Experiment No.

Coupled Pendulum
Introduction :
In a periodic system, the number of vibration frequencies is generally equal to the
number of degrees of freedom, which in turn is the minimum number of co-ordinates needed to
completely describe its motion. For example, a single pendulum that is constrained to pivot in one
plane can have its position specified by a single coordinate (usually angular displacement from
the vertical) and has only one natural frequency of vibration. A spring that can pivot around its
attachment point has at least two degrees of freedom and therefore two vibration frequencies. The
most interesting (and useful) examples of this type are systems with several oscillators that are
somehow coupled together. In this experiment, you will study a system that has two gravity
pendula coupled together by a spring.

Tasks:
Two identical gravity pendula are coupled with a soft spiral spring. You have to record
(using the computer interface) displacement of one of them as a function of time for (i) various
conditions of initial displacements, and for (ii) different coupling lengths in both modes.
· The normal mode frequencies are to be determined from such data.
· The relation between normal mode frequencies and coupling lengths are to be established.

Theory and Principle:

You should be able to work out the theory behind the experiment either from energy
considerations or from force-torque considerations. We outline it briefly using the former though
it s equally easy by both methods.

Fig. 1: Schematic of coupled pendula with coupling spring attached at the end.

43
Given two pendula with the same mass ‘m’ and length ‘L’ attached with a spring. When
the pendula are displaced by angles q1 and q2, the masses are at positions (Lsinq1,-Lcosq1)
and (d + Lsinq2, -Lcosq2), where ‘d’ is the horizontal distance between the pendula
attachment points. Let ‘l’ be the coupling length of the pendula (i.e. the distance of the
point, where spring is attached to the pendula, from the point of suspension). Assuming qI
« 1, or by ignoring vertical motion of the masses so that the amount by which the spring
is stretched is given by

Dx = l(sinq 2 - sin q1 ) » l(q 2 - q1 ). (1)

Combining the potential energy of the two masses with that of the springs gives:
1
U = mgh1 + mgh2 + k (Dx) 2
2
1
= -mLg(cos q1 + cos q2) + kl 2 (q 2 - q 1 ) 2 (2)
2
U& = mgL sin q 1q&1 + mgL sin q 2q&2 + kl 2 (q 2 - q 1 )(q&2 - q&1 ). (3)

Using the small angle approximation again in (3):

U& » mgL (q 1q&1 + q 2q&2 ) + k l 2 (q 2 - q 1 )(q&2 - q&1 ). (4)

The kinetic energy of the masses is

K =
2
(
1 2 &2 &2
mL q1 + q 2 ) (5)

(
K& = mL 2 q&1q&&1 + q&2q&&2 . ) (6)

Conservation of energy E& º U& + K& = 0 then gives

[
E& = q&1 mL 2q&&1 + mgL q 1 - kl 2 (q 2 - q 1 ) ]
+ q&2 [mL 2q&&2 + mgL q 2 + kl 2 (q 2 - q 1 ) ] = 0. (7)

Separating this equation gives

mL 2q&&1 = q 1 ( - mgL - k l 2 ) + ( k l 2 )q 2 (8)

mL 2q&&2 = q 1 ( kl 2 ) + q 2 ( - mgL - kl 2 ) . (9)

Solving for q&&1 gives

æ g kl 2 ö kl 2
q&&1 = - q1 çç + ÷ + 2 q2
2 ÷
(10)
è L mL ø mL

44
 kl 2 æ g kl 2 ö
Similarly, q&&2 = q1 - q ç +
2ç
÷
2 ÷
(11)
mL2 è L mL ø

g k æ l2 ö
using w o2 = and wc =
2
ç ÷ the two equations of motion reduce to:
L m çè L2 ÷ø
q&&1 = - (w 02 + w c2 )q 1 + w c2q 2
q&&2 = - (w 02 + w c2 )q 2 + w c2q 1
Here w c is called the coupling frequency.
Adding and subtracting the above two differential equations decouple them and lead to
the two normal-mode equations
(q1 +q 2 &)& = - w02 (q 1 + q 2 )
(q - q &)& = - ( w 2 + 2 w 2 )( q - q )
1 2 0 c 1 2
having normal-mode frequencies w I =w 0 andw II = w 02 + 2w c2 .

Consider the following special cases:

In all cases the pendula are released from rest.

(I). Asymmetric stretch q1 = q2 = qo, q1(t) = q2(t) = qo cos wot

in place (in phase)

(II). Symmetric stretch -q1 = q2 = qo, q1(t) = qo cos wo2 + 2wc2 t

(Out of phase)

(III). Beat case q1 = qo, q2 = 0

The angular displacement of one pendulum is given by

q1(t) = qo cos (w B t ) ´ cos(w III t )
q2(t) = -qo sin (w III t ) ´ sin(w B t )
where the beat frequency w B and the oscillation frequency w III are given by
æ w 2 + 2w 2 - w ö
wB = ç o c o ÷
ç 2 ÷
è ø
æ w 2 + 2w 2 + w ö
w III = ç o c o ÷
ç 2 ÷
è ø

45
Set Up :

The experimental set up consists of two gravity pendula mounted on a rigid

platform. The angular position of one of them can be recorded using rotary
potentiometer. The voltage drop across the variable slider on the potentiometer is
proportional to the angle by which the pendulum is displaced. The computer interface
helps you record this voltage variation as a function of time and displays it on the screen.
The masses must be placed symmetrically about the strip for each pendulum.
The Instructor in charge of this experiment will show you how this data is
obtained and how to read values of particular points of interest from the display. Do not
change any settings or electrical connections without express permission of the instructor.
However you can choose software parameters according to your needs.

46
 Procedure Observations

Ø Find the time period of a single pendulum For a single uncoupled pendu
without any coupling from a record of -lum ,measure the oscillation
displacement as a function of time. time period. T0
Average over several oscillations and do
this many times. Familiarize yourself with
the data acquisition process at this stage.

Ø Make sure that the pendulum is always For the asymmetric mode,
oscillating in the vertical plane as much measure the oscillation time
as possible. Force the pendulum to the required period TI .
amplitude by using your finger tip at about one
third of the length from the suspension.
Attach the coupling spring at the holes closest to the
bob. Displace both pendula by the same amount in
the same direction (this needs care) to obtain initial
conditions corresponding to asymmetric stretch
(in phase) case. The effect of coupling spring would
be seen to be negligible. Though you would notice
damping as in the case of single pendulum. Obtain
the time period for this oscillation from several
measurements. This is one of the two normal modes.

Ø Now create initial conditions for symmetric stretch For the symmetric mode,
by displacing the pendula equal amounts in opposite measure the oscillation
direction (out of phase). Record enough data points time period TII
from the display to obtain the frequency of
oscillations.

Ø Repeat previous step , with the initial conditions such For the beat case ,
that only one of them is displaced (beat case).Relate measure the oscillation time
the frequency information obtained by you to period TIII and the beat
determine normal mode frequencies and coupling semi time period T Beat / 2
frequencies in each of the above case. corresponding to the envelope
oscillation (time interval
between n th and ( n + 1) th
envelope minimum or
maximum )

Ø Repeat the beat case for different coupling Repeat the beat case
lenghts l ie the position at which the for different l, measuring
Coupling spring is attached to the pendula T Beat / 2 in each case and also
Measure TIII in each case. Do it
at least for 4 coupling lengths (l).

47
All data must be recorded in neat tabular form.

The screen displays of displacement as a function of time must be qualitatively

sketched in your report for each case.

Calculations:
1) Calculate the various oscillation frequencies w 0,w I ,w II , w III and w Beat
2) Compare
a) w I with w 0 and comment.
b) w III + w Beat with wII only for one particular coupling length for which you
have calculated TII .
c) w III - w Beat with w I for all coupling lengths..
3) Plot 2w Beat / w 0 or ( 2T0 ) / TBeat vs l.
This should be a straight line if the coupling is weak. [Prove this]
From this plot determine the spring constant k.
4) Compare the two characteristic frequencies w spring = k / m and w 0 = g / L
in this experiment.

5) Calculate D k from D m where m is the slope of the straight line in the graph.

QUESTIONS:

1. Give at least two real life examples of coupled oscillators.

2. How close is your pendulum to a simple pendulum? What differences you expect in
characteristics from a simple pendulum and why?
3. What other dominant modes of vibration, other than the modes you have tried to
measure, are present in the system. Suggest ways of reducing their effect.
4. How can the system you are using for this experiment be improved?

48