VCDO

[Don Tillman]

   Date: Thu, 08 Apr 1999 09:06:14 -0400
   From: Gene Zumchak <zumchak at cerg.com>

   I have another idea, and there is probably somebody that can
   implement it.  Use a simple voltage controlled high frequency
   oscillator and divide it down by the number of samples you want to
   have per waveform, say 64.  Use a phase-locked loop to lock the 64
   times oscillator to the audio frequency you want to create.  The
   audio frequency is taken from an audio exponential front-end VCO.
   The result will have the same accuracy and stability of the your
   audio VCO.  The audio VCO is not intended to generate tones, just a
   square wave for phase locking and can be consequently as simple as
   possible.  Any comments?

Yeah, this was discussed not too long ago.  Here's my contribution: 

----------------------------------------------------------------
Date: Sun Nov 22 09:21:53 -0800 1998
To: magnus at analogue.org, dhutch at kadets.d20.co.edu, synth-diy at mailhost.bpa.nl
Subject: Re: Fast VCOs for wavetable
From: Don Tillman <don at till.com>

   From: Magnus Danielson <magnus at analogue.org>
   Date: Sun, 22 Nov 1998 05:37:02 +0100

   > "DH" == Drew Hutchison <dhutch at kadets.d20.co.edu> writes:
    DH> It has been fairly well agreed that no VCO we know of can
    DH> handle running at 32-64 times the speed we normally use them 
    DH> with enough stability to reference a table.  Would it be possible 
    DH> to run a VCO (Stopp's?) at a normal speed and send that
    DH> through a multiplier of some form?  I know there are many
    DH> circuits out which do this at MUCH higher frequencies than
    DH> audio, and they seem pretty stable.  It seems like a 
    DH> phase locked loop might do it.  Any thoughts?

   To start of with, the VCO that Gene put in the ASM-1 is good for just
   above 100 kHz (116 kHz, my notes and memory missmatches). It will
   go nonexponential in tracking between 102 and 116 kHz.

A sawtooth VCO is just not going to do well at high frequencies
because the speed and accuracy requirements of the reset transition
get mighty extreme.  Frequency multiplication by manipulating sawtooth
the waveform gets ugly because it multiplies any waveform
imperfections.  And frequency multiplication with a PLL is a pretty
much a trainwreck if you want to use this VCO over musical frequency
ranges (though possibly musically entertaining!).

The most straighforward approach is a triangle VCO.  No reset
transition issues, very simple and well behaved at high frequencies,
just make the caps small and use a really fast comparator.

But Magnus' suggestion got me thinking... you could also but together
something with a regular audio-range sawtooth VCO, a counter connected
to a DAC, and a comparator circuit.  The sawtooth reset pulse zeros
the counter, and as the sawtooth ramps up you compare the sawtooth
voltage with the DAC output, when the sawtooth voltage is delta-V
above the DAC you clock the counter.  Pretty clean, and you get a
simultaneous analog sawtooth.

  -- Don

----------------------------------------------------------------