Strange modulation schemes & Chaos

Dan Slater dslater at
Thu Jun 26 18:04:23 CEST 1997

Paul Perry wrote:
> >>     John Atwood has this General Radio pink-noise generator from
> >>     the 1950s. It uses a 6D4 thyratron surrounded by a big alnico
> >>     permanent magnet. Apparently the field biases the ionized
> >>     gas, producing substantial noise. It is sensitive to the
> >>     magnetic pole placement versus the thyratron internal structure.
> >>     A bizarre scheme, but it apparently works well.
> I think this scheme is pretty well limited to thyratron or other gas
> discharge probably works because the currents flowing through
> the gas snake around because the magnetic field generates sideways movements
> in the ions/electrons, making new paths.....

	I have built similar musical instruments that use vacuum photocells in
a weak magnetic field. These instruments produce various oscillations
and noises and which are controlled by how the instrument is
illuminated. The weak magnetic field causes the electrons emitted by the
photocathode to spiral apparently using a mechanism similar to that of a
microwave magnetron. Because the magnetic field is much lower that in a
microwave magnetron, the oscillations occur in the audio spectrum. As
the system incorporates significant non-linearities and feedback
(apparently produced by the non uniform photocathode illumination and
the electron cloud interactions), the system is chaotic. The sounds from
this instrument vary from simple sine waves to FM modulation sounds to
white noise depending upon the illumination characteristics (light
pattern, intensity, wavelength, etc., on photocathode). Because of the
low signal output levels and sensitivity to 60 Hz interference (both
electrostatic and modulated light), the instrument usually must be
played outside, away from power line and optical interference. Not much
good as a studio instrument but it is fun to play and produces
interesting sounds.

> those 'imitation candle' neon discharge xmas kitch-0-rama bulbs give a bad
> lfo effect too....

	I have built various circuits using flicker neon lamps. The most
interesting so far have tended to be the simplest, just the lamp and a
load resistor. There is a large lamp to lamp variation in the sounds
produced so selection of lamps is required. The flicker lamp circuits
are useful for both audio and CV production.

this idea of cadmium ldr cells stuck to the front of a TV
> was once described in the Journal of Experimental Musical
> Instruments..someone built a box that sat on the tv and made it look a it
> like an octopus with  corrugated arms coming down to suction cups with light
> dependent resistors which you stuck to the screen...a rare case of a tv with
> something decent on it...
> seriously though, i have had very good results with three lfo's in a ring
> configuration, the output of each going to the frequency control of the
> next. i realized this with the voltage controlled lfo's from a roland system
> 100m, but the oscillators re absolutely non-critical, so long as they are
> not square wave, also you have to set at least one fairly slow with respect
> to the others....only drawback, if it is,  being that you can not possibly
> get exactly the same effect twice.....i suspect that this config is the
> minimum component assembly required to generated truly chaotic (as opposed
> to random) low frequency oscillations.....GO FOR IT!

	You can get very good chaotic sound production by cross coupling 2 FM
modulated oscillators (subsonic or audio). With some practic you can
even get very good white noise out of this patch. The non-linear FM
feedback between oscillators forms a classical chaotic system that shows
period doubling, high sensitivity to initial conditions, etc. For this
to work well, it is necessary that there be sufficient gain in the FM
modulation paths, so some amplification may be necessary in some
instruments. Buchla 258 oscillators, EMS AKS and laboratory function
generators generally work quite well in this patch, Serge and Moog VCOs
do not work as well.

	Even a single oscillator with sufficient FM feedback can produce a
chaotic system. The requirement for the production of a chaotic system
is simple, you must have a non-linear feedback system. The feedback is
from the oscillator output to the FM input. The non-linearity is
produced by the FM circuit. The oscillator output is non-linearily
related to the FM input. The Yamaha DX7 and Buchla 700 used this method
to produce white noise like signals without requiring a separate noise

Dan Slater (dslater at

More information about the Synth-diy mailing list