Expo convertors with tubes ?

[Rene Schmitz]

At 23:19 10.12.99 -0800, you wrote:

>It looks like it's going to be tricky to get the VCO's to do the same thing
>as in traditional synthesizers. Wide-range voltage control of frequency with
>the stability to reliably produce musical intervals seems to be too
>important a function to abandon. Using discrete steps with switched passive
>elements, although highly practical, has the disadvantage of not allowing
>things like interval-tuned multiple oscillators, envelope pitch modulation
>and portamento - three highly distinctive attributes of synthesizer playing.
>It would be a shame, however, to have to resort to a hybrid approach whereby
>the VCO's are solid-state and the rest of the signal chain processes these
>function-generated raw waveforms.

I have shown that at least a hybrid VCO is possible, where the high-voltage 
part is done with tubes alone. However I'm not sure how good/bad it
actually tracks. 
(I've only built one of them, and I've not yet tested tracking it with
another VCO, 
however for a single VCO-voice it seems to be OK.) I'll have to make a
second VCO 
and see how good they track. My personal philosophy is to avoid
semiconductors in the signal path. But there is nothing wrong doing control
circuitry with semiconductors. 

> The PLL concept could even have the potential of
>allowing a small amount of pitch wobble or wander, another way of getting a
>unique sound. 

Thats a great idea.

>The thought of a few tube-created oscillator waveforms tuned in
>unison with some portamento through a warm-sounding VCF... this just might
>be one hell of a killer sound.

What I'd like to try is to use VT-diodes, to get a tuneable LPF, maybe just
2 or three poles, these would probably have a different characteristic than
most other filters. 
Neither linear nor exponential.

>Also, the curves of tube-generated envelopes and LFO's could be interesting.
>Subtle variations between these curves from one kind of synthesizer to
>another do tend to distinguish them apart sonically. The difference between
>simulating an ambulance siren with a slow triangle wave into a VCO, and the
>real thing, is primarily the curve of the pitch rise and fall.

Yes, and perhaps it can trigger something in the SS-domain too. I mean we
usually do linear or exponential envelopes, triangle/sine LFOs , but there
are yet other possibilities, like square, squareroot or something else.

Bye
 René

opinionsexpressedher   | uzs159 at uni-bonn.de
einarethatofmyemploy   | http://www.uni-bonn.de/~uzs159
erandnotofmyself....   | http://members.xoom.com/Rene_Schmitz