From obxwindsurf@yahoo.com Fri Oct 03 14:10:35 2003
Subject:Next Generation Vibrato/Chorus Clone

After reviewing a number of design options over the past couple of
weeks (thanks to all who contributed to the discussion), I have
decided on a design and will post a *preliminary* schematic on my web
site in the next week or so before embarking on this project.

As I previously mentioned, Hammond was limited by available
technology based on the era they designed their chorus. Functionally
the Vibrato/Chorus on a Hammond organ has the following attributes:

- tapped phase shift line using inductors and capacitors (this
arrangement served no other purpose than to realize a multi-stage
phase shifter with discrete phase shifts (which were tapped) rather
than a continuously variable phase shift from the phase shifters we
know today)

- discrete phase shifts were selected by the vibrato scanner which
behaved as a sort of switch blending the signal from any given tap
with the next one in line as the scanner's rotor was swinging
through - as the scanner rotated the effective phase shift increased
to the max and back to the min as the scanner made one complete
revolution - it did this 412 times per minute (or so the research has
indicated).

- The V1/C1, V2/C2, V3/C3 switch was used to determine how far the
scanner traversed the phase shift line was used to achieve the
desired effect with the "1" setting using the least length and "3"
used the most (see the B3 schematic I posted this morning for those
who are interested).

- Chorus (the sound we all know and love) was realized by mixing the
dry signal with the phase shifted signal.

- the combination of all of the above produces what we know as chorus.

My design will incorporate the digitally controlled phase shifters of
Dave Amels' circuit, replacing the digi-pots with a binary-weighted 4
resistor serial line and 4 analog switches (in effect, a digitally
controlled resistor) driven off of an up/down counter circuit to
either increase or decrease the resistor in an op-amp phase-lead
configuration by shorting appropriate resistors. This gives 16
discrete phase shifts that when driven by an up/down counter will
scan back and forth as if scanning a multi-tapped delay line. If a
given phase shift is realized by one op-amp, then 2 or 3 times that
phase shift is achieved by cascading the same phase shifter with 2 or
3 identical circuits (as in Dave's circuit which used 4). Each of
the 3 stages' outputs will feed a V1/C1, V2/C2, V3/C3 switch to
determine the amount of chorus or vibrato depth that will be used. A
pot will be available which can be used to adjust the depth of the
chorus effect. It will probably be a trimpot since once the Hammond
chorus sound is achieved you probably don't want to mess with it :)

Individual changes to the digitally controlled resistor (during
scanning) will probably not change instantaneously since a resistor
and capacitor are involved, so I don't suspect switching noise to be
a problem. This will *probably* closely approximate the transfer
function of the scanner as it blends from one tap to the next during
rotation.

The previous discussion on using cross-fading VCAs and how much
overlap is needed to approximate the scanner's transfer function,
while academic, is probably not necessary for this effect based on
the information I have today - Thanks Dave for the innovative design
using discrete scanned phase shifters.

The only issue I might encounter is digital bleedthrough into the
analog signal, but the difference between a digipot and my digital
controlled resistor is that the resistors are not on the same
substrate as the analog switches and may not suffer from the same
behavior. I've used analog switches in similar instrumentation and
industrial control designs before with very little if any noise fed
through.

As soon as I have a schematic I'll post and update the group. Thanks
to all who contributed in any way to the discussion and stimulated my
thinking. Once I've posted the schematic and commenced building,
I'll update with progress and eventually sound samples both dry and
through my Leslie.

Regards,
Kevin