tube VCF

[René Schmitz]

Hi Ken, and List!

At 21:00 09.01.01 +1100, you wrote:
>I remember a lot of people commented on the tube VCF designed by René
>Schmitz. I'm wondering if anybody else has built one too. 

I'm also wondering, but it seems like we're the only ones...

>I finished the tube portion of mine last night, and no doubt it has some
>peculiarities due to the different tubes and transformer I used. I used some
>small mains transformer with a 1k primary, and who-knows-what ratio for the
>feedback. What is interesting is that I am getting way too much
>feedback/resonance, and while this could be easilly fixed, I find it
>worthwhile to leave as is.

What tubes did you use? 

There is a trimmer in that schematic for this reason, but I ended up maxing
out resonance too. If you know the voltage ratings for the windings in your 
transformer you easily get the turns ratio (n1/n2 = v1/v2). The impedance
looking into the plate is then given by 560Ohms * (turns-ratio)^2. Thats
about 50kOhms for the one I used. Of course this is only valid for mid-band
signals, at low frequencies the primary inductance shorts it out...

>With the resonance low, it behaves much like one would expect a filter to
>behave. As resonance is increased, the filter begins to oscillate at higher
>frequencies though not lower frequencies, due to the high pass nature of the
>feedback, as outlined in René's article. The higher the resonance, the
>greater the range of the self oscillation. Okay, so nothing unexpected so
>far. Self oscillation over the full range of the (pot controlled) frequency
>sweep occurs around half way around the resonance pot rotation.

Hm, I've got about the same, maybe a little less than in yours. 
Most likely the gains of the individual stages V1b and V2 are different
too, so that the grid of V2 sees a different voltage. I'd be interested to
compare these figures.

>Push the resonance pot further around, and run an oscillator into the
>filter, and it is possible to get some chaotic responces, period doubling,
>chirping and other peculiar behavior.

Yes, when going beyond the point of selfoscillation very interesting things
happen.

>So far I have only been using front panel controls to adjust it, though feel
>the behavior is worth exploiting musically.

Indeed! Though I've so far only used an Envelope generator besides the
panel controls. 
Btw: A neat addition would be to switch the filament to V2 on and off. (I
noticed this when swapping tubes in the hot circuit...)

 Bye,
  René
-- 
uzs159 at uni-bonn.de
http://www.uni-bonn.de/~uzs159