ADSR circuit

[gstopp at fibermux.com]

     By some strange cosmic coincidence, I was just testing a new ADSR 
     design of mine in the lab when I got up and came to read my email. (No 
     kidding!)
     
     I've looked at your ADSR, and it is quite understandable. I see that 
     you are using the logic that causes the release cycle to take over if 
     the gate goes away during the attack phase, much like most commercial 
     ADSR types.
     
     Since I have ADSR parameters fresh in my mind, I have a couple of 
     observations:
     
     First, there is a 5K resistor in series with the main capacitor. Is 
     this to limit the current through the analog switches? The presence of 
     this resistor means that all time constants will have a minimum value 
     of around 20 milliseconds. This seems like a long time for a minimum 
     setting. Second, the 500K pots mean that the maximum times will be 
     around 2.3 seconds, which seems like a short time for maximum 
     settings.
     
     To give an example, the envelope generators on a Minimoog have a time 
     constant range from 10 ms to 10 seconds. The shortest attack is nice 
     and sharp, and the longest decay allows you to hit a note and then let 
     it sound for a fairly long time.
     
     The design I am working on today is a dual ADSR that is intended to 
     emulate the Minimoog envelope generators using modern components. I've 
     found that the 555 timer chip has everything that you need on it for 
     the attack logic - the attack flip-flop plus the peak-reached 
     comparator. In addition it has a reset pin so that you can cancel the 
     attack phase when the gate goes away. For the dual ADSR I'm using a 
     556, which is a dual version of the 555. For the switching function, 
     I'm using 2/3 of a 4053 triple SPDT analog switch. I use the SPDT 
     function to either charge the cap or discharge it, since you can only 
     do one or the other at a time. The sustain pot is driven by the gate 
     voltage so that when the gate goes away, the sustain voltage goes away 
     also. This allows you to use the DECAY pot as the RELEASE pot as well, 
     like on the Minimoog. You can add a switch and a diode to short out 
     the cap when the gate goes away also, to emulate the "Decay" switch on 
     the Minimoog. With 1 Meg pots and a 10uF cap, I have measured the 
     envelope times to be a couple of milliseconds on the fast side and ten 
     or twenty seconds on the slow side. One thing I do want to investigate 
     is the need for a current limiting resistor on the analog switch since 
     I'm using such a big cap. If I do add one I'll try to keep the minimum 
     times around 10 milliseconds.
     
     I hope this is the kind of feedback you were looking for!
     
     - Gene
     gstopp at fibermux.com


______________________________ Reply Separator _________________________________
Subject: ADSR circuit
Author:  Jonas Karlsson <jonas.k at linkoping.mail.telia.com> at ccrelayout
Date:    5/1/97 10:21 AM


Hi all!
I have been on this list for a while, but this is my first posting. = 
Anyway, i'm designing my own modular synthesizer at the moment, and last = 
night i got the idea of an ADSR-circuit that seems to work quite well. = 
The schematics are on my homepage, = 
http://www.edu.isy.liu.se/~z93jonka/synt.html.
     
I would appreciate any comments and ideas for improvements. Feel free to = 
use the design i you find it useful.
     
Jonas Karlsson