[sdiy] Stupid Sheppard Generator Question

[Chris Crosskey]

Somewhere on my list of things to do I seem to remember that I came up with
a sketch of a Sheppard converter for a normal rising-saw LFO and it was all
analogue...it's a bit of a bodge and was inspired by the sawtooth-chorus
circuit used in the Digisound Waveform Multiplier...basically you take your
basic sawtooth LFO input, use a bank of seven sets of comparators and
voltage shift circuits to get the set of eight 45 degree out of phase
sawtooths all locked to the same frequency, you need to low-pass filter the
waveforms to kill the switching glitch as the compare and shift works on
each one but it's no big deal otherwise...then generate a triangle from your
sawtooth by whatever triangle circuit you happen to favour.....reson why the
heavy filtering on the sawtooth doesn't matter at turnover (if you're
stepping on the switch glith you can bet you're hurting the turnover slope
too) is because at that point in time of that section the triangle is at the
bottom of it's wave and therefore the VCO/VCA combination is at
<quiet>....good thing is that this can be implemented with a single circuit
design repeated eight times, just buld the whole thing around one TL074 per
section....it's fast enough to be the comparator (remember what I said about
the filtering), one section to do the shift and add, one to do the filtering
and one to do the triangle convert....

chrisc
> -----Original Message-----
> From: Magnus Danielson [mailto:cfmd at swipnet.se]
> Sent: 26 September 2002 18:09
> To: thudson at tomy.net
> Cc: pgrenader at mksound.com; synth-diy at dropmix.xs4all.nl
> Subject: Re: [sdiy] Stupid Sheppard Generator Question
> 
> 
> From: Thomas Hudson <thudson at tomy.net>
> Subject: Re: [sdiy] Stupid Sheppard Generator Question
> Date: 25 Sep 2002 16:53:44 -0500
> 
> > On Wed, 2002-09-25 at 10:04, Peter Grenader wrote:
> > > Could one of you guys explain to me what exactly a 
> Sheppard Generator does?
> > > I know, I know...I'm the big know it all ....but this one 
> I must of missed.
> > > They SOUND great (samples on Grant's Page)....but I am 
> not sure what I am
> > > hearing.
> > > 
> > > 
> > > Thanks in advance,
> > > Peter
> > > 
> > Good description here:
> > 
> > http://machines.hyperreal.org/manufacturers/Paia/images/shepard.txt
> 
> The PAIA description and kit seems like a good way to be 
> introduced to it and
> its realization. It wasn't until I saw that description I 
> really grasped it.
> 
> PAIA at least used to have it up on the web. I'm sure you 
> could get it from
> him/them if you ask.
> 
> In essence you have a number of oscillators one octave appart 
> (1 V) in a
> slowly rising (or falling) sawtooth. For each oscillator you 
> have a VCA control
> voltage wich rises and falls (triangle) such that you have 
> full power half the
> way up in pitch. The PAIA kit realises this by use of a 
> counter, XOR gates to
> create the phase variants for each output pair and triangle 
> form for each, and
> then D/A converts this.
> 
> Shockingly simple.
> 
> I once attempted to hack a barberpole in C on my SGI Indy 
> workstation, but
> there was some tiny flaw...
> 
> Cheers,
> Magnus
> 

This e-mail, and any attachment, is confidential. If you have received it in
error, please delete it from your system, do not use or disclose the
information in any way, and notify me immediately.