SV: Re: [sdiy] Much progress on Buchla waveshaper analysis

[Grant Richter]

According to Buchlas original literature, the non-linear wavefolder  
produces the net
effect of dynamic depth linear FM with tracking oscillators of equal  
frequency.

I've seen Music Easels with both versions, the original dynamic depth  
linear FM circuit with two tracking VCOs
and the wavefolder version, and they both produce identical waveforms  
on the scope.
(maybe they have to be thru zero VCOs though?)

The dual oscillator version gets a lot more "whack" though when the  
oscillator calibration drifts a tiny bit.
Very organic sounds of dying Martian Whales (or Wails).

You can make a great Indian tambura like drone with that wavefolder.
Just tune the oscillator low and slowly cycle the input depth (maybe  
between 50 and 70%) into a medium reverb.

The non-linear transform for the Serge version of the wavefolder is  
stored in the Blacet Mini-Wave at Bank 13 Wave 9 of the factory ROM.
Just take a sine or triangle through a VCA or attenuator and vary the  
amplitude into the Mini-Wave input while monitoring the output.
The Mini-Wave should be set to +/- 5 volts.
All the rest of the Serge wavefolder NLTs are stored in the rest of  
Bank 13 including 12 interpolations between the 4 original curves (16  
in all).

Oscilloscope photos of the Serge non-linear transform curves are at

http://www.musicsynthesizer.com/Circuitry/nlt.htm

The Buchla is similar to the "middle section" curve.


On Nov 20, 2005, at 2:23 PM, karl dalen wrote:

> My 5 cent.
>
> Nice samples,...but.... using 3 cascaded stages of
> the http://www.cgs.synth.net/modules/cgs52_folder.html
> R Lockart nonlinear shaper produces lot of folds
> who gives a more gentle, rounded DX'ish FM to it
> and that to a LOT less circuitry compared to
> the Buchla, but with all those stages shouldent
> the Buchla sound produced have a lot more of folds
> during symetry sweep? It sounds a bit hollow and weak!
>
> KD