Back to main activity.

 Invent an Alien - Doppler activity    
 

General information

This activity is not critical to the Invent and Alien activity, but I think it's interesting for the students and does go into a little more detail on how we can actually find exoplanets.

This description assumes you already are familiar with audio spectrums, spectrograms, the basics of sound, etc.

As such I'm only describing my setup, not how to explain it to the students, etc. I wish I could point to a good source for learning the details from but as my work has involved to the underlying concepts for the past 20+ years I'm afraid I'm hard pressed to think of a good source other than starting in Wikipedia and follow the links out from there. http://en.wikipedia.org/wiki/Spectrogram and http://en.wikipedia.org/wiki/Musical_Acoustics

Building the sound source

The picture below illustrates how I put together the beeper. Everything is taped or hot melted to a short loop of nylon rope. I then have a longer price of rope with a dog clip on the end to clip onto this; this allows using the beeper without a long rope attached.

If for some reason this came apart when you were using it the battery would hurt. A small piece of foam or bubble wrap should go around it on the off chance you slip or Murphy otherwise intrudes on your day. (and yes, I've whacked a number of things with it)

 

The duct tape works pretty well as you can cut it off if you need to change the battery. With typical use the battery will drain itself versus you actually running it down.

 

    
 

General plan

   
 

After the police siren example from the power point I fire up the Spectrogram software. I first just run it in FFT mode, and pick a student to do some whistling, first some low notes, then high, then up and down so they get an idea of the "visual" connection between the sounds.

Here's the settings I used; I increase the lowest frequency to 200 Hz as there's not much below that and it just distracts from things. Note I use a LOG scale for this part as IMHO it looks better.

Here's a typical screen shot:

Once they've played a bit I turn on the beeper so they can see that. Then I switch to a spectrogram mode. Initially with the frequency range about 200 Hz at the low end. Later with the beeper we'll zoom in (linear scale) to just around its center frequency to better see the effect. (with this particular piece of software you can't save settings easily so you should make notes about what to change each time).

Example spectrogram (log scale) of whistling,

Here's the setup for the showing the doppler effect by swinging the beeper in front of the microphone (in this case like a pendulum)

And the results

 

 

    
Notes    
 

 

You have to swing the beeper in a large circle (4'+) and reasonably fast. Of course since you're inside the sounds is bouncing off of everything and the effect isn't as clean as it might otherwise be. The the microphone does a pretty good job of showing it.

One good thing to point out is that it's all "relative" - i.e. I can swing the mic instead of the beeper and get the same results. This can then lead to the one person *on* the train vs. the person by the side of the tracks and each hears something different. Then explain that while this is true for sound at low speeds, it's *not* true for light...and then can lead into doppler effect is only an approximation, etc., just like Newton's theory of gravity vs. Einstein's theories, etc.

Though in the case of 4th graders have to keep it all very basic ... well at least that's what I thought, until a student asked "what if you put the microphone in the center and swing the beeper around it in a circle" (i.e. the mic to beeper distance remains constant). In this case the mic measures a constant frequency, but the kids all hear it changing. This really drove home the "relative" nature of it.

Yet another variation is to swing the mic and the beeper like two pendulums, either in sync or in opposite sync.

You can also play the car horn video and run the spectrogram at the same time, it produces a very obvious result.

 

   
 

 

    

 

 

v1.0