[sdiy] Experimental VCA RFC

[Jerry Gray-Eskue]

Good point about the dark current. 

The mirrored Optos is an interesting idea. I would rather keep it to a
minimum component count but it needs to be decent in performance.

I guess the question is how many millivolts of feed through does it take to
be a problem? 
Obviously none is best but what is acceptable?

-----Original Message-----
From: Neil Johnson [mailto:neil.johnson97 at ntlworld.com] 
Sent: Friday, March 18, 2011 10:57 AM
To: Jerry Gray-Eskue
Cc: synth-diy at dropmix.xs4all.nl
Subject: Re: [sdiy] Experimental VCA RFC

Jerry,

> The biggest design challenge is keeping the circuit from being effected by
> temperature changes,

According to the datasheet from Vishay the collector-emitter leakage
current increases by 100x over the 20 to 60 degree C range (although
at Vce = 10V).  That leakage will translate into feedthrough when shut
off, which you seem to be seeing in your Spice plots.

Servoing will help some things, but when the LEDs are off you can't
turn off the transistors any more than that.

Perhaps one option would be to have two more optos mirroring the ones
you have, but always 'off', and then run a diff amp between the two
sets of outputs.  That way any leakage due to thermal effects should
cancel out (phase shifts and component tolerances notwithstanding).

Neil
-- 
http://www.njohnson.co.uk