[sdiy] Moog ADSR

[Gene Stopp]

Pretty sure that was me!

Minimoogs do have a "climbing" attack peak when Attack = 0 and Sustain <>
max. It's easy to hear on recordings, actually, try some old Rick Wakeman
with Yes or how about side 1 of the album "You" by Gong.

I like it - I've been toying with adding it as a feature to my next
synthesizer (whenever I get to it!). It can be used as a playing style.
Basically, here's a description:

Set up the filter envelope with Attack = 0, Decay = 3 seconds, Sustain = 0.
Set the filter with Frequency = 0, Amount of Contour = 5, Emphasis something
halfway between none and oscillation (just to clearly hear where the cutoff
is). Be sure the amplitude decay is turned up and the DECAY switch is off so
that the notes linger a good long time after you let up on all the keys.

<okay everybody, use your imagination now>

Hit a key, the synth will go "BOW".

Hit a key twice in a row, the synth will go "BEOW".

Hit the same key quickly over and over, the synth will go
"BEEEEEEEOOOOOWWWW".

I frankly don't remember if the Moog 911 does the same thing. I'll try it
someday, probably at the same time I try the 904 range switch experiments
that we talked about a few weeks ago :)

Best Regards,

- Gene



-----Original Message-----
From: jhaible at t-online.de [mailto:jhaible at t-online.de]
Sent: Monday, September 24, 2001 9:03 AM
To: synth-diy at dropmix.xs4all.nl
Cc: analogue at hyperreal.org
Subject: [sdiy] Moog ADSR


Hi,

I remember vaguely that there was some odd behaviour of early
Moog ADS(R) circuits. I *think* it was the modular ADSR (but
it may as well have been the Minimoog). I mean there was a mail
to synth-diy or to AH about it some years ago, but I cannot
find it in the archives.
>From memory: it said you can retrigger the envelopes with
rapid sucessive gate signals, so it would go *above* its
nominal level. 
This seams reasonable, as the level detector (for switching
from Attack phase to Decay phase) is AC coupled to the
Flipflop, and there is a small lag (cap to GND) on the other
Flipflop input. So you could somehow "miss" the peak level
with the right (or wrong) succesion of Gate triggers,
and then it would be stuck on the asymptode level (11.x Volts)
until the Gate is turned off. Does this makes sense ?

I tried a clone, and it behaves like this, but I suspect there
are other (unwanted) side effects in my circuit, too. (It
seems too strong an effect, basically doubling the output
voltage, so I'm in doubt.) Also, when Attack and Decay are
both set to minimum, the amplitude in my circuit is smaller
(like 4V instead of 6V), i.e. the level detector triggering
too early. (Overshoot from fast transient?)
I'm quite sure that most of this is not as it should be, but
then I remember this old (lost) email which described at least
one of these effects.

So - how does the original behave ? Anybody on the list who
could try it ?

Thanks,

JH.