Resonance Pot Scaling

jh jhaible at primus-online.de
Tue May 11 22:44:02 CEST 1999



-----Original Message-----
From:	Martin Czech [SMTP:martin.czech at intermetall.de]
Sent:	Tuesday, May 11, 1999 6:17 PM
To:	synth-diy at mailhost.bpa.nl
Subject:	Resonance Pot Scaling


> Comments ?

Yes.
I remember a discussion with Graham Hinton who claimed that
this reverse log pot stuff was all nonsense, and that an ordinary
log pot were the best solution. I don't agree, and I generally think it's
really a matter of taste, and probably a matter of the filter circuit
in question, too. If you have a circuit that will break into wild
self oscillation without soft amplitude limiting, you will naturally need
the best pot resolution near this point where the oscillation starts,
i.e. a reverse log pot, or a linear one with a bypass resistor to the
cw end. This is not *necessary*, of course, but it can make adjustments
easier.
personally, I've tried the following configuarations, and I am quite
satisfied with the results:

Linear pot on SVF (yes, log and reverse log would change places here,
anyway !).

Linear 10k pot with 10k bypass to cw end for a SSM2040 filter (actually
controling a linear 2020 VCA, but this makes no difference)

Linear pot without bypass resistor controlling linear gm cell for 2040-clone.

Reverse log variable resistor (no pot configuration!) in feedback path of Moog ladder. (50k -log)

10k lin pot plus 2.2k bypass resistor to cw end in Sallen / Key filter (MS-20).
But careful, there are othe resistors here that will reduce the effect of the 2k2.

Linear pot for 6-pole filter

JFET (that would be something like x**2, wouldn't it ?) as lower leg of
a feedback divider in diode ladder filter (VCS-3-clone).

Bottom line: While I still haven't tried Graham's rather extreme suggestion,
I am rather content with the solutions described above: Either straight linear,
or (more often) stretching of the upper range (rev log or bypass). In one case,
the Moog ladder, I was not satisfied with a linear pot or bypass resistor
solution at all; the 50k -log pot is just about right.

JH.

 






More information about the Synth-diy mailing list