Alternative to MIDI-CV revisited

[Don Tillman]

   Date: Thu, 09 May 96 15:02:35 PDT
   From: gstopp at fibermux.com

   discharging-cap idea went by the wayside. I had this nagging thought 
   that the discharging-cap method (or linear cap plus log amplifier 
   method for that matter) are "blind" converters - that is, these 
   converters look at the input and put out what they think a VCO should 
   get to match the input frequency. They could be way wrong and never 
   know it. 

Two glib comments:
1. Not that that's necessarily a bad thing...
2. So what, VCO's are "blind" too!

   The PLL method does produce a perfect CV. However the problem is that 
   the correction CV always has some ripple in it, 

This is a classic feedback loop problem.

A way around this problem is to use a "blind" converter for a first
approximation, and then add a PLL loop to tweak that voltage a bit.
You get the best of all worlds.

   I get the feeling that to do this right you will need some very fast 
   circuit components with low leakage to slam the cap to full charge, 
   then let it decay with a good buffer down to ground, and then sample 
   it with a S/H that has as the narrowest possible sample window. 

Actually, I came up with a pretty elegant implementation that uses no
fast components at all.  I'm not going to draw an asciimatic or
anything, but quickly, I noted that you need two caps, one for the
discharging function and one for the S/H-ing function.  So I turned
the problem inside out and had both caps do both jobs alternately.
Soooo, on the first cycle one cap charges for a fixed amount of time
then discharges for the rest of that cycle, and on the next cycle that
cap's voltage is held and fed to the output.  Meanwhile the other cap
is doing the alternate job, holding while the first cap is charging/
discharging, and vice-versa.

The whole thing can be implemented in four chips (one cmos schmitt
trigger, one cmos dual flop, one cmos dual 1/4 mux, and an opamp) and
some random Rs and Cs.

  -- Don