[ARC5] Electret Mikes

[Richard Knoppow]

----- Original Message ----- 
From: "Robert Eleazer" <releazer at earthlink.net>
To: <arc5 at mailman.qth.net>
Sent: Sunday, October 21, 2012 4:05 PM
Subject: [ARC5] Electret Mikes


> Back in the late 80's I bought the plans for a homemade
> aircraft intercom and built one.  It worked great - better
> than any other intercom I have used.  The intercom plans
> also included how to build boom mikes for aircraft use
> employing Radio Shack electret mike elements together with
> a few other parts, including a transistor.
>
> Those mikes worked great and proved to be quite reliable
> as well, with only one failure over a period of 10 years
> or so.  But they were so sensitive that they were useless
> with the intercom.  They turned the intercom "On" at an
> engine RPM of around 1700 and thereafter you got to listen
> to the engine via the intercom.  Even so, I used them
> without the intercom for some years thereafter.
>
> Wayne

     Many microphones for use in aircraft are of the first 
order
differential type. These are essentially bi-directional but
do not have low frequency equalization. As a result they
discriminate against low frequency sounds coming from a
distance of more than about a wavelength.  For close talking
purposes they will roll-off below the point where
diffraction takes over from the differential drive, perhaps
a couple of thousand Hz. In some cases a second order
differential microphone is used. This is essentially two
first order differentials very close to each other and out
of phase. The roll off then is doubled.  This works because
the driving force on the diaphragm of a microphone is
dependent on the difference in pressure on its two sides. In
a non-directional microphone one side is enclosed and
essentially fixed so the diaphragm movement is controlled by
the pressure on one side only. These are known as zero order 
differential microphones.
   In a first order
differential microphone the back side is also exposed to the
sound wave but with a delay.  The delay is caused by the
distance between front and back so there is a difference in 
the two parts of the wave that are compared to actuate the 
diaphragm.  If you picture this you
will see that the force on the diaphragm is dependent on the
wavelength of the sound wave.  The longer the wavelength the
closer in pressure the two sides will be and the less the
force.  Unless equalized in some fashion such a microphone
will have about a 6db/octave roll off below the frequency
where the distance between front and back is equivalent to a
half wavelength.  In a normal broadcast ribbon type
microphone, which is a second order gradient or differential
microphone the pattern is a figure eight since sound coming
from the plane of the ribbon is equal in pressure on both
sides regardless of frequency and the roll off of driving
force is compensated by placing the resonant frequency of
the ribbon below the range of interest.  If there is no
compensation the lows will be rolled off. Now, there is
another important property active here.  That is that the
difference in the pressure between parts of a sound wave
depend on its distance from the source.  As the distance
becomes smaller the sound wave at the microphone becomes
curved.  At a great distance the sound waves are considered
to be plane waves even though they are really still curved.
This is illustrated by the familiar rock dropped into a
pond. Near the rock the waves are definitely circular and
diverge from the point of disturbance.  At a great distance
the curvature is no longer noticeable and the waves appear
to be planar.  Its the same with sound waves.  At the close
distances, where the wave front is noticeably curved the
relationship of pressures within the wave change so that the
pressure difference between segments of the wave increase.
This is because the difference in distance from the source
of different parts of the sound wave are large enough to
cause a pressure difference due to the inverse square law.
When applied to a differential microphone this causes the
drive for  low frequencies to be increased over what it is
for distant waves.  So, when a differential microphone is
used close up the frequency response is flat while for
distant sounds it rolls off significantly for the lower
frequencies.  The point where this begins to happen depends
on the effective path length between the front and rear of
the diaphragm.  For very small microphone elements, such as
electrets, it can be made very short so that the
differential effect begins to be significant at speech range
frequencies.  Such microphones must be used _very_ close to
the source or they will seem deaf to the desired source.
     This long explanation is of why a simple microphone may
seem far to "sensitive" for use in a noisy aircraft
environment.  Its true sensitivity, that is the signal it
produces from the desired source, may not be extremely high
but it can not discriminate against noise coming from a
distance so the desired voice signals will be overwhelmed by
them.
     A simple way to make a second order differential
microphone is to mount two non-directional capsules back to
back and out of phase. Up close only the microphone will
respond to the voice because the waves coming around to the
back will not have the right phase and amplitude
relationship to cancel but sounds coming from a distance
will have. Of course, the overall sensitivity may need to be
adjusted because both the voice and the background sounds
will be of high intensity but this is simply a matter of
adjusting the gain of the amplifiers and perhaps putting
some damping material such as fine silk over the front of
both sides.
     Dr. Harry Olson, of RCA, experimented with high order
differential microphones, some of which are described in the
last edition of his book on acoustics.  Most are difficult
to make work in practice because the balance of energy is so
delicate.


--
Richard Knoppow
Los Angeles
WB6KBL
dickburk at ix.netcom.com