Discussion about the Korg PolySix synthesizer

on 8/26/03 9:08 AM, PolySix@yahoogroups.com at PolySix@yahoogroups.com
wrote:

> 
> Message: 1
> Date: Mon, 25 Aug 2003 17:23:44 -0400 (EDT)
> From: The Old Crow <oldcrow@oldcrows.net>
> Subject: I finally managed to make new a KLM-367
> 
> 
> After years of procrastinating, I spent this past weekend creating
> schematic and PCB files for the KLM-367A.  It has gotten to the point
> where it takes less time to assemble a new board than to repair a damaged
> one.  Thus, I will be offering new replacement boards in the very near
> future.  Here is an image of the layout:
> 
> http://www.oldcrows.net/~oldcrow/klm367a_img.gif

Wow!  My hat is off to you crow, Well done!

(What a monumental undertaking you have done!)

I am better known as a modular fan, but I also have a couple of PolySixes
in my arsenal of goodies. I really love them!

(Basically all I use are my Modcan / Cynthia hybrid modular system with
two Aries analog keybaords, and two PolySixes.)

In fact I've been wanting to add jacks to my P6 to interface it with my
modular...   Simple mods that I've been wanting to do is add an LED for the
third LFO and make the LFO outputs available, as well as converting the
trigger from Moog S-type over to um, "Arp" type.

There are many other mods that would be cool, in fact If anyone has some
cool mods they've done I'd love to hear about them.  (As you can see here I
am not a purist, as far as keeping my Poly's "stock" looking).

(It would be cool if your new PCB layout facilitated a few mods easily?)

It's really refreshing to see this level of commitment to the machine Scott.
I think would like to get a couple of these new boards when you have them
made, and am Very interested in facilitating the MIDI mod as well.

Do you suggest these boards just for units are found broken, or also for
PolySixes that have already had their battery problems repaired?

I hope that I don't upset anyone here with this next comment, but...

Does anyone really ~want~ to deal with that cassette interface?  (Gak!)

I've often thought that that entire section could be replaced with something
much more interesting, (or even a small floppy disk drive or the little
flash memory cards thingies that go into digital cameras in order to save
patches).  

(Personally, I stopped dealing with cassette data interfaces back in 1978
with my old Apple ][+  and I'm not going back!  LOL!)

Thank you for such an inspiring job Crowster!

Cynthia

www.cyndustries.com

(ps, anyone who wants to sell disabled or broken Korg Poly Sixes please
contact me privately thanks!)

> 
> Notice the 3V CR2032-type coin cell socket that replaces the evil, old
> NiCad battery. ;)
> 
> I known this for a long time, but I might remark that the Polysix
> service manual's KLM-367 schematic has a few errors.  My schematic is
> "correct" as it is derived from the manual's schematic but cross-checked
> against the actual board layout.  The result is the exact same circuit as
> is implemented on original KLM-367s.
> 
> Once I build and test a few of these boards, I'll make a few revisions
> to better support adding in Ricard Wolf's MIDI retrofit as well as the
> little bit of hardware needed to support adding in my "2nd VCO PCB" which
> provides 2 VCOs per voice similar to the Trident.
> 
> Crow
> /**/

>  I hope that I don't upset anyone here with this next comment, but...
>  
>  Does anyone really ~want~ to deal with that cassette interface? 
>  (Gak!)
>  
>  I've often thought that that entire section could be replaced with 
something
>  much more interesting, (or even a small floppy disk drive or the little
>  flash memory cards thingies that go into digital cameras in order 
> to save patches).  
>  
> 
>  (Personally, I stopped dealing with cassette data interfaces back 
> in 1978 with my old Apple ][+  and I'm not going back!  LOL!)

Well, since all modern computers have good quality audio I/O, we are 
back to the point where the cassette interface is useful again. True,
it would be much nicer to have a direct MIDI sysex implementation of
the entire set of parameters, but hooking up the cassette interfce to a 
computer can be done with no modifications at all to the P6, so I find 
the beforementioned P6 librarian program a very attractive tool indeed,
and I will be looking forward to evaluate (and probably pay for) a copy.

/Frank

Yup - 
 or even more simply, 
 using a minidisc recorder instead of a cassette recorder
can make all those old cassette interfaces MUCH more
reliable these days! (and automatic track location thrown
in!)

 --- dko4342@vip.cybercity.dk wrote: > >  I hope that I
don't upset anyone here with this next
> comment, but...
> >  
> >  Does anyone really ~want~ to deal with that cassette
> interface? 
> >  (Gak!)
> >  
> >  I've often thought that that entire section could be
> replaced with 
> something
> >  much more interesting, (or even a small floppy disk
> drive or the little
> >  flash memory cards thingies that go into digital
> cameras in order 
> > to save patches).  
> >  
> > 
> >  (Personally, I stopped dealing with cassette data
> interfaces back 
> > in 1978 with my old Apple ][+  and I'm not going back! 
> LOL!)
> 
> Well, since all modern computers have good quality audio
> I/O, we are 
> back to the point where the cassette interface is useful
> again. True,
> it would be much nicer to have a direct MIDI sysex
> implementation of
> the entire set of parameters, but hooking up the cassette
> interfce to a 
> computer can be done with no modifications at all to the
> P6, so I find 
> the beforementioned P6 librarian program a very
> attractive tool indeed,
> and I will be looking forward to evaluate (and probably
> pay for) a copy.
> 
> /Frank
> 
> 
> ------------------------ Yahoo! Groups Sponsor
> 
> To unsubscribe from this group, send an email to:
> PolySix-unsubscribe@egroups.com
> 
> PolySix "Digiest" Page:
> http://www.acc.umu.se/~amber/Poly6 
> 
> Your use of Yahoo! Groups is subject to
> http://docs.yahoo.com/info/terms/ 
> 
> 
> 
> 
>  

=====
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
http://www.felinedream.co.uk

________________________________________________________________________
Want to chat instantly with your online friends?  Get the FREE Yahoo!
Messenger http://uk.messenger.yahoo.com/

Hi Cynthia--I also do design work for Paul Schrieber of SynthTech, so 
I'm also making stuff on the modular front.  Anyway:

On Tue, 26 Aug 2003, Cynthia Webster wrote:

> Wow!  My hat is off to you crow, Well done!
> (What a monumental undertaking you have done!)

  Well, it needed doing a long time ago..but I am very good at
procrastination, especially when there are more pressing engineering tasks
at hand.  The original KLM-367 has a serious design flaw in that 1) they
put a NiCad battery on the pcb and 2) put it near critical circuitry.  I 
finally got fed up with spending an average of 3 hours per KLM-367 repair, 
and decided that the same 3 hours would yield a new, assembled board.

> In fact I've been wanting to add jacks to my P6 to interface it with my
> modular...   Simple mods that I've been wanting to do is add an LED for the
> third LFO and make the LFO outputs available, as well as converting the
> trigger from Moog S-type over to um, "Arp" type.

  Well, 2 LFOs happen to be on the KLM-367: the PWM MG  (Korg calls LFOs 
MGs as you know) and the routable (VCO/VCF/VCA) MG.  Not much would be 
needed to provide outputs for either of these.

> There are many other mods that would be cool, in fact If anyone has some
> cool mods they've done I'd love to hear about them.  (As you can see here I
> am not a purist, as far as keeping my Poly's "stock" looking).
>
> (It would be cool if your new PCB layout facilitated a few mods easily?)

  Well, the initial prototypes are direct KLM-367 clones.  Once I verify 
the clone boards in fact work correctly, I'll start adding in option 
features.  For example, there are two unused MUX S/H outputs that beg to 
have S/H circuitry added so as to create two new control voltage 
parameters.  Ricard Wolf used them for pitchbend and mod wheel CVs, I 
believe.  I would probably provide circuitry for the same.

  Down the road I want to change CPUs as 8748s are getting harder to come
by.  With a CPU change, I can start to do things like add MIDI, integrate
the key assigner (a separate CPU on the KLM-366 board) and generally make
things more efficient such as eliminate the battery-backed SRAM and use
the new MCU's internal user data flashrom area to store patches..many more
than 32 patches, btw.  While doing this I would switch to surface-mount
for most of the board's parts: I can get a factory to build SMD boards
*much* cheaper than to build TH boards.

> It's really refreshing to see this level of commitment to the machine Scott.
> I think would like to get a couple of these new boards when you have them
> made, and am Very interested in facilitating the MIDI mod as well.
> 
> Do you suggest these boards just for units are found broken, or also for
> PolySixes that have already had their battery problems repaired?

  I would suggest the clone boards for repair job first--thats why I made 
them.  For upgrading the Polysix, I would go with the next version of the 
board, once I figure out what to put on it.
 
> I hope that I don't upset anyone here with this next comment, but...
> 
> Does anyone really ~want~ to deal with that cassette interface?  (Gak!)
> 
> I've often thought that that entire section could be replaced with something
> much more interesting, (or even a small floppy disk drive or the little
> flash memory cards thingies that go into digital cameras in order to save
> patches).  

  The cassette interface takes up very little hardware (and not much code
space, either).  It is about the most generic implementation of the Kansas
City Standard FSK data-to-tape I/O routine I've ever seen.  I would leave
it in simply because lots of folks still have patches on tape.

  Anyway, once I have a new board assembled and working I'll figure out a 
repair/replacement scheme.  I can probably still do the $100+old board (I 
would scavenge a few esstentials off the old boards) and send a new board 
to folks...we'll see.

Crow
/**/

Hi,

>
>  Down the road I want to change CPUs as 8748s are getting harder to come
>by.  With a CPU change, I can start to do things like add MIDI, integrate
>the key assigner (a separate CPU on the KLM-366 board) and generally make
>things more efficient such as eliminate the battery-backed SRAM and use
>the new MCU's internal user data flashrom area to store patches..many more
>than 32 patches, btw.  
>
yes, maybe some Kind of a ATMEL MCU like the 8535 would me IMO 
interesting for this board cause of 8K Code Space, 512 Bytes of EEPROM 
(a bit few), 32 I/O-Lines, 3 PWM-Channels and eight 10-BIT A/D\ufffds. 
Additionally you have a SPI for in-System-Programming and afaik a 
programmable Serial Interface (so you can add easy a midi-Solution). 
Other Problem is that he is not Pin-compatible to the 8048, so a 
completely new (or re-wired Socket of IC22) Board would be a -must- for 
changing the MCU. On the other Hand i think that much Parts of the 
KLM-367 can be replaced with this MCU, like the variable Resistors for 
MG Level, PWM Rate etc. You have to use 6 A/D\ufffds for those resistors, one 
for reset_adj and additionally one for the Knob-Board (which needs afaik 
all in all 4 Adress lines; three for the 4051\ufffds, one for changing 
between both 4051\ufffds; and one Data-Line).

For additional functions would be much easier IMO to build a new Board 
from scratch with a new MCU and a own program (presupposed someone knows 
exactly which pin on which connector has which function)....

Otherwise, has anyone ever read out the 8048 and had a look into the 
code? I can program in Assembler; proably i can analyze the Code of the 
8048 and make it compatible to the 8535 with a few re-programming....i 
have two 8048\ufffds from two 367\ufffds here, but i can\ufffdt read them out, cause i 
have no Device to do so.....

Greetings

Stef

I plan to use a PIC18F6520 in the T64 (0.5mm TQFP) package.  It has all 
sorts of internal features: dual UARTs, 4 timer/counters, I2C, SPI, 
multiple A/D..you name it.  It has 32K program flashrom, 1K user data 
flash, and 2K of SRAM.  Plus, I'm used to the part so using it in a patch 
manager board like this will be relatively simple.  If I somehow manage to 
run out of program space, I can go to the 6620 (2x all memories) or the 
6720 (4x all memories).

  It would be a surface-mount revision of the board, with a lot of the old 
hardware going away in favor of PIC-based circuits and code replacements.  
The only things that really need to stay are the analog circuits.

  As far as the code for the KLM-367's 8048, I extracted the code about
ten years ago and have it online as a 1K binary file.  Ricard Wolf did a
complete disassembly on paper for it around the same time; I have a paper
copy of his notes but they're not really mine to hand out--please ask him
for a copy.  (He also disassembled the KLM_366's key assigner 8049).

  FWIW: the key assigner 8049 is also used in the Mono/Poly.  In fact, the 
voice stack/share mode of the Mono/Poly will work in 4 voices on the 
Polysix, provided you hardwire a button in to shunt the proper switch 
scanning pins.

  Also, the 8048 used in the KLM-367 is also used on the KLM-380 for the 
Trident MkII--it is the exact same program.

Crow
/**/

On Wed, 27 Aug 2003, Stefan Rinass wrote:

> Hi,
> 
> For additional functions would be much easier IMO to build a new Board 
> from scratch with a new MCU and a own program (presupposed someone knows 
> exactly which pin on which connector has which function)....
> 
> Otherwise, has anyone ever read out the 8048 and had a look into the 
> code? I can program in Assembler; proably i can analyze the Code of the 
> 8048 and make it compatible to the 8535 with a few re-programming....i 
> have two 8048\ufffds from two 367\ufffds here, but i can\ufffdt read them out, cause i 
> have no Device to do so.....

I forgot to mention, one of the mods I've done is to add a 2nd set of 
VCOs.  Patch info for octave and detune interval is saved in a little PIC
with a given patch (not saved to tape, though).  The VCO#2 circuits are 
sawtooth-only, but they can be turned off, set to the same octave, -1oct 
or -2oct of VCO#1.

  I suppose I could modify the KLM-367 mainly to support the signals I 
need to store/recall VCO#2 settings, but the VCO#2 board wouldn't be part 
of the '367.

Crow
/**/