specific gravity 19.1) and copper (specific gravity
 PROBLEMS 153

8.9). A specimen of this alloy is balanced by 132.0 gm in air and by 121.8 gm when
immersed in water. Determine the percentage, by weight, of gold in the alloy.
24. Some water and oil placed in a U-tube come to equilibrium in the positions
shown in the diagram. From the dimensions given, determine the specific gravity
of the oil.
25. In a test using the apparatus of Fig. 9, a column of water 4.0 cm high was found
to balance h column of carbon tetrachloride 2.5 cm high. From these data determine
the specific gravity of carbon tetrachloride. In the test described, by how much
was the air pressure in the connecting tube reduced below atmospheric pressure?
26. A ball 5 in. in diameter floats in fresh water with exactly one-half of its volume
immersed. Compute the weight of the ball.
27. What is the specific gravity of a sample of wood if a cubical block of it floats
in water with one-fifth of its volume projecting above the liquid surface?
28. Some lead shot are poured into a long glass tube, which floats upright when
immersed in water. If the tube has an external diameter of 2 in. and a length of 3 ft,
and if its weight with contents is 3t lb, how far does the tube project above the water
surface?

Oil Water

PROBLEM 24. PROBLEM 33.

29. A hydrometer floating in a liquid as in Fig. 10 is pushed down slightly and re-
leased. Show that the restoring force then acting upon it is proportional to the dis-
placement and that the instrument will bob up and down with harmonic motion.
30. To measure the surface tension of a liquid a fine wire formed into a loop of
1-cm radius is placed horizontally within it and then pulled upward through the liquid
surface. In testing a liquid advertised as "ST 37" this action requires a force of
465 dynes in addition to the weight of the wire. Determine the surface tension of the
liquid.
31. Compute the height to which water at room temperature will rise by capillary
action in a glass tube 1.5 mm in diameter.
32. Two parallel glass plates of indefinite length are lowered into a liquid and held
so that the plates are vertical and only a slight distance apart. Show that the capillary
rise of liquid between the plates is
h = 2 T cos 81
sdg
where Tis the surface tension of the liquid, 8 the angle of contact, s the separation of
the plates, d the density of the liquid, and g the acceleration due to gravity.
33. The diagram shows a small glass tube fitted with a piston at one end. A bubble
is formed across the other end and the pressure within it is measured by a U-tube
manometer containing water. The bubble has a diameter of 2.5 cm and the water
level in the manometer is 2 mm higher in one arm than in the other. From these data
find an approximate value of the surface tension of the liquid of which the bubble is
formed.
 9
GASES
This chapter continues the study of fluids at rest, but deals with gases in-
stead of liquids. These two phases of matter have much in common, and
their chief difference is that gases are highly compressible, whereas liquids
cannot be compressed except to a slight extent. The words "at rest" applied
to a gas imply an absence of flow and not a standstill of molecules. It is
supposed that the molecules of a gas, like those of a liquid, move about inces-
santly but in a random fashion, and that their speed in this aimless motion
is a property of the temperature. The thermal behavior of gases is taken
up in Chap. 13.
9-1. The Gaseous Phase. Gases, according to the accepted theory, are
c-omposed of molecules which are comparatively far apart, and which, in
their unceasing motion, penetrate the entire space to which they are admitted.
Gases differ from liquids in two respects: first, gases are very compressible;
and second, they completely fill any closed vessel in which they are present.
In most other respects, however, gases resemble liquids: both exert pressure
upon surfaces with which they are in contact, both exert buoyant forces in
accordance with Archimedes' Principle, both adapt themselves to the shape
of the containing vessel, and both, having no elasticity of shape, are unable
to exert shearing forces, except those due to friction, or viscosity.
The term vapor applies to a gas which can be liquefied by pressure alone.
At ordinary temperatures, steam and carbon dioxide are called vapors, but
air, hydrogen, and nitrogen are called gases.
9-2. Kinetic Theory of Gases. Because of their simple structure, gases
are suited to mathematical study; and, from such investigations, a theory
called the kinetic theory of gases has been developed. In accordance with
this theory, the molecules of a gas are all alike and behave like tiny elastic
spheres; they are relatively far apart and, in their random motion, are con-
tinually striking against one another and against the walls of the container.
It is shown later that energy is needed to raise the temperature of a sub-
stance; in a gas, this energy appears as an increase of molecular motion, and
the kinetic theory assumes that the temperature of a gas is directly proportional
to the mean kinetic energy of the molecules. At a given instant, some molecules
are moving one way and some another, some are traveling rapidly and some
slowly, and some may even be momentarily at rest, but the average kinetic
154
§ 9-2 KINETIC THEORY OF GASES 155

energy of all of them eorresponds to the temperature of the gas. It is desired

first to express an intermediate speed which, if possessed by all the molecules,
would correspond to the same t2mperature of the gas. This intermediate
speed v must be such as to impart the same kinetic energy to N molecules,
each of mass m, as that due to their individual speeds, v1, v2, · · · VN, Hence,
from Eq. 6.7,

from which

This equation shows that the multiplicity of individual speeds can be replaced
by a single speed, which is found by squaring the individual speeds, taking
the mean of these squares, and then extracting the square root. The result
is called the root-mean-square, or rms, speed.
Therms speed of the molecules of a gas can be found in terms of the pressure
and density of the gas by summing up the impulses of the molecules against
one of the confining walls; this summation per unit time
and unit area is equal to the pressure, and a rearrange-
ment of terms yields the desired expression. In Fig. 1
is shovm a cubical box of volume V = l 3 which is as-
sumed to contain N molecules of gas, each of mass m
and each moving with therms speed v. The motion of
the molecules within the box can be studied most simply
by supposing one-third of the molecules to be moving
perpendicularly between each pair of opposite faces. FIG. 9-1. Diagram
Such a group of molecules moving bet,veen any face and used in studying the
speed of gas mole-
the opposite one would encounter many collisions along cules.
the way. Upon leaving face A, the first collisions may
be assumed to occur after traveling a small uniform distance x. The number
of molecules in a zone of this thickness would be f N, one-third of which
would strike face A with a velocity v and rebound with a velocity -v. Each
of these molecules would thus undergo a change of momentum 2 mv, which,
according to Eq. 4.6, equals the impulse imparted to the wall by collision.
When all the molecules striking the face are included, the impulse per impact is
xN
Ft= IT 2 mv.

The molecules travel a distance x before collision and travel back to face A,
a total distance of 2 x, in a time 2 x/v, and thus strike the face v/(2 x) times
per unit of time. The total impulse exerted per unit time is equal to the
force, or
 156 MECIIANICS Ch:.tp. 0

xN v Nmv 2
F = -2mv- = - - ,
3l 2x 3l
and the pressure exerted on the face is
F Nmv 2 Nmv 2
p = Z2 = 3l3 = 311"
But the density of the gas is d = Nm/V, and consequently therms speed of
the molecules of a gas under an absolute pressure p is found to be

v=!f· (9.1)

As an illustration, the density of hydrogen is 9.0 X 10-5 gm/cm3 at stand-

ard temperature and under standard atmospheric pressure of 1.013 X 106
dynes/cm2 ; consequently, the rms speed of hydrogen molecules under these
conditions is

v= 3 X 1.013 X 106 dynes/cm 2

9.0 X 10-5 gm/cm3
1.84 X 105 cm/sec,

which is considerably more than the speed of a rifle bullet.

Even though the molecules have high speeds, they zigzag about, and their
advance in a given direction is fairly slow. The reason is that each molecule
is continually impeded by collisions with others, thus altering its speed and
direction. It can be shown that each of the hydrogen molecules referred to
undergoes, on the average, billions of such collisions each second and that itR
mean free path between collisions is only about 0.00002 cm.
Two gases of different density, when admitted to the same enclosure with
the "lighter" one introduced at the top, will be found, after a time, to form a
mixture which is practically uniform throughout. This process, known as
diffusion, is explained by the random motion of the molecules, which causes
each gas to penetrate the other. The same effect is present in liquids, but
takes place at a slower rate.
9-3. Atmospheric Pressure. The earth is surrounded by a layer of air
which extends to a considerable height and which is held to the earth by
gravitational attraction. This body of air, like all fluids, exerts a pressure
that depends upon its height and density. Atmospheric pressure is usually
expressed in pounds per square inch or in dynes per square centimeter. It
is also expressed in a unit called the bar; one bar is equivalent to 106 dynes/cm 2 •
The value of the atmospheric pressure can be measured with a mercury
barometer, shown in Fig. 2. In setting up this instrument, a glass tube,
about three feet long and closed at one end, is completely filled with mercury
and then placed upside down in a vessel of mercury, as shown. The mercury
in the tube will fall somewhat and leave a vacuum above it. The height of
the column is a measure of the atmospheric pressure; a height of exactly 76
cm of mercury (76 cm Hg) at sea level and at 0°C is taken as standard. The
§ 9-3 ATMOSPHERIC PRESSURE 157

corresponding pressure, from Eq. 8.5, is p = hdg, where his the height of the
column, d the density of mercury, and g the acceleration due to gravity.
Hence, standard atmospheric pressure is 76.00 cm X 13.596 gm/ cm3 X 980
cm/sec 2 = 1.013 X 106 dynes/cm 2 ; this is equivalent to 1.013 bars or about
14.7 lb/in. 2 In mks units, standard atmospheric pressure is 1.013 X 105
newtons/m 2•
The aneroid barometer is another device for measuring atmospheric pres-
sure. This instrument consists essentially of a small sealed metal box from
which most of the air has been removed and which has a corrugated face that
moves.in and out as the atmospheric pressure changes. This slight move-
ment is· magnified by a system of levers and is communicated to a pointer
which sweeps across a graduated dial.

15

nh
E
~

t;
0
:;::::

w
>
0
Q)
10

20 40 60 80
Pressure, cm Hg

FIG. 9-2. FIG. 9-3. Variation of atmospheric

The mercury pressure with altitude.
barometer.

Near the earth, the air is weighed down and compressed by that above,
and at greater elevations the strata become rarer, although this effect is
somewhat offset by contraction due to cooling. The pressure does not vary
uniformly with altitude, but changes less rapidly at greater heights, as shown
in the graph of Fig. 3, which is based on actual test. Such graphs show
that it is possible to determine the altitude from a knowledge of the corre-
sponding atmospheric pressure. In fact, the altimeter used on aircraft is
essentially an aneroid barometer calibrated to indicate altitude instead of
pressure.
The force due to atmospheric pressure is often very large, even upon a
surface of moderate size. In most instances, the top and bottom of an object
are subjected to almost the same pressure; consequently, the surface does not
have to sustain a great load.
158 MECHANICS Chap. 9

For example, the force exerted by the atmosphere under standard conditions upon
a table top that measures 3 X 5 ft is

x 144
2
_!Q_
F -- pA -- 14 .7 lil·2 ~· x 15'if:!
1 ~2 ' 2 -
- 31, 750 lb'
or more than 15 tons. The underside is pushed up with an equal force.
9-4. Pressure of a Confined Gas; Boyle's Law. One of the outstanding
properties of a gas is its compressibility. When some gas is confined within
a closed cylinder having a tight-fitting piston, as repre-
Force sented in Fig. 4, and a steady force is applied as shown,
the piston will move inward and come to rest at some new
position where the pressure within the gas sets up a force
equal to that exerted upon it by the piston. In this proc-
ess the pressure of the gas is increased and its volume
reduced. Robert Boyle (1627-1691), English natural phi-
losopher, found a very simple relation between the pressure
and volume of a gas, which is known as Boyle's Law;
namely, that the volume of a confined body of gas varies in-
Fm. 9-4. Diagram
used in discussing versely as the absolute pressure, provided the temperature
Boyle's Law. remains unchanged. If P1 and V1 represent the pressure
and volume of the gas under one condition, and p2 and V 2
its pressure and volume under some different condition, then Boyle's Law
states that V1/V2 = PdP1, or
(9.2)
An ideal gas will be defined for the present as one which conforms to Boyle's
Law.
The kinetic theory, which pictures the pressure exerted by a gas as a con-
tinual bombardment of the enclosing walls by the moving molecules, can be
used to derive Boyle's Law theoretically. Since the temperature of the con-
fined gas is to remain unchanged, the mean kinetic energy of its molecules,
each of mass m, is assumed to remain unchanged. This equality of kinetic
energies for conditions 1 and 2 can be expressed as

where v1 and v2 are the respective rms molecular speeds. From Eq. 9.1:
v = V3p/d, it follows that 3 pi/d1 = 3 p2/d2 • But the density of the gas is
d1 = M/V 1 , and similarly d2 = M/V 2 , where the mass M of the gas is the
same under the two conditions. Therefore,

P1 V1 = p2V2,
which corresponds to Boyle's Law as stated above.
9-6. Pressure Measurement; Absolute and Gage Pressure. A simple form
of gage for measuring the pressure of confined gases consists of a U-shapcd
§ 9-G MIXTURES OF GASES; DALTON'S LAW 159

manometer tube containing mercury or other liquid. Fig. 5 shows manome-

ters of two common types connected to tanks of compressed gas. The pres-
sure forces the liquid down the tube at the near side and up at the far side.
In the open manometer, the difference in height h between the two columns
multiplied by the weight density of the liquid shows how much the gas pres-
sure exceeds that of the atmosphere. This value is known as the gage
pressure to distinguish it from the absolute pressure, which includes the
pressure of the atmosphere, § 9-3. In the closed manometer, the closed end
of the tube· contains air that is at atmospheric pressure when the columns are
level. As the liquid in the tube is forced to the position shown, this air be-
comes compressed, and the absolute pressure in the tank is found by adding
the pressure of the air entrapped in the tube (computed by Boyle's Law) to
that due to the liquid column of height h. The closed manometer is adapted
to the4 measurement of higher pressures than the open type, but it require.s
more calculation and is not so sensitive, especially at high pressures.

II

Frn. 9-5. Devices for measuring pressure: I, open type manometer; II, clm,ed type
manometer.

The Bourdon gage is used extensively for industrial purposes, as in the

measurement of steam pressures. The operating element of this device con-
sists of a bronze tube of elliptical cross-section. The tube, which is curved
into a circular arc, is fixed at one end, and the other is connected by a mechani-
cal linkage to a toothed sector which engages the instrument pointer. When
subjected to internal pressure, the tube tends to assume a circular cross-
section, and in so doing uncoils slightly and moves the pointer over a cali-
brated scale.
9-6. Mixtures of Gases; Dalton's Law. It has been indicated that when
the volume of a given mass of gas is reduced its pressure increases, and
when its volume is increased its pressure becomes smaller. The latter fact
will be of value in studying the mixture of several gases. If two or more
closed vessels, originally containing different gases that do not react chemi-
cally, are interconnected so that each gas has access to all of them, the
molecular motion causes each gas to penetrate the entire volume of all the
containers, and by diffusion the mixture eventually becomes homogeneous
throughout. In this process, each of the constituent gases expands into the
total available volume as though the other gases were not present, and the
160 MECI!ANICS Chap. 9

absolute pressure of each constituent is reduced, in accordance with Boyle's

Law, to a lower value called its partial pressure. If the pressure of the gas
mixture is measured, it will prove to be equal to the sum of the partial pres-
sures of the various constituents. This relation, which was first established'
by John Dalton (1766-1844), English chemist and physicist, may be stated
as follows: A mixture of several gases which do not react chemically exerts a
pressure equal to the sum of the pressures which the several gases would exert
separately if each were allowed to occupy the entire space alone at the given
temperature.
As an illustration, suppose that 2 liters of nitrogen in a container at an absolute
pressure of 3 atmospheres are allowed to mix with 8 liters of air in another container
at an absolute pressure of 5 atmospheres. The nitrogen in expanding from 2 to 10
liters will undergo a corresponding reduction of pressure, and in the final mixture its
partial pressure will be to X 3 = 0.6 atmosphere. Similarly, the partial pressure of
the air will be / 0 X 5 = 4.0 atmospheres. The absolute pressure of the mixture will
be the sum of the partial pressures, or 4.6 atmospheres.
9-7. Density of Gases. For a gas, as for any substance, density is defined
as mass per unit volume. Since the volume of any given mass of gas varies
inversely with the pressure, it follows that the density of a gas will vary di-
rectly with the pressure. The density of a gas is also influenced by the tem-
perature, for gases expand when heated and contract when cooled, § 13-1.
The density of air is of special importance and has been determined with
great care. One liter of air at standard temperature and pressure (0°C and
76 cm Hg) has a mass of 1.293 gm, and thus the density of air is 1.293 X 10-3
gm/cm3 , or 1.293 kg/m 3 • The corresponding weight density is approxi-
mately 0.081 lb/ft 3 • The specific gravity of a gas is usually expressed with
reference to air as a standard, instead of with reference to water as is done
with solids and liquids. Thus, coal gas, which is about four-tenths as dense
as air, is said to have a specific gravity of 0.4 with respect to air.
The densities and specific gravities of a few common gases are given in
Table I. These values are for standard temperature and pressure unless
otherwise noted. The values for specific gravity are stated with respect to
air as unity.
TABLE I. DENSITIES AND SPECIFIC GRAVITIES OF GASES

:Vlass Weight
Specific
Gas Density, Density,
Gravity
gm/liter lb/ft3
----- -----
Air .................... 1.293 0.081 1.000
Air at 20°C ............. 1.205 0.0755 0.932
Carbon dioxide .......... 1.977 0.123 1.529
Hydrogen .............. 0.090 0.0056 0.069
Helium ................. 0.179 0.011 0.138
Nitrogen ............... 1.251 0.078 0.967
Oxygen ................ 1.429 0.089 1.105
Steam at 100°0 ......... 0.598 0.037 0.462
§ 9-9 A l'OGADRO'S LAW; THE MOLE 161

9-8. Buoyancy of the Atmosphere. Archimedes' Principle, which is con-

sidered in § 8-7 in connection ,vith liquids, applies equally well to gases; the
principle states broadly that a body located in any fluid-liquid or gaseous-is
buoyed up by a force equal to the weight of the fluid displaced. As applied to
gases, interest in this principle centers in the buoyancy of the atmosphere.
A balloon filled with gas that is lighter than the surrounding air is raised
by the buoyant force, but as it rises through strata of decreasing density, the
buoyant force also decreases. Finally, a height is reached where this upward
force is equal to the total downward force; the latter includes the weight of
the balloon and its load, and also the weight of the gas with which it is filled.
At that elevation, the balloon is in equilibrium and does not rise higher.
During the ascent, some gas is allowed to escape in order to keep the pressure
inside the balloon about equal to that of the surrounding air and thus pre-
vent bursting of the envelope. To reach greater heights, sand ballast is
scattered overboard and, to descend, more gas is released.
Consider a balloon of 20,000-ft3 capacity, designed to reach an altitude of 10,000 ft
when. filled with hydrogen, and calculate the allowable weight of the balloon and its
load. As the elevation increases from O to 10,000 ft (3.05 km), the atmospheric
pressure, from Fig. 9-3, falls from 76 to 52 cm of Hg. The weight density of the air,
which, from Table I, is 0.081 lb/ft3 under standard conditions, is (52/76) X 0.081
= 0.0554 lb/ft3 at the 10,000-ft elevation, if temperature differences are neglected.
The buoyant force on the balloon at this level is, therefore, 20,000 ft3 X 0.0554 lb/ft3
= 1110 lb. If the hydrogen in the envelope is taken to have the same pressure as
ttie surrounding atmosphere, its weight is 0.069 X 1110 lb = 77 lb, and consequently
the allowable weight of the balloon and load is 1110 - 77 = 1033 lb.

In precise weighings it is sometimes necessary to apply corrections to take

account of the buoyancy of the atmosphere, especially if the object being
weighed and the weights with which it is being compared have markedly dif-
ferent densities. In an equal-arm balance the net downward force at the
end of each arm is the true weight of the object there minus the buoyant force
on it due to the atmosphere; and, when the instrument is balanced, the net
downward forces are equal. These facts make it possible to determine the
true mass of an object.
Find the correct mass, m gm, of a sample of balsa wood (density = 0.12 gm/cm3 )
that is balanced by 200 gm of brass weights (density = 8.5 gm/cm 3 ) at 20°C. The
weight of the wood is mg dynes, its volume is (m/0.12) cm3, and the buoyant force upon
it is (m/0.12) X 1.205 X 10-3 g dynes, where 1.205 X 10-3 is the density of air in
grams per cubic centimeter at the temperature stated. Similarly, the weight of the
brass is 200 g dynes, its volume is (200/8.5) cm3, and the buoyant force upon it is
(200/8.5) X 1.205 X 10-3 g dynes. Consequently,
mg - (m/0.12) X 1.205 X 10-3 g = 200 g - (200/8.5) X 1.205 X 10-3 g,
whence the correct mass of the wood sample is m = 202.0 gm.

9-9-. Avogadro's Law; the Mole. The Italian physicist Amadeo Avogadro
(1776-1856) suggested that, at the same temperature and pressure, equal
162 MECHANICS Chap. 9

volumes of di.fferent gases contain equal numbers of molecules. This hypothesis

has been verified experimentally and is called Avogadro's Law.
Under the kinetic theory this law, like Boyle's Law, could have been pre-
dicted on purely theoretical grounds, a fact which marks another triumph for
the modern theory. To establish the result, two gases will be considered.
The first will be supposed to consist of N 1 molecules each of mass m1 ; the
rms speed of the molecules will be called v1, the absolute pressure of the gas
P1, its volume Vi, and its density d1. For the second gas, the corresponding
properties will be indicated by the subscript 2. If the two gases are at the
same temperature, their molecules have the same mean kinetic energy, or

It follows from Eq. 9.1 that! m1 (3 pi/d1) = ~ m2 (3 p2/d2); and, if the density
dis replaced by Nm/V, this energy equation becomes

But if the gases have the same pressure and the same volume, p1 = p 2 and
V1 = V2; consequently, the foregoing equation reduces to

showing that each gas contains the same number of molecules, as stated in
Avogadro's Law.
This law is used in Physical Chemistry to establish a basic quantity known
as a mole, or gram-molecule. A mole of a substance is a mass in grams
numerically equal to the sum of the atomic weights of the atoms in the
molecule of that substance. The number of molecules in a mole of gas is
known as Avogadro's Number; a mole of any gas contains 6.024 X 1023
molecules. At standard temperature and pressure a mole of any gas occupies
a volume of 22.4 liters; this is called the gram-molecular volume.
The number of gas molecules in even a small volume is so immense as to
be almost beyond comprehension. Even at the lowest pressures achieved
by the modern vacuum pumps, there are still hundreds of millions of molecules
left in each cubic centimeter of space.

PROBLEMS
1. The density of carbon dioxide gas at 0°C and at standard atmospheric pressure
is 1.98 gm/liter. Compute the rms speed of carbon dioxide molecules under these
conditions.
2. Under atmospheric pressure the density of air is 1.293 gm/liter at 0°C; this
value is reduced by one-fourth at 91 °C. Compute therms speed of air molecules under
each of these conditions.
3. Suppose a barometer constructed as in Fig. 2 to use water instead of mercury
and compute the height of the water column in the tube at standard atmospheric
pressure.
 PROBLEMS 16:~

4. Convert a pressure of 150 cm Hg to (a) atmospheres, (b) pounds per square

inch, and (c) newtons per square meter.
5. Estimate the force exerted by the atmosphere under standard conditions on
the top of a bench 2! ft wide X 20 ft long.
6. State Boyle's Law in terms of pressure and density of a gas instead of pressure
and volume.
7. ,A cylinder 40 cm high is full of oil of specific gravity 0.90. A bubble of air
clinging to the bottom becomes detached and rises through the oil. If the bubble has
a volume of 0.05 cm 3 at the bottom, what is its volume just before it escapes at the top?
8. A piston 4 in. in diameter and weighing 30 lb is placed in the top of a vertical
cylinder 10 in. high, containing air. Assume that no leakage takes place and compute
how far the piston will move downward in compressing the gas before it comes to rest.
9. A 20-gal tank contains compressed air at a gage pressure of 30 lb/in. 2 Oil is
pumped into the tank without allowing any air to escape until the gage reads 60 lb/in. 2
How much oil was pumped in?
10. A tire pump with a cylinder 18 in. long is used to pump up a tire in which the
gage pressure is 10 lb/in.2 How far must the piston of the pump be pushed down
before air can enter the tire?
11. How many cubic feet of air reckoned at standard atmospheric pressure must be
pumped into a tire to increase the gage pressure in the tire from 15 to 28 lb/in. 2 ?
Assume the tire to have a constant volume of 0.8 ft 3 •
12. A tank contains 10 ft 3 of air at a gage pressure of 120 lb/in. 2 The tank is
equipped with a reducing valve by which air is removed from the tank at a gage
pressure of 15 lb/in. 2 How many cubic feet of air, reckoned at the reduced pressure,
have been withdrawn from the tank when its pressure gage reads 90 lb/in. 2?
13. A tank contains 20 ft 3 of air initially at a gage pressure of 100 lb/in. 2 If 50 ft 3
of air, reckoned at atmospheric pressure, are withdrawn from the tank, what will then
be the gage pressure in the tank?
14. The two closed tanks A and B shown in the diagram have equal diameters and
each is 4 ft high; they are connected at the bottom by a pipe in which there is a valve.
Initially the valve is closed, A contains water 3 ft deep with compressed air above, and
B contains water 1 ft deep with air at atmospheric pressure above. When the valve
is opened the water level drops 2 ft in A and rises 2 ft in B. What was the initial gage
pressure, in pounds per square inch, of the air in the top of tank A?
15. A tube 40 cm long, which is closed at the top, is pushed vertically downward
into a vessel of mercury as shown in the diagram, until only 30 cm of it project above
the exposed mercury surface. With the tube in this position, how far below the
exposed mercury surface will the mercury level be within the tube?
16. The diagram shows a U-tube in which air, initially at atmospheric pressure, is
entrapped by a column of mercury. From the dimensions shown, compute the gage
pressure and the absolute pressure of the entrapped air. Express the results in
(a) centimeters of mercury, (b) atmospheres, and (c) pounds per square inch.

Mercury

A Air
B

~5tcm
,a-
t
PROBLFJM 14. PROBLEM 15. PROBLEM 16.
164 MECHANICS Chap.9

17. In the open manometer of Fig. 5, suppose that the height his 14 cm and that
the liquid in the tube is water, and compute the gage pressure within the tank. Ex-
press the result in dynes per square centimeter.
18. In the closed manometer of Fig. 5, let the height h be 3 cm, the liquid in the
tube be mercury, and the length of the entrapped air column be 6 cm. Determine
the absolute pressure and the gage pressure in the tank. Express the results in
centimeters of mercury and in pounds per square inch.
19. A tank with a volume of 10 ft3 contains air at an absolute pressure of 80 lb/in. 2
A second tank with a volume of 6 ft 3 contains nitrogen at an absolute pressure of
50 lb/in. 2 If the tanks are connected together, what is the resulting absolute pressure?
20. A 10-gal tank contains air at an absolute pressure of 50 lb/in. 2, and a second
tank of 20-gal capacity contains air at an absolute pressure of 40 lb/in.2 (a) If the
two tanks are connected and the air allowed to mix, what is the resulting absolute
pressure? (b) If a third tank, having a volume of 15 gal and containing air at an
absolute pressure of 100 lb/in.2 is connected to the other two, what is the absolute
pressure in the combination?
21. Estimate the weight of the air in the tank of Prob. 9. Assume a temperature
of 20°C.
22. Estimate the weight of the air in a room that measures 30 X 40 X 15 ft.
Assume a temperature of 20°C.
23. What volume is occupied by 20 lb of oxygen at a gage pressure of 400 lb/in. 2?
Assume standard temperature.
24. What is the rms speed of the molecules of oxygen when the gas is at standard
temperature and at a pressure of (a) 1 atmosphere, and (b) 10 atmospheres?
25. A cork ball has a diameter of 4 cm. Find its apparent weight in air at 20°C as
measured on an equal-arm balance using brass weights. The specific gravity of cork
is 0.24 and that of brass is 8.5.
26. Compute the number of molecules of air in the bubble of Prob. 7. Assume a
temperature of 0°C.
27. A glass tube is evacuated to an absolute pressure of 10-7 mm Hg. Assume a
temperature of 0°C, and estimate the number of molecules per cubic centimeter re-
maining in the tube.
28. Refer to Prob. 1 and compute the mass of a carbon dioxide molecule, also the
average value of its kinetic energy due to molecular motion when the gas is at standard
temperature and pressure.
 10
FLUIDS IN MOTION
The preceding chapters deal with liquids and gases at rest, and the present
one with these mediums in a state of flow. To a large extent, this subject is
based on energy relations and on the energy transformations that occur from
point to point in fluid streams. It deals with the filling of containers and
the rates of discharge from them, the calculation of fluid flow, and the evalu-
ation of frictional forces. The principles apply not only to the motion of
fluids with respect to stationary objects, but also to the motion of objects
through stationary fluids.
10-1. Types of Liquid Pumps. Liquids can be raised from one level to
another by means of pumps. Some types commonly used in pumping water
are shown in Fig. 1. The l~ft pump and the force pump are of the reciprocat-

Suction

Water Surface

Lift pump Force pump Centrifugal pump

Frn. 10-l. Types of liquid pump: I, lift pump; II, force pump; III, centrifugal pump.

ing type, in which valves are provided in the piston or cylinder so that a new
supply of liquid will be furnished for each stroke. In the lift pump, when
the piston is drawn upward, valve 1 closes and valve 2 opens, and the pres-
1mre in the cylinder and pipe below it is lowered. Liquid rises into this space
because of the atmospheric pressure on the water surface below. When the
piston is pushed downward, valve 2 closes and the imprisoned liquid passes
through valve 1 to the upper portion of the cylinder. Upon the next up-
stroke the liquid above the piston flows out of the spout into the upper tank
U, and the cylinder fills again with liquid through valve 2. The operation of
165
166 MECHANICS Chap. 10

the force pump can be described similarly; the liquid is first drawn into the
cylinder as in the lift pump, and then forced to any desired height. In rais-
ing water, valve 2 of either pump must be less than 3-:l: ft above the water
surface, because atmospheric pressure cannot support a column of water
higher than this.
The centrifugal pump is continuous in operation. The liquid enters at the
center of a wheel, or impeller, where it is caught in a series of rotating vanes
and set into rotation. The liquid moves outward between the vanes into the
watertight housing of the pump, and thence into the discharge pipe.
10-2. Work Done in Pumping Liquid. In the process of raising liquid
from one level to another, the liquid gains potential energy, and this energy
is supplied by the pump. The amount of work done by the pump can be
found by supposing that an open-top tank containing liquid is equipped at
the bottom with a cylinder having a tight-fitting piston of area A, as shown
in Fig. 2, and that work is done in pushing the piston inward a distances by
the application of a force F. The force required is F = pA, where p is the

Pump
Reservoir

Fm. 10-2. Diagram used in cal- Fm. 10-3. An elevated

culating the work done in pump- tank being filled from a
ing a liquid. reservoir.

liquid pressure at the pump, and hence the work done is E = Fs = pAs.
During this stroke of the piston the quantity of liquid pushed from the cylin-
der into the tank has a volume V = As, represented by the shaded areas in
the figure. Hence, in forcing a volume V of liquid into a region where the
pressure exceeds that of the atmosphere by p, the work done is
E = pY. (10.1)
In deriving the foregoing expression, it is assumed that the pressure remains
eonstant; if it varies during the process, the work done on an infinitesimal
volume dY of liquid is expressed as dE = p dV, and the total work is therefore
E = fp dV. (10.2)
§ 10-3 SPEED OF FLOWING LIQUIDS 1G7

This expression will be used to find out how much work a pump must do in
order to fill an elevated tank with liquid from a source at a lower level. Fig.
3 shows an open-top tank with vertical sides, which is being filled with liquid
by a pump through a pipe extending from the base of the tank to a reservoir
below. The tank has a horizontal cross-section S, its height is H, and its
base is located a vertical distance B above the liquid level in the reservoir.
\,Vhen the tank is filled to a depth y, as shown, the pressure that the pump
must supply is p = (B + y)w, where w is the weight density of the liquid.
In order to raise the liquid level by an amount dy, the volume of liquid to be
pumped into the tank is dV = S dy, and the work needed is dE = p dV
= (B + y)wS dy. Hence, the total work required to fill the tank from bot-
tom to top is

E f B+y)wSdy
= = wSJ:7B+y) dy = wS [Bu+~J = wSH(B:+~)-
In interpreting the result, it may be noted that the weight of the liquid raised
is wSH, and the height through which its center of gravity is lifted is

(B + ~)-
10-3. Speed of Flowing Liquids. When a liquid flows out of a container
through an orifice it gives up potential energy and
gains kinetic energy. If there is no waste in the
process, the potential energy given up equals the
kinetic energy gained, and this fact makes it pos-
sible to compute the speed of the issuing jet of
liquid.
Figure 4 shows a tank with liquid extending to a
height h above an orifice, and indicates at the sur-
face an infinitesimal layer of the liquid which is
Fw. 10-4. Showing efflux
assumed to have a mass dm. When a mass equal of liquid through an ori-
to dm has issued from the orifice, this top layer has fice.
disappeared, and the potential energy of the liquid
is reduced by an amount dm (gh). If the speed of efflux is v, the gain in
kinetic energy is! dm (v 2), whence
dm (gh) = ! dm (v 2 ),

from which the speed of the issuing stream is found to be

v = v2gh. (10.3)
Since the acceleration due to gravity may be regarded as constant, the result
shows that the speed of efflux depends only upon the height of the liquid
above the orifice and is proportional to the square root of this height.
T'he speed of efflux can be stated in terms of liquid pressure instead of
height of liquid. Since the liquid pressure at the orifice depth is p = hdg,
168 llIEC!IANICS Chap. 10

from Eq. 8.5, it follows that h = p/dg, and therefore the expression v = V2 gh
can be written in the alternate form:

v = v2 p/d. (10.4)
Herein, dis the density of the liquid and pis the liquid pressure at the orifice
if the liquid is discharging from an open-top tank into the atmosphere; in
more general terms, p is the difference in pressure between the inside and
outside of the orifice.
10-4. Discharge Through an Orifice. The volume of liquid discharged per
unit time through an orifice can be calculated by geometry. If an incom-
pressible liquid flows through an orifice of area A cm 2 at a steady speed of
v cm/sec, the amount discharged in 1 sec would fill a cylinder v cm long and
A cm2 in cross-section. Therefore, the volume discharged per second is
Q = Av. (10.5)
A similar conclusion is reached with A in square feet and v in feet per second.
As an illustration, consider an open-top tank from which water is discharging
through an orifice of !-in. radius. At an instant when the liquid surface is 4 ft
above the orifice, what is (a) the speed of efflux, and (b) the discharge rate?
. - I ft
The speed of efflux, from Eq. 10.3, 1s v = V2 gh = '\.12 X 32 - 2 X 4 ft = 16 - .
ft
~ sec sec
The orifice has a radius of .}4 ft and an area A = 1r ( 2\ ) 2 = 0.00545 ft 2 ; consequently,
the discharge rate, from Eq. 10.5, is Q = Av = 0.00545 ft 2 X 16 ft/sec = 0.0872
ft 3/sec.
The efflux speed v and the discharge rate Q given by Eqs. 10.3 to 10.5 are
ideal values; actual values of v are less because of friction, and actual values
of Q are less because of contraction of the jet due to stream-line flow. The
actual discharge rate is found by multiplying the ideal value of Q by a coeffi-
cient of discharge, C. Values of this discharge coefficient, based on experi-
mental tests, are given in Fig. 5 for some orifices.

~ ~ ~
~
C=0.60 to 0.65
~ C=0.82
~ C=0.96
I ll m
Fw. 10-5. Some types of orifice and their discharge coefficients.

When the speed of efflux is not constant, it is useful to express the discharge
rate Q as dV /dt, where V represents volume and t time, and to modify Eq.
10.5 to read: dV /dt = Av. Thus, in emptying a tank through an orifice,
-dV = Av dt,
§ 10-5 EFFECT OF CROSS-SECTION ON SPEED 169

the minus sign being introduced because an increase in time corresponds to a

decrease in volume. From this equation the theoretical time to empty the
tank can be found by integration.
Figure 6 shows an open-top tank with vertical sides; it has a height H, a sur-
face area S, and an orifice of area A in the base.
If the height of the liquid above the orifice is h at
a given instant and is lowered by an amount dh in
time dt, then v = V2 gh and dV = S dh. These
values are substituted in the foregoing equation,
which then contains only two variables, h and t,
as follows:
-S dh = A V2gh dt.
Next, the variables are separated, giving

dt = - S h-112 dh. Fm. 10-6. Diagram used in

AV2g calculating time for emptying
a tank.
To find the time to empty the tank, starting full,
this expression is integrated between the limits h = H when t = 0, and
h = 0 when t = T; thus,

Jo
rodt = - s
AV2g)II
r O
h-1/2 dh.

It follows that [t JoT= -

2g z A
S
V
[hl/2Jo
-1
H
,
2S
or T = -~;;::-
Av2g
(o - VH),
-

whence the time to empty the tank is T = (S/A) V2 H/g.

A 10,000-gal tank with vertical walls 12 ft high has an orifice 3 in. in diameter in its
base. Assume that the tank is full of water, and compute (a) the time for the first
5000 gal to run out, and (b) the time for the remaining 5000 gal to run out of the tank.
. 231 0.. 3 1 ft 3
The volume of the tank 1s 10,000 gal X 1 ~l X 1728 -:,,. - 3 = 1337 ft3, and there-
~ L,
fore its cross-section is S = 1335 ft 3/12 ft = 111.4 ft2. The cross-section of the orifice
)2
is A = =[ X ( 132 n = 0.0491 ft2, and the ratio of surface to orifice area is S/ A
= 2269. The time to empty the tank completely would be 2269 V (2 X 12 ft) /32 ft/ sec 2
= 1965 sec or 32.75 min. Again, solve the problem as though the tank were only 6 ft
high; the corresponding time is 2269 V (2 X 6 ft)/ (32 ft/sec 2 ) = 1390 sec or 23.17 min.
Therefore, the first 5000 gal run out in 32.75 - 23.17 = 9.58 min, and the remaining
5000 gal run out in 23.17 min.

10-5. Effect of Cross-section on Speed. The equation Q = Av, derived

above for the discharge through an orifice, can also be used to calculate the
rate of liquid flow past a point in a pipe line. Moreover, for an incompress-
170 MECHANICS Chap. 10

ible liquid in steady flow the same volume that passes any cross-section of
the pipe in unit time must pass any other cross-section of the pipe in the
same time, or
(10.6)
where the subscripts l and 2 refer to any two points of the pipe line. This
expression, which is called the equation of continuity, shows that where the
area is reduced the speed undergoes a corresponding increase, and vice versa.
For example, if a pipe line contains a constriction, liquid will speed up in
flowing through the constricted portion to such an extent that the speed is
inversely proportional to the cross-sectional area.
Gases, being compressible, do not follow so simple a relation, but the gen-
eral effect is the same; that is, a gas speeds up in flowing through a reduced
area.
10-6. Energy of Moving Liquid; Bernoulli's Principle. A principle pro-
posed by Daniel Bernoulli (1700-1782), Swiss scientist and philosopher,
applies the Law of Conservation of Energy to a flowing stream of liquid, and
is of great importance in hydraulics. A liquid, regarded as incompressible
and frictionless, will be assumed to flow steadily through a tube of varying
elevation and cross-section, as illustrated in Fig. 7. The motion will be

h1

_____l _______________________ t _______ _

Frn. 10-7. Diagram used in the study of Bernoulli's Principle.

taken as sufficiently slow to permit stream-line flow-that is, flow without

wasteful turbulence or eddies-and for simplicity the speed at any cross-
section will be considered uniform across its area. During a short interval
of time, particles of liquid at section 1 and at section 2 will move as indicated
by the arrows. The areas at these sections are taken respectively as A1 and
A 2 , the corresponding speeds of the liquid as V1 and v2, the elevations above a
convenient datum level as h1 and h2 , and the pressure of the liquid, deter-
mined by manometers, § 9-5, or pressure gages; as Pi and p2. Since the
liquid is incompressible, the same mass m will pass any cross-section of the
tube in a given time t; the volume of this mass will be V = m/d, where dis
the density of the liquid. As this mass passes cross-section 1, an amount of
work p1V or p 1m/d is done upon it by the oncoming liquid at the left, in
accordance with Eq. 10.1; likewise, at cross-section 2 it will do work amount-
§ 10-6 ENERGY OF MOVING LIQUID; BERNOULLI'S PRINCIPLE 171

ing to p2m/d on the outgoing liquid at the right. The net work done by the
mass m of liquid is p2m/d - p 1m/d, and this must equal the reduction in its
potential and kinetic energy as it moves from section 1 to section 2. In
mathematical form:
p2m/d - p1m/d = (mgh1 + ! mvi2) - (mgh2 + ! mv22).
With the terms rearranged and each divided by mg, the corresponding expres-
sion per unit weight becomes
2 2
h1 + 2g
~ + J!1 =
dg
h2 +~
2g
+ dg
l!J.. (10.7)

These quantities are spoken of by .engineers as follows: h = elevation head,

v2 p
- = velocity head, and - = pressure head; the sum of the three is called
2g dg
total head. The pressure head can be expressed more simply as 12, where w,
the weight density of the liquid, replaces the product dg. w
Bernoulli's Principle is expressed in mathematical form by Eq. 10.7 and is
phrased concisely by stating that as an incompressible liquid flows, the total
head remains unchanged. It is supposed that waste of energy by friction is
negligible, and that no energy is supplied or expended, as with pumps or
turbines, in the region considered.

-++-fV_e__'_o_ci-ty hea_d=-----
__ 1
- _ - - _ - _ -·: F°""'·~=1

=::- -Ja_nk -- -_::. __ _ _

Pressure Velocity
head Pressure head
head

-L_______L_
Total
head

l __ _:_ej_ __
Fm. 10-8.
Elevation
Elev~Hon
head
p;pe ElevaHon \

Representing flow of liquid through a level, frictionless pipe.

head

The heads referred to are shown pictorially, although not to seale, in Fig.
8, which represents liquid flowing from a tank along a pipe and out of an
orifice, and shows the pressures that would be obtained at several points if
it could flow without friction. Within the tank the speed of the liquid is
negligible, and there exist only the pressure head (assumed to be maintained
constant) and the elevation head. Just outside the tank there is a velocity
head and a consequent reduction of the pressure head. Since friction is
assumed absent, tho pressure head remains unchanged along the line. In
172 MECHANICS Chap. 10

the stream issuing from the orifice the original pressure head is replaced
entirely by velocity head.
With friction present, some of the mechanical energy of a flowing stream
is converted to heat, and there is a gradual reduction of pressure along the
line of flow, as shown in Fig. 9. This pressure reduction due to friction is
conveniently treated as a "friction head," which increases linearly along the
horizontal pipe, thereby reducing the pressure head. At the outlet its effect
is to reduce the velocity head and the speed of efflux.

-~-~
-
- - ----
--- - -I Velocity Friction
- - -- -

-----~l-
- -- --
- - - -- - head
- - - - -
-~--::~~~~---_
------ --:::.::::::.

Pressure
Total head
head :;
- - 1 - - - -~

~'\
A

I]
Elevation
head ".

_________________________ j ---------
Fw. 10-9. Showing the effect of friction in reducing the rate of discharge.

Under certain conditions, the mathematical form of Bernoulli's Principle

becomes simplified. Thus, for flow through a horizontal pipe h1 = h2 , and
Eq. 10.7 reduces to
2 2
!:'.!_ + Pi = ~ + P1. (10.8)
2 d 2 d
Again, for flow in an open channel p1 = p2; consequently,
2 2
h1 + 1!.!_
2g = h + ~-
2g 2
(10.9)

The student is reminded to use consistent units in applying Bernoulli's

Principle. In the pressure head, p/dg or p/w, for example, the following sets
of units will be found consistent:

System of Units p d g w
-----~--
British gravitation al lb/ft2 slugs/ft3 ft/sec 2 lb/ft3
"f t . { mks
., enc
newtons/m2 kg/m 3 m/sec 2 newtons/m3
cgs dynes/cm 2 gm/cm3 cm/sec2 dynes/cms

10-7. Measurements of Flowing Liquids. Some measuring devices for

liquid flow that operate on Bernoulli's Principle will be described.
§ 10-7 MEASUREMENTS OF FLOWING LIQUIDS 173

The Venturi ivleter. The Venturi meter is used for determining the rate of
discharge through a horizontal pipe line. This device introduces a constric-
tion in the pipe and makes provision
for measuring the pressures in the
pipe and in the constriction. A sim-
ple form of the Venturi meter is
shown in Fig. 10. Two cross-sec-
tions, one located in the pipe and
one in the constriction, are num-
bered 1 and 2, respectively, and the
fluid speeds at these locations are
represented by v1 and v2 • The liquid Fm. 10-10. The Venturi meter.
rises in the pressure tubes at these
locations to heights h1 and h2, and the pressure in the pipe is noted to be
greater than in the constriction. From a knowledge of the pressure values
and the dimensions of the meter, it is possible to determine the discharge
rate in the pipe.
In an experimental model of the Venturi meter, the diameter of the pipe is 4 cm and
that of the constriction is 3 cm. With water filling the pipe and flowing at a certain
rate, the height of liquid in the pressure tube is 20 cm at the pipe and 15 cm at the
constriction. What is the discharge rate?
The cross-sectional area of the pipe is A1 = 4 1r cm2 and that of the constriction
is A2 = (9 1r/4) cm2. From the equation of continuity, A 1 v1 = A2 v2, it follows that
V2 = (A1/A2) v1 = 16 v1/9. The pressures at the cross-sections, from Eq. 8.5, are
gm dynes dvnes
p1 = 20 cm X 1 - X 980 - - = 19,600 -·-·- 2- and P2 = 15 X 1 X 980 = 14,700
cm3 gm cm
dynes.
Equation 10.8 expressing Bernoulli's Principle is transposed to read
cm 2

and in this problem becomes

16 V1 ) 2 = 2 X (19,600 - 14,700) dynes

cm2 , or (175) 2
(9 gm 81 v1 = 9800 cm2/sec ,
- V12 2
1
cm3
from which v1 = 67.4 cm/sec. Consequently, the discharge rate is Q = A1 v1
= 4 1r cm2 X 67.4 cm/sec = 847 cm 3/sec.

The Pitot Tube. The speed of flowing liquid can be measured with a
Pitot tube. This device in its simplest form consists of an £-shaped glass
tube, open at the ends, which is placed in the liquid with its orifice turned
upstream. Because of the impact of the current, the liquid in the tube
rises to a greater height than it would reach otherwise; this excess height is a
measure of the speed of the stream. Figure 11 shows a horizontal pipe
filled with flowing liquid, together with a Pitot tube P arranged to measure
the speed of the liquid at the center line of the pipe and a riser tube R to
measure the pressure of the liquid. Bernoulli's Principle is applied to two
174 1\IEeHANICS Chap. 10

locations as follows: (1) a point just inside the orifice of the Pitot tube, and
(2) a point in the pipe at the same level. As applied to these points, Eq.
10.8 becomes
p R O _ h1dg = V22 _ h2dg1
d 2 d
from which 1h = V2 g(h1 - h2) = 'V2gh.
Consequently, the speed of the stream can
be computed directly from Eq. 10.3.
10-8. Illustrations of Bernoulli's Principle.
'(2 Besides the measuring devices just described,
============ there are many applications of Bernoulli's
Fm. 10-11. The Pitot tube. Principle. Among these is a simple arrange-
ment called an aspirator, for exhausting the
air from a vessel. In this device, shown diagrammatically in part I of Fig. 12,
a stream of water is admitted at the top and flows through a small nozzle into
a chamber which connects through a horizontal passage to the vessel being
exhausted. The small size of the nozzle causes the stream to issue from it at
high speed, and the pressure in the chamber is lowered in accordance with
Bernoulli's Principle. Consequently, air is drawn from the vessel and driven
through the outlet by the water stream. Although not nearly as effective
as a vacuum pump, an aspirator will reduce the pressure to about 3 or 4 cm Hg.
The atom1:zer, shown in part II of Fig. 12, is an aspirator in reverse. A

Fm. 10-12. Illustrations of Bernoulli's Principle: I, an aspirator; II, an atomizer.

stream of air is blown at high speed across the top of the upright tube that
leads to the liquid, the pressure at this point is reduced, the liquid is raised
into the air stream by normal atmospheric pressure, and tiny droplets form
the familiar spray.
A tennis ball that is set spinning when struck has a disturbed flight. When
the player "tops" the ball on a ground stroke he sets it spinning about a
horizontal axis, say, he spins it clockwise while it is moving to the right in
still air, as indicated in part I of Fig. 13. The effect is the same as if the ball
were spinning on a stationary axis in a wind directed toward the left, as in
part II of the figure. As the ball spins, a layer of air clings to it and is car-
§ 10-9 VISCOSITY OF LIQUIDS 175

ried around with it, and the velocity of the air at any point near the ball can
be regarded as made up of two components, one due to the wind and the
other to the spinning of the ball. Below the ball, these components have the
same direction, but above they have opposite
directions. It follows that the velocity is
greater at the bottom surface than at the top,
and according to Bernoulli's Principle, the
pressure is reduced at the bottom and in-
creased at the top. Hence, an unbalanced
0- 0 II
force is developed which causes the ball to FIG. 10-13. Diagram used in ex-
curve downward as it moves forward. In plaining the curved flight of a
serving, the ball is usually given a side spin spinning ball. I, The ball mov-
also, which makes it swerve to the right or left. ing through stationary air; II,
Bernoulli's Principle has an engineering ap- the ball imagined to be turning
about a fixed axis in moving air
plication of great importance in the lifting stream.
effect produced upon a moving airfoil. An
airfoil section which is assumed to be moving to the left in still air experiences
the same effects as if it were stationary and the air streamed past it toward
the right, as indicated in Fig. 14. Because of the curvature of the top surface
of the airfoil, the air that flows close to it is deflected as shown, while that at
a higher level is relatively undisturbed. Therefore, the stream of air entering
the region just above the airfoil is forced to flow through a constricted area,
like the water in a Venturi meter, and its speed is increased. In accordance
with Bernoulli's Principle the pressure in this region is lowered, and the
normal atmospheric pressure upon the lower surface forces the airfoil upward.
A further lifting effect is produced by the impact of the air against the inclined
lower surface of the wing. When the air strikes the wing, the air is thrust
downward and the wing upward.

-- --r~--, - .. r=d---7-
I - - c _ .. - - ·; -· .. - ...
-- - - - I . -- - /-
- I - - --1- / - --,--
-'-- __ ,_,___ , __
-- I - . . 1- I - - -·/ - .

- L--~~-J L_=--3-~~' -

FIG. 10-14. The lift on an airfoil ex- FIG. 10-15. Diagram used in ex-
plained by Bernoulli's Principle plaining viscosity.

10-9. Viscosity of Liquids. The property of a liquid that results in a

resistance to flow is called viscosity. If two beakers, one containing some oil
and the other some alcohol, are tilted from side to side, the oil exhibits much
less mobility than the alcohol, and the oil is said to be the more viscous liquid.
The concept of viscosity will be introduced by considering the flow of a
liquid over a flat surface. The layer of liquid particles in contact with the
176 MECHANICS Chap. 10

surface remains stationary because of adhesion, the next layer of particles

moves slowly over the first, the third layer moves with respect to the second,
and so on, the speed of each layer increasing with its distance from the solid
surface. This distribution of speed causes a portion of the liquid that is
cubical in shape at one instant to become rhomboidal at a later instant, as
illustrated in Fig. 15. The layer of liquid that forms the base of the cube
travels from a to b while the upper face travels from c to d. If the speed of
the upper face exceeds that of the lower face by an amount dv, and if the
vertical distance between these faces is dh, the liquid can be looked upon as
shearing at the rate dv/dh, and this rate will be constant as long as the shear-
ing stress to which the liquid is yielding remains unchanged.
The shearing stress acting on the cubical portion of the liquid in Fig. 15
is measured by the force per unit area of the upper (or lower) face and can be
represented by F / A. Experiment shows that the rate of shear dv/dh is pro-
portional in any liquid under like conditions to the shearing stress F / A
which causes the liquid to flow; that is,
F dv F dv
A o: dh ' or A = rJ dh '
where rJ ( eta) is the proportionality constant. This constant is called the
coefficient of viscosity of the liquid; its value is given by

rJ = :V;~h, (10.10)

an expression which is similar to that for shear modulus of elasticity, § 7-4.

In general, the viscosity of a liquid decreases with a rise in temperature.
Equation 10.10 affords a physical idea of what is meant by unit viscosity.
If a layer of liquid 1 cm thick is at rest, and if a plane surface of large area is
drawn across the upper surface of the liquid with a speed of 1 cm per sec, then
the force in dynes per square centimeter of area needed to move this plane
surface in the manner indicated is numerically equal to the coefficient of
viscosity of the liquid. Its unit in the cgs system is the dyne· sec per cm 2, and
is called the poise. The unit for viscosity in the mks system is the newton· sec
per m2, and in the British gravitational system is the lb· sec per ft 2 • Coeffi-
cients of viscosity in poises are given for some liquids in Table I.
TABLE I. CoEFFICIEXTS oF VrncosrrY OF LIQUIDS

Liquid Poises

Alcohol, ethyl. ......... at 20°C 0.012

Benzene .............. . 20°c 0.0065
Glycerine ............. . 20°C 8.3
Mercury .............. . 20°c 0.016
Oil, machine .......... . 19°C 1.02
Water ................ . 20°C 0.010
Water ................ . 100°C 0.0028
§ 10-10 VISCOSITY OF GASES 177

The method of measuring viscosity described in the preceding paragraph

is not easy to carry out directly. A more convenient method is to cause the
liquid under test to flow through a vertical tube of small bore and to measure
the rate of flow. Equation 10.10 is applied to an annular layer and the re-
sult integrated; the analysis yields the following equation for the rate of flow
in cubic centimeters per second through a tube of length L cm and radius
rem:
- 1rpr4 (10.11)
Q - 8 'Y/L'

where pis the liquid pressure in dynes per square centimeter that would exist
at the lower end of the tube if it were closed. This relation is known as
Poiseuille's Law and is named after the French physiologist Jean L. M.
Poiseuille (1799-1869). It applies to small speeds of flow that will not set up
eddies in the liquid.
The Saybolt viscosimeter is widely used in this country for measuring
viscosity. It operates on the principle that the time for a definite volume of
liquid to flow through a short small tube is proportional to the coefficient
of viscosity. In the usual instrument the number of seconds for 60 cm 3 of
liquid to flow out is used as the measure of viscosity. Water at room tem-
perature gives a reading of about 30 sec.
10-10. Viscosity of Gases. Viscosity is present in gases as in liquids,
although to a smaller extent, as would be expected from the increased spacing
of the gas molecules. It not only retards the motion of gases themselves in
flowing through tubes, ducts, and the like, but also retards other bodies in
moving through a stationary gas. Air drag increases with the speed and is
an important factor in airplane construction, as evidenced by the stream-
lining of surfaces to minimize its effects.
A study of scientific interest is the motion of a sphere falling through a gas,
such as a raindrop falling through the atmosphere. Such a drop starts to
fall with constant acceleration under the action of gravity; at first viscous
friction exerts only a small drag upon it and it gains speed quickly, but as it
does so the upward force of friction becomes greater and soon equals the
downward force of gravity. Thereafter the drop descends with uniform
motion. Sir George Stokes (1819-1903), British mathematician and physi-
cist, found that for very small drops of radius rand density d the steady, or
terminal, speed acquired in falling through a medium having a viscosity
coefficient 'Y/ is given by the expression

(10.12)

in which g is the acceleration due to gravity. With cgs units, the value of 'Y/
will be in poises. The coefficients of viscosity for some gases are given in
Table II. It is interesting to note that a rise in temperature causes the
178 AfECIIANICS Chap. 10

viscosity of gases to increase, although the reverse 1s generally true for

liquids.

TABLE IL COEFFICIENTS OF VISCOSITY OF GASES

Gas Poises

Air ................ at 0°0 173 X 10-s

Air ............... . 9900 220 X lQ-6
Ammonia .......... . 0°C 96 x 10--5
Hydrogen ......... . 0°C 87 x 10-5
Mercury vapor ..... . ::ioo 0 c 532 x 10-5
Oxygen ........... . 0°C 189 X 10-ll
Water vapor ....... . 0°C 90 X 10-s
Water vapor ....... . 100°C 132 X 10-s

PROBLEMS
1. How much work is needed to pump 3 ft 3 of water into an open-top tank through
an inlet in the base, if the water in the tank is 5 ft deep? Assume that the tank has
so large a cross-section that the depth of the water is not changed appreciably during
the process.
2. An open-top cylindrical tank has a diameter of 2 m and a height of 3 m; its
axis is vertical and its base is elevated 5 m above a constant-level reservoir. How
much work is needed to fill the tank with water through a pipe extending from the
reservoir to the base of the tank?
3. A cylindrical tank 12 ft in diameter and 15 ft high is open at the top; its axis is
vertical and its base is elevated 30 ft above a constant-level water reservoir. Compute
the amount of work needed to fill the tank through a pipe that extends from the
reservoir to the base of the tank.
4. The tank of Prob. 3 is to be filled by a motor-driven pump that delivers 5 hp,
the pumping unit having an overall efficiency of 45 per cent. (a) Compute the time
needed to fill the tank. (b) What does it cost to fill the tank if electrical energy is
supplied at 3 cents per kilowatt· hour?
5. If the tank of Prob. 3 were to be filled through a pipe that discharged into the
top of the tank, as shown in the diagram, how much work would be needed?
6. A cylindrical can 20 in. high is vented to the atmosphere at the top, and is! full
of kerosene. If a crack develops in the can at a place 3 in. from the bottom, with
what speed will the liquid flow through the leak?

- -~=
- ---- .=.-:
Pump
~- :-=-.12 iii::
Reservoir l__ Orifice -=--::...---=--

PROBLEM 5. PROBLEM 9. PROBLEM 10.

 PROBLEMS 179

7. What is the speed of efflux of water from an open-top tank when the liquid
surface is 2.5 ft above the orifice?
8. Water squirts vertically upward through a leak in the top of a pipe. If the
gage pressure in the pipe is 2 lb/in.2, (a) what is the speed of the issuing stream, and
(b) how high will the stream rise?
9. The diagram represents a fire extinguisher in which the region above the liquid
contains gas under pressure. Suppose the gage pressure in this region to be 3 lb/in. 2
and compute the speed of efflux of the liquid through the orifice. Assume that the
liquid has the same density as water, and neglect the effects of friction.
10. Two open-top tanks are connected at the bottom by a pipe, as in the diagram.
When the depth of the liquid is 48 in. in one tank and 30 in. in the other, what is
the theoretical speed of the liquid in the pipe?
11. A stream 20 ft wide has an average depth of 4 ft and flows at 1.5 ft/sec. If it
falls through a height of 15 ft, (a) how many cubic feet of water pass over the fall
per second, (b) how many foot·pounds of work can this water do in falling, and (c) how
many horsepower are available, theoretically, from the stream?
12. Refer to Prob. 7 and compute the actual discharge rate, in cubic feet per minute,
under the conditions described. Suppose the orifice to have a diameter of ! in. and a
discharge coefficient of 0.82.
13. Suppose that the tank of Prob. 3 is full of water and that it is allowed to empty
through an orifice in its base. If the cross-sectional area of the orifice is 4 in. 2 , in what
time will all the water run out of the tank?
14. Refer to Prob. 13 and compute the time in which the depth of water in the tank
is reduced from (a) 15 ft to 7.5 ft; (b) 7.5 ft to zero.
15. A hollow cone has its axis vertical and its vertex at the bottom. Its vertical
height is H and its radius at the top is R. The cone is open at the top, and initially
is full of liquid. If there is a small orifice of area A at the vertex, show that the time
in which all of the liquid will run out is

t = 1rR2 /2H,
5A '\Jg
where g is the acceleration due to gravity.
16. A horizontal pipe line 14 in. in diameter contains a constriction 12 in. in di-
ameter. If oil flows through the pipe at 3 ft/sec, what is its speed at the constriction?
17. In a horizontal piping system a main 6 in. in diameter feeds two pipes each 4 in.
in diameter. If water filling the main flows at 2 ft/sec, what is its speed in the smaller
pipes?
18. Water flows through a horizontal pipe 5 cm in diameter in which there is a
constriction 4 cm in diameter. If the pressure in the pipe exceeds that in the con-
striction by 1.5 X 104 dynes/cm2, (a) what is the speed of the water in the pipe, and
(b) what is the rate of flow in cubic centimeters per second?
19. vVater flowing in a horizontal pipe line 1 ft in diameter encounters a constriction
9 in. in diameter. If the pressme in the pipe exceeds that in the constriction by
2 lb/in.2 , what is the rate of flow in the pipe in cubic feet per second?
20. Water flows down an open channel that drops 1 ft for each 20 ft of length. If
the stream has a speed of 2 ft/sec at a certain place, what is its speed :.tt a place 10 ft
farther along the channel?
21. To measure the viscosity of a liquid, an open-top can is maintained full of the
liquid while it discharges steadily through a nozzle that extends vertically downward
from the bottom of the can. In a particular apparatus the can is 10 cm high; the
nozzle is E, cm long and has an internal diameter of 1.3 mm. A test of carbon tetra-
chloride in this apparntus r,;howecl a discharge of 100 cm 3 in 29.1 sec at 20°C. The
180 MECHANICS Chap. 10

specific gravity of the liquid at this temperature is 1.595. From these data find the
coefficient of viscosity of carbon tetrachloride at the temperature stated.
22. A vertical tube extending downward from the base of an open-top container is
used in testing the viscosity of liquids. Water discharges through the tube at the rate
of 2 cm3/sec when the temperature is 130°F; at this temperature its density is 0.986
gm/cm3 and its viscosity coefficient is 5.27 X 10-3 poise. Lubricating oil known in
the trade as No. 20 discharges at the rate of 0.4 cm8/sec when the temperature is also
130°F; at this temperature the density of the oil is 0.854 gm/cm3 • In both tests the
surface of the liquid in the container is kept at the same constant level. Determine
the coefficient of viscosity of the oil at the temperature stated.
23. A droplet of oil falling through air is observed to fall 0.5 cm in 0.15 sec at con-
stant speed. If the coefficient of viscosity of the air is 1.82 X 10-s poise at the temper-
ature of the test, what is the weight of the droplet? Take the density of oil to be
0.90 gm/cm3 •
 HEAT

11
HEAT AND HEAT
TRANSFER
Heat has been referred to earlier as a form of energy. When a body is
subjected to heating, the fact that it absorbs energy becomes evident in one
of two ways: (1) a change of temperature or (2) a change of phase, as in the
melting of a solid or the vaporization of a liquid. Associated with these
major changes, the application of heat to a body generally causes an increase
in size and may produce numerous other effects: incandescence, in which the
body emits light; chemical changes, such as oxidation and reduction; changes
in the resistance of electrical conductors; or the generation of an electromotive
force at a junction of dissimilar metals. lvfany of these effects are considered
later in this book; the present chapter is devoted to change of temperature and
expansion, and to methods of heat transfer.
11-1. Nature of Heat. In the early days of science, heat was thought to
be a weightless fluid, called caloric; all bodies were supposed to contain more
or less of this fluid, and the passage of heat from one to another was explained
as a flow of caloric from the hotter to the colder body. It is to the American
scientist Benjamin Thompson, later Count Rumford (1735-1814), that the
present concept of heat as a form of energy is due. He was impressed by the
large amount of heat produced in attempting to bore metals with blunt tools;
since the supply of heat appeared inexhaustible, he concluded that heat could
not be a substance but must be related in some way to motion. His conclu-
sions have been confirmed by other investigators, and now heat is accepted
without question as a form of energy.
W'hen heat is applied to a body, the energy of that body is increased, and
181
182 I!EA1' Chap. 11

since, in general, there is no change in either the kinetic or the potential

energy of the body as a whole, it may be assumed that the energy is given
to the molecules of which it is composed. Molecules are known to possess
kinetic energy, for there is ample evidence of their incessant motion, as
described in Chap. 9. Molecules may also possess potential energy, as do
the molecules of a solid or liquid that has been expanded by heating, for work
must have been done upon the molecules to separate them in opposition to
the forces of cohesion. Gas molecules have but little potential energy, since
they are relatively far apart and hence have only a slight attraction for one
another. The heating of a body, therefore, increases its internal energy,
either in kinetic or potential form, or both.
It is generally known that heat flows from a hot object to a cold one. In
this process, one object loses, and the other gains, internal energy. ·when
an object emits or gives off heat its internal energy is reduced, and when it
absorbs or takes on heat its internal energy is increased.
Also, the transfer of heat from one part of a body to another is explained
by an increase of molecular motion or, for good electrical conductors, by an
fr1crease of electronic motion. If one end of a metal rod is placed in a fire,
the entire rod becomes warmer. It is supposed that at the fire the free elec-
trons present in the metal are set into more rapid vibration, and that these
in striking neighboring electrons impart kinetic energy to them, and so on
throughout the rod. Again, when a vessel containing a gas is heated, the
gas molecules striking the heated sides of the vessel in their incessant motion
rebound with greater speeds; these molecules then strike others, and so on;
in this way the entire gas is heated.
11-2. Temperature Scales. The term temperature is used to express how
cold or how hot a body is; cold implies a low temperature and hot a high tem-
perature. A more definite idea of temperature can be realized by considering
the result of bringing a hot object into contact with a cold one; for example,
plunging a hot steel forging into cold oil. The forging becomes cooler and
the oil warmer, as the hot body gives up some of its internal energy to the
cold one. This process continues until a state of thermal equilibrium is
reached, when the forging and the oil reach the same temperature. A state-
ment of temperature signifies that the molecules of the object referred to
possess a particular amount of energy due to thermal agitation, and an in-
crease of temperature implies a corresponding increase in internal energy.
The terms cold, cool, warm, and hot, although used in everyday speech to
suggest temperatures, do not make it possible to state a given temperature
with definiteness, and this fact has led to the adoption of thermometric
scales. Such a scale is constructed by choosing two standard temperatures,
called fixed points, that can be reproduced easily, assigning arbitrary num-
bers to these temperatures, and dividing the interval between them into an
appropriate number of equal parts, each called a degree ( 0 ) . The divisions
are extended above and below the fixed points, if necessary.
§ 11-2 TEMPERATURE SCALES 183

Four such temperature scales are in rommon use. For each, the fixed
points are taken as the melting point of ice and the boiling point of water,
both at the standard pressure of 76 rm Hg. These scales are shown in Fig. 1,
and Table I lists the data from which they are constructed. The fahrenheit

TABLE I. COMPARISON OF TE:\1PERATURE SCALES

Fahrenheit Centigrade Kelvin Rankine

Boiling point of water .............. 212° F 100° c 373°K 672° R

Melting point of icP ................ 32 0 273 492
Divisions between fixed point;; ....... 180 100 100 180
I
scale, which is used largely for engineering and household purposes, ,ms
named after the German physicist Gabriel D. Fahrenheit (1686-1736), who
made the first mercury-in-glass thermometer. The centigrade scale is due
to Anders Celsius (1701-17 44), Swedish astronomer, and is correctly called
the celsius scale, but this name is not in general use. This scale is universally
used for scientific measurements. The other two scales extend to absolute

Boiling point -...

212°F---- 100°(----- 373°K----- 672°R
of water

Melting point-... 32------ o------- 273------ 492

of ice

Absolute zero _ -460 -273 0 0

of temperature

Fahrenheit Centigrade Kelvin Rankine

FIG. 11-1. Comparison of thermometer scales.

zero: a value that is considered theoretically to be the lowest possible tem-

perature. The one based on centigrade divisions is called the kelvin scale in
honor of the English scientist Lord Kelvin (William Thomson, 1824-1907), and
the other, based upon fahrenheit divisions, is called the rankine scale after
the Scottish engineer William J. M. Rankine (1820-1872). Absolute zero of
temperature is -273.16°C or -459.69°F. For most purposes these values
are taken to the nearest degree; Table I is based on the values - 273 °C and
-460°F.
184 HEAT Chap. 11

It is frequently necessary to convert from one scale to another. In doing

so, it should be observed that a difference of 180 fahrenheit or rankine degrees
is equivalent to 100 centigrade or kelvin degrees, and that the melting point
of ice, which serves as the usual reference temperature, has a different value
on each of the four scales.
For example, to convert a temperature of 70°F to the centigrade scale, find the
number of degrees in excess of 32°F and multiply this number by 100/180; thus,
(70 - 32) X 100/180 = 21.1 °C.
The relation between fahrenheit and centigrade readings can be expressed by
the proportion
F - 32 C (11.1)
180 = 100'
wherein the letters represent the corresponding temperatures.
A rise or fall in temperature can be converted from one scale to another
without reference to any fixed point. Thus, a change of 50 centigrade degrees
corresponds to a change of (180/100) X 50 or 90 fahrenheit degrees. A
change of temperature is indicated by the symbol C 0 or F 0 to distinguish it
from a scale reading, which is designated by °C or °F, as used above. If the
temperature rises from 20°C to 30°C, the rise is 10 C 0 •
11-3. Linear Expansion of Solids. Practically all solids expand when
heated. A metal rod, heated uniformly from end to end, will increase in
length by an amount that depends upon the material of the rod, its length,
and the temperature rise. In symbols, if Le is the length of the rod at the
lower temperature (cold), then the increase in length, AL, due to a temperature
rise At, is proportional to Le and to At; thus,
AL= aLc At, (11.2)
and, therefore, the length at the higher temperature (hot) will be Lh = Le +
aLe At, or
(11.3)
The proportionality constant a in these expressions is called the coefficient of
linear expansion. This coefficient represents the increase in length per unit
length per degree of temperature rise; it has different values for different
substances, and for a given substance varies somewhat over different tem-
perature ranges. Some representative values of a are given in Table II; the
values listed apply to temperatures in the neighborhood of 20°C, except where
particular temperatures are noted.
For iron at ordinary temperatures the coefficient of linear expansion is
given as 0.000012/C 0 or 0.0000067 /F 0 • This means that a 1-cm length of
iron becomes 1.000012 cm long when subjected to a temperature rise of 1
centigrade degree, and similarly a 1-in. length becomes 1.000012 in. long
under the same temperature change; the increase in length would be only
i as much for a rise of 1 fahrenheit degree.
§ 11-3 LINEAR EXPANSION OF SOLIDS 185

TABLE IL COEFFICIENTS OF LINEAR EXPANSION OF SOLIDS

Per C Degree Per F Degree

Aluminum .................... . 24 X 10-5 13 X 10-5

Brass or bronze ................ . 18 X 10-5 10 X 10-s
Copper ....................... . 17 X 10-s 9.5 X 10-s
Glass (soft to hard) ............ . 8 to 9.5 X 10-s 4.5 to 5.3 X 10-s
Ice (range -10° to 0°C) ........ . 51 x 10-s 28 X 10-5
Invar steel (36% nickel) ........ . -0.3 to +2.5 X 10-s -0.2 to +1.4 X 10-s
Iron (wrought) ................ . 12 X 10-5 6.7 X 10-s
Lead ......................... . 29 x 10-s 16 x 10-s
Platinum ..................... . 9 X 10-5 5 X 10-s
Pyrex glass . . . . . . . . . . . . . . . . . . . . 3 X 10-s 1.7 X 10-5
Silica, fused (0° to 30°C) ....... . 0.42 X 10-s 0.23 x 10-5
Silver ........................ . 19 X 10-s 11 X 10-5
Steel ......................... . 11 X 10-5 6.1 X 10-5
Zinc ......................... . 26 X 10-s 14 x 10-s

The importance of expansion can be illustrated by determining the change in length

of steel rails due to a change of temperature. Suppose that the rails are laid when
the temperature is 35°F and that each rail is 40 ft long at that temperature; what gap
should be left between adjacent rails to provide for expansion up to 120°F? The
expansion of each rail over the range from 35 to 120°F-a temperature rise of 85 F 0 -
is given by Eq. 11.2 as AL = (6.1 X 10-6/F 0 ) X 40 ft X 85 F 0 = 0.0207 ft or 0.249
in., and a gap of this length should be provided between adjacent rails.

Thermal expansion can be offset by mechanical compression; the compres-

sive force must be sufficient to shorten the body by the same amount that
heating would elongate it. The value of the force depends upon Young's
modulus of elasticity for the material of the body and can be computed from
Eq. 7.2.
Several examples of linear expansion are illustrated in Fig. 2. In part I

-----___; Bross

a m
FIG. 11-2. Expansion of metal bars: I, additive expansion; II, differential expansion;
III, expansion neutralized; IV, unequal expansion causing bending.
186 HEAT Chap. 11

a strip of brass and another of steel are joined in one line; the total expansion
due to heating will be the sum of the individual expansions of the two strips.
The arrangement in part II of the figure shows these strips joined in another
way to produce differential expansion. By properly choosing the lengths of
the strips, the distance l between their free ends can be kept the same regard-
less of temperature changes. This is the result when the elongation ab Lb At
for the brass strip equals the elongation as Ls At for the steel strip over the
same temperature range At. In part III of the figure is shown schemati-
cally the left half of a clock pendulum made of steel and brass rods. In
order to keep the length L of the pendulum constant at all temperatures, the
lengths of brass and steel must be so proportioned that 2 ab l = a.(L + 2 l).
Part IV of the figure represents a bimetallic strip of brass and steel, riveted
or welded together side by side, their lengths being identical at some initial
temperature. The shape indicated results from heating because brass ex-
pands more than steel. It can be shown that this shape is approximately a
circular arc having a radius R = dj[(ab - a.) At], where d is the thickness
of the brass or the steel strip and At is the temperature elevation. The bend-
ing shown is made use of in thermostats and similar appliances to make and
break an electric contact and thereby control the operation of heaters or
other apparatus.
In order to measure the coefficient of linear expansion of a metal, a rod of
the material two or three feet long is placed centrally within a pipe fitted with
cork stoppers at the ends to form an enclosure through which cold water or
steam can be passed. The rod is allowed to project slightly at the ends so
that its length can be measured, and the measuring d6'Vice is equipped with
a micrometer for direct observation, or with a lever system to magnify the
change of length. From the length measurements made on the rod in cold
water and again in steam, together with the corresponding temperature
readings, a simple calculation gives the linear expansion coefficient of the
metal.
11-4. Surface and Volume Expansion of Solids. The expansion due to
heating affects all of the dimensions of a body. For an isotropic substance;
that is, one which has the same physical properties in all directions, an expan-
sion of 1 per cent in length is accompanied by an expansion of 1 per cent in
width and 1 per cent in thickness.
In order to compute the expansion of a surface, a rectangular plate of
(cold) dimensions a 0 and b0 is assumed to have its temperature raised by
an amount At. The (hot) dimensions then become ah = ac(l + a At) and
bh = bc(l + a At); consequently, the area of the surface at the higher temper-
ature will be Sh = ahbh = acbc(l + a At) 2 = Sc(l + 2 a At + a 2 At2 ), where
S 0 = acbc is the area of the plate at the lower temperature. The term with
the square of the small product a At is negligible with respect to the others,
and the final expression for the area of the plate becomes
Sh = Sc(l + 2 a 11t) = SAl + a' At),
§ l 1-5 EXPANSION OF LIQUIDS 187

where a' = 2 a is the coefficient of surface expansion of the materjal forming

the plate.
An interesting example of expansion is that of a square plate of metal on
which a smaller square is marked out to form a frame of uniform width.
Upon heating, the plate expands, and the inscribed square expands propor-
tionately. Had the inner square of metal been removed before heating, the
expansion of the frame for the same temperature elevation would have been
the same as though the plate "Yvere complete.
The expansion of a volume can be expressed in a manner similar to that of
a length or a surface. When a body of volume Ve is heated through a tem-
perature range JJ.t, its volume will become
(11.4)
where /3 is the coefficient of volume expansion (sometimes called the cubical
expansion coefficient) of the substance. The procedure above, if followed,
will show that, as a close approximation,
/3 = 3 a. (11.5)
It is important to remember this relation in seeking cubical coefficients of
solids in tables of physical constants, since usually only linear coefficients of
such substances are listed.
The cavity in a hollow object expands upon heating as though it were a
solid piece of the surrounding material. Thus, the inside of a hollow brass
ball would expand as though the space were filled with brass.
Suppose a liter flask of glass (a = 8 X 10-s;c0 ) to be correct at 0°C; find its volume
at 100°C. From Eq. 11.4, v\ = 1000 cm 3 X [1 + 3 X 8 X 10-s X (100 - O)J
= 1002.4 cm3 , an increase in volume of 2.4 cm3 •
11-5. Expansion of Liquids. In general, liquids follow the same law of
expansion that solids do. The volume of a liquid at temperature th in terms
of its volume at a lower temperature tc is given by
Vh = Vc(l + /3 JJ.t),
where /3 represents the coefficient of volume expansion of the liquid, and IJ.t is
the temperature difference (th - tc). Table III gives the coefficients of volume

TABLE III. EXPANSION CoEFFICIE:"<TS oF LIQUIDS

I _____ Per C Degree Per F Degree

i
Alcohol (ethyl) ................ · I 110 X 10-5 61 x 10-5
Alcohol (methyl) . . . . . ...
1
122 X 10-5 68 x 10·-5
Ether (ethyl)..... . . . . . . . . . . . . . . 163 X 10-5 91 X 10-5
Glycerine. . . . . . . . . . . . . . . . . 53 X 10-5 29 x 10-5
Knosene. . . . . . . . . . . . . . . . . . . . . . 83 X 10-5 46 X 10-5
.VIercury. . . . . . . . . . . . . . . . . . 18.17 X 10-5 10.09 X 10-5
Sulfuric acid. . . . . . . . . 57 X 10-5 32 X 10- 5
Turpentine.. . . . . 94 X 10-5 52 X 10-5
188 HEA'l' Chap.11

expansion for a few liquids, these being average values over ranges of tem-
perature around 20°C.
In computing the expansion of a liquid, it is usually necessary to remem-
ber that the container expands also when heated. A problem will show how
the expansion of the container is taken into account.
A steel drum is filled level full with 55 gal of kerosene at 60°F. If the temperature
rises to 90°F, how much of the liquid will overflow? For a temperature rise of 90 - 60
= 30 F 0 , Eq. 11.4 shows that the volume of the kerosene increases to Vk = 55 gal X
+
( 1 46 ~ 10- X 30 F 0 ) = 55.7590 gal. Meanwhile the volume of the drum in-
5
0

creases to V d = 55 gal ( 1 + 3 x 61Fox 10-5

X 30 F 0
)
= 55.0302 gal. The overflow
is the difference, Vk - Vd, between these values, namely: 55.7590 - 55.0302 = 0.7288
gal, which is equivalent to 168.4 in. 3
The density of a substance is affected by temperature changes, since the
volume varies with temperature and the mass does not. This fact is par-
ticularly important in dealing with fluids, for their expansion coefficients are
much larger than those of solids. At temperatures th and tc, the volumes of
a fluid of mass mare V h = m/dh and Ve = m/dc, respectively, where the cor-
responding densities are dh and de. From Eq. 11.4,
d de (11.6)
h = 1 +~tit'
which shows that when the temperature of a substance is raised, the density
is reduced.
M 13.60 J - - - - + - - + - - - - + - - + - M 1.000 t:,:a--j-.:;;;::-t----t--+
E
u
E
"-. 13.58 t-------t-'.....-+--t---+- } 0.999 i----t--+-----f'..-+-

-
E
C)

~ 13.56
·;;;
ai 13. 54
t----+--~--
,_I -~----+-----+-
-
·;;;
~
c
C)
~ 0.998 t----+--+--e---~

0.997 :-1----:-1--+--....J.l_---1._
O I I I I
I I I
0 5 10 15 20 0 5 10 15 20
Temperature, 0
( Temperature, 0
(

FIG. 11-3. Typical expan- FIG. 11-4. Anomalous expan-

sion of liquid with rise of sion of water; its density is
temperature, as evidenced by greatest at 4°C.
lowering of density. The
curve shows data for mer-
cury.

As the temperature is raised, most liquids expand at a uniform rate, and

the density falls off in the same proportion. Mercury is a typical example,
and its behavior is represented in Fig. 3; the density diminishes regularly
from 13.5955 gm/cm 3 at 0°C to 13.5462 gm/cm 3 at 20°c.
The behavior of water is exceptional, particularly at temperatures near
§ lJ-7 HEAT UNITS 189

0°C. As water is cooled from room temperature, it contracts fairly steadily

until its temperature reaches 4 °C; at this temperature water has its greatest
density. Upon further cooling to 0°C it expands and its density becomes
slightly less. This irregular expansion of water is represented in Fig. 4.
The density of water in grams per cubic centimeter is 0.99987 at 0°C, 1.00000
at 4 °C, and 0.99823 at 20°C. The anomalous behavior of water causes the
freezing of lakes and ponds to begin at the surface rather than at the bottom.
It is interesting to conjecture what would happen if the opposite were true.
A knowledge of the densities of liquids is of importance in making volume
calibrations. Thus, if mercury is used to calibrate a glass flask that is to
have a volume of 250 cm 3 at 20°C, it is necessary only to pour in 250 X 13.5462
= 3386.55 gm of mercury at that temperature and to place a mark on the
flask at the liquid surface. For accuracy, the shape of the flask should be
such that the mark will come at a position where the area is small.
11-6. Mercury Thermometry. Temperature measurements with the fa-
miliar mercury thermometer are based upon the expansion of mercury or,
more strictly, upon the difference between the expansion of mercury and
that of glass. The thermometer, in its simplest form, is made by fusing a
glass bulb to the lower end of a length of capillary glass tubing. Mercury is
added, the thermometer is boiled and annealed, and the capillary is sealed at
the top. Next, the thermometer is calibrated at two or more points, and,
finally, the graduations and figures are etched on the stem. Since ordinary
glass softens at about 400°C and mercury boils at 356.9°C, it is usual, for
measurements up to 500°C, to make thermometers of borosilicate glass and
to fill them above the mercury column ,vith an inert gas under pressure.
For low temperatures, the use of mercury is limited by its freezing point,
-38.87°C. In thermometers of similar construction the range can be
extended downward by using toluol or pentane instead of mercury; these
liquids freeze at -97°C and -200°C, respectively.
The clinical thermometer used by physicians has a constriction in the stem
near the bulb. Upon heating, the mercury expands through this tiny open-
ing and rises properly on the scale, but upon cooling, the mercury beyond the
constriction is prevented by surface tension from returning to the bulb.
This arrangement permits the indication to be read at any time after the
thermometer is used; vigorous shaking restores the mercury to the bulb.
11-7. Heat Units. In recognition of the fact that heat is a form of energy,
a given quantity of heat could be expressed in foot· pounds, joules, or ergs,
_and no special units for heat energy would be needed. Through the years,
however, such special units have been developed and are widely used. They
are based upon the change of temperature of definite quantities of water.
It is common experience that more heat is needed to bring a kettleful of
water to the boiling point than a cupful, both starting at the same low tem-
perature. Investigation shows that the quantity of heat required to produce
a given temperature rise is directly proportional to the amount of water
190 HEAT Chap. 11

heated. Again, to increase the temperature of any given amount of water

through, say, 60 degrees takes more heat than to raise the same amount
through 30 degrees; in fact, it takes twice as much. Thus, the heat required
is proportional to the mass of water and to the rise of temperature. In conse-
quence, the unit of heat is chosen as the quantity of heat needed to raise the
temperature of a unit mass of water through one degree.
Two units of heat are in common use: the calorie (cal) and the British
thermal unit (Btu). The calorie is the quantity of heat required to raise the
temperature of 1 gm of water through 1 centigrade degree. The British
thermal unit is the quantity of heat required to raise the temperature of l lb
of water through l fahrenheit degree. It will be noted that the pound is
used as a unit of mass in the study of heat. Since a mass of l lb = 453.6
gm, § 4-2, and since a temperature rise of l F 0 = a rise of i C 0 , it follows
that l Btu = 453.6 X i = 252 cal. Once the units of heat are estab-
lished, it is a simple matter to compute the amount of heat needed to pro-
duce a given temperature elevation in any mass of water. Thus, to raise
10 kg of water from 20°C to 70°C requires 10 X 1000 X 50 = 5 X 105 cal,
and to raise 50 lb of water from 40°F to 150°F requires 50 X llO = 5500 Btu.
Although the foregoing statements are sufficiently exact for most practical
purposes, actually, the quantity of heat required to raise unit mass of water
through one degree varies slightly from point to point along the thermometer
scale. The mean calorie is defined as the amount of heat required to raise
l gm of water from 14.5 to 15.5°C; this is almost exactly one one-hundredth
part of the heat required to raise 1 gm of water from Oto l00°C.
11-8. Transfer of Heat. Heat energy can be transmitted from one place
to another in three ways: they are known as conduction, convection, and
radiation.
Conduction. The method of heat transfer called conduction is a point-by-
point process in which one part of a body is heated by direct contact with a
source of heat, and neighboring parts become heated successively. Thus,
when one end of a metal rod is placed in a hot furnace, the heat travels along
the rod by conduction. This is commonly explained, for poor conductors of
electricity, by supposing that within the furnace the molecules of the rod are
in violent vibration and that in jostling their neighbors they set them into
more rapid motion, and this process continues throughout the rod; for good
conductors of electricity the process of heat conduction is carried on predomi-
nantly by the movements of free electrons. Good conductors of heat are
also good conductors of electricity; the free electrons that are present in such
materials play an important part in both types of conduction.
Convection. The transfer of heat by convection refers to the circulation of
warmed fluids and is brought about by the changes of density that accom-
pany changes of temperature. Water in a kettle on the kitchen stove is
heated by convection; the air in a room is heated with a hot stove by the
same process. In the gas water heater, Fig. 5, hot gases from the burner rise
§ 11-8 TRANSFER OF HEAT 191

in a helical path through a central flue and heat the water in the surrounding
tank. As the water near the flue is heated it becomes less dense and rises,
setting up convection currents as shown; in this way all the water in the tank
becomes heated. Hot water is drawn from the
top of the tank as needed, and cold water is sup- Hot water Cold water
plied to the bottom to replace it. ~ r
In the heating of houses, heating units are i ~ '2 ~

placed in the rooms, and the air which comes in ( ~ r l Insulation

~

contact with them circulates the heat by convec- tt tt

tion. The so-called steam or hot-water radiator t t ti
does transfer some heat by radiation, but it trans- t t t t Flue
fers far more by convection. The draft in a chim- t t t i ·.
ney is produced by convection; the heated gases lJ ..........
,___.. \:~ ......- Burner
weigh less than a corresponding column of cold -
air; consequently, the gas in the chimney is buoyed
up in accordance with Archimedes' Principle, as Frn. 11-5. Convect10n cur-
explained in Chap. 9. rents in a gas-fired water
The trade winds are caused hy the continued heater.
heating of the air near the equator and the move-
ment of the air away from the earth's surface. Cooler air from the tropical
belts rushes to take its place, flowing from the north in northern latitudes
and from the south in southern latitudes, and convection currents are set up
which blow in the same direction for long periods. The rotation of the earth
causes some deflection of these winds from the directions mentioned.
Radiation. The transfer of heat by radiant energy is an illustration of
radiation. The process involves the conversion of internal energy into radiant
form at the heater, the transmission of the energy by waves through space,
and the reversion of radiant energy into internal energy wherever the radia-
tion is absorbed. The electric radiator is an appliance operating on this
principle; it has a heating element mounted at the focus of a parabolic reflec-
tor. When heated by an electric current, the element emits not only light
·waves that stimulate the eye, but also infrared waves that do not. When a
body absorbs waves of either kind, there is an increase in its internal energy
that produces the various effects mentioned at the beginning of this chapter.
The purpose of the reflector is to concentrate the waves into a beam, just as
a searchlight reflector does. Radiation impinging on a body may be re-
flected from it, absorbed by it, or transmitted through it, and these effects
may occur singly or in combination. Bodies that absorb radiation become
heated, and their internal energy is increased.
All bodies radiate energy, whether they are hot or cold; the hotter they are
the greater will be the radiation. Furthermore, all bodies receive radiation
from others. This exchange of radiant energy goes on continuously. Accord-
ingly, a body that remains at constant temperature is considered not as
having stopped radiating, but rather as receiving energy at the same rate
192 JIBA'l' Chap. 11

that it is giving off energy by radiation. A body that is a good radiator of

energy is also a good absorber of energy. A black rough surface, such as that
provided by a coating of lampblack, is an excellent radiator (as well as an
excellent absorber) of radiant energy. A highly polished surface has opposite
characteristics.
Radiation is the process by which energy reaches the earth from the sun.
In full sunlight, the energy received above the earth's atmosphere is about
2 cal per min on a surface 1 cm square held perpendicular to the sun's rays;
about one-third of this amount is absorbed as the waves pass through the
atmosphere. The waves that transmit radiant energy, such as light and
electric waves, §§ 25-9, are electromagnetic waves that pass through the
vacuum which exists over most of the distance from sun to earth.
11-9. Heat Conduction. Substances differ widely in their ability to con-
duct heat from one point to another; metals are relatively good conductors,
whereas porous substances in which air is entrapped are poor conductors or
good insulators. For a given material, the amount of heat conducted per
unit time along a specimen of uniform cross-section depends upon the linear
temperature distribution and upon the area of the section. The temperature
gradient is the ratio of the temperature difference between two points along
the line of heat flow to the separation of these points; for a temperature dif-
ference tit over a distance iiL the temperature gradient is

G . t h e 1·1m1t,
= -fit , or, 1n . dt
G=-·
iiL dL
If the sectional area perpendicular to the passage of heat is A and if the rate
of heat transfer is represented by the single symbol q, then, in any substance,
q = kGA, (11.7)
where the proportionality constant k is called the thermal conductivity of that
substance.
The unit for thermal conductivity depends upon the units used for the
other quantities in Eq. 11.7. In metric units, if the rate of heat transfer is
to be expressed in calories per second with the temperature gradient in cen-
tigrade degrees per centimeter and the area in square centimeters, then
cal
§ 11-9 HEAT CONDUCTION 193

ducted in 1 hr through a specimen 1 f t 2 in cross-section and 1 in. thick, when

the opposite faces are kept at a temperature difference of 1 F Average
0
•

values of thermal conductivities of a number of substances at ordinary tem-

peratures in metric and British units are listed in Table IV.

TABLE IV. THERMAL CONDUCTIVITIES (k)

Cal Btu·in.
Substance Sec·cm· C 0 Hr· ft2· F 0

Air .................. . 0.000054 0.16

Aluminum ............ . 0.49 1400
Brass ................ . 0.26 750
Cement .............. . 0.0007 2.1
Copper ............... . 0.91 2600
Cork ................. . 0.0001 0.29
Cotton ............... . 0.0005 1.5
Glass ................ . 0.002 5.8
Ice .................. . 0.005 15
Iron ................. . 0.15 430
Silver ................ . 0.99 2900
Slate ................. . 0.005 15
Water ................ . 0.0015 4.3

In order to measure thermal conductivity, it is necessary to determine the

rate at which heat passes through a specimen of given cross-section under a
known temperature gradient. In testing metals, since their conductivity is
relatively high, satisfactory precision can be attained by using specimens in
the form of rods. In the usual method, the rod under test is fitted with a
steam jacket at one end and with a coil of several turns of small metal tubing
wound around the other end. Between these parts two thermometers are
placed in good thermal contact with the rod and located a known distance
apart. Thermometers are also used to measure the temperature of water
entering and leaving the coil of tubing. The entire rod is surrounded by
suitable insulation to reduce emission of heat from the surface of the rod.
When steam is passed through the jacket and all thermometers have attained
steady readings, the test may be conducted, for then the heat supplied to the
rod is conducted steadily along its length and transferred to the water.
In a typical experiment, a copper rod having a cross-section of 10 cm2 has thermal
contacts 12 cm apart. When one end is heated and a steady rate of heat flow is estab-
lished, the thermometers indicate temperatures of 78 and 52°C at these contacts. At
the other end 3.9 gm of water are heated each second through 5.0 C0 • The average
temperature gradient over the rod length under test is G = (78 - 52)/12 = 2.17
C0 /cm, and the rate of heat transmission along the rod is q = 5.0 X 3.9 = 19.5 cal/sec.
Therefore, the heat conductivity of copper, as given by this test, is k = q/GA
= 19.5/(2.17 X 10) = 0.90 cal/(sec·cm·C 0 ) .

In measuring the thermal conductivities of poor conductors of heat, accu-

racy requires that specimens, if solid, be in the form of sheets so that the heat
194 HEAT Chap. 11

path will be short and of large cross-sectional area, but in other respects the
method of test is the same as for good conductors.
Liquids and gases are poor heat conductors. Air confined in small cells is
a good insulator, because its thermal conductivity is low and because its circu-
lation is limited. Textiles such as wool and felt are good heat insulators,
primarily because of the air entrapped in them. Sawdust, granulated cork,
loose asbestos, and similar porous materials are used in the walls of refrigera-
tors, frame buildings, and ice houses for insulation purposes.
11-10. Heat Conduction Through Cylindrical Walls. The escape of heat
from bare steam pipes, or through insulating coverings around them, and
the heat transfer through boiler tubes or refrigerator piping illustrate the
conduction of heat radially through the wall of a hollow cylinder. If the
wall is imagined to consist of a series of concentric tubes, it will be clear that
the heat flow takes place through layers of varying cross-sectional area. 11 he
temperature difference across such a tube can be stated
in terms of the rate of heat conduction if the tube is
regarded as infinitesimally thin. Then, by integration,
the total temperature difference can be evaluated, and
the rate of heat flow determined.
A pipe, shown in cross-section in Fig. 6, may be sup-
posed to carry steam at a temperature t1, the surrounding
temperature being t2. If dt represents the temperature
FIG. 11-6. Diagram difference over a tube of radius rand infinitesimal thick-
used in calculating the ness dr, the gradient at the tube becomes G = dt/dr.
conduction of heat
The rate of heat conduction through the tube is obtained
through a cylindrical
wall. from Eq. 11.7 as q = -k (dt/dr) A, where k is the
thermal conductivity of the wall material and A is the
area normal to the flow of heat, which for a cylinder of length L is 2 1rrL.
The minus sign is used because, in this example, the quantity dt/ dr is inher-
ently negative: an increase in radius corresponds to a decrease in temperature.
The temperature difference across the tube is, consequently,

q dr
dt =
2 1rrLk

The temperature differences across all the tubes from the inner radius r1
to the outer radius r 2 are summed up to give the temperature difference be-
tween the steam and the surroundings; thus,

_g__ r2
\'irhencc f2 - f1 = - ln -. It follows that the heat conducted through
2 1rLk r1
§ 11-11 EFFECT OF SURFACE FILMS ON HEAT :l'RANSMISSION 195

the walls per unit time is

2 1rLk (ti - t2)
q = In (r2/ri) '
where In indicates the natural or N apierian logarithm.
11-11. Effect of Surface Films on Heat Transmission. In many applica-
tions, heat passes through solid walls by conduction and is transmitted fur-
ther through gases or liquids by convection. In such instances it is necessary
to take account of thin fluid films which adhere to the wall surfaces. These
films transfer heat principally by conduction; and, since fluids as a class have
low conductivities, the films reduce the transfer of heat very materially.
The effects of such films are allowed for in heat calculations by the use of
so-called film coefficients. The determination of these coefficients is beyond
the scope of this book, but it will be of value to know what they mean and
how they are used.
The film coefficient is the ratio of the thermal conductivity of the film
material to the equivalent thickness of the film. Equation 11.7, applied to
the film, becomes q = h Llt A, where h = k/L is the film coefficient.
In the passage of heat through a wall separating two fluids, the heat really
passes through three layers in succession: first, the film on wall surface; next,
the wall itself; and third, the film on the other side. After a steady state is
reached, the rate of heat transfer must be the same for each layer, or

q -_ k i Llti A -_ tC9
1 Llt2 A -_ 1C3
1 /::.t3 A ,
Li ~ L2 L3
where subscripts 1 and 3 refer to the films and 2 to the wall. The temperature
differences across the successive layers are then evaluated and added to give
the overall temperature difference Llt, and from the resulting expression, the
rate of heat flmv is found to be
A Llt (11.8)
q=
1 +L2 + _1'
hi k2 hs
an express10n that reduces to the simpler form of Eq. 11.7 if no films are
present.
As an example, the rate of heat conduction through a plate-glass window having an
area of 15 ft 2 and a thickness of tin. is computed for a temperature of 70°F on the
inside and 20°F on the outside. The film coefficient of the air on either side is found
by experience to be about 2 Btu/(hr·ft2 ·F 0 ) , and the thermal conductivity of glass is
given in Table IV. From these data, the rate of heat conduction through the window,
as calculated from Eq. 11.8, is 718 Btu/hr. If the air films were neglected, the rate
of conduction, by Eq. 11.7, would be 17,400 Btu/hr.

Surface films may reduce the transfer of heat through a wall very greatly,
as the foregoing results show. The recognition of such films is of great impor-
tance in the design of heat exchangers and other apparatus.
196 HEAT Chap. 11

PROBLEMS
1. Comfortable room temperature is 68°F. Express this temperature on the centi-
grade, kelvin, and rankine scales.
2. The temperature difference between the freezing point and boiling point of
mercury is 395.77 C 0 • Express this temperature diffe.rence on the fahrenheit, kelvin,
and rankine scales.
3. At what temperature do the readings of a centigrade and a fahrenheit ther-
mometer have the same value?
4. A wrought-iron steam pipe extends vertically a height of 80 ft. How much does
its length increase when its temperature rises from 40 to 190°:F?
5. A steel rod having a cross-section of 1 in. 2 is cut to a length of 4.002 in. at 60°F.
Find its length when it is cooled to -320°:F by immersion in liquid nitrogen. Assume
that the coefficients listed in Table II apply over this temperature range.
6. A steel tape that is correct at 0°C is used to measure a parcel of land when the
temperature is 29°0. What is the correct length of a line for which the tape reading
is 47.22 ft?
7. With what force must a steel tape i in. wide and 0.015 in. thick be pulled in
order to compensate for a temperature drop of 20 F 0 ?
8. If the rod mentioned in Prob. 5, while still at a temperature of -320°F, is
placed between fixed supports 4 in. apart, what force will it exert upon the supports
when its temperature rises to 60°F?
9. A steel scale that is correct at 0°C is used to measure the length of an aluminum
rod. The scale reading is 90.00 cm at 10°0; what would it be when the measurement
is made at 30°0?
10. In order to fasten a pulley to a steel shaft, the shaft is made oversize and is then
shrunk by cooling until it fits the hole in the pulley. To what size should the shaft
be turned down at 70°:F in order that when the shaft is packed in "dry ice" at -108°:F
its diameter will be exactly 1 in. for insertion in the pulley? Assume the linear co-
efficient of expansion for steel to be 6.1 X 10-6/F 0 over the temperature range involved.
11. An aluminum cylinder has a length of 1.25 rn and a cross-sectional area of 8 crn2
at 0°C. Compute the length, cross-section, and volume of the cylinder at 40°0.
12. Iron is poured into a spherical mold 3 in. in diameter at its melting point, 1530°0.
In solidifying it shrinks 3 per cent in volume, and as it cools to 20°0 it contracts further.
If a for cast iron has the average value of 16 X 10-s;c0 over this temperature range,
what is the volume of the casting at 20°C?
13. A round hole in a steel plate is exactly 4 in. in diameter at 20°0, and a copper
sphere is 3.99 in. in diameter at the same temperature. To what temperature should
both pieces be heated in order that the sphere shall have the same diameter as the
hole? Assume that the values in Table II apply over the temperature range involved.
14. An open steel drum is filled with 55 gal of kerosene at 80°F. If the temperature
falls to 40°F, how much kerosene can be added?
15. A Pyrex glass flask is filled level full with 400 crn3 of glycerine when the temper-
ature is 15°0. How much glycerine will overflow if the temperature rises to 60°C?
16. A crude thermometer can be made by using a glass quart bottle, inserting
through the stopper a small glass tube, and pouring in sufficient kerosene to extend
just into the tube. If the tube has an internal diameter oft in., how much will the
liquid column rise when the temperature goes up 10 F 0 ? Take the coefficient of linear
expansion of glass to be 5 X 10-s/F 0 , and neglect the expansion of the tube because
of its relatively small size.
17. Compute the value of the specific gravity of glycerine at 60°C.
18. Compute the quantity of heat needed to raise the temperature of (a) 1 ft 3 of
water from 40 to 90°:F, and (b) 1 liter of water from O to 80°(;,
 PROBLEMS 197

19. How many British thermal units are needed to heat 20 gal of water from 60 to
150°F?
20. A refrigerator room measures 6 X 12 X 9 ft, and all its walls are insulated
with Foamglas 6 in. thick. The thermal conductivity of this material is 0.46
Btu· in./ (hr· ft2 · F 0 ) . Assume the cross-sectional area of the insulation to be the same
as the surface of the enclosure, and compute the amount of heat conducted through
the walls per hour when the temperature is 32°F inside and 80°F outside.
21. A solid plane wall is constructed of an outer layer having a thickness L1 and a
thermal conductivity k1 , and an inner layer having a thickness L2 and a thermal
conductivity k2 • If the temperature of the outside of the wall is To and that of the
inside of the wall is Ti, show that under conditions of steady heat conduction the tem-
perature at the junction of the two layers is (k1 Ta/L1 + k2 T/L2)/(k1/L1 + k2/L2).
22. A cubical enclosure measuring 1 m along each edge is lined with cork 2 cm thick.
An electric lamp left on continuously keeps the interior at 37°C when the temperature
outside is 20°C. Assume the cross-sectional area of the lining to be the same as the
surface of the enclosure, and determine how many calories per second the lamp must
supply to offset the loss by conduction.
23. A steam pipe with an outside diameter of 2.5 in. is surrounded by a corrugated
asbestos covering 1 in. thick. When the temperatures of the outside surface of the
pipe and of the covering are respectively 190 and 75°F, how much heat is conducted
through the covering per hour for a 30-ft length of pipe? The thermal conductivity
of the covering is 0.6 Btu·in./(hr·ft2 ·F0 ).
 12
HEAT QUANTITIES AND
CHANGE OF PHASE
The application of heat increases the internal energy of a body, and may
produce either a rise of temperature or a change of phase. The first of these
is considered in the preceding chapter, and the second in this one. With the
units of heat established, it is possible to treat both effects quantitatively, and
numerous problems are presented to show just how much heat is needed in
particular instances to bring about the desired change. The principles in-
volved are basic and are used repeatedly in the further study of Heat.
12-1. Heat Required to Change Temperature. Unit quantity of heat, as
explained in § 11-7, is sufficient to raise the temperature of a unit mass of
water one degree; there are two of these heat units: the calorie and the Brit-
ish thermal unit. One calorie is sufficient to raise 1 gm of water through
1 C 0 , and 1 Btu to raise 1 lb of water through 1 F 0 • It must not be thought,
however, that a calorie of heat will raise 1 gm of every substance through a
I-degree interval. Roughly, only 1\ cal is needed to raise 1 gm of copper
through 1 C 0 ; Tltr cal ,vill suffice for platinum, and so on. The heat needed
is a characteristic of the material and gives the value of its specific heat. The
specific heat of a substance is the ratio of the heat supplied per unit mass of the
,'!11h,'!tance to the resulting temperature rise. Thus, if Q cal supplied to a mass m
of a substance cause a temperature rise t, the specific heat of the substance is

c=--·
Q/m
t
The specific heat is numerically equal to the number of calories needed to
raise 1 gm of a substance through 1 C 0 , or, what amounts to the same thing,
the number of British thermal units needed to raise 1 lb of it through 1 F 0 •
The value of the specific heat varies slightly with temperature; thus, for
copper, at 20°C it is 0.093 cal/(gm·C 0 ) or 0.093 Btu/(lb·F0 ) , but at 200°C itt:,
value is 0.096. The specific heats of some solids and liquids are given iu
Table I. These are average values for the temperature ranges listed and the
units are those just mentioned.
From the definition of specific heat it follows that in order to raise a mass
m of a substance having a specific heat c through a temperature rise t, the
198
§ 12-2 CALORIMETRY 199

TABLE I. SPECIFIC HEATS

Temperature, Specific
SubstancP oc Heat
- - - - - - - - - - - - - - -- - - - - - - - - 1 - - - - -
Alcohol (ethyl) .. . 0 - 40 0.59
Aluminum ....... . 0 - 100 0.21
Copper ................... . 0 - 100 0.093
Glass, ordinary ............ . 10 - 50 0.14
Ice ...................... . -20 - 0 0.50
Iron ..................... . 0 - 1000 0.15
Iron ..................... . 0 - 100 0.11
Lead ..................... . 0 - 300 0.032
Mercury ................. . 0 - 100 0.033
Platinum ................. . 20 - 100 0.032

quantity of heat required is

Q = met. (12.1)
This expression also shows the quantity of heat which the body must give up
to lower its temperature by t degrees.
For example: To raise 100 gm of copper from 10° to 70°C requires 100 X 0.093 X 60
= 558 cal. Again, 20 lb of iron in cooling from 100° to 40°F will give off
20 X 0.11 X 60 = 132 Btu.
12-2. Calorimetry. The measurement of heat quantities, called calorim-
etry, is usually carried out by the method of mixtures. In order to apply the
method to two objects, initially at different temperatures, the two are "mixed"
to bring them into good thermal contact with each other, whereupon heat is
transferred from the hot body to the cold body, and the system comprising
the two objects settles to a condition of equilibrium at some temperature
which is uniform throughout. It is important in this process that no heat
be transferred to or from the surroundings; if this condition is satisfied it
follows that
(1) the internal energy of the system, although distributed differently among its
parts, remains constant in amount;
(2) the heat given off by the hot body is equal to that taken on by the cold body.
These two concepts provide two distinct procedures for using the data to
determine some unknown factor in the measurement.
Suppose it is desired to find the resulting temperature when 5 lb of water at 180°F
are mixed with 3 lb of water at 60°F.
By the total-energy concept, the internal energy before mixture is equated to that
after mixture, the energy values being reckoned from any desired datum. In this
problem the energy values, reckoned from an arbitrary datum of 50°F, are as follows:
Before mixture
Hot water 5 X (180 - 50) = 5 X 130 = 650 Btu
Cold water 3 X (60 - 50) = 3 X 10 = 30 Btu
Total 680 Btu
200 HE.AT Chap. 12

After mixture, if the resulting temperature is x, the internal energy of the total
amount of water, reckoned from the same datum, is (5 + 3) X (x - 50) Btu. Since
the total energy is constant, 8 X (x - 50) = 680, whence the temperature of the
mixture is x = 135°F.
By the heat-transfer concept, the resulting temperature x will be such that the heat
given off by the 5 lb of water in cooling from 180° to x equals that absorbed by the 3 lb
of water in rising from 60° to x. The heat equation in British thermal units follows:
5 X (180 - x) = 3 X (x - 60),
from which the temperature of the mixture is found to be 135°F, as before.
12-3. Measurement of Specific Heat. The method of mixtures can be
used to determine the specific heat of a substance; a common form of appa-
ratus for this purpose is shown in Fig. 1. A test sample of known mass is
heated in a steam-jacketed compartment to a measured high temperature;
the calorimeter, a heat-insulated metal vessel containing a known amount
of water at a known temperature, is pushed directly under the heated com-
partment and the sample is allowed to drop into it; then the calorimeter is
quickly moved away to prevent its absorbing heat from the jacket, and the
resulting temperature of the mixture is noted. From the data thus obtained,
the specific heat of the test sample can be computed by either of the fore-
going concepts. The heat quantities involved are expressed by means of
Eq. 12.1.
In a particular test, a 300-gm sample of unknown specific heat c was heated to 98.5°C
and dropped into 600 gm of water at 19.0°C contained in a 140-gm aluminum calorime-
ter, and the resulting temperature was found to be 22.4°C. Determine the specific
heat of the sample.
By the total-energy concept, the internal energy values in calories, reckoned from
water atQ°C as a datum, are as follows:

Before Mixture After Mixture

Sample .... . 300 X c X (98.5 - O) = 29,550 c 300 X c X (22.4 - 0) = 6,720 c
Water ..... . 600 X 1 X (19.0 - 0) = 11,400 600 X 1 X (22.4 - 0) = 13,440
Calorimeter 140 X 0.21 X (19.0 - 0) = 559 140 X 0.21 X (22.4 - 0) = 659
Total. ..... 29,550 c + 11,959 6, 720 c + 14,099

Since the internal energy is constant,

29,500 c + 11,959 = 7620 c + 14,099,
from which c = 0.094 cal/(gm·C 0 ) .
By the heat-transfer concept, the heat given off by the sample is equated to that
taken on by the water and calorimeter; thus,
300 X c X (98.5 - 22.4) = 600 X 1 X (22.4 - 19) + 140 X 0.21 X (22.4 - 19.0)
= (600 + 29.4) X 3.4,
from which the specific heat of the sample is c = 0.094 cal/ (gm· C 0 ) , as before.
The water equivalent of an object is the product of its mass and its specific
heat, and is often useful in calorimetric computations. In the foregoing
§ 12-5 1tlJiJLTING AND FREEZINO 201

illustration, the 140-gm calorimeter is equivalent to 140 X 0.21 = 29.4 gm

of water. This statement means that 29.4 gm of water would undergo the
same temperature rise as the 140-gm aluminum calorimeter for an equal
absorption of heat.

Calorimeter

c D
~100 -----
Steam
~-
.2
jacket ...
0
(!)
0..
E
(!) OA
~
B

Time

Frn. 12-1. Apparatus used in measuring spe- Frn. 12-2. The effect of applying
cific heat. heat to ice.

12-4. Change of Phase. Although it is true that, when an object is heated,

its temperature usually rises, there are instances where this does not occur.
For example, if heat is applied to a block of ice at 0°C, the ice melts and the
resulting water, remaining in contact with the ice, will stay at a temperature
of 0°C until all the ice is melted. Again, if heat is applied to some water at
100°C, the water boils and its temperature remains constant until continued
heating converts it all to steam. From these examples it is seen that the
absorption of heat does not always cause a temperature rise but may serve
instead to change the phase of a substance from a solid to a liquid or from a
liquid to a vapor. The change of phase from ice to water to steam is repre-
sented on a temperature chart in Fig. 2. The horizontal line AB indicates
the heating of a certain mass of ice at 0°C, the inclined line BC indicates the
heating of the resulting water with rise of temperature from 0° to 100°C, and
the horizontal line CD indicates the heating of water at its boiling point. It
should be mentioned that ice melts at 0°C and water boils at 100°C only
when the pressure has the standard value of 76 cm Hg, the boiling point
especially being influenced by the pressure.
12-5. Melting and Freezing. The change of phase which occurs when
matter is transformed from a solid to a liquid is called melting or fusion.
When ice melts, the change of phase occurs at a fixed temperature, provided
the pressure remains constant. The property of having a fixed melting point
is generally characteristic of crystalline substances; in contrast, amorphous
substances, such as tar and glass, pass imperceptibly from one phase to the
202 HEAT Chap. 12

other. During the process of melting the heat energy supplied is used in
separating the molecules against the forces of cohesion, and, in consequence,
the melted substance has a greater internal energy than the same substance
in the solid phase. Fusion is the opposite of freezing, or solidification. A
solid which has been heated to its melting point will melt at that temperature
upon the further application of heat; the corresponding liquid when cooled to
this same temperature will freeze if more heat is taken away. The melting
(or freezing) points of a number of substances at standard atmospheric pres-
sure are given in Table II.
TABLE IL 11ELTING POINTS

Substance oc
Aluminum ............ . 660 1220
Copper ............... . 1083 1981
Hydrogen ............ . -259 -434
Ice .................. . 0 32
Iron ................. . 1535 2795
Lead ................. . 327 621
l\1:ercury ............. . -38.87 -37.97
Platinum ............. . 1755 3191
Tin .................. . 232 450
Tungsten ............. . 3370 6098

Most liquids contract in the process of freezing, but there are some impor-
tant exceptions. Water undergoes a remarkable expansion when freezing,
ice at 0°C having a density only 0.91 that of water at the same temperature.
This expansion accounts for the bursting of water pipes when the water in
them freezes. Type metal, an alloy containing antimony as the principal
constituent, also expands upon freezing (solidifying); this action fills the
mold and produces a good casting.
The presence of impurities usually has the effect of lowering the melting
point of a substance. In calibrating thermometers at 0°C by immersion in
a mixture of ice and water, errors from this cause are avoided by using dis-
tilled water. A useful freezing mixture is formed by adding table salt to ice;
with 1 part of salt to 2 parts of ice by weight, the melting point is lowered
from 32° to about 0°F.
12-6. Heat Required to Melt a Solid. Melting is a process that requires
heat. If a substance is melting and no heat is supplied to it intentionally, it
must be absorbing heat from its surroundings; for this reason melting is
sometimes referred to as a cooling process. Conversely, a liquid in freezing
must give up heat to its surroundings, and so freezing is sometimes called a
heating process. It is reported that this latter fact is utilized in fruit cellars
by placing large vats of water near the fruit; should the temperature fall dan-
gerously low, the water would freeze, and the consequent liberation of heat
would hold the temperature near 0°C and help to prevent freezing of the fruit.
§ 12-6 HEAT REQUIRED TO MELT A SOLID 203

The quantity of heat per unit mass needed to melt a solid without change
of temperature is called the heat of fusion. If the heat of fusion of a solid is
represented by L, then the quantity of heat Q needed to melt a mass mat con-
stant temperature will be
Q=mL. (12.2)

This expression shows the number of calories which must be transferred to

m gm of the solid to melt it, or transferred from m gm of the corresponding
liquid to solidify it, where the heat of fusion L is expressed in calories per
gram. If mis expressed in pounds, then L must be in British thermal units
per pound in order that Q may be in British thermal units.
Experiment shows that at standard atmospheric pressure, 1 gm of ice in
melting absorbs 80 cal from its surroundings; conversely, 1 gm of water in
freezing gives up 80 cal to its surroundings. Hence, the heat of fusion of ice
is 80 cal per gm; this value is equivalent to ! X 80 = 144 Btu per lb. The
heat of fusion of ice can be measured by observing the temperature change
when a definite amount of ice is dropped mto a known amount of hot water,
and using either the total-energy or the heat-transfer concept.
In a typical test, some surface-dried ice at 0°C is dropped into a calorimeter con-
taining water at 24°C; the mass of water, including the water equivalent of the calo-
rimeter, is 300 gm. After all the ice melts, the temperature is 16°C and the calorimeter
is found to contain 25 gm of water more than previously.
By the total-energy concept, the internal energy values in calories, reckoned from
water at 0°C as a datum, are as follows:

Before Mixture After Mixture

Water 300 X 24 = 7200 Water 300 X 16 = 4800

Ice -25 X L = -25 L Melted ice 2,5 X 16 = 400
Total 7200 - 25 L Total 5200

Therefore, 7200 - 25 L = 5200, and the heat of fusion of ice is L = (7200 - 5200)/25
= 80 cal/gm.
Table III gives the values of the heat of fusion for a few sub~tances, all at
the standard pressure of 1 atmosphere.

TABLE III. HEATS OF FUSION

---------------
Substance Cal/gm Btu/lb

Aluminum... . . . . . . . . . . 71 128
Copper. . . . . . . . . . . . . . . . 43.3 77.9
Ice................... 80 144
Lead.................. 5.4 9.7
Mercury. . . . . . . . . . . . . . 2.8 5.0
204 HEAT Chap. 12

12-7. Effect of Pressure on Freezing. It has been stated that the freezing
point of a liquid is influenced slightly by the pressure to which it is subjected.
For liquids which contract upon freezing, an increase of pressure raises the
freezing point; for liquids which expand upon freezing, such as water, an
increase of pressure lowers the freezing point. This property follows from
the fact that pressure on a body tends to prevent its expansion. Conse-
quently, for a liquid that expands upon freezing, an increase of pressure,
which tends to prevent it from expanding, will also tend to prevent it from
freezing; that is, will lower the freezing point.
The behavior of water is represented graphically in Fig. 3, in which pres-
sure is plotted against temperature. The curve slopes sharply upward to
the left and a pressure increase of 100 atmospheres lowers the freezing point

Ice
Ol
:r: Wire
E
u liquid
~-
:)
Solid Ice
V)
V)

,._
Cl)

0.. B

0
Temperature, 0
(

Fm. 12-3. The influence Frn. 12-4. Regelation

of pressure on the freezing experiment.
point of water. For most
liquids the curve slopes
the other way.

by less than 1 C 0 • Most liquids behave in the opposite manner; that is, an
increase in pressure raises the freezing point. A curve such as that shown
represents a state of equilibrium between the solid and liquid phases of the
substance under consideration. Thus, if point A on the curve represents a
mixture of ice and water at 0°C and 76 cm Hg pressure, then, in order to
raise the temperature of the mixture at this pressure to temperature B, it
would be necessary to supply enough heat first to melt all of the ice present
and then to produce the desired temperature i'.ise. Similarly, to lower the
temperature to temperature C, enough heat must be transferred from it first
to freeze all of the water and then to lower the temperature. In the diagram,
therefore, the region to the right of the curve represents the liquid phase and
that to the left the solid phase. The line itself shows the relative conditions
of temperature and pressure under which ice and water can coexist in equi-
librium.
The effect of increased pressure in lowering tho melting point of ice is illus-
§ 12-8 THE PROCESS OF EVAPORATION 205

trated by a classical experiment in which a wire melts its way through a cake
of ice, leaving the ice intact, a process known as regelation. The wire is
placed over the ice and weighted at the ends, as shown in Fig. 4. The
experiment is explained as follows: The pressure beneath the wire is sub-
stantially greater than atmospheric and causes the melting point of the ice
there to be slightly below 0°C. But, since that ice is at 0°C, it is momentarily
above its melting point and a little of it melts, causing a slight reduction of
temperature; the wire settles down and the water formed is squeezed into the
region above it. Here the pressure is standard and the water, which is now
slightly below 0°C, freezes again. During the process the heat given off by
the water in freezing flows downward through the wire and serves to melt the
ice below.
12-8. The Process of Evaporation. The changing of a substance to the
vapor phase is a process called vaporization. This is a general term and
includes: (1) evaporation, vvhich is a conversion from the liquid to the vapor
phase that occurs at the surface; (2) boiling, which is similar to evaporation
but which takes place throughout the interior of the liquid; and (3) sublima-
tion, which is a conversion from the solid directly to the vapor phase, without
passage through the liquid phase. Boiling and sublimation are considered
later in this chapter.
Evaporation goes on at all temperatures, and continues until the liquid
disappears or until the space above the liquid has become saturated with
vapor. In the process of evaporation a liquid is gradually transformed to a
vapor by loss of molecules at its surface. Since the molecules of the liquid
are moving in all directions and with various speeds, some will approach the
liquid surface with sufficient speed to carry them beyond the range of attrac-
tion of the surface layer; these molecules will leave the liquid and become
molecules of vapor.
It will be inferred from this explanation of evaporation that the molecules
which succeed in escaping through the surface are those with the higher
speeds and that, consequently, the average molecular speed in the liquid is
lessened. Such is the case, for evaporation lowers the temperature of the
liquid. As examples, water in a porous jar is cooled by evaporation through
the walls, a person's skin is cooled by the evaporation of perspiration, and
an invalid uncomfortably warm because of fever finds relief when sponged
with alcohol.

FIG. 12-5. Vibration of India-ink drop on a hot plate, photographed from above at speeds
high enough to "freeze" the vibratory motion.
(Courtesy of N. J, Holter and W.R. Glasscock of the Holter Research Foundation).
200 HEAT Chap. 12

An interesting effect of evaporation can be observed in a drop of liquid

resting on a level hot plate. ,vhen the plate is sufficiently heated the drop
actually floats on a film of its own vapor and is found to undergo a series of
radial vibrations in a horizontal plane which cause it to assume such patterns
as shown in Fig. 5.
12-9. Saturated Vapor. A saturated vapor can be produced by a.llowing
a liquid to evaporate into a confined space. As evaporation continues, more
and more molecules leave the liquid and assume the vapor phase. These
molecules of vapor are also in motion, and some of them strike the liquid
surface and return to the liquid. The number which re-enter will increase
as evaporation continues, until finally a state of equi-
librium is reached wherein the molecules pass out of
the liquid and back into it at the same rate; the space
above the liquid is then saturated with the vapor.
Ordinarily the conditions necessary for the production
Frn. 12-6. The produc- of saturated vapor are: (1) the presence of the generating
tionof a saturated vapor. liquid, and (2) a confined space. These are illustrated
in Fig. 6. In the evaporation process, the vapor above
the liquid becomes more and more dense and the pressure due to this vapor
becomes greater, reaching a maximum value at the point of saturation. The
maximum vapor pressure of a liquid is called the saturated vapor pressure.
Thus, for water at 20°C the vapor pressure would build up to a maximum
value of 1.75 cm Hg, which is the saturated vapor pressure of water at this
temperature.
The value of the saturated vapor pressure is practically
unaffected by the presence of other gases, provided no chem-
ical reactions occur. Thus, for water at 20°C, if the evapo-
ration takes place in a vacuum, the pressure builds up to the
value 1.75 cm Hg; but, if the enclosure contains dry air at
the standard pressure of 76.0 cm Hg, the pressure builds up
to an absolute value of 76.0 + 1.75 or 77.75 cm Hg. Of
this total, 76.0 cm is the partial pressure due to dry air and
1.75 is the partial pressure due to the water vapor, § 9-6.
Evaporation continues, regardless of other gases, until a Frn. 12-7. Dia-
partial pressure due to the vapor is established which is gram used in ex-
plaining that the
equal to the saturated vapor pressure of the liquid at the pressure of a sat-
existing temperature. urated vapor is
In describing the production of a saturated vapor, no independent of
reference was made to the size of the enclosure in which the volume.
evaporation occurs. The various effects mentioned take
place whatever the size of the enclosure, but the larger the enclosure, the
1onger the time required for saturation. The pressure of a saturated vapor
is, then, independent of the volume occupied by the vapor. If the evapora-
tion were to take place in an otherwise empty cylinder of variable volume,
§ 12-9 SATURATED VAPOR 207

as in Fig. 7, the piston might be moved slowly up or down without affecting

the pressure in the cylinder. These considerations show that it is impossible
to change the pressure of a saturated vapor by varying its volume. Reducing
the volume causes some of the vapor to condense, and increasing the volume
causes some of the liquid to vaporize, but the pressure remains constant at a
value which depends only upon the material and the temperature. In this
respect a saturated vapor behaves quite differently from an ideal gas which,
according to Boyle's Law, has its pressure doubled when its volume is halved.
The pressure-volume diagrams for an ideal gas and for a saturated vapor are
contrasted in parts I and II of :Fig. 8; the temperature is assumed constant in
both instances.
It is found by test that the saturated vapor pressure of a liquid has mate-
rially different values at different temperatures. For example, the saturated

Ideal gas Saturated

vapor

Volume Volume
II
Fm. 12-8. Pressure-volume relations plotted for an ideal gas and a saturated vapor.

TABLE IV. SATURATED VAPOR PRESSURES OF LIQUIDS

Temperature, Pressure,
Liquid oc cm Hg

Alcohol (ethyl) .......... . 20 4.4

Alcohol (ethyl) .......... . 50 22.2
Alcohol (ethyl) .......... . 78.5 76.00
Mercury ............... . 20 1.2 x 10- 4
Mercury ............... . 100 2.73 X 10-2
Mercury ............... . 356.7 76.00
Mercury ............... . 400 157.4
Octoil .................. . 20 2.43 X 10-s
Octoil .................. . 50 1.09 x 10-5
Octoil .................. . 100 1.57 x 10-4
Octoil .................. . 150 7.00 X 10-a
Water ................. . 0 0.458
Water ................. . 20 1.75
Water ................. . 50 9.25
Water ................. . 100 76.00
Water ................. . 150 357.0
208 HEAT Chap. 12

vapor pressure of water is 1.75 cm Hg at 20°C, 3.17 cm at 30°C, 9.25 cm at

50°C, and 76.00 cm Hg, or exactly 1 atmosphere, at 100°C. These pressure-
temperature values are plotted to scale in part I of Fig. 9. The saturated
vapor pressures of a few liquids at various temperatures appear in Table IV.

160
I
:E> 140
E
"!. 120 I
!:::,
e
"'100 I C>
:c
.5 2.0
0.
... 80 I Q)'
:i 1.6
/
/
&.
.g.~
) /
g 60 a. 1.2 v
-0

~ 40 I 0
0.8
/
- v
_....,,..,,. /
--
:::, >
/
c 20
en "i+- 0.4
e
0
0 20
~

40 60
Temperature, °C
80 100 120 140 -:::,

J!
0
~

:30 40 50 60
Temperature, °F
70 80 90 100

I II
Frn. 12-9. Saturated-vapor data of water for use in humidity calculations I, in metric
units; and II, in British units.

12-10. Atmospheric Humidity. As the result of evaporation, the atmos-

phere always contains some moisture in the form of water vapor. This mois-
ture does not consist of tiny particles of liquid held in suspension in the air,
but is a vapor as truly invisible as the air with which it mixes. The number
of pounds of water vapor per cubic foot, or the number of grams per cubic
meter, is known as the absolute humidity.
Since the atmosphere is a mixture of dry air and water vapor, the total
atmospheric pressure p has two components: the partial pressure due to dry
air, Pa, and the partial pressure due to water vapor, Pw; whence, from Dalton's
Law,
P =Pa+ Pw• (12.3)
As stated previously, water evaporates into the atmosphere until the partial
pressure due to the vapor equals the saturated vapor pressure of water at
the existing temperature. At this point the atmosphere is saturated and
evaporation ceases. Such a condition is uncomfortable in warm weather
because perspiration cannot evaporate and furnish the desired cooling effect.
The pressure of water vapor in the atmosphere at saturation can be read
for various temperatures from the curve in part I of Fig. 8, but in humidity
calculations it is usual to use pressures in inches of mercury and temperatures
in degrees fahrenheit. A corresponding curve with these units is given for
§ 12-10 ATMOSPHERIC HUMIDI'PY 209

the lower temperature range in part II of the figure. The data from which
it is plotted appear in the first two columns of Table V, followed by values of
the absolute humidity for saturation at the various temperatures. Similar
data in metric units are also included.

TABLE v. PROPERTIES OF WATER VAPOR

Pressure of Pressure of
Saturated Absolute Saturated Absolute
Tempera- Vapor, Humidity, Tempera- Vapor, Humidity,
ture, °F in.·Hg lb/ft3 ture, °C cm Hg gm/m 3

0 0.0375 0.675 x 10-4 -15 0.1436 1.32

10 0.0628 1.10 x 10-4 -10 0.2149 2.10
20 0.1027 1.77 x 10-4 -5 0.3163 3.20
30 0.1646 2.78 x 10-4 0 0.4579 4.90
35 0.2036 3.40 x 10-4 5 0.6543 6.72
40 0.2478 4.10 x 10-4 10 0.9209 9.41
45 0.3003 4.91 x 10-4 15 1.279 12.9
50 0.3624 5.88 x 10-4 20 1.754 17.3
55 0.4356 6.99 x 10-4 25 2.376 23.0
60 0.521 8.29 x 10-4 30 3.183 30.4
65 0.621 9.80 x 10-4 35 4.218 39.6
70 0.739 11.5 x 10-4
75 0.874 13.5 x 10-4
80 1.031 15.8 x 10-4
85 1.212 18.4 x 10-4
90 1.421 21.4 x 10-4
95 1.659 24.7 x 10-4
100 1.931 28.5 x 10-4

The atmosphere does not often contain sufficient water vapor to produce
saturation. The ratio of the amount of water vapor actually present to the
amount necessary to give saturation at the existing temperature is called the
relative humidity. Relative humidity can also be expressed in terms of pres-
sures, since the pressure due to the vapor is approximately proportional to
the amount of vapor present; thus, the relative humidity in per cent is
H = Pw 100. (12.4)
Ps
Herein Pw is the pressure of the water vapor actually present in the atmos-
phere, and Psis the pressure that would be exerted by the water vapor if the
atmosphere were saturated at the existing temperature; that is, Ps is the
saturated vapor pressure of water.
With a given moisture content, if the temperature of the atmosphere rises
the relative humidity is lowered, because the saturated vapor pressure has a
higher value; thus, the atmosphere feels drier, although the amount of water
vapor remains the same. On the other hand, a fall of temperature increases
the relative humidity for the same moisture content. If the temperature
falls to a sufficiently low value, the atmosphere becomes saturated and the
210 HEAT Chap. 12

vapor begins to condense, forming dew; accordingly, this value of the tem-
perature is called the dew point.
The dew point can be measured by cooling a small portion of the air until
it becomes saturated and observing the temperature at which condensation
occurs. The device for making the measurement is called a condensing
hygrometer; it consists primarily of a glass or metal plate so arranged that
one face can be cooled, either by circulating cold water or hy evaporating
ether, until dew is observed to form on the other face, which is exposed to the
atmosphere. The corresponding temperature of the air, which is assumed to
be the same as that of the plate in contact with it, is the dew point.
The relative humidity of the atmosphere can be calculated from a knowl-
edge of the dew point by reference to a vapor-pressure curve such as Fig. 9 or
to tabulated data as in Table V. The pressure corresponding to the dew
point is Pw, and that corresponding to the air temperature is Ps; the ratio
Pw!Ps gives the relative humidity in accordance with Eq. 12.4.
For example, if at 75°F the dew point is measured as 60°F, then the relative humid-
ity, from data in Table V, is H = Pw!Ps = 0.521/0.874 = 0.60 or 60 per cent. The
water vapor present in the atmosphere amounts to 8.29 X 10-4 lb/ft3, and that to
produce saturation is 13.5 X 10-4 lb/ft3; the ratio (8.29 X 10-4)/(13.5 X 10-4 ) = 0.61
or 61 per cent, a result that agrees sufficiently with the foregoing.
The relative humidity can be determined experimentally with a psychrom-
eter, which consists of two suitably mounted thermometers; one is exposed
to the atmosphere and has its bulb dry (dry bulb), and the other has cloth,
moistened with \Yater, wrapped around its bulb (wet bulb). The drier the
atmosphere, the more rapid the evaporation of water from the cloth, and the
lower the reading of the wet-bulb thermometer. Tables are available that
give the relative humidity in terms of the dry-bulb reading and the amount
by which the wet-bulb temperature is lower.
Calculations for humidity and heat are fundamental in air conditioning.
The objective in this field is to control the moisture content and temperature
of the air in a room or building and usually also to "wash" the air and circu-
late it, in order to provide conditions that are healthful and comfortable or
that are desired for certain manufacturing processes. It is possible to calcu-
late, for given values of humidity and temperature outdoors and desired
values of these properties indoors, how much water vap.or and how much
heat must be supplied or withdrawn for each pound of circulated air.
Condensed vapor suspended in the air is familiar to the observer as fog
when at the surface of the earth, and as a cloud when at greater altitudes.
Dust particles and ions serve as nuclei for the saturated vapor to condense
upon; as the condensation continues, the water particles grow in size and the
drops so formed fall to the earth as rain. Condensation of water vapor at
temperatures below the freezing point forms frost and snow.
12-11. Boiling. As previously stated, boiling differs from evaporation in
that it occurs within the liquid instead of at its surface. During the process,
§ 12-11 BOILING 211

bubbles of saturated vapor form within the liquid and rise to the surface.
The bubbles would not form if the pressure exerted upon them were greater
than their own internal pressure. The external pressure consists of the
atmospheric or other pressure on the liquid surface plus the pressure, often
slight, due to the liquid above the bubble. The internal pressure is the
saturated vapor pressure of the liquid at the existing temperature. Hence,
a liquid does not boil unless its sat1trated vapor pressure is as large as the pres-
sure exerted on the liquid.
Boiling can be brought about either by increasing the temperature until the
corresponding vapor pressure is equal to the pressure on the liquid (see Fig.
9), or by reducing the pressure on the liquid to the value of the saturated vapor
pressure. Water, initially at 50°C, will serve as an illustration. When
heated in the open air it will boil at 100°C, at which temperature the satu-
rated vapor pressure is 76.0 cm Hg or 1 atmosphere. Or, it can be made to
boil at 50°C by lowering the pressure upon it to 9.25 cm Hg, which is the
value of the saturated vapor pressure of water at 50°C. From these consid-
erations it is seen that a curve of saturated-vapor pressures, as in Fig. 9, can
also be called a boiling-point curve, since it shows the relation between boiling
point and applied pressure. The difficulty of cooking by boiling at high alti-
tudes, where low pressures prevail, may be inferred from such a curve.
Just as the freezing-point curve of Fig. 3 represents the equilibrium condi-
tion between the solid and liquid phases of water, so the boiling-point curve
of Fig. 9 represents the equilibrium condition between its liquid and vapor
phases. The region to the left of the curve represents the liquid phase and
that to the right, the vapor phase. Table VI shows the boiling points of a
number of pure substances at standard atmospheric pressure.
TABLE VI. BOILING Porn·rs

Substance oc OF
Alcohol (ethyl) ........ . 78.5 171.3
Coppn ............... . 2300 4170
Helium .............. . -268.9 -452
Hydrogen ............ . -252.7 -422.9
Iron ................. . 3000 .5400
Lead ................. . 1620 2950
Mercury ............. . 356.9 674.4
Oxygen .............. . -183.0 -297.4
Sulfur ................ . 444.6 832.:{

Impurities in a liquid generally raise the boiling point. Thus, a pinch of

salt added to a saucepan of water will raise the temperature of boiling to a
higher value, an effect sometimes desired in cooking. A thermometer that
is being calibrated at 100°C should be placed in the saturated steam close to
the water surface rather than in the boiling water, for if the water is impure
its temperature is uncertain.
212 HEAT Chap. 12

12-12. Heat Required to Vaporize a Liquid. In the process of vaporization

,vork: must be done to overcome molecular attraction and to increase the vol-
ume, and hence heat must be supplied. The quantity of heat needed per
unit mass to convert the liquid to vapor without ehange of temperature is
called the heat of vaporization. The equation used in § 12-6 for calculating
heat quantities during fusion, namely,
Q = mL,
can also be used to determine heat quantities during vaporization, if L is
taken to represent the heat of vaporization. This equation shows how much
heat must be imparted to a mass m of liquid to vaporize it at constant tem-
perature, or how much heat that mass of vapor must give off in order to
condense under the same conditions.
The heat of vaporization depends upon the temperature at which change of
phase occurs, and the temperature in turn is determined by the pressure.
The valueR of the heat of vaporization L at standard atmospheric pressure
are given for a few liquids in Table VII.

TABLE VII. HEATS OJ<' VAPORIZATION

Liquid Cal/gm Btu/lb

Alrohol (ethyl)... . . . . . . 204 367

Mercury . . . . . . . . . . . . . . 68 122
Oxygen............... 51 92
Water................. 539 970

If heat is applied to some water in an open beaker, the pressure being 7(5
cm Hg, the water will boil at 100°(1 and the heat of vaporization will be 539
cal per gm. If, instead, water in a beaker at 0°C is placed in the receiver of
a vacuum pump and the air pressure therein reduced to 0.46 cm Hg, the
water will boil at that temperature and the heat of vaporization will be 599
eal per gm. For water boiling at high pressures and temperatures, the heat
of vaporization is less than the values mentioned, and at the critical tem-
perature, §13-8, the heat of vaporization is zero.
The heat of vaporization of water can be determined by passing saturated
steam into cool water within a calorimeter, and measuring the change of
temperature. The amount of steam condensed is found by weighing the
contents of the calorimeter before and after the test, and the temperature
rise of the calorimeter and its contents is measured by a thermometer. If
m gm of water are raised from le to th °C as a result of admitting M gm of
steam at T°C, the heat absorbed by the water will be m(th - tc), and that
given off by the steam will be M [L +
(T - th)]. Since these amounts must
be equal, the heat of vaporization Lat temperature T can be found.
Engineers frequently use "steam tables" in determining the performance
§ 12-13 SUBLIMATION AND THE TRIPLE POINT 213
.
of boilers. These tables include the temperature at which change of phase
occurs, the internal energy of the water, and the heat of vaporization. A
few scattered entries from such a table are given in Table VIII to show the
variation of the heat of vaporization with the temperature at which change
of phase occurs.
T ABLJ<J VIII. PROPERTIES OF SATURATJ<JD STEAM

Btu per Lb Above 32°F

Absolute Temper- I Temper- Absolute
Pressure, ature, Internal En- Heat of ature, Pressure,
lb/in. 2 OF ergy of Water Vaporization oc cm Hg

1 101.8 70 1035 30 3.17

2 126.1 94 1022 ,10 5.51
5 162.3 130 1000 50 9.23
10 193.2 161 982 60 14.92
14.7 212.0 180 970 70 23.35
20 228.0 196 960 80 35.51
50 281.0 250 923 90 52.58
75 307.6 277 904 100 76.00
100 327.8 298 888 120 148.9
150 358.4 330 863 140 271.0
200 381.8 355 843 160 463.3
250 401.0 376 824 180 751.4
300 417.3 394 808 200 1165
400 444.6 424 780 220 1739
500 467.0 450 754

The process of heating a house through the use of steam radiators involves
boiling water over a flame in the cellar and condensing the resulting steam in
the rooms above. When a pound of water is boiled at the normal boiling
point it absorbs 970 Btu from the flame and changes to steam at atmospheric
pressure, and when that pound of steam is condensed within the radiators of
the rooms, its heat of vaporization is given off to heat the air in the rooms
by convection. This method of evaporation and condensation is also utilized
in mechanical refrigeration, wherein the heat from foods in storage is trans-
ferred to the surrounding air by evaporating particular liquids at relatively
low pressures in coils in the freezing compartment, and then condensing them
at higher pressures outside.
12-13. Sublimation and the Triple Point. Under the proper conditions of
temperature and pressure, a substance can pass directly from the solid to the
vapor phase, without liquefying as an intermediate step. Such a transition
is called subb:mation. Iodine crystals sublimate under ordinary room condi-
tions, as do also naphthalene "moth balls" and "dry ice" (solid carbon diox-
ide); carbon behaves similarly at 3800°K.
The relation between sublimation, freezing, and boiling is illustrated for
water in Fig. 10. The line F X is a redrawing of the freezing-point rurve
(Fig. 12-3); BX is likewise a redrawing of the boiling-point curve (Fig. 9);
214 I!EAT Chap. 12

and SX is a corresponding curve representing equilibrium between the soljd

and vapor phases-that is, the sublimation curve. The regions separated
by these curves represent the solid, liquid, and vapor phases, as indicated.
The meeting point X of the three curves is of
special interest. Every point on the curve FX
represents a state of equilibrium between the
B liquid and solid phases; similarly, every point
...::i
Q)
on BX represents a state of equilibrium between
..."'"'Q) the liquid and vapor phases; consequently, the
Cl..
common point X denotes a stata of equilibrium
Vapor
among all three phases. This point is called
the triple point, and represents a condition of
s pressure and temperature at which water can
Temperature freeze and boil simultaneously. To reach this
Frn. 12-10. The triple-point
point, the temperature must be lowered suffi-
diagram for water. ciently to cause freezing, and the pressure low-
ered sufficiently to cause boiling. The triple
point for water occurs at a temperature of +0.0098°C and a pressure of
0.46 cm Hg.
The effect of high pressures upon the value of the freezing point has been
investigated by Professor Percy W. Bridgman at Harvard University. The
upper portion of the freezing-point curve FX is found to undergo some
marked changes of shape, and several varieties of ice are formed having
properties different from those of ice at atmospheric pressure. For one
variety the freezing-point line shifts to the right at extremely high pressures
to such an extent that at about 20,000 atmospheres water can be frozen at
70°C.
The boiling-point curve, BX, comes to a definite end at point B, for which
the temperature is 374°0. This is the critical temperature of water (see
§ 13-8); at higher temperatures steam cannot be condensed into water, how-
ever great the applied pressure.
Although water freezes normally at some point on the curve FX, water
which is quiescent and free from impurities can be cooled several degrees
below its normal freezing point without solidification. This phenomenon,
called supercooling or undercooling, represents a condition .of unstable equi-
librium. Upon dropping particles of ice into the supercooled liquid, solidi-
fication takes place immediately and the temperature rises rapidly to the
normal freezing point. Similarly, under certain conditions water can be
superheated several degrees above its boiling point, but when boiling starts
the temperature falls to the normal value. This use of the term "super-
heated" should not be confused with the common use of the word in connec-
tion with steam. Steam is superheated by removing it from the water and
heating it; the steam then becomes an unsaturated vapor that is above its
§ 12-14 CALORIMETRY INVOLVING CHANGE OF PHASE 215

liquefaction temperature, and begins to behave more like an ideal gas. Super-
cooling and superheating are also observed with other substances.
Experiments in artificial rain-making are based on the theory that super-
cooled clouds, which may have temperatures considerably below 0°C, contain
much water but do not form snow crystals because of the lack of nuclei.
Small pellets of "dry ice," dropped through the cloud from an airplane, create
low-temperature zones and cause ice crystals to form spontaneously; as the
crystals grow, the larger ones fall as precipitate. It has also been found that
clouds can be "seeded" from the ground with silver iodide particles, which
serve as excellent nuclei for ice crystals.
12-14. Calorimetry Involving Change of Phase. When substances at dif-
ferent temperatures, and possibly in different phases, are mixed and allowed
to settle to equilibrium, some of them may vaporize or condense, melt or
freeze. In these processes heat is either absorbed or given off and must be
considered in applying the method of mixtures. A hypothetical problem
will illustrate the procedure.
Suppose 25 gm of steam at atmospheric pressure, superheated to a temperature of
115°C, and 30 gm of ice at -10°C to be introduced simultaneously into a copper
calorimeter having a mass of 200 gm and containing 500 gm of water at 20°C. The
specific heats in calories per gram per centigrade degree are as follows for the tempera-
ture ranges involved: steam, 0.48; ice, 0.50; and copper, 0.093. The heat of fusion
of ice is 80 cal/gm, and the heat of vaporization of water is 539 cal/gm. Neglect heat
radiated to the surroundings, and calculate the resulting temperature of the mixture,
by using the total-energy concept, with energy values reckoned from water at 0°C
as a datum.
Before mixture
Steam 25 X 0.48 X (115 - WO) = + 180}
25 X 539 = +13,475
25 X (100 - O) = + 2,500 26,527 cal
Water 500 X 1 X (20 - 0) = + 10,000
Calorimeter 200 X 0.093 X (20 - 0) = + 372
Ice 30 X 0.50 X ( -10 - 0) = -150}
-2,550 cal
-30 X 80 = -2,400
----
23,977 cal
Aft0r mixture thiH amount of internal energy, 23,977 cal, is to be distributed among
25 + 30 + 500 or .5.55 gm of water and 200 gm of copper. Since the 200-gm copper
calorimeter is the equivalent of 200 X 0.093 = 18.6 gm of water, the 23,977 cal will
be imparted to an equivalent of 555 + 18.6 = 573.6 gm of water. This will cause a
temperature rise above the reference level of 23,977 /573.6 = 41.8 C0 , and, since the
datum is 0°C, the final temperature of the mixture will be 41.8°C.
The internal-energy concept is of advantage in solving problems in which
the final temperature is unknown, in that it is not necessary to make any
assumption as to the probable value of this temperature. This concept is
also helpful in problems involving change of phase where the resulting tem-
perature is 0°C with some ice unmelted (total internal energy of system
216 II EAT Chap. 12

above water at 0°C would be negative) or where the resulting temperature is

100°C with some steam uncondensed (total internal energy of system suffi-
cient to raise entire system to 100°C, leaving a surplus to vaporize part of
the water). In such situations the calculation will show the amount of ice
unmelted or the amount of steam uncondensed.
12-15. Heat of Combustion. When a substance is burned or oxidized it
produces heat, and the amount of heat liberated per unit mass is called its
heat of combustion. The heats of combustion of several fuels are listed in
round numbers in Table IX.

TABLE IX. HEATS OF COMBUSTION

Fuel Cal/gm Btu/lb

Coal (anthracite) .......... . 7,600 to 8,400 13,500 to 15,000

Coke ..................... . 6,900 12,600
Fuel oil .................. . 10,300 to 11,000 18,600 to 19,800
GasolinP .................. . 11,000 to 11,400 20,000 to 20,500
Illuminating gas ........... . 5,500 to 6,400 9,900 to 11,500
Wood (various) ........... . 4,000 to 4,500 7,000 to 8,000

The heats of combustion of carbon and of a few organic compounds are

expressed in Table X in calories per mole, § 9-9, but can be converted to
calories per gram without difficulty. In order to show how this is done,
methane (marsh gas) will be used as an illustration. The chemical formula
CH4 signifies that the molecule of methane is composed of 1 atom of carbon
(C) and 4 atoms of hydrogen (H). Sincl:) the atomic weights of these ele-
ments are C = 12 and H = 1, the compound has a molar weight of 12 4(1) +
= 16 gm, and therefore its heat of combustion per gram is 210,800/16 =
13,170 cal. In engineering, heats of combustion are usually considered posi-
tive. In thermochemistry, the opposite practice is followed, heat evolved iu
a reaction being regarded as negative. The use of signs is conventional and
need not cause confusion.

T ABU~ X. MOLAR HEATS OF COMBUSTION"

Substance Formula Cal/mole ·

Acetylene ............ . C21L 312,000

Alcohol (ethyl) ........ . C2H50H 328,000
Alcohol (methyl)- ..... . CH30H 170,900
Carbon .............. . c 97,300
Methane ............. . CH4 210,800

Measurements of heat of combustion are made in combustion calorimeters,

one type of which is shown in cross-section in Fig. 11. The substance to be
tested is placed in a platinum crucible mounted within a strong steel con-
§ 12-15 HEAT OF COMBUSTION 217

tainer, which is lined with platinum, gold, or porcelain to avoid corrosion.

Oxygen under a pressure of several atmospheres is admitted and then the
bomb is sealed tightly. The bomb is placed in a calorimeter containing
water, and an electric current is passed through
a small heating coil which dips into the test
substance. When the wire becomes incandes- -oxygen
cent, combustion takes place within the bomb supply
and the resulting temperature rise of the water
outside is observed.
A different procedure is used for measuring
the heat of combustion of gaseous fuels. The Bomb
gas is burned at a constant rate within a cham-
Heating coil
ber through which water is circulating uni-
formly. From a knowledge of the rate of water Crucible
flow and of the temperature of the water as it
enters and leaves the chamber, the energy liber- Fm. 12-11. Bomb calorimeter
ated per unit volume of fuel can be determined. for measuring heat of combus-
In some fields of study heat quantities are tion of pulverized fuels.
expressed in terms of the kilogram calorie ; this
is also called the large calorie and is equal to 1000 calories. Food values are
stated in the larger unit for convenience; thus, the fuel value of the food
consumed per day by the average American is 3250 kg cal.

PROBLENIS
1. How many calories are needed to heat 300 gm of ethyl alcohol in a glass beaker
of 100-gm mass from 15 to 40°C?
2. How much heat must be withdrawn from 5 lb of aluminum to cool it from 200 to
32°F?
3. Find the resulting temperature when 5 kg of water at 80°C and 1 kg of water at
20°C are mixed.
4. A mixing faucet is supplied with cold water at 40°F and hot water at 160°F.
The cold water alone fills a quart measure in 20 sec, and the cold and hot water mixed
fill a quart measure in 12 sec. Find the temperature of the mixed water.
5. A copper calorimeter of 160-gm mass contains 350 gm of water at 20°C. Into
this are placed 200 gm of lead at 80°C and 150 gm of iron at 70°C. Find the resulting
temperature of the mixture.
6. A steel die block is quenched at 1500°F by plunging it into water at 50°F. For
careful control of the process it is desired to limit the temperature rise of the water to
8 F 0 • If the steel block has a mass of 4.5 lb, what is the smallest mass of water that
may be used? Take the specific heat of steel to be 0.15 Btu/ (lb· F 0 ) over the temper-
ature range involved.
7. A glass beaker of 120-gm mass contains 200 gm of water at 15°C. Into the
water is placed a 60-gm sample of metal, originally at 98°C. If the resulting temper-
ature is 19.8°C, what is the specific heat of the metal sample?
8. A !-lb copper calorimeter contains 1.5 lb of water at 65°F. Into this is dropped
0.4 lb of aluminum at 200°F. The resulting temperature of the mixture is found to
be 72°F. Compute the specific heat of aluminum as given by these data.
218 HEAT Chap. 12

9. How much heat is needed to convert 10 lb of ice at 32°F into water at 120°F?
10. How much heat does 1 ft 3 of ice absorb in melting at 32°F?
11. How much heat must be withdrawn from 10 lb of water, initially at 70°F, in
order to lower its temperature to the freezing point and then freeze half of it?
12. Find the relative humidity for a day when the temperature is 80°F and the dew
point is 60°F.
13. The relative humidity in a room is 60 per cent at a temperature of 70°F. If the
temperature rises to 75°F, what will the relative humidity be?
14. The heat of vaporization of ammonia is 589 Btu/lb and that of sulfur dioxide
is 172 Btu/lb. Express these values in calories per gram.
15. How many calories are required to convert 1 kg of ice at - 30°C to steam at
140°C under atmospheric pressure? Take the specific heat of steam to be 0.48
cal/ (gm. C 0 ) .
16. What will be the result of supplying 500 Btu to 10 lb of water initially at 100°F?
17. Steam of density 0.037 lb/ft3 is supplied at 212°F to a radiator at the rate of
80 ft 3/hr, and leaves the radiator as water at 120°F. At what rate does the radiator
give off heat?
18. A 50-gm sample of an alloy is heated to 100°C and placed on a large piece of ice.
If 10.4 gm of ice melt, what is the specific heat of the alloy?
19. A copper calorimeter of 150-gm mass contains 250 gm of water at 15°C. Into
this are placed 200 gm of iron at 100°C and 10 gm of ice at -20°C. Find the resulting
temperature.
20. An ice cube taken from the freezing compartment of a refrigerator is dropped
into 0.75 lb of water initially at 85°F, contained in a calorimeter that has a water
equivalent of 0.05 lb. After the ice has melted it is found that the calorimeter con-
tains 0.87 lb of water at 57°F. Find the initial temperature of the ice.
21. A glass beaker of 100-gm mass contains 200 gm of water at 20°C. Into the
water are placed 60 gm of ice at 0°C. Determine the final temperature and phase of
the mixture.
22. A 1.25-gm sample of coal is finely pulverized and burned in a bomb calorimeter.
The bomb is immersed in 2500 gm of water, and the water equivalent of the bomb
is 625 gm. If the temperature of the water rises 3.20 C 0 , what is the heat of com-
bustion of the coal?
23. About how many gallons of fuel oil would be equivalent in heating value to 1 ton
of coal? Assume the specific gravity of the fuel oil to be 0.90.
24. The therm is a heat unit used in the gas industry; 1 therm = 105 Btu. (a) Take
the heat of combustion of natural gas to be 1040 Btu/ft3 , and determine how many
cubic feet of this fuel will provide 1 therm. (b) How many gallons of water will
1 therm raise from the freezing point to the boiling point?
 13
THERMAL BEHAVIOR
OF GASES
The effects produced by applying heat to a gas depend upon whether the
gas is allowed to expand or is prevented from expanding, whether the pres-
sure is allowed to change or is kept constant, and whether or not the tempera-
ture is fixed. Even the simple expansion of a gas as its pressure is lowered is
not the same when the temperature is held steady as when the gas is sur-
rounded by insulation through which no heat can pass. In brief, the thermal
action of a gas depends upon the manner in which the gas is constrained while
it is being tested. These effects, together with some related matters, are
considered in this chapter. The subject matter is of great importance in
engineering, for engines and turbines of all sorts, compressors, refrigerating
systems, and air-conditioning equipment depend for their operation on the
thermal behavior of gases.
13-1. The General Gas Law. The physical properties of a gas are con-
trolled by three variables: the pressure of the gas, the volume that it occupies,
and its temperature. For a stated mass of the gas, if two of these factors are
given, the value of the third can easily be computed; or if one is kept con-
stant, the relation between the other two can be found. These relationships
for actual gases are complicated by the interaction of their molecules, par-
ticularly near the temperatures of liquefaction. The analysis is simplified
by imagining a so-called "ideal" or "perfect" gas, § 13-4; such a gas will be
assumed in the following analysis of the three gas variables. Air, hydrogen,
and other fixed gases behave very nearly like an ideal gas over wide tempera-
ture ranges.
The manner in ,vhich these variables are related, for a fixed mass of gas, is
expressed in the General Gas Law as follows:
p1V1 p2V2 (13.1)
=
T1 T2 '
in which p 1 is the absolute pressure of the gas, V1 its volume, and 7\ its abso-
lute temperature, all under some condition designated by the subscript 1;
P2, V2, and T2 have similar meanings but refer to some other condition desig-
nated by the subscript 2.
219
220 HEAT Chap. 13

The form of the foregoing equation shows that the pressures p1 and p 2 can
be expressed in any convenient pressure unit, provided the same unit is used
for both; this is true similarly for the volumes V1 and V2 and for the tempera-
tures T1 and T 2• Indeed, the units for these quantities need not belong to
the same system-for example, the pressures can be stated in atmospheres,
the volumes in liters, and the temperatures in degrees rankine. It is neces-
sary, however, to use absolute values for the pressures and temperatures.
Constant Pressure. Under the condition of constant pressure, Pi = P2, the
General Gas Law takes the simple form:
V1 = V2, (13.2)
T1 T2
which shows that for a fixed mass of gas at constant pressure, the volume
varies directly as the absolute temperature.
A gas could be heated at constant pressure by confining it in a cylinder
equipped with a freely moving piston, as indicated in part I of Fig. 1. With

Q)
E
:::>
0
> -----/IV,
.,,,,-" I o

0
.,,.,,""
1.<::...._ _ _
io oc}
_J....:..___ _ __ . . . _ - : : :
0
-K Temperature
273
II
Frn. 13-1. A gas heated at constant pressure. Its volume varies directly as the absolute
temperature.

this arrangement, the slightest change of pressure would cause the piston to
move in or out, keeping the pressure inside of the cylinder constant at the
value existing on the outside. The relation between volume and temperature
is essentially the same for a gas as given for a solid or liquid in Eq. 11.4. This
equation was based on a tacit assumption of constant pressure, although
pressure changes would have but little effect on the volume of a solid or of a
liquid. The Gas Law for constant pressure, Eq. 13.2, can be arranged in the
form of the earlier one, in order to compare the two, by using temperatures of
0°C and t°C-that is, 273°K and (273 + t) °K-as follows: ~: = ~t, or
Vo = 273Vt+ t, whence Vt = Vo
273 + t 1 t ) , or
= Vo ( 1 + 273
273 273
Vt = Vo (1 + (3t). (13.3)
In this expression Vo and Vt represent the volumes of the gas at 0°C and t°C,
respectively, and (3 is the coefficient of volume expansion of the gas.
§ 13-1 THE GENERAL GAS LAW 221

The coefficient f3 for gases is much larger than the expansion coefficients
for solids and liquids; moreover, it has almost exactly the same value for all
gases. This value is f3 = 0.00367 ( = 1/273) per centigrade degree, which
means that a given amount of gas at 0°C ·will expand by 1/273 of its volume
when heated 1 degree, and will contract by the same amount when cooled
1 degree. It should be noted that this value for /3 applies only when the
volume Vo is taken at 0°C.
The relations described are plotted in part II of Fig. 1, which also shows
that, under constant pressure, the volume of the gas varies directly with the
absolute temperature.
Constant Volume. For the condition of constant volume, V1 = V 2 , the
General Gas Law, Eq. 13.1, takes the simple form:
1!J. ~ (13.4)
=
T1 T2 '
vvhich shows that, for a fixed mass of gas kept at constant volume, the abso-
lute pressure varies directly as the absolute temperature.

...v
::::>
I
I
V)
V) I
...v
a.
I
IP,
I I
I Po I

0 ~----
10 / t °C}
27'--3------'-T--oK Temperature
n
Fm. 13-2. A gas heated at constant volume. Its absolute pressure varies directly as the
absolute temperature.

A gas can be heated at constant volume by confining it in a tight container,

as represented in part I of Fig. 2, the container being made of some material
that has negligible expansion. The increase of pressure ·with temperature is
indicated in part II of the figure. It is seen to be similar to the increase of
volume with temperature at constant pressure and will be expressed in similar
notation. At constant volume, po/To = pe/T, and
Pt = Po (1 + /3t). (13.5)
The constant /3 is called the pressure coefficient of the gas; it has the same
value as the coefficient of volume expansion, 1/273, and is represented by
the same symbol. As before, the zero subscript refers to 0°C or 273°K.
The relations expressed in Eqs. 13.3 and 5, dealing with the behavior of a
gas at constant pressure and again at constant volume, are attributed to the
222 HEAT Chap. 13

French mathematician and physicist Jacques A. C. Charles (1746-1823) and

jointly are known as Charles's Law.
Constant Temperature. The General Gas Law as applied to the condition
of constant temperature, T 1 = T 2 , takes the form
(13.6)
or pifp2 = VdV 1 , which shows that for a fixed mass of gas at constant tem-
perature the volume varies inversely as the absolute pressure. This is a re-
statement of Boyle's Law, as given in § 9-4.

Volume
II
Fm. 13-3. Variation of pressure with volume of a gas at constant temperature.

To demonstrate this law, the gas is enclosed in a cylinder having a piston,

as in part I of Fig. 3, and provision is made to keep the temperature constant.
The pressure is found to vary with the volume as represented by the hyperbola
in part II of the figure.
As a typical problem, suppose that a tank of air has a volume of 2 ft 3 and that the
gage on the tank indicates a pressure of 300 lb/in. 2 If the valve on the tank is opened
until the gage reading goes down to 150 lb/in. 2 , what volume of air, reckoned at atmos-
pheric pressure, will escape from the tank? Assume that the temperature remains
constant throughout.
If the air initially in the tank were subjected to a gage pressure of 150 lb/in.2, it
would occupy a volume V2 such that
(300 + 14.7) X 2 = (150 + 14.7) X V2,
the pressures being in pounds per square inch and the volumes in cubic feet. From
this expression, V 2 = 3.82 ft 3 • Of this amount, 2 ft 3 will stay in the tank, and the
remainder, 1.82 ft3, will escape. The volume, V3, that the air which escapes would
occupy at atmospheric pressure is found similarly; thus, ·
(150 + 14.7) X 1.82 = 14.7 X Va,
from which the desired volume is Va = 20.4 ft3.
13-2. The Gas Constant. The General Gas Law, Eq. 13.1, is frequently
stated in another useful form, which introduces the so-called gas constant.
 TllE GAS CONSTANT 223

The equation referred to is written in the form P; = K, or p V = K T,

and the constant of proportionality K is replaced by the product m R, where
mis the mass of the gas, assumed constant, and R is the gas constant. Hence,
the General Gas T,aw becomes
pV = mRT, (13.7)
and applies to all conditions of pressure and temperature.
The gas constant R is found from the above equation by noting that
R = TP V = _]!_ = J!_Td, where d is the density of the gas. If the subscript s
m Tm _
v
is used to designate standard conditions of temperature and pressure (0°C and
76 cm Hg), then the gas constant can be expressed as

(13.8)

As applied to a gram, kilogram, or pound of gas, R has a particular value

for each gas. For air, in metric units, the density under standard conditions
is 1.293 X 10-3 gm/cm 3 • Standard pressure is equivalent to l.013 X 106
dynes/cm 2 • Therefore, the value of the gas constant for air is

R = (1.013 X 106 dynes) /(273°K X l.293 X 10-3 gm) = 2.87 X 106 ergs _
cm2 cm 3 gm·°K
or 287 l;~~~- Again, for air, but in British units, T. = 492°R; at this tem-
perature and at a pressure of 14.7 lb/in. 2 the weight density of air is 0.081
lb/ft3; consequently, the value of the gas constant for air is
lb
( 14.7 in. in. 2 ) /( 0 lb) ft· lb
R = 2 X 144 ft 2 492 RX 0.081 ft 3 = 53.3 lb· 0R·

As applied to a mole of gas, the gas constant has the same value for all
gases. Since a mass of 1 mole of gas under standard conditions always
occupies 22.4 liters, § 9-9, the density of any gas is d. = 1 mole/(2.24 X 104
cm3), and therefore the universal gas constant is

R = (1.013 X 106 dynes) /(2730K X 1 mole ) = 8.31 X 101 ergs .

cm2 2.24 X 104 cm3 mole·°K
To show the application of the gas constant, find the volume occupied by 10 lb of
air at 200°F when subjected to an absolute pressure of 20 atmospheres. From Eq. 13.7,

ft· lb
mR T 10 lb X 53.3 lb. 0 R X (200 + 460) oR
v = -- = = 8.28 ft3•
p 20 x (14.7 x 144) ~~2
224 HEAT Chap. 13

The General Gas Law in either form justifies the assumption stated in
Chap. 9 to the effect that the temperature of the gas is directly proportional
to the mean kinetic energy of the gas molecules. This kinetic energy,! mv2,.
equals; X ¥,fromEq. 9.1, wheredisthedensityofthegas. Butm/d = V,
the volume of the gas, and therefore the kinetic energy is proportional to the
product p V. The General Gas Law, however, states that for a fixed mass of
gas the p V product is proportional to the temperature and thus agrees with
the kinetic theory.
13-3. The Gas Thermometer. The gas thermometer utilizes the change
of pressure caused by heating or cooling a gas at constant volume as a means
for measuring temperature. The thermometer, Fig. 4, consists essentially
of a thin-walled bulb of glass, metal, or porcelain, in which the gas is con-
fined, and to which a flexible tube containing mercury is attached. The vol-
ume of the gas is kept constant, barring the expansion of the bulb, by keep-
ing the top of the left-hand mercury column in contact with an index, and
the pressure of the gas is obtained from the difference in height of the two
columns as measured on the scale.
The pressure of the gas in the bulb will be represented by po at t0 °C, Pi at
ti °C, and p 2 at t2 °C. Since the volume of the gas is constant, a change in
temperature will produce a proportional change in pressure, or
(ti - to) o: (pi - Po),
and (t2 - to) o: (p2 - po).
The proportionality is the same in both expressions; consequently,
t2--- to= P2 - po .
ti - to Pi - Po
Specifically, if the pressure of the gas is p 0 when the bulb is packed in
melting ice (to = 0°C) and Pi when the bulb is surrounded by steam at stand-
ard atmospheric pressure (ti = 100°C), then, if the pressure of the gas is P2
at some unknown temperature t2 °C, that temperature can be expressed as

t2 = 100 P 2 - Po ,
Pi - Po
in terms of the observed gas pressures.
The gas thermometer is somewhat awkward to manipulate and is not
direct reading, but it is used as a basic standard for the calibration of ther-
mometers which serve as secondary standards. The gas used in the bulb
is usually hydrogen, but for measurements at very low temperatures helium
is preferable.
13-4. Intermolecular Forces. It is, perhaps, natural to inquire \vhether or
not the molecules of a gas attract one another, and the law of conservation
of energy suggests an experiment to help answer this question. When a gas
§ 13-4 INTERMOLECULAR FORCES 225

is allowed to undergo free expansion-that is, to expand without external

opposition-if the gas molecules do attract one another, work would have to
be done upon them as they move farther apart during expansion, and this
work would diminish the internal energy of the gas, with a consequent lower-
ing of temperature.
An experiment of this kind ,vas conducted by Joule, using apparatus essen-
tially as represented in Fig. 5. A tank of gas under pressure ,vas piped to a
second tank from which the air had been evacuated, and the valve between
the tanks was suddenly opened to allow the gas to expand freely into the
vacuum. The tanks ,vere immersed in a water bath, and the temperature of
the water was observed before and after the valve was opened. In this test
the temperature appeared to remain unchanged, indicating that the gas,
taken a:s a whole, was neither cooled nor heated. The experiment was not

Water

Bulb
=-1}Ua+'===t----t==Ji Gas under
pressure

Frn. 13-4. A constant- Frn. 13-5. Joule's experiment on

volume gas thermometer. free expansion.

entirely convincing, however, because the bath contained so much water that
a slight temperature change might readily have escaped notice. Other ex-
periments were devised later in which it was possible to show conclusively,
although indirectly, that an actual gas upon free expansion undergoes a
slight cooling.
The kinetic theory offers a picture of what takes place when an actual gas
expands through a small opening into a region of lower pressure. As any
molecule passes through the orifice it is attracted by all other molecules
within its sphere of action; those ahead attract it forward, and those behind
attract it backward. The predominating effect, however, is due to the
molecules behind, since the pressure, and consequently the molecular den-
sity, is greater in that region. The result is a slowing down of the molecules
as the gas passes through the orifice, and this implies a reduction of kinetic
energy and a lowering of temperature. Thus, the theory shows that when-
ever an actual gas expands, its temperature is lowered because of molecular
attraction. At ordinary temperatures this cooling is very slight but, as the
226 HEAT Chap. 13

temperature is reduced, the molecules come closer together (if the pressure
is kept constant) and the cooling effect is greatly increased.
In the ideal gas, it is supposed that the molecules exert no forces upon one
another. If such a gas existed, it would, therefore, show no cooling during
free expansion; it would have no energy abstracted from it even though its
pressure and volume were both changed. From this property an important
principle follows; namely, the internal energy of an ideal gas remains con-
stant, regardless of changes in pressure and volume, provided the tempera-
ture remains unchanged.
These considerations emphasize the difference between actual and ideal
gases and permit the properties assigned to the ideal gas to be fully stated:
Thus, an ideal gas conforms to the General Gas Law, and its internal energy is
a fv,nction of temperatttre only.
13-5. Constrained Expansion. When a gas is allowed to expand against a
back pressure, the expansion is termed constrained· to distinguish it from free
expansion into a vacuum. During constrained expansion
the gas does external work in pushing back the atmosphere
or other surrounding gases, in addition to the slight (often
negligible) amount of internal work that it does, if an ac-
tual gas, in separating its own molecules.
Area A ~th- The external work done by a gas in expanding can be
ds determined by supposing that a cylinder, Fig. 6, contains
Frn. 13-6. Con-
gas under an absolute pressure p, and that the gas in ex-
strained expan- panding pushes a piston of area A back an infinitesimal
sion of a gas. distance ds. During the process, the confined gas exerts
a force F = pA and does an amount of external work
dW = F ds = pA ds. But A ds = dV, the increase in volume of the gas;
therefore, the external work done by the gas during an infinitesimal expansion
is dW = p dV, and during a finite expansion,
W = fpdV, (13.9)
an expression that is similar to Eq. 10.2 for liquids.
If no energy is supplied to the gas during expansion, the work will neces-
sarily be done at the expense of the internal energy of the gas, and as a result
its temperature will fall. When an actual gas expands against a back pres-
sure, it is cooled by doing external work, and also by the internal ,vork of
separating its own molecules.
The commercial manufacture of ice is based on the expansion of gases.
Refrigeration is often brought about by allowing anhydrous ammonia to pass
through an expansion valve into a region of reduced pressure. The evapora-
tion of the liquid and the consequent expansion of the vapor are both cooling
processes; together they lower the temperature of the ammonia sufficiently
to freeze water placed nearby.
The compression of a gas raises its temperature; the heating of the air in a
§ 13-6 ENERGY CONSIDERA1'ION8; SPECIFIC HEATS OF GASES 2Z7

tire pump is a familiar example. In the Diesel engine, § 14-10, air drawn
into the cylinder is compressed and thereby heated to about 1000°F, which is
sufficiently high to ignite the fuel oil as it is sprayed into the cylinder.
The kinetic theory again offers an explanation of these processes. The gas
molecules in the cylinder, Fig. 13-6, in their incessant motion, strike against
the cylinder walls and piston and rebound from these surfaces. If the piston
is moving outward, as during expansion, the molecules rebounding from this
receding surface lose speed; if it is moving inward, as during compression, the
molecules rebounding from this advancing surface gain speed. Thus, expan-
sion lessens the speed of the gas molecules, reduces their kinetic energy, and
lowers the temperature of the gas; compression produces the opposite effect.
13-6. Energy Considerations; Specific Heats of Gases. In general, when-
ever heat is applied to a gas, two effects are produced: the temperature of the
gas rises, and the gas expands. It follows from the Law of Conservation of
Energy that the energy supplied as heat appears in two forms: (1) in part as
an increase in the kinetic energy of the gas molecules (this increases the
internal energy of the gas and manifests itself as a rise in temperature), and
(2) in part as work done by the gas in expanding. To view the process as a
series of small steps, the heat supplied will be represented by dQ, the increase
of internal energy by dU, and the work done by expansion by dW, all these
quantities being differentials; then,

dQ = dU + dTV. (13.10)
These considerations ·will be applied in studying the specific heats of gases.
The specific heat of a gas, as of any substance, is numerically equal to the
amount of heat necessary to raise the temperature of a unit mass one degree.
For a gas, the specific heat is not single-valued but may have a range of values
depending upon the constraints that are imposed upon the gas while being
heated. In order to raise 1 gm through 1 C 0 , more heat would be required if
the gas were kept at constant pressure than if it were kept at constant volume,
and still different values would be obtained if variations were allowed in both
pressure and volume. The most important values of the specific heat of a
gas are those at constant pressure, cp, and at constant volume, c,,.
The heat required to raise 1 gm of gas through 1 C 0 at constant pressure
(see Fig. 1) is greater than at constant volume (see Fig. 2), because at con-
stant pressure the gas expands and does work, whereas at constant volume
it does not; and the energy needed to do this work must come from the heat
supplied. The increase of internal energy is the same in each process, be-
cause in each the temperature rise is the same. Consequently, the specific
heat of a gas at constant pressure exceeds its specific heat at constant volume.
In Table I are listed the specific heats of several gases at constant pressure,
Cp (for atmospheric pressure), and at constant volume, cv; the values are cor-
rect either in calories per gram per centigrade degree or in British thermal
units per pound per fahrenheit degree.
228 HEAT Chap. 13

TABLE I. RPECIFIC HEATS OF GASES AND v APORS

Temperature,
Substance oc Cp Cv

Air ..................... 0-100 0.242 0.173

Carbon dioxide ........... 0-100 0.208 0.159
Hydrogen ..... . . . . . . . . . . 0-100 3.40 2.40
Oxygen ................. 0-800 0.246 0.176
Water vapor ......... .... 100 0.48 0.34
I
It is evident from the foregoing statements that the difference between the
two specific heats of a gas is virtually the external work done by the confined
gas in pushing back the surrounding gases. To verify this conclusion, it will
be supposed that 1 gm of air is heated from 0° to 1°Cat a constant pressure p
of 1 atmosphere. The initial volume of the air is 1/0.001293 = 775 cm3 ,
and the increase in volume during a temperature rise from 0° to 1°C is
(1/273) X 775 = 2.83 cm3• The external work done during expansion is
dynes
found from Eq. 13.9 to be W = pfdV = 1.013 X 106 X 2.83 cm3 =
cm 2
2.87 X 106 ergs = 0.287 joule. In the next chapter it will be shown that
this result is equivalent to 0.0686 cal. The difference between the specific
heats, cP - cv, as given in Table I for air is 0.242 - 0.173 = 0.069 calf (gm· C 0 ) .
The two results are in approximate agreement. For an ideal gas the agree-
ment would be exact, because the difference between cp and Cv would be due
entirely to the external work done in expansion.
Along more general lines, it can be shown that, for an ideal gas, the quan-
tity Cp - Cv is equal to the constant R in the General Gas Law. In this
derivation it will be supposed that a given mass of gas is heated in two dif-
ferent ways: (1) at constant pressure, and (2) at constant volume, until in
each instance it undergoes the same temperature rise, and the amounts of
energy expended in these tests will be compared.
In the first test, a mass m of gas is heated at constant pressure, as in Fig. 1,
through a temperature rise dT; the energy supplied as heat is m Cp dT. Part
of this energy, dU, increases the internal energy of the gas, and the remainder,
dW, is used in doing the mechanical work of expansion, in accordance with
Eq. 13.10; thus,
m cp dT - dU = dTV = p dV.
The last term can be modified by differentiating the General Gas Law,
p V = mRT, with p constant, from which p dV = m R dT. Hence, the fore-
going equation can be rewritten as
m Cp dT - dU = m R dT.
In the second test, the same mass m of gas is heated at constant volume, as
in Fig. 2, through the same temperature interval as before, but the energy sup-
§ 13-7 ISOTHERMAL AND ADIABATIC PROCESSES 229

plied as heat, m Cv dT, will be only that needed to increase the internal energy
of the gas; that is,
m Cv dT = dU.
The temperature rise is the same in both tests; consequently, dU has the
same value in the two preceding equations, and they can be combined as
follows:
m Cp dT - m Cv dT = m R dT.
By cancellation of m dT from each term, it follows that
Cp - Cv = R, (13.11)
all terms being expressed in the same units. Therefore, the difference be-
tween these two specific heats of a gas is equal to the gas constant.
13-7. Isothermal and Adiabatic Processes. The principle of energy con-
servation is a basic concept that applies to processes of all kinds. It is
applied to gases in Eq. 13.10, dQ = dU +
dW, where dQ represents the quan-
tity of heat supplied to a gas, dU is the increase of internal energy of the gas,
and dW is the work done by the gas in expanding. Two processes, known
by the names isothermal and adiabatic, are fundamental in scientific and
engineering work and will be studied by the aid of this basic equation.
An expansion or contraction of a gas which occurs without change of tem-
perature is said to be isothermal. Since at constant temperature there is no
change in the internal energy of the gas, the quantity dU in the equation is
zero, and hence for an isothermal process
dQ = dW, (13.12)
which shows that the heat supplied during an isothermal expansion is equiva-
lent to the work done by the gas in expanding. Similarly, during an iso-
thermal compression, there must be a continual transfer of heat away from
the gas in order to keep the temperature constant, and the heat transferred
is equal to the work done in compressing the gas. Since the energy from the
heat must penetrate the body of gas during these processes, isothermal
changes usually take place somewhat slowly.
The isothermal process has already been considered in connection with
Boyle's Law; the relation between pressure and volume for an ideal gas
when kept at constant temperature is shown in Fig. 3 and is given by the
expression p 1V 1 = p2 V2. More simply, the equation of an isothermal process
.
IS
p V = a constant. (13.13)
Since expansion is a cooling process, it follows that heat must be supplied to
a gas during an isothermal expansion. This action will take place auto-
matically if the gas is allowed to expand slowly while in good thermal con-
tact with some source of heat which is maintained at the temperature of the
230 HEAT Chap. 13

gas. The cooling tendency during expansion will be offset by a transfer of

heat from the source to the gas, keeping the gas temperature constant.
When a gas expands or contracts without either gaining or giving up heat,
the process is called adiabatic. Such a process would result if the gas were
contained in a cylinder completely surrounded by a perfect heat insulator, so
that no heat could be taken on during expansion or given off during compres-
sion, and dQ in Eq. 13.10 zero; becomes. Hence, for an adiabatic process,
-dU = dW. (13.14)
During an adiabatic expansion some of the internal energy of the expanding
gas is converted into mechanical work, and the temperature of the gas falls;
the reduction in internal energy of the gas is equivalent to the work done
during an adiabatic expansion. Again, during an adiabatic compression the
work done upon the gas causes its temperature to rise. Adiabatic processes
are usually rapid.
The equation of the adiabatic curve can be derived by combining the Gen-
eral Gas Law, Eq. 13.7, with the characteristic equation for the adiabatic
process, Eq. 13.14. Several steps are involved, which result in the following
relation between volume and pressure:
p V'Y = a constant, (13.15)
where the symbol 'Y (gamma) represents the specific heat ratio cp/cv.
The numerical value of 'Y depends upon the molecular structure of the gas.
lVIost of the common gases, such as hydrogen, oxygen, and nitrogen, are
diatomic; that is, the molecule is composed of two atoms; while the inert gases,
such as helium and argon, are chiefly monatomic, the molecule consisting of a
single atom. It can be shown theoretically that for diatomic gases 'Y = l.40
and for monatomic gases 'Y = 1.66. For air, composed largely of diatomic
gases, the value is 1.40.
During an adiabatic expansion or compression there is a change in the
temperature of the gas which, however, does not appear explicitly in Eq.
13.15. But if this equation is combined with the General Gas Law, an
expression for the adiabatic process can be obtained in terms either of tem-
perature and volume or of temperature and pressure. The temperature-
volume relation will be derived to show the procedure.
The equation for the adiabatic process, as expressed for two conditions of a

gas, becomes Pi V/ = p2 V/, from which Pi = (~2 )'(· The General Gas
P2 v1

Law, Eq. 13.1, is written in the form Pl = vV· {JT.\ The ratio Pl is eliminated
P2 1 2 P2
. . .
f rom t h ese expressions, g1vmg v: Y
(V9)' V2T1 h
= Vi X T 2 ; w ence,

T 1 V 1'(-1 -
-
T 2 V 2-y-1 , (13.16)
§ 13-8 LIQUEFACTION OF GASES 231

which is the relation sought.

As an illustration of an adiabatic process, suppose that some air initially occupies a

volume of 2 ft 3 at a gage pressure of 100 lb/in. 2 and at a temperature of 400°F, and that
it expands adiabatically to a volume of 6 ft 3 • Determine the volume and temperature
of the gas after expansion.
The volume is found from Eq. 13.1.5, with subRcripts 1 and 2 used to repreRent the
initial and final conditions; thus, p1 V/ = p 2 V/, from which Pi = (~ 2 )'Y =
P2 T1
(¥)1.
2
40

= 31. 40 • To evaluate this quantity, let 31 · 40 = x, and take logarithms of both sides of
the equation. Then log x = 1.40 log 3 = 1.40 X 0.4771 = 0.6679, and therefore
x = 4.65. It follows that P2 =
(100 +
14.7) lb/in. 2 •
4 _65 = 24.7 lb/m. 2 absolute pressure,

which is equivalent to 10.0 lb/in. 2 gage pressure. The temperature is found simi-
larly from Eq. 13.16, as follows:
T
T: (V)-y-1
v:
= = 3°· 40 = 1.55; consequently,
T2 = (400 + 460)/1.55 = 554°R, which is equivalent to 94°F.
To permit comparison between isothermal and adiabatic processes, an
isothermal curve and an adiabatic curve are drmvn on the same pressure-
volume diagram in Fig. 7. The two curves start
from a common point X, which represents a cer- x
tain condition of pressure and volume of the gas
under consideration. The gas, if expanded adia-
batically to some lower pressure p', at point 1vl, ~-
.,,.
::,
will be cooler and therefore occupy less volume en
e-
than if heat had been supplied to expand it iso- a..
thermally to the same pressure at point N. Con-
sequently, the curve representing the adiabatic
process is steeper than for the isothermal process.
Actual expansions and compressions are neither Volume
isothermal nor· adiabatic, but are intermediate
Frn. 13-7. Isothermal and
between the two. adiabatic processes com-
13-8. Liquefaction of Gases. Since a liquid pared.
differs essentially from a gas in that its molecules
are closer together, it is natural that, in attempting to liquefy a gas, the gas
should be subjected to low temperature and high pressure, both of which cause
it to contract. The behavior of a gas as it approaches liquefaction can be
studied by investigating the relation between its pressure and volume as the
temperature is held at lower and lower values. Such a series of tests was
conducted on carbon dioxide by Thomas Andrews (1813-1885), Irish chemist
and physicist, with results as plotted in Fig. 8.
In one test, the tube containing the carbon dioxide was maintained at a
constant temperature of 48.1 °C while the volume of the gas was reduced from
a large to a small value; during this process the pressure increased approxi-
232 J!EA1' Chap. 13

mately in accordance with Boyle's Law, as shown by the uppermost curve in

the figure. A similar test at a lower temperature, 35.5°C, showed a notice-
able departure from that law. For tests at still lower temperatures the
departure became more marked, and the curve for the test at 31.1 °C shows a
definite point of inflection at E. The curve for any temperature below this
value exhibits a horizontal portion, signifying the presence of liquid as well
as vapor, as described in § 12-9. In this temperature region, a curve such as
LJYINP shows that, as the volume of gas is reduced, the pressure at first in-
creases (LM) much as for an ideal gas; next the point of saturation is reached
and liquid begins to form (M); a further reduction in volume causes more and

p
100

.,, 90
...
(I)

-a.,, 80
(I)

-...
0
E
0
70
(I)'

:::,
en
en
60
....
(I)
a..
50 L

40
0 2 4 6 8 10
Volume, cm3/gm
Fw. 13-8. Isothermal curves for carbon dioxide. A horizontal portion of a curve indi-
cates that some vapor has liquefied.

more of the vapor to liquefy, the pressure remaining constant (MN); even-
tually the liquefaction is complete (N); and from this point on, a great
increase of pressure (NP) is required to produce a small change in volume.
The point E is called the critical point, and the corresponding propertieR of
the substance under test are referred to as the critical values.
The critical temperature is that minimum temperature above which a gas
cannot be liquefied, no matter how much pressure is applied;· for carbon diox-
ide its value is 31.1 °C. The critical pressure is the absolute pressure (of gas
and liquid) at the critical temperature. The critical voliime is the volume of
gas at the critical temperature and pressure which at 0°C and 76 cm Hg pres-
sure would have unit volume.
An experiment on critical temperature can be conducted on some liquid
carbon dioxide sealed in an evacuated tube, the tube having heavy glass
walls to withstand high pressures and still permit the interior to be seen.
At room temperature the liquid rests at the bottom of the tube and the
space above it is filled with saturated vapor. As the tube is heated, the
§ 13-8 LIQUEFACTION OF GASES 233

liquid expands and its density decreases, while that of the vapor increases,
and the critical point will be passed through when the temperature reaches
31.1 °C. At this temperature the line of demarcation between the liquid and
vapor disappears and the tube presents a uniform appearance throughout.
The liquid and vapor then have the same density, and the phases cannot be
differentiated from each other.
All gases show the same general behavior with reference to liquefaction as
described for carbon dioxide. The values of critical temperature, pressure,
and volume, are, however, quite different for different substances, as the
values in Table II will show.

TABLE II. CRITICAL VALUES

Temperature, Pressure, I
Substance oc Atmospheres Volume

Air ................... -140.7 37.2 0.0047

Alcohol (ethyl) ......... 243.1 63.1 0.0071
Ammonia .............. 132.4 111.5 0.0048
Carbon dioxid~ ......... 31.1 73 0.0066
Helium ............... -267.9 2.3 0.0030
Hydrogen ............. -239.9 12.8 0.00264
Sulfur dioxide .......... 157.2 77.7 0.0075
Water ................. 374 217.7 0.00386

In order to liquefy the fixed gases, additional means beyond ordinary cool-
ing and compression are needed, and use is made of the cooling due to expan-
sion. Nevertheless, the critical temperatures are too low to be attained by
simple expansion, and this fact makes it necessary to resort to some cumula-
tive cooling action. In von Linde's process, the air is compressed to about
200 atmospheres, next cooled by means of a freezing mixture, and then passed
through a long tube from the end of which it is allowed to expand to a pressure
of about 15 atmospheres. The air is appreciably cooled by this expansion
and is then allowed to expand again, this time to atmospheric pressure, which
results in a further lowering of its temperature. Some of the air which has
been cooled by each expansion circulates around the tube from which it
issued, in order to cool the air stream before expansion cools it further. Thus,
as the 1;1ystem operates, the cooling action is progressively intensified, and
finally the temperature is lowered sufficiently to cause some of the air to
liquefy as it leaves the tube. The liquid air is collected in large containers
resembling Thermos bottles so that it can be transported without undue loss
by evaporation.
By somewhat similar processes it has been possible to liquefy all known
gases. Helium presented the greatest difficulty in liquefaction, but was
eventually liquefied at -268.9°C by the Dutch physicist H. Kamerlingh
Onnes (1853-1926). By the evaporation of liquid helium, lower and lower
234 HEAT Chap. 13

temperatures have been attained by many investigators; the lowest tempera-

ture reached is only a small fraction of one degree above absolute zero. The
final stages are accomplished by the use of a magnetic salt, which is placed in
liquid helium and magnetized by a strong field, so that its molecules take up
an orderly alignment. Then the field is removed and the salt molecules
assume random positions, the energy needed to produce this motion being
drawn from the surrounding helium.
A recent commercial development in this field is the Collins cryostat, a
machine for liquefying helium and for maintaining temperatures within a
degree of absolute zero. The helium is highly compressed, cooled by water,
and passed through an engine, where its temperature is lowered because of
the work that it does in expanding. The engine has two cylinders so arranged
Lhat the cold exhaust from one cools the intake gas of the other.
13-9. Actual Gases; van der Waals' Equation. It has long been recog-
nized that Boyle's Law, although used as a criterion for an ideal gas, is not
followed exactly, even by the so-called fixed gases. The curves in Fig. 8
show wide deviations from this law as a gas approaches the liquid phase.
Numerous attempts have been made to formulate an expression which would
agree more closely with the facts of experiment; one of these is due to the
Dutch physicist Johannes D. van der Waals (1837-1923). He considered that
in an actual gas the molecular attraction, which is absent in the concept of an
ideal gas, would be equivalent to a slight increase in the pressure applied
to the gas. Any given molecule is affected by a number of others within its
sphere of action, and this number is proportional to the density of the gas.
Further, for a given amount of gas, the number of molecules affected is also
proportional to the density. Hence, the correction should be proportional
to the square of the density, or inversely proportional to the square of the
volume V. For this reason van der ·waals replaced the pressure pin Boyle's
Law by a term ( p +; 2} He also considered that the volume should be
that into which the gas can expand and that, since the molecules themselves
occupy some space, say b, the volume in Boyle's Law should be replaced by a
smaller term (V - b). With these corrections, the relation between the ab-
solute pressure p and the volume V of a mass of actual gas at constant temper-
ature is given by the expression

(p +; 2) (V - b) = a constant,

a cubic equation wherein a and bare constants that can be evaluated experi-
mentally for each gas. For large values of p and V the corrections have but
little effect, and the equation is represented by a curve which is nearly hyper-
bolic. For smaller values of p and V the curve takes the form shown by the
full line in Fig. 9. It conforms almost perfectly to the experimental curve
§ 13-9 ACTUAL GASES; VAN DER WAALS' EQUATION

LMNP of Fig. 8, except that the horizontal portion is replaced by a curved

line.
The corrected relation between pressure and volume applies, therefore, to
either the gaseous or liquid phase of a
substance, and the transitional region in
which the equation does not apply is ~
::,
V)

known to be one of constant pressure. ~

V)

It can be shown for a fluid which satis- a..

fies van der Waals' equation that the

values of critical temperature, pressure,
and volume are given, respectively, by Volume per gram
8a a Ve= 3 b, Frn. rn-9. Graph of van der Waals'
Tc= 27 bR' Pc = 27 b2'
equation.
in terms of the constants a and b of the foregoing equation.

PROBLEMS
1. Some air, initially at 20°C, is heated at a constant pressure of 1 atmosphere
while its volume increases from 1000 to 1200 cm3 • What is the temperature of the air
after expansion?
2. A hand pump is being used to pump air into a tire in which the gage pressure
is 25 lb/in.2 The cylinder of the pump is 16 in. long. How far must the piston be
pushed down before air can enter the tire? Assume the temperature to remain con-
stant.
3. Oxygen is used from a cylinder which has a volume of 0.75 ft 3 and which was
initially at a gage pressure of 2200 lb/in. 2 When the gage reading has been lowered
to 1000 lb/in. 2 , what volume of oxygen, reckoned at atmospheric pressure, has been
removed from the cylinder? Neglect temperature changes.
4. An automobile tire contains air at a gage pressure of 20 lb/in.2 and at a temper-
ature of 60°F; the volume of the tire may be considered constant at 0.8 ft 3 • What
volume of air, reckoned at atmospheric pressure and at 60°F, must be pumped into
the tire in order to increase the gage pressure to 30 lb/in.2? Assume that all the air
is heated 5 F 0 in the process.
5. Suppose that the tire described in Prob. 4 can just withstand a gage pressure of
65 lb/in. 2 and that it contains air at a gage pressure of 30 lb/in. 2 and a temperature
of 65°F. If the temperature of the tire were raised, at what temperature would it
burst?
6. In Prob. 8 of Chap. 9, suppose the temperature of the compressed air to be
60°F and compute the temperature rise that would be needed to restore the piston
to its initial position.
7. A compressor that operates at 90 rev/min takes air from the atmosphere at
standard pressure and at 50°F to supply an industrial process that requires 50 ft 3 of air
per min at 70°F and 80 psig (pounds per square inch, gage). Compute the intake
capacity of the compressor in cubic feet per revolution.
8. A mass of air initially occupies a volume of 2 liters at a temperature of 30°C
and at atmospheric pressure; it is heated at constant pressure until its volume is
3 liters and is then heated further at constant volume until its absolute pressure is
1.5 atmospheres. Compute the final temperature of the air.
236 HEAT Chap. 13

9. Determine the gas constant for nitrogen.

10. A steel tank that has a volume of 5 ft 3 contains air at a pressure of 150 psig
(pounds per square inch, gage) and a temperature of 90°F. How many pounds of
air does the tank contain?
11. Compute the mass of air in the tire of Prob. 5.
12. Find the mass of the air referred to in Prob. 8.
13. How much heat is needed to raise 5 lb of oxygen from 100 to 200°F at constant
pressure and then to raise it from 200 to 500°F at constant volume?
14. In order to cool a large electric generator, the machine is tightly enclosed and a
stream of high-pressure hydrogen is blown through it. If the hydrogen is supplied
at the rate of 200 lb/min and passes through the machine under constant-volume
conditions, and if its temperature is raised 40 F 0 , at what rate does it remove heat?
15. What volume does 1 lb of air occupy at 32°F and 14.7 lb/in. 2 absolute pressure?
What volume would this air occupy if it were compressed to twice this pressure
(a) isothermally, or (b) adiabatically?
16. Some air that initially occupies 2 m 3 at an absolute pressure of 5 X 105
newtons/m2 and a temperature of 200°C expands to a volume of 5 m 3• What is the
absolute pressure of the air after expansion, if the expansion is (a) isothermal, or (b) adi-
abatic?
17. A quantity of air initially at 20°C and-at standard atmospheric pressure is com-
pressed adiabatically to one-tenth of its original volume. Compute (a) the absolute
pressure, and (b) the absolute temperature, of the air after compression.
18. Some air at a temperature of 1000°F and an absolute pressure of 200 lb/in. 2
expands adiabatically to an absolute pressure of 50 lb/in. 2 What is its temperature
after expansion?
19. A mass of air occupies 50 liters at an absolute pressure of 2 atmospheres and a
temperature of 25°C. It is compressed adiabatically to a volume of 10 liters and then
expands isothermally to its initial volume. Show these processes on a pressure-
volume diagram, and determine the final pressure of the air.
20. In a Diesel engine a charge of air is drawn into the cylinder and compressed to
1\ of its original volume. Suppose that the air is initially at 60°F and atmospheric
pressure and that the compression is adiabatic. Determine the gage pressure and the
temperature of the compressed air.
21. Prove that in the adiabatic expansion or compression of a gas the pressure and
temperature vary in accordance with the expression

Pi ,y-1 '1' 2'Y = P2 ,y-1 T 1'Y ,

 14
THERMODYNAMICS
The term "thermodynamics" implies a relation between heat and motion,
which has already been recognized in accepting heat as a forni of energy.
This chapter deals with the Laws of Thermodynamics and the application of
these laws to two practical aspects of the subject: the transformation of work
into heat, and of heat into work. The second of these is of great importance,
for it is the basic principle on which all forms of heat engines operate.
14-1. The Laws of Thermodynamics. The modern concept of heat dates
back to the experiments of Count Rumford in 1798, of Joule from 1840 on, and
of others in several decades following. These studies indicated that in any
process in which mechanical energy disappeared-that is, became unavailable
as mechanical energy-an equivalent quantity of heat appeared. This long
series of experiments finally led to the recognition of heat as a form of energy,
and thus the Law of Conservation of Energy became applicable to trans-
formations involving heat. The science of Thermodynamics is based upon
two laws; the first deals with the quantities involved in such energy trans-
formations, and the second with the direction in which the heat transfer takes
place.
The First Law states that whenever heat energy is transformed into any
other kind of energy, or vice versa, the quantity of energy which disappears
in one form is exactly equivalent to the quantity produced in the other form.
This law is merely a restatement of the Law of Conservation of Energy, § 6-4,
applied especially to heat. Indeed, the energy relation developed as Eq.
13.10,
dQ = dU + dW,
which states that the heat supplied equals the increase of internal energy plus
the external work done, may be regarded as a mathematical statement of
this law.
The Second Law has been expressed in many forms, and its full signifi-
cance involves a study of Thermodynamics beyond the scope of this book.
For present purposes, the following statement will suffice: Heat passes from
a region of high temperature to one of low temperature and can do mechani-
cal work in the process, but heat cannot pass in the opposite direction unless
work is done to accomplish this result. Heat naturally flows down a "tem-
237
238 lJ EA '1' Chap. 14

perature hill," so to speak, but work must be expended to take it up that hill.
In the operation of a steam engine, heat passes from a high-temperature boiler
to a low-temperature exhaust and does work in the process. In order to keep
a room cool by refrigeration, on the other hand, heat is transferred from that
room to a place of higher temperature outside, but work must he done to
effect the transfer.
14-2. Transformation of Work into Heat. Every observer is familiar ·with
the production of heat from other forms of energy: Heat is produced from
mechanical energy in an automobile brake, from electrical energy in an elec-
tric heater, and from chemical energy in many reactions. Heat is also pro-
duced as a by-product in most energy transformations, and in many of them
much of the heat is wasted.
The numerical relation between heat and mechanical energy, which is
implied in the First Law of Thermodynamics, was determined by Joule.
In one method, he placed a paddle wheel in a known amount of water and
set it into rotation by means of descending weights; the work done in stir-
ring the liquid was converted into heat by friction and caused a rise of tem-
perature. By measuring the work W done in turning the paddle wheel and
the heat Q produced in the water, he found that these quantities were in a
constant ratio, namely,
w (14.1)
J =Q,
which is known as the mechanical equivalent of heat. Joule's numerical
results for the ratio J have been modified somewhat in subsequent investiga-
tions, notably by the American physicist Henry A. Rowland (1848-1901);
the present accepted values are given in Table I. Whenever a specified
TABLE I. RELATION OF HEA'r 'rO MECHAKICAL vVORK
Equivalent Amount of
Quantity of Heat Mechanical Work

1 cal ....................... . 4.186 X 107 ergs

1 cal ....................... . 4.186 joules
1 Btu ...................... . 778 ft· lb
0.239 cal ................... . 1 joule

amount of work is transformed into heat, the quantity of heat produced can
be determined from the values tabulated. Thus, the number of heat units
corresponding to 0.287 joule of mechanical work mentioned in § 13-6 is

Q = W = 0.287 ~oule = 0. 0686 cal.

J 4 . 186 Joules
cal
As an example to illustrate the conversion of mechanical energy to heat, consider a
15,000-kg trolley car moving at 12 m per sec, which is brought to rest by the brakes.
§ 14-3 TRANSFORMATION OF HEAT INTO WORK 239

The kinetic energy of the car, by Eq. 6.7, is E" = ~ mv 2 = ! X 15,000 kg X (12 ~ )
sec
2

= 1.08 X 106 joules. This mechanical energy is transformed into 1.08 X 106/4.186
= 258 X 103 cal (or 258 kg cal) of heat. If one-half of this heat (129 kg cal) is ex-
pended in heating 8 iron brake shoes, each having a mass of 14 kg, their temperature
rise can be calculated from Eq. 12.1. The specific heat of iron is taken as 0.11 calco
gm·

( or 0.11 kg cal)
kg·Co , and the temperature rise, assumed uniform, is found to be

t = -Q =
12,9 = 10 .t>_ co •
mc 8 X 14 X 0.11

The trend at present is to express heat quantities directly in mechanical

units, and thus avoid the use of a conversion factor.
14-3. Transformation of Heat into Work. The transformation of heat
into mechanical work, which is the reverse of the process just described, is
fundamental to the operation of every heat engine. Such a transformation is
usually carried out with a gas, by heating the gas and then causing it to do
work as it expands. Usually, as a gas expands,
its pressure becomes lower; this is shown in the M
PM --- I
pressure-volume diagram of Fig. 1, which indi- I
I
cates the expansion of a gas from JJ1 to N, its I
I
pressure falling from PM to PN as its volume I
I
increases from Vu to V N. Such an expansion I
I
can be regarded as composed of a number of PN -i-
i
infinitesimal expansions over any of which the I
I
I
pressure can be assumed constant; one of these
tiny expansions i~ represented in the figure by Volume
the shaded strip. The work done in the infini-
tesimal expansion will then be the product of Frn. 14-1. The area under a
pressure-volume curve represents
the pressure and the increase in volume-that work.
is, dW = p dV, as given by Eq. 13.9. Since
the height of the shaded strip represents the pressure p and its width the
increase in volume dV, the shaded area represents the work done, dW.
Hence, the work done during the entire expansion is represented by the sum
of the areas of all the vertical strips under the curve, namely, the area
MN VNVM· Mathematically, the work done is given by the expression

W = JpdV.

In general, if the expansion or compression of a gas is represented as a

curve on pressure-volume axes, the area under the curve represents either the
work done by the gas during expansion or the work that must be done upon
it during compression. Such an area can be measured with a planimeter, or
can be computed if the equation of the curve is known. The amount of
240 HEAT Chap. 14

work determined in these ways will be expressed in mechanical units, such as

joules or foot · pounds.
When a gas expands and is later compressed so as to return exactly to its
initial condition, it is said to have completed a cycle. In the ordinary heat
engine the cycle is represented on a pressure-volume diagram by a loop, the
area of which indicates the amount of mechanical work per cycle that the
engine delivers in return for the heat supplied.
14-4. Work Done by Expanding Gas. The amount of work done by an
expanding gas in some fundamental types of expansion will be computed by
means of Eq. 13.9; each will be studied with the aid of a pressure-volume dia-
gram. Three processes will be considered: in the first the gas is assumed to
expand without change of temperature (isothermal expansion), in the second
without the application or withdrawal of heat (adiabatic expansion), and in
the third without change of pressure.

p p p

v v y
II m
Fm. 14-2. Work done by a gas expanding I, without change of temperature (isothermal);
II, without the addition or removal of heat (adiabatic); and III, without change of pressure.

Isothermal Expansion. The work done in an isothermal expansion is rep-

resented by the shaded area under the curve in part I of Fig. 2. From
Eq. 13.13, expressed in the form p V = K, the work dW done in an infinitesi-
K
mal expansion dV can be written dW = p dV = V dV, and consequently
during a finite expansion from a volume V 1 to a volume V2 the work is

W = I 1v2dV [ JV2 y2
p dV = K )vi V = K In V vi= K(ln V2 - ln V1) = pV ln Vi·

When converted from natural logarithms (ln) to those of base 10 (log), the
expression for the work done by a gas during an isothermal expansion becomes
V2 (14.2)
W = 2.303 p V log V 1 •

In this expression p represents any known value of the absolute pressure of

the gas and V is the corresponding volume; V1 and V2 represent the initial
§ 14-4 WORK DONE BY EXPANDING GAS 241

and final volumes for the expansion considered. As an alternate form of this
expression, the product p V is replaced by mRT, in accordance with the Gen-
eral Gas Law, Eq. 1.3.7; consequently,
V2 (14.3)
W = 2.303 m R T log Vi,

where m is the mass of the gas, T its absolute temperature, and R the gas
constant.
Adiabatic Expansion. The work done during an adiabatic expansion is
represented by the shaded area under the curve in part II of Fig. 2, and
can be expressed conveniently in terms of the temperature change that takes
place during the expansion. From Eq. 13.14, the behavior of a gas during
an adiabatic expansion is indicated by the expression dW = -dU, or
p dV = -mcv dT, where m is the mass of the gas, Cv is its specific heat at
constant volume, and dT is the temperature change accompanying the
expansion. Hence, the work done by the expanding gas is

W- Jp dV - -m c.JJ~ - -me,, (T, - 1',).

or W = m Cv (T1 - T2), (14.4)

in which T1 is the initial temperature of the gas and T2 its final temperature.
This expression involves the difference between the two temperatures, and
therefore the individual temperatures need not be reduced to absolute values.
Since the equation is expressed in thermal quantities rather than in terms of
pressure and volume, it gives the work in calories or in British thermal units;
nevertheless, the work is correctly represented by the area under the pressure-
volume curve and can be converted into mechanical units by using the
mechanical equivalent of heat.
Expansion at Constant Pressure. A constant-pressure expansion is indi-
cated in part III of Fig. 2; the work done is represented by the area under
the horizontal line. ,vhen the pressure is kept constant, the work done by
the gas during expansion is

W- JpdV-pJ;~,
or W = p (V2 - V1), (14.5)
where pis the absolute pressure of the gas, and V1 and V2 are respectively the
initial and final volume.
Illustrative Problems. Two problems will serve to illustrate the use of the
foregoing equations.
I. Some air which initially occupies 10 ft 3 at a gage pressure of 15 lb/in.2 expands
isothermally to atmospheric pressure and is then compressed to its initial volume at
constant pressure. Compute the net work done by the gas.
242 IIEA.T Chap. 14

The conditions of the problem :1re shown in the pressure-volume diagram of Fig. 3,
in which A.B represents the isothermal expansion and BC the constant-pressure com-
pression. The volume of the gas after expansion, VB, is found from Boyle's Law to be
VB= PAVA/Ps, where PA= (1.5 + 14.7) X 144 = 4277
p lb/ft2, VA = 10 ft3, and PR = 14.7 X 144 = 2117 lb/ft2.
A
Hence, Vs = 20.2 ft3. The work done by the gas during
the isothermal expansion is given by Eq.14.2 as W AB =
lb
2.303 p V Jog (Vs/VA) = 2.303 X 2117 ft2 X 20.2 ft3 X

log 2~~2- The logarithm of 2.02 is 0.3054, and the result

is found to be 3.01 X 104 ft· lb. This work is represented
by the total area under the curve AB. The work done
Q.11. the gas during the constant-pressure compression is

FIG. 14-3. Curves showing

the volume of a gas in-
given by Eq.14.5 as Wsc = pn(Vs - Ve)= 2117 :t X

creased isothermally and (20.2 - 10) ft 3 = 2.16 X 104 ft·lb, and is represented by
then reduced at constant the area under the horizontal line BC. Hence, the net
pressure. work done by the gas amounts to (3.01 - 2.16) X 104 =
8.500 ft· lb.
II. A tank having a volume of 5 ft3 contains air at a gage pressure of 50 lb/in. 2 and
a temperature of 250°F. Compute the work done by the air in expanding adiabatically
to a volume of 10 ft3.
As preliminary steps, it is necessary to find the mass of the air and its temperature
after expansion. Its mass is found from the General Gas Law to be m = p V /(RT),
lli ft·lli
where p = (50 + 14.7) X 144 = 9300 ft 2 , V = 5 ft3, R = ,53.3 lb· oR' and T = 250

+ 460 = 710°R. Hence m = 9300 !~

2 X 5 ft3/( 53.3 1:."!~ X 710°R) = 1.235 lb.
The final temperature of the air is found from Eq. 13.16, T1Vi-y-i = T2Vi-Y-1 , from
which Ti/T2 = (V2/V1)-Y-1 = (10/5)C1.4-1), or T1/T2 = 20-4. To evaluate this expres-
sion, take the logarithms of both sides~thus, log (Ti/T2) = 0.4 log 2 = 0.4 X 0.301
= 0.120. Then T1/'P2 = 1.32, and T 2 = 710/1.32 = 537°R. With these values of m
Btu
and T2 known and the value of Cv taken as 0.173 lb·Ro' the work done by the gas
during the adiabatic expansion is found from Eq. 14.4 to be TV = m c.., (1\ - T 2 )
= 1.235 lb X 0.173 1: ~ 0 X (710 - 537) R 0
= 36.8 Btu; this is equivalent to
36.8 X 778 = 2.86 X 104 ft· lb.
14-6. The Ideal Engine; Carnot Cycle. The French physicist Sadi Carnot
(1796-1832) described an ideal engine that has an efficiency which c~nnot be
exceeded by any heat engine working between the same temperature limits.
Carnot's studies mark the beginning of those considerations which led to the
statement of the Second Law of Thermodynamics. He introduced the con-
cepts of the cyclic process and of reversibility as applied to a heat-engine cycle.
The operating cycle of Carnot's ideal engine will be explained with the aid
of Fig. 4. The engine consists of a cylinder containing a gas as the working
substance, a source of heat at a high temperature T 1, a so-called "refrigerator"
at a lower temperature T2, and an insulating stand. The cylinder walls and
§ 14-5 THE IDEAL ENGINE; CARNOT CYCLE 243

piston are regarded as perfect insulators of heat, § 11-9; and the bottom,
or head, of the cylinder is regarded as a perfect conductor, through which
heat will pass when the temperatures inside and outside the cylinder differ
by the slightest amount. The insulating stand is considered a perfect insu-
lator, and the engine is assumed to operate without friction.
The cycle is started with the cylinder standing upon the source of heat and
with the piston held at rest near the bottom of the cylinder; the gas within
the cylinder at temperature T1 occupies a small volume at high pressure, as
indicated by point A on the diagram. Then the pressure applied to the
piston is slightly reduced to allow the gas to expand; meanwhile a quantity
of heat Q1 passes into it from the source, maintaining the temperature con-
stant. The expansion is represented by the isothermal curve AB. Next,
the cylinder is transferred to the insulating stand and the gas is allowed to

A
Temperature T1
Cylinder ...
G)
B
.,,.,,
:)

...
G) I
c.. I

~- 171
D1 I
I Heat ,c
lout~Q2I

Source lnsula~ng Refrigerator a d b c

stan Volume

Fw. 14-4. The Carnot ideal engine and its operating cycle.

expand further. This expansion BC is adiabatic, since the gas is entirely sur-
rounded by heat insulation, and the temperature falls until at point C it
reaches the value T2. When the gas reaches this temperature, the cylinder
is placed on the refrigerator and the gas is compressed by slightly increasing
the pressure on the piston. The heat generated by compression, Q2, passes
into the refrigerator-this represents the waste in the cycle-and the com-
pression CD is isothermal at the lower temperature T2. Finally, the cylinder
is shifted to the insulating stand again and more work is done on the gas,
compressing it adiabatically along DA, and completing the cycle.
Since the mechanical work in any process is measured by the area under
the corresponding pressure-volume curve, it is seen that during expansion the
gas does an amount of work equivalent to the area ABba + BCcb = ABCca,
and that during compression the work done upon it is given by the smaller
area CDdc + DAad = CDAac. The gas, therefore, does more work than is
done upon it, and the net work done by the gas during the cycle amounts to
ABCca - CDAac = ABCD, which is the area enclosed by the loop.
Throughout the Carnot cycle the gas does not depart appreciably from a
state of equilibrium. The slightest increase of external pressure during the
expansion AB would compress the gas and cause heat to be delivered to the
244 HEAT Chap. 14

source, and the slightest reduction of pressure during the compression CD

would allow the gas to expand, removing heat from the refrigerator. There-
fore, by slight changes of pressure during the cycle the engine could be oper-
ated in the reverse manner, abstracting heat from the refrigerator and deliv-
ering heat to the source. Under such conditions there would be a net amount
of mechanical work done on the gas rather than by it during the cycl0. A
cycle which can be reversed in this manner is called a reversible cycle. If
such a cycle is reversed at any point and returned to the initial state, every-
thing connected with the process is restored exactly to its original condition.
In practice, this ideal is never attained, one reason being that friction, which
is always present, causes the dissipation of some energy. Actual cycles are
thus irreversible cycles.
Efficiency of the Ideal Engine. The energy transformations in the Carnot
cycle will now be considered. During each cycle, the engine receives an
amount of energy Q1 from the source in the form of heat, of which an amount
Q2 is wasted in the refrigerator, and the remainder, Q1 - Q2, is transformed
into mechanical work; the latter is also represented by the area ABCD in
mechanical units or by (ABCD) / J in heat units. Since the cycle is reversi-
ble, friction is assumed absent, and the latter amount of energy, (ABCD)/J,
represents the output of the engine. The efficiency of any device is the ratio
of its output to its input; for the ideal engine operating on a reversible cycle
this is the ratio of the output energy (Q1 - Q2) to the input energy Q1, or
Qi - Q2 (14.6)
Carnot efficiency = Qi ,

and is independent of the working substance used in the engine.

The efficiency of the ideal engine can be expressed also in terms of the
absolute temperatures, T 1 and T2, of the source and refrigerator. Strictly,
this expression is dependent on the use of an ideal gas as the working sub-
stance, because the proof makes use of the General Gas Law, and this law is
rigorously true only for an ideal gas. In the Carnot cycle, the heat transfer
to and from the gas occurs under isothermal conditions, whence, from Eq.
14.3,
Q1 = m R T1 In (VB/VA) and Q2 = m R T2 ln (Ve/VD),
But points l'J and Care on the same adiabatic curve, and po1nts A and D arc
on another such curve, whence, by Eq. 13.16,
T1 V B,y-i = T2 V c'Y-1 and T1 V A,y-t = T2 V D'Y-1•
If the first of these equations is divided by the second, it is found that
(VB/VA)-r- 1 = (Vc/VD)-r-1 • Therefore, VB/VA= Vc/Vv, and consequently
Q1 = m R T 1 In (Ve/VD), Hence, Q1 - Q2 = mR (T1 - T2) ln (Ve/VD), and
the expression for efficiency becomes
.
Carnot effie1eney =
T1 - T2 (14 ·7)
T1
§ 14-6 THERMODYNAMIC TEMPERATURE SCALE 24-i>

In this expression, T 1 and T 2 are the respective absolute temperatures of the

working substance as received and released by the engine. Evidently, the
efficiency can be increased by raising the temperature at which the working
substance is received or lowering that at which it is rejected. Since the effi-
ciency depends only upon these temperatures, all ideal engines operating
between the same temperature limits would have the same efficiency.
Thus, for an engine supplied with saturated steam at an absolute pressure of 150
lb/in. 2 and exhausting into the atmosphere, Table VIII of Chap. 12 gives the supply
and exhaust temperatures as 358.4° and 212°F, whence, by Eq. 14.7, it appears impos-
. .
s1ble to convert more than
+
(358.4 460) - (212
+ + 460) ,.,
= 0.119 or 17.9 per cent of
358.4 460
the heat energy into mechanical form.
If the values for Carnot efficiency given by Eqs. 14.6 and 14.7 are equated,
the result can be reduced to give the proportion

(14.8)

which shows that the quantity of heat received or rejected in the ideal engine
when an ideal gas is used as the working substance is proportional to the abso-
lute temperature at which the heat transfer takes place.
14-6. Thermodynamic Temperature Scale. William Thomson-later Lord
Kelvin-used the concept of an ideal engine to establish a theoretical tem-
perature scale which would not depend upon the physical properties of any
particular substance. In order to establish this scale, he imagined a series of
one hundred ideal engines, using the same working substance, to be so ar-
ranged that the first receives heat at the temperature of boiling water and
the last gives off heat at the temperature of melting ice, with each intermedi-
ate engine having as its source the refrigerator of the one preceding. Further,
it is assumed that all of these engines operate between the same pair of
adiabatic curves, that the heat input of any engine in the series is equal to
the heat given off by the one preceding, and that the output of each engine
is the same. With these conditions e8tablished, the intermediate tempera-
tures between the two fixed points are assigned by considering the difference
in temperature between the source and refrigerator of each engine to be the
same as for the others. If the respective engine outputs are
Q1 - Qz, Q2 - Qa, · · · ,
then the corresponding temperature intervals are

a condition which requires that

T1 T2 Ta
Q1 = Q2 = Q3 ...
246 JJEA.T Chap. 14

The temperature scale constructed in this manner is based solely on thermo-

dynamical principles and is called the thermodynamic scale. Absolute zero
on this scale is that refrigerator temperature at which the ideal engine would
release no heat at all.
It will be noted that the foregoing equation is identical with Eq. 14.8,
which is based on the use of an ideal gas as the working substance. The
thermodynamic scale would be followed exactly by an ideal gas thermometer
and is followed very closely by the actual gas thermometer. Practically, the
thermodynamic scale can be considered identical with the kelvin or absolute
scale described in § 11-2.
14-7. The Rankine Vapor Cycle. The Carnot cycle previously described,
in which a.gas was used as the working substance, does not apply directly to
the ordinary engine using saturated steam because of the difference in behavior
of these substances, § 12-9. It was not until about the middle of the nine-
teenth century that Rankine proposed a modified form of the cycle that was
applicable to vapors.
In Rankine's modification, the engine parts are all the same as before,
Fig. 4, but the cylinder is assumed at the start to contain a small quantity of
water at the boiling point T1. These con-
A B ditions correspond to point A in the vapor
cycle represented in Fig. 5. The addition
~ of a quantity of heat Q1 from the source
::>
"'"'Q)... causes boiling at constant temperature and
a.. under constant pressure, as indicated by the
line AB. At point B the vaporization is
D c complete and the cylinder is transferred to
Volume the insulating stand; an adiabatic expan-
Fm. 14-5. The Rankine vapor
sion of the vapor follows (BC) with lower-
cycle. ing of the temperature to T 2• At point C
the cylinder is moved to the refrigerator,
and the volume of the vapor is reduced at constant pressure (CD); lique-
faction takes place meanwhile without change of temperature, and the heat
of vaporization Q2 escapes to the refrigerator. At a predetermined point D
the cylinder is again moved to the insulating stand and further pressure is
applied, causing adiabatic compression of the vapor (DA) and completing the
cycle.
The Rankine vapor cycle, like the Carnot gas cycle, is reversible and is,
moreover, a close approximation to that of the actual steam engine.
14-8. Entropy. It will be recalled that in the reversible cycle of the Carnot
engine, described in § 14-5, the gas absorbs a quantity of heat Q1 from the
source at temperature T1 and rejects a quantity Q2 to the refrigerator at tem-
perature T2, such that
§ 14-8 ENTROPY 247

and the efficiency of the ideal engine is (Qi - Q2)/Q1 or (T1 - T2)/T1.
In the study of heat changes in thermal processes, a quantity called entropy
is helpful. This quantity, represented by the symbol S, is so defined that
when a substance absorbs or gives out an amount of heat t:.Q at an absolute
temperature T the increase or decrease of entropy of the substance is

t:.S = tiQ_ (14.9)

T

In the brief discussion that follows, reference will be made to the quantities
just mentioned, and also to the mass m of a substance, its specific heat c, its
heat of fusion or of vaporization L, its volume V, and to the gas constant R.
Entropy changes are not restricted to gases

,: ::=~o
but apply also to liquids and solids and to A B
change of phase. The change of entropy ac-
companying change of phase is easily com-
puted, since the change takes place at constant a. I I
temperature. The quantity of heat needed to E I I
~ I I
cause melting, for example, is Q = mL, by I I
Eq. 12.2; consequently, the change of entropy S1 s,
during the process is t:.S = mL/T. In melting Entropy
100 gm of ice at 0°C, the increase in entropy
FIG. 14-6. The Carnot cycle in
is (100 gm X 80 cal/gm)/273°K = 29.3 cal/°K. temperature-entropy coordi-
When heat is absorbed during a change nates.
of temperature, for which Q = m c (T2 - T1),
Eq. 14.9 is used in differential form. Thus, when a 10-lb piece of copper is
raised from 40 to 140°F (500 to 600°R), the increase in entropy is

= 10 X 0.093 X ln (600/500) = 0.170 Btu/ 0 R.

The principal use of entropy is with the behavior of gases. During a re-
versible isothermal expansion or compression the heat absorbed or rejected
is equal to the work done; therefore, Q = m R T ln (Vz/V1), by Eq. 14.3, and
the entropy change is Q/T = m R In (V2/V1). During a reversible adiabatic
process, since no heat enters or leaves the system, tiQ/T = 0, and thus the
change in entropy is zero. These processes can be represented usefully on
temperature-entropy coordinates. Since dS = dQ/T, it follows that dQ
= T dS, and hence the area under a T-S curve represents the quantity of
heat absorbed or rejected during the process that the curve represents. The
Carnot cycle, .Fig. 4, is redrawn as a temperature-entropy diagram in Fig. 6.
Herein, AB represents the isothermal expansion, BC the adiabatic expansion,
CD the isothermal compression, and DA the adiabatic compression. The
248 HEAT Chap. 14

heat absorbed, Q1, corresponds to the area ABS2S1; the heat rejected, Q2, to
the area CDS1S2, and the engine output, Q1 - Q2, to the area ABCD. In
this form, the diagram is very simple, and the effect of temperature on effi-
ciency can be seen easily.
In the cycle of an actual engine there is some waste of energy because
of friction, and the cycle is irreversible. The heat input must be in-
creased, other factors being unchanged, to offset the energy wasted; hence,
(Qi/T1) > (QdT2), with the net result that the entropy increases.
14-9. The Reciprocating Steam Engine. The reciprocating steam engine,
invented by Watt, utilizes the expansion of steam for producing mechanical
work. The simple engine, represented in Fig. 7, has a cylinder with ports
A and B, a piston, and a slide valve. Depending on the position of this

Slide ;Steam from boiler

valve
Steam
f, Exhaust
chest ~~~~~~=J::::::_::Connecting rod

Crank Eccentric
Frn. 14-7. The reciprocating steam engine.

valve, the ports lead either to the steam chest or to an exhaust pipe entering
from the side. The piston is joined to the cross-head, and thence, through
the connecting rod and crank, to the main driving shaft of the engine. Upon
this shaft is mounted the flywheel and an eccentric that controls the slide
valve.
Steam enters the steam chest under high pressure from a boiler and, at the
instant shown, is admitted through port A to one end of the cylinder; the
other end is connected simultaneously through port B to the low-pressure
exhaust. The piston at this position is subjected to an unbalanced or "effec-
tive" pressure and is forced to the right (forward stroke); the valve closes port
A at a suitable point and the stroke is completed by the expansion of the
steam enclosed in the cylinder. The slide valve next interchanges the port
connections, joining port A to the exhaust and port B to the steam chest.
With this position of the valve, the steam drives the piston in the opposite
direction (return stroke). This cycle is repeated over and over, and the re-
ciprocating motion of the piston is converted to rotary motion of the driving
shaft and flywheel.
A diagram showing the relation between the pressure and volume in an
§ 14-10 INTERN AL COMBUSTION ENGINES 249

engine cylinder throughout a cycle can be obtained by an indicator. This

device, which is piped to the cylinder like a pressure gage, has a tracing pencil
which moves up and down as the pressure changes. The pencil rests against
a card which follows the forward and
backward motion of the piston to a re-
duced scale. The vertical motion of
the tracing point combined with the ----->-j

horizontal motion of the card results in I I

I I I
a diagram of the form shown in Fig. 8. . ,I I
The student should correlate the differ- MN,1
I

ent portions of the diagram with the

various steps of the actual cycle of the Volume
engine, and should also note the simi- Frn. 14-8. A steam-engine indicator card.
larity between this diagram and that of
the Rankine vapor cycle. As before, the area of the indicator diagram repre-
sents that portion of the energy supplied during a cycle that is available for
mechanical work.
In the ordinary steam engine the same
Inlet--..... yExhaust ports are used alternately to admit the high-
valve "'I. valve temperature steam and to exhaust this
steam at a lower temperature after expan-
sion. The repeated heating and cooling .0f
the ports is a wasteful process and is avoided
in a later development known as the unijlow
engine. In this engine the steam is ad-
Spark
plug
mitted at the ends of the cylinder and is
exhausted at the center, the piston serving
as an exhaust valve by covering and uncov-
ering the exhaust ports at the proper times.
14-10. Internal Combustion Engines.
The internal combustion engine differs
from the steam engine in that the fuel is
burned directly in the cylinder. This type
of engine, as used for automobile propul-
sion, usually has six or eight cylinders of
the form represented in Fig. 9. The water-
cooled cylinder, fitted with mechanically
Frn. 14-9. An internal-combustion operated valves, encloses a piston that is
engine. connected by a piston rod to the crank
shaft. The fuel used is gasoline, which is
atomized and mixed with air in a carburetor so as to form an explosive mixture.
The mixture is ignited at the spark gap by means of an electrical dis-
charge.
250 HEAT Chap. 14

The complete cycle consists of four strokes. The figure shows the piston
starting downward on the first stroke; the inlet valve is open, and a charge of
fresh fuel is drawn in through it from the carburetor. When the piston has
reached the bottom of its stroke and starts back, the inlet valve closes and
the piston compresses the charge in the upper part of the cylinder. As the
piston reaches the end of its upward stroke, the compressed charge is ignited
and the resulting explosion drives the piston downward during the next, or
working, stroke. On the return upward stroke of the piston, the exhaust
valve opens and the piston forces the burned gases out through the exhaust
pipe, leaving the cylinder ready for the beginning of a new cycle. At each
explosion, the heat of combustion, § 12-15, of the gasoline consumed is liber-
ated and part of this energy is converted into mechanical work.
The Diesel engine eliminates spark plugs and uses fuel oil. Air is drawn
into the cylinder and highly compressed, thus raising it to a high temperature.
A charge of fuel is then sprayed into the cylinder under high pressure and
ignites spontaneously as it mixes with the hot compressed air. Burning
takes place without explosion, and the fuel supply is regulated so that the
pressure remains almost constant during combustion.
14-11. Engine Horsepower and Efficiency. It will be recalled that of the
heat energy supplied each cycle to an engine, the portion that becomes avail-
able for mechanical work is represented by the area of its indicator diagram.
This fact makes it easy to calculate the horsepower of an engine from such a
diagram. Since the area of any figure is the product of its average height
and its base, for an indicator diagram such as in Fig. 8 the area is the average
of all the ordinates MM', NN', etc., multiplied by the base b. The average
ordinate is known as the mean e.ffective pressure-that is, the average differ-
ence in pressure on opposite sides of the piston; this value is denoted by p.
The base represents the volume swept out by the piston during one stroke;
that is, b = LA, where L is the length of the stroke and A the area of the
piston. When p is expressed in pounds per square inch, L in feet, and A in
square inches, the product pLA gives the work in foot· pounds for each work-
ing stroke. If there are N working strokes per minute, the output of the
engine becomes pLAN ft· lb per min. Since this value is obtained from the
indicator diagram, the output in horsepower is known as the indicated horse-
power (ihp), and has the value
. pLAN (14.10)
ihp = 33 ooo·
'
The actual engine may be regarded as a combination of a heat engine and
a mechanical device. First, as a heat engine, the input is the heat absorbed
from the source, which, per unit time, can be expressed in horsepower, and
the output is the indicated horsepower. The ratio of the latter to the former
is known as the indicated thermal efficiency, or
*1-l-12 THE STEAM TURBINE 251

. indicated horsepower
Indicated t h erma1 effi c1ency = horsepower . ·
of heat supply'
naturally this will be less than the efficiency of an ideal engine operating
over the same temperature range, given by Eq. 14.7. Second, as a me-
chanical device, § 6-12, the input is the indicated horsepower and the output
is the horsepower as measured by a brake (bhp). Hence,
. . brake horsepower
Mechamcal efficiency m d'1cat ed h orsepower ·
= .

The overall efficiency is the product of the indicated thermal efficiency and
the· mechanical efficiency.
For the purpose of assigning horsepower ratings to automobile engines,
standard conditions have been adopted that are equivalent to the following:
a mean effective pressure of 67.2 lb per in.2, and an average piston speed of
1000 ft per min. In the automobile engine the piston travels 4 L for each
working stroke; consequently, 4 LN = 1000. From these values, the rated
horsepower per cylinder is found by Eq. 14.10 to be
h _ pLAN _ 67.2 X (1000/4) X (,rrd2 /4) _ !.!!.._
P - 33,000 - 33,000 - 2.5'
where d is the piston diameter in inches. For an automobile engine having C
cylinders, the rating is

hp=-·
a2c
2.5
14-12. The Steam Turbine. The steam turbine makes use of the kinetic
energy of a steam jet rather than the expansion of a vapor as in a reciprocat-
ing engine. High-velocity jets are formed by
Fixed blades
passing tp.e steam through a set of fixed nozzles;
t
these jets impinge against a series of curved
vanes or blades evenly spaced around the rim
of a rotary disk, and set the disk into rapid
motion. The energy of the steam jets cannot
be absorbed by a single row of blades without
causing excessive speed. In the turbine con-
struction shown in Fig. 10, the steam issues
from the fixed nozzles and after passing one
row of rotating blades strikes a corresponding \ J
Rotating
row of fixed blades and is redirected against a blades
second row of rotating blades mounted on the
Fm. 14-10. Arrangement of
same rotor. blades in a turbine.
The theoretical output developed by a tur-
bine is equal to the reduction in kinetic energy of the steam in passing through
the machine. If W lb of steam moving at a speed of v1 ft per sec are supplied
252 HEAT Chap. 14

to the blades int min, and if this steam is discharged at a speed of v2 ft per sec,
then the reduction of energy in foot·pounds is given by Eq. 6.10 as
Wvi2 Wv2 2
2g 2g
This value represents the theoretical energy output of the turbine; the corre-
sponding power output in foot· pounds per minute is consequently
w
p = 2 gt (v12 - V22),

and can be converted to horsepower by dividing by 33,000.

14-13. The Jet Engine. The turbo-jet is a gas turbine system used in air-
planes, in which a powerful jet of combustion products is expelled backward
from the engine, and the forward reaction to this jet is the source of propul-
sion. The air for combustion is rammed into the front of the engine by the
forward motion of the plane; it is compressed by a compressor and discharged
into the combustion chamber. Here fuel oil is sprayed into the air and
ignited, raising its temperature to perhaps 1400°F and increasing its volume
at constant pressure. From the combustion chamber the gases are delivered
to a turbine, which in turn drives the compressor mentioned above. The
gases leave the turbine at considerable pressure and high speed and are dis-
charged at the back of the plane through the exhaust nozzle. Here the gases
expand adiabatically, and their internal energy is largely transformed into the
kinetic energy of the final jet.
If a mass m of air passes through the engine in time t, and if its velocity
relative to the plane is v0 at the front of the engine and v1 at the mouth of the
exhaust nozzle, then the thrust given to the plane is basically
F = (m v.tft) - (m Vo/t),
although this result may be influenced by pressure differences in different
parts of the engine.
14-14. Refrigeration. The manufacture of ice, the cooling of rooms, and
the preservation of food in cold-storage spaces are processes that require
apparatus for the production of low temperatures. In order to cool the
articles under refrigeration, heat must be removed from them and given off
to surroundings which are at higher temperatures. ·
The evaporation of a 1iquid and the expansion of a gas or vapor are known
to be processes in which heat is absorbed. These actions can be illustrated
in the making of carbon-dioxide snow, by allowing some liquid carbon dioxide
at the pressure of its saturated vapor to escape from the containing cylinder
through a small opening into the atmosphere. The resulting evaporation and
expansion take sufficient heat from the issuing stream to cause the C02 to
solidify as snow; at atmospheric pressure its temperature is -78°C.
In the commercial manufacture of ice, mentioned in § 13-::5, anhydrous
ammonia circulates continuously around a closed system, such as that rep-
§ 14-14 REFRIGERATION 253

resented in Fig. 11. The vapor is compressed in a compressor cylinder and

passes through the coils of a condenser. In these coils, which are cooled by
water, the vapor liquefies and the liquid flows to an expansion valve at the
freezing tank. Here the liquid vaporizes into an evaporator or brine coil
and is then drawn into the compressor to repeat the cycle. The evaporation
and expansion which occur at the expansion valve absorb heat from the brine
within the freezing tank and lower its temperature. Cans filled with water
are placed in the brine tank and their contents are frozen. This system can
be regarded as a reversed heat engine, operating as described in § 14-5. The
working substance (ammonia vapor) absorbs heat from the refrigerator
(freezing tank) and gives it off to the higher-temperature source (condenser),
work in driving the compressor being done on the vapor during the process.
Expansion valve

Frn. 14-11. The compression system of refrigeration.

The electric refrigerator for household use operates as a compression sys-

tem, essentially like the ice machine of Fig. 11, but with certain modifications
to adapt it to domestic purposes. Sulfur dioxide or ethyl chloride is com-
monly used as the refrigerant; the evaporator coils are located in the food
compartment, and the condenser coils, which are air cooled, are outside.
The compressor is driven by an electric motor, which is controlled by a
thermostatic switch to start and stop as needed in order to keep the tempera-
ture at the desired value.

PROBLEMS
1. How high would a 1-ton weight have to be raised in order that its potential
energy, if transformed into heat, would be sufficient to convert 1 lb of ice at 32°F to
steam at 212°F?
2. Water flowing through a pipe at the rate of 10 gal/hr is heated by a 1200-watt
immersion heater. Assume that all the heat generated is transmitted to the water,
and compute the temperature rise of the stream.
3. Some gas expands adiabatically, doing 400 ft· lb of work. It then expands
isothermally, doing an additional 900 ft·lb of work. How much heat was supplied
externally to the gas?
4. Heat is supplied to 1 kg of hydrogen at constant pressure, raising its temperature
from 20 to 70°0. Compute (a) the quantity of heat supplied, (b) the work done by
the gas in expanding, and (c) the increase of internal energy of the gas. See Table I
of Chap. 13.
254 HEAT Chap. 14

5. Refer to the expansion described in Prob. 1 of Chap. 13, and compute (a) the
work done by the expanding gas, and (b) the quantity of heat supplied externally to it
during the process.
6. What mass of air will do 5 X 104 joules of work in expanding from 14 to 6
newtons/m2 absolute pressure, if its temperature is maintained constant at 200°0?
7. Air under standard conditions is to be compressed to a pressure of 100 psia
(pounds per square inch, absolute), without change of temperature. Compute the
amount of work needed per pound of air.
8. Refer to Prob. 9 of Chap. 9 and compute the amount of work expended in
compressing the air. Assume no change in temperature.
9. A quantity of gas expands adiabatically, doing 1000 joules of work; it then
expands further isothermally while it is supplied with 200 cal of heat. Determine,
for the entire expansion, (a) how much external work was done by the gas, and (b) how
much its internal energy was changed.
10. Some gas which initially occupies 20 ft 3 at a gage pressure of 100 lb/in.2 expands
isothermally to a volume of 30 ft 3 and then expands further at constant pressure to a
final volume of 50 ft3. Compute the work done by the gas during the entire expansion.
11. Compute the efficiency of an ideal engine that delivers an output of 200 ft· lb for
each British thermal unit of heat absorbed from the source.
12. An ideal engine at the beginning of its isothermal expansion contains 1 liter of
air at an absolute pressure of 10 atmospheres and a temperature of 230°C. The air
expands isothermally to a volume of 3 liters, and then expands adiabatically until its
temperature is lowered to 100°C. At this temperature the isothermal compression
takes place, followed by an adiabatic compression to complete the cycle. Determine
the efficiency of the engine operating under these conditions.
13. Compute the amount of work done by the air during the isothermal expansion
in the cycle described in Prob. 12, and also the quantity of heat absorbed per cycle
from the source.
14. Calculate the mass of air used in the ideal engine described in Prob. 12.
15. Compute the amount of work done during the adiabatic expansion (or com-
pression) in the cycle described in Prob. 12.
16. A steam engine receives steam at 212°F and exhausts into the atmosphere at
fi0°F; it delivers 24 hp and operates at 120 cycles per min. For an ideal engine
operating under these conditions, compute (a) the efficiency, and (b) the quantity of
heat absorbed each cycle from the source.
17. An ideal engine exhausts into a refrigerator at 40°C. (a) What is the tempera-
ture of the source if the engine operates at an efficiency of 20 per cent? (b) To what
temperature must the source be raised in order to increase the efficiency to 30 per cent?
18. In a pressure-volume diagram of a Carnot cycle the areas, converted to
foot· pounds, are as follows: under the isothermal expansion curve, 23,000; under each
adiabatic curve, 15,880; under the isothermal compression curve, 17,920. Determine
(a) the quantity of heat supplied by the source per cycle, and (b) the efficiency of the
engme.
19. Compute the change of entropy of the air in Prob. 7.
20. What is the change of entropy of the hydrogen in Prob. 4?
21. Refer to Prob. 15 of Chap. 12 and calculate the total change of entropy of the
substance.
22. Construct a temperature-entropy diagram for the cycle described in Prob. 12.
23. The area of an indicator diagram as measured by a planimeter is 2.11 in. 2
Each inch of height corresponds to a pressure of 80 lb/in.2 and each inch of length
corresponds to a volume of 0.25 ft 3 • Compute the work in foot· pounds represented
by the diagram.
24. A reciprocating steam engine ]urn a piston diameter of JO in. ::md fl :citroke of
 PROBLEMS 255

12 in. Steam is admitted to the cylinder in such a way as to produce a mean effective
pressure of 110 lb/in. 2 (a) How much work does the engine perform during each
stroke? (b) What horsepower does it develop at a speed of 240 rev/min?
25. A 4-cycle (that is, 4 strokes per cycle) tractor engine has 4 cylinders with 3!-in.
bore and 4i in. stroke. If the mean effective pressure is 90 lb/in. 2, what horsepower
does the engine develop when it operates at 1400 rev/min?
26. A Diei'iel engine operating at 780 rev/min uses 0.625 lb of fuel in 10 min while
developing 58 lb· ft of torque. What is the fuel consumption of the engine in pounds
per horsepower· hour?
27. The unit of refrigeration called a "ton" is a rate of heat absorption sufficient to
produce 1 ton of ice per 24-hr day from water at its freezing point. What is a ton of
refrigeration expressed in (a) British thermal units per minute, and (b) horsepower?
ELECT-RICITY & MAGNETISM

15
ELECTRO ST A TICS
Electricity and Magnetism is that part of Physics which deals with electric
charges at rest and in motion, with magnetism, and with the forces exerted
by electric and magnetic fields. The subject has three broad subdivisions:
Electrostatics-concerned with the production of electric charges and the
action of electric fields upon them; Electric Currents-concerned with the
chemical, heating, and magnetic effects due to charges in motion; and Electro-
magnetic Induction-concerned with the generation of electromotive force
and the behavior of alternating currents. The principles of electricity and
magnetism underlie the operation of electrical machinery, such as generators,
motors, and transformers; also the various systems of electrical communica-
tion, such as the telephone, radio, and television.
Several systems of units can be used to express electrical quantities; in this
book consistent use is made of the meter-kilogram-second system (abbrevi-
ated mks system), but occasional reference is made to the older systems so
that the student can coordinate his references to published material expressed
in electrostatic and electromagnetic units. The mks system incorporates the
practical units used in electrical engineering and eliminates the powers of 10
so often present in equations utilizing other systems. Two forms of the mks
system have developed, one called rationalized and the other unrationalized.
As the rationalized form has been employed widely in advanced texts, it was
chosen for inclusion here. A comparison of the mks units with those of other
systems is given in the Appendix, and conversion factors are given there also.
Although most students of physical science have some familiarity with
electric currents and with certain practical applications of electricity, they
need to learn the basic concepts of electricity and magnetism and the way
257
258 ELECTRICITJ' & MAGNETISM Chap. 15

these concepts are related to one another. As a logical approach to the sub-
ject, it is desirable to begin with the fundamental principles of electrostatics.
15-1. Electric Charges. The earliest electrical experiment ever recorded
is probably that attributed to the Greek philosopher Thales (c 624-c 546 B.c.),
who observed that a piece of amber when rubbed with cloth was able to
attract light objects placed near it. Nowadays, the act of bringing about a
Yery close contact between the amber and the cloth is said to give the amber
a charge of electricity, and the attraction is called an electric or electrostatic
attraction. The term electricity is derived from elektron, the Greek word for
amber.
Many substances can be charged in the above manner. A glass rod will
become charged when it is rubbed with a silk cloth. A hard-rubber rod
stroked with fur becomes highly charged; it can exert sufficient force upon
a meter stick to turn it horizontally about a pivot at its midpoint..
The forces due to electric charges can be demonstrated best by using very
light objects, because the effects produced can then be observed readily.
Two pith balls, each suspended by a thread and hung a few centimeters apart,
will serve very well. When each of them is touched with a charged glass rod
the balls fly apart and remain separated; when both are touched with a
eharged hard-rubber rod they also repel each other. But if one pith hall
that has been touched with the glass rod is brought near one that has been
touched with the rubber rod, they will attract each other. From these tests
it is evident that there is a difference between the electricity on the glass and
that on the rubber. The charge on the glass is called positive and that on the
rubber negative, as originally named by the versatile American, Benjamin
Franklin (1706-1790). Furthermore, the tests show a fundamental fact of
great importance-namely, that like charges of electricity repel each other,
whereas unlike charges attract each other.
The presence of an electric charge on a body can be detected by an electro-
scope, the construction of which is indicated in Fig. 1. Two leaves, ordinarily
of aluminum or gold foil, hang side by side from a metal rod
which passes through an insulating bushing and terminates in
a metal knob outside of the case. If the knob is touched with
the charged body, the leaves will acquire electricity of the
same sign and will repel each other, as represented in the
figure. The larger the quantity of charge, the farther the
Leaves
leaves will stand apart. Thus, a calibrated electroscope can
be used to measure amounts of charge.
Frn. lS-1. A 15-2. Structure of Atoms. A more detailed consider-
simple form
of electro- ation of electric charges requires an insight into the composi-
scope. tion of matter itself. The atoms of matter are extremely
small, and the atoms of one element differ from those of
another. Despite their minuteness, it is known that atoms are made
11p of still smaller particles, some of which have electric charges. These parti-
§ 1D--2 S1'RUC'l'UR1l OF ATOMS 259

cles are called electrons, protons, and neutrons. The electron has a tiny lmt
definite charge, which is negative. The proton has 1836 times as much mass
as the electron but has the same amount of charge; however, the charge is of
opposite sign-that is, positive. The neutron has about the same mass as
the proton and possesses no charge.
The actual structure of the atom is subject to conjecture, but for many
purposes it may be pictured as consisting of a nucleus composed of protons
and neutrons, around ,vhich the electrons whirl in much the same way that
the planets move around the sun. When an atom is in an electrically neutral
or uncharged condition, the number of its electrons is equal to the number of
protons in its nucleus. This planetary picture of the atom is useful in clarify-
ing many phenomena in Physics and Chemistry.
The simplest atom is that of hydrogen; it is pictured as having a single
proton for the nucleus and a single electron whirling around it; the neces-
sary centripetal force on the electron is provided by its attraction to the
oppositely charged nucleus. Next in order of simplicity is the helium atom,
composed of a nucleus and two planetary electrons; the nucleus is regarded as
a stable combination of two protons and two neutrons. The more complex
atoms have more and more protons and neutrons in the nucleus, together
with a corresponding increase of planetary electrons. The electrons are
thought to take positions in so-called shells around the nucleus, their arrange-
ment under normal conditions being as follows:
The hydrogen (H) atom has its one electron in the first shell, and the
helium (He) atom has both its electrons in that shell. This innermost shell
accommodates only two electrons. The lithium (Li) atom, with a total of
three planetary electrons, has one of them in the second shell; beryllium (Be)
has two in the second shell, boron (B) three, carbon (C) four, nitrogen (N)
five, oxygen (0) six, fluorine (F) seven, and neon (Ne) eight; the last number
fills the second shell. The sodium (Na) atom has a total of eleven planetary
electrons, of which two fill the first shell, eight fill the second shell, and the
remaining one is in a third shell. Other elements follow that have additional
electrons in the third shell. By a continuation of this process to include the
more complex atoms, the entire Periodic Table of the elements can be con-
structed (see Appendix).
Certain elements are quite inactive chemically, from which it is concluded
that their atomic structures are inherently stable. Perhaps because of com-
pactness or symmetry, such stability is associated with electron shells that
are completely filled. Helium with its first shell complete, neon with the
first and second shells complete, and other elements similarly located in
the table (argon, krypton, xenon, and radon) are chemically inert.
The ability of atoms to combine and form molecules is determined by the
planetary electrons, and the tendency in combining is apparently to form
arrangements in which the electron shells are completely filled. A lithium or
a sodium atom, with one electron in its outer shell, is in a condition which
260 ELECTRICITY & MAGNETISM Chap. 15

favors losing this electron, while a fluorine or a chlorine atom, with one elec-
tron less than is needed to complete its outer shell, is.in a condition ,vhich
favors gaining one. When sodium and chlorine are allowed to mingle, each
sodium (Na) atom joins a chlorine (Cl) atom to form a molecule of sodium
chloride (NaCl), in which process the loosely held electron of the sodium
atom is transferred to the chlorine atom, thus leaving both atoms with com-
pletely filled shells. The measure of the ability of atoms to form molecules
by combining in this manner is known as valence; for example, sodium,
which has one electron more than is needed to fill its outer shell, is said to
have a valence number of + 1, and chlorine, which has one electron less than
is needed to fill its outer shell, is said to have a valence number of -1.
The charge on the nucleus of an atom is determined by the number of
protons in it, and this number is called the atomic number; it also represents
the number of planetary electrons in the neutral atom. The total number of
protons and neutrons in the nucleus is called the mass number.
The atoms of any one element are not all alike-some are heavier than
others. Elements containing atoms that have the same atomic number but
different mass numbers are called isotopes; they differ in the number of neu-
trons in the atomic nucleus. The average weight of the atoms of any one
element, as it occurs in nature, is nearly always the same. Instead of stating
such quantities in weight units, the value for oxygen is taken as 16.000, and
the values for the other elements are assigned proportionately. The number
thus obtained for any element is called its atomic weight.
Atomic quantities are much too small to be measured directly, but results
of great precision have been obtained from indirect measurements. For
example, the hydrogen atom, which is composed of a proton and an electron,
has a mass of 1.673 X 10-24 gm. Of this amount, the electron forms only a
small part; its mass when at rest is 9.107 X 10-28 gm. The dimensions of
the hydrogen atom, with its components assumed to be spherical, are of the
following order of magnitude: radius of nucleus and radius of electron, each
about 2 X 10-13 cm; least radius of electronic orbit, about 5 X 10-9 cm. A
better appreciation of the relative proportions of these quantities can be
obtained by imagining the atom to be magnified until the electronic orbit is
as large as that of the earth about the sun. The electron would then be rep-
resented by a sphere about the size of the earth itself and would rotate around
a nucleus of approximately equal size.
15-3. Production of Electric Charge. The process of charging a body by
rubbing it with another material may be viewed as a stripping of electrons
from some of the atoms at the contacting surface. Atoms of some elements
release electrons with comparative ease, and others acquire them readily. A
neutral or uncharged body contains equal amounts of positive and negative
electricity; when electrons are added it becomes negatively charged, and
when electrons are removed it becomes positively charged. Thus a hard-
rubber rod, when brought into intimate contact with fur, gains electrons and
§ 1,.~:)-4 CHARGING BY INDUCTION 261

becomes negative, whereas the fur loses these electrons and becomes positive
to an equal extent. A glass rod rubbed with silk loses electrons and becomes
positive, while the silk gains these electrons and so becomes negative.
Examples illustrating the production of electric charge are familiar to
everyone. The effect may be observed in dry weather by passing a rubber
comb through the hair or by shuffling the feet on a woolen carpet. A leather
belt traveling on iron pulleys may acquire sufficient electricity to produce a
spark to a person's finger held near it. The paper in a printing press usually
manifests a charge when it is separated from the rollers, and means are pro-
vided to dissipate· this charge.
It is possible to charge a body from another that is already charged, simply
by bringing the two into contact. Thus, a metal sphere gains negative elec-
tricity if it is touched with a negatively charged rod, Fig.
2. It appears that some electrons leave the rod at the
point of contact by virtue of their mutual repulsion and
attach themselves to the sphere, making it negative also.
Again, if the sphere is touched with a positively charged
rod it becomes positive, because electrons are attracted
away from it to the rod at the point of contact.
It is assumed that the sphere referred to is supported
in such a way that the electricity acquired ,vill not leak
Frn. 15-2. Charg-
away. This can be done by suspending the sphere with ing a metal sphere
a dry silk string or by supporting it on props of mica or by contact.
glass. Evidently materials like these do not transfer or
conduct electricity to any appreciable extent, and they are called insulators
or dielectrics. If the sphere had been suspended by a metallic wire or
mounted upon a metal support, practically the entire charge might escape
to the earth. It can be concluded that metals are good conductors of elec-
tricity. Many substances are neither good insulators nor good conductors
but may be classed in an intermediate group as fair electrical conductors;
for example, the human body, a piece of damp wood, and the earth.
16-4. Charging by Induction. There is a method of charging a conductor
from a charged object which does not require bringing the two into contact;
the process is called induction. In order to charge a conductor by induction,
it is necessary (1) to bring the charged object close to, but not in contact with,
the conductor to be charged; (2) to connect the conductor to ground; (3) to
break the ground connection; and (4) to remove the initially charged object.
The conductor will then have acquired a charge which will be found to be
opposite in sign to the original one.
This process can be explained by reference to Fig. 3, vvherein the conductor
is represented as a brass tube with rounded ends mounted on a glass stem for
insulation, and the charged object as a negative rod. The mechanism of the
process is as follows: (1) When the charged rod is brought near the tube, it
will repel some of the electrons of the tube; this adion makes the distant end
262 ELEC'l'RICITY & MAGNETISM Chap. 15

of the tube negative and leaves the adjacent end positive. A state of equi-
librium will soon be reached in which any other electrons repelled by the rod
are prevented from moving to the distant end of the tube by the repulsion of
the negative electricity already accumulated there. (2) When the tube is
connected to ground, a path is provided for some of the electrons to escape,

+ + +

(1) (2) (3)

d6 (4)

FIG. 15-3. Procedure in charging a metal tube by induction.

and there will be a flow of electrons through the ground connection to the
earth. (3) When the ground connection is broken, the tube is again isolated
and will have a positive charge, since it has lost some electrons. Finally,
(-!) the removal of the inducing object allows the charges on the tube to dis-
tribute themselves in a normal manner, and as a result the tube becomes posi-
tive over its entire surface.
15-5. Forces between Charged Bodies. It has been shown that charges of
like sign repel and those of unlike sign attract each other, but nothing has
been said thus far about the magnitude of the forces of
repulsion or attraction. The first quantitative measure-
ments of these forces were made by the French physicist
Charles A. Coulomb (1736-1806), using a torsion bal-
ance. This instrument, sketched in Fig. 4, has a sta-
tionary sphere A and a suspended element consisting
of two spheres B and C connected by a slender rod, all
within a glass enclosure. When spheres A and B carry
charges of the same sign, the repulsion between them
twists the supporting wire S until the torque balances
that set up in the wire. Coulomb and later investiga-
tors showed that the force between two charged bodies,
whether of repulsion or attraction, depends upon the
distance between the two bodies, their shapes, and the
Fm. 15-4. The tor-
amount an<l di:;tribution of the charge upon each of
sion balance.
them.
When the two bodies are very mnall, :;o that the charges may be regarded
as being concentrated at points, the force between them is directly propor-
tional to the product of the charges and inversely proportional to the square
of the distance between them. If Q1 and Q2 represent the charges, r the dis-
tance separating them, and F the force between them, this basic relation can
§ 15-5 FORCES BETWEEN CHARGED BODIES 263

be stated mathematically for the two charges as

Fa: Qi Qz.
r2
This proportion can be expressed as an equation by introducing a propor-
tionality factor k, thus
F = k Q1Q2. (15.1)
r2
Either form is called Coulomb's Law, and may be stated as follows: The force
of attraction or repulsion between two charges is proportional directly to the prod-
uct of the individual charges and 'inversely to the sqiwre of the distance between
them.
Equation 15.1 is strictly true only for point charges, but it can be applied
to charged bodies of finite size if their dimensions are small compared with
their separation. The force is also influenced by the presence of any medium
or substance between them, for these charges induce others in the molecules
of the intervening matter. Consequently, the equation applies specifically
to charges in a vacuum, but practically it also serves for charges in air at
normal pressures since the results differ only slightly.
Coulomb's Law explains a number of effects, including the action of a
charged rod on the pivoted meter stick mentioned in § 15-1. The rod induces
a charge of opposite sign on the stick, and the two charges attract each other
more strongly as they are brought closer together.
The value of the proportionality factor k depends upon the units chosen
for the physical quantities in Eq. 15.1. If the distance r between two charges
is expressed in centimeters and the force between them is in dynes, then the
unit in which electric charge is expressed can be chosen so that k becomes
unity numerically. Thus, Coulomb's Law with k = 1 defines the unit charge
as the charge that acts on an equal charge at a distance of one centimeter with
a force of one dyne. This unit is the basis of the electrostatic system of units
and is called the electrostatic unit of charge, abbreviated esu of charge. To
balance units in the equation, kin this system has the value
dyne·cm
k = 1 ------"-'----
2

(esu of charge) 2
In the mks system the unit of electric charge is defined on the basis of the
force between two currents, §§ 16-2 and 4, rather than between two charges,
and this unit of charge is called the coulomb. '\Vhen Qi and Q2 are expressed
in coulombs, r in meters, and Fin newtons, then the proportionality factor k
in Eq. 15.1 is
k = 8.988 X 109 newto11 ·m2
coulomb 2
(practically 9 X 109).
264 ELECTRICITY & MAGNETISM Chap. 15

The coulomb is vastly greater than the esu of charge (also called the stat-
coulomb); 1 coulomb = 2.998 X 109 statcoulombs (practically 3 X 109).
To illustrate, determine the force between two charges, one of +300 microcoulombs
and the other of -100 microcoulombs, when the two are 5 m apart in a vacuum.
From Eq. 15.1, the force is
F = 9 X 109 newton ·m.= (300 x·10-s ~ ) (100 X 10-6 ·1::,:;~~2c"'.h..) = 10 8
- 1 ~, 1• 2
0vtt1fht-t......
(5 \
b;(, 2
• newtons.

Consequently, the force between the charges amounts to 10.8 newtons (about 2.43 lb);
since the charges are of opposite sign, the force is one of attraction.
Coulomb's Law can be modified in such a way as to simplify later expres-
sions involving mks units. The change is made by replacing the proportion-
ality constant k by a fraction which includes a parameter of free space. Thus,
1
k=--,
4 1l"Eo
where Eo represents the property of free space known as permittivity. The
equation then assumes the form
F = _1_ Qi Q2. (15.2)
4 1r Ea r 2
From the foregoing, two I-coulomb charges when 1 meter apart in a vacuum
would act upon each other with a force of 9 X 109 newtons; from this con-
clusion the value for the permittivity of vacuum can be evaluated as
Qi Q2 1X 1 1 coulomb 2
Eo=--= = .
41r Fr 2 4 1r X 9 X 109 (1) 2 4 1r X 9 X 109 newton ·m2
The permittivity of air at standard temperature and pressure may often be
taken as equal to that for free space.
15-6. Electric Field Intensity. The region about a charged body is referred
to as an electric field of force or, briefly, an electric fiekl, because any other
charge located in this region will experience a force, either of attraction or
repulsion. The intensity of an electric field at any point is defined as the force
per unit positive charge placed at that point. The unit of electric field intensity
in the mks system is the newton per coulomb. Field inten~ity has direction
as well as magnitude and therefore it is a vector quantity, § 1-5.
In general, if a charge of Q coulombs located in an electric field is acted upon
by a force of F newtons, the field intensity at that place in newtons per cou-
lomb is
F (15.3)
8 = -.
Q
In this basic expression Q represents a charge placed at some point in an
already existing field, and 8 is the field strength at that point before Q was
introduced. If a coulomb of positive electricity experiences a force of 10
§ 15-6 ELECTRIC FIELD INTENSITY 265

newtons toward the right at a particular point in an electric field, then the
field at that point has an intensity of 10 newtons per coulomb and is directed
toward the right.
The direction and strength of the electric field which a concentrated charge
produces in the surrounding region vary from point to point. In Fig. 5 the
field near a charge +Q is indicated at four points,
A to D, by arrows of different lengths, all directed
away from the charge. The value of the electric
field at any point can be found by assuming a unit
positive charge to be placed at the point chosen, and +~-----r----). G).
computing the force acting upon that charge. C
Thus, the force on a unit charge at C, distant r from D
+Q, is found by placing Q2 = 1 in Eq. 15.1; the
result is F = kQ 1 (1/r 2) and hence the field inten- ~
sity at this point is e = kQi/r 2 • In general, if Q Fm. 15-5. Radial electric
represents the charge that establishes the field, then field around a point
charge.
the strength of the field at a distance r from it is

e= k f2.
r2
(15.4)

This expression shows that the field intensity due to a point charge varies
directly with the amount of charge and inversely with the square of the dis-
tance away; furthermore, it is directed radially outward from a positive
charge and radially inward toward a negative one.
The electric field is regarded as playing an important role in the attraction
and repulsion between electric charges. Each charge sets up a field around
it, and any other charge within that field experiences a force. Positive charges
tend to move along the direction of the field, and negative charge8 tend to move in
the opposite directi"on.

n
Frn. 15-6. .Electric field due to two point charges.

Since electric field intensity is a vector quantity, field intensities can be

resolved or combined in the same way as velocities or forces. When the
field intensity at a point is due to several point charges, its value can be
determined by computing the field intensity at the point due to each in turn,
266 HLECTRICITY & MAGNETISM Chap. 15

and then finding the resultant of these individual values. In part I of Fig.
6, the field intensity at point P has one component 8 1 directed radially away
from the positive charge Q1, and another component 82 directed radially
toward the negative one Q2; their resultant field intensity at this pojnt is
represented by the vector e.
As an illustration, suppose the charges shown in the figure to be Q1 = +0.0004
coulomb and Q2 = -0.0001 coulomb, these being 12 m apart in free space. Deter-
mine the resultant electric field intensity at point P located 5 m directly below Q2 ;
this point will be V (12) 2 + (5) 2 = 13.0 m away from Q1 •
With k taken as 9 X 109 newton·m 2/coulomb2, the components of the electric field
at P will be
81 =9X 109 0.0004 = 2.13 X 104 newtons,
(13) 2 coulomb

and 82 = 9 X 109 0.0001 = 3.60 X 104 newtons

(5) 2 coulomb
Next, 8 1 is resolved into a vertical component of fa X 2.13 X 104 = 0.82 X 104 down-
ward, and a horizontal component of g X 2.13 X 104 = 1.97 X 104 to the right, as
in part II. The net vertical field intensity is (3.60 - 0.82) 104 = 2.78 X 104 • Hence
+
the resultant is 8 = 104 V (2.78) 2 (1.97) 2 = 3.41 X 104 newtons per coulomb. The
direction of this resultant field intensity is given by tan() = 197/278; whence() = 35.3°
to the right of the vertical.
The newton per coulomb is smaller than the unit of electric field intensity
in the electrostatic system which is called the dyne per statcoulomb; 1 newton
1
per coulomb = 3 X 104 dynes per statcoulomb.
15-7. Electrostatic Lines of Force. The electric field around an electric
charge has a definite direction and magnitude at every point within it. The
characteristics of the field can be por-
trayed by so-called electric lines of
force, which show by their direction
the direction of the electric field and
by their closeness the magnitude of
the field. These lines of force have
no physical existence, but they are
helpful in visualizing the configura-
tion of the field. Electrostatic
lines of force are imagined to start
at a positive charge and to terminate
at a negative charge. For the field
Frn. 15-7. Some lines of force between around a charge on a sphere isolated
two electric charges. in space the lines are directed radially
outward if the sphere is positive and
radially inward if negative. Divergence of the lines indicates a weaken-
ing of the electric field; convergence a strengthening of the field.
§ 15-7 ELECTROSTATIC LINES OF FORCE 267

The configuration of the field between two equal charges, one positive and
the other negative, is illustrated in Fig. 7 by a few lines extending from one
charge to the other. Such a field can be mapped experimentally by sprinkling
small gypsum crystals or cork filings on a sheet of insulating material placed
in the field.
The spacing of the electric lines of force to indicate intense or weak electric
fields can be made definite by associating a certain number of lines per unit
area of surface, oriented at right angles to the field, with the intensity of field
to be pictured. The number chosen is equal to the product of the permittivity
of free space and the field intensity, thus
D = Eo 8,
where D represents the density of the lines in a region in which the field inten-
sity is 8. If this intensity is the same at all points of a surface of area A that
is perpendicular to the field, the total number of lines through this surface is
'1' = DA.
In situations when the field is not uniform and not perpendicular every-
where to the surface, the total number of lines can be determined by integra-
tion, as indicated by the expression
'1' = fD cos 8 dA,

-vvhere O is the angle between the direction of the field and a normal to the
differential area dA.
The relations given above can be used to find the number of electric lines
of force that extend from an isolated point charge. If a sphere of radius r is
imagined to surround such a charge Q and to be centered upon it, the field
intensity at any place on the spherical surface is given by Eq. 15.4 as
8 = k ~
r
Q 2 , and the lines will extend radially outward through the
= -
4 1rt:or
:,;urface. Consequently, the density of the lines is everywhere the same over
the surface and has the value
D = <=08 = _!l__
4 7rr2

Since the sphere has a surface area of 4 1rr2 , the total number of lines through
the surface is

'1' = DA = 4 ';,.2 4 7rr2 = Q.

This result shows that the total number of lines extending from the charge is
numerically equal to the charge-for example, one line from a charge of one
coulomb. The total number of lines from an isolated charge is independent
of the radius of the sphere chosen, but the density of the lines through its
surface varies inversely as the square of the radius.
268 ELECTRICITY & MAGNETISM Chap. 15

15-8. Electric Potential. From the fact that an electric field exerts a
force upon a charge located in the field, it is evident that work must be done
upon such a charge in moving it against this force. If the field has uniform
intensity, the work done will be the product of the constant force and the
distance the charge is moved, provided that the charge moves along the di-
rection of the force. If the field is not uniform, the force will vary from
point to point along the path of the charge, but the work done can be com-
puted easily by using a concept known as potential.
Potential, in Electricity, refers to electrical level just as in Mechanics it
refers to gravitational level. In Mechanics, work must be done upon a body
to move it from a given level to a higher one; similarly in Electricity, work
must be done upon a + charge to move it from a given potential to a higher
one. Thus, points in space near charged objects have definite potentials,
and work is done whenever charges are transferred from one potential to
another. Although the concept of potential is associated with work done on
a charge, potential characterizes the electric field and not the charge.
The potential at any point is defined as the work per unit charge required to
move a positive charge from an infinitely great distance to the point in question.
If Wis the work done upon a test charge Q to bring it from infinity to a par-
ticular point in an electric field, then the potential at that point is given by
the basic equation

V=-·
w (15.5)
Q
With the work W expressed in joules and the charge Qin coulombs, the poten-
tial will be expressed in joules per coulomb. A potential of one joule per
coulomb is named the volt, after the Italian physicist Alessandro Volta (1745-
1827). Hence, from Eq. 15.5, volts X coulombs = joules.
To derive an expression for the potential at any point due to a specific
charge, reference will be made to Fig. 8, which shows a charge +Q located
at point 0, and a computation will be
made of the work done upon a unit
I I
I I positive charge in moving it from a
+04'0
I
•+
I I remote point at the right up to a point
I I
---,--- -,,.I
i.,-
I I
R distant r from O. · When the unit
-dx charge is at a distance x from 0, it is
F 1 G. 15 • - 8• .
D iagram · obtarnmg
use d 111 · · th e in a field of intensity kQ/x 2 and conse-
potential at a point in space. quently experiences a force (kQ/x 2)
X 1 directed away from the charge Q;
an equal force directed toward Q is required to hold it at this position. Hence
the work done in moving it an infinitesimal distance -dx nearer to Q is
- k~ dx. The total work done in moving the unit charge from an infinitely
x
great distance to point R is found by integrrJ.ing this expression between
§ 15-8 ELECTRIC POTENTIAL 2fi9

the limits x oo and x = r; thus,

f ~
r - k~ dx = [+kQ]r =kQ
x x ~
-·
x
Consequently, the potential at a point R distant r from a point charge Q is
v = kQ_ (15.6)
r
In this expression the potential will be in volts when Q is in coulombs and
r in meter8; with. these units k = 9 X 109 newton·m 2 /coulomb 2 , as before.
Thus, at a distance of 1 meter in free space from a concentrated point charge
of + 1 microcoulomb, the potential is +9000 volts. The volt is smaller than
the unit of potential in the electrostatic system; the latter is called the erg per
statcoulomb, or statvolt; 1 volt = s6 0 statvolt.
Potential is completely expressed by a statement of magnitude; it is there-
fore a scalar quantity. The potential at a single point due to several charges
ean be found by adding algebraically the individual potentials due to the
several charges. The difference of potential between two points is found by
subtracting the potential at one point from that at the other point. This
difference represents the work that would be done in moving a unit positive
charge from one point to the other. The difference of potential between two
points is used more in practice than are their actual potentials.
The work done in transferring a quantity of electricity from one point to
another in an electric field can now be determined readily. Since the work
done in transferring unit charge between two points is equal to the potential
difference V between the points, the work done in transferring any charge Q
between these points is W = QV, as indicated by Eq. 15.5. Furthermore,
the work done depends only upon the positions of the original and final points
and is independent of the path taken by the moving charge.
As an illustration, consider two charges, one of +2 X 10-4 coulomb and the other
of +5 X 10-5 coulomb, initially 20 m apart in a vacuum, and determine the work done
in bringing them closer together so that their final separation is 18 m.
Suppose the larger charge to remain at rest; the potential that it establishes at a
point 20 m away is V 20 = 9 X 109 2 ~io- 4
= 9 X 104 volts, and that at a point 18 m

away is Vis = 9 X 10 9
2 x 10-
4
= 10 5 volts. The work needed to move the other
18
charge, the one of 5 X 10-5 coulomb, through the potential difference 105 - 9 X 104
volts is equal to 104 volts X 5 X 10-5 coulomb = 0.50 joule. The same result would
be obtained if the smaller charge were kept at rest and the other charge were moved
the same distance.
Work must be done to move a positive charge from a point of low potential
to one of higher potential; under the same circumstances a negative charge
would move by itself and would acquire kinetic energy. The free electrons
(§ 15-10) in a conductor will move if there is a potential difference between
270 ELEC1'RICJ1T & MAGNE'PISM Chap. 113

any two points on the conductor. Conversely, if a conductor bears a charge

that is not moving, its surface must have the same potential at all points.
Such a surface is called an equipotential surface, and it follows from the defi-
nition of potential that no work would be required to transport electrical
charges from one point to another on an equipotential surface.
15-9. Field Between Parallel Charged Plates. The electric field between
two parallel plates is uniform, except at the edges; if the plates are large com-
pared with the distance between them, the edge effects may be disregarded.
The intensity of the field between the plates can be computed by considering
the work that must be done upon a charge to move it from one plate to the
other.
In Fig. 9, the work done upon a positive charge in moving it from the
negative plate to the positive one against the direction of the field will be
some force F multiplied by the distance

~ ll!lllf lf~~IIIllllII\IfIf fi~;~t~(I~~~~~~~~f~~~;~i

FIG. 15-9. A uniform field between par- V between the plates. It follows that
allel charged plates. Fs = QV, or
F V (15.7)
Q= -;·
But F /Q, the force per unit charge, is the same as the field intensity 8, by
Eq. 15.3; whence the field between the plates has an intensity of

8 = -.
v (15.8)
s
If V is in volts and s in meters, then 8 is in volts per meter. (The student
should show that this unit is the same as the newton per coulomb mentioned
in § 15-6.)
15-10. Motion of Electric Charges. The transfer of electric charge from
one place to another is called the conduction of electricity. The mechanism
of this transfer differs between solids and fluids. In solids, the closeness of
the atoms to one another makes it possible for the outer electron orbits of
neighboring atoms to overlap. Moreover, most conducting solids are metals
in which the atom has only a few electrons in the outermost shell, a condition
which favors the loss of these electrons. Under such circumstances, and pos-
sibly because some electrons serve in a double capacity in two adjoining atoms,
it is believed that a large number of electrons are comparatively free to move.
When one end of a conducting wire is maintained negative by supplying it
with electrons and the other is maintained positive by withdrawing electrons
from it, a field is set up which is directed away from the positive and toward
the nega~ive end. The free electrons, being negative, are urged in a direc-
§ 15-10 MOTION OF ELECTRIC CHARGES 271

tion opposite to this field, and acquire a definite drift toward the positive end
of the wire. It is this flow of electrons which constitutes an electric current,
§ 16-1. Ordinarily the free charges in a conductor are pictured as being in
a state of unceasing, unordered motion, depending upon the temperature;
when there is current through the conductor, the directed drift is superposed
upon this random motion.
It is interesting to note that in solids the positively charged atoms remain
fixed in position, except for thermal agitation, and that the current is attributed
entirely to a movement of the free electrons. These free electrons are known to
assist also in the conduction of heat. In insulating solids the electrons are
not readily detachable from the atoms. If an insulator is placed in an elec-
tric field, the electrons do not drift through it but merely undergo a slight
shift in position with respect to the positive charges of its molecules.
The conduction of electricity in fluids is attributed to the presence of so-
called ions, a term which is applied to atoms or atomic groups that are
electrically charged. The motions of these ions in liquids and gases are due
to the forces experienced by charges when they are located in an electric field;
because of such forces the charges are set in motion, and in this way electric
currents are established in such mediums.
To demonstrate the flow of electricity through gases, suppose that an
electroscope has been charged so that its leaves stand apart as in Fig. 1, and
that the flame of a gas jet is brought near the electroscope knob. It will be
found that the leaves fall together promptly, showing that the electroscope
has lost its charge. The molecules of the heated gases collide with each
other and with air molecules so forcibly as to knock some electrons out of the
atoms. As a result, the air becomes ionized, and the neutral molecules be-
come interspersed with ions carrying negative and positive charges. If the
electroscope is positive, the field will be directed away from it and negative
ions will move toward it; if negative, the field will be directed toward it
and positive ions will move in this direction. In either event, the charge on
the electroscope will be neutralized and an equal charge dispersed in the sur-
rounding medium.
Again, when a charged electroscope is exposed to the radiation from ura-
nium, radium, or other radioactive substance, the electroscope loses its
charge, because the radiation has ionized the air. Ionization of the air is
also produced by x-rays and ultraviolet light, as well as by cosmic radiation
reaching the earth from remote sources. Whatever the cause, practically
every sample of air will include some ions.
When two highly charged electrodes are separated in air at ordinary pres-
sure, the few ions already present in the air will be accelerated so strongly
that they create others by collision, and these in turn produce still more
ions. By this cumulative action the air becomes highly conductive, and the
charge escapes from the electrodes in the form of a brush discharge or corona
which can be seen in the dark. ·when the electrodes are brought sufficiently
272 ELECTRICITY & MAGNETISM Chap. 15

close to each other, the ionization occurs so rapidly that a spark will jump
between them; this so-called disruptive discharge is accompanied by a sharp,
quick snap or crackle. Disruptive discharge may occur also in insulating
liquids and solids.
Gaseous ionization is used in eliminating smoke from chimneys. In the
Cottrell process the smoke is passed between a pair of oppositely charged
electrodes, one being a wire of small size so that the electric field nearby will
be intense. In this region the gas becomes ionized by collision and the sus-
pended carbon particles acquire charges by contact with the gaseous ions.
If the wire is positive, as in the usual design, the positively-charged smoke
particles will move along the outwardly directed field and be deposited on the
negative electrode.
16-11. Distribution of Charge on Conductors. The shape of a conducting
body has a marked effect on the distribution of electricity over it, and in gen-
eral the amount of charge per unit area will not be uniform over its surface.
The amount of the charge at any spot can be measured with an electroscope
by using a proof plane-merely a small metal disk fastened to one end of an
insulating rod. The disk of the proof plane is applied to the spot under test
and then to the electroscope; the resulting separation of the electroscope
leaves becomes a measure of the surface density of charge.
A charged sphere, whether hollow or solid, when isolated so as to be unin-
fluenced by its surroundings, will show the same separation of the electroscope
leaves when tested with the proof plane at all points on the outer surface; this
result indicates that the surface density of charge on a spherical conductor
is uniform. On an elongated conductor, such as a rod or tube, this density
will be found to be greater at the ends than at the middle, a result which
shows the tendency of the individual charges to repel one another to the great-
est possible distance. The surface density is particularly large at sharp
points, and the charge may escape into the surrounding air. To minimize
such leakage, rounded knobs are used instead of sharp points on electrostatic
apparatus.
The distribution of electric charge can be studied further by repeating an
experiment originally conducted by the British chemist and physicist Michael
Faraday (1791-1867), using a metal ice pail, an electroscope, and a metal
sphere, arranged as in Fig. 10. ·when the sphere is chargecl from an outside
source and lowered into the pail, the leaves of the electroscope diverge. The
sphere may then be moved around inside the pail, touched to its inner sur-
face, and removed, without causing any further change in the electroscope.
After the sphere is touched to the pail and removed, both the sphere itself
and the inner surface of the pail will be found to be entirely free from charge.
To explain this result, it will be assumed that a positively charged sphere
has been lowered to the position shown. The sphere attracts some of the
free electrons in the metal pail to the inner surface, and consequently the
outer surface and the electroscope are left positive. When the sphere is
§ 15-12 ELECTROSTATIC GENERATORS 273

touched to the pail, the induced charge on the inside of the pail and the
inducing charge on the sphere neutralize each other. The charge on the
outside of the pail, which before contact was as large as that on the inside
but opposite in sign, must therefore be equal to the
initial amount on the sphere, and of the same sign.
Consequently it may be concluded that a charge
induces an equal and opposite charge on the sur-
rounding surface. Further, it is seen that a charge
cannot exist inside of a conductor unless this region
also contains an equal and opposite charge; instead,
it will reside on the outer surface. In this experi-
ment it is desirable to have the opening in the
receptacle as small as possible, for the action will + +
not be complete unless the hollow conductor is
Frn. 15-10. Faraday's
entirely closed. ice-pail experiment.
Atmospheric Electricity. Under certain atmospheric
conditions the clouds gather electricity, either by contact of electrically
dissimilar layers of air or by the falling of charged raindrops, and this induces
a charge of opposite sign on the surface of the earth below. If the inter-
vening electric field is sufficiently intense, a disruptive discharge will occur
as a stroke of lightning.
To protect buildings and other structures from damage which may result
from lightning strokes, the structures are equipped with lightning rods.
These are large conductors, well grounded at the bottom and terminating in
sharp points at the top, arranged to form as complete an enclosure of the
building as is practicable. Such an arrangement tends to prevent the interior
of the building from becoming electrically charged, since the conductors form
at least a partial metallic shell around it. Also, the arrangement permits the
electricity induced on the neighboring portion of the earth to escape from the
sharp points and, in this way, tends to prevent the accumulation of sufficient
charge to result in a lightning stroke. In the event of a stroke, the large
conductors provide an easy path for the current and tend to divert it from
the building directly to the earth. Although there is no doubt that buildings
equipped with lightning rods have been damaged by lightning, nevertheless a
,vell-designed and properly installed system of lightning rods is recognized as
a valuable safeguard against such harm.
15-12. Electrostatic Generators. For the operation of discharge tubes, for
the production of x-rays of high penetration, and for the acceleration of
charged particles for research on the atomic nucleus, use if often made of
electrostatic generators to develop the necessary high potentials. The mod-
ern machines are of the type designed by Professor Robert J. Van de Graaff of
the Massachusetts Institute of Technology and developed jointly with his
colleague, Professor John G. Trump.
A typical design is represented diagrammatically in Fig. 11. It consists
274 ELECTRICITY & MAGNETISM Chap. 15

essentially of a hollow metal terminal T mounted on an insulating column C,

within which a motor drives an endless belt B of insulating material for con-
veying electricity between the terminal and ground. The· terminal may be
made either positive or negative; the operation described is for the latter.
A row of metal points represented by P is
directed toward the lower metal pulley L and
receives continuously a small supply of negative
..., electricity from an auxiliary source A. The air
in the vicinity of the points becomes ionized,
and the negative charges are repelled toward
the pulley and deposited on the intervening
insulating belt. This neutralizes the positive
charge that may be on the belt from its down-
ward run and leaves a net negative charge to
c + be carried to the upper terminal.
+
The collector Q transfers this charge partly
to the upper insulated pulley U and partly
(through suitable control devices indicated by
zigzag lines) to the terminal. The charges are
so distributed as to make the pulley more
Ground -:c- negative than the terminal itself and so the
Frn.15-11. The Van de Graaff- pointed electrode R connected to the terminal
Trump type of electrostatic may now be regarded as positive with respect
generator. to the pulley. In consequence, the air be-
tween the two becomes ionized and positive
charges are attracted toward the pulley and deposited on the belt. This action
neutralizes the negative charge remaining on the belt and leaves a residue of
positive charge to be carried downward. It is evident that, as the machine
is driven, there is a continuous transfer of negative electricity up one side of
the belt and positive electricity down the other side. As a result, the ter-
minal acquires a large negative charge, and this brings about a high negative
potential, as desired.
15-13. Measurement of the Electronic Charge. An achievement of great
importance, which furnished an insight into the nature of electricity, was the
determination of the charge of the electron in 1913 by the American physicist
Robert A. Millikan (1868-1953). His method briefly is to charge a tiny oil
drop while in an electric field, to direct the field so as to urge the droplet
upward against the pull of gravity, and to adjust the field strength so as to
hold the droplet stationary. Under this condition the upward force due to the
field equals the downward force due to gravity (the weight of the droplet), a
fact which permits the quantity of electricity on the droplet to be computed.
The apparatus, shown in Fig. 12, consists of a closed chamber C containing
near the top an atomizer A, and near the bottom a pair of parallel plates PP
whieh can he connected to a source of potential B. The plates are separated
§ 15-13 MEASUREMENT OF THE ELECTRONIC CHARGE 275

by a few insulating blocks, not shown, and the air between them is ionized by
x-rays from the tube X. The central region between the plates is illuminated
by the arc lamp L, and observations can be made of that region by means of
a microscope at the front.
Oil is sprayed from the atomizer into the chamber and falls as a mist.
Some of the droplets enter the space between the plates, the upper one of
which is pierced with a pinhole for this purpose. After a droplet is singled
out, a preliminary measurement is made with the plates uncharged to deter-
mine its uniform speed of fall; this is done by noting the time required to fall

B
l l l!
1 1 1 1J11--~1--,
c

Fm. 15-12. :.\iillikan's oil-drop apparatm,.

a small known distance. A knowledge of this speed makes it possible to

determine the radius of the droplet from Stokes's Law, § 10-10, -which gives
the radius as
;g:,;;
r='\J2dg'
where r, is the coefficient of viscosity of the air, d the density of the oil, g the
acceleration due to gravity, and v the terminal uniform velocity of the droplet.
The mass m of the droplet can then be expressed in terms of its radius rand
density d as

The plates PP are next charged from the source B, and the potential dif-
ference between them is adjusted to a value V such that the same droplet
will remain stationary at some convenient point between the plates. The
electric field strength in this region is 8 = V /s, wheres rep"resents the separa-
tion of the plates. If the droplet has acquired an amount of electricity q
from the ionized air, it will experience an upward force F = q V / s. But this
is equal to the downward force on the droplet due to gravity, whence qV /s
= mg, from which the quantity of electricity on the droplet becomes
mgs (15.9)
q= .
v
276 ELECTRICITY & MAGNE1'ISM Chap. 15

Millikan found that the smallest charge ever acquired by a droplet has a
definite value and that all others are exact multiples of this smallest value;
hence it is inferred that electricity consists of tiny discrete charges. Accord-
ing to the latest determinations this elementary quantity, the charge of the
electron, is
e = 4.802 X 10-10 statcoulomb = 1.602 X 10-19 coulomb.
No electrical charge of either sign smaller than this has ever been observed.

PROBLEMS
1. The nucleus of the helium atom contains 2 protons and 2 neutrons. What is
the atomic number of helium, and what is the mass of its nucleus?
2. Sulfur has an atomic number of 16, a mass number of 32, and an atomic weight
of 32.07. How many electrons does a neutral atom of sulfur have, and how many are
there in each shell around the nucleus? What is. the valence. number of sulfur? What
is the composition of its nucleus?
3. A negative point charge of 1 microcoulomb is placed 30 cm away from an
identical charge in air. What force does each charge exert upon the other?
4. At what separation would two electrons repel each other in air with a force of
1 dyne? What gravitational force would exist between these electrons at the same
separation? The charge of the electron is given in § 15-13, and the gravitational
constant is given in § 4-5.
5. Charges of 100 microcoulombs are located in a vacuum at the corners A, B, and
C of an equilateral triangle measuring 4 m on a side. If the charges at A and C are
positive, and the charge at B is negative, what is the magnitude and the direction of
the total force on the charge at C?
6. Assume that the charges in Prob. 5 are all positive and calculate the resultant
force on any one of them due to the other charges.
7. Two tiny pith balls, each having a mass of 0.05 gm, are supported in air from a
common point by silk threads each 12 cm long. When equal charges are placed upon
the pith balls, their supporting threads stand 30° apart. Compute the charge on each
pith ball.
8. What electric field intensity exists in free space at a point 2 m from a concen-
trated charge of -20 microcoulombs? What force will this field exert upon a con-
centrated charge of +s microcoulombs placed at that point?
9. A charged body which has a mass of 0.1 gm is held stationary in space by placing
it in an upwardly directed electric field having an intensity of 105 newtons per coulomb.
Find the charge on the body.
10. At standard pressure, the air near a charged conductor which is flat or only
slightly curved becomes conducting when the field intensity at the'conductor surface
reaches 3 X 106 newtons per coulomb, and the charge then leaks off. What is the
largest charge that can be put on a conducting sphere 2 in. in diameter without exceed-
ing this limiting field intensity?
11. Two equal point charges of 50 microcoulombs, one positive and the other nega-
tive, are located 2 m apart in air. Determine the electric field that they produce
(a) midway between the charges, and (b) at a point equally distant from the charges
and 0.75 m away from the line joining them.
12. A "dipole" formed of charges q and -q separated by a distance 2s is located in
a vacuum. Show that the electric field intensity due to the dipole at 11 distance r from
1'ts m1"dpomt
· an d m· 1·me wit · (36
· l1 t h e ch arges is q r 8) 109 •
r2 - s2 2
 PROBLEMS 277

13. A point charge of +o.2 microcoulomb is located in air 50 cm from a point

charge of +0.5 microcoulomb. At what point on a line connecting the two charges
is the field intensity due to the charges zero?
14. Compute the potential due to the charges described in the problem of § 15-6 at
the point P in the associated figure.
15. Point charges of 1.5, -2.5, and 5 microcoulombs are located in air at the vertices
of a right triangle as shown in the figure, the hypotenuse
being 80 cm long. What is the potential at the midpoint +5
of the hypotenuse?
16. Refer to the charges described in Prob. 13 and find
the field intensity and also the potential at a point midway
between the charges·.
17. Two large conducting plates are supported so that -2.5 +1.5
their planes are vertical, parallel, and 12 cm apart. The PROBLEM 15.
potential difference between the plates is 15,000 volts. A
charged pith ball having a mass of 0.04 gm is suspended by a thread and held in the
air space between the plates, and test showed that the thread supporting the pith ball
is inclined 30° with the vertical. What is the charge on the pith ball?
18. Find the potential at a point 30 cm from a point charge of +4 X 10-7 coulomb
in a vacuum, and also at a second point 20 cm from this charge. How much work
would be necessary in order to move a quantity of 10-9 coulomb from one of these
points to the other?
19. The difference in potential between two electrodes in a certain vacuum tube is
200 volts. If an electron is released with negligible velocity at the negative electrode,
with what speed does it strike the positive electrode?
20. In a certain Van de Graaf-Trump generator the potential of the terminal is
maintained 1.5 million volts different from that of the place where charges are sprayed
upon the upwardly moYing belt. Assume the belt to deliver negative charge to the
terminal, and to remove positive charge from the terminal, both at the rate of 1800
microcoulombs per sec. Compute the power needed to drive the belt against the
electrical forces alone.
21. Two plates, each 20 cm in diameter, are placed horizontally one over the other
and 0.6 cm apart in air. A drop of oil of density 0.86 gm/cm 3 was introduced into
the region between the plates. The diameter of the drop was determined to be
0.00015 cm, and it was found that a potential difference of 111.6 volts between the
p]ates held the drop at rest. How many electrons did the drop carry?
22. Refer to the preceding problem and establish what the limiting speed of the oil
drop in the air between the uncharged plates must have been in order to conclude that
its diameter was the value given. Assume the viscosity of the air to be 1.80 X 10-4
p01se.
23. At what speed must the electron in the hydrogen atom revolve to prevent it
from being pulled toward the nucleus by electrostatic attraction if the radius of its
orbit is taken as 5.29 X 10-9 cm? How many revolutions does it make per second?
24. A particle of radius r carrying a charge q is located in an electric field of in-
tensity 8, the surrounding medium having a coefficient of viscosity 17. Show by means
of Stokes' Law that the speed of the particle due to the field can be expressed as

Sq
v=--·
61rw
State the units for the several entities so that the equation will balance.
25. An electron is shot horizontally at a speed of 1.5 X 108 m/sec into a vertical
electric field having an intensity of 1.2 X 105 volts perm. How far will the electron
278 ELECTRICITY & MAGNETISM Chap. 1!i

be deflected vertically by the field while traveling a horizontal distance of 10 cm

through it? Take the mass of the electron at this speed to be 10.5 X 10-28 gm.
26. Suppose a charged dust particle of 0.001 mm diameter to be located in a hori-
zontal electric field having an intensity of 7500 volts per cm, the surrounding medium
being air at 20°C, for which the coefficient of viscosity may be taken as 1.80 X 10-4
poise. From the expression developed in Prob. 24, find the number of electrons on
the dust particle if it travels a horizontal distance of 3.4 cm in 1 sec.
 16
ELECTRIC CURRENT AND
ELECTROMOTIVE FORCE
Consideration has been given so far to electric charges at rest; attention
will now be directed to charges in motion-that is, to electric currents. Just
as two electric charges in space exert forces upon each other, so two electric
currents in parallel wires exert mutual forces; from this fact is derived the
concept of unit current strength. The effects produced by electric currents
are numerous, but three of them are of outstanding importance; they are:
the chemical effect, heat production, and electromagnetism. The first of
these, to be considered in this chapter, deals with Electrolysis, a process that
is utilized in electroplating, in battery operation, and in the manufacture of
chemicals.
16-1. Electric Circuits. The concepts of electric charge and potential,
developed in the preceding chapter, are also essential in the study of electric
currents. When an extended conductor has different potentials at its ends,
the free electrons of the conductor itself are caused to drift from one end to the
other. In order for this flow to continue, it is necessary that the potential
difference be maintained by some electrical source such as an electrostatic
generator, or, much more frequently, a battery or a direct-current generator.
The wire and the electrical source together form an electric circuit, and the
electrons drift around it as long as the conducting path is maintained. Such
a flow of electrons constitutes an electric current.
Batteries and direct-current generators are sources of potential difference
which urge the electrons around a circuit continually in one direction, pro-
ducing a unidirectional current. For this reason such a source is said to
have fixed polarity; one terminal is called positive and the other negative.
If it is desired to reverse the flow, then the terminals of the circuit must be
reversed with respect to the source.
From the early days of electrical science, current has been regarded as a
flow of electricity from the positive terminal to the negative terminal in the
external circuit connected to a source. It is now known that a current
through a conductor is actually a movement of electrons, and since these
have negative charges, they travel around the external circuit from the nega-
tive terminal to the positive terminal. The electron flow is, therefore,
279
280 ELECTRICITY & MAGNETISM Chap. 16

opposite to the conventional direction of current, making it necessary, in order

to avoid confusion, to distinguish one from the other by name.
16-2. Forces between Conductors. Besides the forces that are exerted
between conductors having static charges of electricity, there are also forces
between conductors that carry electric currents. This action is due to the
magnetic effect of the current, as explained in Chap. 18; it can be demon-
strated by suspending two wires close together from an elevated support and
setting up currents in them. The forces cause the
wires to assume the shapes shmvn in Fig. 1, wherein
the directions of the currents are indicated, but the
circuit arrangements for providing the currents
are not shown. It will be observed that the con-
t ductors are attracted when their currents are in the
same direction, and repelled when they are in
opposite directions. It should be noted that this
statement does not match that expressing the
behavior of electric charges, for charges of like sign
Fm. 16-1. Like-directed repel and those of unlike sign attract.
currents attract; opposing In order to measure the force between two con-
currents repel. ductors, the arrangement should conceivably be a
pair of parallel wires, one very long and the other
movable, so that the force on a definite length of the latter could be measured.
Experiment shows that the force is directly proportional to the product of
the currents I 1 and I 2 in the wires, and inversely proportional to the distance
r between them. The force on a wire of length l can be expressed as
F a: I1I2l/r, or in equation form as
F = 2 K IJ2l, (16.1)
r
where 2 K is chosen as the proportionality factor. This statement for the
force acting between current-carrying wires is called Ampere's Law for paral-
lel conductors; it is named after the French scientist Andre M. Ampere
(1775-1836), who investigated the relation between forces on wires in dif-
ferent configurations.
The practical unit of current is the ampere (amp), w4ich was defined by
the International Committee on -Weights and Measures in 1935, on the basis
of Ampere's Law, as follows: The ampere is a current which, when established
in each of two infinitely long parallel conductors one meter apart in a vacu1tm,
Catlses either conductor to experience a force of 2 X 10-7 newtons per meter of
length. On the basis of this definition, the foregoing expression gives the
force Fin newtons when the currents 11 and I2 are in amperes, the conductor
length l and separation r in meters, and the constant K has the value
K = 10_7 newton.
amp 2
§ 16-3 PERMITTIVITY AND PERMEABILITY OF FREE SPACE 281

From the positions of l and r in the expression, one in the numerator and
the other in the denominator, it is clear that other length units could be used
instead of the meter. Thus the force between parallel wires each carrying
1 amp is also 2 X 10-7 newtons per centimeter of length when the wires are
1 cm apart in free space, and so on.
As a numerical example, consider two parallel conductors carrying currents of
;200 amp each and located 0.2 meter apart in air. Find the force acting on each
conductor per meter of length.
The force is given by Eq. 16.1 as

F = 2 X 10-1 1112 = 2 X 200 X 200 X 1 X 10-1 = 0.04 newton.

r 0.2
The same force would act between 1-cm lengths of two adjacent ,vires 0.2 cm apart
center to center, and carrying the same currents.
Instruments for measuring current can be designed on the basis of the
foregoing relation; fundamentally they take the form of "current balances"
in which two coils of wire are mounted horizontally and coaxially, and the
vertical force between them is measured. In the more usual form, called the
electrodynamometer, the coils are mounted with their planes vertical and at
right angles to each other, and the torque between them is measured. The
commercial instruments for measuring current, called ammeters, are de-
scribed in § 18-9.
16-3. Permittivity and Permeability of Free Space. In § 15-5 the per-
mittivity of free space is given as
1 coulomb 2
Eo = 4 1r 8.988 X 10 newton ·m 2 '
9

and the basic equation of electrostatics, F = ?

le Q1 2, is rewritten in terms of
r-
Eo and becomes
(15.2) F = _1_, Q1Q2.
4 1r1: 0 r 2
This change is made in order to simplify subsequent equations that are more
widely used than the basic equation itself.
The same plan is followed with the basic equation of electrodynamics,
Eq. 16.1, by replacing the proportionality factor K by a fraction that includes
another parameter of free space. Thus,

K= µo'
411"
where µo represents the property of free space known as permeability. The
equation for the force on parallel wires then becomes
F = J!jJ__ 2 Iil2l. (16.2)
4 1r r
282 ELECTRICITY & MAGNETISM Chap. 16

Since two 1-ampere currents in parallel wires 1 meter apart are acted upon in
free space by a force of 2 X 10-1 newtons per meter of conductor length, the
value of the permeability of free space can be evaluated herefrom as
,--- 41rFr _ 4 1r X 2 X 10-7 X 1 _ 4 10_7 newton
µo - 2 Iil2Z - 2 X 1 X 1X 1 - 1r amp 2 •
The permeability of air and of other nonmagnetic substances is practically
the same as for vacuum, § 19-2.
It is shown in more advanced books dealing with electromagnetic theory
that the two parameters Eo and µ 0 for free space are related to the speed of
light c by the equation
1
µoEo=-·
c2
The speed of light is 2.99790 X 108 m/sec for vacuum, § 27-4.
16-4. Relation Between the Ampere and the Coulomb. The current in an
electric circuit is essentially a fl.ow of electrons through it-the greater the
number of electrons passing any point of a circuit in a given time, the greater
is the current. Thus current is the time rate of fl.ow of electric charge, or
I = dq/dt, and conversely
q = fl dt, (16.3)
or, if the current is steady,
q = It. (16.4)
In these expressions, q is the charge or quantity of electricity passing a point
of the circuit in a time of t seconds, and I is the current in amperes. The
coulomb is defined by Eq. 16.4 as follows: the coulomb is the quantity of elec-
tricity transported past a point of an electric circuit in one second by a current of
one ampere.
Since the electron has a charge of 1.602 X 10-19 coulomb, § 15-13, it fol-
lows that the coulomb is equivalent to 1/ (1.602 X 10-19 ) = 6.24 X 1018 elec-
trons.
The relation between the ampere and the coulomb may be made clearer
by analogy with hydraulics. A pump circulates water in a pipe line, just as
a battery or a generator circulates charges in an electrical circuit. The
quantity of water is expressed in gallons, corresponding to electric charge in
coulombs. The rate of flow of water in the piping is expressed in gallons per
second, corresponding to the rate of flow of electric charge in the circuit in
coulombs per second, or current in amperes.
While the system of electromagnetic units (emu) is not used in this text-
book, it is worth while to point out that this system is based on Eq. 16.1 with
a numerical value of unity for the constant K. When the force Fis expressed
in dynes, and the conductor length l and separation r are in centimeters, the
currents are expressed in a unit called the abampere. One abampen1 = 10
amperes; similarly, one abeoulomb = 10 coulombs.
§ 16-6 ELECTROLYSIS 283

16-5. Electromotive Force and Potential Drop. In the study of electro-

statics it is found that work is done when an electric charge moves between
two points which are at different potentials. The work done is equal to the
product of the charge and the potential difference, in accordance with Eq.
15.5. This statement applies also to the circulation of charge around an
electric circuit. The charge does work in traversing the circuit, as evidenced
by the evolution of heat and the operation of electromechanical devices; the
charge has an equal amount of work done upon it as it moves through the
battery or other source of electrical energy. The battery or other source is
said_to have an electromotive force (abbreviated emf), which is measured by
the work done per unit of charge in passing through the source.
· A source within which a charge Q has an amount of work W done on it has
an electromotive force given by
w (16.5)
E = cj"
Also, the work done by a unit charge in passing between two points of a
circuit is equal to the potential drop between those points. If Wis now taken
to represent the work done by the charge Q in moving between two such
points, the potential drop between the points is
w
(15.5) v = -zr
The term potential difference applies to both emf and potential drop; the
practical unit is the volt, § 15-8. The potential difference between two points is
one volt if a charge of one coulomb either acquires or expends one joule of energy
in moving f ram one point to the other.
To give a practical idea of the volt, a few illustrations will be cited. The
emf of the familiar dry cell is 1.5 volts. Practically all electric lamps are
supplied with 110 to 120 volts. Many direct-current railways are operated
on 500 to 600 volts. For the long-distance transmission of electrical energy,
values over 100,000 volts are often employed.
Since the energy which the charge receives from the source must be the
same as that which it expends throughout the circuit, it follows that the emf
of the source must equal the sum of the potential drops around the circuit.
The difference of potential between two points in an electric circuit is
measured by an instrument called a voltmeter, by merely connecting the instru-
ment across the points under consideration. It operates on the same princi-
ple as the ammeter but is designed to take very little current and is calibrated
to read in volts.
16-6. Electrolysis. Liquids can be introduced into electric circuits by
putting them in containers equipped with conducting plates or electrodes for
current entry and exit. These are called electrolytic cells. Such cells, often
of large size, are used for the manufacture of certain gases, for the deposition
284 ELECTRICITY & MAGNETISM Chap. 16

and purification of metals, and for other electrochemical processes. The

solutions used in electrolytic cells are called electrolytes, and the process of
current transmission through them is termed electrolysis.
Other forms of cells, known as voltaic cells, are used to supply electrical
energy. They are called primary cells if the electrodes are consumed in the
process, as in the standard dry cell, and are called secondary cells if the elec-
trode materials can be restored by charging from outside sources of electrical
energy, as in the automobile storage battery.
The conduction of electricity through electrolytes is due to the motion of
ions, some carrying positive and some negative charges, as explained in
§ 15-10. The two electrodes of an electrolytic cell are called the anode and
the cathode; the current enters at the anode and leaves at the cathode. Except
for chemical actions that may occur, the behavior of any electrolytic cell is
based entirely upon the action of the ions in the electric field between the
anode and cathode. Some examples of electrolysis follow.
Electrolysis of Hydrochloric Acid. A solution of hydrochloric acid (HCl)
in water (H 20) presents evidence of containing H ions, each atom having lost
an electron and therefore being positively charged, and Cl ions, each atom
having gained an electron and therefore being negatively charged. In other
words, the acid dissociates into positive hydrogen ions and negative chlorine
ions, or symbolically,
HCl ~ H+ + Cl-.
The ions of a substance do not have the properties of the corresponding atoms.
For example, a solution of H+ ions affects indicators such as litmus; a solu-
tion of hydrogen gas does not. A change of
properties would, indeed, be expected from
the modifications in atomic composition.
Evidence indicates that bare H+ ions do not
Anode H+ions Cathode actually exist in aqueous solutions; instead it
is believed that the carriers occur in a hy-
drated form as hydronium ions, H 30+, each
being a combination of H+ and H20. To avoid
complexity, however, the hydrogen ion will
be represented by the simpler notation H+.
In Fig. 2 two platinum plates A and C are
Frn. 16-2. Conduction in an
shown connected to a battery and dipped into
electrolytic cell. a dilute solution of hydrochloric acid. The
H+ ions are drawn toward the negative plate
C and the Cl- ions toward the positive plate A. The result is a drifting or
migration of ions in both directions through the liquid (this motion being
in addition to the unordered motion of the particles at the existing tem-
perature, § 9-2).
§ 16-6 ELECTROLYSIS 285

Each H+ ion, upon reaching the cathode C, acquires an electron e there and
forms a hydrogen atom:
H+ + e-H.
From these atoms hydrogen molecules are produced, and pass off as gas, H 2•
Each Cl- ion, upon reaching the anode A, gives up its electron and becomes a
chlorine a tom:
Cl- - e - Cl,
the atoms combining to form molecules of chlorine, Cl2, which are liberated
at this plate.
Decomposition of Water. In demonstrating the electrolytic decomposition of
water it is customary to use an electrolyte of dilute sulfuric acid (H2804) and
electrodes of platinum or carbon which are chemically inert. The electrolyte
dissociates into H+ and 80r ions; the H+ ions are drawn toward the negative
plat1:; and the 804= ions toward the positive plate. The result is again a
drifting of ions .in both directions through the liquid.
Each H + ion, upon reaching the cathode, acquires an electron there and
forms a hydrogen atom. These atoms combine into molecules and escape
as gas, that is:
2 H+ +2e- H2 j .
When the 804= ions reach the anode, electrons are given to the electrode and
oxygen is liberated as a gas, while more sulfuric acid is formed. The chemical
equation is
2 804= - 4 e + 2 H20 - 4 H + + 2 804= + 02 j .
2 H2804

The net result is, thus, the decomposition of water into hydrogen and oxygen.
Electroplating. Consideration will next be given to an electrolytic cell in
which the electrolyte is a dilute copper sulfate solution (Cu804) and the
electrodes are of platinum. The following equations represent the reactions
which are believed to occur at the electrodes:
At cathode Cu++ +2e- Cu l.
At anode 80r - 2 e + H20 - 2 H+ + SOr + 0.
'-------y------
H2S04
These show that copper is plated upon the cathode and that the electrolyte
gradually changes its composition.
When copper is substituted for platinum as the material of the anode, the
anode reaction becomes:
286 ELECTRICITY & MAGNETISM Chap. 16

while the cathode reaction remains unchanged. In this process the electrolyte
retains the same concentration, the anode is dissolved, and copper is plated
on the cathode.
Many metals are electroplated under the general conditions just described,
the anode being made of the metal to be electroplated and the electrolyte
being a solution of one of its salts. For example, silver is deposited from
silver nitrate and the anode is of silver. In contrast, chromium is generally
plated from a chromic acid bath and lead is used as anode material; the
chromium content of the electrolyte is replenished when necessary. Zinc
and copper can be liberated simultaneously from a cyanide solution, in which
these elements are present, and produce a plating of brass.
Rubber is deposited electrolytically from rubber latex-the milky juice of
rubber trees. The latex particles are in suspension as a colloid; they have
negative charges and become affixed to the anode of the cell.

Trolley wire

Buried pipe line

FIG. 16-3. Electrolytic corrosion of a buried pipe.

Electrolytic Corrosion. Electrolysis sometimes produces unexpected and

destructive effects. An example illustrating electrolytic corrosion of an
underground pipe from the leakage currents of an electric railway line is
indieated in Fig. 3. This diagram represents a generator at a power station
supplying current through a trolley wire or third rail to the motors on an
electric train. The current is returned to the generator partly by way of
the rails and partly along the grounded pipe line. The pipe is likely to
become corroded at the region marked X, for here the moist earth around the
pipe serves as the electrolyte, and the pipe itself acts as the anode of an elec-
trolytic cell. Long-continued current through this "cell" will wear away
the anode, as previously described. To eliminate this corrosion, the point
X and the negative terminal of the generator should be bonded together by a
wire, thereby providing a metallic path for the current leaving the pipe.
16-7. Faraday's Laws of Electrolysis. It will be observed that, whenever
hydrogen or any metal is liberated in an electrolytic cell, this action, with rare
exceptions, occurs at the cathode, and the liberation of other substances takes
place at the anode. It will also be observed that, if electrolysis is to continue,
the external circuit must continuously supply electrons to the cathode and
remove them from the anode; in other words, there must be a current in the
§ 16-7 FARADAT'S LAWS OF ELHCTROLfSIS 287

circuit containing the cell. Moreover, one electron is required in forming a

hydrogen or other univalent atom at the cathode, two electrons are required
for a bivalent atom, and so on. Similarly, at the anode one electron is re-
leased in forming a univalent atom, and two electrons in forming a bivalent
atom. Eince a flow of 6.24 X 1018 electrons is equivalent to 1 coulomb,
§ 16-4, it follows that in an electrolytic cell 1 coulomb of electricity is capable
of liberating at the cathode 6.24 X 1018 hydrogen or other univalent atoms,
or half this number of bivalent atoms, and so on. A corresponding libera-
tion of atoms will also occur at the anode of the cell.
Quantitative measurements in electrolytic cells are based on two laws of
Faraday which state:
1. The mass of a substance liberated in an electrolytic cell is proportional to
the quantity of electricity passing through the cell.
2. When the same quantity of electricity is passed through different electrolytic
cells, the masses of the substances liberated are proportional to their chemical
equivalents. The chemical equivalent of an element is the ratio of its atomic
weight w to its valence number z; with radicals the chemical equivalent is
the sum of the atomic weights of the component elements divided by the
valence number of the radical.
The atomic weights and usual valence numbers of several elements are
given in Table I.

TABLE I. ATOMIC "WEIGHTS AND VALEKCE NUMBERS

Element Atomic "\Yeight Valence Number

Aluminum ........ . 26.97 +3

Chlorine ......... . 35.46 -1
Copper .......... . 63.54 +2
Hydrogen ........ . l.008 +1
Iron ............. . 55.85 +3
Lead ............ . 207.21 +2
Oxygen .......... . 16.000 -2
Silver ............ . 107.88 +1
Zinc ............. . 65.38 +2

In the application of Faraday's Laws of Electrolysis, it is desirable to

find what quantity of electricity will liberate 1 chemical equivalent of a sub-
stance. Silver, for example, has a chemical equivalent of 107.88 and is
univalent. Because of the latter fact 1 coulomb of electricity will deposit
6.24 X 1018 atoms of silver. But Avogadro's Number, 6.024 X 1023 , is the
number of atoms in a gram-atom of a substance, § 9-9; hence there are
6.024 X 1023 atoms in 107.88 gm of silver. To deposit this amount of silver
therefore requires 6.024 X 1023 /(6.24 X 1018 ) = 9.65 X 104 coulombs. This
quantity of electricity liberates 1 chemical equivalent of any substance.
288 ELECTRICI'IT & MAGNETISM Chap. 16

The quantity of electricity needed to liberate one chemical equivalent of

a substance in an electrolytic cell is called a faraday. For an element of
atomic weight w and valence number z, one faraday, or 96,500 coulombs, will
liberate w/z grams of the element from a solution of one of its salts. It fol-
lows from Faraday's Laws of Electrolysis that some other quantity, q cou-
lombs, will liberate from the same solution a different mass, m grams, such
that
m _ w/z .
q 96,500
This expression may be written in the form
It w (16.6)
m = 96,500 --; '
where the quantity of electricity q coulombs is replaced by the product of the
current I amp and the time t sec.
Determine the amount of copper which would be plated from a solution of a cupric
salt by a steady current of 10 amp for 1 hr.
For copper, the atomic weight is 63.54, and its valence in the cupric salt is 2; con-
sequently, in 1 hr (3600 sec) the mass of copper plated on the cathode is
m = 10 X 3600 63.54 = 11 86 m
96 500 X 2 . g .
'
16-8. Single-electrode Potentials. When a metal plate is dipped into a
solution of one of its salts, a contact is established between the metallic atoms
of the plate and the positive metallic ions of the solution. Two tendencies
exist, either of which may determine the action that results: First, metallic
ions tend to acquire electrons from the plate and deposit themselves on it as
atoms, and second, the atoms of the plate tend to lose electrons and go into
solution as ions. The tendency for metal to deposit on the plate is propor-
tional to the concentration of the ions, and, theoretically, a sufficiently high
concentration could always cause deposition. On the other hand, if the ionic
concentration is low, the tendency of the plate to dissolve will predominate.
At some intermediate concentration value, neither deposition nor dissolution
will occur; it is supposed under this condition that the tendency toward
deposition is exactly neutralized by a so-called solution pressure. Although
the real significance of this term is in doubt, it is usual to think of the solution
pressure of a metal as an agency that tends to throw the metal into solution
in the form of positive ions.
Solution pressure has a characteristic value for each metal. Copper has
a low solution pressure, and, if a copper plate is dipped into a copper sulfate
solution of ordinary concentration, a few Cu++ ions take electrons from the
plate, forming atoms of copper which deposit upon the plate. The action
stops almost immediately, however, because the plate becomes positive and
repels the advance of other positive ions. Again, zinc has a large solution
§ 16-8 SINGLE-ELECTRODE POTENTIALS 289

pressure, and if a zinc plate is dipped into a zinc sulfate solution, a few atoms
of the plate dissolve, leaving their electrons behind, and go into solution as
Zn++ ions. This effect is also of very slight magnitude, because the plate
becomes negative and the throwing off of further positive ions is prevented
by electrostatic attraction between plate and ions. The mass of metal de-
posited or dissolved in these instances is minute. The processes could be
made continuous, however, by removing the charges from the plates as fast
as they are formed.
The foregoing analysis shows that a copper plate placed in a solution of a
copper salt assumes a higher potential than the solution, and that a zinc
plate placed in a solution of a zinc salt assumes a lower potential than the
solution; such potential differences are called single-electrode potentials. When
the two solutions are put in contact, the combination forms a voltaic cell with
the copper positive and the zinc negative. If the small potential difference
at the junction of the solution is neglected, the emf of the cell will be the sum
of the single-electrode potentials of the two metals.

-To potentiometer-

Capillary tube
I
Electrode

x
Solution (
Hydrogen electrode

FIG. 16-4. Hydrogen electrode used in measuring normal electrode potential.

The potential of a given metal with respect to a solution of one of its salts
under specified condifaons can be expressed precisely. Its value ~s called the
normal electrode potential if the solution is "normal"; that is, if 1 liter at
25°C contains an effective concentration of 1 chemical equivalent of the
metallic ions. The normal electrode potential of copper is +0.3448 volt and
that of zinc is -0.7581 volt. Potential values for a number of substances
are given in Table IL
Figure 4 illustrates the apparatus for making measurements of normal
electrode potential. The metal X under test is placed in a normal solution
of one of its salts to form a half cell, a hydrogen electrode is immersed in a solu-
tion of normal hydrogen-ion content to form another, and the two are joined
by a capillary. The hydrogen electrode is commonly used as a reference
electrode; its potential with respect to its own solution is constant and is
arbitrarily taken as zero. Hydrogen, which is nonconducting as a gas, 1s
290 ELECTRICITY & MAGNETISJlf Chap. 16

passed over a plate of platinum covered with platinum black. Here the gas
is adsorbed, that is, condensed and held upon the surface; in this condition,
hydrogen is conducting like a metal electrode. The normal electrode poten-
tial of the metal is found by measuring the potential difference between the
electrodes, ordinarily by means of a potentiometer, § 17-14.
16-9. Electromotive Series of the Metals. The relative chemical activities
of the metals can be indicated by arranging them in tabular form with the
most active one at the top of the series. The ion of each metal listed is dis-
charged under ordinary conditions by any metal element above it in the list
from chemically equivalent solutions of their simple salts; this result is
attributed to differences in solution pressure.
Table II indicates some of the important metals and gives their normal
electrode potentials. Each of these values represents the emf of a cell having
a hydrogen electrode and an electrode of the metal referred to. The emf of
a cell having any two metals for its electrodes in a uniform electrolyte can
be predicted from the individual potential values. In a copper-zinc cell, for
example, copper is +0.34 volt on the potential scale and zinc is -0.76 volt,
each with respect to hydrogen; consequently, the potential of copper is
0.34 + 0.76 or 1.10 volts higher than that of zinc. Thus, the copper-zinc
cell has an emf of this value, and its positive terminal is the one attached to
the copper electrode. The emf of a voltaic cell depends upon the materials
used, and not upon size.

TABLE II. ELECTROMOTIVE SERIES OF THE :METALS

Normal Electrode Potential, Volts

Element Symbol (Referred to Hydrogen Electrode)

Potassium ......... K -2.92 high solution pressure

Sodium ........... Xa -2.71 '
Magnesium ........ Mg -1.55
Zinc .............. Zn -0.76
Iron .............. Fe -0.44
Tin ............... Sn -0.13
Pb -0.12
~t)ROGEK:::: :I H 0.00
Bismuth .......... Bi +0.20
Copper ........... Cu +0.34
Mercury .......... Hg +0.80 ..
,
Silver ............. Ag +0.80 low solution pressure

16-10. The Voltaic Cell. An elementary type of voltaic cell can be con-
structed by placing electrodes of copper and zinc in a dilute solution of sul-
furic acid (H 2S04), as indicated in Fig. 5. Such a cell will set up a current
from copper to zinc in the external circuit R. At the copper plate, because
of the low solution pressure of eopper, the H+ ions of the electrolyte deposit
§ 16-10 THE VOLTAIC CELL 291

themselves upon the copper plate, thereby giving it a positive charge. At

the zinc plate, since the solution pressure of zinc is high, zinc atoms dissolve,
throwing Zn++ ions into solution and leaving electrons on the plate, a process
which makes the zinc plate negative. The electrons on the zinc plate repel
one another along the wire, and are also attracted by the positively charged
copper plate; as a result a current is established through the external circuit.
Any impurities in the negative electrode behave like tiny short-circuited
cells and cause it to be consumed wastefully. This so-called local action can
be reduced by coating the zinc with mercury; the amalgam spreads over the
surface and keeps.the impurities from coming into contact with the electrolyte.
The current supplied by the simple voltaic cell described is quickly reduced
in value during even short periods of use because of the accumulation of
hydrogen on the positive plate. This action not only changes the concen-
tration of the electrolyte, but alters the material of the electrode and sets up
an opposing emf which must be overcome before the cell can supply current
to the external circuit. This effect, called polarization, is
present in many types of cells. In voltaic cells subject R
to polarization, some substance is usually added to combine
chemically with the hydrogen gas as it forms; this substance
is called a depolarizer.
In an electrolytic cell having like electrodes, it frequently
happens that the electrodes become coated with dissimilar
materials. Such a cell then acts like a voltaic cell and
opposes the applied emf; hence its own emf must be over-
come before current can be set up in it. For example, a cell Fm. 16 _5 _ Dia-
with platinum electrodes in a solution of hydrochloric acid gram of a simple
may cease to conduct current, because hydrogen and chlorine voltaic cell.
accumulate on the electrodes and set up an emf of polariza-
tion opposing the outside source. However, so long as both electrodes of
an .electrolytic cell remain of the same substance and the electrolyte is
uniform throughout, whatever electrode potential is developed at one of them
will be neutralized by an equal and opposite potential at the other. The
smallest potential difference applied to the cell will then set up a current in it.
16-11. Primary Cells. The familiar dry cell is the only primary cell in
general use today. It utilizes carbon as the positive and zinc as the negative
electrode; the solution is ammonium chloride (NH4Cl), and the depolarizer
is manganese dioxide (Mn0 2). The construction of the cell is shown in Fig.
6. The zinc container is lined with blotting paper, and the space between it
and the carbon electrode is filled with a mixture of ground carbon and man-
ganese dioxide. The entire cell is moistened with ammonium chloride and
a little zinc chloride. The cell is represented diagrammatically in Fig. 7.
When the cell delivers current, some zinc dissolves, leaving electrons behind,
and these repel one another along the external circuit to the carbon plate.
The Cl- ions in the electrolyte migrate toward the Zn++ ions near the zinc elec-
292 ELECTRICITY & MAGNE'l'ISJI Chap. 16

trade, and the NH4+ions migrate toward the carbon electrode. The formation
of hydrogen gas is prevented to a considerable extent by the reduction of the
manganese dioxide. The initial cell reaction is probably of the form:

The complete reactions within the cell are obscure and apparently are influ-
enced by the conditions under which the cell is used.

Ground carbon·
and Mn0 2 •
Zinc
Pa per carton

FIG. 16-6. Sectional view of the FIG. 16-7. Dia-

dry cell. gram of the dry
cell.

This type of primary cell has an emf of about 1.5 volts and is best adapted
to "open-circuit" work, that is, service in which the cell is called upon to
deliver current only for short periods, its circuit at other times being open.
The depolarizing action in the ammonium chloride cell is imperfect and the
emf falls off with continued use; however, the cell recuperates during periods
of open circuit, because the layer of gas which may have been formed has an
opportunity to disappear. In a typical test, a dry cell delivered current to a
3-ohm load continuously for 1 hr, and during this period its emf was observed
to fall from 1.53 to 1.43 volts, but, upon standing on open circuit for 30 min,
the emf increased to 1.49 volts.
16-12. Storage Cells. Storage batteries are used extensively for purposes
requiring steady and large currents. Typical applications include engine
starting and auxiliary lighting of automobiles, the propulsion of submarines
and trucks, and stand-by service in power substations and telephone ex-
changes to supply power for short periods in the event of temporary power
failures.
When a storage battery is delivering electrical energy, it is said to be dis-
charging. After a period of discharge the battery can be restored to its
original condition by supplying energy to it from an outside source; this
action is called charging. The energy supplied to a battery during the charg-
ing process is not all available on discharge, about 30 to 40 per cent being
 § 16-12 STORAGE CELLS 293

wasted as heat. The rating of a battery is expressed by the number of

ampere· hours it can supply over a stated period.
Lead-acid Cells. The lead storage cell, when charged, consists essentially
of a positive plate of lead dioxide and a negative plate of spongy lead, both
immersed in a water solution of sulfuric acid as an electrolyte. As the cell
discharges, Fig. 8, both plates become coated with lead sulfate, and the sul-
. furic acid is partly converted into water, thereby lowering the specific gravity
of the electrolyte.
During discharge of the lead storage cell, the emf falls quickly from an
initial value of approximately 2.1 volts and remains nearly constant at 2.0
volts throughout most of the discharge period. At the approach of complete

Positive Negative Positive Negative

plate plate plate plate
+ Pb0 2 Pb /
+
+
+
-
-+
H
so:
H2 S04

FIG. 16-8. A lead storage cell on FIG. 16-9. A lead storage cell
discharge. being charged from an outside
source.

discharge the emf falls rapidly from this value to 1.75 or lower. Even when
the cell is standing idle there is a gradual conversion of lead into lead sulfate;
this effect is called self-discharge.
To charge a storage battery, the battery is connected to a source of some-
what higher emf in such a way that a current is established through it in a
direction opposite to the current supplied by the battery on discharge.
This means that the positive terminal of the battery should be joined to the
positive terminal of the charging source. During the charging process, Fig.
9, the lead sulfate on the positive plate is reconverted to lead dioxide and
that on the negative plate to lead; also the electrolyte returns to its original
density. · A little hydrogen and oxygen are liberated, as in the decomposition
of water in an electrolytic cell; this effect, known as "gassing," becomes
prominent near the completion of charge.
The chemical reactions at both electrodes of the lead battery, and for both
the discharge and charge conditions, can be merged into a single equation to
provide a brief statement of the reversible reaction; it is

The state of charge of the cell can be tested by measuring the specific grav-
ity of the electrolyte with a hydrometer, § 8-8. For an automobile battery,
294 ELECTRICITY & MAGNETISM Chap. 16

the specific gravity of the electrolyte is 1.285 when the battery is fully charged,
and when discharged it may be as low as 1.150.
Recent investigations show that the antimony, which is normally alloyed
with the plates of lead storage cells to harden them, is also subject to elec-
trolytic action and hence speeds up self-discharge and corrosion. Replace-
ment of antimony by a much smaller quantity of calcium in batteries for
telephone service has resulted in lowering the loss in electrical charge to
about 4 per cent a month and in increasing the life of the batteries by 50
per cent.
Nickel-alkaline Cells. The nickel-iron storage cell, devised by the Ameri-
can inventor Thomas A. Edison (1847-1931), employs nickel oxide for the
positive electrode and finely divided iron for the negative electrode. These
materials are packed respectively into finely perforated cylindrical tubes and
flat pockets carried by steel grids. The electrolyte is a solution of potassium
hydroxide. As the cell discharges, the nickel oxide becomes reduced and
the iron becomes oxidized, but the electrolyte remains unchanged. The
chemical reactions are reversible and may be represented by the following
composite equation:
Fe + Ni20s + 3 H20 ~ Fe(OH)2 + 2 Ni(OH)2.
The Edison cell was designed particularly for operating vehicles, and for this
reason was made small in volume, light in ·weight, and arranged for regular
cycle service with respect to charge and discharge.
The other alkaline storage cell, designated as the nickel-cadmium type, had
its origin in Europe but is now manufactured in the United States. It em-
ploys nickel hydroxide for the positive electrode and cadmium oxide for the
negative electrode, both materials being packed into identical flat pockets
and carried in steel frames. The electrolyte is a solution of potassium hydrox-
ide, which remains unchanged in concentration during use. The chemical
actions are reversible and can be represented by the following composite
equation:
Cd+ 2 Ni(OH)a·2} H20 ~ CdO + 2 Ni(OH)2·3 Hl).
This cell is larger and heavier than the Edison cell, but has a lower self-
discharge rate; also, it does not need to be operated on regular cycles and
does not gas until shortly after the charging is completed.
The emf of either alkaline cell is 1.2 volts and drops somewhat during dis-
charge; in charging, the impressed potential difference should range from
1.40 to 1.50 volts. Both types are lesiii subject to mechanical derangement
than the lead cell, are not injured by freezing, and can be left in a discharged
condition without damage.
16-13. Standard Cells. Primary cells of special form are used as standards
of emf for measuring potential differences accurately. These measurements
 § H\-13 STANDARD CELLS 295

are made with a potentiometer, § 17-14, under conditions such that the stand-
ard cell delivers only very small currents, and for brief periods of time.
The construction of the standard cell invented by Edward Weston is
shown in Fig. 10. The electrodes are placed in the opposite sides of an
H-shaped glass vessel which holds the electrolyte. The positive electrode is
a paste of mercurous sulfate, and the negative is composed of cadmium amal-
gam. The electrolyte is a solution of cadmium sulfate, which may be kept
saturated by having crystals of cadmium sulfate present.

-Seal
Wax
Cork rcadmium
· sulfate
Porcelain
l solution
Asbestos Mercurous
Cadmium sulfate
amalgam Mercury

Platinum
terminals
Frn. 16-10. Construction of the Weston standard cell.

The emf of the saturated cell is 1.01830 volts at 20°C, and varies slightly
but definitely with temperature changes. These cause slight variations in
the amount of cadmium sulfate which dissolves in the solution, thereby
altering to some extent the concentration and the emf of the cell. In the
unsaturated cell, temperature changes do not affect the concentration of the
electrolyte, and their effect on the emf is so small that it is usually neglected.
With unsaturated cells the emf must be determined for each, usually by com-
parison with an emf of known value.

PROBLEMS
1. The location of two neighboring circuits is such that a conductor from one is
parallel over a length of 8 m to a conductor from the other, and the two conductors
are 5 cm apart in air. The current in one circuit is 250 amp and that in the other is
350 amp. What force is exerted by the currents in the parallel conductors?
2. A twin-conductor lamp cord carries a current of 8.5 amp to and from a load.
The conductors are 0.1 in. apart and the insulation between them has the same
permeability as free space. vVhat force acts on each conductor per foot of length as
a result of the current?
3. What current in each of two parallel conductors 1 in. apart in air would cause
each to be acted upon by a force of l lb per yard of conductor length?
4. Write out the values for the permeability and permittivity of free space together
with the mks units for these quantities, and show that the reciprocal of their product
is the same numerically as the square of the speed of light and that the units are
consistent.
296 ELECTRICITY & MAGNETISJ.11 Chap. Hi

5. A particular pulse of current in a circuit lasts for 1 microsec and has an average
value of 1 microamp. How many electrons are transferred in this pulse?
6. In the manufacture of caustic soda (NaOH) and chlorine gas (Cb) an aqueous
solution of sodium chloride (NaCl) is electrolyzed with mercury as cathode. Write
the overall equation of the electrolysis without regard to the intermediate part served
by the mercury electrode. State what is produced at each electrode.
7. How many coulombs are required to deposit (a) 1 gm of copper, and (b) 1 gm of
chlorine?
8. A current of 50 amp through a solution of a cadmium salt is found to deposit
314.5 gm of the metal in 3 hr. Determine the chemical equivalent of cadmium.
9. How much silver will be deposited from a silver nitrate solution in an electrolytic
cell by a current of 5 amp maintained for 1 hr and 25 min? How many electrons are
transferred per second?
10. What quantity of electricity must be passed through an electrolytic cell in order
to decompose 1 kg of water?
11. A metal article having a surface of 250 cm2 is coated with silver in a plating
bath containing silver nitrate solution, through which a current of 0.3 amp is main-
tained for 15 min. Compute the thickness of the plating. The specific gravity of
silver is 10.45.
12. In copper refining, pure copper is deposited from a copper sulfate solution in an
electrolytic cell using impure copper as the anode. If the potential difference across
the cell is 0.3 volt, how much energy is used in depositing 1 kg of copper?
13. An ammeter was calibrated by connecting it in series with an electrolytic cell
containing copper sulfate solution. During a 15-min test the weight of the cathode
increased by 2.900 gm, and the instrument reading was constant at 10.00 amp. What
correction should be applied to the ammeter at this scale reading?
14. Calculate the number of faradays of electricity required to evolve 1000 liters of
chlorine at standard temperature and pressure.
15. A volume of 10 cm3 of oxygen was produced in 1 min by electrolysis in a test
conducted at a temperature of 25°C and a pressure of 78 cm Hg. How much current
was used in the test'?
16. A current of 55 amp leaves an underground iron pipe and enters the soil. To
estimate the damage to the pipe through electrolytic corrosion, find in what time I lb
of iron would be removed.
17. Determine the least amount of zinc which should be put into a dry cell in order
that it may supply an average current of 0.2 amp for 1000 hr.
18. A certain lead storage cell has 5 positive and 6 negative plates, each measuring
Ji by 7i in. Its rated capacity is 200 amp· hr for an 8-hr discharge and 156 amp· hr
for one of 3 hr. What are the current densities for the positive plate at these discharge
rates?
19. For each ampere· hour delivered by a lead storage battery, how much lead
dioxide (Pb02) of the positive plates is changed to lead sulfate (PbS04)?
 17
RESISTANCE OF CIRCUITS
The second of the three important effects of electric currents-that of heat
production-is considered in this chapter. Underlying this effect is a prop-
erty of electric circuits known as resistance; the larger the resistance of a
conductor the greater is the amount of heat produced in it by a given cur-
rent. The calculation and measurement of resistance is taken up first, and
this is f9llowed by determinations of electrical and heat energies, methods of
connecting resistors, instruments for measuring resistance and potential, and
the analysis of circuits and networks.
17-1. Electrical Resistance. The flow of electricity through a conductor
encounters an opposition analogous to friction, which results in the conver-
sion of some electrical energy into heat. The property of a conductor that
causes heat to be evolved when electrons flow through it is called resistance.
It may be looked upon as the opposition presented by the conductor to the
electrons in weaving their way among its atoms.
Resistance is a general property of all materials. Good conductors have
little resistance and poor ones have much resistance; insulators have very
high resistance. Both conductors and insulators are needed in all electric
circuits, machines, and appliances; the conductor provides the path for the
current, and the insulator prevents its leakage from that path. Copper and
a number of other metals and alloys are used as electrical conductors, while
such materials as porcelain, rubber, and treated fabrics are used as insulators.
17-2. Ohm's Law. Simple experiments with metallic circuits show that
the amount of current established is directly proportional to the potential
difference applied. Thus, if the emf in the circuit is E, and the current pro-
duced by it is J, then I ex: E. This proportionality is best expressed as an
equation, and it can be put in such form by introducing a quantity R so as to
yield
E (17.1)
I=-,
R
where R represents the total resistance of the circuit. If either the emf or
the resistance of a circuit is changed, the current will automatically adjust
itself so that the equation wm remain true. The student is cautioned that
the foregoing expression does not apply to nonmetallic circuits, notably those
which contain gases, and also that E is not necessarily the emf developed by
297
298 ELEC1'RICI'PJ' & MAGNETISM Chap. 17

batteries, for sometimes other emfs are active in circuits. Further, the
total resistance of a circuit must also include any resistance that may be
present within the battery or other source which supplies current to the cir-
cuit; this so-called internal resistance is usually small, but it may not be
negligible.
Equation 17.1 can be modified so as to apply to only part of a circuit. If
in a metallic circuit a potential drop V exists across a part that has a resistance
R, then the current in that part of the circuit is

I=-.
v (17.2)
R
This relation was first given by the German physicist Georg S. Ohm (1787-
1854) and is known as Ohm's Law. It finds wide application in electrical
work of all kinds. The equation can be rewritten in the form V = IR, which
states that the potential drop across any part of a circuit is equal to the
product of the current in that part and the resistance of the same part.
Ohm's Law can also be written as R = V/I; in this form it serves to define
the unit in which resistance is expressed. For if the potential drop across a
conductor is V = 1 volt, and the current in it is I = 1 amp, then the resist-
ance of the conductor must have unit value; this unit is called the ohm. The
resistance of a conductor is one ohm if there is a potential drop of one volt across
it when the current in it is one ampere. The units to be used in Ohm's Law,
whether applied to a complete circuit or to only part of a circuit, are the same;
thus
volts
amperes= -h--.
oms
A few examples will serve to give an approximate idea of the magnitude of
the ohm as the unit of resistance. Each of the following conductors has a
resistance of about 1 ohm: a copper wire of diameter 0.1 in. and length 1000
ft; a copper wire of diameter 0.005 in. and length 2.4 ft; an iron rod of cross-
section 1 cm2 and length 750 m.
As an example of Ohm's Law, consider a storage battery to operate an electrical
instrument such as a relay in a control circuit. Assume the battery to have an emf
of 2.0 volts and an internal resistance of 0.03 ohm, and the relay .to have a resistance
of 0.37 ohm. Determine the current in the relay, the potential drop in the battery,
and the potential difference impressed upon the relay.
The total resistance of the circuit is the sum of the resistance of the relay and the
internal resistance of the battery; hence R = 0.37 + 0.03 = 0.40 ohm. The current
in the circuit is given by Eq. 17 .1 a.s I = 2.0 volts/0.40 ohm = 5.0 amp, and this is of
course the current in the relay. The potential drop in the battery is the product of
the current and the internal resistance, or 5.0 amp X 0.03 ohm = 0.15 volt. Finally,
the potential difference across the relay is 2.00 - 0.15 = 1.85 volts.

17-3. Measurement of Resistance. Ohm's Law provides a direct and

practical method for measuring electrical resistance; it is called the voltmeter-
§ 17-4 FACTORS AFFECTING RESISTANCE 299

ammeter method. The resistance of a device or load can be determined by

connecting it to a battery and using an ammeter to measure the current
through the load, and a voltmeter to measure the potential drop across it.
Figure 1 shows the connections for measuring the resistance R of a load.
The battery is indicated conventionally by using long lines for representing
the positive plates of the cells and short ones
Battery
for the negative plates. The voltmeter V is
.---~11111----,<-o
assumed to have a high resistance so that it
will not divert an appreciable amount of cur-
rent from the main circuit, which includes Load
the ammeter A. vVhen this circuit is closed,
the battery establishes a current through the
load, and readings are taken of the ammeter
and voltmeter. If the readings are I and
V respectively, then the resistance of the
load is R = V/I. An alternate method Frn. 17-1. Connections for meas-
uring resistance with voltmeter
for measuring resistance is described in and ammeter.
§ 17-13.
17-4. Factors Affecting Resistance. The resistance of an electrical con-
ductor depends upon the dimensions of the conductor and the substance of
,vhich it is made, and it is usually affected by temperature. In regard to
si;:r,e, the resistance is found to be directly proportional to the length of the
conductor and inversely proportional to its cross-sectional area, as might be
expected. Two conductors having the same dimensions, but made of dif-
ferent substances, have different resistances; this variation is due to the
different resistivities of the substances.
The resistivity of a substance is the resistance of a sample which has speci-
fied unit dimensions, as measured at a chosen temperature. Consequently,
a conductor of length Z, of cross-sectional area A, and of resistivity p, has a
resistance R which is given by the equation
l (17.3)
R = P -.
A

When the length l is expressed in meters, the cross-section A in square

meters, and the resistance R in ohms, the unit for resistivity will be
RA ohm·m 2
p = -l- = m = ohm ·m.

Copper has a resistivity of 1.72 X 10-s ohm ·m, which means that a specimen
of copper 1 m long and 1 m 2 in cross-section will offer 1.72 X 10-3 ohm resist-
ance between opposite faces of the cube.

To illustrate, calculate the resistance of a copper wire having a length of 1 m and a

diameter d = 0.1 cm.
300 ELECTRICITY & MAGNETISM Chap. 17

The cross-sectional area of the wire is A = 1rd2 / 4 = 0.00785 cm2 • A meter length
of this wire has a resistance of

R = p ~: = 1.72 X 10-s ohm· m X 78 _5 : ~-s m2 = 0.0219 ohm.

The resistance of a conducting medium which has a varying cross-section

affords an interesting application of the
foregoing equation. An example of thiskind
is presented by the earth when used as a
return path for a single line wire, as in
telegraphy. The current leaving one me-
tallic ground electrode diverges through an
ever-increasing sectional area of moist earth,
Frn. 17-2. Diagram used in calcu-
and then converges through a decreasing
lating earth resistance around a area upon the other ground electrode. Fig-
hemispherical ground conductor. ure 2 shows a hemispherical ground elec-
trode of radius a, and an elementary earth
shell of thickness dx at a distance x from the center of the electrode.
Since the shell has a cross-sectional area of 2 1rx2, its resistance is given by
Eq. 17.3 as
dx
dR = p--.
2 1rX2
Consequently, the earth resistance extending from the electrode to an infi-
nitely great distance can be found by integrating this expression from x = a
to x = oo; thus
p f"' dx p
R = 2 1r J
a x2 = 2 1ra.

Most of the resistance of the earth path is localized near the electrodes,
because there the sectional area is relatively small. Hence the foregoing
expression gives approximately the value of the earth resistance near one
electrode.
17-5. Resistance of Wires. In calculating the resistance of wires for indus-
trial purposes, it is common practice in the United States to express the length
of the wire in feet, and the cross-section in units of area called circular mils
(abbreviated CM). One circular mil is the area of a circle 0.001 in. (1 mil)
in diameter. Since the area of a circle varies as the square of its diameter, a
wire having a diameter of 2 mils will have an area of 4 CM, one having a
diameter of 3 mils will have an area of 9 CM, and so on. Thus, to find the
cross-section of a wire in circular mils, express the diameter in mils and square
this number. The circular mil is a convenient unit of area for round wires,
because it avoids the use of the factor 1r/4 in the calculations.
When the length of a wire is expressed in feet and its cross-section in circu-
lar mils, the corresponding unit for resistivity in Eq. 17.3 is the ohm: circular
§ 17-6 INFLUENCE OF TEMPERATURE UPON RESISTANCE 301

mil per foot. Thus, copper has a resistivity of 10.4 ohm· CM/ft, meaning
that a specimen of copper 1 ft long and 1 CM in cross-section will have a
resistance of 10.4 phms. Resistivity in metric units is often expressed in
microhm•cm, the microhm being one-millionth of an ohm; thus, copper has
a resistivity of 1.72 microhm · cm.
Table I lists the resistivities of a number of common materials in both
microhm · centimeters and ohm· circular mils per foot. The resistivity of a
substance is affected somewhat by temperature changes; the values given
apply to temperatures around 20°C. By comparing this table with that in

TABLE I. RESISTIVITIES OF CONDUCTORS

Substance Microhm·cm Ohm·CM/ft

Aluminum .......................... . 3.21 19.3

Carbon .................................. . 4000 to 7000 24,000 to 42,000
Constantan (Cu 60%, Ni 40%) ............. . 49 295
Copper·.................................. . 1.72 10.4
Iron ..................................... . 12 to 14 72 to 84
Lead .................................... . 20.8 125
Manganin (Cu 84%, Ni 4%, Mn 12%) ...... . 43 258
Mercury ................................. . 95.76 575
Nichrome (Ni 60%, Cr 12%, Fe 26%, Mn 2%) 110 660
Platinum ................................. . 11.0 66
Silver ................................. . 1.65 9.9
Tungsten .............................. . 5.5 33
Zinc ..................................... . 6.1 36.7

§ 11-9, it will be observed that those substances which are good conductors of
heat usually have low resistivity and are, therefore, good conductors of electric-
ity as well. Insulators, in contrast, have very high resistivities; for example,
that of Teflon (a polymer of tetrafluoro-ethylene) exceeds 1015 ohm· cm.
Low resistivity is necessary in wires that are used for transmitting elec-
tricity efficiently from. place to place. On the other hand, conductors of
moderately high resistivity are used when the primary purpose of the current
is the production of heat. Thus copper is used almost exclusively for electric
wiring, and alloys such as nichrome are used for electric heaters. Electric
fuses require a metal of relatively high resistivity and low melting point; they
are usually strips of lead alloy connected directly into a circuit. If the cur-
rent becomes too large the fuse melts and breaks the circuit, thereby protect-
ing the devices in the circuit and the wires themselves against overheating.
In commercial fuses the fusible element is mounted within a casing to confine
the arc which is produced upon melting.
17-6. Influence of Temperature upon Resistance. The resistance of a
metallic conductor usually increases as the temperature is raised. The law
governing this change has the same form as that for the expansion of a rod
302 ELECTRICITY & MAGNETISM Chap. 17

(§ 11-3), but the variation of resistance with temperature is usually much

greater than the corresponding change of size. In computations for resist-
ance at specified temperatures, the resistance at 20°C is usually taken as the
reference value; it may be symbolized as R20. When the temperature is
changed by !1t C 0 , the resistance can be expressed as
Rt= R20 (1 + a !1t), (17.4)
where a is a constant of the material called its temperature coefficient of
resistance. Numerically, a represents the increase of resistance of a I-ohm
resistance of the material when subjected to a temperature rise of 1 degree.
The temperature coefficient is positive for metallic conductors, and for all
pure metals the values have the same order of magnitude. Certain alloys
have been developed in which a is very small; these materials are useful for
the resistance elements of measuring instruments in which constancy of
resistance is desired. For practical purposes, where the temperature change
involved is small, the temperature coefficient is usually considered as remain-
ing constant at the value which it has at the lower temperature taken as a
reference point; actually it varies somewhat with the temperature, and for
accurate calculations the average value for the given temperature range
should be used. Equation 17.4 applies also to nonmetallic substances, in-
cluding carbon, liquids, and insulating materials; but for these the tempera-
ture coefficient is usually negative.
Table II lists the values of the temperature coefficients for a number of
materials in common use; these values apply to a reference temperature of
20°c.

TABLE II. TEMPERATURE COEFFICIENTS OF RESISTANCE

Material a for Centigrade Scale

Aluminum ......................... . 0.0038

Carbon (average between Oto 1850°C) .. -0.00025
Constantan ........................ . -0.00004 to +0.00001
Copper ............................ . 0.00393
Iron ............................... . 0.0062
Lead ............................ · - · 0.0043.
Manganin ......................... . 0.000002 to 0.00005
Mercury ........................... . 0.00090
Nichrome .......................... . 0.00017
Platinum .......................... . 0.0038
Silver ............................. . 0.0040
Tungsten .......................... . 0.0045
Zinc ............................... . 0.0037

The Resistance Thermometer. The fact that the resistance of a conductor

changes with temperature is utilized in the so-called resistance thermometer.
§ 17-8 ELECTRICAL ENERGY AND POWER 303

This device consists of a calibrated coil of wire, placed in the location where
the temperature is to be determined, and suitable instruments for measuring
the resistance of the coil. This type of thermometer is useful over a wide
temperature range, is precise, and can be read from a remote point.
A coil of platinum wire used as a resistance thermometer has a resistance of 10.0
ohms at 20°C and of 15.0 ohms when placed in a certain oven. Determine the temper-
ature of this oven.
Assume a to average 0.0038 per C0 over the temperature range involved; then the
temperature elevation lit of the oven above 20°C is given by Eq. 17.4 as
. lit = Rt - Rw = 15.0 - 10.0 = 132 co
aRw 0.0038 X 10.0 '
and the desired. oven temperature is 20 + 132 = 152°C.
17-7. Superconductivity. Certain metals are found to exhibit a remark-
able drop in resistance at very low temperatures; this effect was discovered
by Professor Onnes and is known as superconductivity. As the temperature is
lowered, the resistance of the specimen is observed to decrease in the manner
to be expected from Eq. 17.4; but at a certain temperature, which has a
characteristic value for each material and which is only a few degrees above
absolute zero (0°K or -273°C), the resistance vanishes. For example, the
resistance of lead at its transition temperature of 7.26°K becomes less than
10-12 of its value at 0°C.
Superconductivity has been observed in numerous metallic elements, in-
cluding tin at 3.69°K, mercury at 4.12°K, and niobium (columbium) at
9.22°K. :Most alloys have transition temperatures even lower than those of
their constituent elements, but there are some notable exceptions, such as
niobium nitride which becomes superconducting at about 14°K, a temperature
attainable with liquid hydrogen.
At the borderline between normal conduction and superconduction a slight
change in temperature of the substance causes an enormous change in resist-
ance, and small alternating currents can swing the substance in and out of
the superconducting state with every reversal of current. Hence supercon-
ductivity provides a method other than those considered in § 23-16 for detect-
ing or rectifying alternating currents. The superconductivity of niobium
nitride is made use of in detecting electromagnetic radiation in the infrared
region of the spectrum; an amount of energy as small as 2 X 10-6 erg can be
detected, and the response occurs in less than a thousandth of a second.
17-8. Electrical Energy and Power. It was shown in § 16-5 that the elec-
trical energy supplied to a circuit by a source of emf E in transferring a
charge Q is given by W = E Q. This result can be expressed more usefully
in terms of current than of charge. Since by Eq. 16.4 the quantity of charge
transferred is equal to the average current multiplied by the time of transfer,
or Q = It, the energy may be expressed as
TV= Elt or TV= Vlt; (17 .5)
304 ELECTRICITY & MAGNETISM Chap. 17

the latter form applies to any portion of a circuit across which the potential
drop is V. The energy Wis expressed in joules, the emf E or potential drop
Vin volts, the current I in amperes, and the time tin seconds.
Equation 17.5 is a general expression for circuits in which the energy may
appear in any form, such as heat, mechanical energy, or chemical energy. In
a metallic resistor-the only part of a circuit to which Ohm's Law applies-
electrical energy is converted entirely into internal energy of the metal and
is given off as heat. For such a resistor, V may be replaced by IR, and the
energy equation W = Vlt becomes
W = RI 2t; (17.6)
this represents the energy dissipated by the resistor in the form of heat.
The rate of delivering energy to a circuit is the power supplied to it; this is
found by differentiating Eq. 17.5 with respect to time. Hence power is
dW
dt = EI. Consequently, the power supplied to the entire circuit by an emf
E, or that supplied to a part of the circuit across which there is a potential
difference V, is
P = EI or P = VI. (17.7)
The unit in which power is expressed is the joule per second, or watt. When
the energy is dissipated as heat, the corresponding expression for power
becomes
p = RJ2. (17.8)
The relationship indicated by Eq. 17.6 shows that the heat produced in a
conductor is proportional to the resistance of the conductor, to the square of
the current, and to the time. This statement was first verified experimentally
by Joule; it is known as Joule's Law of electric
Thermometer Switch heating. The heat developed can be expressed
. - - - - - - - -----. 1 in calories by using the ratio between the
-
o->.----'
joule and calorie given in § 14-2; the result is
H = 0.239 RJ2t. (17.9)
Ammeter
Thus, if a current of one ampere is maintained
Calorimeter for one second through a. conductor of one
ohm resistance, the amount of heat produced
Fm. 17-3. Arrangement for
is 0.239 calorie.
measuring the heat produced by
a current through a resistor. Equation 17.9 can be verified experimen-
tally by using a calorimeter containing oil or
other insulating liquid and immersing a conductor in it as indicated in
Fig. 3. The current can be measured by an ammeter, the duration of the
test by a stop watch, and the heat liberated by knowing the water equivalent
of the calorimeter with its contents and observing the change of temperature
(§ 12-3).
§ 17-9 SIMPLE CIRCUIT FORMS 305

Compute the rating of an electric heater which will heat 60 liters of water from
15°C to 65°C in 25 min, on the assumption that no heat is wasted. Also, what is the
current in this heater if its resistance is 16 ohms?
The heat needed to raise the temperature of 60 liters of water through 50 C 0 is
60,000 X 50 = 3 X 106 ca1 = 32 X 106 l/"cal = L.9 5
X 106 JOU
. 1es. Th e power d"1ss1-
.
0. 39 ca JOU1e
pated is 12.5 X 10 6 joules/(25 X 60 sec) = 8333 watts; this is equivalent to a rating
of 8.33 kw. From Eq. 17.8, the power is 8.333 = 16 X I 2 ; whence the current is
I= 22.8 amp.
17-9. Simple Circuit Forms. The simplest forms of electric circuit are
shown in Fig. 4. ·rn part I, a generator G supplies current to a fixed resistor
R 1 and an adjustable resistor R 2 connected in series. In part II, the gen-
erator supplies current to these resistors connected in parallel.

II
Frn. 17-4. Simple forms of electric circuits: I, series; II, parallel.

In the series arrangement the current has the same value throughout the
eircuit; if the circuit is opened at any point, there will be no current in any
part of the circuit. The sum of the potential drops across the resistors is
equal to the emf applied to the circuit.
In the parallel arrangement the current supplied by the generator divides
between the two resistors, and the current in each is determined by its re-
sistance. The potential difference is the same across all branches, and the
currents in the branches add up to the value of the main current. The parallel
connection is widely used in electric lighting and power distribution; any lamp
or motor may be turned on or off without interfering with the operation of
the others joined to the same circuit.
The resistance within the source, whether a battery or a generator, must
often be considered in circuit calculations. This resistance is in series with
the external circuit and carries the same current. If the internal resistance
of the source is r and the current I, the potential drop within the source is
I X r, by Eq. 17.2. ·when a path is traced around a circuit going through
the source in the direction from its - to its +
terminal, there is within the
source a rise of potential amounting to E and at the same time a drop of po-
tential amounting to Ir. Hence the potential difference across the source is

V = E - Ir. (17.10)
30G ELRC'PRICI'IT & lvfAGlv'E"l'ISM Chap. 17

To measure the resistance of a battery, a circuit is set up as shown in

Fig. 5. The reading V1 of the voltmeter in part I is practically equal to the
emf E of the battery B, because the resistance of the instrument is so high
that the current taken by it is too small to produce an appreciable potential

Load

]]

Fm. 17-5. Arrangement for measuring the resistance of a battery.

drop across the internal resistance r of the battery. Next, a load is con-
nected to the circuit as in part II, and the voltmeter reading V2 and ammeter
reading I are taken. The voltmeter reading is less than before by the amount
Ir, and hence the resistance of the battery is
V1 - V2
r=
I
17-10. Series and Parallel Connections. Electric circuits are often com-
posed of resistors that are connected in series or in parallel, or in combinations
of the two. In the study of such groupings, it is often desired to determine
the value of a single resistance that would be equivalent to the entire group.
The symbol R with suitable subscripts ,,·ill be used to represent the resistances

i R
~ ~ ./VVVVV'~~~~~~
:.--v--:
Frn. 17-6. Resistors connected in series, and an equivalent resistor.

of the resistors included in a group, and I and V with corresponding subscripts

will be used respectively for the currents in these resistors and the potential
differences across them. The same symbols will be used for the "equivalent
resistance" but without subscripts.
Series Connection. When the resistors are connected in series, as in Fig. 6,
an application of Eq. 17 .8 indicates that the rate of heat production, or power,
§ 17-10 SERIES AND PARALLEL CONNECTIONS 307

in the circuit is

Since the current in a series circuit is the same throughout, it follo,vs that

R = R1 + R2 + R3 + · · · . (17.11)
This equation shows that the equivalent resistance of several resistors con-
nected in series is the sum of their individual resistances. This result was
tacitly assumed in the solved problem of § 17-2.
When Ohm's Law is applied to each resistor, Eq. 17.11 becomes

but, as before, the current is the same in all of them; consequently,

V = V1 + V2 + V3 + · · · .
Hence, the potential drop across an entire series circuit is the sum of the
potential drops across the component resistors.
Parallel Connection. When the resistors are connected in parallel, as m
Fig. 7, the rate of expending energy in heating the
circuit is again R1

P = Ri/1 2 + RJ2 2 + R3Js2 + · · · = RI2.

If Ohm's Law is applied to each resistor by re-
placing every current term with the fraction po-
tential difference over resistance, this expression
becomes

I R
~
I I
Since the potential difference is the same for all ---v ti
the resistors in parallel grouping, it follows that Fm. 17-7. Resistors con-
nected in parallel, and an
!__l..+l..+l..+ ... (17.12) equivalent resistor.
R - R1 R2 R3 .
This relation shows that the reciprocal of the equivalent resistance of a parallel
circuit is the sum of the reciprocals of the component resistances. It is an
interesting fact that the group resistance is less than the lowest resistance of
the individual resistors.
The connection of only two resistors in parallel is of frequent occurrence in
electrical circuits. For two resistances, R 1 and R2 in parallel, the equivalent
resistance is given by Eq. 17.12 as
308 ELECTRICITY & MAGNETISM Chap. 17

R = R1R2 ,
R1 + R2
that is, the product of their resistances divided by their sum.
17-11. CelJs in Series and in Parallel. Primary and storage cells are fre-
quently connected in series in order to obtain an increased emf. When a
number of cells are connected in series, their combined emf is the sum of the
emfs of the individual cells. Three lead cells or five alkaline cells are joined
in series to form the standard 6-volt storage battery.
Cells are sometimes connected in parallel in order to supply a large current
to a low-resistance load without demanding an excessive current from any
single cell. When a number of cells having equal emfs are connected in
parallel, the emf of the combination is the same as that of any individual cell.
Cells having appreciably different emfs would not be connected in parallel,
because wasteful circulating currents would be set up in the cells themselves,
even if the external circuit were open.
17-12. Sizes of Wires in Distribution Circuits. The wires in an electric
light or power circuit form a line connecting the generator with the lamps,
motors, or other load. The line wires are kept small in order to reduce their
cost. But if they are too small, their resistance will be unduly high, and the
potential drop along the line will be excessive; consequently, the potential
difference across the load will be too low to operate it properly. It is common
practice to restrict the potential drop along the wires to not more than 3 per
cent on lighting circuits, and to from 5 to 10 per cent on power circuits.
A second objection to undersized line wires is the possibility of overheating
by the current, which might cause the surrounding insulation to deteriorate.
A calculation of the proper wire size for an electrical installation affords
practice in applying the principles stated thus far in this chapter. An example
is solved below in detail, but without specific reference to the equations used.
An installation of lamps and motors, taking a total of 10 kw, is to be made at a
place 100 ft from electrical mains at which the potential difference is maintained
constant at 120 volts. What size of copper wire should be used to supply the necessary
current without exceeding a specified potential drop in the line wires of 2.5 per cent?
To answer this question it is necessary to find the resistance of line wires that can
carry the necessary current with the drop in potential stipulated. The potential drop
in the line wires is 120 X 0.025 = 3 volts, and therefore the potential difference
supplied to the load is 120 - 3 = 117 volts. The current taken by the load is
10,000 watts/117 volts = 85.5 amp. This is the value of the current in the line
extending from the mains to the load and back again, a total length of 200 ft. Hence
200 ft of wire should produce a drop of 3 volts with a current of 85.5 amp, and there-
fore the line resistance should be 3/85.5 = 0.0351 ohm. The cross-section of the wire
is 10.4 X 200/0.0351 = 59,300 CM.
17-13. The Wheatstone Bridge. The Wheatstone bridge is an instru-
ment for measuring the value of an unknown resistance by comparing it with
a known standard. This arrangement, devised in 1833 by S. Hunter Christie,
§ 17-13 THE WHEATSTONE BRIDGE 309

was brought to public attention by the English physicist Sir Charles Wheat-·
stone (1802-1875), and has remained associated with his name.
The bridge consists of four resistances, M, N, P, and X, connected to a
battery and to a sensitive current-measuring device called a galvanometer,
as shown in Fig. 8. One of the resist- c
ances, X, is unknown, and the other
three are known and adjustable in value.
In using the bridge, resistances M and N
are given suitable values, and then resist-
ance Pis manipulated until the galvanom-
eter shows no deflection. The bridge is
then said to be balanced, and the resistance D
of X can be found from a knowledge of
the values of resistances M, N, and P. ...______
- +-111 t------.
- ..
To derive the equation of the balanced Battery
bridge, it should be observed that, since
Frn. 17-8. Diagram of a Wheat-
there is no current in the galvanometer, stone bridge.
the potential at point C must be the same
as that at point D. Hence the potential drop across M must be the same as
that across N; also the drop across X must be the same as that across P.
Also under this condition the current in M must be the same as that in X,
say I c, and the current in N the same as that in P, say In; hence it follows
that the potential drops are alike in pairs, namely

IcAf = InN,
and IcX = InP.
The first of these expressions is then divided by the second, term by term, and
rearranged; whereupon
M X (17.13)
N = p'

This is the equation of the balanced bridge, and from it the value of the un-
known resistance X can be found. The resistances M and N which appear
as a ratio in the equation are called the ratio arms of the bridge, and resistance
P is called the rheostat arm.
Commercial bridges are made in various designs. In the form illustrated
in Fig. 9, the resistances M, N, and P are all included within the case, and
binding posts are pro.vided for external connections to the battery and gal-
vanometer and to the resistance being measured. The two ratio arms are con-
nected into the circuit by inserting plugs, and the rheostat arm has five dials
by which it can be adjusted in 0.1-ohm steps from Oto 10,000 ohms. Switches
known as keys are provided in the battery and galvanometer circuits, since
310 ELECTRICITY & MAGNETISM Chap. 17

in the use of the bridge these circuits are closed only momentarily while a
balance is being obtained.

FIG. 17-9. Wheatstone bridge and its internal arrangement.

(Courtesy of Leeds & Northrup Company)

17-14. The Potentiometer. The emf of a voltaic cell or a battery is not

given exactly by the reading of a voltmeter connected across the terminals,
because even a high-resistance voltmeter draws some current. Such emfs can
be compared accurately with the known emf of a standard cell by means of a
potentiometer. An advantage of the potentiometer method is that the meas-
urement is made when the cell is not supplying any current whatever, and
since there is then no potential drop in the cell and no polarization, the value
measured is the true emf of the cell.

Galvanometer

Slidewire
+ -
-------11---------'
B
Frn. 17-10. Elementary potentiometer circuit.

The apparatus used in a potentiometer measurement is indicated in Fig. 10.

The slidewire MN of uniform cross-section is supplied with current from a
battery B. By means of a double-throw switch, either the standard cell S
or the cell X under test is connected through the galvanometer to a slider
which can be touched to any point along the slidewire. The battery B should
have a constant emf, which should be larger than that of either Sor X.
The current l from the battery B establishes a uniform potential drop along
§ 17-15 ELECTRIC NETWORKS 311

the slidewire MN, the potential being higher at M than at N. The cells S
and X both have their positive terminals connected to M, and these terminals,
therefore, have the same potential as M. When the double-throw switch is
closed either way, the slider, if free from the slidewire, will have a potential
which is lower than that of M by an amount equal to the emf of the cell which
the switch has placed into the circuit. There will be some point along MN
that has the same potential as the slider; this point can be found by touching
the slider to different points 'until one is found for which the galvanometer
shows no deflection. Under these conditions there will be no current in the
circuit branch containing the cell and galvanometer, and the potential drop
along the wire to the point of contact will then be equal to the emf of the cell.
Thus, with the double-throw switch connected to the standard cell of emf Es,
a balance will be obtained at some point s such that Es = IRs, where R. is
the resistance of the slidewire from M to s. Then with the double-throw
switch connected to X, another balance will be obtained at some other point x
such that Ex = IRx, where Ex is the emf being measured and Rx is the resist-
ance of the slidewire from M to x. From these relationships,
Ex
- = fRx
-,
E. IR.

whence Ex= E. ~:, (17.14)

which expresses the unknown emf in terms of a known emf and of two known
resistances.
Since the slidewire has a uniform cross-section, its resistance is proportional
to its length; hence the unknown emf may also be expressed as
lx
Ex= E. ls,
where lx and z. are the wire lengths from M to x and from M to s, respectively.
In commercial potentiometers, the slidewire is usually replaced, at least in
part, by resistance coils arranged with dial switches, and potential differences
up to about 1.5 volts can be read directly from the setting of the instrument
at balance. Higher values can be measured by using the potentiometer with
a potential divider; this is an arrangement of resistances for dividing the total
potential difference into two definite parts, one of which is less than 1.5 volts.
17-15. Electric Networks. The circuits employed in electrical work cover
a variety of types; they may be groupings of simple series and parallel paths
like those described in § 17-9, or combinations of such groupings; or be ar-
rangements which take the forms of the letters wye (Y) and delta (Ll), or
repeated groupings in the form of a ladder. The term electric network is used
for the more complicated circuits.
In analyzing electric networks the plan is usually to simplify them step by
step, and then solve for the unknown factors. Figure 11 illustrates the pro-
312 ELECTRICITY & MAGNETISM Chap. l'i

cedure. Part I shows the branches BF and CD in parallel, and this pair con-
nected in series with path AG, which includes the battery. First, the parallel
group of branch BF is simplified, as shown in part II, and then the two parallel
paths BF and CD are merged, as shown in part III. It is possible, by repeated
application of Ohm's Law, to compute the current supplied by the battery,
the current in each of the resistors, and the potential drop across each.

12-volt battery

Ac:r
,-----J-J---, 1
A _ + ,,,, - G A 11 11 G
4
3 7
B F B F

-
10 10
c D c D

j II m
Fm. 17-11. Steps in simplifying an electric network.

The circuit BCDF shown in part I of Fig. 11 is connected across a 12-volt battery
which has an internal resistance of 1 ohm. The Tesistance of each of the component
resistors of the circuit is indicated in ohms. Find the current in each.
In part I, the 4-ohm and 12-ohm parallel combination is equivalent to a single
res1s X 12 = 3 oh ms; t h'1s 1s
. t ance of 4 + ' m' series
' wit
. h the 7-o h m resistor,
. as shown m
.
4 12
part II. The resistances of branches BF and CD are each 10 ohms; when combined
in parallel an equivalent load resistance of 5 ohms is obtained, as indicated in part III.
The current supplied by the battery is I = E/R =¥ = 2 amp. The potential
drop across the 1-ohm battery resistance is V = I X r = 2 X 1 = 2 volts, leaving
12 - 2 = 10 volts across the battery terminals A and G. This same potential differ-
ence is available to the two branches BF and CD, and since each of these has a resist-
ance of 10 ohms, the current in each is 1 amp. In the branch BF, the potential
difference across the 7-ohm resistor is V = I X R = 1 X 7 = 7 volts, and that across
the 4-ohm, 12-ohm parallel combination is V = 1 X 3 = 3 volts. The currents in the
4-ohm and 12-ohm resistors are ]4 = t = 0.75 amp and 112 = i32 = 0.25 amp.

17-16. Kirchhoff's Network Laws. The more intricate networks are not
easily solved by the method used in the preceding section·, and recourse is had
to two generalizations pointed out by the German physicist Gustav R. Kirch-
hoff ( 1824-1887), as follows :
1. At any point in an electric circuit where two or more conductors are
joined, the sum of the currents directed toward the junction equals the sum
of the currents directed away from the junction. This law can be rephrased
to state that the algebraic sum of the currents at a junction equals zero. The
currents toward the junction are taken as positive, and those away from it as
negative; then symbolically
T,[ = 0. (17.15)
 § 17-16 KIRCHHOFF'S NETWORK LAWS 313

2. For any closed path in an electric circuit, however complicated, the sum
of the rises and falls of potential around the path equals zero. Rise of potential
is regarded as positive and drop of potential as negative. When the fact is
introduced that a current I in a resistance R corresponds to a potential differ-
ence IR, this law may be stated as
J;E + J;JR = 0. (17.16)
Thus, in the diagram of the potentiometer, Fig. 10, for the path extending
from M along the slidewire to the contact points at balance and then through
the standard cell S back to M, the second law gives '- I Rs + Es = 0; and for
balance at x with the other cell -!Rx+ Ex = 0. These equations yield
the same result as found in § 17-14.
In the application of these laws to the solution of a network, it is necessary
· to assume some direction for the current in each branch and then to express
Kirchhoff's Laws in as many independent equations as there are unknown
currents. After that the equations are
solved simultaneously to find the cur- c
rents in the several branches of the cir-
cuit. If the result gives a negative value
for any current, the direction of that 3 8
current is opposite to the direction origi- 6
nally assumed.
volts ==:::
To illustrate the foregoing procedure, con-
sider the network of Fig. 12 in which the A D B
numbers on the resistors represent their resist-
ances in ohms. Current is supplied to the Frn. 17-12. A network to be solved
network by two batteries. Find the values by Kirchhoff's Laws.
of the currents marked 11, 12, and 13.
The assumed directions of these currents are shown. From the First Law (at
point C or point D): [ 1 + 1 2 - 13 = 0. From the Second Law (around path ACD):
+18 - 311 - 61 3 = O; also (around path BCD): +4 - 812 - 613 = 0. Thus, the
three simultaneous equations to be solved for the three currents are
11 + 12 - l3 = 0
311 + 613 = 18
812+613 = 4
and the results in amperes are found to be 11 = 2.53, 12 = -0.80, and 13 = 1.73. The
negative sign for 12 shows that the direction of this current should be reversed, but no
other changes are required.

PROBLEMS
1. How much current does a resistor of 4 ohms resistance take when connected to a
6-volt storage battery? and when connected to a 120-volt lighting circuit? If the
largest current the resistor can take without damage is 40 amp, what is the limiting
potential difference that may safely be applied to it?
314 ELECTRICI'l'Y & MAGNETISM Chap. 17

2. A slidewire resistor has 280 turns wound on an insulating cylinder and its total
resistance is 45 ohms. \Vhen a current of 2.5 amp traverses the entire length of the
wire, what is the potential difference between adjacent turns of the winding?
3. A battery having an emf of 6.0 volts and an internal resistance of 0.12 ohm
supplies a current of 2.4 amp to a lamp. Find the potential difference across the
terminals of the battery and also the resistance of the lamp.
4. A hollow metallic cylinder mounted concentrically about a metal rod is fitted
at one end with a circular disk of highly insulating material; the device so formed is
set with its axis vertical so that it may serve as a container for a liquid. With a
particular liquid in place, its resistance is measured between rod and cylinder. Let
the radii of the rod and cylinder be a and b respectively, the depth of liquid be h, and
the observed resistance be R. Derive an expression for obtaining the resistivity of the
liquid.
5. The "third-rail" of an electric railway is made up of 39-ft lengths of steel having
a sectional area of 6.86 in. 2 What is the resistance of one length of rail if the resis-
tivity of the material is 5 microhm ·in.? What is the resistance of the rail per mile
if 3 per cent is added to allow for the resistance of joints and bonds?
6. A helical winding is formed of 90 turns of copper strip that is 1 cm wide and 1 mm
thick, and the average diameter of the winding is 30 cm. What is the resistance of
the winding at 20°C?
7. A resistor is to be constructed of nichrome wire that has a diameter of 0.081 in.
What length of wire is required for the resistor if its resistance is to be 15 ohms at
100°C?
8. A constantan wire 10 in. long and 0.005 in. in diameter serves as an electrical
strain gage in tensile strength tests. In a certain test the wire was stretched and its
resistance was increased by 5 per cent. Compute the initial resistance of the wire,
and the amount the wire was stretched in the test. Assume that when the wire
elongates by a certain percentage, its sectional area is reduced by that same percentage.
9. The conductors of a certain transmission line are copper tubes having an inside
diameter of 1 in. and a wall thickness of t in. What is the resistance of these con-
ductors per mile of length at 40°C?
10. How much current does an electric iron of 30 ohms resistance take when con-
nected to 115-volt service mains? What is its rating in watts? How much heat does
it produce per hour?
11. A 20-hp direct-current motor operating at full load takes a current of 38.5 amp
at 440 volts. What is the efficiency of the motor? How much does it cost to operate
the motor at full load for a day of 8 hr use, if electricity costs 2 cents per k.v · hr?
12. A 32-volt storage battery is charged from 120-volt direct-current supply mains,
a resistor being connected in series with the battery to limit the charging current to
15 amp. Compute the rate at which energy is wasted in heating the circuit. If all
the heat produced could be made available in heating water, how long would it take
to bring 1 kg of water from 15°C to the boiling point? ·
13. An insulated cabinet has a surface area of 50 ft 2 formed of walls 2 in. thick
made of rock wool having a thermal conductivity of 7.2 Btu· in./ (ft2 ·day· F 0 ) . An
electric heater operated continuously within the cabinet keeps the interior at a temper-
ature of 98°F when the surroundings are at 60°F. Find the rating of the heater in
watts.
14. A dry cell develops 1.490 volts across its terminals when on open circuit; this is
lowered to 1.452 volts when the cell delivers current to an external circuit of 3 ohms
resistance. Calculate the internal resistance of the cell.
15. The emf and internal resistance of a voltaic cell are to be measured, but only
an ammeter and two resistors of 5 and 10 ohms resistance are available for the test.
When the cell is connected in series with the ammeter to the 5-ohm resistor, the meter
 PROBLEMS 315

reads 230 milliamp, and when similarly connected to the 10-ohm resistor the reading
is 118 milliamp. What results are derived from these data?
16. Three resistors are available, one has 10 ohms resistance, another 30 ohms, and
the third 60 ohms. Give the values of resistance that can be formed by these resistors
when all are used and connected in series, in parallel, and in series-parallel groupings.
17. Suppose that two resistors of 2 ohms and 6 ohms are connected in parallel and
this combination is joined in series with a 4-ohm resistor and a battery having an emf
of 12 volts and an internal resistance of 0.5 ohm. Find (a) the current in the 6-ohm
resistor, and (b) the power expenditure in the 2-ohm resistor.
18. Single resistors of 6 and 12 ohms resistance are connected in parallel as one
group. Two other resistors of 8 and 24 ohms are connected in parallel to form another
group. These two ·groups are connected in series across a battery having an emf of
24 volts and an internal resistance of 1 ohm. Calculate the current in each resistor
and also the potential drop across each.
19. The accompanying figure shows a delta grouping of three resistors having the
resistances indicated in ohms and connected to terminals A, B, and C; also a Y group-
ing of three resistors a, b, and c joined to similar terminals. What should be the

A A

c
25
PROBLEM 19.

resistances of the latter in order that the two networks may have the same resistances
between corresponding pairs of terminals? Hint. Write an equation for the resist-
ance between terminals A and B for the two networks with terminal C open; repeat
this procedure for the other pairs of terminals.
20. If the three resistors of a Y network have a resistance of 10 ohms each, what
would be the resistance of each of three identical resistors in an equivalent ~ network?
21. Two storage cells connected in parallel supply jointly a current of 12 amp to an
external circuit. The cells have emfs of 2.1 volts and 2.0 volts, and each has an
internal resistance of 0.05 ohm. Find the current in each cell.
22. Two dry cells, each having an emf of 1.5 volts and an internal resistance of
0.2 ohm, are to be connected in such a manner as to supply as large a current as pos-
sible to an external circuit. Should the cells be connected in series or in parallel (a) if
the external circuit has a resistance of 0.1 ohm, (b)
if the external circuit has a resistance of 1.0 ohm?
23. To show why batteries having unequal emfs
should not be connected in parallel, imagine the
batteries in the diagram to have emfs E1 = 6 R
volts and E2 = 5.5 volts and internal resistances r1 r2
of r1 = 0.3 ohm and r2 = 1 ohm respectively, and
· the load to have a resistance R = 10 ohms. Desig-
nate the currents supplied by batteries E1 and E2
as 11 and 12 respectively, and determine the current PROBLEM 23.
in each battery and in the load.
24. A lamp load which takes a current of 30 amp is located at a distance of 225 ft
from a generator that maintains 120 volts across its terminals. What diameter of
316 ELECTRICITr & Jl!IAGNETISivI Chap. 17

copper wire is needed for the line connecting the lamps to the generator, if the allowable
potential drop is 3 per cent?
25. An electric power and light load aggregating 6 kw input is to be operated at
115 volts at a place 200 ft from service mains. What potential difference would have
to be maintained at the mains if copper wires having a resistance of 0.197 ohm per
1000 ft were used for the installation? If at times only one-thirtieth of the total load
is operated on this circuit, how much greater is the potential difference on that load
than it is when the full load is being carried?
26. The slidewire circuit of a potentiometer is supplied with a steady current from a
storage cell. When a Weston standard cell having an emf of 1.01830 volts is bridged
across part of that circuit the galvanometer deflection becomes zero when the setting
(resistance along slidewire from M in Fig. 10) includes ,50.91 ohms, and when a dry
cell is substituted for this standard the setting includes 75.15 ohms. What is the emf
of the dry cell and what is the steady current in the slidewire circuit? The dry cell
is then used in series with a standard I-megohm resistor to test the sensitivity of the
galvanometer, and a deflection of 3.5 divisions is observed on the galvanometer scale.
What is that sensitivity, expressed in microamp per division, if the instrument has a
resistance of 2400 ohms?
27. The accompanying diagram shows a network supplied by two batteries, E1 of
6 volts and E2 of 2 volts, and the numbers on the resistors indicate their resistances
in ohms. Use Kirchhoff's Laws to determine the current in each battery.
40

10 20 R2
Ri R3
+
E2 I
I 1+
30 +
E1 f3
PROBLEM 27. PROBLEM 29.
28. Five resistors are connected in the form of a Wheatstone bridge, such as illus-
trated in Fig. 17-8. They have the following resistances in ohms: M = 400, N = 1000,
G = 600, X = 500, and P = 200. Compute the equivalent resistance of the network
as measured across the terminals to which the battery is connected. Hint. The
Y-~ transformation developed in Prob. 19 will prove useful.
29. Find the current in each of the three resistors of the circuit shown in the ac-
companying figure. Assume R1 = 20, R2 = 5, and R3 = 10 ohms, and E1 = 100,
E2 = 150, and E3 = 200 volts.
30. In the ladder-type network shown in the accompanying diagram, in which the
numbers represent the resistances in ohms, determine the value of· the terminal resist-
ance R that will make the resistance at points cd, reckoned to the right, also equal
to R. When this value of resistance is used at R, what will be the resistance between
terminals ab?
3 3 3 3 3 3 3 3
a

b d

PROBJ:.,EM 30,
 18
ELECTROMAGNETISM
The last mentioned of the three principal effects of the electric current,
namely Electromagnetism, is considered in this chapter. The study begins
with the configuration and densities of the magnetic fields produced by the
current in circuits of simple shapes, and continues with the action of the field
on current-carrying conductors and on moving charges. The results are ap-
plied to the theory of electrical measuring instruments, to the determination
of a fundamental electronic ratio, and to a modern device for accelerating
charged particles.
18-1. Magnets and Magnetic Fields. The property of magnetism has
been known for centuries. The ancients recognized that a black mineral
called lodestone exhibited this property through its ability to attract pieces
of iron and to turn to a particular direction when suspended. The word
magnetism and the name of the mineral-magnetite (Fe304)-were presum-
ably derived from Magnesia, a district in Asia Minor where the ore was
plentiful. The term magnet is applied to bodies which possess magnetism and
can attract iron. Magnetism is not exhibited uniformly over the surface of a
magnet; the regions where its effects are pronounced are called poles.
Magnetism can also be produced by electric currents. A familiar arrange-
ment is a coil of wire wound on an iron core; the iron becomes magnetized
whenever a current is established in the winding. Such an arrangement is
called an electromagnet to distinguish it from the other type-the permanent
magnet-which remains magnetic after having been magnetized.
Just as an electric field exists in the region around an electric charge, so a
magnetic fielil exists in the region around a magnet. Because of magnetic
fields, forces are exerted upon magnets and upon electric currents. The force
considered in § 16-2 acting between two wires carrying electric currents is
attributed to the fact that each cmrrent is located in the magnetic field pro-
duced by the other current.
When a magnetic field acts upon a magnet that is suspended, the magnet
aligns itself with the field. This is the same action that causes a compass
needle to orient itself in the earth's magnetic field; the end of the needle which
points .in the north direction is called the north-seeking pole, or simply north
pole, the other which points south is the south pole.
317
318 ELECTRICITY & MAGNETISM Chap. 18

If a light magnet is mounted on a piece of cork and floated on water while

in a uniform magnetic field, the magnet will rotate in the manner described.
Furthermore, it will have no motion of translation, showing that the uniform
field acts equally, but oppositely, upon the two poles. This is experimental
evidence that the two poles of a magnet are equal in strength.
The magnetic field in a region is regarded as permeated with magnetic flux.
The flux is represented by so-called lines of induction; the direction of these
lines shows the direction of the flux. The lines of induction from a magnet
extend outward from the north pole and inward to the south pole, and con-
tinue through the magnet itself to form complete loops. The number of
lines extending normally through a unit area shows the density of the flux.
The flux density, also called the magnetic induction, is defined as the flux per
unit area.
The symbol for magnetic flux is <I>, and that for flux density is B. If an
amount of flux <I> passes perpendicularly through an area A, then the magnetic
flux density through this area has an average value of

B = !· (18.1)

When the magnetic flux is not perpendicular to the reference area, the com-
ponent of the flux density which is perpendicular to it is B cos O, where () is
the angle between the direction of the flux and a normal to the area. If the
flux density is not uniform, then the total flux can be determined through
integration by means of the expression
<I> = f B cos O dA,
where B is the flux density through the area dA.
The unit of magnetic flux in the mks system is the weber; it is named after
the German physicist Wilhelm E. Weber (1804-1891). The unit of magnetic
flux density is the weber per square meter; there is no special name for this
quantity in this system of units.
In the electromagnetic system of units alluded to in § 16-4, the unit of
magnetic flux is the maxwell, named after the Scottish physicist J. Clerk
Maxwell (1831-1879). The unit of magnetic flux density in this system is
called the gauss, after the German mathematician and phys~cist Karl F. Gauss
(1777-1855). A gauss is a flux density of one maxwell per square centimeter.
For purposes of comparison it is helpful to know that
1 weber = 108 maxwells,
and 1 weber/m2 = 104 gaustoes.
18-2. Magnetic Effect of the Current. The discovery of electromagnetism
was made in 1820 by the Danish physicist Hans Christian Oersted (1777-1851),
who observed that an electric current in a conductor is surrounded by a mag-
netic field. He also found that the direction of the field about a current-
carrying wire is perpendicular to the wire, and that the magnetic induction
§ 18-3 AMPERE'S GENERALIZATION ON MAGNETIC INDUCTION 319

diminishes in receding from the wire. The magnetic flux about an isolated
wire is concentric with it, as can be proved by arranging the wire vertically
and moving a compass around it. Figure 1 shows diagrams of the magnetic
field around a wire; the heavy circles indicate the wire in section and the light
circles represent lines of flux. When the current is directed toward the ob-
server, as represented by a dot within the wire, the magnetic flux is counter-
clockwise, as shown at the left. When the current is directed away from the
observer, as represented by a cross within the wire, the flux is clockwise, as
shown at the right.
A simple rule for determining the relative directions of current and flux
makes use of the right hand, the fingers being curved as though grasping the
wire carrying the current, and the thumb being outstretched. When the
thumb of the right hand point:.; in the direction of the current, the fingers will
indicate the direction of the magnetic flux.

p
p

Current out Current in

Fm. 18-1. Repre,mntatio11 of Fm. 18-2. General-

the magnetic flux around a eur- ized circuit form used
rent-carrying wire. in finding the mag-
netic flux around a
wire.

18-3. Ampere's Generalization on Magnetic Induction. The magnetic

induction around a circuit carrying a current can be computed by means of a
fundamental expression attributed to Ampere. As applied to any shape of
conductor, such as the curved wire pq in Fig. 2, the expression states that the
current in each elementary length of the wire contributes something to the
magnetic field at all points around the wire. Specifically, the contribution
of the element Lis to the flux density at any point P is
LiB = KI sin () Lis, (18.2)
r2
where I is the current in the wire, r the length of a line joining the wire element
and the reference point, () the angle between this line and the direction of the
current, and K a constant which, like k in Coulomb's Law, § 15-5, depend8
320 ELECTRICITY & MAGNETISM Chap. 18

upon the choice of units. The total flux density at the point Pis, therefore,
the summation of the contributions of all conductor elements and can be
expressed as
B = K L I sin () D.s. (18.3)
r2

Its direction is perpendicular to the plane which includes As and r.

That the magnetic induction should vary inversely as the square of the
distance from the current-carrying wire producing it might be expected through
similarity with the electric field intensity established by an electric charge,
§ 15-6. Further, it is reasonable to suppose that doubling the current or
doubling the length of the element should double the induction. The fact
that the flux is concentric about the wire suggests that the induction due to
the element has its greatest value in the direction normal to the element
(e = 90°), and its least value along the direction of the element (() = 0°), but
experimentation was required to prove that it varies as the sine of angle e.
The constant Kin the mks system has the value

K = ..!:!:2_ = 10_7 weber,

4 7r amp·m
where µ 0 is the permeability of free space, § 16-3; hence

0
= 4 7r 10_7 weber
µ amp·m

With the foregoing value for K in Eqs. 18.2 and 18.3, and with the current I
in amperes, and the length As and distance r in meters, the flux density will
be in webers per square meter.
For comparison, in the electromagnetic system the constant K has the nu-
merical value of unity, and with the current in abamperes and the linear
dimensions in centimeters, the flux density will be in gausses.
The Ampere formula for magnetic induction or flux density can be applied
to all forms of circuits, a few of which are treated in the following sections.
The value used for the constant K is that of the mks system, and the other
quantities are in units appropriate to that system.
18-4. Flux Density along Axis of Circular Coil. The simplest shape of
circuit to be considered is a circular loop of wire carrying a current; Eq. 18.3
will be used to determine the flux density, first at the center of the loop and
then at some other point along its axis.
Figure 3 shows a wire forming an arc of radius r centered at P and carrying
current in a clockwise direction, with the wires connecting the arc to the rest
of the circuit placed radially so that they will not affect the field at the center.
From the right-hand rule, it is found that the magnetic field is directed into
the page in the region toward the center of the arc. For all positions of the
element ds the radius r is constant and perpendicular to the element,. and
§ 18-4 FLUX DENSITY ALONG AXIS OF CIRCULAR COIL 321

consequently the angle () is 90°. Hence, the flux density at the center of the
arc produced by a current I in the wire becomes

µal
B = 4--:;
1rr~
J
ds = 4~µal
1rr s,
in which f ds = s, the full length of the arc.
If the arc in Fig. 3 is extended to form a complete circular loop of radius r,
the integral becomes the circumference of the circle, 2 1rr; then the flux density
at the center will be
B = µol.
2 r
A coil of N turns will have N-times the effect, provided the turns are concen-
trated so that the entire coil has little axial length compared with its radius.

FIG. 18-3. Dia- _FIG. 18-4. Diagram for calcu-

gram for evaluat- lating the flux density at any
ing the magnetic point along the axis of a circular
flux density at the coil.
center of a circular
arc of ,vire.

Consequently, the flux density at the center of a flat circular coil carrying a
current I is
NI (18.4)
B = µa-·
2 r
With the current in amperes, the product NI is in ampere·turns; with the
radius of the coil in meters, the fraction is in ampere· titrns per meter. Since
the unit of B is the weber per meter2 , it follows that permeability has as its
um't th e weber
t , as we ll as t h e weber given
. . t h e prece d"mg sect10n.
m .
amp· urn·m amp·m
Finally, it is desired to find the flux density at a point somewhere on the
axis of the circular conducting loop but not in its plane, as at P in Fig. 4.
If z is the distance from the element ds to point P, then from Eq. 18.2 the field
at the point due to the current in the element is
322 ELECTRICITY & MAGNETISM Chap. 18

dB = µol sine ds,

4 1r z2
and it is directed at right angles both to z and to ds. For a complete turn of
wire, the component perpendicular to the coil axis y will cancel, but the com-
ponent along the coil axis will have a definite value which is found by reso-
lution to be (r/z) dB. With these changes, the magnetic induction due to
current in the coil at a point on its axis distant y from its plane is
B = NI µo r2 (18.5)
2 (r2 +
y2)012

18-6. Flux Density Within a Solenoid. A type of winding that is widely

used in magnetic apparatus consists of a coil of wire wound uniformly in a
long helix; such a winding is termed a solenoid. To determine the flux density
along the axis of a solenoid, use is made of Eq. 18.5 for the induction produced
by the turns forming an elementary length of the solenoid, and the resulting
expression is integrated over the entire length of the winding.

1
Center line
N turns of toroid

FIG. 18-5. Diagram of a sole- .Fin. 18-6. Sectional

noid for use in calculating the view of a ring solenoid.
flux density along the axis.

Figure 5 shows a solenoid having N turns distributed uniformly over its

axial length l. The number of turns per unit length is N /l, .and the number
of turns in an element of length dy is N dy/l. The current in this element of
the winding will contribute to the flux density at an axial point P an amount
dB directed along the axis, of which the value is
NI µo r 2
dB = 2 l (r2 + y2)s12 dy,
where I is the current in the solenoid, r is its radius, and y is the axial distance
from the point P to the element.
It is convenient to express y and dy in terms of the angles e and de at point P
as shown in the figure. Thus y = r cote; whence dy = -r d8/sin2e. Con-
§ 18-6 FLUX DENSITY ABOUT A 8TRAIGJ/T CONDUCTOR 323

sequently, the elementary contribution to the field along the axis is

dB = NI µo sm
-2l . 8 de ,

and the total flux density at the point P is

B =
NI
-2l µo
f /i·
asrn 8 d() =
NI µ 0 (cos a
2l - cos f3).

For a long solenoid, the density along the axis at points not too near its ends
is found from this result by taking a = 0° and /3 = 180° as the limits; whence

B = µ0
NI
-. (18.6)
l
A solenoid that is bent around to form a ring is called a ring solenoid; its
core has the shape of a toroid. In order to form the winding it is necessary
to thread the wire repeatedly through the center opening of the toroid. It
will be evident from the preceding paragraph that the flux density at any
point along the center-line of the toroid, such as shown in Fig. 6, is
NI
B = µo -l-,

where N is the total number of turns, I the current through them, and l the
circumferential length of the toroid. Although this expression gives the flux
density at the axis, the density is approximately
uniform over the entire cross-section of the core. A

1-- ----
18-6. Flux Density about a Straight Conductor. r p
The general expression for flux density given by R - f3
Eq. 18.3 w,as developed from experiments by the
Fre.nch physicists Jean B. Biot (1774-1862) and Felix s
Savart b(l 79dl-~84dl) for a_ long fstraighht condductor.
1t can e enve analytica11y or sue a con uctor ds
J__
by letting AB in Fig. 7 represent a portion of a i,--
straight wire that extends indefinitely in both direc-
tions, selecting a point P in space at a perpendic- B
ular distance r from the wire, and solving the
Fw. 18-7. Diagram for
equation for the flux density at this point in the calculating the flux den-
same manner as before. sity near a straight wire.
The field at the reference point may be considered
as produced by the current in elementary wire lengths, such as ds, that
collectively make up the whole wire. In terms of the symbols R, s, 8, and
/3 used in the figure, the flux density at P due to a current I in the elemen-
tary length ds is expressed as

dB _ I sin e ds = µof cos (3 ds

, - /).() 4 1r R 2 4 1rR 2 '
324 ELECTRICITY & MAGNE'Pl8lvl Chap. 18

where {3 = 90° - e. Since s = r tan {3, ds = r d{3/cos 2 {3, and R = r sec {3, it
follows that
dB = µ 0 f cos {3 (r d{3) = µof cos {3 d{3.
4 1rr2 sec 2 {3 cos 2 {3 4 1rr
The total flux density at P is then

B = µof
4 1rr
f +!
_,i:
2
cos {3 d{3.
2

Upon integration, this expression gives the flux density at a distance r from
a long straight wire carrying a current f as

(18.7)

With the current expressed in amperes, the distance r in meters, and the
permeability in webers per ampere· turn· meter, the flux density will be in
webers per turn per meter2 • This unit agrees with that previously given for
B, since one turn of wire is implied even though the wire extends indefinitely
in both directions, for the current must have a return path.
Compute the magnetic induction due to a current of 8 amp in a long straight wire,
at a point 5 cm in air from the wire.
The value for the permeability of air is given in § 18-3 as µo = 4 1r X 10-7
newton/amp 2, which is the same as 41r X 10-7 weber/(amp·turn·m). Hence, by
Eq. 18-7, the flux density at the desired point is

4 7r 10_7 weber
X 8 amp
amp·turn·m _ _4 2
B= 2 1r X O.OS m - 3.20 X 10 weber/m .

18-7. Force on a Conductor in a Magnetic Field. The force exerted be-

tween the currents in two parallel wires, described in § 16-2, can now be
related to the magnetic field in which they are located. With the value of
K appropriate for mks units, this force is expressed as

F = .J!:2-. 2 fif 2 l,
4 1r r

where /1 and / 2 are the currents in the two wires of length l that are separated
a distance r. If Eq. 18.7 for the flux density near a straight wire is now
applied to the one with current Ii, then the flux density at a distance r from
this wire is
B -_ µof1
2 1rr'

which is, of course, the density of magnetic flux where the other wire is
lorated. A combination of these equations shows that the force on the second
§ 18-8 THE GALYANOMETER 325

wire can be expressed as F = BI2l. The result may be written without sub-
scripts as
F = BIZ. (18.8)

Although this force equation was derived for a particular conductor con-
figuration-that of parallel wires-it is perfectly general and can be applied
to all conductors that are located normal to the magnetic flux. If the con-
ductor is not at right angles to the flux, but in-
stead makes an angle e with it, then the effective
length is the component l sine, and the force is
expressed by
F = BIZ sine. ~18.9)
Equation 18.8 states that the force on a wire
carrying a current and located in a magnetic field
is equal to the product of the flux density, the
current, and the length of the wire. With Bin
webers per square meter, I in amperes, and· Z in
meters, the force F is in newtons. The direction Frn. 18-8. Perspective view of
of the force relative to the flux and current is a conductor located at right
angles to the magnetic flux;
shown in Fig. 8, in which the quantities involved the force is exerted perpendicu-
are shown in perspective along mutually perpen- lar to both when current is es-
dicular axes. tablished in the conductor.
The fact that a force acts on a current-carrying
wire located in a magnetic field is utilized in many electrical machines and
instruments.
18-8. The Galvanometer. Most instruments for measuring electric current
depend for their operation on the force exerted on conductors within magnetic
fields. In the usual form, a coil of wire is supported in the field and comes
to rest when the torsion of its suspension wire, or restraining springs, is re-
lieved. When current is established in the coil the forces exerted on its con-
ductors by the field produce a deflection of the coil from its rest position, and
the amount of deflection serves as a measure of the current. This principle
applies to most ammeters and voltmeters used in electrical testing, and also
to the d' Arsonval galvanometer used for the measurement of small currents.
The galvanometer consists essentially of a movable coil placed between the
poles of a permanent magnet, as shown in Fig. 9. The current is conducted
to the coil by a flat metal ribbon which serves as the suspension, and is led
out by a flexible helix below the coil. The moving system carries a mirror M
so that its deflections can be read with a telescope and scale.
It will be assumed that there are N turns on the coil, that it is located in a
uniform field of flux density B, and that a current I circulates around the coil
in a clockwise direction. Each side of the coil, of length a, will experience
a force given by Eq. 18.8, amounting to
326 ELECTRICITY & MAGNETISM Chap. 18

F = BINa.
These forces act in opposite directions, but combine to produce a torque in a
common direction. The result is that the right side of the coil moves forward
(out of the page) and the left side moves backward (into the page). When
the coil is in the plane pictured in the figure, the deflecting torque will be

::sd = 2 BIN at= BINS,

where S = ab is the surface area of the coil face. As the coil turns in response
to this torque the lever arm becomes less than b/2, and ultimately the coil
would come to rest in a position at right angles to that shown; however, the

F
B

N s N •I Ix S

FIG. 18-9. A galvanometer coil within a, magnetic field shown in elevation at the left and
in plan view at the right.

suspension exerts a restoring torque ::Sr on the coil and brings it to rest before
that extreme position is reached. The value of this restoring torque is pro-
portional to the twist cf, of the suspension wire, or
3r = Tcf,,
where T is a constant determined by the stiffness of that wire. Obviously,
the rest position is reached where ::Id = ::Ir, and hence
Tcf, = BINS.
This expression shows that, for a given instrument (having fixed values of
B, N, S, and T), the twist cf, is proportional to the current I. With a parallel
field, as assumed in Fig. 9, this expression would be true only for small de-
flections.
In the galvanometer, as actually constructed, a stationary cylindrical core
is mounted within the moving coil and the pole pieces are usually curved; as
a result the magnetic field is radial, and as the coil swings it always moves
perpendicularly to the flux. By this means the twist is made proportional to
the current through a wide range.
§ 18-9 AMMETERS & YOL'1'1~,fETERS 327

Another form of this galvanometer, called a ballistic galvanometer, measures

electric charge rather than current. The instrument has a coil of considerable
width in order to give it a relatively high moment of inertia. When a small
quantity of electricity is discharged through such a galvanometer, a deflecting
torque is set up for a brief interval. The coil continues to deflect after the
discharge has been completed, and the maximum deflection or "throw" is
read. This deflection is proportional to the charge, and hence the instrument
can be calibrated to measure quantity of electricity.
18-9. Ammeters and Voltmeters. The instruments used for measuring
current and potential difference in commercial testing are galvanometers of a
portable type, in which the coil is pivoted and fitted with a pointer that moves
over a suitably calibrated scale. Current is supplied to the coil through flat
spiral springs; these serve the same purpose as the suspension wire of the
galvanometer in rendering the deflection proportional to the current, and in
returning the pointer to the zero position when the current ceases. There is
no structural difference between an ammeter and a voltmeter, and both deflect
because of current in the movable coil. They differ, however, in electrical
resistance; in this respect the instruments are
designed so that their introduction into a cir-
cuit for purposes of measurement will not
change appreciably the quantities they are
intended to measure.
In the ammeter, Fig. 10, the coil is provided
with a by-pass of fixed resistance, called a
shunt, connected directly across the terminals;
as a result the current in the coil will be only
a small but definite part of the entire current
in the circuit. In order that the ammeter may
have a small potential drop, its resistance is
purposely made small. Naturally, care must Terminals
be taken to have enough resistance elsewhere in Fm. 18-10. Basic parts of an
the circuit to limit the current to a value within ammeter.
the range of the instrument. An ammeter is
connected in series with a circuit, never across the service mains.
A voltmeter is designed to be connected directly across a generator or
service mains; it has a high resistance in order that it may take as little cur-
rent as possible. This result is obtained by connecting an appropriate re-
sistor in series with the coil. Because of its high resistance the voltmeter
provides its own protection; nevertheless care must be exercised not to apply
potential differences exceeding the range of the instrument. In order to
extend the range of a voltmeter, an auxiliary resistor, called a multiplier, is
connected in series with the instrument.
The resistance of the shunt or series resistance for a particular instrument
can be found by applying Ohm's Law, as the following examples will show.
:l28 ELECTRICITY & MAGNETISM Chap. 18

An ammeter has a coil of 5 ohms resistance, and it deflects across the entire scale
when carrying a current of 0.01 amp. Find the shunt resistance necessary to change
this instrument into a 15-amp ammeter, that is, one which deflects full scale when the
current being measured is 15 amp.
At full-scale deflection, the potential drop across the coil is V = I R = 0.01 X 5
= 0.05 volt, and this must also be the potential drop across the shunt. The current
in the shunt is 15 - 0.01 = 14.99 amp. Consequently, its resistance should be
R = V /I = 0.05/14.99 = 0.003336 ohm.
Again, a voltmeter has a coil of 5 ohms resistance, and it deflects full scale when
carrying a current of 0.01 amp. Compute the resistance that must be connected in
series with the coil if this instrument is to have a range of 150 volts-that is, to deflect
full scale when 150 volts are impressed across its terminals.
Since the current is the same in the coil and series resistance, the resistance of the
entire instrument is R = V /I = 150/0.01 = 15,000 ohms. Inasmuch as the coil has
a resistance of 5 ohms, the series resistance must have the value of 15,000 - 5
= 14,995 ohms.
18-10. Wattmeters and Watthour Meters. Since power is the product of
potential difference and current, § 17-8, an instrument for measuring power
must be arranged so that the deflection of the coil will be proportional to the
potential difference as well as to the current. The permanent magnet com-
mon to the voltmeter and the ammeter is replaced in the wattmeter by a heavy
stationary winding in series with the
load circuit; in this way the flux den-
sity B within this winding is made
proportional to the load current.
The coil that moves within this flux
is connected in series with a high
resistance across the circuit under
measurement, and the current I in
To line To load
this coil is thereby made propor-
Potential tional to the potential difference
coil
across the load. Since the force on
the coil is proportional to B X I,
Fm. 18-11. Connection diagram of a watt- according to Eq. 18.8, it will be pro-
meter. portional to the product of the am-
peres in the load and the volts across
it. Therefore, the instrument can be calibrated directly to read watts. The
arrangement is essentially that of the electrodynamometer mentioned in
§ 16-2; its connections are shown in Fig. 11. The "current winding" is indi-
cated by the fixed coils in series with the load, and the "potential winding"
by the movable coil bridged across the lines.
The metering of electrical energy for determining the cost of electric serv-
ice is not done with wattmeters, for these instruments indicate merely the in-
stantaneous rate at which the energy is being used; the meter employed for
this purpose is called a watthour meter. It is similar electrically to the watt-
meter but the coil is arranged to rotate continuously at a rate dependent
§ 18-12 ELECTRONIC MEASUREMENTS 329

upon the power expended in the load circuit, and the cost of service is based
upon the number of revolutions made.
18-11. Action of Magnetic Field on Moving Charge. It has been shown
that a current-carrying conductor located transversely in a magnetic field ex-
periences a force; its value is given by Eq. 18.8 as F = BIZ, where Bis the flux
density of the magnetic field, I the current in the conductor, and l the length
of the conductor. Since a moving charge constitutes a current, it is evident
that a force will act upon a charge while it is moving in a magnetic field. The
magnitude of this force can be found by considering a charge q to be trans-
ferred along a conductor of length l in the time t; the speed of transferring the
charge will be v = l/t, and the current I = q/t. If the values of l and I
hercfrom are substituted in the foregoing equation, the force upon the moving
r harge becomes
F = Bqv. (18.10)
The direction of this force is at right angles to both the motion of the charge
and the direction of the field. If the charge is not moving in a direction per-
pendicular to the magnetic flux but at an angle e with the flux instead, then
the force will be
F = Bq v sine .
•
These equations give the force in newtons when the flux density B is in
webers per square meter, the charge q in coulombs, and the speed v in meters
per second. The equations can also be used with electromagnetic or electro-
static units. In the emu system B is in gausses, q in abcoulombs, v in centi-
meters per second, and F in dynes. In the esu system B is in gausses, q in
statcoulombs, and v in centimeters per second, but the product of these quan-
tities must be divided by 3 X 1010 to obtain the force in dynes. I'he ad-
vantage of the mks units is the merger with the practical electrical units and
the simplicity of the more frequently used equations.
The sidewise thrust upon charged particles, such as electrons, protons
(hydrogen nuclei), and alpha particles (helium nuclei), is utilized in determin-
ing electronic mass and in accelerating the heavier particles for bombarding
atomic nuclei.
18-12. Electronic Measurements; Ratio of Charge to Mass. The prin-
ciples of electromagnetism and electrostatics enabled Sir Joseph J. Thomson
(1856-1940), a leading English scientist, to determine the ratio of the charge
of the electron to its mass. He used an evacuated tube of the form shown in
Fig. 12, which contains a cathode C and a tubular anode A near one end, and
a fluorescent screen S at the other. Near the middle of the tube are placed
two charged plates PP for establishing an electric field vertically directed,
and two coils MM for setting up a magnetic field horizontally directed.
When a potential difference is applied across the electrodes C and A, elec-
trons issue from C; while most of them fall upon the anode, a number pass
through the hole in it. In the absence of the electric and magnetic fields,
330 ELECTRICITY & MAGNETISM Chap. 18

these electrons would travel undeviated to the screen, where they would pro-
duce a bright spot at 0. With the plates charged as shown, each electron
passing between them would be subjected to an upward force of Se newtons
according to Eq. 15.3, where 8 is the electric field intensity between the plates
and e the charge of the electron. In consequence, the spot on the screen would
be displaced upward to some points. Instead, with the coils energized, each
electron would experience a vertical force of Bev newtons, according to Eq.
18.10, where B is the flux density of the magnetic field, and v is the electron

Fra. 18-12. Thomson's apparatus for making electronic measurements.

speed. If the current through the coils is directed as shown this force would
be downward, and the spot on the screen would be deflected to some point m.
Finally, with both the electric and magnetic fields acting simultaneously,
their strengths are adjusted so that the upward and downward forces on the
electrons become equal, causing the bright spot on the screen to return to
the undeflected position 0. When the forces balance, 8 e = Bev, from which
the electron speed can be determined as
0
v = -.
B
With a potential difference V between electrodes C and A, an electron of
mass m, in moving from one to the other, has an amount of work Ve done
upon it. This appears as kinetic energy when the electron leaves A, whence
ve = mv2, and the speed of the electron is
2

v ~2:e.
=
§ 18-13 THE CYCLOTRON 331

,v-hen the two equations for v are combined, the ratio of the charge c of the
electron to its mass mis found to be
e 82 (18.11)
m = 2 VB 2 •

The result is expressed in terms of the electric field intensity 8, the magnetic
flux density B, and the potential difference V between the electrodes. With
8 expressed in volts per meter, Vin volts, and Bin webers per square meter,
the ratio e/m will be in coulombs per kilogram .
. The numerical value of e/m, based on a number of recent measurements,
is 1.759 X 1011 coulomb/kg. When this result is combined with the value of
the eiectronic charge, e = 1.602 X 10-19 coulomb (§ 15-13), the mass of the
electron is found to be 9.107 X 10-31 kg.
It is customary to express the energy of high-speed particles in terms of the
potential difference through which an electron must accelerate in order to
acquire the particular energy value. The energy acquired by an electron in
accelerating through a potential difference of one volt is called an electron•volt.
Since the work done in transferring 1 coulomb of charge through a potential
difference of 1 volt is 1 joule, it follows that

1 electron-volt= 1.602 X 10-19 joule.

18-13. The Cyclotron. The application of electromagnetic principles to

the problems of modern research is illustrated in a machine developed by
Professor Ernest 0. Lawrence, of the University of California, and called the
cyclotron. The purpose of this apparatus is to give charged particles very
high speeds and a large amount of kinetic energy, so that they can be used for
the bombardment of atoms in the investigation of
nuclear structure and artificial radioactivity, § 35-11.
Essentially, the cyclotron consists of a pair of ,.,.,.- c-,,
hollow semicircular metal segments called dees, shown / /' -,, \ I l
I I 01 \ I I
at X and Yin Fig. 13, and at the center a source of I
I
I\.\
\"'I'_/ I/
1-'I ) , /

positive ions such as protons or alpha particles. \' ',-~ __,,. ,

'~ /
The space within and around the <lees is evacuated, ' ..... ,,~

and a uniform magnetic field is set up at right angles

to the plane of the segments. In operation, one of l<'rn. 18-13. Path of
the dees is charged positively and the other nega- pa1ticles in a cyclotron.
tively, and then the polarity of these charges is The magnetic field is
reversed again and again in rapid succession by con- perpendicular to the
page.
nection to an oscillator-a source of high-frequency
alternating current.
The positive ions are released at a and are attracted into dee X when it is
momentarily negative. Each ion is continously deflected by the magnetic
field, which is perpendicular to the page, and moves in a circular path until
it leaves that dee at b. At this instant the polarity of the dees is reversed
 ELECTRICITY & MAGNETISM Chap. 18

by the oscillator. The difference of potential that exists between them ac-
celerates the ion and causes it to enter dee Y at a higher speed, after which
it travels again in a circular path of larger radius until it emerges at c. By a
repetition of these events the ion moves along the dotted path, traveling each
half-revolution with higher speed, until it reaches its outermost orbit; then it
is deflected out of the magnetic field at Z.
It is of interest to note that the ions travel around any semicircle of the path
in the same time regardless of the radius. The force exerted by the magnetic
field of flux density B upon a charge q moving with a speed v is F = Bqv; this
force acts continuously at right angles to the direction of motion of the charge
and becomes the centripetal force of its circular motion. It may be expressed
by Eq. 4.9 as mv 2 /r, where mis the mass of the charged particle and r is the
radius of its path. Hence
mv 2
Bqv = -.
r
But the speed of a particle which moves around a semicircular path of radius r
in a time tis
1rr
v = -.
t
·when this value is substituted in the foregoing equation, the time interval
can be expressed as
1rm
t=-,
Bq
and is independent of r. Thus, as the ion moves, it completes each half-
revolution in the same time interval. The intensity of the magnetic field
and the reversal frequency of the electric field are adjusted· to values suitable
for the particular particles being accelerated. The apparatus is capable of
building up extremely high ionic speeds without the use of unduly high po-
tentials.
The kinetic energy possessed by a particle of mass m and charge q as it
leaves the cyclotron is

the result is expressed in joules when the mass is in kilograms, the charge in
coulombs, the radius of the largest ion path in meters, and the flux density in
webers per square meter.
As an illustration, consider a cyclotron to have <lees of 40 cm radius, through which
the flux density is 1 weber/m2, and that the machine is used to accelerate hydrogen
nuclei. Determine the energy of the issuing proton beam.
Each proton has a mass of 1.672 X 10-27 kg; the radius of the outermost path
. . . . (1.602 x 10-19 x 0.4 x l)~
IS 0.4 m. Hence the kmetic energy of the proton Is Ek = 2 X 1. 672 X 10_27

=l.228X10-12 i'oulc. 11 . . . l
'rn,18eqmvaentto 1. 228 x ,o-12 = "'('"'
l.i, X 10. 6 cccron·vo
1 t 1ts.
l.CO~ X 10-]!I
 PROBLEMS 333

PROBLEMS
1. A flat circular coil of 40 turns that have a radius of 8 cm is placed flat on a
table top, and a current of 3 amp is set up in the winding in a clockwise direction as
viewed from above. Determine the magnetic flux density produced by the current:
(a) at the center of the coil, and (b) at a point along the axis of the coil and 6 cm
from its plane.
2. To produce a uniform 1nagnetic field over a limited region it is customary to
use two large circular coils mounted with their axes aligned and separated by a dis-
tance equal to their common radius. Derive an expression for the flux density at the
axial point midway between the coils when the same current traverses both coils
in the same direction.
3. A solenoid is 0.8 m long and 5 cm in diameter; its winding contains 720 turns
closely wound over its entire length. (a) What magnetic flux density is produced at
the center of the solenoid when 50 milliamp are passed through the winding? (b) If
there were only one turn of the same size, what density would the same current pro-
duce at the center of that turn?
4. In the solenoid of Prob. 3, what would be the flux density on its axis at the
·endpoints of the winding, assuming the current as unchanged?
5. Two long parallel wires are 1 ft apart and each carries 250 amp. Find the mag-
nitude of the flux density at either wire due to the current in the other. What force
do these wires exert upon each other over a 10-ft length?
6. A straight wire, located within a magnetic field directed from left to right,
carries a current of 60 amp directed from A to Bas shown in the diagram. The field
extends over an area 20 cm high and 30 cm broad, and the flux totals 0.6 weber and
is uniform. Determine the force on the wire due to the magnetic field, both in magni-
tude and direction.
B

A 5 25 +
PROBLEM 6. PROBLEM 11.

7. A wire bent into the shape of a hairpin that measures 16 cm from side to side
is lowered into the magnetic field of Prob. 6 until its circular portion is well within
the flux, while its sides are kept vertical. If the plane of the hairpin is perpendicular
to the flux, what force is exerted on the wire when it carries a current of 25 amp?
8. A galvanometer coil 3 cm long and 1.2 cm wide, wound with 200 turns, is sup-
ported as shown in Fig. 9. The magnetic field has a density of 0.1 weber/m2 and the
suspension requires a torque of 8 X 10-9 newton· m to twist it through 1 radian. What
steady current maintained in the coil will cause it to deflect through an angle of 1°?
9. A particular galvanometer produces a deflection of one division on its scale when
25 microvolts are impressed upon it. The resistance of the instrument is 400 ohms.
What deflection will the galvanometer produce when it is connected in series with a
dry cell having an emf of 1.5 volts and a resistor having a resistance of 1 megohm?
10. The moving coil of an instrument has a resistance of 5.0 ohms and yields a full-
scale deflection with a current of 10 milliamp. Compute the resistance of a shunt
334 ELECTRICITY & MAGNETISM Chap. 18

for the coil which would cause this instrument to deflect full-scale with a current of
2.5 amp.
11. The coil of Prob. 10 is to be used in forming a two-range ammeter with shunts
as connected in the accompanying diagram. What should be the resistances of shunts
R1 and R2 for a meter which can read up to 5 amp on one scale and to 25 amp on
the other, depending upon the terminals used?
12. A millivoltmeter of 1.25 ohms resistance yields full-scale deflection when 50
millivolts are impressed across it. What must be the resistance of an external shunt
to convert this millivoltmeter into a 10-amp ammeter?
13. Two resistors, one having a resistance of 1100 ohms and the other of 500 ohms,
are connected in series across 120-volt constant-potential service mains. ·when a
voltmeter is connected across the 1100-ohm resistor it reads 80 volts; what is the
resistance of the meter?
14. A meter movement has a resistance of 10 ohms and produces a full-scale deflec-
tion when the current in it is 0.005 amp. (a) Find the shunt resistance necessary to
make this movement serve for a 5-amp ammeter. (b) Find the resistance of a series
resistor that is needed to make this movement serve for a 150-volt voltmeter.
15. Thomson's apparatus, Fig. 12, is used in a test without a magnetic field. When
a charged particle is projected horizontally into the electric field, it experiences a
vertical force and travels along a parabola. Show that a charged particle (having
mass m and charge q) while traveling horizontally a distance l in the electric field will
be displaced vertically by an amount s = :~, where V is the accelerating potential
in volts, 8 is the intensity of the electric field in volts per meter, and l and s are in
meters.
16. An electron that is traveling at a speed of 5 X 106 m/sec enters perpendicularly
into a magnetic field in which the flux density is 0.001 weber/m2 • Determine the
radius of its path.
17. How many reversals per second should be made in the electric field between the
dees of the cyclotron mentioned in § 18-13 in order to accelerate protons? How
many reversals to accelerate helium nuclei?
18. The diagram shows a convenient arrangement for measuring the charge-to-
mass ratio of ions. The ions emitted by source Care accelerated to electrode A and,
after passing through its aperture, follow
a straight path with uniform speed until
they go perpendicularly into a magnetic
field that is localized in the triangular area.
The ions are deflected around a circular
path within the field and then proceed in
PROBLEM 18. a straight line to the detecting device D.
Show that the charge-to-mass ratio of the
ions is equal to 8 V/ B 2d2, where V is the accelerating potential, d is half the distance
from A to D, and B is the flux density perpendicular to the page.
19. A cyclotron is used to accelerate deuterons; these have the same charge as
protons but twice their mass. What value should the flux density have if the polarity
of the dees is reversed 14 million times per second?
 19
MAGNETS AND MAGNETIC
CIRCUITS
Magnetic fields for most purposes are developed through electric circuits
and magnetic materials. Of these materials, iron and steel are the most
important, although many other magnetic substances have been introduced.
The basic properties of the materials used for electromagnets and permanent
magnets are considered in this chapter. Included also are the calculations of
magnetic circuits and some facts about the earth's magnetism.
19-1. Magnetic Intensity. In the foregoing chapter were derived a num-
ber of expressions for the magnetic induction or flux density produced by
currents in simple forms of circuits located in nonmagnetic surroundings. To
study the effect of other mediums on the flux density, consideration will be
given t9 an electric circuit in the form of a ring solenoid, § 18-5, for the mag-
netic flux is practically uniform through the turns of the winding. When the
space enclosed by the winding is a vacuum, and when the winding has a total
of N turns and the flux path a length l, then, as shown before, a current I in
the winding establishes a flux of density
NI
(18.6) B = µo - l •

where µ 0 is the permeability of free space and has the value 4 1r 10-1 and the
umt. newton or
O
weber
t . a1so su b st an t.1a11y th e va1ue for air.
; t h"1s 1s .
amp- amp· urn·m
The method of measuring flux density utilizes the principle of electromag-
netic induction which is considered in the next chapter, but the experimental
details are given here. The toroid is equipped with two windings as shown in
Fig. 1; the primary winding is connected through a rheostat and a switch to a
battery, and the secondary winding is connected directly to a ballistic gal-
vanometer. When the switch is closed the current from the battery estab-
lishes magnetic flux through the toroid, the amount depending upon the
current as adjusted by the rheostat. Upon opening the switch, the current
ceases, and the magnetic flux falls quickly to zero. In doing so, the changing
flux "induces" an emf in the secondary winding which causes the galvanom-
eter to deflect, and that instrument can be calibrated so that its throw indi-
335
336 ELECTRICITY & MAGNETISM Chap. 19

cates the change of flux density through the toroid. When the test is made
with vacuum or air as the toroid, the measured value of the flux density
agrees with the value obtained for B from the equation above. But when
the toroid is formed of iron or other magnetic material, without any change
in the winding or current, the test value
Galvanometer
for the flux density is found to be vastly
Secondary
greater.
This increase of flux density within cores
of magnetic materials is attributed to the
motion of the electrons within the atoms
of these materials, for it is known that a
moving electron constitutes an electric
current and that a current produces a
~rimary magnetic effect. On the basis of spec-
.
FIG. 19-1. T est connect10ns f trum analysis and in keeping with the
or meas-
uring flux density. planetary picture of the atom, §§ 15-2,
31-3 and 34-5, it is believed that the elec-
trons not only revolve about the atomic nucleus but also spin about axes
through their own centers. Further, it is believed that in magnetic sub-
stances each atom has more electrons spinning in one direction than in the
other.
The increment in flux density brought about by electronic motion within a
magnetic material depends upon the external field to which it is subjected.
If the material forms the toroid of a ring solenoid, this field is produced by the
current in the winding, and the increment can be expressed as
NI
B = x-,
l
where x (chi) represents the magnetic susceptibility of the material. The total
flux density produced in the material is obtained by merging this result with
Eq. 18.6, hence
NI
B = (µo + x) - l · (19.1)

For a vacuum x is zero, and the equation reduces to its initial form.
It is convenient to introduce another magnetic quantity, a vector called
magnetic intensity, and symbolized by H. Its definition for a vacuum is the
ratio of B to µo, that is,
B (19.2)
H=-·
µo

A general interpretation of the meaning of H within matter is quite involved

and depends greatly upon the shape of the material. For a toroid the field
intensity can be expressed as the number of ampere· turns of the winding per
unit length of flux path, that is,
§ 19-3 },1AGNETIC SUBS'PANCES 337

NI (19.3)
H=-·
l
The unit of H in the mks system, the amp· turn/m, is obvious from this ex-
pression; it can also be found from Eq. 19.2, since the unit of flux density B
is the weber/m2, and that for permeability µo is the weber/(amp·turn·m),
§ 18-3. The magnetic intensity in a region can be represented by lines of
magnetic force, similar to those representing flux density.
19-2. Permeability. The magnetic character of a material can be expressed
in terms of the flux density and the magnetic intensity within it. These
quantities are given in the preceding section for a toroid by Eqs. 19.1 and 19.3,
and a merger of them results in
B = (µo + x) H;
this can be shortened to
B = µH, (19.4)
where µ represents the sum of µo and x and is called the permeability of the
material. Its unit is the same as that for the permeability of a vacuum, the
weber/(amp·turn ·m).
The permeabilities of substances are frequently compared with the perme-
ability of free space (vacuum), and a term expressing the permeability of
a substance with respect to that of vacuum is of great convenience. The
relative permeability µr is defined by the relation
µ
µr = - , (19.5)
µo

where µ is the permeability of the substance, and µ 0 the permeability of a

vacuum. Since both of these quantities have the same unit, the relative per-
meability of a substance is a pure number. For air and other nonmagnetic
substances µr is practically 1, while for magnetic materials its value extends
into the hundreds and the thousands, and moreover depends on the flux
density.
19-3. Magnetic Substances. Substances when classified as to their mag-
netic behavior form three groups, according to the values of permeability.
'fhe relative permeability µ, is less than unity for diamagnetic substances,
greater than unity for paramagnetic substances, and very much larger for
ferromagnetic substances. The values of µr for all known diamagnetic sub-
stances are but slightly less than unity; bismuth has the lowest known value,
0.99998. The values of µr for paramagnetic substances are but slightly greater
than unity; for example, platinum has a value of 1.00002. Iron, nickel, and
cobalt are the ferromagnetic elements, and a number of alloys are also fer-
romagnetic.
Table I lists a number of ferromagnetic materials with their maximum
values of relative permeability and the corresponding flux densities. Inas-
338 ELECTRICITY & MAGNETISM Chap. 19

much as definite figures depend upon the purity of the substance, the method
of preparation, and the thermal treatment, the values listed are only approxi-
mate.

TABLE I. RELATIVE PERMEABILITIES OF MAGNETIC MATERIALS

µ, B
Maximum Value webers/m2

Cobalt ..................................... . 170 0.3

Heusler alloy (Cu 60%, Mn 24%, Al 1GS{) ..... . 200 0.2
Iron-cobalt alloy (Co 34%) ................... . 1:3,000 0.8
Iron, purest commercial annealed ............. . GOOO to 8000 0.6
Nickel .................................... . 400 to 1000 0.1 to 0.:1
Permalloy (Ni 78.5%, Fe 21.5%) .............. . over 80,000 0.5
Perminvar (Ni 45%, Fe 30%, Co 25%) ........ . 2000 0.0004
Silicon steel (Si 4%) ......................... . 5000 to 10,000 0.6 to 0.8
Steel, cast .................................. . 1500 0.7
Steel, open-hearth ........................... . 3000 to 7000 0.6

The Heusler alloys were the first magnetic alloys formed of elements that
are themselves nonmagnetic. In contrast, there are varieties of manganese
steel which are practically nonmagnetic; in these, the carbon content is about
1 per cent, and the manganese varies between 10 and 15 per cent. Iron
begins to lose its magnetic properties as the temperature is raised, and ceases
to be ferromagnetic at 770°0.
The magnetization of ferromagnetic substances is accompanied by slight
changes in size, an effect called magnetostriction. Changes in the length of
steel and iron rods caused by magnetostriction are utilized for the production
of ultrasonic waves, § 26-19.
19-4. The Magnetization Curve. The permeability of a ferromagnetic sub-
stance does not have a fixed value, even for a particular specimen, but de-
pends upon the previous magnetic history of the specimen and the flux density
in it. The dependence of relative permeability µr upon magnetic intensity H
is usually shown indirectly by so-ealled magnetization curves, which coordinate
the flux density B with the magnetic intensity H.
Figure 2 shm,vs a typical magnetization curve for silicon steel, a material
that is widely used in electrical apparatus and machinery. For low magnetic
intensities, the flux density B increases in direct proportion to H, but for high
intensities the specimen becomes "saturated" with flux, and a large increase
of H causes only a slight increase of B. In order-to find the permeability of
the steel for a given magnetic intensity, the corresponding value of the flux
density is read from the curve, and the values of B and Hare substituted in
Eq. 19.4. Thus at magnetic intensity H = 1600 amp· turn/m, silicon steel
has a flux density B = 1.45 webers/m2, and a permeability µ = B/ H =
§ 19-4 THE MAGNETIZATION CURVE 339

1.45/1600 = 0.00091 weber/(amp·turn·m). Its relative permeability is

µr = µ/ µo = 0.00091/ (L_l: 7r 10-7) = 725.
A magnetization curve can be obtained for a sample of iron or other mag-
netic material by preparing a specimen in the form of a toroid, and winding
upon it a coil for magnetizing the specimen and another coil joined to a bal-
listic galvanometer for measuring the flux produced, as explained in § 19-1.
The iron must be demagnetized before the test is started. Then, with the
rheostat adjusted to give a small current, the primary circuit is closed and the
maximum deflection of the galvanometer is observed. The primary current
is then increased in small steps, without opening the primary circuit, and the
galvanometer throw is read for each current increment. This process is con-

I
c
I
I
I
I

l B
Q) 1.2 ,___,___ _ _ _--,
<I)
I
I
I
I
I
..!)
I
f
:i, 0.8 l - l - - - + - - - - - 1 - - - - l
H
·;;;
c
~ 0.4 H - - - + - - - - 1 - - - - - l
x
::>
ti: , . _ __ _ J __ _- - ' - - - - - '
H
0 800 1600 2400 C'
Mmf, amp·turns/m
Frn. 19-2. Magnetization Frn. 19-3. Hysteresis loop for a
curve of silicon steel. sample of iron.

tirmed until the desired maximum field intensity is reached. The values of
the magnetic intensity H inside the specimen are computed from the primary
currents by Eq. 19.2, and those of the flux density B are computed from the
throws of the galvanometer. From these values the magnetizing curve
for the iron can be plotted.
If, after magnetizing the iron ring and plotting its magnetization curve,
the current in the winding is decreased step by step, then it will be found that
the curve representing demagnetization will not coincide with the fir1-,t curve.
The result is indicated in Fig. 3, which depicts the demagnetization curve CBr
intercepting the B axis above the origin 0. This shows that when the current
has been reduced to zero (making H = 0) there will still be flux in the ring,
and the flux density for this res,idual magnetism ,vill be B,.. For the ring to be
completely demagnetized the current would have to be reversed and then
increased in strength sufficiently to produce a field intensity He, an amount
called the coercive force.
340 ELECTRICITY & MAGNETISM Chap. 19

When the magnetization is increased further in the reversed direction and

the current is allowed to reach the same maximum value as before, the curve
representing the behavior of the iron is extended to C' in the figure. There-
after the current is reduced to zero, then its direction is reversed, and finally
it is increased again to the initial maximum value. The complete mag-
netization cycle for the sample under test is indicated by the closed curve CC',
called a hysteresis loop. It can be shown that the area of such a loop is pro-
portional to energy per unit volume in changing the magnetization; this
statement can be verified by multiplying the units of the two coordinates
B and H-thus, (webers/m2) X (amp·turn/m) = weber·amp·turns/m 3 =
volt·amp·sec/m3 (Eq. 20.2) = joules/m3 (Eq. 17.5).
If alternating current were used to magnetize the sample, the energy repre-
sented by one loop would be wasted in each cycle of current values. For a
lesser alternating current the loop would be smaller, as shown by the dotted
curve, and there would be less waste. Charles P. Steinmetz (1865-1923), an
American electrical engineer, showed that the waste of energy due to hysteresis
is proportional to the 1.6 power of the maximum flux density attained during
the current cycle.
In electrical machinery, use is made of soft iron or silicon steel, for these
materials have low hysteresis and high permeability under operating con-
ditions. On the other hand, permanent magnets are made of materials that
have a high residual magnetism and high coercive force.
19-5. Permanent Magnets. Many magnetic materials have been used
over the years for permanent magnets. The basic material is iron, to which
various percentages of other elements are added. These elements include
aluminum, carbon, chromium, cobalt, copper, manganese, molybdenum,
nickel, and tungsten. Several nonferrous alloys are magnetic. The appro-
priate forming and machining operations and the proper heat treatment of
suitable compositions of the alloying elements, within or without magnetic
fields, are important factors in the manufacture of permanent magnets.
The evaluation of material for permanent magnets is based upon the shape
and size of the hysteresis loop, and specifically upon the form and position of
the demagnetization portion of the loop-the part in the second quadrant.
The reason for forming a judgment on this basis is the fact that a permanent
magnet when in service is subject to demagnetizing influenees; these are due
chiefly to the presence of an air gap in its magnetic circuit, and also to heating,
mechanical vibration, and stray magnetic fields. As these influences change,
the magnet operates along the demagnetization portion of the hysteresis loop.
This portion of a typical loop is shown in part I of Fig. 4, with magnetic
intensity H as abscissas and flux density B as ordinates. Since the magnet
operates by itself (that is, without aid from the source which brought about
the magnetization), it always functions at some point along the curve between
the point of residual flux density Br and the point of maximum coercive
force He.
§ 19-6 SIMILARITY OF ELECTRIC AND MAGNETIC CIRCUITS 341

It was pointed out in the preceding section that area on a B-H diagram
represents energy; the area under the curve in the second quadrant represents
energy stored in the field of a magnetic substance. Since this area depends
upon the magnitudes of both H and B, the energy will depend upon their
product. This "energy product" for a material indicates its suitability for
use as a permanent magnet. Part II of Fig. 4 shows a graph of such products
for the magnet material represented by the loop in part I; a high value for
the maximum product, (BH)max, is the best single criterion of good material
for permanent magnets.

8 8

----
N
8, "'
'$
11.2
..a

i-
(1)

(1)

~
0.8 I
v (1)
..0
(I)
~
i- 0.8
1.2 1------'1'""""-=:---:-f----l

>----+--,t"---+--+----1
·;;;

"O
i:
(1) I ·;;;
i:
(1)
-u
~ 0.4 ~ 0.4 ! - - - - - + + - - - + - - - ~
u:
He
6xl0 4 4 2 0 0 2 4 6x10 4
Field intensity H Energy product BH
amp·turns/m joules/m 3
I JI
Frn. 19-4. Magnetic behavior of Alnico V: I, demagnetization curve; II, magnetic energy
curve.

An iron alloy called Alnico V (Al 8%, Ni 14%, Co 24%, Cu 3%), when
cooled in a magnetic field, has a very high energy product-40,000 joules/m3 •
Lately permanent magnets have been made from ultra-fine iron powder and
have yielded values for (BH)max surprisingly high for this technique-8500
joules/m3 •
Permanent magnets are used to hold objects in machines or attract them
in magnetic ore separators, to produce a magnetic drag in watthour meters,
to develop emf in light-weight generators such as magnetos and phonograph
pick-ups, to reproduce music or speech in telephone receivers and radio loud-
speakers, and to establish constant flux in a variety of electrical measuring
instruments.
19-6. Similarity of Electric and Magnetic Circuits. The fact that mag-
netic flux assumes the form of closed loops accounts for the application of the
term magnetic circuit to the path that the flux loops follow. There are many
points of similarity between electric circuits carrying current and magnetic
circuits carrying flux, and a counterpart of Ohm's Law can be set up which is
helpful in 1mlving magnetic circuits.
342 ELECTRICITY & MAGNETISM Chap. 19

Reference again to the ring solenoid of § 19-1 shows that the magnetic
intensity in the toroidal core is
H =NI,
l
where N is the numher of turrn; in the magnetizing winding, I the current in
it, and l the length of the core. The magnetic flux through the core may be
taken as uniformly distributed over its cross-sectional area, and hence the
flux density has the uniform value of
<I>
(18.1) B = A'

where <I> is the total amount of flux, and A is the sectional area of the core.
The relationship between the two equations, expressed by Eq. 19.4, is
B = µH,
whereµ is the permeability of the core under the existing magnetization. A
combination of these equations, with rearrangements, yields
NI NI
<I> = BA = µHA = µA -l = l/ µA

A comparison of this result with Ohm's Law for a metallic circuit, namely
I = E/ R, reveals the analogy between electric and magnetic circuits. The
electromotive force E in the electric circuit is replaced by the product NI;
it is called the m.agnetom.otive force (mmf) in the magnetic circuit. The
resistance R of the electric circuit is replaced by the fraction Z/µA; it is called
the reluctance of the magnetic circuit. The value of reluctance of a particular
circuit depends upon its material and dimensions in much the same way that
resistance of a circuit depends upon these factors. The symbol for mmf is IT
and that for reluctance is CR; hence the expression for the flux produced in a
magnetic circuit can be written as

q:, = .'.!... (19.6)

(R

In the mks system, the unit for mmf is obviously th ampere· turn from the
equation IT = NI, and the unit for reluctance from the equation CR = l/ Aµ
is the meter/(meter2 X w;ber t ) = ampere ·turn/weber. A sub-
ampere · urn· me er
stitution of these units in Eq. HJ.6 will show that they are consistent. A
reluctance of 1 ampere· turn per weber represents the reluctance of a path
1 meter long and 1 square meter in cross-section within a medium of unit
permeability. A vacuum having these dimensions would have a reluctance
of 107/(4 1r) ampere·turns/weber. A mmf of 1 ampere·turn would establish
a magnetic flux of 1 weber in a circuit of unit reluctance.
It may be of interest to mention that in the rgs system the unit of magnetic
 § 19-7 MAGNETIC CIRCUIT CALCULATIONS 343

intensity is called the oersted, after the discoverer of electromagnetism, and

that the unit of mmf is called the gilbert, after the English physicist William
Gilbert (1540-1603). One oersted = 79.6 ampere·turns per meter, and 1 gil-
bert = 0. 796 ampere· turn. The gilbert is the mmf that would establish a
magnetic flux of 1 maxwell in a path 1 centimeter long and 1 square centimeter
in cross-section within a vacuum.
19-7. Magnetic Circuit Calculations. The parallelism between electric and
magnetic circuits is illustrated in Fig. 5. At the left is shown an electric
'circuit of resistance R, in which a battery of emf E sets up a current I. At
the right is shown a toroidal core of reluctance CR, surrounded by a coil
of wire carrying a current; this magnetizing agency sets up a magnetomotive
force 5 to establish a flux cp in the toroid. The circuits are compared in
parallel columns:
Ewctric Nlagnetic
emf ~"l ux mmf
Current = --.- - .r =
resistance reluctance
E ;}
l=- cp = --
R CR
volts ampere· turns
amperes= - - we b ers =
ohms ampere· turns per weber
Resistance of a wire of length Z, of sec- Reluctance of a flux path of length Z, of
tional area A, and of resistivity p is sectional area A, and of permeabilityµ is
pl CR = _l . (19.7)
(17.3) R = -.
A µA
Although the parallelism between the two circuits is seemingly exact, there
are important differences that make the calculation of magnetic circuits less
direct. In the electric circuit, the re-
sistance does not usually depend upon
the current, except insofar as it is in-
fluenced by heating. In the magnetic
circuit, on the other hand, the reluc-
tance depends upon the magnetic flux,
because the permeability is not constant.
Also, in the electric circuit, the current Fw. 19-5. Electric and magnetic cir-
is practically confined to the conduc- cuits compared.
tors, whereas in the magnetic circuit
there is usually some magnetic leakage through the surrounding medium,
because no material is known that insulates magnetic flux.
The derivation of Eq. 19.6 is based on a ring solenoid in which the magnetic
flux is confined to a path through the core. It may be applied to other shapes
if magnetic leakage is negligible. For circuits containing a magnetic material,
it is necessary to compute the magnetic intensity H, given by Eq. 19.3 as
344 ELECTRICITY & MAGNETISM Chap. 19

NI/l, and to refer to the magnetization curve of the material in order to

determine its permeability under given conditions of magnetization.
Suppose an inverted U-shaped electromagnet with its contacting member or arma-
ture, Fig. 6, to have a core length (indicated by the dotted line) of 34 cm, and a core
cross-section of 8 cm2 • The winding consists of 500 turns
and the current in it is 1.5 amp. Assume Fig. 2 to repre-
sent the magnetization curve of the iron, and determine the
reluctance of the core and the amount of flux established
N
in it.
The mmf of the winding is g: = NI = 500 X 1.5 = 750
_____
.... ;
amp· turns, and the magnetic intensity in the iron is
H = NlI = 750 = 2200 amp· turns/m. The magnetization
0.34
Fw. 19-6. Electromag- curve indicates that this intensity develops a flux density of
net with armature forms B = 1.49 webers/m 2 ; hence the permeability of the core
aclosedmagneticcircuit. at this magnetization is µ = B/H = 1.49/2200 = 6.77 X
10-4 weber/(arnp·turn·m). Consequently, the reluctance
of the core is CR = ZA = 6 77
µ .
°· 34
x 10-4 x 8 x 10- 4 = 6.28 X 10 amp· turns/weber,
5

and the magnetic flux is <I> = g: /CR = 750/(6.28 X 105) = 1.194 X 10-3 weber. As
a check on the latter result, the total flux is also given by Eq. 18.1 as <I> = BA =
1.49 X 8 X 10-4 = 1.194 X 10-3 weber.

If the flux path is made up of two or more successive sections of different

size or material, the reluctance is computed for each section separately and
the sum of the results is used for CR in Eq. 19.6 to determine the flux produced
by a given magnetomotive force.
19-8. Domain Theory of Magnetism. The electrical origin of magnetism,
alluded to in § 19-1, has been studied for many years. Recent developments
show that a ferromagnetic substance is made up of a multitude of regions of
microscopic size called domains, and that each domain consists of many billions
of atoms so aligned that the magnetic fields produced by their spinning elec-
trons are parallel. When the substance is unmagnetized, the direction of the
field of one such saturated domain differs from that of the next, and with
the domains in a variety of directions the net magnetic field of even a small
piece of the substance is zero.
When the substance is located within an external magnetic field its behavior
depends upon the intensity of that field. In a weak field, the magnetic direc-
tions of the domains turn more nearly toward the direction of the external
field, and the domain boundaries themselves shift, causing the domains in
which the magnetic direction lies close to the direction of the external field to
become larger while those more inclined become smaller. In a stronger field,
entire domains shift suddenly through large angles in tending to align their
magnetic directions with the external field. In very strong fields the magnetic
directions of the domains turn farther to become parallel with the external
field until the entire sample becomes saturated. The behavior of the domains
in the magnetization process is pictured in Fig. 7.
§ 19-9 ACTION OF A BAR MAGNET IN A MAGNETIC FIELD 345

The stages of magnetization described merge into each other and account
for the shape of the magnetization curve, such as that shown in Fig. 2. The
initial rapid rise of the curve corresponds to the movement of the domain
boundaries and the sudden rotation of the domains; the later sloping of the
curve toward the horizontal shows the approach to saturation.
It has been found that the magnetization of ferromagnetic crystals with
cubic symmetry, like iron, is due almost entirely to electron spin; for example,
the iron atom has four more electrons that spin one way than spin the other.
Ferromagnetism is normally associated with atoms that have unfilled inner
shells of electrons.

'
-- -
Unmagnetized Shift of domain Elimination Alignment with
state boundaries of unfavorably external field
oriented domains

Fw. 19-7. Stages of the magnetization process.

19-9. Action of a Bar Magnet in a Magnetic Field. A bar magnet differs

from the magnetized iron toroid considered thus far in that the magnetic flux
leaves the magnet over one portion of its surface and enters at another. If
the bar is slender, like a compass needle, practically all the flux is directed
into or out of the magnet at its ends, and these regions behave as magnetic
poles. The strength of the poles depends upon the degree of magnetization-
that is, upon the number of uncompensated electron spins which remain in
alignment after the bar has been magnetized.
When a oar magnet is placed in an external magnetic field it will turn until
it becomes parallel to the field; this it does for the same reason that a current-
carrying coil will turn in an external field until its axis becomes parallel with
the field. The torque acting on such a coil is expressed in § 18-8 as
J = BINS,
where Bis the flux density of the field, I the current in the coil, N the number
of turns in the coil, and S the surface area of the coil face. Hence, if the
moving electrons within the atoms of the bar magnet are assumed individu-
ally to represent a current i in a loop enclosing an area a, and 1£ there are n
such loops in unit volume of the magnet, then the torque on a magnet of
length l and cross-sectional area A is
J = Bi(nlA)a,
where the flux density B of the external field in air is equivalent to the product
of the magnetic intensity H and the permeability µ 0 •
346 ELECTRICITY & MAGNETISM Chap. 19

The group of terms represented by µoina is designated as the intensity of

magnetization, and the product of this quantity and the area A of the end face
of the magnet is designated the pole strength of the magnet. When the pole
strength is symbolized by m, the torque exerted on the bar magnet to align
itself with the field becomes
3 = Hml. (19.8)

It can be shown from the foregoing that the pole strength may be expressed
in webers. If H is in ampere· turns per meter, m in webers, and l in meters,
then the torque will be in newton· meters.
The product of the pole strength m of a magnet and the distance l between
its poles is called the magnetic moment (M) of the magnet; this is a property
of a magnet that can be measured directly.
19-10. The Earth's Magnetic Field. The fact, mentioned in § 18-1, that
a suspended magnet orients itself in a particular direction at every point on
or near the earth, shows that the earth is surrounded by a magnetic field.
The distribution of the field is such as might be produced roughly by a huge
bar magnet within the earth, located about 17° away from its axis and having
a length much less than the earth's diameter. The two places at the earth's
surface where the field is vertical are called its magnetic poles. The region
of the magnetic pole in the northern hemisphere was long believed to be on
Boothia Peninsula in Northern Canada (latitude 70°N, longitude 96°W), but
explorations in 1947 indicated that this region possesses three poles, the major
one on Prince of Wales Island and the others on Boothia Peninsula and
Bathurst Island. The magnetic pole in the southern hemisphere is near Ross
Sea, Antarctica (latitude 72°S, longitude 157°E). The magnetic equator is
a line on the earth's surface connecting points where the field is horizontal;
the line is irregular and varies in latitude from 16°S in South America to
11 °N in Africa.
The fact that a pivoted magnet aligns itself with the earth's magnetic field
makes possible its use in the form of a compass for guidance in navigation.
Such use dates back a very long time, for it appears that pieces of lodestone
were suspended for this purpose by the Chinese about ten centuries B.c. In
the modern magnetic compass the magnets and the scale are supported by a
spheroidal air vessel that floats in a mixture of water and alcohol.
The compass points along the direction of the earth's magnetic field, but
since the magnetic polar region in the northern hemisphere is far away
(1400 mi) from the north geographic pole, the compass does not, in general,
point true north. The angle that it makes with the geographic meridian at
any place on the earth's surface is called the magnetic declination or variation
of the compass at that location.
The magnetic variation at any place does not remain the same year after
year, but changes slightly over long periods of time. Besides these so-called
secular cha11gcs, there arc fluctuations within the year and also changes of
§ 19-11 THE COMPASS AND THE MAGNETOMETER 347

small extent throughout the day. Large erratic irregularities occur during
"magnetic storms"; these are often concurrent with the appearance of sun-
spots. Much work has been done in attempting to explain terrestrial mag-
netism and to account for its variations, but too little is known at present
about the magnetic sources within the earth and about atmospheric currents
of electricity to establish a satisfying theory of the earth's magnetism.
The direction of the earth's magnetic field is not horizontal (except at the
magnetic equator), as can readily be observed by balancing a magnetized
needle on a horizontal spindle, and placing it so that its vertical plane of
movement includes the direction of the field. The horizontal and vertical
components of the earth's field can be expressed
in terms of the magnetic intensity and the angle True North
of dip f3 as I
Hh = H cos /3 and H 0 = II sin /3,
respectively, where His the total intensity of the
field. The horizontal component is most fre-
quently used in magnetic measurements; its value
in the United States ranges from 10.4 ampere·
turns per meter at Gull Island in Lake Superior
to 22.3 like units near Brownville, Texas. These
values correspond to 0.13 and 0.28 oersted, re- I\
spectively, § 19.6. I \
I \
I \
At New York City the magnetic declination is I \
I \
12° west and the dip is 72° down from the hori-
zontal. These angles are shown in perspective in Frn. 19-8. Magnetic declin-
Fig. 8. The earth's magnetic intensity is H = ation and dip at New York
45 ampere· turns per meter at this location; its City.
components are indicated by Hh and llv.
19-11. The Compass and the Magnetometer. The direction of a magnetic
field can be measured by observing the rest position assumed by a compass
needle. When disturbed from that position, the needle will oscillate, but will
eventually resume the same position. The torque that causes the needle to
return to the rest position depends upon the magnetic moment of the needle,
the magnetic intensity in the region where the compass is placed, and the
direction of the needle with respect to the field. If M represents the magnetic
moment of the needle, H the magnetic intensity of the field, and 8 the angle of
displacement from the rest position, the restoring torque is found to be
~ = MH sine.
This torque is such as to tend to lessen the magnitude of angle e, and the
needle, because of its inertia, oscillates until it comes to rest in the direction
of the field.
The torque varies from zero when the needle is aligned with the field to a
348 ELECTRICITY & MAGNETISM Chap. 19

maximum~ = MH when the needle is at right angles to the field. The latter
form of the equation makes it possible to define the magnetic moment of the
needle as the torque per unit field intensity when the angle between the needle
and the field is 90°.
Measurements of field intensity can be made with a magnetometer, an instru-
ment consisting of a small compass pivoted to swing over a horizontal scale.
The field to be measured is arranged at right angles to another that is takep.
as a standard; the latter may be the earth's field if its'
value is definitely known, or else may be produced by
R </> Hh an electric current under known circuit conditions. Thus,
to measure the magnetic field X shown in the plan view,
-----"*p Fig. 9, this field is arranged horizontally at right angles to
X the earth's field of known horizontal intensity Hh. The
Fm. 19-9. Vector magnetometer is placed at P; its compass needle will point
diagram showing in the direction of H h when the "unknown" field X is
how magnetic m-
t ens it y can be
absent, and will point in the direction of the resultant R of
measured. the two fields when both act. The angle cp through which
the needle deflects in moving from one position to the
other is observed, and the magnetic intensity of the field under test is then
given by -~
X = Hh tan cp. (19.9)
Another way to determine magnetic intensity with a magnetometer is to
set the compass needle of the instrument into oscillation and measure its
frequency. It can be shown in a manner similar to that for a physical pendu-
lum (§ 7-15) that the period of oscillation for small amplitudes is

T=21r/l,
'\JMH
where I is the moment of inertia of the needle about its axis, Mis its magnetic
moment, and H is the magnetic intensity sought. The intensity of one field
can be expressed in terms of another by observing the frequencies of oscillation
of the needle in the two fields. For conditions 1 and 2, it follows from the
foregoing expression that
H1 N
H2 = h2 '
since the frequency f is the reciprocal of the period T of the oscillations.

PROBLEMS
1. An iron toroid having dimensions as shown is closely wound with 1800 turns
of wire. (a) What current is necessary in the winding to produce a magnetic intensity
of 2000 amp· turns/m along the toroid axis shown dotted in the figure? (b) Assume
the magnetic intensity to be uniform throughout the cross-section of the toroid, and
consider the relative permeability of the iron to be 600 under the conditions stated.
What is the value of the total flux?
 PROBLEMS 349

2. A ring solenoid surrounds an iron toroid that has an average diameter of 10 in.
and a square cross-section measuring 2 in. on a side. The winding has 750 turns of
wire and carries a current of 1.2 amp. Consider Fig. 2 to be the magnetization curve
for the material of the toroid, and determine the magnetic intensity and the total
magnetic flux in it. What is the relative permeability of the iron for the given mag-
netization?
3. The power loss in a core of sheet steel because of hysteresis is 75 watts when
there is a 25-cycle current (that is, one that reverses 25 X 2 = 50 times each second)
in the winding around it which yields a maximum flux density of 0.7 weber/m2 • What
will the power loss in this core be when the winding is energized by a 60-cycle current
which produces a m:,i,ximum flux density of 1.0 weber/m2 ?

rwN
16
L_
mm s
2

~ - - - - 30 - - - - - - - " > 1
PROBLEM 1. PROBLEM 5.

4. An iron rod 2 ft long and 1 in. in diameter is bent and welded to form a closed
ring. Upon this core is wound a coil of 1600 turns of wire. What current is needed
in the winding to establish a flux of 5 X 10-4 weber within the core, if the relative
permeability of the iron under these conditions is 2000?
5. The accompanying cross-sectional view shows a magnetic shell with its di-
mensions in centimeters. Consider the portions a and c of the flux path to have con-
stant cross-sections and average lengths of 8 cm, and the portion b to have a variable
cross-section depending on the distance from the center. If the relative permeability
of the iron has an average value of 500, what is the reluctance of the iron part of the
flux path?
6. Compute the reluctance of the entire magnetic circuit of Prob. 5. The flux
density at the central air gap is to be 0.2 weber/m2 ; this value is consistent with the
permeability assumed for the iron. Windings are inserted in the annular space, as
indicated at one side of the diagram, in order to magnetize the shell. What mmf
should be provided?
7. The magnet and armature described in § 19-7 are arranged to have an air gap
between them of 1 mm at each side. Neglect the change that this adjustment makes
in the permeability of the iron, and compute the magnetic flux density in the core.
8. Show that the unit of magnetic pole strength in the mks system is expressed in
webers.
9. A compass needle 5 cm long, magnetized to have a pole strength of 10-5 weber,
is held crosswise in the earth's field where the horizontal intensity is 14.0 amp, turns/m.
What force does this field exert upon each pole of the needle?
10. A flat coil 20 cm in diameter and composed of 30 turns of wire has its plane verti-
,cal and aligned with a compass needle that is placed at the center of the coil. It is
found that a current of 40 milliamp causes the needle to deflect through 20°. Calcu-
late the horizontal component of the earth's magnetic intensity at the place where the
test is being made.
350 ELECTRICITY & MAGNETISM Chap. 19

11. The field produced by an electric current is established horizontally at right

angles to the earth's field of horizontal intensity 15 amp· turns/m and causes a compass
needle to deflect through an angle of 30°. If the applied field is produced by a circular
coil of wire aligned initially to include the compass in its plane and at its center, and
if the coil has 20 turns 0.3 m in diameter, what current in the coil would produce the
stated deflection?
12. Near the Malay Peninsula, the horizontal component of the earth's magnetic
field has its greatest value, namely 32.6 amp· turns/m, and the angle of dip there
averages 10°. Compute the vertical component and the resultant magnetic intensity
at this location.
13. It is desired to neutralize, at a particular point, the vertical component of the
earth's magnetic field; this component has a value of 55 amp· turns/m and is directed
downward. For this purpose a flat circular coil is mounted horizontally with its center
at the reference point. If the coil has 6 turns and a diameter of 2.5 ft, what current is
necessary and in what direction should it be established in the winding?
14. A pivoted magnet makes 1 vib in 2 sec in a field having a magnetic intensity of
15 amp· turns/m. What is the intensity in a region where this magnet makes 2 vib
per sec?
 20
ELECTROMAGNETIC
INDUCTION
The companion effect to the production of magnetic flux around current-
carrying conductors, considered in Chap. 18, is the generation of electromotive
force in a circuit by means of magnetic fields. However, such an emf is
generated only while the magnetic flux through the circuit is changing. The
explanation of the process and the manner in which it is applied to the pro-
duction of electric power is given in this chapter.
20-1. Induced Electromotive Force. An emf can be developed in an elec-
tric circuit through relative motion between the circuit and a magnet; this
fact was discovered in 1831 by Faraday. The action, called electromagnetic
induction, will be illustrated by describing some fundamental experiments.
Figure 1 shows a bar magnet NS held near a
coil of wire, the coil being in a vertical plane and
having its circuit closed through a galvanometer.
No action is observed as long as the devices re-
main stationary. But when the magnet is moved Magnet

-
horizontally toward the coil, the galvanometer [S N]
deflects, showing that an emf is being induced in
Motion
the coil and that this emf sets up a current in
the closed circuit. The faster the magnet is
moved, the larger the deflection will be; when
FIG. 20-1. Inducing an emf
stopped, the deflection falls to zero. When the in a coil by moving a magnet
magnet is moved away from the coil, the results nearby.
obtained are similar but the deflections are re-
versed; this fact indicates that the emf is now in the opposite direction. The
same effects may be produced by moving the coil instead of the magnet,
by using an electromagnet instead of a permanent magnet, or by turning the
coil. These and similar tests lead to the conclusion that an emf is induced
in a coil of wire whenever there is a change in the magnetic flux linked with the
coil.
The direction of the induced electromotive force was established experi-
mentally by the German physicist H. F. Emil Lenz (1804-1865), and the
generalization of the results is called Lenz's Law. This law states that when-
351
352 ELECTRICITY & MAGNETISM Chap. 20

ever a current is set up by a change of magnetic flux through a circuit, the

direction of current will be such as to oppose the act that caused it. Under the
conditions shown in Fig. 1 the act that induces the emf is the movement of
the N pole of the magnet toward the coil. In order to oppose this act, the
adjacent face of the coil must also acquire N polarity, by having the magnetic
flux due to the current directed outward from this face of the coil. The hand
rule described in § 18-2 shows that the current to produce this flux direction
must be counterclockwise as viewed from the magnet.
20-2. Factors Affecting the Induced Emf. The magnitude of the emf in-
duced in a circuit can be found by considering the energy relations that are
involved when a portion of the circuit moves through a magnetic field.
Figure 2 is a plan view of a loop circuit formed
,,..M====x===C=t=:1::'x==N of two bare parallel rails MN and PQ; these
x x x lx x are joined at the left by the fixed conductor
I
x x x1--o-1 ---';.. F MP, and at the right by the bare wire CD
x x x ix x which can be moved along the rails, suppo-
p x D ' x Q sedly without friction. When a magnetic field
\ds exists in the region of wire CD and that wire
is moved, an emf is induced in it and causes
FIG. 20-2. Circuit for evaluat-
current to be set up in the loop circuit; it is
ing the emf induced in a wire
when it moves through a mag- desired to determine the value of the induced
netic field. emf.
The magnetic flux is directed into the page
as indicated by crosses in the figure, and the wire CD is assumed to have a
length l from rail to rail. If a force F is applied to the wire and causes it to
move a distance ds to the right in the time dt, the work done in this inter-
val is dW = F ds. It should be noted that this motion causes a change of
flux through the loop and therefore sets up a current in it. Since the mova-
ble wire carries current and lies in a magnetic field, a force will act upon it;
this force is given by Eq. 18.8 as the product of the flux density B, the
current I, and the conductor length l. Consequently, for uniform motion,
the applied force F must have a value of BIZ and be opposite in direction
to the electromagnetic force. Hence the work done can be expressed as
dW = BIZ ds.
But l ds represents the amount by which the area of the circuit changes, and
this area multiplied by the flux density B gives the change in the magnetic
flux linking the circuit, namely def,. Therefore, the work done is
dW =Id¢.
The work can be expressed independently in terms of the emf induced in
the circuit during the time interval dt; if this emf is represented by e, then,
as explained in § 17-8, the work done is
dW = le dt.
§ 20-3 GENERATION OF ELECTROMOTIVE FORCE 353

The two expressions for dW represent the same quantity of work and may be
equated; ,vhen this is done I d<I> = I c dt, and the induced emf is
d<I> (20.1)
c = di'
an equation known as Faraday's Law of Electromagnetic Induction. When
the rate of change of flux is expressed in webers per second, the emf is in volts.
Equation 20.1 applies to any circuit-for example, to a coil in which an emf
is induced by motion of a magnet. A coil wound with several turns close
together, so that all are subject to the same variation of flux, would have equal
emfs induced in all the turns, and these would be added to obtain the total
induced emf. Consequently the emf induced in a coil of N turns while the
flux through it is changing at the rate of ~~ webers/sec is given in volts by
d<I> (20.2)
e = N-·
dt
Another way of regarding the process of electromagnetic induction is to
consider that an emf is induced in a circuit whenever any of its conductors cut
magnetic flux. To develop this view, the simplest way is to combine two of
the preceding equations concerning the moving conductor-one of them
. ds ds
dW = Bll ds and the other diV = I e dt. The result 1s e = Bl dt · But dt
is the speed of the conductor; if this is designated by v, the emf induced in the
conductor becomes
e = Blv. (20.3)
If B is in webers per square meter, l in meters, and v in meters per second,
e will be in volts.
The foregoing expressions are important equations in electrical engineering,
and their simplicity and directness can be attributed to the use of mks units.
·when used with electromagnetic units, d<I>/dt is in maxwells per second, Bin
gausses, l in centimeters, and v in centimeters per second, but the results fore
are in abvolts rather than volts. Since 1 weber = 108 maxwells, § 18-1, it
can be inferred that 1 volt = 108 abvolts.
Whether the emf is regarded as due to a change of flux through the loop or
to the cutting of flux by a conductor, its direction can be found by Lenz's Law.
The cause of the induced emf in the circuit of Fig. 2 was the motion of con-
ductor CD toward the right, a motion which increased the flux linking the
circuit. To oppose this action, the induced emf must be so directed that it
tends to set up a current that would decrease the flux linking the circuit. The
direction of this current would necessarily be from D to C, and that is the di-
rection of the induced emf.
20-3. Generation of Electromotive Force. The production of an electro-
motive force in a coil of wire by changing the magnetic flux through it is the
354 ELECTRICITY & MAGNETISM Chap. 20

basic principle of operation of all types of electric generators. The physical

structure of a generator is indicated in Fig. 3. A coil ABCD is mounted
on an axis in its own plane and at right angles to the flux passing between the
poles N and S of an electromagnet. The
ends of the coil are joined to slip rings,
and brushes X and Y press against them,
so that the coil, when rotating, may re-
main connected electrically to the exter-
nal circuit R.
It is apparent that the emf generated
in the rotating coil will not be the same
from instant to instant. As the coil
moves through the position shown, no emf
will be generated, since at this instant
neither coil-side AB nor CD is moving
Fw. 20-3. Elementary alternating- across the magnetic flux. vVith motion
current generator. of the coil in counterclockwise direction
from this position, AB moves upward
and CD downward through the flux, setting up small emfs directed respec-
tively from A to B and from C to D; these add together and make brush Y
positive and brush X negative. The emf keeps increasing in magnitude
until the coil is horizontal, at which instant the conductors have their great-
est speed at right angles to the flux. As the coil moves farther, the emf
becomes smaller, and finally reaches zero again when the coil-sides AB and
CD have interchanged places. During the second half-revolution the same
effect is produced, but the emf is in the opposite direction because AB is then
moving downward and CD upward through the flux. As the action continues
it results in an alternating emf, and this sets up currents which traverse the
load circuit, first in one direction and then in the other, reversing direction
with every half-revolution of the coil. Most generators today produce cur-
rents of this kind; they are called alternating-current generators, or simply
alternators.
Alternating-current generators are considered more fully in Chap. 22, to-
gether with alternating-current circuits, motors, and transformers.
20-4. Method for Generating Direct Current. The simple form of alter-
nator described in the preceding section can be converted into a direct-current
generator of rudimentary form by replacing the slip rings with a commutator.
This is a device for reversing the connections between the coil and the load
circuit at the instants when the emf reverses in the coil. The construction
is represented in part I of Fig. 4, in which the commutator is shown as a split
metal tube, one part connected to conductor AB and the other to conductor
CD of the coil. In the position shown, these conductors are connected to the
load R through brushes Y and X, respectively. Part II of the figure illus-
trates the action of the commutator by showing the alternating emf developed
§ 20-5 DIRECT-CURRENT GENERATORS 355

in the coil at different positions and the unidirectional emf available at the
brushes.
As the coil moves from the position shown, the performance during the first
half-revolution, represented between O and 1 in part II, is the same as that
of the alternator; this motion produces the positive lobe of the emf curve.
Then the commutator interchanges the connections between coil and load,
joining AB to X and CD to Y; consequently, during the second half-revolution
(from 1 to 2 in the figure), the motion produces another positive lobe, instead
of a negative one as produced by the alternator. Therefore, as the coil of
the direct-current generator continues to rotate it supplies a pulsating but
unidirectional current to the load.
20-5. Direct-current Generators. The modern generator used for furnish-
ing large currents at constant potential is a development of the simple machine
illustrated in Fig. 4. It consists essentially of a stationary field structure com-
posed of electromagnets, and a rotating armature carrying a number of coils,
together with a multi-segment commutator and a set of brushes.

B~

-0 +- · 1 - - - - ~ - - ~ - - - - ~ -

-
o.
_g - Unidirectional

II
F1(!. 20-4. I, Elementary direct-current generator; II, curves of emf in armature and in
load circuit.

The field structure may have one or more pairs of poles. A bipolar field,
with one N and one S pole, is shown in the two preceding figures. A typical
4-pole construction is illustrated in Fig. 5; the poles are alternately N and S,
and there are four flux paths, as shown by the four groups of flux loops.
The armature is wound to conform to the number of poles of the field
structure. For a bipolar machine an easily understood form of armature is
shown in Fig. 6. Its coils form a continuous closed winding upon an iron
ring, and are connected to the segments of the commutator C. Brushes press
against the commutator at points midway between the pole pieces N and S,
and constitute the positive and negative terminals of the machine. The mag-
netic flux follows the iron ring because of its low reluctance, and the space
inside of the ring is left practically free from flux. When the armature is
rotated, the conductors on the outer cylindrical surface of the ring cut the flux
as they sweep past the poles and thereby develop emf.
356 ELECTRICITY & MAGNETISM Chap. 20

There are two paths for the current through the armature of Fig. 6 from
one brush to the other, and one path has the same number of conductors as
the other. The emfs induced in the conductors are both directed to make
the bottom brush positive and the top one negative. These paths are in
parallel, hence the total emf generated by the armature is the sum of the emfs
developed in the conductors on one side only. The armature coils, as they
reach the positions at the top and bottom, are momentarily short-circuited
by the brushes, but this does no harm, since emfs are not being induced in
them at these positions.

N s

Fm. 20-5. Four-pole field Frn. 20-6. Ring-type armature.

structure showing paths of
magnetic flux.

A considerable part of the winding on a ring armature is ineffective, because

the conductors on the inside do not cut flux. In modern machines the arma-
ture takes the form of a drum-shaped iron core with the conductors laid in
slots along the cylindrical surface, each pair of conductors forming one loop
of a coil that spans the distance from pole to pole. As the armature turns,
the flux from the poles links with coil after coil of the armature, and conse-
quently all the conductors that make up the coils are effective in producing emf.
In practice, generators have a large number of coils distributed evenly
around the armature and a correspondingly large number of commutator seg-
ments. The advantage of this construction can be seen from Fig. 7, which
shows the effect of connecting in series two coils, 1 and 2, which are located
so that when the emf in one of them is a maximum, the emf in the other coil
is zero. The instantaneous values of these emfs are indicated by the curves;
their resultant is represented by the heavy line in the figure. It will be seen
that the fluctuation of emf is less in the resultant of the two curves than in
either one by itself. By the use of a large number of coils, the fluctuation is
nearly eliminated.
The iron structure which supports the armature coils is made of thin sheets
or laminations, in order that the motion of the armature through the flux may
not induce currents to any gn'at extent in the iron itRelf. Currents induced
§ 20-5 DIRECT-CURRENT GENERATORS 357

by flux changes in such metal masses are called eddy currents; they waste
energy in heat, and thus reduce the efficiency of the machine.
The field winding is usually supplied with current from the armature of the
machine itself. This is done in the shunt-wound generator by bridging the
field coils across the armature terminals B and D, as shown in Fig. 8. The
magnetic flux is adjusted by controlling the current in the field circuit with a
rheostat. When a shunt-wound generator is producing an emf E, the po-
tential difference across the machine terminals will be smaller than E by the

+
-- 0
.,,
0
11)
..r:
....
Resultant
......
E Coil 1 Load
UJ
;-,-Coil2
I

Time Shunt circuit Main circuit

F10. 20-7. Fluctuations of emf and their Fm. 20-8. Connections of a shunt-wound
reduction. generator.

amount of the potential drop in the armature itself. If the current is Ia in

an armature of resistance Ra, the potential difference at the terminals will be
(20.4)
The current in the field coils of resistance R1 will be
v
11 = R1
when all the resistance of the field rheostat is cut out of circuit. Equation
20.4 shows that as the load on the generator increases, thereby increasing the
armature current Ia, the potential difference V across its terminals becomes
less; it is lessened further by the reduction in flux brought about by the con-
sequent lowering of the field current.
The emf of any direct-current generator can be computed by finding the
average emf induced in each armature conductor and multiplying this value
by the number of conductors in series in each path through the armature.
In a generator that has P poles and is driven at n rev per sec, the armature
will make one revolution in 1/n sec, and the time taken for any one of its
conductors to pass one pole is in sec. If the magnetic flux from each pole

is <I> webers, the average rate of cutting flux is <I> I in' and hence the average
358 ELECTRICITY & MAGNETISM Chap. 20

emf induced in each armature conductor is if>Pn volts. With S conductors

on the entire armature and p current paths in parallel through it, the number
of conductors in series in each path will be S/p. Usually the armature is
wound to have as many paths as the field winding has poles, so that p = P.
Hence the emf of the generator in volts will have an average value of
E = if>nS. (20.5)
For a given generator, the number of conductors is fixed; consequently the
emf generated is determined by the magnetic flux if> and rotational speed n.
In practice, the emf of a generator is adjusted by changing the flux, for the
speed is determined by that of the driving engine or other prime mover.
20-6. Direct-current Motors. The electric generator, described above, is
a mechanically driven machine which can produce current in a circuit con-
nected to it; in contrast, an electric motor is an electrically driven machine
which can do mechanical work by operating all sorts of machinery coupled
to it. The two machines are alike in physical construction, except for minor
differences in design and adjustment. \Vhen such a machine is driven as a
generator it converts mechanical into electrical energy, and when operated as
a motor it converts electrical into mechanical energy. In each process some
energy is of course wasted in heat.
The fundamental principle upon which every electric motor operates is
stated in § 18-7; briefly it is that a conductor carrying a current and appro-
priately placed in a magnetic field experiences a sidewise thrust. The mag-

I
I
To
supply Field
mains
I
I

Fw. 20-9. Current, flux, Fw. 20-10. Connections of a eihunt-

and torque relations in a wound motor.
motor armature.

nitude of this force on a conductor of length Z, carrying a current I, and placed

at right angles to a magnetic flux of density B, is F = BIZ. The direction
of the force can be determined by reference to Fig. 18-8; with current in the
wire directed toward the observer and the magnetic flux directed from left to
right, the wire will be forced upward.
The essential elements of a bipolar motor are illustrated in Fig. 9. A com-
mutator (not shown) keeps the armature current in the right-hand belt of
conductors directed toward , the observer, and that in the left-hand belt
directed away from him. For the direction of the field shown, the armature
§ 20-7 COUNTER ELECTROMO'l'IVE FORCE 359

conductors at the right are forced upward and those at the left downward.
Each conductor experiences a torque determined by the force acting upon it
and the lever arm of this force about the axis of rotation. The torques for
all the conductors have the same direction, counterclockwise in the present
instance, and are added to obtain the total torque developed by the armature.
The torque developed for driving a mechanical load depends in any particu-
lar motor upon two factors: the current I a in the armature conductors, and
the density B of the flux in which they are located. The horsepower output
of a motor can be found by measuring the torque with a band brake placed
around the pulley of the motor and measuring the speed of its shaft with a
revolution counter or tachometer.
A direct-current motor in which the field coils are connected in parallel with.
the armature, as shown in Fig. 10, is said to be shunt wound. In the larger
sizes, the current is supplied through a starting device (not shown), which
first connects the field winding across the service mains and then connects the
armature to them through starting resistors; these are gradually cut out of
circuit as the motor gains speed.
20-7. Counter Electromotive Force. As the armature of a motor rotates,
its coils move through the magnetic flux from one field pole after another.
As a result, emfs are induced in the motor armature exactly as in a generator
armature. If the currents in the motor armature are directed as shown in
Fig. 9, driving it in a counterclockwise direction, then the emf that is induced
in the armature conductors is opposite in direction to the currents indicated
in them. Therefore, the induced emf acts in opposition to the emf supplied
to the motor and is aptly termed a counter emf.
The value of the counter emf in a motor depends upon its speed of rotation,
and can be calculated in the same way as is the emf of a generator. If the
motor is connected across supply mains which provide a difference of potential
Vat the motor terminals, and if Ee is the counter emf, then the net potential
difference acting to produce current in the armature will be V - Ee, and the
current through the armature of resistance Ra will be

(20.6)

The field current can be computed directly by Ohm's Law, by dividing the
potential difference V across the field by the field resistance R1 ; that is,

I1
v
= -.
R1
If the mechanical load on a motor is increased, the machine automatically
develops a larger torque. This action may be explained as follows: The in-
creased load exerts an increased backward torque on the armature, causes it
to slow down somewhat, and thereby reduces the counter emf induced in it.
As a result, the net potential difference becomes greater and causes the arma-
360 ELECTRICITY & MAGNETISM Chap. 20

ture current to increase. The motor, therefore, draws a greater armature

current, thereby developing more torque, and drives the increased load, al-
though at a slightly reduced speed.
As a typical example, a shunt motor operating on a 230-volt line has an armature
resistance of 0.15 ohm and a field resistance of 100 ohms. At light load the motor
rotates at a speed such that its counter emf is 228 volts; at an inrreased load this emf
falls to 223 volts. Compare the change of armature current with the change of speed.
The armature current at light load is Ia = (230 - 228)/0.15 = 13.3 amp, and at
increased load is (230 - 223)/0.15 = 46.7 amp. The speed change that causes a
lowering of counter emf from 228 to 223 volts is a 2.2 per cent decrease; the resulting
change of armature current from 13.3 to 46.7 amp is a 250 per cent increase. The field
current remains constant at the value 11 = 230/100 = 2.30 amp.

Just as a motor when revolving develops a counter emf, so also a generator

when delivering current sets up an opposing torque. This fact can be verified
easily by noting that in a generator under load, the armature conductors carry
current and are located in a magnetic field; a study of the distribution of flux
will show that the torque set up is opposite to the direction of rotation. This
torque must be overcome by the driving engine, and the engine must do more
work as the electrical load on the generator is increased.
20-8. The Induction Coil. If two circuits are close together and the mag-
netic flux around one of them changes because the current in it has been
altered, then this change of flux affects the
G
other circuit and develops an emf in it.
This is the operating principle of the induc-
tion coil. Figure 11 shows two windings
surrounding a straight core composed of
soft-iron wires. The primary winding,
represented by the heavy line, is connected
in series with a battery, a switch S, and an
interrupter; the secondary winding, repre-
Fm. 20-11. Circuit of an induction
sented by the light line, is a coil of many
coil. turns terminating in a spark gap G. The
interrupter is an armature mounted on a
flat spring s which carries a hard metal contact a; this periodically touches
a similar but stationary contact b when the spring is set into vibration.
When the switch is closed, the current through the primary winding mag-
netizes the core, inducing an emf in the secondary winding. The armature
is promptly attracted to the core and the contact between a and b broken.
By this action the current in the primary winding is interrupted and the
magnetism in the core falls to a low value (the residual magnetism), again
inducing an emf in the secondary winding. A capacitor C is bridged across
the contacts of the interrupter in order to hasten the breaking of the circuit,
§ 21-8. The armature is then released from the core and springs back, again
closing contact between a and b. The cycle of operations is repeated as long
§20-8 THE INDUCTION COIL 3fil

as the switch is closed, at a rate determined by the stiffness of the spring and
the mass of its armature.
The emfs induced in the secondary winding produce sparks across the gap
in its circuit. However, the emf is much greater when the primary contacts
open than when they close; hence for the longer gap lengths the discharge is
unidirectional. The negative terminal of the gap is found to be the one near
which the spark is brighter.
PROBLEJ11S
1. A flat circular coil of 30 turns has a radius of 10 cm. It is placed with its plane
horizontal and the N pole of a magnet is brought downward toward the coil. This
action causes the magnetic flux through the coil to increase by 5 X 10-4 weber in
0.2 sec. What is the magnitude and direction of the emf induced in the coil?
2. A copper wire 15 m long has a resistance of 0.3 ohm. The wire is looped into a
flat circular coil of 24 turns and the ends of the wire are joined to form a closed circuit.
The N pole of a magnet is then brought toward the coil. If, in consequence, the flux
through the coil changes by 1 milliweber, what quantity of electricity circulates in
the winding?
3. The flux density in the air gap of a large electromagnet is to be measured with
the aid of a flat exploring coil connected to a ballistic galvanometer. The coil consists
of 50 turns of fine wire, and the coil diameter is 1 in. It is placed perpendicular to
the magnetic flux in the gap and then quickly pulled out of the field. As a result, a
quantity of electricity of 0.0012 coulomb is observed to flow through the galvanometer
circuit of which the total resistance is 30 ohms. What is the flux density in the air gap?
4. The wire illustrated in Prob. 6 of Chap. 18 is moved in 0.1 sec from the top
to the bottom of the magnetic field shown. Determine the emf generated in the wire
by means of both Eqs. 20.1 and 20.3.
5. To find the flux density of the earth's magnetic field in a horizontal direction,
a coil is mounted on a vertical axis with the plane of the coil also vertical, and a
measurement is made of the emf induced in the coil when turned in the field. The
coil is square, measuring 10 in. on a side, and composed of 40 turns of wire. With
the plane of the coil initially perpendicular to the earth's field, it is turned through
one-quarter of a revolution in 0.018 sec; this resulted in an average induced emf of
3 millivolts. What value is obtained for the horizontal component of the earth's
flux density at the place of test?
· 6. The coil of Prob. 2 is suspended so that its plane lies in the magnetic meridian
of the earth; it is then turned through 90° about a vertical axis in lo sec. What is
the average emf induced in it if the horizontal intensity of the earth's field is
16 amp·turns/m?
7. In the air gap of a certain bipolar generator each armature conductor generates
an average of 1 volt when the armature revolves at 1800 rev /min. The field poles
are square and measure 20 cm on a side, and the conductor cuts all of the flux from
these poles when it revolves. What is the flux density in the air gap?
8. In a simple generator, such as shown in Fig. 3, suppose the coil to be a 10-cm
square and to be turned between poles of the same size at 100 rev /min in a uniform
field of 0.2 weber/m2 • Construct a graph showing the relation to scale between the
emf generated and time.
9. An armature is placed in the bipolar field of the generator of Prob. 7. The
armature winding consists of 150 conductors arranged in two parallel paths. Deter-
mine the average emf generated by the machine when its armature revolves at 2400
rev/min.
3G2 ELECTRICITY & MAGNETISM Clmp. 20

10. A 4-pole generator has pole pieces with a cross-section of 100 in. 2 and an arma-
ture with a total of 160 conductors connected in 4 parallel paths. The flux extending
from each pole piece is 0.075 weber. At what speed will the armature generate an
average emf of 240 volts?
11. A 120-volt shunt-wound generator has an armature of 0.090 ohm resistance and
a field winding of 100 ohms. What are the currents in these two parts of the machine
when it delivers 5kw? What emf is generated in the armature by electromagnetic
induction?
12. The armature of a bipolar shunt-wound motor is 24 cm in diameter and supports
on its surface 100 conductors each carrying 20 amp. It revolves at 1200 rev/min,
between square pole faces measuring 24 cm on each edge, in a field of flux density
0.8 weber/m 2 • Assume that each conductor moves at right angles to the flux, and
compute (a) the torque developed by the armature, (b) the work done in one revolu-
tion, and (c) the horsepower supplied by the motor.
13. A shunt-wound motor has an armature resistance of 0.25 ohm and a field resist-
ance of 150 ohms. It is connected to 120-volt supply mains and is generating a counter
emf of 114 volts. Compute (a) the armature current, (b) the field current, and (c) the
total current taken by the motor.
14. A shunt-wound motor, which has a field resistance of 160 ohms and an armature
resistance of 0.33 ohm, takes a current of 36.5 amp when connected across 240-volt
service mains. Under these conditions the motor output is 10.0 hp. Compute (a) the
field current, (b) the armature current, (c) the counter emf, (d) the efficiency of
the motor, (e) the power wasted in heating the field circuit, (f) the power wasted in
heating the armature circuit, and (g) the power wasted elsewhere in the motor.
15. The armature of a shunt-wound motor has a resistance of 0.15 ohm and takes
12 amp from 120-volt supply mains when turning at 1080 rev/min. Upon increasing
the load the motor slows down to 1070 rev/min; what will be the increase in armature
current?
 21
INDUCTANCE AND
CAPACITANCE
The production of magnetic flux around a circuit by establishing a current
in it, and the production of emf in a circuit by changing the magnetic flux
around it, are the basic facts that were developed in the last two chapters.
Because of these facts, important effects are produced in every circuit through
its magnetic field whenever the current in the circuit is changing. It will also
be found that comparable effects are produced in a circuit through its electric
field whenever the applied potential difference is changing. These effects are
considered in this chapter: the influence of magnetic fields under the heading
of inductance, and that of electric fields under capacitance. Inductance and
capacitance are the properties of electric circuits that permit energy to be
stored respectively in magnetic and electric fields; these properties are alto-
gether different from resistance, wherein the energy is dissipated in the form
of heat. With alternating currents the properties of inductance and capaci-
tance play a dominating role in circuit behavior, because the current and emf
are continually changing in magnitude and direction.

INDUC'rANCE

21-1. Mutual Induction. Any change of current in an electric circuit pro-

duces an alteration in the magnetic field around it, and the accompanying
change in flux sets up an emf in whatever circuit may be located nearby.
Such changes in current may occur from a variety of causes, such as closing
and opening the circuit or varying the load; with alternating current, even
when steady conditions are maintained, the current continually changes its
value from moment to moment. Often the emfs induced in neighboring cir-
cuits by these current changes are sought for, and the circuits are then
designed to develop particular emfs for the purposes intended; in other situ-
ations the emfs induced disturb the normal operation of the neighboring
circuits, and hence provisions are made to minimize these disturbances.
The interaction between two circuits will be studied by considering a pair
of coils in an environment free from magnetic materials. Figure 1 shows the
coils wound side by side on a cardboard tube; coil 1 is connected to battery B
through a rheostat R, and coil 2 is connected directly to galvanometer G. So
363
364 ELECTRICITY & MAGNETISM Chap. 21

long as the current in coil 1 remains steady, the magnetic flux extending
through it and coil 2 will be constant, and the galvanometer will show no
deflection. If the current is changed by moving the slider along the rheostat
the flux through both coils will change, and the galvanometer will deflect as
a direct result of changing the flux in coil 2. In a short interval dt, if the
current changes in coil 1 by the amount di1, and meanwhile the flux linking
coil 2 changes by the amount difJ, there will be induced in coil 2 an emf that
is proportional to difJ/dt, as expressed by Eq. 20.1. But the flux is proportional
to the current that produces it, and hence the time rate of change of flux is
. ~ ~
proportional to the rate of change of current, or dt a: dt · Therefore, the
induced emf in coil 2 will be proportional to the rate of current change in coil 1,
or in equation form
(21.1)

where M is a factor, called the coefficient of mutual induction, that depends

upon the magnetic coupling of the two coils.
The expression above indicates that the more rapidly the current changes,
the greater will be the emf induced; this can be verified experimentally by
moving the rheostat slider at different speeds. Also, if the battery (with
rheostat) and galvanometer are interchanged, it will be found that an emf is
induced in coil 1 when the current is altered in coil 2, and its value will be

where M has the same value as before and di2/dt is the time rate of change
of current in coil 2.
The coefficient of mutual induction, also called the mutual inductance, is
expressed in terms of a unit called the henry; it is named after Joseph Henry
(1797-1878), one of the earliest American physicists. Two coils have a mutual
inductance of one henry when a current change of one ampere per second in one
coil causes an emf of one volt to be induced in the other.
If an iron core is inserted in the coils of Fig. 1, the results will be different
in two respects: first, the mutual inductance of the coils will be greatly in-
creased, and second, it will not have a fixed value, because a given change in
current will not in general cause a proportional change in the magnetic flux
in the iron, § 19-4.
The direction of the induced emf can be found by Lenz's Law. Thus, an
increase of current in coil 1 is equivalent to moving this coil mechanically
toward coil 2-an action that causes the emf induced in coil 2 to be opposite
in direction to the current in coil 1. Also, a decrease of current in coil 1
causes the emf induced in coil 2 to have the same direction as the current in
coil 1.
§ 21-2 SELF-INDUCTION 365

21-2. Self-induction. As just stated, when two circuits are close together,
like the coils illustrated in Fig. 1, a change of current in one of them pro-
duces an emf in the second, because of the accompanying change in flux link-
ing it. A change of current in one coil, however, will cause a change of flux
not only through the other but also through the very coil in which the current
is changing. Hence, any change of current that occurs in a coil will cause an
emf to be set up by electromagnetic induction in that same coil. As before,
the emf will depend upon the rate of current change, and its value is
di
e = L dt, (21.2)

where Lis a factor, like Min the preceding equations, which is characteristic
of the circuit itself and its magnetic environment. The emf is called the emf
of self-induction and its direction is always such as to oppose the change of
current that caused it. The factor Lis called the coefficient of self-induction,
or simply the inductance, of the circuit.

Coil

Lamp
R

Switch
Fm. 21-1. Circuits coupled Frn. 21-2. Experiment toil-
magnetically to demonstrate lustrate effect of inductance
mutual induction. in a circuit.

Inductance, like mutual inductance, is expressed in henrys. A circuit has

an inductance of one henry if a current change of one ampere per second causes
an emf of one volt to be induced in it. A circuit in which a large counter emf
of self-induction is set up for a given rate of current change is said to have a
large inductance.
A coil of many turns of wire has more inductance than the same wire when
unwound so as to form only a single loop, because the emf induced depends
not only upon the rate of change of flux but also upon the number of turns
through which this change occurs. The inductance would also be greater if
the coil were wound upon an iron core.
The effect of inductance can be illustrated by the circuit shown in Fig. 2,
in which a coil of wire, with many turns wound upon an iron core, is shunted
across a lamp and joined to a direct-current generator, some resistance being
included in the circuit. The coil has a vastly greater inductance than the
lamp, but much less resistance. When the switch is closed, the lamp will flash
brightly for a brief interval and then become dim. This action is explained
366 ELECTRICITY &: MAGNETISM Chap. 21

by the fact that the magnetic field had to be established around the coil, and
that during its formation the emf of self-induction hindered the growth of
current in it, hence the coil did not serve as an effective shunt on the lamp.
Again, when the switch is opened quickly, the lamp will flash even more
brightly, because the large emf induced in the coil by the rapid decay of the
flux then establishes a strong pulse of current through the lamp.
For some purposes it is necessary to use coils of wire with very little or no
appreciable inductance; coils of this type will have very little magnetic flux
around them. Such so-called noninductive coils are wound by arranging the
wire in a long "hairpin" loop and winding the two conductors so formed side
by side until the coil has the desired resistance; in this way the flux due to
current in one conductor neutralizes that of the other. The resistance coils
of Wheatstone bridges and similar instruments for electrical measurements
are wound noninductively.
21-3. Inductance of a Solenoid. The inductance of a coil of wire depends
upon the amount of magnetic flux that a given current in the coil can produce.
This fact can be verified by considering the current in a coil of inductance L
to change from one value to another, thereby inducing in its winding an emf
which at any instant is given bye = L ~; volts, as in the preceding section.
But any change of magnetic flux through the coil of N turns induces in it an
emf which has an instantaneous value of e = N ~: volts, by Eq. 20.2. Ob-
viously, these equations express the same action in different ways, and there-
fore
L di= N def>_
dt dt

. express10n,
F 1 om th1s . L -- N dd;
~. . On the assumption that the flux grows
uniformly with the current, and that a value of flux if> will be reached when
the current is I, the inductance of the coil becomes
Nif> (21.3)
L=-,
I
where the product of the number of turns N and the number of flux loops if>
linked with them is spoken of as flux-linkages. It follows that a circuit which
has an inductance of 1 henry will have one flux-linkage per ampere of current
in the circuit.
Equation 21.3 can be applied to a ring solenoid in which a coil of N turns
is wound upon a toroidal core of length l and cross-sectional area A. The
magnetic flux set up in this core by a current of I amperes in the coil is
<I> = µHA, where µ is the permeability of the core and H is the magnetic

intensity within it. This intensity is given in § 19-1 as H = ~I, and conse-
§ 21-4 GROWTH AND DECAY OF CURRENT 367

quently the inductance can be expressed as

L =NAµ NI= N 2 Aµ
I l
showing that the inductance of a solenoid varies directly with the square of
the number of turns, and depends upon the cross-section, length, and perme-
ability of the core. When the dimensions of the core are expressed in meters,
the inductance will be in henrys.
The expression indicates that the unit of permeability may be expressed in
simpler units than before; the permeability unit is the henry per meter.
In summary, the different units in which permeability can be expressed are
as follows:
weber = newton = Q_enry.
µ = ampere· turn· meter ampere 2 meter
21-4. Growth and Decay of Current in Inductive Circuits. The current in
a circuit containing inductance and resistance does not instantly assume its
ultimate value when connected to a unidirectional source of emf, because of
the counter emf of self-induction. Although the entire current growth usually
occurs in a fraction of a second, the current rises rapidly at first and then
builds up more and more slowly as it approaches its final value.
- A solenoid of 10 ohms resistance and 2 henrys inductance is being connected across
120-volt direct-current supply mains. Determine the momentary values of the cur-
rent when it is increasing at 50 amp/sec and at 10 amp/sec.
At an instant when the rate of current growth is 50 amp/sec, the counter emf has

a value given by Eq. 21.2 of e = L ~: = 2 X 50 = 100 volts, and hence the current
in the solenoid is (120 - 100)/10 = 2 amp. When the rate of current growth has
fallen to 10 amp per sec, the counter emf is momentarily 2 X 10 = 20 volts, and the
current value is (120 - 20)/10 = 10 amp. The current will have the final value of
V /R = 120/10 = 12 amp.
The foregoing illustration also indicates that Ohm's Law in its simple form
applies only to steady currents. A broader statement of the law includes the
counter emf of self-induction, and the current at any instant can be ex-
pressed as

E- Ldi
dt (21.4)
i =----,
R
where E is the emf impressed upon the circuit. At the instant that an in-
ductive circuit is connected to an electrical source, all of the emf causes the
current to grow, for then the current i is zero and E = L ~;; in contrast,

whei1 the current has reached its final value I, then~! = 0 and E = RI.
368 ELECTRICITY & MAGNETISM Chap. 21

To solve the differential equation above, the variables are separated, and
the resulting equation is multiplied through by - R and integrated. Thus,
dt = L di = __!!_ - R di ,
E - Ri -RE - Ri

and
r
• dt = -
L s-Rdi
R E - Ri ;

hence t = LR In (E - Ri) + constant.

The value of the constant of integration must be such that i = 0 when t = 0,
and therefore the constant is ~ ln E. Then -t = ~ ln (E - Ri) - ~ lnE
= L ln E - Ri_ ·when transposed to exponentials, the result is
R E
E - Ri _LRt

E =e '

where e is the base of natural logarithms and has the value 2.7183. Conse-
quently, the current in amperes at an instant t seconds after an emf of E volts

.. -----------
...
c:
Cl)

:::>
u

Time

Fru. 21-3. Graph of current growth and decay in an inductive circuit.

is impressed upon a circuit having a resistance of R ohms and an inductance

of L henrys will be

i = ! ( e-P).
1 -

A similar method of analysis shows that when the applied emf is suddenly
withdrawn and replaced by a short-circuit, the current does not fall to zero
instantly. The current values during this period are given by that part of
the foregoing equation which is subtracted from E/R, that is,
E _lli
i = - e L·
R
In these expressions the ratio of the inductance L to the resistance R deter-
mines the rate of current growth and decay in the circuit. For this reason
the ratio L/R is called the time constant of the circuit. If the time is chosen
so that t = L/R, then the exponent of e becomes -] , and the latter equa-
 § 21-5 ENERGJ' OF A MAGNE'PIC FIELD 369

tion indicates that the current falls to 1/e = 0.368 of its steady value in
this time.
Figure 3 shows a graph of current in an inductive circuit from the instant
it is connected to a constant source of supply until the current subsequently
falls to zero after the source is replaced by a short-circuit. At instant t1 a
constant emf is applied to the circuit, and at instant t2 the source is replaced
by a short-circuit.
21-5. Energy of a Magnetic Field. In order to produce current in an induc-
tive circuit, work must be done against the emf of self-induction so that the
magnetic field can be created around the circuit. When the current has the
, momentary value of i and the emf impressed upon the circuit is E = L ~; + Ri
(§ 21-4), the energy supplied in a time dt is
Ei dt = Li di + Ri 2 dt,
which shows that part of the energy is converted into heat (Ri 2dt) and the rest
is used to build the magnetic field. Consequently, while the current is reach-
ing its final value I, the energy expended in the magnetic field, as found by
integration, is

W = I t=t
i e dt =
fi=IiL d"..!:. dt = [L -JI _i
2
,
=0 i=O df 2 0

or W = ! L/2• (21.5)

The2nergy of the magnetic field, W, will be in joules if the inductance of the

circuit is in henrys and the current through it is in amperes. Once established,
the field requires no further energy to maintain itself; all energy supplied to
the circuit subsequently is dissipated in heating.
Equation 21.5 is analogous to the expressions for the kinetic energy of a
moving mass, § 6-5, and affords an energy concept of the unit of inductance.
A circuit of one henry inductance, carrying a current of one ampere, would
have ! joule of energy stored in its magnetic field.
The energy per unit volume of the magnetic field can be expressed in terms
of magnetic quantities rather than in quantities of the associated electric
circuit. Let it be assumed that the energy is stored in a toroidal core of
length land cross-sectional area A, and that the core material has a perme-
ability µ when the magnetic flux is <I>. Then, since the inductance of a ring
solenoid is L = i<I>' the energy per unit volume of the magnetic field is:

1. LJ2 Nl<I>
Energy density = 2
lA = 2 lA ·

But the number of ampere· turns NI of the winding equals the product of the
l
flux <I> and the reluctance µA of the core, and the flux per unit area <I>/ A
370 ELECTRICI1T & MAGNETISM Chap. 21

equals the flux density B. Hence, the energy per unit volume of the mag,netic
field is:
B2
Energy density (21.6)
= 2 µ'
and the unit is the joule per cubic meter.

CAPACITANCE

21-6. Capacitance of an Isolated Conductor. The study of capacitance is

based on the principles of electrostatics, developed in Chap. 15. There, con-
sideration is given to electric charges in space, to electric fields near them,
and to the potentials at points within such fields. It is pointed out, for
example, that a charged sphere isolated in space has an equi-potential sur-
face, and that the electric lines of force around it are arranged radially,
exactly as though the charge were concentrated at a point.
The potential at any distance r from a point charge Q in free space is given
by Eq. 15.6 as V = k Q,
r
where k has the numerical value 9 X 109 • This
would be the potential also of the surface of a sphere of radius r which carries
the same charge. Basically, the equation shows that the potential V is
directly proportional to the amount of charge Q on the sphere. This con-
clusion is found to be true for an isolated conductor of any shape.
The reverse of this conclusion is also true--that is, the charge on an isolated
conductor is proportional to its potential; hence Q o: V, or
Q = CV, (21.7)
where C is a proportionality constant determined by the size and shape of the
conductor and is called the capacitance of the conductor. In this expression,
when Q is in coulombs and V in volts, the capacitance unit is the coulomb per
volt; it is called the farad, derived from the name of Faraday.
For an isolated conducting sphere of radius r, the foregoing equation gives
the capacitance as
Q Q r (21.8)
c= v= 9x 109 Q = 9 x 109 '
r
which shows that the capacitance of the sphere is directly proportional to its
radius. When another conductor is brought nearby, the capacitance of the
sphere is increased.
21-7. Dielectrics. In the earlier consideration of electric fields near and
between charged bodies, the medium was restricted theoretically to free space,
but practically includes air also. Attention is now given to the presence in
electric fields of insulating substances-so-called dielectrics. The molecules of
such substances are normally uncharged, that is, the positive component
particles neutralize the negative ones. But in an electric field, these charges
§ 21-7 DIELECTRICS 371

are displaced and the molecules become polarized. Thus, when a sheet of di-
electric is placed in a uniform electric field between parallel plates, the entire
sheet becomes polarized and one surface becomes positive and the other
negative.
The effect of dielectric polarization is shown in Fig. 4. In part I the electric
field between parallel plates in a vacuum is directed toward the right from the
positive to the negative plate. In part II the space between the plates is
almost filled by a dielectric sheet, and charges of opposite sign are induced
on its adjacent surfaces. The field set up within the dielectric by these bound
surface charges is directed toward the left and is opposite in direction to the
initial field, causing the field in the dielectric sheet to be less intense than in a
vacuum. This effect is allowed for through the concept of permittivity.

Plate s Dielectric

.. - +-.

+ , ... - + ..
--+ -- -t-----,--
1. . - +--
-
.. - + ..

ll
Fm. 21-4. Uniform field between parallel plates--I, in vacuum; II, with dielectric.

The electric field produced at a point near a charged sphere depends upon
the amount of charge on the sphere and the distance of the point from it.
If the sphere is located in a vacuum and undisturbed by neighboring charges,
the field intensity at a distance r is
8 = k Q_ = - 1~ Q ,
r2 4 7rf:o r 2
as in § 15-6, where Eo is the permittivity of free space. If similarly located in
a medium of permittivity E, the field intensity can be expressed as
8 = _l_Q_
4 7rE r 2
The ratio of the permittivity E of the medium to the permittivity Eo of free
space may be called the relative permittivity of the medium and defined by
E
E,. = -. (21.9)
Eo

The relative permittivity Er is a pure number and has a characteristic value

for every imbi-tanee. Th1rn, for a medium having a relative permittivity of 2,
372 ELECTRICITY & MAGNETISM Chap. 21

the permittivity e is twice that of vacuum, and hence the intensity of the elec-
tric field at the reference point above, distant r from a charge Q, is half as
much as it would be in free space. The relative permittivity of vacuum is
unity, and for air at standard temperature and pressure is 1.000586.
The displacement of the charges that occurs in the molecules of a dielectric
when polarized in an electric field corresponds to a so-called displacement
current. It persists only as long as these charges are shifting-that is, while
the electric field across the dielectric is changing.
21-8. The Capacitor. A capacitor or condenser consists essentially of two
conductors separated by some insulating medium; the conductors are often
called the plates and the insulation the dielectric. Capacitors are used chiefly
to reduce arcing at contact points, to neutralize the effects of inductance, to
tune radio circuits, and to obtain pulses of current for various purposes. In
an automobile ignition circuit the condenser is ordinarily composed of two
long strips of tinfoil separated by waxed paper and rolled into cylindrical form.
The type commonly used in radio reception consists of two sets of aluminum
plates separated by air, one set being mounted on a shaft so it can be rotated
for changing the overlap or effective area of the plates.
A capacitor can be charged by connecting its plates to the terminals of a
battery or other source of direct current. This will cause electrons to leave
one plate and flow to the other through the source and associated circuit, until
a state of equilibrium is reached in which the plates have a potential difference
equal to the emf of the battery. In this process the plates will receive equal
charges, one positive and the other negative. If a battery of higher emf were
used, the charges would be correspondingly larger; consequently, the ratio of
the charge on either plate to the potential difference between the plates would
remain the same. This ratio of the charge on one of the plates to the potential
difference between them is the capacitance of the condenser; it is given by
Eq. 21.7 as

It follows from this expression that a condenser has a capacitance of one farad
when a charge of one coulomb of electricity produces a potential difference of one
volt across its plates. The farad is an enormously large unit of capacitance,
and, for convenience, smaller units, the microfarad (abbreviated µf) and the
micro-microfarad (µµf), are generally used. One farad = 106 µf = 1012 µµf.
Oceanic cables have large amounts of capacitance. The conductor at the
center serves as one plate of a capacitor; the sheathing, together with sea
water, forms the other plate; and the gutta percha or other insulating layers
between them serve as the dielectric. Such a cable 2000 miles long has a
capacitance of about 0.001 farad.
21-9. The Parallel-plate Condenser. The capacitance of a parallel-plate
condenser is determined entirely by the dimensions and properties of the di-
§ 21-9 THE PARALLEL-PLATE CONDENSER 373

electric and is not affected, for instance, by the materials used for the plates,
provided only that these are electrical conductors. This will be proved for a
capacitor, Fig. 5, which consists of conducting plates separated by a dielectric
layer of thickness s, effective area A, and
permittivity e. It is assumed that the
plates are large compared with the thick- +
ness of the dielectric.
When the plates are charged and the
potential difference between them is V,
the electric field in the dielectric is uni-
form and has the value given by Eq. 15.8 FIG. 21-5. Dimensions of a parallel-
as 8 = V /s. The polarization of the di- plate condenser.
electric produced by this field is repre-
sented by the number of lines of electric flux '¥ that extend from one plate to
the other per unit of area; this is the same as the electric flux density

Since by § 15-7 the number of flux lines is numerically equal to the charge Q
on either plate, and the flux density is D = E8, it follows that D = Q/A and
D = eV /s. Hence
v
-Q = e-·
A s
Finally, the capacitance of the condenser, being the ratio of the charge on one
of the plates to the difference of potential between them, is

c = -
Q= -
eA
.
V 8

If the dimensions of the condenser are given in meters, the capacitance will
be in farads.
The ratio of the permittivity e of a dielectric to the permittivity Eo of empty
space,- called the relative permittivity (Er), is also known as the dielectric con-
stant of the dielectric; the dielectric constant of a material is defined as the
ratio of the capacitance of a condenser with that material as dielectric to its
capacitance when the dielectric is a vacuum. When e is replaced by e0 Er, the
expression for the capacitance of a condenser becomes

C=EoErA, (21.10)
8

where = 1
eo 4 1r 9 X 109 numerically.

The result also affords a simpler unit for permittivity than used previously.
Since the dielectric constant is a pure number, the permittivity unit is the
374 ELECTRICITY & MAGNETISM Chap. 21

farad per meter. In summary, the different ways in which the unit of per-
mittivity can be expressed are as follows:
coulomb coulomb2 farad
€ = = = --·
volt· meter newton· meter 2 meter
Some typical values of dielectric constant are given in Table I.

TABLE I. DIELEC'l'RIC CONSTANTS

Glass, crown ....... . 5 to 7

Glass, flint ......... . 7 to 10
India rubber ....... . 2.1 to 2.3
Mica .............. . 5.7 to 7
Paper, dry ......... . 2 to 2.5
Paraffin wax ....... . 2 to 2.3
Teflon ............. . 2.0 to 2.05
Water (pure) ....... . 81

Find the capacitance of a condenser formed of 21 square metal plates measuring

10 cm along each edge, separated by sheets of mica 0.01 cm thick and having a dielec-
tric constant of 6. Alternate plates are connected to one terminal of the capacitor
and the remaining plates are connected to the other terminal.
The 20 dielectric sheets may be regarded as equivalent to a single one having an area
of 20 X 10 X 10 = 2000 cm2 = 0.2 m2 • The dielectric thickness is 10-4 m. Hence
th e capac1'tance of t h e con d enser 1s
. given
. b y E q. 21 .10 as 7T' X 6 XX 0. 2 X _4
4 9 109 10
= 0.106 X 10-6 farads = 0.106 µf.

By the use of a ballistic galvanometer, § 18-8, the capacitance of a con-

denser can be compared experimentally with that of a standard capacitor.
The capacitors are charged separately from the same battery and each is dis-
charged in turn through the instrument, the maximum throw being noted for
each. With constant potential difference the deflections are proportional to
the charges; these, in turn, are proportional to the capacitances, and therefore
the capacitances are in the same proportion as the deflections.
21-10. Energy of a Charged Capacitor. When a condenser is connected
across a battery, it charges very quickly, and the current falls from an initially
high value to zero within a fraction of a second. Its value at any instant
during this brief period is
. - cJ!J.
i - dt'

dq being the small amount of charge that is transferred in the time interval dt,
when this interval is vanishingly small. Since the capacitance C of the con-
denser is constant, Eq. 21.7 shows that this current can be written as

• = C dv (21.11)
" dt'
§ 21-11 CAPACITORS IN PARALLEL AND IN SERIES 375

proving that the momentary value of the charging current is proportional to

the time rate of change of potential difference, dv/dt, across the capacitor.
The amount of energy supplied to a capacitor in charging it can be deter-
mined by evaluating the work dW done in transferring a small charge dq from
one plate to the other when the potential difference between them is v; this is
found to be dW = v dq according to Eq. 15.,5. If a series of amounts like this,
or its equivalent v i dt, is summed up during the entire charging period t in
which a potential difference V is being established across the condenser, the
result is given by

W =
t=t
( vi dt =
Jt=O
f i•=V
vC dv dt = [Cv
V=O dt 2
2
J,
V

or w = ~ cv 2• (21.12)

When the capacitance C of the condenser is expressed in farads and the

potential difference V is in volts, the energy W stored in the condenser will be
given in joules.
Since the energy of a charged condenser resides in its electric field, it is of
interest to express the foregoing result in terms of field quantities rather than
circuit quantities. On the assumption that the energy is stored uniformly in
a dielectric of area A, thickness s, and permittivity E, the energy per unit
volume of the field is:
1. CE 2 Q V
Energy density = ~ = .!. - -,
As 2 A s
where the fraction Q/ A is the same as the density D of the electric lines of
force, and the potential gradient V / s is the same as the electric field in-
tensity 8. Finally, since D and 8 are related by D = E8, it follows that in an
electric field the
n2 (21.13)
Energy density = ~ '

and is in joules per cubic meter when D is expressed in

coulombs per square meter.
21-11. Capacitors in Parallel and in Series. Conden-
sers ~re often connected in parallel in a circuit in order
to increase the capacitance of that circuit. Let G\, C2 ,
C:i, · · · be the capacitances of several capacitors that are F1u. 21-6. Capaci-
com1ected in parallel, as in Fig. 6, to form a condenser of tors connected m
equivalent capacitance C. The potential differences parallel.
across the condensers will be represented by V1, V2, V3,
· · · , and the charges on them by Q1, Q2, Q3, · · · . The corresponding
potential difference and charge for the equivalent capacitor will be taken as V
and Q, rrn;pectivcly.
376 ELECTRICITY & MAGNETISM Chap. 21

For the parallel connection, each condenser has the same potential differ-
ence as the source, or
V = V1 = V2 = V3 = · · · ,
and the total charge is distributed among them, or

hence, by division,
Q = Qi + Q2 + Q3 • • • = Qi + Q2 + Q3 +
V V Vi V2 V3

But C = ~' Ci = ~:, and so on, from Eq. 21.7; and consequently,

C = Ci + C2 + C3 + ··· , (21.14)
which shows that the combined capacitance of several condensers connected
in parallel is equal to the sum of the individual capacitances.
Condensers are sometimes connected in series in
order to lessen the potential difference across each of
them; the result of this arrangement is to reduce the
capacitance. In the series connection of Fig. 7 the
same momentary flow of electrons occurs in all of
the condensers; this gives each an equal charge, or
Fm. 21-7. Capacitors
connected in series. Q = Qi = Q2 = Q3 = ' ' ' ,
and the applied potential difference is divided among the individual con-
<lensers:
V = Vi + V2 + V3 +
Then
V1 + V2 + v3 + . ' • = V1 + V2 + v3 +
Q Q1 Q2 Q3

or l=l..+1-+1..+ ... (21.15)

C C1 C2 C3 '
an expression which shows the relation between the capacitances C1, C2,
C3, · · · of the individual condensers and the equivalent capacitance C, when
these are connected in series.
Two condensers having capacitances of 2 µf and 6 µfare connected in series across
a 24-volt direct-current source. Find the charge of each capacitor and the potential
difference across each.
The combination is equivalent to a single condenser of capacitance C such that
~= i +~'whence C = 1.5 µf. The charge of this capacitor would be Q = 1.5 X 24
= 36 microcoulombs, and this is the charge of each individual capacitor in the series
§ 21-11 CAPACITORS IN PARALLEL AND IN SERIES 377

connection.
- . . . 36 microcoulombs
The potential difference across the 2-µf condenser 1s 2 . f d
micro ara s
= 18 volts, and that across the 6-µf condenser is obtained similarly as 6 volts.

In verifying experimentally the potential distribution in circuits containing

condensers in series, it is necessary to use voltmeters that do not require
current for their operation; a useful instrument for this purpose is the so-called
electrostatic voltmeter in which the deflection depends upon the force action
between charged plates.

PROBLEMS
1. The current in a particular power circuit falls uniformly from 1100 amp to 200
amp within a period of 0.6 sec. The effect of this change in a nearby circuit is to
induce an emf of 7.8 volts during the period. What is the mutual inductance of the
two circuits?
2. The mutual inductance of two circuits is 0.05 henry. Assume the current i
(amperes) in one of them to vary with time t (seconds) in accordance with the relation
i = 20 sin 400 t. What is the expression for the emf e (volts) induced in the other
circuit?
3. A coil of wire has an inductance of 0.15 henry and a resistance of 4 ohms. What
emfs will be induced in this coil when the current in it changes at the rates of
(a) 50 amp/sec, and (b) 20 amp/sec?
4. A long coil formed of a single layer of wire has an inductance of 5 millihenrys.
The wire is removed and replaced with a single layer of wire having one-third the cross-
sectional area of the removed wire. If the new winding has the same axial length as
the original one, how much inductance will it have?
5. An induction coil has a primary winding of 180 turns and a secondary winding
of 27,000 turns; the latter surrounds the former and the constant of magnetic coupling
will be taken as 0.8. A current of 5 amp in the primary sets up a flux of 0.004 weber
in the iron core, and the current is repeatedly interrupted in order to produce a suc-
cession of sparks across the secondary terminals. Determine the mutual inductance
of the windings, and the average emf that is produced in the secondary winding when
the primary current falls to zero in 0.002 sec.
6. Compute the inductance of the ring solenoid of Prob. 1 in Chap. 19.
7. An iron core, forming a closed magnetic circuit such as shown in Fig. 19-6, has a
cross-section of 2 square in. and an average length of 39.4 in. The winding consists
of 800 turns of wire, and it is energized by a current of 2.5 amp. If under these con-
ditions there is a magnetic flux of 0.006 weber in the core, what is the inductance of
the winding, and the relative permeability of the iron?
8. Show that if a number of air-core solenoids are geometrically similar in all
respects, including the size and spacing of the wires, then the inductances of these
solenoids are proportional to their linear sizes, as implied by the unit for permeability
-the henry per meter. Illustrate.
9. How fast will the current grow in the coil of Prob. 3 at the instant: (a) when it is
connected to a 12-volt storage battery, and (b) when the current has reached one-third
of its ultimate value?
10. A circuit which has 10 ohms of resistance and 1 henry of inductance has im-
pressed upon it a unidirectional potential difference of 100 volts at the instant t = 0.
What will be the current in this circuit when t = 0 sec, 0.1 sec, 0.3 sec, and 1.0 sec?
Plot the results.
378 ELECTRICITY & MAGNETISM Chap. 21

11. The field coil for one pole of a 4-pole generator has a resistance of 24 ohms, and
has an inductance of 3.6 henrys when carrying a current of 1.3 amp. (a) How much
energy resides in the magnetic field of the generator when the current has the value
stated? (b) How much energy is wasted in heating the field coils every second?
12. Show that the time constant L/ R of a circuit can be expressed in seconds.
What is the time constant of the field winding of the preceding problem?
13. What is the energy density of the magnetic field in the core of Prob. 7?
14. The energy associated with the magnetic field around a large electromagnet
carrying a current is 400 joules. The current is then reduced to half its former value,
and this brings about an increase of 30 per cent in the inductance of the coil. By what
percentage has the energy of the field been reduced?
15. A ballistic galvanometer is used to measure the capacitance of a condenser.
The galvanometer is first calibrated and found to produce a throw of 180 divisions of
1 mm length for a quantity of 0.001 coulomb. The capacitor is charged to 120 volts
and then discharged through the galvanometer; the throw is measured as 64 divisions.
Compute the capacitance.
16. The earth may be regarded as a conducting sphere that is isolated in space.
Refer to Prob. 25 of Chap. 3 and determine the capacitance of the earth.
17. Two metal spheres are isolated from each other and individually charged. One
sphere, of radius 5 cm, is given a charge of + 1.3 X 10-7 coulomb and the other, of
radius 10 cm, is given a charge of +.s X 10-s coulomb. The spheres are next brought
into contact, so that they assume the same potential, in which process one transmits
charge to the other. If care is taken that no charge is lost, what is the resulting
charge on each when they are separated?
18. The capacitance in farads of a condenser formed of two concentric spherical
metal shells, measured in meters, is found to be
c= f:r r1 r2 '
9 X 109 (r2 - r1)
where e, is the relative permittivity of the dielectric between the shells, and r 1 and r2
are the radii of the inner and outer shells respectively. (a) Show that this expression
is consistent with Eq. 21.10 when the effective radius of the dielectric is the geometric
mean of the radii of the shells. (b) Show that this expression is consistent with
Eq. 21.8 when the outer shell is of infinite size.
19. Calculate the capacitance of a condenser formed by two circular metal plates
1 m in diameter with a uniform layer of air between them 1 cm thick.
20. A small tuning condenser consists of 29 semicircular plates 1.50 in. in diameter
and the air spacing between the plates of opposite polarity is 0.011 in. What is the
maximum capacitance of this condenser, obtained when the plates are fully inter-
leaved?
21. A 2-µf condenser of the type used in telephone circuits is formed of two strips
of tinfoil 3.5 in. wide and 23 ft long and two strips of treated paper 0.0005 in. thick.
The four strips are placed one upon another and rolled into a compact spiral, the tinfoil
forming one terminal being insulated from the other everywhere by one thickness of
paper. What is the dielectric constant of the treated paper?
22. A condenser of 1 µf capacitance is charged by a 100-volt source, which is then
removed. This condenser is next connected across another one identical with the
first, except that it is initially uncharged. Precautions are taken to prevent loss of
charge by leakage during the experiment. Compare the initial and final states of the
two condensers with regard to charge, potential difference, and stored energy. If
there is a change in the total energy, how can the change be accounted for?
 PROBLEMS 379

23. A condenser consists of two brass plates each 20 cm long and 15 cm wide,
separated by a layer of air 2.5 mm thick. The condenser is permanently connected
across a 4000-volt battery. Find (a) the charge on the condenser and (b) the energy
stored in it. A large sheet of glass 2.5 mm thick and having a dielectric constant of
7.0 is next inserted between the plates, completely filling the space between them.
(c) How much additional charge will the condenser now take from the battery, and
(d) Vl"hat will be the total energy stored in it?
24. A 12-µf and a 4-µf condenser are placed in series and connected across a 600-volt
source. Determine the charge on each condenser and the potential difference across
each one.
25. Two condensers, one of 2 µf and the other of 6 µf capacitance, are connected in
parallel, and this combination is connected in series with a third condenser of 4 µf
capacitance across 240-volt direct-cunent service mains. Find the charge of each
condenser and the potential difference across each.
26. A capacitance of 3 µf is desired in a certain test, but the capacitors on hand are
only of 2 µf capacitance. What is the smallest number that would be needed, and
how would they be connected to obtain the desired capacitance?
 22
ALTERNATING CURRENTS
Alternating currents are of special importance because most electric power
stations nowadays supply alternating-current service. The reason for this
lies in the great usefulness of two pieces of equipment: the transformer and
the induction motor. Although very common, these are nevertheless re-
markable devices-the transformer because it operates without any moving
parts and the induction motor because, in its simple form, it involves no
electrical connection whatever to the rotating part. The fundamentals of
alternating currents-their production, characteristics, circuit relations, and
utilization-are presented in this chapter. Included are the basic principles
of some alternating-current machines and instruments.
22-1. Production of Alternating Emf. The development of an alternating
emf in a coil of wire when rotated in a magnetic field is explained in § 20-3.
During a half-revolution of the coil each conductor moves through the mag-
netic flux in one direction and in the next half-revolution it moves through it
in the opposite direction, and hence the emf generated in the coil is alternat-
ing. The coil itself and the load circuit connected to it by means of slip
rings form a closed circuit, in which the alternating emf sets up an alternating
current.
In the bipolar field each revolution of the coil causes all its conductors to
sweep past the north pole as well as the south pole of the field structure, and
this action generates a complete set of emf values, called a cycle. The time
required to produce a cycle is known as the period of the emf, and the number
of cycles completed per second is called the frequency.
In a four-pole magnetic field only half a revolution is needed for the con-
ductors of the coil to sweep past a north and a south pole, and so a cycle of
emf values will be produced in a half revolution, or two cycles in a revolution.
Hence, in general, with a field structure having P poles (necessarily an even
number) and with the coil making n revolutions per second, there will be
P /2 cycles per revolution, and the frequency will be
f = Pn (22.1)
2
cycles per second. In the United States the standard frequency is 60 cycles
per second.
380
§ 22-2 SINUSOIDAL EMF' AND CURRENT 381

It makes no difference in the end result whether the coil rotates in a sta-
tionary magnetic field or the magnetic field rotates with respect to a sta-
tionary coil, for the emf is induced in the coil whenever there is relative
motion between the two. Most alternators today have revolving field
structures and stationary armature coils. This construction makes it easier
to provide adequate insulation for the armature windings, in which high
potentials must be developed so that electric power can be transmitted eco-
nomically.
22-2. Sinusoidal Emf and Current. To investigate how the emf of an
alternator varies from moment to moment during a cycle, attention will be
directted to a single coil of wire revolving
at constant angular speed in a bipolar
D
field. In Fig. 1 the coil is shown end-on
with its two conductors at A and D,
moving around the dotted circle between
the poles NS. As already indicated in s
connection with Fig. 20-3, the induced
emf will be zero when the coil passes posi-
tion AD, and will have a maximum value A
when it passes the position 90° farther Fw. 22-1. Coil rotating in a uniform
along. At some intermediate coil posi- magnetic field.
tion such as A'D', making an angle f3
with the initial position AD, the induced emf will have a value between these
extremes of O and Em; this can be found by resolving the linear velocity v of
the conductor into two components. Only the component v sin f3 is effec-
tive in causing the conductor to cut magnetic flux; hence the emf generated
as the coil moves through position A'D' is
e = Em sin /3,
where e denotes the instantaneous emf value.
If the time required for the coil to make one complete revolution is T, and
the time for it to turn through the angle f3 is t, then
t . /3
T = 2 ,/
and the uniform angular speed of the coil in radians per second can be ex-
pressed as
/3 2 7l"
w = t
= T = 2 7l"f,

where f is the frequency of the emf. Consequently f3 = wt = 2 1l"jt, and the

instantaneous emf generated can be written as
c = Em sin wt = Em sin (2 1l"jt). (22.2)
This equation shows that the emf generated by a coil rotating at constant
382 ELECTRICITY & MAGNETISM Chap. 22

speed in a uniform field can be represented by a sine curve plotted with respect
to time.
When such an alternating emf is generated in a circuit, and when no other
emfs are acting in it, the current established has the same frequency as the
emf and undergoes similar variations in intensity. It follows that the instan-
taneous current i is related to the maximum value Im in the same way-that
is, i = Im sin {3, or
i = Im sin wt = I rn sin (2 1r.ft). (22.3)

Hence a sine curve, as shown at the left in Fig. 2, may be used to represent'
either the alternating emf of the generator or the alternating current in the
circuit.
A sine curve can be constructed by moving a point uniformly around a
circle and projecting its successive positions upon the vertical diameter of
the circle, § 7-9; the projections will then determine the corresponding ordi-

] I Effective
:::, value
u T
0 271"

c
......
G)

:,
v
.......
0
...0
G)

::,
O"
(/)

0 71" 271"
0° 180° 360°
Fm. 22-2. Sine wrve of alternating Fw. 22-a. Diagram uc;etl in determin-
emf or current plotted on axis of time ing the effective value of alternating
or angle. current.

nates of the curve. It is usual to represent an alternating emf or an alternat-

ing current with the aid of a radius extending to such a rotating point. The
radius is a vector having a length equal to the maximum ordinate of the sine
curve, and is assumed to turn counterclockwise, making one revolution per
cycle. The rotating vector R in Fig. 2 represents the sine curve shown at the
left; its vertical projection, after a rotation of {3 from the zero position at A,
gives the value of the emf or current at instant t, as shown.
A 60-cycle alternating current of sinusoidal wave shape has a maximum value of
30 amp. Find the momentary value of this current at an instant 0.0025 sec after
passing in a positive direction through the zero value.
Since the angular velocity w = 2 7l" X 60, the instantaneous current value for
§ 22-3 EFFECTIVE VALUES 383

t = 0.0025 is given by Eq. 22.3 as i = 30 sin (2 1r 60 X 0.0025) = 30 sin 0.3 1r =

30 sin 54.0° = 24.3 amp.
22-3. Effective Values. One might well inquire how a definite numerical
value can be given to an alternating current, when it actually has positive
and negative values extending from zero to the maximum value that corre-
sponds to the extremes of the sine curve. An evaluation is made possible by
observing the heat produced by the current, because the heating effect does
not depend upon current direction. In such a test a direct current of 1 amp
is passed through a resistor immersed in water, and the heat produced in a
given time is measured. If the test were repeated with alternating current
of different magnitudes, a value could be found which would produce the
same amount of heat in the resistor during the same time interval. This
alternating current is then said to have an effective value of 1 amp. Thus,
in any given conductor, one ampere of alternating current will produce the
same amount of heat in a given time as one ampere of direct current.
Jhe rate of developing heat in a circuit is proportional to the square of the
current, and, since the value of an alternating current at any instant is
i = Im sin {3, the instantaneous rate of heating will be proportional to
i2 = I m 2 sin2 /3.
The curves of current and current-square for a cycle are shown in Fig. 3 at
the top and bottom respectively; the current-square curve is similar to the
current curve but of double frequency, and its time or angle axis is displaced.
The average amount of heat developed is proportional to the average ordi-
nate of the current-square curve, and this value is obtained by calculating
tlie area of one lobe of the curve and dividing the result by the width of the
lobe. The lobe area is

1;2 l~
d{3 = m2 sin2 {3 d{3 = I m2 [~ - sin42 /3I = ~ I m2·

Since the width of the lobe is 1r radians, the average ordinate is! I m 2• Hence
in Fig. 3, the average value of the current-square curve is shown as half the
maximum ordinate. The square of the steady current I, which would pro-
duce the same heating effect as the alternating current of IJ?.aximum value
Im, is

and the square root of 12 is the so-called effective value or root-mean-square

(rms) value of the sinusoidal current. Hence the effective value is

I=~·

Alternating emfs are related in the same manner; if the maximum value of a
sinusoidal emf is represented by Em, the effective value is E,,,/y2. In sum-
384 ELECTRICITY & MAGNETISM Chap. 22

mary, the effective values of current and emf are

I=0.701Im, (
(22.4)
E = 0.707 Em. )

Whenever alternating electrical quantities are expressed it is understood

that effective values are meant; thus, an alternating current of 10 amperes
means an effective current of 10 amperes (its maximum or peak value is
14.1 amp, if sinusoidal). Alternating-current ammeters and voltmeters are
calibrated to indicate effective values.
22-4. Phase Relations. The application of an alternating emf to a circuit
that contains resistance only causes an alternating current to be set up that
keeps step with the emf at every instant. When
E the emf is zero, the current will be zero; when the
emf has its maximum positive value, the current
will also have its maximum positive value, and so
on-the current is in phase with the emf. This sit-
e
w
I
uation is shown in Fig. 4, wherein the curves of
90° 270° emf E and current I cross the time axis at the
same instants and in the same direction. The
Frn. 22-4. Phase relations of
time axis may, therefore, be marked in degrees,
current and emf in a circuit
with only resistance. with one cycle of emf (or current) values repre-
senting 360°. Since the sinusoidal curves repre-
sent cyclic motion with constant angular speed, the angle swept through is
proportional to time, for t = (} / w as in § 22-2.
In most circuits the current set up by an alternating emf, although of the
same frequency, docs not rise and fall at the same instants as the emf applied
to the circuit. As a result there is a time interval between the maximum

....s::
.. E

1
R L G)

::,

c£::J
u
"ti Time
s::
c '' / O I
..... '' /
/

E #

I 11 m
Fm. 22-5. Inductive circuit; the current lags the impreRsed emf.

positive value on the emf cycle and the maximum positive value on the cur-
rent cycle; also, between the zero values as the curves sweep through them in
the same direction. This time interval represents an angular separation of
the emf and current curves and is called a phase difference; it is commonly
expressed in degrees.
§ 22-5 INDUCTIVE CIRCUITS 385

In a circuit that contains a resistance R and an inductance L, such as that

shown in part I of Fig. 5, the current I lags the impressed emf E, as indicated
in part IL In a circuit that contains a resistance Rand a capacitance C, such
as shown in part I of Fig. 6, the current I leads tho impressed emf E, as indi-
cated in part II of that figure. The angle of lag or lead is represented by 8.
The phase relations shown by the sine curves in parts II of Figs. 5 and 6 can
be indicated more conveniently by rotating vectors, as shown in part III of

-......
c: I E

c~
(I) _,

---y;
/
:,
u /
/
'' Time 0 I
"tJ '

l
c:
0
.....
E
w O

I Il m
Fw. 22-6. Capacitive circuit; the current leads the impressed emf.

these figures. The vectors are usually drawn to scale to represent the effec-
tive values
_., rather than the maximum values of emf and current, and they
are assumed to rotate counterclockwise about point O with the uniform angu-
lar speed w = 2 1rf. As they rotate, the angle () between them remains un-
changed; it represents the angle of lag or lead of the current with respect
to the emf.
22-5. Inductive Circuits. It was pointed out in § 21-2 that, when the cur-
rent in an inductive circuit is changing, an emf is induced in that circuit.
This emf, called the emf of self-induction, depends upon the inductance of the
circuit and the rate of change of current. Its value at any instant is ex-
pressed by the equation eL = L ::, where Lis the circuit inductance, and ~:
is the instantaneous rate of current change. With an alternating current in
the circuit, the continuous changing of the current causes the emf of self-
induction to be alternating also. To show that this is so, the expression for
an alternating current will be substituted in the foregoing equation and then
differentiated with respect to time. Since, from Eq. 22.3, the instantaneous
current is i = Im sin wt, it follows that :: = Im w cos wt, and consequently
the emf of self-induction becomes
CL = wLI m cos wt.
This emf equation represents a cosine curve, which has the same shape as
the sine curve of current, but is displaced 90° from it. To verify this fact,
observe that any value of time t which would make sin wt a maximum in the
equation for current would make cos wt zero in the foregoing equation for
emf, and vice versa. By Lenz's Law, the emf of self-induction is directed in
386 ELECTRICITY & MAGNETISM Chap. 22

such a way as to oppose the change of current which produces it; thus, when
the current is increasing, this emf is opposite in direction to the current, and
when the current is decreasing, the emf has the same direction as the current.
Therefore, the emf of self-induction lags the current by 90°.
Evidently an inductive alternating-current circuit contains two emfs in
series: e from the alternator that supplies the circuit, and eL resulting from
the inductance of the circuit. The current at any instant is equal to the
resultant emf divided by the resistance of the circuit. This statement may be
+
wn"tten i · = e R eL ore = i "R - eL, b ut It
· must b e remem b ered t h at t h"IS IS
· a
vector equation and means that e is the resultant of iR and -e£. The re-
lations among these quantities can best be explained by a diagram like that
in Fig. 7, in which the current vector i is taken as a datum, or reference

c E

_k
-el Cl)
....
....
::,
u
""O
iR c
0 I
....E
0

el w -
I
oo 180°
FIG. 22-7. Vector dia- Frn. 22-8. Phase relations in a circuit
gram for an inductive with only inductance.
circuit.

direction, along the horizontal axis. The potential drop iR due to resistance
is in phase with i; the emf of self-induction eL lags i by 90°, as shown; -eL is
equal and opposite to eL and therefore leads i by 90°; and finally the resultant
of iR and -eL gives the supply emf e. It will be observed that the supply
emf leads the current-that is, the current lags the supply emf.
Inductance in a circuit not only makes the current lag the emf but chokes
it down to a smaller valv,e than if the inductance were not present. The chok-
ing effect will be computed by imagining a circuit to contain inductance only,
the resistance being negligible. In such a circuit the current I lags the sup-
ply emf by 90°, as pictured in Fig. 8, and the emf of the source is equal and
opposite to the emf of self-induction. Therefore, the supply emf is numeri-
cally e = wLI m cos wt, and its maximum value Em occurs when cos wt = 1,
whence Em = wLI m· A similar relation will be true for the effective values
of emf E and current I, or E = wLI. Since w = 2 1rf, § 22-2, it follows that
the current in amperes established in an inductance of L henrys by an emf of
E volts and frequency f cycles per sec is

I= _!L_
2 1rJL
§ 22-7 THE SERIES ALTERNATING-CURRENT CIRCUIT 387

The quantity 2 1rJL has the same position in this equation as has resistance
R in Ohm's Law, and it accounts for the choking action produced by induc-
tance. It is called the inductive reactance of the circuit; in symbols
XL= 2 1rf L, (22.5)
where the inductive reactance XL is expressed in ohms when the inductance
L is in henrys.
22-6. Capacitive Circuits. A condenser that is connected across a source
of alternating emf becomes charged alternately in opposite directions, and
electrons surge to and fro in the connecting wires. This means that there
is an alternating current in them, and for this reason alternating current is
said to "flow through" a capacitor, even
though the capacitor is supposed to i: +
~
have a perfectly insulating dielectric. :5
u
An alternating-current ammeter, § 22- -u 0
c
13, placed in such a circuit would show c
a steady deflection. The capacitor ] - ~90'~
presents a certain reactance to an I
alternating current which is somewhat 00 270°
similar to the· reactance of an induc- Frn. 22-9. Phase relations in a circuit
tance coil; it is called capacitive reac- with only capacitance.
tance, and its value can be derived
froni Eqs. 21.11 and 22.2 in a manner similar to that used in the preceding
section. The capacitive reactance is found to be
1 (22.6)
Xe= 2 1rJC'
and is expressed in ohms when the capacitance C is in farads and the fre-
quency f is in cycles per second.
If the resistance of the circuit is disregarded, then the current I in the
capacitor will lead the emf E of the source by 90°, for when this emf starts
to decrease from its maximum value, the condenser will begin to discharge in
the opposite direction. These relations are shown in Fig. 9, from which it
will be seen that the current changes from + to - at the instant the emf
recedes from its maximum + value.
22-7. The Series Alternating-current Circuit. It has been shown that the
alternating current in a circuit containing only resistance is in phase with the
emf of the source, that the current in a circuit containing only inductance
lags that emf by 90°, and that the current in a circuit containing only capaci-
tance leads that emf by 90°. Where all three of these elements of a circuit
are combined in series, the current will be the same throughout; its value will
depend upon the resistance and upon the net reactance of the circuit.
Figure 10 shows in part I a series alternating-current circuit, having the
constants R, L, and C, connected to an alternator of emf E; and in part II
388 ELECTRICITY & MAGNETISM Chap. 22

the potential differences across these elements drawn with respect to the
phase of current as the datum. The vector terminals are indicated by num-
bers which correspond to those on the circuit diagram; thus, vector 2-3 rep-
resents the potential difference IXL across the circuit points 2-3 between
which is connected the inductance L.
The potential difference across the resistance R is in phase with the current
and is shown as IR, that across the inductance L leads the current by goo and
is shown as IXL, and that across the capacitor C lags the current by go 0 and is
shown as IXc. The net reactive drop in potential IX is the difference be-
tween the capacitive drop and the inductive drop, and is either IXc - IXL
or IXL - IXc, depending on which is the greater. This reactive drop is
added at right angles to the resistance drop to obtain the entire potential
drop marked IZ, which is equal to the emf E of the source.

E IX,
.....-------< ~ I------~ 4 IXL

Phase of
-------~
current
2
I II
Frn. 22-10. Series circuit that contains resistance, inductance, and capacitance, together
with vector diagram of potential differences.

The total opposition presented by the circuit may be expressed in a man-

ner similar to Ohm's Law as the ratio of the magnitudes of emf and current,
namely
E
z = -, (22.7)
I

where Z is called the impedance of the circuit and is expressed in ohms. I ts

value can be obtained from the figure as

Z = VR 2 + X 2 = VR 2 + (XL - Xc) 2•
The magnitude of the current in the circuit is given by the general equation

I= E (22.8)
VR 2 + (XL - Xc) 2
and lags the emf by an angle
a = t _
v ___ [1,n 1
XL R:
- Xe,. (22.9)
§ 22-8 THE PARALLEL ALTERNATING-CURRENT CIRCUIT 389

The values of the individual reactances are given in the two preceding
sections as XL = 2 1r f L and Xe = 2 1r\ c' where f is the frequency of the
source of potential E. It is interesting to note that, in circuits having large
values of L, the reactance XL is large and the current is small, but that in
circuits having large values of C, the reactance Xe is small and the current is
correspondingly large.
To illustrate the solution of a circuit problem, consider a 100-volt, 500-cycle source
of alternating emf acting in a series circuit having a resistance of 60 ohms, an induct-
ance of 0.012 henry, and a capacitance of 20 µf. Find the impedance of the circuit,
and determine the current and its phase.
The inductive reactance is XL = 2 1r 500 X 0.012 = 37.7 ohms, and the capacitive
reactance is Xe = 1/(2 1r 500 X 0.00002) = 15.9 ohms; hence the net reactance is
37.7 - 15.9 = 21.8 ohms (inductive). The impedance of the circuit is Z =
V (60) 2 + (21.8) 2 = 63.8 ohms. The current established in the circuit is 100/63.8
2 8
= 1.57 amp, and it lags the emf by the angle() = tan-1 ~~ = 20°.

An inspection of Eq. 22.8 shows that when the inductive reactance and
capacitive reactance of a series circuit are equal, that is, when
1
2 1rfL = 2 1rJC ,

the current in the circuit will have its greatest value, namely I = E/R, and
_will be in phase with the supply emf. Such a condition is called resonance.
The frequency at which resonance occurs is evidently
1
(22.10)
f = 2 1rVLC

and is called the natural frequency of the circuit. When a circuit is in reso-
nance the drops in potential across the inductive and capacitive elements of
the circuit may be considerably greater than the emf applied to the circuit.
For this reason resonance is to be avoided in power circuits: In radio receiv-
ers, however, the condenser is adjusted for resonance in order to give maxi-
mum signal strength.
22-8. The Parallel Alternating-current Circuit. The simplest way of
handling a parallel circuit with alternating currents is to find the current in
each branch by the procedure outlined in the previous section, and then add
the several currents vectorially to find the total current of the parallel com-
bination. Finally the impedance of the combined circuit is found by dividing
the emf impressed upon the circuit by the total current.
The two-branch circuit of Fig. 11 will serve as an example. Branch 1 contains a
resistance R1 of 60 ohms and an inductance L of 0.012 henry, and branch 2 contains a
resistance R 2 of 60 ohms and a capacitance C of 20 µf. The emf E applied to the circuit
has a magnitude of 100 volts and a frequency of 500 cycles/sec. Determine the cur-
390 ELECTRICITY & MAGNETISM Chap. 22

rent in each branch circuit, the total current taken by the parallel circuit, and the total
impedance.
The reactance of branch 1 is X 1 = 37.7 ohms (inductive) and that of branch 2 is
X2 = 15.9 ohms (capacitive), just as in§ 22-7. The impedances of these branches are
respectively Z 1 = V (60) 2 + (37.7) 2 = 70.8 ohms, and Z 2 = V (60) 2 +
(15.9) 2 = 62.1
ohms. Hence the currents are 11 = 100/70.8 = 1.41 amp, which lags Eby the angle
01 = tan-13;~7 = 32.1°, and 12 = 100/62.1 = 1.61 amp, which leads Eby the angle
() 2 = tari-1 15.9 = 14.80 . The vector sum of these currents is indicated in part II and
60
is found by trigonometry to be 2.77 amp at an angle lagging Eby 7°. Consequently,
the impedance of the parallel circuit is Z = E/1 = 100/2.77 = 36.1 ohms.

"-' E

--- -
n
FIG. 22-11. Parallel cireuit that contains resistance, inductance, and C'apacitance, together
with vector diagram of component currents.

If, in a parallel circuit of the type considered, the inductive reactance of

one branch equals the capacitive reactance of the other, the circuit is said to
be in parallel resonance. \iVith this condition, the currents in the two branches
will be nearly opposite in phase and the total current will be very small; con-
sequently, the impedance of the circuit as a whole will be much greater than
that of either branch.
22-9. Power an~ Power Factor. 'The power developed at any instant in
an alternating-current circuit is the product of the instantaneous emf and
the instantaneous current. For practical purposes it is necessary to obtain
the average power developed and to express it in terms of the effective values
of emf and current.
Figure 12 shows the emf and current curves for typical alternating-current
circuits. The phase relations in a purely capacitive circuit, wherein the cur-
rent I leads the emf Eby 90°, are shown in part I. From a to b, the emf and
current are both positive, and when their instantaneous values are multiplied
point by point along the time axis, a positive lobe of the power curve P is
obtained; this lobe represents energy stored in the capacitor. From b to c,
the emf is positive but the current is negative, and their product gives a
negative lobe in the power curve; this lobe represents energy being returned
to the circuit as the capacitor discharges. During the negative half of the
§ 22-9 POWER AND POWER FACTOR 391

emf cycle from c to d to e, the situation is similar; consequently, the power

curve is alternately positive and negative, and, since the lobes have the same
area, the net energy expended in the circuit is zero. A comparable result is
obtained with an inductive coil of negligible resistance; here the current lags
the emf by go0 • Evidently, a current that is go0 out of phase with the emf
results in zero power expenditure.
A different condition exists in a circuit that contains only resistance. The
current and emf in such a circuit are in phase, as shown in part II of the
figure, and whether both instantaneous values are positive or both are nega-

...
Cl)
...
Cl)
... +
Cl)
3: 3: 3:
0
a. + 0 0
a. a.
"'CJ
c: -0
c: "...:fc::
--...
0
c:.
~
I
I '
~
0
c:
!...
0

i
::> ,t
v ::> \"'-. _/;I'/ ::>
.....-
v \ v
....: _
' .... .... / ...:-
w
E E E
w w

II Ill
Fw. 22-12. Curves of emf, current, and power plotted with respect to time for: I, capaci-
tive circuit; II, resistive circuit; and III, inductive circuit ,vith resistance.

tive, their product will have a positive value. The pm,ver curve is similar to
that shown in part I but is entirely above the axis, and the area under it rep-
resents energy that is dissipated as heat.
For circuits in which the current is neither in phase nor 90° out of phase
with the emf, the power curve will have positive lobes that are larger than
the negative lobes, as in part III; the difference between their areas represents
the amount of energy expended.
The corresponding vector diagrams for these typical circuits are given
respectively as parts I, II, and III of Fig. 13, in which E and I represent the
effective values of emf and current. The average power supplied to a circuit
.,,,

I lcoso£E
---
8 .... ,
I
E

II Ill
Fw. 22-13. Vector diagrams corresponding to the sine curves of Fig. 12 for alternating-
current circuits.
392 ELECTRICITY & MAGNETISM Chap. 22

is found much more readily from such vector diagrams than from the areas
of the power curves indicated in the preceding figure. It is necessary only
to compute the component of the current which is in phase with the emf and
to multiply it by that emf in order to determine the power. In part I the
current I has no component in phase with E, and hence the power supplied
is zero; in part II the current I is in phase with E and the power is P = E I;
in the general case of part III the component of current I which is in phase
with E is I cos e, and consequently the power is
P =EI cos fJ. (22.11)
Naturally this equation applies to the preceding circuits for which e is 90°
and 0° respectively.
Since cos e is a factor by which the product EI must be multiplied in order
to give power, it is known as the power factor of the circuit. Hence,
p
Power factor = cos e = EI, (22.12)

so that the power factor is defined either as the cosine of the angle of lag or
lead, or as the ratio of the watts to the volt· amperes. Reference to Fig. 10
shows that the power factor of a circuit can be expressed also as the ratio of
its resistance R to its impedance Z. For a circuit containing resistance only,
the phase angle is 0, so that cos e = 1; hence the power factor is unity, and
the power is P = EI, as for direct-current circuits.
A highly inductive circuit has a low power factor, as the current lags the
emf considerably; a highly capacitive circuit also has a low power factor, as
the current leads the emf by a large angle. Low pmver factor is a disadvan-
tage, because on the ordinary constant-potential circuit it necessitates a
relatively large current in order to supply a given amount of power, and the
large current causes more waste of energy in heat. The rating of electrical ma-
chines is limited chiefly by their ability to dissipate the heat produced in them;
for this reason alternating-current machinery is rated in kilovolt· amperes
(abbreviated kva) instead of kilowatts.
An inductive load takes 8 amp when connected to 2200-volt, 60-cycle supply mains.
The power factor has the low value of 0.75, and improvement is sought by connecting
a 2-µf capacitor in parallel with the load. Find the resulting power factor of the
combination.
To help in visualizing the problem, the student is advised to construct a vector
diagram and to use the applied emf E as a basis of reference. The current in the
inductive load lags the applied emf by the angle cos-1 0.75 = 41.4°. It has a compo-
nent of 8 cos 41.4 ° = 6.00 amp in phase with E, and a component of 8 sin 41.4 ° = 5.29
amp, which lags Eby 90°. The capacitor has a reactance of 1/ (2 1r X 60 X 2 X 10-6)
= 1327 ohms and takes a current of 2200/1327 = 1.66 amp, which leads E by 90°.
The resultant current has two components. One of these is 6.00 amp in phase with E.
The other is found by subtracting the capacitor current from the lagging component
of the inductive load current; it equals 5.29 - 1.66 = 3.63 amp and lags Eby 90°.
§ 22-10 THE ALTERNAT/NG-CURRENT GENERATOR 393

The resultant current thus lags Eby the angle tan-1 (3.63/6.00) = 31.2°; hence the
power factor of the combination is cos 31.2° = 0.855.
22-10. The Alternating-current Generator. The stationary and rotating
members of an alternator are called respectively the stator and the rotor;
generally the stator is the armature and the rotor is the field structure.
Alternators must provide an emf that has not only the desired magnitude but
also the desired frequency-a characteristic that is controlled by the speed
of the rotor and the number of poles. High-speed alternators usually have
two poles, while those driven at slow speeds have many pairs of poles. The
rotational speed at which an alternator gives its rated frequency is called
the synchronous speed; its value is given by Eq. 22.1 as n = !f rev per sec,
where f is the frequency in cycles per second, and P is the number of field poles.
The armature is designed to have a sufficient number of coils or conductors
to develop the desired emf at the synchronous speed. The value of the emf
generated in each conductor is obtained from Eq. 20.3 in terms of the mag-
netic flux density B, the length l of the conductor, and the linear speed v of
cutting the flux. This emf has the maximum value of Em = Blv and is
assumed to be sinusoidal; consequently, the effective value of the generated
emf per conductor is
E -_ V2,
Blv (22.13)

where Bis in webers per square meter, l in meters, and v in meters per second.
The simple form of alternator shown in Fig. 1 generates a single alternating
emf; it is called a single-phase machine to distinguish it from the more usual
type of alternator that generates two or three alternating emfs at the same
time and which is called a polyphase alternator.
These emfs have the same magnitude and have +
definite phase relations with each other.
The two-phase alternator consists of two inde- !!
pendent armature windings with a definite angular g
separation; in a bipolar tield they are at right angles 'i
to each other. When such a machine is driven, an w
alternating emf is generated in each winding; and,
because of the relative positions of the windings,
the emfs are go 0 apart in phase. These emfs are
Fm. 22-14. Emf curves
plotted with respect to time in Fig. 14. The ma- of a two-phase alterna-
chine has four terminals in order that it can supply tor.
a ~ur-wire circuit; the potential difference across
one pair of wires is displaced go 0 from that across the other pair, as indica-
ted. Sometimes one wire is common to the two phases.
The three-phase alternator consists of three like windings symmetrically
394 ELECTRICITY & MAGNETISM Chap. 22

placed on the armature so as to produce three equal emfs that are 120° apart
in phase, as shown in Fig. 15. The windings are connected in Y or in Li,
§ 17-15. With either connection, the total
+ power developed in a three-phase alternator,
when the load is the same on all three phases,
is given by
P = V3 El cos e, (22.14)
where E represents the emf acting across any
pair of lines, I is the line current, and () is the
phase angle between E and I.
Fm. 22-15. Emf curves of a three- 22-11. The Induction Motor. The most
phase alternator. widely used alternating-current motor is
the polyphase indnction motor. It consists
essentially of a stationary field structure and a rotating element resembling
a squirrel cage. The rotor is usually formed of heavy copper bars welded to
end rings and has no connection with the supply circuit. Part I of Fig. 16
shows the arrangement of a two-phase induction motor. The field windings
on alternate poles are connected to one phase, and the windings on the
others are connected to the other phase of the supply circuit. The direc-
tions of the currents through the motor during their positive half-cycles are
indicated by the arrows.

...
c:
...~
a... ~-+---+--+--
_g
Phase A 0
V)

I
I
I
I

Phase B
i
I II
Frn. 22-16. Diagram of a two-phase induction motor and its current and flux relations.

The motor currents are shown graphically by curves IA and le in part II

of the figure, and the direction- of the magnetic flux which they set up through
the rotor is represented by the arrows below the curves. At the instant indi-
cated by 0, I A = 0 and le has its maximum negative value; hence only the
horizontal poles (phase B) will be energized and the current through their
windings will produce a fl~x that is horizontally directed toward the right.
After l- cycle, le = 0 and IA has its maximum positive value, at point 1 on
§ 22-12 TRANSFORMERS 395

the curves; now only the vertical poles (phase A) will be energized and the
flux will be directed upward. At the instant corresponding to 2 on the curves,
IA = 0 again and IB is positive, directing the flux horizontally toward the
left, and so on. In this manner, the same effect is produced upon the rotor
as if the flux through it were rotating mechanically. The result is to induce
currents in the rotor bars, and these currents, reacting with the flux, develop
a torque which makes the rotor follow the rotating field.
The rotor does not rotate as fast as the field; if it did, no magnetic flux
would be cut by the rotor bars and no current would be set up in them. The
difference between the actual rotor speed and the synchronous speed of the
rotating field is called the slip of the motor. As the load on the motor is
increased, the machine slows down a little, thereby increasing the slip and
causing larger currents to be induced in the rotor; these larger currents set
up a greater torque, which enables the motor to drive the increased load.
The three-phase induction motor is similar in construction and performance
to the two-phase machine described, but it has three sets of windings for con-
nection to a three-phase supply circuit; it is the type most frequently used.
The polyphase induction motor operates as a constant-speed machine like
the direct-current shunt motor. The power supplied to an induction motor
can be computed from Eq. 22.14.
22-12. Transformers. In transmitting electrical power over long dis-
tances it is of advantage to use large potential differences, because a given
amount of power can be transmitted with a correspondingly small current.
This results in reducing the energy wasted in heating the transmission lines
and permits the use of relatively
small line wires. Transformers are Core
used to change the potential differ-
ence from one value to another; in To supply
power transmission they step it up mains Load
to a high value at the alternator end
of a line and then step it down at
Primary Secondary
the other end to a value suitable for
the apparatus in the consumer's Fm. 22-17. Connections of a transformer.
premises.
The transformer consists essentially of two coils of wire, entirely separate
electrically but wound upon the same core of laminated iron, as represented
in Fig. 17. The primary winding is connected to the power supply mains,
and the se~ondary winding is connected to the load circuit. The alternations
of the primary current set up an alternating flux in the core, and the continual
building up and collapsing of this flux induces an emf in the secondary coil.
The value of this emf depends upon the number of turns Ns of the secondary
coil and upon the rate of change of flux ~f webers per second through its
turns, in accordance with Eq. 20.2, which gives e. = Ng ~f volts.
396 ELECTRICITY & MAGNETISM Chap. 22

The variations of flux which produce the secondary emf also affect the
primary coil of NP turns and induce in it an emf ep = NP~~ volts. This
emf, by Lenz's Law, opposes the impressed potential difference, in somewhat
the same manner as the counter emf of a motor, § 20-7. This explains the
fact that a transformer under no load takes very little current from the supply
circuit, for the emf induced in the primary coil is very nearly equal to the
potential difference of the supply; for practical purposes the two may be con-
sidered equal. The foregoing expressions can be combined and changed to
effective values, yielding

(22.15)

which shows that the emfs in the transformer coils are directly proportional
to the numbers of turns on their windings.
Upon connecting a load across the secondary winding, the emf induced in
that coil will set up a current. This current will oppose the magnetizing
effect of the primary current and will reduce the flux in the transformer core
slightly. As a result, the counter emf induced in the primary winding is
lessened, and more current is taken from the supply mains. In this manner
the input to a transformer automatically accommodates itself to the output.
The efficiency of a transformer may be expressed as the ratio of the output
volt· amperes to the input volt· amperes, since the transformer itself has
little effect upon the power factor of the circuit. Hence
,
Effi ciency Es f s
= -.- - . (22.16)
Eplp
The efficiency of transformers is very high, as might be expected from the
absence of moving parts, and usual values range well over 95 per cent. If
the efficiency were taken as 100 per cent, the input and output volt· amperes
would be the same, and it follows that the ratio of primary and secondary
currents would be inversely proportional to the numbers of turns on the re-
spective windings.
22-13. Alternating-current Measurements. The instruments most com-
monly used in alternating-current circuits are those which measure current,
potential difference, and power. The d' Arsonval type of instrument described
in §§ 18-8 and 9 is not suitable for use with alternating current, for the coil
tends to follow the rapid reversals of the current and the pointer merely
oscillates near the zero mark of the scale.
The iron-vane ammeter makes use of the fact that iron is attracted by a
coil carrying current regardless of the direction of the current, and hence will
be attracted if the current is alternating. Current in a stationary coil attracts
a soft-iron vane, which, as it moves, swings a pointer over a scale. The
iron-vane voltmeter differs from the ammeter in that it has a high resistance
in series with the coil to permit direct connection across the supply mains.
§ 22-14 THE SYNCHRO 397

The dynamometer type of ammeter has two current coils in series, one of
which is stationary. The other coil carries a pointer and is pivoted at right
angles to the field of the first, being normally held in this zero position by a
coiled spring. An alternating current through the instrument reverses
simultaneously in both coils, and the resulting torque on the moving coil is,
therefore, always in the same direction. When a high resistance is con-
nected in series with the coils,· this type of instrumen'.t can be used as a volt-
meter.
The power delivered to a load connected in an alternating-current circuit
cannot be measured with a voltmeter and an ammeter; for this purpose it is
customary to use a wattmeter. This instrument, described in § 18-10, oper-
ates with alternating as well as with direct currents. The torque producing
the deflection is proportional to the instantaneous values of potential differ-
ence e and current i, and hence is proportional to the product of e and i.
With sinusoidal waves of emf and current, and the latter lagging by the
angle (), the instantaneous values are e = Em sin wt and i = Im sin (wt - ());
hence the torque is proportional to
e i = Em Im sin wt X sin (wt - e).
The deflection of the instrument is proportional to the average product of ei,
which, following the procedure in § 22-3 by integrating over a half-cycle, is
found to be

(ei)av = Em2Im COS()= EI COS()= P,

and shows that the deflection is proportional to the power delivered.

The power factor of a motor or other load circuit can be determined by
measuring the power with a wattmeter, the current with an ammeter, and

Wattmeter
Ammeter Current coil

Potential
coil Load

High resistance

FIG. 22-18. Connections for measuring the power factor of a, load.

the potential difference with a voltmeter, connected as shown in Fig. 18.

The power factor is obtained by dividing the watts by the volt· ampere
proquct.
22-14. The Synchro. The field of instrumentation concerns itself not only
with .instruments for measuring physical quantities, such as those considered
398 ELECTRICITY & MAGNETISM ,Chap. 22

in the preceding section, but also devices for controlling industrial processes.
A typical example of these devices is the Selsyn, a self-synchronizing arrange-
ment for controlling angular position at a distance; it is commonly called a
synchro.
The synchro is a combination of two or more identical units, each of which
has a stator and a rotor as in an induction motor. The connection plan for
two units is indicated in Fig. 19. The stators are wound for three phases,
and the terminals A, B, and Care connected to each other in proper order.
The rotors R 1 and R2 are single phase and connected to a common supply of
alternating current in order to develop magnetic fields that are synchronous.

c B
8 c
120V l20V
Frn. 22-19. Scheme of Synchro as position controller. The rotors are supplied from a
common supply of alternating current.

When the rotor R1 of the sending synchro has assumed a definite position,
determined by some machine element to which it is coupled, emfs will be
induced in the stator windings by transformer action. These set up currents
in the three-phase circuit that extends to the stator of the receiving synchro
and produce a torque that will cause its rotor R2 to turn. This movement
continues until the rotor has the same position as that of the sending synchro.
When this position is reached, the emfs developed in the two stators will be
equal, and there will be no current in the connecting lines. When the
machine element moves and turns rotor R1 through some angle, then rotor
R2 will turn through the same angle, just as though both were mounted upon
a common shaft that might be a few miles long. Commercially available
synchros have an accuracy of about one minute of arc.

PROBLEMS
1. A 24-pole alternating-current generator delivers a sinusoidal curren\ having a
maximum value of 250 amp. What will be the instantaneous current value after the
machine has turned through -,h rev from the position at which the current has its
maximum value?
2. What is the period of an alternating current that has a frequency of 25 cycles/sec?
If this current were passed through the coil of an electromagnet, how many times per
second would the electromagnet tend to attract its armature?
3. The effective value of a sinusoidal alternating emf is 120 volts. Over what emf
range do its instantaneous values extend?
 PROBLEMS 399

4. The effective emf and current values of the alternating-current supply to a

capacitive circuit are respectively 120 volts and 10 amp, and once each cycle their
+
instantaneous values are 120 volts and -10 amp at the same moment. What is the
phase relation of the emf and current waves?
5. It is desired to have a resistor develop heat at the rate of 70 cal/sec when con-
nected across 120-volt alternating-current mains. What should its resistance be?
6. Two alternating-current generators operating at the same frequency are con-
nected in series. One generates 120 volts and the other 50 volts. What is the resultant
emf of the combination if the smaller emf (a) is in phase agreement with the larger,
(b) is in phase opposition to the larger, and (c) lags the larger by 90°?
7. A circuit that has an inductance of 0.1 henry carries a 60-cycle current. What is
the co}mter emf of self-induction in that circuit at an instant when the current has
(a) it::,jmaximum value of 5 amp, and (b) half its maximum value?
8. An inductive coil of negligible resistance draws a current of 3.5 amp when con-
nected across 120-volt, 25-cycle supply mains. What will be the current in the coil
when it is connected across 240-volt, 60-cycle mains?
9. A coil of 0.15 henry inductance and 15 ohms resistance is connected across a
220-volt, 60-cycle line. Find (a) the reactance of the coil, (b) its impedance, (c) the
current in it, and (d) the angle by which this current will lag the applied potential
difference.
10. A 2-µf condenser is used in a telephone set. Find its capacitive reactance
(a) for the "ringing current" of 16i, cycles/sec, and (b) for the "speech current"
averaging 1000 cycles/sec.
11. A 60-cycle potential difference of 120 volts is impressed across a condenser of
20-µf capacitance. Compute the reactance of the condenser and the current in the
circuit.
12. What should be the capacitance of a condenser so that it will have a capacitive
reactance of 5 ohms at (a) 500 cycles/sec, (b) 500 kilocycles/sec, (c) 500 mega-
cycles/sec?
13. At a certain frequency a series circuit has a capacitive reactance of 30 ohms, a
resistance of 80 ohms, and an inductive reactance of 90 ohms. What is the current
in the circuit when an alternating emf of the same frequency and of 240 volts effective
value is impressed on the circuit?
14. What is the impedance at 1 megacycle/sec of a series circuit containing a 1-µµf
condenser and a I-megohm resistor? What will be the result when the frequency is
10 kilocycles per sec?
16. A series circuit has a resistance of 25 ohms, an inductance of 0.01 henry, and a
capacitance of 2 µf. Calculate the impedance of the circuit at 1000 cycles/sec.
16. Determine the natural frequency of the circuit of Prob. 15, and also the current
that the circuit will take at that frequency with an impressed emf of 90 volts.
17. It is desired to vary the natural frequency of an oscillating circuit from 200 to
1000 kilocycles/sec. The circuit has an inductor of 0.5 millihenry and a variable
condenser. What range in capacitance should the condenser have?
18. What inductance connected in series with a 10-µf condenser will produce reso-
nance at a frequency of 25 cycles/sec?
19. Across 60-cycle supply mains are connected in series a condenser and an in-
ductive coil which have reactances respectively of 75 and 100 ohms. What should be
the capacitance of a series-connected condenser which, when added to the circuit,
would establish resonance?
20. In the solved problem of § 22-8, suppose the frequency of the applied emf to be
240 cycles/sec. Determine the current taken by the parallel circuit and the total
impedance.
400 ELECTRICITY & MAGNETISM Chap. 22

21. Find the power developed in a coil having a resistance of 150 ohms and an in-
ductance of 0.50 henry when connected across a 220-volt, 60-cycle supply line.
22. Find the power factor of each branch circuit and of the entire circuit for the
solved problem of § 22-8, the frequency being 500 ~ycles/sec.
23. A fluorescent lamp unit takes 0.75 amp when connected across 115-volt
alternating-current mains and draws 53 watts. When equipped with a condenser for
power factor improvement, the unit takes 0.50 amp and draws the same power.
Determine the power factor of the unit with and without the condenser.
24. The alternators at the hydroelectric power station at Bonneville Dam have
96 poles and develop 13,800 volts at 60 cycles/sec. At what speed do these water-
wheel generators revolve?
25. A three-phase balanced inductive load of 80 per cent power factor is supplied
with 60-cycle power over a three-wire circuit. The potential difference between the
wires is 220 volts at the load, and the current in each wire is 15 amp. What power is
supplied to the circuit? What is the angle between the current and emf for any phase?
26. A three-phase induction motor operating on 440-volt supply mains has a power
factor (lagging) of 0.82 and an efficiency of 86 per cent when delivering 30 hp. What
is the line current?
27. The alternators mentioned in Prob. 24 are three-phase machines and most of
them are rated at 60,000 kva. How much current will a generator of this rating supply
to the lines at full load? How much power can each supply with rated current when
the currents are 15° out of phase with the generated emfs?
28. A 10-kva transformer, having 800 primary and 80 secondary turns on its wind-
ings, is connected to 2200-volt service mains and delivers full-load currents to a non-
reactive load. Determine the currents in the windings. Neglect losses.
29. A coil of wire and a capacitor are connected in series to an alternator that
generates 120 volts at 1000 cyles/sec. At that frequency the coil has a resistance of
40 ohms and a reactance of 60 ohms, and the capacitor has a reactance of 90 ohms.
The impedance of the alternator itself may be neglected. (a) What should be the
reading of a wattmeter when connected to this circuit to measure the power delivered
by the alternator? (b) What should be its reading when the coil and capacitor are
connected in parallel?
 23
ELECTRONICS
Of the various divisions of physical science, perhaps no division of this
vast subject has developed faster and brought about more industrial applica-
tions than that dealing with electronic behavior. This field of Electronics
includes Thermoelectricity, which deals with contact potentials and their
depJ,idence upon temperature, Thermionics, which deals with the emission
of electrons from heated cathodes, and the new branch dealing with Semi-
conductors. The basic concepts and important applications of these branches
are presented in this chapter. Other topics that come within the scope of
Electronics, such as photoelectric and photovoltaic action, radio and micro-
waves, television and radar, are considered in other chapters.
THERMOELECTRICITY

23-1. Thermoelectric Effects. When two different metals are placed in

contact with each other they assume slightly different potentials. This
phenomenon, discovered by Volta, is now explained as a transfer of certain
of the free electrons present in all metals across the boundary between the
two; it is found that more electrons pass in one direction than in the other,
and consequently one metal becomes negative and the other positive. The
difference of potential between the metals, spoken of as contact potential dif-
ference, depends in amount upon the metals used and is also influenced by
the temperature of the junction.
This contact potential difference provides an explanation for three thermo-
electric effects which have been known for years. The most important of
these states that an emf will be produced in a circuit formed of two wires of
different metals when their junctions are at different temperatures. Another
states that an unequal heating of the junctions of a circuit formed of two
different metals will result when a current traverses the circuit. The third
states that an emf will be developed in a circuit formed of a single metal when
there is a temperature gradient along its length.
23-2. The Seebeck Effect. The first of the thermoelectric effects was dis-
covered by the German physicist Thomas J. Seebeck (1770-1831). This
effect is the production of an emf in a thermocouple formed of two different
metals when one of their junctions is at a higher temperature than the other.
401
402 ELECTRICITY & MAGNETISM Chap. 23

A circuit formed by joining an iron and a copper wire is shown in Fig. 1; a

galvanometer is included in order to indicate the amount of emf produced.
It will be assumed that the low-temperature junction is kept at 0°C while the
temperature of the other junction is raised. When the latter is, say, at room
temperature, the thermal emf developed sets up a current which, for the
particular circuit shown, is directed clockwise. As the second junction is
heated more and more, the thermal emf increases to a maximum value, then
diminishes to zero, reverses in direction, and again increases in magnitude.
The values of emf developed in such a copper-iron thermocouple are plotted
in Fig. 2; the curve is approximately a parabola.
The thermal emf for any specific couple between any two given tempera-
tures is always the same, provided the metals are of the same purity and
crystalline state. As implied, the emf is affected by any structural changes
that may be produced by mechanical and heat treatments.

+ 1.5
-
]
Cl)

+l.O
I
I
v
,,.... ........
'' ,,
',,..
E +o.5
Galvanometer ....- I I\
\
E O
UJ

i
''\
-0.5 '
I\

-1.0 \
0 200 400 600
Temperature, °C

Fw. 23-1. Thermoelectric circuit of two Jcrn. 23-2. Thermal emf diagram of a
metals. copper-iron couple.

The temperature at which the thermal emf of a couple has a maximum (or
minimum) value is called the neutral point, and the temperature at which
this emf reverses is called the inversion point. Each pair of metals has its
own characteristic neutral and inversion temperatures; for the copper-iron
couple these temperatures are shown in the figure to be approximately 205
and 480°C, respectively.
The slope of the thermal emf curve at any point is called the thermoelectric
power of the thermocouple at that temperature-an unfortunate use of the
term "power." For the copper-iron couple the thermoelectric power is 13.7
microvolts per degree at 0°C, and 3.7 microvolts per degree at 150°C.
The value of the emf produced by a couple formed of any two metals can
be computed from a knowledge of the thermoelectric powers of these mate-
rials with respect to some metal chosen as a standard, such as lead or plati-
num. For this purpose graphs of thermoelectric power against temperature
are useful. Graphs for a few metals against lead are shown in Fig. 3; they
§ 23-2 THE SEEBECK EFFECT 403

are straight lines over the temperature range shown. An example will clarify
the procedure.
Compute the emf developed by an iron-copper couple over the range from Oto 150°C.
Reference to Fig. 3 shows that the thermoelectric power for iron falls from a value
of 16.5 microvolts per degree at 0° to 8.1 at 150°C, and that the value for copper rises
from a value of 2.8 microvolts per degree at 0°C to 4.4 at 150°C. Hence the total emf
generated by an iron-lead couple over this temperature range is the average of 16.5
and 8.1 microvolts per C 0 multiplied by 150°, or 12.3 X 150 = 1845 microvolts; simi-
larly the total emf of a copper-lead couple over the same range is 3.6 X 150 = 540
microvolts. Consequently, the emf of an iron-copper couple would be 1845 - 540
= 1305 microvolts; also see Fig. 2.
The same result is obtained by computing the area between the graphs for iron and
copper in Fig. 3 over the temperature range from O to 150°C and observing that each
square represents 5 microvolts. Over this range iron is positive with respect to copper.

20 r,... ..._
~- - ... - ~- --
I I I

})
..._ ,.._
~--
fro,.,

- --
r-.. '--
,_.
J Use scale
I lat righf
-
.- -cadmium - -- -
..r.
:::,

-Copper -......
. . . . . 10 ,.._ ....... 50 ~

-- - -
Antimony
-- -
l
i-- i.- r-. al
'--
-c
...... .._ c
I I 0
.E o ...... Lead !'"---
5
>,.

- 0
E
f
u
'.E-10
Plati~um :.::
c
<(

so]
u
Cl)
f Use scale })
Bismuth lat right
1E - ........
!?
... 0
Cl)
i=-20
~
i--
-- -- -- i-- 1
Nickel
100 §
~
>

-,--.r--.1o-.

-- - - - ---
-30
0 100 200 300
Temperature, °C

Frn. 23-3. Thermoelectric powers of several metals ·with respect to lead (Computed from
data given in the International Critical Tables).

Measurements of radiant heat energy may be made with a multiplicity of

couples formed of two metals and placed close together; such a compact group
is called a thermopile. This device is often constructed of short lengths of
bismuth and antimony connected in series alternately, and put together in
zigzag fashion so that alternate junctions can be exposed to the radiation
while the intermediate ones are shielded from it.
The hot-wire ammeter, which utilizes a thermal junction in contact with a
hot wire, is often used for measuring small alternating currents. The heat
404 ELECTRICITY & MAGNETISM Chap. 23

evolved in the wire by the current which is being measured causes the thermo-
couple to generate an emf, thereby producing a deflection in a calibrated
instrument connected across the couple. Increased sensitivity can be
obtained by placing the wire and couple in a vacuum.
Since the emf of a thermocouple depends in a known manner upon the
temperatures of the junctions, the thermocouple can be used as a pyrometer,
for the measurement of temperature. Such an instrument is commonly used
to measure the temperatures of furnaces and of molten metals, care being
exercised to protect the thermal junction from furnace gases and from direct
contact with the molten substances.
23-3. The Peltier Effect. The French physicist Jean C. A. Peltier (1785-
1845) discovered that in a circuit formed of two different metals, the main-
tenance of current from an outside source of emf caused one of the junctions
to be heated and the other cooled. As compared with the Seebeck Effect,
the hot and cold junctions are interchanged for the same direction of current.
In the copper-iron couple of Fig. 1, when a current traverses the circuit at
ordinary temperatures in a clockwise direction, the left junction will become
heated and the right junction cooled. The extent to which the junctions are
heated or cooled by a given current depends solely upon the metals used.
For a copper-iron junction at room temperature, about 4 calories are de-
veloped per hour per ampere of current.
The heating or cooling due to the Peltier Effect is directly proportional to
the current which is passed through the junction. If a current of I amp is
maintained fort sec, and if the contact potential difference in volts caused by
electron diffusion at the junction is represented by the symbol II, then the
heating or cooling is given in joules by W = IIIt. This effect is small and is
often masked by the heating of both metals brought about by the Joule Effect,
which for a junction of resistance R ohms is given by Eq. 17.6 as W = Rl2t.
The combination of both effects at a junction can be expressed as
P = RI2 ± III;
where Pis the power in watts; the plus sign is used for Peltier heating, the
negative sign for cooling.
23-4. The Thomson Effect. A thermodynamic analysis of the foregoing
effects prompted William Thomson to predict that an emf must exist between
different parts of the same metal if they are at different temperatures. He
demonstrated that, if a uniform metal bar is heated at the middle and a
current is sent through it from end to end from an external source, the heat
is conducted unequally along the two halves. In a copper bar, Fig. 4, the
region A where current is directed from a colder to a hotter part will be cooler
than it would be if there were no current, and the region B where current is
directed from a hotter to a colder part will be warmer; thus B is warmer than
A. The same is true for cadmium, silver, and zinc, but the effect is reversed
in iron and nickel, to mention but a few metals. This effect and the emf in-
volved are named after Thomson. Lead shows no appreciable Thomson
§ 23-5 THERMIONIC EMISSION 405

Effect, and this accounts for its frequent use as a reference metal in Thermo-
electricity (see Fig. 3).
The evolution or absorption of heat at the junctions of a thermoelectric cir-
cuit demonstrates that there must be a difference of potential at places where
two dissimilar metals are in contact. If both junctions are at the same tem-
perature, the thermal emfs at the two junctions are in opposite directions and
annul each other; but if there is a temperature difference, these two emfs do
not balance, and their resultant, together with the Thomson emfs, establishes
a current in the circuit.

FIG. 23-4. Experiment to FIG. 23-5. Two-element

show the Thomson Effect. electron tube circuit.

THERMIONICS

23-5. Thermionic Emission. A discovery of far-reaching importance was

made by Edison in 1883. He experimented with an evacuated tube contain-
inf a heated filament and a separate electrode, and found that a current would
be set up between them if the electrode were positive with respect to the fila-
ment, but not if it were negative. In the light of present knowledge this
current is explained by the emission of electrons from the filament into the
space surrounding it, and the attraction of these emitted electrons to the
nearby electrode. The escape of electrons from the surface of a metal is
comparable in many respects to the escape of molecules from a liquid during
the process of evaporation. The situation is a little different, however, be-
cause an electron about to leave the metal induces a positive charge on the
surface behind it and hence is attracted backward toward the metal. This
action gives rise to the idea of a potential barrier, somewhat like surface ten-
sion, which must be overcome before the electron can escape. When the
metal is heated, the electrons in it are given more kinetic energy, which
assists them in passing the potential barrier at the surface. For each metal,
there is a definite minimum amount of energy needed to release an electron
from the surface; this energy is known as the work function of the metal.
The thermionic emission of electrons by a hot body is the operating principle
of the modern electron tube. The simplest of these is the two-element tube,
or diode. This device consists of an evacuated bulb or tube with a filament
somewhat like that of an incandescent lamp, and a separate metal plate, as
shown in Fig. 5. When the filament is heated to incandescence by battery A,
406 ELECTRICITY & MAGNETISM Chap. 23

it will emit electrons. These will be attracted to the plate when it is main-
tained positive by battery B, as indicated, and the galvanometer G will show
a deflection. If the plate were made negative by reversing battery B, the
electrons evaporated from the filament would be repelled by the plate and,
since no electrons would be emitted from the cold plate, the galvanometer
would not show a deflection. Hence the electrons can flow only from filament
to plate, or, what corresponds to the same thing, the conventional direction
of current, § 16-1, can be only from plate to filament. Consequently, this
electron tube acts as a rectifier, the plate being the anode and the heated fila-
ment the cathode.
The number of electrons emitted from the filament in a unit of time depends
upon the substance of which it is made and upon its temperature. The rate
of electron emission is generally expressed as the current per unit of surface
area of the hot body; at absolute temperature T this current density is ex-
pressed by
b
J = AT2 e-T (23.1)
'
as given by the English physicist Owen W. Richardson (1879-1945). Herein
J is expressed in amperes per square centimeter, Tis in degrees K, e is the
base of natural logarithms, and A and b are constants. The value of A is
60.2 for the materials used as emitters. The constant b incorporates the work
function of the material; its values found experimentally for three metals are:
molybdenum 50,900, thorium 38,900, and tungsten 52,400; dimensionally
these values of b are in degrees K.
The current from plate to filament of an electron tube, as expressed by the
foregoing equation, is the steady value that results when the plate potential
is high enough to sweep all the electrons from the region around the filament
as fast as they are liberated; this is the so-called saturation current. At lower
potential differences between the two electrodes of the tube the current will
be less, because some of the evaporated electrons are driven back into the
filament by the negative charge that builds up in the space near the plate.
The accumulation of electrons in this region of the tube is called the space
charge. In most types of electron tubes it is not expedient to measure the
saturation current, because its value is so large as to change the emitting
conditions or damage the tube.
When there is a copious supply of electrons in a diode and the potential
difference across its electrodes is much below the value needed to produce the
saturation current, then the current is limited by the space charge; it can be
expressed in terms of the applied potential difference E by means of the gener-
alized expression
(23.2)
where K and n are constants determined by the geometry of the tube. For
a vacuum diode with plane parallel electrodes, the value of n is l That this
§ 23-6 RECTIFIER TUBES 407

figure is appropriate can be appreciated from the facts that the speed of the
electrons reaching the anode is proportional to the square root of the potential
difference between the electrodes, § 18-12, and that the number of electrons
that are within the interelectrode space to neutralize the field around the
anode is proportional to the potential difference.
23-6. Rectifier Tubes. The two-element tube permits current to be estab-
lished through it in one direction only, and hence can serve as a rectifier of
alternating currents. The relation between the current in such a tube and
the potential difference between its filament and plate is indicated graphically
in Fig. 6. The curves show the operating characteristics of a vacuum diode
and a gas diode, the latter being used for rectifying the larger amounts of
current.
Gas diode

-......
c
(J)
T
:::> Vacuum
u
diode AC~
2 Input~
0
a:

2
Plate potential DC load circuit
FIG. 23-6. Characteris- FIG. 23-7. Full-wave tube rectifier
tic curves of vacuum and circuit.
gas diodes.

A rectifier that eliminates every other half-cycle of the alternating-current

wave is called a half-wave rectifier, and one which utilizes both half-cycles is
called a full-wave rectifier. The circuit of the full-wave rectifier is shown in
Fig. 7; it comprises an iron-core transformer T with three windings, and a
double-anode tube. The primary winding joins to the source of alternating
yurrent, the secondary winding 1-2 connects with the two anodes P 1 and P 2,
lnd the third winding supplies current to heat the filament F. ]\!Jid-taps 3
and 4 on the latter windings lead to the load circuit. At an instant when
terminal 1 of the lower winding is positive, the current is directed from P 1
to F, to 4, through the load, to 3 and back to 1. When terminal 2 is positive,
the path of the current will be from P2 to F, to 4, through the load, to 3 and
back to 2. Consequently, the load current will be unidirectional. Since this
current would pulsate between zero and a maximum value, a filter is used to
make the current steady, § 24-2. High alternating potentials can be rectified
similarly; a tube for this purpose bears the trade name of Kenotron.
In the gas-filled diode the gas atoms are bombarded by the electrons emitted
from the cathode and become ionized when electrons are knocked out of them.
These gas ions are positively charged and tend to neutralize the space charge
about the cathode; hence the emission of electrons from this electrode is in-
408 ELECTRICITY & MAGNETISM Chap. 23

creased. The Tungar rectifier and the mercury-vapor rectifier are examples.
The Tungar rectifier tube contains argon or other inert gas at low pressure,
a cathode of tungsten coiled into a closely wound spiral, and an anode of
graphite having a relatively large area. Tungar rectifiers are available for
half-wave as well as for full-wave rectification, and are often used for charging
small storage batteries from alternating-current service mains.
The mercury-arc rectifier contains an electrode of mercury and one of graph-
ite, within an evacuated container of glass or steel. To start the action, an
auxiliary electrode is touched for a moment to the mercury pool and then with-
drawn, striking an arc. Thereupon mercury vapor is formed which becomes
ionized by electrons proceeding from the mercury surface, and a current is
established between the electrodes. When the rectifier is connected to alter-
nating-current mains, there will be current only during those intervals when
the mercury electrode is negative; this fact accounts for the rectifying action
of the device. Two anodes are used for full-wave rectification from a single-
phase supply; more are used when converting from polyphase systems. Steel-
tank rectifiers, designed for currents up to several thousand amperes, are used
in supplying power for traction systems.
Other types of rectifiers, utilizing the properties of semiconductors, are
considered in § 23-16.
23-7. Three-element Electron Tubes. The idea of adding another elec-
trode to the two-element electron tube, in order to control conveniently the
number of electrons passing from the filament to the plate, came from the
American inventor Lee DeForest. The introduction of this so-called grid
electrode between the filament and the plate makes the electron tube more
versatile by enabling it to serve a number of
functions, especially in telephone and radio
circuits. The three-element tube and its
circuits then appear as in Fig. 8.
Input The effect of the grid is like that of a
shutter which, opening and closing, controls
the flow of electrons going through it from
the filament to the plate. This control is
accomplished by changing the potential of
Frn. 23-8. Three-element electron the grid. When the grid is positively
tube circuit. charged, it attracts electrons and increases
their flow from the filament to the plate, for
most of them pass through the relatively wide spaces between the grid wires.
When the grid is negatively charged it repels the electrons and they cannot go
to the plate. Consequently, when the grid G is made alternately positive and
negative by joining the input terminals to a source of alternating potential,
the electron flow from F to P is increased and decreased accordingly,
thereby varying the direct current in the plate circuit. The grid potential
might change thousands or millions of times per second and the plate current
§ 23-7 THREE-ELEMENT ELECTRON TUBES 409

would change accordingly. Actually, the grid is not made positive with
respect to the filament, but only more or less negative. This is done by in-
serting a so-called C battery, as shown, to "bias" the grid negatively; when
so biased there will be no current in the grid circuit. Thus, the grid serves as
a gate-valve to control the plate current while taking practically no power
itself.
The cathode of the tube is often a thin metal sleeve coated with thorium
or other material having a low work function; a heating coil of tungsten wire
is mounted within but separated from the sleeve. This construction makes
it possible to heat the cathode with alternating current without introducing
disturbing effects.

~ 1.2
Q)
a.
E 1.0 t----+--1--+
.Q
e o.s t-.--ii----+-+-+

t...... 0.6 t--+-+--1----f-l---+--+-+-

:::,

-
~
c
0.4 t-1---+--+--+----+-t-------+--H'--+-I--+-

0:: 0.2 1----.'I---V--t-r----+-,<-+---,l'+----,lf---'-,1'-

0 80 160 240 320 40(

Plate potential, volts

Fm. 23-9. Characteristic curves of a triode for a number of grid potentials.

The behavior of a three-element tube or triode is best portrayed by curves

of plate current plotted against plate potential for different grid potentials.
Such a family of curves is shown in Fig. 9 for a Type 6SF5 tube with grid bias
potentials E 0 ranging from O to -4 volts; the heater of the cathode of this
tube takes 0.3 amp with an applied emf E1 = 6.3 volts.
As mentioned in § 23-5, the space between the cathode and the plate has a
negative charge, because of the presence of electrons. The more negative the
grid, the more it assists this space charge in limiting the electron flow. The
curves show, for example, that, with the plate potential maintained at 200
volts, the plate current is 0.90 milliamp with -1.5 volts on the grid, 0.31 milli-
amp with -2.0 volts on the grid, and 0.10 milliamp with -2.5 volts on the
grid. This illustrates the large variation in plate current that can be obtained
by relatively small changes of grid potential.
Three constants of a tube are useful in determining its performance near
the operating point chosen from its characteristic curves. The first of these
is the amplz(ication factor, symbolized byµ, and is determined by the geometry
of the electrodes. The amplification factor is defined as the ratio of a change
 410 ELECTRICITJ' & MAGNETIS11J Chap. 23

in plate potential to the corresponding change in grid potential in the opposite

direction, such that the plate current will remain unchanged. Its value can
be written as
dEp' (23.3)
µ = -dE 0 ip const,

where EP and E 0 are the plate and grid potentials, and ip is the plate current.
Find the amplification factor of the triode having the characteristics shown in Fig. 9.
Choose the operating point to have potentials of -2.0 volts on the grid and 250 volts
on the plate, both with respect to the cathode; for these conditions the plate current
will be 0.90 milliamp. Again, with -1.0 volt on the grid and 150 volts on the plate,
the plate current will also be 0.90 milliamp. Thus, an increase of 1 volt on the grid
yields the same output with a decrease of 100 volts on the plate. The ratio of a
potential change of 100 volts on the plate to 1 volt on the grid gives the amplification
factor µ a value of 100.
The second constant of a tube is the plate resistance, rP; it is defined as the
ratio of a change in plate potential to the change in plate current when the grid
bias is kept constant. The third constant, called the transconductance, gm, is
defined as the ratio of the change in plate current to the change in grid bias
when the plate potential is kept constant. The three tube constants are re-
lated by the expression µ = r P gm·
The current that is drawn from the cathode in a three-element tube can be
expressed in the same way that Eq. 23.2 gives it for a diode, also under the
condition that the current is limited by the space charge. This means that
the current in a triode is determined by the electrostatic field near the cathode,
. and since this field in the presence of a grid is dependent both upon the grid
potential E 0 and, to a lesser extent, upon the plate potential EP, it is found
that the plate current can be expressed by

ip = K ( E0 + IP: )3 12
when·the quantity within the parentheses is positive; when this quantity is
negative, however, the current is zero.
23-8. Amplifiers. The three-element tube is widely used for amplification
purposes in communication circuits. The current to be amplified is supplied
suitably to the input or grid circuit of the tube, and the amplified current is
made available in its output or plate circuit; the additional energy is supplied
by the battery in the plate circuit.
In analyzing the operation of the tube a:s an amplifier, an alternating emf
e0 of the form Em sin wt ( § 22-2) is considered to be applied between the cathode
and grid of the triode shown in Fig. 8. At instants when e0 is positive, more
electrons travel to the plate, and the current in the output circuit increases.
Conversely, at instants when e0 is negative, the output current decreases.
The effect is the same as adding an alternating emf to that of the battery in
the plate circuit. If the amplification factor of the tube is µ, the equivalent
§ 2:3-9 GAS 'PUBES 411

value of this alternating emf in the plate circuit is µe 0 , and its frequency is
the same as that of e0 •
The output circuit may then be regarded as a simple series circuit in which
the alternating emf is µe 0 , the plate resistance is rp, and the load R is a pure
resistance. Hence the alternating current in the load is
. - µeq
ip - rp +R
and the alternating potential difference across the load is
.R µe 0 R
eR = ip = rp + R.
Consequently, the power supplied to the load is ipcR = iiR, or
p = R(
rp
µcg
+R
)2· (23.4)

The quantities ip and CR are alternating and should not be confused with the
unidirectional current and emf produced by the battery in the plate circuit.
If the load has an impedance Z rather than merely a resistance R, then in
Eq. 23.4 the denominator should be replaced by (rp + R) 2 + X2, where Xis
the reactance of the load at the frequency of the applied potential c0 •
In communication circuits the gain in power produced by an amplifier is
measured in terms of a unit called the decibel (abbreviated db). One decibel
represents a ratio of the two power values of 10°·1 and 10° or unity-a ratio
of 1.26 to 1. When the gain of an amplifier is N db, the ratio of the output
P2 to the input power P 1 is 10°-IN, or
P2 100.lN.
P1 -
From this relation, the gain in decibels becomes

N = 10 log-.
P2 (23.5)
P1
The decibel scale is also used in the measurement of sound and noise levels
and is explained more fully in § 26-3.
As illustrations of the decibel scale in expressing amplification, an amplifier will be
considered to have an input of 5 microwatts. A gain of 10 db would give a power
ratio PdP1 = 10°·1x1o = 10; consequently, there would be a 10-fold increase and the
output would be 50 microwatts. Similarly, a gain of 60 db would signify a 106-fold
power increase to 5 watts. Again, if the output is 300 microwatts, then the gain in
decibels is found from Eq. 23.5 to be N = 10 log (300/5) = 10 log 60 = 10 X 1.778
= 17.8 db.

23-9. Gas Tubes. The addition of a grid to a gas-filled diode results in a

tube which has found unusually wide application in electrical communication,
control systems, and power circuits. Such a tube is often called a gas dis-
412 ELECTRICITY & MAGNETISM Chap. 23

charge tube; one of the trade names is the Thyratron. The advantage it has
over the vacuum triode is the very much larger plate current that it yields
for the same potential difference.
The outstanding feature of the three-element gas tube is the fact that the
plate current is not established until the plate potential reaches a certain
critical positive value (determined by the characteristic curve of the tube and
the amount of grid bias) at which the gas ionizes. After the current is started,
its value does not change when the grid potential changes, as it does in a
vacuum triode, but continues irrespective of grid bias until the plate potential
drops practically to zero or the current falls so low that ionization of the gas
ceases. The difference of potential between plate and filament need be only
about 15 volts to maintain ionization, and this potential difference is not
appreciably affected by the plate current.

c
Load

r, L
Filament
transformer c
Fra. 23-10. A thyratron circuit for con- Fra. 23-11. Electron-tube oscillator
trolling current. (Hartley type).

The Thyratron tube and its circuits are shown in Fig. 10. The usual plate
battery of a triode is replaced by an alternator, and the filament F is heated
by alternating current supplied through a transformer. The grid circuit ex-
tends from the transformer tap a through the bias battery Ee and the resistor
Ra to the grid G. This resistor serves to limit the grid current when the gas
ionizes; it is necessary even when the grid bias is negative, for then the grid
draws positive ions that result from the ionization. With the grid bias posi-
tive, as shown, the grid current is that due to electron flow. Current through
the Thyratron is established only during the half-cycles of emf from the
alternator which make the plate P positive, but the current through the load
does not commence in each half-cycle until the plate potential reaches the
critical value for the grid bias employed.
With alternating potentials on both plate and grid, the latter electrode can
regain control once each cycle and delay the starting of current in the plate
circuit as long as the grid is sufficiently negative. Therefore, the phase rela-
tion between these two potentials fixes the point in each cycle at which current
§ 23-10 ELECTRON-TUBE OSCILLATORS 413

is established and thereby determines the average amount of current in the

plate circuit.
The I gnitron resembles a mercury-arc rectifier utilizing an auxiliary elec-
trode of silicon carbide or other high-resistance material for initiating ioniza-
tion during each positive half-cycle. The tube current is controlled by chang-
ing the point in the cycle when "ignition" occurs (that is, when ionization
begins); this is done by connecting a Thyratron to the auxiliary electrode and
shifting the phase of grid potential on the latter tube. Ignitrons are used for
controlling large currents, such as are necessary in electric welding.
23-10. Electron-tube Oscillators. Alternating current can be generated
with the three-element vacuum tube by coupling the plate and grid circuits
so that some energy in the plate circuit is fed back to the grid circuit. The
alternating currents produced are usually oscillations of high frequency, and
consequently the tube with its circuit is called an oscillator. Fig. 11 shows
the form of oscillator devised by Ralph V. L. Hartley; the condenser C and
the coil L constitute the oscillating circuit.
The operation of the oscillator may be explained as follows: The filament F,
heated by current from the battery A, emits electrons, and these pass to the
plate P, establishing a current in the plate circuit and developing a magnetic
field in coil L2. The growing field about L2 cuts the turns of the other half
of the coil, inducing an emf in L 1 exactly as in a transformer. The momentary
direction of the counter emf in L2 is such as to make the intermediate point O
positive with respect to b. Similarly the emf induced in L1 will be such as to
make point a positive with respect to 0, and therefore positive to b also.
Hence, as the current in the plate circuit grows, it is aided by the positive
potential pulse imparted to the grid G by L1, § 23-8. As the plate current
reaches a limiting value and no longer continues to rise, however, the emf
induced in L1 becomes less, and very soon falls to zero. Under these condi-
tions the plate-circuit current must decrease, because the positive potential
which was available at the grid is now lacking. As the plate current falls
in value, the emf induced in L1 is reversed in direction and increased in mag-
nitude. This reversal of the potential difference across L1 aids the reduction
of the plate current until it reaches zero. When this value is attained, the emf
induced in L 1 is again zero and the plate current begins to rise, because the
negative grid potential necessary to keep the plate current zero no longer
exists. Once more the current rises in the plate circuit and produces an emf
in L 1 directed to make G positive; this action aids the current to increase until
the steady value is reached. Thereafter the cycle of events is repeated, and
continuous oscillations are established.
The capacitor c is a by-pass around the battery B for the high-frequency
pulses to the plate. The battery Ee is added to the circuit to permit selection
of the most efficient operating condition-one that results in a minimum loss
in both grid and plate circuits.
The frequency of alternating rnrrcnt produced by the oscillator is deter-
414 ELECTRICITY & MAGNETISM Chap. 23

mined by the inductance Land capacitance C of the oscillating circuit and is

given by Eq. 22.10 for the natural frequency of the circuit.
To illustrate, suppose the oscillating circuit to be composed of a coil of 0.06 milli-
henry inductance and a variable condenser. What should be the capacitance of the
condenser to generate alternating current of 600,000 cyles per sec?
The required capacitance is

C = (2 :/)2£ (2 7r 600,000)\ 0.06 X 10-3 = 1.17fi X 10-9 farad = 1175 µµ£.

If the alternating current from an oscillator is to be used for electrical com-

munication, it is necessary to vary the current in accordance with the sounds
to be transmitted. In this process, called modulation, § 24-3, the alternating
current is molded by varying the amplitude, or the frequency, or the phase.
Most radio broadcasting stations use amplitude modulation, and some use
frequency modulation.
23-11. Multigrid Tubes. The electrodes of electron tubes behave exactly
like the plates of a condenser, and the vacuum or gas within the tube serves
as the dielectric. Of these small interelectrode capacitances, the one of great-
est importance is the capacitance between grid and plate, because it is most
disturbing in amplifier operation.

I -...
c:
Q)
...
:::,
R u
Input 2
0
a:

Plate potential

FIG. 23-12. Connections of a four- F1G. 23-13. Volt-ampere char-

element tube (tetrode). acteristic of a tetrode.

The grid-plate capacitance of a triode can be reduced by placing an addi-

tional grid between the original grid and the plate. The original one is then
called the control grid and the added one the screen grid. In such four-element
tubes or tetrodes, the screen grid virtually shields the control grid electro-
statically from the plate and lessens the feedback of energy and the inter-
ference.
The elementary circuit of a tetrode is shown in Fig. 12, wherein K is the
eathode, P the plate, CG the control grid, and SG the screen grid of the tube;
R is the load; and EB is the battery for the plate circuit, Ec1 that for the
eontrol grid, and Ec2 that for the screen grid. The operation is similar to
that of a triode u:sf'd as an amplifier.
§ 2~~12 'l'JJB RLBC'l'RON GUN 415

The relation between the output current and the potential difference across
filament and plate of a tetrode is indicated by the curve of Fig. 13. The slope
of this curve between lines a and b shows that the plate circuit of the tube
manifests a negative reRistance, for the plate current fallR with the rising plate
potential; this characteriRtic makes for good operation of the tube as a stable
oscillator. The region of the curve beyond c is the one which affords good
amplifier action, for there the characteristic is practically straight, and hence
very little distortion of the input occurs.
The negative-resistance part of the characteristic curve reRults from the fact
that the electrons from the cathode arrive at the plate with such high speeds
that they knock a considerable number of electrons out of it, and these so-
called secondary electrons· are drawn over to the screen grid. To eliminate
the dip in the characteristic curve and improve the amplifier action, still
another grid, called a suppressor grid, is added to the tube. This is placed
between the screen grid and the plate and connected to a point of low po-
tential, usually to the cathode itself. Its purpose is to repel the secondary
electrons, to drive them back into the plate, and thereby prevent them from
reaching the screen grid. Five-element tubes containing the three grids
mentioned are called pentodes and are widely used in radio circuits.
The three constants of a triode defined in § 23-7 apply also to multigrid
tubes, but their definitions then specify that the potentials on the additional
electrodes be kept constant. The amplification fador naturally refers to the
ratio of a change in plate potential to the change in control grid potential.

ELECTRONIC APPLICATIONS

23-12. The Electron Gun. An arrangement for emitting and accelerating

electrons, and for directing and controlling the electron beam, is called an
electron gun; it comprises a cathode, a control grid, and one or more anodes.
The electron gun is used in a variety
of devices, such as the cathode-ray oscil- ,-----------------------
I'
I
lograph, the betatron, the electron micro- I

scope, and television tubes. !I

A typical structure of an electron gun is I
I ~CG
shown in Fig. 14. The cathode K is an \ ,_

oxide-coated emitter, heated internally,

and equipped with a heat shield. The
control grid CG has the form of an envel- +
oping cylinder and serves the same func- Fm. 23-1-L Electrode structure of an
tion as that of the control grid in the electron gun.
ordinary electron tube. The first anode
A1 is called the accelerating electrode and the second A2 the focusing elec-
trode; both are cylindrical in shape and have concentric apertures. The
potentials of these electrodes are adjusted by means of the potential divider P
to the values suitable for the specific purpose of the gun. The intensity of
41G ELECTRICITY & MAGNETISM Chap. 23

the beam is controlled by adjusting the potential of the control grid, and the
focusing is usually accomplished by varying the potential of the first anode
and keeping the second one at a constant and higher potential.
The electrodes of the electron gun act on an electron beam in much the same
way that a lens system acts on a light beam; for this reason the electrostatic
field developed by applying potentials to neighboring electrodes is called an
electron lens, and the study of such lenses is included in the field of electron
optics. The structure shown in the figure is equivalent to two converging
electron lenses; the first results from the form of the electrostatic field be-
tween the control grid and the first anode, and the second from the field
between the first and second anodes. To understand how the electron lens
operates it should be remembered that when an electron moves within an
equipotential region it is not acted upon by any forces, and hence its path will
be straight; but when it travels from one such region to another which is at a
different potential, the path will change abruptly.

l@lr,
N
I
I
I
I

~~r,
j I
~
I v,
2 v V2
1'
I
I
I
N
s
II
Frn. 23-15. I, Refraction of an electron beam; II, electrostatic electron lens.

The path of an electron beam is shown in part I of Fig. 15 as extending from

a region of constant potential V1 to a similar one of potential V2. If the latter
region is more positive than the first, then an electron will change its direction
toward the normal NN in crossing the potential boundary, because the com-
ponent of its velocity along the normal will be increased, whereas the compo-
nent along the separating plane will be unchanged. The electron velocities
in regions 1 and 2 are designated respectively as v1 and v2, and their unchanged
velocity component as v. By the law of refraction, § 28-5, it follows that
sin i = v/v 1 and sin r = v/v 2, where i and rare respectively the angles of in-
cidence and refraction of the electron beam at the plane separating the two
equipotential regions. Consequently,

sin i
- . - = -.
v2
Sln r V1

Now the relation between the velocities of the electron beam and the po-
tentials of the two regions is found by determining the work done on an
§ 23-13 CATHODE-RAY OSCILLOSCOPE 417

electron of mass m and charge e in crossing the boundary; this is

e (V2 - V1) = ! mv2 2 - ! mvi2.
If the work done upon the electron in bringing it initially into the zone of
potential V1 is eV 1 = ! mv12, it follows that eV2 = ! mv22 ; hence by division
VdV1 = v22/vi2. When this result is merged with the earlier one, the velocity
ratio becomes
(23.6)

In conclusion, the ratio of electron velocities in the two regions is proportional

to the square root of the potential ratio; also to the inverse ratio of the sines
of the angles that the electron beam makes ·with the normal to the plane
separating the potential regions.
Although the abrupt bending of an electron beam as described is analogous
to the refraction of light on entering a medium of greater refractive index, in
practically all arrangements applying electron optics the potential zones
change gradually, and consequently the electron paths shift gently from zone
to zone. Thus, part II of Fig. 15 shows a simple arrangement, consisting of
two cylindrical tubes T1 and T2 charged to different potentials, which is
equivalent optically to a converging lens and serves as an electron microscope
of low power. The curved lines are sections of symmetrical equipotential
surfaces which spread into the tubes, the potential gradient being greatest at
the midplane of the gap where these surfaces are closest together. The
electron lenses in Fig. 14 are the areas on either side of anode A 1.
23-13. Cathode-ray Oscilloscope. A vacuum tube utilizing an electron
gun to produce a beam of electrons and including means for deflecting the
beam and rendering its deflections visible is called a cathode-ray tube. This
tube forms the principal part of an instrument called a cathode-ray oscilloscope,
which is widely used in electrical measurements for showing the forms of
alternating and transient emfs or currents.

Frn. 23-16. Cathode-ray oscilloscope.

The cathode-ray tube is similar to the evacuated tube used in Thomson's

experiment for determining the charge-to-mass ratio of the electron, § 18-12.
Its appearance is indicated in Fig. 16, wherein the electron gun is symbolized
418 ELECTRICITY & MAGNETISM Chap. 23

merely by the cathode Kand anode A, and the electron deflecting system by
two pairs of plates P1 and P2 arranged in planes at right angles to each other.
When the filament is heated and a difference of potential is maintained be-
tween it and the anode, an electron stream or cathode ray passes through
the anode aperture, travels straight ahead, and impinges upon a fluorescent
screen S, where it produces a tiny luminous spot. When the plates P 1 are
charged, the electric field established between them acts on the electron stream
and deflects it upward or downward, and when a field is established between
plates P2, the electron stream is deflected to one side or the other. If the
potential differences on the deflecting plates vary, the electron stream will be
deflected accordingly, and the spot on the screen will move about with corre-
sponding rapidity, tracing a luminous line that reveals the character of these
potential differences.
When used as an oscilloscope to show the wave form of an alternating emf,
one pair of plates, P 2 in the figure, is periodically charged at a uniform rate
from a capacitor and then suddenly discharged, each time sweeping the lu-
minous spot across the screen along the dotted line and making it hop back
to the starting point again. The alternating emf is applied to the other pair
of plates, P 1, thereby producing vertical displacements of the spot; the com-
bination of the two displacements gives a picture showing the wave shape
with respect to time. The fre-

-r
Screen _r_
quency of recurrence of the sweep
motion can be varied over a wide
range; with an appropriate rate, a
steady image of the wave shape will
s Axis be obtained on the screen.
It will be shown that the vertical
deflection produced by the plates
P1 is proportional to the potential
Frn. 23-17. Path of electron beam in cathode- difference V between them. These
ray tube. plates will be assumed to have a
length l and a separation s, and to
have their midpoint at a distance L from the screen, as shown in Fig. 17.
The work done on an electron of charge e in moving it from cathode to anode
in the electron gun can be expressed as eV', where V' is the potential differ-
ence between these electrodes. This work will appear as kinetic energy of
the electron; whence eV' = ! mvx2, where m is the mass of the electron and
v,,, is its (horizontal) velocity along the tube axis. This velocity will then be

Vx = ~ 2 ::''·
The time t required for the electron to travel the length l of the electric field
between plates P 1 is l/vx. While in this field, the electron experiences an
upward force F = e8, where the electric field strength 8 = V /s, by Eq. 15.8.
§ 23-14 X-RA Y8 AND THEIR PRODUCTION 4:H)

Finally, the electrons have an acceleration a = !!._ = eV in the direction of

m ms
the force, and in time t they acquire a (vertical) velocity of
eV l
vy =at=--.
msvx

Upon leaving the region between the plates the electron will travel in a straight
line at an angle to the horizontal axis given by the slope vy/Vx, This line is
found to intersect the axis at the midpoint of the plates, which is at a distance
L from the screen. Hence, the ratio of the screen deflection to this distance
is the same as the ratio of Vy to Vx, As a result, the deflection of the electron
at the screen is

D=~Y=_l!:_V (23.7)
ms v x 2 2 s V' '
which shows that this deflection is proportional to the potential difference
V between the deflecting plates.
23-14. X-rays and Their Production. An invisible radiation of great pene-
trating power was discovered in 1895 by Wilhelm K. Rontgen (184,5-1923),
German physicist; he called the rays x-rays because their nature was not
understood at first. These rays are now known to be of the same nature as
visible light, hut their wavelengths are very much shorter.

7i
] Tube

focusing cup

IX-Rays\
/////.!I!\:'.
'1 I ' ' I I

Fm. 23-18. X-ray tube and its connPctions.

X-rays are produced wherever cathode rays (electrons) strike material sub,
stances at sufficient speeds. It is found that they are produced more copi-
ously if the substance upon which they impinge has a high atomic weight, and
that they are more penetrating if the speed of the bombarding electrons is
high. The electrons are usually derived from a hot cathode and are focused
upon a metal target which forms the anode. A large difference of potential
maintained between these electrodes give the impinging electrons speeds up-
ward of one-tenth that of light, which is 186,000 miles per second.
The general design of a modern x-ray tube is indicated in Fig. 18. The
420 ELECTRICITY & MAGNETISM Chap. 23

anode A is a target of tungsten or other heavy metal attached to a metal

sleeve which helps to radiate the heat evolved, and the cathode K is a filament
of tungsten wire wound into a spiral and heated by current supplied through
a step-down transformer T2, as shown. These electrodes are connected to a
source of high-potential direct current obtained from a step-up transformer T 1
through a mechanical rectifier or from a Kenotron.
When different substances are interposed between a photographic plate or
film and a source of x-rays, the radiation penetrates them to different extents,
according to their densities, and the plate or film, upon development, shows
the shadows of the objects interposed. When the hand is so placed near an
x-ray tube, the plate is affected less behind the bones than behind the flesh,
because the bones are more opaque to the radiations. A print made from
such a plate gives shadows of the hones and a faint outline of the flesh.
Broken bones and foreign objects in the body can be located accurately in this
manner.
X-rays may be used for the direct examination of objects by the use of a
fluoroscope screen that is coated with certain crystals, such as platinobarium
cyanide or calcium tungstate, which fluoresce under the action of x-rays. If
the hand be interposed between such a screen and the tube, the shadow of the
bones can be plainly seen.
Physicians use x-rays not only in making examinations but also for their
curative action. Animal tissue suffers a change in structure after exposure
to x-rays and is destroyed by prolonged exposure. In x-ray therapy, appro-
priate dosages of the rays are concentrated upon abnormal or diseased tissue
that is to be destroyed.
The radiation from an x-ray tube covers a wide range of frequencies, the
range extending to higher frequencies (shorter wavelengths) as the potential
difference between the target and cathode is increased, § 34-3. It can be
shown that the penetrating power of x-rays depends upon their frequency.
Equipments have been developed for high-speed x-ray photography, and for
high-penetration analysis such as inspecting welds in thick steel plates and
locating defects in heavy castings.
When a gas is exposed to x-rays it becomes ionized, and the amount of
ionization produced serves as a measure of the intensity of the rays. Such a
measurement is carried out with an ionization chamber and a charged electro-
scope; see Fig. 34-7. The chamber consists of a closed metal box, having a
window of thin celluloid or aluminum, and containing an insulated metal
electrode to which the electroscope is connected. When the rays are ad-
mitted through the window, and a difference of potential is maintained
between the electrode and the box, the ions produced by the x-rays sepa-
rate; those of one sign gather on the electrode while those of the other sign
collect upon the chamber. The rate at which the electroscope discharges is
an accurate measure of the intensity of the x-rays.
§ 23-15 SEMICONDUC110RS 421

SEMICONDUCTORS

23-16. Semiconductors. A review of the mechanism for conducting elec-

tricity through various kinds of matter shows that in electrolytes and in gases
conduction occurs through the motion of ions, that in metals conduction takes
place through the motion of electrons, and that in insulators there is no con-
duction but only a slight displacement of the charges within the atoms them-
selves. There is still another kind of matter in which conduction takes place
by electrons just as in metals, but, contrary to the behavior of metals, a
substance of this kind exhibits an increase of resistance as the temperature
falls. Such a substance is called a semiconductor, and at the absolute zero of
temperature it would be an insulator. Among the examples of semicon-
ductors are silicon and germanium.
The variation of resistance with temperature is explained as follows: In a
metal only a very few electrons are free to move upon the application of a
potential difference, and, as the temperature of the metal is lowered, the ther-
mal vibration of its atoms is reduced; as a result the atoms interfere less with
the motion of the electrons, and consequently the resistance is lowered.
Those electrons free to move in a metal are in semiconductors bound loosely
to the atoms. At absolute zero a semiconductor has no current carriers; as
the temperature is raised, more and more of the loosely bound electrons are
released by the thermal energy and conduction is improved, which means that
the resistance is lowered as the temperature rises.
The current carriers in a semiconductor may be supplied by an impurity.
For example, an arsenic impurity in silicon supplies one loosely bound excess
electron for each atom of arsenic dissolved, and hence conduction is due to a
transfer of excess electrons; such a semiconductor is said to be of the n-type,
because the carriers are negative. In contrast a trace of boron in silicon re-
moves one electron for each atom of boron dissolved, and the "hole" left in
the electronic structure of a silicon atom provides a type of conduction, called
hole or deficit conduction, by transfer from atom to atom of electrons into
available holes. A semiconductor in which the conduction is due to holes is
said to be of p-type, because the carriers act like positive charges, for the
"hole" travels in a direction opposite to that of the electron filling it.
The importance of semiconductors, at least in present applications, arises
from their behavior when in contact ,vith a metal. When the two substances
are placed in contact, the free electrons in the metal and those in the semi-
conductor come into equilibrium. Since their numbers are unequal, and
since the forces that bind them are unequal, there is a transfer of electrons,
which continues until the charge accumulated is large enough to repel a fur-
ther transfer of electrons. The accumulation of charge at the interface acts
as barrier layer, so called because it interferes with the passage of current.
The barrier at the interface between conductor and semiconductor is
422 ELECTRICITY & MAGNETISM Chap. 23

decreased when potential is applied in one direction and is increased when

potential is applied in the opposite direction. Hence, the current will rise
sharply with increase of potential in one direction, called the "forward" direc-
tion, but will remain small with increase of potential in the other direction.
This result is indicated in Fig. 19, which shows the volt-ampere characteristic
of a barrier layer. Different scales are used for the forward and reverse
directions so as to show the current variation with potential difference in both
directions on the same chart.
A device that utilizes a semiconductor in which the changes of current are
markedly out of proportion to the changes of potential is called a varistor.
The resistance of a varistor at any point of the characteristic curve, as at P
in the figure, is defined as the ratio of its abscissa OA to its ordinate OB; this
value, commonly referred to as the direct-current resistance, is important in

0.1,.___----'-----'-----'----'
-100 -50 O 1 2
Potential, volts

Fm. 23-19. Characteristic curve of a Fm. 23-20. Full-wave rectifier

varistor for both current directions. connections with four varistors.
Note that the scale above zero current
is 100 times as large as that below.

the application of unidirectional potentials. With alternating potentials the

ratio of change in potential, dE, to change in current, dl, is more important
and is known as the alternating-current or dynamic resistai:i-ce; for the varistor
biased to potential OA this resistance is the reciprocal of the slope of the curve
at point P.
23-16. Contact Rectifiers. Varistors are widely used to rectify alternating
currents. They may be connected in series or in parallel. If the units have
identical characteristic curves they will divide the potential difference or the
current equally. "When two varistors are connected in parallel and with
opposite polarity, the volt-ampere characteristic of the pair becomes sym-
metrical. Such pairs are used in protective circuits to present high resistance
to the low potentials normally applied, but low resistance to abnormally high
potentials, thus les::,cning the damaging effects of high transient potentials.
§ 23-lG CON'l'_'iC'J.' iUCCTIFI ERS

Most power rectifiers on single-phase circuits are of the full-wave type.

With vacuum or gas diodes, either two tubes or one double-anode tube may be
used; both require a center-contact transformer connected as in Fig. 7. With
varistors, four units are generally used, and the connection is similar to that
of a w·heatstone bridge, § 17-13, as shown in Fig. 20; it does not require a
center tap on the transformer. Each varistor is pictured as a transverse dash
designating the cathode and a triangle representing the anode, with the tri-
angle pointing in the forward current direction. vVhen the upper transformer
terminal is positive the current traverses varistor 1, the load circuit, and
varistor 4; when the lower terminal is positive current traverses varistor 3,
the load circuit, and varistor 2; hence at all times the current traverses the
load in the same direction.
Two classes of contact rectifiers are in common use; the area contact variety
called metallic rectifiers, such as copper oxide, selenium, and copper sulfide;
and the point contact rectifiers, such as silicon and germanium. Electronic
rectifiers based on other principles are considered in § 23-6.

lead or Barrier
f>lated metal layer
V Catwhis.ker

1=~
Copper-oxide
-Barrier
layer
Silicon

Copper oxide Selenium Silicon

I II Ill
Frn. 23-21. Rectifiers utilizing different semiconductors. Arrows indicate the direction of
current.

The copper-oxide rectifier has an electrode of copper on which a layer of the

oxide is produced by heat treatment. Contact with the exposed surface of
the oxide layer is usually made in small rectifiers by evaporating a layer of
silver or gold on the oxide, in intermediate units by clamping a lead washer
against the oxide face, and in large rectifiers by reducing the top surface of
the oxide to copper and then nickel plating it. The rectifier is pictured
sectionally in part I of Fig. 21; the barrier layer is located between the copper
and the oxide. Electrons pass easily from the copper to the oxide, and hence
the direction of the conventional current is from oxide to copper. Small recti-
fiers are used in communication circuits, and the larger ones as power rectifiers.
Copper-oxide and selenium rectifiers form the basis of the photovoltaic or
barrier-layer type of photoelectric cell, §§ 27-8 and 34-2.
The seleni'um rectifier is made by depositing a thin coating of metallic
selenium on steel or aluminum and, after heat treatment, applying an alloy
having a low melting point. The barrier layer is located between the selenium
and the alloy, as indicated in part II of Fig. 21. Electrons pass easily from
424 ELECTRICITY & MAGNE'l'ISM Chap. 23

the alloy to the selenium and hence the "forward" current direction is from
selenium to alloy. A selenium rectifier does not require as many series-
connected units as the copper oxide rectifier for a given potential difference
and hence has the advantage of smaller size and weight in power applications.
Tiny selenium rectifiers are used in telephone switching for contact protection;
in this application two rectifiers are connected in series but in opposed relation
to suppress sparking across a pair of contacts.
The silicon rectifier is a development of the early crystal detector with the
"catwhisker" contact. The schematic arrangement is indicated in part III
of the figure, the barrier layer being at the contact point. The silicon is in
the form of a wafer with one terminal soldered to it and the other touching it;
the latter is the catwhisker usually formed of tungsten, platinum, or phos-
phor-bronze wire. The forward direction of current is from silicon to cat-
whisker. To give an idea of size, the silicon wafer for a rectifier that operates
at microwave frequencies up to 25,000 megacycles per sec is 0.05 in. square
and 0.01 in. thick.
A modification of the silicon rectifier which is suitable for high potentials
makes use of silicon-carbide-that is, carborundum. The granules are mixed
with a binder and are fired to form ceramic-like rods or disks, contact with
which is made by plating opposite faces. Within these forms the individual
crystals constitute a series-parallel grouping of rectifier units connected with
random polarity, and for that reason form a symmetrical varistor. These
rectifiers are used in protective circuits, such as lightning arresters, and are
known under the trade name of Thyrite.
The germanium rectifier, which resembles the silicon rectifier in construction,
is the most widely used point-contact varistor; because it serves the same
purpose as a diode tube, it is generally called a germanium diode. Its forward
current direction is usually from catwhisker to germanium. Most of these
diodes can withstand reverse potentials of 75 volts, and some twice that.
Germanium diodes are widely used in communication circuits and in industrial
electronics.
A possible competitor of the germanium and silicon rectifiers is a recently
developed and inexpensive semiconducting compound formed of aluminum
and antimony.
E c 23-17. Transistors. The aim to utilize
semiconductors as amplifiers was realized
IZ:'Z:Z?;Z~ n-type R by John Bardeen and Walter H. Brattain
Input 8
at the Bell Telephone Laboratories
~ - -- - - . . i
through the invention in 1948 of the trans-
istor. This device is similar to the triode
FIG. 23-22. Input and output cir-
cuits of a transistor. in action, and can be used as an oscillator
as well as an amplifier.
A transistor consisting of a tiny block of n-type germanium with three
electrodes is shown schematically in Fig. 22. Two of the contacts are rectify-
§ 23-17 'l'RANSIS'l'ORS 425

ing points, such as those described in the preceding section; they are the
emitter E, connected to the input circuit, and the collector C, connected to
the output circuit R. The third electrode is the base B and is common to
both circuits. When the emitter is biased to establish current in the forward
direction, as shown, and the collector is close by, the input current from Eis
found to control the output current to C in such a way as to produce amplifi-
cation.
It will be recalled that there are n-type and p-type varistors, and that
conduction may take place by means of electrons or by "holes" that are
equivalent to positive charges; the action of the transistor depicted can be
explained by considering the emitter to inject holes into n-type germanium
(which otherwise would contain only electrons) and the collector to draw them
over to it, thus augmenting the output current many-fold.

,i() 8

-.,,
0
>
30
~
a. 6
E

~ E

-..
I
20 .~ 4
0
:.::: c

-
c Cl)
Q)
0
a. 10 02

0 4 8 12 0 20 40 60
Current-ic, milliamp Potential Ep,volts
I Transistor II Triode
Frn. 23-23. Comparison of transistor and triode characteristics. Signs on symbols apply
ton-type transistor.

The performance of a transistor is compared with that of a triode in Fig. 23,

which shows in part I a few of the collector circuit characteristic curves of a
particular transistor, and in part II a few of the plate circuit characteristic
curves of a particular triode. While the families of curves are similar in
appearance, the difference in the variables should be emphasized, for the
transistor curves show the effect of one current upon another, whereas the
triode curves show the effect of one potential upon another.
Like the triode, the transistor has several operating factors; for example, its
amplification factor measures the effect which the emitter current ie has upon
the collector current i 0 and is defined by

die I
a = die Ve const.
42G ELECTRICITY & 1v!A.GNETISM Chap. ~3

This factor can be likened to the amplification factor µ of a triode, for the
application of positive current bias to the emitter shifts the collector circuit
curve of the transistor to the right just as negative potential bias shifts the
plate circuit characteristic of the triode to the right.
Another form of transistor has been developed by interposing a thin p-type
layer serving as base electrode between two sections of n-type germanium
which form the emitter and collector. Such units show better operating char-
acteristics, have exceedingly small power consumption, possess mechanical
strength, and give high amplification.

PROBLEMS
1. Compute the emf that is generated by a copper-nickel thermocouple when one
of the junctions is at 0°0 and the other is at 300°C.
2. What emf is produced by an antimony-bismuth thermocouple when the temper-
atures at the junctions are maintained at 20°0 and 175°0?
3. The thermoelectric powers of aluminum and silver, both against lead, are ex-
pressed by the following equations over the temperature range under consideration:
Aluminum E = -0.8 + 0.004 t Silver E = 2.1 + 0.15 t,
where Eis in microvolts per C degree and tin degrees C. Find the emf produced by an
aluminum-silver couple when the temperature of one junction is 0°C and the other is
(a) 100°C, and (b) 200°0. •
4. A thermopile being designed for use in fire detection is to have the same number
of antimony-bismuth junctions on both faces. It is desired to generate a total emf
of at least 0.10 volt when the exposed face of the thermopile reaches 160°C, the other
face remaining at the room temperature of 15°C. What is the smallest number of
junctions that may be used on the exposed face?
5. A thermopile has 64 antimony-bismuth junctions on each face. When radiation
falls upon one face while the other is at the room temperature of 20°0, the thermopile
generates an emf of 0.18 volt. 'What is the temperature of the thermopile face that
receives the radiation?
6. A demonstration magnet for operation by a thermocouple formed of copper and
nickel is able to exert a large attractive force when one junction is heated by a Bunsen
burner. If the winding consists of 8 massive turns and has a resistance of 50 microhms
(including the thermocouple), and if the junction temperatures are 450 and 60°0, how
much magnetomotive force is developed by the winding? Extrapolation of the curves
of Fig. 3 gives the thermoelectric powers at 450° for copper as 7 .5 and for nickel as
-32.5 microvolts per 0°.
7. Compute the rate of electron emission in amperes per square centimeter from
a tungsten filament operating at 2300°0.
8. A thoriated-tungsten cathode is operated at a temperature of 1900°K and pro-
duces an electron emission of 0.70 amp/cm2 • What value do these data give for bin
Richardson's equation?
9. The constant bin Richardson's equation is found by theoretical analysis to equal
the ratio of the work function of the metal to the gas constant per molecule. This
gas constant is obtained by dividing the universal gas constant by Avogadro's Number.
Calculate the work function for molybdenum; express the result in electron· volts.
10. The potential difference applied to a diode has a steady value of 250 volts and
the current through the tube is 0.10 amp. (a) What is the value of the tube constant
Kin Eq. 23.2? (b) How many electrons arrive each second at the anode? (c) With
what velocity do the electrons impinge upon that electrode?
 PROBLEMS 427

11. Determine the amplification factor of a Type 6.T5 tube having the characteristics
shown in the accompanying figure. Choose
the operating current to be 7.0 milliamp.
12. A triode, for which the amplification Type 6J5
E1=6.3 volts
factor is 70 and the transconductance 1600
micromhos, is connected to a load circuit of
0.1 megohm resistance. What power is sup- a.
E
plied to the load when an emf of 1 volt rms
value is applied to the input circuit?
J_, 6 f-------1----+1----1-----.~
"i§
13. For a triode that has an amplification
factor of 10, the plate current is found to c
be 8 milliamp when the grid and plate po- ~ 4>----<-
::,
u
tentials are respectively -10 and +250 2
volts. On the assumption that the space 0
charge is maintained, estimate the value of c: 2 t-----t-,1----t+----,~-----l
the current when the potentials on the grid
and plate are respectively O and +200 volts.
14. A telephone transmitter generates
10-11 watt of power when placed 1 ft from a 0 50 100 150 200
Plate potential,volts
person speaking in a moderate tone of voice.
If 25 watts are required to operate a public- PROBLEM 11.
address system outdoors, what gain must be
provided in an amplifier connected between transmitter and loudspeakers to meet this
requirement?
15. The gain of an amplifier is found to be 15 db. What is the ratio of the power
output to the power input of the amplifier?
16. The oscillatory circuit of a vacuum-tube oscillator operating at 1 megacycle per
sec has an inductance of 0.15 millihenry. Calculate the capacitance of the condenser
in this circuit.
17. An electron beam travels at an angle of 30° to the interface between two regions
of different potentials, leaving one region of positive potential and entering the other,
also of positive potential but 50 per cent greater in value. Determine the angle
through which the beam is bent.
18. In an oscilloscope tube the deflecting plates have a length of 1.2 cm in the axial
direction of the electron beam and a separation of 0.5 cm in the transverse direction.
The screen, assumed plane, is located at a distance of 22 cm from the center of the
plates. What will be the deflection of the luminous spot on the screen when the ac-
celerating and deflecting potentials are 1000 and 200 volts, respectively?
19. 1n a cathode-ray tube designed for deflecting the electron beam by a mrq~net.ic
field transverse to the axial direction of the beam, show that if the field has a uniform
flux density B webers/m2 limited specifically to an axial length l cm, the deflection of
the beam will be given in centimeters by
D - 2 96 BlL 10s
- · VV' '
where L is the distance from the center of the field to the screen in meters, and V' is
the accelerating potential in volts. Refer to § 18-12.
20. For the varistor having the characteristic curve shown in Fig. 19, estimate the
direct-current resistance and the dynamic resistance for operation at point P.
21. Estimate the amplification factor of the transistor having the characteristic
curves shown in part I of Fig. 23, assuming the operating point to be at the center of
the middle curve.
 24
RADIATION FROM
CIRCUITS
All electrical circuits in which the current varies will radiate energy into
space unless they are shielded. The amount of energy radiated from low-
frequency power systems is almost negligible, but is large from circuits such
as radio antennas, which are comparable in size with the lengths of the waves
they transmit or receive. With extended circuits or with higher frequencies,
radiation becomes increasingly important. This chapter presents the physical
concept of electromagnetic waves, the phenomena of their generation and
reception, and the basic circuit elements used in modulating the waves, in
amplifying the emfs, and in detecting the signals in such systems as radio,
television, and radar.
24-1. Electromagnetic Radiation. The energy that resides in the electric
field of a capacitor or in the magnetic field of a coil remains with the circuit
when changes of current occur slowly. But when such current changes occur
rapidly, some of the energy escapes from the circuit as waves of electromagnetic
energy. The discovery of electromagnetic radiation and its transmission
through space was made about 1887 by the German physicist Heinrich R.
Hertz (1857-1894); he also showed that such radiations have measurable
velocity and wavelengths. The Italian scientist Guglielmo Marconi (1874-
1937) applied this discovery to electrical communication; he may be regarded
as the founder of the radio arts.
In radio, the waves are launched at the transmitting or broadcasting station
from an aerial system, or antenna, and are received at a similar antenna con-
nected to the receiving set. The antenna usually consists of two wires, but
one may be replaced by a ground connection. At the transmitting station a
high-frequency source of alternating current is joined between the two parts
of the antenna, and this forces electrons back and forth, charging each of the
antenna wires alternately plus and minus. The two parts of the antenna be-
have like the plates of a capacitor, and the region between them is under
electrostatic stress.
The region around an antenna that utilizes ground for one part may be
pictured as in Fig. 1; at the left the antenna current i is upward and at the
right it is downward. The electrostatic field is directed from the antenna
428
§ 24-1 ELECTROMAGNETIC RADIATION 429

wire to ground or vice versa, as depicted by the lines of electric flux, and the
magnetic field encircles the wire, as indicated by the lines of magnetic flux
shown in section by the dots and crosses. The charges surge up and down the
antenna wire as the current reverses, and, since the lines of the electric field
terminate on the charges themselves, the ends of these lines at the wire also

Electric
lines of
force

Frn. 24-1. Magnetic and electric fields around a vertical antenna wire.

move up and down. At the same time, both the magnetic and electric fields
are pushed outward by successively new fields as the current changes with
each cycle. Figure 2 shows the propagation of the composite field in cross-
section; the lobes embracing the electric lines are concentric with the antenna
wire and the magnetic lines are horizontal circles, both expanding outward

Antenna

I I
f'T1m me mm fT1e
Me Mm M.i A1m
---·-·--A---~~
,.._ _______
~'---~-
Frn. 24-2. Electromagnetic waves leaving a transmitting antenna.

into space. The electromagnetic waves travel in free space with the speed
of light-approximately 300,000 km per sec.
The length of the waves depends upon the frequency of the alternating
current in the circuit-the higher the frequency the shorter the waves. The
wave length >. of the waves in meters is determined by dividing the speed of
.light by the frequency, and can be expressed as
>.. = 300,000,000 = 300,000 =-,
300 (24.1)
f !Kc !Mc
430 ELECTRICITY & MAGNETISM Chap. 24

where f is the frequency in cycles, !Kc in kilocycles, and !Mc in megacycles, all
per second. This equation is generalized in the next chapter for waves of all
kinds. If the frequency of the alternating current in the antenna wire of
Fig. 2 is 600 kilocycles/sec, then the wavelength of the waves emitted from
it is 500 m.
In order to radiate as much energy as possible, the current in the antenna
is made large by adjusting the inductance and capacitance in the circuit so as
to produce resonance at the frequency desired. The strength of the waves
radiated from the antenna is measured by the intensity of the electric field
produced in space and is usually expressed in microvolts per meter, § 15-9.
At any place its value is the same as the emf that would be induced in a short
conductor per meter of length by the magnetic flux of the waves when their
configuration, § 25-1, sweeps past it, perpendicularly, with the speed of light.
Naturally, the strength of the waves diminishes as they recede from their
source, and other factors such as atmospheric conditions also have an influ-
ence.
24-2. Resonance. It is shown in §§ 22-7 and 22-8 that when an emf is
impressed upon a circuit of low resistance and the frequency is such as to make
the inductive and capacitive reactances of the circuit equal, then the circuit
is in resonance with the frequency of that emf, and the resonant frequency is
1
(22.10)
f = 2 1rVLC 1

where Lis the inductance of the circuit in henrys and C is its capacitance in
farads. At resonance, with these two circuit elements connected in series, the
impedance of the circuit has its least value; with them in parallel, the im-
pedance of the circuit is close to its greatest value.
When an alternating emf Eis impressed upon a series circuit of resistance R,
inductance L, and capacitance C, the current at resonance is E/R; upon
making the frequency of the emf higher or lower, the current becomes less,
because of the effects of inductance and capacitance. A circuit in which the
current falls off sharply with change of frequency is said to have high selec-
tivity, because it can better discriminate between emfs of comparable ampli-
tudes but of different frequencies.
The energy wasted in the circuit as heat increases as the frequency of the
alternating current is raised. These losses are due: (1) to the crowding of the
current in conductors toward their surfaces, a phenomenon called skin effect;
(2) to the dielectric losses associated with the capacitance between the turns
of coil windings; and (3) to the eddy currents in metal parts, such as shields,
in the neighborhood of the coils. To take these heating effects into account,
it is common practice to define the so-called effective resistance R of an alter-
nating current circuit as the quotient of the total power loss P due to heating
by the square of the current I in it, so that P = RI2, as in Eq. 17.8 for direct
current.
§ 24-3 MODULATION 431

The performance of a series circuit near resonance is determined largely by

the ratio of its inductive reactance XL to its effective resistance R, a ratio
symbolized by Q and expressed as
Q = XL = 2 1rfL.
R R
It is called the quality factor of the circuit, or circuit Q. This quantity tends
to remain constant with frequency near resonance, for as the frequency in-
creases and makes the inductive reactance larger, the effective resistance also
becomes larger, because of the increased losses.
Analysis of a series circuit shows that when the frequency of the applied
emf deviates from the resonant value by an amount 2~ of the resonant fre-

quency, the current is very nearly 1/V2 of its value at resonance; this result
makes it possible to estimate the sharpness of resonance afforded by a circuit
when only the circuit Q is known. In a parallel resonant circuit such as used
in broadcast receivers it is found that the impedance at resonance is equivalent
to an effective resistance that is very closely Q times the impedance of either
branch circuit.
The coil of a series circuit that is resonant at 1 megacycle/sec has an inductance of
0.15 millihenry and an effective resistance of 10 ohms. How much must the frequency
change in order to reduce the current in the circuit to 70.7 per cent of its value at
resonance?
The circuit Q is practically the same as its value for the coil alone; consequently,
. 2 7r X 106 X 0.15 X 10-3
it has a value of 10 = 94.2. Hence the frequency change to
reduce the current to 70.7 per cent or 1/'\12 of its value at resonance is 106/(2 X 94.2)
= 5310 cyles/sec.

Resonant circuits are extensively used in electric filters to eliminate currents

of unwanted frequencies and pass along those of frequencies that are wanted.
The magnetic coupling (§ 21-1) of two circuits designed for resonance at an
appropriate frequency provides a means for forming a band-pass filter which
permits the transmission of currents within a specified frequency range.
24-3. Modulation. The primary purpose for producing electromagnetic
waves is to transmit information through telegraph signals, telephone com-
munication, or radio and television programs. The wave forms of the cur-
rents in these communication circuits are molded to the particular forms
needed to convey the desired information. This molding of wave form is
called modulation. The process is used not only with the radiated waves in
electrical communication but also with the guided waves associated with
telegraphy and telephony over wires.
Perhaps the simplest type of modulation is the formation of the dots and
dashes of a telegraph code by closing a line circuit for short or longer periods
of time. Every time the circuit is closed by means of a telegraph key a direct
432 ELECTRICITY & MAGNETISM Chap. 24

current is established over the line, causing a distant electromagnet, called a

sounder, to attract its armature and produce a click. Since the current re-
ceived on a long telegraph line is small, a more sensitive instrument, called a
relay, is used in the line circuit in place of the sounder. A relay has a light-
weight armature and functions merely to open and close a local battery circuit
which includes a sounder, rather than to produce an audible response itself.
In the transmission of speech by telephone, first achieved by the American
inventor Alexander G. Bell (1847-1922), a direct current is modulated by the
sound waves to be transmitted. This is now accomplished by a telephone
transmitter that consists of carbon granules confined between two electrodes;
one electrode is rigid and the other is flexibly mounted on the diaphragm
against which the voice is directed. Variations of air pressure on the dia-
phragm cause changes in the contacting areas of the granules and the resist-
ance of the transmitter is altered correspondingly; as a result the current
supplied by a battery to the receiving instrument is changed.

~-~---------
Dot Dash Sinusoid

Carrier wave ''

3..... //

II
FIG. 24-3. Modulation of a carrier wave by: I, telegraph signal8; ll, sound tones.

The telephone receiver comprises a small electromagnet combined with a

permanent magnet, and a thin diaphragm of magnetic material. Variations
of current in the electromagnet cause variations of the same frequency in the
attraction of the diaphragm, thereby setting it into vibration to produce sound
in conformance with the variation of air pressure at the transmitter. ·without
the permanent magnet, the receiver diaphragm would produce sounds of
double pitch, for the diaphragm would be attracted once for the positive lobe
of an alternating-current cycle and once for the negative, yielding two com-
plete vibrations of the diaphragm for each current cycle. But with the per-
manent magnet present, a magnetic "bias" is produced, so that the resultant
flux from both magnet and winding does not change its direction; conse-
quently, the diaphragm is attracted just once for each cycle of current, and
frequency doubling does not occur.
In transmitting telegraphic signals or telephonic information by electromag-
netic waves, the characteristics of such communication methods must be im-
parted to the waves themselves and be carried by them to the rece1vmg
§ 24-3 MODULATION 433

station. A customary way of shaping the carrier waves is to modulate their

amplitude in conformance with the particular signals; this procedure is called
amplitude modulation. Part I of Fig. 3 indicates how the carrier wave is
modulated for a dot and dash signal in carrier telegraphy, and part II for a
simple tone in telephony. While the sounds of speech and music form com-
plicated wave shapes, they may be regarded as summations of sinusoidal
components of different frequencies and continually varying amplitudes
(§ 26-4). The effective modulation of the carrier wave in radio broadcasting
by means of a microphone is essentially the same as in telephony. This is true
also in television, were the carrier wave is modulated to conform to the light
intensities of the successive elements of the picture or scene to be telecast.
An analysis of a sinusoidally modulated carrier wave reveals that it is
equivalent to three separate waves. If f is the frequency of the carrier and
A is its amplitude, then the displacement at an instant t after passing through
the zero value is given by Eq. 22.2 as y = A sin 2 1rft; and if the amplitude
of the carrier is modulated by the fraction m at a frequency fm, then this
amplitude is expressed as
A = Ao+ mAo sin 2 1rfmt,
where Ao is the average amplitude. When these equations are combined, the
result is
Y = (Ao + mAo sin 2 1rfmt) sin 2 1rft
= Ao sin [2 1rft] - m:o cos [2 1r (f + f m) t] + m:o cos [2 1r (f - f m) t].
The first term on the right-hand side of the final expression is the carrier wave
itself, and the other two are side band waves which are alike in magnitude but
different in frequency, one side band being higher in frequency than the
carrier by an amount equal to the modulation frequency, and the other lower
by the same amount. When the degree of modulation m is zero, the side
bands vanish and the carrier wave is unaffected. The side bands can be
measured by separating their component frequencies from each other and
from the carrier by suitable filter circuits.
When the carrier wave is modulated by the complex sounds transmitted in
telephony, which embrace frequencies say from 20 to 10,000 cycles/sec, then
the modulated wave contains a pair of side bands for each component fre-
quency. Consequently, a high-fidelity broadcasting station would require a
frequency band extending 10,000 cycles on each side of the carrier. The band
of frequencies utilized by a transmitting station is termed a radio channel.
Beside amplitude modulation there are other ways of molding the wave
form; among these are frequency modulation, § 24-6, and various types of
pulse modulation.
To regain the information transmitted by a modulated carrier wave it is
necessary, when receiving the waves, to demodulate the carrier; this process
is usually called detection. The basic devices and their circuits for generating,
434 ELECTRICITY & MAGNE'l'ISM

modulating, and detecting the waves of a radio transmission system are con-
sidered in the next section.
24-4. Radio Transmission and Reception. The operation of a radio system
involves a transmitting station and one or more receiving stations.
Radio Transmitter. The function of the transmitting station is to produce
electromagnetic waves as the carrier and to modulate them in accordance with
the signals to be transmitted. The
high-frequency alternating currents
H are generated by an electron-tube os-
cillator at a frequency determined by
Mr{,J/1 T
a so-called tank circuit consisting of
inductance and capacitance. Figme
4 shows a radio transmitter that
employs a Hartley oscillator (Fig.
23-11) coupled to a grounded antenna
circuit composed of coil L and the
Frn. 24-4. A simple radio telephone aerial conductors.
transmitter. Modulation may be accomplished
by inserting a microphone and a trans-
former in the grid circuit of the oscillator, as shown. This arrangement, called
grid modulation, is very simple and effective in low-power circuits. The sound
waves impinging upon the microphone M are converted into corresponding
electrical current variations in the local circuit H, and these voice currents
produce varying emfs between the grid G and cathode K by means of trans-
former T. Thus, the grid is subjected to two sets of potential variations, one
being the very rapid radio-frequency oscillations of the carrier wave and the

Antenna

IIQ
Oscillating crystal

Frn. 24-5. A crystal-controlled radio telephone transmitter.

other the relatively slow audio-frequency variations of the voice wave. As a

result, the circuit oscillations, substantially constant in frequency, are varied
in amplitude in accordance with the sound wave. Capacitors are shunted
across the battery connected to the plate P and across the secondary winding
of the transformer T to by-pass the high-frequency currents. Another modu-
lation procedure is to couple the microphone through an electron tube to
the plate circuit of the oscillator, as in Fig. 5; this method is called plate
modulation.
§ 24-4 RADIO TRANSMISSION AND RECEPTION 435

The frequency of a radio transmitter can be controlled by the use of crystals,

like quartz, which exhibit the piezoelectric effect. If a small slab is cut in a
particular way from a crystal, it is found that distorting the slab mechanically
causes each face to exhibit a charge, and that an opposite distortion causes
this charge to be reversed. Conversely, if the slab is located in an alternating
electric field, it will contract and expand periodically and set up mechanical
vibrations of constant frequency. The vibration frequency will vary in-
versely with the thickness of the slab, a typical value for a 1 mm thickness of
quartz being about 3 megacycles per sec. When a crystal plate of appropriate
thickness is placed between two electrodes that are connected in the grid
circuit of an electron-tube oscillator, with the plate circuit tuned to the reso-
nant frequency of the crystal, the frequency of oscillation will be kept practi-
cally constant regardless of load changes. Figure 5 illustrates a simple crystal-
controlled transmitter using plate modulation; the circuit elements indicated
have the usual significance. Coils L1 and L2 are reactors which serve respec-
tively as radio-frequency and audio-frequency chokes.

H
c
Modulated input emf

Frn. 24-6. A simple electron-tube receiver. Fm. 24-7. The operation of a diode
detector in demodulating a voice signal.

Radio Receiver. The functions of the receiving station are to absorb energy
from the electromagnetic waves passing its antenna, to amplify the weak high-
frequency oscillations that develop in the antenna, and to demodulate the
carrier to obtain the signals desired. The receiver is tuned to a particular
transmitting station by adjusting the capacitance or inductance of its antenna
and associated circuits. Figure 6 shows a receiver with a simple diode de-
tector. Changes of potential across the variable capacitor C act upon the
tube, and the impulses are passed along to the telephone headset H through a
circuit consisting of a capacitor C1 and a resistor R. The capacitance is
chosen so that the reactance will be low at the frequency of the carrier and
high at the frequency of modulation.
Each positive peak of potential that is applied to the tube causes capacitor
Ci to charge, and its potential will rise to a value somewhat less than that
applied, because of the drop through the tube itself caused by the charging
current. Before the next peak of potential is applied some of the charge will
leak off through the resistor R. This process continues, with the results
indicated in Fig. 7; the light curve shows the modulated input. potential from
capa.eitor C, and the heavy curve the output potential available acrm,s R.
436 ELECTRICITY & MAGNETISM Chap. 24

It will be seen that the latter curve is unidirectional, although it has a radio-
frequency ripple; nevertheless it has the same modulation envelope as the
applied signal.
24-5. Radio- and Audio-amplification. In order that an electron-tube
detector may operate to the best advantage, it is necessary that the potentials
impressed upon its grid circuit be reasonably large. Since the available
energy for the antenna to receive is usually too small to develop sufficiently
large potentials, it is customary to amplify the received radio-frequency im-
pulses before they are applied to the detector. This process is termed radio-
frequency amplification and is accomplished by one or more electron tubes
used as amplifiers. The detector then converts the modulated radio-frequency
(r-f) oscillations into signals of audible frequencies.

I
I
I
• I
Signal
input
I
I
I
C

~--1-------l.--l+ 1J1I _
E
Frn. 24-8. Radio-frequency amplifier cou- Frn. 24-9. Audio-frequency amplifier with
pled to tuned load. resietance coupling.

Amplifiers used at radio frequencies generally include a multi-grid tube and

a tuned circuit. Figure 8 shows such a circuit that employs a tetrode having
the cathode K, plate P, and two grids. Coils L 1 and L2 serve to couple the
plate circuit to the tuned circuit formed of coil L2 and variable capacitor C2.
The signal is applied between the control grid CG and ground. The potential
supplied to the screen grid SG by battery Eis reduced by the potential drop
caused by the screen grid current through resistor R. Capacitor C1 by-passes
r-f currents from this grid to the cathode K, while capacitors C by-pass r-f
currents in the plate and grid circuits. The drop across resistor r provides
the negative unidirectional bias potential for the control grid CG relative to
the cathode. The magnetic flux caused by the amplified r-f current in coil L1
induces a corresponding emf in coil L 2 • Capacitor C2 tunes the circuit to the
desired signal frequency. The emf established across C2 is applied between
ground and the control grid of the next amplifier tube. All interfering signals
finding their way into the input of the amplifier produce only a small effect
on the potential difference across C2 • Thus, the action builds up the desired
signal for the next stage of amplification and reduces the effect of interfering
signals.
§ 24-0 FREQUENCY MODULA'l'ION 437

After one or more stages of r-f amplification, the output of the detector is
still small, and further increase is necessary through two or more stages of
audio-frequency amplification. The last stage of audio-frequency (a-f) ampli-
fication employs a tube of larger size, a so-called power tube, in order to yield
sufficient volumes of sound in a loudspeaker.
Figure 9 shows the circuit of an a-f amplifier that utilizes resistance coupling
between the stages. The tube represented is a triode, but multi-electrode
tubes are also used. The coupling elements are resistors Re and capacitors Cc,
the latter being introduced so that the only unidirectional potential applied to
the control grid is the bias. Resistor R1 of high resistance provides a return
path for any electrons from the cathode K which strike the control grid. The
other circuit elements have the same significance as before.
Some types of receivers also employ amplifiers between the radio-frequency
and audio-frequency parts of the circuit to achieve high selectivity; they are
called intermediate-frequency amplifiers. In the so-called superheterodyne re-
ceiver a local oscillator is used to produce beats (§ 25-14) with the incoming
carrier wave, and the intermediate-frequency (i-f) stages amplify one of the
beat frequencies. The amplitude of the local oscillations is kept constant,
and the i-f signals are modulated the same as the received wave. The i-f
amplifiers are designed for a constant frequency and hence can be arranged
for sharp tuning and high gain.
24-6. Frequency Modulation. Another method of modulating radio waves,
in which the carrier wave is modified in frequency rather than amplitude, was
devised by the American inventor Edwin H. Armstrong (1890-1954) and is
called frequency modulation. In this process, the amplitude of the radio wave
emitted by the transmitter is always constant. Frequency-modulated broad-
casting stations operate at very high frequencies-of the order of 100 mega-
cycles/sec. The disturbing effect known as static, which produces extraneous
noise in amplitude-modulated receivers, is not carried through the circuits of a
frequency-modulated receiver.

Frn. 24-10. Graph of a frequency-modulated carrier wave.

In frequency modulation the basic principles are: (1) the frequency of the
wave is varied over a range proportional to the intensity of the sound, and
(2) the rate at which this variation takes place is equal to the frequency of
the sound. Figure 10 represents a frequency-modulated wave on a time axis.
The frequency (and therefore wavelength) of the wave varies with the in-
tensity of the sound being carried, and the frequency range expands and con-
tracts at a rate which depends upon the pitch of that sound. Thus, louder
438 ELEC'PRICITY & MAGNETISM Chap. 24

sounds produce larger changes of wavelength, and higher-pitched sounds

produce more rapid stretching of the wave pattern.
A numerical illustration will help to clarify the principles. Suppose a frequency-
modulated transmitter operating at a carrier frequency of 45 megacycles/sec to be
transmitting a 1000-cycle tone of such intensity as to require a frequency change of
10 kilocycles/sec above and below the normal value. The frequency of the trans-
mitter will then vary from 44.99 to 45.01
megacycles/sec, at a rate of 1000 times a
------r--· - second. This situation is depicted by a
Frequency sine curve in Fig. 11 in which the ordi-
deviation nates show frequency deviation and the
I
I
abscissas indicate time. Since 1000 com-
plete sets of frequency values are to be
>..
swept out per second for a 1000-cycle tone,
u one set must be completed in 0.001 sec, as
cG)
shown.
5-44.99
e
u.. --------0.001 sec-------
Next, suppose the intensity of the same
tone to be doubled. The deviation of the
carrier frequency would be doubled, rang-
Frn. 24-11. Diagram to explain fre- ing from 44.98 to 45.02 megacycles/sec.
quency modulation. Note that the ordi- Should the original intensity of the tone
nates signify frequencies and not ampli- be tripled, the frequency would vary from
tudes. 44.97 to 45.03 megacycles/sec. Thus, the
frequency deviation for the three intensity
values of the 1000-cycle tone would be, respectively, 10, 20, and 30 kilocycles/sec.
For all, the frequency would vary from its minimum to its maximum and back to the
minimum frequency at the rate of 1000 times per second. Of course, the amplitude of
the radiated wave would remain constant throughout.
Finally, suppose the tone to change from a pitch of 1000 to one of 4000 cycles/sec,
while maintaining the same intensity as originally employed. The radiated wave
would vary over the same range (from 44.99 to 45.01 megacycles/sec) as before, but
now at a rate of 4000 times per second.
Reception of frequency-modulated signals requires the use of specially con-
structed receivers. In addition to amplifiers, such a receiver includes one or
more limiter tubes and a discriminator or other form of detector for frequency
modulation. The limiter sets an upper limit to the plate current at a value
that corresponds to a low signal potential on the grid of the tube; it thereby

Antenna
Mixer or
converter Limiter Discriminator

r-f amplifier
Loudspeaker
Local
a-f amplifier
oscillator

Frn. 24-12. Block diagram of a frequency-modulated receiver.

§ 24-7 TELEVISION 439

eliminates any amplitude variations in the incoming signals due to static or

other causes. The discriminator transforms the frequency-modulated signals
to corresponding ones of audio frequency.
Figure 12 shows the general circuit of a frequency-modulated receiver by
means of a "block" diagram; in addition to a local oscillator it makes use of
radio-, intermediate-, and audio-frequency amplifiers, symbolized as before.
In the present allocation of channels, each station is allowed a band extending
75 kilocycles per sec on each side of the carrier.
24-7. Television. The basic principles of "vision at a distance" by elec-
trical means involve: (1) scanning the scene spot by spot in an orderly fashion;
(2) establishing a current that varies with the light intensities of these spots;
(3) amplifying these currents and transmitting them over wire lines or by
radio waves; (4) converting the electrical pulsations into corresponding light
fluctuations at a receiving station; and (5) arranging these light patches upon
a screen to form an image of the original scene. Furthermore, the entire scene
must be scanned repeatedly at least 15 times per second in order that its
successive images may merge to produce a steady result, like that of motion
pictures.
In the television system devised by Vladimir K. Zworykin of the Radio
Corporation of America, a narrow electron beam is used to scan the image in
a photoelectric tube of special design called an iconoscope. This is an evac-
uated glass tube arranged as illustrated in Fig. 13. The electron beam is
produced by an electron gun, § 23-
12, which consists of a heated ca-
thode K, a grid G, and the anodes
A1 and A2. The beam is directed
upon a screen S formed by a thin
sheet of mica having a metal film
on one side and on the other a
tremendous number of tiny pho-
tosensitive particles resembling a
mosaic. The scene to be trans- 2
mitted is imaged upon the mosaic
by a lens system denoted by L, and
as a result the particles are illumin- Fm. 24-13. Iconoscope for converting a light
ated to various intensities. Each image into a varying current.
particle may be regarded as a tiny
photoelectric cell, § 34-2, that also forms a capacitor with the metal film on
the other side of the mica sheet, and hence each one acquires a charge by
photoelectric emission that depends upon the intensity of light upon it. The
electron beam is defleeted from one side to the other along a line, and then
deflected from line to line by means of a magnetic field that is symbolized by
Fin the figure. In this way the entire image is scanned and the charges on
the particles of the mosaic are released in succession, thereby establishing a
440 ELECTRICITY & MAGNE'l'ISJ1 Chap. 24

current in the output circuit 1-2 that fluctuates in accordance with the illum-
ination of the mosaic elements. This so-called video signal current is am-
plified by electron tubes and transmitted to the distant station by methods
already described in this chapter.
The television receiver comprises tube circuits for receiving, amplifying,
and synchronizing the signals, and a large cathode-ray tube (§ 23-13) called a
lcinescope. It converts the video impulses from the transmitting station into
minute luminous spots on a fluorescent screen.

2nd anode, a metal

film on inside of
glass envelope
Accelerating
electrode efu Fluorescent
screen on
inside of glass
---,--.--- ,,. ....

~~Deflecting
Grid 1st anode Electron
beam

coils \ 2nd anode terminal

Frn. 24-14. The components of a kinescope.

The design of the kinescope is indicated in Fig. 14, the electrodes at the left
being the electron gun. The amplified video signals are applied to the grid,
and through it control the intensity of the electron beam. The other elec-
trodes focus the beam so that it produces a luminous spot on the screen. The
magnetic fields developed by the currents in the deflecting coils cause this
spot to move in the same manner as the beam that sweeps over the mosaic
of the iconoscope. As the spot on the kinescope moves across the screen, the
brilliance of fluorescence changes from point to point under the action of the
signals impressed upon the grid. These changes conform to the variations of
light and shade in the picture on the iconoscope mosaic and, under proper
synchronization of the scanning operation with that at the transmitter, the
picture is reconstructed on the kinescope screen. Because the scanning of a
scene in the iconoscope and its reconstruction in the kinescope proceed at a
high speed, the eye sees only the composite picture and not the individual
luminous points of it. This blending takes place within the eye in the same
manner that a number of rapidly changed still pictures give the impression of
movement in motion pictures, and is aided by the continuing but brief fluo-
rescence of the screen after the beam has moved on.
Figure 15 gives a general impression, through a block diagram, of a typical
television receiver. The antenna receives both the amplitude-modulated pic-
ture signals and the frequency-modulated sound signals that are transmitted
on carriers which differ sufficiently in frequency so that they may be separated
 Antenna Sound
Discrim- Audio
i-f Loudspeaker
inator amplifier
amplifier

Mixer
r-f
or
amplifier
converter

..... Picture Horizon-

..... 2nd Video
,...... i-f tal
detector amplifier
amplifier deflection

Video Synchron-
Oscillator izing
automatic
volume separator Kinescope
control
Vertical
Rectifier
deflection

FIG. 24-15. Block diagram of a typical television receiver. Arrow at right leads to a source of alternating current.
442 ELECTRICITY & MAGNETISM Chap. 24

in the receiver. The signals are passed along to the radio-frequency selector,
by means of which the desired station is tuned in. The output of this ampli-
fier and that of a local oscillator are fed to the mixer or converter; from it
proceed the frequency-modulated signals of the sound channel in the upper
part of the diagram, and the amplitude-modulated signals of the video channel
in the lower part. These signals are amplified and separated by the circuit
blocks indicated and finally reach the loudspeaker and kinescope, respectively.
The block marked "Rectifier" represents sources for supplying appropriate
potentials to the anodes of the kinescope and to the plates of the various
tubes in the television receiver.
24-8. Microwaves. Electromagnetic waves of 30-cm or of shorter wave-
length have been found useful in signaling schemes that require highly direc-
tional characteristics. Such waves, known as microwaves, have frequencies
upward of 1000 megacycles/sec. At such high frequencies the rapidly chang-
ing magnetic field causes the current in a conductor to be restricted almost
entirely to its surface, a fact which must be considered in the design of micro-
wave circuits and equipment.
The product of the inductance and the capacitance of a circuit must be very
small in order that resonance may occur at microwave frequencies. Equation
22.10 shows that even for the longest waves in this range-those having a
frequency f = 1000 megacycles/sec-the product of the circuit constants L
and C would not exceed

= ( -1
2
LC -) 1.._ X 10-18 henrys X farads
=
27rf 40
= 0.025 (microhenrys X micro-microfarads).
This product is less than can be realized by the usual form of electron-tube
oscillators even of minute size; consequently, other methods for generating
microwaves must be utilized.

I JI __
I
t9ft' I
I
.,.,. ,..
:......... -- ,,,,-

II m
FIG. 24-16. Successive stages in the development of a resonant cavity.

The evolution of a resonant circuit for microwaves from the conventional

tuned circuit is indicated pictorially in Fig. 16. In part I a circuit is shown
formed of a parallel-plate capacitor and a coil of several turns; with respect to
terminals at the top and bottom this combination is a parallel circuit, and the
impedance between these terminals is very high at resonance, § 24-2. In part
II the capacitance is decreased by separating the plates, and the inductance
§ 24-8 MICROW A YES 443

is decreased by reducing the coil to a single turn of wire. In part III the
further reduction of the LC product is indicated, the plates being separated
farther and the single wire connecting them being replaced by four in parallel,
with these in strip form to gain more surface for conducting charges between
the plates. In part IV the strips are widened to the full dimensions of the
plates to achieve the maximum conducting surface; the result is a completely
closed box.
A picture of the electric and magnetic field distribution of the simplest mode
will be considered for an energized enclosure such as that developed above.
In Fig. 17 the tuned circuit is again shown as part I; the upper diagram repre-
sents the + and - charges residing on the plates of the capacitor at a certain
instant during the cycle, and the lower one the flow of charge from one plate
to the other at a different instant. In the latter the flow is indicated by a

II Ill

~1
Frn. 24-17. Development of electromagnetic fields within a resonant cavity.

single arrow for the current in the coil and by several tiny arrows for the dis-
placement current in the dielectric, § 21-7. The corresponding properties for
the closed box are shown as part II; the upper diagram represents the charges
on the horizontal faces, and the lower one the current in the walls and dielec-
tric. The associated electric and magnetic fields are represented, respectively,
in the upper and lower perspective pictures of part III; in the latter the mag-
netic flux is shown encircling the displacement current only. If a line con-
centric with these flux lines were drawn around the outside of the box, it would
embrace both the currents in the walls and the displacement current, but since
these are equal and opposite, there could be no magnetic flux along such a line.
Hence, both the electric and magnetic fields are confined to the enclosure; the
energy during a cycle is stored first in one field and then in the other, and no
energy is radiated. In short, the enclosure is a resonant cavity for some
particular frequency of oscillation in the microwave range.
The size of the enclosure has a bearing on the resonant frequency, for the
wave must "fit" into it. For the mode pictured in part III, the electric field
is greatest at the vertical centerline and tapers off in strength to a small value
444 ELECTRICITY & MAGNETISM Chap. 24

at the edges where the walls virtually short-circuit the charged surfaces. In
order to attain such a field distribution the width of the box should be greater
than half the wavelength corresponding to the frequency of oscillation.
Microwaves can be transmitted inside hollow pipes called waveguides. Fig-
ure 18 shows a waveguide of rectangular cross-section supplied at the left
with energy from a microwave generator through a coupling loop. The closed
magnetic flux lines encircle the electric flux lines and both progress together
toward the right. Microwaves can be radiated into space from antennas just
as can the longer waves considered earlier in this chapter. Because of the
short wavelengths it is possible to construct antennas with superior directional
properties. Some antennas consist of a dipole radiator mounted in front of
a parabolic reflector.

Frn. 24-18. Propagation of microwaves along a hollow tube.

24-9. The Magnetron. The generation of microwaves is accomplished by

electronic devices which make use of resonant cavities. They include the
magnetron, the klystron, and the traveling-wave tube; only the first of these
will be described.
The magnetron has an anode shaped like a cylinder and provided with a
number of resonant cavities, a cathode in the form of a wire mounted axially
within the anode, and a steady magnetic field directed axially through the
space between these electrodes. The details are pictured in Fig. 19. Part I
shows the electrodes in cross-section with the field directed at right angles to
the page, and several possible paths of electrons in their motion from the

Anode cavity

I Il
FIG. 24-19. Diagrams of a magnetron. I, Possible paths of electrons in the space be-
tween cathode and anode; II, anode block with its several resonant cavities.
§ 24-10 RADAR 445

heated cathode to the positively charged anode. With no magnetic field the
electrons would follow a radial path as shown by line a, with weak fields
the electrons would be deflected along curves like arc b, and with intense fields
they would be deflected more strongly and form closed paths as indicated by c.
The electrons gain kinetic energy as they are accelerated by the electric field
between the cathode and the anode. If they reach the anode, the electrons
waste their energy by heating that electrode. The anode potential is ad-
justed so that the electrons do not reach the anode, and hence the energy
associated with them is kept in a useful form. The electrons collectively
constitute a cloud that pirouettes in the space around the cathode. The
mechanism by which the energy in the cloud is utilized is explained in the next
paragraph.
Part II of the figure shows the resonating chambers of the anode; each con-
sists of a gap which extends back from the axis and terminates in a cylindrical
hole parallel to the axis. The gap between adjacent segments of the anode
may be regarded as the capacitance and the hole as a single-turn inductance;
together they constitute a tuned circuit at microwave frequencies. Between
the segments there exist alternating electric fields which are approximately
parallel to the path of the electrons moving around the cathode. These fields
speed up or slow down the electrons, depending upon the momentary direction
of the fields at the time the electrons are near the gaps; the result is a bunching
of the electrons which ultimately concentrates their energy into dense clouds
to be delivered as "bundles" to the resonating chambers passed later. This
energy is made available not by transferring electrons directly from cathode
to anode but by inducing alternating currents of comparable frequencies in
the resonant circuits. Electromagnetic fields are thereby sustained within
the cavity resonators, and the microwave energy from all of them in the mag-
netron may be led out through a single loop coupler, such as indicated in the
figure.
24-10. Radar. An effective method of locating objects by radio techniques
is called radar, a contraction of radio detection and ranging, which was greatly
improved during World War II. The basic principle of radar is the scanning
of an area by a beam of microwaves from a station of known position, and
the detection at the same station of the waves reflected from the objects to be
located. Since the speed of the waves is known, it is only necessary to
measure the time which elapses between the transmission of microwave pulses
and the arrival of the reflected waves in order to locate, in both direction and
range, the object which caused the reflection.
Radar equipment includes an antenna, the transmitter and its modulator,
the receiver and its indicator, and the usual power supply systems; these
elements are indicated in the block diagram of Fig. 20. The antenna as-
sembly is rotated continuously when in use. The modulator causes the trans-
mitter to radiate microwaves in short pulses, usually of 1 microsecond dura-
tion or less, with a recurrence rate of about 1000 pulses per second. The wave
446 ELECTRICITY & MAGNETIS1.lI Chap. 24

pulses are of high intensity; in fact, the power transmitted in each one may
be of the order of a megawatt.
When the transmitter is active, a pulse of power is conveyed to the antenna.
Then a rapidly acting electronic
Dipole antenna and switch transfers the antenna con-
parabolic reflector
nection to the receiver circuit.
Indicator When the echo signal is received
from an obstacle, it is rectified and
Receiver
0 impressed upon the cathode-ray
tube of the indicator, causing a
bright spot to appear on the screen.
Transmitter
The scanning is radial, for the
sweep starts from the center of the
oscilloscope screen and radiates out-
Modulator wardly at a constant rate and in a
direction matching that of the radio
Frn. 24-20. Principal parts of a radar set. beam from the antenna, and then
returns quickly to the center. The
distance of the bright spot from the center of the screen bears a fixed relation
to the distance of the obstacle, and the angular position of the spot indicates
the direction of that obstacle.
Another application of microwaves is the radio altimeter which gives the
elevation of an airplane above the ground or a mountain. Pulses of micro-
waves are sent from the plane toward the earth's surface, and are there
reflected; the time taken for them to return is measured electronically. Thus,
the instrument automatically indicates the height of the airplane above the
terrain.
A new type of weapon for all-weather defense against aerial attack of cities
is the Nike, a guided missile named after the Greek goddess of victory. The
missile itself is about 20 ft long and contains an explosive "warhead," a rocket
propulsion unit, and guidance equipment. When warning is received from
air-defense radar sets that a hostile aircraft is approaching, the enemy target
is picked up and is tracked electronically by the Nike control system; when
the plane comes within a predetermined range the missile is fired skY'vard hy
remote control, and reaches supersonic speed within seconds after launching.
It converges upon the target aircraft, regardless of evasive action, and de-
stroys it upon interception.

PROBLEMS
1. The "ultra high frequencies" (UHF) authorized in 1952 by the Federal Com-
munications Commission for television channels extend from 470 to 890 mega-
cycles/sec; previously television stations were limited to the "very high frequencies"
(VHF) ranging from 54 to 216 megacycles/sec. What are the corresponding wave-
length ranges?
 PROBLEMS 4·17

2. How much inductance should be used with a 0.02-µf capacitor in order that the
circuit so formed shall be tuned to a 25,000-cycle carrier frequency?
3. The antenna circuit of a broadcasting station operating at a carrier frequency
of one megacycle per second has in series with the antenna a resonating inductance of
0.044 millihenry. What is the effective capacitance of the antenna?
4. When the current in the antenna circuit of Prob. 3 is 10 amp, the power radiated
is 20 kw. What is the effective or so-called "radiation resistance" of the antenna?
What is the emf across the resonating inductance?
5. A tuned circuit to be used in an amplitude-modulated receiver must respond to
a band of frequencies 10,000 cycles/sec wide with a response always at least 70 per cent
of that at resonance. If the carrier frequencies to be received are between 500 and
1500 kilocycles/sec, what is the highest Q the circuit may have at each extreme of the
carrier frequencies?
6. The tuned circuit of Prob. 5 is made up of a coil and a variable capacitor with a
minimum capacitance of 40 µµf. What is the inductance and effective resistance of
the coil if the circuit is tuned to 1500 kilocycles/sec when the capacitance has its mini-
mum value? What is the capacitance required to tune this coil to 500 kilocycles/sec?
7. The frequency of vibration in megacycles per second for a certain "cut" of
quartz slab is found to be 65.4 divided by the thickness of the slab in thousandths of
an inch. Calculate the thickness of such a slab for operation at a wavelength of 80 m.
8. The electric field intensity produced at a certain location by a frequency-
modulated broadcasting station is found to be 200 microvolts/m when the wave is not
modulated. What will be the field intensity at this location when the wave is modu-
lated?
9. The picture of a checkerboard having as many rows and columns as there are
scanning lines yields an image which has the maximum detail that can be transmitted
by a television system. Each traverse of the scanning spot over two adjacent squares
of the checkerboard must constitute at least one cycle of the video signal. The present
black-and-white television picture is produced by 525 scanning lines and is scanned
30 times per sec. 'IVith these facts and assuming that the picture is square, determine
the lowest frequency that may be used in connection with the camera tube.
10. A radar echo is received at a ship's antenna 25 microsec after the transmitting
pulse leaves that antenna. How far from the ship is the obstacle that caused the echo?
11. (a) How many cycles are there in a radar signal pulse lasting 40 microsec when
the station operates on 1750 kilocycles/sec? (b) How many cycles constitute a signal
pulse lasting 1 microsec when waves 10 cm long are used in the transmission?
 WAVE MOTION & SOUND

25
WAVE MOTION
Familiarity with wave motion is desirable as preparation for the study of
Sound and Light. Sound is propagated by wave motion through material
mediums, while light is transmitted through empty space as electromagnetic
waves. Some aspects of the latter have been considered in the last chapter
under radiation from electric circuits. The present chapter takes up some
general features of waves, shows how to determine the speed of waves of cer-
tain types, and introduces the concepts of reflection, refraction, and inter-
ference.
25-1. Some Types of Waves. In the study of wave motion it will become
apparent that waves are of many kinds and are of common occurrence.
Water waves sweep across the surface of the ocean, distortional waves surge
to and fro within vibrating bodies, sound waves carry tones and noises
through the air, and electromagnetic waves transmit radio and television
programs, light, and x-rays.
Probably everyone has thrown a stone into quiet water and watched a
group of ever-widening circular waves spread over the surface from the point
of impact. The water does not move as a whole, but some particular con-
figuration of the surface does. In transmitting the wave the individual par-
ticles of water must move, but their motion takes place over rather short dis-
tances. Another example of wave propagation is the motion of a disturbance
along a taut cord. If the cord is stretched horizontally by hand and the hand
is moved up and back suddenly, a wave will start along the cord. 'When
launched in this way, the wave consists practically of only one crest and is
called a single-pulse wave. If the hand is moved up and down repeatedly,
449
450 WAVE MOTION & SOUND Chap. 25

going through the same motion each time, a train of waves will be set up in
the cord. These waves all have the same configuration and are called periodic
waves. Unless specially mentioned, all of the waves to be studied herein are
of the latter kind.
The motion of the prongs of a tuning fork sets up periodic waves in the
surrounding air. Each forward movement of a prong compresses the air in
front of it, and each backward movement rarefies the air. These conditions
are transmitted outward from the fork as a wave disturbance comprising so-
called condensations and rarefactions. Upon entering the ear, these waves
produce the sensation of sound, and the waves themselves are called sound
waves.
Waves can be produced by vibrations other than those of material particles.
As discussed in § 24-1, the current in the antenna circuit of a radio station
sets up a magnetic field and an electric field in the region around it. As the
current oscillates, these fields continually build up and collapse, and in so
doing set up electromagnetic waves which spread outward from the antenna.
These waves are not transmitted by the motion of air particles but by changes
in the magnetic and electrical conditions of free space.
Most waves can be classified as either longitudinal or transverse. Alongi-
tudinal wave is one in which the vibrating particles move forward and back-
,vard parallel to the direction in which the wave is propagated. The sound
wave produced by the tuning fork referred to above is longitudinal. In a
transverse wave, the particles vibrate at right angles to the dfrection of
propagation. The wave moving along the cord previously mentioned is
transverse. Electromagnetic waves of all types behave like transverse waves.
25-2. Wave Motion and Particle Motion. The progress of any wave in-
volves two distinct motions. The wave itself in a homogeneous medium
moves forward with constant speed, which means that the configuration
advances equal distances in equal periods of time. Meanwhile, the particles
of the medium that conveys the wave vibrate to and fro in harmonic fashion.
Their locations at successive moments depend upon the period, amplitude,
and phase of the vibration. These terms have been used in Chap. 7 in the
study of harmonic motion; their definitions are restated here, together with
those of some other terms commonly applied to wave motion.
The period of a vibrating particle is the time in which it completes one
vibration, and the frequency is the number of vibrations completed per unit
time. The amplitude of vibration is the maximum displacement from the
undisturbed position. Two particles vibrating with the same frequency have
definite phase relations. They are in phase when they continually pass
through corresponding points of their paths at the same times. Otherwise,
they are out of phase; for the particular condition where they reach their
maximum displacements in opposite directions at the same instant, they are
in phase oppos£tion. The wavelength is the distance, measured along the
§ 25-3 THE MECHANISM OF WAVE PROPAGATION 451

direction of propagation, between two points that are in phase on successive

waves in a wave train.
Figure 1 represents a particle P in harmonic vibration of amplitude r,
together with the corresponding reference circle, as in Fig. 7-12, and also
shows part of the wave train due to the vibration. In one period T the
radius of the reference circle makes a complete revolution, and in the same
time the wave advances uniformly a distance equai to its wavelength A.

Fm. 25- l. A vi bra ting source P and the wave produced by it.

Therefore, the speed of the wave is v = >-/1'; it is expressed more usefully in

the form
v = f>-, (25.1)
where the reciprocal of 1' is replaced by the frequency f. These expressions
are applicable to all wave motions. The speed v of a wave is determined
completely by the properties of the transmitting medium, and the frequency
f of the source may be assumed to remain constant; it follows that, if a wave
from this source enters a new medium wherein it moves slower or faster, there
will be a corresponding change in wavelength to satisfy the equation.
25-3. The Mechanism of Wave Propagation. The process by which a
wave advances will be explained by referenre to a mechanical model, Fig. 2,

1 ,l
I
I
O t--
1
I
2 t
II
Fm. 2fi-2. '\fochanical wave model illustrating a transverse wave moving toward the right.
452 WA VE MOTION & SOUND Chap. 25

consisting of a long coiled spring with small masses A, B, C, which rep-

resent the vibrating particles of the medium. The spring is fastened at the
distant end, and particle A is given harmonic motion along the path 0-1-2-0.
During the time that the end particle A is moving up to position 1, particle
B is pulled upward by the spring tension, and proceeds in that direction, as
shown in part I of the figure. Because of its inertia, B continues to move
upward when A reverses its direction. As A moves to 2, a downward force
acts on B which soon arrests its upward motion and causes it to move down.
The same behavior is repeated at the lower end of the path, and it follows
that B will have the same kind of motion as A but will reach a corresponding
point of the path a little later than A will. As a consequence of the inertia
of the weights and the elasticity of the spring, similar motion will be imparted
successively to all parts of the spring, and the weights will reach their maxi-

A B c D E F G H J K

I
I
I I
I I I
I I I
I I I
I ' I I I
I g I. lk
I c \I
I
I
\

'

Frn. 25-3. A longitudinal wave and its transverse representation.

mum displacements in the sequence A, B, C, and so on-in other words, a

wave is set up which advances toward the right. At the instant represented
in part II of the figure, A has been given a complete vibration and the wave
has advanced to J; the wave along the spring has a crest at G and a trough
at C.
The same mechanical model will also serve to illustrate the motion of a
longitudinal wave, as represented in Fig. 3. Motion of A toward the right
compresses the spring, and motion toward the left extends it. The spring
acts on B and gives it the same kind of motion as A, except for a slight lag in
phase. Similarly, B produces a corresponding motion of C, and so on. The
result is a series of condensations in which the weights are close together,
separated by rarefactions in which the weights are farther apart. Both con-
figurations move to the right and constitute the advancing wave. At the
instant shown in the figure the wave in the spring has advanced to J.
The wave form of a longitudinal wave is not apparent but can be made so
by laying off the displacements of the particles at right angles to the direc-
tion in which they actually occur. Such a construction is shown at the
bottom of the figure. In this diagram the normal rest positions of the par-
ticles are indicated by corresponding lower-case letters, and the displace-
ments along the axis are shown turned counterclockwise through 90 degrees
by means of arcs centered at these points.
§ 25-3 THE MECHANISM OF WAVE PROPAGATION 453

From the behavior of the mechanical wave model, some general inferences
can be made. \Vave motion is evidently not due to bodily transfer of the
medium through which the wave advances. It is caused by vibrations of
individual particles over short ranges about normal rest positions, all the
particles having the same kind of motion, but with a progressive change of
phase along the direction of propagation. Moreover, mechanical wave
motion requires that the transmitting medium possess both inertia and elas-
ticity; for electromagnetic waves these properties are replaced by their elec-
trical equivalents, inductance and capacitance.

1 ~ ..... .....
\
I
\
\
\
\
\
' I
I
I
I
I
I
I

..J..
I
I
I

I
I
I
I

..... I
I
"
I
I

2 ~· \
~
I
• I I I I

-•
\ \ I I I I I

-- -
I \ \ I I I I
I I \ I I I I
3 •I
I
\
\
\
I
\
\
' I
I
I I
I

-- --
I \ I I \ I I
4 .J..
I
I
I
I
I
I
\
\
I
I
I
\
I
I
•I
I

\
+ I
I
I I \ \ \

•
I
I
..J.. \ \ I I
5 •I ~
I
I
·I
I
I
I
\
I
I
I
I
' \

--
I I I I I I \
6 ...,_ I
I
I
..,,... I
• I \ \
~

-
I I \ \
I I I I I \
I I \ \
I I
7 ~I I
I
I
I
..... I
I
I

• I
......\
I
I
\
I

-
I I I I I I

--
I I I I I
I
\ I

•
I
8
I
\ +I ...,.. I
I
....,.. I
I
I
I

- -
\ I I I I I
\ I I I I I \

•
\ I I I I
9 -l.. •

Fm. 25-4. Motion of air particles transmitting a sound wave. The wave is moving
toward the right, and the arrows represent the velocities of the particles.

The progress of a sound wave in air is represented in Fig. 4. Line 1 shows

the positions and velocities of the vibrating particles at a particular instant,
line 2 shows them t of a period later, line 3 shows them t of a period later
still, and so on. It will be observed that each particle moves back and forth
with harmonic motion, that there is a progressive change of phase in consid-
ering one particle after another, that condensations and rarefactions are
formed, and that these configurations advance in a definite direction-toward
the right in this example. In the condensations the particles move in the
same direction as the wave, while in the rarefactions they move the other way.
Water waves are of interest in that the particles which transmit the dis-
turbance move longitudinally as well as transversely. The particles follow
elliptical or circular paths; the wave is neither longitudinal nor transverse,
but a combination of the two. A typical construction for deep water waves
454 WA VE MOTION & SOUND Chap. 25

is shown in Fig. 5; the progressive change of phase in the motion of the par-
ticles and the resulting wave shape are indicated.
25-4. Energy Transmission by Waves. Whenever waves pass through a
medium, energy is transmitted through it in the direction of propagation.
This fact can be illustrated by the mechanical model of Figs. 2 and 3, which
show waves progressing to the right. For either type of wave, each section
of the spring exerts a force on the weights at its ends, and the weights at the
right move in the direction of this force, but those at the left are constrained
to move in opposition to it. Each spring section thus does work on the weight
ahead of it and has work done upon it by the weight behind it, § 6-1, and each
weight, in turn, performs a similar action on the adjoining spring sections.
Hence, a continuous transfer of energy takes place in the direction of wave
travel. In the model described, if no energy were wasted as the wave ad-
vances, the amplitude of vibration would be the same for all of the weights.

Frn. 25-5. The displacement of water particles in a surface wave.

On the other hand, in a wave that spreads out as it advances, such as a cir-
cular wave on the surface of water, the amplitude of vibration diminishes as
the wave progresses, since the energy is spread out over a larger and larger
surface. The energy of a wave may be transformed in various ways; for
example, that of a sound wave may be converted into mechanical energy in
setting the ear drum into vibration.
When a wave encounters a medium of a different character, some of its
energy will be reflected back into the initial medium, and the rest will be
transmitted into the second medium; also, as the wave advances, part of its
energy will be absorbed. For a light wave that impinges upon a sheet of
glass, most of the energy is transmitted to the region beyond the glass, part
is returned by reflection at the surfaces, and a small portion is absorbed within
the glass itself. When a light wave strikes black velvet, practically no light
is reflected from it nor transmitted through it; the velvet absorbs the energy
and its temperature is raised very slightly.
25-5. The Wave Equation. It has been repeatedly pointed out that wave
motion is associated with vibrations of the particles in the transmitting
medium. To investigate how these two motions are related, it is convenient
to imagine a source vibrating harmonically and producing a sinusoidal wave.
If the source vibrates with frequency f and amplitude r, its displacement at
any instant is given by the expression
Ys = r sin 2 1rft, (25.2)
where tis the time reckoned from the instant when the vibrating source passes
§ 2!i-G WA VE IN A STRETCHED CORD 455

through the midposition of its path in a positive direction. This motion is

imparted to the particles of the surrounding medium, and as the disturbance
moves forward it will reach these particles one after another and give each
the same type of motion as the source. If the disturbance moves with a
speed v, then to travel a distance x will require a time x/v, and consequently
the displacement of a particle in the medium at a distance x from the source
will be

y P = r sin 2 1rf ( t - : } (25.3)

Although this expression gives the displacement of the particle in the medium,
tis still reckoned from the time the vibrating source passes the midposition of
its path. It is clear that the displacement of the source at a given instant
will be duplicated by the vibrating particle after a time interval x/v; thus, if
t is given· a particular value in Eq. 25.2 and is increased by x/v in Eq. 25.3
the values of y. and YP will be identical.
The wave equation, 25.3, shows how the displacements of the vibrating
particles vary with respect to their location and also with respect to time.
By selecting a particular value for t and pl<?tting the values of y for different
values of x, a sine wave is obtained which represents a "snap shot" of the
wave in space; if this process is repeated for a slightly greater value of t, it
will be found that the whole wave profile has moved away from the source.
The displacement of any particle of the medium can be found by selecting
the appropriate value for x to represent its distance from the source, and by
plotting y against t. This also yields a sine curve and represents the har-
monic vibration of the particle with respect to time.
25-6. Wave in a Stretched Cord. The speed of a transverse wave in a
stretched cord is determined by the stretching force and the mass of the
cord per unit length. The relation among these quantities can be found by

Cord Cord F
.,,,_-----f---r':'-....
s~s
s
~,
I
p

II
M

Frn. 25-6. Detail of a transverse wave in a stretched rope.

supposing a cord of length l and mass m to be stretched horizontally and

then given an impulsive blow at one place; the result can be pictured as a
"dent" having the shape shown in Fig. 6. This displacement sets up a dis-
tortional wave which travels along the cord with a speed v, and the shape of
the dent is maintained as it moves forward. This dent could be kept sta-
456 WAVE MOTION & SOUND Chap. 25

tionary in space by imagining the cord to be of unlimited length and to be

propelled with the same speed v in the opposite direction. With this artifice,
one section of the rope after another moves around the dent; perhaps the
reader has observed the maintenance of such shapes in the movement of belts
between pulleys. The force acting upon the curved section of the cord can
be evaluated if the dent, or a small part of it, is regarded as circular, for the
force then becomes the centripetal force of circular motion.
In part I of the figure, the short central portion of the dent /::i.Z is considered
as a circular arc of radius r subtending an angle !::i.8 at the center. If this
portion of the cord has a mass !::i.m, the centripetal force acting upon it is
2
F = !::i.mv and is provided jointly by the forces S, S, each of which represents
r
the stretching force in the cord. These forces are drawn from a common
point P in part II of the figure and, since t:,.8 is very small, it folJows as a
close approximation that F /S = !::i.l/r; whence
!::i.mv
2 !::i.l S
-;:s = -;, or v2 = !::i.m/ t:,.(

The mass of the cord per unit of length, namely D.m/!::i.l, may be designated as
m1. Consequently, the speed of the cord, and its equivalent the speed of
the wave along the cord, will be

(25.4)

In mks units, Sis expressed in newtons and m1 in kilograms per meter; with
these units v will be in meters per second. In engineering units, S is in
pounds and m1 in slugs per foot; then v will be in feet per second.
As an illustration, determine how long it takes a wave to travel the length of a 50-ft
rope that weighs 4 lb and is stretched with a force of 30 lb.
The mass of the rope is 4/32 = 0.125 slug, and its mass per unit length is 0.125/50
= 0.0025 slug/ft. The speed of the wave will be
I 3o 1b ~
v = '\) 0.0025 slug/ft = 109 ·0 ft/sec,
and the time required to travel the length of the rope will be 50 ft/(109.5 ft/sec)
= 0.457 sec.
26-7. Propagation of Sound. The waves of sound do not pass through a
vacuum. This fact can be demonstrated by operating an electric bell within
the glass chamber of an air pump and then withdrawing the air; the hammer
can be seen striking the bell but no sound will be heard. The physical
medium required for the propagation of sound may be a solid, a liquid, or a
gas. Consideration will first be given to propagation within a liquid, and
the analysis will show that the speed of sound is determined by the elasticity
and density of the medium.
§ 25-8, SPEED OF SOUND IN GASES 457

The speed of sound in a liquid is found by applying Newton's Second Law

to a portion of the medium in which a wave enters. Figure 7 pictures this
portion of liquid as a segment of a cylinder with a cross-sectional area A. and
under a pressure p. For simplicity, a single-pulse wave will be considered,
which is produced by increasing the pressure at the left end of the liquid seg-
ment by an amount t:..p. When this extra pressure is applied, the left end
moves a distance t:..x in a time t.
Within this time the wave advances
toward the right with a speed v and
travels a distance x through the
liquid, and therefore the time inter-
() I
I
)
p

x----'
val can be expressed as t = x / v.
·
D urmg h" · l · h Frn. 25-7. Diagram used in deriving the
t IS mterva, mt e segment speed of a sound wave in a liquid.
of length x, the left end moves a dis-
tance t:..x while the right end remains stationary, and consequently the center
of mass of this volume moves through a distance t:..x/2. The acceleration of
the center of mass, from Eq. 3.11, is t:..x/t2 , where t is x/v. The liquid seg-
ment under consideration has a volume A.x and a mass A.xd, where d is the
density. Its center of mass has an acceleration (~:) 2 ' and the force that
produces the acceleration is A.t:..p. From Newton's Law of Motion, F = ma,

At:..p = A.xd Lx~:) 2}

and therefore v2d = ~- In this equation, the denominator of the right-

.A. t:..x
Ax
hand member is the ratio of the decrease in volume ( -t:.. V) to the initial
volume (V), and the entire right-hand member is therefore the bulk modulus B
of the liquid, as defined in§ 7-3. Hence, the speed of the wave disturbance is
(25.5)

In cgs units, B is expressed in dynes per square centimeter and din grams
per cubic centimeter; then v is in centimeters per second. Similarly, in
engineering units, B is in pounds per square foot, d in slugs per cubic foot, and
v in feet per second.
The foregoing treatment may also be applied to the transmission of sound
along a solid rod. The result will be the same as that just obtained, except
that the bulk modulus B will be replaced by Young's modulus Y of the
medium.
25-8. Speed of Sound in Gases. The method used in the preceding sec-
tion for finding the speed of a sound wave in a liquid, which resulted in the
equation v = ~:, can be used also when the medium is a gas. The elastic
458 WA VE MOTION & SOUND Chap.25

modulus B = A~/V can be written as - V ~~ or, in calculus notation, as

- V i· In order to apply this modulus to gases, it is necessary to know how

the volume V of the gas is affected by changes in the pressure p, and this
effect depends upon the restraints imposed upon the gas during the process.
If the temperature is kept constant, then in accordance with Eq. 13.13 the
relation between pressure and volume can be expressed as p V = k, where
both p and V are variables, and k is a constant. Upon differentiation, the
result is p dV + V dp = 0, or p = - Vi· A comparison of this equation
with the one above shows that p and B are identical. Hence, if the com-
pressions and expansions were isothermal, the speed of the wave would be
v = Vp/d.
In the passage of a sound wave, the medium undergoes compression and
expansion at a rapid rate, usually several hundreds of times each second.
Throughout the range to which the ear responds, it has been found for gaseous
mediums that the heat generated during compression is not dissipated quickly
enough to maintain uniform temperature, and that the compressions and
expansions should be considered as adiabatic rather than isothermal. In an
adiabatic process, by Eq. 13.15, p V'Y = constant, where 'Y represents the ratio
of the two specific heats, cp/Cv, By differentiation it is found that P'Y V-r-1 dV +
v dp
V 7 dp = 0, whence - dV = 'YP, or B = 'YP· Consequently, the speed of a
sound wave in a gas becomes
(25.6)

the units being the same as before. For air the value of 'Y is 1.40; if the
values of p and d for standard conditions of temperature and pressure are
taken from §§ 9-3 and 7, the corresponding value for the speed of sound is
= /1.40 X 1.013 X 106 dynes/cm2 = 33 130 I
v '\} 1.293 X 10-3 gm/ cm3 ' cm sec.
The speed of sound in the atmosphere is unaffected by changes in the
barometric pressure because the density is changed in the same proportion,
and for this reason the ratio p/d remains unchanged. Changes in the tem-
perature of the atmosphere, however, affect the density without influencing
the pressure, and hence cause a change in the speed of the wave. An inspec-
tion of the General Gas Law, as given by Eq. 13.8, shows that the ratio of
the pressure p to the gas density d can be expressed as p/d = RT, where R
is the gas constant and T is the absolute temperature. If this product is
substituted for p/d in Eq. 25.6, the speed of sound becomes
v = V 'YRT, (25.7)
§ 25-9 ELECTROMAGNETIC WAVES 459

which shows that the speed varies as the square root of the absolute tempera-
ture. An example will show that R must be expressed in absolute units in
order to be consistent with the units for v.
Determine the speed of sound in the stratosphere where the temperature is 400° on
the rankine scale.
The gas constant is found from § 13-2 to be
R = 53 3 ft·lb X 32 pounds= 1710 ft·lb = 1710 ft2 .
· pound · 0 R 1 slug slug· 0 R sec 2 • 0 R'

therefore the speed of sound is Vl.40 X 1710 X 400 = 978 ft/sec.

Values of the speed of sound in a few mediums are given in Table I. For
temperatures near 0°C, the speed of sound in air increases by about 2 ft/sec
or O.G1 m/sec for a temperature rise of 1 C 0 •

TABLE I. SPEEDS OF SOUND

Temper- Speed
ature
Medium ac m/scc ft/sec
-----
Air .. .. . . . . . . . . . . . . . . . . . . . .
' 0 331.3 1087
Aluminum ...... . . . . . . . . ... . ... 5100 16,700
Copper ..... . . . . . . . . . . . . . . . . .. 3970 13,000
Hydrogen ... . . .......... . ... 0 1286 4220
Iron and steel . .. . . . . . . . '. . . .I 4900 to 5100 16,000 to 16, 700
Lead ............ . . . . . . . . . . . . . . . . I • 1
.. 1230 4040
Water ....... ... I 15 1450 4760
I

For <'alculations in which the effect of temperature need not be considered,

the speed of sound in air may be taken in round numbers as 1100 ft/sec or
335 m/sec.
25-9. Electromagnetic Waves. Although many waves are transmitted by
the mechanical vibration of physical particles, others, such as light and radio,
are propagated by means of periodic variations in electric and magnetic
fields; these are called electromagnetic waves. In Fig. 8 such a wave is

Frn. 25-8. The varying electric and magnetic field intensities of an electromagnetic wave.
460 WA VE MOTION & SOUND Chap. 25

shown advancing along the axis to the right; the graph represents the instan-
taneous values of the associated fields. The electric field intensity 8 and
the magnetic intensity H repeatedly increase and decrease, but in mutually
perpendicular planes.
Theory shows that the speed of electromagnetic waves in a medium of
permittivity e and permeability µ is given by
1 (25.8)
v = Veµ.

For free space the permittivity and permeability are designated by e0 and µ 0 ;
when the values of these constants (§ 16-3) are inserted, the speed is found
to be
1 1
c = ~ = --;==========================================
V eoµo / 1 coulomb 2 X 4 1r X 10_7 newton
"\J 4 1r X 8.988 X 109 newton· m 2 amp 2
= 2.9979 X 108 m/sec,
where c is the usual symbol for the speed of electromagnetic waves in free
space. This speed is close to 300,000 km/sec or 186,000 mi/sec, and these
values may be used for air without appreciable error for most purposes.
Electromagnetic waves extend over a wide range of wavelengths, and spe-
cial names are given to the different wavelength regions, as listed in Table II.
In general, the methods of production and detection of the waves differ from
one class to another.
TABLE IL SUBDIVISIONS OF THE ELECTROMAGNETIC SPECTRUM

Approximate
Wave Classification Wavelength Range

Radio, low frequency .................. . 30,000 to 600 m

broadcast frequency ............. . 600 to 200 m
high frequency .................. . 200 to 0.3 m
microwaves .................... . 30 to 0.08 cm
Infrared .............................. . 800 to 0.75 µ
Visible light .......................... . 0. 75 to 0.38 µ
Ultraviolet ........................... . 0.38 to 0.01 µ
X-rays ............................... . 10-5 to 10-10 cm
Gamma rays (natural radioactivity) ..... . 3 X 10-9 to 5 X 10-11 cm

1µ = 0.001 mm = 10-4 cm = lQ-6 m.

25-10. Wave-front Construction; Huygens' Principle. A wave front is a

surface of which all the points are vibrating in phase with one another. The
"points" may be the vibrating particles of the medium, such as the oxygen
and nitrogen molecules in sound transmission through the air, or may be the
instantaneous electric and magnetic field intensities for a light or a radio
wave. Figure 9 shows a wave spreading horizontally from a source S, and
§ 25-10 WA VE-FRONT CONSTRUCTION; HUYGENS' PRINCIPLE 461

the line W, representing a particular wave front, connects particles of the

medium that are momentarily at their greatest upward distances from the
undisturbed positions. In a homogeneous medium the wave front from a
point disturbance is spherical; however, at con-
siderable distances from the source, small portions
of such a wave front may be regarded as plane.
From a study of the changes in wave fronts which
occur as a wave advances, it is possible to predict
the effects that will be produced when a wave en-
counters an obstruction, or when it is reflected or
refracted. This study is greatly assisted by use of
a principle credited to the Dutch scientist Christian
Huygens (1629-1695).
Huygens' Principle states that every point on a
wave front acts as though it were itself a center of
disturbance, sending out little wavelets of its own, the
collective effect of which constitutes a new wave front. FIG. 25-9. Representation
For a wave front in any given position, these of a wave front through ad-
jacent particles that are in
wavelets may be represented by arcs of equal phase.
radius drawn from various points on the wave "·
front as centers, since the radius represents the distance the wavelets would
advance in some specified time. A line or surface tangent to these arcs on
the side toward which the wave is advancing shows the new shape and loca-
tion of the wave front at the end of the time interval selected.

p P1
W1 W2 W3 W4

) .,,

I D
Frn. 25-10. Illustrations of Huygens' construction for locating successive wave fronts.

This construction is applied in part I of Fig. 10 to a plane wave front P

and to a spherical wave front S, both advancing toward the right to new posi-
tions at P1 and 8 1, respectively. The progression of a wave around an ob-
struction is illustrated in part II of the figure. The same procedure is followed
462 WA VE MOTION & SOUND Chap. 25

in constructing the successive wave fronts W1 to W4, but the time intervals
between them are purposely made different.
25-11. The Inverse Square Law. As a wave spreads out uniformly in all
directions from a source, the energy of the wave is distributed over a wider
and wider area. The time rate at which the energy is
transferred per unit of cross-sectional area of the wave
front is called the intensity of the wave. At a distance
r from a source the wave front is spherical and has an
area 4 1rr2 ; hence the intensity of the wave at that dis-
tance is
p
I=--,
Frn. 25-11. Diagram 41rr 2
used in proving the
Inverse Square Law.
where P represents the rate at which the source emits
energy. Appropriate units for wave intensity are the
erg per second per square centimeter, and the watt per square meter.
Figure 11 represents a source at O and two concentric wave fronts. At a
distance r1 from the source the wave front has a spherical surface of area
4 1rr12, and at a distance r 2 the area is 41rr2 2. If the intensities at these dis-
tances are symbolized as / 1 and / 2 , respectively, and the wave progresses
through a medium without loss of energy, then the product of the surface
area and intensity must remain the same, or

Hence, the ratio of the intensities is

11 _ r22
(25.9)
12 - r12
and shows that the intensity of a wave varies inversely as the square of the
distance from its sQurce. This relationship, called the Inverse Square Law,
has a wide application in physics and engineering.
25-12. The Law of Reflection. The behavior of a wave upon striking a

Frn. 25-12. Reflection of a plane wave at a plane surface.

§ 25-13 THE LAW OF REFRACTION 463

reflecting surface can be determined by Huygens' construction. In Fig. 12

a plane wave front, represented in cross-section by the trace AB, is shown
impinging upon the plane surface NIN, through which it cannot pass. If
this surface had not been present, the wave would have advanced without
change in direction and, in a certain time interval, would have reached the
position CD. The presence of the reflecting surface, however, causes a
change in the direction of the wave front. In the particular time interval
selected, different points on the wave front move as follows: B advances
directly to D; A cannot move to C and travels an equal distance above the
surface to some point on the arc CQ of radius AC; any other point E advances
without obstruction to F and, since it is unable to continue to G, travels an
equal distance to some point on the arc of radius FG. By Huygens' Prin-
ciple, the line DH tangent to the arcs is a trace of the wave front at the end
of the specified time; the wave has been reflected back into the region above
the surface MN.
The line AB represents the incident wave front, and DH represents the
reflected wave front. The angles i and i' that these wave fronts make with
the reflecting surface are called the angles of incidence and of reflection,
respectively. The relation between these angles can be shown by drawing
the radius AH to the point of tangency H, and recognizing that BD = AC
= AH; the right triangles ADH and DAB are equal, and consequently the
angles i and i' are equal. These angles are also coplanar, since the points
A, B, D, and H lie in the same plane. These facts are embodied in the law
of reflection, which states that, when a wave incident upon a plane surface is
reflected, the angles of incidence and of reflection are equal and lie in the same
plane.
26-13. The Law of Refraction. A wave which enters another medium
obliquely will undergo an abrupt change in direction if the speed of the wave
in the second medium is different from that in the first; this phenomenon is
called refraction. In Fig. 13 the incident wave front AB, moving with speed

Medium l

Medium 2

Fm. 25-13. Refradfon of a plane wave at a plane surface. The reflected wave is not
shown.
464 WA VE MOTION & SOUND Chap. 25

V1 in medium 1, encounters the interface 11!IN at an angle of incidence i.

Here it is partly reflected back into medium 1 and partly transmitted through
medium 2. The transmitted portion advances with speed v2. The reflected
portion behaves as explained in the preceding section.
In the time required for B on the wave front to advance in the first medium
a distance BD to the interface at D, point A does not move an equal distance
to C, but moves a distance (v2/v1) X AC = AH in the second medium to
some point on the arc centered at A. Any other point E advances to F in
the first medium and then moves a distance (vdv1) X FG in the second
medium to some point on the arc centered at F. The line DH drawn tan-
gent to these arcs represents the refracted wave front at the end of this time
interval, and the angler which it makes with the interface is called the angle
of refraction.
It was discovered by the Dutch astronomer and mathematician Willebrord
Snell (1591-1626), and by others more or less simultaneously, that in the
passage of light from one medium to another the ratio of the sine of the
angle of incidence to the sine of the angle of refraction is constant for waves
of definite wavelength. The relation between these angles can be found by
drawing the radius AH to the point of tangency H, and noting that sin i
= .BD/AD and sin r = AH/AD. Consequently,
sin i BD BD
--=-= '
sin r AH (v2/v1)BD
or sin i v1 (25.10)
-.-=-·

This relation expresses the law of refraction, which states that when a wave
travels obliquely from one medium into another, the ratio of the sine of the angle
of incidence to the sine of the angle of refraction 1·s the same as the ratio of the
respective wave velocities in these mediums, and is a constant for two particular
mediums. The angles of incidence and refraction lie in the same plane.
26-14. Interference of Waves. Two waves moving simultaneously through
the same region will advance independently, each producing its own dis-
turbance of the medium. The combined action of the two waves can be
pictured by adding the ordinates of the component waves algebraically,
point by point.

I 11
FIG. 25-14. The interference of two waves of the same frequency: I, reinforcement; 11,
annulment.
§ 25-14 INTERFERENCE OF WAVES 465

Two waves of the same frequency, in phase with each other, and moving in
the same direction, produce constructive interference-that is, reenforcement.
This result is shown in part I of Fig. 14, in which the individual waves are
represented by A and B. The resultant wave R is in phase with the com-
ponent waves and has an amplitude equal to the sum of their amplitudes.
Two waves of the same frequency, in phase opposition, and moving in the
same direction, produce destructive interference-for short, interference; if
they also have equal amplitudes the result is a complete annulment as is
represented in part II of Fig. 14, where the amplitude of the resultant R of
the individual waves A and B is zero at all points. In the wave-front con-
structions in Figs. 12 and 13 the multiplicity of wavelets, typified by the
two arcs in each diagram, also interfere with each other, and it can be shown
that these wavelets annul one another except at the outermost points where
the envelope representing the new forward position of the wave front is
located. It can also be shown that there is no back wave.

Frn. 25-15. Interference of two waves of slightly different frequencies to produce beats.

Two waves of slightly di.ffering frequencies produce a type of pulsating inter-

ference that sets up throbs called beats. The effect is most pronounced when
the individual waves have equal amplitude. Two such waves, A and B, are
shuwn in Fig. 15 as originating from vibrating sources at M; the frequency
of the source generating A will be assumed to be greater than that of B by
one vibration per second. Each source sets up a train of waves; the dis-
placements at a particular instant are shown in the figure and their resultant
is indicated by the heavy line R. At the instant shown, the sources are
vibrating in phase, and the waves produce reenforcements at Mand P and
an annulment at N. At an instant ! sec later, the sources will be in phase
opposition, and the waves A and B will be displaced with respect to each
other by one-half wavelength from the positions shown, producing annul-
ments at M and P and a reenforcement at N. At another instant ! sec later
still, the sources will once more vibrate in phase, and the relative positions
of the waves will again be as shown in the figure. These changes will con-
tinually recur as long as the sources are kept in vibration.
Pulsating interference is particularly noticeable with sound waves; here the
resulting sensation is alternately loud and soft. If the waves in Fig. 15 rep-
resent sounds, an observer at any point along the path MNP will notice a
reenforcement and an annulment of sound each second-that is, he will hear
1 beat per sec. If the difference in frequency of the sources were d vibrations
466 WA VE MOTION & SOUND Chap. 25

per sec, then at a given point a reenforcement would change to an annulment

and back to a reenforcement again in 1/d sec, and the observer would hear
d beats per sec. When the beat frequency lies within the audible range of
the ear, § 26-12, the beats are heard not as separate pulses but as a distinct
tone.
25-15. Stationary Waves. Another type of interference is produced when
separate waves of equal frequency and equal amplitude pass through a
medium in opposite directions. Such waves set up a stationary or standing
wave system in which there are points, lines, or surfaces where some charac-
teristic of the wave field, such as displacement or pressure, has zero ampli-
tude. The places where the amplitude is zero are called nodes, and where it
is a maximum are called antinodes.

FIG. 25-16. Construction of two traveling waves to show the location of fixed nodes in
the resulting stationary wave.

The curves of Fig. 16 represent two waves traveling along a cord, wave B
moving to the right and wave C to the left. In part I of the figure the waves
coincide momentarily and yield a resultant R of double amplitude. In part
II both waves have advanced i wavelength in their respective directions, and
their resultant R is reduced but crosses the axis (the cord) at the same places.
In part III, the waves have made a similar advance and are now in phase
opposition, and the resultant R is a straight line along the axis. Thus, as
the waves progress, the nodes N of the resultant remain fixed in space !
wavelength apart, and the maximum variation of displacement occurs at the
intervening antinodes A.
§ 25-15 STATIONARY WAVES 467

The production of stationary waves can be demonstrated with a cord

fastened to a wall at one end and moved transversely with appropriate fre-
quency at the other. The outgoing waves are reflected at the wall with but
little reduction of amplitude, and consequently the two wave trains set up a
stationary wave in the cord.
Figure 17 illustrates the appearance of the cord when it vibrates in three
segments or loops; it is stationary at the nodes N, and it whips back and forth
at the antinodes A.
In conducting Melde's experiment a cord is attached to one prong of a vibrating
tuning fork and stretched along the direction of the prong; the stretching force is
adjusted so that stationary waves are produced in the cord. Suppose that the cord is
0.7.5 m long and has a mass of 0.12.5 X 10-3 kg, and that the fork is kept vibrating
electrically at 264 vib per sec. With what force should the cord be stretched so that
it will vibrate in 3 loops?

Fm. 25-17. Nodes and antinodes set up in a cord by a stationary wave.

The length of each loop is 0.2.5 m, the wavelength is 0.50 m, and consequently the
Yelocity of propagation is 264 X 0 ..50 = 132 m/sec. The required stretching force in
. 0.12,5 X 10-3 kg ( m)2
the cord is found by Eq. 2.5.4 to be S = m1 v2 = 0 _75 m X 132 sec = 2.90
newtons (about 0.65 lb).
Another interesting way in which the production of stationary waves can
be illustrated makes use of a flat dish containing mercury, and a tuning fork
with a stylus on each prong. When
the fork is set into vibration and
each stylus is adjusted to touch
the mercury, circular waves will
proceed from the points of contact.
These are represented at some par-
ticular instant by the concentric
circles in Fig. 18, the full lines repre-
senting the crests and the dotted
lines the troughs. At points where
full lines intersect or dotted lines
Frn. 25-18. Wave interference pattern pro-
intersect, the waves from the two duced upon a liquid surface by two vibrating
sources meet in phase, and the dis- sources.
turbance of the surface by one
source is reenforced by that of the other. Between these points the inter-
sections of full lines with dotted lines represent the meeting of a crest from one
::;ource with a trough from the other and are, therefore, points where destruc-
468 WAVE MOTION & SOUND Chap. 25

tive interference occurs. The surface of the mercury is thus divided into
regions of maximum disturbance (shown by the heavy lines), separated
by lanes where the surface remains almost perfectly at rest.

PROBLEMS
1. Compute the length of the sound waves produced in air by a piano string that
vibrates 264 times per sec. Take the speed of the waves to be 1100 ft/sec.
2. Compute the length of the waves used in broadcasting a radio program at a
frequency of 660 kilocycles/sec. Assume the waves to travel at a speed of 3 X 108
m/sec.
3. The tip of a reed vibrates with harmonic motion at a frequency of 300 vib per
sec. Determine the shortest time after the reed passes through its midposition that
the displacement of the tip will be one-half of its amplitude.
4. A string 200 m long has a mass of 120 gm. Determine the speed of a trans-
verse wave in this string when it is stretched with a force of 9 newtons.
6. A stretching force of 0.25 newton is applied to a long wire which has a mass of
5 gm per meter of length. If the wire is set into vibration with a frequency of 15
vib/sec, what is the wavelength of the transverse wave set up in it?
6. If a wire 120 cm long which has a mass of 0.60 gm is stretched between two
points, to what tension in dynes must the wire be subjected in order that a transverse
wave set up in it may travel from one end to the other in 0.004 sec?
7. A rope weighing 0.1 lb per ft is fastened at one end to a support, and to the
other end is tied another rope weighing t as much per foot. The ropes are subjected
to a tension of 20 lb and the lighter one is set into vibration with a frequency of
5 vib/sec. There will be three distinct waves: one advances tmvard the knot, another
is reflected back at the knot into the lighter rope, and the third is transmitted onward
into the heavier rope. Find the frequency, speed, and wavelength of each.
8. Derive an expression for the speed of a transverse wave in a stretched wire in
terms of the radius of the wire, the density of the material of which it is made, and the
force with which it is stretched.
9. A longitudinal wave sent from a ship to the bottom of the ocean returns after
a lapse of 1.62 sec. Take the elasticity of sea water to be the same as that of pure
water, and compute the depth of the ocean at the ship's location.
10. An observer 1.25 mi away from a gun sees the flash when it is fired and hears
the report 5.8 sec later. The air temperature at the time of the test was 22°C. From
these data compute the speed of sound at 0°C.
11. A road extends east 1 mi from A to Band then north 2 mi from B to C. Clocks
at B and C strike the hour of one at the same instant. What time interval will elapse
between the two sounds as heard by a listener at A?
12. Compute the speed of sound in oxygen under standard conditions of temperature
and pressure.
13. A brass pipe 120 ft long is struck at one end, and sound waves travel to a listener
at the other end both along the pipe and through the air. Take the specific gravity
of brass to be 8.5, and compute the time interval between the two sounds heard by the
observer.
14. Find the increase in the speed of sound in the atmosphere due to a temperature
rise from O to 20°C.
15. (a) How long would it take a sound wave to travel 75 ft in an auditorium at
20°C? (b) How far would a radio wave travel in the same time?
16. A listener outdoors who is 40 ft away from a steady source of sound moves to a
new position 10 ft nearer the source. Compute the ratio of the sound intensity at the
new position to that at the initial position.
 PROBLEMS 469

17. A light wave traveling in air impinges on a glass plate, making an angle of in-
cidence of 60°, and the angle of refraction is found to be 35.3°. Find the speed of light
in the glass.
18. When a light wave is directed upon the surface of water at an incident angle of
30° the angle of refraction is 22°. If the angle of incidence is increased to 4Ei 0 , what
will the angle of refraction become?
19. In the diagram S is a steady source of sound mounted in a sound-proof enclosure
E from which two tubes A and B extend to an observation point O where the resulting
sound is detected by means of a stetho-
scope. The length of tube A is fixed at
3 ft and that of tube B can be varied. s
When B is 3 ft long the intensity at O is a •
maximum, and when B is lengthened to
4.25 ft the intensity at this point becomes
a minimum. Compute the wavelength of
the sound and the frequency of thf PROBLEM 19.
source.
20. Two harmonically vibrating sources of sound when placed side by side are found
to produce 20 beats in 15.4 sec. If one source vibrates 528 times per sec, what are
the possible frequencies of the other source?
21. Two identical horns, operated electrically at a frequency of 330 vib/sec, are
placed several feet apart, and stationary sound waves are formed in the intervening
space. Compute the distance between adjacent nodes near the midpoint between the
sounding sources.
22. A cord 2 m long has a mass of 2.5 gm and is stretched with a force of 100 newtons.
Two waves of equal amplitude, set up by pulses having the same frequency, move
along the cord in opposite directions. What must the frequency be in order that t·he
nodes in the resulting stationary wave shall be separated by a distance equal to one-
third the length of the cord?
23. In Melde's experiment a cord hangs vertically from a horizontal vibrator which
has a frequency of 120 vib/sec. If the cord is 90 cm long and has a mass of 0.2 gm,
what mass should be attached to its lower end in order to make the cord vibrate in
three loops?
 26
SOUND
The term "sound" is used in two senses: subjectively, it signifies the sensa-
tion experienced in the act of hearing, and objectively, it signifies the vibratory
motion which gives rise to that sensation. It is used in the latter sense in
the science of Acoustics, which is concerned with vibratory motion and the
production of waves, the propagation of sound waves in different mediums,
and the effect of discontinuities in the wave paths. The preceding chapter
deals with some of the topics that are common to sound, light, and other
wave disturbances. The present chapter deals with the characteristics of
sound and their measurement, basic forms of sounding sources, musical scales
and instruments, human organs of speech and hearing, rudiments of archi-
tectural acoustics, and ultrasonic waves.
26-1. Characteristics of Sound. Many terms are employed in ordinary
language to convey impressions of sounds; these include howl, whistle, squeal,
rustle, rumble, and hum. Most of these sounds would be classed as noises,
in contrast with those that are spoken of as musical tones. The distinction
is based largely upon the regularity of vibration of the source and the degree
of damping, as well as upon the ability of the ear to recognize components
that have a musical sequence. When a stick is dropped upon a table top,
the sound produced is definitely a noise, but when a number of sticks of dif-
ferent lengths are dropped one after another the effect of musical tones is
clearly evident. The complicated sounds of speech are formed by grouping
the more-or-less sustained tones of vowels and the impulsive launching and
quenching of these tones by the consonants.
The ear can distinguish tones that differ in pitch, in loudness, and in quality.
Each of these characteristics is associated primarily with a single property
of the sounding body or of the waves which it produces. Thus, pitch is
determined chiefly by the frequency of vibration, loudness by the intensity
of the sound, and quality by the nature of the vibrations as revealed by wave
shape. In Fig. 1 are several curves that illustrate both the shapes of sound
waves and the characteristics of the vibrations producing them. The sine
curves A and B differ in frequency, the sound represented by B having the
higher pitch. Curves A and C differ only in amplitude, A representing the
louder sound; with the amplitude of A twice as great as C, the sound will
have 4 times the intensity of the latter, although the loudness sensation will
470
§ 26-2 INTENSITY OF SOUND 471

not in·crease in the same proportion. Curves A and D differ in shape, D

having high-frequency components that are not present in A; the curves
represent sounds of different quality.
The pitch of a tone depends primarily upon the frequency of the sounding
source; pitch and frequency are commonly regarded as having the same value,
although pitch is influenced somewhat by sound intensity and wave form.
The dependence of pitch on the frequency of the source can be shown experi-
mentally by means of a siren, a device which consists of a disk with regularly
spaced holes through which air is blown. As the disk revolves, the ,air
stream is repeatedly interrupted, and the resulting puffs in the medium be-
come condensations and rarefactions, which are transmitted to the ear.
The rise in pitch of the sound with increased speed of the disk can be exhib-
ited strikingly in this manner.

Il m
Fm. 26-1. Sound waves which differ in frequency, amplitude, and wave form.

26-2. Intensity of Sound. The term sound intensity used in the preceding
section stems from the broader concept of wave intensity explained in § 25-11.
For sound, the intensity is defined as the time rate of transmitting vibratory
energy per unit of sectional area of the sound wave. The intensity of a
sound depends upon certain physical characteristics of the sound wave, spe-
f'ifically its amplitude and frequency of vibration.
In order to establish the relationship between sound intensity, amplitude,
and frequency, it will be assumed that the vibrating particles have harmonic
motion. In a thin layer of the medium of thickness dx that lies in the wave
front, all particles transmitting the sound wave have the same momentary
displacement. If their amplitude, or maximum displacement, is r and their
greatest speed is v, and if the layer considered has an area A and a mass m,
then the maximum kinetic energy of the particles is
Addx
Ek = ! mv2 = - 2- (21r f r)2,

where d is the density of the medium and .f is the frequency of the sound
wave. The energy of the la:ver will all he of kinetic form when the particle::;
sweep through their equilibrium positions and will all be of potential form
472 WA VE M01'10N & SOUND Chap. 26

when they have their maximum displacements; at other times the energy
will be of both kinds but the total will always be as just expressed, provided
no energy is wasted. In consequence, the total energy per unit volume of the
medium is

(26.1)

and may be termed the energy density of the wave. With cgs units, the
result will be in ergs per cubic centimeter. As the layer mover. with the
speed v cm per sec in the direction of propagation, the energy of its particles
is passed along, and the time rate of transmission of energy per unit of area
of the wave front is v times the foregoing energy density. This product is a
measure of the physical intensity of the sound wave, and therefore the inten-
sity of sound in ergs per second per square centimeter is
I = 2 rr 2v f2 r 2 d. (26.2)
This result shows that the intensity of a sound in a given medium is propor-
tional to the square of the amplitude and also to the square of the frequency
of vibration.
Equation 26.1 for energy density applies to a plane wave or to a spherical
wave at a great distance from its source. At any distance from a point
source, the energy density in the wave varies inversely as the square of that
distance, § 25-11; consequently, it may be concluded that the amplitude of
vibration r varies inversely as the distance from the source.
Associated with the intensity of a sound is the difference between the
pressure in a condensation and that in a rarefaction. As the wave moves
farther from the source, this pressure difference becomes less and less until,
finally, the condensations cannot be distinguished from the rarefactions and
the sound has disappeared. It can be shown that the intensity is propor-
tional to the square of the pressure variation regardless of the frequency.
Thus, by measuring pressure changes, the intensities of sounds having dif-
ferent frequencies can be compared directly, and instruments which make
such measurements are preferred to those that measure amplitude.
26-3. Intensity Levels. A law that is applicable within limits to all human
senses was expressed by the German physiologist Ernst H. Weber (1795-
1878); it states that the smallest noticeable change in the stimulus is a con-
stant fraction of that stimulus. Hence, if I represents the stimulus and ill
the least change that can be noticed in it, then
M cc I.
The chief importance of this law is the use that was made of it by Gustav T.
Fechner (1801-1887), German psychophysicist. He concluded that as the
stimulus is increased, the sensation produced by that stimulus can be assumed
to increase by amounts which are proportional to ill/ I. Thus, in calculus
§ 26-3 INTENSITY LEVELS 473

notation, the increase in the sensation can be expressed as

dl
d8 = K-,
I
where K is a constant. By integration the sensation is found to be

8=K LO
I dl I
-=Kln-,
I Io
where Io is the least perceptible or threshold stimulus. Therefore, the sensa-
tion is proportional to the logarithm of the stimulus.
With reference to sound, if 11 and I2 represent the intensities of two sounds,
and 81 and 82 the corresponding sensations of loudness produced by them,
then the difference in loudness is
12 12
a = 82 - 81 = K ln - = C log -.
11 11
The values of a are commonly spoken of as intensity levels and are expressed
in bels (after Alexander G. Bell) if C is taken as 1, and in decibels if C is
taken as 10. The decibel (abbreviated db) is one-tenth as large as the bel;
it is the standard unit for intensity level and for attenuation and power gain,
§ 23-8. The change of intensity level in decibels that corresponds to a
change in intensity from l 1 to 12 is therefore
12 (26.3)
a= 10 log 11.

The significance of 1 decibel can be gained from this equation by letting

a = 1 and solving for the intensity ratio; the result is 12/11 = 10°· 1 = 1.26,
which shows that 1-db rise in the intensity level corresponds to a 26 per cent
increase of sound intensity. If a particular sound is strengthened so that
it has 10 times its original intensity, then the intensity level has been in-

20 -!-- -

I.;

"'
Gi /
I/
..0 15 ....
·uG> I/
-0 /
v
G) I/
> 10

-
..!?
>-
·;;; I/
I/

-
E
c: 5
G>

/
/
v
....
/

0v
1 2 4 6 810 20 40 60 100
Intensity ratio

Frn. 26-2. The decibel scale of sound intensity levels.

474 WA VE MOTION & SOUND Chap.26

creased by an amount given by Eq. 26.3, namely, 10 log 10 = 10 X 1 = 10 db.

Again, if the sound is strengthened to 100 times its original intensity, then the
level has been increased by 10 log 100 = 10 X 2 = 20 db. Intermediate
values can be read from Fig. 2.
Just as ground elevations are measured from sea level as a datum, so in-
tensity levels are ordinarily measured from that of a particular sound intensity
chosen arbitrarily as a reference standard. If the reference intensity is taken
as Io and the intensity of a sound under consideration is taken as I, then the
intensity level of that sound can be expressed by
I
a= 10 log-.
Io
A commonly used reference datum 10 is 10-12 watt per meter2 ; another is
1 microwatt per centimeter2 •
26-4. Quality of Sound. The tones produced by lightly struck tuning
forks have wave shapes approximating the sine waves A, B, and C in Fig. 1
and are often ref erred to as pure tones. The
tones produced by most sources can be repre-
sented by composite waves, in which the fun-
damental mode of vibration, that corresponds
to the lowest pitch, is accompanied by others
I f=lOO I I
of higher frequency. Herman L. F. Helmholtz
I I I (1821-1894), German physiologist and physi-
c I I I
I 5µ d (\ I
cist, showed that the quality of a tone depends
upon the number of modes in which the source

.~
-u
~ p~
I f =300 I I
is vibrating, and upon their relative frequen-
cies and intensities. It is this quality char-
acteristic that distinguishes tones of like pitch
~ I I
·- I II and loudness when sounded on different types
~ 2·5 p--T\ d (\ (\ I of musical instruments.
°' 1Vv OVvj
J f=500 I I
It is possible to produce a tone of any desired
I I Resultant I quality by combining pure tones in suitable
10 I - - I I proportions. For example, the various vowel
sounds, which are known to have complicated
I I I wave forms, can be duplicated by combining
Time the pure tones of particular organ pipes.
Figure 3 represents the compounding of three
sine curves of different frequencies; the resul-
tant curve at the bottom is obtained by add-
Fw. 26-3. Synthesis of a wave
from its harmonic components. ing the ordinates at each point along the
time axis. The resultant can also be expressed
as the sum of three terms, each having the form given by Eq. 25.2; for the
amplitudes and frequencies noted in the figure the equation of the resultant
is as follows :
§ 26-5 VIBRATING STRINGS 475

y = 10 sin (2 1r 100 t) + 5 sin (2 1r 300 t) + 2.5 sin (2 1r 500 t),

in which y is the instantaneous displacement of the resultant at any time t,
reckoned from the origin.
The reverse process, namely that of resolving a sound wave into its com-
ponents, can also be carried out. Curves of sound-wave shapes can be ob-
tained experimentally in several ways (§ 26-14) and can be measured by
suitable instruments. A mathematical analysis for determining the relative
amplitudes and phases of the component pure tones is based upon a general
principle stated by Jean B. J. Fourier (1768-1830), French mathematician,
to the effect that any periodic wave form can be resolved mathematically
into component sine curves, of definite amplitudes and phases, and having
frequencies in the proportion 1, 2, 3, · : · n. Several types of machines
patterned after the planimeter enable an operator to determine the sinusoidal
components of complicated wave shapes; these are called harmonic analyzers.
26-6. Vibrating Strings. The bowing of a string that is under tension sets
up disturbances which travel to the ends of the string and are there reflected
back again. At either end the incident and reflected waves have the same
frequency and essentially the same amplitude and, since they move in oppo-
site directions, establish a stationary wave in that portion of the string,
forming nodes at intervals of one-half wavelength from that end. The same
effect is produced at the other end of the string. Since the reflected waves
travel from end to end of the string repeatedly, the nodes produced by reflec-
tion from both ends will coincide only for definite wavelengths. If the string
were set vibrating with a multitude of frequencies at the same time, most of
these vibrations would annul one another, and only a few would persist.
These are called free vibrations; they are the ones for which the length of the
string is an exact number of half wavelengths. The free vibrations of a
string are transverse and set up condensations and rarefactions in the sur-
rounding air; these pressure variations proceed away from the string as a
longitudinal sound wave.
The fundamental mode of vibration of a string is that for which the ends
of the string are the only nodal points, and the center is the only antinode.
This condition produces the fundamental tone of the string, as is represented
in Fig. 4. Herein ABCDE represents a sine wave of length }.., and AC repre-
sents the string of length Z. During its vibration, the string travels periodi-
cally from its extreme upper position ABC to its extreme lower position
AB'C, and forms a single loop, which can be observed by the blurred pattern
"'-~~~~B~~~~-x~~~~~~~~--'~:

I
----------- ---------:.J--
:c~~-----------~~; E
K-~~~~8_'_ 1 : D

Fw. 26-4. Relation between string length and wavelength.

47G WA VB MOTION & SOUND Chap. 2G

it makes. Thus, for the fundamental mode of vibration the string forms one
loop and its length is a half wavelength.
Additional modes of vibration develop when nodes are established between
the ends of the string. For a uniform and flexible string vibrating between
fixed supports, these additional modes, together with the fundamental, are
called harmonics, and their frequencies are in the ratio of the integers: l, 2,
3, 4, and so on. Figure 5 illustrates the simpler modes of vibration, in which
the string forms one, two, and three loops. For a string of length l that
vibrates in n loops, the wavelength of the corresponding harmonic will be

x= ~-
n
(26.4)

The fundamental is also called the first harmonic, the mode having twice the
frequency is called the second harmonic, and so on. The frequency f of the
string is obtained from Eq. 25.1 for the speed of the wave, namely v = P,,.

N6cles at ends only A= 2 /=~

1
IE ==:>I _ _
Fundamental .or
first harmonic

0 ne 'mtermed.1ate nod e ~
"= =2 , " ~ secon dh armonic·
2 / ~
~1-1 I
Two intermediate nodes A= 3 ~ Third harmonic

Frn. 26-5. Harmonic vibrations of a string.

The vibration rate can be expressed in terms of the constants of the string
by combining the foregoing equations with Eq. 25.4. Thus, the frequency
in vibrations per second becomes

J = !!:...
2 Z\)~
rs, (26.5)

where n is the number of loops in the string of length Z, Sis the stretching
force of the string, and m1 is the mass of the string per unit of length.
J
As an illustration, determine the tension needed in the aluminum-covered stee
A-string of a violin to give it a fundamental vibration rate of 440 per sec. The string
is 32.7 cm in length, 0.0523 cm in diameter, and its density is 3.5 gm/cm3 •
Since m1 = 11'r2d, the stretching force becomes

S = 4 71'd(lrf) 2 = 4 71' 3.5 -gm3 ( 32.7 cm X 0.02615 cm X 440 vib)2

-
cm sec
= 62.3 X 105 dynes
This is equivalent to 14.0 lb.
§ 26-6 VIBRATION OF RODS AND PLATES 477

An experiment to verify Eq. 26.5 can be carried out with a sonometer,

which consists of a string or wire mounted over a sounding board and resting
on movable bridges. One end of the string is rigidly fastened and the other
passes over a pulley and carries a weight to put the string in tension. By
bowing the string; at chosen places and by touching it lightly at appropriate
points, the string can be set into vibration in a number of modes, as illus-
trated in Fig. 5. The stretching force can be varied by altering the weight,
and the length by shifting the movable bridges.
26-6. Vibration of Rods and Plates. Vibrations of a longitudinal character
can be set up in a solid rod by stroking it lengthwise with a small rosined pad.
The possible modes of vibration depend upon the manner in which the rod
is supported.

X=41=~12,
" ll~ ~
Ii h First
armonic
I ------!I
I
I I , I
I l-3--l
I

<,,____..;.>\<:::J
I

X=~=11.~~ Third
3 % 1 harmonic
I I
I I
I I _!___J
I r-- 2 I
>..= 21,.21
1
~t~
I
i~ First .
harmonic
I I I
Frn. 26-6. Modes of vibration of a rod.

When the rod is supported at one end, it will vibrate in its fundamental
mode with a node at that end and an antinode at the other. This longitu-
dinal vibration is shown at the top of Fig. 6 in the usual transverse fashion;
the rod forms a half loop, and the wavelength would extend over four rod
lengths or 4 Z. For the next higher mode of vibration there is one intermediate
node, which is located two-thirds the way from the fixed end, as shown at the
center of the figure; the wavelength is 4 Z/3. The bottom diagram shows a
rod supported at the center so as to form two half loops; the corresponding
wavelength would be 2 Z. In general, if there are n loops, the wavelength of
t h e v1"b ration
. 1s 2 .l Theory shows t h at t h ese statements about v1brat-
. 'A = - .
n
ing rods apply only to wavelengths which are large as compared with the
thickness or width of the rod.
478 WA VE MOTION & SOUND Chap. 26

The frequency of the longitudinal vibrations in a rod is found by combining

Eqs. 25.1 and 25.5 to be
(26.6)

where Y is Young's modulus of elasticity of the rod, and d is its density.

With cgs units l is in centimeters, Yin dynes per square centimeter, and din
grams per cubic centimeter; with mks units l is in meters, Y in newtons per
square meter, and din kilograms per cubic meter. A rod fixed at one end or
at the center can vibrate only in the odd-numbered harmonics.
Transverse vibrations in a straight rod or bar are more complicated than
longitudinal vibrations, because the end segments are different fractions of
a loop in the various modes and consequently the higher frequencies are not
exact multiples of the fundamental frequency. If a vibrating rod has two
nodes and is bent at the center, then, as the bending
Prongs progresses, the nodes will approach each other, and in
the final U-shape the nodes will be so close together that
the rod vibrates practically as two separate bars, each
fixed at one end. Figure 7 illustrates such a bent bar
with a stem attached to the center to form a tuning fork,
the nodal points being indicated. Transverse motions of
the prongs PP cause an up-and-down motion in the
stem, and this motion can be arranged to impart vibra-
Fw. 26-7. Nodal
points of a tuning tion to a sounding board and thence to a column of air
fork. to intensify the sound produced by the fork. Tuning forks
are usually made of steel, aluminum, or magnesium.
The vibrations of plates can be investigated experimentally by supporting
them horizontally at the center or edge, sprinkling fine sand upon them, and
setting them into vibration by mechanical or electrical means. The sand

FIG. 26-8. Nodal lines of vibrating plates.

particles will hop about and accumulate in places of least motion, thereby
indicating a series of nodal lines. Patterns of such lines are called Chladni
plate figures; a few are shown in Fig. 8 for plates clamped at the center. A
§ 26-7 VIBRA1'IONS OF AIR COLUMNS 479

great many modes of vibration are possible, depending upon the manner of
supporting the plate and of setting it into vibration. The diaphragms of
telephone transmitters and receivers can be tested similarly.
A bell may be regarded as a combination of a plate and a cylinder, with
relatively more metal near the center. Many higher frequencies accompany
the fundamental vibration of a bell when it is struck, and it is the aim of
bell-founders to attain certain relationships between them in order to produce
pleasant tones.
26-7. Vibrations of Air Columns. Disturbances in air or other gaseous
mediums are propagated as condensations and rarefactions in all directions
in open space. When the medium has the form of a column within a rigid
tube, a disturbance produced at one end travels to the other end, is there
reflected, travels back to the initial
end, is reflected again, and so on. Sta-
tionary waves are set up in the gas col-
umn and one or more nodal points are
established within the tube. Figure 9
is used to illustrate how an air column
can be set in vibration by a jet of air
impinging against one side of the tube
at a. The condensation so produced Fw. 26-9. Method of setting an air col-
travels down to the closed end of the umn into vibration.
tube and is reflected. Upon returning
to the upper end it pushes the jet aside, as shown at b, and causes a rare-
faction to start down the tube. This is reflected at the bottom and retraces
its path; upon arrival of the rarefaction at the top of the tube, the condition
represented at a is restored. This process goes on rapidly and repeats itself
as long as the air jet is maintained. The closed end of the tube becomes a
node, the open end becomes an antinode, and the tube length constitutes a
quarter wavelength for the fundamental mode of vibration.
A somewhat similar behavior is observed for a tube open at both ends; at
each end a condensation is reflected as a rarefaction and both open ends be-

Closed tube Open tube

Fm. 26-10. Harmonic vibrations of air columns.

480 WAVB MOTION & SOUND Chap. 26

come antinodes. With one nodal point between these ends, the air column
will vibrate in its fundamental mode, and the tube length constitutes a half
wavelength.
The harmonics produced by air columns of constant cross-section and the
relations between their wavelengths;\ and the length l of the column are indi-
cated in Fig. 10. Their frequencies are given by the following expression:

(26.7)

where v is the speed of the sound wave in the air, or other gas, and n is the
number of loops in the stationary wave. For closed tubes n = 1/2, 3/2,
5/2, · · · , whereas for open tubes n = 1, 2, 3, · · · .
This equation for air columns does not apply perfectly unless an "end cor-
rection," depending upon the cross-section of the column, is made at the
open end. For cylindrical tubes having a radius r that is small in comparison
with the wavelength, a distance about equal to 0.6 r should be added at an
open end to locate the antinode more exactly.
Compute the frequency of the fundamental tone of a tube 8 ft long and 4 in. in
diameter. Take the speed of sound in air as 1100 ft/sec.
If the end correction is neglected, the frequency is f = ! X 1100 ft/sec/(2 X 8 ft)
= 34.4 vib/sec when closed, and f = 1 X 1100 ft/sec/(2 X 8 ft) = 68.8 vib/sec when
open. With the end correction of 0.6 X 2 in. included, the effective tube length is
8.1 ft when closed at one end, and 8.2 ft when open at both ends. The corresponding
frequencies are 33.9 and 67.0 vib/sec.

26-8. Resonance. The production of sound by some vibrating systems,

such as strings, rods, and air columns, has been considered in the three pre-
ceding sections. It was assumed that these systems were caused to vibrate
without constraint, thereby producing free vibrations-that is, vibrations
having frequencies determined entirely by the constants and properties of
the vibrating bodies themselves. Such bodies may also be forced to vibrate
Fork
by periodic impulses imparted by outside agents;
then the bodies are said to execute forced vibra-
~
tions. When conditions are so adjusted that the
Air column forced vibrations have the same frequency as the
Vessel free vibrations of the body upon which they are
impressed, the free vibrations reinforce the re-
ceived ones--an effect known as resonance. When
the impressed vibration has a different frequency
from that of the free vibrations of a body, the
received impulses sometimes help and sometimes
Frn. 26-11. Resonance of an d d ff h
air column. hinder the free vibrations, an o not a ect t e
free vibrations appreciably.
A simple way to illustrate the phenomenon of resonance is to hold a tuning
fork over an air column, as shown in Fig. 11, and to adjust the level of water
§ 26-0 MUSICAL SCALES 481

in the cylindrical tube by raising or lowering the associated vessel. Upon

sounding the fork and altering the height of the air column, resonance is
attained when the audible response is a maximum. This experiment affords
a simple means of determining the speed of sound in air; in such a test a
tuning fork of known frequency is used and the length of the column at
resonance is measur d.
For example, a stan rd tuning fork vibrating 440 times per sec is observed to pro-
duce resonance when eld above an air column 18.0 cm long, and again when the
length of the air colum is increased to 55.5 cm. Find the speed of sound in air.
The wavelen th is (55.5 - 18.0) = 75 cm = 0.75 m. The speed of sound is given
by Eq. 25.1 as ( 40 vib/sec) X 0.75 m = 330 m/sec.
There are many applications of the phenomenon of resonance. Tuning
forks are often mounted on top of wooden boxes closed at one end and open
at the other; the length of such boxes must be about one-quarter wavelength
of the sound emitted by the fork in order to produce a loud response. Reso-
nance is utilized in the tuning of a radio receiver by adjusting its inductance
or capacitance so that the circuit frequency is the same as that of the radiation
from the desired broadcasting station, § 24-2. There are instances where
resonance may build up free mechanical vibrations of such amplitudes in
structures as to produce dangerous effects in them. Soldiers break step in
marching across bridges, for example, in order to eliminate the possibility of
setting up destructive vibrations in the various members of such structures.
26-9. Musical Scales. The charm of music is based upon the blending
and the succession of sounds to give pleasing auditory sensations. The char-
acteristics of the ear impose certain physical restrictions on the frequencies
of the sounds to be combined to obtain harmonious effects. The ear would
interpret quite differently a frequency increase of 50 vib per sec in an initial
sound having a pitch of 200 vib per sec and a like increase in a sound having
a pitch of 400 vib per sec, but would give the same interpretation to an
increase of 100 vib per sec in the sound of higher pitch. Thus, the ear recog-
nizes two sounds to have the same tonal interval as two others, if the fre-
quency ratios, rather than the frequency differences, are the same for the two
pairs. Hence, with tones of frequencies !1 and f2, the musical interval is
.h/.f1 and not h - f1.
Further, two simultaneous sounds that have nearly the same frequency
produce pulsations in their combined intensity. Two tuning forks, one
making 200 and the other 205 vib per sec, set up 5 beats per sec if sounded
together (§ 25-14), and produce an unpleasant musical effect. But if the
frequency difference of two sounds is large, the beat frequency is perceived
as a new tone, called a combination tone.
The complete sequence of tones used for musical selections constitutes a
scale, and the tones are designated as notes of the scale. Experience shows
that tones having frequency ratios of 2 to 1, 3 to 2, 4 to 3, 5 to 3, 5 to 4,
and 6 to 5 produce a pleasing effect, and musical scales are based upon these
482 WA VE MOTION & SOUND Chap. 26

ratios. The scales are formed by using three combinations called triads,
each of which is a chord formed of three tones. In such a chord, the octave
of a tone may accompany or replace the fundamental without altering the
nature of the chord.
The major diatonic scale of eight notes, beginning with middle C as the key-
note, includes notes D, E, F, G, A, and B of successively higher pitches in
reaching the octave c. The frequencies of these notes are determined by the
major triads CEG, FAc, and GBd, which notes have frequency ratios of
4: 5: 6, and also by the pitch of some one note regarded as standard. Based
upon the standard "concert pitch" of 440 vib per sec for A in the treble clef,
the notes of the major scale have the following frequencies and intervals:
c D E F G A B c d
6
Triads

Name
{4
do re
5

mi
4

fa
4
sol
5

la
5
Sl
6

do
6
re
Frequency 264 297 330 352 396 440 495 528 594
Intervals -9 10
- 16 9
-8 10
-9 16
8 9 15 9 8 15
The intervals 9/8 and 10/9 are called full tones, and the interval 16/15 is
called a half tone. A study of the frequencies tabulated shows that the
tones of a triad as well as their harmonics produce no disturbing beat notes.
For example, the third harmonic of C coincides with the second of G, the
fifth of C coincides with the fourth of E, and the sixth of E coincides with
the fifth of G, and so on.
The minor scale is built upon three minor triads for the same notes as the
major scale, but having frequency ratios of 10 :12 :15. The intervals have
the same values as before but have a different sequence. When the calcula-
tions have been made it will be found that three additional notes are needed
to produce the minor scale; these are somewhat lower than E, A, and B of the
major scale. Music rendered in the minor scale usually has a more plaintive
character than that in the other scale.
To accommodate different instruments and voices, it is desirable to have
sufficient tones to permit changing the keynote from C to some other note.
If, for example, the successive notes were to be determined for the key of D,
the procedure followed above would indicate the successive frequencies for
the major scale to be 297, 334, 371, 396, 4--!5, 495, 557, and 59--1 vib per sec.
These frequencies agree with those previously found for notes G and B as
well as for the keynote, and approximate them for notes E and A; but two
are quite distinct-these notes are designated as F sharp and C sharp. Simi-
lar computations for other keys in both scales would show the necessity of
having a large number of separate notes if it were desired to render a selection
in any key.
To avoid the multiplicity of notes and yet provide sufficient flexibility in
§ 26-10 MUSICAL INSTRUMENTS 483

musical instruments such as the piano and organ, that produce sounds of
fixed frequency, a scale has been developed which has 12 intervals and has
the same frequency interval between consecutive notes. This so-called
tempered scale has a frequency interval of the twelfth root of 2, namely,
1.0.595. With 440 vib per sec for A, the scale begins with C at 261.6 vib per
sec. The tempered scale is now universally used, and only persons with well-
trained ears can detect the slight changes of pitch from the natural scales.
26-10. Musical Instruments. Instruments for the production of music
utilize various forms of vibrating bodies as the sound-generating elements,
and some instruments have sounding boards or their equivalents to reinforce
the tones produced. The following list gives some examples of the instru-
ments that make use of the usual acoustic elements:
Air columns, conical Bassoon, cornet, oboe, saxophone
Air columns, cylindrical Clarinet, flute, organ, piccolo, trombone
Plates and membranes Cymbal, drum
Reeds and rods Reed organ, xylophone
Strings, bowed Violin, viola, violoncello, double bass
Strings, plucked Guitar, harp, mandolin, ukulele
Strings, struck Piano

The excitation of the air columns in some of the so-called wind instruments
is accomplished by a thin stream of air, or a reed which often has no definite
frequency by itself. The air column then vibrates by resonance at a fre-
quency determined by its length, and reacts upon the exciter to keep its
vibration rate unchanged. This method with reeds is utilized in the clarinet,
saxophone, and some organ pipes, and with the human lips as reeds in the
cornet and trombone. In some other wind instruments, such as the flute
and piccolo, the air is set into vibration without reeds. The length of the
air column is varied in the flute and clarinet by opening side holes in the tube,
in the cornet by adding lengths of tubing by means of valves, and in the
trombone by sliding one tube within another.
The stringed instruments have various numbers of strings that are con-
stant or alterable in length. In the piano each string is of definite length;
the string for the highest tone is usually from .5.0 to 5 ..5 cm long, and the
others are progressively longer. There are 88 notes extending over more
than seven octaves from 27! to 4186 vibrations per sec. The strings of
various diameters are set into vibration through key action by felt-nosed
hammers at places approximately one-eighth the string length from one end.
The lower tones have many harmonics, whereas the high tones have few; the
middle tones of the scale have at least ten with well-proportioned intensities.
In the violin family of instruments there are four strings of equal length
stretched over a resonant box of particular shape and construction. The
G (low), D, A, and e strings are tuned to fifths, that is, the relative frequencies
of two neighboring strings are 2: 3. Other tones are produced by varying the
lengths of the strings by pressing them against the flat fingerboard with the
48-i WA VE MOTION & SOUND Chap. 26

fingers. The strings are set into vibration by drawing a bow strung with
rosined horsehair across them, the vibration being a succession of alternate
forced and free movements as the string adheres to the bow because of the
pressure applied, and slips back again by virtue of its elasticity. The third,
fourth, and fifth harmonics are particularly prominent in the tones of the
violin.
In recent years several types of electrically operated instruments have
become available which produce the musical effects of the organ. In one of
these, the Hammond organ, a series of
toothed disks is driven by a constant-
Co;I ~
speed electric motor, one disk being
Magnet
provided for the fundamental tone cor-

~~ responding to each key on the instrument.

The action will be explained by reference
to Fig. 12, which represents a steel disk
Fro. 26-12. Emf-inducing ele-
ment of the Hammond organ. rotated near the tip of a permanent mag-
net, on which there is a coil of wire. The
disk is shaped with teeth and slots, and as
these move in succession past the magnet tip, variations are produced in the
magnetic flux and an alternating emf is induced in the coil. An electron-
tube circuit amplifies this emf
and operates a loudspeaker.
The instrument provides for
superimposing additional emfs
upon that of the fundamental to
give desired tone qualities.
26-11. Vocal Organs. The
organs of speech are composed
of the vocal cords, through
which the lungs force streams Lips
of air, and of the resonating
chambers formed by the throat, Larynx Food passage
mouth, and nasal cavities. Fig- Vocal cords ~ (oesophagus)
ure 13 shows a sectional view of Windpipe
these parts of the neck and head. (trachea)
The larynx is the valve at the F!G. 26-13. Organs of speech.
entrance of the windpipe; it con-
sists of a framework of cartilages
connected by ligaments, including two fibrous bands called the vocal cords.
These form a straight slit (glottis) from 11 to 15 mm long, and when the breath
passes through it the cords are set into vibration and send a succession of
air puffs to the chambers above it. The vibration rate is determined prin-
cipally by the size of the glottis opening and to some extent by the temiion of
§ 26-12 THE EAR 485

the vocal cords. The lips, tongue, and teeth modify the shape of the vocal
passages; certain tonal characteristics are impressed by this action upon the
air puffs and they emerge from the mouth as speech or tonal sounds. When
relaxed, the vocal cords are farther apart and form a V-shaped aperture; in
this condition the passage of air to and from the lungs in normal breathing
occurs without the emission of sound.
The sounds of speech are complicated tones that have many harmonics.
As with musical tones, the quality is determined by the relative intensities
of these harmonics, that is, by the distribution of power with respect to fre-
quency. Measurements on the power of speech sounds show that the aver-
age power for conversational speech is about 10 microwatts. Talking as
loudly as possible raises this average to about 1000 microwatts and talking
as softly as possible without whispering lowers the average to about 0.1
microwatt. This range corresponds to 40 decibels, § 26-3.
The pitch of the voice in singing ranges generally from about 80 to 300 vib
per sec for bass voices, and from about 250 to 850 vib per sec for soprano
voices, but these ranges are extended considerably by many individuals.
26-12. The Ear. In the process of hearing sound, the acoustic waves enter
the auditory canal of the outer ear and fall upon the eardrum; the vibration
of this membrane is transmitted through the middle ear to the inner ear and
received by nerve endings, which in turn send nerve impulses to the brain
and produce the sensation of hearing.

Semicircular Tectorial
canals membrane

Corti fibers

,« Scala Basilar
membrane
. ~ Eustachian \ .... tympani ,.··
tube ~~ #'
-·-· ·.·.,......

Frn. 26-14. General view of the ear and transverse section of the cochlea.

The principal parts of the ear are illustrated in Fig. 14, in which the inner
ear is much enlarged with respect to the outer ear, and the sectional view of
the end-organ (cochlea) is further magnified. The cavity beyond the ear-
drum, called the tympanum, connects with the upper part of the throat
through the Eustachian tube which opens when swallowing occurs, thereby
equalizing the pressure with that of the outside air. The tympanum houses
three tiny bones in the middle ear, the hammer, anvil, and stirrup; these are
coupled and the base of the stirrup is applied to a membrane which closes
486 WAVE MOTION & SOUND Chap. 26

an oval window, called the fenestra ovalis. There is also another membrane
at that side of the tympanum which closes a round window, called the fenestra
rotunda. Both of these membranes transmit the pressure variations of sound
waves in the air to the liquid of the inner ear. This part of the ear, which is
encased in solid bone, can be subdivided into three sections: the semicircular
canals (which do not contribute to the process of hearing but serve as an
organ of balance), the vestibule, and the cochlea.
The cochlea has the form of a spiral of nearly three turns and is the organ
in which the vibrations of sound are translated into nerve impulses. It is
divided along its length into three parallel canals, as shown in the sectional
view at the right of Fig. 14, but the upper two, which are separated by a very
thin flexible membrane, act mechanically as though they were one. The
scala vestibuli and scala tympani have at their ends the oval and round win-
dows, respectively, for communication with the tympanum, and are separated
by a bony projection for about half their length and a flexible membrane,
called the basilar membrane, for the other half. That the cochlea is very
small can be judged from the facts that the length is about 3.5 cm when
straightened out and that the average cross-section of either canal is about
1 mm2 •
The terminal organs of hearing are the Corti fibers which are nerve terminals
in the form of rods with small hairs that extend from one side of the basilar
membrane into the canal of cochlea. Opposite them is a soft loose membrane
called the tectorial membrane. In receiving sound, these two membranes move
relatively to each other and stimulate the hair-like nerve endings, thereby
causing the sound to be heard. Experimental evidence shows that the basilar
membrane responds to different frequencies at different points along its
length, each fiber responding more strongly to a certain band of frequencies
than to others. In the reception of a complex sound, certain areas of the
membrane are activated more than others and the fibers transmit a cor-
responding pattern to the brain.
The range of sounds that can be heard varies with the individual, but the
average range extends from about 20 to 20,000 vibrations per sec. If the
ear is tested with tones of any one frequency and the intensity is changed,
it will be found that the auditory sensation ceases when the intensity has
been reduced to a sufficiently low level called the threshold of audibility, and
also that the sound produces the sensation of feeling and begins to be painful
when the intensity has been increased to an appropriately high level called
the threshold of feeling. Such a series of tests on the ear yields full informa-
tion on its frequency and intensity ranges. Figure 15 shows the results of
carefully controlled experiments on the loudness judgments of persons with
normal hearing. The lowest curve represents the minimum and the highest
curve the maximum auditory response; these and the intermediate curves are
called equal loudness contours. The scale of abscissas is a logarithmic one of
frequencies extending from 20 to 20,000 cycles per second, and the seale of
§ 2G-13 SOUND RECORDING AND REPRODUCTION 487

ordinates is a linear one of intensity levels extending from -10 to 130 db,
the zero value having an intensity of 10-12 watt/m2 , § 26-3. Each contour
represents the sounds of different frequencies that appear to be of equal
loudness to the average normal ear. The chart shows that at low intensities
tones of high frequency seem louder than those of low frequency, and that
at high intensities all tones sound almost equally loud.
Most persons can estimate ,vith considerable accuracy the direction from
which a particular sound comes through the so-called binaural sense. The
waves received by the two ears vary slightly in intensity and phase, and from
early childhood these differences have come to be associated with particular
directions. Improved accuracy may be obtained by use of an instrument

~
Cl)
~120
u

~
Cl)

Cl)
>
..!!
100
-
:-, -,..
I

Feel inc I / --
-- - :-...... _1,.1
.,
~ .... ~
L/

-
v
,/

...:::>
Cl)
,..
......
80 ,r-,... r,...i,..~
~ - ,...... j
~

..."'"'
Cl) 60 '"~ I'.-. '~ --- i---~
-.......... ....
~j
v
Q.. r-... '~
...0
a;
>
40 '
""
!'I'.. f",.."-
"i-,..

r,...t-,
-- .I
/

-
..!!
>- 20
-~ - -
Minimum
I'"
'r--. -~ - .I
/

-QI
.£:
0
20
audibility

50 100
11 I
'"'"

500 1000
""r--
-""'500010,000
Frequency, vib/sec

Frn. 26-15. Sound contours of equal loudness.

for sound ranging which in effect increases the base line between the ears;
this device consists of a pair of microphones mounted in positions correspond-
ing to the ears on an imaginary head of large size. The microphones are
connected through separate circuits tu corresponding receivers ,vorn by the
observer. The direction from which the sound comes can be located by turn-
ing the device until the source appears to be directly in front. The same
result can also be attained by shifting the phase relation of the currents in
the two circuits to balance the phase difference in the sound waves at the
microphones.
26-13. Sound Recording and Reproduction. The recording of sound was
first accomplished by Edison in 1877. His arrangement consisted of a dia-
phragm with a cutting stylus at the back, and a cylinder of tinfoil that
could be rotated and also moved laterally while in contact with the stylus.
A person speaking in front of the diaphragm caused indentations to be pro-
duced in the cylindrical surface, and in this way a record was made of the
speech sounds uttered. Upon returning the cylinder to its original position
488 WA VE MO'PION & SOUND Chap. 2G

and moving it as before, the diaphragm was set into vibration by a needle
that followed the surface indentations, and the diaphragm reproduced the
original sounds. The success of this experiment led to the development of
the dictating machine, which employs cylinder3 of hard wax, and of the
phonograph, which reproduces records made on plastic disks.
In making a phonograph record, the modulated current in the microphone
circuit controls an electromagnet, and this in turn controls a stylus that cuts
a groove in the phonograph disk. The appearance of the groove, when mag-
nified, is a wavy line that represents the composite wave of the sound incident
upon the microphone. In the electrical reproduction of a phonograph rec-
ord, the motion of the needle causes a tiny iron armature to move within a
coil of wire. The resulting disturbance of the magnetic flux induces an emf
in the coil and sets up a proportional current; this is amplified by electron
tubes and operates a loudspeaker.
The sound record of a talking motion picture consists of a narrow strip or
"sound track" which extends along one edge of the film and is usually crossed
with lines of varying density. The record is made by passing a beam of
light through a "light valve" formed by a pair of metal ribbons suspended
in the narrow space between the pole pieces of an electromagnet; the amount
of light reaching the film through a transverse slit depends upon the deflec-
tion of the ribbons, and this in turn is determined by the current supplied to
them by the microphone. In reproducing the record a steady light is di-
rected upon the sound track through a similar slit, and by this means the
transmitted beam is varied in strength in accordance with the varying film
density along the track. This beam strikes a photoelectric cell ( § 3-±-2) and
causes proportional changes in the current through it. The varying current,
properly amplified, operates a loudspeaker and reproduces the original sound.

Oxide::,. gs NS N

Fra. 26-16.
-
~F==============~s===N=s===N===i\Tape
Plastic base/ Motion
Magnetic recording of sound on a tttpc.

The recording of sound is now also done magnetically by using plastic tape
or motion-picture film stock coated with a thin layer of finely divided mag-
netic iron oxide. A record is made by passing the tape or film along a tiny
electromagnet energized by the amplified current from a microphone, and
the oxide becomes magnetized in proportion to the varying current; Fig. lG
shows the usual arrangement. The reproduction is made by drawing the
tape or film over the poles of a similar magnet, thereby inducing emfs in its
eoil that are proportional to the varying magnetizations, and then amplifying
§ 26-14 SOME MEASUREMEN1'S IN SOUND 489

the resulting current sufficiently to operate a loudspeaker. The record can

be erased from the oxide by passing a current of suitable intensity and of
ultrasonic frequency through a companion magnet, thus permitting the tape
to be used repeatedly. The recording operation also utilizes an ultrasonic
(inaudible) bias-a current having a frequency of from 30 to 100 kilocycles
per sec upon which the microphone current is superimposed-to avoid the
distortion due to the bend in the magnetization curve of the oxide at low
excitation.
26-14. Some Measurements in Sound. Systematic progress in sound con-
trol requires a knowledge of acoustic quantities based on actual test. Some
of the experimental methods used are described in the following paragraphs.

1,1/ 00~ ~
1'2:> 25 8 e: c
l,35§g~2
2,3x~~~o
Frn. 26-17. Lissajous figures for different frequency ratios.

Pitch. The pitch of a tone is usually measured by comparison with a

standard. The standard tone may be generated by a card held lightly
against the rim of a rotating toothed disk, or produced by a siren, § 26-1; its
frequency can be computed from a knowledge of the speed of rotation and
the number of teeth in the disk or holes in the siren wheel. By regulating
the speed, the pitch of the standard tone is made equal to that of the tone
being measured. The ear can detect this equality directly if the two sources
are sounded alternately, or by the elimination of beats if they are sounded
together.
An optical method is used for making precise comparison of the pitches of
two tuning forks, one being a standard. This method applies where the
pitches are very nearly in simple proportion, for example, 3 :1. Each fork
has a polished spot on one prong to serve as a mirror, and the two forks are
mounted in such a manner that one fork will vibrate in a horizontal plane
and the other in a vertical plane. A narrow beam of light projected upon
one fork is reflected to the other and then to a screen. When the forks are
190 WAVE MOTION & 80 UN D Chap. 26

vibrated electrically, the beam is given two harmonic motions at right angles
to each other and a figure is traced upon the screen; the shape of the figure
depends upon the frequency ratio of the vibrations. If one of the frequencies
is an exact multiple of the other, the figure will appear stationary; otherwise
progressive changes of phase cause it to pass through a succession of shapes,
returning to the original shape when one fork has lost or gained exactly one
cycle with respect to the other. Figure 17 shows some typical examples of
these shapes, known as Lissajous figures. Each row represents a particular
value for the nominal frequency ratio, as indicated, and the diagrams across
the row show the shapes for increasing phase differences up to ! cycle between
the two sources.
Suppose that a standard fork, which vibrates exactly 200 times per second, is com-
pared with another that vibrates very nearly 600 times a second, and that the figure
on the screen goes through a cycle of shapes in exactly 4 sec. Determine the fre-
quency of the second fork.
The fork under test completes either 2399 or 2401 vib in the 4-sec interval, hence its
frequency is either 599.75 or 600.25 vib per sec. To determine which of these is
correct, the test must be repeated with a material like wax stuck to one prong of the
test fork; this addition will lower its frequency and indicate which result to accept.

Wave Shape. Present methods of wave-shape measurement employ a

cathode-ray oscilloscope (§ 23-13) actuated by a microphone. The sound
waves striking the microphone are converted into equivalent electrical vi-
brations and these are rendered visible upon the fluorescent screen of the
oscilloscope.
Intensity and Loudness. The classical method for measuring sound in-
tensity is due to Lord Rayleigh (John W. Strutt, 1842-1919), English physi-
cist, who found that a suspended disk tends to set itself at right angles to the
to-and-fro motion of the air particles transmitting a sound wave. The disk
is supported by a torsion suspension and is placed where an antinode will be
located in a resonating tube, with its plane at an angle of 45° with the axis of
the tube. When a sound wave is established in the tube, the disk will be
deflected by a torque that is proportional to the intensity of the sound.
The loudness of a noise can be measured by the audiometer, a device de-
signed primarily for the measurement of deafness. This instrument produces
tones of numerous frequencies, the intensities of which can be adjusted to
various sensation levels. The sound is produced in a telephone receiver sup-
ported at a fixed distance from the ear, and its loudness is regulated until it
is just masked by the noise being measured. The difference between this
setting and a similar setting made in a quiet place is a measure of the loudness
of the noise.
Speed. The speed of a sound wave in the open air can be determined by
observing the motion that produces a distant sound, and measuring the time
interval which elapses before the sound reaches the ear, as well as the distance
between source and observer. The time required for light to travel this dis-
§ 26-14 SOME MEASUREMENTS IN SOUND 491

tance is so short that it can be neglected, and the speed of sound can be
computed directly, although the result is subject to considerable personal
error. The distance to a lightning flash can be estimated in this way by
counting seconds until its thunder is heard, allowing 5 sec to the mile.
In the Kundt method for measuring the speed of sound, longitudinal vibra-
tions are set up in a metal rod so placed as to produce stationary waves in
air or other gas within a glass tube. Lycopodium powder or cork filings,
spread over the interior of the tube, permit the nodes and antinodes to be
located. The arrangement is shown in Fig. 18, wherein A represents the

d d
I
I 1
l<- .,I"', )": 1
P
n
(Ej F

•
11 ·Ill ·Ill· ·ilh ,,h ·•h ·•h
D

.FIG. 26-18. Kundt's-tube method for measuring the speed of sound in a gas.

rod clamped at its center near one end of the tube, and P represents a plunger
at the other end. The rod is set into longitudinal vibration by stroking with
rosined chamois and the plunger is moved until the powder shows distinct
differences in appearance along the tube. At the nodal points the powder
will be lumped together, and at the antinodes it will assume sharply defined
ridges across the tube. A node will form at P, and there will be an approxi-
mate node at the disk D fastened to the end of the rod. The rod has a
length Z, and the pitch f of its fundamental vibration is measured by compari-

Osei II oscope

'----.---o-1---1,,o
HOV <>-!--+<>

Oscillator Amplifier

Frn. 26-19. Oscilloscope arrangement for measuring the speed of sound.

son with a siren or equivalent means. The longitudinal wave in the metal
rod has a wavelength 2 Z, and a speed Vm = 2 fl, as explained in § 26-6. If
the average distance measured between successive antinodes is d, the sound
wave in the gas has a wavelength 2 d, and a speed v0 = 2 f d.
Another method utilizes the Lissajous procedure and employs electronic
instruments (Chap. 23). The scheme is outlined in Fig. 19, wherein S is a
source and R is a receiver, both fitted with reflectors arranged coaxially to
concentrate the acoustic power, and one of them movable along a scale so
492 WAVE 1110TION & SOUND Chap. 26

that their separation can be changed and measured. The source of sound
waves is an earphone activated by a vacuum-tube oscillator, and the receiver
is a microphone that supplies its output to a vacuum-tube amplifier. Signals
from source and receiver are fed to the horizontal plates Hand vertical plates
V of an oscilloscope, and adjusted to produce comparable deflections in the
two directions. The result is a 1: 1 Lissajous figure which changes in phase
as the receiver is moved.
In a measurement of the speed of sound in air by this method, the oscillator was set
to produce a tone of 4000 cycles/sec. When the receiver was moved a distance of
4.25 cm, the Lissajous figure on the oscilloscope changed by the full range indicated
at the top of Fig. 17.
The shift of the receiver indicates that 1 wavelength is 2 X 0.0425 = 0.085 m.
Hence the speed of sound in air at the temperature prevailing is v = JX = 4000 X 0.085
= 340 m/sec.

26-15. Doppler's Principle. A person standing near a railroad track can

observe a distinct lowering of pitch in the whistle of a train as the train passes
him. This observation illustrates a, principle applicable to all wave motion
that ,vas developed by the Austrian physicist Christian Doppler (1803-1853).
Applied to sound, this principle states that the apparent frequency of the
sound heard differs from the frequency of the vibrating source from which it
originates whenever the listener or the source moves. This difference is
quite marked even when the speed of motion is only a few per cent of the
speed of sound, which in air may be taken as 1100 ft per sec.
In considering specific applications of this principle, the air will be regarded
as stationary, and motion will be restricted to the line connecting the source
and listener. The following symbols will be used: f for the frequency of the
source, v for the speed of wave propagation, S for the speed of the source, if
moving, and L for the speed of the listener, if moving. Whenever the source
or listener, or both, move, the apparent frequency heard by the listener can be
found by dividing the speed with which the waves pass him by their wavelength.
Source Moving. In one second the moving source emits f waves, and these
spread out into the surrounding medium; in the region in front of the source
these f waves will be crowded into a distance equal to v - S. To a listener
in this region the wavelength will be v f 8 , and hence the apparent frequency
of the sound heard when the source moves toward the observer will be the
quotient of the speed v and this wavelength, or

f'=f(v vs} (26.8)

Similarly, when the source moves away from the listener, the apparent fre-
quency will be
f' = t(-v
v+S
,)·
§ 2G-16 SOUND REVERBERATION 493

When the sounding source is stationary and the listener

Listener 111 oving.
approaches it with a speed L, then the waves pass him with a speed (v + L),
and the wavelength of these waves will be v/f. Hence to the listener the
apparent frequency of the sound will be

.f' = .f (v +v L)· (26.9)

If the listener moves away from the source, the sign of L will be negative.
Source and Listener Moving. ·when both the source of sound and the
listener are moving, the apparent frequency will be changed in the proportion
v ; S by motion of the source, and in the proportion v ~ L by motion of
the listener.
As an illustration, consider a whistle to be moving with a speed of 50 ft per sec
toward a stationary listener, while emitting a sound having a frequency of 300 vib/sec.
Determine the value of the apparent frequency from fundamentals, rather than by
substituting the data in Eq. 26.8.
In 1 sec the whistle will produce a train of 300 waves, the first of which will advance
1100 ft while the last one is just being emitted from the source at a point 50 ft away
from the location it had when the first wave was emitted. This train of waves occupies
a length of (1100 - 50) ft; therefore, the wavelength of the sound is (1100 - 50)/300
= 3.50 ft. The apparent frequency is the wave speed divided by the wavelength, or
1100/3.50 = 314.3 vib/sec.
Next, determine the apparent frequency when the 300-cycle whistle is held station-
ary, and the listener approaches it with a velocity of 50 ft per sec.
The speed with which the waves pass the listener is (1100 + 50) ft/sec. Since the
wavelength of the sound in air is (1100/300) ft, the apparent frequency becomes
(1100 + 50)/(1100/300) = 313.6 vib/sec.
It is interesting to note that the frequency increase in the two illustrations
is not the same, even though the source and listener have the same relative
motion.
26-16. Sound Reverberation. Almost everyone has noticed the persistence
of sound, particularly in a large empty hall. This effect is caused by echoes
repeated in rapid succession and is called reverberation. It may be explained
by considering the effect of sustaining a tone within a room from which no
acoustic energy escapes. When the source is set into vibration, the sound
waves emitted travel to the walls and other surfaces, and are then reflected
back and forth from one surface after another. If none of the acoustic
energy were absorbed, the resulting sound intensity would increase indefi-
nitely. Absorption does take place at the surfaces, however, and hence the
intensity will reach a steady value when the rate of energy absorption equals
the rate of emission. After this condition has been attained and the vibrat-
ing source is stopped, the sound does not cease instantly but dies away as
the acoustic energy within the room is absorbed. The intensity I at an
instant t sec after stopping a sound is given in terms of the maximum intensity
In by the equation:
494 WA VE MOTION & SOUND Chap.26

(26.10)
where e is the base of natural logarithms and k is taken as a constant for the
room under consideration.
The time in which a sound diminishes until it can no longer be heard has
been termed the reverberation time. The concept was made more definite
by Professor Wallace C. Sabine (1869-1919) of Harvard University, a pioneer
in acoustics; the reverberation time is now defined as the interval during
which the intensity of the sound diminishes to one millionth of its initial
value. On this basis the reverberation time can be found by taking Io/I as
1()6 and following the procedure in § 26-3; then its value is

t = .!k ln IoI = 2.303 log 106

k
= 2.303 X ti
k =
13.82
-k-.

The reverberation time can also be expressed as the time in which the inten-
sity level of a sound is lowered 60 db by absorption.
A sound produced in a small room diminishes very rapidly because the re-
flections and absorptions at the walls are repeated at short intervals. In an
auditorium, however, the reverberation time may be much longer, making it
necessary for a speaker to talk very slowly to prevent confusion between the
syllable being spoken and the echoes of the preceding one. On the other
hand, the sound should not die away too quickly, for the room would then be
judged too "dead." Values of the reverberation time which are regarded as
acceptable for different sized rooms are given in Table I.

TABLE I. ACCEPTABLE REVERBERATION TIMES

Volume in ft3 Time in sec

10,000 ............... . 0.7 to 1.2

40,000 ............... . 0.9 to 1.4
100,000 ............... . 1.1 to 1.6
300,000 ............... . 1.3 to 1.8
1,000,000 ............... . 1.5 to 2.0

Choice within the time ranges indicated is governed chiefly by the intended
use of the room or auditorium; the shorter times are more appropriate for
speech (lecture rooms) and the longer ones for music (concert halls).
26-17. Acoustic Absorption. The persistence of sound in a room is due to
incomplete absorption upon successive reflections at its surfaces. The in-
terval between reflections is longer in a large room than in a small one; con-
sequently, the reverberation time varies directly with the volume. For a
similar reason this time varies inversely with the speed of sound. Further-
more, it is influenced by the character of the room surfaces, being reduced if
they are good sound absorbers. An open window returns none of the energy
§ 26-17 ACOUSTIC ABSORPTION 495

that reaches it, and for this reason the opening acts like a perfect absorber.
On this basis the absorbing power of a room may be expressed as an equivalent
number of square feet of open window. These characteristics make it pos-
sible to evaluate the factor kin the expression for the reverberation time given
in the preceding section.
Sabine showed that if the volume of the room in cubic feet is represented by
V, the speed of sound in feet per second by v = 1100, and the absorbing power
of the room in effective square feet by a, then
V
-1 = C-,
k va
in which the proportionality constant C has a value close to 4 for the units
employed. Hence the reverberation time in seconds becomes
t _ 13.82 _ 13.82 X 4 V
- k - 1100 a '

or approximately t = 0.05 v.
a
(26.11)

A number of modifications of this formula exist for representing the acoustical

conditions of a room more accurately, but they are too complicated for in-
clusion here.
The absorbing power of the room, including furnishings and audience, is
found by adding a number of terms for the various surfaces, each term con-
sisting of the area s multiplied by an appropriate absorption coefficient a; thus,
a = a1S1 + a2S2 + a3S3 + · · · .
A more exact appraisal of the absorbing power of a room must take account
of the manner in which its various absorbing surfaces are distributed.
The absorption coefficients of materials are expressed as fractions of the
perfect "absorption" by an open window of equal area. A felt surface is a
good absorber because it contains many tiny channels in which the air vibra-
TABLE II. SOUND ABSORPTION DATA

Absorption Absorbing
Coefficient Power, a
Material a Items sq ft Each

Open window ............... 1.00 Auditorium chairs, wood 0.25

Acousti-Celotex ............. 0.80 Auditorium chairs, upholstered 2.8
Brick wall, unpainted ........ 0.03 Church pews, wood 0.4
Carpet, felt lined ............ 0.37 Audience, per person 4.0
Draperies, heavy ............ 0.50
Felt ....................... 0.70
Glass ...................... 0.027
Linoleum or wood flooring .... 0.03
Plaster, smooth ............. 0.03
Wood paneling .............. 0.06
-
496 WAVE MOTION & SOUND Chap. 26

tions are damped out and their energy dissipated as heat; its coefficient of
a = 0.70 indicates that 1 ft 2 of felt is equivalent in quenching reverberation
to 0.70 ft 2 of open window space. On the other hand, glass and metals are
poor absorbers. Absorption coefficients vary somewhat with the frequency
of the sound and with the condition of the surface. Some values suggestive
of the order of magnitude appear in Table II.
As an example of sound absorption, determine the reverberation time in an audi-
torium measuring 100 ft X 70 ft X 25 ft and having the following surface areas:
6000 ft 2 of Acousti-Celotex, 7000 ft2 of plaster, 3000 ft2 of wood flooring, 2000 ft 2 of
wood paneling, and 500 ft 2 of glass. Assume the room to have 800 seats and an
audience of 500 persons.
The absorbing power of the room, based on the foregoing data, is:
Acousti-Celotex ........... . 6000 X 0.80 = 4800
Plaster ................... . 7000 X 0.03 210
Wood flooring ............. . 3000 X 0.03 90
Wood paneling ............ . 2000 X 0.06 120
Glass .................... . 500 X 0.027 = 14
Seats (empty) ............. . 300 X 0.25 75
Audience ................. . 500 X 4.00 = 2000

Absorbing power 7309 ft 2

Then, by Eq. 25.11, the reverberation time is
100 X 70 X 25
t = 0.05 7309 = 1.20 sec,

which is an appropriate value, according to Table I.

26-18. Acoustics of the Auditorium. The suitability of an auditorium for
public speaking or for musical presentations depends primarily upon two
factors, reverberation and interference. Reverberation is responsible for the
slowness with which sounds die away, and may be controlled by selecting wall
materials and coverings which absorb sound energy to the desired extent, as
described in the preceding section. •
Interference arises from the reflection of sound waves by the walls and
ceiling, as a result of which the waves tend to reenforce one another at certain
regions and to destroy one another at other places. Sabine in some of his
investigations discovered regions within an auditorium where the loudness of
sound was greater than if near the source, and found these regions to be
separated by zones of almost complete silence. In particular, large curved
surfaces tend to focus the sound at definite spots. If such a surface is to be
used in the architectural design of a room, it should be shaped so that the
point of concentration will lie outside the room; this would avoid the effects
observed in so-called ,vhispering galleries.
The reflection of sound waves from the walls of a room is pictured in Fig. 20.
'The dimensions of the room are assumed to be 40 X 24 ft, and the wave fronts
indicated show their positions 0.02 sec after they leave source S.
§ 26-19 ULTRASONIC VIBRATIONS 497

The late Professor Arthur L. Foley of Indiana University perfected a

method for showing the reflected wave fronts in a model of the room to be
tested. The sound disturbance is produced by an electric spark at a chosen
point and spreads outward as a single wave pulse. The advancing compres-
sion is illuminated for an instant by a second spark after any desired time
interval, and the shadow cast on a photographic plate shows the position then
occupied by the wave front.

Frn. 26-20. Sound reflection within a room. The dotted lines are drawn perpendicular to
the wave fronts and indicate from which walls the reflected waves come.

26-19. Ultrasonic Vibrations. Longitudinal vibrations beyond the upper

limit of hearing have been produced and applied; this range, called the ultra-
sonic frequency region, extends at present up to 5 X 108 vib per sec. Such
vibrations can be produced by a quartz crystal mounted between metal elec-
trodes, making use of the piezoelectric effect described in § 24-4. When an
alternating emf of appropriate frequency is impressed upon the electrodes,
the crystal is set into mechanical vibration, which it communicates to the
surrounding medium. Ultrasonic vibrations can also be produced by a mag-
netostrictive rod when placed within a solenoid that is energized by a high-
frequency alternating current, § 19-3.
Because of their relatively short wavelength, ultrasonic waves do not spread
much by diffraction, § 32-6, but remain localized in narrow beams; this
ability to concentrate power is utilized in the hydrophone for under-water com-
munication between ships. The transmitting crystal, supported outside of
the ship's hull, sets up a longitudinal wave in the water. This wave acts
upon a similar crystal at the receiving station, setting it into vibration and
causing it to generate an emf of the same frequency between its electrodes.
The water wave corresponds to the carrier wave in radio telephony, and the
same methods of modulation and reception are used for both, §§ 24-3 and 4.
Ultrasonic waves, as well as those of somewhat greater wavelength, are also
used for echo ranging and depth sounding. Instruments for these purposes
are designed to transmit a signal and receive its echo, to repeat these opera-
tions many times per second, and to give visible indications of the distances
498 WA VE M01'10N <~ 80UND Chap. 26

sought. A method known as "Sonar" was used in World \Var II to locate

enemy submarines; it utilized crystals of ammonium dihydrogen phosphate
and operated on frequencies ranging from 10 to 30 kilocycles per second. The
ultrasonic waves were sent out from the ship and the reflected waves from the
submarine revealed its direction and range.
A depth-sounding instrument called a Fathometer sends a signal to the bottom and
receives the echo 1/80 sec later. Assume the speed of transmission in sea water to be
1500 m/sec, and determine the depth at the place of observation.
The depth is ! X 1500/80 = 9.38 m. Since 1 fathom = 6 ft = 6 X 0.305 = 1.83 m,
the result is 9.38/1.83 = ,5.13 fathoms.
The molecules of a medium transmitting ultrasonic waves are caused to
move with great rapidity between condensation and rarefaction, and hence
their accelerations are exceedingly large. Vibrations of moderate power at,
say, one megacycle per second can easily develop accelerations a million times
the acceleration due to gravity. It follows that suspended particles are sub-
jected to high-speed bombardment; they tend to break apart and form col-
loidal solutions or emulsions. Another effect produced by ultrasonic waves in
liquids is called cavitation-a rushing of molecules toward condensations so
rapidly that holes are virtually left in the liquid. The forces exerted in this
process are sufficient to erode metals and to produce such biological actions
as the destruction of bacteria and insects. Ultrasonic waves of high power
can also produce the opposite effect-coagulation, and this property is now
being applied in smoke precipitation.
The passage of ultrasonic waves through a medium sets up planes of con-
densation and rarefaction which may be so close together that the medium acts
as a diffraction grating, § 32-9, causing a beam of light perpendicular to the
direction of wave propagation to be spread into spectra of several orders.
This phenomenon has been used in the measurement of the moduli of elasticity
of liquids and solids. Ultrasonic waves are also used in the nondestructive
testing of castings, forgings, and concrete structures to locate hidden flaws.
26-20. Flight above Sonic Speeds. The speed of sound has a bearing on
the flight of airplanes, projectiles, and rockets. When any one of these is in
motion, the air disturbances produced move away from it with a speed equal
to the speed of sound. In air at standard pressure and 0°C, this sonic speed
is 1087 ft/sec, § 25-8, or almost 750 mi/hr; in the stratosphere it is about
10 per cent less. If the body is moving at a speed less than the sonic speed
in the undisturbed air, it will not overtake the disturbances that it produces.
Therefore, for steady flight conditions at such speeds the air undergoes only
gradual changes in speed and pressure as a body moves through it.
When the body travels at speeds greater than the local sonic speed in the
undisturbed air, the disturbances cannot move ahead of the body. Conse-
quently, the air in front of the body is unaffected until it encounters a shock
wave, across which the speed and fluid properties vary abruptly. This shock
wave originates either on or close to the front of the body, depending upon the
§ 26-20 FLIGHT ABOVE SONIC SPEEDS 499

shape of the nose of the body. Figure 21 shows the shock waves formed in
air by a blunt body moving faster than the speed of sound. Across such a
shock wave, part of the kinetic energy
of the body is transformed into irre-
coverable heat and, therefore, the
entropy of the affected air increases,
§ 14-8. The result is an increase in
the drag on the body. This so-called
wave drag adds to the drag of the
body caused by the small but finite
viscosity of the air. It can be shown
that the drag increases approximately
as the square of the ratio of the speed
of the body to the local speed of sound
in the undisturbed air, a ratio known
as the Mach Number.
In addition to the shock wave in the
neighborhood of the nose of the mov- Frn. 26-21. Shock waves formed by mov-
ing projectile. (Photograph by Ballistic Re-
ing body, other shock waves can search Laboratories).
occur in regions where the local air
speed is high. For example, shock waves are found in the flow near a body
which is moving at a speed less than that of sound if the surface curvature
of the body causes the local air speed to be greater than sonic speed.
An interesting phenomenon is observed when shock waves become detached
from a moving body as it maneuvers or accelerates and then reach the earth's
surface. The energy of these detached waves can produce a "thunder clap"
sufficiently strong to break windows. To avoid such damage it is necessary
to restrict the flight and maneuvering of supersonic aircraft to high altitudes.

PROBLElvlS
1. A siren disk has 36 holes; at what speed should it be driven in order to produce
a tone frequency of 528 vib/sec?
2. Two point sources of sound, A and B, are adjusted so that the frequency of A
is three times that of B. \Vhen spaced 20 ft apart, the sources produce equal sound
intensities at a point midway betv.reen them. How far from source A, on a straight
line joining the sources, will the intensity due to source A be twice as great as that due
to source B?
3. Two sounds have intensity levels of 40 and 50 db above a reference intensity
of 10-10 microwatt/cm2 • If these sounds are produced together, what is (a) the
intensity of the resulting sound? and (b) the intensity level of that sound?
4. Compute the change in intensity level for a sound that is (a) doubled in in-
tensity, and (b) reduced tot of its original intensity.
5. A listener shifts his position from one place outdoors to another three times as
far away from a steadily sounding source. (a) How does the sound intensity at the
new position compare with that at the original position? (b) What change in inten-
sity level is caused by moving to the new position?
500 WAVE MOTION & SOUND Chap. 26

6. A public address system raises the intensity level of a speaker's voice 20 decibels.
(a) If the speaker's voice has a power of 10 microwatts, what is the power of the sound
produced by this device? (b) If a listener could just hear the speaker at a distance
of 15 ft without the address system, at what distance could he just hear the speaker
when the device is used?
7. What are the frequencies of the fundamental and the second harmonic of a
copper string 50 cm long? Take the speed of transverse waves in the string to be
180 m/sec.
8. Compute the frequencies of the lowest and the next two higher pitched tones
that can be produced by the vibration of a stretched string 2 ft long. Take the speed
of sound in the string to be 840 ft/sec.
9. The highest tone on a piano is produced by a steel string vibrating with a fre-
quency of 4186 vib/sec. If the manufacturer specifies this string to have a length of
2.1 in. and to be stretched with a force of 160 lb, what should be its diameter? Take
7.84 as the specific gravity of steel.
10. A slender rod 3 ft long is clamped at the center and set into longitudinal vibra-
tion. Compute the wavelengths corresponding to the lowest and next two higher
pitched tones that the rod can give forth.
11. An aluminum rod 2 ft long is clamped at one end and is set into longitudinal
vibration by stroking with rosined chamois. Compute the frequency of the lowest
and next higher pitched tones that can be produced by the rod when so clamped.
12. An organ pipe closed at one end is to produce a tone of 352 vib/sec. How long
should it be? Neglect the end correction.
13. Two organ pipes, each closed at one end, are set into vibration in their funda-
mental modes. When the pipes are sounded together, they produce 45 beats per
minute. If the shorter pipe is 24 in. long, what is the length of the other? Neglect
the end correction.
14. What length of air column in a tube, open at both ends, would vibrate in reso-
nance with the rod of Prob. 11? Suppose each of the sounding bodies to vibrate in its
fundamental mode.
15. At what rate should impulses be applied to the air column within a chimney
100 ft tall in order to set it vibrating in its fundamental mode? Assume the chimney
to be closed at the bottom.
16. The speed of sound is measured by the method illustrated in Fig. 11, using a
glass tube 30 in. long and an electrically operated tuning fork which vibrates 440 times
per sec. As the water is lowered in the tube, two points of resonance are observed,
the heights of the water column being 23.1 in. and 8.1 in. Compute the speed of
sound in air from these data.
17. A tuning fork which has a frequency of 396 vib/sec is mounted on a wooden box
to reenforce its sound. How long should the sounding box be if it is open at one end?
18. Compute the frequencies of the tones on the minor scale for the octave beginning
with C = 264 vib/sec.
19. In measuring the speed of sound by Kundt's method, a nickel rod 80 cm long
clamped at the center produces the same pitch as that of a siren driven at 1866 rev /min,
the siren disk having 100 holes. The powder heaps within the tube are 5.3 cm apart.
Compute the speed of sound in the nickel rod and also in the gas inside the tube.
20. A train is traveling at 60 mi/hr and its whistle is sounding steadily with a fre-
quency of 400 vib/sec. Find the apparent frequency of the tone heard by a man
standing near the track (a) as the train approaches him, and (b) after the train has
passed and is receding from him.
21. A police car is sounding its siren at 500 vib/sec while overtaking a speeding
motorist. If the speedometer in the police car indicates 80 mi/hr and that in the
other car 65 mi/hr, what is the apparent frequency of the siren to the motorist?
 PROBLEMS 501

22. A freight train moving forward at 30 ft/sec is sounding its whistle continuously
with a frequency of 300 vib/sec. Compute the apparent frequency of the tone heard
(a) by a stationary listener standing behind the train, (b) by a person walking forward
beside the train at 4 ft/sec, and (c) by a brakeman running forward on top of the train
at 8 ft/sec.
23. Find the reverberation time for a room in which a sound decays by absorption
to ! of its initial intensity in 0.06 sec.
24. If the reverberation time for an empty hall is 2.0 sec, in what time will the
intensity of a sound in this room diminish by absorption to ro of its initial value?
25. In a room for which the reverberation time is 1.2 sec, find the time interval in
which the intensity level of the decaying sound is lowered by (a) 20 db, and (b) 40 db.
26. An auditorium has a floor area 40 X 60 ft and is 18 ft high. It has a linoleum
floor and a plaster ceiling. The walls have a wood wainscot 4 ft high, above which
the windows have an area of 300 ft2; the remaining surface is covered with acoustic
material having an absorption coefficient of 0.60. Compute the reverberation time
for the empty room.
27. A guided missile "whistles" through the air and gives rise to a tone having a
frequency of 600 vib/sec. If the missile approaches its target at a speed for which the
Mach Number is 0.8, what will be the apparent frequency of the sound caused by
the missile as observed by people near the target?
 OPTICS

27
LIGHT AND VISION
Light is sometimes defined as the aspect of radiant energy that affects an
observer through the visual sensation resulting from the stimulation of the
eye. From this point of view, light is radiant energy that is evaluated visu-
ally and hence involves certain phases of Physiology and Psychology as well
as Physics. The visible radiations are bordered on both sides by invisible
ones, and the study of light includes them also; this broader scope is indicated
by the more inclusive term Optics. Therefore, Optics treats not only visible
light of all wavelengths, but also the shorter-wavelength radiations of the
ultraviolet and x-ray regions and the longer-wavelength infrared rays which
extend to the microwave region of radio.
The present chapter deals with some of the broader aspects of light, with
light sources and their evaluation, with receptors of light, especially the
human eye, and ,vith the production and measurement of illumination.
27-1. The Nature of Light. Aside from wavelength differences, the field of
Optics can be divided into three branches; in the order of their acceptance
and utilization these are Geometrical Optics, Physical Optics, and Quantum
Optics. The distinction among them js based on the concepts used in their
development: Light in geometrical optics is usually regarded as made up of
rays, in physical optics as waves, and in quantum optics as a combination of
waves and corpuscles known as photons.
Geometrical optics is the branch ,vhich gives a satisfactory description
of the reflection of light rays within a medium and the refraction of such rays
as they pass from one medium into another. The concept of rays dates back
to the early theories of light, in which streams of corpuscles were thought to
originate in sources of light, to travel outward along straight lines and, upon
striking the eye, to stimulate the sense of sight. The design of optical instru-
ments falls almost completely within the compass of geometrical optics.
Physical optics treats light as electromagnetic waves; the wave concept
gives the most straightforward explanation of the optical effects known as
503
504 OPTICS Chap. 27

interference, diffraction, and polarization, to be considered later. This treat-

ment fits well with the electromagnetic theory and the growing accumulation
of experimental data in electricity and electronics.
Quantum optics originated when the quantum theory of radiation was ap-
plied to explain an interaction of light with matter known as the photoelectric
effect, and also to explain the regularities in the wavelengths of the radiations
or "spectrum lines" given off by atoms. At first this theory seemed incom-
patible with the idea of wave motion, and it became the custom to employ
whichever one of the three concepts--rays, vmves, or quanta-fitted best the
particular task at hand and simply to overlook the others. This apparent in-
consistency has been resolved on theoretical grounds, and it is now appropriate
to consider only the wave nature of light when it is traveling in free space
and its quantum or corpuscular nature when it interacts with matter. These
concepts are to be regarded as different aspects of a basic, if complicated,
underlying truth.
27-2. Sources of Light. All sources of light radiate energy as a result of
the excitation of atoms and molecules of matter by thermal, electrical, and
sometimes chemical means. Whatever the means, the radiation from any
source usually extends beyond the wavelength range that the eye can per-
ceive. It follows that the greater the relative amount of energy that does
fall within the visible region, the greater, in general, is the so-called luminous
efficiency of the source.
Our greatest primary source of light is, of course, the sun, a thermal source
with a surface temperature of about 6000°K. Our eyes are most sensitive
for wavelengths that lie at or near the energy peak of the sun's emission curve;
hence the sun is regarded as an efficient radiator-in fact, it is more efficient
than any of the artificial thermal sources because of its higher surface temper-
ature. However, since only a fraction of its total radiation lies within the
visible range, the sun is not a perfectly efficient source of visible light.
At night the moon may be an important source of light. But since it
merely reflects the sun's light it is really a secondary source. As a matter of
fact, the way we see the moon is actually the way we see most objects, that is,
by reflected light. The reflectors, or secondary sources, are such things as
the sky, trees, buildings, and the walls and ceilings of rooms.
Artificial sources of light, in the order of their development through the
centuries, include the torch, the oil lamp, the candle, the gas lamp, the carbon
arc lamp, the incandescent electric lamp, vapor and gas tubes, and the fluo-
rescent lamp. Little need be said about the several flame sources; these
thermal sources are very inefficient because they operate at relatively low
temperatures and hence radiate only weakly, and mostly in the infrared
region.
The carbon arc is the most efficient artificial thermal source of light because
of its high operating temperature; it is commonly used in motion-picture pro-
jectors and searchlights. Ordinarily it consists of a pair of carbon rods con-
§ 27-2 SOURCES OF LIGHT 505

nected in series with a resistor across direct-current service mains. The arc
is started by bringing the carbons into contact and then separating them.
As the carbons burn away during operation, they are fed toward each other
so as to keep the gap between them fairly constant in length. Most of the
light comes from a crater that is formed in the tip of the positive carbon; its
temperature is about 3600°K.
The most convenient type of artificial light source is the incandescent
electric lamp. This consists essentially of a conducting filament enclosed
within a glass bulb and heated so intensely by an electric current that it emits
light. The original lamp of this type, brought out by Edison in 1880, used a
filament of carbonized bamboo mounted within an evacuated bulb to prevent
oxidation. About the year 1910, when it became possible to draw tungsten
into fine wires, lamp filaments were made of this metal and were operated in a
vacuum at temperatures of around 2500°K. The gas-filled lamps of the
present day operate at temperatures up to 3400°K, which is close to the melt-
ing point of tungsten, 3655°K. Incandescent lamps are used for the illumi-
nation of all sorts of interiors and for much highway lighting; they generally
operate across 110-volt service mains.
The process by which light is produced in the sources considered thus far is
thermal. The high temperatures are attained through nuclear reactions in
the sun, through combustion processes in the flame sources, and by electrical
means in the arc lamp and the incandescent lamp. Other electrical methods
are now used to produce light, and usually result in higher luminous efficiencies
than are obtained with purely thermal excitation. A vapor or a gas at low
pressure is sealed within a glass enclosure provided with electrodes between
which an electric field is established. Electrons and ions accelerated by the
field bombard the gas molecules and cause the gas to emit strong radiation
in the visible region. Practical lamps include the mercury vapor lamp, the
sodium vapor lamp, the luminous tubes so common in "neon signs," and
fluorescent tube lamps. 1\/Iany of these modern types of lamps are charac-
terized by large area and lower surface brightness rather than by the concen-
trated, almost point-source nature of the carbon arc and the tightly coiled
tungsten filament.
The early mercury-vapor lamps, developed by the American inventor Peter
Cooper Hewitt, were long tubes filled with mercury vapor at low pressure.
The present type of lamp contains mercury vapor plus a little argon gas and
operates at pressures of a few atmospheres; its construction is shown in Fig. 1.
The inner or arc tube contains the two main electrodes, each consisting of a
coiled tungsten wire coated with barium oxide; it also contains a starting
electrode connected with a resistor to control the starting current. ·when the
lamp is connected to an alternating-current supply, electrons are drawn from
the upper main electrode by the starting electrode whenever the latter is
positive. This stream of electrons ionizes the argon, and a blue glow fills
the tube. The discharge serves to heat the mercury vapor to the point of
506 UPTICS Chap. 27

ionization, and then conduction is established between the main electrodes.

Such mercury-vapor lamps give off light of a bluish-green color, but are
satisfactory for street and industrial lighting.
The sodium-vapor lamp, now used for highway lighting, has two glass en-
closures, the outer one evacuated. The inner one has two electrodes and con-
tains some metallic sodium and a little neon gas. When the lamp is started,
the discharge passes through the neon
and produces the dull red glow that can
Leads be observed during the first few min-
utes of operation. The heat developed
by the neon discharge evaporates the
Resistor
sodium, and the lamp soon produces
Supporting the characteristic yellow light of sodium
band vapor.
Starting Luminous tubes containing gas at low
electrode
pressures are widely utilized for display
Main lighting. Neon is used for orange-red
elect.rodes HE---1tt-1--=- Lead and
light, argon and mercury for blue light,
support
wires and helium for pinkish-white light.
These long-tube units are operated
from 110-volt alternating-current service
-+--+-- Inner tube
through step-up transformers that pro-
duce the high potentials needed to ionize
Supporting --T"'-n--4----..J the gas.
band
Outer tube The fluorescent lamp consists of a
long glass tube with a thin layer of
fluorescent material, called a phosphor,
Frn. 27-1. Modern mercury vapor on its inner wall; it contains argon or
lamp krypton gas and mercury vapor, at
low pressure. An electric discharge
through the gas is initiated between the electrodes, and the resulting ultra-
violet radiation energizes the phosphor, ·which then becomes a brilliant source
of visible light. The color produced depends upon the phosphor used: calcium
tungstate for blue light, magnesium tungstate for blue-white, zinc silicate for
green, and cadmium borate for pink. In starting the lamp, current is estab-
lished in a small coil of tungsten wire at each end of the tube and, when these
coils have become heated, the connection is automatically broken between the
two and they then serve as electrodes for the gaseous discharge. The lamp
with its auxiliary equipment is designed for connection to 110-volt alternating-
current circuits, and operates at a relatively low temperature.
27-3. The Transmission of ·Light. To a close approximation, light travels
through a homogeneous medium in straight lines. Such rectilinear trans-
mission of light is recognized, at least subconsciously, by almost everyone, in
sighting to decide whether objects are "lined up," whether edges are straight,
§ ......
~/-1;)
') 1'/lE TRA.NSllHSSION OF LIGJI1' 507

and whether surfaces are flat. People are so accustomed to thinking that
objects are where they appear to be that it is hard to believe that this is not
always true, and they can easily be deceived by tricks played with mirrors.
The radiation from a point source of light is often indicated by straight
lines called rays diverging from this spot. When such divergent rays enter
the eye, they are rendered converging by its lens system and brought to a focus
upon the retina, where they stimulate the nerve endings and produce the
sensation of sight-that is, of "seeing" the spot from which the rays appear
to diverge. A narrow cone of rays diverging from a common point is called a
pencil of light, and a group of nearly parallel rays is called a beam of light.
Rays indicate the direction of transfer of energy by the light beam; ordinarily
they are perpendicular to the wave fronts, Fig. 25-10.

Extended source

Umbra
11
I
I
I
I
I
I
I

J
I
I

ID ~luminatlon t Illumination
I
I
Umbra ' ' \ Umbra' f'
· Penumbra
Frn. 27-2. Part I, sharp shadow from point source; II, diffuse shadow from extended
source. Parts III and IV, representation of screen illumination across the shadow regions

The fact that shadows have the same shape as the objects that cast them is
commonly used as a proof that light travels in straight lines. Observation
shows that a shadow cast when the light source is small is much sharper than
when the light source is large. For instance, the shadow of a telephone pole
produced by an unshaded street lamp is quite sharp, whereas that produced by
sunlight has fuzzy edges because the sun is relatively large, even as seen from
the earth. A fluorescent lamp would be considered a large or extended source
in a small room, but would be regarded as a point source if far away.
Figure 2 illustrates the difference in the shadows cast by point and extended
sources. With the latter there is an area of partial shadow, called the pe-
numbra, surrounding the central area of complete shadow, called the umbra.
No light from any part of the source reaches the umbra, but some does reach
the penumbra, in an amount that varies from a minimum at the edge of the
umbra to a maximum at the outer edge of the shadow pattern. The illumi-
.508 OP1'1CS Chap. 27

nation of the screen is plotted in the lower parts of the figure, for both types
of source.
Solar and lunar eclipses provide large-scale examples of the formation of
shadows and hence give another proof of the rectilinear propagation of light.
A solar eclipse occurs when the moon comes between the sun and the earth;
the eclipse is complete when the observer is located within the umbra or com-
plete shadow of the moon as cast by the sun, but is partial when the observer
is within the penumbra, for then the sun's disk is only partially blocked off
from view by the moon. A lunar eclipse occurs when the moon passes
through the shadow of the earth-that is, when the earth comes between the
sun and the moon.
27-4. The Speed of Light. A natural question to ask about any process
that transfers energy through space is its speed of action. If people in ancient
times asked themselves this question about light, they probably answered it
by saying that light moved with infinite speed. In fact, the earliest recorded

Jupiter /0
I
I
/ Earth ~ Sun
One of
Jupiter's
Moons
\ --fp---c 1
I
BIB'
~

Fm. 27-3. Roemer's astronomical method for measuring the speed of light

attempt to measure the speed of light, made by Galileo, failed because of the
enormous value, although the method was based on valid physical prin-
ciples. The experiment was carried out at night-time, between two observers
a few hundred yards apart. The first observer uncovered a lantern and, when
the second one saw its light, he immediately uncovered his own lantern; this
procedure enabled the first observer to measure the elapsed time for the light
to make the round trip. What he actually measured was principally the
sum of the "reaction times" of the two observers, which total was much
larger than the transit time of the light itself. Apparently Galileo understood
this and abandoned further attempts to measure the speed of light.
The first successful measurement of the speed of light was made by the
Danish astronomer Ole Roemer (1644-1710), from observations on the revolu-
tion of one of the moons of the planet Jupiter. This method is illustrated in
Fig. 3, which represents Jupiter and the earth in their orbits about the sun.
Roemer found that the time interval between two eclipses of the satellite
(about 40 hr) was the same when measured with the earth at A or at C, but
that the interval was lengthened when the earth was at Band shortened when
§ 27-4 THE SPEED OF LIGHT 509

at D. He judged correctly that these variations were not due to irregular

motion of the satellite but rather to the finite speed of light. Thus, if the
earth were at B at the beginning of a measurement, it would move away to
some position B' during the 40-hr interval between eclipses, and the light
would have to travel the additional distance BB' to mark the conclusion of
the test, thereby increasing the elapsed time.
From measurements made with the earth at A, Roemer predicted the time
at which the moon would emerge from behind Jupiter for a particular transit
when the earth had moved to C (about 6 months later) on the supposition
that the speed of light is infinite; he found that the transit took place several
minutes later than predicted. He recognized that the difference between the
observed and the calculated times represented the time required by the light
to traverse the diameter of the earth's orbit. From more recent measure-
ments, this time difference is approximately 1000 sec and, since the diameter
of the earth's orbit is 186,000,000 mi, the speed of light is about 186,000 mi
per sec.
In view of the tremendous speed of light, its determination must require
the mea~urement either of very long distances or of very short time intervals.
The former apply to the early astronomical
methods, and the latter to the more recent S
terrestrial methods. The first successful terres-
trial determination was made by the French
scientist Augustin H. L. Fizeau (1819-1896), :
using a toothed wheel as a timing device. Ad- ,- ~
1
ditional measurements of the same type were
M
made by him with his countryman Jean B. L.
T
Foucault (1819-1868), using rotating mirrors.
This technique was repeatedly refined through 1-+-----d----
the long-term efforts of the American physicist Frn. 27-4. Michelson's terres-
Albert A. Michelson (1852-1931). The essen- trial method for measuring the
tial parts of his apparatus are indicated in Fig. speed of light
4. Light from an intense source S strikes one
face of an octagonal mirror R and is reflected to a distant plane mirror M; the
returning ray is reflected from another face of the octagonal mirror into the
observer's telescope T. vVhen mirror R is set into rotation in a clockwise
direction, flashes of light will strike the distant mirror and the reflected beam
will fall to the left of the telescope. If the octagonal mirror is turned faster,
a speed will be reached such that in the time required for the light to travel
to mirror M and back again, mirror face 2 will have advanced to the posi-
tion shown at 3. At this speed the light will be reflected into the telescope
again for observation. If n represents the rotational speed of the mirror in
revolutions per second, the time taken by it to turn through one eighth of a
revolution is i of (1/n) sec. In this time the light travels to mirror Jl{ and
back, a distance 2 d; consequently the speed of light is given by
510 OPTICS Chap. 27

v = 2 d / 81n = 16 nd.

In Michelson's experiment, the revolving mirror was located at the astro-

nomical observatory on Mt. Wilson, California, and the stationary mirror was
installed on Mt. San Antonio; the distance of about 22 mi between the two
points was measured with great precision by the U. S. Coast and Geodetic
Survey. The angular velocity of the revolving mirror was determined by
comparison with an electrically operated tuning fork; the fork in turn was
calibrated against a free pendulum, and this was standardized by an astro-
nomical clock. Michelson subsequently made a long series of measurements
of the speed of light within an evacuated tube one mile long and obtained
comparable results. From both direct and indirect methods, the current
"best value" for the speed of light in a vacuum is 299,790 ± 1 km per sec.
The symbol c is ordinarily used to represent the speed of light; in most calcu-
lations, the speed is taken as c = 300,000 km per sec, or 186,000 mi per sec.
So great is this speed that light emitted by the sun reaches the earth in only
8.3 min, whereas a jet airplane flying at the speed of sound would require over
14 years to travel the same distance. Even with this great speed, the light
from the stars takes years to reach the earth; for the nearest of them, a in the
Centauri group, the time is about 3f years. The distance to this star is con-
veniently expressed as 3.5 light-years; the Great Nebula in Andromeda is more
than 800, 000 light-years distant.
The speed of light in the earth's atmosphere is only slightly less than in
a vacuum; the difference, for standard atmospheric conditions, amounts to
about 90 km per sec, or 0.03 of 1 per cent. Therefore, air and vacuum values
of the speed of light can ordinarily be used interchangeably. In other medi-
ums the speed of light is considerably less than it is in a vacuum; moreover,
it is different for different wavelengths-blue light travels slower than red
light-a fact that gives rise to the phenomenon of dispersion, § 28-14.
The speed of light is one of the cornerstones in the theory of relativity and
in electromagnetic theory. The speed of light can in fact be computed from
purely electromagnetic data, as is done in § 25-9, and the value obtained is in
good agreement with the experimental values.
27-5. The Eye as a Receptor of Light. Vision is the most highly valued of
human senses, and the eye is a marvelously versatile and adaptable sense
organ. Its description as an optical instrument is given in § 30-1, but
some of its more general properties as a receptor of light are considered here.
The eye may be thought of as a readily adjustable, rapid-acting color camera
that functions over an exceedingly wide illumination range. The surface
that corresponds to the film in a camera is the retina, located at the back of
the eyeball, Fig. 30-1. The light-sensitive receptors of the retina are of two
types: the cones, which are color sensitive, and the rods, which are only black-
and-white sensitive. In viewing an object, the eye automatically rotates in
its socket to foC'us the object on the fovea-the area of most acute vision.
§ 27-;j 'PlIE EYE AS A. RECEP'l'OR ()Ji' LIGHT 511

Here only cones are present, but outward from this area there are more and
more rods in proportion to cones.
The relative sensitivity of the eye to various colors of light has been the
subject of much investigation. In the experiments pure colors are used,
and these are identified by wavelength. The familiar color band of the rain-
bow is an example of a spectrum, in which the colors are arranged in order by

I I I I

Ultraviolet - - - - - V : t1
:.1s1'bl e spec l
rum - - - - - Infrared
I I I I
I I I I

300 350 400 450 500 550 600 650 700 750 800
Wavelength, millimicrons

Fw. 27-.5. The wavelength bands associated with the spectral colors

wavelength as in Fig. 5, wherein the wavelengths are indicated in millimicrons

(mµ). Actually, the colors blend gradually into their neighbors, and the
wavelength boundaries between them are not sharp.
It has been found that under bright illumination, the eye is most sensitive
for yellow-green light of wavelength 555 mµ, and its sensitivity falls off rap-
idly both for longer and for shorter wavelengths. The relative sensitivity
of the average eye under bright illumi-
nation, representing so-called photopic
vision, is shown in Fig. 6. A relative
sensitivity curve obtained under sub-
0
t
0.8 t--i,-----,----,.--+-+-+---+------1
dued lighting, and representing so-called c
. ·E o.6
scoptic vision, differs from t he curve of ::,
this figure in two respects. First, the ~ 0.4 l-----'f---i1J----+--+--+--+------l
·.;:
eye, having become at least partially a
"dark-adapted," exhibits a much greater Ji 0.2 I - - - - + - - + + - - - + - - - + - - - > - + - - - - <
sensitivity than in bright light. Second, 0.....-=-.,._ _.___-'---'--'--~
the wavelength of maximum sensitivity 400 500 600 700
shifts toward the blue end of the spec- Wavelength, millimicrons

trum to about 505 mµ; this phenomenon Fm. 27-6. Relative sensitivity of the
is called the Purkinje effect after its eye for the spectral colors, in photopic
discoverer, the Czech physiologist Jan vision
E. Purkinje (1787-1869).
When the eyes are well dark-adapted, more sensitive vision is afforded by
the rods than by the cones. However, in rod vision the various colors are
not discerned, but only different shades of gray. In order to see lights as
colored, the cones must come into action, and higher brightnesses are needed;
it is found that, from the violet to about the orange of the spectrum, approxi-
512 OPTICS Chap. 27

mately a hundred times the brightness is needed for cone vision as for rod
v1s10n. The rod and cone sensitivities become equal at a wavelength of about
580 mµ; at the red end of the spectrum the situation is reversed, and the cones
are more sensitive than the rods. Thus red lights have much lower rod than
cone visibility; this fact was applied widely during w· orld War II in the use
of red lighting aboard warships that were "blacked out" at night to avoid
enemy detection and to preserve the dark adaptation of the ship's crew.
Advantage was also taken during the war of the relative "red-blindness" of
the rods through the extensive use of red dark-adaptor goggles; these transmit
only deep orange and red light of wavelengths longer than 600 mµ. If look-
outs or others requiring the keenest
100 night vision wore these goggles for about
+-
c l.,..---
~ 80 half an hour before going on duty, it was

S 60 I I
not necessary for them to pre-adapt their
eyes by spending that period in complete
c
darkness; instead they could stay in
~ 40
·e lighted compartments, move about freely,
~ 20 and even read. Goggles of this type are
e
I- .)
)
also useful for persons who must pass
0
400 500 600 700 back and forth frequently between lighted
Wavelength,millimicrons offices and photographic dark rooms.
Frn. 27-7. Transmission curve of red Figure 7 shows the transmission curve
dark-adaptor goggles of the dark-adaptor goggle, with its sharp
"cut-off" in the orange-red. The location
of the cut-off represents a compromise between transmitting enough light
to permit photopic vision for reading, and excluding the light of short wave-
length that retards dark adaptation.
Another important property of the eye is the rapidity of perception upon
exposure to light and the persistence of vision for a short time after its re-
moval; this property is utilized in viewing motion pictures. In the ordinary
motion-picture projector, each "frame" or individual picture is flashed briefly
on the screen and the beam of light is then cut off momentarily by an opaque
shutter while the film is jerked ahead to the next frame. If the interruption
rate is high enough, the sensation of flicker disappears and the impression of
continuous screen illumination is produced. An interruption rate upwards of
35 per second is necessary, but the critical value varies from person to person
and depends also on the brightness of the screen. Modern motion-picture
projectors usually give 48 flashes per second; the film is advanced during
alternate dark periods.
Still another property of the eye is its sensitiveness in detecting the motion
of objects in the field of view-this ability is pronounced even in the peripheral
portions of the retina, where visual acuity is low. A basketball player, for
example, can follow the movements of the other players without turning his
head, by looking "out of the corners of his eyes."
§ 27-6 RADIANT AND LUMINOUS FLUX 513

27-6. Radiant and Luminous Flux. The importance of the human eye as
the primary receptor of light is recognized in illumination engineering by the
reliance placed upon the sensation of brightness evoked by light. The con-
cept of brightness can be interpreted in terms of the relative lightness of a
series of gray areas uniformly graded from a pure white to a jet black which
serves as a "gray scale"; moreover, colored samples can be matched for
brightness quite accurately with one of the gray areas of the scale. The
terms brightness and lightness apply to the subjective concept of an ob-
server-"how bright it looks to him," although they are of course related to
the physical intensity of the light waves that reach the eye from the object.
Objective measurements in optics are made with instruments that do not
require the judgment of the observer; for example, radiation is measured by
the optical pyrometer, § 34-1, and the result can be expressed in power units.
In contrast, subjective measurements in optics are related to the stimulation
of the eye and the sensation produced thereby; the results are expressed in
arbitrary units of light that are widely accepted. The field of the objective
measurement of radiation is called radiometry, and that of its subjective
measurement is called photometry. There is no simple relationship between
the radiometric term radiant flux, which is the time rate of energy flow in
radiation, and the photometric term photometric brightness (sometimes called
luminance), which is the luminous intensity per unit of projected area of an
extended light source, whether primary or secondary; such a relationship
depends largely upon the eye's response to light of different wavelengths.
Luminous flux is defined as radiant flux evaluated visually, that is, by
reference to the relative luminosity curve, Fig. 6, for the so-called "standard
observer" in photometry. The relationship between radiant and luminous
units can be established by defining the lumen as the luminous flux that cor-
responds to 1/680 = 0.00147 watt of radiant flux, assuming monochromatic
radiation of the wavelength for which the photopic eye is most sensitive,
namely 555 mµ. Conversely, a luminous flux of 680 lumens is equivalent to
one watt of radiant flux for this monochromatic radiation. The lumen is
defined in § 27-9 in terms of the "candle" as a unit of luminous intensity.
The ratio of luminous flux to radiant flux is called the luminous efficiency and
is expressed in lumens per watt:
. . luminous flux in lumens
Lummous efficiency = rad'1ant fl ux m . watt s

Light that has 680 lumens per watt has the maximum possible luminous
efficiency; light of any other composition evokes a smaller brightness sensation
per watt of radiant energy, and hence has a smaller luminous efficiency.
Blue-green light of wavelength 500 mµ has a luminosity relative to 555 mµ yellow-
green light of 0.323, from Fig. 6; therefore its luminous efficiency is 680 X 0.323 = 220
lumens per watt. The radiant flux in a 100-lumen sample of this monochromatic blue-
green light would be 100/220 = 0.454 watt, whereas only 100/680 = 0.147 watt of
555 mµ radiation would produce the same brightness sensation.
514 OPTICS Chap. 27

27-7. Efficiency of Light Sources. The term efficiency is applied to light

sources in the same sense as to machines, § 6-12, namely to indicate the ratio
of output to input. The output of a light source is reckoned in lumens of
luminous flux, as defined in the foregoing section, and the input is reckoned in
units of power supplied. Hence the efficiency of a light source is given by
. _ lumens output
Source efficiency - watts mpu
. t

The values of efficiency for actual light sources make allowance for heating
and other losses incidental to their operation, as well as for the imperfect
conversion of radiant flux to luminous flux. The candle has an efficiency of
about 0.1 lumen per watt, the acetylene flame 0.7, the carbon-filament lamp 3,
and present-day tungsten-filament lamps 10 to 20 lumens per watt and higher.
Tungsten has a slightly lower melting point than carbon, but it can be
heated to a higher temperature without excessive evaporation, especially
within bulbs containing inert gases. At its melting point of 3655°K, tungsten
would have an efficiency of 50 lumens per watt. Such a thermal source ap-
proaches the theoretical concept of a black-body radiator (§ 34-1); computa-
tions show that a thermal radiator would have its greatest efficiency (93 lu-
mens per watt) at a temperature of 6500°K, which is somewhat above that of
the sun's surface.

TABLE I. EFFICIENCIES oF LIGHT SouRCJ<JS

Lumens per Watt

Kerosene lamp (estimated) ............... . 0.3

Tungsten-filament lamp (100-watt size) .... . 15
High-intensity tungsten-filament lamp ..... . 30
"Photoflood" lamp ...................... . 36
l\1:ercury-vapor lamp, low pressure ......... . 25-40
Mercury-vapor lamp, high pressure ........ . 40-65
Sodium-vapor lamp ...................... . 50
Fluorescent lamp, "white" ................ . 18-65

A source with an energy-wavelength distribution corresponding to the sensi-

tivity curve of the photopic eye, Fig. 6, would have a luminous efficiency of
about 200 lumens per watt; this figure might be considered as the legitimate
performance target of practical light sources. The sun has less than half
of this potential efficiency, and the best tungsten-filament lamp only about
one sixth. The main reason for these "poor performances" is that much of
the energy from thermal sources lies in wavelength ranges outside the visible
spectrum. For a standard 100-watt tungsten-filament lamp, only 10 per cent
of the energy radiated is within the visible range, and the remaining 90 per
cent is in the infrared region. The newer types of light sources have higher
luminous efficiencies. Sodium and low-pressure mercury lamps have intense
§ 27-8 PHYSICAL RECEPTORS OF LIGll'I' 515

spectrum lines, § 31-2, in the visible region, and fluorescent lamps employ
phosphors to convert the ultraviolet radiant energy into luminous energy.
The higher efficiency of a 40-watt fluorescent lamp over a 40-watt tungsten-
filament lamp can readily be inferred from its much higher light output and
lower surface temperature. The efficiencies of several types of light sources
are given in Table I.
The drawback of the efficient sodium and mercury lamps is their color dis-
tortion, for no light whatever is emitted in certain color regions. Hence the
use of these sources will probably be restricted to the lighting of highways,
parking lots, and other exteriors, where color discrimination is not important.
27-8. Physical Receptors of Light. The most important physical receptor
of light is the photographic emulsion on glass plates or plastic film. It is useful
over a wider wavelength range than is the eye, since its sensitivity extends
somewhat further into the infrared region-to 1200 mµ under the best con-
ditions-and all the way through the ultraviolet spectrum to the x-ray and
gamma-ray regions (see Table II in § 25-9). Photographic emulsions also
differ markedly from the eye as to variation of sensitivity with wavelength.
Ordinary (undyed) photographic plates or films are sensitive between about
500 and 230 mµ; at the shorter wavelength limit the absorption of the pro-
tective gelatin coating over the emulsion becomes the limiting factor. Color-
sensitized (panchromatic) emulsions can be used through the yellow, orange,
and red regions, and into the near infrared. The foregoing facts refer to
"black-and-white" photographic plates or film. Color film, which repro-
duces natural colors directly, involves three separate black-and-white emul-
sions plus appropriate color filters. The photographic plate also differs from
the eye in that the former can integrate an exposure over a period of time,
and the result, is roughly equal to the product of light intensity and exposure
time. Thus, when astronomers desire to observe stars that are much too
faint to be seen by the eye, they resort to night-long photographic exposures.
The specifications of photographic plates for scientific use should include the
granularity of the emulsion, the associated resolving power (ability to show
fine picture detail), and the rendition of contrast.
Physical receptors of light are used in present-day techniques for the deter-
mination of light intensities; they include photoelectric cells, photomultiplier
tubes, photoconductive cells, bolometers, and thermocouples or thermopiles.
There are two common types of photoelectric cells: (1) the photoemissive,
alkali metal or oxide type (sodium, potassium, or cesium), § 34-2, with which
an external source of electromotive force must be employed; and (2), the
photovoltaic or barrier-layer selenium or cuprous oxide cells, which generate
their own electromotive force. Modern cells of the photoemissive type are
sensitive to light of all wavelengths from the deep ultraviolet region to the
near infrared, say to 1200 mµ. Because of their rapid response to incident
radiation and the ease with which they can be coupled into electronic amplify-
ing circuits, such phototubes are important in optical instrumentation. While
5Hi OPTICS Chap. 27

photoemissive cells are regarded as better for linearity and dependability,

modern photovoltaic cells give quite satisfactory performance if treated prop-
erly. The ultimate in sensitivity for a photoelectric cell is the photomultiplier
tube; this is a photoemissive cell that in effect includes the first few stages
of electronic amplification in the same vacuum-tight envelope as the photo-
sensitive surface, thereby reducing initial amplifier noise and permitting cou-
pling through less sensitive amplifiers to low-impedance recorders.
The physical detectors of radiant energy described thus far are used chiefly
in the visible and ultraviolet regions of the spectrum; none of them operates
satisfactorily very far into the infrared. Sensitivity to longer wavelengths is
obtained by employing still another type of photoelectric cell, which is based
on photoconductivity, that is, the lowering of electrical resistance of certain
materials upon the absorption of radiant energy, § 23-15; the action is closely
related to that of the photovoltaic cell. The lead sulfide photoconductive cell
is sensitive to a wavelength of about 3 microns (3000 mµ), whereas lead sele-
nide and lead telluride cells, when cooled by dry ice or liquid air, are useful
to 7 microns.
Instruments called bolometers can be used as radiation detectors from the
longest infrared to the shortest ultraviolet rays. They are ordinarily em-
ployed only in the infrared, however, because they are less sensitive than
photocells in the visible and ultraviolet regions. The metal bolometer com-
prises a short, very thin strip of platinum or nickel that is blackened so as to
absorb all the electromagnetic radiation that strikes its surface. As radiation
is absorbed, the temperature of the strip rises, and its electrical resistance of a
few ohms increases slightly. The bolometer is usually connected as one arm
of a Wheatstone bridge, so as to facilitate measuring the change of resistance
of the strip. A thermistor bolometer is a flake of a heat-treated mixture of
the oxides of cobalt, nickel, and manganese, with electrical leads soldered to
its ends. The mixture has a large negative temperature coefficient of re-
sistance and, although the resistance of the unit at room temperature is very
high-a few megohms-it drops rapidly when radiation is absorbed. This
device is useful for detecting minute amounts of infrared energy; for large
objects at or near room temperature it can measure temperature changes of
one thousandth to one millionth of a centigrade degree.
Another class of infrared detectors includes the thermocouple and the multi-
junction thermopile, both of which utilize the Seebeck effect, § 23-2. In addi-
tion to the type widely used as thermometric elements for ovens and furnaces,
thermocouples can be constructed that are tiny in size and have extremely
small heat capacity, so that measurable signals are given upon the absorption
of minute amounts of radiant energy.
27-9. Luminous Intensity and Illumination. In the preceding sections a
distinction was made between the physical concept of radiant flux and the
C'orresponding photometric concept of luminous flux, the unit of luminous flux
§ 27-9 LUMINOUS INTENSITY AND ILLUMINATION 517

was defined and evaluated in terms of power, and the efficiencies of light
sources were considered; but nothing has been said as yet about the luminous
intensity of light sources themselves.
The luminous intensity of a source of light is expressed in terms of a source
chosen as a standard; originally the flame of a spermaceti candle burning at
the rate of 120 grains per hour was selected for this purpose and was regarded
as having an intensity of one "candle" when viewed in a horizontal plane.
Other light standards have been employed subsequently, first flames of speci-
fied oil lamps, then carbon-filament incandescent lamps, and later, tungsten-
filament lamps. Such so-called primary standards are in turn used to cali-
brate secondary and tertiary standards, and
To
the last-named are used as working standards photometer
for testing commercial lamps.
A new primary standard of luminous inten-
sity, developed by the National Bureau of
Standards in Washington, was adopted by the
International Committee on Weights and

~~r~i; p;~:!E;. ;t:i;:~~:~:f:;~:f:~s}; ,:--Nf\t;:.::::

shown in cross-section in Fig. 8. The plati- ·(/ - - __ ?\/
num, contained in a crucible of fused thorium >}·/ ~ -- _-_- (/\. Sight tube
oxide surrounded by heat insulation, is placed /)/: \; · - - ·......
in an alternating magnetic field and melted by :/):· - -t, - - \// Fused

:~I;cEit::t!f~f[i~f~;:i:;::;u:1;~~:::;:
oxide and containing some of this material in a
;?i;;\;;":;?:?J}t!. ~~~.m
1

finely ground state, extends into the molten Frn. 27-8. Primary standard of
photometric brightness; 60 can-
metal and serves as a black-body radiator. dles per cm2 under prescribed
The photometric brightness within this tube is conditions
taken as 60 candles per square centimeter when
the metal, in cooling slowly, reaches its solidifying temperature.
This newest standard is not really one of luminous intensity (as for a point
source), but rather of photometric brightness-that is, luminous intensity
per unit of projected area. The standard defines a "new" candle, which is
slightly smaller than the old, or international, candle that was specified in
terms of carbon-filament incandescent lamps.
The luminous intensity of a point source is conveniently expressed in terms
of the number of lumens of luminous flux radiated into unit solid angle in the
space surrounding the point source. Solid angle is defined as the ratio of
the area normal to a radius, to that radius squared; it is measured in steradians.
The total solid angle about a point is 4 7f" steradians. The lumen is defined
518 OPTICS Chap. 27

as the flux per steradian from a source of unit luminous intensity, one candle.
Therefore, the luminous intensity of a point source can be expressed as lu-
mens per steradian, which is numerically equal to candles.
If a point source of luminous intensity I candles could radiate equally in
all directions (a practical impossibility), it would radiate a total luminous
flux F = 4 1r I lumens. For an actual light source the flux is not uniform in
all directions, and the luminous intensity in a particular direction is given by
I_ dF (27.1)
- dw'
where dF represents the luminous flux crossing any section of a cone of solid
angle dw with its apex at the source, and I is the luminous intensity along the
axis of the cone.

A 2 =1 m 2
1----{ E2 =1 lumen/m
2
: =1 meter-candle
:I =l lux

Fw. 27-9. Illumination produced by a source of unit luminous intensity (one candle).
The solid angle shown is one steradian and the luminous flux within it is one lumen

When luminous flux falls upon a surface, it is illuminated. The illumina-

tion (also called illuminance) of the surface is defined as the ratio of the
incident luminous flux to the area upon which it falls. A measure of the
illumination E at any point is given by the flux dF that strikes the small
area dA surrounding the point according to
dF (27.2)
E = dA.

The relation between illumination and luminous intensity is readily seen in

Fig. 9. If the solid angle is one steradian, the area intercepted at a distance
of one foot is one square foot; at one meter it is one square meter, and so on.
Therefore, with a source having a luminous intensity of one candle (one lumen
per steradian), the illumination at a distance of one foot is one lumen per
square foot, and is called the footcandle; at a distance of one meter the illumi-
nation is one lumen per square meter, and is called the meter-candle or lux.
One footcandle is hence equal to 10.76 meter-candles. The illumination E
in Eq. 27.2 will be in footcandles or meter-candles, according to whether
the area is expressed in square feet or square met<Jrs.
§ 27-10 P flOT01vlETR l' 519

The value of the illumination of a surface produced by a light source is

determined only by the intensity I of the source and its distance r from the
surface, provided the rays of light strike the surface normally. As for all
radiation, the illumination varies inversely with the square of the distance in
accordance with the Inverse Square Law, § 25-11. Hence, the illumination
varies as l/r 2 , which can be expressed as E = kl/r 2 , where k is a propor-
tionality factor. If the luminous intensity l of the source is in candles, the
distance r in feet, and the illumination E of the surface in footcandles, the
factor k becomes unity, and therefore
l (27.3)
E = -.
r2
In metric units, E is in meter-candles if r is in meters. This equation holds
when the light flux is everywhere normal to the surface, or when the dimen-
sions of the surface are small relative to the distance between it and the
source, and the source is on the normal to the surface. In general, when
the normal to the surface makes an angle i with the incident rays, the illumi-
nation at the center of the area is:
E = I cos i. (27.4)
r2

This is the basic equation for the illumination produced on a surface by a

point source of light.
27-10. Photometry. A photometer is a device for measuring the luminous
intensity of lamps and other light sources. The eye by itself would hardly
serve as a photometer, for it could give only a rough measure of the compara-
tive intensities of two sources that are viewed at the same time. The eye is
entirely unreliable if the two sources are viewed at different times, since its
memory for brightness is prl;\ctically nil. However, the eye is able to deter-
mine rather accurately whether two neighboring surfaces are equally bright,
and on this fact is based the operating principle of the visual photometer.
In the photometer, two lamps are placed a suitable distance apart with a
screen between them, so that each side of the screen is illuminated normally
by one of the sources. The screen is moved back and forth until the bright-
ness on both sides is observed to be the same, and the distances from the
screen to the lamps are measured. If the two sides can be assumed to have
equivalent reflecting properties, the observed equality of brightness indicates
an equality of illumination. Then from Eq. 27.3 it follows that
11 l2
-=-, (27.5)
r1 2 ri
where 11 and 12 are the luminous intensities of the sources in candles, and r 1
and r2 are their respective distances from the screen. From this equation,
either intensity can be computed if the other is known.
.520 OPTICS Chap. 27

In the Bunsen "grease-spot" photometer, the screen consists of a small sheet

of paper with a waxed spot near the center; a pair of mirrors is provided so
that an observer can see both sides of the paper at the same time. The
Lummer-Brodhun photometer has an improved optical system for accomplish-
ing this purpose. Its elements, shown in Fig. 10, comprise an opaque white
screen S, an observing telescope T, reflecting prisms P and P', and a com-
pound glass cube C. The cube is made up of two prisms cemented together,
one having a design etched in its principal face to give the field of view the
appearance shown. The paths of light along which both sides of the screen
are observed in the telescope are also indicated. From the side illuminated
by source 1, the rays pass to P and are reflected to C, and the light passes
through to the telescope without obstruction except where the cube is etched.
From the other side, illuminated by source 2, the rays pass to P' and are re-

1
s
2

~~~--,L..,L.L1~~~~-r-'.:2-V-V/_V,_:,;,o_:,;,o_U,ZZV/V?aT/,o/,'7~

c b

Field of view

Frn. 27-10. Diagram of the Lummer-Brodhun phot.ometer

fleeted to C, where the light is reflected at the etched portions into the tele-
scope. The field of view is thereby illuminated at zones a and b by lamp 1
and at zones c and d by lamp 2. When the screen is so placed as to be
equally illuminated by the sources, the pattern vanishes and the field appears
uniform throughout. This arrangement, whereby a zone illuminated by one
source is completely surrounded by another illuminated by the other source,
makes it easy for the eye to judge equality of brightness and permits the
luminous intensities of the light sources to he compared with precision.
In accurate visual photometry, a substitution method is employed so as to
avoid possible differences between the two sides of the photometer. One of
the two sources, say lamp 1 of Fig. 10, is used as a comparison source; first the
standard lamp is used as source 2, and then the "unknown" or test lamp,
and Eq. 27.5 is used twice.
Difficulties are encountered when it is attempted to compare two lamps of
different color, since the eye cannot judge accurately the equality of bright-
ness of two surfaces unless their colors match approximately. Lamps of dif-
ferent color can be compared by means of a so-called flicker photometer which,
§ 27-10 PHOTOMETRY ,521

by a rotating prism or other suitable optical arrangement, enables the observer

to view first one side of the screen and then the other alternately and in rapid
succession. Upon increasing the frequency of alternation, a value is found
for which the flicker due to color difference disappears, and the colors of the
two sources will appear to blend into a single resultant color. If the fre-
quency is not too high, however, the flicker due to brightness difference re-
mains. The photometer screen is then moved until this flicker also disap-
pears, yielding a brightness equality; the luminous intensity of the test lamp
is computed in the usual way. Flicker photometry does not give reliable
results for strongly colored lights.
In photometry requiring the highest accuracy, and particularly where
sources of different color are to be compared, the human eye is not sufficiently
trustworthy; thermopiles are used instead, because of their freedom from
color selectivity and their linearity of electrical output with light intensity.
This device receives the radiant flux from the lamp to be calibrated, and the
electrical response is compared ,vith that given by a standard lamp of known
intensity. Under controlled conditions such radiation receivers do not vary
with time; hence the two lamps may be measured in sequence, and a compli-
cated photometer head like that of Fig. 10 is not needed. The reduction from
radiant to luminous flux units is effected automatically by placing before the
thermopile a color filter which at each wavelength transmits light in propor-
tion to the relative luminosity factor of the "standard observer" for that
wavelength; see Fig. 6.
In the photometric methods described, the intensity of a light source is
measured in only one direction at a time. When it is desired to know the
total luminous flux output of a lamp, use is made of a sphere photometer-
a hollow sphere, sometimes as large as 15 ft in diameter, with its inside surface
painted matt white. The lamp to be tested is placed near the center of the
sphere; the light emitted in all directions from the lamp reflects around inside
the sphere in a random fashion and gives it uniform illumination. This illu-
mination value is measured by a photocell placed in the wall of the sphere, and
shielded from the direct radiation of the lamp. The response given by the
photocell with the test lamp in place is compared with that due to a standard
lamp.
Actual measurements of luminous intensity are of interest chiefly to
manufacturers of electric lamps and lighting fixtures. Much more common
are measurements of the illumination produced by artificial lighting in homes
and offices, in factories, and on streets and highways. One instrument for
this purpose is the illuminometer, a portable photometer for comparing the
brightness of the surface under measurement with that produced by a small
incandescent lamp of known luminous intensity on a screen within the instru-
ment. By the use of suitable accessories, illumination rather than bright-
ness can be measured. The lamp is mounted within a light-tight enclosure
and can be moved until it produces the same brightness as that of the surface
522 OPTICS Chap. 27

being tested. Since each position of the standard lamp corresponds to a

particular value of illumination, a scale can be provided on the instrument to
give the result directly in footcandles. The illumination of a surface can also
be measured readily with a photovoltaic cell that is connected to a microam-
meter which is calibrated to read directly in footcandles or other desired units;
the exposure meter used by photographers operates on this principle. Typical
values of illumination recommended by illuminating engineers are listed m
Table II.

TABLE II. lLLUl\IINATION VALUES

Footcandles
Class rooms and laboratories ........... . 20-50
Corridors and stairways ............... . 5
Drafting rooms ....................... . 30-50
Flood lighting ........................ . 5
Hospital operating tables .............. . 200-2000
Library reading rooms . . .............. . 30
Offices .............................. . 30-50
Retail stores ......................... . 20-50
Show windows ....................... . ,50-200
Tennis courts ........................ . 20
Theatre and church auditoriums ........ . 5
Waiting rooms ....................... . 20

27-11. Illumination of Interiors. For correct interior lighting a sufficient

amount of light should be provided to give the desired illumination, and the
lighting fixtures should be so designed and placed as to make the illumination
reasonably uniform, without glare. Three methods of lighting are in com-
mon use, the so-called direct, indirect, and semi-indirect methods. With
direct lighting, the lamps project light directly upon the desired working
surfaces. With indirect lighting, the lamps are concealed entirely, and the
illumination is produced by reflection from the walls and ceiling of the room.
With semi-indirect lighting, the illumination is produced in part by light re-
flected from the walls and ceiling, and in part by direct transmission, usually
through translucent glassware.
A large amount of light is absorbed by the walls and ceiling, especially if
these are dark colored; this is particularly true for small rooms, where the wall
and ceiling surfaces are relatively large in comparison with the surface to be
illuminated. It follows that a comparatively small fraction of the light flux
generated at the lamps is received at the working plane; this fraction is called
the coefficient of utilization, and its value varies with the type of lighting
fixture, the color and finish of walls and ceiling, and the size of the room.
Such coefficients have been measured under a wide variety of conditions; some
values suggestive of their order of magnitude are given in Table III. The
§ 27-11 ILLUMINATION OF INTERIORS 523

values listed are based on average conditions of cleanliness of the lighting

fixtures and are appropriate for a 12-ft ceiling height.
The usual method of determining the size of lamps to be installed is as
follows: First, it is necessary to calculate the amount of light flux which must
reach the working plane in order to produce the desired illumination. Next,
a suitable coefficient of utilization is applied in order to obtain the number of
lumens that must be generated in the lamps. Finally, from the efficiency of
the lamps, the number and size of lamps to be used can be determined.

TABLE III. COEFFICIENTS OF LIGHT UTILIZATION

Floor Area, Square Feet

Type of Color of Both 1600

Lighting Unit Walls and Ceiling 200 400 800 and over

Direct Light 0.30 0.34 0.38 0.41

Dark 0.28 0.32 0.36 0.39
-----
Semi-indirect Light 0.16 0.20 0.23 0.26
Dark 0.095 0.12 0.14 0.16

Indirect Light 0.12 0.15 0.18 0.20

Dark 0.045 0.065 0.080 0.090

The procedure is illustrated by calculating the number of 100-watt incandescent

lamps needed to give an illumination of 20 footcandles with i;;emi-indirect lighting units
in a room measuring 20 ft X 40 ft and
having light-colored walls and ceiling.

1--' ·
Figure 11 :;;hows a suitable division of the
room into eight sections, with a ceiling i · i I •

,--~--r-~--r-~--r-~---
fixture at the center of each; this arrange- I I I
ment should afford reasonable uniformity I I I

of illumination. The desired illumination 20111

is equivalent to 20 lumens/ft 2 and, since
the area to be illuminated is 20 X 40 = --
•
I
I
I
I
I
800 ft2, it follows that 20 X 800 = 16,000
lumens must reach the working plane. The Fm. 27-11. Design of room lighting
coefficient of utilization for the conditions
of the room is found from Table III to be 0.23; thus 16,000/0.23 = 69,600 lumens
must be generated in the lamps. Therefore, each unit must furnish 69,600/8 =
8700 lumens. Table I shows that a 100-watt tungsten-filament lamp has an effi-
ciency of 15 lumens/watt; for this efficiency the input to each unit must be
8700/15 = 580 watts. Hence, six 100-watt lamps are needed in each of the eight
lighting fixtures.

PROBLEMS
1. An apparatus for measuring the speed of light employs a revolving mirror, and
a stationary mirror 40.0 km distant. If an eight-sided mirror is used, what must be
its Rpeed of rotation?
524 OPTICS Chap. 27

2. The luminous efficiency of a sample of radiant flux is 272 lumens/watt. The

sample consists of 20 watts of radiant flux of monochromatic light of wavelength less
than 555 mµ. (a) Compute the luminous flux in the sample. (b) What is the wave-
length of the light?
3. A beam of colored light comprises 15 watts of monochromatic light of wave-
length 480 mµ and 20 watts of monochromatic light of wavelength 610 mµ. Determine
(a) the luminous flux, (b) the radiant flux, and (c) the luminous efficiency.
4. Compute the luminous flux in a beam of light of relative luminosity 0.30, if the
radiant flux is 6 watts.
5. A surface of 0.8 in. diameter is illuminated normally by a 20-watt sodium-vapor
lamp that is 4 ft distant. Compute the luminous flux incident on the surface.
6. A photovoltaic cell with a sensitive surface 4 cm in diameter delivers 180 X 10-6
amp to a microammeter when the cell is illuminated by a source of luminous intensity
150 candles at a distance of 25 cm. Determine the sensitivity of the apparatus, in
microamperes per lumen.
7. The minimum luminous flux required to operate a certain photoelectric circuit
is 2 lumens. The diameter of the sensitive surface of the photocell is 2 in. Located
2.5 ft from the cell is a uniform point source. Compute its minimum luminous
intensity for the operation of the circuit.
8. Show that 1 footcandle is equivalent to 10.76 meter-candles.
9. The illumination on a section of a horizontal surface is 30 footcandles, and is to
be increased to 70 footcandles by means of an additional 25-watt "photo.flood" lamp
placed directly above the illuminated section. Assume that the lamp radiates uni-
formly in all directions. Compute the height, from the center of the section, at which
the lamp should be placed.
10. Derive Eq. 27.4 by considering the light flux intercepted by a surface so inclined
that a normal to the surface makes an angle i with the direction of the incident light
rays.
11. Two lamps, A and B, are located 15 ft apart. The luminous intensity of A is
25 candles and that of B is 100 candles. (a) At what position along a straight line
between the lamps will equal illumination be produced? (b) Compute the illumina-
tion at this position, on an area perpendicular to the line between the lamps.
12. A standard lamp, of luminous intensity 100 candles, is at one end of a photome-
ter bench, and a test lamp is at the other end. The final setting of the photometer
head is two thirds of the way from the standard lamp to the other. Compute the
luminous intensity of the test lamp.
13. A standard lamp of luminous intensity 37 candles and a test lamp of unknown
luminous intensity are placed 200 cm apart on a photometer bench; they produce equal
illumination on the photometer screen when it is 86 cm from the standard lamp.
Compute (a) the illumination produced at the screen by each lamp, and (b) the lu-
minous intensity of the test lamp.
14. A vertical sign 6 ft high is illuminated by two floodlights placed 4 ft in front of
its upper edge and 6 ft apart. The luminous intensity of each floodlight is 780 candles.
Compute the illumination at a spot on the sign that is half way between top and
bottom and is equally distant from the two lights.
15. A work bench is illuminated by two lamps, each of luminous intensity 225
candles; they are 3 ft above the bench and are spaced 4 ft apart. Calculate the
illumination on the bench at a point directly under one lamp.
16. In employing the "substitution" method of photometry mentioned in § 27-10,
the following data are obtained: (a) When the comparison source, Lamp 1, is tested
against the standard source, Lamp 2, of luminous intensity 50 candles, r1 = 80 cm
and r2 = 120 cm. (b) When the "unknown" lamp is substituted as Lamp 2, r1 = 110
cm and r2 = 90 cm. Find the luminous intensity of the "unknown" lamp.
 PROBLEMS 525

17. A 15-watt high-intensity tungsten-filament lamp is held at the center of a spheri-

cal enclosure 4.5 ft in diameter. Determine the average luminous flux received per
square foot of illuminated surface.
18. A "photoflood" lamp is supported at the center of a spherical enclosure 4 ft in
diameter. The average amount of luminous flux received on each square foot of the
inner surface of the sphere is 36 lumens. Determine the power rating of the lamp,
in watts.
19. A library reading room measuring 24 ft X 30 ft has light-colored walls and
ceiling and is equipped with six symmetrically spaced ceiling outlets. Direct lighting
units with daylight fluorescent lamps of efficiency 48 lumens/watt are to be installed.
Compute the number of 40-watt lamps required at each outlet in order to provide
suitable illumination.
20. A hospital operating room 20 ft square has light-colored walls and ceiling and
has twelve symmetrically spaced ceiling outlets. The recommended illumination
value is 200 footcandles. Semi-indirect lighting units with fluorescent lamps of effi-
ciency 60 lumens/watt are to be installed. Compute the number of 100-watt lamps
required at each outlet.
 28
REFLECTION, REFRACTION,
AND DISPERSION
The direction of a beam of light can be altered by reflection and by refrac-
tion, processes upon which are based the design and operation of most optical
instruments. The angle through which light is deviated in reflection depends
only upon the inclination of the reflecting surface, but in refraction it depends
also upon the nature of the optical mediums involved and the ·wavelength of
the light. A beam of blue light is refracted through a larger angle than is a
beam of red light; this action, called dispersion, is widely used in the spectro-
scope and related instruments. The treatment of reflection, refraction, and
dispersion in this chapter is restricted to plane surfaces, and monochromatic
light is generally assumed in the sections on refraction.

REFLECTION

28-1. Regular and Diffuse Reflection. The behavior of a plane wave

front upon striking a plane surface is treated in § 25-12, wherein Huygens'
s s
Q

I /Ip
I I I
I I I
I ., / I
I I I I
f, I
I I
}'
I
FIG. 28-1. Reflection of spherical Fw. 28-2. Reflection
wave at plane surface represented by rays

construction is used to determine the orientation of the reflected wave front.

This procedure may be applied to the reflection of light, as is illustrated in
Fig. 1. A source of luminous flux is located at 8, in front of the plane mirror
2YI. A wave front that originated at 8 would have advanced to position WW
had it not been for the reflecting surface. At the positions shown, each
point on this wave front between A and B has reached the reflecting surface
526
§ 28-3 IMAGES FORMED BY PLANE MIRRORS 527

and returned to some position along the arc of a secondary wavelet; several
such wavelets are represented by the circles in the figure. A curve tangent
to these arcs shows the position of the reflected wave front R; the reflected
wave appears to have come from I, and this point is called the image of S.
By symmetry, the image I is located on a line through S normal to the mirror
surface and is as far back of the mirror as the source is in front of it. Thus,
a ray of light SP upon striking the mirror is redirected along PQ as though it
had come from the image I.
The rays SP and PQ, together with the normal connecting the source Sand
the image I, are transferred to Fig. 2 for clarity. Since the source and image
are equally distant from the mirror, SN = NI, and the triangles SNP and
INP are similar, the side NP being common. It follows that angle i at S
and angle i' at I are equal, and these in turn are equal to the angles similarly
marked at P. Hence the angle of incidence i between the incident ray and
the normal is equal to the angle of reflection i' between the reflected ray
and the normal.
"Reflection of light from a smooth surface takes place along a definite direc-
tion, as described, and is designated as regular or specular. In reflection from
a rough or matt surface such as that of plaster
or blotting paper there is regular reflection for
each tiny element of the reflecting surface; but
since these elements are disarrayed, reflection
occurs in a great many directions for any one
direction of the incident beam, as suggested in
Fig. 3, and is said to be diffuse or scattered. It
Frn. 28-3. Surface irregular-
is by diffuse reflection that nonluminous objects ities magnified to show diffuse
become visible. reflection
28-2. The Law of Reflection. A comparison
of the ray treatment in the preceding section with the wave-front treatment
in § 25-12 shows that the angle of incidence can be regarded either as the
angle between the incident wave front and the reflecting surface, or more
simply as the angle between the incident ray and the normal to the reflecting
surface; similarly for the angle of reflection. From either point of view, the
angle of incidence i is equal to the angle of reflection i'. The generalized
statement of this equality is of great importance in Optics, and is usually
called the Law of Reflection. The complete statement of the law includes the
fact that the two rays and the normal to the surface at the point of incidence
all lie in the same plane. Although this law has been used thus far only for
reflection at plane surfaces, it applies equally to reflections from curved
(usually spherical) surfaces, and this explains why the angles are more gen-
erally expressed with respect to the normal.
28-3. Images Formed by Plane Mirrors. The image of a point object
formed by reflection in a plane mirror is located the same distance behind
the reflecting surface that the object is in front of it, § 28-1. For an extended
528 OPTICS Chap. 28

object, each image point is similarly located, and so the complete image can
readily be constructed. Figure 4 shows an object 0, and its image I formed
by reflection in the mirror 111. Rays are drawn for the extreme object points
and are extended until they strike the eye of the observer. It will be noted
that the angles of incidence, as at i1 and i2, are equal respectively to the angles
of reflection, as at 1"/ and iz'. The rays from each point enter the eye as a
diverging pencil, and to the observer it appears that the object is located at
the place where the prolongations of these rays intersect: at I, the image.
Two plane mirrors that are inclined to each other yield multiple images of
an object placed between them. These images can be located by the method
just explained.

y'
:Frn. 28-4. Image formation by a :Frn. 28-5. Effect pro-
plane mirror duced on reflected ray
by rotating a mirror

28-4. Rotation of Reflected Ray. The rotation of a plane mirror upon

which a ray of light falls causes the reflected ray to rotate also; moreoever, the
amount by which the reflected ray turns is twice that of the mirror. To prove
this relation, an incident ray of light from a source S may be considered to
fall upon a plane mirror M, Fig. 5, at an angle i1 with the normal N; it is re-
flected along OY at an equal angle i 1', and the angle SOY = i1 + ii' = 2 i1.
As the mirror is rotated through an angle f3 to M', the normal undergoes an
equal rotation to N'; the angle of incidence increases to i2 = i1 + {3, and the
angle of reflection iz' increases equally, so that the reflected ray is rotated to
the position shown at Y'. The angle SOY' = i2 + iz' = 2 i2 = 2(i1 + {3).
Therefore
+
YOY' = SOY' - SOY = 2(i1 f3) - 2 i1 = 2 f3.
Thus the reflected ray turns through twice the angle that the mirror does.
This fact is used in amplifying the deflections of galvanometers and other
sensitive instruments.
In order for a navigator to determine the latitude of a particular position,
it is necessary for him to measure the angle between the sun and the horizon
as subtended at his position on board ship. This he does with a sextant, a tele-
scopic instrument which is held in the hand while the observer brings images
of the sun and horizon into coincidence in the field of view of a small telescope.
§ 28-5 'THE LAW OF REFRACTION 529

The sextant consists essentially of two mirrors called the index glass M and
the horizon glass m, Fig. 6, supported perpendicularly to the plane of the
sextant on a metal framework which also carries the telescope and a scale.
Mirror mis fixed in position and is clear over half its surface, so that the ob-
server whose eye is at E can see the ho-
rizon beyond B directly, thatis, without
reflection at mirror m. Mirror M is
carried on an arm which is pivoted at A
P and is fitted with a vernier V at its
other end so as to permit the position
of the index glass to be read accurately B
E
on scale S. When the two mirrors are
parallel, the observer will see the hori-
zon also via the path AMmE, and this
image will blend with the direct one
obtained along BE; the vernier for this
position of mirror M will read zero. In FIG. 28-6. Arrangement of a sextant
viewing the sun at an angle e degrees
above the horizon, the arm carrying the mirror will have to be turned through
an angle 8/2 so that the rays from the sun coming along the line CM will be
reflected along the line .Mm and again reflected into the telescope by the
silvered portion of the horizon glass m. The image of the sun is then aligned
with the image of the horizon, and the altitude of the sun will be indicated by
twice the angle through which the vernier arm is turned. Each half-degree
division on scale S is marked as one degree; this makes it possible to read the
angular elevation of the sun directly.
REFRACTION
-
28-5. The Law of Refraction. Refraction is treated for wave motions in
general in § 25-13 by applying Huygens' Principle to a plane wave front, and
the wave is found to undergo an abrupt change of direction when it passes
obliquely from one medium into another. In the study of the refraction
of light it is usually more convenient to deal with rays than with succes-
sive positions of a wave front, and, furthermore, it is preferable to measure
the angles of incidence and refraction from the rays to the normal rather than
from the ,vave fronts to the surface. A ray is bent toward the normal when
it passes into a medium in which the speed of light is reduced and is bent
away from the normal when it passes into a medium in which the speed is
increased. Only homogeneous or so-called isotropic mediums are considered,
except in describing atmospheric refraction effects.
In Fig. 7 a ray of light is shown striking the boundary surface between two
mediums, at an angle of incidence i. Some light will be reflected at angle i'
and, if the second medium is transparent, some light will enter it, making an
angle of refraction r with the normal N to the surface. The relative amount
530 OPTICS Chap. 28

of light reflected and refracted is discussed in § 28-10; the present concern is

only with the direction of the refracted beam. If the speed of light in Medium
1 is indicated by v1 and the speed in :Medium 2 by 112, then, since Fig. 7 is drawn
for V1 > v2, the refracted ray is deviated toward the normal, and r < i.
The ratio vi/v 2 is called the refractive index of the
N second medium with respect to the first and is given
the symbol n 12, where the order of the subscripts indi-
cates the direction the light travels. As the name
suggests, the refractive index is a measure of the devi-
ation of a ray of light as it traverses the boundary sur-
face between the two mediums. It is usual practice to
quote the refractive indexes of optical materials with
respect to a vacuum; that is, to regard a vacuum as
the other medium involved. A vacuum refractive
FIG. 28-7. Reflection . d · · · h b · d · 11 d
and refraction at a m ex IS written wit out a su script .an 1s ca e "the
plane surface refractive index of the substance." If the speed of light
in a vacuum is indicated by c, and the speed of light in
the substance by v, then the index for that substance is n = c/v. Since
c > v, the value of the index n is always greater than unity, except of course
for a vacuum, for which n = 1.
The amount of refraction is indicated by Snell's Law, which states the
constancy of the ratio of sin i to sin r for two mediums as
sm i Vi
(25.10) -.-=-·

This equation can be combined with that for refractive index by dividing
numerator and denominator of the right-hand fraction by c:
sin i vi/c
--=-=--=-,
l/n1 n2
sin r vdc l/n2 n1
where n1 is the refractive index of the first medium (the one in which the light
approaches the interface) and n 2 is the refractive index of the second medium.
Upon simplification this equation yields the Law of Refraction in easily re-
membered form as:
. . .
n1 sin i = n2 sin r. (28.1)

Again, the complete statement of the law includes the fact that the two rays
and the normal lie in the same plane-the so-called plane of incidence.
28-6. Refractive Index. It is stated in § 27-4 that the speed of light in air
differs only very slightly from that in vacuum, the difference being about 0.03
of 1 per cent for air at standard conditions. Therefore, the vacuum refractive
indexes can be used for refraction between any substance and air, with negli-
gible error except in very special situations(§ 28-13). The vacuum refractive
§ 28-7 APPARENT DEPTH OF A SUBMERGED OBJECT 531

indexes of a few common substances are listed in Table I; values are given for
gases at standard conditions-0°C and 760 mm Hg pressure; the values for
liquids and solids are for a temperature of 20°C.
TABLE I. REFRACTIVE INDEXES

Substance n Substance n

GASES SOLIDS
Air ...................... 1.000293 Diamond ................. 2.417
Carbon dioxide ............ 1.000450 Fused silica ............... 1.4.58
Hydrogen ................. 1.000132 Glass (crown) ............. 1.48 to 1.62
Glass (flint) . . . . . . . . . . . . . . . 1.54 to 1.80
LIQUIDS
Lithium fluoride ........... 1.39
Carbon disulfide ........... 1.62.5 Rock salt ................. 1.544
Ethyl alcohol .............. 1.360 Thallium bromide-iodide .... 2.60
Water .................... 1.333

Since the speed of light in any substance varies somewhat with the wave-
length of the light, the refractive index of that substance will vary with
wavelength. Hence, the ,vavelengths should be indicated when quoting re-
fractive indexes. If no such indication is given, as in Table I, it can be
assumed that the values apply for the wavelength of the characteristic yel-
low radiation from glowing sodium vapor, 589.3 mµ.
It will be realized that the larger is the refractive index of a medium, the less
rapidly will light travel in it. For example, in glass of index 1.50 the speed
of light is only two thirds of its value in air or in a vacuum. Therefore, it will
take light 1.5 times as long to traverse 1 cm of glass as 1 cm of air or, what
amounts to the same thing, light can travel 1.5 cm in air in the same time that
it travels 1 cm in glass. The numerical value of the refractive index of a
substance hence indicates the air distance that is equivalent to unit length
of the substance; this is the basis for the so-called "optical path length" used
in certain lens computations. Refractive index is sometimes called "optical
density," but the use of the latter term is not encouraged.
Since the speed v of light changes abruptly when the light ray passes from
one medium to another, it is natural to ask whether in the fundamental equa-
tion of wave motion, v = fA, the frequency or the wavelength changes when
refraction takes place. The answer that is clearly given by experimental facts
and associated reasoning is that the frequency remains unchanged, while the
wavelength shortens when the speed of light lessens. This constancy of fre-
quency indicates that frequency is a more fundamental quantity in wave
motion than is wavelength. It must be remembered that the wavelength
values previously mentioned assume light to be traversing a vacuum (or air).
28-7. Apparent Depth of a Submerged Object. Because of refraction an
object immersed in a medium of higher refractive index than air appears
532 OPTICS Chap. 28

nearer the surface than it actually is. In Fig. 8, the object point O may repre-
sent a stone in water a distanced below the surface S, with light rays traveling
upward to the eye, as shown. For the ray OXY, the angle of incidence i is
equal to GOX and the angle of refraction r is equal to
'/IX, so that
sin GOX 1
sin GIX n'
where n is the refractive index of water. For small angles
(smaller than those shown in the diagram) the sine may
be equated to the tangent, whence
tan GIX GX/d' d
n= = -,
tan GOX GX/d d'

or d' = 1..
n
(28.2)
FIG. 28-8. Effect of
refraction on appar- Equation 28.2 shows that the apparent depth d' of an ob-
entdepth of an object ject when viewed from above is the quotient of the actual
under water
depth d divided by the refractive index of the medium.
Although this equation was developed for the special case where air or vac-
uum is the upper medium, it also applies more generally; in the general case
n is replaced by the relative refractive index n 12 between a medium of larger
refractive index n 1 and one of smaller index n2-the convention here is the
same as that used in § 28-5.
28-8. Total Internal Reflection. A ray of light traveling in a medium of
higher refractive index and directed toward one of lower index passes into that
p
Q

Air (medium 2)
S--t--t----,i<---,1:--'-',.._,.._--------

0
Frn. 28-9. Illustrating total internal reflection

medium only if the angle of incidence is not too large. If the incident ray
is inclined more and more away from the normal to the interface, a position
will be reached for which the ray does not enter the second medium, but is
totally reflected at the interface.
A source 0, located in water (Medium 1) of refractive index n, may be
supposed to emit light in all directions; a few of the rays are indicated in Fig. 9.
§ 28-9 REFLECTING PRISMS 533

The light emitted along OA is mostly refracted into the air (Medium 2) along
AP in such a direction that n sin i = sin r, but a small part is reflected along
AW. Similarly, the light emitted along OB is partly refracted along BQ and
partly reflected along BX.
The ray OC strikes the surface S at such an incidence angle ic that the
refracted ray CR grazes the surface, the angle of refraction being 90°; the
reflected ray travels along CY. Any ray such as OD, for which the angle of
incidence is greater than ic, does not emerge into the air at all, but is totally
reflected along line DZ. The angle i,, is called the critical angle of incidence;
it is the angle of incidence in the more highly refractive medium for which the
angle of refraction in the other medium is go 0 , and its value is evidently given
by
1 (28.3)
Sln ic =
n
In general, total reflection will occur at the boundary separating any two
mediums when a ray in the medium of higher index n 1 is directed toward one
of lower index n2 at an angle of incidence greater than the critical angle.
Hence the critical angle is determined by
. . n2 (28.4)
Sln i 0 = -.
n1
As an example, a ray directed from glass of refractive index 1.50 into air will be
totally reflected internally ,vhen the angle of incidence is greater than

1- = 41 8°
i c = sin-1 -1.50 . .

Instruments called refractometers are used for determining the refractive

index of solids and liquids, by measuring the critical angle of light passing
from a "standard" prism of high-index glass to the material under test. The
range of such instruments is limited by the optical density of the standard
prism material, usually dense flint glass of refractive index 1.7 or higher.
28-9. Reflecting Prisms. When it is desired to deflect the light beam in
optical instruments through an angle of 90° or 180°, use is made of reflecting
prisms which have angles of 45°, 45°, and 90°. Such totally reflecting prisms
are preferred to simple mirrors because of freedom from double images, higher
reflectivity, and superior mechanical sturdiness. In Fig. 10 a back-silvered
mirror, in part I, is compared with the simple reflecting prism in part II.
Both give the beam a deviation of go 0 when they are oriented as shown with
respect to the incident light. The ordinary mirror is silvered on the back
surface, and the reflecting layer is protected by a coat of paint. A small
amount of light is reflected from the front (glass) surface; for normal incidence
(i = 0; i' = 0) this weak reflected beam is not noticed since it coincides with
the stronger beam reflected from the back surface, but for other than normal
incidence the beams are separated, although parallel. This donhle-image
534 OPTICS Chap. 28

effect can be avoided by silvering the front rather than the back surface of
the mirror, as shown in part III of the figure. However, the bare surface of
silver (or aluminum, more commonly) is soft and fragile, and, therefore, front-
surfaced mirrors are used only in very special circumstances, usually inside
permanently sealed instruments.

SUve.ed ~

n m
Fw. 28-10. Deviation of 90° produced by a back-silvered mirror, a reflecting prism,
and a front-silvered mirror

In the reflecting prism the angle of incidence at the entrance and exit sur-
faces is zero, and hence there is no doubling of the beam. The angle of in-
cidence at the reflecting surface is 45° and consequently exceeds the critical
angle for glass-to-air of about 42°. Since total internal reflection occurs, it
is not necessary to silver this reflecting surface, but it is important to keep
the surface clean and free from dust.
The 45°-45°-90° prism can be used also as a so-called Porro prism to give a
deviation of 180°, by reorienting the prism so that the light is incident on its
hypotenuse face, as shown in part I of Fig.
11. W~en so used the prism can be regarded
as two smaller 90° prisms in succession, as
indicated in part II; sometimes it is con-
structed in this manner. A third way to use
a 45°-45°-90° prism is mentioned in § 28-11.
I II 28-10. Reflection Losses. Reference has
Fm. 28-11. Deviation of 180° pro- been made to the fact that, when a beam of
duced by reflecting prisms light strikes the boundary between two
mediums of different refractive index, some
of the light is reflected and some is refracted. The distribution of intensity
between the reflected and refracted beams depends upon the refractive indexes
of the mediums, the angle of incidence, and the direction of traversing the in-
terface. In Fig. 12 the percentage of light reflected is indicated for the
typical situation of an interface between air and glass. of refractive index
1.50. Curve A applies when the beam of light is traveling in air and inci-
dent upon the interface, whereas Curve B applies when the beam is traveling
in the glass and incident upon the interface. Total internal reflection
occurs for an angle of incidence of 42°, as computed in § 28-8. Curve
A shows that the reflectance loss at an air-to-glass surface is constant at
about 4 per cent for angles of incidence up to perhaps 45°, and then in-
§ 28-11 REFRACTION BY A PARALLEL PLATE 535

creases at first gradually and then more rapidly to 100 per cent for "grazing
incidence."
The laws governing this distribution of light intensity were worked out by
the French mathematician, physicist, and engineer, Augustin Fresnel (1788-

-..
cCl)
100

80 J
u
Cl)
a. 60
B A/
Cl)
u
I
- c
0
u
Cl)
.::;::
Cl)
40

20
/
~
/ __. . /

0° 10 20 30 40 50 60 70 80 90°
Angle of incidence
Frn. 28-12. Per cent reflectance at interface between air and glass of refractive index 1.50,
plotted against angle of incidence. Curve A, light traveling in air before reflection;
Curve B, light traveling in glass before reflection

1827), and are rather complicated, especially since the polarization (Chap. 33)
of the incident light is an important factor. For unpolarized light at or near
normal incidence, and for a medium of index n with the other as air or vacuum,
Fresnel's equation reduces to
(n - 1) 2
Per cent reflectance = (n +
l) 2 X 100. (28 ..5)

With light reflected normally in air by glass of refractive index 1..50, Eq. 28 ..5
gives (0 ..5/2 ..5) 2 X 100 = 4 per cent. This loss of light by reflection, although
small, becomes important in the more complicated optical instruments that
have a number of lenses, prisms, and windows, and hence many air-glass
surfaces.

a
w

g
a

Frn. 28-13. Light ray displaced but not deviated in passing through parallel-sided slabs

28-11. Refraction by a Parallel Plate. A ray of light in passing through

one or more parallel-sided slabs and emerging into the original medium is dis-
placed laterally but not deviated. This result is shown in Fig. 13, where
536 OPTICS Chap. 28

a, w, and g represent any three different mediums, such as air, water, and glass.
The angles x, y, and z are called the angles of incidence, refraction, and emer-
gence respectively. A single slab, as in part I of the figure, is considered first
and Snell's Law is applied to each surface of the glass; then
Slll X Va
--=-=na
sin y Vg g,

sm y _ ~ _
sin Z - Va - nga,

If these equations are multiplied, member by member, the product of the

second members becomes unity; thus:
. .
sm x sm y
-.- X -.- = 1 = nag X nga•
sm y sm z
The first two terms of this equation show that sin x = sin z, whence x = z;
consequently the incident and emergent rays are parallel. The last two terms
of the equation show that

that is, the refractive index of glass with respect to air is the reciprocal of the
refractive index of air with respect to glass.
,,...
/
/ ''

~ '
' , _____ -....... /
'~---~
' / /

II
PIG. 28-14. Part I, the Dove, or rotating, prism; part II, optical equivalence of this prism
to a plane-parallel plate

When the same procedure is applied to two slabs, as illustrated in part II

of the figure, it is found that the angle of incidence x and the angle of emer-
gence z are again equal, and also that

Hence

·which shows that the relative refractive index of any substance g with respect
to another substance w is equal to the air index (or the absolute or vacuum
index) of substance g divided by the air index of substance w.
The 45°-45°-90° prism of § 28-9 may be used as a so-called Dove prism to
§ 28-12 DEVIATION BY A PRISM 537

turn an image upside-down but not left-to-right, as indicated in part I of Fig.

14. It has the useful property, when rotated through an angle e around an
axis parallel to its hypotenuse face AB, of rotating the image through an angle
2 e. This action is similar to the behavior of mirrors, § 28-4. Part II of the
figure shows that the Dove prism is equivalent to a parallel plate (compare
part I of Fig. 13) in displacing the light beam without changing its direction;
hence the prism when used in this way will not disperse light rays of different
wavelengths (§ 28-14).
28-12. Deviation by a Prism. Whereas the refractive index of a transpar-
ent substance can be evaluated by measuring the angles of incidence and re-
fraction of a light beam at its surface in a
situation like that of Fig. 7, a more accurate
determination is made by measuring the
deviation of the beam in traversing a prism
of the substance. In general, light is devi-
ated whenever it passes through a refracting
body having nonparallel sides; the Dove
prism is an exception. The amount of de-
viation produced by a prism depends upon Fm. 28-15. Ray passing through a
the angle of the prism, the refractive index 60° prism at minimum deviation
of the substance, and the angle of incidence.
It can be shown, both experimentally and mathematically, that the devia-
tion has a minimum value when the ray passes symmetrically through
the prism, as shown in Fig. 15.
If a beam of light is considered to strike one face of the prism at an angle i
with the normal N, and to be refracted at an angler with this normal, it will
proceed through the prism and emerge from the second face, making angles
of incidence i' and refraction r' with the normal N'. From geometry, the
angle A of the prism is equal to the angle A' between the two normals and,
since the latter angle is the exterior angle of a triangle of which r and i' are
the acute interior angles, it follows that A = r + i'. The angle of overall
deviation o between the incident and emergent rays is the sum of the devia-
tions a and a' produced at the two faces individually, or

o= a + a' = (i - r) + (r' - i') = i + r' - A.

If n is the refractive index of the prism material relative to air, it follows from
the Law of Refraction, Eq. 28.1, that sin i = n sin rand that n sin i' = sin r',
so the deviation is

o= sin-1 (n sin r) + sin- 1 [n sin (A - r)] - A.

In order to find the relation between the angles rand i' which will make the
deviation a minimum, the foregoing expression is differentiated with respect
to r; the result is then equated to zero. Thus,
538 OPTICS Chap. 28

do n cos r
------- n cos (A - r) = 0.
dr (1 - n 2 sin 2 r)1/2 [1 - n 2 sin 2 (A - r)J 112 '

cos 2 r 1 - n 2 sin2 r
whence
cos 2 i' 1 - n 2 sin2 i' '
and it follows that r = i'.
For the minimum-deviated ray-the one that passes symmetrically through
the prism-the angles r and i' are equal, and the angles i and r' are equal;
hence the angles a and a' are also equal. Then the angle of minimum devi-
ation is
Om = a + a' = 2 a = 2 (i - r).
The refractive index n of the prism material can be expressed in terms of the
prism angle A and the angle of minimum deviation om by stating the angles
of incidence and refraction in terms of these quantities, and applying Eq. 28.1.
From the foregoing, the angle of refraction is r = A/2, and the angle of
incidence is
i = Om +r = Om + A.
2 2
From the Law of Refraction, sin i = n sin r, there results

. (om. +
Sll1 2
A.)
n = ---'----~ (28.6)
sin (t)
which is the desired expression for the refractive index.
An accurate determination of the refractive index of a substance can be
made from measurements of A and om taken on a prism of that substance.
A 60°-60°-60° (equilateral) prism is chosen for optical glass and common
optical crystals with refractive indexes in the usual range of 1.4 to 1.75, but
for more highly refractive materials a smaller angled prism of the isoceles
shape is usually employed. Since only the two refracting faces and the angle
A between them are involved in the measurements and calculations, the base
of the prism does not need to be polished. A hollow glass prism is used to
hold a liquid to be measured for refractive index; the faces of the prism, being
parallel-sided slabs, produce no deviation and thus do not influence the result.
Angular measurements are made with a spectrometer, the essential parts
of which are shown in Fig. lG. The test is conducted most simply with mono-
chromatic light, corresponding to a particular color of the Hpectrum. The
light enters the instrument through a narrow slit, is formed into a parallel
beam by a lens, and falls upon the prism. Here it undergoes two refractions
and upon emergence is received by a suitably placed telescope, through which
the observer sees an image of the slit. The prism and telescope are turned
by trial until the angle of deviation ois observed to be a minimum; the value
§ 28-13 ATMOSPHERIC REFRACTION 539

of this angle is then read from the circular scale. The prism angle A is usually
measured optically by reflecting light from one face of the prism, as from a
plane mirror, and then rotating the prism until the second prism face occupies
the same position; the prism angle is found by subtracting the angle of rotation
from 180°.

Fm. 28-16. Deviation of light measured with a spectrometer

28-13. Atmospheric Refraction. Probably most persons have noticed the

apparent quivering of an object when observed across the top of a hot radiator.
This effect is explained by the fact that the heated air has a slightly lower
refractive index than cool air, and the mixing of the two in the process of
convection produces an irregular bending of the transmitted light rays.
Light from outer space encounters air of increasing density and increasing
refractive index as it approaches the earth. Consequently, a beam of light
slanting downward through the atmosphere is bent more and more toward the

Observer

Frn. 28-17. An example of atmospheric refraction: The moon can be seen even though it
is below the geometrical horizon

vertical as it advances. For this reason, the stars are not seen in their true
positions unless they are directly overhead. The same phenomenon also ex-
plains the fact that the entire disk of the sun or moon can be seen for a short
time when it is geometrically below the horizon, where it would not be visible
at all except for atmospheric refraction. This effect is pictured in Fig. 17.
Another terrestrial phenomenon produced by atmospheric refraction is the
mirage; for example, the traveler on a desert seems to see a lake ahead, but it
is really a patch of blue sky. The heated air near the surface of the ground
has a slightly smaller refractive index than normal and hence forms an "air
540 OPTICS Chap.28

prism" oriented with its apex down and its base up. Through it the sky light
reaches the observer seemingly from a point on the ground, some distance
ahead, where he imagines there is water. This same effect accounts for the
large "puddles of water" that are imagined on a warm day to lie a few hundred
yards ahead on a smooth highway; again, they are reflections of the blue sky.
DISPERSION

28-14. The Phenomenon of Dispersion. White light, such as that from

the glowing carbons of an arc lamp, is in reality composed of many colors
blended together. When a narrow slit is illuminated by white light and the
issuing beam is refracted through a prism, the colors are spread out into a
brilliant array, merging insensibly into one another and forming a spectrum.
Although there are hundreds of distinct hues in the spectrum, they may be
grouped broadly into six principal colors. These are red, orange, yellow,
green, blue, and violet, in the order of increasing deviation by a prism, as is
shown in Fig. 18. Actually the spectrum extends at both ends into regions

,,,fa,~~3:::===~~==Red
Orange
Yellow
Green
Blue

Violet
Frn. 28-18. Dispersion of white light by a prism

invisible to the eye, the infrared being beyond the red end, and the ultraviolet
beyond the violet end of the visible spectrum, as is indicated in Fig. 27-5. The
spreading out of a light beam into its component colors is known as dispersion.
Color is determined by the frequency of vibration and the associated length
of the light wave, the wavelength being greatest for red and least for violet.
When the light refracted through a prism is passed through a second prism
identical with the first and arranged to produce deviation in the same direc-
tion, a spectrum of greater length will be produced but, when the second prism
is opposed to the first, the colors will be recombined almost perfectly into
white light.
It has been noted that the deviation produced by a prism is greater for
light of shorter wavelength. There is no simple mathematical relationship
between deviation and wavelength, however, and prisms of different sub-
stances spread out the component colors of the spectrum to somewhat different
extents. If two such prisms are arranged one above the other in such a man-
ner that their spectra match exactly at two colors, say at the extremes of the
red and violet portions, it wil1 be found that the intermediate colors do not
§ 28-14 'THE PHENOMENON OF DISPERSION 541

register point by point. A few substances exhibit what is known as anomalous

dispersion; prisms of such substances do not disperse white light in the regular
color sequence. Thus, fuchsine deviates red and orange light more than blue
and violet and absorbs the middle portion of the spectrum.
The angular dispersion of light depends upon both the refractive index and
the geometrical shape of the material. For a prism such as the one shown in
Fig. 15, angular dispersion is defined as the rate of change of deviation angle
with wavelength, and is expressed as do/d';,... A more direct way is to specify
the dispersion of a substance, as dn/d';,... The variation of refractive index with
wavelength is not simple; several approximate formulas have been suggested
to portray it. One of these is:
b c
n = a + xz + }..4 , (28.7)

where a, b, and c are constants characteristic of the substance, and can be

evaluated if the value of n is known for each of three values of A. If less ac-
curacy is acceptable, only the first two terms on the right-hand side of the
equation are used, and only two constants need be determined.

1.80 ..---,----,----,.-,---,---

~i---4-.L..:::I Crystalline
quartz
1.50 t - - - - - + - - - + - - - - 1 Ophthalmic
crown glass
1---I--..Q Fused silica
r:G:,-:Fi---=+--=-1 Calcium fluoride
1.40 '--''-'----'--'----''---U--1.L--J
400 500 600 700
Wavelen·gth, millimicrons
Frn. 28-19. Dispersion curves of some optical glasses and crystals. The location of the
most prominent Fraunhofer lines is indicated

The general form of the relation between n and';,.. is shown in so-called dis-
persion curves such as those of Fig. 19 for the visible spectrum. If the curves
are extended into the neighboring infrared and ultraviolet regions, they be-
come much steeper because so-called absorption bands are being approached.
It is within these absorption bands that the refracting material exhibits
anomalous dispersion.
542 OPTICS Chap. 28

The dispersion, dn/ d)o.., of a substance at any wavelength can be obtained by

measuring the slope of the dispersion curve at that wavelength. Often, an
inspection of carefully plotted dispersion curves of various materials gives the
desired information as to the relative dispersions in the wavelength regions of
interest.
28-15. Fraunhofer Lines. In the earliest study of the sun's spectrum, the
English chemist and natural philosopher William H. '\Vollaston (1766-1828)
noticed dark lines crossing the otherwise continuous band of colors. Later
the German optical physicist Joseph Fraunhofer (1787-1826) studied these
dark lines with improved spectroscopic apparatus and assigned letters to the
more conspicuous ones; these designations have remained in popular use ever
since. The wavelengths of the principal lines are given in Table II, and the
most prominent are indicated in Fig. 19. Some of the Fraunhofer lines were
subsequently shown to be not sharp, single lines but groups of lines; the aver-
age wavelength of the group is quoted for these, but only to three figures.
TABLE II. THE CHIEF FRAUNHOFER LINES

Designation Wavelength Color

B £\87 mµ red
c 656.3 red
D 589.3 yellov.
E 527 green
F 486.1 hlue-green
G' 434.0 violet
G 430.8 violet
h 410.2 violet
H 396.8 deep violet

Wavelengths of spectrum lines are measured in several units. The most

popular unit in spectroscopy has been the angstrom unit or angstrom, named
in honor of the Swedish physicist Anders J. Angstrom (1814-1874), one of
the earliest workers in spectroscopy; the angstrom, now designated by A,
is equal to 10-10 meter or 10-s cm. In this textbook the millimicron (mµ) is
ordinarily used as the wavelength unit: 1 mµ = 10-:i µ = 10-6 mm = 10-a m
= 10 A. The millimicron is chosen because it now has the widest acceptance
in all scientific fields except the academic spectroscopy of the visible and ultra-
violet regions.
28-16. Dispersive Power. A measure of the dispersion produced by an
optical material is given by its dispersive power, expressed in terms of the re-
fractive indexes at three wavelengths-ordinarily those for the red, yellow,
and blue light specified by the C, D, and F Fraunhofer lines. The definition
of dispersive power is given by:
nF - nc (28.8)
w = '
nv - 1
and its physical significance is explained in the next section.
§ 28-17 DEVIATION AND DISPERSION BY THIN PRI8Jv!S

Table III gives values of refractive indexes for some common optical materials
for the wavelengths of the C, D, and F lines. For ophthalmic crown glass,
the dispersive power is w = (1..5293 - 1.5204)/(1..5230 - 1) = 0.0170.
The reciprocal of w is often used, especially in glasR catalogs and handbooks,
to indicate dispersive power. This reciprocal is variously termed the recipro-
cal diRpersion, the constringence, the nu-value (after the symbol v which is
ordinarily used for it), and the Abbe number (named after the German optical
designer Ernest Abbe, 1840-1905). The reciprocal dispersion of crown glass
is 1/0.0170 = 58.8. It should be noted that the mediums which spread the
colors into a wider spectrum have the larger values for w and the smaller
nu-values.

TABLE III. REFRACTIVE INDEXES OF OPTICAL MATERIALS

Material Refractive Index

Cline D line F line

Ophthalmic crown glass ......... . 1.5204 1.5230 1.5293
Dense barium crown glass ....... . 1.6079 1.6110 1.618:3
Extra dense flint glass ........... . 1.7131 1.7200 1.7377
Crystalline calcium fluoride ...... . 1.4326 1.4339 1.4371
Crystalline quartz (ordinary ray) .. 1.5419 1..5442 1.5497

28-17. Deviation and Dispersion by Thin Prisms. The dispersion pro-

duced by a prism is conveniently expressed by the angular separation of rays
of different color or wavelength. For rays that traverse the prism at or near
minimum deviation, Eq. 28.6 can be used to compute the deviation angles o
if the refractive indexes n are known. The deviation angles can also be
measured on a spectrometer, provided a light source is employed which gives
sharp spectrum lines or sufficiently narrow bands of color. Prisms with re-
fracting angles of the order of 10° or smaller-so-called thin prisms-provide
a simple means for studying prismatic dispersion. For such prisms the ap-
proximations can be made that the sine of an angle is equal to the angle in
radians and that each ray passes through the prism at minimum deviation.
Then the equation reduces to n = ( om +
A)/ A, whence the deviation pro-
duced by the thin prism of angle A is
o= (n - 1) A, (28.9)
where the subscript of o is dropped in accordance with the approximation.
In order to give a physical interpretation of dispersive power, Eq. 28.9 can
be written separately for the C, D, and F Fraunhofer lines, thus:
oc = ( nc - 1) A,

on = (nn - 1) A,
OF = (nF - 1) A..
544 OPTICS Chap. 28

The overall angular separation, or dispersion, between the F and C lines is

oF - Be = (nF - 1) A - (nc - 1) A = (nF - nc) A.
When this is divided by the mean deviation, say that for D light, the result is
the dispersive power:
(nF - nc) A nF - nc
=
nn - 1
= w
·
(nn - 1) A

28-18. Achromatic Prisms. One of the early misconceptions in optics was

the belief that deviation and dispersion changed proportionately, for different
optical mediums. It was thought that if one medium produced a larger
deviation than another, its dispersion would be larger in proportion. There-
fore, it was considered impossible to obtain prisms and lenses which did not
show color-fringe effects due to dispersion. The fact is that the fractional
change in dispersive power is often much greater than the fructional change

F
c D
D
F
c

Crown Crown
Fro. 28-20. Achromatic com- FIG. 28-21. Direct-vision
bination of thin prisms, with combination of thin prisms,
deviations exaggerated with deviations exaggerated

in deviation, between one type of glass and another. As a consequence, it

is possible to design and construct achromatic combinations in which two
prisms (or lenses) of different materials are used together in such a way that
each offsets the dispersion of the other, but a net deviation remains. This
result is obtained by putting a crown glass and a flint glass prism in opposite
orientation, as shown in Fig. 20. The greater dispersion of the flint compo-
nent (of smaller prism angle) neutralizes the dispersion of the crown compo-
nent, but there remains a deviation in the "crown direction," which for these
two kinds of glass is about half the deviation that would be caused by the
crown component alone. This smaller deviation indicates the cost of ridding
the transmitted beam of its colored fringes.
In problems that involve prism combinations, confusion may arise be-
tween the "C" and "F" that stand respectively for crown and flint glass
and for the C and F Fraunhofer lines ordinarily used as the color limits for
achromatization in the visible region. Therefore, unprimed symbols are used
for angles and refractive indexes for the crown component, and primed sym-
bols for the flint component.
§ 28-19 DIRECT-VISION PRISMS 545

Given an ophthalmic crown glass prism of apex angle 10°, (a) how large a prism of
extra dense flint glass is needed to achromatize it for C and Flight? (b) What is the
deviation of the resulting combination for D light? Use the values of refractive index
given in Table III.
(a) For achromatization,

(nF' - 1) A' - (nc' - 1) A' = (nF - 1) A - (nc - 1) A;

(nF' - nc') A' = (nF - nc) A,
(1.7377 - 1.7131) A'= (1.5293 - 1.5204) X 10,
and the required apex angle of the flint prism is A' = 0.089/0.0246 = 3.62°.
(b) For deviation,
On - On' = (nn - 1) A - (nn' - 1) A'
= 0.5230 X 10 - 0.7200 X 3.62 = 2.62°,
which is the net deviation of the prism combination for D light.

28-19. Direct-vision Prisms. Another example of the combination of two

thin prisms is the so-called direct-vision assembly used in small hand spectro-
scopes for viewing spectra of light sources. The advantage of this arrange-
ment is that the instrument is pointed directly at the source, instead of at the
angle that would be required by the "standard" spectroscope, Fig. 16. The
direct-vision prism is designed so that some color near the middle of the visible
spectrum, say the yellow of the D line, is undeviated, but the spectrum will
still be adequately dispersed. This prism arrangement is shown diagram-
matically in Fig. 21; the resulting dispersion is in the "flint direction." In
such a direct-vision assembly the deviations of D light produced by the crown
and flint prisms are equal; then, since the prisms are oppositely oriented, the
deviations cancel each other.
Design a direct-vision prism combination with a 10° crown prism, as in the preceding
example, and use the same type of flint glass. (a) What is the apex angle of the flint
prism? (b) What is the resulting angular separation between red C and blue Flight?
(a) For zero net deviation of D light,
(nn' - 1) A'= (nn - 1) A,
0.7200 A' = 0.523 X 10,

and the required apex angle of the flint prism is A' = 7 .27°.
(b) The angular separation between blue and red light is

(o/ - Oc') - (oF - oc) = (n1/ - nc') A' - (nF - nc) A

= 0.0246 X 7.27 - 0.0089 X 10
= 0.090°.
Combinations of thin prisms, especially those of the achromatic type, are
analogous in many respects to the compound lenses discussed in Chap. 29.
546 OPTICS Chap. 28

PROBLEMS
1. A man 6 ft tall wishes to see himself from head to foot in a vertical mirror, when
he is standing 8 ft in front of the mirror. Determine the minimum possible vertical
dimension of the mirror. If the man moves to a position 3 ft from the mirror, would
its minimum size be different?
2. The "rear-view" mirror in an automobile is 2 ft in front of the driver's eyes and
6 ft from the rear window, which measures 12 in. vertically and 30 in. horizontally.
In order that the image of the window, as seen by the driver, may just fill the mirror,
what size mirror is required?
3. In a wall-type galvanometer, light from a small source located 50 cm directly
in front of the instrument is reflected by a mirror attached to the galvanometer coil;
the light falls on a cylindrical scale of radius 50 cm, centered on the galvanometer coil.
vVhen a current is sent through the galvanometer, the scale reading changes from O
to 25 cm. Through what angle (in degrees) does the mirror turn?
4. A light ray strikes an air-liquid surface at an angle of incidence of 60°. By what
amount is the ray bent toward the normal to the surface? The speed of light in the
liquid is 2.10 X 108 m/sec.
5. A tiny object is covered by a 1-cm plate of the most refractive flint glass, on
top of which is a 3-cm plate of rock salt. At what depth does the object appear to
be, when viewed vertically downward?
6. A layer of water covers a plate of thallium bromide-iodide that is 5.0 mm thick.
How thick does the plate appear to be when observed normally? Assume that the
eye, even though "under water," can still focus adequately.
7. A light ray, traveling in air, strikes the surface of a 10-cm plate of the least
refractive flint glass at an angle of incidence of 60°. Compute the lateral displacement
of the ray in passing completely through the glass.
8. A diamond is submerged in water. Find the greatest angle with the normal at
which a light ray in the diamond could pass through the diamond-water surface.
What is the name given to this angle?
9. A piece of fused silica is submerged in water. At what minimum angle will the
light in the silica undergo total internal reflection?
10. What percentage of the incident light passes all the way through the air-filled
absorption cell shown in the diagram, if the end plates of the cell are made of (a) thal-

II
lium bromide-iodide, (b) lithium fluoride? Assume all
the light loss is due to reflection.
11. Light of 100 units intensity is incident normally
;. -- - ;, on a transparent plate. The intensity of the light after
passing completely through the plate is 81 units. Deter-
mine the refractive index of the plate; assume all light
losses are due to reflection.
P1wBLE:11 10 12. Determine the angle of minimum deviation of
sodium light for an equilateral prism of fused silica.
13. A beam of sodium light traverses a 30°-60°-90° prism at minimum deviation.
The incident light strikes the prism face bounded by the 30° and 90° angles. Com-
pute (a) the angle of minimum deviation, and (b) the angle of incidence.
14. A light ray strikes the first surface of a 50° lithium fluoride prism at an incident
angle of 25°. Determine the angle made by the emerging ray with the normal to the
second surface of the prism.
15. (a) Evaluate the constants a and b of Eq. 28.7 for dense barium crown glass
in terms of the C and F lines. (b) Give the units of these constants. (c) Compute nn.
16. Compute the dispersive power and the nu-value for crystalline quartz.
17. A 9° prism is made from glass of nu-value 60.0 and nn = 1.660. Compute the
angular separation produced between C and F light.
 PROBLEMS 547

18. Express 505 mµ in angstroms and in microns; express 10-3 mm in microns and
in millimicrons; express 4000 A in centimeters and in millirnicrons.
19. A dense barium crown prism of apex angle 10° is to be combined with a prism
of extra dense flint glass so that there will be no resulting deviation of the D line.
(a) Find the angle of the flint prism. (b) Find the resulting angular separation
between C and F light.
20. A dense barium crown prism of apex angle 10° is to be combined with a prism
of extra dense flint glass so that the combination is achromatized for C and· F light.
(a) Find the angle of the flint prism. (b) Find the net deviation of the combination
for D light.
 29
LENSES AND CURVED
MIRRORS
Light naturally diverges from all sources, and hence some means must be
employed to bring it to a focus in optical instruments. This chapter deals
with two devices for changing the divergence or convergence of a beam of light.
One of these is the lens, which in its simplest form is a circular disk of trans-
parent material having at least one curved surface. The other is the curved
mirror-the plane mirrors discussed in Chap. 28 change the direction of a
light beam but do not affect its convergence. The curved surfaces of lenses
and mirrors are usually spherical, because spherical surfaces are generally as
satisfactory as any other and are the easiest to manufacture.
For lenses or mirrors to be perfect is a theoretical as well as a practical im-
possibility, so their inherent shortcomings should be understood. These
so-called aberrations can be controlled within limits but cannot all be removed
entirely. Therefore, the intended use of a lens or mirror should be known in
order to design it to give the best results under the prescribed working con-
ditions.

LENSES

29-1. The Action of a Lens. The immediate purpose of a lens is to alter

the convergence or divergence of a light beam that passes through it, for the
ultimate purpose of producing a suitable image of an object. Typical lens
action is illustrated in Fig. 1, vvherein a diverging beam of light directed from
left to right is rendered converging by the lens. Two ways of indicating the
action are shown: by wave fronts and by light rays. Since these are equiva-

Wave fronts

Axis

Fw. 29-1. The action of a. converging lens illustrated with wave fronts and light rays
.1548
§ 29-] THE ACTION OF A LENS 549

lent, either one may be chosen, but the usual choice is the ray representation
because it is easier to draw. In the figure, the lens is indicated by its cross-
sectional shape, with geometrical center at L. The centers of curvature C
of the two spherical surfaces determine the axis of the lens: the line that
passes through these two points. The point V, where a lens surface intersects
the axis, is called the vertex of that surface. The source of light or object
point is located at P, and the image is formed at point Q. The distance PL,
measured along the axis from the object to the lens, is called the object dis-
tance and marked p, and the distance LQ from the lens to the image is called
the image distance and marked q. The distances p and q are called conjugate
distances.
Fundamentally, a lens renders equal all the rays that diverge from the ob-
ject point P and converge at the image point Q. This equality is based on
time, for it takes all rays the same time to travel from P to Q. The ray that
travels along the axis is slowed, by its passage through the greater thickness
of glass, just enough to compensate for the longer air path of a ray that travels

1:111~r~ 1---f---l
O b j e c t ~ Image

I
I p ;, I<
I
q ------;..;

Fm. 29-2. Illustrating the focal point FIG. 29-3. Formation of image of an
and focal length of a converging lens extended object by a converging lens

through the lens near its periphery. This agrees with the concept of optical
path length, § 28-6, defined as the product of distance along the path in any
homogeneous medium and the refractive index of the medium-the optical
path lengths of all the rays going from P to Q are equal.
Often the object is so far away that the rays extending from it to the lens
are effectively parallel. The point at which the lens converges a parallel beam
directed along the lens axis is called the focal point or principal focus of the
lens. This focusing action is indicated in Fig. 2; the principal focus is marked
F and the distance LF from the lens to the focal point is called the focal
length of the lens and marked f.
The action of the lens in forming an image of an object of extended size is
indicated in Fig. 3. Each point of the image is formed by wave fronts or light
rays coming from the corresponding point of the object, as explained in con-
nection with Fig. 1, but only the rays from the ends of the object and which
pass through the center L of the lens are shown in the later figure. Obviously,
the relative sizes of object and image are determined by the object and image
distances, p and q. In Fig. 3, the image is inverted with respect to the object,
but in other situations the image may be erect, that is, oriented like the object.
There are important relationships between the conjugate distances and the
550 OPTICS Chap. 29

focal length of the lens, and between the focal length and the radii of curvature
of the lens surfaces. Before these are developed it is desirable to introduce a
classification of lenses and establish a set of sign conventions that will permit
a common procedure to be followed with lenses of all forms.
29-2. Types of Lenses. As classified by their action on a parallel beam of
light, there are two general types of lenses: converging and diverging. Each
may take several forms, as is indicated in Fig. 4. The converging lenses in
part I are designated as double-convex (radii of curvature not necessarily

Converging lenses
TIIT[ l
Diverging lenses
II
Frn. 29-4. Types of lenses, and their symbols

equal), plano-convex, and convex meniscus, in the usual order; all are thicker
at the center than at the periphery. The diverging lenses in part II are the
double concave, the plano-concave, and the concave meniscus; all are thinner
at the axis than at the periphery. Converging lenses are sometimes called
plus and diverging ones minus, because of the algebraic sign carried by their
focal lengths (see § 29-4). It is convenient in diagrams to represent lenses by
bold double-headed arrows, fashioned as shown at the right of part I for con-
verging lenses, and at the right of part II for diverging lenses. The arrows
cross the axis at the center of the lens.
29-3. Sign Conventions for Lenses and Mirrors. The successful solution
of lens and mirror problems requires careful adherence to a set of sign con-
Direction of light

+p

- --
-q
-f
+q
-p
+f
v ~cc 4 +hj I
v ~

Object, image,and Radii of Object and Angles

focal distances curvature image heights
Frn. 29-5. Schematic representation of sign conventions for lenses and mirrors

ventions for the quantities involved. Several such sets of rules are in use;
the one employed in this textbook is chosen to make the object and image
distances p and q positive for the most frequently occurring situation-a con-
verging lens forming a real image-which is illustrated by Figs. 1, 2, and 3.
The rules follow; it is recommended that they be used consistently.
(a) Consider light to be directed from left to right, and construct all figures
accordingly.
§ 29-4 FOCAL POINTS AND FOCAL LENGTHS 551

(b) Regard the object distance p as positive when the object is to the left
of the lens or mirror, and negative when it is to the right.
(c) Take the image distance q as positive when the image is to the right of
the lens or mirror, and negative when it is to the left. Since focal lengths are
in fact image distances, they follow this same rule.
(d) Regard the radius of curvature r as positive when the center of curva-
ture C is to the right of the vertex V, and negative when to the left of the
vertex. If the radius of curvature of a lens or mirror surface is measured
from the surface to the center of curvature, then measurements made with
the light are positive, and those against the light are negative. Lens surfaces
and their corresponding radii of curvature are ordinarily numbered to follow
the order in whirh light strikes the surfaces.
(e) Take transverse dimensions, such as object and image heights h, as
positive when measured upward and negative when measured downward.
(f) Consider angles as positive when the slope of the ray with respect to the
axis is positive, and vice versa.
These sign conventions and rules are illustrated schematically in Fig. 5.
29-4. Focal Points and Focal Lengths. The action of a converging lens in
bringing a parallel beam of light to a focus is illustrated in Fig. 2; the focal
length LF or f is simply the image distance q
that corresponds to an infinitely great object
distance p. Figure 6 illustrates the action of a
F
diverging lens on a parallel incident beam. In
part I the plane wave fronts striking the lens
are slowed more at the periphery than at the cen-
ter, and spherical wave fronts emerge with a
curvature that is the same as if they had origi-
nated at F, the principal focus of the lens. The
focal length f is the distance LF from lens cen-
ter to focal point, but it carries a negative sign,
according to rule (c) of the sign convention, since
F is to the left of the lens. In the ray repre- I
I
sentation of part II of the figure, the incident f--f
rays parallel to the axis are shown to be rendered Il
diverging, as if they came from point F.
The fact that the diverging light in Fig. 6 seems Fm. 29-6. Illustrating the fo-
cal point and the focal length
to come from an imaginary rather than a real of a diverging lens
source leads to the designation of such image
points or extended images as virtual images, in contrast to the real images
illustrated previously for converging lenses. It is useful, in constructing lens
diagrams, to draw as dashed lines the rays that extend backward toward
the virtual image and have no real existence-this practice facilitates the
recognition of virtual images as such.
Part II of Fig. 6 illustrates the practical value of the lens symbol in estab-
552 OPTICS Chap. 29

lishing the deviation plane for the rays and the location of the center L of the
lens. The use of lens symbols is permissible only for so-called thin lenses for
which the focal length is much greater than the maximum lens thickness;
the deviation produced by each face of a thin lens need not be considered
separately.
The focal length of a thin lens is the same regardless of whether light trav-
erses the lens from left to right or from right to left; for example, see Prob. II
of § 29-6. Thus, if a point source of light is placed on the axis at a distance f
to the left of a converging lens, Fig. 2, the rays, after refraction by the lens,
will be parallel to the axis. Likewise, converging rays directed from the right
toward the principal focus of a diverging lens, Fig. 6, will be rendered parallel
by refraction through the lens.
29-6. Equations for Lenses. The focal length of a thin lens is determined
by the radii of curvature of its two surfaces and the refractive index of the lens
material. The equation that relates these four quantities
is called the Lensmaker's Equation-it will now be de-
rived, along with the Lens Equation which relates focal
r length and conjugate distances. The derivation of these
equations is easier with wave fronts than with light rays,
...
, -----+.--.__.,---1 but the wave-front method involves a geometrical rela-
~r-s s
tionsh ip among the dimensions of a segment of a circle
that must first be established.
In Fig. 7, the circular segment has a radius of length r
and a chord of length 2 y; the distance s from chord to
FIG. 29-7. Sagitta
arc, measured along the radius that is perpendicular to
of an arc the chord, is called the sagitta of the arc. In the right-
angled triangle shown, r 2 = y 2 + (r - s)2, or y2 = 2 rs - s 2 •
For chords that are short in comparison with the radius, s 2 is negligibly small
with respect to 2 rs, and consequently the length of the sagitta can be written
as
y2 (29.1)
s =-·
2r

This relation between the sagitta and the semi-chord and radius of the arc
is useful in many optical derivations.
In order to derive these important equations a diverging spherical wave
front is considered to fall upon a thin converging lens and to be brought to
convergence after refraction. In Fig. 8 light issues from a source P on the
axis of a converging lens, which produces an image of the source at Q. The
lens has surfaces of radii r 1 and r 2 , as shown, and the distance from its axis
to its edge E is represented by y. Incident and emergent wave fronts, re-
spectively W1 and W2, are shown just touching the lens surfaces at V1 and V2,
and the extreme rays PE and EQ are drawn to the edge of the lens, intersect-
ing these wave fronts at A and B, respectively.
§ 29-15 EQUATIONS FOR LENSES .5.53

As the wave advances, the disturbance at any point on wave front TV1
moves to the corresponding point on W2 in the same time, as discussed in
§ 29-1. Consequently, at the edge of the lens the distance AEB (entirely in
air) is traversed in the same time that is required at the center to traverse
the distance V1V 2 (entirely in the lens material). If the refractive index of the
lens is nit follows that AE + EB = n(V1 V2). The extreme rays may be sup-
posed to make small angles with the lens axis, so that points A' and B' on the
wave fronts at a distance y from the axis can be substituted for A and B
without appreciable error. With this approximation, the lens surfaces and
the wave fronts can be represented by arcs having the same semi-chord y.

p
Q

~----p-----k-------q-------
Fm. 29-8. Used in deriving the lens equation

The sagittas of these arcs are marked in the figure as 81 and 82 for the lens
surfaces, and 81 and 82 for the wave fronts. When these substitutions are
made, the foregoing equation becomesA'E +EB'= n(V1V2), or81 + 81 + 8 2
+ 82 = n(81 + 82), whence
81 + 82 = (n - 1) (81 + 82).
Each of the sagittas can be expressed in terms of the semi-chord and radius
in accordance with Eq. 29.1, giving

-2y2p + -2y2q = (n - 1)
( y2
-
2 r1
+ -2y2r2 ) ,
where p = PVi may be taken as the object distance and q = V2 Q as the image
distance, and r 1 and r2 are the radii of curvature of the left and right lens sur-
faces, as shown in the figure. Because of the sign convention for radii of
curvature given in rule (d) of § 29-3, the algebraic sign of r2 must be minus.
If this change is made, and each term is divided by y 2/2, the relation becomes

-p1 + q-1 = (n - 1) (1- - -1) ·

r1 r2
(29.2)

From the definition given in § 29-4, the image distance will be equal to the
focal length when the object distance is infinitely great; that is, q = f when
554 OPTICS Chap. 29

p = oo. When these values are used, Eq. 29.2 reduces to the Lensmaker's
Equation:
!
.f
= (n - 1) (.!.r1 - r2.!.)· (29.3)

If now Eqs. 29.2 and 29.3 are combined so as to eliminate the right-hand
terms, tho Lens Equation is obtained:

.!. + ! = !. (29.4)
p q .f
Equation 29.3 makes it possible to calculate the focal length of a thin lens
in air, if the radii of curvature of the surfaces and the refractive index of
the lens material are known. Eq. 29.4 is useful in locating images formed
by thin lenses, as is illustrated in the following section. It should be re-
membered that these equations contain certain approximations, and apply
accurately only to ideally thin lenses and for rays that make small angles
with the lens axis. These equations give correct results only if the sign con-
ventions of § 29-3 are adhered to.
29-6. The Analytical Location of Images. In lens problems, the object
distance and focal length are ordinarily known, or the latter can be computed
from Eq. 29.3. The solution of the Lens Equation, Eq. 29.4, then gives the
image distance, q. If this should be a positive number, the image is located
to the right of the lens and is real. A real image is one through which the rays
actually pass after refraction by the lens; its existence can readily be demon-
strated by placing a screen of white paper or ground glass in the image plane.
If the value of q should be negative, the image is located to the left of the lens,
and is virtual, § 29-4. A virtual image is not actually formed by converging
rays, but light seems to diverge from it, as seen by the eye when appropriately
located; that is, on the other (right-hand) side of the lens. The presence of a
virtual image cannot be demonstrated by placing a screen in the image plane.
Several examples follow, to illustrate the analytical location of the images
formed by a single thin lens. Graphical solutions of these problems are given
in the following section.
I. A double-convex lens is made of crown glass of refractive index 1.52. The radii
of curvature of the lens surfaces are 11.6 and 9.45 cm. It is desired to calculate the
focal length and to determine where images will be located when the object is first
40 cm and then 8 cm from the lens.
Since r 1 = 11.6 cm, r2 = -9.45 cm, and (n - 1) = 0.52, it follows from Eq. 29.3
that

-1 = 0.52 ( - 1 - -1- ) = 0.100,

f 11.6 -9.45
whence the focal length f = 10.0 cm. For an object distance p = 40 cm, Eq. 29.4
gives 1/40 + 1/q = 1/10, from which the image distance is q = 13.3 cm. Similarly,
when p = 8.0 cm, the value of q is -40.0 cm. The first of the images is real, and the
second is virtual.
§ 29-7 THE GRAPHICAL LOCATION OF IMAGES 555

II. The curved face of a plano-concave lens of flint glass of refractive index 1.63 has
a radius of 9.45 cm. Determine the focal length of this lens, and the location of the
image of an object placed 40 cm from it.
Here r1 = oo, r 2 = 9.45 cm, and (n - 1) = 0.63; it follows that the focal length f
= -9.45/0.63 = -15.0 cm. The same result is obtained if the lens is reversed; that
is, if r1 = -9.45 and r2 = oo. For an object distance p = 40 cm, Eq. 29.4 gives
1/40 + 1/q = 1/(-15). Therefore the image distance is q = -10.9 cm and the
image is virtual.
While "objects" in lens problems are ordinarily bodies that exist, it is
possible, in certain combinations of lenses (or of lenses and mirrors), to have
what are called virtual objects. These are associated with negative values of
the object distance, and hence must be to the right of the lens. Thus, they
can only be images formed by a previous lens-one farther to the left. Virtual
objects are considered in § 29-9 under lens combinations.
29-7. The Graphical Location of Images. Graphical constructions are very
helpful in checking analytical solutions of the lens equation. Although the
same order of accuracy is not ordinarily achieved graphically, errors made in
the algebraic solution will be revealed readily. The graphical solution also
indicates whether the image is real or virtual, erect or inverted, and whether
it is enlarged or diminished with respect to the object.

Object

Lens
Fw. 29-9. Illustrating the graphical location of lens images

The procedure is illustrated in Fig. 9, wherein a number of rays are shown

extending from the head-end of the object (represented by an arrow) to vari-
ous parts of the lens, and thereafter refracted to a common point, which is the
image of that object point. Rays could be drawn similarly for every other
point of the object, and the complete result would be the image. Thus, every
portion of the lens contributes its share to the production of each part of the
image. A fragment of a broken lens, but with portions of its refracting sur-
faces in good condition, will produce a clear image, but it will not be as bright
as the image that would be produced with the entire, unbroken lens.
In order to locate a point in the image, only two rays need be drawn; the
two chosen are those easiest to construct-they are shown as heavier lines in
Fig. 9. One ray is drawn parallel to the axis from the head-end of the object
until its strikes the lens, and then through the principal focus F; this con-
struction is based on the definition of the focal point of a lens, and is illustrated
556 OPTICS Chap. 29

in Fig. 2. The other ray is drawn from the same point of the object through
the center L of the lens; it is undeviated because planes tangent to the lens
surfaces where this ray meets them, near the axis, are parallel and the lens at

Frn. 29-10. Image construction for a diverging lem

that place acts merely as a thin, parallel-sided slab of glass. The intersection
of these two rays, or of their prolongations backward, locates the image of
the head-end of the object.

F I
0 I I
f=10.0---,.J I
I
I
,f-E--------p=40.0----------q=13.3--i

=---r----~---=- -=::-:....::::- -
I
I --:::---
1 ------
1 ~~~--
1 ·-.-===---==-t----
I O
I ~p=8.0 I
I I
' - - - - - - - q =-40.0--------,----f=l o.o--1
....

O FI I
--- -- L

: : l--q=-10.
I
I
~f=-15.Q----,,..,>1I
I I
~--------p=40.0--------
Frn. 29-11. Graphical solution of lens problems in § 29-6.
All dimensions are in centimeters
§ 29-8 THE SIZE OF IAL10ES: MAGNIFICATION 5,57

The procedure is the same, whatever the type of lens and location of object.
If the construction is carried out for both ends of a straight object, the entire
image can be definitely located with four rays, as is illustrated in Fig. 10 for
an object that is inclined to the axis of a diverging lens. The incident rays
parallel to the axis appear after refraction to come from F, but actually do
not; in consequence the image I is virtual and is shown as a dotted rather
than a solid arrow.
Since objects and their images are ordinarily symmetrical with respect to
the lens axis, it is not necessary, for image location purposes, to show the
object extending both above and below the axis, as is done in Figs. 9 and 10.
Only half the arrow need be used, say the part above the axis, and this pro-
cedure is followed to simplify the drawings.
In Fig. 11 are given the graphical solutions of the illustrative problems of
§ 29-6; the upper and middle parts refer to Prob. I, and the lower part to
Prob. II. Solid lines are used only for rays that actually follow these paths;
broken lines indicate the backward projections of rays and show the position
of the -image as seen by an eye located at the right of the lens.
29-8. The Size of Images: Magnification. A lens can produce an image of
any size desired, if there is freedom in choosing the object and image distances.
The diagrams of Fig. 11 indicate that when the image distance is greater than
the object distance, the image is larger than the object, and vice versa. The
ratio of image size to object size, as measured transversely to the axis, is called
the lateral magnification of the lens and symbolized by m.

p-------

Fm. 29-12. Lateral magnification produced by a lens

In Fig. 12 a converging lens at L produces a real and inverted image of an

object. The distances of object and image from the lens are designated p and
q, as before, and their respective heights are shown as ho and hi, Since the
object and image subtend equal angles at the lens, it is evident from similar
triangles that p/q = h /hi. This leads to a simple expression for lateral mag-
0

nification, namely,
q
m- -hi--
ho - --,
(29.5)
p
where the minus sign is introduced in conformance with the sign conventions.
Although half-arrows are used for simplicity in the figure, the equation applies
equally to objects and their images which extend above and below the lens
ax1s.
558 OPTICS Chap. 29

Equation 29.5 shows that with real objects and thin, single lenses, the lateral
magnification is negative for real (inverted) images and positive for virtual
(erect) images. A value of m larger than unity indicates an enlarged (often
rather loosely called "magnified") image, while a value smaller than unity
indicates a diminished image. When m = ±1, the image has the same size
as the object, and hence object and image must be equidistant from the lens.
In addition to the lateral magnification of an object, a lens also produces a
longitudinal magnification of any object dimensions that are parallel to the
axis. The lateral and longitudinal magnifications are seldom equal, a fact
which accounts for the distortion of image space by instruments such as field
glasses and telescopes.
While the concept of lateral magnification is sufficient in many situations
-for instance, the enlargement of the motion-picture film to its image on the
screen-it fails for visual instruments such as telescopes and microscopes,
where the object or image cannot be measured because one or both may be
inaccessible or virtual. In such situations the concept of angular magnifica-
tion is used, to give a measure of the apparent magnification seen by the
observer. Angular magnification 'Y is defined as the ratio of the angles sub-
tended at the eye by the image and the object, and is conveniently expressed
as the ratio of the half-angles:
Ui
'Y = -1-lo ' (29.6)

wherein sign convention (f) of § 29-3 applies.

29-9. Lens Combinations. Most optical instruments have two or more
lenses arranged on the same axis so that the image of an object produced by
one lens serves as the object for another lens. The location and size of the
final image formed by such a lens combination can be determined by the re-
peated application of the analytical or the graphical methods described pre-
viously for single lenses. If the converging rays from the first lens to the first
image are intercepted by the second lens, that image then serves as a virtual
object for the second lens; in the calculation for the second image, the object
distance for the second lens is taken as negative.
To illustrate, two converging lenses of focal lengths 10 cm and 8 cm are aligned
axially and their separation is specified. The original object is 15 cm to the left of the
first lens and is 1 in. high. It is desired to find the position, size, and character of
the final image for three locations of the second lens: numerically 40, 36, and 14 cm
to the right of the first lens. Subscripts 1 are used to designate the quantities for the
first lens, and subscripts 2 for the second.
For the first lens, p1 = 15 cm, f 1 = 10 cm, and the lens equation shows that q1
= 30 cm. From Eq. 29.5 the image height is -2 in. This image is real, and is in-
verted and enlarged with respect to the object.
A. Separation between lenses = 40 cm. Then P2 = 40 - 30 = 10 cm, J2 = 8 cm,
and the lens equation gives q2 = 40 cm. The image height is +sin. This final image
is located 80 cm to the right of the first lens and is reai; it is erect and enlarged with
respect to the original object.
§ 29-9 LENS COMBINATIONS 559

B. Separation between lenses = 36 cm. Therefore, P2 = 36 - 30 = 6 cm; q2

= - 24 cm, and the image height is - 8 in. This image is virtual, inverted, and
enlarged.
C. Separation between lenses = 14 cm. The object distance is P2 = 14 - 30 =
-16 cm; the negative sign indicates that the object is virtual. The lens equation
gives 1/( -16) + l/q 2 = 1/8; whence q2 = +5.3 cm. The final image height is
-0.66 in. This image is real, inverted, and diminished.
The graphiral solution of this three-part prohlem is given in Fig. 13; the
upper, middle, and lower diagrams refer respectively to parts A, B, and C.
12

I
I
I
!""""-Pi= 1 5 - - - - - qi =30

I
I
I : I
i fi=lO~ :
: I
: I //
: I // I
I hV/ I I
r--pi=15-----~~-· 91=30--ip2=6~
~q2=-24~

I
I
I
I
I
I
I
I
f..---p1 = 15-,-.;----

Fw. 29-13. Graphical solution of lens combination problems of § 29-9. All dimension,;
are in centimeters
560 OPTICS Chap. 29

29-10. Thin Lenses in Contact; Diopters. Thin lenses which are in con-
tact provide a type of lens combination that is of special interest. It will be
shown, by applying the analytical methods of the preceding section, that there
is a simple relationship between the focal lengths of the individual lenses and
the focal length of the combination. Although the process is carried through
for only two lenses, it is applicable to larger numbers as long as the approxi-
mations hold that the lenses are in contact and that the combination can be
regarded as a thin lens.
For the first of the two lenses, the object distance p1, image distance qi,
and focal length J1 are related, as before:

.!_ + _!_ = l.
P1 qi f1
Since the second lens, of focal length .f2, is in contact with the first, the object
distance is p2 = - q1; hence for the second lens,
1 1 1
-q1 + q2 = I
The two equations can be added to give
1 1 1 1
P1 + q2 = Ji + -.r;.·
This equation shows that the single lens which is equivalent to these two lenses
in contact will have an image distance q2 for an object distance P1, and hence
a focal length f given by

-P11 + -q21 = -,
1
f
1 1 1
whence
f
-= -+-,
J1 f2
(29.7)

which expresses the focal length of the lens combination in terms of the focal
lengths of its components.
Equation 29.7 suggests that if each lens is characterized by the reciprocal
of its focal length, these reciprocals can simply be added to obtain the re-
sultant effect of combining lenses. This is done through the introduction
of the concept of "dioptric power." Lenses with large dioptric power con-
verge (or diverge) light more than do those with small dioptric power. The
unit is the diopter, which is defined as the reciprocal of the focal length meas-
ured in meters. Or, since the centimeter is the more usual unit in which to
express focal lengths, the power of a lens in diopters is given by
D = 100. (29.8)
fem
Hence, if Eq. 29.7 is multiplied through by 100 and each fraction is expressed
in diopters, it reduces to
§ 29-11 'J'HICK LENSES ,561

(29.9)
where D is the dioptric power of the combination of thin lenses in contact,
and D 1 and D 2 are the powers of the individual lenses. This expression holds
for diverging lenses as well, but the values off and D then carry negative
signs .. Equation 29.9 is widely used by optometrists and ophthalmologists
in eye testing and in writing prescriptions for spectacle lenses; those with
large dioptric powers are often called "strong" lenses, and those with small
dioptric powers "weak" lenses.
As an example, determine the focal length for a combination of three lenses in
contact; one lens has a dioptric pO\ver of +4 and the other lenses have focal lengths
of -30 cm and +200 cm. For these lenses

D1 = +4; D 2 = ~02o = -3.33; D3 = t~~ = 0.50.

The dioptric power of the combination is D = 4 - 3.33 + 0.50 = +1.17 diopters.

Therefore, f = 100/D = 100/1.17 = 85.5 cm (converging) is the focal length of the
combination.
The concept of dioptric power can be applied also to single refracting sur-
faces. In a plano-convex lens, for example, wherein the plane surface does
not contribute to changing the convergence or divergence of light, the dioptric
power must necessarily come from the curved surface. This fact is utilized
by the optometrist in a handy measuring instrument, called a lens measure
or dioptometer, which reads dioptric power for each surface of a lens; the
two readings are added algebraically to obtain the power of the lens in diop-
ters. This instrument actually measures the sagitta (see Fig. 7) of the arc
formed by the lens surface, by means of two fixed pins and a retractable
plunger. It can be calibrated to read diopters directly only for one refractive
index-usually that of ophthalmic crown glass, nn = 1.5230.
29-11. Thick Lenses. The lenses considered thus far were assumed to be
so thin that no appreciable errors are introduced in calculations for focal
lengths and image distances by taking the measurements to either lens surface.
The entire deviation of a ray was considered as occurring at a plane repre-
sented by the double-headed arrow symbol rather than partially at each lens
surface; the refracting properties of a thin lens are hence determined by the
location of two points, the lens center L and the principal focus F. But such
approximations are not valid for a thick lens, one which has a thickness that
cannot be neglected in comparison with the focal length.
In the most general situation, when there is a different refractive medium
on the two sides of the lens, six cardinal points are necessary in order fully to
specify a thick lens. These points all lie on the lens axis; they are: the two
focal points, F1 and F 2 ; two so-called principal points, P1 and P2; and two
so-called nodal points, N1 and N2. Their positions can be located with
respect to the lens surfaces by experiment or by the use of explicit formulas
which involve the radii of curvature, the thickness, and the refractive indexes.
5G2 OP'I'lCS Chap. 29

The nodal points can be thought of as the generalization of the lens center L;
a ray incident on the lens and directed toward one nodal point will leave the
lens in a parallel direction, but displaced as if it came from the second nodal
point. The principal points locate the intersection with the lens axis of
the two (transverse) principal planes; these are the generalization of the lens
symbol that has been used in graphical constructions, and are planes of unit
magnification. That is, the deviations at the two lens surfaces are equivalent
to a single deviation at one principal plane. When there is the same refrac-

I I
I I
I :
I I
I I
: I
I I
I I
: I
I I
:l
• I

Frn. 29-14. Location of the principal planes and principal points of some thick lenses

tive medium on both sides of the lens, as for a glass lens in air, the two nodal
points coincide with the two principal points. The locations of the principal
planes and principal points of some converging and diverging thick lenses are
indicated in Fig. 14.
The second focal point F2 of a thick lens is the point on the lens axis through
which a ray passes that was incident on the lens in a direction parallel to the
axis. The first focal point F 1 is that object point on the lens axis for which
the ray emerges parallel to the axis (see Fig. 15); F 1 could be located by
directing a parallel beam onto the lens from right to left and noting ·where it

I I
I I

Frn. 29-15. Graphical location of image formed by a thick lens

crosses the axis. The second focal length, called the equivalent focal length
is the distance from the second principal point P2 to the second focal point F2,
and hence corresponds to the focal length used in problems on thin lenses.
The first focal length is the distance from the first principal point P1 to the
first focal point F1 ; if there is the same medium on both sides of the lens,
the two focal lengths of a thick lens are equal.
Object and image distances are measured to the first and second principal
points, respectively, rather than to the lens surfaces. If the location of these
§ 29-12 SPHERICAL 1\IIRRORS 563

points is known, the images formed by a thick lens can be determined in posi-
tion and size in a manner similar to that employed for thin lenses. Equations
29.4 and 29.5 can be used, if the proper interpretations are given to the sym-
bols p, q, and f.
A plano-convex lens with an axial thickness of 3.0 cm will serve as an illustration.
One principal point is located at the vertex of the convex surface; it will be assumed
that the other principal point is 1.0 cm to the right of this vertex and that the second
focal point is located 11 cm to the right of the vertex. Determine the position and
magnification of the image of an object located 20 cm to the left of the vertex.
The thin lens equation can be used if the object and image distances and the focal
length are all measured to the appropriate principal points. Thus p = 20 cm,
f = 10 cm, and from Eq. 29.4, q = 20 cm. The image is hence located 20 cm to the
right of the second principal point, or 18 cm beyond the plane face of the lens. The
lateral magnification is found from Eq. 29.5 to be -1; the image is the same size as
the object, is real, and is inverted. The image can be located graphically by the use
of any two of the three rays shown in Fig. 1.i, by applying the definitions given above
for the focal points and the nodal points of a thick lens.

MIRRORS

29-12. Spherical Mirrors. Curved mirrors alter the divergence or con-

vergence of beams of light, as do lenses, and also deviate the light beams
through about 180°. The mirror surfaces are ordinarily spherical in figure,
and they may be classified as concave or convex, depending on whether the
reflecting surface is on the inside or outside of the spherical shell.

Concave Convex
FIG. 29-lo. Reflection of light by spherical mirrors

Figure 16 shows a concave and a convex mirror, each with a parallel beam
of light incident upon it. The line connecting the center of curvature C with
the center of the mirror surface is called the principal axis; its point of inter-
section with the mirror is called the vertex, V. The distance VG is equal to
the radius of curvature, r, of the mirror.
Most spherical mirrors used for optical purposes are comparatively flat-
that is, the dimensions of the mirror are small in comparison with the radius
of curvature of the surface. A mirror of this sort is said to have a small
aperture. With such a mirror, incident rays parallel to the principal axis
will pass through a common point F after reflection, if the mirror is concave;
or the rays will diverge after reflection as though they originated from a com-
564 OPTICS Chap. 29

mon point F, if the mirror is convex. The point F is the focal point or prin-
cipal focus of the mirror, and its distance from the mirror, VF, is the focal
length, f, of the mirror.
The principal focus of a spherical mirror is located on the principal axis half
way between the center of curvature and the mirror surface. This relation can be
proved by reference to the diagrams in Fig. 17. In each of these, an in-
cident ray AP parallel to the axis strikes the mirror at P and is reflected along

v F c

:FIG. 29-17. Used in locating the focal point of a spherical mirror

the line PF. If i is the angle of incidence between the ray and the normal PC
to the mirror surface, the angle of reflection i' at P is equal to the angle of
incidence i; also, the angle made at C by the normal PC with the axis is equal
to i and to i', since ray AP is parallel to the axis. The angle between the
reflected ray and the axis at Fis 2 i, for the same reason. For mirrors of slight
curvature, the angles i are small and the mirror surface PV may be considered
to be perpendicular to the axis. Hence, as an approximation, PV = CV tan i
= FV tan 2 i. But the tangents of small angles may be set equal to the
angles themselves, whence CV(i) = FV(2 i), or CV = 2 FV. This shows
that the focal length of a mirror is equal to half its radius of curvature, or
r
f = -·
2
29-13. Equations for Mirrors. Spherical mirrors form images of objects
much as do lenses, and for mirrors of small aperture there is a simple equation
which relates the image and object distances to the radius of curvature.
This equation can be derived by reference to Fig. 18. The object and image
are, respectively, the half-arrows labeled O and I. Their head-ends are con-

X'I
I
I
I
I
I
I

:Frn. 29-18. Image construction for a concave mirror

§ 29-14 IMAGES FOR1vfED BY SPHERICAL lvIIRRORS 565

nected by three rays: a ray parallel to the axis XV which, after reflection,
passes through the focal point F; a ray through the center of curvature C
which is reflected back upon itself, since it strikes the mirror normally; and
a ray to the vertex V which is reflected so that the incident and reflected rays
make equal angles with the axis, which of course is normal to the mirror
surface. These rays form two pairs of similar triangles. From one pair,
OXC and IYC, it follows that 7: = ~~ = ; ~; the other pair, OXV and

IYV, gives OIYX = P...

q
Hence P..
q
= ~'
r - q
or qr + pr = 2 pq. If this expres-

sion is divided by pqr, there results !p + !q = ~-

r
This derivation was purely
trigonometric, without reference to algebraic signs. When the sign con-
ventions of § 29-3 are used, this expression becomes the Mirror Equation:
1
- -
1
- =
2
--· (29.10)
p q r

Since it is shown in the preceding section that f = r /2, the mirror equation
could be written with ( -1/f) as the right-hand member. If this were done,
the result would appear so similar to the lens equation that confusion might
result. Hence the above form is recommended; the presence of r in the
equation should suggest its application to spherical mirrors.
The lateral magnification of mirrors can also be defined as the ratio of the
image height hi to the object height h and the similar triangles that relate
0,

these heights to the image and object distances in Fig. 18 are IYV and OXV.
The expression thus obtained for the lateral magnification produced by a
mirror is identical to Eq. 29.5, except for the sign of q/p:

m = h; = !1. (29.11)
ho P
Although Eqs. 29.10 and 29.11 were derived for a concave mirror, they hold
equally well for a convex mirror. Since the focal length is the image distance
that corresponds to an object distance of infinity, it follows by rule (c) of
§ 29-3 that the focal length of a concave mirror is a negative number, and that
the focal length of a convex mirror is positive.
29-14. Images Formed by Spherical Mirrors. As with lenses, the images
produced by spherical mirrors may be larger or smaller than the object and
may be either real or virtual. The images can be located analytically by
the use of Eq. 29.10, and their lateral magnifications can be computed from
Eq. 29.11; the solutions can be checked graphically by using the first two of
the construction rays discussed in the preceding section and illustrated in
Fig. 18. For the usual case, with real objects, the object distances are posi-
tive; the image distances associated with real (inverted) images are negative,
whereas those for virtual (erect) images are positive. Thus, the outstanding
566 OPTICS Chap. 29

difference between the analytical use of the lens and the mirror equations
shows up in the reversal of signs for q and f for otherwise similar pairs: con-
verging lens and concave mirror (which usually form real images), and diverg-
ing lens and convex mirror (which form only virtual images of real objects).
This similarity of lens and mirror performance is sometimes presented by treat-
ing reflection as a special case of refraction, wherein the refractive index is set
equal to -1.
Figure 19 illustrates the graphical solutions of two problems on mirrors. In each,
the object distance is 36 cm, and the radius of curvature has a numerical value of
24 cm, but the mirror at the left is concave, and that at the right is convex. For the
former, p = 36 cm, r = -24 cm, and the equation becomes 1/36 - 1/q = -2/(-24)
= 1/12, whence q = -18 cm; the image is real, inYerted, and diminished to one half
the size of the object. For the convex mirror, p = 36 cm, r = 24 cm, and the mirror
equation gives q = +9 cm; the image is virtual, erect, and diminished to one quarter
the size of the object.

0
v
V Ii lF c
I I I
I I I
I
I
I
I
-----p=36---~ I
I I I

I
J+f=-12~
I
~q=9~ I
"i+-q=-184 i--f=12~
----~p=36----
Fm. 29-19. Graphical solution of mirror problems in § 29-14. All dimensions are in
centimeters

Mirrors can be used in combination with lenses, although for practical

reasons the mirror should be the extreme (right-hand) optical element. The
computations and constructions involved are similar to those for lens combina-
tions, § 29-9.

ABERRATIONS

29-15. The Aberrations of Lenses and Mirrors. The analytical treatment

of lenses and mirrors given in this chapter involves certain approximations;
it becomes less exact for rays that make appreciable angles with the axis
and for object and image points that do not lie on the axis. The practical
result is that lenses do not form images that are as sharp as the objects. This
effect should not be attributed to lens or mirror "defects," for the spherical
surfaces may be ground and polished to high accuracy. Instead, the term
aberrations is employed to indicate the departures from perfect image forma-
tion that are inherent in spherical lenses and mirrors. There are five aber-
§ 29-16 SPHERICAL ABERRATION AND COMA 567

rations which occur even for light of a single wavelength; these monochromatic
aberrations are: spherical aberration, coma, astigmatism, curvature of field,
and distortion. Other, so-called chromatic, aberrations apply to lenses only;
they are caused by the variation of the focal length with the wavelength of
the light. Troublesome colored fringes border the images formed by a lens
that is not corrected for chromatic aberration, but such fringed images are not
produced by mirrors, since reflection angles are independent of wavelength.
Freedom from chromatic aberration is one of the great advantages of mirrors
over lenses.
29-16. Spherical Aberration and Coma. ·when the rays from an object to
a spherical mirror make large angles with the principal axis, the images formed
are confused and imperfect. For example, rays issuing from a point source
on the axis do not come to a focus at a com-
mon point; the rays reflected from the outer
parts of the mirror cross the axis nearer to
the mirror than do those reflected from the
central portion. This imperfection is called
p v
spherical aberration; it is illustrated in Fig.
20, where the symbols C and V have the
same meanings as before. The image of the
source P formed by this large-aperture mir-
ror is not localized at a point but is drawn Frn. 29-20. The spherical aberra-
out along a surface generated by the inter- tion of a mirror
secting reflected rays; a cross-section of this
surface is a line called the cailstic of the reflecting surface. The effect can be
observed on the surface of milk in a glass when illuminated obliquely by
a distant source-the glass above the milk acts as the reflector and the milk
as a screen.
It is possible to design a reflecting surface of such shape that all rays from a
definite object point will be brought to a common focus. For an object point
at infinity, such a mirror surface would be a paraboloid. This form of mirror
is often used in searchlights and automobile headlights-the lamp is placed at
the focus and the light directed toward the mirror is reflected outward in a
parallel beam. ·while the use of a paraboloidal -mirror in image-forming
instruments such as astronomical telescopes gives perfect imagery for object
points on the axis, the image deteriorates very rapidly in quality for object
points receding from the axis, and thus the useful field of view is limited.
Spherical aberration is also present with lenses. Incident light rays paral-
lel to the axis and passing through zones near its periphery are deviated more
than are the rays that pass through zones near the axis. The effect is illus-
trated in a somewhat exaggerated form in part I of Fig. 21, for a plano-convex
lens. The parallel rays from a distant source intersect at various points
along the principal axis from F' to F. If a screen is placed at these points, a
blurred image of the source will be obtained, but a screen position can be
568 OPTICS Chap. 29

found where the least blurring will occur. The figure shows this position
to be at C- - -C, where the rays constitute a circle of least confusion. The
amount of spherical aberration produced by a lens is usually measured by
the axial distance FF' between the intersections of the central and marginal
rays. In the design of lenses, spherical aberration can be reduced by a proper
choice of radii for the surfaces. For example, in part II of the figure the
plano-convex lens is reversed; the deviation of the rays is now shared by the
two refracting surfaces, and the spherical aberration is reduced considerably.
While spherical aberration deals with light rays to points on the axis of the
lens or mirror, the associated, and somewhat similar, aberration concerning
rays to points that are not on the axis is called coma, perhaps after the comet-
like distribution of light in the image of a point-source object. Coma can
also be reduced by employing proper lens shapes. A lens that is free from
both spherical aberration and coma is called an aplanatic lens.
29-17. Astigmatism and Curvature of Field. Another aberration caused
by the spherical surfaces of lenses and mirrors and producing indistinctness

II
Frn. 29-21. The spherical aberration of a lens is reduced if the deviation is distributed
between the refracting surfaces

of images is called astigmatism. Rays of light from an object point remote

from the axis that strike the reflecting or refracting surfaces obliquely do
not converge to a common image point. Figure 22 shows side and top views
in part I of several rays from the upper end of a long slender object 0, in their
passage through a converging lens. Viewed from the side, the rays converge
upon A1, and viewed from the top, they converge upon A2. This shows that
the emergent wave front possesses different curvatures in the two planes, the
curvature in the vertical plane (side view) being the greater. Part II of the
figure gives a perspective view of a pencil of the beam around ray P. It
shows that the rays in vertical planes such as ab intersect at points along a
line at A 1, while those in horizontal planes such as ac intersect along a line
at A 2 • These lines, at right angles to each other (see part I) are called focal
lines. If a screen is placed at either focal line, an elongated patch of light
will be observed, but somewhere between the two focal lines, as at C, there
will be a patch of light roughly circular in shape, again called the circle of least
confusion. This is the nearest approach to a point image that the lens pro-
duces of an astigmatic beam of light. Rays of light that are incident obliquely
upon a spherical mirror behave similarly after reflection.
§ 29-18 DISTORTION 569

The amount of astigmatism of a lens or mirror for any object point is indi-
cated by the distance between its focal lines as measured along the middle
ray from that point. Astigmatism can be corrected by the use of two lenses
with appropriate separation.
It follows from the above discussion that the best image of a two-dimen-
sional object will lie on the surface drawn through the circles of least confu-

c
0

Side view

'
I
I
A2
I
I
I
I
o~1~~~~~~~-t-t~~~~~~~-+----======,....,,======-

Top view I
Fm. 29-22. The astigmatism of a lens

sion, C (part II of Fig. 22). Such a surface, which is often called the "image
plane," will not in general be flat, but will be curved, and usually concave
toward the lens. The lack of flatness constitutes another aberration, curva-
ture of field, which is particularly objectionable in cameras and motion-
picture projectors, because the two-dimensional image must be flat so that
all portions of the picture will be sharply focused. Curvature of field can
be corrected by the use of suitable compound lenses, such as are described in
Chap. 30.
29-18. Distortion. The image produced by a lens or mirror may also be
imperfect because of the distortion produced by the variation of lateral mag-
nification in the image plane at different distances from the axis. If the

IT III
Fm. 29-23. I, Undistorted image; II, pincushion distortion; III, barrel distortion
570 OPTICS Chap. 29

magnification increases with increasing distance from the axis, so-called pin-
cushion distortion results, in which the outer portions of the image, although
sharp, are disproportionately enlarged; the opposite effect is known as
"barrel" distortion. These effects are illustrated in Fig. 23. This type of
aberration is objectionable in lenses used for aerial mapping and photoengrav-
ing, where high fidelity of reproduction of lateral dimensions is required, and
where straight lines must be imaged as straight lines. A two-lens system with
an axial aperture called a "stop" located midway between the lenses is used
to reduce distortion.
29-19. The Chromatic Aberrations. The variation of focal length of
lenses with wavelength gives rise to chromatic aberration, an effect from which
mirrors are entirely free. For two wavelengths near the extremes of the
visible spectrum, a lens would yield a red image at a different place from the

Blue
Red

Red
Blue

FIG. 29-24. The longitudinal chromatic aberration of a lens

blue image, and because the image distances are different the images are not of
the same size. The former effect is called longitudinal chromatic aberration
and is illustrated in Fig. 24. The latter effect is called lateral chromatic aber-
ration or lateral color; it is particularly objectionable in the eyepieces of tele-
scopes and microscopes.
The achromatic doublet, a combination of two simple lenses of different
types of glass placed in contact, is used to correct longitudinal chromatic
aberration; it provides such a simple and inexpensive remedy
that it is employed even in low-cost optical instruments.
A typical achromatic doublet lens is shown in cross-section
in Fig. 25; the converging component is double-convex, with
the two radii of curvature not necessarily equal, and is usu-
ally made of crown glass. The diverging component is of
Crown Flint flint glass, and is often chosen to be a plano-concave lens,
FIG. 29-25. An for simplicity of manufacture. The doublet lens is almost
achromatic always designed to have the two surfaces that are in contact
doublet lens of the same curvature, to simplify the cementing opera-
tion.
The basic condition to be satisfied by an achromatic doublet is to have its
focal length the same for the two wavelengths of light for which it is being
corrected. The color correction for visual instruments is computed on the
basis of the C and F Fraunhofer lines; the condition is f c = f P, where f is the
§ 29-19 'THE CJIJWJ11A.'l'IC A.HERRA.TION8 571

focal length of the doublet and is related to the focal lengths of the component
lenses by Eq. 29.7:
1 1 1
J = J;_ + i·
The focal lengths f 1 (crown component) and !2 (flint component) can each
be computed from the Lensmaker's Equation, Eq. 29.3, using the proper re-
fractive index for the glass and wavelength involved. This basic condition
therefore takes the following form:
1 1 1 1 1 (29.12)
f = fw + fw = J1F + /zF.
The Lensmaker's Equation, with appropriate values of n, r1, and r2, is substi-
tuted for each of the four reciprocal-focal-length terms. If all radii of curva-
ture are known, the terms in brackets can be represented by constants:
( r1! - !)
r2
= k1; (1.r3 _1)
r4
= k 2, and this simplifies at least the appearance of
the solution. Ordinarily r 2 = r 3, and often r 4 = oo, as noted above.
Photographic lenses, when ordinary blue-sensitive rather than panchro-
matic film is being used, are ordinarily corrected for the D and G' lines, with
the F line serving as the median wavelength.
The similarity between achromatic lenses and achromatic prisms (§ 28-18)
is now apparent. Both depend on the fact that the deviation produced by
flint glass is larger than by crown glass, whereas the dispersion of the former is
very much larger than that of the latter. By the use of a strong crown and a
weak flint component oppositely oriented (converging and diverging lenses),
the dispersions can be canceled, while still leaving a net deviation, or dioptric
power. Generally speaking, the focal length of an achromatic doublet is about
twice that of the crown component alone. Thus, in order to obtain color
correction, about half the power of the crown lens has been neutralized by the
weaker flint component.

PROBLEMS
1. A converging meniscus lens is made of ophthalmic crown glass, and the radii of
curvature of the lens surfaces have absolute values of 5 cm and 6 cm. Calculate the
focal length of the lens for D light.
2. A diverging meniscus lens has a focal length of -30 cm, and the radii of curva-
ture of the lens surfaces have absolute values of 10 cm and 6 cm. Calculate the
refractive index of the glass.
3. An object is placed 8 cm in front of a converging lens of focal length 12 cm and
an image is formed. (a) Locate the image with respect to the lens. (b) Characterize
the image by stating whether it is real or virtual, erect or inverted, and enlarged or
diminished.
4. A converging lens of focal length 20 cm forms a real image that is twice the size
of the object. (a) Find the object and image distances. (b) Characterize the image.
(c) Construct a diagram to scale, to check the analytical solution.
572 OPTICS Chap. 29

5. Solve Prob. 4 for the formation of a virtual image.

6. A diverging lens forms an image that is half the size of the object, when the
object distance is 20 cm. (a) Compute the focal length of the lens. (b) Make a
construction diagram to scale, to check the numerical solution.
7. Two thin lenses, one converging and of focal length 15 cm, and the other di-
verging and of focal length 10 cm, are in contact. (a) Find the focal length of the lens
combination from Eq. 29.7, and its dioptric power from Eq. 29.8. (b) Find the
dioptric powers of the individual lenses, and then that of the combination from
Eq. 29.9.
8. An object 3 cm high is located 15 cm in front of a lens having a power of +10
diopters. A lens of -5 diopters is located 40 cm to the right of the first lens. (a) Lo-
cate the image formed by light from the object that passes through both lenses.
(b) Determine the size of the final image. (c) Characterize the final image, comparing
it with the original object.
9. A lens of focal length+ 10 cm is located 5 cm to the right of an object 1 in. high;
a second lens, of focal length -15 cm, is located 20 cm beyond the first one.
(a) Locate the final image formed by this combination of two lenses. (b) Determine
the size of the final image: (c) Characterize the final image, comparing it with the
original object.
10. An object 1.5 in. high is located 10 cm in front of a lens of power -10 diopters;
a second lens, of power +6.67 diopters, is located 25 cm beyond the first lens. (a) Lo-
cate the final image formed by the two lenses. (b) Determine the size of the final
image. (c) Characterize the final image, comparing it with the original object.
11. The diverging lens of Prob. 8 is moved to a position 20 cm closer to the first lens.
Solve the problem for the new conditions.
12. A thick lens of the converging meniscus type has an axial thickness of 4 cm.
The first and second principal points are 1.5 and 0.5 cm, respectively, in front of the
vertex of the first surface of the lens (see Fig. 14), and the second focal point is 19.5 cm
to the right of that vertex. For an object located 26.5 cm to the left of the vertex
locate the image, (a) with respect to this vertex, (b) with respect to the second lens
surface.
13. The axial thickness of a piano-convex lens is 2.4 cm. One principal point is
located at the vertex of the convex surface, and the other is 0.8 cm to the right of the
vertex; the second focal point is 30 cm to the right of the vertex. An object is located
40 cm to the left of the vertex. (a) Locate the image, with respect to the vertex.
(b) How far is the image to the right of the second lens surface?
14. An object 5 cm high is located 6 in. to the left of a spherical concave mirror;
the radius of curvature of the mirror has an absolute value of 8 in. (a) Calculate the
position and size of the image formed. (b) Make a graphical solution. (c) Charac-
terize the image.
15. An object 2 cm high is located 12 in. to the left of a spherical convex mirror,
the radius of which has an absolute value of 8 in. (a) Calculate the position and size
of the image formed. (b) Make a graphical solution. (c) Characterize the image.
16. An object 1.5 cm high is located 60 cm in front of a lens of focal length +20 cm,
and a concave mirror with a radius of -12 cm is located 40 cm beyond the lens.
(a) Locate the final image, formed after light passes through the lens a second time.
(b) Determine the size of this image. (c) Characterize the final image by comparing
it with the original object. (d) Construct a diagram to scale, to check the solution of
part (a).
17. An object 3 cm high is located 20 cm in front of a lens having a power of -5
diopters, and 40 cm from a concave mirror having a radius of -20 cm that is beyond
the lens. (a) Locate the image formed by the light from the object that passes
through the lens and is reflected by the mirror, but does not pass through the lens a
 PROBLEMS 573

second time. (b) Determine the size of this image. (c) Characterize this image,
comparing it with the original object.
18. An object 3 cm high is located 20 cm in front of a lens of focal length + 15 cm.
At a distance of 15 cm beyond the lens is a concave mirror of radius -30 cm. (a) Lo-
cate the final image, formed after light from the object passes through the lens, is
reflected by the mirror, and then passes through the lens a second time. (b) Deter-
mine the size of the final image. (c) Characterize the final image, comparing it with
the original object.
19. A lens of power +6.25 diopters is located 10 cm in front of a convex mirror of
radius 20 cm. Parallel light from a distant star is refracted by the lens and is then
reflected by the mirror back through the lens. (a) Locate the final image thus
formed. (b) Characterize this image, in terms of the distant star object.
20. A converging ophthalmic crown glass lens and a diverging extra-dense flint glass
lens in contact form a converging achromatic doublet of focal length 50 cm. Find
the focal length of each component lens for C, D, and F light. Use the refractive
indexes given in Table III of Chap. 28; give answers in centimeters.
21. Continue Prob. 20 by assuming that the flint component is a plano-concave lens
and that the contacting lens surfaces have equal radii; evaluate the radius of each lens
surface~
22. Check the answers for Prob. 20 by using Eq. 29.7 for C and Flight.
 30
OPTICAL INSTRUMENTS
The most important optical instrument of all is the human eye. l\iiany
man-made optical instruments are designed to extend, amplify, or supple-
ment the abilities of the eye-these include magnifying glasses and micro-
scopes; surveyor's transits, binoculars, and other telescopes. While it is
commonly thought that the principal purpose for using these optical instru-
ments is to further magnify the images over their size as formed by the un-
aided eye, the basic improvement sought is often an increase in the ability
to see fine detail-to achieve a higher "resolving power." Another class of
optical aids includes lantern-slide and motion-picture projectors; also there
are the military instruments such as rangefinders, gunsights, and bombsights,
most of which are special-purpose adaptations of telescopes. Independent
of the properties of vision are such optical instruments as the camera and
the electron microscope.
30-1. The Eye as an Optical Instrument. The human organ of vision con-
sists of the eyeball cushioned in a bony socket, a muscular system for moving

Vitreous humor
Aqueous humor
Direction of light
Pupil ~Cone
Optic Temporal side
{~

(~Rod
Axis Nasal side
Optic nerve
Choroid
Fibers to intermediate
Sclerotic coat
cells,and from them
Suspensory ligament to optic nerve
II
Fm. 30-1. Part I, horizontal section of right eyeball; part II, schematic of rods and cones

it, and lachrymal glands and ducts for moistening its anterior portion. The
eyeball comprises a lens system for producing images of objects under obser-
vation, a retina for the reception of the images and their conversion into
nerve impulses, and the optic nerve for transmitting these impulses to the
brain to produce the sensation of vision.
574
§ ~{0-1 TIIR EYE AS AN OPTICAL INS'l'RUMEN'l' 575

The eyeball has two coatings outside of the retina, as illustrated in part I
of Fig. 1. The outermost is the white fibrous envelope ,vhich, because of
its toughness, is called the sclerotic coat. Its front portion, transparent and
more curved than the "white of the eye," is called the cornea because of its
horny texture. The inner layer or choroid is composed of numerous blood
vessels that nourish the eye, and of pigment cells that shield the retina from
stray light. This dark coat extends forward to the colored iris diaphragm
which, by involuntary muscular control, regulates the amount of light ad-
mitted through its circular aperture or p11pil.
Within the eye is the crystalline lens, which is made up of flexible layers of
different refractive indexes. The chamber in front of the lens contains a
watery substance (a weak salt solution) called the aqileous humor, and the
chamber behind it is filled with a gelatinous substance called the vitreous
humor. The dimensions and refractive indexes of the optical parts of the
eye have been the subject of a great deal of study and measurement, Rnd
have been found to vary over a considerable range, from person to person.
Also, the several optical mediums are not accurately homogeneous, and the
surfaces are not perfect spheres. Therefore, no simple optical system can
be devised to represent the eye exactly, but recourse is had to the use of
average values, in what are termed schematic eyes, which closely simulate the
optical behavior of the living eye. The data given in Table I are taken from
the schematic eye proposed by Helmholtz.

TABLE I. CoKSTANTS oF THE RELAXED EYE

Radius of
Curvat.iirc, mm Thickness Refractive
Along Axis, Index
Front Back mm

Cornea ...................... 7.7 6.8 0.5 1.376

Aqueous humor ....... ........ 3.1 1.336
Lens ......................... 10.0 6.0 3.6 1.413
Vitreous humor ............... 12.0 15.1 1.336

The eye can properly be regarded as a thick lens (see § 29-11). The six
cardinal points of the relaxed schematic eye are located as follows on the
optic axis, all distances being from the vertex of the front surface of the
cornea: First focal point, 13.7 mm to the left; second focal point, 22.8 mm to
the right, at the surface of the retina; principal points, 1.8 and 2.1 mm to the
right; nodal points, 7.0 and 7.3 mm to the right. The principal points lie
within the aqueous humor, and the nodal points are located near the back
surface of the lens, as shmvn in part I of Fig. 2. A surprisingly good approxi-
mation to the actual refractive action of the relaxed eye is given by the
reduced eye, shown in part II, which contains no lens and is filled with a single
liquid of refractive index equal to that of water, 1.333 = -!. All the refrac-
57G OP'l'ICS Chap. 30

tion occurs at the single interface of the "reduced cornea" having a radius of
5.0 mm, and the axial length is 20 mm. The first focal point is located 15
mm in front of the vertex of the "reduced cornea," and the second focal point
is 20 mm behind this vertex. The two principal points fall together at the
vertex of the cornea, and the two nodal points fall together at the center of
curvature of the cornea, ,5 mm to the right of its vertex. Calculations .such
as those on the size of retinal images are greatly
simplified by the use of the reduced eye.
The ability of the eye, not possessed by any
other optical instrument, to focus automati-
I cally upon objects at different distances from
I
I it is called accommodation. This result is
I
I accomplished through changes in the shape
I
I of the lens, chiefly in the curvature of its front
I
I surface, caused by the action of the muscles of
I
I the ciliary body as communicated to the peri-
I f1
phery of the lens by the suspensory ligament,
11 Fig. 1. It is believed that in accommodat-
ing for a near object these muscles contract
Fm. 30-2. Cardinal points of and hence relax the ligaments around the rim
the relaxed schematic and re- of the lens, thereby permitting the lens to
duced eyes thicken by its own elasticity. For an eye ac-
commodated for close work such as reading,
the radii of curvature of the front and back surfaces are approximately 6 mm
and 5.5 mm, respectively, rather than the values of 10 mm and 6 mm given
in Table I for the relaxed or unaccommodated eye, which is focused for the
parallel rays from distant objects. The faculty of accommodation diminishes
with increasing age.
The end organs of sight are the microscopic elements of the retina called
rods and cones, shown diagrammatically and greatly enlarged in part II of
Fig. 1. Some of the properties of these highly specialized receptors are

+
F1a. 30-3. To demonstrate the existence of the retinal blind spot

described in § 27-5. The most acute vision is associated with the fovea
centralis, an area of about 0.25 mm diameter where only retinal cones are
present, and no rods. The fovea centralis and the immediately surrounding
portion of the retina are of a yellowish color and have no blood vessels,
except for the finest capillaries.
No sensation of vision is produced by light which falls upon that part of
the retina knmvn as the blind spot, where the optic nerve enters the eyeball.
§ 30-2 DEFECTS OF VISION 577

The existence of the blind spot can be demonstrated by Fig. 3, as follows:

Close the left eye and look at the cross steadily with the right eye; move the
page slowly closer to and farther from the eye. When the page is about 10
in. from the eye the black disk will disappear, since its image falls entirely on
the blind spot.
It is well known that the size of the pupil of the eye increases as the ambient
illumination is decreased, typical values being a pupil diameter of 2 mm in
very bright light and 7 mm in complete darkness. This gives a range in
pupil areas and hence in retinal illumination of (7 /2) 2 or about 12, but the
range of field brightnesses for which the eye can adapt itself, if given sufficient
time, is of the order of 1,000,000 to 1-from brightest sunlight to the very
low-level illumination of a moonless night. Thus, it is apparent that there
must be some factor other than pupil area which accounts for this remarkable
ability of the eye for dark adaptation (see discussion in § 27-5); the nature of
the mechanism is an important topic of investigation in the field of physio-
logical optics.
30-2. Defects of Vision. In the passage of light through the cornea,
aqueous humor, lens, and vitreous humor to the retina of the eye, the largest

Emmetropic or
"normal" eye

11 Myopic eye

Ill Hypermetropic eye

Frn. 30-4. Correction of myopia and hypermetropia by spectacle lenses

deviation occurs at the front surface of the cornea, where there is the greatest
difference of refractive index between neighboring mediums. In the dia-
grams of Fig. 4, which show parallel beams of light entering the pupils of
schematic eyes, all the refraction is regarded for simplicity as occurring at
the surface of the cornea. Part I represents an eye which, when entirely
relaxed, forms an image of a remotely distant object point upon the retina;
it is called a normal or ernmetropic eye. Nearer object points can be focused
upon the retina when the lens system is rendered more converging through
578 OPTICS Chap. 30

accommodation. Part II of the figure represents at the left a relaxed myopic

eye; the parallel rays from a remotely distant object point focus in front of
the retina, usually because the eyeball is too long. Such an eye cannot focus
a distant object ,vithout a correcting lens, and can see only near objects dis-
tinctly. The vision is said to be near-sighted, and a diverging lens is needed
to counteract the excessive refracting power of the eye, as shown at the right.
Part III illustrates the hypermetropic eye, in which the eyeball is too short
or the refractive system is too weak, so that the intersection point of parallel
rays would be behind the retina. ·with some accommodation, distant objects
can be brought to a focus, and for this reason the vision is called far-sighted.
In viewing nearer objects, however, the accommodative function is over-
loaded, and headaches may develop. Thus, the far-sighted person should
wear correcting spectacles with converging lenses, as shown in the figure at
the right. Since the only easily noticeable symptom of hypermetropia may
be an inability to see very close objects clearly, this defect is less likely to be
detected than is myopia, in which distant objects can never be seen clearly.
The closest distance of acute vision for the normal eye of the average adult
is taken as 10 in. or 25 cm. Objects that are brought closer to the eye may
be seen more clearly, but the accommodative mechanism may be strained.
The flexibility of the crystalline lens decreases with age, and hence the so-
called near point of the eye recedes. This process is termed presbyopia and
should not be regarded as a visual defect, since its rate is about the same for
all normal eyes; it usually becomes evident between the ages of 40 and .50.
Converging lenses, such as are worn to correct farsightedness, permit an
older person to see objects clearly at the usual reading distance of 12 to 1-4: in.,
but he cannot see distant objects clearly through the same lenses. There-
fore, combination spectacles called b4ocals may be worn: the upper part of
each lens is corrected for distance vision, and the lower part has additional
converging power so as to permit reading and other close visual tasks.
If the lens system of the eye is more converging in one meridian than in
another, an object point will be imaged along two focal lines, much as in the
lens aberration of astigmatism, § 29-17. This defect of vision, also called
astigmatism, is usually due to unsymmetrical curvature of the front surface
of the cornea. It is corrected by so-called cylindrical lenses which have dif-
ferent dioptric power in different meridians, and which are so arranged that
the convergence produced by eye and spectacle lens together is the same in
all planes transverse to the optic axis.
The phorias, another class of eye troubles, result from the lack of balance
among the several muscles that hold the two eyes in place and turn them
about their spatial axes. In extreme cases these imbalances produce such
obvious effects as "cross-eyes," known technically as convergent strabismus;
in milder cases eyestrain and headaches may be caused. Phorias can often
be corrected by muscle exercises, or by wearing before the eyes weak prisms
§ 30-3 TIIE CAMERA 579

so oriented as to deflect the lines of sight from a distant object along tho
optic axes of the two eyes when they are positioned normally.
The meniscus form of lens is ordinarily used in spectacles. The basic
curvature is usually that of a 6-diopter surface-------the spherical surface which,
combined with a plane surface, gives a lens of 6 diopters power. Both con-
verging and diverging spherical lenses are manufactured by making the other
surface more or less strongly curved than the 6-diopter surface; "cylindrical"
lenses have a different curvature in different meridians, and prismatic lenses
have differing thickness across the lens. Bifocals have a roughly semicircular
section at the bottom called an "add," wherein positive dioptric power is
added to the distance prescription.
30-3. The Camera. A picture of an object can be produced on a ·photo-
graphic plate or film merely by interposing a screen having a tiny aperture.
Rays of light from each object point proceed through the aperture in straight
lines to a corresponding point on the plate, as illustrated in part I of Fig. 5.

I II
Frn. 30-5. Action of the photographic camera. I, the pinhole camera; II, the simple
box camera

This simple device is termed the pinhole camera. If the aperture is made
larger to let more light fall upon the plate, a diverging pencil will proceed
from each object point and produce a patch of light upon the plate; these
patches will overlap and yield a blurred picture. Thus picture brightness is
attained at the expense of definition. ·with a lens placed at the aperture, all
the rays from each object point that pass through the lens will be brought
to a focus upon a corresponding image point, as shown in part II of the figure.
The usual photographic camera makes use of this arrangement, and also
includes facilities for regulating the effective aperture of the lens by stops,
for employing various exposure times, and for varying the distance between
lens and plate to permit focusing objects at different distances.
The brightness of the image on the plate depends upon the lens aperture
used, that is, upon the diameter d of the stop in the diaphragm placed ttt
the lens. But the light-gathering power of the lens depends also upon its
focal length f, and the two factors can be combined into the "relative aper-
ture," d/f. However, the reciprocal of this quantity is in more common
use; it is called the !-number, focal ratio, or "speed" of the lens, and is de-
580 OPTICS Chap. 30

fined as the ratio of the focal length to the useful aperture or free diameter:
f / d. Thus, a lens set at f /8 is stopped down to a diameter one eighth of its
focal length. If the stop is changed to f /4, the free diameter is doubled
and four times as much light reaches the plate; the corresponding exposure
time is about one fourth the previous value. I
The simplest type of camera lens is of the converging
I meniscus type and is fitted with an aperture in front of
the lens in order to reduce curvature of field-a necessity
because of the flatness of the film upon which the image
is formed. This type of lens can be used at photographic
speeds up to about f /11 before spherical aberration be-
comes excessive. Chromatic aberration can be corrected
by using a cemented achromatic doublet meniscus lens; the
Fm. 30-6. The h . , b · · 1 · ·h
Tessar, a photo- c romat1c correct10n ecame 1ncreasmg y important wit
graphic anastig- the introduction of panchromatic emulsions some years ago.
mat lens lVIost fixed-focus cameras such as the inexpensive box camera
are equipped with meniscus lenses. The useful field of view
is about 40° total: 20° on each side of the axis. This angular field is much
greater than that attained in telescopes, where the total field rarely exceeds
10°, and hence photographic lenses must be more highly corrected than tele-
scope objectives.
A frequent requirement of camera lenses is high speed, in order to make
possible the photography of fast-moving objects with short exposures, or the
taking of pictures under dim lighting with longer exposures. Another re-
quirement for many uses is that the camera be reasonably light in weight.
These requirements can be met by the use of more complicated lens designs
and by utilizing lenses of short focal lengths, the lenses being made up of
several components of different optical glass to reduce the aberrations dis-
cussed in Chap. 29. One of the best camera lenses is the so-called anastigmat
lens, which is satisfactory over a field of 60° at aboutf/5, and is well corrected
for chromatic and spherical aberration, astigmatism, and curvature of field.
Figure 6 illustrates such a lens, the Tessar; each of its four components is
made from a different optical glass. The spacing of the components of such
a compound lens is very critical. Speeds off/3.5 are commonly attained in
the small amateur cameras that use 35-mm film. Lenses used in motion-
picture cameras are designed to operate at speeds up to f /2 or higher, but
the requirements for high-quality imagery are much less stringent than for
"still" cameras. A somewhat different set of requirements is placed on lenses
used in aerial cameras and in photoengraving, where distortion must be held
to a minimum, § 29-18.
30-4. The Simple Microscope or "Magnifying Glass." The extreme rays
from an object form an angle at the eye called the visual angle subtended by
that object; this angle determines the apparent size of the object. In the
 § 30-4 THE SIMPLE MICROSCOPE OR "MAGNIFYING GLASS" 581

upper part of Fig. 7, a is the visual angle subtended by the object O at the
distance d, and also by the inverted retinal image I. A similar object twice
as long and located twice as far away vrnuld subtend the same visual angle
and would produce the same retinal image; consequently, it would have the
same apparent size. As the object is
brought nearer to the eye, the visual r----aT::=::~~::r::::~J
angle and apparent size increase, but a O
practical limit is reached when d = 25 cm, I
I
the "closest distance of distinct vision," ---d=25 cm--'>!
I
§ 30-2. I
I
I
The visual angle of an object can be I
I
increased even further by bringing the
object nearer to the eye and interposing
1------- l
a converging lens to form a virtual im- 1,
age of the object farther from the eye,
~
! ------;~' I

at a distance equal to or greater than , ____

I
1 --
.,.,...-

25 cm. As shown in the lower part of .....

Fig. 7, the object O is moved to position Fw. 30-7. Visual angle is increased
O', which increases the visual angle from by the simple microscope or "magni-
a to {3. If the lens is assumed to pro- fying glass"
duce an image 1 1 of 0' at the initial
distance d, the magnification afforded by the lens over the naked eye will
be the ratio of the apparent sizes of the object in the two positions, that is,

whence (30.1)

The converging lens thus acts as a simple magnifier, and has a magnification
that can be expressed either as the ratio of image size I1 to object size O
(lateral magnification, m), or as the ratio of the angles {3 and a subtended at
the eye (angular magnification, "f).
The magnification produced by the magnifying glass can be expressed in
terms of its focal length by applying the procedures of § 29-.5 and § 29-8.
The object O may be assumed to be located at the distance p from a converg-
ing lens L of focal length .f, as shown in the upper part of Fig. 8. With p
numerically less than .f, a virtual image I 1 will be formed on the same side of
the lens asp, and at a distance q such that
1 1 1
p = .f - q.
This relation, when combined with Eqs. 30.l and 29.5, gives the magnification
~s
582 OPTICS Chap. 30

m = ~1 = -; = -q (}-i) = -J + 1.
The closest that O should be brought toward the eye is the position for
which I1 is at the distance of most distinct vision, q = -25 cm. For this
condition, a lens of focal lengthf cm produces a magnification (either angular
or lateral) of
25
1=m=-+l. (30.2)
fem

The value of magnification computed herefrom is not actually achieved

because of the practical impossibility of placing the magnifying glass at the
principal points of the eye.
Another important adjustment with a magnifier is that for which the image
is at its most remote position, namely, (minus) infinity. To obtain this image
location, the object distance p is made equal to the focal length f. Consid-
erations of object and image size and of lateral magnification then no longer
hold, but angular magnification is still pertinent. A comparison is to be
made between the apparent size of the object 01, of height h when it is 0,

located at the distance of most distinct vision and viewed with the naked eye,
and the apparent size of the image of the same object formed by the lens,
when the object, 0 2, is located at the first focal point of the lens. With ref-
erence to the lower part of Fig. 8, the tangent of the half-angle Uo subtended

r---
f I
1 lI
-----
-----F-~~
' I
I I
: i-s--p
--q=-25cm------,>,

F'
k-p=f
1-<s------ p=25 c m - - - - - . -

Fiu. 30-8. l~sed in deriving the magnification of a simple microscope

in the former case is tan u 0 = h0 /25. The half angle 1li subtended by the
virtual image located at - co is the same as that subtended by the object 02
located at a distance f cm from the lens, whence tan lli = ho/f. The angular
§ 30-5 EYEPIECES 583

magnification is obtained by applying Eq. 29.G, with the approximation that

the tangents of small angles may be replaced hy the angles:

Ui
'Y=-=--,
ho/f
U0 ho/25
25 (30.3)
whence 'Y = - .
fem

Rince a normal eye can focus sharply for any distance from 2.5 cm to in-
finity, the magnification produced by a simple microscope can have either of
the values given by Eqs. 30.2 and 30.3, or any intermediate value. For
example, the magnifications given by a lens of focal length 10 cm lie between
the extreme values of 2.5 and 3.5. Actually, the aberrations become objec-
tionable if a simple converging lens is used to give magnifications greater
than about 2. For magnifications up to 20, two- or three-element compound
lenses are used in 1Yhich chromatic and spherical aberrations, as well as coma
and distortion, are corrected. In using either simple microscopes or more
complicated optical instruments, the recommended setting is that for which
the image is formed at infinity. Although at this setting the magnification
is less than the maximum obtainable, the eye is relaxed (unaccommodated),
and there is the least possibility of inducing eyestrain.
A simple microscope of high magnification was used by the Dutch biologist
Antonie van Leeuwenhoek (1632-1723), who is sometimes, although incor-
rectly, called the inventor of the microscope. He ground accurate surfaces
on glass beads that were less than l mm in diameter. The very short focal
lengths of these tiny lenses gave magnifications up to 275, and the quality of
the surfaces was so high that he could separate details as closely spaced as
0.001 mm. He was able to observe such small objects as bacteria and red
blood corpuscles; he described details which had never before been seen by
man, and the existence of which was not confirmed until the more powerful
compound microscopes were built many years later.
30-5. Eyepieces. The magnifier used in optical instruments such as com-
pound microscopes and telescopes for viewing the real image formed by the
objective lens of the instrument is called an eyepiece or ocular. It forms a
final virtual image, very much as discussed in the preceding section, and con-
tributes to the total magnification of the instrument. Most eyepieces con-
sist of two component lenses, which are usually single lenses, but may be
compound. The front component, which is nearer the objective lens, is
called the field lens; the rear component is called the eye lens. In use, the
eye lens is located a fmv millimeters in front of the cornea of the observer's
eye-this short distance is called the eye relief.
The two most common types of oculars are the so-called Huygens eyepiece,
illustrated in Fig. 9, that is used in microscopes, and the Ramsden eyepiece
shown in Fig. 10, ordinarily used in telescopes. The Huygens or negative
584 OPTICS Chap. 30

eyepiece can be used only to examine images, and these must be located be-
tween the field lens and the eye lens. These two lenses have focal lengths in
the ratio of about 3 to 1 and are separated by a distance equal to half the
sum of the focal lengths. The Ramsden eyepiece comprises two simple
plano-convex lenses of equal focal length, separated by a distance of about
two-thirds this focal length. The Ramsden is sometimes called a positive
eyepiece since it can be used to examine real objects located in front of the
field lens, as can be seen from Fig. 10.
The use of two lenses in eyepieces reduces spherical aberration, both be-
cause of the lens orientation employed (see Fig. 29-21) and because the total
deviation of the rays is spread over four refracting surfaces. Also, it is pos-
sible by this construction to reduce the chromatic aberrations substantially.
At the field lens blue light is deviated more than is red, but the eye lens re-

Frn. 30-9. The Huygens eyepiece Frn. 30-10. The Ramsden eyepiece

ceives the first of these rays at a point nearer the center and for this reason
deviates it less than the other. With lenses of correct proportions and spac-
ing, the total deviation can be made the same for both colors, which renders
the eyepiece achromatic.
An advantage of the Ramsden type is that crosshairs can be employed;
they are located in the telescope tube at the position of the real image formed
by the objective lens of the telescope, as at I in Fig. 10. Crosshairs cannot
conveniently be used with the Huygens eyepiece, since the image formed by
the objective lens must be located between the two eyepiece lenses, as at I
in Fig. 9. This image serves as a virtual object for the field lens, which forms
the real image I'. The eye lens forms an enlarged (virtual) image of I', and
this is the final image seen by the eye. Crosshairs can be placed in the plane
of I', but they are viewed through only the eye lens and hence their image
shows different and greater aberrations than does the image formed by the
entire instrument.
In optical instruments designed to cover a wide field of view, eyepieces are
used which provide better correction of aberrations; one or both of the lenses
are compound rather than simple, or an even more complicated design is
employed, as illustrated in Fig. 21.
30-6. The Compound Microscope. vVhen magnifications greater than
those obtainable with a simple microscope or magnifier are required, it is
necessary to use a compound microscope, usually called just a microscope.
§ 30--G 'l'HE COMPOUND MICROSCOPE 585

This instrument employs two lenses or lens systems; the first, called the
objective, forms a real, inverted, enlarged image of the object under observa-
tion, and the second or eyepiece forms a virtual, erect, enlarged image of
this image, as discussed in the two preceding sections.
The construction of the images formed by the compound microscope is
illustrated in Fig. 11, wherein each lens system is represented by the usual
converging-lens symbol, and certain dimensions are exaggerated for clarity.
The object O is just outside the first focal point Fo of the objective, and this
lens produces a real image of the object at 11. This image falls between the
eyepiece and its first focal point Fe, and the eyepiece is adjusted to form
the final image 12 at about 25 cm from the eye, which is placed very close
to the eyepiece. The final image 12 is virtual and is inverted and greatly
enlarged with respect to the original object.

Eyepiece
Objective

I
I
I
I
I
fe---'>t

Frn. 30-11. Optical diagram of the compound microscope

The total, or overall, magnification M produced by the compound micro-

scope is taken as u'/u, where u' is the angle subtended at the eye by (half) the
final image, and u is the angle that would be subtended at the unaided eye by
(half) the object if placed 25 cm from the eye. This total magnification can
be evaluated as the product of the lateral magnification produced by the
objective lens and the angular magnification produced by the eyepiece. The
first term is given as - L/fo, where fo is the focal length of the objective and
L is the optical tube length: the distance between the facing focal points of
the objective and the eyepiece, as shown in Fig. 11. The distance L is
approximately 16 cm in most standard microscopes; it is taken as a positive
number in these calculations. Since image I 1 is close to the first focal point
Fe of the eyepiece and since f o is relatively much smaller than shown in the
figure, the image distance q1 can be taken as equal to L; and since the object
is so close to Fo, the object distance P1 is nearly equal to f 0 • With these
586 OPTICS Chap. 30

approximations, Eq. 29.5 gives the magnification of the objective as m =

-qi/pi = -L/fo·
The angular magnification produced by the eyepiece is (25/J. + 1), from
Eq. 30.2, where fe is the equivalent focal length of the eyepiece. Thus, the
total magnification of the microscope is given approximately by

M = _ L (25
Jo fe
+ i). (30.4)

In this equation, L and f o must be in the same units, and J. must be in cen-
timeters. The focal length of the objective lens is engraved (usually in
millimeters) on its metal sleeve, and the objective may also be marked with
its lateral magnification when used at normal working distance; eyepieces
are marked to indicate angular magnification. A microscope is usually pro-
vided with several sets of oculars and objectives which are readily interchange-
able so as to make available a wide range of overall magnifications. For
example, a laboratory microscope might be supplied with interchangeable
objectives of focal lengths 16 mm, 4 mm, and 1.8 mm. When used with a
10 X eyepiece, the total magnifications computed from Eq. 30.4 are 100,
400, and 890, respectively. The more accurate magnifications, as measured
by the manufacturer, are 100, 440, and 950.
Microscope objectives are among the most highly corrected lenses in gen-
eral use-special attention is given to the reduction of chromatic and spheri-
cal aberrations, and coma. Figure 12
shows the lens system of a 16-mm,
l 10 X objective at the left, and a 1.8-mm,
! j 95 X "oil-immersion" objective at the
I
l C::=::l right; each has achromatic doublets. The
l 62) ) tiny hemispherical lens and the meniscus
' ' - ~ ..../
lens above it are both of the aplanatic
I II type and introduce the minimum amounts
Fm. 30-12. Microscope objectives: of spherical aberration and coma (see
I, low magnification, II, high mag- § 29-16). Such high-magnification lenses
nification, oil immersion have very short working distances; that
is, they must be brought to within a
fraction of a millimeter of the object under study. To obtain the best
performance, the space between the bottom lens surface and the glass cover
slip over the object is filled with a liquid such as cedarwood oil of about the
same refractive index as the glass.
Proper illumination of the sample under observation is important in the
use of a compound microscope. A plane or a concave mirror mounted under
the microscope stage can be used to reflect light from a window, or from a
specially provided electric lamp, through the sample in the proper direction.
When the microscope is used at high magnification, it is necessary to employ
what is known as a substage condenser; this is a multi-element, short-focus lens
§ 30-6 THE COMPOUND MICROSCOPE 587

system that resembles the objective of the microscope, but is built on a much
larger scale. These two methods of microscope illumination are shown

Fw. 30-13. Micro- Frn. 30-14. :\ficroscope

scope illumination with illumination with a sub-
a concave mirror stage condenser lens

diagrammatically in Figs. 13 and 14. ·when an oil-immersion objective lens

is used, a similarly short-focus substage condenser should be employed, and
the space between the upper surface
of the top condenser lens and the glass
microscope slide should be filled with
immersion oil.
Figure 15 shows a modern com-
pound microscope, fitted with three
interchangeable objectives, and with
substage condenser in place. This par-
ticular instrument is equipped with a
rotatable stage, to facilitate the exam-
ination of objects in polarized light
(see Chap. 33). A permanent record
of the highly magnified microscope
image can be obtained by using a
special camera attachment called a
photomicrographic camera located be-
yond the eyepiece of the microscope,
as indicated at the right in Fig. 17
(although inverted from the usual op-
tical microscope position).
High-magnification microscopes en-
able an observer to recognize objects
down to about 0.0002 mm in size; Fm. 30-15. A modern microscope
such dimensions are comparable to (Courtesy of American Optical Company)
the wavelength of light. Smaller par-
ticles (to about 0.000005 mm) can be discerned by the light they scatter
588 OPTICS Chap. 30

against a dark background, if the particles are illuminated by rays suffici-

ently oblique that they do not enter the objective. When so used, the in-
strument is called an ultramicroscope. The positions of minute organisms
and colloidal particles can be observed in this way, and Brownian movements
can be studied; this method is used in the Millikan oil-drop experiment,
§ 15-13.

.
30-7. The Electron Microscope. A new and quite different type of instru-

...
...
......
.
ment forms images with electrons instead of light rays, and brings the elec-

[] D [il] ...
tron beams to a focus by deflection
in electrostatic or magnetic fields .
... With this device, useful magnifica-
...
... tions as high as 50,000 can be ob-
tained, and details can be revealed
that are one one hundredth the size
II
of the smallest details that can be
Fm. 30-16. Magnetic lenses for the elec- resolved by optical microscopes.
tron microscope: I, without pole pieces,
long-focus condenser; II, with pole pieces,
Thus, for the first time man is able
short-focus objective and projection lenses to "see" individual molecules-large
organic molecules which are made
up of many atoms. His knowledge of minute living organisms such as bac-
teria, chromosomes, and viruses has been greatly increased during the decade
or two of availability of the electron microscope.
The great resolving power obtained with the electron microscope depends
on the wave nature of fast-moving electrons and the very short wavelengths
associated therewith (§ 34-10). The laws of image formation, however, are
based upon the principles of electricity and magnetism. Electrostatic
"lenses" may be used, such as the one discussed in § 23-12 and illustrated in
Fig. 23-15. Greater practical success has been achieved with magnetic
lenses, which comprise a cylindrical-shell electromagnet giving an axially
symmetrical magnetic field; these are illustrated in part I of Fig. 16. Elec-
trons shot from a point source slant through the field; they are acted upon
by a force that is always transverse to their motion and to the field, and this
causes them to follow helical paths which converge to a point on the axis.
By the use of specially designed pole pieces, such as those shown in part II of
the figure, the convergence of the electrons can be brought about much more
rapidly, an effect which corresponds in optical terms to a shorter focal length.
The similarity in function between the lenses of an optical microscope and
those of its electron counterpart is illustrated in Fig. 17. The source is an
electron gun of the type described in § 23-12, which utilizes an accelerating
potential difference of the order of 50,000 volts. The resulting electron beam
is collimated by the "condenser," an electron lens of long focal length. The
collimated beam strikes the object 0, which must be a specially prepared, very
thin film, because of the high stopping power of matter for electrons; the
difficulty of preparing appropriate test samples is, in fact, one of the serious
§ 30-8 ASTRONOMICAL TELESCOPES 589

limitations of electron microscopy. After passing through the object, where

they are differentially absorbed by its various details, the electrons pass
through the "objective lens,'' which is of the short-focus type shown in part
II of Fig. 16, and the intermediate image 11 is formed. Further magnifica-
tion is given by the projection lens, and the final image 12 falls either on a
fluorescent screen for examination through a viewing window, or on a photo-
graphic plate, for making a permanent picture. The similarity of the over-
all system to that of a conventional photomicrographic outfit is readily seen;
in the latter, shown at the right in Fig. 17, the eyepiece has been replaced by
a projection lens, so as to form a final real image on a photographic plate
instead of a virtual image that can only be examined visually.

Electron gun
WLight source

~--+--.+-.-Condenser
lens

-....--. Objective
lens

Window

Screen
or plate --=-=====~- Photographic
plate

:F'ra. 30-17. ::VIagnetic electron microscope and its optical counterpart

The electron microscope will operate only if the entire electron path is in
a high vacuum, and this requirement involves special design considerations
and the provision of a considerable amount of accessory equipment. A
simpler and less expensive type of electron microscope employs fixed-magnet
lenses and can achieve magnifications up to 6000.
30-8. Astronomical Telescopes. A telescope, as the name implies, is an
instrument for viewing distant objects, such as landscapes, survey marks, and
stars. There are two types of astronomical telescopes: the refractor and the
reflector. The former employs only refraction-produced by lenses-in
590 OPTICS Chap. 30

forming images, and its optical diagram, Fig. 18, is quite similar to that of the
compound microscope, Fig. 11. In each instrument an objective lens forms
a real, inverted image which is viewed through an eyepiece. The major
difference between the two is in the objective lens; in the microscope this is
a lens of very short focus and small size, appropriate for studying minute
detail, and in the telescope is a lens of long focal length, often of large diameter
so as to have great light-gathering power.
Figure 18 is the optical diagram of a refracting astronomical telescope.
Since the object is located far to the left of the objective lens, the image 11
is formed by this lens at its second focal point. The eyepiece is placed far-
ther to the right, at such a distance that the final virtual image 12 will be

Objective
Eye

a ____

I
1///
/3 ,:--'~/!'
../,V'/ I
I
::;,.,-...- I I
~/ I I
,,,;:,;y I J

-9'?' : :
/.~ I I
I //?'
///
:
I
I
I
I /// I
/// : I
I /,/// I I
12 I //// I :
(atinflnity) I
///
/,;;;i/ i ,
I ///
I
I
I
I
I
I
I/// I I I
t?'/ fo I fe-""j
FIG. 30-18. Optical diagram of the refracting astronomical telescope

located somewhere between 25 cm to the left and an infinite distance away-

this is the range through which it can be viewed by the eye placed close to
the eyepiece, as shown. When the final image is located at infinity, so that
the eye is unaccommodated, the intermediate image 11 must be located ex-
actly at the first focal point of the eyepiece. Thus, the overall length of the
telescope is equal to the sum of the focal lengths of objective and eyepiece:
Cfo + fe),
The angular magnification of the telescope is the ratio of the angle (or
half-angle) subtended by the final image 12 to the angle (or half-angle) sub-
tended by the original object as seen with the unaided eye. These half-
angles are marked fJ and a, respectively, in the figure. Since the image l 1
is located at the second focal point of the objective lens, a is equal to -li/fo·
(The negative sign is a consequence of applying the sign convention for
angles: item (f) of § 29-3.) Since 11 is also located at, or very near to,
the first focal point of the eyepiece, the angle {3 is measured by (3 = li/fe·
§ 30-8 ASTRONOMICAL TELESCOPES 591

The angular magnification can hence be expressed in terms of the two focal
lengths as follows:

whence (30.5)

From this equation it is seen that large magnifications are given by telescopes
with objectives of long focal length and short-focus eyepieces. But the
eyepiece focal length cannot be made too short without encountering in-
creasing aberrations, with their attendant difficulties, and the field of view
of a telescope is also limited by the eyepiece. These two factors account for
the complicated eyepiece design used in some modern optical instruments,
such as the prism binoculars of Fig. 21.
Although the name "astronomical telescope" may suggest a large instru-
ment installed in its own dome-topped building, refracting telescopes are
made that are only a few inches in length, for example, those used in the
physics laboratory to read the scales of wall galvanometers.

Frn. 30-19. Reflecting Newtonian astronomical telescope

The reflecting telescope has a large mirror for collecting light, and uses the
conventional two-lens eyepiece. Figure 19 shows a reflecting telescope of
the Newtonian type, in which a small plane mirror (or a prism) on the axis
shifts the real image formed by the concave mirror to a position outside the
telescope tube. This image can then be viewed with an eyepiece. In the
Cassegrainian type, also widely used, the rays from the large mirror are
doubled back by a small convex hyperboloidal mirror on the telescope axis and
pass to the eyepiece through a central hole in the large mirror.
Refracting telescopes are in use that have objective lenses up to a diameter
of '10 in., and some reflecting telescopes are much larger. The 100-in. tele-
scope at Mount Wilson Observatory in California has a mirror of 42.3-ft
focal length. The largest is a 200-in. mirror at Mount Palomar Observatory,
also in California. This mirror was cast of a special Pyrex glass having a
temperature coefficient of linear expansion of only 0.00000245 per centigrade
degree. It can gather four times as much light as the 100-in. instrument
and theoretically can provide greater resolution of detail; it is expected to
penetrate hundreds of millions of light-years into space.
When a telescope is used for making a photographic record, the eyepiece
592 OPTICS Chap.30

is removed and the sensitized plate is located at I, where images are formed
by the objective. The optical arrangement is then substantially the same
as in a camera, except that the focal length is much longer; the large astronom-
ical telescopes are in fact used chiefly as cameras.

Objective lens

Frn. 30-20. The terrestrial telescope

30-9. Terrestrial Telescopes. Although the inverted images formed by the

telescopes described in the preceding section are not objectionable in astron-
omy, they are quite unsatisfactory for terrestrial use. There are three
common telescopes which give erect images. One is the so-called terrestrial
telescope, which has an erecting lens system added between the objective and

Fw. 30-21. A modern, 8-powcr binocular, with optical details shown on one side
(Courtesy of Bausch and Lomb Optical Company)

the eyepiece, as shown in Fig. 20. With a single erecting lens of focal length
f, the length of the telescope would be increased by Jf, which is the minimum
possible sum of the object and image distances of a lens. Actually, in order
§ 30-9 TERRES TRI AL TELPJSCOPES r,93

to keep aberrations to a practical minimum it is necessary to use two achro-

matic doublets, as shown in the figure; this further increases the overall
length of the telescope. The objective lens is also shown in its practical form
-an achromatic doublet. The final virtual image seen by the observer is
formed by the eyepiece, with the real image 12 serving as the object. Cross-
hairs can be located either at 11 or at 12. In the telescope diagrammed, the
erecting lenses contribute nothing to the overall magnification of the instru-
ment; images 11 and 12 are equal in size, although inverted with respect to
each other. Alternatively, the locations of the elements of the erecting lens
system can be changed so as to vary telescope magnification.
The second type of terrestrial telescope is the common prism binocular; it
comprises two telescopes mounted side by side, with the lateral separation of
the eyepiece axes adjustable to the so-called interpupillary distance between
the visual axes of the observer's eyes. The image is rendered erect by the
use of two 180°-deflection prisms in each telescope, the two prisms being
oriented at 90° to each other. 1_:hese are the Porro prisms described in

Eyepiece

Fm. 30-22. The Galilean telescope

§ 28-9; in Fig. 21 they are shown mounted in a complete instrument. One

prism turns the image right-side-up, and the other turns it left-to-right; the
two together give complete image erection. Since the light beam doubles
back in the prism train, and since no additional lenses are involved in the
erecting system, this telescope is much shorter than a terrestrial telescope of
the same magnification.
The third way to obtain erect telescopic images is the oldest historically-
it was developed by Galileo over 300 years ago. The so-called Galilean tele-
scope uses a diverging lens as the eyepiece; as shown in Fig. 22, the eyepiece
lens is placed in the converging beam coming from the objective lens, and
forms a virtual image l of its virtual object 0', the image that would be formed
by the objective lens. Equation 30.5 for magnification applies also to this
optical arrangement, but since f e for a diverging eyepiece is negative, the
magnification 'Y is positive (erect image). Also, the overall length is less than
that of the corresponding astronomical telescope by 2fe· The disadvantages
of a Galilean telescope are the small field of view and the low magnification.
It finds common use only when mounted in pairs in low-power "sport glasses"
or opera glasses, where light weight and compactness are more important
than wide field or large magnification.
594 OPTICS Chap. ;30

30-10. Projectors. A projector for lantern slides or motion-picture films

consists optically of a condensing lens for illuminating the glass slide or film,
and a projection lens for forming an enlarged image of that object upon the
screen. A typical arrangement (but of very low magnifieation) is illustrated
in Fig. 23, wherein the condenser consists of two plano-convex lense:c;, and the
projection lens is a highly correrted compound lens of 8evera1 clementr,;.
The object 0, a lantern slide or film frame, is oriented upside-down and left-
to-right in order that its image I on the Acreen will be erect. In projecting
objects that cannot be inverted, Aurh as cells of liquids for demonstrating
capillarity and electrolysis, it is customary to use an erecting prism j u8t
beyond the projection lens. The magnification, h,/ho, is equal numerically
to the ratio of image distance to object di8tance, and depends upon the foc::tl

I
Projection
lens
Light
source

Fw. 30-23. Optical system of the projector

length of the projection lens. The source of illumination may be an incan-

descent lamp with it8 filament concentrated in a small flat zone, or an arc
lamp.
As an example, a motion-picture projector in an auditorium throws a picture upon
a vertical screen distant 100 ft horizontally from the projection lens, which has a focal
length of 4.5 in. = 0.375 ft. To find the magnification produced, it is necessary first
to find the object distance for an image distance of 100 ft, using Eq. 29.4. Thus,
1/p = 1/f - 1/q = 1/0.375 - 1/100 = 1/0.376, which shows that the film is just
outside the principal focus of the lens. The magnification, by Eq. 29.5, ism = -q/p
= -100/0.376 = -266; the minus sign indicates that the image is inverted with
respect to the object.
In television receivers where a large image is desired for viewing by many
persons, the projection system of Fig. 24 may be used. The picture surface
of the cathode-ray tube is directed downward toward a spherical mirror, and
a flat mirror is used to reflect the image to a vertical screen. Light from an
object point O is collected by the spherical mirror and focused at I on the
screen, only the extreme rays being shown in the figure. The optical system
is designed for very large relative aperture (see § 30-3) in order that the screen
may be as bright as possible. For the large apertures attained, up to a
nominal value off/0.7, the spherical aberration of the mirror is excessive and
must be corrected. This can be accomplished by the method devised by the
Estonian astronomer Bernard Schmidt. A specially figured plate is placed
§ 30-10 PROJECTORS 595

in the light path, as shown. This plate or lens does not alter the focus or the
magnification of the system to an appreciable extent, but it deflects rays dif-
ferently from the different zones of the spherical mirror, in effect introducing
spherical aberration which is just equal and opposite to that of the mirror.
Schmidt plates are difficult to manufacture because
one surface is aspherical and cannot be produced by
the machine methods commonly used in generating Screen
the spherical surfaces of lenses and mirrors. There-
fore, the production of these plates in glass is very
expensive. An alternative, often used, is to manu-
Cathode-ray
facture them in plastic, forming the surfaces with tube
accurate glass master molds; plastic Schmidt
c::j:::i~t::\=i-Schmidt
plates up to 22 inches in diameter are in use. corrector plate
·with such highly corrected projection systems, 0
screen brightnesses can be obtained that are equiv-
alent to a relative aperture off/0.9 in a solid system
not pierced by the hole for the cathode-ray tube. Spherical
mirror
Schmidt plates have proved useful in correcting the
spherical aberration of reflecting astronomical tele- Fm. 30-24. Television
projection with a cor-
scopes like that shown in Fig. 19, and they can be rected spherical mirror
adapted also to refractinµ; systems. Such corrected
telescopes give good star imagery over much wider fields than previously
eovered, and make it possible to obtain photographic plates with a greatly
increased number of useful star images.

PROBLEMS
1. The greatest distance at which objects can be seen distinctly by a near-sighted
person is 2 ft. Determine the power of the spectacle lenses that will enable this person
to see remotely distant objects clearly.
2. An elderly musician is both far-sighted and presbyopic. He uses glasses having
a power of +0.75 diopter for distance vision, and by wearing "trifocals" he is able
to read comfortably (a) a musical score at 50 cm, and (b) a book at 30 cm. What
are the dioptric powers of the regions of his spectacles utilized in the two reading
operations?
3. A person with astigmatism wears for distance vision spectacle lenses with powers
of +0.50 diopter in the horizontal meridian and -1.33 diopters in the vertical me-
ridian. On a sketch similar to those of Fig. 4 indicate the location of the sharp images
formed by his relaxed eye of distant (a) horizontal lines, (b) vertical lines.
4. An f /4.5 camera lens has a maximum free diameter of 2.5 cm. (a) Calculate
the focal length of the lens. (b) If for a certain scene the proper exposure is 0.010 sec
with the f / 4.5 lens, what exposure should be used if the free lens diameter is reduced
to 1.5 cm with an iris diaphragm? (c) What is the !-number of the lens with this
diaphragm adjustment?
5. Under certain conditions a 0.01 sec exposure gives a good camera picture when
the focal ratio (!-number) is f /3.5. What is the correct focal ratio for an exposure
time of 0.04 sec, other factors being unchanged?
6. When the lens of a camera is at the farthest possible distance from the film,
596 OPTICS Chap. 30

objects at 15 ft are in focus. By placing a converging lens called a portrait attach-

ment in contact with the camera lens, nearer objects are brought into focus. If the
focal length of the added lens is 3.0 ft, what is the distance of the nearest objects that
are brought into focus?
7. A simple magnifier has a focal length of 11 cm. Compute its angular magnifica-
tion for two conditions: (a) when the image is at the distance of most distinct vision;
(b) when the image is at infinity.
8. A magnifying glass held 3 cm from a printed page gives a magnification of 4.
Determine the focal length of the lens, and characterize the image formed.
9. The most powerful simple microscope in common use gives a maximum magni-
fication of 20. Through what range of object distances can it be used as a magnifier?
10. Use the method of § 29-9 to show that the focal point of a Ramsden eyepiece
is located a distancef/4 in front of the field lens (see Fig. 10), wherefis the focal length
of each lens. Hint. Reverse the eyepiece, and have parallel rays incident upon the
eye lens.
11. A crude compound microscope is assembled in the laboratory of two spectacle
lenses, one of +25 diopters power for the objective and one of + 10 for the eyepiece.
What magnification is obtained with the instrument for an optical tube length of
16 cm?
12. The focal length of the objective lens of a compound microscope is 1.15 cm, and
the object under examination is 1.22 cm from this lens. The final virtual image is
25 cm from the eyepiece, and the eyepiece is 24 cm from the objective lens. Treat
objective and eyepiece as thin lenses. (a) Find the focal length of the eyepiece.
(b) Compute the approximate overall magnification of the microscope from Eq. 30.4.
(c) Compute the exact overall magnification by using object and image distances as
in § 29-9.
13. If the eyepiece of a compound microscope is drawn farther from the objective
lens, the final image becomes real rather than virtual, and photomicrograms can be
taken with a suitable camera, as illustrated at the right in Fig. 17. (a) How much
must the eyepiece of the microscope of Prob. 12 be moved so as to form a real image
30 cm beyond the eyepiece? (b) What is the total magnification of the equipment?
(c) Make a sketch to show the relative positions of the object, both images, and the
focal points of the two lenses.
14. Another way to form a real final image with a compound microscope is to move
the entire instrument farther from the object, so that the image formed by the objective
lens lies ahead of the front focal point of the eyepiece. (a) How much must the
microscope of Prob. 12 be moved in order to form a final image 30 cm beyond the eye-
piece? (b) What is the total magnification?
15. The overall length of a small astronomical telescope is 160 cm, and the focal
length of the objective lens is 7 times that of the eyepiece. (a) Find the focal lengths
of the objective and the eyepiece, treating them as single thin lenses. (b) Find the
angular magnification of the telescope.
16. A crude astronomical telescope is made in the laboratory from two spectacle
lenses, of powers +2 diopters and +5 diopters. (a) What is the magnification of
the telescope? (b) What is its overall length?
17. All the light that enters the objective lens of a telescope or a compound micro-
scope passes subsequently through a small axial area or disk located a short distance
beyond the eyepiece. This disk, called the "exit pupil" of the instrument, is the image
formed by the eyepiece of the objective lens aperture. The axial distance from eye
lens to exit pupil is the "eye relief" mentioned in§ 30-5. (a) Prove, by using Eqs. 29.4
and 30.5, that the magnification of an astronomical telescope is given by the ratio of
the free diameter of objective lens and the diameter of the exit pupil. (b) Compute
 PROBLEMS 597

the exit-pupil diameter and the eye relief of the telescope of Prob. 16, assuming the
free diameter of the objective lens to be 40 mm.
18. A simple Galilean telescope is to be constructed so as to give the same magnifi-
cation as the astronomical telescope of Prob. lG, and the same objective lens is to be
used. (a) '\Yhat is the focal length of the lens to be used as the eyepiece? (b) What
is the overall length of the Galilean telescope? (c) Comment on the differences be-
tween the final images formed by the two telescopes.
19. Prism binoculars are usually specified by their overall magnification and the free
diameter of their objective lenses; thus, 8X, 30 mm (abbreviated 8 X 30) binoculars
have a magnification of 8 and objective diameters of 30 mm. (a) What is the diameter
of the exit pupil of these binoculars? (b) Compute the exit-pupil diameters of the
following standard types of binoculars: 6 X 30, 7 X 50. Which of the three would
be best for night use? (See § 30-1.)
20. The field of view of a telescope is usually determined by the design of the eye-
piece, § 30-5. Good eyepiece design gives an "apparent field" of about 50°: This is
the angle 2 f3 subtended at the eye by the final image in Fig. 18. Compute the values
of the corresponding "real field" angles, 2 a, for the three binoculars of Prob. 19.
21. The telescope of Prob. 16 is to be modified for terrestrial use by the addition of
a single erecting lens having a power of +8 diopters, but the total magnification is to
be the same. (a) What is the overall length of the modified telescope? (b) Make
n, sketch showing the positions of the three lenses, their focal points, and the three
images.
22. A Newtonin,n telescope is to be converted into n, Cassegrainian telescope by
cutting a central hole in the large mirror, and replacing the small plane mirror by a
convex one. The convex mirror is to be placed 18 ft from the large mirror, of focal
length 20 ft, and is to form a real image 1 ft behind the large mirror. Calculate the
radius of the convex mirror and make a sketch of the arrangement. (For this calcula-
tion, assume that the convex mirror is spherical.)
23. Find the focal length of the projection lens that will give images 1.00 m square
on a screen 4.00 m from the lens, if the useful area of a lantern slide is 35 mm square.
'\Vhat is the distance from lantern slide to lens?
24. The lens in a home projector has a focal length of 6 in., a free diameter of 1.5 in.,
n,nd is just filled with light from its condensing lens. The lamp filament of the pro-
jector may be considered to be a circular light source of 0.50 in. diameter. The useful
area of a lantern slide measures 1 in. X 1.5 in. Treat both lenses as thin. (a) If the
image on the screen measures 2 ft X 3 ft, what is the distance from screen to projection
lens? (b) What is the distance from the projection lens to the lnntern slide?
(c) What is the focal length of the condensing lens, if it is 1.7,i in. from the projection
lens?
 31
SPECTRA, COLOR, AND
A TM O SPHERIC OPTICS
The study of spectra is a well-established field of physics, and the spec-
troscopes and spectrographs employed therein constitute a class of optical
instruments in addition to those discussed in Chap. 30. Spectra find their
practical applications in spectrochemical analysis, which supplements the
conventional chemical methods for determining the elementary and molecular
constitution of solids, liquids, and gases. Mankind throughout the centuries
has responded to color in its broader sense-beyond the sharp color bands of
spectrum lines-as evidenced in painting and the decorative arts. Color
has become increasingly important in everyday life, in industrial products,
clothing, and home furnishings, and its accurate specification is essential.
Through recent advances in the sciences of colorimetry and spectrophotom-
etry, colors can now be specified with sufficient accuracy for most practical
purposes. Certain atmospheric phenomena in optics are described in this
chapter; others have been treated in foregoing chapters.

SPECTRA
31-1. Spectroscopes and Spectrographs. The phenomenon of dispersion
described in § 28-14 is employed in prism spectroscopic instruments to sepa-
rate the component colors of a heterochromatic beam of light. This separa-
tion can also be effected by diffraction gratings, as explained in § 32-9.
The common laboratory type of spectrometer used for measuring deviation
angles is shown diagrammatically in Fig. 28-16. The process of measuring
angles of minimum deviation for more than a very few spectrum lines is
tedious and is employed only when high accuracy in determining refractive
indexes is required. In the study of spectra, visual observation of the entire
spectrum of colors is often desired, and a spectroscope is employed. This
instrument may be similar to the spectrometer but without the accurately
divided circular scale, or it may be a simple hand-held instrument built
around the "direct-vision" prism assembly of Fig. 28-21. Such a spectro-
scope produces a rather low-dispersion spectrum, and provides information
suitable for preliminary survey purposes.
Permanent photographic records are obtained through the use of a spec-
trograph. The viewing telescope of the spectrometer or spectroscope is
598
§ 31-1 SPECTROSCOPES AND SPECTROGRAPHS 599

replaced in the spectrograph by a camera so arranged that the spectrum is

brought to a focus all along the photographic film or plate. Spectrographs
are commercially available in many different forms, sizes, and varieties; the
user's selection is made on considerations such as the wavelength region to
be covered, the dispersion and resolving power to be achieved, the photo-
graphic speed desired, the type of photographic material to be used, and
whether qualitative or quantitative data are desired. In some modern
instruments the photographic process is avoided entirely, and spectrum line
intensities are measured by photocells coupled to electronic integrating and
recording circuits. Figure 1 is a schematic diagram (top view) of a conven-

~
\ ~~ Collimating
Source \\ lens
Plate
Slit
Fm. 31-1. Schematic diagram of a quartz prism spectrograph

tional prism spectrograph for use in the ultraviolet and visible spectral
regions, and hence incorporating lenses and a prism made of crystalline quartz;
Fig. 2 is a photograph (side view) of such an instrument. It consists of an

Frn. 31-2. A commercial quartz spectrograph. The plateholder has been removC'd
from the camera end and is shown leaning against the instrument
(Courtesy of Bausch and Lomb Optical Company)

equilateral prism of quartz mounted between two quartz lenses, together

with an adjustable slit and a photographic plateholder. Light coming
through the narrow slit is made parallel by the collimating lens (Fig. 1), and
after dispersion by the prism is brought to a focus upon the sensitized plate
by the camera lens. The plate, which may be slightly curved, is inclined
considerably to the dispersed beam in order to accommodate the different
focal lengths of the lens for radiations of wavelengths from 200 to 800 mµ.
The exposure time is suited to the intensity of the source, the width of the
slit, and the speed of the photographic plate. Provision may be made for
600 OPTICS Chap. 31

photographing next to the spectrum a scale calibrated in wavelengths, so

that the approximate wavelengths of individual lines can be read off directly.
Figure 3 is a set of spectrograms of five samples of a zinc alloy, in the wave-
length region 237 to 310 mµ. It can be seen that the strengths of the lead
impurity lines increase from the top to the bottom spectra. Such a series
of spectra could be used in setting up a spectrochemical method for the
determination of lead in zinc.

Frn. 31-3. Emission spectrograms of five samples of a zinc alloy, m tho ultraviolet
region, taken with the spectrograph of Fig. 2
(Courtesy of Bausch arul Lomb Optical Company)

31-2. Types of Spectra. Spectra can be classified as emission or absorp-

tion, and as continuous or discontinuous; the last-named can be sub~ivided
into line and band spectra. Emission spectra are given off by solids, liquids,
and gases, when they are raised by high temperature to a state of thermal
emission or otherwise have their atoms or molecules "excited," often by
electrical means. Absorption spectra are produced when light from a hot
source that gives a continuous spectrum-a glowing solid, for example-
passes through a cooler gas, liquid, or transparent solid; the spectrum lines
are dark against the bright continuous spectral background. In general,
line spectra are given off by atoms, while the fluted band spectra are char-
acteristic of molecules; each may occur either in emission or in absorption.
Emission spectra are illustrated in Figs. 3 and 4. They consist of bright
lines of definite wavelengths characteristic of the atoms emitting the light;
each line is an image of the slit of the spectroscope or spectrograph. They
appear in the figures as dark lines corresponding to the "negative" photo-
graphic plate images. Figure 4 shows the bright-line spectra of three ele-
ments over the range of the visible spectrum and the near ultraviolet; other
lines may be found in the deep ultraviolet and infrared regions. Sodium
vapor yields two bright lines in the yellow part of the spectrum, but they are
so close together that they appear as one line unless the slit is very narrow
and the resolving power is sufficiently high. Mercury vapor yields several
bright lines, the most conspicuous being in the green and blue regions. A
continuous spectrum such as that of Fig. 27-5 is an uninterrupted series of
images of the illuminated slit.
§ 31-2 TYPES OF SPECTRA 601

The absorption spectra of gases consist of dark lines that are identical in
wavelength to the corresponding bright lines of the emission spectra of the
same substances. The production of a dark-line spectrum is illustrated by
a well-known experiment showing the reversal of the sodium lines. An arc
lamp is arranged to project an intense beam of light upon a prism, the beam
being directed through a cloud of glowing sodium vapor produced by heating
sodium chloride in a Bunsen flame. A narrow slit is placed near the prism in
the light path. In conducting the experiment, the glowing sodium vapor is
first used alone; the resulting spectrum consists of the two bright yellow lines
characteristic of sodium, Fig. 4. Upon starting the arc lamp and shining its
rays through the sodium vapor, a continuous spectrum is formed which has
two dark lines at exactly the positions previously occupied by the bright
lines. This reversal of the lines is explained by supposing that the sodium
vapor, which emits these particular lines when excited, responds to the cor-
responding frequencies of vibrations, and absorbs the energy of these par-

Sodium vapor

Hydrogen gas
11111111111 I I

Mercury vapor
II I I 1111 II II I I II 11 II
300 350 400 450 500 550 600 650- 700 750
Wavelength, millimicrons
FIG. 31-4. Bright-line spectra in the visible and near ultraviolet spectral regions

ticular vibrations from the beam proceeding from the arc. This vapor,
considered as a source, has low luminosity as compared with the arc because
of its relative coolness, and hence the lines that it produces alone are less
bright than the rest of the arc spectrum. An explanation in terms of the
quantum theory of radiation is given in Chap. 34.
The Fraunhofer lines discussed in § 28-15 serve as another example of an
absorption spectrum. The source of the continuous spectrum is the sun
itself, its surface temperature being about 6000°K; absorption at definite
wavelengths occurs in the relatively cooler gases of the sun's atmosphere.
The analysis of such spectra, and comparisons with emission spectra pro-
duced in the laboratory, have yielded much information regarding the chemi-
cal constitution of the sun and the stars; the same elements exist there as
are found upon the earth. Helium was in fact first discovered in the atmos-
phere of the sun, and its name was derived from the Greek word Helios: sun.
Some stars give absorption spectra and some emission spectra. In the
spectra of certain stars, the lines are displaced slightly from their normal posi-
tions. This displacement indicates an apparent change in frequency, as
explained by Doppler's Principle, § 26-15, and reveals the fact that there is
602 OPTICS Chap. 31

relative motion between the earth and the star under observation. A shift
toward the red end of the spectrum, for example, represents an apparent
increase in wavelength or lowering of frequency and shows that the star and
earth are receding from each other. It is from measurements of this kind
that the motion of the solar system in the galaxy was first computed.
The absorption spectrum of a gas may also take the form of fluted bands,
which under sufficient dispersion are found to consist of closely spaced dark
lines arranged in an orderly manner, as shown in Fig. 5. In the figure, the
lines are lighter than the background, as they are on the photographic plate.
Band spectra are emitted only by molecules, whereas line spectra are emitted
by single or uncombined atoms. From the study of band spectra much has
been learned about the structure of molecules and the forces acting ,,rithin
them. Bands occasionally appear in bright-line or emission spectra, but less
commonly than in absorption spectra because the intense excitation used in

Frn. 31-5. Absorption bands of the oxygen molecule in the vacuum ultravioleL region,
179 to 187 mµ. The pressure of the oxygen in the absorbing path is greater for the upper
spectrum
(Courtesy of Professors J-J. P. Knauss and S.S. Ballard)

producing emission spectra is likely to break up the molecules into their

component parts.
The absorption spectra of liquids and transparent solids do not usually
have the sharp lines and bands that characterize gases, but have broader,
structureless regions of absorption which, although characteristic of the
absorbing substance, are less accurately defined.
31-3. Spectrum Analysis and Spectrochemical Analysis. The light sources
used in spectroscopy are similar to those discussed in § 27-2 and include
flames, electric arcs and sparks, and electric discharges in gases under reduced
pressure. The spectra of some elements, especially the alkali metals and the
alkaline earths, can be excited by introducing their salts into a Bunsen flame,
either as solid powders or as solutions vaporized into the gas supply; the
flame glows with a characteristic color which, upon examination with a
spectroscope, is found to consist of the spectrum lines of one or more of the
constituents of the material added. A more vigorous type of excitation is
obtained with electric sources such as the arc or spark, in which the electrodes
may be rods of the metallic samples under spectroscopic examination. Non-
conductors can be placed in powder or liquid form into a cavity within the
§ 31-4 SPECTROSCOPY G03

lower (vertical) electrode, a rod of carbon or metal, and the resulting spec-
trum will show emission from both the sample and the electrode material, and
perhaps also from the surrounding air. The gas discharge lamps (mercury
vapor, sodium vapor, neon, etc.) described in § 27-2 may be used to show the
characteristic spectrum of the gas or vapor enclosed. The spectra of other
gases can be obtained by putting them at reduced pressure into a so-called
spectrum tube of glass or quartz, and exciting them by applying a high elec-
tric potential.
The analytical study of the characteristic spectra of substances is called
spectrum analysis; it was carried out vigorously during the period 1900 to
1930, and made major contributions to our knowledge of the structure of
atoms and also of molecules, especially those of the diatomic gases.
The spectroscope is also used for the analysis of materials, in the rapidly
growing field called spectrochemical analysis. ,Vith flame or arc excitation
and a visual spectroscope it is possible to identify two or three dozen of
the elements very rapidly. With recording spectrographs which cover the
ultraviolet as well as the visible region, and the technique of photographic
photometry, many elements can be identified and the analysis can be made
quantitative rather than merely qualitative. It has been found that spec-
trochemical methods complement the standard chemical methods of analysis
in a most fortunate fashion. That is, some of the elements that are difficult
to analyze chemically can be handled with ease by the spectrograph, and,
further, the spectrochemical methods are accurate with very small concen-
trations of the unknown, whereas the chemical methods are more accurate
at larger concentrations. Spectroscopy is now used in industry to test the
purity of raw materials and of finished products, and to serve as a production-
control technique. An important feature of routine spectrochemical analysis
is the high speed with which results are obtained.
The use of spectroscopy as an analytical tool for compounds is not as
advanced as it is for elements, but since about the year 1940 great strides
have been made in this direction. Absorption spectra are employed, and the
"unknowns" are usually in the form of liquids or gases. Certain large organic
molecules, for example vitamin A, have characteristic ultraviolet absorption
spectra around which quantitative analytical methods can be developed.
Many other types of molecules, again chiefly organic, have characteristic
absorption spectra that lie in the infrared region, in the wavelength range 2
to 40 µ. The hydrocarbons, and industrial materials such as synthetic rub-
ber and petroleum products, can be handled by these techniques, and infrared
instruments have been developed especially for such analyses, both qualita-
tive and semi-quantitative.
31-4. Spectroscopy in the Ultraviolet and Infrared Regions. A conven-
tional spectrograph such as that of Fig. 2 is ordinarily equipped with quartz
optics so as to cover both the visible and the ultraviolet spectral regions;
glass is opaque for wavelengths shorter than about 350 mµ, but crystalline
604 OPTICS Chap. 31

quartz is transparent to at least 185 mµ. At the latter wavelength air itself
becomes opaque in long paths, and so a vacuum spectrograph, from which
the air can be exhausted, must be employed if shorter wavelengths are to be
photographed. Crystalline calcium or lithium fluoride prisms can be used
almost to 100 mµ. For still shorter wavelengths, diffraction gratings must
be used rather than prisms as dispersing elements; spectra through the
ultraviolet and into the x-ray region can be studied with vacuum grating
spectrographs. Ordinary photographic plates are more sensitive in the near
ultraviolet than they are in the visible region, but they lose their sensitivity
at a short-wavelength limit of about 230 mµ. Special types with very thin
gelatin, or ordinary plates treated with a fluorescent oil, are used at shorter
wavelengths and through the vacuum ultraviolet region.
Electric arc and spark sources produce spectra which lie in both the visible
and the ultraviolet regions, but high-potential sparks must be used to obtain
the shorter-wavelength spectra of the ionized atoms. Electrical discharges
in hydrogen or helium gas, or high-pressure mercury lamps, give the con-
tinuous spectra needed for ultraviolet absorption spectroscopy.
Although glass cannot be used in the infrared region beyond about 2.5 µ,
there are several crystalline materials which have wide transmission regions
at longer wavelengths. Prisms of natural or synthetic rock salt (sodium
chloride) can be used to 17 µ, those of potassium bromide to 30 µ, and those
of the newer optical materials such as thallium bromide-iodide extend this
range farther into the infrared. Since the infrared region is very broad-
a frequency range of 5 to 6 octaves as compared with the single octaves of
the accessible ultraviolet and the visible regions-the use of achromatic lenses
is not practical, and mirrors are required.
Infrared spectra are almost always studied in absorption rather than in
emission, because of the very low intensity of the emission spectra. The
sources of continuous spectra are low-temperature incandescent-filament
lamps, or heated electrical resistors that have a radiation maximum at a
wavelength of about 2 µ. The characteristics of available sources limit the
useful analytical infrared region to wavelengths shorter than about 50 µ-
for longer wavelengths the sources are weak, and the scattered light (espe-
cially of shorter wavelength) is very troublesome. Photography can be used
only to a wavelength of 1.2 µ, and some photoconductive cells are sensitive to
about 7 µ, so the usual infrared detectors are thermocouples, thermopiles, or
bolometers of the types discussed in § 27-8.

COLOR

31-6. Color and Colorimetry. Color is a concept that is seemingly under-

stood by persons with normal vision, but it is difficult to describe in scien-
tific terms. Though great advances have been made during the last thirty
or forty years in the field of color measurement-colorimetry, there are still
~ome 1msolved problems. The distribution of radiant energy with wave-
§ 31-5 COLOR AND COLORIMETRY 605

length in a light beam can be measured readily and accurately with modern
spectrophotometers (§ 31-10), but does not determine uniquely the subjec-
tive interpretation of an observer-that is, "what color it seems to him to
be." Again, both physical and physiological-psychological considerations
are involved, as in sound and hearing. An important distinction is that
between the color of a beam of light and the color of an object. It is the
former which is most easily specified; the latter depends both on the nature
of the surface of the object and the spectral distribution of the light that
illuminates it; see § 31-8.
Color is ordinarily described in terms of three characteristics: dominant
wavelength, purity, and luminous flux. These three purely physical quan-
tities give rise to the perceptual sensations of hue, saturation, and bright-
ness. Dominant wavelength and purity, taken together, comprise what is
called the chromaticity of the light. The dominant wavelength and related
hue are described by comparison with spectral colors and represent what is
commonly regarded as "color"-red, or yellow, or green, for example. A
pure spectral color has the highest possible purity; the addition of white light
"dilutes" the color and reduces its purity and associated saturation. These
two concepts, dominant wavelength and purity, play the main part in the
modern method of color specification that is described in § 31-9. The con-
cepts of luminous flux and of brightness, discussed in § 27-6, apply both to
colors and to neutral tints.
For many years scientists have attempted to understand the physics and
physiology of the color sensitivity of the eye and to formulate a scientific
theory of color vision, but a complete explanation of all observed phenomena
has not yet been given. Studies have been made not only on normal vision,
but also on eyes that are color deficient or "color blind" to a greater or less
extent. It is estimated that about 8 per cent of the male population has
anomalous color vision to some degree, and about one-half of one per cent
of the females, although about 14 per cent of all females are genetic "carriers"
of color blindness even though they do not exhibit it themselves. The two
common forms of color deficiency are protanopia, sometimes called "red
blindness," and deuteranopia, inaccurately called "green blindness." In
both forms there tends to be a confusion of colors at the long wavelength end
of the spectrum-that is, the colors that appear red, orange, yellow, and
green in normal color vision. The luminosity curve (Fig. 27-6) of the pro-
tanope differs from that of the normal person by being much lower for wave-
lengths longer than about 550 mµ; this means that the yellow, orange, and
especially the red colors are much less bright than normal. The deuteranope
has a luminosity curve practically identical with that of a normal person,
although he still confuses the long-wavelength colors. Many cases of color
deficiency are not nearly as pronounced, but are intermediate between these
extremes and normal color vision; the persons so afflicted may be unaware
of their visual anomaly.
606 OPTICS Chap, 31

31-6. Color Mixture by Addition. When a spectrum is produced by pass-

ing white light through a prism, and then all the colors of the spectrum are
caused to overlap, their effects are added together and produce the sensation
of white light. This result can be observed by allowing the composite beam
to enter the eye directly, or by projecting it upon a white screen. It is also
found that by combining only the yellow and blue-violet portions of the spec-
trum, a white light is produced which the eye cannot distinguish from the
other. There are several pairs of colors that yield white light when added;
they are called complementary colors. Yellow and blue-violet are comple-
mentary, as are also red and blue-green. Furthermore, the light from a
sodium vapor lamp appears much like that of illuminating gas, but the former
has primarily the yellow spectrum lines, whereas the latter has the continuous
spectrum from heated carbon particles. Evidently, a particular color sensa-
tion can be produced in a variety of ways, and the
G composition of the radiant beam cannot be deter-
mined by its appearance but only by an analysis
of its spectrum.
It is possible to match many colors by the
additive mixture of red, green, and blue lights in
suitable proportions. Consequently, these three
B Purple R are called the additive primary colors. This
(magenta) process can be demonstrated by the Maxwell
Fw. 31-6. Maxwell color tri-
color triangle, Fig. 6. · A white triangular surface
angle. The additive prima- is illuminated by overlapping light beams of the
ries are indicated by R, G, primary colors just mentioned, issuing from
and B colored lamps at the corners. The illumination
due to the red lamp at R is regarded as varying
from a value of 100 at that corner to zero all along the opposite side BG;
similarly the green and the blue each vary from 100 at the corresponding
corner to zero all along the opposite side. At the corners of the triangle
appear the saturated hues of the primary colors. Along the upper sides
are the colors in the spectrum order, with yellow midway between R and G,
and green-blue (or blue-green) midway between G and B. Along the base
BR, where only blue and red are present, are the hues of purple, which
are not present in the spectrum. It will be seen that the complement of
each primary color appears at the center of the opposite side. The central
point, W, where the three primary colors are added in equal proportions, is
white. Lighter tints appear throughout the central part of the figure,
where the saturation is reduced by the effective addition of white light.
31-7. Color Mixture by Subtraction. The light incident upon a non-
luminous object is partly reflected and partly absorbed, and the remainder is
transmitted. The apparent color of the object depends upon the composition
of the incident light and the extent to \vhich the various component colors
are reflerted, absorbed, and transmitted.
 MAGENTA

GREEN

ADDITIVE PRLMARIES SUBTRACTIVE PRIMARIES

BLUE YELLOW
/ .~
: "":
• '1

t~
GREEN MAGENTA

RED CYA N

Courtesy Modern Collrgr Physus 2nd Erl., 1953 - TI. E. W hite

§ 31-8 COLOR BY REFLECTED LIGJIT; ME'l'ALLIC COLORS G07

A sheet of ordinary glass held in front of a source of white light appears

colorless by light transmitted through it, because it transmits practically all
of the light falling upon it. On the other hand, a sheet of red glass appears
red because red, orange, and some yellow are transmitted, while green, blue,
and violet are subtracted by absorption. Similarly, a blue transmitting
medium absorbs red, orange, yellow, and green, and a green glass absorbs red
and orange, blue and violet. Light from the three spectral regions corre-
sponding to the absorptions just listed is seen as a blue-green called cyan, a
yellow, and a red-purple called magenta, respectively. Cyan, yellow, and
magenta, called the subtractive primary colors, are the complements of the
additive primaries, as shown in Fig. 6; their superposition gives black.
The subtractive primaries cover much broader regions of the spectrum than
do the additive primaries, and hence are more complex colors. The additive
and subtractive primaries, and their simple combinations, are illustrated in
the color plate.
The property of absorbing certain colors and transmitting others is called
selective absorption. Ordinary uncolored glass, which transmits the entire
range of wavelengths of the visible spectrum, is opaque to most ultraviolet
and infrared radiation, and thus exhibits selective absorption outside of the
visible range. The glass of a greenhouse transmits radiant energy over the
visible and near-infrared ranges to the plants within, where it is largely con-
verted into heat and chemical energy. The radiation from the interior is
principally in the far-infrared region (so-called radiant heat), because the
surface temperature is so much lower than that of the sun, see Fig. 34-2; and
since the glass is opaque at these longer wavelengths, the interior of the
enclosure remains warm.
The light given out by the ordinary incandescent electric lamp is redder
than sunlight; that is, the spectrum is relatively more intense toward the red
than toward the violet end as compared with that of sunlight. The incan-
descent "daylight" lamp has an envelope of blue glass which, by selective
absorption, removes some of the radiation from the red end of the spectrum,
and the transmitted light, although slightly dimmed, conforms more closely
to sunlight in the relative proportions of the component colors.
The colors of paints and inks depend largely on selective absorption by
tiny particles embedded in a colorless, transparent medium, and on color mix-
ture by the subtractive process. For example, if there are yellow particles
that absorb blue and violet but transmit the rest of the spectrum, combined
with cyan particles that absorb only red and orange, then the only colors
transmitted by the paint are green and yellow-other colors are absorbed
by one pigment or the other.
31-8. Color by Reflected Light; Metallic Colors. Opaque bodies are seen
by light reflected diffusely from them. An object may be examined in white
light containing all wavelengths of the visible spectrum: If the object reflects
no light, it appears black; if it reflects all of the light incident upon it, it
G08 OPTICS Chap. 31

appears white; and if it reflects only part of the light, but reflects all wave-
lengths in equal proportions, it appears gray. Most objects that appear
colored when viewed by reflected light do so because of selective absorption,
as discussed in the preceding section. Reflection is not strictly a surface
phenomenon, for the light seems to penetrate a short distance beneath the
surface before reflection. A surface painted blue-green absorbs the com-
plementary red and neighboring colors, and reflects blue and green (see Fig.
6). Similarly, a surface painted yellow absorbs blue and violet and reflects
all the colors from red to green. If blue-green and yellow paints are mixed,
the only color reflected by both is green; the blue-green paint absorbs the
red, orange, and yellow, and the yellow paint absorbs the blue and violet.
This result is quite different from that produced by the additive mixing of
yellow and blue light mentioned in § 31-6.
An object exhibits its true color when examined by light that contains
all the wavelengths of the visible spectrum. If certain colors are absent
from the incident light, the apparent color of the object may be quite dif-
ferent from its true color. Thus, dark blue cloth appears nearly black when
examined under an incandescent electric lamp because of the deficiency of
blue in the incident light. Since the light from the mercury vapor lamp con-
tains no red radiations, a red object examined in its light appears black.
For many types of lighting this is an objectionable feature, and for this
reason mercury vapor lamps are sometimes used in combination with neon
lamps, the red light from the glowing neon offsetting the deficiency of red
light in the glowing mercury vapor.
A few instances occur of bodies that exhibit a surface color, like the luster
of metallic surfaces, that is apparently due to selective reflection. Gold
exhibits a yellow surface color, but this is not because the other colors are
absorbed; in fact, the surface color itself is the one that is absorbed most
strongly and re-emitted by a resonance effect. This can be shown by pass-
ing white light through a sheet of very thin gold foil; since the yellow and
orange have been absorbed, it will be found that the transmitted light is
bluish green.
31-9. Modem Color Specification. There are no three real colors that can
be combined in suitable proportions so as to match all possible colors. Red,
green, and blue do as well as any combination, however, and hence are usu-
ally chosen as the "primary" colors, § 31-6. The particular red, green, and
blue colors to be used were standardized by the International Commission on
Illumination in 1931; they can be defined mathematically in terms of spec-
trum colors, but they are not additive mixtures of spectrum colors and hence
are not real colors. Nevertheless, they form a proper basis for color specifi-
cation. The amounts of these three primaries that must be combined in
order to match a given color are termed the "tristimulus values." In order
to simplify the representation from three to two dimensions, the so-called
trichromatic coefficients x, y, and z were introduced; they are related to the
§ 31-9 MODERN COLOR SPECIFICATION 609

tristimulus values in a simple algebraic fashion, and by definition x + y + z

= 1, so there are only two independent variables. Any two of the three
variables suffice to specify a color, and the ones usually chosen are x and y;
when plotted against each other they give a two-dimensional plot called the
chromaticity diagram. Figure 7 shows this diagram with the locus of pure
spectrum colors plotted as ROYGBV, and wavelengths indicated at frequent
intervals. It also shows the straight-line "purple boundary" VPMR which
connects the ends of the spectrum locus. White light is represented by the
point W, near the center of the diagram. Actually, the term "white light"
is far from specific, and for this reason the International Commission on
Illumination found it necessary to define "whites." The most useful white

520
G

0
600
610
,,-r-------~ 620 R
700

v x
Frn. 31-7. Chromaticity diagram, showing the spectrum locus ROYGBV, the purple
boundary VPMR, and the "white point" W. Wavelengths in millimicrons are given
around the spectrum locus

is that specified as Illuminant (or Luminator) C, which has a spectral dis-

tribution approximating that of a black body at 6500°K, and is equivalent
to "average daylight"; it is the one located at Win the figure.
The chromaticity of the various real colors can be represented by points
that lie within the spectrum locus on the chromaticity diagram. For a color
sample, represented say by point 1, the dominant wavelength (hue) is ob-
tained by extending the line that connects the white point and this color
point until it strikes the spectrum locus, and reading off the wavelength.
The purity (saturation) of the color sample is measured from the diagram as
the fractional or the percentage distance of the color point from the white
point, taking the distance from the white point to the spectrum locus (or
610 OPTICS Chap. 31

purple boundary) as unity or as 100 per cent. The purity of a spectrum color
is of course 100 per cent, and the purity of white is zero. If the extended
white-point-to-color-point line strikes the purple boundary, the color is not
a pure spectral color but a purple or a magenta. To obtain a numerical hue
specification for a purple, the line is then extended back through the white
point until it strikes the spectrum locus, and the wavelength so determined
is that of the complementary hue.
Three examples are shown in Fig. 7: Point 1 represents an orange or red
color of dominant wavelength 620 mµ and purity 55 per cent; point 2 repre-
sents a green of dominant wavelength 532 mµ and purity 37 per cent; and
point 3 represents a blue-purple having as its complementary color a yellow-
green of dominant wavelength 553 mµ-the designation of point 3 is hence
553c mµ-and a purity of 73 per cent. These three points do not neces-
sarily represent narrow color bands such as spectrum lines. Indeed, the
various component ·wavelengths that go to making up a sample of colored
light cannot be shown on the chromaticity diagram; all that can be indicated
thereon is the single spectral wavelength which when mixed with white light
will match the sample color.
Points 4 and 5 in Fig. 7 represent a complementary blue and yellow pair,
and the line connecting the points passes through the white point. The
additive mixture of these two colors will produce white or perhaps a neutral
gray, since, as noted above, the chromaticity diagram depicts only two of
the three defining characteristics of color, and a third dimension would be
necessary in order to indicate luminous flux (brightness) also. When two
colors such as those of points 6 and 7 are mixed by addition, the point that
represents the chromaticity of the mixture lies somewhere on the straight
line joining points 6 and 7, at a position depending on the relative amounts
of the two component colors in the mixture.
31-10. Spectrophotometry. The spectral composition of a sample of light
flux can be measured, wavelength by wavelength, with an instrument called
a spectrophotometer. This device does not make absolute measurements, but
only relative measurements, usually of the light transmitted by a test sample
or the light reflected from its surface. In transmission tests, comparison is
made against the unobstructed beam from a light source, or for liquid or gas
samples the comparison reading is taken with an empty absorption cell in the
beam. In reflection measurements the "unknmvn" sample is tested against
a standard white area such as a freshly prepared surf ace of magnesium oxide
made by burning magnesium ribbon in the air, or more conveniently, against
a calibrated plate of white glass. The early instruments were of the visual
type in which the eye was used to make a brightness match between two
adjacent fields. In modern spectrophotometers the eye is replaced by a
photoelectric cell, and often a recording mechanism is employed, so that the
spectrophotometric curve is drawn automatically and rapidly. The quan-
tity plotted for each wavelength is spectral transmittance (the ratio of the
§ 31-10 SPECTIWPIIOTOJ1ETRY Gll

radiant energy transmitted by the sample to that incident upon it), or spec-
tral reflectance (the ratio of the radiant energy reflected from an object to
that incident upon it). Transmittance and reflectance values are usually
expressed in per cent.
The schematic optical system of a typical spectrophotometer is shown in
Fig. 8. The path of light through the instrument is as follows: from the
light source to the reflecting mirror; through the slit into the spectrometer,

Light source

Phototube

FIG. 31-8. Schematic diagram of a non-recording spectrophotometer arranged for making

transmission measurements

where it is collimated by the spherical mirror and reflected toward the 30°-
600-900 dispersing prism, which has a mirror coating on its second surface.
Light of the selected wavelength is reflected back through almost the same
path and is again reflected from the spherical mirror, through the slit. It
now passes below the reflecting mirror and through the absorption cell or
other sample into the phototube; the reading of the galvanometer connected
to the phototube is proportional to
the intensity of the beam. This c lOO
Cl)
spectrophotometer does not record ::'. 80 1---1----+----1---+----b.,,c----1
Cl)
readings automatically, but operates a.
on the ''point-by-point" method; a ~ 60
narrow wavelength band is selected ~ 401---+---1<-+-----.+---,'-----v--------1
by proper rotation of the prism -~
and setting of the slit width and a I-- 20 e
transmittance or reflectance meas- 0 !:::=.:-=:::~~~~.::~!""~~~~~~~
500 600 700
urement is made for that wavelength 400 Wavelength, millimicrons
band, and so on through the spectral
range under study. Other instru- FIG. 31-9. Spectrophotometric curves rep-
resenting the spectral transmittance of: 1, a
ments are designed to scan the red color filter; 2, a green filter; 3, a purple
spectrum automatically and record filter
data on a moving chart.
In Fig. 8 the instrument is shown as set up for making transmission meas-
urements of a liquid or gas contained in an absorption cell. The transmit-
tance of a solid sample such as a color filter can be measured by putting it in
the beam in place of the absorption cell. Reflection measurements are made
by using additional accessories. An instrument of this general type is
G12 OP'PICS Chap. 31

suitable for measurements in the visible and the ultraviolet regions if quartz
optical elements are used; a rather similar instrument, but with its optical
elements made of rock salt or some other infrared-transmitting material, can
be used in the infrared region. The general considerations discussed in
§ 31-4 regarding optical materials, light sources, and detectors apply to
spectrophotometers as well as to spectroscopes and spectrographs.
Figure 9 shows typical transmission curves for a red, a green, and a purple
color filter, such as might be obtained with a direct-recording spectropho-
tometer that covers the visible spectrum. The light transmitted by these
three filters, with white light incident, might well give the chromaticities
plotted as points 1, 2, and 3 in Fig. 7.
31-11. Color Filters. Color filters are used in such diverse applications as
optical instruments and stage lighting to isolate desired portions of the visible
spectrum. The amount of light transmitted by glass or plastic color filters
depends on the nature of the absorbing substance in the glass or plastic, and
on the thickness of the filter. The former factor accounts for the general
shape of the transmittance-wavelength curve, and the latter factor affects
the transmittance at all wavelengths.
The absorption of light of a given color by a homogeneous, isotropic mate-
rial can be treated as follows. If the entire medium is regarded as made up
of a number of layers of equal thickness, each one absorbs the same fraction of
the light entering it. Thus, if 100 lumens impinge upon the first layer and
if each absorbs 1/10 of the light that reaches it, the amount incident upon
the second layer will be 90 lumens, upon the third layer 81 lumens, upon the
fourth layer 72.9 lumens, and so on. This behavior has been verified by
experiment. Thus, an infinitesimally thin layer cuts down the intensity by
an amount that, expressed as a fraction of its value, is directly proportional
to the thickness of the layer. If I is the intensity at a particular layer of
infinitesimal thickness dx, and -dI the change of intensity in this layer, then
the foregoing statement can be expressed mathematically in the form
dI
-1 = k ax,
where k is a constant of the material called its absorption coefficient. This
expression can be integrated to give the intensity I of a light beam, after
passing through a thickness x of a medium having an absorption coefficient
k, in terms of the intensity Io of the entering beam, by the equation:
I = Io e- kz, (31.1)

where e is the base of natural logarithms.

Because of the exponential nature of Eq. 31.1, it is convenient to introduce
a quantity that is linearly proportional to the thickness of the filter. This
physical quantity, called optical density, is defined as the common logarithm
of the reciprocal of the transmittance, I/ I O of the filter, and may apply either
§ 31-12 THE RAINBOW 613

to overall white-light transmittance or to transmittance at any particular

,vavelength. Thus,
Optical density = log ~0 • (31.2)

The correspondence of optical density values to decimal and percentage

transmittance is as follows:

Transmittance

I Optical Density
!_ X 100
Io Io

1 100 per cent 0

0.1 10 1
0.01 1 2
0.001 0.1 3
0.0001 0.01 4

If a filter 3 mm thick has an optical density of 2, a similar filter 6 mm thick will

have a density of 4--a transmittance of 0.01 per cent.
It should be noted that the simple relation between optical density and
filter thickness holds accurately only after transmittance values have been
corrected for surface reflection losses, Eq. 28.5.

ATMOSPHERIC OPTICS

31-12. The Rainbow. One of the most spectacular phenomena of atmos-

pheric optics, the rainbow, is caused by the dispersion of sunlight within
raindrops. The ray of light that strikes a drop is refracted at its surface,

Secondary rainbow

Primary rainbow

Fig. 31-10. Illustrating the rays that form the primary and secondary rainbows.

reflected within it, and refracted again (and dispersed) upon emergence.
Murh of the light is scattered in various directions, but the rays near the
path of minimum deviation from a series of drops merge into a fairly intense
beam along the directions shown in the lower part of Fig. 10. Dispersion
614 OPTICS Chap. 31

spreads the incident white light into the spectral colors, with angles between
the incident and emergent rays of about 40° for violet and 42° for red. An eye
at E receives light of different colors from a multitude of drops, the red color
appearing uppermost. Drops similarly located with respect to sun and
observer but not in the plane of the page produce a corresponding effect, and
the collective action yields the familiar rainbow. A secondary rainbow is
sometimes formed outside the first, and is much fainter; it is produced by
a double reflection within the drops, as shown in the upper part of the figure.
The order of the colors is reversed from that of the primary bow, and the
angles between incident rays and those emerging after the double internal
reflection vary from about 50° for the red to 54 ° for the violet.
31-13. Atmospheric Scattering. The daylight sky is bright because of the
scattering of sunlight by molecules of air, and by dust particles and other
impurities contained in the atmosphere. Without this scattering of the
sun's rays by the atmosphere, the sky would appear black. It has been
established that the intensity of scattered radiation varies inversely as the
fourth power of the wavelength of the incident light, and for this reason the
short waves, corresponding to the blue and violet colors, are scattered more
than the longer waves of red light; this accounts for the prevailing blue color
of the sky. The light coming directly from the sun when it is near the
horizon is predominately orange and red, because the blue and green portions
have been scattered to a large extent from the direct beam in traversing the
relatively long paths through the atmosphere. The colors of sunsets and of
the clear blue sky are thus aspects of the same atmospheric scattering phe-
nomenon.
31-14. Transmission of Light through the Atmosphere. The answer to the
practical question of how far one can see at a certain time depends on the
properties of the atmosphere, on the eye, and on the objects that are in
the field of view. In common practice the term visibility refers to more-or-
less horizontal viewing, while the term ceiling refers to vertical viewing. The
latter term is being replaced in aircraft operations by the more applicable
concept of glide-path visibility, at an angle of 8 ° to the horizontal. Since the
atmosphere tends to be stratified, that is, to contain fairly horizontal layers
of differing wind and cloud composition, the situation is more complicated
than the simple term suggests.
Atmospheric haze reduces the visibility of all distant objects by decreasing
their apparent contrast-the difference between the apparent brightness of the
object and that of its background. For objects seen against a horizon-sky
background, this reduction of contrast results from a change in the apparent
brightness of the object. This change is governed by two processes which
operate concurrently: (1) The beam of light coming from the object is gradu-
ally weakened by scattering and by absorption; (2) daylight is scattered
toward the observer, all along the line of sight. The balance .between that
portion of the light from the objert which reaches the observer, and the
§ 31-14 . TRANSMISSION OF LIGII'P THROUGH TIIE ATMOSPIIERE 615

space light contributed from the intervening air, determines the apparent
brightness of the object. For example, the apparent brightness of a dark
mountain increases with distance because of the addition of space light, and
the mountain finally is invisible because its apparent brightness matches that
of the sky background. In a like manner, the apparent brightness of a sunlit,
snow-clad peak decreases with distance until, at a sufficient range, it too
matches its sky background; but here the reduction in apparent brightness is
due to the weakening of the light beam from the mountain, by scattering
and by absorption.
In estimating so-called meteorological range, meteorologists customarily
make a visual judgment of the optical effect of the atmosphere by inspecting
distant black objects seen against the horizon sky. By international agree-
ment the daylight visual range is the distance at which a large dark object on
the horizon is just recognizable. Here a "large dark object" is taken to
mean an object that subtends so large an angle at the observer's eye that if
the subtended angle were greater, the reported value of daylight visual range
would not be changed. Because the distant dark objects available to mete-
orologists are rarely of sufficient angular size to fulfill this requirement, the
reported "visibility" is usually somewhat less than the true daylight visual
range-often only about three fourths of the correct value.

PROBLEMS
1. Classify the sun's spectrum in terms of the spectrum types listed in § 31-2.
2. Derive an expression for the apparent change in frequency of a line in a star's
spectrum, due to a relative speed u between the star and the earth.
3. Measurements in the laboratory show that the red line of hydrogen has a wave-
length of 656.28 mµ. Compute the wavelength of this line as observed in the spectrum
of a star that is moving away from the earth at a radial speed of 120 km/sec.
4. The hydrogen red line in the spectrum of a star is observed to have a wavelength
of 656.24 mµ. Is the star moving away from or toward the earth? What is the
relative speed of star and earth?
5. In a group of 400 people, half men and half women, how many of each sex would
be expected to be "color-blind?"
6. Name the color that designates a filter that partially transmits each of the
following: (a) Red, orange, blue, and violet, (b) yellow, green, and blue.
7. What is the resulting color, when the following pairs of colors are mixed by
addition? (a) Green and purple, (b) green and red, (c) green and yellow.
8. What is the resulting color, when the following pairs of colors are mixed by sub-
traction? (a) Green and purple, (b) green and red, (c) magenta and yellow.
9. A piece of cloth is purple in sunlight. What color would it appear to have (a) in
sodium light, (b) in light from a mercury vapor lamp, (c) in light from a hydrogen
discharge tube? (Compare Figs. 4 and 9.)
10. Trace the chromaticity diagram from Fig. 7, and locate upon it points to repre-
sent the following: (a) a very light pink, (b) the spectral yellow of the sodium D lines,
(c) a color of dominant wavelength 555 mµ and purity 80 per cent.
11. On the figure of Prob. 10, plot a point to represent a color of 70 per cent purity
that is complementary to a green of dominant wavelength 500 mµ. ·what color name
applies? How is its chromaticity designated?
616 OPTICS Chap. 31

12. On the figure of Prob. 10, draw a line that represents all the possible chro-
maticities that can be obtained from the additive mixture of the following two colors:
480 mµ, 75 per cent; and 540c mµ, 25 per cent.
13. Of what optical materials could the prism of a spectrophotometer be made for
use in the following wavelength ranges? (a) 130 to 350 mµ (vacuum instrument),
(b) 2.5 to 15 µ, (c) 15 to 30 µ?
14. In measuring weak light intensities with the spectrophotometer, the slit is
opened wider so as to obtain a larger phototube response. Comment on how greater
slit widths affect the amount of detail shown in the spectrophotometric curves, es-
pecially ·when narrow absorption bands are involved.
15. Sketch spectrophotometric curves, similar to those of Fig. 9, illustrating the
transmittance of a red color filter, an orange filter, and a yellow filter.
16. A glass filter 1 mm thick has a transmittance of 10 per cent at a certain wave-
length. A filter of similar glass, but 2 mm thick, is placed in contact with the first.
Determine the transmittance of the combination at the same wavelength, neglecting
the effects of surface reflection.
17. A glass filter 1.5 mm thick has a transmittance of 50 per cent at a certain wave-
length. What thickness of filter, made of similar glass, would have a transmittance
of 30 per cent at the same wavelength? Neglect the effects of surface reflection.
18. Repeat Prob. 17, taking into account a reflection loss of 4 per cent at each
glass surface.
19. Draw to scale the four rays coming from a rainbow to the eye in Fig. 10, using
horizontally incident light and the numerical values given in § 31-12. Note that the
dispersion in the secondary bow is about twice that in the primary bow.
20. If shorter waves are scattered to a much greater extent than longer waves,
should not the clear sky appear violet rather than blue? Explain.
 32
INTERFERENCE AND
DIFFRACTION
Experiments on interference and diffraction accounted for the abandon-
ment, in the early years of the 19th century, of the corpuscular theory of
light. This theory, vigorously supported by Isaac Newton, had failed to
explain such experimental results as interference fringes and other small-
scale deviations of light from straight-line travel, whereas the wave theory
accounted for these effects quite readily. With the theoretical work of
James Clerk Maxwell in connecting light and electricity, the electromagnetic
wave theory of radiation seemed adequate to explain all experimental facts.
But this satisfactory state of affairs was brought to a sudden end by certain
experiments performed near the close of the century, and these led to the
formulation of the quantum theory of radiation. The wave theory can still
be used, however, in describing and explaining the phenomena of interfer-
ence, diffraction, and polarization, which comprise the major part of the field
of physical optics. Though this field has been chiefly academic in interest,
its practical applications are becoming increasingly important in science and
industry.
32-1. Interference. When two beams of monochromatic light from a
source fall upon a screen, it would ordinarily be expected only that the
illumination would be greater than for one beam alone. Under certain spe-
cial experimental conditions, however, the illumination is far from uniform,
and there are areas of the screen that are much brighter than would have
been- expected, and others that are perfectly dark. This is an 'example of
the interjerence of light. The superposition of the two light waves upon
arriving simultaneously at a given point produces a resultant effect that de-
pends upon their wavelengths, amplitudes, and phases. With monochromatic
light, waves of equal amplitude reenforce each other if they arrive in phase
agreement and annul each other if they arrive in phase opposition, as de-
scribed in § 25-14. When they annul and produce zero illumination at cer-
tain points there are other points where the illumination is increased, since
the total energy of the waves remains unchanged. With white light, in
which many colors are blended, the annulment of one color at a particular
point still leaves illumination by the other colors.
The interference of light was originally demonstrated and explained by
617
(HS OPTICS Chap. 32

the brilliant English physician and physicist, Thomas Young, who used
an arrangement as shown in Fig. 1. Light from a monochromatic source is
directed upon a small opening 0, usually a narrow slit, in the first shield;
this serves as a line source for illuminating two slits }.f and N in the second
shield. From these apertures waves spread out in all directions; a few are
indicated in the plane of the light source. vVaves leaving M and N in phase
agreement and proceeding to the screen reenforce each other and produce
brightness at certain places, BBB; at other places, DD, they annul each other
and produce darkness. Reenforcement occurs where the two waves arrive
at the screen in phase agreement, both waves having traveled the same dis-
tance or one of them having traveled one or more complete wavelengths
farther than the other. Annulment occnrs where the two waveR arrive at

Light
source

Fm. 32-1. Reenforcement and annulment of light waves

the screen in phase opposition, one wave having traveled an odd number of
half wavelengths farther than the other. The result on the screen is an inter-
ference pattern of lines, called fringes, which are alternately bright and dark.
The figure shows a sectional view of only the zone containing the interference
pattern; this region is actually very small, and is greatly exaggerated later-
ally, for clearness.
The factors determining the spacing between fringes will be investigated
with the aid of Fig. 2, which represents a sectional view of two parallel-sided
slits M and N separated by a distance d, and a screen at a distance L from
the plane of the slits. The axis AB is drawn perpendicular to the screen
from A, midway between the slits; the rays shown are those that go from the
apertures to a point P on the screen, located at a distance x from point B.
Light waves issuing from the two slits travel different distances in reaching P.
By drawing MQ so as to make QP = J.vfP, the path difference can be expressed
as s = NQ. A line connecting points P and A is very nearly perpendicular
§ 32-1 INTERFERENCE 619

to MQ, and hence the angles NMQ and PAB are almost equal. Since these
angles, both marked e, are small, s/d, which is approximately the sine of one
angle, may be equated to x/ L, which is the tangent of the other, giving the
simple relation

(32.1)

As point Pis moved along the screen, nearer to or farther from B, the dis-
tance x changes and so does s. The resultant effect at P depends upon the
number of wavelengths X contained ins: '\¥hen s = mX, ,vhere mis an integer,
reenforcement occurs at P; when s is an odd number of half wavelength:cs,
annulment occurs. For either situation, the distance Lix on the screen be-

I
L
I
p

I"'~IAJ
--f
x
Light B _J_
source

i N+r Screen

Frn. 32-2. The double slit

tween ceuters of adjacent bright or of adjacent dark fringes corrm;ponds to a

value of Lis equal to ;\, so that Lix = 't..L/d, from Eq. 32.1. This expression
can be solved for wavelength to give
d
X = Lix y; (32.2)

Thus, the wavelength of the monochromatic light can be determined experi-

mentally by measuring the fringe spacing Lix, the slit separation d, and the
slit-to-screen distance L.
For example, in a certain experimental arrangement the centers of the two slits are
1 mm apart and the screen is 1 m away, and the distance between centers of neighbor-
ing bright fringes, as measured with a traveling microscope, is 0.546 mm.
With this arrangement d = 0.1 cm, L = 100 cm, and Lix = 0.0546 cm. Hence
the wavelength of the light, from Eq. 32.2, is X = 0.0546 X 0.1/100 = 0.000054ti
cm= 546rnµ.
When white light is used instead of monochromatic light, each color pro-
duces its own interference fringes; those for red light are spaced about twice
as far apart as those for violet light, and the _resulting pattern combines the
effects for all colors. Such brilliantly colored designs may also be observed
by looking at a distant white-light source through a piece of silk or a fine-mesh
screen, since the meshes act as closely spaced apertures that transmit inter-
fering beams of light.
620 OP11JCS Chap. 32

32-2. Fresnel's Biprism and Lloyd's Mirror. An early criticism of the

double-slit experiment was that the fringes might be caused by a modification
of the light beams at the edges of the slits rather than by true interference,
and the new wave theory of light was hence questioned. This suggestion was
refuted by producing interference fringes by additional methods that did not
involve narrow slits. In one of these, introduced by Fresnel, the pair of
apertures is replaced by a thin prism, known as a biprism, having an angle
of almost 180°. Such a prism, shown in Fig. 3, refracts the light from a
narrow source A so that it appears to come from two virtual sources 1"11 and N.
The screen is illuminated by source M over the region GH and by source N
over the region EF. The interference fringes occur near B, at the center of
the overlapping region GF, and can be observed with a low-power telescope.

Screen

E
G
G
Ml F
F N I.-.--
-- ---
Mirror I
I Screen
H I
Bi prism y
Fw. 32-3. The Fresnel biprism FIG. 32-4. The Lloyd mirror

Still another method for producing interference fringes is the so-called

Lloyd's mirror experiment diagramed in Fig. 4. Light from slit J.l1 can reach
the screen either directly, or after reflection in the mirror as if proceeding
from the virtual image N. The slit and its virtual image thus serve as two
sources and produce a pattern of interference fringes in the region GF of over-
lapping light.
The difference in the nature of the interfering light sources in the three
devices should be noted: In the double slit both are real, in the bi prism both
are virtual, and in Lloyd's mirror one is real and the other is virtual. In all
three, however, the light comes from the same original source, and hence the
two interfering beams have a constant mutual phase relationship that does
not vary with time, even on the finest scale. Such beams are said to be
coherent; only coherent beams can produce interference.
32-3. Thin Films. An example of the interference of beams of light from
extended rather than slit sources is given by thin films. Colors are frequently
observed in films such as soap bubbles, thin layers of oil on water, and coatings
of oxide on heated metal. Such films are usually observed by reflected light,
and the colors are due to the interference of the light waves reflected at the
front and back surfaces. Figure 5 represents a film of refractive index n,
illuminated by a beam of monochromatic light incident upon the surface in
the direction A. An eye placed as shown receives light reflected partly at B
§ 32-3 THIN FILMS 621

(ray 1) and partly at C (ray 2). In order to compare the optical paths of
these rays, a line FG is drawn perpendicular to the direction in which they
approach the eye. It is seen that the only differences in path occur between
the point B and the line FG, and that the second ray travels farther than
the first by an amount (2 a - b), where a represents the distance BC or CF
and b represents the distance BG. The distance 2 a ,vithin the film is equiva-
lent to an air distance of 2 na, and hence the equivalent light path of ray 2
exceeds that of ray 1 by the amount (2 na - b). For normal incidence, the
distance a is the same as the thickness t of the film and the distance b becomes
zero, whence the optical path difference is 2 nt; in the general case it can be
shown to be 2 t-V n 2 - sin 2 i, where i is the angle of incidence.
It might be expected that as the film thickness t approaches zero, the two
rays would come into phase and reenforce each other, since their optical paths
would then approach equality. It is found in experiments with soap bubbles,
however, that a black spot appears where
the film becomes so thin that it is about to
A
rupture, and this shows that destructive in-
terference of light rather than reenforcement
occurs when the film has negligible thick-
ness. In order to explain this effect, it
must be recognized that one reflection takes Bi
I
place within a medium (air) of low refractive n I
I
index at the boundary of a medium (soap I tC
film) of higher index, whereas the other I I
I I
I I
reflection takes place oppositely. Under 1/
j(
these circumstances there is always a rela- I

tive phase change equivalent to a path Frn. 32-5. Interference produced

difference of a half wavelength between the by a thin film
two reflected waves, in addition to the phase
displacement due to difference in optical path. It can be shmvn by the
Lloyd's mirror experiment that the phase change of 180° occurs when light
is traveling in the medium of lower index and is reflected at the surface of
the more refractive medium: If the screen is placed in contact with the end
of the mirror, at position X---Y in Fig. 4, the edge Z comes at the center of
a dark rather than a bright fringe.
In the thin-film interference of Fig. 5, the retardation of ray 2 with respect
to ray 1 for normal incidence equals 2 nt because of path difference, and the
phase displacement due to reflection is equivalent to a further retardation of a
half wavelength, X/2. Consequently, the total retardation of ray 2 with
respect to ray 1 is:

Retardation = 2 nt +t (32.3)

·when the ret~rdatiop. is an odd number of half wavelengths, there is destruc~

622 OPTICS Chaµ. 32

tive interference and darkness; when it is an even number of half wavelengths,

there is maximum reenforcement and brightness. Thus, for the first reen-
forcement 2 nt + >../2 = 2 >../2, whence the minimum thickness of film is
t = >../4n. With white-light illumination, the thinnest film to show inter-
ference color by reflection is that for which t = >../2n for violet light; this film
will have a residual reddish color. Thick films do not appear colored, because
so many wavelengths satisfy the conditions for reenforcement, namely,
t = A/4n, 3>../4n, 5A/4n · · · , that the reenforced waves upon merging pro-
duce the effect of white light.
An interesting experiment on the interference produced by films makes use
of an optically flat plate and a plano-convex lens of large radius of curvature,
with the two placed together so that a wedge-shaped film of air is formed
between them. When the apparatus is illuminated by monochromatic light
and examined in reflection, the observer
sees an interference pattern consisting of
Partially
silvered a series of bright and dark rings con-
centric around the point of contact.
This phenomenon was first described by
Extended
Newton, and the rings are called by his
source name. The effect, which is due to the
interference of light reflected from the
glass-air and air-glass boundaries of the
air film, can be observed in an experi-
Frn. 32-6. Arrangement for observing mental arrangement such as is shown
Newton's rings in Fig. 6. Light from an extended
source strikes a partially silvered mir-
ror, and is reflected to the air wedge between the lens and the plate. After
reflection at the wedge, some of the light returns through the mirror to the
observer's eye. From Eq. 32.3 it follows that the variation in film thickness
corresponding to a shift from one bright (or dark) fringe to the next sim-
ilar fringe is t = >../2n. If the wedge-shaped film is an air wedge, this
becomes t = A/2; thus simply by counting the fringes one can determine
minute variations in film thickness. This interference effect is widely used
by the optical industry in testing the surfaces of newly manufactured optical
elements for correct shape. The surfaces are examined, in monochromatic
light, by placing test glasses of known, standard curvature against them.
The fringe pattern shows immediately whether the element has been brought
to the correct figure, and reveals any slight residual surface irregularities.
In testing plane surfaces, so-called optical flats are used as standards; devia-
tions from perfect flatness of as little as one twentieth of a wavelength (about
three millionths of a centimeter) can readily be detected by this extremely
sensitive method.
32-4. Reflection-reducing Films. The lenses, prisms, and windows in
optical instruments are often coated with thin transparent films in order
§ 32-4 REFLECTION-REDUCING FILMS 623

to minimize reflection from the surfaces, and thereby increase the amount of
light transmitted. The films are of such materials and thicknesses that the
light reflected from the air-film surface annuls the light reflected from the film-
glass surface. In order to produce destructive interference, the two reflected
waves should be in opposite phase and should have equal amplitudes.
The same type of reflection occurs at each interface, because the ray is
directed from a less to a more refractive medium; therefore no relative phase
difference is introduced between the two reflected rays. The condition for
their destructive interference, with monochomatic light incident normally,
is hence that the equivalent air thickness of the film should be an odd num-
ber of quarter wavelengths. For the minimum film thickness t, and for a
wavelength >.., it follows that
(32.4)

where the subscript f refers to the film.

The requirement that the waves reflected from the glass and the film shall
have equal amplitudes can be stated as

(32.3)
where n1 and nv are the refractive indexes of film and glass, respectively.
This condition follows from Fresnel's treatment of reflection referred to in
§ 28-10, but will not be derived here.
A suitable film material of sufficiently lo,v refractive index to satisfy Eq.
32.5 has not been found. With the available film materials, the effectiveness
of reducing reflection is much poorer for crown glass surfaces than for the
higher-index flint glass. This fact is illustrated in Table I, which gives the
reflection loss for white light incident normally at an uncoated glass sur-
face as computed from Fresnel's equation (Eq. 28.5), and the loss expected
when the glass surface is coated by evaporation in vacuum with a quarter-
wavelength retardation film of magnesium fluoride, the best available ma-
terial. For crown glass the reflection loss is reduced to i by the coating, as
shown in the last column, whereas for dense flint glass it is reduced to y1 .

TABLE I. EFFECTIVENESS OF A J\fAGNESIUl\f FLUORIDE FILM IN REDUCING

REFLECTION AT AN AIR-TO-GLASS SURFACE

Refractive Index Reflection Loss, Reflection Loss, Relative Reduction,

of Glass Uncoated Surface Coated Surface Coated to Uncoated

1.52 4.3 per cent 1.5 per cent 3

1.60 5.3 1.0 5
1.70 6.7 0.6 11

Equation 32.4 indicates that a certain film thickness produces maximum

reduction of reflection only for a single wavelength, and hence that the re-
624 OPTICS Chap. 32

duction for other wavelengths is less complete. Therefore, optical elements

for use in visual instruments are coated with films of such thickness as to be
most effective in reducing the intensity of the green light for which the eye
is most sensitive. The lack of perfect annulment illustrated in Table I can
be attributed partly to the lower efficiency for wavelengths well removed
from the green.
The improvement in light transmission brought about by coating with
reflection-reducing films depends of course on the number of air-glass surfaces
in the particular optical instrument under consideration. For a simple tele-
scopic instrument like the prism binocular, Fig. 30-21, the gain in transmit-
tance is only about 35 per cent of the uncoated value, whereas for the com-
plicated submarine periscope it may approach 200 per cent, so the overall
transmittance is three times that of the uncoated instrument. In addition,
there is a reduction in the intensity of ghost images and of veiling glare over
the principal image, which may be even more important than the increase
in overall transmittance of the instrument.
32-5. Interferometers. The interferometer makes use of interference in
measuring the wavelength of light in terms of a standard of length, or in
measuring an unknown length in terms of a known wavelength of light. The
essential parts of the instrument devised by Michelson are arranged as shown
in Fig. 7; they comprise an extended light source, two plane mirrors A and B,
and a glass plate M, of which the lower surface is lightly silvered. One of
the mirrors (A, in the figure) is mounted on a fine-pitch screw, and the silvered
plate is set at 45° to the directions Alvf and Blvf. Monochromatic light from
the source strikes plate 111 and divides into two beams, 1 and 2; these advance
to the mirrors A and B, respectively, are reflected back to M, and thence
proceed to the eye. Beam 1 passes through the half-silvered plate M three
times, whereas beam 2 passes through it only once; a second plate N of the
same thickness and inclination as Mis hence introduced to equalize the two
paths. If the waves arriving at the eye from 111 are in phase, they recombine
in constructive interference and the center of the field is bright, or if they are
in phase opposition, the field appears dark along the axis Al\1. These effects
can often be observed more conveniently through a small telescope placed in
front of the eye.
In a certain adjustment, ,vi.th light of wavelength>.._ directed upon the half-
silvered plate, the distances to mirrors A and B are such as to cause the field
of the telescope to be dark. If the movable mirror is now advanced slowly, a
movement of >../4 causes a change of path of >../2 and the field will appear
bright; a further movement of >../4 will make it dark again, and so on. An
accurate scale is provided for measuring the distance through which this
mirror is moved, and consequently the wavelength of the incident light can be
determined directly by counting the number of light annulments. Thus, if m
successive interferences are observed while the mirror is moved a distance Z,
the wavelength is found to be
§ 32-5 I 1\/ 'PH RF fiJ RO Jl ET E RS 625

r. = ~- (32.6)
m
Conversely, if X is known, the distance l can be determined with great ac-
curacy.

Frn. 32-7. The Michelson interferometer

Usually the mirrors A and Bare not exactly at right angles to each other,
as implied in the foregoing description, but depart slightly from this condition.
·with the latter adjustment, the field of view is crossed by dark interference
fringes, separated by bright reenforcement fringes, and the movement of
mirror A can be observed by counting either the bright or the dark fringes as
they sweep past the cross hair of the telescope.
The interferometer can be used to measure extremely small distances-for
example, the expansion of crystals under slight temperature changes. Michel-
son also used this instrument by a
doubling process to measure the
From
length of the standard meter in terms extended
of the wavelength of the prominent source
red line in the cadmium spectrum,
§ 1-3 ;this measurement fixes definitely To screen
the length of the standard in terms of
an unvarying unit. The diameters of
a few of the largest stars have been de-
termined by adapting the interfero- Frn. 32-8. Illustrating the multiple reflec-
meter to the astronomical telescope, tions in the Fabry-Perot interferometer
§30-8.
There are several other types of interferometers, the most generally useful
of which is probably the one developed by the French scientists Fabry and
Perot and illustrated diagrammatically in Fig. 8. The Fabry-Perot inter-
626 OPTICS Chap. 32

ferometer consists essentially of two very flat plates of glass, lightly silvered
on the inner, facing surfaces, and held rigidly parallel to each other by an
appropriate separator. An extended source of light is used, and the inter-
ference pattern is focused on a screen by a lens, or vimved directly. The in-
cident light beams, of which only one is shown in the figure, reflect b_ack and
forth several times between the two plates. At each reflection there is also
partial transmission, so that each point on the screen is illuminated by many
interfering beams rather than by just two as in the Michelson interferometer.
As a result (the explanation of
,vhich is beyond the present treat-
ment) the interference fringes are
much sharper than those formed by
J only two interfering beams; the
comparison is illustrated in Fig. 9.
The sharper fringes of multiple-
rr beam interferometers have several
important applications, for example,
in spectroscopy in revealing very
Fm. 32-9. Comparing the fringes formed by close spectrum line spacings that
two interfering beams (part I) with the
are much too fine to be resolved
sharper fringes of a multiple-beam interfer-
ometer (part II) by ordinary spectroscopes. A re-
cent application is in the so-called
interference filters which provide a relatively inexpensive means of isolating
narrow bands (5 to 15 mµ wide) of colored light. These bands are much
narrower in wavelength than are the transmission bands of ordinary glass or
plastic color filters, and have high peak transmission. A set of interference
filters can be substituted for the much more expensive monochromators or
wavelength spectrometers used in certain laboratory techniques to select
narrow regions of the spectrum. These filters are solid, nonadjustable Fabry-
Perot interferometers in which the spacing is accomplished by evaporating
a thin transparent film, often of magnesium fluoride, over the lightly sil-
vered surface of one plate; the film thickness is adjusted during the evapor-
ization process so that an interference fringe occurs at the desired wavelength.
Ordinarily, only one such fringe lies in the visible region, but if there are others
they can be suppressed by the use of conventional color filters.
32-6. Diffraction. The interference devices described in the five preceding
sections all involve the interaction of only two beams of light, with the ex-
ception of the Fabry-Perot interferometer. The term diffraction is usually
employed when the number of interfering beams is very large. These may
be discrete beams, as in a diffraction grating, or they may be the rather
artificial subdivisions of an extended wave front used in the study of the
diffraction of light by an aperture, and be infinite in number.
The names of two early workers are associated with specific treatments of
diffraction: Fresnel and Fraunhofer. In so-called Fresnel diffraction, which
is the general treatment, the wave fronts need not be plane, and hence the
§ 32-7 DIFFRACTION BY A SLIT 627

light beams may be convergent or divergent; in Fraunhofer diffraction the

beams must be collimated and the wave fronts plane-thus the latter type
may be considered as a special case of the former. Since the resulting pat-
terns are not too different, and since the mathematical treatment of Fraun-
hofer diffraction is much simpler, only that type of diffraction is treated
here. Hence, collimating lenses are employed, and lenses are used to focus
the light on the screen (or on the retina, by the optical system of the unac-
commodated eye).
Diffraction explains the fact that light does not always travel in exactly
straight lines but can "bend around corners" very slightly. These effects are
on such a small scale that they are seen but rarely and only when they are
being looked for specifically. Nevertheless, diffraction is the controlling fac-
tor in the amount of magnification that can be employed in an optical instru-
ment, as discussed in § 32-11.

L Screen

z1w
TM
Lens
p

B
1x
u v
II
0
III

:r CD
aA
l.N m=2
\ 3
4
5 --
6

N
Fm. 32-10. The diffraction of light by a slit

32-7. Diffraction by a Slit. Because of diffraction, a parallel beam of

monochromatic light passing through a slit toward a screen will ordinarily
produce on the screen a bright band somewhat wider than the slit, and this
band will be bordered at the edges by a fow narrower bands which are al-
ternately dark and bright. This effect can be explained with the aid of Fig. 10,
which shows a single slit MN of width a located a distance L from a screen.
Parallel light incident on the slit is focused on the screen, after diffraction, by
a converging lens. The line AB, which is also the lens axis, is drawn per-
pendicular to the screen from the middle of the slit, and P represents a point
on the screen distant x from the point B. The line J.11Q is drawn at right
angles to a line connecting A and P to show the difference in path length
of. the extreme rays NP and MP; this difference, NQ, is marked s.
If the path difference of the rays from the slit edges is one wavelength of
light (s = X), P is a point of darkness rather than brightness. This can be
proved by imagining the slit to be composed of a series of much narrower slits
extending uniformly from .kl toN, and investigating their combined effect in
628 OPTICS Chap. 32

producing illumination at the point P. The intensity at P due to the upper-

most of these slits can be represented by vector u in part II of the figure; the
intensity at the same point due to the next lower slit is then represented by
vector v, which is slightly out of phase with u because the light has to travel
farther from this slit to reach the point. Similarly, the intensity due to the
third slit is represented by w, and so on. The resultant intensity can be found
by placing the individual vectors end to end and drawing the closing line of
the figure, as shown by R for three vectors. If this process is continued from
M to N, then since s = >.. the extreme vectors are again in phase, and the re-
sultant of them all is zero, as shown in part III. This sets the limit of the
central bright band by determining the location of the first dark band above
and below the center. Brightness occurs again if the point P is so located
that s = 3 >../2 (represented in part IV in the figure), darkness if it is so lo-
cated that s = 2 >.., and so on. In general, for annulment or reenforcement,
the path difference for the extreme rays of the slit is s = m>../2; destructive
interference occurs for even values of m and reenforcement for odd values.
It can be shown that the locations of the fringes on the screen above and below
B are given by x = m>..L/2a; annulment occurs when m = 2, 4, 6 · · · and
reenforcement occurs when m = 3, 5, 7 · · · , although the intensity peaks of
the secondary bright bands are actually displaced slightly toward the center of
the pattern. Hence the diffraction pattern consists of a bright central band
having a width 2 x = 2(2 >..L/2a) = 2 XL/a, bordered on each side by nar-
wwer and weaker bands, alternately dark and bright. The relative intensities
at different points are indicated by the curve in part I at the lower right-hand
corner of the figure, wherein the rectangular-shaped "geometrical optics"
intensity pattern of the slit is shown for comparison.
The breadth of the central bright band, or central maximum, of the diffrac-
tion pattern is its most important feature and is usually designated by the
angular half-width a, where a is the angle subtended at the slit by the distance
on the screen from the center of the pattern to the first minimum on either
side. From the foregoing analysis this angle can be given in terms of sand a,
when s = X. The condition for the first intensity minimum is hence given
by
. >.. (32.7)
Sln a = -.
a
When the slit width is much larger than the wavelength of the light, this ex-
pression can be reduced to a = >../a. The linear width, 2 x, of the central
diffraction maximum on the screen is then given by the product of twice
the angle a and the focal length L of the lens. If the slit width is very small,
approaching the magnitude of the light wavelength, the diffraction pattern
will spread through a large angle. In the extreme case a = X, and from
Eq. 32.7, a = 90°; the central diffraction maximum spreads around the
entire 180°.
§ 32-8 INTERFERENCE AND DIFFRAC'l.'ION COMBINED 629

32-8. Interference and Diffraction Combined. The distribution of light

by diffraction, just described, occurs whenever the beam passes through a
narrow aperture. A more complete treatment of the double-slit interference
pattern must hence include the effects of the diffraction produced at each of
the two slits. It is convenient to examine the results of interference and
diffraction separately, and then to combine them to give the final distribution
of light on the screen. The bright and dark interference fringes are equally
spaced, § 32-1; their intensity contour can be shown to follow the cos 2 law,
as illustrated in part I of Figs. 9 and 11. The spacing on the screen can be
given in terms of the angle {3 subtended at the double slit by the centers of
successive bright fringes as {3 = X/d, from Eq. 32.2: The angular spacing is
-9 -8 -7 -6 -5 -4 -3 -2 -1 0 2 3 4 5 6 7 8 9

MIBMMNllMMillV\N\,
D'

ill

-9 -8 -7 -6 -5 -4 -3 -2 -1 0 23456789
FIG. 32-11. Illustrating the combined effect of double slit fringes (part I) and the diffrac-
tion envelope of the single slit (part II), to give the final intensity pattern (part III)

seen to vary directly with wavelength and inversely with distance between
slit centers.
The diffraction pattern of the single slit differs greatly from the interjerence
pattern of the double slit. As shown in part I of Fig. 10 and in part II of
Fig. 11, by far the most prominent feature of the diffraction pattern is its
large central bright zone; the secondary maxima on the two sides are much
less intense and have only half the angular width of the central maximum.
The angle between the axis and the direction to the center of the first dark
fringe on either side is a, which from Eq. 32.7 is given approximately by
'A/ a. It is assumed that the two slits of the double slit are of equal width,
and hence that their diffraction patterns are identical and are superimposed
on the screen.
The intensity distribution pattern actually observed on the screen is the
combination of the interference and diffraction patterns-it represents the
product of the two component intensity factors for points on the screen re-
630 OPTICS Chap. 32

ceding from the axis. The resultant pattern is given in part III of Fig. 11; it
shows the basic interference fringe spacing, but with the fringe intensity
altered in accordance with the diffraction envelope. Thus the third, sixth,
ninth, etc., fringes are missing, because they occur at positions of zero in-
tensity (dark fringes) in the diffraction pattern. These "missing orders"
occur wherever a, the angular constant of the diffraction pattern, is an inte-
gral multiple of /3, the angular spacing of the interference fringes. From the
relationships given, the ratio between these angles is
a 'A/a d (32.8)
~ = X/d = ~·

The location of the missing orders, and hence the number of interference fringes
located within the central diffraction maximum, depend only upon the rela-
tive magnitudes of double-slit spacing and slit width.
For example, a certain double slit measures 1.2 mm between the centers of the two
slits, each of width 0.3 mm. Since d/a = 1.2/0.3 = 4, the missing orders of inter-
ference fringes are ±4, ±8, ±12 · · · . There are 2 X 4 - 1 = 7 bright interference
fringes in the central group of greatest intensity. Each side group, corresponding to
a secondary maximum of the diffraction envelope, contains 3 interference fringes of
low intensity.

32-9. The Diffraction Grating. Another example of the combined action

of diffraction and interference is provided by the diffraction grating. This
device may be thought of as an extension of the double slit to an array of
several hundred or several thousand parallel, similar slits that are very closely
spaced. The location of the centers of the bright fringes formed on a screen
can be determined by an extension of the treatment of § 32-1, even though
larger angles from the axis are involved and a lens is used to bring the light
to a focus on the screen. The former treatment yielded Eq. 32.1, namely:
s/d = x/L, wheres is the path difference of the rays from the two slits to a
point on the screen that is a distance x from the axis and L is the distance
from slits to screen. This relationship is unchanged by the presence of the
additional slits in the grating, so long as they are regularly spaced at the dis-
tance d between centers, now called the "grating space." For constructive
interference (bright fringes), s must be an integral number m of wavelengths 'A.
From Fig. 2 it is seen that m'A/d = sin 8, which gives the "grating equation" as:
m'A = d sin 8. (32.9)
This figure also shows that the distance x = BP from the axis to the center
of the bright fringe is given in terms of 8 and 'L by x/ L = tan 8. In the usual
grating arrangement, Fig. 12, the diffracted light is focused on the screen by
a lens of focal length nearly equal to L.
There are certain differences between the fringes formed by a double slit
and a grating. The latter are much brighter, because of the larger amount of
light transmitted by the many additional slits, and are much sharper. The
§ 32-9 THE DIFFRACTION GRATING 631

explanation of the difference between the sharpness of these fringes and those
of the double slit is similar to that for the difference between multiple-beam
interferometer fringes and two-beam fringes, § 32-5, and the contrast in
appearance is similar to that illustrated in Fig. 9.
Diffraction gratings are used in spectrometers, Fig. 28-16, and in spectro-
scopes and spectrographs, § 31-1, to disperse light into its component color
bands; wavelengths of discrete lines can be computed from Eq. 32.9, if the
other factors can be measured. A grating spectrograph contains the same
basic parts as the prism instrument of Figs. 31-1 and 31-2. Light from the
source that passes through the slit spreads out in a diverging beam that is
rendered parallel by a collimating lens. This parallel beam strikes the grating
and is broken up by interference and diffraction, somewhat as shown in Fig. 12

-- i_
I
I
L

-
Screen
di
T1
--
-- I m=O

---
I
I m=l
I
I I
I
I
m=2
I
I
I

.FrG. 32-12. Plane diffraction grating of the transmission type, with parallel light incident
normally

for light of a single wavelength. But instead of the screen, a spectrograph i~

equipped with a photographic plate or film holder, and a spectroscope or
spectrometer with a viewing eyepiece. The figure shows at m = 0 the "cen-
tral image" of the slit formed by rays of equal path length and hence zero
relative retardation. Bright fringes for light of wavelength >-. occur on both
sides of the axis, at angles 01 for which m = 1 in Eq. 32.9; these are called
"first-order" fringes. For a relative retardation of two wavelengths between
neighboring beams, the "second-order" fringes are located at 02, and so on.
The figure is drawn for light from a monochromatic source; if other wave-
lengths were present, additional fringes would be formed at other angles,
as shown by the grating equation.
As an example, the light from magnesium vapor formed in an electric arc of the type
mentioned in § 31-3 falls normally upon a diffraction grating that has 5000 lines per cm,
and a spectrum is formed which includes the three prominent green lines of magnesium,
of wavelength 516.7, 517.3, and 518.4 mµ. The angular positions of the intermediate
of these spectrum lines are to be calculated, for the various orders observed with a
grating spectrometer, and also the separation of the outer green lines in the second-
order spectrum observed on a screen parallel to the grating and distant 80 cm from it.
The grating space d = 1/5000 = 0.0002 cm, and the angular positions, reckoned
632 OPTICS Chap. 32

from the normal to the grating surface, are found from the following relations: sin 81
= A./d = 0.00005173 X 5000 = 0.2587 for the first-order spectrum, sin 02 = 2 A.jd
= 0.5173 for the second-order spectrum, and sin 8s = 3 A/d = 0.7760 for the third-
order spectrum. Because mX/d is greater than unity for values of m larger than 3,
under the conditions given, there are no spectra of higher order. The corresponding
angles of diffraction are as follows: 01 = 15.0°, 02 = 31.2°, and 03 = 50.9°. For the
second-order spectrum and shortest-wavelength green line, 2 X 0.00005167/0.0002 =
sin 82, whence 82 = 30.11°. For the line of longest wavelength, 02' = 30.23°. The
distances from the axis (the normal to the grating surface) are given by tan 8 = x/L;
tan 30.11 ° = x/80, whence x = 48.27 cm for the shortest-wavelength line. Similarly,
x' = 48.50 cm for the other, and the separation of the outer lines on the screen is
0.23 cm.

The fringes formed in spectroscopic instruments by diffraction gratings

(and also by prisms) are often called spectrum "lines" because of their sharp-
ness. The lines are actually diffraction patterns of the spectroscope slit, and
have the general form shown in part II of Fig. 11, but greatly enlarged verti-
cally and compressed horizontally. Diffraction also determines the dimen-
sions of the overall intensity envelope, as discussed for the double slit in
§ 32-8. The critical dimension here is the width a of the individual slits of
the grating; ordinarily this dimension is very small, so that the diffraction
pattern is broad, extending through almost 180° if a approaches the wave-
length of light, § 32-7. In the coarse grating pictured in Fig. 12, however,
d = 3 a, and the third orders on each side of the axis are missing because of
diffraction.
Gratings may be of the reflection type, in which the incident light strikes
a metallic surface upon which many parallel grooves have been ruled; the
smooth spaces between grooves function as the "slits." The most useful
gratings are those in which the metallic surface has the shape of a concave
spherical mirror; since the slit is imaged directly as spectrum lines, with such
a grating no collimating or camera lenses are needed. The wavelength range
through which this type of grating can be used is much greater than that ob-
tained when light must be transmitted through a refracting material, and
there is complete freedom from chromatic aberration. Thus, the concave,
reflecting grating has found wide use in spectroscopy, both in instruments for
academic research and in the spectrographs used in the routine analysis of
materials.
32-10. Diffraction by a Circular Aperture. Light is diffracted when it
passes through a circular aperture, much as it is in passing through a narrow
slit, and a spot of light is produced that is somewhat larger than would be
indicated by the rectilinear propagation of light. For the same reason,
parallel light passing through a lens produces at the focal plane a small disk
of light rather than the ideal point of light that is indicated by the geometrical
lens constructions of Chap. 29. It should be emphasized that this result is
not due to an imperfection of the lens but is a consequence of the nature of
light itself. The smallness of the disk image determines the ability of the
§ 32-11 RESOLVING POWER 633

lens to register detail in the object under observation, that is, it determines
the resolving power of the lens.
The treatment of diffraction by a circular aperture can be carried out in a
fashion similar to that of § 32-7, by dividing the circular disk into a number
of imaginary strips of equal width, and calculating their effect. The results
of this rather laborious computation show that the diffraction pattern con-
sists of the central, very bright disk surrounded by alternate dark and faintly
bright rings, and gives a, the angular half-width of the central bright disk, as
. 1.22 :\ (32.10)
s1na = - - ,
D
where D is the diameter of the aperture. Thus diffraction by a circular
aperture gives results quite similar to those with a rectangular slit; the same
general relationship holds for the width of the central diffraction maximum,
but the former is about 22 per cent larger, for a given wavelength and for equal
values of aperture diameter and slit width. Although the central maximum
of the diffraction pattern of the circular aperture is surrounded by several
much weaker fringes, about 85 per cent of the total light in the pattern is
concentrated in the central spot-the so-called Airy disk. The spacing of
the surrounding rings is not uniform, as it is for the secondary fringes of the
single slit diffraction pattern shown in part II of Fig. 11.
When the aperture-to-screen distance L is known, the dimensions of the
diffraction pattern projected on a screen, as in Fig. 10, can readily be deter-
mined from Eq. 32.10. In Fraunhofer diffraction, a lens of focal length equal
to Lis always located just to the right of the aperture, or the "aperture" may
be simply the free diameter of the lens, as in optical instruments. The di-
ameter of the central maximum of the diffraction pattern is hence given as
2 r = 2 f tan a, where r is its radius, or the half-width of the Airy disk, and f is
the focal length of the lens. Since a is a very small angle for lenses of ordi-
nary size, this expression gives for the radius, r = fa.
32-11. Resolving Power. The use of large-diameter lenses (or mirrors) in
astronomical telescopes makes it possible to obtain bright images and small
diffraction patterns. Such patterns reveal fine image detail, and the instru-
ment is said to have large resolving power, or good angular resolution. As an
illustration, two object points such as close stars that are being viewed with
a telescope may be considered. Their images are not discrete points, but
are diffraction patterns of the type discussed in the preceding section; the
width of each pattern is determined by the limiting aperture of the telescope,
usually the free diameter D of the objective lens. If the objects are too close
together, the diffraction patterns overlap and merge to such an extent that
they appear to be a single pattern. If the points are separated widely enough,
there is little overlapping of the diffraction patterns and the "images" are
well resolved and clearly distinguished as two, as shown in part I of Fig. 13.
The practical criterion for the limit of resolution of two close diffraction pat-
634 OPTICS Chap. 32

terns was first suggested by Lord Rayleigh, and usually bears his name.
It is that two diffraction patterns can just be resolved when the central max-
imum of one falls at the first minimum (dark fringe) of the other, as illus-
trated in part II. In part I the separation is three-halves that of the Rayleigh
criterion, and in part III it is one-half. The angular separation specified by
this criterion for image points is of course equal to that of the barely resolved
object points. This limiting angle of resolution, </>, for a telescope can be
written as
1.22).. r (32.11)
</> = - - = -,
D f
where the symbols have the same meanings as before.

Il m
FIG. 32-13. Diffraction patterns (dashed lines) that are (I) well resolved; (II) jrn,t resolved,
in accordance with the Rayleigh criterion; and (III) not resolved. The solid line in each
part indicates the resultant intensity envelope of the two patterns

If a telescope is to be used visually, it is useless for its resolving power to be

greater than that of the eye. The resolving power of the eye can be considered
as being determined by the diameter of the pupil--about 2 mm in bright sun-
light. The corresponding angle of minimum resolution </> is about 1 minute
of arc, which is the angle subtended at 25 cm (the "distance of most distinct
vision") by two points that are 0.07 mm apart. The centers of their images
(diffraction patterns) on the retina are about 0.005 mm apart, and this is
about twice the distance between adjacent cones in the fovea centralis, § 30-1.
Thus there is one unstimulated cone between the two cones that are stimulated
in seeing the finest possible detail.
The finite limit of resolution of the eye is taken into account in designing
optical instruments for visual use. The resolving power is established by the
diameter of the objective lens, and the overall magnification of the instrument
is adjusted, say by choosing the proper eyepiece, so that the final image is
"blown up" to a size where its resolvable detail just matches the resolving
power of the eye. The magnification that will effect this match is called the
normal magnification of the instrument; larger values result in some so-called
empty magnification which makes the image larger and perhaps easier to
§ 32-12 RESOLVING POWER OF THE MICROSCOPE 635

view but reveals no additional detail. The normal magnification of an

optical instrument such as a telescope or a microscope can be computed
from an appropriate, specific formula, which is usually derived in terms of
matching the size of the exit pupil of the instrument to the (entrance) pupil
of the eye.
The objective lens of the large refracting telescope at the Yerkes Observatory has
a diameter of 102 cm; with this lens two point sources of light 1 in. apart can be dis-
tinguished a,s separate objects at a distance of over 20 miles. This statement can be
verified by the application of Eq. 32.11, in which 'A may be taken as 0.00006 cm for
the average wavelength of light. The limiting angle of resolution of the lens is
<p = (1.22 X 0.00006)/102 = ·7.17 X 10-7 radian or about t second of arc. If the
sources are separated by 2.54 cm, they subtend an angle of 7.17 X 10-7 radian at the
lens when distant from it by an amount 2.54/(7.17 X 10-1 ) = 3.54 X 106 cm. This
is equivalent to (3.54 X 106 )/(30.48 X 5280) = 22.0 mi. In contrast, two objects
1 in. apart can be resolved by the eye at a maximum distance of about 100 yd.
32-12. Resolving Power of the Microscope. The treatment of the resolv-
ing power of the microscope differs from that for the telescope because interest
centers on the smallest distance by which two object
points can be separated and still be seen as two, rather I
than on their angular separation. Again, the limiting
aperture that determines the width of the diffraction
patterns is usually the free diameter of the objective
lens. The object is very close to the objective lens of
the microscope, and the angle subtended at the object
by the lens is usually designated by the half-angle i,
Fig. 14. The object O yields an image I that is actu-
ally a diffraction pattern, the dimensions of which are
determined by the diameter D = AB of the objective
lens. Now if O' is a neighboring point in the object
plane a distance w from O such that the geometrical
image I' is located at the first dark ring of the diffraction
pattern centered at I, then the angle a is equal to </J, the
angle of minimum resolution of the Rayleigh criterion;
see Eq. 32.11. The difference between the two paths Frn. 32- 14· Used toil-
lustrate the numerical
O'AI and O'BI is 1.22 A, and it can be shown that aperture and the re-
O'A is longer than O' B by 2 w sin i, whence 2 w sin i solving power of a mi-
= 1.22 }.,, or w = 1.22 }../ (2 sin i). croscope objective
The above result depends on the assumption that the
light beams from object points O and O' are independently self-luminous and
are not coherent. But such is not the case in an actual microscope, where
the object is illuminated with light from the substage mirror or condenser
(Figs. 30-13 and 30-14), and hence there is some phase correlation for various
points in the object. This complicated problem was investigated theoreti-
cally by Abbe, who concluded that the best practical value for w, the min-
imum separable distance in the object, is given by omitting the factor 1.22
636 OPTICS Chap. 32

from the above equation. In working with high-magnification objectives, the

space between the lens and the cover glass over the specimen is of ten filled
with a fluid of refractive index n (see § 30-6). Thus the final express10n
for limit of resolution, or smallest resolvable object detail w, is

w = - . (32.12)
2 n sin i

The product (n sin i) for a given objective is called the numerical aperture
(ordinarily abbreviated NA), and microscope objective lenses are commonly
specified in its terms. The resolution limit of a microscope can be written
most simply as w = X/(2 NA).
Lenses with short focal lengths, which in turn have high magnifications,
have large values of NA. The upper limit for a "dry" microscope objective
is 1.0, but with oil immersion the highest value obtainable is about 1.6. For
white light, or monochromatic light of wavelength near the middle of the
visible spectrum, say at 555 mµ, the smallest detail that can be resolved with
an oil-immersion objective of numerical aperture 1.6 is w = 'A./(2 NA) =
5.55 X 10-5 /(2 X 1.6) = 1.72 X 10-5 cm, or about one third the wavelength.
This limit can be decreased by using light of shorter wavelength, and for ultra-
violet microscopes, which employ radiation in the wavelength range from 400
to 200 mµ, it is about half as great; the images must of course be recorded
photographically. The use of x-rays, with wavelengths several hundred times
shorter, has often been suggested, but the difficulty of focusing these very
penetrating radiations has only recently been overcome. It is stated in
§ 34-10 that fast-moving electrons have a wave nature and a very short
"wavelength," and they are used in image formation in the electron micro-
scopes described in§ 30-7. Even though only very small numerical apertures
can be obtained with electron lenses, see Fig. 30-17, still the "wavelength"
of the electrons is so short, a fraction of a millimicron, that the resolution
limit is reduced to at least a hundredth of that for visible light.
32-13. Resolving Power of the Diffraction Grating. An important prop-
erty of a diffraction grating is its resolving power: its ability to separate
spectrum lines that have nearly the same wavelength. The treatment of
the resolving power of diffraction gratings is closely related to the considera-
tions of the last two sections, and Fig. 13 is applicable here also. A diffraction
pattern is formed as the image of the slit of the spectroscope, and there are
many of these patterns-one for each spectrum line or discrete wavelength in
the radiation from the light source. The spectral resolving power R is defined
in terms of the smallest wavelength interval that can be detected, as:
'A. (32.13)
R = -,
~>..
where 'A. is the average of the two wavelengths and ~Xis their difference. For
example, the Fraunhofer D lines have an average wavelength 'A. = 589.3 mµ
§ 32-13 RESOLVING POWER OF THE DIFFRACTION GRATING 631

and a separation LlX = 0.6 mµ; a spectroscope having a resolving power of

about 1000 will show these as two lines rather than one.
It can be shown that the theoretical resolving power of a grating is given
by the simple expression
R = mN, (32.14)
where mis the order number, as in§ 32-9, and N is the total number of grating
lines being used-the number of lines per cm times the width of the ruled area
in cm. Resolving powers up to 200,000 are obtained with the largest and best
gratings; for still better resolution, interferometers must be used.
The theoretical resolving power of a grating with 5000 lines per cm, as in
the example of § 32-9, and a ruled surface 4 inches wide, is 50,800 in the first
order, 101,600 in the second order, and so on. Although a good grating may
perform at its theoretical resolving power in the first order, in the second order
it may resolve only about half again as well, and in the third order perhaps
only twice as well as in the first.

PROBLEMS
1. In an experimental arrangement similar to that of Fig. 2, the distance between
centers of the slits is 0.20 mm and the distance from the slit plane to the screen is 80 cm.
Determine the distance on the screen between the centers of adjacent bright fringes,
when the slits are illuminated with monochromatic light of wavelength (a) 450 mµ,
(b) 700 mµ.
2. The arrangement shown in Fig. 2 is used to measure the spacing of a double
slit. On a screen 72 cm from the slits, the measured distance between centers of ad-
jacent bright fringes is 0.060 cm when the wavelength of the incident light is 500 mµ.
Compute the slit separation.
3. Determine the minimum thickness of a water film, such as a soap bubble, that
will give constructive interference (brightness) with light of wavelength 550 mµ.
4. When a thin water film is illuminated with light of wavelength 750 mµ, maxi-
mum reenforcement occurs; when the same film is illuminated with violet light, there
is destructive interference (darkness). (a) Determine the minimum possible thick-
ness of the film. (b) Compute the wavelength of the violet light. (Neglect the dis-
persion of water.)
5. In a Newton's rings experiment, the lens and plate are in "optical contact" at
the center, and the air film has zero thickness at that point. Determine the thickness
of the air wedge at the third bright fringe for light of wavelength 500 mµ.
6. What are the best values of the refractive index and thickness of a reflection-
reducing film for a rock-salt lens to be used with sodium light?
7. A simple optical instrument contains a total of 5 lenses and prisms, and hence
has 10 air-glass surfaces. Assume that loss of light is due only to reflection, and use
the data of Table I to compute the overall transmittance of the instrument (a) for un-
coated surfaces; (b) for coated surfaces, with all lenses and prisms of crown glass,
n = 1.52; (c) for coated surfaces, but all the optical elements of flint glass, n = 1.70.
8. (a) For what refractive index of glass will a reflection-reducing coating of re-
fractive index 1.30 be most effective? (b) For what wavelength will an 80 mµ thick-
ness of this coating be most effective?
9. While mirror A of the Michelson interferometer of Fig. 7 is moved through ex-
actly 0.20 mm, how many bright fringes of sodium light sweep past the cross hair
of the telescope?
638 OPTICS Chap. 37

10. A plate of thallium bromide-iodide is tested for thickness in a Michelson inter-

ferometer, using sodium light of wavelength 589.3 mµ. The plate is placed per-
pendicularly in the optical path of beam 2 in Fig. 7, and to compensate for this path
change the movable mirror A is advanced until 17,260 bright fringes have swept by
the reference cross hair. What is the thickness of the plate?
11. In an arrangement such as that of Fig. 10, a diffraction pattern is formed on a
screen by monochromatic light passing through a slit of width 0.015 cm and through
a converging lens of focal length 120 cm. Compute the linear width of the central
diffraction maximum for light having a wavelength of 600 mµ.
12. In a diffraction pattern formed in the focal plane of a positive lens of focal length
40 cm, the separation between the fourth minimum on one side of the central maximum
and the fourth minimum on the other side is 0.48 cm. The slit is illuminated with
monochromatic light of wavelength 450 mµ. Determine the slit width.
13. A double slit consists of two identical slits of width 0.08 mm, with centers
separated by 0.24 mm. The slits are illuminated by monochromatic light of wave-
length 480 mµ. After passing through the slits, the light falls on a screen located
60 cm beyond the double slit. (a) Find the distance on the screen from the center
to the first minimum of the single-slit diffraction pattern. (b) Which of the first ten
bright interference fringes on one side of the center are missing, because of diffraction?
(c) Illustrate the answer to part (b) by plotting a graph of screen intensity versus
distance from the axis, going out to at least the tenth fringe on one side and to the first
diffraction minimum on the other.
14. A double slit consists of two identical slits with centers separated by 1.00 mm,
and is illuminated by monochromatic light of wavelength 720 mµ. A screen is located
200 cm beyond the double slit. (a) Find the spacing of the interference fringes on
the screen. (b) If the central diffraction maximum contains 9 interference fringes,
what is the width of each slit? (Assume that the slit separation is an integral number
of slit widths.)
15. Light of wavelength 600 mµ falls normally on a diffraction grating having 7000
lines per cm; bright lines are observed on the screen located 40 cm from the grating,
as in Fig. 12. Compute the distance on the screen from the central image to the
second-order bright line.
16. A diffraction grating having 5000 lines per cm is illuminated with monochro-
matic light, and bright lines are observed on a screen 60 cm from the grating. The
distance on the screen between the central image and the first-order bright line is 10 cm.
(a) Compute the wavelength of the incident light. (b) Determine the highest order
image that can be observed for this wavelength with this grating.
17. A grating has 6000 lines per cm, and the ruled surface is 4 cm square. Parallel
light in a spectrometer is arranged to fall normally on this grating. (a) How many
orders of violet light of wavelength 450 mµ can be seen? (b) The angle between a
second-order red line and the central image is 60°; find the wavelength of this line.
18. A circular aperture 0.01 cm in diameter is illuminated with monochromatic light
of wavelength 600 mµ, and a bright disk is formed by a lens on a screen located 45 cm
from the aperture. Compute the diameter of this central diffraction disk.
· 19. Light of wavelength 700 mµ falls on the circular aperture of a converging lens,
and a central diffraction disk l.50 cm in diameter is forrned on a screen 60 cm distant.
Determine the diameter of the aperture.
20. An observer looks from a distance at two incandm;cent lamps that are separated
by 3 feet. At what maximum distance, measured normally to the plane of the lamps,
can he resolve them?
21. An aerial camera lens has a free diameter of 4 in. and a relative aperature of f/8.
How close are the smallest resolvable details in the image plane of the lens, for an
average wavelength of 550 mµ?
 PROBLEMS 639

22. The objective lens of a small telescope is 3 cm in diameter. (a) What is the
limiting angle of resolution, for a wavelength of 550 mµ? (b) What is the "normal
magnification" of the telescope: the magnification that utilizes the full resolving power
of the eye?
23. The smallest object detail that can be resolved with a certain microscope
measures 4.0 X 10-5 cm. The wavelength of the light used is 600 mµ. (a) Find the
numerical aperture of the objective when used "dry." (b) What numerical aperture
would be obtained by using an immersion oil of refractive index 1.48?
24. (a) Compute the theoretical resolving power, in the second order, of the grating
of Prob. 17. (b) Determine the wavelengths of two spectrum lines, with average
wavelength 650.00 mµ, that could just be resolved in the second order with this grating.
 33
POLARIZATION
The phenomena of interference and diffraction establish without doubt the
wave nature of light, but they do not reveal the character of the wave motion.
The fact that light can be polarized shows decisively that light waves are
transverse, that is, have displacements which are at right angles to the direc-
tion of propagation of the disturbance. The effects of polarization are
utilized not only in the scientific laboratory but also to an increasing extent
in everyday life.
33-1. Polarized Light. A longitudinal wave motion such as sound always
exhibits symmetry around the direction in which the waves are traveling,
since the vibrations are along the direction of propagation. In transverse
waves the vibrations are at right angles to the direction of propagation, and
there may not be symmetry around this direction. Waves lacking such sym-

tt 11 • ! ! II.
I II m
Fm. 33-1. Symbols used to represent beams of linearly polari:r.ed light (I and II), and
unpolari:r.ed light (III and IV)

metry are said to be polarized. Although the unaided eye cannot distin-
guish between polarized and unpolarized beams of light, simple accessories
can demonstrate the difference, and these show that light is often partially
polarized. The vibrations may be elliptical, circular, or linear, the first-
named being the general case and the last two being important special cases.
The polarization of light can be explained in terms of a wave theory, as is
done in Chap. 32 for interference and diffraction. By the electromagnetic
theory, the waves that transmit energy by radiation are composed of electric
and magnetic fields that vary continuously and rapidly with time; the electric
field-designated briefly as the electric vector-is the one used to specify the
vibration. When the electric vector varies in magnitude only, and not in
direction, the light is said to be linearly polarized (or plane polarized, as it is
sometimes called). Such light can be represented by double-headed arrows
or short lines drawn perpendicular to the direction of propagation, as shown
in part I of Fig. 1; these arrows represent the amplitude of the electric vector
640
§ 33-2 POLARIZATION BY REFLECTION 641

and indicate by their direction the polarization axis of the beam. The dots of
part II are the "end views" of such arrows that are oriented normal to the
page. Representations of unpolarized light are shown in parts III and IV of
the figure.
Unpolarized light may also be represented as in Fig. 2, wherein part I
suggests that unpolarized light may be considered as being made up of many
linearly polarized waves, with random orientation. Moreover, the several
component waves are assumed to be noncoherent (§ 32-2), that is, irregular in
their space and time phase relationships. Even though the representation
may be artificial, it is useful in picturing how linearly polarized light may be
formed from unpolarized light-by
sorting out the linear vibrations
along a selected direction. Since
the vectors represent amplitudes of
the vibrations, components of all
vectors along any desired direction,
@+ Il
Fm. 33-2. End-view representations of
m
such as the direction of "permitted beams of unpolarized light (I and II), and
vibration" in a linearly polarizing of partially linearly polarized light (III)
device, may be taken in order to
compute the amplitude and hence the intensity of the light transmitted by
the device. In such analytical treatments the multiplicity of vectors of part l
can be replaced satisfactorily by the two mutually perpendicular vectors of
part II, again with the requirement of noncoherence; this explains the repre-
sentation of parts III and IV of Fig. 1. Part III of Fig. 2 shows a convenient
way to indicate light that is partly linearly polarized and partly unpolarized.
There are four different mechanisms by which linearly polarized light can
be produced from natural, unpolarized light; these are described in the follow-
ing four sections. Unless specifically stated otherwise, as in § 33-9, the term
"polarized light" will be used to signify linearly polarized light.
33-2. Polarization by Reflection. The polarization of light by reflection
was first noted by the French engineer Etienne Malus (1775-1812), who dis-
covered in 1808 that with natural light incident in air on a glass surface, the
reflected beam is partly polarized. It can easily be shown that the amount
of polarization varies with the angle of incidence and is a maximum for a
certain value called the polarizing angle. At this angle the linear polarization
of the reflected beam is complete, and the direction of vibration is normal to
the plane of incidence. The value of the polarizing angle was found experi-
mentally to depend upon the refractive indexes of the reflecting material and
of the medium in which it is located; the discovery was made in 1812 by the
Scottish physicist Sir David Brewster (1781-1868). The relationship, called
Brewster's Law, expresses the tangent of the polarizing angle ip as the ratio
of the two indexes, thus
(33.1)
642 OPTICS Chap. 33

where n1 is the refractive index of the medium in which the incident ray
travels, and n2 is the refractive index of the medium at the surface of which
reflection takes place. Brewster made the further discovery that when light
is incident at the polarizing angle, the reflected and refracted rays make an
angle of 90° with each other, as shown in Fig.
3. This fact is useful in deriving Eq. 33.1,
and confirms the experimental observation
that the reflected ray is totally linearly polar-
ized, with the electric vector perpendicular to
the plane of incidence. The refracted ray is
only partially polarized, with a preponder-
ance of light polarized in the plane of incidence.
In order to obtain a stronger reflected beam,
and also a higher degree of polarization of
FIG. 33-3. A ray of light is po-
larized by reflection at the po- the refracted beam, a pile of several similar
larizing angle glass plates can be used.
The amount of light reflected at a boun-
dary between two mediums of different refractive index depends on the
state of polarization of the incident light beam as well as on the angle of inci-
dence. Figure 4 gives Curves P and Q which show, as a function of angle of
incidence, the relative reflection in air at a glass surface for incident light
polarized perpendicular and parallel, respectively, to the plane of incidence.

100
-...
c:
Cl)
v
80
J.1
Cl)
a. 60 IZ
/ '/
II)

v//
v

-c:
0 40

-- lo
v
Cl)
c.::
Cl) 20
y /.
1./';,\

- ~ --- ~~y
IX

0
00 30° 60° 90°
Angle of incidence
FIG. 33-4. Per cent reflectance at interface between air and glass of refractive index 1.50.
Curve Pis for incident light polarized perpendicular to plane of incidence; Curve Q is for
incident light polarized parallel to plane of incidence; Curve A is for incident unpolarized
light

The curves indicate clearly that for an angle of incidence equal to the polar-
izing angle of about 56°, no light is reflected that is so polarized that the
electric vector lies in the plane of incidence; thus all the light in the reflected
beam is polarized at 90° to this plane, as noted above. The dashed curve,
intermediate between the other two, is Curve A of Fig. 28-12, for unpolar-
ized incident light.
§ 33-3 POLARIZA'l'ION BY DOUBLE' REFRACTION 643

33-3. Polarization by Double Refraction. Many crystals possess the prop-

erty of double refraction or birefringence, by which a single incident beam is
split into two refracted beams. The natural crystals calcite (or Iceland spar
-CaC03) and quartz (Si02) are doubly refracting, as are such synthetic
crystals as sodium nitrate (NaN03) and ammonium dihydrogen phosphate
(NH4H 2P04). ·when a beam of unpolarized light is incident normally upon
the end face of a calcite crystal, the ray is split as shown in Fig. 5. Ray O
passes through the crystal without deviation-it is called the ordinary ray
because it obeys the law of refraction. Ray Eis deviated despite its perpen-
dicular incidence and hence is called the extraordinary ray. The emergent
beams are parallel to the incident beam and to each other; if the crystal is
rotated around the O beam, the E beam rotates with the motion. Each beam
is polarized, and the two polarization axes are mutually perpendicular.

',--+-+--+--~ E
~-1--~4-J.~=::;:::=-~~--~-1----~--.o
\ \
\ \
\ \
\
\

FIG. 33-5. Double refraction in calcite. The direction of the optic axis is denoted by
the dashed lines

The calcite crystal has six faces that form a rhombohedron, and the three
obtuse face angles that meet at the two opposite blunt corners are 102°. A
line through one of these blunt corners that makes equal angles with the three
faces meeting there, or any line parallel to it, is called the optic axis of the
crystal. The dashed lines in Fig. 5 and in subsequent figures indicate the
optic axis of calcite; it is to be noted that this is a direction in the crystal
rather than a specific line. A plane that contains the optic axis and the
normal to any face is called a principal section; the view of the calcite crystal
at the left in Fig. 5 is a principal section.
The direction taken by the rays within the crystal depends upon the angle
of incidence and the orientation of the optic axis. The speed of the ordinary
ray does not depend upon its direction, and the refractive index, no, of cal-
cite for this ray has the constant value of 1.658, as measured for the sodium
D line. The speed of the extraordinary ray varies with its direction in
the crystal. If this ray happens to advance along the optic axis, its speed
is the same as that of the ordinary ray, and the two rays coincide. If it
travels in any other direction its speed is greater, with a maximum value in a
direction at right angles to the optic axis. The refractive index of calcite for
the extraordinary ray, as measured for the D line, varies from 1.658 along the
optic axis to 1.486 at right angles to the axis. The refractive index for the
644 OPTICS Chap. 33

extraordinary ray is customarily defined as the value that is farthest from

the refractive index for the ordinary ray, and this is the value quoted in
tables for nE.
The shape and location of the wave fronts within the crystal can be found
by using secondary wavelets, following Huygens' construction. For the or-
dinary beam the wavelets are spherical. For the extraordinary beam they are
ellipsoids of revolution (that is, spheroids) in which the minor axis of the
ellipse is equal in magnitude to the radius of the sphere and is parallel to

II m
Frn. 33-6. Ordinary and extraordinary wave fronts in different planes within calcite
crystals that are cut parallel and perpendicular to the optic axis. In parts I and III,
the optic axis (indicated by dashed lines) is in the plane of the figure; in part II it is per-
pendicular to that plane and is indicated by dots

the optic axis, in which direction the wavelets for the two beams coincide.
The relative shapes, in calcite, of the O and E wave fronts in three mutu-
ally perpendicular planes are shown in Fig. 6; the traces of the wave fronts
are seen to be either circular or elliptical. A more general situation is
shown in Fig. 7, wherein the optic axis is neither parallel nor perpendicular
to a crystal face. A plane wave AB is shown advancing toward the crystal
surface MN, and the wave fronts for the ordinary and extraordinary rays are
constructed as shown at CD and CF, tangent respectively to the spherical

E O
Frn. 33-7. Double refraction of an unpolarized plane wave incident obliquely on the
surface of a calcite crystal. The ordinary ray is perpendicular to the corresponding wave
front; the extra,orqinary ray is not
§ 33-3 POLARIZATION BY DOUBLE REFRACTION 645

and ellipsoidal wavelets sent out from each point where the wave strikes the
surface. The corresponding rays are directed from A to the points of tan-
gency; for the extraordinary wave, the rays are not necessarily perpendicular
to the wave front.
In crystals of certain doubly refracting substances, such as quartz, the speed
of the extraordinary ray (except along the optic axis) is less than that of the
ordinary ray, instead of greater. For these crystals, the ellipsoid representing
the extraordinary wave front has its major axis coincident with the radius
of the spherical wave front of the ordinary beam and lies inside the sphere.
These are called positive uniaxial crystals and those of the calcite type are
called negative uniaxial crystals. Refractive indexes of crystals of both types
are given in Table I. Certain so-called biaxial crystals have two optic axes.
and represent the general case of optical anisotropy.
TABLE I. REFRACTIVE INDEXES OF UNIAXIAL BIREFRINGENT
CRYSTALS, FOR SODIUM LIGHT

Crystalline Material no

Calcite ........................... . 1.658 1.486

Quartz ........................... . 1.544 1.553
Sodium nitrate .................. . 1.587 1.336
Ammonium dihydrogen phosphate ... . 1.525 1.479
Sapphire ......................... . 1.760 1.769
Ru tile ........................... . 2.616 2.903

Linearly polarized light can be produced by a doubly refracting crystal if

one of the beams can be separated entirely from the other. This was first
accomplished by the Scottish physicist William Nicol in 1828. The Nicol
prism is prepared from a calcite crystal of the shape shown in Figs. 5 and 7,
and about three times as long as it is wide, by resurfacing the end planes of the

Fm. 33-8. The Nicol prism (principal section)

crystal from an angle of 71 ° to one of 68°, sawing the crystal in two, diagonally
at right angles to these end planes as shown in Fig. 8, and cementing the two
parts together with an adhesive called Canada balsam. The refractive index
of the balsam for D light is 1.54, a value intermediate between the ordinary
and extraordinary refractive indexes, no and nE, of calcite. The action of
the Nicol prism is as follows: An incident ray of unpolarized light is separated
646 OPTICS Chap. 33

as usual into ordinary and extraordinary rays. The ordinary ray strikes
the balsam interface at an angle greater than the critical angle, and hence
this ray is totally reflected; it passes to one side, out of the prism, where
it is absorbed by a coat of black paint or a piece of black velvet. The extraor-
dinary ray is transmitted as polarized light, its vibrations being in the plane
of the principal section, as shown.
The limitations of the Nicol prism are its restricted cross-sectional area
due to the scarcity of large calcite crystals, its narrow angular field of useful
operation, and the increase of optical path length, which may cause defocusing
when the prism is introduced into the light beam. Other types of polarizing
prisms have similar limitations. Theoretically, the Nicol prism transmits
about 46 per cent of the incident light, half being lost by the deviation of the
ordinary beam out of the light path and about 4 per cent by reflection at
the entrance and exit surfaces. Its transmittance is uniform across the visible
spectrum.
33-4. Polarization by Dichroism. Some doubly refracting crystals absorb
one transmitted ray more than the other. Thus in a crystal of sufficient
thickness, one ray is almost completely absorbed whereas the other is trans-
mitted with little reduction in intensity, as illustrated in Fig. 9. Such crys-
tals are called dichroic; an example is tourmaline, which absorbs the ordinary
ray more strongly than the extraordinary ray. Tourmaline crystals serve as
satisfactory demonstration pieces to illustrate linear polarization, but they
are seldom used in polarizing instruments because of the rarity of large, clear
samples, and because the transmitted beam is of ten highly colored.

Frn. 33-9. Illustrating the production of a linearly polarized beam from natural light by a
dichroic crystal

Synthetic dichroic polarizers of large area have been developed by the

American scientist and inventor Edwin H. Land, founder of the Polaroid
Corporation; this development was undertaken while he was an undergraduate
at Harvard College. Sheet polarizers have been produced commercially since
about 1936, and their availability has done much to expand the control of
light and to provide practical applications of polarization that are impossible
with the older crystal polarizers.
The earliest type of synthetic polarizer depends for its action on dichroic,
§ 33-5 POLARIZATION BY SCATTERING 647

submicroscopic needles of a crystalline form of quinine sulfate per-iodide

named herapathite after its discoverer, the English physician and chemist
William B. Herapath (1796-1868). The needles are put into suspension in a
viscous medium that is then caused to flow through a long, narrow slit. This
process yields a sheet in which the needles are oriented parallel to a common
direction determined by the streamlines in the flow process. The dried po-
larizing film, only about 0.001 in. thick, is immediately placed between pro-
tecting cover sheets of plastic or glass.
A later type is made by absorbing iodine in a stretched sheet of polyvinyl
alcohol, a clear, colorless plastic which becomes rubbery when heated. The
stretching process produces the required orientation, and the absorbed poly-
meric iodine the dichroism. Such polarizers, in which the polarizing unit is
molecular in dimensions, scatter light to
a much less degree than do the earlier 50 I \
microcrystalline polarizers. A third c
~ 40
I '\ Nic~I I
type of sheet polarizer is made by heat- ...Q)
. /n
I
I
~
\
_, prism_......
ing a stretched film of polyvinyl alcohol ~ 30
u /sheet I
in the presence of an active dehydration c po Iarizer
. · II
~ 20 \
catalyst, whereupon the film darkens .E \
I \
slightly and acquires dichroism, because Ill I \
§... 10 I
of the formation of a long-chain conjug- I- \
ated single- and double-bond molecular 0 ·"'
/
/
...
rs.,

system. This polarizer is very stable 400 500 600 700

Wavelength, millimicrons
against the action of high temperature
and humidity and hence is selected for Frn. 33-10. Transmission curves for a
many outdoor uses. Nicol prism and a sheet polarizer (not
corrected for surface reflection). The
A dichroic polarizer transmits 25 to dotted curve shows at half scale the rela-
40 per cent of the incident light; its tive sensitivity curve of the photopic eye,
transmittance is somewhat lower than for comparison
that of the Nicol prism because the
transmitted component is partly absorbed. This absorption varies with wave-
length through the visible region, giving a transmission curve such as that of
Fig. 10; the nearly straight-line curve of the Nicol prism is also shown.
33-6. Polarization by Scattering. An examination of the blue sky with a
polarizer shows that light from the sky is partially polarized, and that the
effect is greatest when the observer is looking at 90° from the direction to
the sun. The polarization results from the scattering of light by submicro-
scopic particles, such as dust specks or gas molecules, in the atmosphere. In
Fig. 11 a beam of light from the sun is shown incident upon scattering par-
ticles, and the scattered light is tested at 90° to the incident beam. Since
there can be no component of displacement (of the electric vector) along the
path of the light, either from the sun to the scattering particles or from the
particles to the eye, the vibration direction of the scattered light is established
by this simple test. In fact, such an observation, made with a polarizer
648 OPTICS Chap. 33

placed before the eye, provides the simplest fundamental method for deter-
mining the "transmission direction" or polarization axis of the linear polarizer.

From sun

Fm. 33-11. Determination of the vibration direction of scattered light.

The dots represent scattering particles

33-6. The Law of Malus. Polarizing elements, such as Nicol prisms and
Polaroid disks, are often used in pairs as variable-density light filters. One
is held fixed and the other is rotated around the beam of light, and the inten-
sity of the beam transmitted by the pair can be varied between a maximum
and a minimum value, with successive maxima and minima occurring go0
apart. The maximum intensity of the light transmitted through two perfect
polarizers would be 50 per cent of the incident (unpolarized) value, except
for reflection losses at the four surfaces.
The variation of the transmittance of a pair of polarizers with the angle of
rotation was first expressed by Malus, and its mathematical statement usually
bears his name. For perfect polarizers, the intensity of the light transmitted is
Io = Im cos2 e, (33.2)
where e is the angle between the polarization axes of the two polarizers, and
Im is the (maximum) intensity transmitted for () = 0, and cos () = 1. This
law can be derived by taking the component of the amplitude of the electric
vector transmitted by the first polarizer along the transmission direction of the
second and squaring amplitudes to obtain intensities.
Actual polarizers do not reduce the intensity of the transmitted beam quite
to zero when they are in the crossed position. If Ix denotes the transmittance
fore = go 0 , and Im the transmittance for the "parallel" position (e = 0), then
the more accurate relationship is
Io = Ix+ Um - Ix) cos 2 e. (33.3)
The term Ix is usually so small for practical polarizers-from 0.1 to 0.001
of 1 per cent-that this more exact equation is used only when fJ is very close
to 90°.
Because of the low transmittance of a pair of crossed polarizers, it is con-
venient to use a logarithmic scale for the ordinate of the transmission curve,
or to express transmittance in terms of optical density, § 31-11. The curves
§ 33-7 DOUBLE-I1lIA.GE PRISMS G49

of Fig. 12 illustrate the transmittance of a modern polyvinyl alcohol-iodine

polarizer: Curve A is for a single filter, and the other curves for a pair of
filters, B when in parallel orientation, and C when crossed. The higher trans-
mittance of the crossed pair at the
short-wavelength end of the visible 5 0.001
spectrum accounts for the deep blue
v_. c c: -
extinction color that is typical of ·;;;>-
this polarizer.
4

~ 3/
'U
-
./
O.Dl (!)u

0.1 Cl)
...(!)
a.
33-7 Double-image Prisms. u
c:
0
Some direct-reading spectrophotom- ·¥ 2 ....0
::
a. E
eters, and certain other photome- 0
r---.... B 10 "'
c:
tric instruments, utilize polarization .___ 0
....
I-
for comparing light intensities. Use 0 A 100
is made of double-image prisms that 400 500 600 700
Wavelength, millimicr9ns
split a single beam of natural light
into two beams polarized at right Fm. 33-12. Transmission curves for a typ-
angles to each other. These prisms ical sheet polarizer, with optical density as
are ordinarily made of calcite or of the ordinate. Curve A is for a single polar-
izer, Curve B for two with polarization axes
crystal quartz. Each prism is cut parallel, and Curve C for two with polariza-
into halves, and the two parts may tion axes crossed. Per cent transmittance
be cemented with glycerine or some values are given at the right
other material that transmits in the (Co·urtesy of the Polaroid Corporation)
ultraviolet, rather than with Canada
balsam, to cover a wider wavelength range. Two common types are the
Rochon prism and the W ollaston prism, shown diagrammatically in Fig. 13.
Each is constructed of two pieces of calcite properly cut and cemented so that
the optic axes are at right angles to each other. The student should examine

E E
--,,.....,_;~---1"----.......-==~-1---1--~o

. . .· : .. ..
0

Rochon Wollaston
Fm. 33-13. Double-image prisms of calcite. The direction of the optic axis is indicated
for each half by dashed lines or dots

the figure to discover why the Rochon prism gives one deviated and one
undeviated beam, whereas the W ollaston prism gives two beams that are de-
viated oppositely and equally. With the Rochon prism the undeviated beam
is achromatic, that is, free from colored fringes when white light is employed.
·when theE beam is screened off, the Rochon prism serves as a compact device
for producing polarized light.
Photometric measurements are made by placing a fixed polarizer in the light
G50 OPTICS Chap. 33

beam beyond the Rochon prism and before the eye, or a photoelectric cell
that serves as the indicator. The intensity of the light transmitted by the
system depends on the angle between the transmission directions of prism and
polarizer, as given by Eqs. 33.2 and 33.3. The use of a Wollaston prism
permits the "unknown" beam and the "standard" or comparison beam to be
conveniently located side by side. An intensity match is made after the light
has passed through the samples under test, by rotating the second polariza-
tion device through the appropriate angle. Often "null" devices are employed
in which the preferred setting is that for minimum light intensity, correspond-
ing to the crossed position of the polarizers.
33-8. Interference of Polarized Light. When two aligned linear polarizers
have their polarization axes at 90° and unpolarized light is incident on the
first polarizer, no light is transmitted by the second.
However, if a doubly refracting crystal or other birefrin-
E
I gent material such as Cellophane is inserted between
I
I the two polarizers, light will ordinarily be transmitted by
the second polarizer. If several sheets of Cellophane are
inserted together, the transmitted light may be highly
colored. These results are caused by the interference of
polarized light, a phenomenon somewhat similar to the
interference of unpolarized light treated in Chap. 32, but
with the added complication that the direction of polari-
Fw. 33-14. Resolu-
tion, by a doubly
zation must be considered, since two beams polarized at
refracting crystal right angles to each other will not interfere.
plate, of incident To explain how interference occurs with polarized light,
linearly polarized the beam normally incident upon a birefringent plate will
light. The com- be resolved into two components, one parallel to the optic
ponents are polar-
ized parallel and
axis and the other at right angles thereto. In Fig. 14
perpendicular to the amplitude of the incident beam is represented by rand
the optic axis of the its vibration direction by AB, the angle between AB and
crystal the optic axis being represented bye. When the beam is
resolved into components, the amplitudes of the ordinary
and extraordinary rays, A.O and AE, are found to be, respectively,

ro = r sin e,
and rE = r COS 8.

These rays travel through the crystal with different speeds and are out of
phase with each other when they emerge, because their optical paths within
the crystal are different.
In Fig. 15 the two polarizer disks P 1 and P2 are shown in part I in the
crossed position, centered on a beam of light. A crystal plate C is placed
between them and turned so that its optic axis makes an angle e with the
vibration direction of the incident polarized light, as previously. If the am-
§ 33-8 INTERFERENCE OF POLARIZED LIGHT 651

plitude of the incident beam is r, its components will have amplitudes of ro and
rE, as indicated in part II. When these components strike the second polar-
izer, P 2, they make angles of 8 and (90° - 8), respectively, with the trans-
mission direction of that polarizer. The amplitudes of the rays transmitted
by P 2 are obtained by again taking components of the amplitudes along the
transmission direction. The results are
E' = ro cos 8 = r sin 8 cos 8,

and E" = rE sin 8 = r cos 8 sin 8,

showing that the components are equal in magnitude.

P, c

FIG. 33-15. Interference of polarized light, produced by a birefringent crystal. Part I

shows a crystal plate placed between crossed polarizers, and part II shows front views
of these optical elements

The vibration directions of the two beams transmitted by P2 are along the
same line-the transmission direction of P2; the rays are coherent, since they
come from the same original beam, and hence they are in a condition to inter-
fere. The results of this interference depend upon the two refractive indexes
of the crystal plate, the thickness of the plate, and the orientation of its optic
axis to the incident polarized light. For certain values of these variables,
and for a certain wavelength, the final beams E' and E" may be exactly 180°
out of phase, and there is destructive interference. But the refractive indexes
vary with wavelength and, for some other wavelength, the t\vo beams may
reenforce each other. Thus, with white light incident on the first polarizer,
brilliant colors may be produced on a screen located to the right of the second
polarizer; when the second polarizer is rotated through 90°, the comple-
052 OPTICS Chap. 33

mentary colors appear. Many beautiful effects can be produced with inter-
ference by using appropriate birefringent materials between two crossed or
parallel polarizers-a so-called "polarizer and analyzer" pair.
The arrangement described is useful in the examination of crystals and in
the study of thin rock sections. When viewed with a microscope, the latter
usually appear colored irregularly and as a result reveal details and structural
differences not readily observed by other means. Such examinations enable
the mineralogist to determine whether a crystal is uniaxial (like quartz) or
biaxial (like mica), and whether it is optically active (§ 33-10) and how much
its plane of polarization rotates; also, he can measure its refractive indexes.
\Vith this information he is able to specify the crystal system to which the
sample belongs, and is assisted in identifying it.

Fm. 33-16. Strain pattern of a bar in flexure, as shown in polarized light

(Courtesy of the Westinghonse Electric Corporation)

A plate of singly refracting, or isotropic, material placed between two

polarizers produces no interference effects, since the incident ray is not sepa-
rated into two rays. Many singly refracting materials, such as glass, celluloid,
Bakelite, and certain plastics, become doubly refracting when strained. Con-
sequently, these substances can be tested with polarized light, and the presence
of strains, not apparent when viewed in natural light, will be revealed by the
appearance of interference fringes. Either white or monochromatic light may
be used in the tests. Figure 16, for example, shows the strain pattern set up
when a rectangular bar, supported at the ends, is bent in flexure by a force
applied at the center.
The Scottish physicist John Kerr (1824-1907) discovered that some trans-
parent insulators become doubly refracting in a strong electric field. This
electro-optic effect is unusually large for nitrobenzene, and hence this liquid
is ordinarily used in so-called Kerr cells, in which a suitable length of the
liquid is placed between two condenser plates. When the cell is located in
a light beam between two crossed polarizers and an electric field is estab-
lished between the two plates, the liquid behaves optically like a birefringent
crystal with the optic axis parallel to the field direction. Since light passes
transverse to this direction, interference effects occur which are similar to
§ 33-9 ELLIPTICALLY AND CIRCULARLY POLARIZED LIGHT 653

those discussed above, and under suitable conditions the presence of the field
causes light to be transmitted by the second polarizer. With a rapidly
alternating field the arrangement acts as a fast-operating light shutter; the
effect has been used in measuring the speed of light and for many other pur-
poses. Certain crystals such as ammonium dihydrogen phosphate also ex-
hibit this electro-optic effect and can be used in optical shutters operating at
frequencies up to at least 100 megacycles per second.
33-9. Elliptically and Circularly Polarized Light. Attention has been
given thus far exclusively to linearly polarized light, because of its relative
simplicity and its many important applications. But elliptically polarized
light is perhaps equally common. The light leaving the crystal plate shown
in Fig. 15 is actually polarized elliptically, since the components represent
two vibrations of the same frequency but of different amplitudes-r 0 and rE;
these are coherent and hence can combine in the manner discussed in § 26-14.
The result of the combination is a 1 :1 Lissajous figure, as shown in the top
row of Fig. 26-17. If the amplitudes of the two components happen to be
equal and if, in addition, the thickness of the crystal is such that the phase
difference between the two vibrations is 90° or 270° (11'/2 or 3 1r/2 radians), a
circular vibration results, and circularly polarized light is produced. If the
phase difference is 180° or 71' radians, the result is linearly polarized light with
the electric vector oriented at 90° to that of the linearly polarized light inci-
dent on the crystal plate.
· The path difference .Ii between the two beams transmitted by a crystal
plate arranged as in Fig. 15 is given by

.Ii = t (nE - no), (33.4)

where t is the thickness of the crystal, and nE and no are the extraordinary
and ordinary refractive indexes, as before. The term in parentheses will be
positive for positive birefringent crystals (§ 33-3) and negative for negative
crystals. The phase difference ois given in radians by 21r..i/>.., or
2 71'
o= T t (nE - no), (33.5)

where~ is the wavelength of the light. If the plate is of such a thickness that
it produces a path difference of one quarter of a wavelength, it is called a
qiwrter-wave plate. When such a plate is oriented so that its optic axis is 45°
to the vibration direction of the incident linearly polarized light, the ampli-
tudes of the two transmitted beams are equal and, since these are 90° out of
phase, they produce circularly polarized light. It is clear from Eq. 33.5 that
quarter-wave and other retardation plates give perfect results only for a single
wavelength. When white light is to be used, a plate thickness appropriate
for a wavelength in the green is ordinarily chosen, and the effect is less com-
plete for colors at the ends of the visible spectrum.
(354 OPTICS Chap. 33

33-10. Optical Rotation. It has been stated that when light travels along
the optic axis of a birefringent crystal, no double refraction takes place since
the O and E beams traverse the crystal at the same speed. While this is
true for most crystals, a few exhibit optical rotation or rotatory polarization,
an effect discovered early in the 19th century. If a beam of linearly polarized
light passes along the optic axis of crystalline quartz, the plane of polarization
is rotated about the transmission direction; the amount, for sodium light of
wavelength 589.3 mµ, is 21.72° for each
mm of thickness. For some samples of
.......
quartz the rotation is right-handed, and
for others it is left-handed. The amount
of rotation depends upon the wavelength
of the light, being greater for the shorter
. )400V wavelengths; this variation brings about
a so-called rotatory dispersion, Fig. 17.
.. . : ./ ..
Although the term optic axis does not
-~· apply to liquids, there are some that
demonstrate optical rotation-they are
----·-- - - ~ - - - - - -
often termed "optically active" liquids.
Frn. 33-17. Illustrating the rota- Examples are turpentine, in which the
tory dispersion of crystalline quartz.
The vertical arrow represents the
plane of polarization is rotated 37° for a
orientation of the incident polar- 10-cm path, and aqueous solutions of
ized light, and the other arrows the sugar, in some of which the rotation is
relative amounts of rotation for the right-handed ("dextrose") and in others
several wavelengths of light, with left-handed ("levulose"). The amount of
letters indicating colors. The dots
represent the optic axis
rotation by a sugar solution is measured by
its specific rotation or rotatory power, p,
which is defined as the rotation produced by a 10-cm column of liquid that
contains 1 gm of sugar per cubic centimeter of solution. Hence, the amount
of rotation <p that is produced is
pld (33.6)
<p = -,
10

where l is the length of the light path in the liquid, and d the concentration
in grams of sugar per cubic centimeter of solution. Since the angular rotation
depends markedly upon wavelength, measurements are ordinarily carried out
with monochromatic light-usually that of the sodium D line. The value of
p for cane sugar with sodium light is 66.45°.
Instruments called polarimeters are used in measuring the optical rotation
produced by solids or liquids. Those especially designed for testing sugar
are known as saccharimeters, and are arranged to pass the polarized light
through a tube filled with the aqueous sugar solution. If the polarizer and
analyzer are crossed initially, giving a dark field, the introduction of the tube
containing the sugar solution requires that the analyzer must be turned to a
§ 33-11 APPLICATIONS OF POLARIZED LIGHT 655

new position to restore the dark field. The angular difference between these
positions is the angle through which the plane of polarization has been rotated,
or differs from it by 180°. Sugar is rated commercially on the basis of purity
tests made with such saccharimeters.
Faraday discovered in 1845 that a block of glass becomes optically active
and rotates the plane of polarization of light when located in a strong magnetic
field. The effect is greatest for dense flint glass, but it has been observed
for many other substances. The Faraday effect can be observed by placing
the specimen between the poles of an electromagnet, the pole pieces having
been bored so that the polarized light can travel in a direction parallel to the
magnetic field. This magneto-optic effect can be used in fast-acting light
shutters and gives results similar to those obtained with electro-optic shutters,
§ 33-8.
33-11. Applications of Polarized Light. Techniques involving polarized
light have been employed in scientific laboratories for many years in po-
larimetry and photometry and in studying the Zeeman, Stark, and Raman
effects in spectroscopy. But since the advent of large-area polarizers of rela-
tively low cost, applications have become numerous in everyday life. Some
of these require only a single linear polarizer, whereas others employ pairs.
Unwanted reflected light is effectively controlled by a single polarizer in
sun-glasses and automobile day-driving visors. The polarizer is mounted
with its transmission direction vertical, so that it is crossed to the partly
polarized component of the reflected, glare light. The attenuation of this
component may greatly increase both the comfort of outdoor vision and the
efficiency of seeing, and is most noticeable when smooth surfaces such as wet
pavements are involved. A properly oriented polarizer attached to a reading
lamp similarly removes the glare reflection from the printed page. The illu-
mination of pictures in art galleries with properly oriented linearly polarized
light does away with the troublesome reflections which may mask the beauty
and detail of the subject. Photography of store-window displays is made
easier and more effective when the unwanted reflection from the window glass
is removed by properly orienting a polarizer placed in front of the camera.
Polarizing camera filters make it possible to take striking cloud pictures, for
the light from the clouds is unpolarized, whereas that from the blue sky back-
ground may be partly polarized: the sky is "darkened" by the filter and the
clouds stand out in exaggerated contrast. The partial polarization of sky
light is utilized to advantage in the polarization sky compass, a device that is
particularly valuable in the Arctic regions, where the magnetic compass is
unreliable and where sun and stars may be invisible for weeks at a time.
With this compass the direction to the sun may be determined with an ac-
curacy of 1°, even when the sun is below the horizon.
The most familiar use of polarizers in pairs is the so-called variable-density
unit that controls the intensity of a beam of light in accordance with the La11.·
of Malus, Eq. 33.2. This unit is used in photometry and also in many mili-
656 OPTICS Chap. 33

tacy optical instruments to decrease the brightness of the field of view. In

the sextant, § 28-4, it provides a convenient method for reducing the sun's
apparent brightness to that of the horizon sky. Variable-density sun-glasses
are available for persons who wish to adjust the apparent brightness to suit
their convenience, and some luxury trains are fitted with large windows of
this type, so that the passengers can choose the daylight illumination they
desire.
Polarized light is used in engineering and other fields for the rapid study of
stress distribution in structural and mechanical components. Scale models
are built from a plastic material that becomes birefringent when strained,
and these models are subjected to the loads and stresses to be investigated.
The strain patterns, such as that illustrated in Fig. 16, can readily be studied
either visually or photographically. In some types of this so-called photo-
elastic analysis the model is mounted between linear polarizers, while for
other types of analysis quarter-wave plates are added to produce circularly po-
larized light. Studies of the flow birefringence of liquids are carried out by
similar methods. Strain testing in glass, which is somewhat birefringent
when insufficiently annealed or slightly stressed, is a standard technique in
the glass industries.
The polarizing microscope, referred to in § 33-8 for use in mineralogy,
is also important in the medical sciences, since it has been found that speci-
mens viewed in polarized light often show
details that are quite invisible in ordinary
light. Any microscope can be converted to a
polarizing microscope by inserting one polar-
izing disk beneath the stage and another
within or directly above the eyepiece. A
microscope designed for use with polarized
light has a rotating stage, as illustrated in Fig.
30-15, and a removable polarizer between the
objective lens and the eyepiece.
The use of polarizing filters in front of the
headlights and on the windshields of automo-
biles in order to eliminate headlight glare at
night is illustrated in Fig. 18. The filters at-
tached to the headlights are oriented with their
Frn. 33-18. The use of linear polarization axes at 45° to the vertical, and a
polarizers in eliminating head-
light glare in night driving
windshield visor or special polarizing spectacles
worn by the driver are oriented identically, so
that the driver can see the road clearly by the
light of his own headlamps. The approaching car is similarly equipped, but
because of its reversed orientation the polarizers over its headlamps are
crossed with the polarizer in the windshield of the first car. Thus, to each
driver the headlights of the other car appear very dim, even when on "high
§ 33-11 APPLICA'l'IONS OF POLARIZED LIOII'l' Gi'i7

beam," and they do not contribute to poor visibility either of the approach-
ing car or of the road ahead.
Polarization is employed in obtaining three-dimensional views in photog-
raphy by using a double camera with the lens of each fitted with a polarizing
filter, but with the two filters oriented at right angles to each other. The
double camera takes views of the object from two slightly different positions,
as do the two eyes in stereoscopic vision. Prints from the two negatives are
spaced appropriately, perhaps by projection on a non-depolarizing screen, and
are viewed through polarizing filters that are properly oriented so that each
eye sees the appropriate image, thereby recreating the "depth" effect. This
technique, illustrated in Fig. 19, is used both in still and in motion-picture
photography.

Non-depolarizing
Polarizing
screen
viewers

~- --
--- --- --~
Twin projectors

Frn. 33-19. Illustrating the projection of three-dimensional pictures in polarized light

PROBLEMS
1. In order to produce complete linear r.olarization by reflection from a plate of
thallium bromide-iodide, at what angle of incidence should a beam of yellow light
strike the surface of the plate?
2. The polarizing angle for a certain sample of glass is 60°. Find the refractive
index of the glass. What type of glass is it?
3. A plate of glass of refractive index 1.52 is immersed in a liquid bath. Light
incident on the glass-liquid surface is reflected at an angle of 48° to the normal to the
surface, and is completeiy linearly polarized. Find the refractive index of the liquid.
4. Derive Brewster's Law, making use of the fact that the reflected ray and the
refracted ray are at right angles to each other.
5. From the curves of Fig. 4, calculate the ratio of intensities of the beams reflected
from a glass plate of refractive index 1.50 for two orientations of the linearly polarized
incident beam: perpendicular and parallei to the plane of incidence. Make computa-
tions of the intensity ratio for the following angles of incidence: (a) 10°, (b) 56°, (c) 75°.
6. Compute the speed of the ordinary light wave and also the maximum speed of
the extraordinary light wave in sodium nitrate, for sodium light.
7. Select one of the three "views" of Fig. 6 to represent the action of rutile and of
ammonium dihydrogen phosphate and draw to scale the ordinary and extraordinary
wave fronts emanating from a point source of light in each crystal. Indicate and label
the optic axis.
8. Justify the statement: "The Nicol prism transmits about 46 per cent of the in-
fi58 OPTICS Chap. 33

cident (unpolarized) light, on the assumption that there are no losses due to absorp-
tion inside the prism."
9. Draw figures similar to those of Fig. 6 but for a positive birefringent crystal.
Label the O and E wave fronts.
10. Refer to Fig. 10 and use the relation 2 T 2 for the transmittance of two parallel
polarizers; refer to curve C of Fig. 12 for the transmittance of two crossed polarizers.
Compute for each of the following wavelengths the ratio of the light intensity trans-
mitted by two pieces of sheet polarizer when parallel to that when crossed: (a) 420
mµ, (b) 525 mµ, ( c) 650 mµ.
11. A polarizer, Y, is placed bet,Yeen the observer and the scattering medium
of Fig. 11. Another polarizer, Z, is placed between the sun and the scattering med-
ium. Both polarizers are initially oriented so that the observer sees a bright field.
(a) Polarizer Y is rotated through 90°; is the field seen by the observer bright or dark'!
(b) Polarizer Z is now rotated 90°; is the field bright or dark?
12. Derive the expression for the Law of Malus, and make an appropriate diagram.
13. Two polarizers are oriented with their polarization axes at an angle of 60°.
Compute the intensity of the light transmitted, on the assumption that the intensity
of the transmitted beam is 30 arbitrary units when the polarization axes are parallel.
14. Two sources of light, A and B, are viewed one after another, through two Nicol
prisms. The intensities of the emerging beams in the two observations are equal when
the angles between the transmission axes of the prisms are 45° and 30°, respectively.
Find the ratio of the intensity of source A to that of B.
15. With two imperfect polarizers in the crossed position the transmittance I"' is
0.1 per cent, and in the parallel position the transmittance Im is 40 per cent. Find
the intensity when the angle between the polarization axes of the two polarizers is 85°,
using (a) Eq. 33.2, and (b) Eq. 33.3. Find the percentage error of answer (a) with
respect to answer (b).
16. Solve Prob. 15 for an angle of 60° between the polarization axes.
17. Explain why the left-hand half of the Rochon prism of Fig. 13 is not made of an
isotropic material such as glass.
18. "As the light beam goes from one half to the other half of a Wollaston prism, the
O beam becomes an E beam and vice versa." Is this statement true or false? Explain.
19. Compute the thickness of quarter-wave plates (for sodium light) made of
(a) quartz, (b) calcite.
20. Compute the minimum thickness of the ammonium dihydrogen phosphate plate
which when placed between two "parallel" polarizers will cause annulment of sodium
light. This is a half-wave plate.
21. A thin plate of birefringent material acts as a quarter-wave plate for light of
wavelength 750 mµ when inserted into the light beam just beyond a linear polarizer,
and with its optic axis at 45° to the transmission axis of the polarizer. (a) Describe
the state of polarization of the light leaving the plate. (b) Assume that the variation
of refractive index with wavelength can be neglected, and that monochromatic light
of wavelength 375 mµ is employed, and describe the state of polarization of light
leaving the plate.
22. An aqueous solution of cane sugar contains 200 gm of sugar per liter of solution.
Through what angle would a 20-cm column of this solution rotate the plane of polariza-
tion of sodium light?
23. In a common type of saccharimeter, the optical rotation produced by the sugar
solution is balanced by placing in the light path an appropriate thickness of crystalline
quartz. What thickness of "left-handed" quartz would just annul the rotation pro-
duced by the dextrose solution of Prob. 22?
24. Turpentine produces a "left-handed" rotation of linearly polarized light. What
l'Oncentration of cane sugar will produce an equal but opposite rotation?
 MODERN PHYSICS

34
QUANTUM OPTICS
An introduction to Modern Physics is conveniently divided into two parts:
Atomic Physics and Nuclear Physics. The first part, covered in this chapter,
deals only with those properties and behaviors of atoms in which the atomic
nucleus can be considered as stable and indivisible. Nuclear physics is the
newer area of investigation and deals with phenomena that concern subdivi-
sions of the nucleus; it forms the subject matter of Chap. 35. Almost all of
the phenomena of atomic physics involve the interaction of light with matter
and hence lie in the domain of Quantum Optics. The term "light" must,
however, be interpreted in its broader sense, as including radiation of all
wavelengths between those of gamma rays and of microwaves. The dual
nature of light is frequently illustrated-its behavior as a wave motion in
free space and in interaction with other light waves, and its behavior as a
particle when interacting with matter.
34-1. The Radiation Laws. The portion of the broad spectrum of electro-
magnetic radiation (see Table II in § 25-9) of greatest importance to man-
kind is that through which life-supporting energy reaches the surface of the
earth from our primary source of energy, the sun. The earth's atmosphere
is transparent to such radiation from a wavelength of about 300 mµ in the
ultraviolet, where ozone commences to absorb strongly, to from 1000 to
5000 mµ (1 to 5 µ) in the infrared, where the absorption is due to water vapor
and carbon dioxide. Energy reaches the earth from the sun by thermal
radiation; this type of transfer is discussed briefly in § 11-8. The laws of
radiant energy were evolved only after many years of experimental and theo-
retical study, and these investigations led at the beginning of the twentieth
century to the formulation of the quantum theory.
659
660 MODERN PHYSICS Chap. 34

·when radiant energy falls upon the surface of an object, some of the energy
is reflected and the remainder is partly transmitted and partly absorbed.
The part absorbed is converted into internal energy and raises the tempera-
ture of the object. ·with opaque bodies (for which the transmittance is
zero), the nature of the surface determines the relative amounts of energy
reflected and absorbed. A perfect absorber, one with zero reflectance, is
called a black body; it is approached in practice by a rough, dark surface, or
even better by a small hole in a hollow object~energy entering the hole has
an extremely small chance of being reflected out again; therefore almost
complete absorption results. It is found experimentally that the relative
amounts of energy radiated by hot objects at a given temperature depend
similarly on the nature of their surfaces: those with shiny surfaces are poor
radiators, whereas those with rough, dark surfaces radiate more effectively.
Kirchhoff's Law. The correspondence between good absorbers and good
radiators was first stated by Kirchhoff. The relationship, now known as
Kirchhoff's Law, states qualitatively that good absorbers are good radiators,
and vice versa. Thus a black body is not only a perfect absorber but is also
the best possible radiator at the given surface temperature; its radiation
depends in amount and wavelength distribution only on its temperature.
The other radiation laws stated below apply accurately to this most efficient
radiator, the black body. Objects that absorb or radiate less effectively, but
in constant proportion to black-body performance throughout the spectrum,
are called gray bodies. Objects for which the absorption varies within the
visible spectral range appear to the eye as colored, § 31-7.
The Stefan-Boltzmann Law. The intensity of the total radiation from a
black body over all wavelengths is given by
I= <rT4, (34.1)
where I is the energy radiated per second per unit area of radiating surface,
and T is the absolute temperature in degrees Kelvin. This relationship was
first stated by the Austrian physicist Josef Stefan (1835-1893), but was de-
rived independently by Ludwig Boltzmann (1844-1906), professor in the
University of Vienna; it is now known as the Stefan-Boltzmann Law.
Stefan's constant, <r, has been evaluated experimentally as 1.35 X 10-12 to
give I in calories per second per square centimeter, or 5.67 X 10-12 in terms
of joules instead of calories. The law can also be applied to gray bodies if
Stefan's constant is multiplied by the emissivity of the body; this factor
specifies the ratio of the total emission of the particular gray body to that
of the black body.
Equation 34.1, often called the Fourth-power Law, indicates the important
dependence of radiation intensity on the temperature of the radiating sur-
face. It is applied in the optical pyrometer used to measure the temperature
of very hot surfaces. In this instrument, shown schematically in Fig. 1, a
small telescope permits the observer to sight on the surface under measure-
§ 34-1 THE RADIA'l'ION LAWS 661

ment; this may be some relatively inaccessible area, such as the inside wall
of a furnace. A tungsten filament is located at the focal point of the objec-
tive lens of the telescope. The current through this filament, and thus the
filament brightness, can be regulated by the rheostat shown. The current is
varied until there is a brightness match between the filament and the object
being viewed, whereupon the filament disappears against its background.
The ammeter in the filament supply circuit is calibrated to give directly the
temperature of the object under test. To reduce errors due to color differ-
ence between test object and filament, a red filter may be inserted in the
eyepiece of the telescope.

-"'
.§ 200 t----1-+-......_---+-----+----l
~

Telescope -..
e
:.0
..-
0
Q)

i1ooi---tt-+-+-+----+-.,.._---~--1
-
c
0

l
0.5 1.0 1.5 2.0
Wavelength, microns

Fm. 34-1. The optical pyrometer FIG. 34-2. Spectral distribution curves
of the radiation emitted by black-body
sources at different temperatures. The
peak of the curve shifts to shorter wave-
lengths for higher temperatures

Wien's Displacement Law. Figure 2 shows curves, for several tempera-

tures, of the power radiated by a black body in a narrow wavelength band
~A, plotted against the wavelength at the center of that band. Upon meas-
urement, the areas under the curves are found to be proportional to the
fourth power of the absolute temperature of the surface, in accordance with
Eq. 34.1. The maximum point of the curve-that is, the wavelength Amax
for which the radiant power is greatest-shifts toward shorter wavelengths
as the surface temperature is increased; this relation was stated in quantita-
tive terms by the German physicist Wilhelm Wien (1864-1928) and is known
as the Wien Displacement Law:
T Amax = 2.898 X 106, (34.2)
where T is the temperature in degrees Kelvin and A is the wavelength in
millimicrons.
662 MODERN PIIYSICS Cha1). 34

Evidence of the shift of the radiant energy curve toward shorter ,vave-
lengths is familiar from the color changes of heated objects as their tempera-
ture is raised: the object first glows with a dull red luminosity, then a bright
red, then orange, then yellow, and finally approaches white.
Planck's Law. Many attempts were made to find the formula for the
black-body spectral distribution curves, such as those of Fig. 2. Success
was finally achieved in 1900 by }\,fax Planck (1858-1947), then Professor of
Physics in the University of Berlin, who expressed the intensity of radiatjon
per unit wavelength increment at absolute temperature T and wavelength >.
in the form:
(34.3)

The product of I and LiX is the energy per second per unit area of radiating
surface that is radiated in a narrow wavelength band LiX centered at>.. The
"radiation constants" k1 and k2 involve such universal constants as Avogadro's
Number, the general gas constant, the speed of light, and a new constant h
that carries Planck's name.
In order to derive Eq. 34.3, Planck made the radical assumption that a
thermal radiator behaves as if it contains a large number of harmonic electro-
magnetic oscillators, each of which has an energy that is an exact multiple of
a certain smallest energy determined by the frequency of the oscillator.
This so-called quantum of energy is given by
E = hf, (34.4)
where Eis in joules, his Planck's constant with the value 6.62 X 10-34 joule·
sec (or 6.62 X 10-27 erg· sec), and f is the frequency in vibrations per second.
This quantum hypothesis of the nature of energy, so at variance with classical
ideas, was first applied by Planck only to the possible energy states of the
harmonic oscillators and to their absorption and emission of energy. Its
extension to radiant energy itself constituted the birth of the Quantum Theory
of Radiation, which is so important in the interpretation of atomic and
molecular processes.
Although Planck's Law was the last of the radiation laws to be formulated,
it is the fundamental one from which the others can be derived: the Stefan-
Boltzmann Law is obtained by integrating Eq. 34.3 over all wavelengths, and
the Wien Displacement Law by differentiating it and solving for the wave-
length for which the radiation is a maximum.
34-2. The Photoelectric Effect. During the experiments which led in 1887
to the discovery of electromagnetic radiation, § 24-1, Hertz noticed that a
spark discharge occurred more readily across a gap when the surfaces of the
facing metals were illuminated than when they were in darkness. A year
later the German scientist Wilhelm Hallwachs (1859-1922), conducting
experiments with a zinc plate attached to a gold-leaf electroscope, observed
§ 34-2 THE PHOTOELECTRIC EFFECT 663

that, when light from a nearby electric arc was directed on this plate, it lost
its charge if charged negatively but not if charged positively; and also that
it acquired a slight positive charge when originally uncharged, upon irradia-
tion by light from the arc. He concluded that the illumination caused nega-
tive electricity to leave the plate. This effect, ordinarily called the photo-
electric effect, is now interpreted as the emission of electrons from the metal
surface as a result of its illumination. It is more properly called the photo-
emissive effect, in order to distinguish it from two other related effects of the
illumination of materials-namely, the photovoltaic effect and the photo-
conductive effect, § 27-8.
In one of the earliest experiments, a glass plate was placed between the arc
lamp and zinc plate and no photoelectric effect vms observed; this result indi-
cated that ultraviolet light of wavelength shorter than 350 mµ, the short-
wavelength transmission limit for glass, is needed in order to produce the
photoelectric effect in zinc, and hence suggested that the effect depends upon
wavelength.
The influence of the wavelength of the irradiating light on the emission of
so-called photoelectrons can be investigated with an apparatus similar to
that shown in Fig. 3. Light of controlled wavelength and intensity passes
through a quartz window into an evacuated chamber and strikes a freshly
prepared metal surface. The potential difference between the metal surface
and the collecting plate is regulated in direction and magnitude by a poten-
tiometer circuit. Appropriate instru-
ments measure the potential differ- Window
ence between plate and metal surface,
and measure the current carried by
photoelectrons inside the tube and
by the wire circuit outside. The Metal
Plate
maximum speed of the photoelectrons surface
emitted from the metal surface is
determined by putting a small nega- Potentiometer
tive potential on the plate and in-
rreasing its magnitude until it reaches
a value called the "stopping poten- Reversing
tial," for which the current drops to switch
zero, thus indicating that even the
most energetic photoelectrons have FIG. 34-3. Apparatus for studying the
been repelled. Many such experi- photoemissive effect
ments were conducted by the late
Professor R. A. Millikan, who reported his results in 1916. He found that
the stopping potential varies directly and linearly with the frequency of the
monochromatic light illuminating the surface, regardless of the intensity of
the light beam. The maximum kinetic energy of the photoelectrons is thus
proportional to the frequency of the incident light and is not a function of its
664 MODERN PHYSICS Chap. 34

intensity. For each metal surface there is a minimum frequency (or maxi-
mum wavelength) of the incident light beyond which the photoemissive
effect does not occur; for most metals the effect can be produced only by
ultraviolet light, but for the alkali metals it extends into the visible region,
and for cesium almost into the infrared. The results obtained in experi-
ments on photoemission depend greatly on the nature of the metal surface
and, for the best results, a fresh metal surface must be prepared just before
data are taken-in Millikan's experiments this was done by an ingenious
device that has been called a "lathe in a vacuum."
Another important experimental result was obtained when the potential
difference between plate and metal surface was increased to draw over to the
positively charged plate all the photoelectrons emitted-the condition of
saturation. The number of electrons emitted per second, as measured by the
circuit current, was found to be linearly proportional to the intensity of the
light beam, with wavelength held constant.
No adequate explanation of these experimental findings could be given in
terms of classical electromagnetic theory, but a satisfactory solution was
offered by the German theoretical physicist Albert Einstein in 1905 with the
announcement of his now-famous photoelectric equation:
hf= W + t mv2max· (34.5)
Here his Planck's constant, f the frequency in vibrations per second of the
monochromatic light striking the metal surface, m the mass of the electron
(9.107 X 10~31 kg), and Vmax the speed of the fastest-moving photoelectron in
meters per second. The symbol w denotes the work function of the surface-
the energy in joules necessary to remove a photoelectron from the metal
against the electric forces thus created. The work function can be expressed
as
w = hfo, (34.6)

where fo is the "threshold frequency" of the incident light-the frequency

corresponding to the longest wavelength that will produce the photoemissive
effect. Such threshold wavelengths for several metals are given in Table I.
These values apply when the metal surface is at a very low temperature; at

TABLE I. PHOTOE"MISSIVE THRESHOLDS

Threshold
Metal Wavelength

Tungsten ............... . 230 mµ

Zinc ................... . 302
Silver .................. . 325
Potassium .............. . 436
Sodium ................ . 583
Cesium ................. . 681
§ 34-2 THE PHOTOELECTRIC EFFECT 665

higher temperatures the electrons in the metal possess some thermal energy,
and a very few of them may be ejected as photoelectrons by light of fre-
quency smaller than the threshold frequency.
In photoelectricity experiments, the wavelength of the light is measured
rather than its frequency, but the frequency can be found from the relation
v = fX, Eq. 25.1, with v taken as the speed of light: c = 2.99790 X 108 m/sec
in a vacuum, § 27-4.
In his explanation of the photoemissive effect, Einstein extended Planck's
quantum ideas from the electromagnetic oscillators to the radiation itself and
postulated that electromagnetic radiation is emitted discontinuously by a
source, and that it travels through space in bundles, now called photons.
Each photon possesses an amount of energy dictated by its frequency as
E = hf, where h is Planck's constant and f is the frequency of the (mono-
chromatic) light. He further postulated that the absorption of radiant
energy in the photoelectric effect is quantized, meaning that the total energy of
a single photon can be communicated to a single electron, which may, in
consequence, escape from the metal surface as a photoelectron. These radi-
cal hypotheses were verified experimentally during the first few years of the
present century.
An alternate form of the photoelectric equation, in terms of the stopping
potential, V, is,
hf - hfo = t mv2max = Ve, (34.7)
where, to conform with mks units, V is in volts and e is the charge on the
electron: 1.602 X 10-19 coulomb.
Practical tubes based on photoemission, § 27-8, do not ordinarily employ a
pure metal as the cathode material but rather a combination of two metals,
sometimes separated by an oxide layer; higher sensitivities and perhaps lower
work functions are achieved in this way. Thus the antimony-cesium surface
used in most "blue-sensitive" phototubes gives high sensitivity in the region
around 400 mµ, and the bismuth-cesium surface gives good response toward
longer wavelengths. The best sensitivity in the far-red and near-infrared
regions is given by a thin surface of cesium oxide overlaid by evaporated
cesium, all on a metallic silver base; the sensitivity of this cathode extends to
a long wavelength limit of about 1200 mµ. Most phototubes are highly
evacuated, but a ten-fold increase in sensitivity is obtained by filling the
tube with an inert gas, such as argon, at a pressure of a fraction of 1 mm Hg.
The speed of response is decreased, however, by an amount which becomes
important at light modulation frequencies of several kilocycles per sec.
The second type of phototube-the photoconductive type-is also men-
tioned in § 27-8, where it is stated that detection sensitivities to infrared
wavelengths of about 7 µ can be obtained by the use of such cells. The third
general type of commercial phototube is the barrier-layer or photovoltaic
cell of the selenium or the cuprous oxide types, § 23-16; these cells, used
666 MODERN PHYSICS Chap. 34

chiefly in the visible spectral region, have the advantage that no external
source of emf is needed.
34-3. The Inverse Photoelectric Effect: X-ray Production. X-ray tubes
and circuits are discussed briefly in § 23-14, and details of the bright-line type
of x-ray spectrum are given later in this chapter. Such spectra can be pro-
duced by using a crystal as a diffracting grating, and may be recorded on a
photographic plate or measured with an ionization chamber. Although the
most striking features are the emission lines characteristic of the target mate-
rial, there is also a continuous spectrum with an intensity-wavelength dis-
tribution rather similar to the curve shape of Fig. 2. An important difference,
however, is that the x-ray continuous spectrum has a sharply-defined short-
wavelength limit, Amin, which is related to the potential difference V across
the tube by the simple expression
he
Ve = - - = hfmax , (34.8)
Amin

where the other symbols have their usual meanings. The equation does not
include a term representing the work function of the target material since
this is negligible in view of the large potential differences employed with x-ray
tubes.
A confirmation of the quantum nature of radiation is found in the field of
x-rays, wherein the physical interpretation of x-ray production is as follows:
Electrons emitted from the heated filament (cathode) in an x-ray tube are
accelerated toward the target (anode) by the large potential difference be-
tween target and filament. The greate8t kinetic energy that an electron can
acquire is given by !mv2max = Ve, similar to Eq. 34.7. When an electron
strikes the target, its kinetic energy will most likely be converted into heat.
But, as a result of the rapid deceleration, the energy of some few electrons is
converted to electromagnetic radiation (x-rays). The conversion of the
kinetic energy of the fastest-moving electrons to photons gives the shortest-
wavelength x-rays, and it is their wavelength and frequency that are given
by Eq. 34.8. This phase of x-ray production is hence seen to be the inverse
of the photoelectric effect: Photons are emitted when a metal surface is
bombarded with energetic electrons, and the kinetic energy of the electrons
reappears as that of the photons.
34-4. The Hydrogen Spectrum. The production of spectra from glowing
vapors is discussed briefly in § 31-2, but no explanation is given there of
the origin of the specific spectrum lines or bands. Although a complete
explanation is still lacking for the complicated spectra from the heavier ele-
ments, the quantum theory affords a partial understanding of this atomic
process for the lighter elements.
The Swiss scientist Johann J. Balmer (1825-1898) found a simple formula
in 1885 which gave the wavelengths of the several lines of the visible spectrum
of atomic hydrogen shown as the central spectrum in Fig. 31-4. This formula
§ 34-4 TIIE JIYDROOEN SPECTRUM 6G7

was entirely empirical, but its success in providing a mathematical connec-

tion among these lines initiated a vigorous program of study of the spectra
of other elements. A modern statement of his formula, extended to cover
not only the so-called Balmer series in the visible spectrum, but also other
series discovered later in the ultraviolet and infrared regions, is as follows:

f = 3 .288 x 1015 (__L

1vf.2
- N_!_),
2
(34.9)

where f represents the frequencies of the individual lines of the hydrogen

spectrum in vibrations per second, and M and N are integers. V{ avelengths
corresponding to the frequencies can be computed from the usual expression,
c = f>-..
The lines of the Balmer series are obtained when Jvl = 2 and N = 3, 4,
5 · · · ; those of the Lyman series in the deep ultraviolet region (discovered
by Professor Theodore Lyman of Harvard University) are obtained when
M = 1 and N = 2, 3, 4 · · · ; and those of the Paschen series in the near
infrared region (discovered by the German physicist Friedrich Paschen) when
M = 3 and N = 4, 5, 6 · · · . Lines belonging to three more infrared series,
for M = 4, 5, and 6, have been observed at longer infrared wavelengths than
those of the Paschen series.
A theoretical explanation of the discrete wavelengths of the Balmer series
was given in 1913 by Professor Niels Bohr of the University of Copenhagen.
He based his explanation on the so-called planetary model of the atom,
according to which the electrons whirl about a central massive nucleus in
orbits, § 15-2; classical theory would require an electron moving with such a
(centripetal) acceleration to radiate electromagnetic energy and hence spiral
into the nucleus. Bohr made two postulates which were contrary to classical
ideas, and which constituted a further extension of the quantum theory of
radiation already advanced by Planck and Einstein: first, that certain sta-
tionary states exist in which an electron can revolve about the central nucleus
without radiating energy, and second, that the atom emits or absorbs energy
only when an electron jumps from one of these orbits to another.
The particular electron orbits allowed are those for which the angular mo-
mentum of the rotating system is an integral multiple of Planck's constant
h divided by 2 1r. That is, the permitted orbits are those for which

nh (34.10)
mvr = - ,
2 11'
where mis the mass of the electron, vis its linear speed in the circular orbit of
radius r, and n is an integer called the principal quantum number.
A brief derivation yields an expression which gives the energy of the atom
for each permitted electron orbit. According to classical theory, the elec-
tron, of charge -e, is held in its path by the electrostatic force of attraction
between it and the nucleus, of charge +e. By Eq. 15.2, stated for the use of
668 MODERN PIIYSICS Chap. 34

rationalized mks units, the attractive force is 9 X 109 e2/r 2 and serves as the
centripetal force, mv 2 /r. When these expressions are equated, the result is
mv2/r = 9 X 109 e2/r 2 • This equation and the mathematical statement of
Bohr's first postulate, Eq. 34.10, can be solved simultaneously for r and v,
g1vmg:
1 n 2h2
r = --,
9 X 109 X 4 1r 2 me 2
e2
and v = 9 X 109 X 2 1r nh •

Now the total energy E of the orbital electron is partly kinetic energy and
partly (electrical) potential energy, thus:
mv2 e2
E = - - 9 X 109 - .
2 r
If the above values for r and v are substituted in this expression, the result is:
2 2 e4
E = -(9 X 109) 2 1r m (34.11)
n n2h2 '
where the subscript on E indicates tha.t a different total energy value is asso-
ciated with each integral value of the quantum number, n. This simple
expression is made more accurate by replacing the electron mass m by the
so-called reduced massµ, defined by

.! = .! + .l,
µ m mi

where mi is the mass of the nucleus. This corrects for the fact that the elec-
tron actually rotates about the center of mass of the nucleus-electron system
rather than about the center of the nucleus. When mks values of the physical
quantities e, h, m, and mi are substituted, the energy values of the several al-
lowed stationary states are given by
En = -
2.178 x 10-18 •
JOU
1
e. (34.12)
2
n
The amount by which the energy of an atom changes when an electron
jumps from one permitted orbit to another is given by the difference between
the energies of the two states, and this change is equivalent to a photon energy
hf, where f is the associated frequency. Thus Bohr's second postulate, as
applied to emission lines, can be stated as
(34.13)
where E2 is the energy of the atom for the outer orbit and E1 for the inner
orbit. A comparison of this equation with Eq. 34.9 shows the similarity be-
tween the results given by the quantum theory and the earlier, empirical
theory: The frequency of a spectrum line is given as the difference between
two terms.
§ 34-4 THE HYDROGEN SPECTRUM 669

Hydrogen gas at room temperature is in the unexcited state, so that in

most of the atoms the electron revolves in the smallest permitted orbit: that
for which n = 1. When energy is supplied by thermal or electrical excita-
tion, the electrons in many of the atoms are moved to orbits of larger radius,
corresponding to n = 2, 3, 4 · · · . An electron stays in such an excited orbit
for a time of approximately 10-s sec and then jumps to a smaller orbit, radiat-
ing energy as a photon in accordance with Eq. 34.13. The apparently steady
spectrum lines seen in a spectroscope are the result of great numbers of such
identical jumps taking place simultaneously. The relative intensities of the
several lines thus produced depend upon a number of factors, such as the
Energy
.n Joule Electron·volts
00 0
6 - 60.6 x 10·21 -0.38
5 - 87.1 x 10·21 -0.54
c(!)
4 (!)
..c ·;::
u
"' -136x10'21 -0.85
"'
0
a.
(!)

"'
3 -242 x 10·21 -1.51
"'
(!)
·;::

.
(!)

"'
(!)
Q _§
c: 0
(!)
2 .....0 , co
a. -544 x 10·21 -3.40
c
Q
"'
(!)

~ ·;::
(!)
c
0 "'c:
0
E
_,>-

- 2178 x 10· 21 -13.60

Fw. 34-4. Energy levels for hydrogen

nature of the excitation conditions and the likelihood of transitions occurring

between any two particular states.
The hydrogen atom can be represented as a series of the energy states or
levels given by Eq. 34.12, as is done in Fig. 4 (not drawn to scale). Energies
may be given in joules, but more often they are expressed in electron· volts,
§ 18-12. The three series of lines discussed in connection with Eq. 34.9 are
shown as transitions between the various energy levels in Fig. 4. The
ionization potential is the energy, usually expressed in electron· volts, neces-
sary to remove the outermost, or valence electron entirely away from the
atom; it is indicated by the long arrow at the left side of the figure. Because
670 MODERN PHYSICS Chap. 34

the ionization potential can be unambiguously determined, the corresponding

upper energy level, for which n = oo, is usually set at zero, and all other
energy levels are hence given negative values. The energy difference in
electron· volts between the lowest or ground level and the first excited level is
called the resonance potential; it corresponds to the energy of what is usually
the strongest line in the spectrum and the line that can be obtained most
readily in absorption. It will now be clear why the absorption spectrum of
an unheated gas contains only the lines originating on the ground level
(n = 1, for hydrogen).
34-5. Spectra of Other Elements. The Bohr theory applies literally only
to the common hydrogen atom, but it can be adapted to give the spectrum
of deuterium, a heavy isotope of hydrogen. For this atom, ;JIT has about
twice the hydrogen value, soµ and also the numerical constant of Eq. 34.12
are increased by about 1 part in 4000. Hence the spectrum lines of deuterium
have slightly shorter wavelengths than those of ordinary hydrogen. The
observation of such lines in 1932 established the existence of this heavy isotope
of hydrogen.
Ionized helium atoms have about four times the mass of hydrogen nuclei
and carry twice the charge. The frequencies and wavelengths of the ionized
helium lines can be obtained from Eqs. 34.11 and 34.13 by replacing min the
former equation by the appropriate value of µ, and multiplying the energy
values (Fig. 4) by 4, the square of the atomic number of helium. For doubly
ionized lithium, the latter factor is 9.
The spectra of other elements are complicated by the presence of one or
more closed shells of electrons around the nucleus, § 15-2. These shells have
the effect of neutralizing some of the positive charge and thereby weakening
the electrostatic attraction between the nucleus and the valence electrons.
But if the outer electrons travel in elliptical paths rather than circles, they
may penetrate these inner shells at least partially, and hence be more strongly
attracted by the nucleus during part of each circuit. The effect is to in-
crease the energy that binds the electron to the nucleus, and this gives a
lower position to the energy level, Fig. 4. In order to take account of the
several discrete possibilities for elliptical orbits, a second quantum number l
is introduced; it is called the quantum number of orbital angular momentum
and can have integral values from Oto (n - 1). The two quantum numbers
n and l can be used to give a satisfactory accounting for the gross features
of the spectra of the alkali metals (lithium, sodium, potassium, etc.), with
their several interlocking series of lines.
A feature not explained by the above theory is the doubling of many of the
individual lines of alkali-metal spectra; the best example is the well-known
sodium D "line," which is actually a doublet with component wavelengths
589.0 and 589.6 mµ. An explanation given by the Dutch physicists Samuel A.
Goudsmit and George E. Uhlenbeck in 1925, and supported by many subse-
quent experiments) assumes that the eledron itself spin15 on its axis (§ 19-1),
§ 34-G THE PAULI EXCLUSION PRINCIPLE G71

and at such a rate as to have an angular momentum of ! h/2 1r. A third

quantum number is introduced by this concept; it is called the quantum
number of electron spin and given the symbols; it has the numerical value!.
The spinning electron constitutes a moving charge and hence produces a
magnetic field, which field can align itself either parallel or anti-parallel to
the magnetic field set up by the orbital motion of the electron, and this
explains the doublets of the alkali-metal spectra. Atoms with more than
one valence electron may have triplet, quartet, or even more complicated
multiplet energy levels and a correspondingly complex "fine structure" of
their spectrum lines.
A fourth quantum number, m, is introduced for assignment to the definite
orientations that the electron orbits take when the light source is placed in a
magnetic or an electric field. This quantum number can have the values O,
±1, ±2, · · · ±l.
The so-called "vector model" of an atom is built up by using the three
quantum numbers, l, m, and s, for each of the valence electrons of the atom,
and combining them by vector addition so as to give resultant vectors with
integral or half-integral values. The results of this process are in qualitative
agreement with many observed details of the spectra of the 90-odd elements.
This method has been of great value, both in explaining the relationships
existing among the permitted energy levels of an atom and in predicting
others for spectra not previously analyzed. The analysis of complex spectra
is, however, still largely an empirical process. The Bohr theory contributed
the concept of finite energy levels and the idea of measuring the energy dif-
ference between two of them by the frequency of the emitted or absorbed
spectrum line. This idea follows from Eq. 34.13, which may be rewritten as

(34.14)

The observed quantities are, of course, the wavelengths of the lines. From
these the frequencies are computed, and from the frequencies it is possible
to deduce the energy levels if intelligent use is made of the quantum numbers
of the various combinations suggested by the vector model. Some theoretical
basis for this process has been furnished by studies in the field of quantum
mechanics.
34-6. The Pauli Exclusion Principle and the Structure of Atoms. Great
assistance in understanding atomic structure was given in 1925 when the
German theoretical physicist ·wolfgang Pauli announced the exclusion prin-
ciple that now bears his name. This principle states that no two electrons
in an atom can have identical sets of quantum numbers, namely n, l, m, and
s. This statement was first made on empirical grounds but has since been
derived from quantum mechanics. Its realm of application has been much
broader than the field of atomic spectra.
As an example, the exclusion prinriple determines the number of electrons
672 MODERN PHYSICS Chap. 34

in each of the completed shells discussed in § 15-2. For the innermost elec-
trons of a heavy atom, the principal quantum number n has the smallest
possible value, 1. Thus, the orbital quantum number must be O for both
electrons, and the magnetic quantum number m must also be 0. The elec-
tron spin can have values of either +! or -!, and hence only two electrons
are permitted in this innermost shell, ordinarily called the K shell. For the
next shell, n = 2, and l may equal 1 or 0. For the latter, m = 0 and again
there are two possibilities, one electron with s = +! and the other with
s = -!. For l = 1, rn may equal +1, 0, or -1, and for each value of m, s
may be +! or -!. Therefore, there are six possibilities for l = 1 and a
total of eight possibilities for n = 2, so the L shell contains 8 electrons.
The situation becomes more complicated for the outer shells, because the
energy differences for various values of n become small, and of ten less than
the energy differences for the several permitted values of Z. Therefore, elec-
tron configurations do not build up so that their quantum numbers progress
in a regular fashion beyond argon (atomic number 18). Nevertheless, it is
possible to state the electron configuration and the corresponding permitted
energy states, with all quantum numbers specified, for at least a part of the
energy-level diagram of all the atoms for which spectra have been examined.
Such important features of the periodic table arrangement as the location of
the inert gas elements with their high ionization potentials and negligible
chemical activity, and the chemical similarity among the rare earth ele-
ments, are readily explained in terms of the electron configurations deduced
from spectrum analysis.
34-7. X-ray Spectra. The existence of the strong emission lines of the
so-called characteristic x-ray spectra in addition to a continuous spectrum
is mentioned in § 34-3, and the apparatus with which such spectra are ob-
served and measured is discussed in § 34-8. The bright-line spectra are
observed when the potential difference across the x-ray tube is sufficiently
high, and the lines are characteristic of the target material. Ordinarily this
is a heavy element such as tungsten, but special tubes have been made in
which targets of different metals can be used interchangeably.

Gold Molybdenum Copper

II I,
0.02 0.04 0.06 0.08 0.10 0.12
II,
0.14
I,
0.16
Wavelength, millimicrons

Frn. 34-5. Some K x-ray spectra

X-ray spectra are much simpler than the optical spectra produced by the
valence electrons of the atom. The spectra with the shortest wavelengths
(around 0.01 to 0.5 mµ), called the K spectra, have a typical triplet configura-
tion, in which the longest-wavelength component is separated from the other
§ 34-8 THE WAVE NATURE OF X-RAYS 673

two and is much more intense, Fig. 5. The K spectra of all the heavier ele-
ments are similar in appearance, but are located at progressively shorter
wavelengths for the higher atomic numbers and are increasingly compressed,
as shown in the figure for three metals.
The accepted explanation for the production of characteristic x-ray spectra
is that a fast-moving electron in the x-ray tube knocks one of the most closely
bound electrons of a target atom out of the atom. An electron from a larger
orbit, and correspondingly higher energy level, jumps into the vacancy with
the emission of a photon of frequency given by Eq. 34.14; since the energy
differences between these orbits near the nucleus are very great, the photons
have high frequency and short wavelength. The vacancy thus created in a
larger orbit is filled by a jump from a still larger one. This cascading process
continues, with the emission of additional lines, until it can go no further,
and the atom is left in an ionized state.
The K spectra are those resulting from jumps into the K shell, after one of
the two K electrons has been dislodged from its orbit. Jumps from an outer
shell to the L shell give the so-called L spectra, which are more complicated
and may consist of half a dozen components. In a sense, the Lyman series
can be considered to be the K x-ray spectrum of hydrogen, and the Balmer
series the L spectrum, although this terminology is not ordinarily applied to
the lighter atoms.
The regularity of the shift toward shorter wavelengths in the K spectrum,
mentioned above, was put in quantitative terms in 1914 by the English
physicist Henry G-J. Moseley (1887-1915). On making a systematic study
of the x-ray spectra of a large number of elements, he found that the fre-
quency of the strongest line, called the K-alpha line, is proportional to the
square of the atomic number of the element reduced by 1; thus, the fre-
quency ratio for two chemical elements denoted by subscripts 1 and 2 is

(34.15)

where Z is the atomic number. This relation, known as Moseley's Law, has
been used to place elements in their proper sequence in the periodic table in
certain questionable cases. For example, for elements number 26 to 28, the
sequence, as given by atomic weights, would be iron-nickel-cobalt, whereas
the proper sequence, indicated by x-ray spectra, is iron-cobalt-nickel, and
this result is confirmed by spectrum analysis.
34-8. The Wave Nature of X-rays. The physical nature of x-rays re-
mained in doubt for several years after their discovery. Their great pene-
trating power set them off from known particles and waves; they showed no
deflection in strong electric or magnetic fields; in early experiments they
could neither be reflected nor refracted. In 1912, however, the German
theoretical physicist Max von Laue suggested a more searching experimental
test of their wave character, using natural crystals as closely spaced, three-
674 MODERN PHYSICS Chap. 34

dimensional diffraction gratings. He proposed that a beam of x-rays be

directed upon such a crystal plate, and that a photographic plate be placed
beyond the crystal, oriented so as to receive the rays scattered by the crystal.
If x-rays are waves, there should be selective scattering in certain directions,
whereas if they are particles, the scattering should be more uniform. The
former was found to be the case; a diffraction pattern characteristic of the
regularities of structure of the scattering crystal was obtained, similar to that
shown in Fig. 6, and it was thus proved that x-rays are waves.

FIG. 34-6. Laue Hpot diagram of beryl. During most of the exposure the central beam
was masked off by a piece of lead, to prevent excessive blackening of the plate
(Courtesy of the Polaroid Corporation)

X-ray diffraction patterns are now widely used in determining the micro-
structure of matter in the solid state, for example, in metallurgy for the
study of the mechanical processes of rolling, hardening, and annealing. In
such studies the x-rays are made monochromatic by passage through filters
of aluminum or other light metal which transmit the strong radiations char-
arteristic of the target material of the x-ray tube but absorb the weaker
radiations.

Lead
plates

~ ~

Fm. 34-7. The Bragg x-ray crystal spPctrometcr

§ 34-S THE WAVE NATURE OF X-RAYS G7;j

A few months after the Laue experiment, a second application of crystal

diffraction of x-rays was made by the British father-and-son team of Sir
William H. Bragg (1862-1942) and Sir Lawrence Bragg, later Cavendish
Professor of Physics at Cambridge University. The arrangement consists of
an x-ray spectrometer, as shown in Fig. 7, in which a fine beam of x-rays,
collimated by slits in the two lead plates, strikes the flat face of a plate of
calcite, rock salt, or other suitable eryf,talline material. Most of the beam
passes through the crystal undeviated, hut, as before, some x-rays are scat-
tered by the atoms of the crystal or, more accurately, by the electrons of
those atoms. The crystal holder is arranged so that it can be rotated through
small glancing angles cf> (measured from beam direction to crystal face), and
the arm holding the ionization chamber, § 23-1-1, rotates through twice that
angle, because the angle of incidence and the angle of "reflection," or maxi-
mum scattering intensity, are equal.
Figure 8 shows rays incident on atoms A and B in neighboring crystal layers.
Scattering will take place in all direetfons from the atoms, hut consideration
is given to the direction at angle cf>.
The path difference between the
lower "reflected" ray and the upper
~"
ray shown is the sum of the distances d
.11:JB and BN, and this total distance ...,_ -+-M- -N----. _ _t_
B
determines the phase difference be- e O + 0 •
tween the two scattered rays. The I -sodium atom
phase difference will be O or a mul- o • + • .,._-Chlorine atom
tiple of 2 1r radians if the path Frn. 34-8. X-ray diffraction by a rock salt
difference is an exact multiple, k, crystal
of the wavelength 11.. The path
difference is 2(AB sin cf>), where AB is the spacing of the crystal layers. If
this spacing is denoted by d, then the necessary condition for constructive
interference between the two rays of wavelength 11. is given by
. k11. (34.16)
sm cf> = 2 d'

known as Bragg's Law. The two length quantities, 11. and d, should of course
be expressed in the same units.
In an actual crystal, with its thousands of layers of atoms, scattering takes
place at many atoms, and there are many resultant beams; nevertheless, the
geometry is unchanged and Eq. 34.16 still applies. It is more appropriate to
use here the term diffraction, which refers to the interference of a large num-
ber of contributing beams. And because of the many beams, there is an
increase in the intensity of the radiation measured. The wavelength and
intensity spectrum of the incident x-ray beam can be determined with an
x-ray spectrometer, if the crystal spacing is known. Or, conversely, crystal
spacing can be measured if monochromatic x-rays of known wavelength are
676 MODERN PHYSICS Chap. 34

used. In modern x-ray spectrometers the ionization chamber may be re-

placed by a photographic plate set at right angles to the diffracted beam, and
much less time is then required for making a complete spectrum measure-
ment.
When x-rays of wavelength 0.1542 mµ are diffracted by planes parallel to the cubic
lattice faces of a rock salt crystal, the smallest glancing angle for reenforcement
(k = 1) is found experimentally to be 15.83°. The spacing between the successive
planes of atoms in the crystal is given by Eq. 34.16 as

d = ~ . </>
= 0.1542 X 10-1 = 2 83
. 15 .83°
2 sm x 10-s cm .
2 sm ,l

An independent evaluation of this spacing can be made from the density of N aCI,
2.163 gm/cm3, and its molecular weight, 58.45. From Avogadro's Number, there are
6.02 X 1023 molecules in 58.45 gm of NaCl; this gives 1.03 X 1022 molecules/gm or
1.03 X 1022 X 2.163 = 2.23 X 1022 molecules/cm3 • With two atoms to the molecule,
there will be 4.46 X 1022 atoms/cm 3 , and if the crystal lattice is a multitude of cubes
v
with atoms at the corners, there will be 4.46 X 1022 = 3.54 X 107 atoms in a row
1 cm long. The spacing between atoms and also between rows is therefore d
= 1/(3.54 X 107) = 2.83 X 10-s cm. This result verifies the other value and indi-
cates the correctness of the method using x-rays.

X-ray diffraction can be applied also to crystal specimens in powder form,

because the vast number of small crystals which make up the specimen have
a random orientation and make every
conceivable angle with the incident beam.
The technique employed is analogous to
the simple Laue method and gives diffrac-
tion patterns such as that of Fig. 9. The
powder technique serves for the identifica-
tion of crystals rather than for the analysis
of their structure.
In recent years it has been found that
x-rays can be reflected, refracted, and
polarized, using specially adapted tech-
niques. Interference effects have also been
demonstrated. A conventional ruled grat-
ing will cause the diffraction of x-rays, if
the beam is incident on the grating surface
Fm. 34-9. X-ray powder diffraction
pattern of beryl. The central beam at a grazing angle, and x-ray diffraction
was largely masked off, as in Fig. 6 grating experiments have yielded accurate
(Courtesy of the Polaroid Corporation) values of such important physical con-
stants as Avogadro's Number.
34-9. The Corpuscular Nature of Radiation. The interference and diffrac-
tion of light-including x-rays-are strong indications that light is a wave
motion, and this was the favored theory during the latter decades of the nine-
teenth century. But the wave theory could not account for the photoelectric
§ 34-9 THE CORPUSCULAR NATURE OF RADIATION 677

effect, whereas the quantum theory did so successfully. The quantum theory
of radiation also gave a satisfactory explanation of the absorption and emis-
sion of light in a discontinuous fashion in the production of atomic spectra.
Additional evidence for the corpuscular nature of light of x-ray wave-
lengths was reported in 1923 by Professor Arthur H. Compton, then of the
University of Chicago. He directed a beam of monochromatic x-rays on a
scattering substance, such as a block of carbon, and found in the scattered
radiation a component with wavelength different from that of the incident
radiation. The wavelength of this component was found to be larger by an
amount AA that a.epends on the angle () between the direct and the scattered
beams, according to the relation
AA = 0.0024 (1 - cos 0), (34.17)
where wavelength is measured in millimicrons. This equation, which states
the Compton Effect in quantitative terms, has since been derived theoretically,
employing the mechanism of an elastic collision between the x-ray photon
and an electron in the scattering
substance, as diagramed in Fig. 10.
The photon, of energy hf, communi-
cates some of its energy to the recoil-
ing electron, and the scattered photon
Incident photon
has energy hf', where f' < f, and
hence has a correspondingly longer E=hf
wavelength. The derivation makes
use of the principle of the conser- Fm. 34-10. Scattering of an x-ray photon
vation of linear momentum and the in the Compton Effect
definition of the momentum of the
photon given in Eq. 34.20 as h/A. The numerical constant of Eq. 34.17
has been shown to be equal to h/mc, where m is the mass of the electron.
Further confirmation was given to the proposed collision mechanism by
observing and measuring the direction of recoil of the electrons in a Wilson
cloud chamber, § 35-6.
An effect somewhat similar to the Compton Effect, but observed in the
visible and ultraviolet spectral regions for light scattered by molecules, was
reported in 1928 by Professor C. V. Raman of the University of Calcutta.
In addition to its theoretical significance, the Raman Effect has been put to
practical use in unraveling the structure of complicated organic molecules
and in analyzing organic liquids.
Thus, the view that light has both wave and particle properties seems to
be well established today. The particle or photon nature is most apparent
when light interacts with matter; the photon energy, E = hf, is proportional
to the frequency, which provides the best tie-in between the wave and cor-
puscular aspects. Electromagnetic radiation exhibits its strongest wave-
like characteristics when the wavelength is long-of the order of millimeters
678 MODERN PHYSICS Chap.34

to kilometers, as in radio waves; the frequency is low, and the energy of an

individual photon is so tiny that its detection as such may be impossible.
At the other extreme are x-rays and gamma rays, with very short wavelengths
and high frequencies. The energy of a single x-ray photon is so large that
its corpuscular properties predominate, and in fact its wave nature is diffi-
cult to demonstrate.
Experiments on the pressure of light have shown that a light beam has a
measurable momentum, and hence it follows that photons must have mass.
According to relativity theory, § 35-2, they possess this mass only because of
their great speed: the speed of light, c. Thus, all the energy of a photon is
in the form of kinetic energy, given by
Kinetic energy = hf = mc2, (34.18)
where the symbols have their usual meanings. The equivalent mass of the
photon can be obtained from Eq. 34.18 as
hf (34.19)
Mass= m = -. c2
When the photon is absorbed by matter, its energy is entirely transferred,
usually to an electron in the form of kinetic energy; its (relativistic) mass dis-
appears entirely, and the photon simply ceases to exist.
The momentum of a photon can be expressed in classical terms as the
product of its mass and its velocity and can further be given in the following
terms:
Momentum = me = hf = 1!.. (34.20)
c A
34-10. The Wave Nature of Particles. The duality of radiation as waves
and particles, discussed in the preceding section, has a counterpart in the
duality of matter. In 1924 the French theoretical physicist Louis de Broglie
suggested that particles of matter possess a wave nature, and that the wave-
length of this new type of wave is inversely proportional to the momentum of
the particle: thus
h (34.21)
-;\=-,
mv
where mis the particle mass and vis its velocity. Equations 34.20 and 34.21
are similar, despite the fact that the former pertains to the electromagnetic
waves of light and the latter to "matter waves."
The effects of matter waves should be best observable for the lightest of
particles-electrons-because these should have the longest wavelengths.
The discovery of their wave character was made, in 1927, at the Bell Tele-
phone Laboratories by Drs. Clinton J. Davisson and Lester H. Germer.
They observed diffraction effects when a single crystal of nickel was bom-
barded by a stream of electrons; the electrons were reflected differentially by
the crystal planes, much as in the Laue experiment, § 34-8. Independent
§ 34-10 Tl!E WA. VE NATURE OF PARTICLES 670

experimental evidence was given by the British physicist George P. Thomson;

who shot beams of high-speed electrons through a very thin film of metal.
Figure 11 is an example of a pattern formed on a photographic plate or on a
fluorescent screen by the diffracted electrons; since the metallic film consists
of many small crystals in random orientation, the pattern is very similar to
that of an x-ray diffraction pattern of a powder, Fig. 9.
The wave nature of particles is utilized in the electron microscope, § 30-7.
If the electron gun accelerates the electrons through a potential difference of
only 25 volts, their wavelength (as computed from Eq. 34.21 and the expres-
sion Ve = ! mv2 )· is about 0.24 mµ, which is a thousandth that of the shortest

FIG. 34-11. Electron diffraction pattern of magnesium

(Courtesy of Tufts CoUeue l'hy.~ical Laboralory)

ultraviolet rays that can conveniently be used in microscopy. ·with accel-

erating potentials in the hundreds and thousands of volts, mueh smaller
wavelengths, and correspondingly greater resolving powers, are obtained,
§ 32-12.
The concept of waves being associated with matter necessitates a re-
examination of the laws of mechanics, at least for very light and very fast
moving particles. This study was first called wave mechanics and is now
well established under the name of quantum mechanics; it has made major
contributions to our understanding of the duality exhibited by waves and
particles and has helped explain many perplexing atomic processes. For
example, the de Broglie wavelength given by Eq. 34.21 is consistent with the
quantum condition introduced by Bohr to fix the energy levels of the normal
hydrogen atom, namely, that the angular momentum of the electron moving
in its circular orbit of radius r must be an integral multiple of h/2 7T', Eq.
GSO MODERN PHLr;JOS Chap. 3-!

34.10. The wave character of the electron should require that the length of
its path in any stable orbit will contain an integral number of wavelengths,
just as with vibrating strings in setting up stationary waves, § 25-15. Con-
sequently, an orbit of radius r and length 2 1rr should contain an integral
number of wavelengths ;\.; for the first orbit this condition is that A = 2 1rr,
or r = ;\./2 1r. The angular momentum of the electron in the orbit is given
by mvr, and from Eq. 34.21, mv = h/;\.. These values for mv and for r can
be combined to give the mathematical statement of Bohr's first postulate:
h
Angular momentum = 2 1r ·

The modern concept of matter waves is that they guide the motions of the
particles with which they are associated, as in the electron orbits of atoms or
the focused electrons of the electron microscope. The precise location of the
particle is hence somewhat uncertain because of the presence of the wave
train; the interpretation given by quantum mechanics is that the square of
the wave amplitude at any point represents the probability of finding the
particle at that point. A statistical rather than an absolute situation is
suggested, and in fact the specific nature of Newtonian Mechanics must give
way to the probability picture of Quantum Mechanics when light, fast-moving
particles are dealt with, as in atomic and nuclear physics. This principle of
indeterminacy was stated by the German physicist Werner Heisenberg in
1925, even before the discovery of the wave properties of electrons. The
Heisenberg Uncertainty Principle states, in effect, that it is impossible to
know precisely both the position and the velocity (momentum) of a moving
particle. Each can be known within a certain range of uncertainty, and the
produrt of the two uncertainties, of position and momentum, can never be
less than h/2 1r. The philosophical implications of this principle, and of
other novel features of the quantum mechanics, are beyond the scope of
this textbook; but, from what has been stated, it will be evident that the
quantum nature of energy and matter has been well established.

PROBLEMS
Express wavelength answers in millimicrons.
1. A filament is 15 cm long and 0.020 mm in diameter. Compute the power in
watts required to keep the surface temperature of the filament at 2000°K. Assume
the emissivity of the filament material to be 0.60.
2. Calculate the centigrade temperature for which the peak of the black-body
radiation curve is located at a wavelength of 1 micron.
3. Assume the sun to be a black body at 6000°K, and determine the peak wave-
length of its radiation curve.
4. The two radiation constants of Eq. 34.3 are related to fundamental atomic con-
stants as follows: k1 = 8 1rhc; k2 = hcN/R, where N is Avogadro's Number and R is
the gas constant per mole. Use the values given in the Appendix to compute the
magnitude of the radiation constants in mks units. State the resulting units for each.
5. The photoemissive threshold of a surface is 600 mµ. Compute the minimum
 PROBLEMS 681

work in joules necessary to eject an electron from this surface, and the work functio11
in volts.
6. If light of wavelength 200 mµ falls upon the photoelectric surface described in
Prob. 5, what is the maximum kinetic energy of the escaping electrons?
7. What opposing potential would be needed to stop the escaping electrons in
Prob. 6?
8. What is the maximum frequency of the x-rays that can be excited by 15,000
volts applied to an x-ray tube? What is the corresponding minimum wavelength?
9. A convenient form of Eq. 34.8 for the rapid calculation of the minimum x-ray
wavelength Xmµ in terms of the potential difference V volts applied to the tube is
Amin = const/V. Evaluate the constant and use the expression to check the answer
for Prob. 8.
10. Use the expression of Prob. 9 to compute the smallest potential differences
necessary to produce x-rays of wavelength (a) 0.5 mµ, (b) 0.02 mµ.
11. Compute the radius of the smallest Bohr orbit, and the speed of the electron in
this orbit.
12. Use Eq. 34.9 to compute the frequencies of the following spectrum lines of
atomic hydrogen: (a) Second line of the Balmer series, (b) second line of the Lyman
series, (c) first line of the Paschen series. Also determine the corresponding wave-
lengths. Put the answers in table form.
13. Calculate the resonance potential and the ionization potential of atomic hy-
drogen.
14. The reduced mass of the valence electron in ionized helium is only 1 part in
2000 larger than for hydrogen. Compute the wavelengths of the lines of ionized
helium that lie in the red, yellow, and green portions of the spectrum. With which
hydrogen lines might each of these helium lines be confused?
15. How many electrons are in the completed M shell, for which n = 3?
16. If the wavelength of the K-alpha x-ray line of tungsten is 0.021 mµ, what is
the wavelength of the K-alpha line of copper? See Appendix for the atomic numbers
of the elements.
17. Verify :Moseley's Law by plotting a curve with the square root of the frequency
of the K-alpha x-ray line as ordinate and (Z - 1) as abscissa. Use the following
elements and wavelengths: Ca 0.335 mµ, Ni 0.166, Mo 0.071, Sn 0.049, W 0.021.
18. X-rays of wavelength 0.045 mµ incident upon a crystal are strongly scattered
at an angle of 10° for second-order "reflection" in a spectrometer of the type shown in
Fig. 7. Compute the spacing of atom layers in the crystal.
19. A parallel beam of homogeneous x-rays of unknown wavelength is directed upon
a rock salt crystal in the manner shown in Fig. 8. If the first-order diffracted ray
appears when the glancing angle is 5.2°, what is the wavelength of the x-rays?
20. The constant of Eq. 34.17 can be shown to be h/mc. Substitute values for
these physical quantities, evaluate the constant numerically, and check its units.
21. (a) What change in wavelength is produced in the Compton Effect by a head-on
collision which causes the scattered x-ray photon to rebound along its original path?
(b) If the original wavelength of the x-ray photon is 0.020 mµ, how much energy is
imparted to the recoil electron?
22. Calculate the Compton-effect change in wavelength of x-rays scattered at angles
of 20°, 90°, and 120° from the direction of the incident beam.
23. Compute the kinetic energy, the mass, and the momentum of an x-ray photon
having a wavelength of 0.020 mµ.
24. Compute the wavelength of electrons that have been accelerated through a
potential difference of (a) 100 volts, (b) 2500 volts.
25. Determine the wavelength of protons that have been accelerated through the
same potential differences as in Prob. 24.
 35
NUCLEAR PHYSICS
Much of the research effort in present-day Physics is being focused on the
nucleus of the atom, for the nucleus, because of its definite electric charge
and its concentration of mass, is the most important component of an atom.
The nuclear charge, which establishes the identity of the atom, is altered
spontaneously in a few elements through the process of natural radioactivity.
All atoms can now be artificially transmuted by the recently developed tech-
niques of Nuclear Physics. The alchemist's dream-of making gold from
the baser metals-has indeed come true; not a transmutation by chemical
means, which affect only the valence electrons of the atom, but by more
powerful physical methods for penetrating the nucleus and changing its
composition.
35-1. The Discovery of Natural Radioactivity. The French scientist
Antoine H. Becquerel (18,52-1908), v1-hile experimenting in 1896 with fluo-
rescence produced hy x-rays, discovered that uranium salts emit unceasingly
a type of radiation that is able to blacken photo-
graphic plates; this phenomenon is called radio-
Electroscope
activity. The French husband-and-wife team of
Pierre Curie (1859-1906) and Marie Curie (1867-
1934) continued these investigations and discov-
ered the more strongly radioactive elements radium
and polonium, the latter named after Madame
Radioactive Curie's native country, Poland.
\.,ul• o"A G' a,/ sample Investigations of the properties of radioactivity
showed the strength of the radiation to be inde-
Fw. 35-1. A metal-leaf
electroscope, adapted for pendent of chemical combination. and of any ex-
radioactivity measurements ternal physical influences such as temperature,
pressure, exposure to light, and x-ray or cathode-
ray bombardment. Thus, radioactivity became recognized as a property of
the element itself, or more properly, of the nucleus of the atom.
It was soon found that the gold-leaf electroscope was more sensitive for
measuring radioactivity than the photographic plate. This device, § 15-1,
adapted for the study of radioactivity, is shown in Fig. 1. Here the support-
ing rod for the thin metal foil projects through an insulating collar in the
bottom of the metal chamber and terminates in a metal plate. The shell of
682
§ 35-2 THE THEORY OF RELATIVITY 683

the electroscope surrounds this plate, and the radioactive material to be

tested is placed at a known distance below the plate. The radioactive emis-
sions ionize the air within the instrument, and the electroscope, originally
charged, is observed to discharge. Quantitative measures of intensity can
be made by timing the discharge rate for different distances between the
sample and the plate.
\Vith the electroscope it was discovered that three kinds of radiation are
given off, but that not every radioactive material emits all three. The
strongest radiation, named alpha rays, was found to give measurable effects
up to sample-to-electroscope distances in air of one to three inches, depending
upon the material being tested. At slightly greater distances the discharge
rate decreased markedly; the results suggested specific ranges in air for these
rays. \Vith the sample located beyond the terminal range of the alpha
rays, a weaker emanation, called beta radiation, was still observed. The
effect of these rays was lessened by interposing metal foils a millimeter or
two thick bet,veen the sample and the electroscope. A third type of radia-
tion, named gamma radiation, was then measured; it ,vas much weaker than
the other two but very penetrating, and could only be stopped by absorp-
tion in several centimeters of lead.
Experiments conducted in a uniform magnetic field, by methods such as
those described in § 18-12, established the following facts: alpha particles
carry a positive charge twice as large as the negative charge on the electron
and have a mass about four fames that of the hydrogen atom; beta particles
are negatively charged with exactly the charge of the electron but have a
mass which may be many times that of the electron; gamma rays exhibit
neither charge nor mass. Diffraction experiments, § 34-8, show that gamma
rays are electromagnetic waves having wavelengths shorter even than x-rays.
The mass and charge found for alpha particles suggested that these might
be helium nuclei-that is, helium atoms with the two valence electrons re-
moved. This surmise was confirmed by an experiment in which a spectrum
discharge tube was exhausted until the electric discharge through it ceased,
and then a small amount of alpha-emitting radioactive substance was placed
just outside a thin window in the tube. When a potential difference was
again applied to the tube a few days later, it glowed with a characteristic
yellow-orange color, and the glow was found upon examination with a spec-
troscope to comprise the well-known spectrum lines of helium. This test
showed that alpha particles from the radioactive material had penetrated
the window of the discharge tube, had picked up two electrons each, and had
become neutral helium atoms.
35-2. The Theory of Relativity. In measurements to determine the mass
of beta particles, the results obtained, although never less than the mass of
the electron, covered a wide range. This apparent inconsistency was ex-
plained by the Specia.l Theory of Relativity proposed by Einstein in 1905.
The theory postulates (1) that the determination of absolute motion is impos-
684 MODERN PHYSICS Chap. 35

sible and that only relative motion has definite meaning, and (2) that the
speed of light in free space is independent of the motions of the light source
and the observer. The first postulate leads to the generalization that the
laws of physical phenomena (for example, the laws of Mechanics) are unaf-
fected by uniform straight-line motion of the system of coordinates to which
the physical quantities are referred. The second postulate results from the
failure of an experiment to detect with an interferometer any difference
between the speed of light along the direction of the earth's motion and along
a direction at right angles thereto. The theory of relativity shows that the
mass of a body increases as its speed becomes greater, but the change is
appreciable only at high speeds. The mass of a body at great speeds is
called its relativistic mass, and its value is given by the expression
mo (35.1)
m= '
Vl - v2/c 2
where mo is the rest mass of the body, v its speed, and c the speed of light.
Thus, a beta particle moving with nine tenths the speed of light has a rela-
tivistic mass equal to about 2.3 times its rest mass of 9.11 X 10-31 kg.
An additional result of the theory is that the Newtonian expression for
kinetic energy, ! mv 2, does not apply for high speeds, and that the exact
expression for it is
Ek = (m - mo) c2 ; (35.2)
the Newtonian form is a special case, for low speeds. In generalized form,
this expression is Einstein's Mass-Energy Equivalence:
E = ~m c2, (35.3)

where E is the energy, of whatever type, and ~m is the change in mass.

Surprising though this concept may be, it has been verified experimentally
in nuclear transmutations such as those discussed in §§ 35-9 and 12. In
purely chemical reactions the Principle of the Conservation of Matter (or
Mass) gives sufficiently accurate results, but in nuclear transformations it is
necessary to extend the principle to the Conservation of Matter plus Energy.
The energy supplied to New York City by the Consolidated ,Edison Company
during an average December weekday in 1952 was 40 million kilowatt·hours. If this
energy could be obtained by the conversion of matter, how much mass would have to
be annihilated? Assume 100 per cent conversion efficiency.
To perform the calculation in mks units, energy must be expressed in joules; thus
40 X 109 watt·hr X 3600 sec/hr = 144 X 1012 watt·sec or joules. From Eq. 35.3,
E 144 X 1012
~m = c2 = 9 X 1016 = 1.60 X 10-a kg = 1.60 gm.

As more and more energy is applied to accelerate a particle, its speed is

brought up very close to that of light, and thereafter the relativistic mass of
the particle is of importance. Hence all exceedingly energetic particles
§ 35-3 RADIOACTIVE DECAY 685

travel with very nearly the speed of light, and differ only in mass, or per-
haps also in charge. If the particle does work, it loses energy both as mass
and as velocity, and finally returns asymptotically to its rest mass at zero
velocity. The concept of changing the kinetic energy of a moving particle
through a change in mass is quite foreign to Newtonian mechanics.
35-3. Radioactive Decay. The radiation from all radioactive materials
decreases in intensity with time; some materials lose most of their activity in
a period of days, hours, or even minutes. It is customary to specify the
time rate of decay by means of the so-called half-life, which is the time for
the activity to drop to one-half its value at the beginning of the considera-
tion. Figure 2 shows the deray curve for any radioactive material dropping

100

l!
'i:
:)
75
...0>.
- ...
.:.0
...
0

- ;;:. 50
'>
:;:
v
0
0
:.s0
0::
25
\ I~ ,-.....__
0 2 3 4 5 6
Time, half-lives
Fm. 35-2. Decay curve of a radioactive element

from 100 to 50 per cent in a period of one half-life, then dropping from 50
to 25 per cent in the next half-life, and so on. For example, the radioactive
gas radon, formed when a radium atom emits an alpha particle, has a half-
life of 3.82 days; the activity decreases to -l- the original value in 2 X 3.82
= 7.64 days, tot in 11.46 days, and so on.
The radioactive decay curve is recognized as being of the logarithmic or
inverse exponential form, and its equation can be derived as in §§ 21-4 and
31-11. The change in activity dl, that occurs in a short time interval dt, is
proportional to the number of radioactive atoms present in the sample, and
hence is proportional to the intensity I at that time; this relationship can be
expressed as dl = -k I dt, where k is a constant of proportionality called the
radioactive decay constant, and the minus sign is introduced to indicate that
the activity decreases as time increases. This expression can be rewritten
as dl / I = - k dt, and upon integration it becomes ln I = -kt + C. The
constant of integration, C, can be evaluated by setting the activity equal to
GSG MODERN PHYSICS Chap.35

Io when t = 0. It is found that C = ln Io, and hence ln (I/Io) = -kt.

This expression is ordinarily used in the exponential form I/ I O = e-kt;
when solved, the intensity at time tis
I = Io e-kt. (35.4)
I and Io can be measured in any convenient but like units, and the unit of k is
reciprocal time. It can be shown that the radioactive decay constant, k, and
the half-life of the radioactive material are related by the expression
H lf n _ 0.693 (35 ..5)
a-1e- k ·

35-4. Isotopes and the Mass Spectrograph. The concept that the atom
contains equal amounts of positive and negative charge (§ 15-2) slowly
gained acceptance, and the distribution of the two was determined later by a
series of scattering experiments conducted by the British physicist Ernest
Rutherford (1871-1937). Alpha particles were directed through a thin metal
foil, and were received upon a photographic plate to record them, or observed
individually with a microscope by counting the scintillations produced when
they struck a fluorescent screen of zinc sulfide. The nucleus of the atom
was "discovered" by these experiments, as follows: Most of the alpha rays
passed through the metal foil in straight lines from the radioactive source, as
would be expected. But a very few, only about one particle in 8000, were
deflected through angles as great as 90°. The only acceptable explanation
for such scattering requires the assumption that the positive charge of an
atom is concentrated in a relatively massive nucleus of small volume, and
that alpha particles which happened to approach very close to one of these
nuclei were repelled, in accordance with Coulomb's Law, § 15-5, and given
the large deflections observed. Thus Rutherford, about 1911, outlined the
so-called planetary model of the atom with its central, relatively heavy, posi-
tively charged nucleus and the surrounding lightweight, negatively charged
electrons.
The atomic number and mass number of an element are defined in the
preliminary consideration of atomic structure given in § 1.5-2. The atomic
number, Z, is the magnitude of the (positive) charge on the nucleus, in terms
of the electronic charge as a unit; it is equal to the number of protons in the
nucleus, and also represents the number of planetary electrons in the neutral
atom. The mass number, A., is the total number of protons and neutrons in
the nucleus. The tenn nuclide has been introduced to designate a species of
atom characterized by the constitution of its nucleus-that is, the numbers
of protons and of neutrons. Tests show that the atoms of a given element
(same value of Z) may have differing values of A because of different numbers
of neutrons in the nucleus-these various atoms are called isotopes of that
element. The average atomic weight used in chemical calculations repre-
§ 35-5 itADIOACTIVE SERIES 687

sents the average of the one or more isotopes that exist for each element,
adjusted for the relative abundance of the several isotopes.
The masses of the several kinds of atoms in an element can be measured
by an instrument called the mass spectrograph; the particular type shown in
Fig. 3 was developed by Francis W. Aston at Cambridge University. An
important part of the instrument, although not shown in detail, is the ion
source wherein the element is suitably vaporized and ionized by thermal or
electrical means. A beam of the positively charged ions from the source is
accelerated through an arrangement similar to the electron gun of Fig. 23-14.
The beam is collimated by two aligned slits, and proceeds through an electric
field maintained between two condenser plates, and then through a magnetic
field oriented at right angles to the electric field. The electric field deflects
the rays in one transverse direction and the magnetic field deflects them
oppositely; in both fields the ions of higher speed are deflected less than the

Magnetic
field
Fm. 35-3. One type of mass spectrograph

slower ones. The field intensities are adjusted so that, no matter what speeds
the ions have, the rays will converge to a common focus on the photographic
plate. Ions having different masses or charges will, of course, be deflected
differently, and only those having the same ratio of charge to mass will
arrive at a given point on the plate. "\Vhen the plate is developed, a line
resembling a spectrum line appears for each isotope of the element, and the
location of this line permits the atomic mass of that isotope to be determined
accurately~to within a few parts in a hundred thousand.
It is becoming the practice to express the mass of any individual isotope in
atomic mass 'anits (amu) based on the most plentiful oxygen isotope, which is
taken to have a mass of exactly 16 units. Thus, mass numbers A_ are the
atomic masses in amu rounded off to the nearest integer.
35-5. Radioactive Series. In each step of the nuclear decay of a natu-
rally radioactive element, either an alpha particle or a beta particle is emitted,
and as a result the atom is transmuted to one of a different element. The
emission of an alpha particle reduces the atomic number by 2 and reduces
the mass number by 4; the emission of a beta particle increases the atomic
688 MODERN PHYSICS Chap. 35

number by one unit but leaves the mass number unchanged. A gamma ray
may accompany the emission of either of these particles, but since it carries
neither charge nor mass the effect need not be considered in studying the
overall features of nuclear transformations. The nucleus resulting from
radioactive decay is not necessarily stable, but may itself be radioactive and
have an emission intensity and a half-life different from that of the parent
atom. The nucleus resulting from a second disintegration may in turn be
radioactive; the process continues until a stable nucleus is formed.

230 ,___ _,__ __,____ ___,__ _...___ _ _ _2.35x 10 5

y1ears
41
"'( 8.0xlO years
_B 226 t - - - - + - - - + - - - - - - 1 - - # - - - - + - - - - + - - - l
E
::, 1.62 x 103 yea rs
c
= 2221----+----!,---#---~--l----J.--___J

82 83 84 85 86 87 88 89 90 91 92
Pb Bi Po At Rn Fr Ra Ac Th Pa U
Atomic number,Z
Frn. 35-4. The uranium-radium radioactive series

Years of study have disclosed that there are now three series of naturally
occurring radioactive elements. These are the uranium series (which con-
tains radium as an important member), the actinium series, and the thorium
series. All three terminate in different stable isotopes of lead, of mass
numbers 206, 207, and 208, respectively. At one time there may have been
a fourth radioactive series which started with the element plutonium and
had neptunium as its longest-lived member; but if so all members of this
series must have turned into a stable isotope of bismuth in the millions of
years since the formation of the earth.
§ 35-(3 MEASURING 'POOLS FOR 'THE NUCLEAR PHYSICIST 689

Figure 4 is a diagrammatic representation of the uranium-radium radio-

active series; the dots represent the nuclei of the member elements. Along
the axis of ordinates are their mass numbers, and along the abscissas are
their atomic numbers and chemical symbols as they appear in the periodic
table. A longer arrow slanting down to the left indicates the emission of an
alpha particle, and a shorter arrow directed toward the right indicates the
emission of a beta particle. Many of the radioactive nuclei formed in such
a series are given special designations (not shown), although they are in fact
isotopes of some of the more common elements. The half-lives of the vari-
ous nuclei are indicated on the graph.
The age of the earth can be measured by making determinations of the
ratio of uranium to lead (isotope 206) in igneous rocks presumed to have
solidified at the time the earth was formed. The assumption must be made,
of course, that no lead occurred naturally in that particular mineral. The
data can be checked by similar calculations made after determining the
amounts of actinium and thorium in the rocks, in relation to the appropriate
lead isotopes. These measurements are best made with a mass spectrograph.
A different method for measuring the earth's age involves melting igneous
rocks in a vacuum furnace and extracting the helium gas absorbed in them.
The helium is presumed to have resulted from alpha particle emission in the
course of radioactive decay. The ratio of the mass of helium to that of
uranium is measured, and the age of the rocks computed. The several
methods agree as to the order of magnitude of the geological age of the earth,
and place it in the neighborhood of two billion years.
35-6. Measuring Tools for the Nuclear Physicist. In addition to electro-
scope and scintillation methods for detecting radioactivity, some other
methods were used in obtaining the results mentioned thus far in this chapter.
As early as 1908, Rutherford and his
German assistant, Hans Geiger, used t-,\\\\\\\\\''''''''"'''''''''''''''\'\'''''''''''''''''W·'''''"""''"''li
pulse ionization chambers to count + V!!//J/J/JJ!!!!J/!JJ!JJJ!JJ/JJJJJJJJJ/J!J!!JJJJJJJJJJJJJJJJJJ,
alpha particles. These chambers were
steadily improved, and were brought Fm. 35-5. A Geiger-Muller tube
to their present highly sensitive form
in 1928 by Geiger working in collaboration with another German scientist,
W. Muller; the complete equipments are known as Geiger-Muller counters,
G-M, or just Geiger counters. Figure 5 shows a typical G-M tube; it con-
sists of a metal cylinder within a glass envelope, and a wire or rod extend-
ing along the axis of the cylinder. The tube is usually filled with air or
argon at a pressure somewhat below atmospheric. A unidirectional poten-
tial of several hundred volts is applied, with the central wire as the anode,
the exact value being adjusted somewhat below the breakdown potential of
the gas. When an ionized particle passes through part or all of the tube and
produces ion pairs, the positive and negative ions are drawn toward the
appropriate electrodes, and the resulting pulse of current through the tube
690 MODERN P!IYSICS Chap. 35

is measured by an electroscope or an electronic counter. 'With proper facili-

ties for quenching the discharge rapidly, pulses can be counted up to a few
thousand per second.
The Geiger tube is readily adapted to modern vacuum tube circuits and
automatic counters. It is used principally in studies of gamma rays and of
cosmic rays, § 3.5-9; when equipped with a thin window in the glass envelope
it can be used in counting beta particles, but it does not distinguish the
various types of rays or particles from one another.
Another instrument for detecting and measuring moving charged particles
is the cloud chamber developed hy Charles T. R. Wilson in the CavendiHh
Laboratory and put into substantially its present form about 1911. A typi-

Stereoscopic
cameras

Light
source
Window

U
11111!11 I 1111 11J111Jl(llll(HII
Piston
t2zzzzzzz:zzz:ZZ1.

i
Frn. 35-6. "Wilson cloud chamber

cal cloud chamber is shown in Fig. 6; it consists of a cylinder containing a

piston and filled with saturated water vapor and alcohol vapor; a glass plate
is placed across the end for viewing. If the piston is suddenly retracted, the
vapor in the chamber becomes super-saturated and readily condenses on any
charged particles or dust specks that may be present. ·when dust-free air is
used, the only condensation nuclei will be those produced by the passage of
charged particles through the chamber. If an alpha particle passes, a dense
fog trail will be produced that will persist for several seconds, until swept
away by an applied electric field or by gravity. Alpha particles produce
tens of thousands of ion pairs, and their tracks are easily seen in the cloud
chamber; beta particles produce only about a hundredth as many, and leave
thin tracks which are harder to see; gamma rays are even poorer ionizers and
hence are usually detected by other means. The use of two cameras, as
indicated, provides stereoscopic pictures and permits the study of track
orientation in three dimensions. By the use of a trigger device set off by an
appropriately located Geiger counter, it is possible to have the expansion of
§ 35-6 MEASURING TOOLS FOR THE NUCLEAR PHYSICIST 691

the vapor take place only when an interesting event is occurring; this tech-
nique is much used in cosmic ray research.
The range and momentum of particles can be determined with the cloud
chamber, and the angles of the paths can be measured, for collisions and dis-
integrations. Curved paths are followed by charged particles if a magnetic
field is directed through the chamber; the direction of curvature indicates the
sign of the charge. Figure 7 shows the typically heavy tracks of alpha rays
from a radioactive source placed inside a cloud chamber, and also a branched
track that is discussed in the next section.

Fm. 35-7. Alpha-particle tracks in a cloud chamber; the radioactive source is at the left
(Courtesy of Professor W. D. Harkins)

Although photography played a leading role in the discovery of radioactiv-

ity, it has not been used much in quantitative work until recently. It is
now appreciated that charged particles of various types, in passing through a
photographic emulsion produce a latent image which can be brought out hy
the usual chemical development. Special photographic materials with thick
emulsions are used to introduce a third dimension. The type of record
obtained is similar to that given by the cloud chamber, but is especially
suited for alpha particles, protons, and the high-energy particles in cosmic
rays, because the "stopping power" of the photographic emulsion is about
two thousand times that of air. Nuclear track photography, as the method
is called, has the advantage of being continuously sensitive, of giving a
permanent record, of recording individual events, and of being much simpler
and less expensive than other methods. Emulsions have been developed
which are sensitive enough to record electrons.
The scintillation method used in counting alpha particles has its modern
counterpart in the so-called scintillation counter used with gamma rays and
x-rays of short wavPlength. These counters arc based on the scintillation
692 MODERN PHYSICS Chap. 35

produced in an organic crystal such as anthracene, or an inorganic crystal

such as sodium iodide that contains a small impurity of thallium. The pas-
sage of a gamma ray photon causes a tiny flash of light that is too weak to be
seen by the eye but strong enough to affect a sensitive photomultiplier tube,
§ 27-8. Such a tube can readily be incorporated into electronic amplifying
and counting circuits to yield a compact and automatic device for counting
particles at rates up to a million per second.
Another recent development is the use of crystal counters that employ
circuits similar to those of the Geiger counter. The crystals used, such as
diamond or silver chloride (at low temperature), are ordinarily poor con-
ductors of electricity, but become conducting when exposed to ionizing radia-
tions. These new counters are small and compact, and their action is very
rapid.
36-7. The First Artificial Transmutations. During the experiments on
alpha particle scattering described in § 35-4, Rutherford bombarded sub-
stances such as paraffin which contain many hydrogen atoms. Besides the
ejection of alpha particles from the paraffin, other particles were observed to
be ejected with a range about four times that of the bombarding alpha par-
ticles. These were judged to be recoil hydrogen nuclei (protons) and to
have the greater range by virtue of their smaller mass, about one-quarter
that of the alpha particle; this conclusion was confirmed by the use of a
cloud chamber.
The bombardment of nitrogen gas with alpha particles, in 1919, also re-
sulted in some of the long-range ejected particles. Since there were no
hydrogen atoms in the nitrogen gas, the only possible explanation seemed to
be that protons had been knocked out of the nitrogen nuclei. This surpris-
ing suggestion was confirmed later by cloud chamber photographs, such as
that of Fig. 7. In this photograph the branched track illustrates the rare
occurrence of the capture of an alpha particle by a nitrogen nucleus. The
thin upper branch is that of the emitted proton, and the heavier branch
below is that of the resultant nucleus, identified as a rare oxygen isotope of
mass number 17. This nuclear reaction can be written in a form similar to
a chemical equation, but with a superscript giving the mass number of the
nucleus and a subscript giving the atomic number, as follows:
(35.6)
where 1H 1 represents the proton-the nucleus of the hydrogen atom. Since
the symbol of the element and the subscript have the same meaning, both
need not be given, and a shorthand form can be used:
(35.7)
which expression indicates that the nucleus of a nitrogen isotope of mass
number 14 captures an alpha particle, emits a proton, and the nucleus be-
comes that of the oxygen isotope of mass number 17.
§ 35-7 THE FIRST ARTIFICIAL TRANSMUTATIONS 693

This nuclear reaction was the first artificially produced transformation of

a nucleus to that of another element; previously, only the naturally occurring
radioactive transformations had been known. The terms transmutation
and disintegration are often applied to such nuclear processes.
Rutherford and his colleagues at Cambridge University continued with
disintegration experiments for several years, and found that with the most
energetic alpha particles available from natural radioactivity-those of tho-
rium C'-transmutations could be produced in elements up to that with
atomic number 19: potassium. Apparently the alpha particles could not
penetrate nuclei of atoms with higher atomic number because of the greater
Coulomb repulsion, and hence the bombarding particles were simply scat-
tered.
It was realized about 1924 that higher particle energies were needed in
order to extend transmutation experiments to elements farther along the
periodic table, and several laboratories undertook the development of high-
potential machines for accelerating protons or other particles. At Cambridge
University, John D. Cockroft and Ernest T. S. Walton applied conventional
methods of potential multiplication by charging condensers in parallel and
discharging them in series, the switching being done by vacuum-tube cir-
cuits. By such "cascade rectification" they were able to obtain potential
differences up to 800,000 volts from a transformer with a 100,000-volt sec-
ondary rating; in other laboratories this limit has been increased to about
3 million volts.
Rutherford suggested, early in 1932, that the electrical potentials then
available (about 300,000 volts) should be used to accelerate protons against
a target containing lithium atoms, rather than wait for the attainment of
still higher potentials. Protons were selected because they should be able
to penetrate the nucleus more readily than the heavier, doubly charged alpha
particles. The studies in quantum mechanics that were being carried out
concurrently indicated that occasional penetrations of nuclei might be pos-
sible with projectiles of kinetic energy lower than the known potential barrier
imposed by Coulomb repulsion. A trial ,vas made, and Cockroft and Walton
in 1932 produced the first nuclear transmutation by entirely artificial
means, that is, using artificially accelerated protons rather than projectiles
from naturally radioactive substances. The reaction has been shown to be
the following:
(35.8)
The result of the bombardment is not beryllium of atomic number 4 and mass
number 8, because it is unstable and immediately breaks up into two alpha
particles, or helium nuclei; this has been confirmed by cloud chamber studieR.
Particle energies in nuclear physics are ordinarily expressed in multiples
of a million electron· volts, abbreviated Mev, which is an energy unit 106 timef;
that of the electron·volt, § 18-12; hence
G94 MODERN PHYSICS Chap. 35

1 Mev = 1.602 X 10-13 joule.

In being accelerated through a potential difference of one million volts, a
proton or an electron acquires 1 Mev of energy; an alpha particle, because its
positive charge is twice that of the electron, acquires 2 Mev of energy, or
3.204 X 10-13 joule. The thorium C' alpha particles referred to above have
an energy of 8.76 Mev.
X-ray photons of very short wavelength, such as those of the K-alpha line
of uranium, have an energy of only 0.12 Mev; this indicates that all energies
greater than about this amount are to be associated with the nucleus rather
than with the extranuclear electrons of the atom.
35-8. Machines for Producing High-energy Particles. In most ion accel-
eraton, developed for nuclear research, the individual particle having a
charge Q is accelerated through a potential difference V, and thus its energy
is increased by the amount VQ. The acceleration may be applied in a single
step, as in the voltage-multiplier apparatus mentioned above and the elec-
trostatic generator of Van de Graaff and Trump, or it may be applied in
repeated smaller steps of lower potential difference, and thereby reduce the
hazards and insulation problems. The effect of the acceleration is to increase
the kinetic energy of the particle, at first by increasing its speed. But for
electrons even at low energy and for heavier particles such as protons at
higher energies, the speed soon approximates that of light, whereupon the
supplying of additional energy increases the relativistic mass of the particle
in accordance with Eq. 35.2. For this reason the high-energy machines are
more properly called ponderators than accelerators.
The electron is not suitable for causing nuclear disintegration because of
its relatively large size. The particle usually chosen is the proton, although
the deuteron (the nucleus of heavy hydrogen, of mass number 2) is often
used; the alpha partirle is less frequently employed.
The Electrostatic Generator. The electrostatic generator, described in
§ 15-12, gives a steady beam of ions accelerated through a potential difference
which can be controlled to within a fraction of one per cent. Such a mono-
energetic beam of electrons or protons is valuable for accurate quantitative
work; for example, in scattering experiments. Million-volt electron genera-
tors are now available commercially for industrial x-ray ,york. A 3.5-Mev
machine was built to inject protons into the proton synchrotron described
below, and generators of about 12 Mev are being constructed for use in nuclear
research.
The Linear Accelerator. The so-called linear accelerator was first developed
in 1931 by Professor Ernest 0. Lawrence and David H. Sloan at the University
of California in Berkeley, to accelerate the ions of a heavy element such as
mercury. The arrangement consists of a series of successively longer tubes,
as illustrated in Fig. 8, connected to a r-f oscillator. The frequency of the
oscillator is adjusted so that the polarity of the second tube becomes negative
by the time the ions have passed through the first one. Therefore they are
§ 35-8 J1ACHINES FOR PRODUCING HIGH-ENERGY PARTICLES 695

accelerated in traversing the gap between the two tubes, after which they
"coast" along the axis of the second tube. The polarity reverses again and
the ions are further accelerated in the space between the second and third
tubes. The energy imparted to the ions in this way depends only on the
applied potential difference of the oscillator and the number of so-called drift
tubes.
The energy limitation of the initial accelerator of this type was imposed
largely by the maximum frequency available, so that only heavy ions could
be accelerated, and even for these unduly long drift tubes were needed.
Developments during World War II in high-frequency power oscillators, such
as the magnetron tube, § 24-9, caused interest in linear accelerators to be
revived. In a later machine 40 ft long, using an oscillator frequency of 200
million cycles per second, protons can be accelerated to 32 Mev. By using a
different principle, in which the ions are propelled by a traveling wave within
the tube, and ,vith frequencies as high as 3 billion cycles per second, it is

To oscillator
f Vacuum chamber)
Ion source --'-(£3.a=::\~ e~---t---f"'"}---=--=-~-.~---_3_-t_)
_ ____________-+_-_t-~Ln Target
I "--c I ,u
t
To oscillator
Fm. 35-8. A linear accelerator

hoped to obtain electron energies of more than 1000 Mev in an accelerator

designed by the late Professor William W. Hansen and now being constructed
at Stanford University.
The Cyclotron. The cyclotron, developed in 1931 by Professor Lawrence
and his co-workers, especially M. Stanley Livingston, is described in § 18-13.
This device is somewhat similar to a linear accelerator except that the charged
particles spiral inside two metal dees; equal times are taken for each semi-
circular transit, and these times are adjusted to the half period of the r-f
oscillator. The cyclotron cannot be used ,,·ith electrons because their rela-
tivistic mass increase would destroy this resonance condition. But the in-
creasing mass is not serious "·ith the heavier protons or deuterons up to
energies of about 20 l\Iev, such as are obtained with a cyclotron having pole
pieces of GO in. diameter. At these energies the speed of a deuteron is about
one seventh that of light, and its mass is increased only about one per cent
over its rest mass.
In 1H40, construction was started in Berkeley on a cyclotron with magnet
pole faces of 184 in. diameter, but the work was interrupted when the finished
magnet was used to separate uranium isotopes, § 3,5-13. Upon resuming con-
struction in 1945, it was suggested independently by Professor Edwin ::vr.
l\foMillan of the University of California and V. Veksler of Russia that the
696 MODERN PHYSICS Chap. 35

applied r-f potential should have its frequency lowered slightly as the mass
of the accelerating particle increased, in order to maintain resonance. This
principle was applied, and resulted in a so-called frequency-modulated cy-
clotron or synchrocyclotron. In it the particles go through their spiral path
in bursts as the oscillator frequency is varied, and then emerge in pulses at
the rate of about 60 per second instead of in a continuous beam. The Berke-
ley machine produces 350-Mev protons.
The Proton Synchrotron. The limit on particle energies that can be reached
with the cyclotron is set by the size of magnet required to produce a uniform
field across the chamber in which the particles are accelerated. The giant
magnet of the Berkeley cyclotron contains 3700 tons of steel and 300 tons of
copper, and its power requirements are correspondingly large. Still higher

Vacuum chamber Coils

-----~----- ''
''
T
/// ,,---1 I____ _ ' \
\
C-magnet
I

/ // [eJ ', \
I

0, { /
11 1
\ 11 I
I
R-f oscillator \ l
I I
I I I
I I I
I I I I I I
L J L .J I I
'' Target I I
Ion injector ,.'__,__ r 11 I
I 11 I
I 11 I
I JI I
I
/ 11 I

~
r ___ ,., , /

wII
Frn. 35-9. The cosmotron. I, plan view; II, sectional view, showing shape of magnets
and location of windings

particle energies are needed, however, in order to study the nature of the
forces acting within the nucleus-energies estimated to be of the order of a
billion electron· volts. This objective can be attained by accelerating pro-
tons in a so-called proton synchrotron.
Protons are now being accelerated to these high energies by the proton
synchrotron completed in 1952 at the Brookhaven National Laboratory on
Long Island. The shape of this "proton race track" of 60 ft diameter is
indicated in part I of Fig. 9; it consists of a vacuum chamber having four
curved portions with straight segments between them. In one straight sec-
tion, monoenergetic protons are injected by an electrostatic generator. These
fast-moving particles follow the desired path in the evacuated tube because
of the action of magnets which surround it; their coils are energized by a
motor-generator to give the correct magnetic flux density for the particular
§ 35-8 MACHINES FOR PRODUCING HIGH-ENERGY PARTICLES 697

particle speed at any instant. As the group of protons passes through the
fourth straight section it is accelerated by an applied r-f electric field, some-
what as in the cyclotron and linear accelerator. The protons continue to go
around in the chamber, gaining energy each time they pass this accelerating
section.
Part II of the figure shows how the C-shaped magnetic pole pieces and ,vind-
ings surround the vacuum chamber, of cross-sectional dimensions 6 in. X 36 in.
As first operated in 1952, a burst of about 10 12 protons is injected into the
chamber with an energy of 3.5 Mev; the protons are accelerated over a total
period of almost pne second, during which they make some 3 million revolu-
tions and travel 130,000 miles. ·within this period they reach an energy value
of 2300 Mev, after which they strike the target, inserted in the straight section
opposite the injector, and produce the effects sought. The frequency of the
oscillator varies between 350 and 4100 kilocycles per second during one ac-
celeration period. In this period the magnetic field must be increased in
order to keep the accelerating ions in the desired path; this is done by sending
a 7000-ampere pulse into the windings from a 40,000-kw motor-generator set.
Some of this energy is returned to the set and stored in its flywheel, for use
in the next pulse about 5 seconds later. The machine is aptly called a
cosmotron, because it gives energies approaching those found in cosmic rays.
An even larger proton synchrotron, called the bevatron, is under construction
at Berkeley; it is designed to give energies in the range 6000 to 10,000 ::\le\':
6 to 10 billion electron· volts. Design work is under way on a still larger
machine, one having a track a mile in circumference, which is expected to yield
particle energies of 100 billion electron· volts. This new accelerator will em-
ploy a recent development known as the "strong-focusing principle," and will
use a tube of smaller diameter and relatively smaller magnets and coils than
in the Brookhaven cosmotron.
The Betatron. A machine for accelerating electrons to very high speeds
was devised in 1940 by Professor Donald W. Kerst at the University of
Illinois. It is named the betatron after beta rays, which are ejected from
radioactive substances at comparable speeds. The machine is used chiefly
for producing x-rays of great penetrating power. The essential elements of
the betatron are shown in Fig. 10. It consists of an evacuated doughnut-
shaped glass tube placed between the poles N and S of a powerful electro-
magnet which is energized by alternating current in the circular coils. An
electron gun, not shown in the figure, is arranged to project the electrons
tangentially into the tube. Here they are acted upon by two forces: a radially
inward force that makes them follow the curvature of the tube, and a tan-
gential force that gives them higher speeds.
The electrons are first projected into the tube for a few microseconds just
as the magnetic field starts to build up, then they move at increasing speed
around the tube, and finally at a chosen instant they are deflected from their
circular path to the x-ray target. All this occurs within a quarter of a cycle,
698 MODERN PHYSICS Chap. 35

before the field has reached its maximum value; nevertheless, the electrons
travel around the tube many thousands of times during this period. In
succeeding cycles these operations are repeated over and over again.
The 100-Mev betatron constructed by the General Electric Company has a
"doughnut" of about 70-in. diameter. Electrons are accelerated to a speed
more than 99.99 per cent that of light, and achieve a relativistic mass about
200 times the rest mass. They are not accelerated in a steady stream as
in the electrostatic generator, but, owing to the nature of the acceleration
mechanism, come out in pulses at a rate of 60 per sec.

Vacuum chamber

Fw. 35-10. The betatron.

A limitation predicted theoretically and proved experimentally in betatron

operation is that the electrons at such high speeds lose energy by radiation,
because of their centripetal acceleration. Hence to achieve higher energies,
paths of larger radius should be used, and these involve a larger and heavier
magnet; further, there should be fewer revolutions between the injection of
electrons and the end of their path at the target. These principles have been
applied in designing the 300-Mev betatron being built under the direction of
Professor Kerst at the University of Illinois.
Another method for obtaining high electron energies is to apply the syn-
chrotron principle, as was done with the cyclotron, and to use C-shaped mag-
nets, as in the cosmotron. These modifications, and others, were employed
in the construction of two rather different 300-Mev electron synchrotrons,
one at the University of California and the other at the University of Michi-
gan.
35-9. Cosmic Rays, Positrons, and Mesons. Work leading to the discov-
ery of a penetrating radiation called cosmic rays extended over a number of
years. As early as 1900 observations showed that charged electroscopes even-
§ 35-9 COSMIC RAYS, POSITRONS, AND MESONS 699

tually became discharged, regardless of precautions taken. The effect was

first ascribed to radioactivity in the earth or the air, but extensive tests showed
that the electroscopes discharged more slowly when lowered into lakes or
taken into deep mines, and more rapidly when sent aloft in balloons. It was
suggested in about 1912 that the ionizing radiations which caused electroscope
discharge had their origin not in the earth but elsewhere, and the name cosmic
rays was adopted to denote this fact.
It took even longer to establish the nature of cosmic rays. Because of their
extremely high penetrating power, they were thought at first to be similar to
x-rays and gamma rays but of shorter wavelength and higher energy. Tests
conducted at various latitudes, however, showed that the rays are deflected
by the earth's magnetic field as they approach the earth from outer space,
causing their sea level intensity to be greater at the poles than at the equator.
Therefore they must be charged particles, at least in part. So-called "cosmic-
ray telescopes" were constructed by placing two or three Geiger counters in
alignment and connecting them in a coincidence circuit so that pulses would be
recorded only when a ray traversed all counters. Experiments with these
instruments established the fact that more cosmic rays come into the earth's
atmosphere from the west than from the east, and this indicates that the
charge of the particles is predominately positive.
The establishment of the exact nature of cosmic rays was hindered because
upon reaching the earth's atmosphere they produce "secondary" particles,
and the secondaries which arrive at the earth's surface greatly outnumber the
"primary" particles. It is now believed that primary cosmic rays are largely
fast-moving protons but may also include nuclei of some of the lighter atoms.
The mean energy of these particles is in the neighborhood of 6000 Mev, and
some particles may have a million times as much energy.
Absorption experiments show that cosmic rays can be classified into two
general types, the hard components and the soft components, according to the
rather artificial specification of whether they will or will not penetrate 10 cm
of lead. Since the earth's atmosphere is equivalent in its absorbing power for
cosmic rays to about 100 cm of lead, it is apparent that some hard cosmic rays
must penetrate the atmosphere entirely, in order to account for the large
number of softer secondaries found near the earth's surface. In fact, the hard
component of cosmic rays is thought to be produced by the primary radiation,
and the soft component by this secondary radiation.
The cloud chamber soon became the chief tool for studying cosmic rays.
When such a chamber is placed in a strong magnetic field, the direction of
curvature of the track produced by a particle indicates whether it is positive
or negative. Chambers were designed to be mounted on edge with a hori-
zontal lead plate across the center to slow down fast particles passing through
it. By noting the difference in curvature on the two sides of the plate, the
direction of motion of the particle can be established, and more information
can be obtained on its mass and charge. The use of coincidence-connected
700 MODERN PHYSICS Chap. 35

G-M tubes around the chamber gives automatic operation when an event of
predetermined magnitude occurs. Photographic records are made, sometimes
in stereoscopic pairs.
With such cloud chamber apparatus, Carl D. Anderson of the California
Institute of Technology observed in 1932 a track which could be accounted for
only by a particle with about the mass of an electron and with an equal but
positive charge. This constituted the discovery of the positron. The posi-
tron is now known to have the same rest mass as the electron but to have a
very short half-life, something of the order of a billionth of a second. When
the positron disappears it probably does so by uniting with an electron in
what has been called a "suicide pact"; the charges of the two particles neutral-
ize each other, and their combined mass is annihilated and reappears as its
energy equivalent (Eq. 35.3) in the form of one or two very penetrating
gamma rays.
The inverse of this mass annihilation process has also been confirmed experi-
mentally; it is called pair creation. The energy of a penetrating gamma ray
"materializes" into an electron-positron pair, and the tracks of the two par-
ticles may be recorded in a cloud chamber. No net electric charge is created
in the process, since the charges on the electron and the positron are equal and
opposite, but the event must take place in the neighborhood of a heavy nucleus
in order that momentum may be conserved. The minimum gamma ray
energy necessary for pair creation can be computed from Eq. 35.3, using form
the rest mass of the electron; thus for each particle
E = mc2 = (9.11 X IQ-31 ) X (3 X 108 ) 2 = 0.82 X 10-13 joule = 0.51 Mev.

If the energy of the gamma ray is greater than 1.02 Mev, the excess appears
as kinetic energy, shared equally between the electron and the positron. For
instance, if a gamma ray photon from naturally radioactive thorium C",
having an energy of 2.62 Mev, produces a positron-electron pair, 1.02 Mev
of the total is materialized as mass, and 2.62 - 1.02 = 1.60 Mev is divided
between the two ·newly created particles as kinetic energy.
When a fast-moving electron passes very close to an atomic nucleus in air
or in a denser medium it may be slowed down abruptly because of the force of
attraction between the two charged particles. It is predicted by electromag-
netic theory that the acceleration or deceleration of a charge causes the
radiation of energy, and this fact is borne out not only in the production of
the continuous spectrum from x-ray tubes, but also by the observation of
gamma rays produced when cosmic ray electrons are rapidly decelerated.
This type of radiation has been given the German name Bremsstrahlung
(meaning braking radiation). If the gamma ray thus produced has energy
greater than 1.02 Mev, it may in turn produce an electron-positron pair,
and in doing so multiply the number of particles.
An even greater production of particles occurs in what are known as cosmic
ray showers. The effects of such showers may be observed in a cloud chamber
§ 35-10 NEUTRONS AND NUCLEAR STRUCTURE 701

when a very energetic electron strikes the upper side of a lead plate and causes
dozens or hundreds of particle tracks to emerge from the lower side of the
plate. This multiplicity of particles is thought to be produced by repetitions
of the Bremsstrahlung-pair-creation process just described. Tests with widely
spaced, coincidence-connected G-M tubes indicate that such showers occur
occasionally in the upper atmosphere and cascade toward the earth's surface,
where they cover an area perhaps a thousand feet in diameter. The energy of
the primary particle that sets off such a sequence of events must be of the
order of a billion Mev or more, and represents the greatest energy thus far
encountered for .a single particle.
In 1935 the Japanese theoretical physicist Hideki Yukawa suggested the
existence of a charged particle with about 140 times the rest mass of the elec-
tron and a half-life of about a microsecond; this idea was introduced in an
attempt to explain the forces that act within the atomic nucleus. A year or
two later several cosmic-ray workers photographed cloud-chamber tracks that
were characteristic of neither an electron nor a proton, but perhaps of a par-
ticle of intermediate mass. Further research has confirmed the existence of
such heavy particles, and they have been called mesons; some are positive, some
negative, and some without charge, but all have an exceedingly short life,
perhaps of the order of lQ----6 to 10-9 sec. Distinction has been made between
two general groups of mesons, those with about 215 and those with about 285
times the rest mass of the electron, called respectively µ and 1r mesons. It
has been established that mesons are a major component of secondary cosmic
rays, and that mesons account chiefly for the "hard" cosmic rays. Addi-
tional particles of this intermediate range have subsequently been discovered.
Mesons were first produced artificially in 1948 with the synchrocyclotron
at the University of California, by using a beam of 380-Mev alpha particles.
The ability to produce mesons in the laboratory, particularly by the higher-
energy bevatrons and cosmotrons, is expected to reveal the nature and prop-
erties of these particles and to throw light on the little-understood forces that
bind the nucleus together.
35-10. Neutrons and Nuclear Structure. The neutron, another elementary
particle, was discovered in 1932 by a co-worker of Rutherford at Cambridge
University, James Chadwick. He established the fact that the penetrating
radiation resulting from the alpha-particle bombardment of various sub-
stances was a new particle with about the mass of the proton but with zero
charge. He produced these neutrons by bombarding beryllium metal with
alpha particles from polonium. ·while the neutrons emerging from the beryl-
lium could not be detected directly since they lacked electric charge, they
could by elastic collisions eject hydrogen nuclei from substances such as
paraffin that contained many hydrogen atoms, and these high-speed protons
could be detected with an ionization chamber or by other means, § 35-6. The
nuclear transformation is as follows:
(35.9)
702 MODERN PHYSICS Chap. 35

where the symbol 0n 1 denotes the neutron with its zero charge and its mass
number of 1. The resulting atom is the common isotope of carbon. In the
shorthand notation mentioned in § 35-7 this transmutation would be written
as Be9 (a, n)C 12 • This particular reaction is still used for the production of
copious quantities of neutrons; an appropriate radioactive material mixed
with beryllium powder provides a compact, inexpensive neutron source.
Neutrons can also be produced in the cyclotron, by bombarding a substance
containing heavy hydrogen with fast deuterons. The reaction is
(35.10)
where the symbol 1H2 represents both the heavy hydrogen atom and its nu-
cleus, the deuteron; the resulting atom is a rare isotope of helium, of mass
number 3.
Neutrons are very useful in producing nuclear disintegration; because they
have no charge they are able to approach and penetrate even the most heavily
charged nuclei without encountering repulsive forces. The Italian physicist
Enrico Fermi carried out extensive experiments with neutron reactions in the
years just preceding World vV ar II and produced many new isotopes, some
of which were radioactive. He found that "slow" neutrons were even more
efficient than "fast" neutrons in producing nuclear disintegration.
Neutrons cannot be slowed down by the means ordinarily used for charged
particles, such as the production of ion pairs; consequently the principle of
elastic impact is employed. In a head-on collision between two particles
of equal mass, one moving and one stationary, the kinetic energy of the former
is transferred entirely to the latter. Hence neutron-slowing devices, called
moderators, consist of hydrogenous materials like water, paraffin, or other or-
ganic compounds, because the energy is transferred efficiently from the neu-
trons to the almost equally massive hydrogen nuclei. Neutrons which have
been slowed down until their kinetic energy is a fraction of an electron· volt
are called thermal neutrons, because their energy is about equal to the average
kinetic energy of atoms and molecules at ordinary room temperatures. In
shielding workers in nuclear laboratories, water tanks can be used to provide
protection from neutrons, whereas lead blocks are needed for protection from
gamma rays.
According to the currently accepted theory of nuclear struct1,1re, the principal
components of the nucleus are protons and neutrons. The major unanswered
question concerns the forces that hold the nucleus together against the electro-
static forces of repulsion among the protons. The forces of attraction are
thought to be associated with mesons, but their exact nature is not yet under-
stood. These forces must be very large, for the distances involved are of the
order of 10-13 cm.
The helium nucleus consists of two protons and two neutrons; it must
be a very stable combination, since the alpha particle is ejected as a single
entity from nuclei in many radioactive disintegrations. The definite ranges
§ 35-11 ARTIFICIAL RADIOACTIVITY 703

(energies) of alpha particles suggest the existence in the nucleus of finite

energy levels, somewhat similar to those of the valence electrons, § 34-4, and
this idea is confirmed by the definite energies possessed by the gamma rays
that sometimes accompany the emission of alpha particles. The emission of an
alpha particle appears to leave the nucleus in an excited state, and the sub-
sequent emission of the gamma ray appears to restore the nucleus to its stable
ground state of lowest potential energy. The analogy with atomic spectra is
not simple, for the beta particles from a radioactive element are not emitted
with a definite energy, but may have energy values within a rather wide con-
tinuous band. One explanation suggested for the continuous beta-ray spec-
trum is that another particle is emitted simultaneously with the beta ray, and
carries away the energy difference between a specific intra-nuclear value
and the observed energy continuum. This hypothetical particle, named the
neutrino, must have zero charge, a very small mass (probably zero rest mass),
and practically no interaction with matter. Although numerous indirect ex-
periments have been suggested to demonstrate its existence, the results are
not conclusive. Energy and momentum must be conserved in beta-ray emis-
sion, if these basic principles are not to be abandoned, and the neutrino offers
a possible mechanism for their conservation.
The question may well be asked how beta emission could take place if
electrons are not thought to exist within the nucleus-as a matter of fact,
the electron is much too large to fit inside the nucleus. It has been suggested
that if, within the nucleus, a neutron changes into a proton spontaneously,
then in order to conserve charge an electron should be emitted. Presumably
the electron materializes on the surface of the nucleus, and carries away the
charge as well as some energy. Conversely, if a nuclear proton changes into
a neutron, a positron should be emitted. The process of positron emission
has been observed frequently in artificial or induced radioactivity, a subject
described in the following section.
The neutron is not a stable particle when outside the nucleus; it is believed
to have a half-life of about 20 min, and to be converted into a proton with
the emission of an electron, and also a neutrino.
35-11. Artificial Radioactivity. The radioactivity considered earlier in this
chapter is that which occurs naturally and spontaneously in a few dozen
nuclides, chiefly of the heaviest atoms. But in 1934 Frederic Joliot and his
wife Irene Curie-Joliot, a daughter of the Curies who discovered radium, de-
termined, by using projectiles from natural radioactivity, that some of the new
nuclei resulting from transmutation processes also are radioactive, emitting
either a positron or an electron. Trouble had been experienced in several
nuclear laboratories with the electroscopes used in detecting ionizing particles;
although the instruments functioned properly at first, they soon developed a
definite discharge rate, as if they had somehow been contaminated by a
radioactive substance. The Curie-Joliots correctly attributed this effect to
the bombardment of the electroscope casing, usually of aluminum, by alpha
704 MODERN PHYSICS Chap. 35

particles and to the formation of a radioactive material according to the

nuclear equation:
(35.11)
where the symbols 2H 4 and 0n 1 have their usual meanings. The phosphorous
nucleus is not stable, but emits a positron to form a stable nucleus:

(35.12)

where the symbol 1e0 represents the positron; its charge is +1 and its mass
negligible. The half-life of the decay of the phosphorus is 14 days.
The new subject of artificial radioactivity-so-called induced radioactivity
-was investigated vigorously, and all elements were ultimately produced in
radioactive form, by the bombardment of appropriate· substances both with
natural alpha particles and with protons and deuterons accelerated in high-
energy machines.
An interesting variant to positron emission in artificial radioactivity is an
electrically equivalent process called K capture, wherein one of an atom's two
most tightly bound electrons in the K shell ( § 34-7) is captured by the nucleus
as a radioactive process. This unusual event can be verified by observing
the characteristic K x-ray lines of the stable element thus formed.
Whether an electron or a positron is emitted in artificial radioactivity is
apparently determined by the direction in which the nuclear charge (atomic
number) must be changed in order to reach a stable nucleus. Alpha particles
are emitted only by the heavier nuclei and only in natural radioactivity, with
a very few exceptions, such as the disintegration indicated in Eq. 35.8, wherein
the resultant beryllium nucleus is apparently so unstable that it breaks up
immediately into two alpha particles.
J\fany practical applications were found for artificial radioactivity, as soon
as its nature was understood. It is used as a substitute for radium and re-
lated elements in treating diseased body tissues. The advantages of the new
materials center around the fact that their life is so short that they need not
be recovered from the body. For example, radioactive iodine, isotope 128,
has a half-life of 25 minutes, and radioactive sodium 15 hours. The quantity
of radioactive material used, rather than the length of treatment, then, deter-
mines the therapeutic dose. Such materials can be administered orally or in
any other convenient way, and can usually be localized more efficiently in the
part of the body in which treatment is needed.
Artificial radioactivity is much used for its "tracer" properties. Radio-
active phosphorus has been employed in studying the mechanism by which
plants obtain the important nutritional element phosphorus from the soil and
utilize it in building plant tissues; it is assumed that the radioactive phos-
phorus atoms follow the same path within the plant as do the normal atoms.
Studies of human nutrition have been made with radioactive iodine and so-
dium; a sensitive electroscope or a Geiger counter is used to trace the passage
§ 35-12 NUCLEAR ENERGY 705

of the "tagged" atoms of these elements through the biological system and to
measure their ultimate distribution and accumulation tendencies.
An element for use as a tracer need not be radioactive. A rare but stable
isotope of a common element, for example carbon 13 or oxygen 17, can be
produced through the transmutation process and then added, in suitable
chemical form, to a biological or other system. Subsequent analysis by the
mass spectrograph indicates the relative percentage of the unusual isotope,
anc;I hence permits tracing its path through the system. The secret of photo-
synthesis, by which the sun's energy is utilized in growing plants, is being
probed vigorously with tracer techniques.
35-12. Nuclear Energy. In determining the gain or loss of mass and energy
in nuclear transformations, two units are useful in expressing the quantities
concerned. The first of these is the atomic mass unit (amu) defined in § 35-4
as 1\ the mass of the oxygen atom of mass number 16; this is found to be
1.66 X 10-27 kg. The second is the electron· volt defined in § 18-12; its mul-
tiple, a million electron·volts (Mev), is given in§ 35-7 as 1.602 X 10-13 joule.
The Einstein Equivalence, E = mc2 , may be applied, and if values accurate
to five figures are used form and c the energy equivalent to 1 amu is found to
be 1.4916 X 10-10 joule. Hence the relationship between the two units is
1 amu = 931 Mev,
or 1 Mev = 1.074 X 10-3 amu.
In reactions such as those illustrated by Eqs. 35.6 and 35.9, not only can
the charge and mass numbers be balanced, but a balance can also be made of
mass plus energy; to do so requires that the various nuclide and particle masses
be known accurately, and that the kinetic energy of the particles and the
recoil nuclei be taken into account.
The historic Cockroft-Walton disintegration experiment described in § 35-7 will
serve as an example of nuclear calculations. The bombardment of lithium by protons
resulted in the formation of unstable nuclei that split into alpha particles, according
to the relation
+
1H1 3LF -----t 2 2He4 •
The masses of these nuclear quantities on the two sides of the equation are:
Left Right
Proton 1.0076 amu Helium 4.0028 amu
Lithium 7.0165 amu Helium 4.0028 amu
8.0241 amu 8.0056 amu
Hence in the reaction there is a disappearance of mass amounting to 0.0185 amu.
The annihilation of this mass represents an energy equivalent of 17.2 Mev. When
an allowance of 0.3 Mev is made for the minimum kinetic energy of the proton neces-
sary to cause transmutation, the amount of mass converted into energy and carried
away as kinetic energy by the two alpha particles is 16.9 Mev. This result is confirmed
by cloud chamber photographs, which show that the range of each alpha particle
corresponds to an original kinetic energy of 8.5 Mev.
706 MODERN PHYSICS Chap. 35

The mass of a nucleus is smaller than the sum of the masses of its constitu-
ents or so-called nucleons: protons and neutrons. This mass difference is
referred to as the binding energy of the nucleus and may be thought of as the
mass equivalent of the potential energy of the nuclear structure, or as a
measure of the work that would have to be done to break up the nucleus into
its constituent particles. The alpha particle is an especially stable combina-
tion of nucleons; its binding energy can be computed by taking the sum of
the masses of the four constituent particles and subtracting the mass of the
alpha particle itself. For the protons, 2 X 1.0076 = 2.0152 amu; for the n~u-
trons, 2 X 1.0089 = 2.0178 amu; this gives a total of 4.0330 amu. Mass
spectrographic determinations give the value 4.0028 amu for the alpha par-
ticle, so its binding energy has the relatively large value of 0.0302 amu, or
28.1 Mev per atom. Thus, in the formation of 4 gm (1 gram atom) of helium
from its elementary constituents an extremely large amount of energy is re-
leased: 6.02 X 1023 X 28.1 = 1.69 X 1025 Mev = 2.71 X 1012 joules, or about
750,000 kw· hr. The binding energy of helium can also be expressed as
290 billion Btu per pound.
It is believed that the source of the sun's heat is a reaction in which four
protons combine to give one alpha particle, with the emission of two positrons.
Although the exact nature of the nuclear processes is not known, it seems
probable that the physical conditions of great pressure and exceedingly high
temperatures within the sun are such
>(I)
10.0 that nuclear reactions can take place
~ directly. Hence it appears that the
c0 8.0?/
---
r---.
(I) :/-c sun is a tremendous nuclear furnace in
u:> I
'°1'- He which mass is converted into energy;
: 6.0 this energy radiates into space as heat
(I)
0.
>- and light.
E>
Q)
4.0 An important consideration in nuclear
c
Cl) computations is the binding energy
g> 2.0 per nucleon. The variation of this
-=oc
quantity with mass number or number
iii
0 50 100 150 200 250 of particles (nucleons) in the nucleus is
Mass number shown in Fig. 11. Since larger values of
F ra. 35 . d. 1
· - 11 . B m 1ng energy per nuc eon, binding energy indicate. more stable nu-
for the elements of the periodic table clei, the curve shows that nuclei of mass
numbers in the medium range of 40 to
120 are the most stable, and that the stability falls off slowly for the heavier
nuclei and very rapidly for the lighter ones. The very stable helium nucleus
has a value lying above the curve, as do also, but to a less extent, the nuclei
of carbon and oxygen with mass numbers, respectively, of 12 and 16-multi-
ples of four. (The nucleus of mass number 8 does not exist naturally, as
has been mentioned.)
An examination of the curve of Fig. 11 suggests two processes for obtaining
§ 35-13 NUCLEAR FISSION 707

energy in nuclear reactions: by the fusion of lighter elements into the more
stable, heavier ones; and by fission of the heaviest elements into lighter frag-
ments. As stated above, the fusion of protons into alpha particles is thought
to be the source of solar energy; this process probably accounts also for the
energy of many stars. The ability to produce controlled fusion reactions with
protons or other light nuclei would lead to an attractive source of power. If
the reactions proceed at an explosive rate, a new, concentrated, and very pow-
erful type of ordnance is possible-the so-called hydrogen or fusion bomb.
Since the curve drops off from a maximum of about 8.8 Mev per nucleon to
only 7.4 Mev for tl,le heaviest naturally occurring nucleus, it would seem that
the fission process would furnish less energy than the other. The nucleus of a
heavy element, however, contains many nucleons (uranium has over 230), so
the fission of one such nucleus into two nuclei of about half the mass (and with
binding factors of 8.4) should release more than 200 Mev of energy.
35-13. Nuclear Fission. That the process of nuclear fission can indeed be
produced in uranium and some other elements is now common knowledge.
Fission was first discovered in 1938 by two German nuclear chemists, Otto
Hahn and Fritz Strassmann, who found traces of barium in uranium that had
been bombarded by neutrons. The explanation was not immediately ap-
parent to them, and was given a few months later by two German physicists,
Lise Meitner and Otto R. Frisch. The reality of the fission process was con-
firmed within a few weeks in four different laboratories in the United States.
Apparently the entrance of a neutron into the uranium nucleus renders that
nucleus so unstable that instead of emitting one or more particles it breaks up
into two fragments of approximately equal size. Such fragment nuclei have
since been identified as elements between selenium and lanthanum, of atomic
numbers 34 and 57, respectively. Since the fragments may contain too many
neutrons for stability, they are themselves very likely radioactive. To cause
the fission of the most abundant isotope of uranium, that of mass number 238,
requires bombardment with neutrons of energies greater than 1 Mev; this is
true also for the fission of protactinium 231 and thorium 232. In contrast,
thermal neutrons, with an energy of a fraction of an electron· volt, will cause
the fission of the uranium isotope of mass number 235, which isotope occurs
naturally with a relative abundance of only 0.7 per cent.
Of primary importance is the fact that one or more neutrons are emitted in
the fission process. If these neutrons can be utilized to produce additional
fissions, then the process becomes sustaining and a "chain reaction" results.
Such a reaction cannot occur with natural uranium for the reason that not
enough high energy neutrons are produced in the fission process. But the
possibility of success is much better with uranium 235, if it can be obtained in
sufficient quantity and be separated from the more abundant isotopes. This
type of separation is difficult to carry out, because physical rather than
chemical methods must be used. Nevertheless, several effective but quite
expensive physical separation processes have been devised-one was the use
708 MODERN PHYSICS Chap. 35

of the 184-inch cyclotron magnet mentioned in § 35-8 as a mammoth mass

spectrograph called the Calutron.
For the chain reaction to continue even in uranium 235, the neutrons must
be captured by the uranium nuclei, rather than by impurities or by uranium
238 nuclei, and must be captured before they escape from the reacting mass.
The first condition imposes problems of chemical purity; the second suggests
that there is a critical size for the reacting mass and that, for smaller masses,
too many neutrons escape through the surface and the reaction stops. Since
uranium 238 nuclei do not capture slow neutrons, the chain reaction can be
maintained even in the presence of natural uranium if the kinetic energy of the
fission-produced neutrons is quickly reduced. This can be done by adding
moderators such as heavy water or very pure graphite, which take energy
from the neutrons by elastic collisions but do not capture them. These con-
cepts were utilized during the period 1940-42 by a group of scientists working
at the University of Chicago under the leadership of Professor Fermi, who
had come from Italy. On December 2, 1942, this group succeeded in pro-
ducing the first self-sustaining nuclear reaction by using the uranium 235
atoms in natural uranium oxide as the "fuel."
This development in nuclear physics created a new type of heat engine; it is
called a pile or nuclear reactor. Many piles have since been constructed, and
tests are being made to explore their possibilities as sources of domestic and
industrial power. The pile consists of a large cube, several feet on a side, of
very pure graphite, into which is placed uranium oxide as chunks or in easily
replaceable rod form. Various apertures may be provided for inserting ma-
terials for neutron irradiation. When enough uranium has been put in place,
the chain reaction sets in and the pile commences to operate. The reaction
rate is amenable to fine control by inserting rods of cadmium steel to different
distances into holes in the pile~the cadmium nuclei are excellent absorbers
of slow neutrons. But the heat produced in a pile is at such a low temperature
that thermodynamic difficulties stand in the way of its utilization for ordinary
power purposes, such as generating electricity or propelling ships and aircraft.
Also, because of the strong radioactivity, adequate shielding must be provided
for persons operating the pile, and care must be taken in the disposal of waste
gases, of various by-products, and of the water used for cooling. Fortunately,
the half-life of most of the radioactive elements involved is relatively short, so
that the disposal problems are not insuperable.
Nuclear reactors are used primarily at present to produce another readily
fissionable element, plutonium, and to produce many other artificially radio-
active nuclides for use in tracer studies, radio-therapy, and nuclear research.
Piles are also used in studying the effects of neutron bombardment on struc-
tural materials. The large reactor at the Brookhaven National Laboratory
is operated at about 27,000 kw, but most of the energy is carried away as heat
by streams of cool air forced through the pile by large fans.
If two masses of fairly pure uranium 235 are each below the critical size
§ 35-13 NUCLEAR FISSION 709

mentioned above, but jointly exceed this size, then when they are brought
together the nuclear reaction will proceed with explosive violence. This is the
basis of operation of the so-called atomic bomb. More than one useful neu-
tron is produced by the fission of each uranium 235 nucleus; in that way the
process multiplies at an exceedingly rapid rate, to explosion.
Fission by slow neutrons can also be produced in nuclei of an isotope of the
element plutonium, atomic number 94, and this element has the advantage
that it can be separated and purified by chemical methods which are much less
expensive and less laborious than the physical methods needed for purifying
uranium 235. The plutonium isotope that is fissionable by slow neutrons has
a mass number of 239. It is produced by irradiating uranium 238 with fast
neutrons; this process gives some isotopes of mass number 239, and the excess
energy is carried away by gamma rays. The new nucleus is an unstable elec-
tron emitter with a half-life of 23 min and which disintegrates into neptunium
239, a nuclide that is also radioactive, with a half-life of 2.3 days; this emits
another electron and becomes plutonium 239, the desired product. This
series of nuclear events can be expressed in equation form as follows:
on1 + g2U23s----+ 92U239----+ 93Np239 + -1eo;
(35.13)
The plutonium 239 nucleus is an alpha emitter with a half-life of 24,000 years.
Plutonium is "manufactured" by the processes indicated by Eq. 35.13, and
subsequently purified by chemical means, in mammoth atomic energy plants
such as the Hanford Engineering Works in the State of ashington. ,:v
Thus one hears of "uranium" and "plutonium" bombs, both of vrhich
operate by the fission process, and also of fusion bombs, in which the fuel is
probably a rare isotope of hydrogen.
Plutonium is one of the so-called transuranic elements-those with atomic
number greater than 92. Such elements are not found in nature, either be-
cause they are so radioactive that all traces of them have disappeared or
because they have never existed in the natural state. The discovery of ele-
ments 93 to 97, all strongly radioactive, was first reported by Fermi during his
neutron transmutation investigations of 1934-38. Questions were raised as
to the correctness of his identification, particularly when it was realized that
the fission process could take place with uranium, and that highly radioactive
fragments w~re produced therein. But the identity of the elements nep-
tunium, plutonium, americium, and curium has now been firmly established,
chiefly through the experiments of Professors Edwin M. JVIcMillan and Glenn
T. Seaborg and their co-workers at the University of California in Berkeley
during the period 1940-44. Several isotopes of each new element have been
identified; all are radioactive, and neptunium 237 has an estimated half-life
of over two million years. Two additional elements, numbers 97 and 98,
were subsequently discovered at this university in 1949-50 and were appro-
priately named berkelium and californium; the discovery of elements 99 and
100 ·was announced early in 1954.
710 MODERN PHYSICS Chap. 35

PROBLEMS
1. Compute the kinetic energy in joules of an electron moving at a speed of
8 X 107 m/sec.
2. How fast must an electron move to quadruple its mass?
3. The kinetic energy of a certain electron is 5.00 X 10-14 joule. Find the rela-
tivistic mass of the electron.
4. Write Eq. 35.2 in a form that expresses the mass increase of a particle due to an
increase in speed, in terms of the kinetic energy of the particle and the speed of light.
For an electron of energy 6 X 10-14 joule, compute the increase in mass (over the rest
mass) and also the relativistic mass.
5. How many grams of matter would have to be annihilated in order to furnish
electrical power of 360 million kilowatts for one minute? Assume 100 per cent con-
version of matter to energy.
6. How many kilowatt·hours of energy would result from the annihilation of 3 mg
of matter? Assume 100 per cent conversion of matter to energy.
7. The radioactivity of radium C is reduced to Th of its original value in 140
minutes. Determine the half-life of radium C.
8. Derive the relationship between radioactive decay constant and half-life, start-
ing with Eq. 35.4.
9. The radioactivity of radon, s6Rn222 , is reduced to -h of its original value during
a certain experiment. Compute the duration of this experiment.
10. Determine the numbers of protons and of neutrons in the follmving nuclides:
He4, 0 16 , Si28 , Cu65 , Cd 112 , Ba137 • Present the data in table form, and make a plot with
the number of protons as abscissas and the number of neutrons as ordinates. Com-
ment on the trend of this graph.
11. When uranium 238 undergoes spontaneous disintegration, the following particles
are emitted in order: ex, /3, {3. Write equations for these disintegrations, and identify
the resulting nuclides as to atomic number, chemical name, and mass number.
12. The thorium radioactive family sequence has the following nuclide names and
atomic numbers: Thorium (90), mesothorium 1 (88), mesothorium 2 (89), radiothorium
(90), thorium X (88), thoron (86), thorium A (84), thorium B (82), thorium C (83),
thorium C' (84), lead (82). Starting with 90Th2.12 , list in table form the nuclide name,
atomic number, chemical name, mass number, and emission (ex or /3 particle) that pro-
duces the next following member of the family.
13. Express the kinetic energy of the electron of Prob. 1 in millions of electron· volts
(Mev).
14. The frequency of the potential difference applied to the <lees of a cyclotron is
7 X 106 cycles/sec. Determine the magnetic flux density necessary to accelerate alpha
particles.
15. Solve Prob. 14 for the acceleration of protons.
16. Compute the wavelength of a 1.80-Mev gamma ray; e~press the result in
millimicrons.
17. The gamma ray of Prob. 16 creates an electron-positron pair, and its energy is
divided equally between the two particles. (a) Compute the kinetic energy of each
particle, in Mev. (b) Compute the mass of each.
18. A positron and an electron, both having negligible kinetic energy, meet in free
space and are annihilated; two photons of equal energy are created in the process.
Find the wavelength, in m,u, and the momentum of each photon.
19. Write Eqs. 35.10 and 35.11 in "shorthand" form.
20. Isotope 75 of selenium becomes converted to a stable nucleus by the process of
K capture. Identify the nuclide produced, and write the nuclear equation for the
process.
 PROBLEMS 711

21. Take the mass of the deuteron as 2.0142 amu, and compute the binding energy
of the deuteron, in Mev.
22. The binding energy per nucleon is 7.6 Mev for uranium 235, and is about
8.7 Mev for the most stable nuclei. Use these data and compute the maximum
amount of energy that could come from the fission of a U 235 nucleus. Suggest reasons
why the actual energy release is only about 200 Mev per nucleon.
23. Compute the kinetic energy with which the positron of Eq. 35.12 is emitted,
using the following mass data: phosphorus atom, 29.9873 amu; silicon atom, 29.9832
amu; positron, 0.0005 amu.
 APPENDIX I
TABLE 1. CONVERSION FACTORS

Length
1 mile (mi) = 1760 yards (yd) = 5280 feet (ft) = 1609 meters (m).
1 foot = 12 inches (in.) = 0.3048 m.
1 meter = 100 centimeters (cm) = 3.281 ft = 39.37 in.
1 inch = 2.540 cm.
1 micron (µ) = 10-3 millimeter (mm) = 1000 millimicrons (mµ).
1 meter = 1000 mm = 106 µ = 109 mµ.
Volume
1 gallon (gal) = 231 in. 3 = 0.1337 ft3 = 3.785 liters.
1 liter = 1000 cm 3 = 61.0 in. 3

Speed
1 mile/hour (mi/hr) = 1.4G7 feet/second (ft/sec).
30 mi/hr = 44 ft/sec.
Angle
1 radian = 57.296 degrees (0 ).
1° = 0.01745 radian.

Force and Weight

1 newton = 105 dynes.
1 pound (lb) = 16 ounces (oz) = 4.445 newtons = 32 poundals.
1 cubic foot of water weighs G2.4 lb.

Pressure
1 atmosphere = 14.70 lb/in. 2 = 1.013 X 106 dynes/cm2 = 1.013 X 105
newtons/m2 •
Mass
1 kilogram (kg) = l 000 grams (gm) = 2.205 pounds.
1 slug = 32.17 pounds = 14.59 kg.

Energy
1 joule= 1 newton·m = 107 ergs= 0.738 ft·lh = 0.239 calorie (cal).
1 calorie = 4.186 joules.
1 foot· pound = 1.355 joules.
1 British thermal unit (Btu) = 252 cal = 1055 joules = 778 ft· lb.
1 electron· volt = 1.t302 X 10-19 joule.
1 Mev = 106 electron· volts.

Power
1 horsepower (hp) = 33,000 ft-lb/min= 350 ft·lb/sec = 746 watts.
1 kilowatt (kw) = 1000 watts = 1.340 hp.
713
 TABLE 2. CONSTANTS

Mathematical
1r = 3.1416 7r2 = 9.8696 l/1r = 0.3183

Physical
Atomic mass unit A = 1.66 X 10-27 kilogram.
Avogadro's Kumber N = 6.024 X 1023 •
Charge of the electron e = 1.fi02 X 10-19 coulomb.
Mass of the electron (at rest) m = 9.107 X 10-31 kilogram.
Mass of the proton (at rest) 1.G72 X 10-27 kilogram.
Planck's constant h G.62 X 10-34 joule· second.
Speed of light in vacuum c 2.9979 X 108 meters/second.
Universal gas constant per mole R = 8.31 joules/degree C.

TABLFJ 3. GEOMETRIC RELATIONS

Triangle
Right triangle of sides a and b and hypotenuse c: a2 = b2 + c2.
Any triangle of base b and height h: Area = ! bh.
. 1e of semi-. perimeter
Any tnang .
s =
a+b+c
2
Area = Vs(s - a)(s - b)(s - c).
Circle
of radius r: Circumference = 2 1rr; Area = 1rr2 •

Trapezi1.lm
of altitude h and bases b and b': Area = h ('> ~ I/).
Cylinder
Right cylinder of altitude h and base area A: Volume = Ah.

Sphere
of radius r: Surface = 4 1rr2 ; Volume = { 1rr3•

Pyramid or cone
of altitude h and base area Li: Volume = } Ah.
714
 TABLJ<J 4. MA TIIF;MATICAL RELA'l'IONS

Powers
1
- = a-x.
ax

Quadratic Equation
- b± -V b2 - 4: ac
Solution of ax 2 + bx + c = 0: x=
2a

Derivatives and integrals

(C and n are eom;tants; 1t, v, and :rare variables).
du dv
v--u-
d du dv d u dx dx
- (u
dx
+ v) = -
dx
+ -·
dx d:r v v2

-d cu = cdu
d du
, -· - /ln = nun-I - ·
dx dx dx dx
dC= 0 d . du
- Slll U = COS U - ·
dx · dx dx
d dv du d . du
dx (uv) = 11 d.r +- v rLc. - eos u = - su1 u - ·
dx d.t

fdu =it+ C. J du
-
u
= lnu + C.
JC du= Cfdu. fsin udu = -cos u + C.
fcosudu = sin1t + C.
TABLE 54 GIUJEK ALPHABET

A, a. Alpha K, Nu
-
jJ.

B, /3. Beta A ~- Xi
r, 'Y. Gamma 0, 0. Omicron
~' 0. Delta II, ?r. Pi
E, f. Eps17on P, p. .. Rho
Z, r. Zeta L
'
(}". Sigma
H, rJ • Eta T, T. Tau
e, e. Theta T, v. Upsilon
I, L ... Iota cf>, <p. Phi
K, K. Kappa X, x. Chi
A, 'A. Lambda 'V, it . Psi
M, µ. J.v!u n, w. Omega
715
 TABLE 6. NATURAL TRIGONOMETRIC FUNCTIONS

Angle Sine Cosine Tangent Angle Sine Cosine Tangent

---- - - -- - -
oo 0.000 1.000 0.000
10 .018 1.000 .018 46° 0.719 0.695 1.036
20 .035 0.999 .035 47° .731 .682 1.072
30 .052 .999 .052 48° .743 .669 1.111
40 .070 .998 .070 49° .755 .656 1.150
50 .087 .996 .088 50° .766 .643 1.192
!
50 .105 .995 .105 51° .777 .629 1.235
70 .122 .993 .123 52° .788 .616 1.280
go .139 .990 .141 53° .799 .602 1.327
go .156 .988 .158 54° .809 .588 1.376
10° .174 .985 .176
i.
55° .819 .574 1.428
110 .191 .982 .194 56° .829 .559 1.483
12° .208 .978 .213 57° .839 .545 1.540
13° .225 .974 .231 58° .848 .530 1.600
14° .242 .970 .249 59° .857 .515 1.664
15° .259 .966 .268 60° .866 .500 1.732

16° .276 .961 .287 61° .875 .485 1.804

17° .292 .956 .306 I 62° .883 .470 1.881
18° .309 .951 .325 63° .891 .454 1.963
190 .326 .946 .344 540 .899 .438 2.050
20° .342 .940 .364 65° .906 .423 2.145
I
21 ° .358 .934 .384 66° .914 .407 2.246
22° .375 .927 .404 67° .921 .391 2.356
23° .391 .921 .425 68° .927 .375 2.475
24° .407 .914 .445 69° .934 .358 2.605
25° .423 .906 .466 70° .940 .342 2.747

26° .438 .899 .488 71° .946 .326 2.904

27° .454 .891 .510 72° .951 .309 3.078
28° .470 .883 .532 73° .956 .292 3.271
29° .485 .875 .554 74° .961 .276 3.487
30° .500 .866 .577 75° .966 .259 3.732

31 ° .515 .857 .601 76° .970 .242 4.011

32° .530 .848 .625 77° .974 .225 4.331
33° .545 .839 .649 78° .978 .208 4.705
34° .559 .829 .675 79° .982 .191 5.145
35° .574 .819 .700 800 .985 .174 5.671

36° .588 .809 .727 81° .988 ,156 6.314

37° .602 .799 .754 82° .990 .139 7.115
38° .6Hi .788 .781 83° .993 .122 8.144
39° .629 .777 .810 84° .995 .105 9.514
40° .643 ' .766 .839 850 .996 .087 11.43

41° .656 .755 .869 86° .998 .070 14.30

42° .669 .743 .900 87° .999 .052 19.08
43° .682 .731 .933 88° .999 .035 28.64
44° .695 .719 .966 89° 1.000 .018 57.29
45° .707 .707 1.000 90° 1.000 .000 00
I I
 TABLE 7. TRIGONOMETRic RELAT!ONS

Right Triangle (C = 90°)

. A = -·
Slll
a .
Slll
B b
= -·
c' c
b a
a cos A = -· cos B = -·
c' c
a b
tan A=-· tanB = -·
b' a
b cos B = sin A.

Any Triangle

Law of Sines:
a b c
sin A = sin B = sin C.
Law of Cosines:
+
c2 = a2 b2 - 2 ab cos C.

Trigonometric Formulas
sin ()
sin 2 () + cos 2 () = 1; tan()=--·
cos()
sin (() ± cf>) = sin () cos cf> ± cos () sin cf>.
cos (() ± cf>) = cos () cos cf> =F sin () sin </>.

Approximations for Small Angles

sin() = () (radians) = tan 0.

Functions for Angles Larger than 90°

For angles beyond the first quadrant, the foregoing table of trigonometric
functions can be used by applying the following relations:

Second Quadrant Third Quadrant Fourth Quadrant

sin (90 + 0) = cos() sin (180 + 0) = -sin () sin (270 + 0) = -cos()

cos (90 + ()) = -sin () cos (180 + ()) = -cos() cos (270 + ()) = sin ()
tan (90 + 0) = -cot() tan (180 + 0) = tan() tan (270 + 0) = -cot()

717
 TABLE 8. LOGARITHMS TO BASE 10

N 0 1 2 3 4 5 6 7 8 9
-1.0 .0000 .0043 .0086 .0128 .0170 .0212 .0253 .0294 .0334 .0374
1.1 .0414 .0453 .0492 .0531 .0569 .0607 .0645 .0682 .0719 .0755
1.2 .0792 .0828 .0864 .0899 .0934 .0969 .1004 .1038 .1072 .1106
1.3 .1139 .1173 .1206 .1239 .1271 .1303 .1335 .1367 .1399 .1430
1.4 .1461 .1492 .1523 .1553 .1584 .1614 .1644 .1673 .1703 .1732
1.6 .1761 .1790 .1818 .1847 .1875 .1903 .1931 .1959 .1987 .2014
1.6 .2041 .2068 .2095 .2122 .2148 .2175 .2201 .2227 .2253 .2279
1.7 .2304 .2330 .2355 .2380 .2405 .2430 .2455 .2480 .2504 .2529
1.8 .2553 .2577 .2601 .2625 .2648 .2672 .2695 .2718 .2742 .2765
1.9 .2788 .2810 .2833 .2856 .2878 .2900 .2923 .2945 .2967 .2989
2.0 .3010 .3032 .3054 .3075 .3096 .3118 .3139 .3160 .3181 .3201
2.1 .3222 .3243 .3263 .3284 .3304 .3324 .3345 .3365 .3385 .3404
2.2 .3424 .3444 .3464 .3483 .3502 .3522 .3541 .3560 .3579 .3598
2.3 .3617 .3636 .3655 .3674 .3692 .3711 .3729 .3747 .3766 .3784
2.4 .3802 .3820 .3838 .3856 .3874 .3892 .3909 .3927 .3945 .3962
2.5 .3979 .3997 .4014 .4031 .4048 .4065 .4082 .4099 .4116 .4133
2.6 .4150 .4166 .4183 .4200 .4216 .4232 .4249 .4265 .4281 .4298
2.7 .4314 .4330 .4346 .4362 .4378 .4393 .4409 .4425 .4440 .4456
2.8 .4472 .4487 .4502 .4518 .4533 .4548 .4564 .4579 .4594 .4609
2.9 .4624 .4639 .4654 .4669 .4683 .4698 .4713 .4728 .4742 .4757
3.0 .4771 .4786 .4800 .4814 .4829 .4843 .4857 .4871 .4886 .4900
3.1 .4914 .4928 .4942 .4955 .4969 .4983 .4997 .5011 .5024 .5038
3.2 .5051 .5065 .5079 .5092 .5105 .5119 .5132 .5145 .5159 .5172
3.3 .5185 .5198 .5211 .5224 .5237 .5250 .5263 .5276 .5289 .5302
3.4 .5315 .5328 .5340 .5353 .5366 .5378 .5391 .5403 .5416 .5428
3.6 .5441 .5453 .5465 .5478 .5490 .5502 .5514 .5527 .5539 .5551
3.6 .5563 .5575 .5587 .5599 .5611 .5623 .5635 .5647 .5658 .5670
3.7 .5682 .5694 .5705 .5717 .5729 .5740 .5752 .5763 .5775 .5786
3.8 .5798 .5809 .5821 .5832 .5843 .5855 .5866 .5877 .5888 .5899
3.9 .5911 .5922 .5933 .5944 .5955 .5966 .5977 .5988 .5999 .6010
4.0 .6021 .6031 .6042 .6053 .6064 .6075 .6085 .6096 .6107 .6117
4.1 .6128 .6138 .6149 .6160 .6170 .6180 .6191 .6201 .6212 .6222
4.2 .6232 .6243 .6253 .6263 .6274 .6284 .6294 .6:304 .6314 .6325
4.3 .6335 .6345 .6355 .6365 .6375 .6385 .6395 .6405 .6415 .6425
4.4 .6435 .6444 .6454 .6464 .6474 .6484 .6493 .6503 .6513 .6522
4.6 .6532 .6542 .6551 .6561 .6571 .6580 .6590 .6599 .6609 .6618
4.6 .6628 .6637 .6646 .6656 .6665 .6675 .6684 .6693 .6702 .6712
4.7 .6721 .6730 .6739 .6749 .6758 .6767 .6776 .6785 .6794 .6803
4.8 .6812 .6821 .6830 .6839 .6848 .6857 .6866 .687.5 .6884 .6893
4.9 .6902 .6911 .6920 .6928 .6937 .6946 .6955 .6964 .6972 .6981
6.0 .6990 .6998 .7007 .7016 .7024 .7033 .7042 .7050 .7059 .7067
5.1 .7076 .7084 .7093 .7101 .7110 .7118 .7126 .7135 .7143 .7152
6.2 .7160 .7168 .7177 .7185 .7193 .7202 .7210 .7218 .7226 .7235
6.3 .7243 .7251 .7259 .7267 .7275 .7284 .7292 .7300 .7308 .7316
5.4 .7324 .7332 .7340 .7348 .7356 .7364 .7372 .7380 .7388 .7396
6.5 .7404 .7412 .7419 .7427 .7435 .7443 .7451 .7459 .7466 .7474
6.6 .7482 .7490 .7497 .7505 .7513 .7520 .7528 .7536 .7543 .7551
5.7 .7559 .7566 .7574 .7582 .7589 .7597 .7604 .7612 .7619 .7627
5.8 .7634 .7642 .7649 .7657 .7664 .7672 .7679 .7686 .7694 .7701
5.9 .7709 .7716 .7723 .7731 .7738 .774,5 .7752 .7760 .7767 .7774
-N 4 6
0 1 2 3 5 7 8 9
 'l'ABLl<J 8. LOGARI'r"HMS TO BASJ<~ 10 (Continued)

0 1 2 3 4 6 I 6 7
N 8 9
-6.0 .7782 .7789 .7796 .7803 .7810 .7818 .7825 .7832 .7839 .7846
6.1 .7853 .7860 .7868 .7875 .7882 .7889 .7896 .7903 .7910 .7917
6.2 .7924 .7931 .7938 .7945 .7952 .7959 .7966 .7973 .7980 .7987
6.3 .7993 .8000 .8007 .8014 .802] .8028 .8035 .8041 .8048 .8055
6.4 .8062 .8069 .8075 .8082 .8089 .8096 .8102 .8109 .8116 .8122
6.5 .8129 .8136 .8142 .8149 .8156 .8162 .8169 .8176 .8182 .8189
6.6 .8195 .8202 .8209 .8215 .8222 .8228 .8235 .8241 .8248 .8254
6.7 .8261 .8267 .8274 .8280 .8287 .8293 .8299 .8306 .8312 .8319
6.8 .8325 .8331 .8338 .8344 .8351 .8357 .8363 .8370 .8376 .8382
6.9 .8388 .8395 .8401 .8407 .8414 .8420 .8°!26 .8432 .8439 .844,5
7.0 8451 .8457 .8463 .8470 .8476 .8482 .8488 '.8494
.8500 .8506
7.1 .8513 .8519 .8525 .8531 .8537 .8543 .8549 .8555 .8561 .8567
7.2 .8573 .8579 .8585 .8591 .8597 .8603 .8609 .8615 .8621 .8627
7.3 .8633 .8639 .8645 .8651 .8657 .8663 .8669 .8675 .8581 .8686
7.4 .8692 .8698 .8704 .8710 .8716 .8722 .8727 .8733 .8739 .8745
7.5 .8751 .8756 .8762 .8768 .8774 .8779 .8785 .8791 .8797 .8802
7.6 .8808 .8814 .8820 .8825 .8831 .8837 .8842 .8848 .88.54 .8859
77 .8865 .8871 .8876 .8882 .8887 .8893 .8899 .8904 .8910 .8915
7.8 .8921 .8927 .8932 .8938 .8943 .8949 .8954 .8960 .8965 .8971
7.9 .8976 .8982 .8987 .8993 .8998 .9004 .9009 .9015 .9020 .9025
8.0 .9031 .9036 .9042 .9047 .9053 .9058 .9063 .9069 .9074 .9079
8.1 .9085 .9090 .9096 .9101 .9106 .9112 .9117 .9122 .9128 .9133
8.2 .9138 .9143 .9149 .9154 .9159 .9165 .9170 .9175 .9180 .9186
8.3 .9191 .9196 .9201 .9206 .9212 .9217 .9222 .9227 .9232 .9238
8.4 .9243 .9248 .9253 .9258 .9263 .9269 .9274 .9279 .9284 .9289
8.5 .9294 .9299 .9304 .9309 .9315 .9320 .9325 .9330 .9335 .9340
8.6 .9345 .9350 .9355 .9360 .9365 .9370 .9375 .9380 .9385 .9390
8.7 .9395 .9400 .9405 .9410 .9415 .9420 .942.5 .9430 .9435 .9440
8.8 .9445 .9450 .9455 .9460 .9465 .9469 .9474 .9479 .9484 .9489
8.9 .9494 .9499 .9504 .9509 .9513 .9518 .9523 .9528 .9533 .9538
9.0 .9542 .9547 .9552 .9557 .9562 .9566 .9571 .9576 .9581 .9586
9.1 .9590 .9595 .9600 .9605 .9609 .9614 .9619 .9624 .9628 .9633
9.2 .9638 .9643 .9647 .9652 .9657 .9661 .9666 .9671 .9675 .9680
9.3 .9685 .9689 .9694 .9699 .9703 .9708 .9713 .9717 .9722 .9727
9.4 .9731 .9736 .9741 .9745 .9750 .9754 .9759 .9763 .9768 .9773
9.5 .9777 .9782 .9786 .9791 .9795 .9800 .9805 .9809 .9814 .9818
9.6 .9823 .9827 .9832 .9836 .9841 .9845 .9850 .9854 .9859 .9863
9.7 .9868 .9872 .9877 .9881 .9886 .9890 .9894 .9899 .9903 .9908
9.8 .9912 .9917 .9921 .9926 .9930 .9934 .9939 .9943 .9948 .9952
9.9 .9956 .9961 .9965 .9969 .9974 .9978 .9983 .9987 .9991 .9996

-;;,I 0 1 2 3 4 6 6 7 8 9

log = logarithm to base 10 log 403 = 2.6053

In = logarithm to base e log 40.3 = 1.6053
e = 2.7183; In 10 = 2.3026 log 4.03 = 0.6053
In N = 2.3026 log N log 0.403 = 9.6053 - 10
log N"' = x log N log 0.0403 = 8.6053 - 10

719
 TABLE 9. PERIODIC TABLE OF THE ELEMENTS

PERIOD GROUP O GROUP I GROUP II GROUP Ill GROUP IV GROUP V GROUP VI GROUP VII GROUP VIII

1
I HYDROGEN
H 1.0080

2 3 4 5 6 7 8 9
II HELIUM LITHIUM BERYLLIUM BORON CARBON NITROGEN OXYGEN FLUORINE
He 4.003 Li 6.940 Be 9.013 B 10.82 C 12.010 N 14.008 0 16.000 F 19.00

10 11 12 13 14 15 16 17
Ill NEON SODIUM MAGNESIUM ALUMINUM SILICON PHOSPHORUS SULPHUR CHLORINE
Ne 20.183 Na 22,997 Mg 24.32 Al 26.98 Si 28.09 P 30.975 S 32.066 Cl 35.457

18 19 20 21 22 .23 24 25 26 27 28
ARGON POTASSIUM CALCIUM SCANDIUM TITANIUM VANADIUM CHROMIUM MANGANESE RON COBALT NICKEL
IV A 39.944 K 39.100 Ca 40.08 Sc 44.96 Ti 47.90 V 50.95 Cr 52.01 Mn .54.93 Fe 55.85 Co. 58.94 Ni 58.69
29 30 31 32 33 34 35
~~PPER63.54 ~:c 65.38 8!LLl~~72 i!RMA;~~: ~!SENI\4.91 ~:LENI~~96 ::oMr::.916

36 37 38 39 40 41 42 43 44 45 46
KRYPTON RUBIDIUM STRONTIUM YTTRIUM ZIRCONIUM NIOBIUM* MOLYBDENUM TECHNETIUM RUTHENIUM RHODIUM PALLADIUM
V Kr 83.80 Rb 85.48 Sr 87.63 Yt 88.92 Zr 91.22 Nb 92.91 Mo 95.95 Tc [99) Ru 101.7 Rh 102.91 Pd 106.7
47 48 49 50 51 52 53
-l SILVER CADMIUM INDIUM TIN ANTIMONY TELLURIUM IODINE
~ Ag 107,880 Cd 112.41 In 114.76 Sn 118.70 Sb 121.76 Te 127.61 I 126,91
0
54 55 56 57-71 72 73 74 75 76 77 78
XENON CESIUM BARIUM !LANTHANIDE HAFNIUM TANTALUM TUNGSTEN RHENIUM OSMIUM IRIDIUM PLATINUM
VI Xe 131.3 Cs 132.91 Ba 137,36 RARE EARTHS) Hf 178.6 Ta 180.88 W 183.92 Re 186,31 Os· 190.2 Ir · 193,1 Pt 195,23
79 80 81 82 83 84 85
GOLD MERCURY THALLIUM LEAD BISMUTH POLONIUM ASTATINE
Au 197.2 Hg 200.61 Tl 204.39 Pb 207.21 Bi 209.00 Po 210 At [210]

86 87 88 89 90 91 92
VII RADON FRANCIUM RADIUM ACTINIUM THORIUM PROTACTINIUM URANIUM TRANS-URANIUM ELEMENTS
Rn 222 Fr [223) Ra 226.05 Ac 227 Th 232.12 Pa 231 U 238.07

57 58 59 60 61 62 63 64
SERIES
ll! LANTHANIDE
VI LANTHANUM CERIUM PRASEODYMIUM NEODYMIUM PROMETHIUM SAMARIUM EU~OPIUM GADOLINIUM
La 138,92 Ce 140.13 Pr . 140,92 Nd 144.27 Pm [145] Sa 150.43. Eu 152.0 Gd 156.9
t 65 66 67 68 69 70 71
:;S TERBIUM DYSPROSIUM HOLMIUM ERBIUM THULIUM YTTERBIUM LUTETIUM
Tb 159.2 Dy 162.46 Ho 164.94 Er 167,2 Tm 169,4 Yb 173.04 Lu 174.99

VII
=
III

iii: ACTINIDE SERIES

89
ACTINIUM
90
THORIUM PROTACTINIUM
91 , 92
URANIUM
93
NEPTUNIUM
94
PLUTONIUM AMERICIUM
95 96
CURIUM
97
BERKELIUM
98
CALIFORNIUM
Ac 227 Th 232.12 P_a 231 U 238.07 Np [237) Pu 1242) Am (243) Cm (2431 Bk (245) Cf (244]

Above the name of each element is its atomic number; below, its chemical symbol and atomic weight.
A figure within brackets indicates the mass of the most stable known isotope.
Courtesy of General Electric Company. * Also known as Columbium (41).
 APPENDIX II

Relation Between Mks and Other Metric Units

The meter-kilogram-second system of units was devised to be of service in

all branches of physics, but primarily to lessen the confusion that results from
the intermingled use of three kinds of units in Electricity and Magnetism-
the electrostatic and electromagnetic units of the centimeter-gram-second
system, and the practical system. The mks system was proposed by Professor
Giovanni L. T. C. Giorgi in 1903, and was adopted by the International Com-
mittee on Weights and Measures in 1935.
In Mechanics~ the mks system introduces only one new unit-the newton;
the newton is the force needed to give a mass of 1 kg (1000 gm) an accelera-
tion of 1 m (100 cm) per sec per sec. In the cgs system, the dyne is the force
needed to give a mass of 1 gm an acceleration of 1 cm per sec per sec. Hence,
1 newton = 1000 gm X 100 cm/sec2 = 105 dynes. The unit of work in the
mks system is the work done by a force of 1 newton acting on a body through
a distance of 1 min the direction of the force; it is the newton· meter. Since
the newton is 105 dynes, and the meter is 102 cm, the newton· meter of work is
equivalent to 107 dyne· cm = 107 ergs = 1 joule. Consequently, the newton·
meter is the same as the joule. The mks unit of power is the newton· meter
per second, which is the same as the joule per second, or the watt.
In Electricity and Magnetism, the mks system incorporates the widely used
practical electrical units, such as the ampere, volt, ohm, joule, and watt; and
it eliminates the powers of 10 that occur in the relations between the cgs units.
The older systems have an advantage in the fact that for vacuum the permit-
tivity is unity with electrostatic units, and the permeability is unity with
electromagnetic units. Thus, in the electrostatic system the unit of electric
charge is defined as one which in a vacuum would repel an equal charge at a
distance of 1 cm with a force of 1 dyne. In the mks system the unit charge
(the coulomb) is one which in a vacuum would repel an equal charge at a
distance of 1 m with a force of 9 X 109 newtons; use of this value in Coulomb's
Law shows that the permittivity of vacuum not only has a numerical value
different from unity, namely 9 ) 109 in unrationalized units, but is a quan-
tity expressed in coulomb2/newton ·m2 • Similarly, the value of the perme-
ability for vacuum in the unrationalized mks system is 10-7 newton/amp2•
Although these facts must be considered in the fundamental definitions, the
expressions derived from them, and used much more often than the funda-
mental definitions themselves, have less numerical complexity; also, the units
are out in the open where they can be multiplied, divided, balanced, and can-
celed as simply as in the field of Mechanics.
The mks system described thus far in this Appendix simplifies the electrical
equations but falls short in simplifying the magnetic equations, for the factor
4 1r is still present, as it is with tho cgs units. So _that this factor may be
eliminated from the more frequently used equations, another version of the
721
mks system, called its rationalized form, assigns values to both permittivity
and permeability of vacuum which differ from those stated above in the
ratio 4 1r to 1. Consequently, the expression in which Eo and µo appear have
different values when stated in terms of the rationalized and the unrationalized
units, but the two forms of the mks system are otherwise the same. The
values of these constants for free space are given in the following table for all
the unit systems considered.

PERMITTIVITY AND PERMEABILITY FOR VACUUM

{- nit .-lystem Eo µo

{ r,mfoMli,OO _1_ } oool=b/(volt·m) 10 ., } w,OO,j(omp·fam·m:

9 X 10 9 or or
mks coulomb•/ (newton· m 2) newton/amp'
Rationalized
1 or or
4 " X 9 X 10' farad/m 4" 10-, henry/m

1 -- 1- sec'/cm2
{ Rl,d.,oo<o<i,
cgs 9 X 10 20
Electromagnetic - -1- sec 2/cm2 1
9 x 1020

The International Committee, in adopting the mks system, did not indicate
a preference between the rationalized and unrationalized forms. A survey of
the more advanced texts in electrophysics published since the adoption show
a greater use of the rationalized form. Only the rationalized form is used
in the body of this book.
Some of the commonly used units of the various systems are given in the
following table, and the relations among them are indicated by the con-
version factors listed in the last column. Electromagnetic units for the elec-
trical quantities are given with the prefix ab (meaning absolute) and the name
of the practical unit. Electrostatic units are indicated in a similar way by the
prefix stat-for example, statamp for the current unit.
By way of illustrating the use of the table, suppose that an iron core 0.25 m
long has a winding that provides 250 ampere· turns and sets up a flux density
through the core of 2 weber/m 2 • In rationalized mks units the mmf is 250
amp· turns, the field intensity is 250/0.25 = 1000 amp· turns/m, the perme-
ability is (2 webers/m2 )/(1000 amp·turns/m) = 0.002 weber/(amp·turn·m),
and the permeability relative to free space is 0.002/ (4 1r 10-1) = · 1590. In the
electromagneti<' <'gs units the mmf is ( ~;) X 2:)0 = 100 1r gilberts, the field

intensity is ( 1~;0 ) 1000 = -1 1r oersteds, the flux density is (10 4) X 2 = 20,000

. . . 20,000 gausses _ h f . th
gausses, an d t h e permea b111ty is 1 t d = 1b 90 ; t e actors m . e
-:1: 1r oers e s

large parentheses are taken from the figures in the lower right corner of the
table. The final result in either system is 1590 for the (relative) permeability
of the iron core under the conditions stated.
722
 I\Iks Rationalized
Entity Symbol or Unrationalize<l Practical Cgs Conversion Factons

Length ............... l meter ........ cm 1 meter - 100 cm

M&1111 ......•......•.. m kg ........ gm 1 kg - 1000 sm
Time ... . . . . . . .. . . . . . I aec ........ aec
Force ................ F newton ........ dyne 1 newton - 1()1 dynee

-
Work, energy ......... W,E joule joule erg 1 joule - 107 erp
Power ................ p watt watt !:!3 1 watt • 107 !!I!
800

Electrostatic Electromagnetic

Charge .............. . (J coulomb coulomb atatcoulomb abcoulomb 1 coulomb • 3 X 1()1 11tatcoulomb • 0.1 abcou-
lomb
Current ............. . I ampere ampere 11tatamp abampere 1 ampere - 3 X 10' 11tatamp • 0.1 abamp
1
Emf, potential. poten- E, V volt volt atatvolt abvolt 1 volt • 300 atatvolt - 1()1 abvolt.a
tial difference
Reaiatance ........... . R ohm ohm 11tatohm aboh:n 1 ohm - - -1- - atatohm - 1()1 aboh1n11
9 x 1011
Capacitance ......... . c farad farad atatfarad abfarad 1 farad - 9 X 1011 atatrarad - _!_ abfarad
1()1
1
Inductanee .......... . L henry henry atathenry abhenry 1 henry - - - - atathenry • 10' abhenrya
9 x 1011
Electric field intensity
volt newton ........ dyne 1 ~ _ 1 newton ___1_ dyne
6 - - ,w---
meter coulomb atatcoulomb meter coulomb 3 X 10' atatcoulomb
Magnetic flux ........ .
Magnetic flux density
•
B
weber
weber
. .......
........
ma,,well
gallilS
1 weber - 108 maxwell.a
1 weber _ lo< g a -
meterl meter4

Rationaliaed U nrationali-1

Magnetomotive force .. ~ ampere•turn ..L ampere•turn . .. . . ... ................ gilbert 1 rationalized mka unit - 4 • unrationalized mks
4 ..
units • 4 .. gilberts
10
Reluctanee ........... (R ampere·turn 1 ampere·turn ........ . ............... gill>crt 1 rationalised mks unit • 4 • unrationaliaed mks
weber 4 .. weber maxwell unite - ~ gilbert
10' maxwell
Magnetic field in ten- II ampere·turn 1 ampere·turn . . .. . . . . ................ oerste<l 1 rationaliaed mb unit - 4 • unrationalized mka
eity meter 4r meter
unit.a - ..!...!... oerated
1000
 APPENDIX III
The following sections in the book may be omitted in the shorter colle.ge
courses:
ltf cchanics Eledrici.ty & Magnetism Optics
Chapter Sections Chapter Sections Chapter Sections
1 11 15 12 27 8, 11
2 9 16 8, 13 28 7, 13, 19
3 3 17 7, 12 29 11, 17, 18
4 14 18 6, 10, 13 30 7, 9
5 5, 9, 10 19 5, 8, 11 31 3, 4, 8, 1-1
6 13, 14 20 5,8 32 2, 13
7 u, 15 21 3, 5, 10 33 7, 10
8 9, 12 22 8, 11, 14
9 8 23 3, -1, 8, 11
10 -1, 7, 10 2-1 5, 9
Heat 17ave Motion & Sound Modern Phys·ic~~
Cha pt.er Sections Chapter Sections Chapter Sections
11 10, 11 25 5, 12, 13 34 7
12 10; 13, 15 26 9, 10, 11, 12 35 5
13 3,8, 9
14 6, 7, 12

724
 ANSWERS TO ODD-NUMBERED PROBLEMS
MECHANICS

Chapter 1-1. 640 acres. 3. 75.8 ft. 5. 315 mi; 18.3° west of south. 7. 5.00 lb;
60.0° down from vertical toward right. 9. P, 2660 lb; R, 910 lb. 11. N, 1157 ft;
W, 1379 ft. 13. 4.66 lb; horizontal to right. 15. 15.64 mi; 32.7° south of east .
. 17. 7.66 lb·ft. ·
Chapter 2-1. 80.0 lb in a; 60.0 lb in b. 3. 32,&l.5°; 13.04 lb. 5. 848 lb. 7. 0.667 lb.
9. 9.43 lb; 58.0° down from horizontal toward right; 8.25 in. from D. 11. 22.0 lb
vertically downward; 1.636 ft from A and 1.364 ft from B. 13. 61.9 in. in front.
15. 11.55 lb. 17. Tie rod, 227 lb; pin reaction, horizontal 160.7 lb, vertical 311 lb.
19. AB, 1000 lb tension; AC, 1000 lb compression; AD, 1732 lb tension. 21. Top,
150.0 lb horizontally to left; bottom, 250 lb, 53.1 ° up from horizontal toward right.
23. A, 4000 lb; B, 5000 lb; C, 2000 lb.
Chapter 3-1. 2.50 min. 3. 9.1 ° from vertical in direction of flight. 5. 4.77 sec.
7. 4.17 hr. 9. (a) Derivation, (b) displacements: 80.0 ft, 144.0 ft, 80.0 ft, all
upward; velocities: 64.0 ft/sec upward, 0, 64.0 ft/sec downward. 11. 2.50 ft/sec2.
13. 4.10 ft. 15. 16.00 ft; 32.0 ft,/sec. 17. 34.7 ft/sec upward. 19. 1120 ft.
21. (a) 3900 ft, (b) 2.70 X 104 ft. 23. 917 ft/sec, 19.1° down from horizontal. 25.
1.152 mi. 27. 8.17, 4.71, and 3.49 radians/sec. 29. (a) 1.745 X 10-3 radian/sec,
(b) 8.73 X 10-4 in./sec.
Chapter 4-1. 15.00 lb. 3. 200 gm. 5. 4.05 m. 7. (a) 0.500 slug, (b) 2.54 slugs/ft3.
9. 6.53 m/sec. 11. 309 lb. 13. 4.57 ft/sec 2 up along plane. 15. 285 newtons.
17. 1.117 sec. 19. 92.4 m/sec. 21. 8.54 rev/min. 23. 8.00 ft/sec.
Chapteru-5.97 X 105 gm·cm 2. 3. -0.785newton·m. 5. 14.14lb·ft. 7. (a) Der-
ivation, (b) 3000 gm·cm2. 9. Derivation. 11. 0.0100 slug·ft2. 13. 107 gm·cm2.
15. 20.8 in. 17. 0.240 m. 19. 7740 gm· cm2. 21. 4.82 ft/sec. 23. 21.4 rev/min;
explanation.
Chapter 6-1. 2700 ft· lb. 3. 2.7 4 X 107 ergs. 5. 400 ft· lb. 7. 1579 ft· lb. 9.
1.786 X 104 ft·lb. 11. 20.0 ft/sec. 13. (a) 173.2 ft-lb, (b) 90.0 ft·lb, (c) 83.2
ft·lb. 15. (a) 9.28 ft, (l>) yes; 10.46 ft/sec 2 • 17. 0.255. 19. 0.0718 hp. 21. (a)
2370 hp, (b) 384 hp. 23. (a) 8.00, (b) 5.00, (c) 62.5 per cent.
Chapter 7-1. Yes; 0.01623 in. 3. 2.51 X 105 lb/in. 2 5. 1.019 X 10-3 in. 7. (a)
10.00 in. 2, (b) 0.0896 in. 9. 0.0265°. 11. (a) 140.0 cm/sec toward block, (b) 0,
(c) 140.0 cm/sec away from ball. 13. 37.7 cm/sec. 15. 625 gm. 17. 1134 lb/in.
19. 2.98 vib/sec. 21. (a) 1.273 vib/sec, (b) 2.21 ft/sec. 23. 1.582 slug·ft2.
Chapter 8-1. 8.56. 3. 33.9 ft. 5. (a) 624 lb/ft2, (b) 811 lb/ft2. 7. (a) 250 lb/ft2,
(b) 4000 lb. 9. (a) 2550 lb, (b) same. 11. 3.27 ft. 13. 66.2 lb. 15. 2270 lb·ft.
17. (a) 4.49 X 104 lb, (b) 1.796 X 105 lb·ft. 19. 625 lb; 16.00. 21. (a) 8.20
gm/cm 3, (b) 0.793 gm/cm3 • 23. 58.5 per cent. 25. 1.600; 3920 dynes/cm2. 27.
0.800. 29. Derivation. 31. 2.04 cm. 33. 61.3 dynes/cm.
Chapter .9-1. 3.92 X 104 cm/sec. 3. 33.9 ft. 5. 52.9 tons. 7. 0.0517 cm 3• 9. 8.03
gal. 11. 0.707 ft3. 13. 63.3 lb/in. 2 15. 6.74 cm. 17. 1.372 X 104 dynes/cm2 •
19. 68.8 lb/in. 2 21. 0.614 lb. 23. 7.97 ft 3 • 25. 7840 dynes. 27. 3.54 X 109 •
Chapter 10-1. 936ft·lb. 3. 3.97 X 106 ft·lb. 5. 4.76 X l06 ft·lb. 7. 12.65ft/sec.
9. 22.5 ft/sec. 11. (a) 120.0 ft3/sec, (b) 1.123 X 105 ft·lb, (c) 204 hp. 13. 65.7
min. 15. Derivation. 17. 2.25 ft/sec. 19. 9.18 ft3/sec. 21. 9.57 X 10-3 poise.
23. 2.01 X 10-11 dyne.
725
 HEAT

Chapter 11-1. 20.0°C; 293°K; 528°R. 3. -40.0°. 5. 3.993 in. 7. 20.6 lb. 9.
90.023 cm. 11. 1.2512 m; 8.015 cm 2 ; 1002.9 cm3• 13. 440°C. 15. 9.38 cma.
17. 1.221. 19. 1.501 X 104 Btu. 21. Derivation. 23. 1844 Btu/hr.
Chapter 12-1. 4780 cal. 3. 70.0°C. 5. 23.1°C. 7. 0.222 cal/(gm·C 0 ) . 9. 2320
Btu. 11. 1100 Btu. 13. 51 per cent. 15. 7.53 X 10 5 cal. 17. 3140 Btu/hr.
19. 17.77°C. 21. 0°C with 6.50 gm of ice unmelted. 23. 197.7 gal.
Chapter 13-1. 78.6°C. 3. 61.2 ft3 • 5. 476°F. 7. 3.44 ft3/rev. 9. 2.97 X 106
ergs/(gm·°K). 11 . .5.76 X 10-a slug. 13. 387 Btu. 15. 12.35 ft 3 ; (a) 6.18 ft3,
(b) 7.53 ft 3 • 17. (a) 25.1 atmospheres, (b) 7:36°K. 19. Diagram; 3.81 atmospheres.
21. Derivation.
Chapter 14-1. 503 ft. 3. 1.157 Btu. 5. (a) 20.3 joules, (b) 17.08 cal. 7. 5.03 X 10 1
ft·lb. 9. (a) 1837 joules, (b) reduced 1000 joules. 11. 25.7 per cent. 13. ] 113
joules; 266 cal. 15. 661 joules. 17. (a) 118.3°C, (b) 174.1°C. 19. -0.1314
Btu/0 R. 21. 2160 cal/ 0 K. 23. 6080 ft·lb. 25. 26.0 hp. 27. (a) 200 Btu/min,
(b) 4.72 hp.

ELECTRICITY & MAGNETISM

Chapter 15-1. 2; 6.69 X 10-24 gm. 3. 0.1000 newton repulsion. 5. 5.63 newtons
parallel to AB and directed from A to B. 7. 7.49 X 10-9 coulomb. 9. 9.80 X 10-9
coulomb. 11. (a) 9.00 X 105 newtons/coulomb directed away from positive charge,
(b) 4.61 X 10 5 newtons/coulomb similarly directed along line parallel to line of
charges. 13. 19.4 cm from smaller charge. 15. 9.00 X 104 volts. 17. 1.808 X
10-9 coulomb. 19. 8.39 X 106 m/sec. 21. 5. 23. 2.19 X 106 m/sec; 6.59 X 1015
rev/sec. 25. 0.407 cm.
Chapter 16-1. 2.80 newtons. 3. 786 amp. 5. 6.25 X 106 • 7. (a) 3040, (b) 2720.
9. 28.5 gm; 3.12 X 1019 • 11. 1.156 X 10-4 cm. 13. -0.22 amp. 15. 2.70 amp.
17. 244 gm. 19. 4.46 gm.
Chapter 17-1. 1.500 and 30.0 amp respectively; 160.0 volts. 3. 5.71 volts; 2.38
ohms. 5. 342 microhms; 0.0477 ohm. 7. 147.1 ft. 9. 0.1053 ohm. 11. 88.1 per
cent; $2.71. 13. 83.5 watts. 15. 1.212 volts; 0.268 ohm. 17. (a) 0.500 amp,
(b) 4.50 watts. 19. a = 3.00, b = 5.00 and c = 7.50 ohms. 21. In 2.1-volt cell
7.00 amp, in other 5.00 amp. 23. 11 = 0.827 amp; 12 = 0.252 amp; load current
is 0.575 amp. 25. 119.11 volts; 3.98 volts. 27. In E1 0.0966 amp; in E 2 0.0207
amp. 29. 11 = 2.86 amp downward; 12 = 1.429 amp downward; Ia = 4.29 amp
upward.
Chapter 18-1. (a) 9.43 X 10-4 weber/m2, (b) 4.82 X 10-4 weber/m2 • 3. (a) 5.66 X
10-5 weber/m2, (b) ] .257 X 10-5 weber/m2• 5. 1.639 X 10-4 weber/m2 ; 0.1250
newton. 7. 40.0 newtons up or down depending on current direction. 9. 24.0
divisions. 11. R 1 = 8.02 X 10-a ohm; R2 = 2.00 X 10-3 ohm. '13. 11,000 ohms.
15. Derivation. 17. Protons 3.05 X 107 per sec; helium nuclei 1.525 X 107 per sec.
19. 0.918 weber/m2 •
Chapter 19-1. (a) 1.047 amp, (b) 4.26 X 10-3 weber. 3. 319 watts. 5. 5.65 X 10 1
amp· turns/weber. 7. 0.359 weber/m 2• 9. 1.400 X 10-5 newton. 11. 0.1299 amp.
13. 6.99 amp; counterclockwise as viewed from above.
Chapter 20-1. 75.0 millholts; counterclockwise as viewed from above. 3. 1.421
webers/m2 • 5. 2.09 X 10-5 weber/m 2 • 7. 0.417 weber/m2 • 9. 100.0 volts. 11.
Armature 42.9 amp, field 1.200 amp; 123.86 volts. 13. (a) 24.0 amp, (b) 0.800 amp,
(c) 24.8 amp. 15. 7.33 amp.
Chapter 21-1. 5.20 millihenrys. 3. (a) 7.50 volts, (b) 3.00 volts. 5. 17.28 henrys;
4.32 X 104 volts. 7. 1.920 henrys; 1850. 9. (a) 80.0 amp/sec, (b) 53.3 amp/sec.
726
 11. (a) 12.17 joules, (b) H\2.2 joules. 13. 4650 joules/m3 • 15. 2.96 µf. 17. Small
sphere 0.0600 microcoulomb, other 0.1200 microcoulomb. 19. 694 µµf. 21. 2.31.
23. (a) 0.424 microcoulomb, (b) 8.48 X 10-4 joule, (c) 2 ..53 microcoulombs, (d)
5.94 X 10-3 joule. 25. On the 2-, 4-, and 6-µf condensers respectively the charges
are 160, 640, and 480 microcoulombs, and the potential differences are 80, 160, and
80 volts.
Chapter 22-1. 125.0 amp. 3. ±169.7 volts. 5. 49.1 ohms. 7. (a) 0 volt, (b) 163.2
volts. 9. (a) 56.5 ohms, (b) 58.5 ohms, (c) 3.76 amp, (d) 75.1°. 11. 132.6 ohms;
0.905 amp. 13. 2.40 amp. 15. 30.1 ohms. 17. From 50.7 to 1266 µµf. 19.
106.1 µf. 21. 125.1 watts. 23. 92.2 and 61.5 per cent respectively. 25. 4.57 kw;
current lags 36.9°. 27. 2510 amp; 58,000 kw. 29. (a) 230 watts, (b) 110.8 watts.
Chapter 23-1. 8.40 millivolts. 3. (a) 1.02 millivolts, (b) 3.50 millivolts. 5. 44.4°C.
7. 0.567. 9. 4.39. 11. 21.5. 13. 12.3 milliamp. 15. 31.6 to 1. 17. 15.0°. 19.
Derivation. 21: With 20 volts at collector a = 2.4.
Chapter 24-1. UHF 63.8 to 33.7 cm; VHF 5.56 to 1.39 m. 3. 576 µµf. 5. 50.0 and
150.0 respectivel:v. 7. 0.0174 in. 9. 4.13 megacycles/sec. 11. (a) 70.0, (b) 3000.

WAVE MOTION & SOUND

Chapter 2u-1. 4.17 ft. 3. 2.78 X 10-4 sec. 5. 0.471 m. 7. First and second:
5 vib/sec, 160 ft/sec, 32 ft; third: 5 vib/sec, 80 ft/sec, 16 ft. 9. 3830 ft. 11. 5.93
sec. 13. 0.0982 sec. 15. (a) 0.0666 sec, (b) 12,390 mi. 17. 2.00 X 105 km/sec.
19. 2.50 ft; 440 vib/sec. 21. 1.667 ft. 23. 117.6 gm.
Chapter 26-1. 880 rev/min. 3. (a) 1.100 X 10-5 microwatt/cm2, (b) 50.4 db.
5. (a) l/9 as large, (b) -9.54 db. 7. 180 and 360 vib/sec respectively. 9. 0.0299
in. 11. 2088 and 6264 vib/sec. 13. 24.13 in. 15. 2.75 per sec. 17. 8.33 in.
19. Rod 4980 m/sec, gas 330 m/sec. 21. 511.2 vib/sec. 23. 1.196 sec. 25. (a)
0.400 sec, (b) 0.800 sec. 27. 3000 vib/sec.

OPTICS

Chapter 27-1. 469 rev /sec. 3. (a) 8230 lumens, (b) 35.0 watts, (c) 235 lumens/watt.
5. 0.01736 lumen. 7. 573 candles. 9. 1.338 ft. 11. (a) ,5.00 ft from A, (b) 1.000
footcandle. 13. (a) 50.0 meter-candles, (b) 65.0 candles. 15. 30.4 footcandles.
17. 7.07 lumens. 19. 5.
Chapter 28-1. 3.00 ft; no. 3. 14.32°. 5. 2.50 cm. 7. 5.26 cm. 9. 66.1 °. 11.
1.925. 13. (a) 17.1°, (b) 23.6°. 15. (a) a= 1.5953; b = 5444, (b) a, no unit;
b, (mµ) 2 , (c) 1.6110. 17. 0.099°. 19. (a) 8.49°, (b) 0.1049°.
Chapter 29-1. 57.4 cm. 3. (a) 24.0 cm to left of lens, (b) virtual, erect, enlarged.
5. (a) 10.0 cm and -20.0 cm, (b) virtual, erect, enlarged, (c) diagram ... 7. (a)
-30.0 cm; -3.33, (b) +6.67 and -10.0; -3.33. 9. (a) 10.0 cm to left of'second
lens, (b) 0.667 in., (c) virtual, erect, diminished. 11. (a) 20.0 cm to right of second
lens, (b) 12.0 cm, (c) real, inverted, enlarged. 13. (a) 108.9 cm to the right, (b)
106.5 cm. 15. (a) 3.00 cm to right of mirror; 0.500 cm, (b) diagram, (c) virtual,
erect, diminished. 17. (a) 15.0 cm to left of mirror, (b) 0.75 cm, (c) real, inverted,
diminished. 19. (a) 3.81 cm to left of lens, (b) real, inverted, diminished. 21.
Crown: +20.53 cm and -36.23 cm; flint: -36.23 cm and oo.
Chapter 30-1. -1.64 diopters. 3. Diagram. 5. f/7.0. 7. (a) 3.27, (b) 2.27. 9.
O to 1.25 cm. 11. -14 (inverted image). 13. (a) 1.63 cm, (b) 88.2, (c) diagram.
15. (a) 140 cm and 20 cm, (b) -7. 17. (a) Derivation, (b) 16.0 mm and 28.0 cm.
19. (a) 3.75 mm, (b) 5.00 mm and 7.14 mm; the 7 X 50 glasses. 21. (a) 120 cm,
(b) diagram. 23. 13.53 cm; 14.0 cm.
Chapter 31--1. Continuous emission with absorption lines. 3. 656.54 mµ. 5. 16
men; 1 woman. 7. (a) White, (b) yellow, (c) yellow-green. 9. (a) Black, (b) blue,
727
 (c) reddish-purple. 11. Diagram; magenta; 500 c mµ, 70 per cent. 13. (a) Cal-
cium fluoride, lithium fluoride, (b) rock salt, (c) potassium bromide, thallium
bromide-iodide. 15. Sketch. 17. 2.61 mm. 19. Diagram.
Chapter 32-1. (a) 1.80 mm, (b) 2.80 mm. 3. 1.031 X 10-5 cm. 5. 6.25 X 10-5 cm.
7. (a) 64.4 per cent, (b) 86.0 per cent, (c) 94.2 per cent. 9. 679. 11. 0.960 cm.
13. (a) 0.360 cm, (b) 3rd, 6th, 9th, (c) graph. 15. 61.9 cm. 17. (a) 3 on each side,
(b) 722 mµ. 19. 6.83 X 10-3 cm. 21. 2.11 X 10-4 in. 23. (a) 0.75, (b) 1.11.
Chapter 33-1. 69.0°. 3. 1.37. 5. (a) 1.0, (b) oo, (c) 4.5. 7. Figures. 9. Figures.
11. (a) Dark, (b) dark. 13. 7.50 units. 15. (a) 0.304 per cent, (b) 0.403 per cent,
(c) -24.6 per cent. 17. If isotropic material were used, the undeviated beam
would be chromatic. 19. (a) 1.64 X 10-2 mm, (b) 8.57 X 10-4 mm. 21. (a) Cir-
cularly polarized, (b) linearly polarized at 90° to the polarization axis of the incident
beam.

]VloDERN PHYSICS

Chwpter 34-1. 5.12 watts. 3. 483 mµ. 5. 3.31 X 10-19 joule; 2.07 volts. 7. 4.14
volts. 9. 1240; 0.826 mµ. 11. 0.0527 mµ; 2.19 X 106 m/sec. 13. 10.20 and
13.60 electron·volts. 15. 18. 17. Graph. 19. 0.0513 mµ. 21. (a) 0.0048 mµ,
(b) 1.92 X 10-15 joule. 23. 9.93 X 10-15 joule; 1.105 X 10-31 kg; 3.31 X 10-23
kg·m/sec. 25. (a) 2.86 X 10-3 mµ, (b) 5.72 X Jo-4 mµ.
Chapter 35-1. 3.14 X 10-15 . 3. 14.67 X 10-31 kg. 5. 0.240. 7. 20.0 min. 9. 22.9
days. 11. Equations; 92Th 234 , 91 Pa 234 , 9oU234 . 13. 0.0196. 15. 0.459 weber/m2 •
17. (a) 0.390, (b) 16.06 X 10-31 kg. 19. H 2 (d, n)He3 ; Al2 7 (a, n)P 30 • 21. 2.14.
23. 3.35.

728
 INDEX
Abampere, relation to ampere, 282 Ammeter: alternating-current, 396;
Abbe, Ernst: dispersive power, 543; re- direct-current, 327; hot-wire, 403
solving power, 635 Ampere, Andre M.: force between con-
Abcoulomb, relati,on to coulomb, 282 ductors, 280; magnetic induction, 319
Aberrations, lens and mirror, 566 Ampere, the, 280
Absolute: pressure versus gage pressure, Amplification factor: of electron tube,
159; pressure within liquids, 138; zero 409; of transistor, 425
of temperature, 183 Amplifiers: audio- and radio-frequency,
Absorption: acoustic, 494; coefficient, 436; electron tube, 410
612; light, selective, 607 Amplitude: modulation, 433; of vibra-
Acceleration: angular, 52; constant, 42; tion, 127, 450
due to gravity, 45; force to produce, Anderson, Carl D., positron discovery,
62; in circular motion, 77; in har- 700
monic motion, 127; linear, 40; torque Andrews, Thomas, liquefaction of gases,
to produce, 82; units of, 41 231
Accelerators: cyclotron, 331; high- Aneroid barometer, 157
energy, 696; linear, 694 Angle: plane, units of, 4, 51; solid, defi-
Accommodation of the eye, 576 O nition of, 517
Achromatic: doublet lens, 570; prism, Angstrom, Anders J., wavelength unit,
544 542
Acoustics: of auditorium, 496; science Angstrom unit, the, 542
of, 470 Angular: acceleration, 52; measure-
Action and reaction, 62 ment, 4, 51; momentum, 91; motion,
Adhesion and cohesion, 146 50, 130; quantities, relation to linear,
Adiabatic process: meaning of, 229; 53; quantities, vector representation
work done in, 241 of, 92; velocity, 51
Advantage, mechanical, 111 Anode and cathode: of cells, 284; of
Air conditioning, 210 tubes, 406
Aircraft, supersonic flight, 498 Antenna, radiation from, 429
Airfoil, lift on, 175 Aperture: numerical, of lens, 636; of
Airplane: forces on, 61; jet propulsion, mirror, 563; relative, 579
252 Archimedes, principle of buoyancy, 142,
Airy, George B., disk of light, 633 144, 161
Alnico magnet steel, 341 Architectural acoustics, 496
Alpha rays, 683 Armature of generator, 355
Alternating current: character of, 381; Armstrong, Edwin H., frequency modu-
circuits, 384; effective value, 383; fre- lation, 437
quency, 380; generator, 393; ma- Aspirator, 174
chinery rating, 392; measurements, Astigmatism of: lens, 569; the eye, 578
396; power, 390; resonance, 389; root- Aston, Francis W., mass spectrograph,
mean-square value, 383 687
Alternating electromotive force-see Al- Astronomical telescope, 589
ternating current Atmospheric: buoyancy, 161; electric-
Alternator: elementary, 354; single- and ity, 273; humidity, 208; optics, 613;
polyphase, 393 pressure, 156; refraction, 539; scat-
Altimeter: pressure, 157; radio, 446 tering, 614
729
730 INDEX

Atomic: bomb, 709; mass unit, 687, 705; Bragg, Sir Lawrence, x-ray crystal dif-
number of element, 260, 686; pile, fraction, 675
708; weight of element, 260, 287 Bragg, Sir William H., x-ray crystal
Atomizer, 174 diffraction, 675
Atoms, structure of, 258 Brattain, Walter H., transistor, 424
Attraction: electrical, 262; gra vita- Bremsstrahlung, 700
tional, 64; magnetic, 280, 324 Brewster, Sir David, polarizing angle,
A.twood, George, machine, 69 641
Audibility range of ear, 487 Bridgman, Percy W., high pressures,
Audio amplification, 437 214
Auditorium, acoustics of, 496 Brightness: of color, 60.5; photometric,
A.vogadro, A. madeo: law of gases, UH; 513
number of molecules, 162 British: gravitational units, 63; thermal
Avogadro's Number, application of, 287 unit, the, 190
Brodhun, Eugen, photometer, 520
Brown, Robert, movement of particles,
Balance: equal-arm, 25; sensitivity of, 145
26; spring, 4, 117 Bulk modulus of elasticity, 119
Ballistic galvanometer, 327 Bunsen, Robert W., photometer, 520
Balmer, Johann J., hydrogen spectrum, Buoyancy of: atmosphere, 161; liquids,
666 142
Banking of curves, 79
Bardeen, John, transistor, 424 Cables, capacitance of, 372
Barometer: aneroid, 157; mercury, 156 Calorie, the, 190
Barrier: layer, 421; potential, 405 Calorimetry, 199, 215
Battery: primary, 290; storage, 292 Calutron, 708
Beats by interfering waveR, 465 Camera: photographic, 579; photomi-
Becquerel, A.ntoine H., natural radio- crographic, 587
activity, 682 Candle, the, 517
Bell, Alexander G.: amplifier gain, 411; Capacitance of: circuits, 387; condenser,
sound intensity levels, 473; tele- 372; isolated conductor, 370 .
phony, 432 Capacitors: connection of, 375; energy
Bernoulli, Daniel, fluid flow, 170, 174 stored in, 374; parallel-plate, 372
Beta rays, 683 Capillarity, 149
Betatron, 697 Carbon arc lamp, 504
Bevatron, 697 Cardinal points of thick lens, 561
Bifocal spectacles, 578 Carnot, Sadi, ideal engine and cycle, 242
Binaural sense, 487 Cascade rectification, 693
Binding energy: of nucleus, 706; per Cassegrain, N., reflecting telescope, 591
nucleon, 706 Cathode-ray oscilloscope, 417
Binocular, prism, 593 Caustic, in spherical aberration, 567
Biot, Jean B., electromagnetism, 323 Cavitation and coagu'Iation, 498
Biprism, 620 Cell: electrolytic, 283; local action in,
Birefringence, 643 291; photo, 515; polarization in, 291;
Black body, meaning of, 660 primary, 291; resistance of, 306; series
Blind spot of retina, 576 and parallel connection of, 308;
Bohr, Niels, hydrogen spectrum, 667 standard of emf, 294; storage, 292;
Boiling: point, 211; process of, 205, 210 voltaic, 284, 290
Bolometer, 516 Celsius, A.nders, temperature scale, 183
Boltzmann, Ludw1:g, radiation law, 660 Center of: curvature of mirror, 563;
Bomb: atomic, 709; fusion, 709 gravity, 23; pressure, 140
Bourdon, Louis, gage, 159 Centigrade temperature scale, 183
Boyle, Robert, law of gases, 1.58, 222, 229 Centrifugal: force, 76; pump, 165
 INDEX 731

Centripetal force, 76 Components: in study of equilibrium,

Cgs unit system, 63; relation to other 18; resolution of vectors into, 9
systems, 723 Compressibility of substances, 120
Chadwick, James, neutron, 701 Compton, Arthur H., x-rayscattering, 677
Chain reaction in nuclear fission, 707 Condensations and rarefactions, 450
Charge: action in electric field, 265; ac- Condenser: microscope, 587; projector,
tion of magnetic field on, 329; distri- 594
bution of, on conductors, 272; force Condensers, electric-see Capacitors
on, by another, 258; measurement of Conduction: deficit, 421; electronic,
electronic, 27 4; production of, 260; 270; heat, 190, 192; ionic, 271, 284
space, in electron tube, 406; to mass Conductivity: electric, 297; super-, 303;
ratio of electron, 329; unit of, 263; thermal, 192
work done in moving, 269 Conductors and insulators, 261
Charles, Jacques A. C., law of gases, Cones, retinal, 510, .576
222 Conjugate distances for lens, 548
Chemical equivalent of element, 287 Conservation of: angular momentum,
Christie, S. Hunter, bridge circuit, 308 91; energy, 102; linear momentum,
Chromatic aberration, 570 75; matter plus energy, 684
Chromaticity, 605, 609 Constants, physical, 714
Circuit: alternating-current, 385; cal- Contact rectifiers, 422
culating wire size of, 308; capacitive, Continuity, equation of, 170
37 5, 387; direct-current, 279; flux Convection of heat, 190
density near, 320; forms, 305, 311; Converging lenses, 550
impedance of, 388; inductive, 363, Conversion factors for: a.ngles, 51; force,
385; magnetic, 341; power factor of, 63; heat energy, 238; lengths, 4; mag-
390; quality factor, 431; radiation netic quantities, 343; mass, 60; me-
from, 428; reactance of, 387; time chanical energy, 99; speed, 35; tem-
constant of, 368 peratures, 183; unit systems, 713, 723
Circular: mil, the, 300; motion, 76 Conversion of units, method of, 13
Cloud chamber, 690, 699 Copper-oxide rectifier, 423
Coated lenses, 622 Corona discharge, 271
Cockroft, John D., artificial transmuta- Corrosion, electrolytic, 286
tion, 693, 705 Cosmic rays, 698
Coercive force, 339 Cosmotron, 696
Coherence of light beam:-, 620 Cottrell, Frederick G., smoke elimination,
Cohesion and adhesion, 146 272
Collins, Samuel C., cryostat, 234 Coulomb, Charles A., force between
Collision: elastic and inelastic, 123; mo- charges, 262
mentum conservation in, 74 Coulomb, the, 263, 282
Color: blindness, 605; by diffraction, Counter: electromotive force, 3.59;
630; complementary, 606, 609; filter, Geiger, 689
612; metallic surface, 607; mixture, Couple, definition of, 13
606; specification, 608; triangle, 606; Crane., traveling, 37
vision, anomalous, 605; ,vith polar- Critical: angle of reflection, 533; entities
ized light, 650 of gases, 232
Colorimetry, 604 Cryostat, 234
Coma, 567 Crystal: counter, 692; dichroic, 646;
Combination tones, 481 frequency-control, 434; ordinary and
Combustion, heat of, 216 extraordinary rays in, 643; principal
Commutator, purpose of, 354, 358 section of, 643; uniaxial and biaxial,
Compass: gyro, 94; magnetic, 347; po- 645~ x-ray analysis of, 674
larization sky, 655 Curie, Pierre and Marie, natural radio-
Complementary colors, 606, 609 activity, 682
732 INDEX

Curie-Joliot, Irene, artificial radioac- Dioptometer, 561

tivity, 703 Dioptric power, 560
Current: alternating, 380; balance, 281; Dip, magnetic, 347
decay in inductive circuit, 367; direct, Direct-vision prism, 545
279; displacement, 372; effects pro- Discharge coefficients, orifice, 168
duced by, 279; growth in inductive Dispersion of light, 540
circuit, 367; heat produced by, 383; Dispersive power, 542
magnetism produced by, 318; meas- Displacement: current, 372; of position,
urement of, 325; saturation, of elec- 2; versus path length, 33; with har-
tron tu be, 406; versus electron flow, monic motion, 126
279 Dissociation, electrolytic, 284
Curvature of field, 569 Distortion by lens and mirror, 569
Cycle: Carnot heat, 242; irreversible Distribution circuits, calculation of, 308
and reversible, 244; of current and Diverging lenses, 550
emf values, 380; Rankine, 246 Domain theory of magnetism, 344
Cyclotron accelerator, 331, 695 Dominant wavelength of color, 605
Doppler, Christian, frequency change
Dalton, John, mixture of gases, 160, 208 in: sound, 492; stellar spectra, 601
D' Arsonval, Jacques A., galvanometer, Double: image prisms, 649; optical slit,
325 618; refraction, 643
Davisson, Clinton J., electron diffrac- Dry cell, 292
tion, 678 Dynamo-see Generator or Motor
Daylight visible range, 615 Dyne, the, 63
de Broglie, Louis, matter waves, 678
Decay constant, radioactive, 685 Ear, human, 485
Decibel: definition of, 411; scale of Earth: age of, determined from radio-
sound level, 473 activity, 689; magnetic field of, 346
Declination, magnetic, 346 Eclipses, solar and lunar, 508
De Fore st, Lee, grid electrode, 408 Eddy currents in generators, 357
Delta and wye connection, 311, 394 Edison, Thomas A.: electric lamp, 505;
Density: definition of, 60; energy, 369, sound recording, 487; storage cell,
375, 472; measurement of, 143; of 294; thermionic emission, 405
electric charge, 272; of electric lines Efficiency, luminous, 504, 513
of force, 267; of gases, 160; of liquids Efficiency of: ideal engine, 244; ma-
and solids, 135; of magnetic flux, 318; chines, 110; reciprocating engine,
optical, 613; thermionic current, 406; 250; transformer, 396
weight, 136 Efflux of liquid, speed of, 167
Depth soundings, 498 Einstein, Albert: mass-energy equiva-
Deuterium spectrum, 670 lence, 684; photoelectric effect, 664;
Dew point, 210 relativity theory, 683
Diamagnetic substances, 337 Elastic limit and yield point, 118
Dichroic crystals, 646 Elasticity: concept of, 116; moduli of,
Dielectric constant, 373 118
Dielectrics, 261, 370 Electric charge-see Charge
Diesel, Rudolph, engine, 227, 250 Electric circuit-see Circuit
Differentiation, concept of, 39 Electric current-see Current
Diffraction grating, 630; resolving Electric field: between charged plates,
power of, 636 270; intensity of, 264; lines of force
Diffraction of: electrons, 679; light, 626, in7 266
632; x-rays, 674 Electric: filter, 431; lamps, 505; light-
Diffusion of: gases, 156; light, 527 ing, 522; networks, 311; organ, 484;
Diode electron tube, 405, 411 potential, 268
Diopter, the, 560 Electric power-see Power
 INDEX 733

Electric vector in polarized light, 640 moving liquid, 170; potential, 104;
Electrical energy-see Energy transformation of, 102; transmission
Electrical resistance-see Resistance by waves, 454
Electrodynamics, 280 Engine: Diesel, 227, 250; efficiency and
Electrodynamometer, 281 horsepower, 250; ideal, 242; internal
Electrolysis, 283 combustion, 249; jet, 252; recipro-
Electrolytic: cells, 283; corrosion, 286 cating steam, 248; uniflow, 249
Electromagnet, 317, 344 Entropy, 246
Electromagnetic: induction, 351; radi- Equilibrant of forces, 22
ation, 428; waves, 459 Equilibrium: conditions of, 17, 19; posi-
Electromagnetisi:µ, 317 tion in harmonic motion, 126; theo-
Electromotive force: concept of, 283; rem of, 20; types of, 25
counter, 359; generation of, 353; in- Equipotential surface, 270
duced, 351; measurement of, 310; Erg, the, 99
of cells, 292; of self-induction, 365, Evaporation, process of, 205
385; of thermocouple, 402 Exclusion principle, 671
Electromotive series of metals, 290 Expansion: adiabatic, 229, 241; at con-
Electron: charge of, 259, 274; charge to stant pressure, 220, 241; at constant
mass ratio of, 329; diffraction, 679; temperature, 222; constrained and
emission, thermionic, 405; flow versus free, 225; differential, 185; isother-
current, 279; gun, 415, 588; lens, 416; mal, 229, 240; of liquids, 187; of
mass of, 329; microscope, 588, 679; mercury, 188; of solids, 184; of water,
optics, 416; refraction, 416; shells in 188
atoms, 259, 672; spin, 670; synchro- Eye: accommodation of, 576; as light
tron, 698 receptor, 510; constants of the, 57 5;
Electron tube: amplifiers, 410; con- dark-adapted, 511; relative sensitiv-
stants of, 409; diode, 405; gas-filled, ity of, 511; relief, 583; schematic, 575
411; multigrid, 414; oscillator, 413; Eyepiece, Huygens and Ramsden, 584
oscilloscope, 417; pentode, 415; tet-
rode, 414; triode, 408
Electron· volt, the, 331 Fabry, Charles, interferometer, 625
Electronics, 401 Fahrenheit, Gabriel D., temperature
Electro-optic effect, 652 scale, 183
Electroplating, 285 Falling bodies, 45
Electroscope, 258, 682 Farad, the, 370
Electrostatic: attraction, 258; electron Faraday, Michael: capacitance unit,
lens, 416; generator, 273, 694; induc- 370; electromagnetic induction, 351;
tion, 261 ice-pail experiment, 272; laws of elec-
Electrostatics, 257 trolysis, 286; magneto-optic effect,
Elements, periodic table of, 720 655
Emission: spectral, 660; thermionic, 405 Faraday, the, 288
Emmetropic eye, 577 Far-sightedness, 578
Emulsions, photographic, 515 Fathometer, 498
Energy: binding, of nucleus, 706; con- Fechner, Gustav T., stimulus change, 472
cept of, 101; conservation of, 102; Fermi, Enrico: neutron experiments,
density of electric field, 37 5; density 702; nuclear reactor, 708; transuranic
of magnetic field, 369; density of elements, 709
waves, 472; electrical, 303; heat, me- Ferromagnetic substances, 337
chanical equivalent of, 238; internal, Field: around antenna, 429; curvature
of gas, 227; kinetic, 103; level in of optical, 569; structure of generator,
atoms, 669; method of solving prob- 355; within resonant cavity, 443
lems, 106; nuclear, 705; of electric Fields-see Electric field or Magnetic
field, 374; of magnetic field, 369; of field
734 INDEX

Film: light interference with, 621; re- rent, 380; of circuit, natural, 389; of
flection-reducing, 623 harmonic motion, 126, 450; of sound-
Filter: electric, types of, 431; optical, ing bodies, 475; threshold, in photo-
626 electric effect, 664
Fission, nuclear, 707 Fresnel, Augustin: bi prism, 620; diffrac-
Fizeau, Augustin H. L., speed of light, tion, 626; reflection losses, 535, 623
.509 Friction: coefficient of, 71; fluid, 73;
Flight: of projectile, 48; of spinning force of, 4, 70; work done against, 105
ball, 174; supersonic, 498 Fringes, interference, 618
Floating bodies, stability of, 144 Frisch, Otto R., nuclear fission, 707
Fluids-see Gases and Liquids Fuses, 301
Fluorescent lamps, 506 Fusion: bomb, 709; heat of, 203; nu-
Fluoroscope screen, 420 clear, 707; process of, 201
Flux: density in magnetic materials,
336; dielectric, 370; linkages, 366; Gages, pressure, 158
magnetic, 318; ndiant, 513 Galileo, Galilei: falling bodies, 45; speed
!-number of lens, 579 of light, 508; telescope, .593
Focal: length of lens, 549, 562; length Galvanometer, 325
of mirror, 564; lines, astigmatic, 569; Gamma radiation, 683
points of thick lens, 561; ratio of lens, Gas: actual, 234; constant, 222; general
579 law of, 219; ideal, 158, 22G; ther-
Foley, Arthur L., acoustic wave fronts, mometer, 224; turbine, 252
497 Gases: density of, 160; diffusion of, 156;
Footcandle, the, 518 electric conduction in, 271; expansion
Foot·pound, the, 99 of, 220; ionization of, 271; kinetic
Force: acceleration by, 62; and trans- theory of, 154; liquefaction of, 231;
lation, 58; between current-carrying specific heats of, 227; speed of sound
conductors, 280; between electric in, 457; viscosity of, 177
charges, 258, 262; between molecules, Gauss, Karl F., flux density, 318
145; centripetal, 76; coercive, in mag- Gauss, the, relation to the weber per
netism, 339; definition of, 4; electro- square meter, 318
motive, in circuit, 283; exerted by Geiger, Hans, pulse ionization cham-
liquid, 139; impulsive, 73; magneto- bers, 689, 699
motive, 342; moment of, 11; of fric- Generator: alternating-current, 393;
tion, 4, 70; of gravity, 4, 65; on direct-current, 355; electrostatic, 273,
conductor in magnetic field, 324; res- 694
olution of, 10; unbalanced, 61; units Geometric relations, 714
of, 4, 63 Germanium rectifier, 424
Forces: on airplane, 61; vector addition Germer, Lester H., electron diffraction,
of, 5 678
Foucault, Jean B. L., speed of light, 509 Gilbert, the, relation to ampere· turn,
Fourier, Jean B . .!., wave components, 343
475 Gilbert, William, magnetomotive force,
Fps unit system, 63 343
Franklin, Benjamin, names of charges, Goggles, dark-adaptor, 512
258 Goudsmit, Samuel A., electron spin, 670
Fraunhofer, Joseph: dark spectrum Gratings, diffraction, 630
lines, 542, 601; diffraction, 627 Gravitation, universal, law of, 64
Freezing: effect of pressure on, 204; Gravitational: constant, 65; system of
process of, 201 units, 63
Frequency: change due to relative mo- Gravity: acceleration due to, 45; center
tion, 492; control by crystals, 434; of, 23; force of, 4, 65; specific, 135
modulation, 437; of alternating cur- Greek alphabet, 715
 INDEX 735

Guided missile, 446 Hydraulic brakes, 142; press, 142

Gun, electron, 415 Hydrogen: atom, dimensions of, 260;
Gyration, radius of, 88 electrode, 289; spectrum, 666
Gyroscope, the, 94 Hydrometer, 144
Hydrophone, 497
Hahn,. Otto, nuclear fission, 707 Hypermetropic eye, 578
Half-life, 685 Hysteresis, magnetic, 340
Half-wave plate, 658
Hallwachs, Wilhelm, photoelectric ef-
Ice, manufacture of. 252
fect, 662
Hammond, Laurens, electric organ, 484 Iceland spar, polari~ation with, 643
Hansen, William lV., linear accelerator Iconoscope for television 439
.
Igmtron tube, 412 '
695 '
Ha:monic: analysis, 474; angular mo- Illumination, 518; of interiors, 522; rec-
ommended levels of, 522
t~on, 130; linear motion, 124; vibra-
tions, 476 Illuminometer, 521
Image: in plane mirrors 527 · location
Hartley, Ralph V. L., oscillator, 413, 434
Head, elevation, pressure, and velocity 554; production, 548, '565 ' '
171 ' Impedance of circuit, 388
Impulse and momentum 73 91
Head.lamps, polarized, 656
Hearmg: organ of, 485; range of, 488 Incandescent lamps, 505 ' '
He~t: conduction of, 190, 192; convec- Inclined plane, mechanical advantage
t10n of, 190; from electric current, of, 112
304, 383; mechanical equivalent of; Indeterminacy principle, 680
238; nature of, 181; of combustion, Indicator, steam-engine, 249
216; of fusion, 202; of vaporization Inductance: mutual, 363; of solenoid,
212; Peltier effect, 404; radiation of'. 366; self-, 365
191; specific, 198; to raise tempera- Induction: charging by, 261; coil, 360;
ture, 198; transfer, 190; transforma- electromagnetic, 351; lines of, 318;
tion into work, 239; transmission motor, 394; mutual, 363; self-, 365
affected by surface films, 195· units of Inertia: concept of, 58; moment of, 83
190 ' ' Insulators and conductors 261
Heisenberg, Werner, indeterminacy prin- Integration, concept of, 39'
ciple, 680 Intensity: levels of sound, 472; lumi-
Helmholtz, Herman L. F.: schematic nous, 517
Interference: and diffraction combined
eye, 57,5; tone quality, 474
629; filters, 626; in thin film 620· of
Henry, Joseph, inductance, 364
Henry, the, 364, 36,5 light, 617; of polarized light: 650; of
Herapath, William B., synthetic polar- sound, 496; of waves, 464
izer, 647 Interferometer: Fabry-Perot, 625; Mi-
Hertz, Heinrich R.: electromagnetic ra- chelson, 624
diation, 428; photoelectric effect, 662 Internal: combustion engine, 249; en-
Ileusler, Conrad, magnetic alloys, 338 ergy of gas, 227
Hewitt, Peter Cooper, mercu~y-vapor Inverse square law, 462, 519
lamp, 505 Ion, meaning of, 271
High-energy machines for nuclear re- Ionization: in solutions , 284·, of gases,
search, 694 271; potential, 669
Hooke, Robert, law of elasticity, 116 Isothermal process, 229, 240
Horsepower, the, 108 Isotopes of element, 260, 686
Hue of color, 605
Humidity, atmospheric, 208 Jet engine, 252
Hu¥gens, Christian: eyepiece, 583; prin- J oliot, Frederic, artificial radioactivity,
ciple of wave propagation, 460 703
736 INDEX

Joule, James P.: gas expansion, 225; Linear: accelerator, 694; motion, 33, 40
law of electric heating, 304; mechani- Lines of: force, electric, 266; magnetic
cal equivalent of heat, 238 induction, 318
Joule, the, 99 Liquefaction of gases, 231, 252
Liquid: discharge rate of, 168; energy of
K-capture, 704 flowing, 170; force and torque due to,
Kelvin, Lord (William Thomson): tem- 139; ideal, 135; pressure, 137; speed
perature scale, 183, 245; thermoelec- of flowing, 167
tric effect, 404 Liquids: at rest, 135; electric conduc-
Kenotron rectifier, 407 tion in, 271; expansion of, 187; speed
Kerr, John, electro-optic effect, 652 of sound in, 457; surface phenomena
Kerst, Donald W., betatron, 697 of, 146; viscosity of, 175
Kilogram calorie, the, 217 Lissajous, Jules A.: sound speed, 491;
Kinescope for television, 440 vibration figures, 489, 653
Kinetic: energy, 103, 105; relativistic Livingston, M. Stanley, cyclotron, 695
energy, 684; theory of gases, 154 Lloyd, Humphrey, mirror, 620
Kirchhoff, Gustav R.: electric networks, Load, electrical, 299
312; radiation law, 660 Logarithms to base 10, 718
Kundt, August A. E., sound speed, 491 Longitudinal and transverse waves, 450
Loudness: contours, 486; of sound, 470,
Lamps, types of, 505 490
Land, Edwin JI., sheet polarizers, 646 Lumen, the, 513, 517
Laue, Max von, x-ray diffraction, 673 Luminous: efficiency, 513; flux, 605; in-
Lawrence, Ernest 0.: cyclotron, 331, 695; tensity, 517; -tube lamps, 506
linear accelerator, 694 Lummer, Otto, photometer, ,520
Length, units of, 3 Lyman, Theodore, hydrogen spectrum,
Lens: achromatic doublet, 570; action 667
of, 548; anastigmat, 580; camera,
580; combinations, 558; condenser for
microscope, 587; cylindrical, 578; Mach, Ernst, speed ratio, 499
electron, 416; equation, 554; eye, 575; Machines, simple, 110
focus of, 549; magnetic, 588; mag- Magnetic: attraction, 317, 341; circuitt3,
nification, 557; measure, 561; micro- 341; declination, 346; effect of cur-
scope, 576; oil immersion, 586; prin- rent, 318; field, action of bar magnet
cipal points and planes of, 561; in, 345; field around circuits, 320;
spectacle, .579; telescope, 590; thick, field, energy of, 369; field of earth,
561; thin, 552; types, 550 346; flux, concept of, 318; induction,
Lensmaker's Equation, 554 Ampere's generalization, 319; inten-
Lenz, JI. F. Emil, interaction of current sity, 335; moment of magnet, 346;
and magnetism, 351, 364 poles of earth, 346; quantum number,
Lever: arm, meaning of, 12; mechani- 671; recording on tape, 488; reluc-
cal advantage of, 112; optical, 119, tance, 342; su bstarices, 337; suscepti-
121 bility, 336
Light: nature of, 503; physical receptors Magnetism: residual, 339; theory of,
of, 514; ray, 507; shadows, 507; 344
sources for illumination, 504; sources Magnetization: curve, 338; intensity of,
for spectroscopy, 602, 604; speed 346; process of, 345
of, 508; utilization coefficients of, 523; Magnetometer, 347
-year, the, 510 Magnetomotive force, meaning of, 342
Lightning: arrester, Thyrite, 424; pro- Magneto-optic effect, 655
tection, 273 Magnetostriction, 338
Linde, Karl R. von, liquefaction of Magnetron, 444
gases, 233 Magnets: behavior in magnetic field,
 INDEX 737

345; classification of, 317; permanent, Mirage, 539

340 Mirror: Equation, 565; focus of, 564;
Magnification: by lens, 557; by mirror, Lloyd's, 620; plane, 526; spherical,
565; normal, 634; of compound micro- 563
scope, 586; of electron microscope, Mks unit system: forms of, 257; rela-
588; of simple microscope, 582; of tions to other systems, 721
telescope, 591 Moderators for neutrons, 702
Magnifying glass, 581 Modulation of waves: amplitude, 431;
1vfalus, Etienne: law of, 648, 655; light frequency, 437
polarization, 641 Mole, the, 161
Manometers, 159 . Molecular: forces and motion, 145;
l{ arconi, Guglielmo, radio transmission, velocities in gases, 155
428 Moment: magnetic, of magnet, 346; of
Mass: definition of, 59; -energy equiva- force, 11; of inertia, 83
lence, 684; number of isotope, 260, :Momentum: angular, 91; conservation
686; of electron, reduced, 668; of of, 75, 91; linear, 73
photon, 678; relativistic, 684; spec- Moon as light source, 504
trograph, 687, 708; units of, 60 1\foseley, Henry G-J., x-ray spectra, 673
Mathematical constants and tables, 714 Motion: accelerated, 40, 52; angular
Matter waves, 678 uniform, 50; circular, 76; combina-
Maxwell, James Clerk: electromagnetic tion of uniform and accelerated, 4 7;
theory, 617; magnetic flux, 318 harmonic, 124; molecular, 145; New-
Maxwell, the, relation to weber, 318 ton's laws of, 60; of falling bodies, 45;
McMillan, Edwin M.: synchrocyclo- of precession, 92; of rotation and
tron, 695; transuranic elements, 709 translation, 16; periodic, 124; -picture
Measurement--see the specific properties projector, interruption rate, 512; pro-
Measurement, physical, 1 jectile, 48; relative, 35; uniform,
Mechanical: advantage of machines, linear, 35
111; efficiency of engines, 250; equiv- Motor: alternating-current, 394; direct-
alent of heat, 238; power, 109 current, 358
Mechanics: classical, 60; quantum and Muller, W., G-M counters, 689, 699
wave mechanics, 679 Multiplier for voltmeter, 327
Meitner, Lisa, nuclear fission, 707 Musical: instruments, 483; scales, 481
Melde, Franz E., wave experiment, 467 ::V[utual induction, 363
Melting points, 202 Myopic eye, 578
:\1ercury: barometer, 156; thermom-
etry, 189
~Iercury-vapor: lamp, 505; rectifier, Near-sightedness, 578
408 Network calculations, 311
Meson, 701 Neutrino, 703
Metacenter, meaning of, 145 Neutron, 701; relative mass of, 259
Meteorological range of vision, 614 Newton, Sir Isaac: interference rings,
Meter-candle, the, 518 622; law of gravitation, 64; laws of
Metric units, 63 motion, 60; reflecting telescope, 591
Mev (million electron· volts), 693, 705 Newton, the, 63
Michelson, Albert A.: interferometer, Nicol, William, polarizing prism, 645
624; speed of light, 509 Nike guided missile, 446
Microscope: compound, 584; electron, Nodal points of thick lens, 561
588, 679; polarizing, 656; simple, 581 Nodes and antinodes, 466, 475
Microwaves, 442 Nuclear: energy, 705; fission and fusion,
Millikan, Robert A.: electronic charge, 707; reactor, 708
27 4; photoelectric effect, 663 Nucleon, 706
Millimicron, the, 542 Nuclide, 686
738 INDEX

Numerical aperture, 636 Periodic: motion, 124; table of ele-

Nu-value of glass, 543 ments, 720
Permeability: concept of, 281; of free
Objective lenses, 585, 590 space, 320; relative, 337; units, equiv-
Ocular, 583 alence of, 367
Oersted, Hans C.: electromagnetism, Permittivity: concept of, 264; of free
318; magnetic intensity, 342 space, 281; units, equivalence of, 37 4
Oersted, the, relation to ampere· turn Perot interferometer, 625
per meter, 343 Phase: change of, 201; change on re-
Ohm, Georg S., law of electric circuit, flection, 621; definition of, 127, 450;
297, 304, 388 of interfering waves, 465, 618; rela-
Ohm, the, 298 tions in alternating-current circuits,
Oil-drop experiment for measuring 384
charge, 274 Phonograph and records, 488
Onnes, II. Kamerlingh: liquefaction of Phorias, 578
gases, 233; superconductivity, 303 Photoconductive cells, 516
Opera glass, 593 Photoelasticity, 652, 656
Optic axis of crystal, 643 Photoelectric: cells, 515; effect, 662; in-
Optical: density, 612; flat, 622; instru- verse effect, 666
ments, 574; lever, 121; path length, Photoelectron, 663
531; 549; pyrometer, 660; rotation, Photoemissive threshold, 664
654; tube length, microscope, 585 Photography, nuclear track, 691
Optics, electron, 416 Photometers, 520
Organ, electric, 484 Photometry, 513, 519
Orifice coefficients of discharge, 168 Photomicrographic outfit, 589
Oscillators: electron tube, 413; micro- Photomultiplier tube, 516
wave, 444 Photon, kinetic energy, mass, and mo-
Oscilloscope: description of, 417; for mentum, 665, 678
measuring speed of sound, 491; for Photopic vision, 511
sound wave shapes, 490 Phototubes, 665
Physical constants, 714
Piezoelectric effect, 435, 497
Pair creation, 700 Pile, atomic, 708
Parallel connection of: capacitors, 375; Pitch of sound, 470, 489
cells, 308; circuits, 307, 389 Pitot tube, 173
Parallel forces, resultant of, 21 Planck, Max: constant, 662, 664; radi-
Parallelogram method of vector addi- ation law, 662
tion, 6, 11, 17, 68, 265, 390 Planetary picture of atoms, 259
Paramagnetic substances, 337 Plates, mechanical vibration of, 477
Particle accelerators, 331, 694 Plutonium, 709
Particles and rigid bodies, 59 Poise, the, 176 .
Pascal, Blaise, transmission of pressure Poiseuille, Jean L. M., viscosity, 177
change, 141 Polarimeter, 654
Paschen, Friedrich, hydrogen spectrum, Polarization: axis, 641; by dichroism,
667 646; by double refraction, 643; by
Path length: optical, 531; venms dis- reflection, 641; by scattering, 647; in
placement, 33 electrolysis, 291; of dielectrics, 371;
Pauli, Wolfgang, exclusion principle, rotatory, 654
671 Polarized light: applications, 655; cir-
Peltier, Jean C. A., thermoelectricity, cularly and elliptically, 653; linear,
404 640; interference of, 650; nature of,
Pendulums, 129 640
Period of vibration, 126, 450 Polarizing angle, 641
 INDEX 739

Pole strength of magnet, 346 Quartz crystal, frequency control by,

Poles, magnet, 317 434
Polygon method of vector addition, 8
Polyphase: alternator, 393; motor, 394 Radar signaling, 445
Ponderator, 694 Radian, the, 51
Position and displacement, 2 Radiant flux, 513
Positron, 700; emission of, 704 Radiation: from circuits, 428; laws, 659;
Potential: barrier, 405; contact, 401; of heat, 191
difference, 269, 283; drop in circuit, Radio: altimeter, 446; amplification,
283, 298; electric, at a point, 268; 436; communication, 433
energy, 104; ionization, 669; measure- Radioactive: decay, 685; series, 687
ment of electric, 310; resonance, 670; Radioactivity: artificial, 703; natural,
single-electrode, 288 682
Potentiometer, 310 Radiometry, 513
Poundal, relation to pound, 713 Rainbow, 613
Power: concept of, 107; electric, 303; Raman, Sir Chandrasekhara V., light
factor of circuit, 390, 397; from scattering, 655, 677
steam, 250; from the sun, 192; in al- Ramsden, Jesse, eyepiece, 583
ternating-current circuit, 390, 396; Rankine, William J ..~!.: temperature
supplied to induction motor, 395; scale, 183; vapor cycle, 246
thermoelectric, 402 Rayleigh, Lord (John W. Strutt): re-
Precession, motion of, 92 solving power criterion, 634; sound
Presbyopia, 578 intensity, 490
Pressure: absolute, 138, 158; atmos- Reactance of circuit, 387
pheric, 156; center of, 140; critical Receiver: radio, 435, 438; telephone,
of gases, 232; effect on boiling, 211; 432; television, 440
effect on fusion, 204; gage, 158; Rectifier: contact, 422; copper-oxide,
liquid, 137; partial, 160; saturated 423; electron tube; 407; germanium,
vapor, 206; solution, 288 424; Kenotron, 407; mercury-arc,
Primary colors: additive, 606; subtrac- 408; selenium, 423; silicon, 424
tive, 607 Reflection: control with polarized light,
Prism: binocular, 593; combinations, 655; law of, 462, 527; losses, 534;
544; deviation by, 537; double- phase change on, 621; -reducing films,
image, 649; Dove, 536; Porro, 534, 623; total internal, 532
,593; reflecting, 533; Rochon, 649; Refraction: atmospheric, 539; by a
W ollaston, 649 parallel plate, 535; double, 643; law
Projectile motion, 48 of, 463, 529; of electron beam, 416
Projector: lantern-slide, 594; motion- Refractive index, 530, 538, 543, 643
picture, 594; television, 595 Refrigeration, process of, 252
Proof plane, use of, 272 Regelation, 204
Proton: charge of, 259; relative mass of, Relative: aperture of a lens, 579; mo-
259; synchrotron, 696 tion, 35
Psychrometer, 210 Relativistic mass, 684
Pulley, mechanical advantage of, 112 Relativity, special theory of, 683
Pumps, liquid, 165 Reluctance of magnetic circuit, 342
Purkinje, Jan E., dark-adapted vision, Resistance: calculation of, 299; concept
511 of, 297; effective, 430; factors affect-
Pyrometer, optical, 660 ing, 299; measurement of, 298, 309;
of ground electrode, 300; plate, of
Q factor of circuit, 431 electron tube, 410; temperature co-
Quantum: mechanics, 679; numbers, efficient of, 302; thermometer, 302
667 Resistivity, meaning of, 299
Quarter-wave plate, 653 Resistors, connection of, 306
740 INDEX

Resolution: limit, criterion, 634; method Selective absorption, 607

of vector addition, 9, 19, 72 Selenium rectifier, 423
Resolving power, 633 Self-induction, 365
Resonance: acoustic, 480; electrical, Semiconductors, 421
389, 430, 442; potential, 670 Series connection of: capacitors, 376;
Resonant cavity, 442 cells, 308; circuits, 306, 387; resistors,
Resultant: meaning of, 5; of parallel 305
forces, 21 Series: electromotive, 290; radioactive,
Reverberation of sound, 493 687; spectral, 667; x-ray, 672
Richardson, Owen W., thermionic emis- Servo-mechanism, Selsyn, 398
sion, 406 Sextant, 529
Rigid bodies and particles, 59 Shadows, light, 507
Rigidity, coefficient of, 121 Shear modulus of elasticity, 120
Rochon, A. M., prism, 649 Shock waves, 498
Rods: retinal, 510, 576; vibration of, Shunt for ammeter, 327
477 Sign conventions for lenses and mirrors,
Roemer, Ole, speed of light, 508 550
Rontgen, Wilhelm K., x-rays, 419 Silicon rectifier, 424
Root-mean-square values, 155, 383 Sky: coloring of, 614; compass, polar-
Rotation: effect of torque on, 82; mo- ization, 655
tion of, 16, 50; of reflected ray, 528 Sloan, David H., linear accelerator, 694
Rotatory: dispersion, 654; polarization, Slug, the, 60
654; power, 654 Smoke elimination, 272
Rowland, Henry A., mechanical equiv- Snell, Willebrord, law of refraction, 530,
alent of heat, 238 536
Rubber, deposition of, 286 Sodium-vapor lamp, 506
Rumford, Count (Benjamin Thompson), Solenoid: flux density within, 322; m-
heat experiments, 181 ductance of, 366
Ruther!ord, Ernest: alpha particle scat- Solid angle, 517
tering, 686; artificial transmutation, Solids, electric conduction in, 270
692; atomic model, 686; neutron Solution pressure, 288
discovery, 701 Sound: absorption of, 494; characteris-
tics of, 470; intensity of, 470, 490;
Sabine, Wallace C., reverberation time, measurement of, 489; pitch of, 470,
494 489; propagation of, 456; quality of,
Saccharimeter, 654 470, 47 4; ranging, 497; recording and
Sagitta of arc, 552 reproduction, 487; records on film,
Saturated vapor pressures, 206 488; reflection of, 497; speed in gases,
Saturation of color, 605 457, 490; reverberation, 493
Savart, Felix, electromagnetism, 323 Space charge, 406
Saybolt, George M., viscosimeter, 177 Specific gravity of: gases, 160; liquids
Scalar quantities, 5 and solids, 136
Scales, musical, 481 Specific heat: definition of, 198; meas-
Scattering, atmospheric, 614 urement of, 200; values of, 199, 228
Schmidt, Bernard, corrector plate, 594 Specific rotation of polarized light, 654
Scintillation counter, 691 Spectra, x-ray, characteristic, 672
Scotopic vision, 511 Spectrochemical analysis, 603
Screw jack, mechanical advantage of, Spectrograph, 598, 604; mass, 687
113 Spectrometer, prism, 539
Seaborg, Glenn T., transuranic elements, Spectrophotometer, 610
709 Spectroscope, 598
Seebeck, Thomas J., thermoelectricity, Spectroscopy: infrared, 604; ultraviolet,
401 604
 INDEX 7-11

Spectrum: absorption, 600; analysis, lean, 593; Newtonian, 591; terrestrial,

603; bright-line, 600; continuous, 511, 592
600; dark-line, 601; electromagnetic, Television, 439
460; emission, 600; lines, 632 Temperature: coefficient of resistance,
Speech: organs of, 484; sounds, 485 302; critical, of gases, 232; gradient,
Speed: angular, 51; linear, 34; of efllux, 192; scales, 183, 245
167; of electromagnetic waves, 459; Tension, surface, 146
of lens, 579; of light, value of, 282, Tetrode electron tube, 414
510; of longitudinal waves, 457; of Thales of Miletus, early experiments,
transverse waves, 455; synchronous, 258
of alternator, 393; versus velocity, 33 Thermionics, 405
Spherical aberrat'ion, 567 Thermocouple, 401, 516
Spring constant, 99, 128 Thermodynamic temperature scale, 245
Stability: of equilibrium, 25; of floating Thermodynamics, laws of, 237
bodies, 144 Thermoelectricity, 401
Standard voltaic cell, 295 Thermometer: clinical, 189; gas, 224;
Statcoulomb, relation to coulomb, 264 mercury, 189; resistance, 302; ther-
Statics, principles of, 16 moelectric, 403; wet-and-dry bulb,
Stationary: states of hydrogen atom, 210
667; waves, 466 Thermopile, 403, 516
Steam: engine, 248; saturated, prop- Thompson, Benjamin (Count Rumford),
erties of, 213; turbine, 251 heat experiments, 181
Stefan, Josef, radiation law, 660 Thomson, George P., electron diffrac-
Steinmetz, Charles P., magnetic hys- tion, 679
teresis, 340 Thomson, Sir Joseph J., ratio of elec-
Steradian, the, 517 tronic charge to mass, 329
Stokes, Sir George, motion m viscous Thomson, William (Lord Kelvin): tem-
medium, 177, 275 perature scales, 183, 245; thermo-
Storage cells, 292 electricity, 404
Strabismus, 578 Three-dimensional pictures, 657
Strain and stress, 117 Thyratron tube, 412
Strassmann, Fritz, nuclear fission, 707 Thyrite rectifier, 424
Stretch modulus of elasticity, 118 Time: constant of circuit, 368; measure-
Strings, vibrations of, 475 ment of, 34; reverberation, of sound,
Strutt, John W. (Lord Rayleigh), sound 494
intensity, 490 Torque: and rotation, 82; definition of,
Sublimation, 205, 213 11; developed by motor, 359; due to
Sun as light source, 504 liquid, 139; on bar magnet in mag-
Superconductivity, 303 netic field, 345
Supercooling and superheating, 214 Torsion: balance, 262; pendulum, 130
Supersonic flight, 498 Total internal reflection, 532
Surface: color, 608; tension, 146 Tourmaline crystals, 646
Susceptibility, magnetic, 336 Tracer studies with artificial radioac-
Synchro, position indicating, 397 tivity, 704
Synchronous speed, 393 Transconductance of electron tube, 410
Synchrotron: proton, 696; electron, 698 Transformers, 395
Transistor, characteristics of, 425
Translation, motion of, 16
Telegraphy, modulation in, 431 Transmittance of light, 612
Telephone receiver, 432 Transmutation, artificial, 692
Telephony, modulation in, 432 Transuranic elements, 709
Telescope: astronomical, 589; Casse- Transverse and longitudinal waves, 450
grainian, 591; cosmic ray, 699; Gali- Trichromatic coefficients, 608
742 INDEX

Trigonometric functions, 716 Vision: defects of, 577; persistence of,

Triode electron tube, 408, 412 512; photopic and scotopic, 511
Triple point, 213 Visual angle, 581
Trump, John G., electrostatic genera- Vocal organs, 484
tor, 273, 694 Volt, the, 268, 283
Tungar rectifier, 408 Volta, Alessandro: contact potentials,
Tuning fork, 478 401; potential in circuit, 268
Turbine: gas, 252; steam, 251 Voltaic cells, 290
Voltmeter, direct-current, 327

Uhlenbeck, George E., electron spm,

670 Walton, Ernest T. S., artificial transmu-
Ultramicroscope, 588 tation, 693, 705
Ultrasonic vibrations, 497 Water: decomposition of, 285; equiv-
Uncertainty principle, 680 alent, 200; vapor in atmosphere, 208;
Uniflow engine, 249 waves, 454
Units: for basic quantities, 63; handling Watt, James, steam engine, 248
of, 13, 35; systems of, for electrical Watt, the, 108, 304
entities, 257 Watthour meter, 328
Uranium-radium radioactive series, 688 Wattmeter: alternating-current, 397;
direct-current, 328
Wave: energy density of, 472; equation,
Valence, number of element, 260, 287 454; -front construction, 460; me-
Van de Graaff, Robert J., electrostatic chanics, 679; Melde experiment on,
generator, 273, 694 467; modulation, 431; motion, 449;
van der Waals, Johannes D ., actual propagation, mechanism of, 451;
gases, 234 shape, measurement of, 490; side
van Leeuwenhoek, Antoni, simple micro- band, 433
scope, 583 Wavelengths: light, 600, radio, 429;
Vapor: definition of, 154; saturated, 206 sound, 480
Vaporization: heat of, 212; process of, Waves: electromagnetic, speed of, 459;
205 energy transmission by, 454; inter-
Varistor, characteristics of, 422 ference and reenforcement of, 465,
Vector: addition, 5; model of atom, 671; 617; light, speed of, 282, 510; longi-
quantities, 5; representation of angu- tudinal and transverse, 450; reflec-
lar quantities, 92; resolution, 9; sub- tion and refraction of, 462, 526;
traction, 7 shock, 498; sound, speed of, 456, 490;
Veksler, V., synchrocylotron, 695 stationary, 466; synthesis of, 474;
Velocity: angular, 51; in accelerated types of, 449
motion, 43; in circular motion, 77; in Weber, Ernst H., stimulus change, 472
harmonic motion, 127; instantaneous, Weber, Wilhelm E., magnetic flux, 318
38; linear, 35; of light, 282, 510; Weber, the, 318
versus speed, 33 Weight, concept of, 65
Venturi meter, 173 Weston, Edward, standard cell, 295
Vertex of optical surface, 548, 563 Wheatstone, Sir Charles, bridge, 308
Vibration: free and forced, 475, 480; White light, significance of, 609
harmonic, 126, 476; longitudinal and Wien, Wilhelm, radiation law, 661
transverse, 450; of air columns, 479; Wilson, Charles T. R., cloud chamber,
of plates and rods, 477; of strings, 690
475; ultrasonic, 497 Wollaston, William H., absorption lines,
Virtual object, 545 542; prism, 649
Viscosity of: gases, 177; liquids, 17 5 Work done: against friction, 105; by
Visibility through atmosphere, 614 expanding gas, 240; in pumping
 INDEX 743

liquid, 166; in rotation, 101; in trans- production of, 419, 66G; wave nature
ferring charge, 269; in translation, 98 of, 673
Work: function of surface, 405, 664;
transformation into heat, 238; versus Y-connection, 394
energy, 101 Yo,ung, Thomas: light interference, 618;
modulus of elasticity, 118
X-ray: characteristic spectrum, 672; Yukawa, Hideki, meson, 701
continuous spectrum, 666; crystal
spectrometer, 674; diffraction, 674; Zworykin, Vladimir K., television, 439