Hurray a question!

[Harry Bissell]

I tried the linearize diodes again in the MS-20 filter (VCF8e) as well as a
bunch of
other limiters in the feedback loop (diodes, zeners, etc...) There seem to be
three distinct
regions of operation...

1) Normal resonance (the usual)
2) Abnormal resonance (the MS-20 is self-oscillating and locking and unlocking
in sync with the input waveform
3) Over-Resonance (the wave has a single overshoot, whose width varies with the
filter
cutoff frequency...

This is a NASTY filter design... I can't stand to listen for very long in modes
2 or 3...
kind of like listening to a ring-mod all night for kicks... Maybe I'm too old
(or too laid back).

Anyway the linearizing diodes do...
Nothing on cutoff frequency...
Little on Feedback... instead of the soft roll you get a hard clip, which semms
to have the effect of killing most of the resonance outright (at high levels) I
presume the clipping is making the feedback basically DC (like not there...)

BTW the Hipass is one of the nicest sounding I've heard...

I have you build a nice smooth filter next... maybe a Moog ???

H^) harry


René Schmitz wrote:

> Hi all,
>
> My point to this is that you *need* some means of limiting in a positive
> feedback loop, to avoid running into the rails. You would get pulse-ish
> oscillations instead of (somewhat distorted) sinusodial with
> selfoscillation. The unlinearized S-shaped response has less AC-gain near
> the saturation than around GND. (The limiting diodes of the MS-20 and MS-50
> filters come to mind.) Gain falls off, and the oscillation is amplitude
> limited.
>
> Bye,
>  René (it could get a little louder here!)
>
> opinionsexpressedher   | uzs159 at uni-bonn.de
> einarethatofmyemploy   | http://www.uni-bonn.de/~uzs159
> erandnotofmyself....   | http://members.xoom.com/Rene_Schmitz
>
>
>
>