home-made vca

[gstopp at fibermux.com]

     The Analog Devices app note for the MAT-04 quad matched NPN hints that 
     the biggest effect of transistor mis-matching in VCA designs is 
     distortion. The specs for their example circuit are as follows:
     
     Frequency response - 20Hz-20KHz flat
     Noise level - 110dB below full output
     Distortion - under 0.03%
     
     The MAT-04 transistors are said to be matched to within 200 microvolts 
     offset voltage of each other, and the Betas within 2%. The CV range of 
     the example circuit looks to be in the negative region but a simple 
     level-shifter can probably move it into a more synthesizer-oriented 
     region.
     
     It seems tempting to try this out, but as long as I spec a part from 
     AD perhaps it might as well be an SSM2018!
     
     - Gene
     gstopp at fibermux.com
     
     p.s. in the same app note, it also says that most available transistor 
     arrays (hinting at the 3046 etc. I think) were more developed to save 
     PCB space than to be matched inside.
     
     p.p.s. how well do the CEM 3330/3360 measure up performance-wise? 
     Anybody have an opinion?


______________________________ Reply Separator _________________________________
Subject: Re: home-made vca
Author:  Joachim Verghese <jocke at netcontrol.fi> at ccrelayout
Date:    2/20/97 6:09 AM
     
I may be wrong, but I don't think the diodes need to be matched 
to the transistor pair unless you want gm to be insensitive to 
temperature changes.
     
> Now the really good thing is that these matched transistors have a much 
> better CV rejection than a 3080, 13700 or the like. You still need an
> offset trimpot, but on a 3080, you still have a remaining "thump", as 
> the offset compensation only works for one specific current. I don't 
> know why, but the discrete transistors are far better here. I can 
> really make the "thump" inaudible by carefully adjusting the trimpot. I 
> don't know what's the reason for this.
> Maybe it's due to the different location of the trimpot (I change the Rc 
> balance, like in the Minimoog VCA, not the Vbe (which is the only method 
> that works on a 3080). Don't know why (anybody knows?) - but it works.
     
Could it have something to do with hfe non-linearity - hfe varying 
with Ic? If the two transistors in a pair exhibit different hfe-Ic 
responses, then there's bound to be problems, especially in current 
mirrors, where hfe is an important parameter.