[sdiy] VCO reference voltages

[David G. Dixon]

> David,
> > PSU load is largely fixed
> > unless modules are added or removed from the cabinet
> 
> But what about transient loads that other modules can create? Say,
> other VCOs which are not nicely decoupled...

If modules are changing the rail voltages, I'd say one has bigger problems
than drifty VCOs!  But, then again, maybe I'm smokin' crack.  I only have my
own system to judge by.

> > the rail voltages should remain fixed to
> > within 1 or 2 mV during actual operation.
> 
> Out of curiosity (pedantry?), where are you getting this figure from?

Just my own experience.  Every time I've ever measured my rails, they've
always been within 1 mV of the last time I measured them.

> > It is only when modules are added
> > that the rail voltages may change measurably, and only then when the PSU
> is
> > at or near its drive capacity.
> 
> You are here assuming a perfectly, or even properly, working
> PSU/synth. It might be a deadly mistake ;-) But then you can ask
> yourself how far do you want to go to correct the problems of someone
> connecting a toaster to the same voltage rails as your VCO.

Exactly.

> Either way: at A5 this voltage difference on the exp input is a
> difference of frequency of over 1 Hz in comparison to an instrument
> which doesn't experience this. I'm just putting this into musical
> wording - I believe this is much better than many VCOs
> 
> Good (and important) point with the knob range though. I wonder if
> there is some tricky/smart way to use voltage references and still
> maintain full range. I'm thinking of using a range switch rather than
> knob, then you don't need a continuous transfer, that might make it
> simpler to implement a technique based on voltage references..

Actually, as Ian Fritz (gently) pointed out to me in a private email, it was
actually a rather stupid point, since all one has to do is use a smaller
input resistor.

There is actually no good justification for not regulating the rails out to
the panel pots, unless a) you just don't care that much about these small
fluctuations, or b) you do care a lot about saving a buck or two on extra
parts or extra layout hassle.  In any case, regulating the tuning pots is
very easy to do, and removes one more source of drift and uncertainty.

> What range are you thinking of for your coarse tuning potentiometer?
> Full range? Or just, say, +/- 6 semitones?

Mine are always full range.

> If full range - are there many practical applications of such a huge
> range? I'm not saying there aren't, just wondering. The only thing for
> which I really use full range knobs on VCO frequency is adding a
> little audio-rate AM to the patch and changing the general setting
> from the LFO area to the audio-rate area while modulating the AM
> depth, for musical expression. This gives an effect similar to what
> happens when the player of a string instrument pulls up a string to
> change the speed of beat between his note and another instrument on
> the same note.

It's just what I started out doing, and I've seen no reason to change it.

> Seriously - how many people active in this field do you think are
> truly masters of their art?

Well, having just read some very interesting interviews and book chapters
about Moog, Buchla and Oberheim, I found it quite reassuring that these guys
basically had no clue what they were doing when they started.  We all learn
by doing, trying, failing, succeeding, failing, two steps forward, one step
back.  Some of us learn faster than others, and some of us persist in
spectacular stupidity longer than we have any right to, but still, we're all
works in progress.  Can any of us really lay claim to being "master of his
art"?

In my experience, greater master is almost always accompanied by greater
humility.