Article on Parabolic Microphone Gain (1 Viewer)

[Lerxst]

I wrote this last month and forgot about it. Simply meant to be a reference for anyone bored enough to read it. Some years ago I was looking for papers on this and only found one that really touched on the topic (it is cited) in a non-trivial way. But it was also pretty dense at points and hard to follow. Moreover, I wanted an excuse to do some animations/simulations. If you do read this and have comments for improving it, I am all ears. It is probably way too long.

The Physics of Parabolic Microphones: Frequency Dependence of Gain - Michael Hurben, PhD

Introduction Parabolic microphones are known for their extreme sensitivity, and the origin of their acuity isn't difficult to guess. It is the most obvious thing about them, which can also make them a liability for field use, namely, their considerable size. Just as a large amount of weak light...

legallyblindbirding.net

 

[Jon.Bryant]

I will read the article more thoroughly, but note that in the introduction you mention how parabolas provide more gain at higher frequencies as if this is a negative, and that recordings are more tinny. This is true, but one of the counters to this argument, is that as higher frequency sound attenuates more quickly than lower frequency sound, and that this attenuation is inverse to the gain provided by a parabola. Therefore the unequal gain of the parabola can be beneficial by 'recreating nearness'. Under this argument, the 'tinny' sound that some people don't like, is then the natural sound of the bird when up close and personal - we then get into the argument of whether we want a 'human ear' or 'birds ear' recording.

I find it interesting that if we are in the rain forest and point a shotgun at a bird singing in the canopy 30m overhead, physics dictate that it will always have the frequency balance as if it is 30m away, no matter how much gain we apply or how loud we play the recording. If we want to recreate nearness, and hear what the bird would sound like, if we were perched on a branch close by, then we need to either apply equalizer or alternatively use a parabola.

I recall that attenuation of higher frequencies is temperature and humidity dependent. Frequency gain from a parabola is fixed for a particular dish. I have therefore been thinking for a while of doing some approximate math to work our the 'impact' of a parabola at different distances, and weather conditions - i.e. what is the apparent shortening of distance to the target, that is created when using a dish? I also think that there must be a minimum distance we should use a parabola, so that you are only recreating nearness, as opposed to over emphasizing the high frequencies. However, as the main reason for using a parabola is for distant work, a 'minimum distance' may not be that important (and I suppose you can always try to fix the an over tinny recording by applying equaliser in post).

 

[Lerxst]

I will read the article more thoroughly, but note that in the introduction you mention how parabolas provide more gain at higher frequencies as if this is a negative, and that recordings are more tinny. This is true, but one of the counters to this argument, is that as higher frequency sound attenuates more quickly than lower frequency sound, and that this attenuation is inverse to the gain provided by a parabola. Therefore the unequal gain of the parabola can be beneficial by 'recreating nearness'. Under this argument, the 'tinny' sound that some people don't like, is then the natural sound of the bird when up close and personal - we then get into the argument of whether we want a 'human ear' or 'birds ear' recording.

I find it interesting that if we are in the rain forest and point a shotgun at a bird singing in the canopy 30m overhead, physics dictate that it will always have the frequency balance as if it is 30m away, no matter how much gain we apply or how loud we play the recording. If we want to recreate nearness, and hear what the bird would sound like, if we were perched on a branch close by, then we need to either apply equalizer or alternatively use a parabola.

I recall that attenuation of higher frequencies is temperature and humidity dependent. Frequency gain from a parabola is fixed for a particular dish. I have therefore been thinking for a while of doing some approximate math to work our the 'impact' of a parabola at different distances, and weather conditions - i.e. what is the apparent shortening of distance to the target, that is created when using a dish? I also think that there must be a minimum distance we should use a parabola, so that you are only recreating nearness, as opposed to over emphasizing the high frequencies. However, as the main reason for using a parabola is for distant work, a 'minimum distance' may not be that important (and I suppose you can always try to fix the an over tinny recording by applying equaliser in post).

Thank you Jon, those are all very good points, and I will work to incorporate discussion of them into my rev2.

Myself I like the "tinniness" because at my age my high frequency content is disappearing. And even the best ears are not immune to the nonlinearity of the Fletcher Munson curves! Years ago I played with equalization, just for fun, that tried to undo those effects, in order to hear what sounds "really sound like" without having to go through the very imperfect filtering and distortion innate to human ears. This has always fascinated me.