Noise floor of dynamic and ribbon mics

[leswatts]

MEMS is micro-silicon, right? So the largest structure is about as small as I can see. My naive thought is that this is too small to approximate human hearing without some significant re-writing of laws of nature.

They are pretty small...fraction of a mm. Most used is probably the Knowles units. Condenser with 4 micron diaphragm/backplate spacing and 11 volt polarization. I extrapolate the self noise at 40-50 dba... pretty bad. But typical of mems. They also have a 20+db peak between 10k and 20kHz.

I don't think you'll see these as studio mics. The noise sources are mostly force noise and head amp noise, and I don't see how that could be improved much. All you can do really is limit the bandwidth.
It would take about 60 of them in parallel to get reasonable noise performance, but they overload at a little over 105 db spl...so it would still be a poor performer.

However they have sold nearly a billion of them since '03. While initially more expensive, they can be pick and place mounted before solder reflow which apparently is a cost saving.

Les
L M Watts Technology

 

[xmvlk]

My dad just bought a 60 year old altec (western electric) 639  ribbon/dynamic microphone and it works perfectly, so it must have microwelded metal/metal contact still.

I do not think microwelding or some kind of contact formation.
When I had tried this I had not succeed.
May be it was kind of chemical preprartion or elsewhere.
I consult it with technology department and they say that
this is not trivial thing.

It is strange storry with ribbon microphon. It was developped firstly
in Dresden laboratory by Gerlach (the acoustician) and Schottki (the physician) in 1920-ies
After some decade (1930-ies) Olson make modification of pole pieces to perform better acoustically.
(For some Radio-Hall, New-York, 1931??? it was discussed somewhere)

Schottki also invented Schottki-diode, maybe due to study contact in the ribbon microphone.
Maybe they had measured noise by some 10 dB higher than nature, maybe they solved
it and Olson have electric part and technological part as solved-
and over the years preparation technique gets unknown.
And here is problem of "noisy" ribbon mics. Maybe Olson does not know this technique
in the 1920-ies and it was one of reason to construct so called "moving-conductor mic" where
ribbon was substituted by copper wire in the center of long diaphragm.

It is too late to answer Olson or Schottki, maybe someone s dad know the right. I do
know nothing, only, that construction of ribbon mic is not so simple how it looks.

 

[PRR]

There was a hearing-aid mike. May have been from Knowles. Tuned to ring at 6KHz, good hearing-aid design. The company across the street got some samples wrung-up to 20KHz. They rang viciously if you lost a little felt damper... ever hear a large PA system take-off at 13KHz feedback? Not spectacularly low-noise neither. Nevertheless bucket-loads of Crown PZM were sold and used.

"force noise"?

Ya, such things have been scaled to conversational adequacy. Cellphones, where you have well over 75dB SPL near the mouth and can tolerate 25dB S/N. Hearing-aids, though not the best sort. Crown had the advantage of "full size" FETs (they didn't come microscopic yet) and ample power so a reasonable noise voltage.

> overload at a little over 105 db spl...

The cheap low-power headamps are limited by Vth of the FET. If you re-rig as a cathode follower with a good negative supply, much higher outputs are possible. And you may be able to pull more current, lowering your noise voltage.

The earliest PZMs had the mike shell floating at some DC voltage; if you got this shorted to ground they shut-up. Crown soon revised the plan, but I have a couple oldies in a coffee-can somewhere.

This is all related to making "good" mikes out of the $2 Panasonic electrets. Within limits, very excellent music recording is possible, especially with head-amp mods. I've also pondered arrays, planar, semi-hemi, and bi-di. But while pondering, the price of quasi-decent condensers fell below the cost of dozens of $2 capsules plus sum/diff amplification.

Exponential horns are impractical, except to boost a top octave. But conical horns can work in sizes similar to the ones used with PZMs when flown away from walls. Consider three 2-foot disks at mutual right angles. You have 8 inside corners. Put a small mike at each corner. This will give 1/8th-space directivity above 300Hz, and 9dB gain above free-space, 6dB above a baffle, cheap S/N improvement! Recovering all 8 signals gives a sound field similar to SoundField(R), though less elegant (and possibly patent-infringing). Using just two octants gives a stereo-pair with very uniform directivity mids to highs. In either case, the bass goes omni and you must bump-up the lost gain. The stereo version might work very well for some recital halls as a fixed installation. A little cheating can bring the inside corners 6" apart, which might improve imagining for slightly off-center source locations.