In the interest of completeness, I'll share what little else I
happened to think of after posting my first message.

        I have also read about the experiments in "QST" and they do
use AM CW, or at least they did for one test.  The light source was a
laser which was pulsed on and off at some audio frequency.  All the
detector had to do was, you guessed it, detect the light.  The signal
could be fed in to a sensitive amplifier and then to a speaker or
recording device.

        Sony used to make a pair of infrared wireless headphones that
worked great as long as you were within about 30 feet of the
transmitter and had an unobstructed view between you and it.  The left
and right channels were FM and were in the 2 megahertz region.  They
sounded great unless you were almost out of range.  They also worked a
lot better if there wasn't any direct Sunlight shining on the headphones.

        The transmitter was a column of infrared LED's and a few red
ones mixed in so that one could see the transmitter was on.

        One thing to keep in mind is that LED's, lasers and gas
discharge tubes can cycle on and off rather quickly so people have
sent audio and data over them for years as more of an interesting
experiment.

        The trouble with LED's and to an even greater extent lasers,
is that they aren't linear devices at all.  Lasers just quit or go in
to a sort of LED mode if they aren't driven with enough current.
LED's dim out very nicely, but they have the same non-linearity of any
semiconductor device.

        I sort of lied a bit when I said you couldn't modulate
incandescent light bulbs.  While the filament does take enough time
to heat and cool that it can't track a human voice very well, people
have modulated the light from a bulb with devices that either wiggle
back and forth or change the polarization of the light so as to act as
a kind of shutter.  Texas Instruments has a chip with thousands of
little mirrors on it that is the heart of their new DLP or digital
Light Processing systems used in projection TV setups.

        If you really want to mess with a cheap and dirty light
modulator, shine a bright light through a spinning fan.  It'll
certainly flicker off and on as each blade breaks the beam.

        I used to work for our Audio Visual Department, fixing stuff.

        Some of that stuff was 16-millimeter sound projectors.  It
isn't hard for a person who is blind to fix a lot of the transport
problems because the things that mess up the picture often also either
mess up the sound or cause the machine to make bad sounds that
certainly let you know that something is terribly wrong.

        Sound movie projectors have a little stripe of black and clear
wavy lines along one edge of the film.  On a 16-millimeter film, the
audio goes down the edge of the film away from the sprockets.

        The projector usually has a lamp called an exciter lamp which
looks like a slightly-oversized pannel lamp.

        That lamp is a regular incandescent bulb and some of the
better projectors run it on either DC or high-frequency AC, maybe
around 20 to 50 KHZ.  The idea is that light shines from the bulb
through a little cylinder containing a couple of lenses to focus the
light in to a slit-shaped beam that is extremely narrow.

        The beam hits the edge of the film and tries to shine right
through it to a detector which is mounted so that it can see the sound
track, but hopefully, none of the pictures.

        The film can actually have two different types of sound track,
depending upon the equipment the movie company used to expose it.  One
type is called variable density and makes the film darker or lighter
according to the vibrations in the sound.

        The other is called variable width and is completely black.
As the sound varies, the black part of the stripe gets wider and
narrower which lets more or less light through to the detector.

        Either way, the detector sees a light that gets brighter or
dimmer with variations in the sound.  The signal coming out of the
detector is just like the signal coming from a microphone or tape head
or phono pickup.  It contains a somewhat faithful representation of
the sound of the actors voices when they spoke in to the microphone
originally.

        Now, does any of this relate to ham radio?  You bet.  That
modulated light source is an AM signal pure and simple.  When nobody
is talking and there is no sound, what do you think the signal level
is on the film?

        I'm going to let you guess.  Some of you will be surprised,
but think about it.  What does silence actually mean?  Remember that
this is an AC signal.

        Oh yes, the reason that some projectors run the excitor lamp
on DC or high-frequency AC is because an excitor lamp running on AC
still gets a tiny bit brighter and dimmer with each half-cycle.  That
causes a hum at 120 Hertz.  The fancier projectors take the trouble to
make sure there is nothing audible on the power supply to the exciter
lamp.

        So now, let's see how many of you can answer the question as
to what is silence when you have an AM signal?  Think carefully.

Martin McCormick WB5AGZ  Stillwater, OK
OSU Information Technology Division Network Operations Group