> This (Analog Shift Register) looks like another great potential MOTM
> although I'd really need about six more VCOs to make it really worth
> >How does the Serge do it ?
> Yeah! Good question!
I have no idea how Serge did it either, but reading the Serge catalogue has
inspired me to design the circuit shown in the adress above, and it works
nicely. The underlying idea is to pass the sampled voltage from left to
but the trigger pulse from right to left, therefore no auxiliary stages are
required. As you can see, I've done it with cheap monoflops, for both the
3 sample pulses, and the 2 "guard intervalls" between these pulses.
Later someone (if memory serves that was Gene Stopp) showed a slightly
more elegant version, replacing all the monoflops with a single shift
In practice (now, that would be making music, not designing circuits
the great benefit of the ASR is that you have a set of exactly correlated
random step functions (if you feed in a random signal). And there is the
other application, of course, which is also explained in the Serge
Feed in the keyboard CV to the first stage, and the key trigger to the clock
and you get canon-like structures when you control a set of 3 VCOs from
the 3 outputs.
And the best: You can even play chords with a monophonic keyboard
(or a single CV midi converter): Play a fast roll over the 3 notes you want
to use for the chord (starting with the lowest note if the keyboard has
high note priority), and then hold the chord as if you were playing a
This works remarkably well in practice, the only drawback being that
you can only play 3-note chords without side effects. If you play
only two notes after releasing the previous 3 ones, the last of these
3 will continue to sound. But you can even use this to your advantage,
because you have a means of mixing the polyphonic and the "canon"