[sdiy] Fun with LFSRs and such

[Ove Ridé]

On 24 March 2012 21:48, Matthew Smith <matt at smiffytech.com> wrote:
> One thing I haven't come across before is the concept of a NAND synth and
> now wonder if, that part of it too, could not be implemented in programmable
> logic. I can't see any logical (ha!) reason why not, but others may see what
> I'm missing.

The NAND synth consists of four oscillators, where each one acts as a
gate (starts or stops the oscillation) for the next one. Each
oscillator is equivalent of a schmitt trigger similar to the on
pictured here
<http://mysite.du.edu/~etuttle/electron/circ130.gif>
but with a nand gate instead of an inverter.

And therein lies the problem. The CPLD likely doesn't have schmitt
inputs, so it will not oscillate in the way God intended. Instead,
this feedback configuration would make it oscillate at a high
frequency and small amplitude, around the 0->1 transition point of the
input. With a schmitt input on the other hand, the capacitors need to
charged/discharged a certain amount before the breakpoint is reached
and output flips.

-- 
/Ove

Blog: <http://blog.gg8.se/>

"Here is Evergreen City. Evergreen is the color of green forever."