voltage controllable harmonic generator idea ( Was Re: clock multiplier )

[Martin Czech]

> >>At 12:57 PM 15/04/98 +0200, Martin Czech  wrote:
> >>>Another idea for multiplying frequency:
> >>>
> >>>A sawtooth input wave is converted with a thermometer code ADC
> >>>in order to get equidistant time markers. 
> >>>
> >>>This sounds horribly complicated and expensive, but it is not.  It
> >>>takes just a handfull of LF311 comparators
> >>
> >>....this is good.... but I think possibly what people were wanting is
> >>to multiply the clock rate, which will be changing with tempo.. so you

Ok, I missed the wanted purpose a bit.
This was only for audio enjoyment, not for clock multiplying.  And
doing some more (c-prg) experiments one finds, that this will give a
kind of slave osc. , very economic, and you can sweep it like hard
syncing a second vco. This would mean voltage controlled overtone
generation. It would be possible to start with 1:1 frequency ratio
(this looks much like pwm there) and then goes up to n/2:1 (n: number
of comparators) in a single sweep.


And now comming back to the original topic of clocks:

If you don't want to use a pll for clock multiplying (large run-in time
constant, but on the other hand... if the clock rate were very high...)
you have to use some kind of time constant element, i.e. RC combination
etc. in your circuit. And this constant will of course not track with
the input clock rate. So it will work for ONE given input clock rate,
but not for others far away from that.  But what does it mean: "not
work"?  It might yield very interesting results, since to outgoing
clock starts to "limp", i.e. the pulses are not evenly distributed. And
the user could regulate this with a pot. If he knew the the incomming
clock rate will rise by a large ammount, he could adapt the circuit to
that.  Most pieces of music won't change the tempo more then some %
anyway.

All this leads to an extension of the original idea:  The incomming
clock could be used as discharge trigger for a very simple sawtooth
generator (just a integrator with resest transistor).  The integration
time constant could be voltage controlled.  With several comparators
(which have their own reference circuitry, and reference knob), one
could "place" notes into the sequence.  This would be a 100% analog
sequenzer idea (if a comparator is an analog device):  The usuall
sequencer step counter is replaced with the sawtooth integrator, and
the different gate pulses come from severall (window) comparators, the
comparators could also enable a preselected voltage to the pitch
control voltage output. Hey, when I started this mail, I thought this
is a rather boring subject, but now I see that this could be a very
economic and elegant design!  And it gives more than a digital counter
sequenzer:  All "notes" of the sequence could have different length and
arbitrary position in the "bar". And all these parameters could be
easily controlled with pots or even external control voltages, with
"infinite" resolution. Somehow a nice counterpart to the voltage
controlled clock divider schem (LM3900 App. Note).

Sigh, again no time for soldering.

m.c.