NetNews Usenet Archive 1992 #31/NN_1992_31.iso/spool/rec/audio/pro/2681

Internet Message Format | 1993-01-02 | 5.0 KB

Path: sparky!uunet!cis.ohio-state.edu!ucbvax!AppleLink.Apple.COM!JOSEPHSON
From: [e-mail] (Josephson Engineering,VCA)
Newsgroups: rec.audio.pro
Subject: Microphonology
Message-ID: <[e-mail]>
Date: 2 Jan 93 22:58:00 GMT
Sender: [e-mail]
Lines: 79

In <[e-mail]>, Scott Dorsey asks (in response to my
small commercial plug):

>(David, do you guys make any decent 200V condenser
>mikes with a cardioid or hypercardioid pattern?)
>--scott

No, we don't, and no one else that I know of does either. Problem is, cardioid
and hypercardioid mics require that the diaphragm be pretty floppy (that means,
high compliance, born of thin diaphragm materials stretched loosely) in order
to have its motion controlled primarily by the resistive damping of the
backplate and the phase-shifted sound arriving from the rear and/or side ports.
Also, it means that the diaphragm-to-backplate spacing must be pretty small.
This is even more critical for a hypercardioid (being further toward the
figure-eight end of the spectrum than toward the omni) than it is for a
cardioid. Resonant frequency for a cardioid or hypercardioid diaphragm is
typically in the 1-6 kHz range; for an omni it's typically 10-20 kHz.

With the diaphragm parameters practical for hypers and cardioids, it typically
means max polarize voltages of 50-60V. You can get higher voltages at the
expense of LF response; if you'd accept -10 dB at 200 Hz, you can go to 90 or
100 volts, and these are very clean sounding mics (I haven't made any recently,
but this type, a KA-400, was the vocal mic used for Shirley Horne's 1990/1
Billboard top-of-the-charts jazz release). If you standardize on a fairly low
polarize voltage (45V +/- 1V in our case) you can reach a lot lower in the LF
by allowing a range of capsules whose diaphragms would stick to the backplate
at voltages much over that.

A comment for those who wonder why this is a big deal, or why Scott would want
such a thing. Higher voltage on the element makes higher output, which means
better signal-to-noise ratio everywhere. Higher diaphragm voltage often makes
subjectively a cleaner sound, due probably to the diaphragm being pulled more
strongly toward the backplate which probably lowers the Q of its tympani-like
resonant modes. (Note in passing that the Bruel & Kjaer high voltage studio
mics (40xx series) are electrets and thus use no polarizing voltage; they offer
a high voltage (130V) version where the HV goes to the preamp only for
increased headroom, so the comment about higher diaphragm tensions does not
apply to them).

Omni mics are easier to make at high standoff voltages because their resonant
frequency is higher and the diaphgram motion is only damped by the airspaces
behind it, rather than also be controlled for directional purposes by sound
arriving from these spaces.

Now here is where the story really starts. After some years of chasing this
technology, and reading texts and papers from several countries, it's amazing
to me how little hard information there is available in print on the art of
microphony.  More than half of the books available on the topic are simply
wrong about major aspects of mic design, having been written by people who have
never taken one apart, much less made one. (See exceptions below.) I've met a
few people at mic companies in the US, Japan and in Europe who know a lot more
than I do on these topics, but the articles they publish don't address these
issues (you only get one guess to figure out why this might be.) I _don't_ know
of anyone doing any research in the .edu community on microphone design. I've
discussed this briefly with the folks at CCRMA at Stanford and CNMAT at
Berkeley, and will be following up with them in the months to come to try to
inspire some work. The technical/engineering problems are daunting enough, but
each mic manufacturer has come up, trial and error, with secret sauces and
snake oils to make things work, converging on the same basic mic designs that
aren't too different from the basic Bell Labs and Nordwest Deutsche Rundfunk
designs of the 30's. There are significant exceptions to the dearth of research
to be found, notably in the design and public release through patent disclosure
of higher order microphone designs, principally at Bell Labs. But these mics
are seldom ideal for wide range music recording, being dependent on arrays that
would just get too big to have the desired qualities over the eleven octaves or
so that we want in our ears. Again, no mystery about this either; there are a
thousand telecommunications microphones for every music microphone in the
world, and some real challenges to make even speech range mics work, for
instance in noisy enviroments for speech recognition.

So, consider a gauntlet thrown down; anyone care to reveal pockets of hidden
microphonology? Any receptive grad programs where folks might want some
problems to tackle?


Cheers...
David Josephson/Josephson Engineering/San Jose CA
[e-mail]