[sdiy] VCO saw integrator cap questions

[David G. Dixon]

> In my opinion all those cap discussions are a bit exaggerated. Low leakage
> for sample and hold is good, but as you don't build S&Hs for scientific
> purposes but for musical applications this is not too important, I would
> say. Temparature stability of caps for VCOs I haven't heard of yet as this
> is solved by temperature resistors or transistor arrays as I know.

I would tend to agree with you, somewhat.  My first VCO was built with an
ultra-cheap orange pancake ceramic cap as the timing cap (I didn't know any
better), and it sounded great!  However, one does want to do the right
thing, generally.

> AND IF YOU WANT REAL MATHEMATICAL STABILITY USE A DCO INSTEAD OF AN VCO.
> Its
> just curious. All those people out there spending tons of effort searching
> for analogue precision to solve a mathematical problem: The exact doubling
> of frequencies and the constant generation of them.

Again, I think you're probably right.  However, in this case, going analog
is more of a philosophical stance.  Having said that, if I'm not satisfied
with this new 2164-based-expo VCO I'm currently laying out, I may do as you
say and go digital.

> Not to be misunderstood: Sound forming modules like filters work best as
> analogue solutions, as the non mathematical nonlinearities make the
> specific
> sounds of them, and the wondeful timbre of a moog ladder just before self
> oscillation cannot be computed by a DSP. But some mathematical tasks like
> octave control should be solved digitally in my opinion.