[sdiy] [AH] Buchla 192 design [was: Buchla used for Silver Apples of the Moon]

[Aaron Lanterman]

On Jun 26, 2010, at 4:58 PM, cheater cheater wrote:

>> Putting some "thing" in the negative feedback loop kind of undoes the effect of that "thing" - there's good discussion of this notion in Horowitz and Hill.
> 
> ah ok, makes sense. Just to be 100% sure, we're talking about the HPF
> being in the negative input feedback loop, right? I've seen designs
> which use both positive and negative feedback.

Yeah.

Positive feedback seems to show up more often in op amps being used as *comparators,* so in such cases your "golden op amp rules" don't apply. The original Hammond VCF patent does use (or at least claims to use) both positive and negative feedback, with the negative feedback having a stronger effect so things stay linear, I suppose... but that seems dicey in general, and the whole thing oscillated like crazy when we tried it on the breadboard. Notably, the final Hammond circuit in the schematics on Bob Weigel's site do *not* have the positive feedback part.

> If I have an HPF in the negative feedback of an op amp with no
> positive feedback (like the hammond filter I assume), will the
> resonance be "normal" in that case, or will it be "wrong"/"out of
> phase" like when you mix an inverted HPF output with the original
> signal in order to create an "LPF"?

I don't really know - you could work out the transfer function and plot it and see. In general, this trick does not give you the same effect as mixing an inverted HPF with the original signal - that's a fairly tricky way to try making an LPF because of the phase issues. In my experience that of thing winds up sounding more like a phaser than an LPF, for instance.

- Aaron