VC ADSR

[Fraser, Colin J]

> -----Original Message-----
> From: macdonald at evenfall.com [mailto:macdonald at evenfall.com]
> Sent: 04 May 1999 17:35
> To: synth-diy at mailhost.bpa.nl
> Cc: Colin.Fraser at scottishpower.plc.uk
> Subject: Re: VC ADSR
> 
> > This VCADSR is almost identical to my own design from a 
> couple of years ago.
> 
> Wow that's great, I would love to see the schematic!

You really don't need to, they are so close. 
I'll see if I have a legible schematic around :-)
The input to the 4052 that selects either the attack or decay/release phase
comes from the flip-flop, the gate input goes to the set input of the
flip-flop, and the comparator on the output resets the flip-flop when the
output reaches the peak for the attack phase.
Aside from that, they are as good as identical. (cue short sample of
Twilight Zone theme...)

> > The inputs to the 4052 also have op-amps in front of each 
> that mix a front
> > panel control with a modulation input for each parameter.
> 
> Good point, my circuit lacks these.  But I had it in mind for use in a
> polyphonic voice card so I didn't mind so much.

If you want to reduce parts count way down, and you plan to use a processor
to control the EGs, you could cut out the 4052 altogether, and have the
processor select the cv for the rate and level of each stage directly. It
would just need to know the state of the gate, and whether the attack peak
was reached.

> I have read the archives on this. Previous circuits I have tried did
> seem to suffer from the offset voltage changes.  This circuit seems to
> perform acceptably to me though.  ???

Mine doesn't noticeably misbehave too much, either. You only notice it under
certain conditions.
If you haven't already, read Juergens explanation at
http://www.it.kth.se/~e93_mda/synths/friends/haible/hj_vc_hadsr.html

> After having experimenting a bit with VCADSR circuits, I see why
> manufacturers went to computer EGs!  :-)

It's just a pity they went to software EGs before the processors were
capable of matching the performance of a good analog design.


Colin f