Physical pendulum
Number 135610-EN Topic Mechanics, rigid bodies
Version 2016.08.11 / HS Type Student exercise Suggested for grade 12+ p. 1/5
Objective
To study the physical pendulum – i.e. a composite, rigid body – comparing measured and
calculated values of moments of inertia.
Principle
The moment of inertia is determined by measuring the period for the physical pendulum.
The emphasis is on symmetric mass distributions which places the pendulum’s centre of mass in
the centre of the pendulum. This simplifies the calculations.
The experiment can be extended to arbitrary mass distributions.
Equipment
(See Detailed List of Equipment on the last page)
The physical pendulum 218100 consists of a steel rod with a row of holes which are used both as
pivots and for attaching weights. The fixed part of the bearing is formed by a knife edge which is
attached to stand material or even better clamped to a table edge.
The equipment is provides with 4 disks of steel and 2 of aluminium. These are used in pairs – placed on each side of
the steel rod with a bolt. Hereafter, two disks, a bolt and a nut is termed a weight.
The centre of mass, the pivot and the moment of inertia can be varied in countless ways. The calculations make good
use of the parallel axis theorem and a range of formulas for inertial moments of different parts of the pendulum.
These are found in the Moments of inertia section. Practical calculations are best performed in a spread sheet.
(Two extra sets of nuts and bolts are provided to use as “trimming weights” for use in connection with the Bessel
Pendulum – see experiment 135630-EN. The Bessel Pendulum also uses four washers with the large weights.)
Measuring the Period of the Pendulum With data logging
Select one of the four different methods given below. Place a motion sensor close by the pendulum, prefer-
ably pointing at a weight. (It takes some luck to hit the
With a stopwatch narrow rod, but it can be done.) Adjust the software to
You must measure the time for a number of complete log position with a sample rate of 100 Hz. Check that
swings and divide by the number. Precision increases data follows a sine curve reasonably well – large spikes
if the watch is started and stopped when the pendu- indicate that the sensor misses the target.
lum passes its lowest point where its speed is highest. Measure for “sufficiently long time” – about a minute.
Use a fixed point behind the pendulum as a reference
and don’t move your head between start and stop. Fit the data with a damped harmonic oscillation. Make
sure that the fit parameters are shown with sufficient
Realistically, you cannot hope for uncertainties less number of digits.
than 0.2 s.
If you aim for a 0.5 % precision, the total measuring Depending on the software, you get the period T
time must therefore be at least 40 s. directly, alternatively ω = 2π/T .
Frederiksen Scientific A/S Tel. +45 7524 4966 [email protected]
Viaduktvej 35 · DK-6870 Ølgod Fax +45 7524 6282 www.frederiksen.eu
135610-EN Physical pendulum p. 2/5
With a photogate and a timer With SpeedGate
Let the pendulum hang motionless. Position the SpeedGate (197570) has two light rays; in this experi-
photogate so that the light ray “touches” the edge of ment we use the one marked “X”.
the rod – see photo. A status indicator in the display is active when the light
Photogate 197550 has a green LED that goes off when ray is blocked. With a motionless pendulum, the “X”
the light ray is blocked. light ray should just graze the rod.
With the pendulum swinging (small amplitude!) the Use only small amplitudes: The light ray must be
light ray must be blocked for a complete half period blocked for one half of the period. Select Period and
and pass through for the other half. This way a period Mean Period (not Pendulum Period).
is exactly the time from one blocking to the next.
1. Start the pendulum with a small amplitude
With timer 200250 the procedure is like this: 2. Press Reset
Plug the photogate into DIN socket A. 3. Read the mean period when the chosen
1. Pull the pendulum a little away from the light measuring time expires.
ray during the following points
2. Press Select until the lamp next to Period
turns on
3. Wait until the lamp
Continuous turns on,
then press
Memory/Continuous
4. At last, press Start/Stop
5. Now release the
pendulum
Results are displayed as the
average of two periods – write
down.
Continue for sufficiently long
time. Calculate the mean value.
A bit of theory can be found via the parallel axis theorem (aka the
We will consider a rigid body as composed of a large Huygens–Steiner theorem):
number of small parts.
The moment of inertia of the body with respect to a
Here, M is the mass of the body in question and a is
given axis of rotation is then the sum of the contribu-
the distance between the two axes.
tions from the individual parts, each of the form
∙
This theorem is extremely useful for calculating mo-
ments of inertia, except from the most simple cases.
where mj is the mass of part no. j and rj is the The term “physical pendulum” is used when a rigid
distance from this part to the axis of rotation. body is suspended from an axis that does not go
The total moment of inertia is then given by the sum through its centre of mass.
The period of the physical pendulum is given by
2
In practice, moments of inertia are found by integra-
tion. If the shape of the body is simple enough this can
be done analytically. A number of relevant examples where I is the moment of inertia with respect to the
of this can be found in a later section. axis of rotation, M is the total mass, a is the distance
Let the moment of inertia for a body with respect to from the centre of mass to the axis, and g is the
an axis through the centre of mass of the body be acceleration due to gravity.
called IG . From this, the moment of inertia I , with
respect to an arbitrary axis, parallel to the other one,
135610-EN Physical pendulum p. 3/5
Moments of inertia
It follows from the definition that the total moment of
inertia of a rigid body can be found as the sum of the
moments of inertia of the individual parts of the body
(with respect to the same axis).
In this specific case, we will divide the pendulum rod in
a rectangular section and two semi-circular ends.
(From this we must subtract the material from the 11
quadratic holes, if you want to be very precise.) The
mass of the rod should be distributed among its three
parts proportional to their area.
The large disks are cylindrical with a cylindrical hole in
the middle. Each weight is fixed by a bolt that we will
consider as a point mass. (This means that its moment
of inertia with respect to its own centre of mass is 0.)
Below you find some formulas for moments of inertia.
Rectangle with dimensions b x d and mass m:
∙
12
Semi-circle with radius r and mass m , displaced from
the axis by a distance D :
8
∙
2 3
Cylinder with outer radius r2 , inner radius r1 and
mass m:
∙
2
The practical calculations on the different parts of the
pendulum are best done in a spreadsheet.
Physical pendulum – Procedure 1 – Moment of inertia of the rod alone
For these experiments, you don’t use the washers or Measure the average period over at least 20 periods
the extra bolts. There are still more than enough ways for each of the 5 possible pivots (the middle hole is not
to combine the parts! used here).
Experiments 1, 2 and 3 below are mutually Measure the distances from the centre of the rod to
independent. each of the pivots meticulously. The centre is marked
by a thin line. It may be useful temporarily to extend
The metal disks are fixed by the bolts – it is sufficient this line across the centre hole with a strip of adhesive
to tighten the nut hard with your fingers. tape. (These distances are also used in the following
For simple demonstration experiments, the knife-edge experiments.)
bearing can be mounted in a retort stand with a heavy
“A” foot or a table clamp. 2 – Symmetrical mass distribution
More precise results are achieved by clamping the Chose a position of the iron (black) weights – e.g. hole
bearing directly to a table edge. Preferably directly no. 2 from the ends. The positions must be symme-
above a table leg. And if possible, use a table that is trical, thus keeping the centre of mass identical to the
bolted to the wall. centre of the rod.
To keep the pendulum from “capsizing” when using
Note: There is a position difference when a hole is
the centre hole, let the two bolts point in opposite
used as a pivot (upper corner), resp. is used for placing
directions
a weight (centre). Take care to use the right values in
calculations. Measure the period as an average over at least 20
periods for each of the 5 possible pivots (including the
The amplitude of the oscillations must be small. About
middle one).
half a centimetre is fine.
135610-EN Physical pendulum p. 4/5
If you don’t already have these values: Measure the 2 – Symmetrical mass distribution
distances from the centre of the rod to each of the The formula for the period is re-written as
∙
pivots meticulously. (See 1 above.)
Repeat eventually with the weights in a new – but still 4
symmetrical – position. As the weights are placed symmetrically, the centre of
mass is at the centre of the rod; thus all values on the
3 – Arbitrary mass distribution
left side are known. Right side can be re-written using
We will now no longer demand that the mass
the parallel axis theorem:
∙
distribution is symmetrical. We will therefore have to
determine not only the moment of inertia of the pen-
!"
This means that if the quantity ∙
dulum but also its centre of mass.
#$"
is plotted as
If you don’t already have these values: Measure the
2
distances from the centre of the rod to each of the a function of a the results will lie on a straight line
pivots meticulously. (See 1 above.) with the total mass M as the slope and the moment
of inertia of the pendulum IG (with respect to its
In order to specify positions unambiguously, we define centre of mass) as the intersection with the y axis.
a coordinate axis along the rod with origin at the
centre and positive direction upwards. Positions below This measured value of IG can be compared with one
calculated from theory:
the centre of the rod are negative.
Name the position of the pivot xO . The positions of
2%
% &%
the weights are termed xA resp. xB and their masses where IR is the moment of inertia of the rod with
mA resp. mB . respect to its centre, IW is the moment of inertia of
Carefully write down the experimental conditions and one weight with respect to its centre axis, mW is the
measure the periods as averages over at least 20 mass of a weight (incl. bolt set), and xW is the distance
oscillations. between the centres of the rod and the weights.
IR and IW are calculated using the formulas in the
Calculations section Moments of inertia. For IR , see also 1 –
Moment of inertia of the rod alone. The weights are
1 – Moment of inertia of the rod alone treated as two cylinders (i.e. extended bodies) and a
The formula for the period is re-written as nut and bolt that can be considered as point masses.
∙
If more than one measurement series is performed,
4 the graphs will be parallel.
As the centre of mass is at the centre of the rod, all
values on the left side are known. The total mass M is 3 – Arbitrary mass distribution
simply the mass of the rod mR . For the rod alone, the As the centre of mass of the rod has coordinate 0, the
position of the centre of mass xG reduces to:
' ∙ &'
( ∙ &(
right side can be re-written using the parallel axis
&
∙
theorem:
The distance from the pivot to the centre of mass is
!"
This means that if the quantity ∙ " is plotted &) * &
#$
as a function of a2 the results will lie on a straight line The moment of inertia of the pendulum (with respect
with the mass of the rod mR as the slope and the to the pivot) is found by adding contributions from the
moment of inertia IR (with respect to the centre of rod and the two weights
∙ &)
the rod) as the intersection with the y axis.
'
' ∙ &) * &'
(
( ∙ &) * &(
where IA and IB denotes moments of inertia of
weight A and B with respect to their centre axes.
Now, a theoretical value for the period can be found
For comparison, the moment of inertia of the rod can and compared with the experimental results.
be calculated by dividing it into a rectangular piece
and two semi-circular ends – see the “Moments of Discussion and evaluation
inertia” section. The error of dropping a full treatment In all three experiments it is possible to compare
of the quadratic holes is quite small – but subtract experimental and theoretical values.
their area when distributing the mass of the rod In this treatment of the results, it will be natural to pay
proportional to the areas of the three parts. attention to the experimental uncertainties of the
measured quantities.
135610-EN Physical pendulum p. 5/5
Teacher’s notes Detailed equipment list
Concepts
Centre of mass Specifically for the experiment
– presumed known 218100 Physical pendulum / Bessel-pendulum
Moment of inertia 001510 Clamp
Parallel axis theorem Larger equipment
Period for physical pendulum Option: Timing with SpeedGate
– formulas are given 197570 SpeedGate
Option: Timing with photogate and timer
Mathematical skills 200250 Universal counter/timer
Solving equations, trigonometrical functions,
197550 Photogate
use of spreadsheet, graphs
Option: Timing with a datalogger
In order to deduce the formulas of the moments Motion sensor
of inertia, calculus (integrals) must be known
Logger or link to PC
These experiments generally demand overview
and systematic work Standard lab equipment
(Depending on the timing equipment)
Didactic considerations 001600 Table clamp
002310 Bosshead, square (1-2 are used)
The calculation of the moment of inertia of the rod 000850 Retort stand rod 25 cm
can be simplified by partially ignoring the holes. 000820 Retort stand rod 75 cm
Meaning that we consider the rectangular part of the 000100 Retort stand Base
rod to be homogenous, but slightly lighter than it
would be without holes.
This results in slightly more than 1 % error for the
moment of inertia with respect the rod’s centre.
Calculated periods for the rod alone will be too large
with up to 0.5 %. (Largest for the centre hole.) With
the weights mounted, the deviation will be reduced.
An even more obvious simplification is to consider
nuts and bolts as point masses.
The moment of inertia is then off by less than 0.08 % ,
the period by half of this value.
These errors can presumably be ignored compared to
the measurement uncertainties and other error when
working with the general physical pendulum.
(Note: When used as a reversion pendulum, all of
these errors are completely irrelevant!)
It might seem tempting to expand experiment 2 –
Symmetrical mass distribution to determine the
moment of inertia of the weights with respect to their
symmetry axis. This requires subtraction of two almost
identical quantities and will be subject to large
uncertainties. It can only be recommended if you wish
to treat uncertainty calculations in detail – or else the
deviations from theory will only cause frustrations.
The best you can do is to repeat experiment 2 for all
possible values of xW , and then plot IG as a function
of xW2 .
The intersection with the y axis will then be IR + 2 IW .
On www.frederiksen.eu you can find complete
spreadsheets for determining moments of inertia etc.
Search for item number 218100
