One of the things which set sound effects recordists apart from many other professionals who record sound, is the special requirements we have for the gear we use. Case in point: microphone frequency response. I don't know how many online recording-related discussions have been derailed the moment someone asks about the feasibility of 192 kHz recording and ultrasonic capability of certain microphones as means for better sounding recordings. Countless hapless n00bs have been told off about the limits of human hearing and the dismal frequency response of most speakers and headphones. I usually pull out before it goes all tar-and-feathers; I can no longer be bothered piping up about sound design and pitch manipulation.
Most microphones supposedly useful in pro audio are not spec'ed for ultrasonic capture. Googling the phrase "ultrasonic microphone" mostly yields bat-detectors and components for 3D positioning. Famous exceptions are the Sanken CO100 and the MKH80x0 series. But maybe spec sheets don't tell the whole truth. Enter cheap electret capsules. After reading this article on a DIY solution to ultrasonic recording using Panasonic capsules, I looked at my portable Sony PCM M10 with fresh eyes. Would it's internal mics (supposedly Primo EM 172's) also be capable of ultrasonics? The datasheet isn't much help - it looks like someone took a wild guess with a pencil and a ruler. But then it dawned on me that both Earthworks and DPA use electrets in mics capable of beyond 20kHz recording. What about true condensers like the MKH series?
Ok, so It wasn't the most linear train of thoughts I have ever had, but I basically decided to test my hypothesis that ultrasonics were not a selling point when the original MKH series came out, and so was ignored in the specs. I wanted to know if my MKH60/30 combo might have that much coveted extended response.
Enter the key-jangle test.
A big reason for wanting the really hi-freq stuff is to record metal objects and then pitch those sounds down in post - it's a great way to design huge mechanical sounds that don't sound strangely dull when their pitch is lowered. So for that reason, a good test might be to jangle a few steel keys (and similar objects) in front of the mic:
This was recorded using the MKH60, at 96kHz sampling frequency. Loading it into iZotope RX, we see a lot of high-frequency action going on:
Zooming in, it becomes obvious that those vertical spikes (the keys hitting each other) go all the way up to 40kHz and beyond. Since the brightest areas in this spectrogram correspond to the loudest frequencies, it's clear that most of the energy lies between 8-20kHz, but the sound doesn't exactly drop off dramatically in the frequencies above that:
Here's a clip of the same recording slowed down to a quarter of it's original speed:
Not shown (or heard) in these clips is the response of my MKH30 which was also recording at the time. Being setup in M/S, it was off-axis to the source, which would attenuate the higher frequencies to begin with, but even then there was still quite a bit going on over 20kHz (you'll have to take my word for it).
So if you own a mic from the old MKH series, you can perhaps forget your 80x0 envy for a while; at least as pertains to ultrasonic recording.