# How would I use an oscilloscope to measure sound signals? [closed]

I want to detect an audio frequency, something about 10 kHz in order to observe the doppler effect. How would I do it? Do you guys have any recommended reading about it?

## closed as unclear what you're asking by Chris Stratton, winny, RoyC, Dmitry Grigoryev, pipeNov 1 '18 at 10:46

Please clarify your specific problem or add additional details to highlight exactly what you need. As it's currently written, it’s hard to tell exactly what you're asking. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

• 10 MHz is not audio frequency. 10 KHz might be, though quite high in pitch. Or maybe you mean that you want to downmix the frequency shift from a radar to audio, something not uncommonly done ("Audio Doppler"). You may find that a computer sound card is more useful for this than a scope - and usually more available. The biggest problem is that your question is nowhere near clear enough to be answerable. Perhaps you should start with some instructional materials or online videos on basic use of a scope if you have one, or play with audio software and the sound card otherwise. – Chris Stratton Oct 28 '18 at 3:37
• I wrote it wrongly, you're correct. i want to use a frequency generator and detect this generated signal. Mechanicaly i will pull the frequency generator in a certain velocity, observing the doppler effect. – Hugo David Costa Pereira Oct 28 '18 at 3:41
• Look at the lab experiment starting on page M6.4 here. – jonk Oct 28 '18 at 3:53
• A microphone converts sound to electrical waveforms which can possibly be observed on an oscilloscope directly, but would better be amplified first to make them easier to see. – mkeith Oct 28 '18 at 6:22
• Do you merely wish to see doppler effect? Or actually make measurements? Seeing is easier than making measurements. – glen_geek Oct 28 '18 at 15:54

you need a capacitor microphone and some parts on a circuit like this:

You'll need a microphone and a preamp. You can make or buy if one is not available. There are cheap preamp modules available and little boxes. SSM2167 is one preamp chip. Another one (Sparkfun BOB-12758) is easily available (at least in North America) overnight and includes an electret mike.

If your oscilloscope has FFT capability you may be able to directly observe the doppler shift in the frequency domain. Otherwise, your oscilloscope may have a frequency or period measurement function that will ease the measurement of the waveforms coming from your microphone/preamp.

• Or triggering off the source signal may show Doppler as the received signal rolling across the screen. Since the movement will be constrained it will be hard to continue the doppler shift in time, but on a dual trace scope the received signal might be seen to roll in phase and arrive at a new one when movement stops. – Chris Stratton Oct 28 '18 at 11:31

Choosing a high audio frequency is key, since audio doppler effect is small. But the transducers you choose influence optimum frequency, since some have poor response at higher frequency. Selecting appropriate transducers, and having a capable oscilloscope makes this project easy.

## Equipment list:

• Function generator HP3310A (a generic sine/square/triangle function generator)
• PC dynamic loudspeaker (ripped from an old PC 2.5" diameter)
• Passive Piezo audio transducer
• Oscilloscope (Hantek DSO5202P)

The piezo on hand is not optimum (not quite as shown below). Its plastic shell resonates at 3.596 kHz, which is where this experiment was conducted. The piezo was connected directly to one oscilloscope input, and provided ample signal amplitude. Its resonance adequately filtered out room noise:
Function generator drives the electrodynamic loudspeaker directly with a sine wave (annoyingly loud) at the resonant frequency of the piezo transducer: 3.596 kHz. The piezo provides adequate voltage amplitude to make measurements on the oscilloscope, even at 1 meter separation from the loudspeaker. The piezo is small enough that it can be moved closer or further from the loudspeaker - oscilloscope probe lead is long enough and flexible enough as well.

simulate this circuit – Schematic created using CircuitLab
Oscilloscope set-up is key to seeing, or measuring the doppler effect. The change in frequency is not large as the piezo moves. Trigger level was set to DC, and carefully adjusted to 0V, where the sine wave crosses the axis. AC trigger is probably good too. Trigger source was from the piezo (CH2).
To see the small change in wavelength that doppler produces, sweep speed was set to 20 us/div. This gives decent time resolution.
Now here's the critical part: This oscilloscope allows display of the waveform well after the trigger point. Horizontal position was set so that the trigger point was 10.00 milliseconds before the display center. That means nearly 36 cycles are not displayed (they are invisible, off to the left side) in the following plot. Waveform display persistence has been set to one second, so that variations of the sinewave period are visible. This is really not necessary, you may be able to judge these variations by eye:

If the piezo is held motionless 20 cm away from the speaker, a single sinewave is visible, with no variation in period. As the piezo is moved closer or further, the period changes (shorter, or longer). This oscilloscope plot shows period variation of about 28 microseconds. This resulted from moving the piezo an estimated +/- 0.2 m/s. That's due to doppler. (My ears are still ringing).

OK, experiment done - let's calculate. Moving the piezo receiver caused a period shift of 28 microseconds after 10,000 microseconds from the trigger point. That's due to moving closer in addition to moving away. That's one part in $$\2.8 x 10^{-3} \$$.
Speed of sound in air 343 m/s.
Speed difference of piezo (moving in + moving out) = $$\ 343 * 2.8 x 10^{-3} \$$
Speed of piezo = $$\ \frac{0.96} {2} \$$ meters per second, assuming speed in = speed out. So my estimate of 0.2 m/s was too slow.
Note that the frequency of 3595 Hz only changes by a small amount. It is very hard to see, if you're only looking at one cycle. The display time offset provided by the 10 milliseconds of pre-trigger makes a measurement possible.

You can use a mixer, homodyning the return signal (which has some tiny Doppler shift) with the transmitted signal.

Assuming the Mixer output is DCcoupled (the NE602 active mixer is DCcoupled), you will notice some change in the (differential) output voltages as your target moves.

If you have large Dopper shift, you will see beatnotes from the Mixer, not just shifts in voltage.