Korg Polysix Mystery

[tomorrowstops]

Who doesn't love spending money on a mystery??

BACKSTORY

Took a gamble and bought this Polysix on eBay, listed as non-functioning. The description was as follows:

"Removed old battery, repaired corroded PC Board traces. Installed new battery holder, NMH batteries. Tried loading programs, LEDS act normally when loading. No output."

So it appears the infamous battery leakage issued had been addressed. If you've read anything about these boards and the battery issues, you also know that the repair requires a moderately skilled individual, but isn't necessarily the toughest thing to accomplish, which means hacks like me would give a go and possibly miss something as the acid from the batteries continually travels through the board, eating traces and solder points.

However, the seller had zero pictures of the condition of the internal boards and never responded to my inquiries - a tech had done all this repair work and gave up? Or did we hit a budget limit in the repair of this Polysix (seems to be a used music shop selling)? Either way, it implies there was no obvious solution to regaining output. Anyways, it was listed at less than half the price of a fully functioning unit in 2022, so I said 'screw it,' and bought it - interior sight unseen, questions unanswered.

I've been down this road before - I actually bought a Polysix a few years ago with similar issues and ended up replacing the battery board with Old Crow's pre-populated clone board - a brilliant decision on my part at the time (zero patience to learn how to rebuild the original board). Drop in and go, worked perfectly. This go around, I'm in no particular rush to get this working. I'd like to properly diagnose and fix this one myself, with my new-found interest in the subject.

PRELIMINARY INSPECTION

Unboxing was unremarkable, the exterior of the synth is in average condition and relatively clean. I plugged it in and powered it on; all leds are working, but sure enough no audio comes from the output nor the headphones out. If I max the volume and crank up the mixer its plugged into, I hear steady pink noise. No hum.

Popped open the chassis, and everything looks super clean in there as well. The infamous KLM-367 (battery) board obviously was worked on, and it looks like someone knew what they were doing. They installed a separate battery holder and ran leads to the board, all very neat and clean.

Hmm. Now what? I’ve poked around with my multimeter at various test points: +15v, -15v, +5v rails all within spec. Sig out from the voice board (KLM-366b) tested fine. I don’t have an oscilloscope yet, so I can’t test waveform output.

SPECULATION

I know I shouldn't, but I'm assuming:

A: if everything lights up and voltage tests positive, the power supply and distribution of power are fine.
B: if everything lights up and programs can be selected, written and saved the klm-367 board is fine.

My next best guess is that something is wrong with key bed assembly/harnesses. Would there be another way to trigger audio (bypass the need for keys to engage key contacts) to prove that?
 If not, whats the best probing method to verify continuity with all the connections related to the key bed?

I’m still totally new to testing and probing, so please fire off any obvious suggestions that you may have to look at. I’ve been reading the service manual and sloooowly wrapping my head around the supplied schematics.

Overall it seems like its fixable, just need to chase down the cause. If it takes me a year, it takes me a year. Its been fun so far....

[upunkt]

If you suspect the keyboard there is a simple test. Look at the voice board, each voice has a diode D10 installed. This lights up and indicates when a single voice is actually played. You can put the P6 in unison mode to have them light up all the same time. This checks functionality of the keyboard and helps identifying issues with single voices. It's the same procedure on a Poly-61, btw.

[madtheory]

Strange choice of replacement battery.

[tomorrowstops]

Alright. I fixed the 'output' issue - after taking apart the key bed and cleaning the contacts, it now generates sound on every key. All the voice diodes light up appropriately.

However, tuning has been a nightmare. Right out the gate, voice 5 was out of tune with the rest of the voices by a whole tone, but as you went up the key bed, the tuning gap between voice 5 and the rest got smaller, to about under a semitone. So scaling is an issue as well. Following the standard tuning procedure (that a couple of bloggers/kiwisix have written in an easier to understand format) its appearing that the 'high tune' global trimmer (VR2) isn't doing a damn thing. So I went ahead and replaced all the recommended components (optocoupler, transistor and related op amps) related to tuning. Still no change in behavior.

Quite frankly, at this point, I'm ready to throw in the towel, pack it all up and sell it to another brave/foolish soul.

[tomorrowstops]

update: spent a couple mornings working the tuning procedure from scratch using just the service manual instructions - finally got somewhere this morning with it.

However

In putting it all back together this morning, I somehow shorted something and the beast is down again. Just low freq blips coming from it, regardless of VCO pitch settings and key presses. The new optocoupler might be sitting too high on the voice board, and shorting itself on a diode thats underneath the keybed. I dunno. Had to walk away from it for a bit!

who said it was a good idea to get back in the vintage synth game????

oh thats right, it was me

[tomorrowstops]

decided to throw the old optocoupler back in to see if that would work - and it did (although the pitch drifts like crazy). So, I've got a new VTL5C9 coming and I'll install it with more care this time around to avoid shorting again!