[sdiy] Suitcase synth updates

[Neil Johnson]

I wrote:

> > 3819 VCO -- it would be better to run each VCO on its own TL084,
> > rather than have each VCO use two halves, otherwise you'll get
> > soft-syncing between the two (caused by the common supply connections
> > within each chip to the two oscillators, and susceptable to glitches
> > on the output of the comparator).

Tim wrote:

> Would not some stout bypass caps eliminate most of this problem?? The TL
> series are kinda low power, at least compared to LM324s.

The trouble is that you have two comparators running two different
oscillators on the same chip sharing the same supply pins and tracks on
the chip, so any supply rail glitch caused by one comparator swinging its
output from -ve to +ve (or vice-versa) into its load capacitance will be
seen by the other comparator, and _possibly_ cause it to sync, especially
if they are both very close in frequency.  You'll notice this if you try
to detune one oscillator by a small amount---it just won't as the slower
oscillator will be soft-syncing to the faster one (or vice-versa
depending on how the glitch affects the offsets within the comparator).

As this is all on-chip there's not a lot you can do, other than physically
place them on separate chips.

Cheers,
Neil

--
Neil Johnson :: Computer Laboratory :: University of Cambridge ::
http://www.njohnson.co.uk          http://www.cl.cam.ac.uk/~nej22
----  IEE Cambridge Branch: http://www.iee-cambridge.org.uk  ----