[sdiy] more sh-101 mods?

[harrybissell]

WeAreAs1 at aol.com wrote:

> It seems to me that it should be possible to rig two LDR's into
> three-terminal variable resistor configuration, but probably with
> rapidly diminishing returns in terms of cost effectiveness and
> performance.  It sure would be nice to figure out a kind of universal
> (and cheap) voltage control circuit for filter resonance/feedback that
> doesn't eat up an OTA, though.

Or... does not have the signal distortion, noise, or offset problems of
the OTA.

The problem with 'universal' is the difference in Moog ladder style
feedback
(no feedback = no resonance) and State Variable Filter feedback (no
feedback =
highest resonance)

In either case, a simple vactrol might be used... but in an attenuator
in the Moog
case, and in the feedback in the SVF case.  The SVF case needs to get
down to .707
of the other resistor values (often 10K - 100K) so that part is easy.
High resonance will
come at very high resistance values, which may be hard to control
accurately.  DC
coupling is necessary, no problem for the vactrol.

Moog case will need an attenuator that goes near zero.  You could do
series
(work against a fixed resistor to ground, or actually the large
capacitor that bypasses the bottom of the ladder) or use the vactrol to
shunt the signal and use a fixed series
resistor. Each has its disadvantages, especially because of the AC
couping that is needed.  Probably a buffer stage after the attenuatoir
woulb be a good idea.

I made a circuit for the SVF case using two photocells illuminated by a
single LED.
OK I lied.  They were supposed to be illuminated by a single LAMP...
only at very
low lamp currents, the lag it took the filiment to turn on caused
oscillation.  It was my
FIRST experience with a servo loop... and I tried to stabilize the
circuit with capacitors
(LOL).  Of course the problem was it was too SLOW... so using LEDS (one
poined each way) was the right answer.

In happier days, there were vactrols with two matched independant cells
inside
(you could find them in Mutron III for instance).  This would be ideal.
One cell is used
in feedback to control the LED current, the other is free to be your
variable resistor.
The technique is found in National Semiconductor AN-20 "analog
multiplier".  It works
really well.

You (Mssr. WeAreAs1 :^) made the wish too easy for the Genie... you
should have wished for a really good universal optical coupled
potentiometer... both resonance
circuits are really two terminal (resistor) applications.

In my crystal ball... I see motorized potentiometers on a chip made with
nanotechnology.  Maybe even a "Fender-Rhodes-on-a-chip" with the tines
and pickups micromachined right in the silicon. Do you DOUBT the word of
the
great and powerful OZ ???   There has been a 'nanoguitar' already...
with strings
that vibrated at 500KHz (iirc).

H^) harry  (the great and powerful :^)