Archive of the former Yahoo!Groups mailing list: MOTM

Hi all.
I am a guitarist and engineer from Champaign, IL with an interest in
analog and guitar synths. My main interest is in the old Roland
GR-300 analog guitar synth which tracks faster than any other guitar
synth out there. Over the past five years or so, I have gone through
periods of interest and activity concerning this synth, figuring out
why it works so well and designing circuits for interfacing it to
external analog equipment and getting more sounds out of it. I worked
on this a bit last year, when I bought my first (and so far only)
MOTM module, an ADSR. The stuff has been sitting untouched for quite
a while, but I'm starting to get excited about working on it again,
especially in light of the appearance on the synthtech web site of
CEM 3391 dynamic waveform processor chips, which can greatly aid in
designing a hugely souped-up GR-300.

Since building all the analog stuff, I have also figured out that the
same scheme can be implemented quite a bit better using the Reaktor
software synth (please, no boos from the diehard analog-heads!), but
it is a processor hog and needs a computer. I still want to finish an
analog version because I already have the thing partially built.

Some of the interface circuits I have designed are as follows:

1. Log(1/x) converter to turn the GR-300's linear pitch (actually
PERIOD) control voltages into 1 volt/octave CV's that other
analog gear can use. I based this on a circuit from Electronotes
and added temperature regulation with a heated 3046 (Paul HATES
this, as I found out when I talked to him last year.).

2. Comparator with envelope-controlled threshold to turn
GR-300's envelope follower outputs into triggers to fire an
ADSR. The envelope helps with note retriggering without reducing
sustain.

3. Circuit to turn the GR-300's weird sawtooth waves (whose amplitude
is proportional to the period of the guitar string) into a pulse
wave with the same duty cycle no matter what note you play.

4. Loudest-string-priority circuit to produce a monophonic output for
diving an aanlog synth like my ARP Axxe. This one needs some work.

Anyway, my idea now is to use the CEM chips to add separate filtering
and envelopes to each string without building insane amounts of
analog hardware. I am interested in hearing from other guitar synth
experimenters out there. I've been somewhat secretive in the past
about this stuff, but what the heck.

Mark