[sdiy] poor man's alternative to a CS-80?

[Sean Costello]

Hi Antti:

Get yerself a copy of Hal Chamberlin's Musical Applications of
Microprocessors right this instant. In it, he shows several digital
derivatives of the SVF, but with various nonlinearities embedded. The
results are very interesting - lots of chaotic filter behavior, but no more
unstable then the standard Chamberlin filter.

Also, the Chamberlin filter tends to behave better in a fixed point
environment with saturation math. I implemented one at work for the Blackfin
(in double precision 1.31 format), and it is much easier to work with than
the floating point one I did for the SHARC, which needs all sorts of
checking of the parameters to be stable.

I have heard of people putting simple compressors in various parts of the
Chamberlin circuit, and getting some very nice behaviors.

Plus, most of the classic state-variable circuits I have seen that use the
CA3080 (such as the Electronotes circuits) use unlinearized OTAs. So you
might want to check your models.

Sean Costello

----- Original Message ----- 
From: "Antti Huovilainen" <ajhuovil at cc.hut.fi>
To: "JH." <jhaible at debitel.net>
Cc: <synth-diy at dropmix.xs4all.nl>
Sent: Thursday, October 20, 2005 11:56 AM
Subject: Re: [sdiy] poor man's alternative to a CS-80?


> On Thu, 20 Oct 2005, JH. wrote:
>
> > In no way would I claim that this covers the actual filter in all
aspects.
> > It does not cover any nonlinearities, for instance. Maybe the actual
circuit
> > is quite linear (what I would guess - but it's just a guess), then it
should
>
> I would assume it is linear or almost linear. I did some simulations of
> SVF using non-linearized OTAs and as soon the nonlinearities became
> significant, the filter went unstable. It is easy to see why:
> Q is inversely proportional to feedback and nonlinearities limit feedback
> for high signal levels -> Q increases with signal level, which is the
> exact opposite of Moog derivatives (moog, ssm, cem etc). The fact that
> CS-80 filter deliberately avoids self-oscillation also points to this
> direction.
>
> Also all the SVF circuits I've personally seen have used linearized
> multipliers (JFET or OTA).
>
> Antti
>
> "No boom today. Boom tomorrow. There's always a boom tomorrow"
>    -- Lt. Cmdr. Ivanova