From pete1000@waitrose.com Fri Mar 03 09:07:44 2000
Subject:Re: [Voce] Outboard Chorus for V3/V5

Thanks, Bruce.
The chorus module within the Digitech Studio Quad (II) that I'm using
allows you to give its delay sweep a value of zero, which is how I'd got
it set. However, I've been playing about with flanger modules today, and
am getting happier with the results!!
So far as the brightness of the effect goes, I've just added a crude
(but effective) boost at 4kHz, which seems to improve things when going
through my Rotosphere. I'll try something more sophisticated at a later
date...
Cheers,
Pete W

Bruce Wahler wrote:

> From: "Bruce Wahler"
>
> Pete,
>
> OK, now we're getting into the juicy parts! I've made a "95% plus"
> chorus vibrato simulation on of my Alesis Quadraverb II, and it makes
> a
> world of difference on some songs -- Santana's "Smooth" comes to mind,
>
> as does anything I do by the Young Rascals. Here are the important
> parts:
>
> 1. Use a flanger algorithm, not chorus. Why? Chorus units
> usually delay +/- some amount around a fixed point (ex: sweep
> from 20-40 mSec and back), while a Hammond scanner vibrato
> runs from no delay to full delay and back again. But guess
> what? That's EXAXCTLY how real flanging works: You start
> with two tape decks that are in sync (zero delay), and then
> drag your thumb on the flange of one of the tape reels (hence,
> the name) to increase the delay, then let go and the tape
> tensioner catches back up.
>
> Make sure you use little or no feedback on the flanger block,
> or you'll get too much high end.
>
> 2. To really nail the sound, your EFX unit must be able to produce
> a sine wave LFO, not a triangle wave. The value of the elements
> in the scanner loop were chosen to approximate a sine wave.
>
> Now for the most important part --
>
> 3. The brightness that you speak of is caused, in a roundabout way,
> by the frequency response of the scanner. The rotating scanner
> is a bunch of series inductors with shunt capacitors -- a
> classic low-pass filter.
>
> To compensate for the high-frequency loss of the scanner, the
> folks at Hammond used "pre-emphasis" to boost the highs ahead of
> the vibrato, leaving a mostly flat response when it was all over.
> The fallout of using this approach is that the highs overwhelm
> the lows inside the vibrato circuit, adding brightness.
>
> I simulated this approach by putting a high-pass filter in
> front of the flanger block, and a mirror-image low-pass filter
> after it. It seems to do the trick. I didn't want to go to
> the trouble of trying to figure out the frequency response of
> the scanner -- too many components -- so I used 800Hz as a
> starting point. That's the crossover of a Leslie horn, and I
> knew that the scanner vibrato was Hammond's answer to the
> Leslie speaker -- the very first Leslie clone! Maybe it
> was insight, maybe it was dumb luck, but that value worked
> fine, and I never touched it again.
>
> Here's the overall block diagram of my Quadraverb2 program:
>
> IN ---> HP -> Flange -> LP1 -> LP2 -----> Rev ---> LP3 -> LP4 -> OUT
> | 800Hz 800Hz ^ | ^
> | +12dB -12dB | | |
> \-------------------------------/ \--------/
>
> HP and LP1 are "shelving EQ's" for the the pre-emphasis and
> compensation. In the QV2, the break point defines the START of HP/LP,
>
> not the point where it reaches +/-12dB; other units may use a
> different
> scheme -- the QV2's sibling, the Quadraverb, does -- and so might
> require a different value.
>
> LP2 was needed to counter the drop in the volume of the lowest octave
> or
> so of 16' tones when the chorus is on (phase cancellation, maybe?).
> This is one of those, "I don't know why it helps, but it helps,"
> things.
>
> LP3 and LP4 put a 3dB down point at ~5-6KHz, the same as a Leslie has.
>
> If you're using a Leslie amp (I use a Hafler MOSFET power amp) or the
> PRO-3T (not the -3) you may not want those blocks in there, or you
> might
> want to raise their frequency some.
>
> There's also some on/off (chorus) and level elements (vibrato depth,
> reverb) in the patch that I don't show here.
>
> Best regards (and happy programming),
>
> -BW
>
> --
> Bruce Wahler
> Electrical/Computer engineer
> Keyboard player
> Voce Support Group moderator
> bruce@ashbysolutions.com
>
> *********************************************************************
> The Internet isn't "free."Ý YOU just don't always have to pay for it.
> *********************************************************************
>
> -