Variable log/expo converter

[René Schmitz]

Hi all,

The function y=x^m could be implemented as exp(ln(x)*m) for positive x
only. As a circuit it would be a log-er followed by a VCA, going into a
expo convertor with voltage output. This looks fine in theory, while there
may be problems when x is near zero. (The output of the ln stage would try
to rise to infinity, and be stopped at the positive rail.)

Bye,
 René



At 10:06 22.04.00 -0700, Toby Paddock wrote:
>Is there a fairly simple circuit (don't laugh) that can be adjusted to 
>go from log to linear to exponential? Set by a pot or CV? 
>The tiny bit I understand about log/expo circuits 
>makes me guess NO, but you never know till you ask.
>I was thinking of using it to change the response curve 
>of a non-keyboard CV controller. Change the "feel".
>It would not have to be very accurate, or thermally stable 
>because it would no need to accurately scale a VCO.
>
>Actually, the function y=x^m, where m is adjustable might be better.
>I can see using a multiplier that you put x into both inputs and get x^2.
>Put x into one input and "1" into the other will give an output of x^1 or
>linear.
>An m of 0.5 would be square root, but it's beyond me how to do that 
>or how to make it variable. 
>
>Does this even make sense?
>
>Playing in water over my head,
> - -- -  Toby Paddock
>
>

transconductance | uzs159 at uni-bonn.de
isfutilepreparet | http://www.uni-bonn.de/~uzs159
obeassimilated.. | http://members.xoom.com/Rene_Schmitz