[sdiy] square to sine

David G Dixon dixon at mail.ubc.ca
Wed Oct 10 18:48:35 CEST 2012

> > > I've looked at V to F to a square, then integrating to a triangle
> > > but the amplitude of the triangle varies with frequency.
> > > Sure, you could correct for that with an AGC, but then your fast 
> > > tracking is lost.
> >
> No, because I'm going to use two 2164 expo converters: one 
> for the V/F 
> >converter, and one for the integrator, both controlled by the same 
> >voltage (remember, 2164s pass bipolar current).
> But if you have a fast, accurate integrator, why not use that 
> directly for a Tri VCO core?
> Ian 

That's what I'm going to do, but I'm going to use the ultra-accurate V/F
converter to drive it.  In this way, the frequency of the oscillator will be
disassociated from a schmitt trigger threshold, and there will be no way
that the integrator can influence high-frequency tracking.  The only concern
will be triangle amplitude control.  This should not be a problem if the
integrator is driven by an expo at the same gain as the V/F converter.  Slew
rate limitations may cause a slight loss of amplitude at high frequencies,
but that remains to be seen.  I'm going to breadboard this mofo on Sunday,
and I'll report back with my findings.

More information about the Synth-diy mailing list