Minimoog VCF

[Sean Costello]

Rene Schmitz wrote:

> Unlike the circuit found at Sean Costellos website, this one oscillates
> reliably

Oh, gee, thanks.  I guess NOTHING I do is good enough, huh? Is that what
you are trying to say?  Awww, ol' Sean just went and RUINED CHRISTMAS!! 

OK, total joke.  I didn't know that the version on my site didn't work
reliably (I couldn't get it to work, but I suck at analog stuff). Mind
if I put a link to your schematic on my page, in addition to the old
links?

> though the amplitude of the oscillation rises with frequency. I
> don't know if this is part of the minimoog sound or not? (Any opinions?)
> I think this is due to the high-pass(es) in the resonance loop (the 10uF
> cap with the Q-pot in series, the .22 decoupling caps).

This is definitely a part of the Minimoog sound, at least according to a
very informative post by Michael Caloroso a few years back.  As the
resonance amplitude decreases with lower frequencies, the amount of bass
frequencies cut out below the resonance decreases also.  For lower
cutoff frequencies, not only is the cutoff lower, but the energy below
the cutoff point increases.  This is probably why the Minimoog is such a
great bass synth: for resonant filter sweeps, there will be more
"bottom" with a given amount of resonance than with a filter with
constant resonance amplitude.  There will be lots of resonance for the
attack and decay of the sound (the "zap" or "dmmp" part of the sound, if
you will), but the sustain will have less resonance, and more bass.  The
Moog Taurus' filter had a similar mechanism going on, while the Taurus
II had a constant amplitude resonance filter; the original Taurus is
supposed to be much "fatter," perhaps due to the filter quirks.

Interestingly enough, the standard RESON filter in digital synthesis
languages has a similar quirk.  At frequencies around 1/2 the Nyquist
frequency, the response looks like an ideal bandpass filter.  However,
when the filter cutoff frequency gets closer to 0, the response below
the cutoff point increases so as to let more bass through. Computer
music textbooks treat this as a problem to be solved, but I view it as a
nice feature that makes the filter far more useable for me.

Sean Costello