AW: Re: PCB for MAT-04 VCA

[gstopp at fibermux.com]

     It looks like this kind of VCA design gets around the problem of input 
     limiting or pre-distortion by modulating the standing current in the 
     differential pairs with the input signal, and then controlling the 
     gain by modulating the bases with the control voltage.
     
     So - some questions for you transistor-theory experts (which I am 
     not!):
     
     Does this mean that the control voltage input is exponential? It 
     appears that way on the scope - most of the effect of rotation of a 
     linear potentiometer providing the CV between +5 and ground, seems to 
     occur in the top half, near maximum gain. By looking at the circuit I 
     suspect that driving the base directly will result in exponential 
     current ratios. It is also interesting to note that although the 
     signal rapidly diminishes to a flat trace with the scope set to 5V/cm 
     vertical, as you "zoom in" downwards to 20mV/cm vertical you can 
     follow the waveform as it gets smaller and smaller, in a very smooth 
     way. (BTW Juergen, perfect waveshape is preserved all the way to 20mV 
     - no distortion at all as the CV changes!)
     
     Is this the reason that this circuit is referred to as a "Voltage 
     Controlled Attenuator", rather than a "VCA" - because the control 
     voltage curve is not intended to follow a precise scale, but yet at 
     the same time allows for clean attenuation? When the CV controls the 
     standing current and the signal modulates the base (in its tiny linear 
     range) you sacrifice low noise for precise and non-drifting control, 
     but when you swap these functions you now have precise waveform 
     preservation and low noise but less control over the CV scale?
     
     I pointed the heat gun at the circuit last night, to raise the 
     temperature of all the components in a (hopefully) uneven manner for 
     worst-case effect, yet no variation in gain or waveshape resulted. 
     This is a good thing, I think.
     
     There are a few points about the existing circuit that need to be 
     addressed if this design is to be used as the basis for a standardized 
     synthesizer module:
     
     1. Input impedances of signal and CV are not 100K, but rather 4.7K for 
     the CV and somewhat less than 30K for the signal.
     
     2. Neither input allows for summing (of course input summers would 
     solve this as well as #1 above).
     
     3. Signal inputs above around 8V p-p do not cleanly limit or clip, but 
     rather get a very nasty distortion as the top of the waveform gets 
     "cut and pasted" into the middle! This makes a sine wave look 
     something like a strange function out of a trigonometry textbook. 
     Maybe useful for something, but in your final signal path it might be 
     rather rude. Some kind of input limiter function may be needed.
     
     4. When the CV travels too far in the negative direction some DC 
     component does affect the output (a simple diode may prevent this).
     
     Chris Crosskey had an interesting thought when he told me that perhaps 
     it would be a good idea to split the VCA into two categories - 
     CV-modulating VCA's, for control of DC control signals and some audio, 
     and audio-only VCA's to allow for clean and noiseless amplitude 
     moduation on the hi-fi side of things.
     
     Well, on to the day's experiments, wherever they may lead me.... 
     things are slow at work these days so I'm taking advantage of the lab 
     bench space. (Thanks for the flattering image of me, Don, I only wish 
     I did have a stereo nearby!)
     
     - Gene
     gstopp at fibermux.com


______________________________ Reply Separator _________________________________
Subject: AW: Re: PCB for MAT-04 VCA
Author:  Haible Juergen <Juergen.Haible at nbgm.siemens.de> at ccrelayout
Date:    2/25/97 2:06 AM