Oscillations

"Physics is experience, arranged in economical order."

E. Mach

OBJECTIVES
To study some simple oscillatory systems.

THEORY
Typical dictionary definitions of the verb "oscillate" are "to move or travel back and forth
between two points" and "to vary above and below a mean value". Physics uses the word in
essentially the same way, and recognizes a wide range of mechanical and electrical systems that
oscillate. One of the simplest possibilities is a mass connected to a spring in such a way that the
mass can move back and forth if pushed away from an equilibrium position. This system
contains the two essential elements of an oscillator: Something to move back and forth, and some
mechanism to provide a restoring force. Another mechanical example would be a solid bar,
clamped at one end. It will vibrate if struck, but here the moving mass and the spring are the
same material. A capacitor and an inductor provide an electrical realization of exactly the same
equation as the mass and spring.
In this exercise we will study the motion of a mass suspended from a springy material, as
shown schematically in Fig. 1. We assume that the mass can only move vertically, and take the
origin at the point where the spring is not stretched at all. The equation of motion, including both
the gravitational and spring forces, is

!
"

Fig. 1 Idealized mass and spring oscillator.
PHYS 111 Oscillations 2

!
!
"
!#
!
! $ !
%""#
&
(1)

When the mass is not moving it will be at an equilibrium position !
"
given by the solution of

!!"
#
" ! $% (2)

If the mass moving we expect it to oscillate around the equilibrium position, and if we knew the
form of #$!% we might be able to solve Eq. 1 to find a description of the motion.
Physics textbooks usually consider the case of a "simple harmonic oscillator". The simple
harmonic oscillator is constructed with an ideal massless spring, for which the force is a linear
function of the stretch

!!"" ! #" (3)

The constant of proportionality & is called the spring constant. This assumption is made mostly
because it is fairly easy to solve Eq. 1 for this force, although it is also possible to manufacture
springs which are approximately linear. The equation of motion is then


!
!
"
!#
!
! $ !
%
&
" (4)

The solution for the stretch as a function of time is

! ! "!"#$!# ""% " !
$
(5)

where A and ! are constants determined by the initial conditions, !
"
is the equilibrium position,
and " is the angular frequency of oscillation, given by

! ! ! ! " (6)

A derivation of the result is in your text, or you can show that it is correct by substituting ! from
Eq. 5 into Eq. 4.
When #$!% is not so simple it may or may not be possible to write down a general
analytic solution to Eq. 1. We can still make progress by considering small-amplitude
PHYS 111 Oscillations 3
oscillations, for which ! remains close to !
"
. Expanding #$!% in a Taylor series around !
"
gives
the expression

!!"" ! !!"
#
" " $
#
!" ! "
#
" (7)

where we have neglected all terms of the infinite series beyond the first two. The constant &
"
is
given by

!
"
!
#$!%"
#%
%
"
(8)

Substituting this approximation for F(x) into Eq. 1, and simplifying with relation 2, we get the
approximate equation of motion


!
!
"
!#
!
!
$
%
"
%
&
!
$
%
&
" (9)

Since we kept only the linear variation of the force, the equation of motion has exactly the same
form as for simple harmonic motion, and will have the same solution. We therefore expect that
small oscillations about the equilibrium position will occur with frequency

! ! !
"
! # (10)

when the effective spring constant &
"
is found from the slope of the #$!% curve at !
"
according to
Eq. 8.
One final complication is that the system may have friction of some sort. This could be
due to air resistance or to some internal process in the spring material. Whatever the cause, the
conversion of mechanical energy to some other form will cause the oscillations to gradually
decrease in amplitude. For many circumstances, the decrease is exponential, leading to a solution
of the form

! ! "#
!"$
!"#$#$ " $% " !
#
(11)

where ! is a constant determined by the nature of the energy loss. There will also be a decrease in
the oscillation frequency, but it is usually small if the damping is weak enough that oscillation
persists for at least several cycles.
PHYS 111 Oscillations 4

EXPERIMENTAL PROCEDURE
1. Data acquisition and analysis
The overall experimental setup is shown in Fig. 2. A small bucket to hold various weights
hangs from a spring. The spring, in turn, is held by a force probe on a support arm which is
clamped to a lab table. Stretch is measured by a sonic ranger which sits on the floor directly
under the weight bucket, pointing up. A box with a wire mesh window protects the ranger from
falling weights.
Data acquisition is accomplished with LoggerPro software. Start the program by double-
clicking an icon labeled Oscillators.cmbl, or load the file after the program is running. This will
give you a good starting point for this exercise.
To test the process, you can suspend the bucket, loaded with a few hundred grams, from
the helical metal spring. If you displace the bucket vertically and then start data acquisition you
should get a plot of stretch vs time showing decreasing oscillations, and a plot of force vs
position. The plots may become irregular if the sonic ranger is confused by other obstacles, if it
is not under the bucket, or if the bucket swings sideways too much.
Before taking data, calibrate the force probe on the 10N range. It is also convenient to set
the zero of the position plot at the equilibrium position for the mass and spring. Go to Experiment
> Zero... or click on the zero icon on the tool bar, unselect the force probe, and click OK. The
sonic ranger will run very briefly, and the menu will close. The distance plotted is now relative
to the equilibrium position of the bucket.
Once you have a clear and properly-zeroed display of the oscillation, you can do a
quantitative comparison with Eq. 11. Using the Damped sine function
spring
bucket with weights
sonic ranger
under shield
support
bar
lab table
force probe

Fig. 2 Physical arrangement for oscillation measurements.
PHYS 111 Oscillations 5

A*exp(-B*t)*sin(C*t+D)+E (12)

in the automatic curve fit, click Try Fit as usual to let the program optimize the parameters shown.
If the results are unreasonable, edit the parameters to give a more reasonable starting point, click
Automatic, and Try Fit again. When the fit is satisfactory, record the oscillation frequency for later
analysis.
You will also want to measure the slope of the force vs position curve to get an effective
spring constant to compare with your oscillation frequencies. To do this, simply determine the
slope of the force-position graph for each equilibrium position you use.

2. Measurement program
Two springs, with different characteristics, are available. One is a wire helix, for which
the force is supposed to be approximately linear. The other is a large rubber band. You should
determine the oscillation frequencies and effective spring constants of each spring for several
weights, corresponding to a range of !
"
. You can then compare the prediction of Eq. 10 that

!
"
! #!
!
(13)

with the values of #
"
that you measure directly from the force vs position curve. If you choose to
use rather small masses the agreement might be improved by accounting for the mass of the
spring, a portion of which is in motion. The usual claim is that you add 1/3 of the mass of the
spring to the overall mass in Eq. 13.
For the metal spring the maximum stretched length to avoid damage is about 80 cm,
while the rubber band will only stretch about 75 cm before breaking. Without exceeding those
limits, determine the oscillation frequency and effective spring constant for several different
masses for both types of spring.
Your data will probably show that #
"
is nearly constant for the metal helix, but varies
significantly with !
"
for the rubber band. Obtain a force vs position curve for the full range of
positions from unstretched to maximum stretch for each spring, and use the curves to explain the
different behaviors of #
"
.
For the rubber band you will probably also note that the force vs position curve is
different for extension and contraction, a phenomenon called hysteresis. Since the mechanical
work done is different for the two directions of motion, mechanical energy will be converted to
heat over an oscillation cycle, accounting for the strong damping you observed. The metal helix
has little or no hysteresis and correspondingly small damping.