[sdiy] Polyphonic keyboard scanner

[David G. Dixon]

Thanks to Veronica, and everyone else, for the brilliant replies!  As
several have mentioned, the scanning, detecting, addressing and storing of
keypresses is not that big of a challenge.  I'd already gotten that far just
lying awake in bed and mulling it over.  It's the assigning of notes to
voices that is the real challenge.

The reason I'm interested in using logic is because of the challenge itself,
and because I want to become more familiar with logic design in general.  I
will probably do it with a MCU or FPGA later, but for me it will simply be
more fun to work with the logic chips first.  I guess I do want the
"bearskin and buckknife" experience, or whatever, at least this once.


> The DPX is a TOG and divider based synth. Every key will produce a
> note. Some nice notes in the construction articles on how this is done
> to make reasonable sound. Of note, for it's day, the chorus was really
> lush without being wobbly. Had one once.
> 
> The Transcendent polysynth inspired my to do a polysynth in days gone
> by. Like that one, mine was done in logic. The basic idea is to use a
> ram as a down counter with addresses corresponding to keys, so as the
> keys are scanned as an address, so is the RAM. The RAM starts off with
> all ones once a key down start is detected and then each time round
> the loop the RAM count is decremented until the key is down, giving
> the time of flight or velocity. The address gives the pitch, the RAM
> content the velocity and the keydown event the gate. Each key moves a
> contact between two rails such that down, up and in flight give
> different voltages. A window comparator gives the logic signals. At
> one point I was thinking about making the ADSR release proportional to
> keyup flight but never did it. The complexity comes in deciding which
> voice for which note when all the voices are used. I can't remember
> what the Transcendent used but I used a wrapping counter in the note
> assigner but like the transcendent, I had a switch for either stealing
> the oldest note or no stealing. The voices in the Transcendent were
> CEM based. I used CA3080s and transistors for mine.
> 
> My logic board was a very messy affair, but once I got it working, it
> was reliable. I wouldn't do it that way again although I learnt a lot
> doing it. I would either use logic in an FPGA or do it in software.
> 
> The originals where black and white to. Very rarely did ETI have
> colour inside as I remember.
> 
> Claim to fame; I was offered a job about 15 years ago now at Akai in
> Oxford by Tim Orr who designed the Powertran/ETI Transcendent stuff.
> 
> They ETI articles are at CAG.
> 
> Veronica
> 
> On 31-Jul-09, at 1:24 PM, Mike Gorman wrote:
> 
> > David,
> >
> > There are at least 3 designs from ETI that should fit the bill here.
> >
> > The Polyphonic keyboard from July 1979,
> > The Transcendent DPX from August to November 1979 (I think this was
> > the one
> > that Tony mentioned)
> > The ETI Polysynth from December 1980 to March 1981
> >
> > The last one might be the best bet, as I know this could be expanded
> > to 8
> > voices (Mostly CEM based voice modules IIRC).
> >
> > I think the ETI articles are all available on CAG, though the scans
> > may only
> > be monochrome, which can be awkward at times.
> >
> > Regards
> >
> > Mike Gorman
> >
> > -----Original Message-----
> > From: synth-diy-bounces at dropmix.xs4all.nl
> > [mailto:synth-diy-bounces at dropmix.xs4all.nl] On Behalf Of David G.
> > Dixon
> > Sent: 31 July 2009 17:53
> > To: 'synth diy'
> > Subject: [sdiy] Polyphonic keyboard scanner
> >
> > I know this will earn the scorn and derision of the programmers on
> > the list,
> > but I'm wondering:  Does anyone here know of a polyphonic keyboard
> > scanner
> > circuit based on logic IC's (non-programmable)?  Is this even feasible
> > without a massive pile of chips?  If so, I'd be very interested to
> > learn
> > what the algorithm is.  The approaches I've been thinking about are
> > all
> > fairly complicated (one scanner (counter + 2 multiplexers), but 8
> > separate
> > latches and ladders with data comparators to compare the count to the
> > latched data for preventing double latching and controlling note-off
> > -- I
> > haven't worked out all the gory details yet).  I'd like to design
> > something
> > that will send out 8 CV's, 8 triggers, and one gate.
> >
> >
> > _______________________________________________
> > Synth-diy mailing list
> > Synth-diy at dropmix.xs4all.nl
> > http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
> >
> > _______________________________________________
> > Synth-diy mailing list
> > Synth-diy at dropmix.xs4all.nl
> > http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
> 
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy