Recording ultrasonic sounds

  • Thread starter Thread starter brun0181
  • Start date Start date
B

brun0181

New member
I realize that this forum is dedicated mostly to audible sound, however, I need expert advice on recording technique and equipment. I figure this is the place to go. I am a graduate student and for my dissertation research I will be recording the echolocation calls of bats. This not something new, many researchers record echolocation calls and analyze them. This is how it works: The calls for the species of bat that I work with are a single call that sweeps from about 20kHz to 80kHz in a matter of miliseconds. A bat detector is used to slow down the call and then the call is recorded on digital audio tape... typically. From the tape, the call is then downloaded onto a computer and there is software for sound analysis. This is what I am wondering: is there technology that would allow me to skip over the digital audio tape? Is there some form of digital recorded that could save the call as a wav file or something? What kind of a sampling rate to these digital recorders have? Then, I will also need to playback some of these calls. I have been looking for loudspeakers which produce ultrasonic sounds, but I can't seem to find one which does not taper off in intensity at very high frequencies (100kHz). Any suggestions?

Any advice, suggestions, etc... would be greatly appreciated. Thanks!

Anja
 
Now that's interesting...

...you say these ultrasonic calls are recorded to digital tape?? I assume you mean a specialized recorder - since for example, your run-of-the-mill DAT machine will have it's frequency response clipped quite significantly via the Nyquist Theorem....

And you would still have a big problem on playback if it were a WAV file with that kind of frequency range...

You may want to ask at rec.audio.pro... there may some specific gear rated to accommodate that kind of response.

Bruce
 
If the bat-detector slows down the calls, then it will shift the frequency spectrum. If the resulting spectrum is between 0 and 20KHz it would be possible. BUT, you say the the freq range is 20kHz to 80kHz... So it would still require sampling at 120 kHz... And this means specializesd equipement.

You could record it directly to a computer using a data-acquisition card with enough bandwidth. You should check electronics-related site for this. The problem you actually have then, is how to get the data in the right format, which can be handled using a fairly simple C program.

And offcourse you need the right mic and signal conditioning before the card.

Playback is also possible using some of these cards.(arbitrary wave generation...) Here the problem is the amplifier and speaker...

It really shouldn't be a problem finding these cards. We've got a hardware AWG (arbit. wave generator) that can 'play' samples at a frequency bigger than 1 GHz at work.

Check what the software that comes with these card allows you, input/output levels, maximum frequency, AC/DC coupling, ... Try to get an engineer in electronics to help you out.
 
Back
Top