[sdiy] Analog division

[Olivier Gillet]

This is also what I was thinking about... If you look at the upper
part of the schematics in the Mike Irwin article, you'll see that one
of the nodes of the circuit outputs a voltage proportional to log(V_in
/ Vcc).

Using exp ( log a - log b ) = a / b, maybe it would be worth trying
using two servo VCA+Op-amp cells like that, one of them into an
inverter, then a summer the result into the CV input (or are the CV
input of the SSM2164 summing nodes?)  of another SSM2164 with a
constant input - this VCA playing this time the role of an exponential
current source. And a final op-amp to convert the output current into
a voltage.

Would this work on four quadrants? Probably not, that's the tricky part!

Straightforward, random, and probably wrong idea :D

On Mon, Jul 19, 2010 at 7:50 PM, David G. Dixon
<dixon at interchange.ubc.ca> wrote:
>> I'm thinking of the SSM2164 as well... it looks like it could go
>> in an opamp feedback loop.
>
> That's how we "linearize" it, actually.
>
> I've used two linearized 2164 VCAs (which means all four VCAs on the chip)
> to make a very nice four-quadrant multiplier.  It's very straightforward.
> I'm not sure about division -- never given it much thought I'm afraid.
>
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>