Tim,
At 8:30 PM +0000 8/26/03, olympus-digest wrote:
>Date: Tue, 26 Aug 2003 00:36:35 -0700 (PDT)
>From: Hughes <hi100@xxxxxxxxxxxxx>
>Subject: [OM] [Exceptionally OT] Condensor mikes
>
>I hate to encourage these really OT threads but having machined and built some
>condensor mic's and
>designed some condensor amplifier electronics, I feel I can comment...
>Joe and others commented
> >>
>These used a 300-volt power supply, to polarize the capacitor microphone
>element, and to power the
>built-in miniature tube preamplifier.
> >>Battery powered mikes were the result of semiconductor advances, that
> >>allowed lower PS voltages. I think the earliest ones were FET, but there is
> >>probably someone more electronically aware that can explain this.
><<
>JFET preamps became common as soon as the JFET availability became good. I was
>using them for
>building my own homemade units in the late 60's and I would guess they were
>already common before
>that in commercial products. The performance of JFET's is so much better than
>vacum devices due
>to not being microphonic themselves,having much lower noise figures and not
>suffering from heater hum.
>Very early on a popular method was to use an RF capacitance bridge circuit
>rather than DC HV
>polarization. This has a number of advantages, for example no 1/f noise,
>closer diaphragm spacing
>(higher sensitivity) being another, along with some disadvantages (RF intermod
>etc) and circuit
>complexity.
I knew about the RF capacitance bridges, and have built a few, but didn't want
to go that deep, and wanted to show the parallel to electret mics.
The RF capacitance bridges I built were of the transformer design, with
phase-sensitive detectors, and were used to track the motions of well-drugged
laboratory rats, in an instrument intended for pharmacological research in the
1970s. As a capacitance bridge it worked very well, but for tracking rats it
was a bust, because the response varied as the inverse cube (not square) of
distance from rat to planar antenna, so nearby tail-twitching looked as large
as real motion of the rat's body.
> >>
>The most basic condensor microphone is a piece of metalized mylar stretched
>over a grounded metal
>ring, with the metallization side against the ring metal. A nearby perforated
>metal plate is
>charged to about 300 volts DC through a large value resistor. (The plate is
>perforated to allow
>air to move through it, so the film can move freely.)
><<
>
>Early professional mikes were always metal diaphragms, not metalized Mylar.
>This is because the
>plastic diaphragm sensitivity can be strongly dependent on humidity and
>temperature. The highest
>quality industrial reference microphones (B&K) are still metal for this
>reason.
I didn't know that all-metal was used. Do you know the material and thickness?
I guess the issue is accuracy versus good sound, accuracy being by far the more
difficult requirement.
>As the electronics
>improved and went solid state the voltages dropped. My microphones ran on 90V
>(with amplifier NF
>of <0.1dB). With a plastic diaphragm there is more of a stability issue since
>the DC polarization
>attracts the diaphragm in a non-linear manner for large displacements, so DC
>voltages dropped for
>that reason as well.
OK. Makes sense.
>The rear plate is actualy usually NOT perforated (except for a pressure
>equalization hole) but contains a special pattern of small blind holes that
>act as a lossy spring
>for damping diaphragm movement.
I think this varies. I've seen mostly designs with holes, perhaps small enough
to give some viscous resistance. But a tuned backplate also makes sense.
>Dual diaphragm cardiod mikes I believe, may have perforations that
>go through to back both diaphragms.
I think they have to, as each diaphram in a cardiod mic needs access to the air
on both sides.
> >>
>Electret microphones have now pretty much displaced all other kinds.
><<
>I believe dynamic microphones still have a following with bands because they
>are very robust and
>"they like the trad sound". National Semi-conductor believes it can replace a
>large percentage of
>the enormous consumer electret market with micromachined integrated
>amplifier-sensors and
>including replacing the discrete JFET amplifier. (The National part is not as
>good ! but they are
>cheaper.) See their website for information. I recently bought 40 electret
>capsules for an
>experimental directional microphone I am building, it is amazing how
>inexpensive they are. (Jameco
>on the web here in the US has good volume prices.)
I still have an old dynamic ribbon mic from the 1960s, but I don't know if it
still works. Probably not, due to atmospheric corrosion. Back then, it was
the best of the common microphone types.
As for the kind of cheap electret mics that National Semiconductor hopes to
displace, they may well succeed, as the market is hundreds of millions of units
a year, and the quality requirement isn't exactly stringent.
Joe Gwinn
< This message was delivered via the Olympus Mailing List >
< For questions, mailto:owner-olympus@xxxxxxxxxxxxxxx >
< Web Page: http://Zuiko.sls.bc.ca/swright/olympuslist.html >
|