<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META content="text/html; charset=us-ascii" http-equiv=Content-Type>
<META name=GENERATOR content="MSHTML 8.00.6001.23562"></HEAD>
<BODY>
<DIV dir=ltr align=left><SPAN><SPAN class=819221521-11082024> >>
</SPAN>Each ADSR requires one 2164</SPAN></DIV>
<DIV dir=ltr>
<DIV class=gmail_quote>
<DIV><BR></DIV>
<DIV><SPAN class=819221521-11082024> > </SPAN>Wouldn't one VCA per
envelope generator be enough? Depends on the design, of course.</DIV>
<DIV> </DIV>
<DIV><SPAN class=819221521-11082024>No, because all of the controls must be
voltage controlled. I tried switching the A, D and R voltages to a single
VCA, but it was not practical. What I ended up doing was using separate
VCAs for A, D and R. The control voltages from the front panel are locally
buffered and sent through three channels of a 4066 analog switch. When
each switch is off, the voltage at that VCA's control pin is pulled up to about
+5V by a 10k resistor to the 15V rail. This effectively turns that VCA
off. All three VCAs are in the feedback loop of the EG integrator
(remember, 2164 VCAs must terminate to virtual ground). Believe me, I
tried many different ways to do this, and this was the only way that worked
perfectly. The fourth VCA of the 2164 is a linearized VCA (the log amp is
on the front panel) which controls the amount of the envelope being sent to its
destination.</SPAN></DIV>
<DIV><SPAN><SPAN class=819221521-11082024> </SPAN></SPAN></DIV>
<DIV><SPAN><SPAN class=819221521-11082024>>> </SPAN>Will there be a
problem mixing 48 key outputs together? Will clipping be a problem?
Obviously, all 48 keys will never be played at the same time, but
still.</SPAN></DIV>
<DIV><BR></DIV>
<DIV><SPAN class=819221521-11082024> > </SPAN>Gainstaging a synth
with lots of voices is a nontrivial problem. But at least it's just one gain
factor that needs to be decided upon..</DIV>
<DIV> </DIV>
<DIV><SPAN class=819221521-11082024>I may have to come back here for some advice
when I get to the point of adding all of the voices together. Right now,
I'm just thinking of summing them all at unity gain at a single opamp (i.e., 48
10k resistors into the inverting input, and a 10k feedback resistor).
Maybe I'll have to make sure that the voicecard outputs are suitably attenuated
so that I can add many of them together without fear of clipping. This is
something about which I haven't given much thought yet. How does
an electronic organ manage this problem?</SPAN></DIV></DIV></DIV></BODY></HTML>