> record frequencies up to at least 40kHz
> as insensitive to sound-source direction as possible
> with a fairly flat frequency response
That's an interesting project! With those requirements this effort is going to be an exercise in making compromises. You don't say what your requirements are for noise. That will be key in selecting the appropriate tradeoff between noise performance and directivity.
If you can allow that the microphone only needs to be omnidirectional in the horizontal plane you can orient the microphone pointing upwards and then the response will be the same in every horizontal direction. But if it is even 1/2" in diameter there will be significant roll-off due to the finite diaphragm size. Note that, if you know what the response of your microphone is it will be relatively easy to equalize it to have flat axial response. But there's no way to equalize the difference between the axial response and the off-axis response.
One response might be to use either a 1/4" or 1/8" instrumentation microphone. For reference, a 1/4" microphone has a diffuse-field response that is about 6 dB down at 40 kHz relative to its axial response and an 1/8" microphone has a diffuse-field response that is about 3 dB down at 40 kHz relative to its axial response. But these microphones are really too noisy for ordinary recording activities.
I've used a Knowles FG capsule for such recordings in the past, not the Avisoft microphone but the raw capsule that is used in it. The single sample that I used had an axial response that was up about 5 dB at 12 kHz, back down to 0 dB at 20 kHz, and about -10 dB at 40 kHz. The self noise of the capsule was 24 dB, A weighted which is really quite good considering that the capsule is only 2.5 mm in diameter. I didn't measure the polar patterns or diffuse-field response since such a microphone is clearly good enough for ordinary audio bandwidth recordings. I would expect the performance of the Avisoft product to be similar.
Another alternative would be to use one of the Earthworks omni microphones. These are all 1/4" in diameter, and as such the 90 degree off-axis response is about 3 dB down at 20 kHz relative to on axis, so clearly not as good for your purposes as an 1/8" or 1/10" micrphone. But they are rugged and have good performance.
I hope this helps.
Eric Benjamin
From: Nicol Harper <nicol.harper@xxxxxxxxx>
To: AUDITORY@xxxxxxxxxxxxxxx
Sent: Wed, April 18, 2012 9:09:51 AM
Subject: Ultrasonic natural sound recordings
Hello. I am new to this list.
I am making at set of natural sound recordings. I would like to be able to record frequencies up to at least 40kHz, with a microphone that is as insensitive to sound-source direction as possible and also preferably with with a fairly flat frequency response. I will probably also want to avoid sounds from vibrations and knocks to the microphone during movement of the microphone.
I am currently using a zoom H2 recorder on mono mix with a 96kHz sampling rate, and carrying it carefully or using a tripod.