[sdiy] Simple MIDI-syncable LFO...?

[Richard Wentk]

At 17:26 13/10/2004 -0700, Tim Parkhurst wrote:
> > > Anyone out there built something like this? I think this might be a neat
>way
> > > to do a quadrature LFO too (one clock, four counters & four D/As, pretty
> > > cheap and easy).
> >
> > You're missing it.  You don't need D/As,  that's what those resistor
>networks
> > are for.  You also don't need four counters,  just four resistor networks.
> > Keep the impedance high enough and they won't interact with each other.
>Or
> > use some gates/buffers to isolate them,  lots simpler than duplicating all
> > that other stuff.  Don Lancaster touched on some of this in his CMOS
> > Cookbook,  and appears to have a bunch of related material on his web site
>as
> > well.
> >
>
>Right, I see what you mean. Well, actually by "four D/As" I meant four R2R
>networks. You could very easily use one counter and then wire up the R2R
>networks so that your outputs were 90° out of phase, but I was thinking of
>using four counters so that you could preset/reset them at different times
>so that you could change the phase 'on the fly.'


It's at times like these a PIC starts to make a lot of sense.

If you want your MIDI sync to actually sync with anything, you need to 
reset it with MIDI start messages. At this point discrete logic starts to 
get messy. With a simple PIC you can program a range of waveforms, allow 
for phase offsets and divide-down ratios, and probably run a couple of LFOs 
on a single board.

Something with more power could use interpolation to produce waveforms with 
12 or even 16-bit resolution without breaking into too much of a sweat, 
eliminating the need for filtering/smoothing.

Richard