[sdiy] The PPPPPulser

[jhaible]

I almost forgot:

You _must_ try this with aftertouch!
Gives the wildest oscillator-sync-like orgies from a single VCO.

JH.

-----Ursprüngliche Nachricht-----
Von: jhaible <jhaible at debitel.net>
An: <synth-diy at dropmix.xs4all.nl>; Ian Fritz <ijfritz at earthlink.net>
Gesendet: Sonntag, 8. Juni 2003 21:43
Betreff: Re: [sdiy] The PPPPPulser


> > I took an hour yesterday and white-boarded the variable multi pulser I
> > proposed a while ago.  To review, it uses an LM3914 bar graph chip with
a
> > constant (0-1V) saw drive and a variable (5-.5V) control drive on the
> > comparitor resistor chain.  The odd-numbered 3914 outputs are summed
into
> a
> > comparitor to make a pulse train. The output starts out as a square wave
> > and continuously adds more pulses, up to five, as the external CV varies
> > from 0 to 5V.  The whole circuit just takes the 3914 and two op amps.
> >
> > The results are pretty interesting, I think.  The unit actually has a
lot
> > of the characteristics of a filter.  There is a lot of phase shifting
> going
> > on, along with large and rapid changes in the harmonic content.  A sweep
> > through control voltages sounds somewhat similar to a sweep of a high-Q
> > bandpass filter.  Here's an MP3 demo, which is just a sweep up and down
in
> > CV, with no other processing.
> > http://home.earthlink.net/~ijfritz/xfer.htm
>
> Sounds nice!
>
> I have a similar circuit, built from LM339 quad comparators, in
> my JH-720:
> http://home.debitel.net/user/jhaible/jh_720_vco_scanner.gif
>
> I've used a 3080 to control the input level. Your method
> of applying a variable Voltage to the resistor divider chain
> is certainly more elegant!
>
> As an ordinary comparator chain produces a "bar" rather than a
> "dot" code, I had to find a way to convert the bar to dot.
> At first I thought about some logic after the comparator outputs,
> but as I'm also only taking every second output (to get that
> multiple pulse waveform), I could simply wire the comparator
> outputs together in pairs, with the comparator inputs having
> opposite polarity. As we're going up the "thermometer code",
> the first comparator of a pair "sets" the dot, and the second
> "clears" it. Then the same thing will start again with the next pair
> of comparators.
>
> Of course these LED driver chips have all the required logic
> already built-in. But if someone is looking for a "standard chip"
> version, you may find the LM339 version interesting.
>
> JH.
>
>
>