[sdiy] THAT 2181C VCA's

[Craig Critchley]

You can linearize a VCA response by putting a second VCA in feedback loop -
EDN has a short article on doing this with the SSM2164 which also has
exponential response. Tracking is very good, especially if you use two VCAs
on a single chip, but of course you double the number of VCAs you need.

The article may be found here:

http://www.reed-electronics.com/ednmag/index.asp?layout=article&articleid=CA233728&rid=0&rme=0&cfd=1

For VCFs and VCOs, having an exponential response might get rid of needing
an external expo converter, but you'll still have the temperature
coefficient to deal with (at least, that's what it looked like with the
2164.) I'm not sure how good the log conformance is, I haven't tried myself.

Roman Sowa's site has VCFs and VCOs built with the SSM2164 here:

http://www.sowa.synth.net

I would imagine these circuits could be adapted to other audio VCAs.

                ...Craig

----- Original Message ----- 
From: <WeAreAs1 at aol.com>
To: <snd_audio at yahoo.com>; <synth-diy at dropmix.xs4all.nl>
Sent: Monday, November 03, 2003 8:41 PM
Subject: Re: [sdiy] THAT 2181C VCA's


>
> In a message dated 11/3/03 8:06:44 PM, snd_audio at yahoo.com writes:
>
> << Hello,
>  Wondering if anyone here has worked with the THAT
> 2181 VCA's. I have 200+ of these and would like to use
> a couple of them for VCF's and VCO's as well as VCA's.
> Any idea's, suggestions...? >>
>
> Hi Steve,
>
> Isn't the 2181 an exponential-only control response?  That makes it very
easy
> to use it in audio apps such as compressors and volume control, but makes
it
> a little complicated for use in filters, VCO's, or for attenuating control
> signals such as envelopes and LFO's.  Yes, it's true that we want
exponential
> response on our VCF's and VCA's, but we need to have a lot of control over
the
> design parameters of the exponential amp, and if I'm not mistaken, the
2181
> doesn't really give you much access to the expo amp section.  If you want
to use
> it a VCA for attenuating (DC) control signals, it's almost always better
to
> have a linear response VCA for that purpose.  There might be some way to
> "linearize" the control input on those chips (by building some kind of
external
> anti-log amp), but that also seems like a lot of trouble, and I'm not sure
you could
> ever really get the linearity to exactly track.
>
> Maybe you could build a nice analog mixer fader automation system with all
> those chips!
>
> Michael Bacich