Physics 2233 : Chapter 16 Examples : Sound

Generic relationships: k = 2π/λ ω = 2π/T f = 1/T v = λ/T = λf = ω/k

Sound is a longitudinal (pressure) wave
Range of sound audible to humans: 20 Hz to 20, 000 Hz
Speed of sound in air: v ≈ (331 + 0.60T ) for T in deg C (about 343 m/s at STP, i.e. 20o C)
Displacement: D(x, t) = A sin (kx − ωt) which implies Pressure: ∆P = −∆Pmax cos (kx − ωt)
where: ∆Pmax = BAk = ρv 2 Ak = 2πρvAf
Intensity (power/area): I = 2π 2 ρvA2 f 2 I = (∆Pmax )2 /(2vρ)
Intensity in Decibels: β = 10 log10 (I/Io ) where Io = 1 × 10−12 W/m2
Stringed Instruments (fixed at each end):
q
2L v
λn = n
= v/fn and fn = n( 2L ) for n = 1, 2, 3, ... where v = FT /µ
Wind Instruments:
2L v
Open pipe: λn = n
and fn = n( 2L ) (for n = 1, 2, 3...)
4L v
Closed pipe: λn = n
and fn = n( 4L ) (for n = 1, 3, 5, ...)
Interference (beats) : favg = 21 (f1 + f2 ) fbeat = |f2 − f1 |
Doppler Effect (using book’s conventions) :
f ′ = f · (v ± vobs )/(v ∓ vsrc ) where v = sound speed, vobs = observer speed and vsrc = source speed.
Upper sign if moving toward other; lower sign if moving apart (treat each separately).

Intensity of Various Sounds

Source Sound Level Intensity Material Properties
(dB) W/m2 at 20o C and 1 atm
jet plane at 30m 140 100 Material Sound speed Density
pain threshhold 120 1 m/s kg/m3
loud rock concert 120 1 Air 343 1.20
siren at 30m 100 1 × 10−2 Helium 1005 0.18
truck traffic 90 1 × 10−3 Pure Water 1440 1000
busy street 80 1 × 10−4 Sea Water 1560 1030
noisy restaurant 70 1 × 10−5 Iron, steel 5000 7800
talk at 50 cm 60 3 × 10−6 Glass 4500 2800
quiet radio 40 1 × 10−8 Aluminum 5100 2700
whisper 30 1 × 10−9 Hardwood 4000 1000
rustling leaves 10 1 × 10−11 Concrete 3000 1400
threshhold of hearing 0 1 × 10−12
These were old homework problems from the previous version of the textbook that relate to the
material from this chapter. (The previous book covered some of this material differently and
developed some different equations since many quantities are related and the same equations can be
written many different ways. Hopefully I’ve tracked down and replaced everything using equations
and terms used in our current textbook.)

Example 3 : Consider a sound wave in air that has displacement amplitude 2.00 × 10−2 mm.
Calculate the pressure amplitude for frequencies of (a) 150 Hz, (b) 1500 Hz, and (c) 15000 Hz. In
each case, compare the result to the pain threshold, which is 30 Pa.

The pressure is related to the amplitude of the waves by pmax = BkA where B is the bulk modulus
of the material. For air, at 20o at sea level, this is 1.42 × 105 P a. We need the wave number k in
this equation though. An equation that directly links wave number and frequency is: v = ω/k so
k = ω/v. But ω = 2πf so k = 2πf /v. We know this sound wave is propagating through the air, so
we’ll take the wave speed to be 344 m/s (although technically it varies with temperature, and they
didn’t give us a temperature here).
Finally the amplitude was given in millimeters. There are 1000 mm in one meter, so this displace-
ment amplitude represents A = 2 × 10−5 m.
(a) At f = 150Hz we have k = (2)(π)(150)/(344) = 2.74 rad/m. For the given displacement
amplitude of the wave, then, the maximum pressure this sound generates is pmax = BkA = (1.42 ×
105 )(2.54)(2 × 10−5 ) = 7.78 P a. (All the individual terms were in proper metric units, so the final
answer will come out in the proper metric units for pressure, which is pascals (i.e. newtons per
square meter). This is well below the pain threshold.
(b) At f = 1500 Hz we are multiplying the value of f by 10 compared to part (a). Since k = 2πf /v,
that means the wave-number will also be 10 times larger. Since pmax = BkA , that means the
maximum pressure will also be ten times larger so here pmax = 77.8 P a, which is over the pain
threshold of 30 P a.
(c) At f = 15000 Hz we are multiplying the value of f by 10 compared to part (b), so by the same
arguments given there, we’ll end up multiplying the maximum pressure by yet another factor of
ten, resulting in 778 P a, or about 26 times the pain threshold.
One thing to note here. 15, 000 Hz is within the range of human hearing. Even a minuscule
displacement amplitude of 0.02 mm would produce a disabling pressure. So the actual sounds
we hear in everyday life must have displacement amplitudes much smaller than this, showing how
sensitive the ear is.
Example 4 : A loud factory machine produces sound having a displacement amplitude of 1.00µm,
but the frequency of this sound can be adjusted. In order to prevent ear damage to the workers
(which will certainly occur by a pressure of 30.0 P a), the maximum pressure amplitude of the
sound waves is limited to 10.0 P a. Under the conditions of this factory, the bulk modulus of air is
1.42 × 105 P a. What is the highest frequency sound to which this machine can be adjusted without
exceeding the prescribed limit? Is this frequency audible to the workers?
Similar to the previous problem, we have pmax = BkA, k = 2π/λ = 2πf /v so pmax = 2πf BA/v
or rearranging, f = vpmax /(2πBA). The machine is producing sound with A = 1.00µm which is
A = 1 × 10−6 m. The bulk modulus of air is 1.42 × 105 P a, and the speed of sound is 344 m/s.
So for the given desired maximum pressure of 10 P a, we can calculate the desired frequency to be
(344)(10)
f = (2)(π)(1.42×10 5 )(1×10−6 ) = 3856 Hz.

The range of human hearing generally runs from 20 Hz to 20, 000 Hz, so this frequency will be
audible.
Note: The maximum pressure is proportional to the frequency in pmax = 2πf BA/v so any higher
frequency will produce pressures above the limit of 10 P a. (And lower frequencies will produce less
painful pressures...)

Example 5 : (a) In a liquid with a density of 1300 kg/m3 , longitudinal waves with a frequency of
400 Hz are found to have a wavelength of 8.00 m. Calculate the bulk modulus of this liquid.
(b) A metal bar with a length of 1.50 m has a density of 6400 kg/m3 . Longitudinal sound waves
travel from one end of the bar to the other in a time interval of 3.90 × 10−4 s. What is Young’s
modulus for this metal?
This problem basically shows a method of calculating physical properties such as the bulk modulus
of a liquid, or the Young’s modulus of a metal using sound.
q
(a) In a liquid, the wave speed is given by v = B/ρ. We know the density of this liquid, so if we
can find the wave speed, we can solve for B: B = ρv 2 . We do know the wavelength and frequency
of these waves, so we can find the wave velocity to be v = λf = (8 m)(400 s−1 ) = 3200 m/s. Thus
B = ρv 2 = (1300)(3200)2 = 1.33 × 1010 P a. (All the individual terms were in proper units, so B
will come out in proper metric units as well, and for a bulk modulus, those units are pascals.)
(Note: the given density sounds like a lot, but consider pure water. It has a density of 1 gram per
cubic centimeter. Converting to standard metric units, this is 1000 kg/m3 , so this liquid is only 30
percent denser than water. One cubic meter of water (a cube only about three feet on each side,
would have a mass of 1000 kg which is a weight of 2200 pounds - over a ton. Imagine the force an
entire swimming pool full of water on the roof of a building is producing.)
(b) Here we are given information that lets us calculate the wave speed. The ‘sound’ took 3.90×10−4
seconds to travel 1.5 m so the wave speed must be v = d/tq= (1.50 m)/(3.90 × 10−4 s) = 3846 m/s.
The wave speed in this type of material is given by v = Y /ρ though so we can rearrange this to
find Y = ρv 2 . For this particular metal, then: Y = (6400)(3846)2 = 9.47 × 1010 P a. (Again, if we
use proper metric units for all the individual terms, the final answer will come out in the proper
metric units for a Young’s modulus, which is pascals or newtons/meter.)
Example 7 : A submerged scuba diver hears the sound
of a boat horn directly above her on the surface of the
lake. At the same time, a friend on dry land 22.0 m from
the boat also hears the horn . The horn is 1.20 m above
the surface of the water. What is the distance (labeled
by the question mark in the figure) from the horn to the
diver? (Both air and water are at 20o C.)

For the person on the shore, the sound travels entirely through the air. For the person in the water,
the sound travels the first 1.2 m in air, then the rest of the time it’s under the water. Since for the
first 1.2 m, the sound is traveling through the air for both listeners, let’s discard that time interval
and just look at the rest of the interval. Now we have sound traveling 22 − 1.2 = 20.8 m through
the air to the land-based listener, and some distance d entirely under the water to get to the diver.
The time interval t = d/v is the same for each listener for this part of the path. For the land-
based listener, t = (20.8 m)/(344 m/s) = 0.0605 s. For the diver, t = d/v but now we know that
t = 0.00605 s and we also know that v = 1482 m/s (the speed of sound in water at 20 degrees)
so (0.0605 s) = (d)/(1482 m/s) or d = 89.6 m. We can avoid some potential round off error by
writing this a little differently and bypass calculating the time interval itself. The time intervals
are the same so da /va = dw /vw where ‘a’ represents the 20.8 m the sound is traveling in the air to
the land-based listener during this interval and ‘b’ represents the unknown distance the sound is
traveling in the water to the diver. Then dw = da × vvwa = (20.8 m) 1482
344
= 89.61 m.
The problem didn’t ask for how deep the diver was, though. They want to know how far the diver
is from the horn. The direct distance from the horn to the diver then is the 1.20 m from the horn
to the surface of the water, plus the 89.6 m that the diver is under the water, for a total distance
of 90.8 m.
Example 11 : An 80.0-m-long brass rod is struck at one end. A person at the other end hears two
sounds as a result of two longitudinal waves, one traveling in the metal rod and the other traveling
in the air. What is the time interval between the two sounds? Take the speed of sound in air to be
344 m/s. (Use 8600 kg/m3 for the density of brass and 9.00 × 1010 P a for the Young’s modulus of
brass.)

The distance the sound travels is related to the velocity and time by d = vt, or t = d/v. In air, the
sound wave travels 80.0 m at a speed of 344 m/s so t = (80.0 m)/(344 m/s) = 0.2326 s.
In the brass rod, the sound will travel the same distance q
but at a considerably
q different speed. The
10
speed of sound in the metal rod will be given by v = Y /ρ = (9 × 10 )/(8600) = 3235 m/s
(nearly ten times the speed of sound in air). The time in the brass rod then is t = d/v =
(80.0 m)/(3235 m/s) = 0.0247 s.
The time interval between these two sounds then is (0.2326 s) − (0.0247 s) or 0.2079 s.
Example 14 : Use information from the table to answer
the following questions about sound in air. At 20o C the
bulk modulus for air is 1.42 × 105 P a and its density is
1.20kg/m2 . At this temperature, what are the pressure
amplitude (in Pa and atm) and the displacement ampli-
tude: (a) for the softest sound a person can normally
hear at 1000 Hz and (b) for the sound from a riveter
at the same frequency. (c) How much energy per second
does each wave deliver to a square 5.00 mm on a side?

We have equations that directly relate intensity to the

2
pressure amplitude: I = 2p√max
ρB
and to the displacement
q
amplitude I = 21 (ρB)ω 2 A2 . Rearranging these equa-
tions to solve for the pressure and displacement ampli-
tudes, we
q have:
√ q
pmax = 2I ρB and A = ω1 √2IρB .
Throughout, we have a frequency of 1000 Hz which rep-
resents an angular frequency of ω = 2πf = 6283rad/s.

(a) The softest sound a person can hear has an intensity of 1 × 10−12 W/m2 . We have all the
other constants we need, so we can calculate the pressure amplitude to be pmax = 2.9 × 10−5 P a.
One atmosphere is a pressure of 1.013 × 105 P a so this represents pmax = 2.8 × 10−10 atm . The
displacement amplitude for this sound would be 1.1 × 10−11 m.
√
(b) For the riveter, we can take aqshortcut and notice that A is proportional to I so the new
3.2×10−3
amplitude will be (1.1 × 10−11 m) 1.0×10 −12 or 6.2 × 10
−7
m.
√
The maximum pressure pmax isqalso proportional to I so the value here is related to the value in
3.2×10−3
(a) by pmax = (2.9 × 10−5 P a) 1.0×10 −12 or 1.6 P a which is equal to 1.6 × 10
−5
atm.
(c) The intensity is the amount of power per area, so if we multiply the intensity by the area of
interest, we find the amount of power passing through that area. The 5.00 mm size is roughly an
ear hole, although here we’re looking at a tiny square that is this length along each side. The area
of this square shape is A = L2 = (5 × 10−3 m)2 or 2.5 × 10−5 m2 . For the softer sound, then, this
represents a power of (I)(area) = (1 × 10−12 W/m2 )(2.5 × 10−5 m2 ) = 2.5 × 10−17 W . For the
riveter, this represents a power of (I)(area) = (3.2 × 10−3 W/m2 )(2.5 × 10−5 m2 ) = 8.0 × 10−8 W .
Example 20 : (a) When four quadruplets cry simultaneously, how many decibels greater is the
sound intensity level than when a single one cries? (b) To increase the sound intensity level again
by the same number of decibels as in part (a), how many more crying babies are required?
(a) In terms of decibels, the intensity is β = (10 dB) log(I/Io ). Comparing two intensity levels, we
derived that: ∆β = β2 − β1 = (10 dB) log(I2 /I1 ). In this case, we are told that I2 (the intensity
with four crying babies) is exactly four times the initial intensity, so ∆β = (10 dB) log(4I/I) =
(10 dB) log(4) = 6.02 dB. A 6 dB increase in the sound level means the underlying intensity went
up by a factor of 4.
(b) In (a), we found that we have to multiply the number of babies by four to produce a 6 dB
increase in the sound intensity. To get yet another 6 dB increase, we have to multiply by another
factor of 4 giving us 16 babies altogether (an additional 12).

Example 21 : A baby’s mouth is a distance of 30 cm from her father’s ear and a distance of
1.50 m from her mother’s ear. What is the difference between the sound intensity levels heard by
the father and by the mother?
As we change the distance from the source, the intensity falls off as the square of the distance:
I = P/r2 . Comparing two intensities then: I2 /I1 = r12 /r22 . The difference in two intensities in terms
of decibels, though, is given by ∆β = β2 −β1 = (10 dB) log(I2 /I1 ), so here ∆β = (10 dB) log(r12 /r22 ),
or in the more traditional form: ∆β = (20 dB) log(r1 /r2 ).
The intensity is going to be higher at the person who is closer, so let’s label the father as ‘2’ and
the mother as ‘1’ (that way ∆β = β2 − β1 will be a positive number). So here r1 = 1.50 m and
r2 = 0.30 m so r1 /r2 = 5 and ∆β = (20 dB) log(5) = 13.98 dB.
A useful rule of thumb comes from this equation. Each time we double our distance from a source,
the sound intensity drops by a factor of 6.02 dB (usually just rounded off to 6 dB). Each time we
cut the distance from a source in half, the sound intensity goes up by 6.02 dB.

Example 24 : The fundamental frequency of a pipe that is open at both ends is 594 Hz. (a) How
long is the pipe? (b) If one end is now closed, find the wavelength of the new fundamental. (c) If
one end is now closed, find the frequency of the new fundamental.
v
(a) For an open pipe, the fundamental frequency is f1 = 2L . Here we are given f1 and we know
the speed of sound to be 344 m/s so the only unknown is the length of the pipe. Rearranging the
equation, we have L = 2fv1 = (344)/(2 × 594) = 0.290 m.
(b) If we close one end, we must have a node at that end, and an anti-node at the open end (with
no other nodes or anti-nodes between, since this is the fundamental). The distance between a node
and an anti-node represents one quarter of a wavelength. So here this quarter of a wavelength must
be L meters long: λ/4 = 0.290 m or λ = 1.16 m.
v
(c) If we close one end, the fundamental frequency of this ‘closed pipe’ is given by f1 = 4L . But we
1 v
can write this as f1 = 2 × 2L . That last fraction is just the fundamental frequency for the OPEN
pipe that we were given though, so for the close pipe, f1 = 21 (594) = 297 Hz.
Example 27 : The human vocal tract is a pipe that extends about 17 cm from the lips to the
vocal folds (also called ‘vocal cords’) near the middle of your throat. The vocal folds behave rather
like the reed of a clarinet, and the vocal tract acts like a closed pipe. Estimate the first three
standing-wave frequencies of the vocal tract. Use v = 344m/s. (The answers are only an estimate,
since the position of lips and tongue affects the motion of air in the vocal tract.)

v
For a closed pipe, the fundamental frequencies are given by fn = n 4L for n = 1, 3, 5, .... The lowest
frequency then would be f1 = (344 m/s)/(4 × 0.17 m) = 505.88 Hz. The next two would be
f3 = 3f1 = 1517.6 Hz and f5 = 5f1 = 2529 Hz.
(So here we have a series of particular frequencies that will cause standing waves to form. For these
frequencies, then, we have nodes putting pressure on the vocal tract at fixed locations. I imagine
then it might be more difficult for a singer to maintain one of these frequencies, than to maintain
other frequencies. Any singers want to comment? Are there certain frequencies that are harder to
maintain for a long time? On the other hand, the throat is not a perfect cylinder, so this effect may
be blurred out..)

Example 28 : The auditory canal of the ear is filled with air. One end is open, and the other end
is closed by the eardrum. A particular person’s auditory canal is 2.40 cm long and can be modeled
as a pipe. (a) What is the fundamental frequency and wavelength of this person’s auditory canal?
Is this sound audible? (b) Find the frequency of the highest audible harmonic of this person’s canal.
Which harmonic is this?

(a) We will model the ear as a closed pipe of length L = 0.024 m. For a closed pipe, λ1 = 4L
v
so in the ear canal, λ1 = (4)(0.024 m) = 0.096 m. The fundamental frequency will be f1 = 4L =
344
(4)(0.024)
= 3583.33 Hz. This is in the audible range of 20 to 20, 000 Hz.
v
(b) For closed pipes, the fundamental frequencies are f1 = 4L and fn = nf1 where n = 1, 3, 5, .... If
we set fn to 20, 000 Hz then nominally n = fn /f1 = (20000)/(3583.33) = 5.58. Well n can’t be a
fraction, so it looks like n = 5 will be the highest (odd) value that n can have and still produce a
frequency that is inside the human hearing range. (That means that there are only three frequencies:
f1 = 3583.33 Hz, f3 = 3f1 = 10, 750 Hz, and f5 = 5f1 = 17, 917 Hz which will produce standing
waves in the ear canal. (Something special should probably happen at these frequencies. Perhaps
they would be more painful since we would have standing waves in the ear canal, producing pressure
maxima at fixed locations. A real ear canal isn’t a perfect cylinder of the given length, so these
variations probably eliminate whatever effect would be produced here.)
Example 31 : You blow across the open mouth of an empty test tube and produce the fundamental
standing wave of the air column inside the test tube. The speed of sound in air is 344 m/s and the
test tube acts as a closed pipe. (a) If the length of the air column in the test tube is 14.0 cm, what
is the frequency of this standing wave? (b) What is the frequency of the fundamental standing wave
in the air column if the test tube is half-filled with water?
With a closed pipe, the fundamental wavelength will be λ1 = 4L =
(4)(0.14 m) = 0.56 m. This implies a frequency of f1 = v/λ =
(344 m/s)/(0.56 m) = 614.3 Hz. Half filling the test tube with water
cuts the length of the air column in half to the wavelength of the first fun-
damental will be cut in half, meaning that the frequency will be multiplied
by 2, producing a sound of frequency (2)(613.4 Hz) = 1228 Hz.

Example 38 : Two guitarists attempt to play the same note of wavelength 6.50 cm at the same
time, but one of the instruments is slightly out of tune and plays a note of wavelength 6.52 cm
instead. What is the frequency of the beat these musicians hear when they play together?

The beat frequency created by two nearby frequencies is fb = |f1 − f2 | and we could convert each of
the given wavelengths into frequencies and just compute it directly. Since we often have sounds in
terms of wavelengths instead of frequencies, it is convenient to derive a general equation in terms
of the wavelengths directly.
Since v = f λ, we have f = v/λ. So here, fb = |f1 − f2 | becomes fb = | λv1 − λv2 | or fb = v| λ11 − λ12 |.
But we can write this as fb = v| λλ21−λ
λ2
1
| or since the wavelengths are positive, finally: fb = v |λλ21−λ
λ2
1|
.
(0.0002)
Here, λ1 = 0.0650 m and λ2 = 0.0652 m so fb = (344) (0.0650)(0.0652) = 16.2 Hz.

Example 39 : Two organ pipes, open at one end but closed at the other, are each 1.14 m long.
One is now lengthened by 2.00 cm. Find the frequency of the beat they produce when playing
together in their fundamental.

The fundamental frequency of a closed pipe is f1 = v/(4L). The beat frequency we usually write as:
fb = |f1 − f2 | where f1 and f2 label the two frequencies involved. We’ve already used the subscript
1 to represent the fundamental though, so let’s rewrite this to avoid confusion as: fbeat = |fa − fb |
where fa and fb are now the two frequencies under consideration.
Since the closed pipe has fundamentals related to the length as f1 = v/(4L) we can rewrite the
beat frequency in terms of the length of the pipes directly (since that’s what we’re changing):
fbeat = v4 | L1a − L1b | or further as fbeat = v |L4La −L b|
a Lb
.

In this case, La = 1.14 m and Lb = 1.16 m so fbeat = v |L4La −L b|

a Lb
(0.02)
or fbeat = (344) (4)(1.14)(1.16) =
1.3 beats/s.
Example 41 : On the planet Arrakis a male ornithoid is flying toward his mate at a speed of
25.0m/s while singing at a frequency of 1200 Hz. If the stationary female hears the sound at a
frequency of 1240 Hz, what is the speed of sound in the atmosphere of Arrakis?

(NOTE: updated to reflect this book’s version of the Doppler equation)

The Doppler equation relates the frequencies of the source and listener to their speeds (assuming
the air is stationary and not moving also): f ′ = f · (v ± vobs )/(v ∓ vsrc ) where v = sound speed,
vobs = observer speed and vsrc = source speed. Upper sign if moving toward other; lower sign if
moving apart (treat each separately).
Here, the listener is not moving so vobs = 0. The source is moving TOWARDS the listener, so we’ll
use the upper sign on that term, leading to:
f ′ = f · (v + 0)/(v − vsrc )
The source is moving at a speed of vsrc = 25 m/s, the source frequency is f = 1200 Hz which the
listener hears as f ′ = 1240 Hz so:
v
1240 = 1200 v−25
(25)(1240)
Mutiplying both sides by (v − 25) and collecting terms, we can rearrange this into v = 1240−1200
=
775 m/s
Example 42 : A police car with its 300 Hz siren is moving away from a warehouse at a speed of
20.0m/s. What frequency does the driver of the police car hear reflected from the warehouse? (Use
344 m/s for the speed of sound in air.)

There are two steps here. First, the siren is moving away from the wall, so AT THE WALL, there
will be a frequency of some (lower) value impinging on the wall. That same frequency will then
bounce off the wall and head off towards the police car. Now we have THAT frequency heading
towards the police car, which is moving away (producing a yet lower frequency when that sound
reaches the moving car).
Here, we will need to apply the Doppler equation twice: first to figure out what frequency is hitting
(and then radiating from) the wall, then what frequency that turns into when it arrives at the ears
of the policeman.
What frequency arrives at (and reflects from) the wall?
f ′ = f · (v ± vobs )/(v ∓ vsrc ) where v = sound speed, vobs = observer speed and vsrc = source speed.
Upper sign if moving toward other; lower sign if moving apart (treat each separately).
Here the ‘listener’ is the (stationary) wall, so vobs = 0. The source is emitting a frequency of f = 300
and is moving away from the listener, so we’ll use the lower sign in the doppler equation on that
term:
f ′ = (300 Hz) · (344 + 0)/(344 + 20) = 283.5 Hz.
Frequency heard at the moving police car.
Now we have a 283.5 Hz source (coming from the stationary wall) and we need to find out what
frequency is received at the moving police car.
f ′ = f · (v ± vobs )/(v ∓ vsrc ) and now f = 283.5 Hz; the source is not moving so vsrc = 0; the listener
is moving away from the source so we’ll use the lower sign on that term, leading to:
f ′ = (283.5 Hz) · (344 − 20)/(344 + 0) = 267.0 Hz.
We could take this problem one step further. A person inside the police car is being hit by two
frequencies: the original 300 Hz siren, and the 267 Hz sound reflected from the wall of the building.
This will produce a beat frequency of fb = |f1 − f2 | = |300 − 267| = 33 Hz.
Example 43 : Two train whistles, A and B, each have
a frequency of 392Hz. A is stationary and B is moving
toward the right (away from A) at a speed of 35.0m/s. A
listener is between the two whistles and is moving toward
the right with a speed of 15.0m/s. No wind is blowing.
Take the speed of sound to be 344 m/s. (a) What is
the frequency from train A as heard by the listener? (b)
What is the frequency from train B as heard by the lis-
tener? (c) What is the beat frequency detected by the
listener?
(a) Train A:
f ′ = f · (v ± vobs )/(v ∓ vsrc ) where v = sound speed, vobs = observer speed and vsrc = source speed.
Upper sign if moving toward other; lower sign if moving apart (treat each separately).
Here the source is not moving, so vsrc = 0. The observer is moving away from train A, so we’ll use
the lower sign on that term, leading to: f ′ = (392 Hz)(344 − 15)/(344 + 0) = 375 Hz.
(b) Train B: This time, we have a source emitting f = 392 Hz. The listener is moving towards the
source of the sound, so we’ll use the upper sign on the vobs term. The source is moving away from
the listener, so we’ll need the lower sign on the vsrc term. This leads to:
f ′ = (392 Hz)(344 + 15)/(344 + 35) = 371 Hz
Note that in both cases here, the frequency heard by the (moving) person in the middle was less
than the frequency the whistles were putting out.
(c) The beat frequency created here will be fb = |f1 − f2 | = |375 − 371| = 4 Hz.

Example 44 : A railroad train is traveling at a speed of 25.0 m/s in still air. The frequency of the
note emitted by the locomotive whistle is 400 Hz. (Assume the speed of sound in air is 344 m/s.)
(a) What is the frequency and wavelength of the sound waves in front of the locomotive? (b) What
is the frequency and wavelength of the sound waves behind the locomotive?

(a) In this case, the listener is not moving so vobs = 0. The source is moving towards the listener,
so we’ll use the upper sign on the vsrc term:
f ′ = f · (v ± vobs )/(v ∓ vsrc ) becomes: f ′ = (400 Hz)(344 + 0)/(344 − 25) = 431 Hz.
f = v/λ so λ = v/f = (344)/(431) = 0.798 m.
(b) In this case, the listener is not moving so vobs = 0. The source is moving away from the listener,
so we’ll use the lower sign on the vsrc term:
f ′ = f · (v ± vobs )/(v ∓ vsrc ) becomes: f ′ = (400 Hz)(344 + 0)/(344 + 25) = 373 Hz.
λ = v/f = (344)/(373) = 0.922 m.
Example 46 : A sound source producing 1.00 kHz waves moves toward a stationary listener at
one-half the speed of sound. (a) What frequency will the listener hear? (b) Suppose instead that
the source is stationary and the listener moves toward the source at one-half the speed of sound.
What frequency does the listener hear? (c) Why do these differ?

(a) In this case the source is emitting a frequency of f = 1000 Hz. The observer is stationary so
vobs = 0. The source is moving towards the listener, so we’ll use the upper sign on the vsrc term.
f ′ = f · (v ± vobs )/(v ∓ vsrc ) becomes: f ′ = (1000 Hz)(344 + 0)/(344 − 177) = 2000 Hz.
(b) In this case, the source is emitting a frequency of f = 1000 Hz. The source is stationary, so
vsrc = 0. The listener is moving towards the source so we’ll use the upper sign on the vobs term.
f ′ = f · (v ± vobs )/(v ∓ vsrc ) becomes: f ′ = (1000 Hz)(344 + 177)/(344 + 0) = 1500 Hz.
(c) Let’s think of what’s happening from the point of view of the listener in each case. In (a), the
air isn’t moving and the source is moving towards the listener, producing the higher frequency. In
(b), the listener is moving so he is feeling the air rush past him at half the speed of sound. So to
the listener, the source is moving towards him AND the air that sound is in is also moving towards
him, resulting in a different frequency.

Example 50 : In the not-too-distant future, it should be possible to detect the presence of planets
moving around other stars by measuring the Doppler shift in the infrared light they emit. If a planet
is going around its star at 50.0km/s while emitting infrared light of frequency 3.330 × 1014 Hz, what
frequency light will be received from this planet when it is moving directly away from us? (Note:
Infrared light is light having wavelengths longer than those of visible light.)
q
c−v
The Doppler equation when the wave speed is the speed of light is: fR = f ,
c+v S
so here: fR =
q
3.0×108 −50000
3.0×108 +50000
(3.333 × 1014 Hz = 3.329 × 1014 Hz.
For velocities that are small relative to the speed of light (which 50, 000 m/s certainly is), this is
not a very convenient (or accurate) way to do this calculation. It relies on the calculator retaining
a lot of digits.
More normally, we do these by calculating the change in frequency. So we let fR = fS + ∆f
r . Inside
1−v/c
the square root, we divide the numerator and denominator by c, resulting in: fR = f .
1+v/c S
1 1−v/c
But from a Taylor’s series analysis, 1+x
= 1 − x + (higher order terms) if x is small, so 1+v/c
is
q
approximately equal to (1 − v/c)2 . Going back to our equation for fR , then: fR = (1 + v/c)2 fS
or fR = (1 + vc )fS = fS + vc fS and finally we can pick off ∆f = vc fS . This lets us calculate the
CHANGE in the frequency with much greater precision.
Example 64 : Not test material, but an interesting case in how the properties of a
material limit what we do do with it.
One type of steel has a density of 7800 kg/m3 and a breaking stress of 7.0×108 N/m2 . A cylindrical
guitar string is to be made out of a quantity of steel with a mass of 4.00 g. (a) What is the length
and radius of the longest and thinnest string that can be placed under a tension of 900 N without
breaking? (b) What is the highest fundamental frequency that this string could have?

(a) The ‘breaking stress’ tells us what tension a wire of a given cross sectional area can maintain
without breaking. If a single wire can support some weight before breaking, then two identical
such wires, or one wire with twice the area of the first, would be able to support twice the weight
without breaking. So (breaking stress) × (area) = (maximum tension bef ore breaking). For this
particular type of steel: (7 × 108 N/m2 )(A) = (900 N ) from which A = 1.2857 × 10−6 m2 . But
A = πr2 so we can convert this to a radius of r = 6.40 × 10−4 m (0.64 mm). Anything thinner and
the force/area would exceed the breaking stress. This is the thinnest we can make this wire and
have it (barely) survive a tension of 900 N .
The mass of this ‘cylinder’ of steel will be its density times its volume: M = ρ(πr2 )(L) or M = ρAL.
It is more accurate to use this second form because we calculated A, then used A to calculate r.
Each time we do a calculation, we introduce more numerical and round-off error. M = ρAL so
0.004 kg = (7800 kg/m3 )(1.2857 × 10−6 m)(L) = 0.399 m (or 40 cm, which is a bit shorter than
the length of a typical real guitar string).
v
(b) A guitar
q
string is locked down at both ends, so the lowest
q
frequency mode here will be f1 = 2L
q q
v F 1 F 1 900
but v = F/µ so f1 = 2L m/L
This simplifies to: f1 = 2 ML
or f1 = 2 (0.004)(0.399)
= 375.5 Hz.
If we lower the tension in the string, it will produce a lower frequency. If we raise the tension
it would produce a higher frequency, but we did this calculation for the string being under the
maximum possible tension before it breaks, so this string can’t be adjusted to produce any higher
frequency. If we do need to tune it to a higher frequency, we’ll need to replace the string with
something made of a different material.
 Interference Examples : Not on this test - we’ll come back to this topic in a later
chapter

Example 33 : Two loudspeakers, A and B, are driven

by the same amplifier and emit sinusoidal waves in phase.
Speaker B is 2.00 m to the right of speaker A. Consider
point Q along the extension of the line connecting the
speakers, 1.00 m to the right of speaker B. Both speakers
emit sound waves that travel directly from the speaker
to point Q. (a) What is the lowest frequency for which
constructive interference occurs at point Q? (b) What is
the lowest frequency for which destructive interference
occurs at point Q?
Constructive interference occurs when the difference in the distances of each source from the listener
is exactly an integer number of wavelengths. The interference is destructive when this difference
of path lengths is half a wavelength plus an integer number of wavelengths. The lowest frequency
implies the largest wavelength. Here the listener is 3 m from speaker A and 1 m from speaker B,
so the difference in the paths is 2 m.
(a) In the case of constructive interference, we need to fit some integer number of wavelengths in the
path difference, so 2m = nλ so λ = (2m )/n. The frequency is f = v/λ so this gives us frequencies
of f = (344)/(2/n) or 177n. The lowest frequency then will be just 177 Hz.
(b) In the case of destructive interference, we need the path difference to represent a half a wave-
length plus some integer number of waves or (2 m) = (n + 12 )λ or λ = 2/(n + 12 ) for n = 0, 1, 2, ....
The corresponding frequencies then will be f = v/λ = (344)/(2/(n + 21 )) = (n + 12 )(344/2) or
f = (n + 12 )177 Hz. Since our series runs from n = 0, 1, 2, ..., the lowest frequency then will be (at
n = 0) 177/2 or 86 Hz.
The book argues this slightly differently using a different series for n. They argue that for destructive
interference, the path difference has be half a wavelength plus multiples of a full wavelength, so the
path difference of two meters represents the series: 12 λ, ( 12 + 1)λ = 23 λ, ( 12 + 2)λ = 52 λ, and so on.
They then write this as n λ2 for n = 1, 3, 5, .... These have to equal the path difference of 2 m so we
can write n λ2 = 2 for n = 1, 3, 5, ... or λ = 4/n for n = 1, 3, 5, ....
v
The corresponding frequencies then are: f = v/λ = (4/n) = nv/4 = (n)(344)/(4) = 86n, again for
n = 1, 3, 5, .... Again, the lowest frequency comes out to be (at n = 1 here): 86 Hz.
Example 34 : Two loudspeakers, A and B, are driven
by the same amplifier and emit sinusoidal waves in phase.
Speaker B is 2.00 m to the right of speaker A. The fre-
quency of the sound waves produced by the loudspeakers
is 206 Hz. Consider point P between the speakers and
along the line connecting them, a distance x to the right
of speaker A. Both speakers emit sound waves that travel
directly from the speaker to point P. (a) For what values
of x will destructive interference occur at point P? (b) For
what values of x will constructive interference occur at
point P? (c) Interference effects like those in parts A and
B are almost never a factor in listening to home stereo
equipment. Why not?

Constructive interference occurs when the difference in the distances of each source from the listener
is exactly an integer number of wavelengths. The interference is destructive when this difference
of path lengths if half a wavelength plus an integer number of wavelengths. The listener is located
BETWEEN the two speakers, so the values of x are limited to the range from 0 to 2.0 m. Sound
travels from A to the listener, and from B to the listener. The path length from A to the listener
is x. The path length from B to the listener is L − x (where L is the distance between the two
speakers, here 2.0 m). (The path length is always some positive number of meters: the DISTANCE
from the source to the listener). The difference in the path length (we’ll call that ∆l) then is
∆l = (L−x)−x = L−2x or rearranging x = (L−∆l)/2 or for this specific situation, x = (2−∆l)/2
or x = 1−0.5∆l. It’s the difference in path length (∆l) that has to be any multiple of the wavelength
(for constructive interference) or equal to half a wavelength plus some multiple of wavelengths (for
destructive interference). The multiples in each case can be either positive or negative, representing
either one of the speakers being further away. Once we have ∆l from those considerations, we can
find the corresponding values for x.
The frequency of 206 Hz represents a wavelength of λ = v/f = (344)/(206) = 1.67 m.
(a) For destructive interference, ∆l = (n + 21 )λ. For n = 0, we have ∆l = 21 λ = (0.5)(1.667) =
0.833 m. This gives x = 1.0 − ∆l/2 = 0.58 m. For n = 1, we have ∆l = (3/2)λ = 2.50 m so
x = −0.25 m which is outside the line between the two speakers. Any higher positive values of n
will just put this node further and further away. For n = −1, we have ∆l = (−1 + 12 )λ = −0.833 m
from which x = 1.0 − ∆l/2 = 1.42 m. You should check but any other values again put x outside
the range of [0, 2 m] that is required for the point to be between the two speakers.
So we have exactly two points between the speakers where destructive interference will occur:
x = 0.58 m and x = 1.42 m. Note that the second position is exactly 0.58 m to the LEFT of the
right speaker, so we have symmetrically located positions here.
(b) For constructive interference, the path difference has to be any integer multiple of the wave-
length, that is: ∆l = nλ where (n = 0, ±1, ±2, ...). n = 0 implies ∆l = 0 (i.e. there are the exact
same number of waves from A to the listener as there are from B to the listener, so the listener must
be the same distance from each speaker.) But x = 1.0 − ∆l/2 so ∆l = 0 implies that x = 1.0 m.
(This is exactly at the midway point between the two speakers, so that makes sense. The sound
from each speaker arrives at the same time - the waves will always be in phase, so they will definitely
be constructively interfering.)
For n = 1 we have ∆l = (1)(1.667) = 1.667 m so x = 1.0 − (1.667)/2 = 0.167 m. For n = −1 we
have ∆l = (−1)(1.667) = −1.667 m so x = 1.0 − (−1.667)/2 = 1.883 m. Other values of n put the
values of x outside of the range [0, 2 m].
(Again, note that these two points are symmetrically located: the first is 0.167 m to the right of
the left speaker, and the second is 0.167 m to the left of the right speaker.)
(c) Treating speakers as point sources is a poor approximation for these dimensions, and sound
reaches these points after reflecting off the floor, ceiling, walls, and any objects in the room, so it is
not likely one could perceive these interference effects.
Example 70 : (Not for the first test; we’ll see this again later.) Two
identical loudspeakers are located at points A and B, 2.00 m apart. The
loudspeakers are driven by the same amplifier and produce sound waves
with a frequency of 784 Hz. Take the speed of sound in air to be 344 m/s.
A small microphone is moved out from point B along a line perpendicu-
lar to the line connecting A and B (line BC in the figure). (a) At what
distances from B will there be destructive interference? (b) At what dis-
tances from B will there be constructive interference? (c) If the frequency
is made low enough, there will be no positions along the line BC at which
destructive interference occurs. How low must the frequency be for this
to be the case?
(a) Destructive interference occurs when the path difference is equal to half a wavelength (plus any
positive or negative integer number of wavelengths). The wavelength of this sound is λ = v/f =
(344)/(784) = 0.439 m.
Let’s set things up symbolically at first. If the separation between the speakers is denoted √ by h,
then the distance from B to√the listener is x and the distance from A to the listener is h2 + x2 .
The path difference then is h2 + x2 − x (which √ will always be positive) and that result must be
an odd-integer multiple of the wavelength. So: h2 + x2 − x = n(λ/2) where n = 1, 3, 5, .... We’re
looking for the values that x can have. If we add x to both sides of this equation and square the
resulting equation, we have h2 + x2 = (x + nλ 2
)2 = x2 + x nλ + ( nλ
2
)2 . We can now subtract x2 from
h2
each side and rearrange to solve for x: x = nλ − nλ
4
. Plugging in the specific values for h and λ here:
x = 9.112
n
− 0.1098n. n = 1 gives x = 9.00 m. n = 3 gives x = 2.71 m. n = 5 gives x = 1.27 m.
n = 7 gives x = 0.53 m. n = 9 gives x = 0.026 m. Any larger values for n produce negative values
for x (points to the left of the speakers, and again we were only interested in the x > 0 solutions.
So those five points are the only places where destructive interference will occur.
(b) For constructive interference, the
√ path difference must be equal to any integer number of full
wavelengths (even or odd), so: So: h2 + x2 − x = nλ where n = 1, 2, 3, .... (n = 0 means there are
zero wavelengths between the two, which would imply they are on top of one another, so we know
this series has to start with at least n = 1.) Doing the same sort of algebraic trickery above, we
h2
arrive at x = 2nλ − nλ2
= 4.556
n
− 0.2195n.
At n = 1 we have x = 4.34 m. At n = 2 we have x = 1.84 m, and so on with the remaining
solutions being 0.86 m and 0.26 m.
(c) It is claimed that if we make the frequency low enough, there will not be any points where
destructive interference occurs. The hand-waving argument here proceeds as follows: The lowest
frequency for which destructive interference will occur at x = 0 will be when h is exactly half a
wavelength. If λ/2 is any larger than h we can’t have any destructive interference at all. A more
mathematical approach would be to take the final equation we derived for the locations of the points
h2
x for destructive interference: x = nλ − nλ4
and couple it with the fact that x has to be positive.
Then for no solutions to exist, we would want the right hand side of that equation to be less than
h2 h2
zero. nλ − nλ
4
< 0 implies that nλ < nλ4
. Rearranging this we arrive at: 2h
n
< λ. We want to find
the lowest frequency, but that implies the largest possible λ. The largest λ occurs when n = 1 so
λ > 2h is the solution again. f = v/λ to larger values of the wavelength imply lower values for f .
v
So f < 2h of f < 86 Hz here.