Sliders galore

[Doug Ferrell]

Tony,
	Brings back memories of 2-D Walsh-Hadamard transforms in image
processing...  This sounds like a potentially very interesting "oscillator"
module.  32 CV inputs makes me drool (1 freq plus 31 coeff).
	My big question is, what would such a beast actually sound like when the
coeffs are modulated.  I can imagine all kinds of ways to configure the
thing (imagine 31 EGs or LFOs, or sub-harmonic audio-rate modulation of
certain coeffs, or a 16x32 sequencer...).  I'm tempted to write a silly
program to allow simulation of such things with WAV output files.  Anybody
done this already?  Did your 1986 prototype function well enough to produce
output?
	Is it just me, or does it seem like a single Walsh VCO ought to be able to
produce formant-like (as in human speech) output?

Doug (who only has his ASM-1 built into a cardboard box at the moment)
Ferrell

p.s.  The chip count is almost as high as the jack count--scary.


-----Original Message-----
From: owner-synth-diy at mailhost.bpa.nl
[mailto:owner-synth-diy at mailhost.bpa.nl]On Behalf Of Tony Allgood
Sent: Sunday, August 01, 1999 12:51 PM
To: synth diy
Subject: Sliders galore


Question: What consists of three 14" x 4" Veroboard panels stuffed to
the gills in components, has 31 sliders, and generates thousands of
waveforms?

Answer: My Walsh VCO. Basically, I am trying to rebuild/redesign my
third year university project from 1986. Except, this one's going to
work properly. Its the standard set up, HFVCO based around a LM13700 and
precision TC expo, a counter, a load of ex-or gates, a d-latch to get
the waves in sync, a whole heap of op-amps and analogue switches for the
31 voltage controlled amps, and a simple voltage summer to add it all
together. The volume of each Walsh function is controlled by a voltage
generated by a set of 31 sliders. In the future I hope to have this
under software control so I can have dynamic changes in timbre;
wavetable and merge functions all under voltage control.

If people are more interested in this sort of thing then see Neil
Johnson's website, where a block diagram of my (and most others) set up
can be seen.

http://members.xoom.com/Neil_Johnson

I have done all the logic by 74HC ics. Neil prefers programmable logic
arrays. I have no problem with 74HC, since the real chip count comes
from the VCAs. They are not real VCAs, because we can simplify things,
since the waveform is either high or low. Again, Neil's website gives
details of his circuit, which does bear more than a passing resemblance
to mine.

The problem with any Walsh circuit is the huge number of interconnects.
31 of everything. You could reduce the number of functions to say 15,
but you lose waveform resolution. But if the whole lot was on a PCB,
then that would make it a lot easier. So....

If anyone is interested in a PCB for the Walsh VCO, complete with
voltage controlled 'amps' for 31 wals, then give me a shout. If I get
enough interest then I will proceed and lay out a board. I will put the
full circuit up on the archive, if Tony Clark will allow. I guess price
will be about 25 to 30UKP per PCB depending on how many want one.

If there is enough interest I will then do an output board for it as
well. This will contain a 4 pole filter, and output VCA with two ADSRs.
Thus one complete Walsh voice could have two PCBs.

Regards,

Tony Allgood, Cumbria, UK

Rack mounted moog filter and the TB3030 SuperBassline projects:

http://aupe.phys.andrews.edu/diy_archive/schematics/oakley/