SergeModular

A member of the Blacet list has brought up a desire of his, to be
able to tap a tempo into his modular system.. Obviously there are
devices to do this, and easier ways then trying to patch it up
manually, but the challenge remains. . Is it possible? Barring the
PSIM, and any other microcontroller approach, how,if at all, can this
be done in the analog domain? Ive been thinking about it for a few
days on and off.. have a few ideas. It seems like Serge territory
to me, so Ill lay it on you guys.

Anyone with any ideas how to get a manually tapped, tempo synced,
clock signal from an analog modular?

Manual trigger / synced slow rising ramp / SH / VCLFO precise
calibration of levels. . getting a time dependent voltage to control
a clock, calibrated to exactly match the duration between
triggers. . might work within a narrow band of tempos..

gated comparator in there perhaps.. ouch, my brain : )

trigger press1, clock pulse one, trigger press2, clock pulse
two, now how do we get this same exact string of pulses to
contiune on?

fun to ponder,
~Steve

>Anyone with any ideas how to get a manually tapped, tempo synced,
>clock signal from an analog modular?
>
>Manual trigger / synced slow rising ramp / SH / VCLFO

That'd be the way I'd copy the usual digital implementation. maybe demux
the incoming pulses into even and odd, and drive two similar chains. that
is:
even pulses reset ramp1 and sample ramp2, odd pulses reset ramp2 and
sample ramp1. these could be sampled by separate analog shift registers so
that a weighted average of the last several readings could be used to drive
the VCLFO. I geuss that's not much different than running one chain, just
thought some timing conflicts might arise, resetting and sampling at once

The other thing that occurs to me is that the slope of the sampled/reset
ramp could be voltage controlled itself, then all that remains is to
maximize
the sampled CV by varying the slope CV. maybe random voltage whose time
correlation increases as the sampled CV increases, zeroing in on the right
slope
to have the ramp reset closest to when it'd naturally reset.
It'd probably take many more taps to get a good signal, but I like this
version
better, it's more like a person in way, the mistakes it'd make would be nice
ones. the sampled CV would be maximized just as well at multiples of the
base
frquency, so it'd work reliably over a wide(r) range

Yasi