M6.

Lab M6: The Doppler Effect

Introduction

The purpose in this lab is to teach the basic properties of waves (amplitude,
frequency, wavelength, and speed) using the Doppler effect. This effect causes the
frequency of sound waves to be higher for a source that is approaching you and lower for
a source that is moving further away.
Sound is a pressure wave in air. When we hear a sound, we are sensing a small
variation in the pressure of the air near our ear. The speed of a sound wave in air, vsound,
is about 340m/s; sound travels 1 mile in about 5 seconds. This speed depends only on the
properties of the air (temperature, composition, etc.) , not on the frequency or wavelength
of the wave and not on the motion of the source or receiver.
Consider a sinusoidal sound wave in air with frequency f and wavelength λ. The
speed vsound is related to f and λ by

(1) vsound = f λ.

To see where this relation comes from, think:

change in distance
speed = .
change in time

The time it takes for one wavelength of the sound to pass by is the period T, so v = λ/T.
But f = 1 T so v = f λ. Note that as f increases, λ goes down, but the speed v stays the
same. The frequency range of human hearing is about 20 Hz to 20,000 Hz. (The upper
end drops as we age; for people over 60, it is 8 -13 kHz, while dogs can hear up to about
35 kHz.)
When the siren from an ambulance passes by a listener, the listener reports a
change in the pitch of the siren. This is the Doppler effect for sound waves. A source of
sound moving away from the receiver (listener) has a lower apparent frequency (tone or
pitch), while a source moving toward the receiver has a higher frequency. The frequency
shift is approximately proportional to the relative speed of the source and receiver.
To understand the Doppler effect, consider a stationary receiver detecting the
sound from a moving source, a source which is emitting a sound wave with constant
frequency fo. Let us consider the case in which the source is moving directly toward the
stationary receiver with a speed v. In this context, “stationary” and “moving” mean at
rest or not with respect to the air which carries the sound.
The diagram below shows a series of successive wavefronts emitted when the
source was at positions 1, 2, 3, 4, and 5. . The time between successive wavefronts
emitted by the source is the period To = 1 , and during the time To, the source moves a
fo
1
distance ∆x = v ⋅ To = v ⋅ .
fo

Fall 2004
 M6.2

wave 1

wave 2

wave 3
wave 4
1 2 3 4 v

∆x stationary receiver

moving source, speed v, frequencyo f

If the source were not moving, the spherical wavefronts would be concentric, centered on
the stationary source, and the distance between successive wavefronts would be the
wavelength λo. The speed of sound vsound, the wavelength λo , the frequency fo, and
period To are related by equation (1)

λo
v sound = = λ o fo .
To

Because the source is moving, the wavefronts are crowded together on the side near the
receiver, and the distance between wavefronts, on the receiver side, is

v
λ' = λ o − ∆x = λ o − v ⋅ To = λo −
fo
(2)
v ⎛ v ⎞
= λo − λo = λ o ⎜1 − ⎟.
v sound ⎝ v sound ⎠

These wavefronts strike the receiver at intervals of T ' seconds with a frequency f ' = 1' ,
T
'
which is related to vsound and λ by [equation (1) again]

λ'
v sound = = λ' f ' .
'
T

Fall 2004
 M6.3

Solving for f ' , we have

v sound v sound fo (receiver stationary,

(3) f' = = =
λ' ⎛ v ⎞ ⎛ v ⎞ source approaching).
λ o ⎜1 − ⎟ ⎜1 − ⎟
⎝ v sound ⎠ ⎝ v sound ⎠

The frequency is shifted up and the receiver hears a higher pitch. With some algebra, one
can show that the frequency shift ∆f = f ' − f o is given by

⎛ ⎞
⎜ v ⎟
= fo ⋅ ⎜ ⎟
v
(4) ∆f sound
,
⎜ v ⎟
⎜1− ⎟
⎝ v sound ⎠

which, if v << vsound, is approximately

v
(5) ∆f ≅ fo ⋅ .
v sound

From eq’n (3), solving for v sound , we have

⎛ f' ⎞ ⎛ f' ⎞
(6) v sound = v ⎜⎜ ⎟
⎟ = v ⎜⎜ ⎟⎟ .
⎝ f '
− f o⎠ ⎝ ∆f ⎠

Thus, one can determine v sound from measurements of the frequency shift and the speed
of the source.

If the source is receding from, rather than approaching the stationary receiver,
then the (−) sign in eq’n (4) is replaced with a (+) sign, and the frequency is shifted
down. If the source is stationary, and the receiver is moving, then a new derivation is
required and one obtains a different formula (not shown here).

Radar guns use the Doppler effect, but with electromagnetic waves, rather than
sound waves. The policeman’s radar gun acts as both a transmitter and receiver. It emits
a radar signal of known frequency, which is reflected from a moving car, and the
Doppler-shifted reflected signal is detected.

Fall 2004
 M6.4

Experiment
The source is an ultrasonic speaker, which emits a tone at about 40 kHz − well
above the range of human hearing. The source rides on an air track glider, whose speed
is determined with a pair of photogates and a timer. An ultrasonic receiver detects the
sound from the source and the frequency is displayed on a high-precision frequency
counter. Both the transmitter and the receiver signals are displayed on the oscilloscope.
The experiment is to measure the frequency shift as a function of the speed of the source,
and, using these data, to compute the speed of sound.

A schematic of the experimental apparatus is shown below. Begin by leveling the

air table. After turning on the air supply, place the glider midway between the
photogates, and adjust the height of the leveling foot until the glider remains stationary.
Make sure that the blocking card attached to the glider interrupts the photogates. The
timer starts when the card interrupts gate 1, stops when the card interrupts gate 2, and
displays the time interval ∆t . The resolution of the timer can be set to either 1 or 0.1
msec resolution, but with 0.1msec resolution, the timer overflows at only 2 sec; use 1
msec resolution to avoid overflow.
By measuring the distance ∆x between the two photogates, the speed v of the
∆x
glider can be computed from v = . Make sure that nothing interferes with thin wires
∆t
leading to the source, as the glider travels back and forth on the air track.

On the oscilloscope, observe both the signal to the source (on channel 2) and the
signal from the receiver (on channel 1). The ultrasonic transducers used for the source
and receiver resonate at a fixed frequency of around 40 kHz; they do not function well at
frequencies far from 40 kHz. With the source stationary, turn the fine adjust knob on the
sine-wave generator to tune the frequency of the source for maximum signal on the
receiver. After allowing the sine-wave generator several minutes to warm up and
stabilize, record the frequency fo of the source with the frequency meter. With the glider
stationary, there is no Doppler shift, and the frequency of the source and receiver are
identical. Hold the glider stationary and read fo from the receiver frequency meter,
which has a 1 Hz resolution. Check fo frequently; if it is drifting you will have to record
it each time you take data.

Make sure that the receiver is directly in front of the transmitter, at the same
height, etc. so that the source moves directly toward the receiver.

The TTL output of the sine-wave generator is used to externally trigger the
oscilloscope. This ensures that there is always a stable display on the oscilloscope.

Fall 2004
 M6.5

TTL fo
f adjust out

Thin floppy wire sine-wave generator

to source

timer ext
Ch.1 trig
oscilloscope
∆x Ch.2

gate 1 gate 2
ultrasonic
card to trip photogates receiver
glider amplifier

air track transmitter delay

39.890kHz f'
leveling foot HOLD RESET

frequency meter

Procedure

Part 1. Speed of sound from the measured Doppler shift

Your task is to measure the frequency f ' at the receiver and hence the frequency
shift ∆f = f ' − f o , as a function of the speed of the source. From these data, you will
compute the speed of sound, as shown below. Reset the timer to zero and push the glider
toward the receiver so that it passes through gates 1 and 2 at a uniform speed. Record the
frequency f’ on the frequency meter and the photogate time interval ∆t . Repeat this
procedure for several (at least 10) trials, varying the speed of the glider over as large a
range as possible. You cannot let the speed of the glider go much faster than 1 m/sec
(which is pretty fast!) or the frequency counter won’t have time to record the frequency.
The frequency meter works by counting cycles for one second (frequency in Hz =
number of cycles in one second). There is a little red light which is on while the meter is
counting. To prepare the frequency meter to take a reading, make sure the FREQ button
is depressed, turn the DELAY knob to the HOLD position, and then press the HOLD
button. When you are ready to record the frequency, press the RESET button. As soon
as you press RESET, the meter will count cycles (the red light will come on) for 1 sec
and then the frequency will be displayed (in kilohertz, kHz). When taking measurements
at faster speeds, you should press the RESET button immediately after pushing the glider
so that the 1 second count time is up before the glider reaches the end of the track.
On the table near the track there is some sound-absorbing foam material which
serves the important function of preventing reflected sound (echoes) from reaching the
receiver. Make sure the foam covers the side of the track. If the sound absorbers were
not present, then the receiver would receive not only the direct-path sound from the

Fall 2004
 M6.6

transmitter but also various reflected-path echoes. The echoes can interfere destructively
with the direct-path sound causing a loss of signal at the receiver.
transmitter direct path receiver
air track

Table top echo

If the receiver signal drops below about 0.2V peak-to-peak, then the frequency counter
cannot count correctly and the displayed frequency will be too low. Check that the
receiver is getting adequate signal by observing the oscilloscope output as you slide the
transmitter along the whole length of the track. Stop the transmitter at any points along
the track where the receiver signal is unusually small, and check that the frequency
counter is still displaying fo.

Plot your measured frequency shift ∆f = f ' − f o vs. the speed v of the source.
You should get a very straight line, as predicted by eq’n (5). From the appearance of the
graph, decide whether any of the data points should be thrown out. (If one point deviates
substantially from the others, it is usually an indication that a number was recorded
incorrectly.) For each of the good data points, compute the speed of sound from eq’n (6).
Using your several values of vsound , compute the mean, the standard deviation, and the
standard deviation of the mean. (To avoid subscript chaos, we write “vs” for vsound
below.)

∑ ( vsi − vs) 2
σ
N∑ i
vs = 1 vs σ= i
σ mean =
i
N −1 N

Part 2. Comparison with the known speed of sound

The speed of sound in air depends on the temperature according to

(7) v( m / s) = 3315
. + 0.607 ⋅ T

where the temperature T is in degrees Celsius. This expression is only accurate for
temperatures near room temperature. (You can read the temperature from the big dial
thermometer on the wall in the lab). Compare this theoretical vtheory computed from
eqn(7) with your measured vmeas from Part 1 .

Part 3. Calculation of the wavelength of the ultrasonic wave

Fall 2004
 M6.7

From the known frequency fo and your measured vsound from Part 1, compute the
wavelength λ o of the sound. Also compute the uncertainty δλ o .

Part 4. Interferometry and wavelength

In an interferometer, a signal that has taken a direct path from a source to a
receiver is compared with a signal that has taken an alternate path. On channel 1 of the
oscilloscope is the electrical signal directly to the transmitter. On channel 2 is the signal
from the receiver that “hears” the sound after passing through about a meter of air. If this
air path is a whole number (N) of wavelengths, these two signals will be in phase and
will add. If the air path is N + 1/2 wavelengths, the signals will be out of phase and will
tend to cancel. The subtraction will not give exactly zero unless the two signals are
exactly the same amplitude. The time for the electrical signals to travel can be ignored in
comparison with the time for the sound to travel through air.
Adjust the Channel 1 and Channel 2 volts/division knobs on the oscilloscope so
that the two signals are nearly three divisions in height (peak-to-peak). (You may have to
fiddle with the little "cal" knobs in the middle of the bigger knobs.) Above the channel 2
knob there is an ADD-ALT-CHOP switch that you should move to the ADD position.
Now the oscilloscope displays the sum of the direct signal and the signal that has gone
through the air. Move the glider with your hand and notice the change in amplitude of the
summed signals. Switch briefly back to the ALT or CHOP position to see the signals
move from in phase to out of phase. Return to the ADD position and measure with a ruler
the distance that you must move the glider to pass through 10 to 20 maxima or minima.
Divide this distance by the number of maxima or minima to determine the wavelength
from interferometry λint. What is δλint? How does it compare with the wavelength λo
that you calculated in Part 3? Calculate the difference between the two measurements ∆λ
= λint – λo and δ∆λ = [δλint2 + δλo2]1/2.

Fall 2004
 M6.8

PreLab Questions:

1. What is the approximate frequency of the ultrasonic transmitter used in this

experiment? What is the wavelength of sound at that frequency?

2. (Counts as two questions) Four things are measured directly in this lab: ∆x , the
distance between the photogates, f0 , the frequency of the source, ∆ti, the time for the
glider to travel between the photogates (trial i), and f i' , the frequency at the receiver (trial
i). Using Mathcad statements, show how to compute vi , the speed of the glider (trial i),
and v_soundi, the speed of sound (trial i), from these measured quantities.

3. Show that eq’n (6) follows from eq’n (3).

4. Eq’n (4) is an exact expression for ∆f , while eq’n (5) is only approximate. Using
v sound = 345 m / s , v = 0.30 m/s, and fo = 40.0 kHz, compute ∆f using both equations
(4) and (5). By how much do the two values differ? What is the fractional difference?
(Since we want to compare the two values, display several digits for the two ∆f ’s, rather
than rounding to significant figures.)

5. Sketch the graph of ∆f vs. v . What is the approximate slope of this graph? Give a
algebraic expression and a numerical value for the slope.

6. How is it determined whether the air track is level?

7. Assume that the values of vsound , δvsound , fo , and δfo are known. Show how to
compute δλ o , the uncertainty in the wavelength.

8. On Jan. 17, 1930, it was -33F (-36C) in Boulder, with an uncertainty of ±1 oC. What
was the speed of sound outside on that day (include the uncertainty in the speed of
sound)?

9. The air track in this experiment is 2 meters long. If the glider goes faster than 2 m/s,
this experiment won’t work, at least not with the existing equipment. Why not? [Hint:
how is f ' measured?]

10. When the source is receding from a stationary receiver and the speed of the source
approaches the speed of sound, what happens to the Doppler shift ∆f? What happens to
∆f when the source is approaching a stationary receiver and its speed approaches the
speed of sound.

Fall 2004