[sdiy] Chinese MG-1s??!?

[Scott Gravenhorst]

"Tom Corbitt" <tom.corbitt at gmail.com> wrote:
>Looking at the design, the only gotcha I see is the 4 TOG chips, which
>brings up a question I've seen asked over and over again on the list
>archives.
>
>What's the best way to recreate a TOG?
>
>Best for me means:
>
>#1 Cheap. if a chip costs over $10-$15 or so, I start thinking ""there
>has to be a cheaper way  to do this"
>#2 Simple. it doesn't have to be a drop in replacement, but I don't
>want to replace 1 part with a sprawl of twenty+ parts.
>#3 Future proof. I want something that I can carry forward in future
>design. Sooner or later everything shuffles off to that great fab in
>the sky, so I want to be able to move my process on to the next big
>thing.
>
>Looking at past answers, the general trend seems to be towards using
>micros like pics to handle the dividing. I've seem the "old crow"
>code, but I've yet to see a post where anyone indicated that they'd
>taken this idea to completion and used it successfully to replace a
>common top octave chip.  It would seem to me that this approach would
>result in a square wave out, that you'd still need to filter back into
>a sine before using it musically.
>
>Am I wrong? What am I missing?

All TOG devices produce rectangular outputs.  A TOG synth also prevents subtle tuning
changes as one moves from one octave to another.  These tuning changes can change and
enhance the character of the synth.

If you're looking to repair a broken vintage synth that contains TOG ICs, I think #2
is doable, I believe someone here has tested a CPLD (or 2) and others have suggested
PIC or other microprocessor.  However, you can still run into problems interfacing a
modern device into and older design because of differences in expected voltages -
adding to the cost.

The key is to find a single programmable digital device large enough to handle the
hundreds of flipflops inside the TOG.  This will determine minimum the number of
devices required to emulate the original TOG.  This and the fact the field
programmable devices (CPLD, PIC, etc) require a programmer affect the cost.

IMO, #3 is unattainable, nothing is future-proof.

If you're doing something new, then you might consider a full digital approach, such
as an FPGA which can be reprogrammed to become whatever you need at the moment.  Both
poly and monosynths are possible in an FPGA.  Not "cheap", but when you consider the
fact that the device is re-usable for many purposes, the moderate cost of a dev board
isn't that much.

>On Fri, 07 Jul 2006, Bob Weigel posted a followup  to a thread titled
>"Top Octave Synthesizer" that ended with the following statement:
>"There are much cheaper solutions for sdiy projects so if there's no
>particular reason to need *that* one I'd look into something like
>that. -bob " (He was referencing the S50240 chip)
>
>What are the cheaper solutions he's speaking of? The one thing the TOG
>seem great at it is keeping the system in global tune without a lot of
>tricky sync work between subsystems.

Maybe Bob can point you in the direction of these "much cheaper solutions".

Digital synths typically have this "in tune" character.

-- ScottG

-----------------------------------------------------------------

-- Scott Gravenhorst
-- GateMan-III - FPGA Based Monophonic MIDI Synthesizer with SVF
-- PolyDaWG/8 - FPGA Based 8 Voice Polyphonic MIDI Synthesizer
-- phLUTe - FPGA Based Monophonic Physical Model MIDI Synthesizer
-- FatMan: home1.gte.net/res0658s/fatman/
-- NonFatMan: home1.gte.net/res0658s/electronics/
-- When the going gets tough, the tough use the command line.