Ann Arbor, Michigan, USA
Poly-61 Episode III: Into the Guts
StoreTags: poly61, symth, repair, project, korg
Author: jcd on September 09 2006
Viewed 3912 times. 4 people liked this blog. You can rate it below if you haven't already.
People who enjoyed reading this: Zanf, Fah, PAWEL, astroid
--> Notes from my first stab at fixing the old battleaxe:

Opened up case.

Found chips which recurred in sixes, gouged out furthest one right. No effect.

Gouged out nearby chip. No effect.

Played keyboard while top up, noticed small lights firing in sequence from left to right. Dumb luck; furthest right set does indeed seem to be the one with the problem.

Played some more while dicking with the connection to the furthest right set. Get some random patches by powercycling, also every so often a note begins very low (so low that the individual clock cycles of the waveform can be heard as distinct ticks), then raising up slowly through octaves. No amount of keypressing seems to affect the note during one of these episodes. It is profoundly creepy.

I get the idea that i should be recording this. I hook up a tape recorder, but nothing creepy/cool happens after I do that.

Eventually pin down the continuously-sounding note; a small clip bringing wires from leftmost board seems loose, and I can make the note sound/not sound by manipulating it. Unfortunately, I cannot see an easy fix to get it into the same shape as the others, and it is hard to get at without dismounting the whole inner central circuit board, something I am wary of doing.

I try disconnecting this clip. Now no note sounds, but now instead of a continuous tone which changes pitch every sixth keypress, I have one note out of six simply not being sounded. Still; progress!

NOTE small round pot looking things near labels like RES ADJ and VCA ADJ. possible pots for direct control? Very exciting, although a little manipulation seemed to produce no results

NOTE when a set is lit (an oscilator engaged), subsequent key presses trigger the DCOs around the engaged DCO without skipping a beat. this may come in handy when devising a plan to bypass the damaged 6th DCO. How to trick DCO6 into always being engaged, but without sounding a note? Perhaps rewiring?

NOTE pulling the rainbow-wired harness out of the leftmost board seems to trigger all notes, like a severed finger's nerves firing in volleys. Playing keyboard differs the pitch in some way, but there seems to be no 'note off' command. perhaps it triggers all available DCOs at once? but how does it know when to sound (and when to stop sounding) a particular note when the harness is plugged in, but it doesn't when it's not plugged in?

More to come.

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

Previously:

Episode I: The Phantom DCO
link

Episode II: Attack of the Tones
link
Read jcd's other blogs.jcd's Recent Blogs
Comments

i love fiddling with synths!
take pictures
record sounds coming from it that aren't supposed to sound

episode II had some samples. i even made a song out of them. then my computer died. gots a new one now so ill finish and post in a few weeks.

Couldn't find my camera = no pictures. I'll def get some once I find it.

I'm not quite sure what you meant by "trigger around the engaged voice", but the idea is to get the CPU to think the voice is always on and mute its output, since it's probably old enough not to have voice-stealing. Is that what you're aiming for? You really should get schematics!!

I used to have an Oberheim OBXa, and there was a dip switch to enable/disable each voice card. It could still be there in the Korg, except hard-wired for the voice count, so it's something to look for if the design actually does use voice-stealing.

just FYI: You don't trigger a DCO. They are continuously on. You trigger an amplifier using an envelope, which allows the signal to pass through. Since what you are describing sounds like a droning voice that the CPU is not aware of, it might actually be the enevelope bias voltage being incorrectly sampled from the MG(software envelope) for that voice. Probably just need to replace a transistor and maybe whatever is coupling it.

Thanks, RogerRoger. That's is exactly the kind of thing I hoped to learn when I began this undertaking.

Apart from messing with keyboards and asking specific questions on this board, are there any books/websites you can recommend for someone looking to learn basic electronics and synth design?

Also I found my camera. So, soon to come: Pics (or STFU).

Can't really help much, as I got my electronics training from the US government. I do know that Korg, being the lame-o's they are, will mail you a photocopied service manual of the Poly-61 for about $20. Kenton also makes a MIDI retro-fit for it: link

You might also try setting up a pseudo unison mode by pressing Hold, playing the same note six times and pressing chord memory. See if the note still sticks.

BTW, I did some looking online, and I was wrong about the Poly-61 using software envelopes: there are actually six SSM2056's located near the LED's that light up for each voice. Your stuck voice can be one of these IC's gone bad, or more likely just the circuit between it(the wire clip you mentioned) and the voice's VCA, or maybe one of the envelope's input signals is off. The latter seems plausible since you said the envelope affected oscillator pitch on startup(slow rise), once or twice.
The EG's output is pin 10 according to its data sheet: link

The source I was reading also mentioned that the keyboard itself must be cleaned regularly or it will have intermittent contact problems, and that the Poly-61 seems to play better from something triggering the arpeggiator, than just playing the keys alone.

Beyond that, I'm not going to be much help, sorry.

Great link, thanks.

Yeah, that chip looks like one of the ones that *ahem* I pulled off the board in the naive assumption that it would make the bad noise go away. Gotta break a few eggs, right? I'll do some reading there and see what I can cook up.

I think that would qualify as a malfunctioning EG. ;)

jcd said: "Are there any books/websites you can recommend for someone looking to learn basic electronics and synth design?"


Otterfan just posted this link
As for synth design, I'm certainly not the person to ask, but I suppose you could start with synth kits.
link link link link link

nevermind...that was poly6 stuff not poly61

Hey jcd...

I found this blog after looking for some tips and tricks on the Poly61...anyways to cut a long story short, I just acquired one and the keys were misfiring....basically the keyboard contacts were dirty...other than that tho, it was in perfect condition....I opened it up last night to clean the contacts, took the synth apart and then once the contacts were clean I tested it......it was working perfectly...I put it all back together and bam, I got the "problem you were talking about...every sixth note palyed goes squwacky... ?!

I thought to myself dammit, 2 days old and ive blown the board....so i took it all apart again thinking that maybe the keyboard chassis was too close to the actual circuit board and was causing interference....I unscrewed it all and it started working fine again ??? huh ???

Anyways it turns out that there was a membrane type wire not sitting properly in its connection...if you take a look at you keyboard, see where the earth from the keyboard chassis goes to the actual keyboard chassis ??? there will be 2 membrane like flatwires connecting the 2 boards together...makes sure that the "flatwires" are sitting properly in their connectors....once I did that, screwed everything together again, it was working perfectly

This may not be the same problem you have but I had the same symptoms so check it out....maybe itll work for you to
Oh I took some pics of the "flatwires" if you need a better description...

Laters


Register / login
You must be a member to reply or post. signup or login