[sdiy] CV inputs with bipolar "amount" knobs

[Mattias Rickardsson]

On 27 July 2010 08:34, Matthias Herrmann-DEU
<Matthias.Herrmann at jeppesen.com> wrote:
>
> in my experience a potentiometer with center detent could cause another problem: you might not always have the same resistance value at the detented position. it may be different each time you set it back to zero. however, for a VCFs control voltage input it might not be crucial, and it may depend on the quality of the potentiometer - I tried alpha and alps with no luck.

Sounds likely to me. On the other hand, as you say - for control
voltages it's probably not a big issue. On audio it would be much
worse. But how often is it important to set an audio input level knob
to exactly zero? Not very often. :-) The problem might most likely
arise when patching audio-rate modulation to a CV input, or similar.
If sidebands etc are created, they can be audible in low levels,
whereas a leaking envelope modulation of the same level would probably
not.

Btw, the flattening discussed above will reduce this problem.

> I think this is why harry suggested a center tapped potentiometer. I never had one, though.

Me neither. But I guess it has a little range in the middle where it's
zero, and hence gives a "glitch" in the value while turning?

With a center-detented but NOT center-tapped pot, on the other hand, I
would expect it to be hard to find really really small values, since
the pot axis falls into the detent.

> I believe you could alternatively use diodes at the potentiometers wiper to create a kind of 'window', but than you would lose a lot of control range where you need it most probably.

No, then you would distort the input signal, wouldn't you? :-)
For a pure CV-generating knob it would be possible though - but that's
not the issue here.

/mr