Korg Poly 61 pitch issue consistently up 5 semitones.

[Midphase]

OK, so this means +13.4V at the "right hand" end of R34. Are there any large (negative?) voltages at the right-hand ends of R33 or R22?

If it's possible to measure any & all voltages around these resistors and Q7, that would help. (Why didn't they put these in the service manual???)

I'll measure those asap.

BTW, do you think IC2 might be playing a role in this? My general presumption and experience is that 74xx___ IC's rarely fail (except in some notable cases) and so far swapping out all three of the opamps in the circuit or the two transistors has not yielded any solutions. I keep thinking that the problem might be staring at me right in the face and be something relatively simple. I have to admit that out of all the synths I have repaired, this one is presenting quite a challenge.

 

[capacitorless]

If you've measured this already, my apologies (crunched for time right now). With the synth powered up, is there any appreciable voltage between pin 2 on the mod (Y) pot and ground? What was said already about the resistance reading when powered up, but I really didn't expect there to be a difference. (edit: I'm making the assumption here that the mod section should normally output 0v dc and 0v ac, when the joystick is at the resting position).

I may print out an enlarged schematic of that section of the schematic, annotate the various readings gathered up to that point, and share it. Update: I'm slammed with work related stuff so it would be more like mid-week at this point. I would suggest someone do that so that each of us (including the OP) don't have to individually go through the various posts to try to pick up on the latest. I've found this exercise to really speed up the process, especially when multiple people are trying to help

 

[Midphase]

OK, so this means +13.4V at the "right hand" end of R34. Are there any large (negative?) voltages at the right-hand ends of R33 or R22?

If it's possible to measure any & all voltages around these resistors and Q7, that would help. (Why didn't they put these in the service manual???)

Sorry for the delay -- bit of a crazy day: Here are my findings:

You are correct that (when the mod wheel pin 3 is disconnected) I get +13v on the right side of R34 and about 3.9V on the left side

R33 is giving me -1.3V on the right side and +3.9V on the left side

R22 is giving me 3.9v on the right side and about +/- .01v on the left side

R36 is giving me 3.9V on the right side and .06 on the left (ground?) side.

R27 is giving me 13V on the R side and 3.9 on the left

R37 is giving me 15.3V on the left side and 14.9V on the left

D5 and D4 I already checked and they tested good, 14.9 going in, 13.9 out of D4 and 13.4 our of D5

R35 is giving me 14.9 on the left and 13.9 on the right

Getting 15.2V on all three pins of the tuning pot VR4 and same on the Trim fine adjustment for the tuning.

As I said, I swapped out Q7 and Q2, and the old components seemed to test fine on my diode measuring mode.

I already swapped out IC6 quad opamp and no difference.

A head scratcher for sure!

 

[Midphase]

With the synth powered up, is there any appreciable voltage between pin 2 on the mod (Y) pot and ground?

Here's what I'm picking up on both pots on the joystick (with all 4 connectors hooked up):

Pot 1 X-axis (pitch bend) pins 1-4 = -7v, -4.5v, -7v, -7v

Pot 2 Y axis mod/LFO pins 1-4 = no appreciable voltage on any pin, whether pin 3 is connected or not.

 

[Midphase]

Ok...hold on....I feel like I'm doing this all wrong. If we are to try to figure out exactly what is happening, I feel like all of the measurements that I give you have to be with all pins of the pots connected, so effectively when the malfunction is present, otherwise it's all kinda moot.

Let's start again, this time with all pins of the X and Y pots on the joystick connected (i.e. with the pitch being wrong):

IC8 4558 dual opamp: p1 = .01v, p2 = .001v, p3 = .0001v, p4 = -15.4v, p5 = .0001, p6 = .0001, p7 = .0001, p8 = +15v

IC7 4558 dual opamp: p1 = square wave at 6.8hz with 13.4v RMS, p2 = 2.4v RMS and weird spikes on the scope, p3 = triangle/swaure wave at 6.8hz and 6.7v RMS, p4 = -15.5v, p5 = .0004, p6 = .001v, p7 = triangle at 6.7hz and 6.7v RMS, p8 = +15.3v

R33 is giving me -1.1V on the right side and +.3V on the left side

R34 is giving me .01V on the right side and +.3V on the left side

R35 is giving me 13.6V on the right side and 14.8V on the left side

R36 is giving me .3V on the right side and .001 on the left (ground?) side.

R37 is giving me 14.3V on the right side and 15.3V on the left

R22 is giving me .3v on the right side and about -.0001v on the left side

R27 is giving me .001V on the right side and .3v on the left

D4 14.3v going in, 13.6v out, D5 13.6 in and 13.0 out.

Do any of these look like what they're supposed to be?

 

[Midphase]

I can't shake the feeling that the key red flag here is that I'm not getting +/-4v at pin 7 of IC8. I think that's got to be the key to figuring this out...but swapping out the opamp for another one doesn't fix the issue, and there are no other soldering issues that I can see around IC8 so I'm pretty sure the actual opamp is fine. The main question is -- what is causing pin 7 not to output the right voltage? By comparison pin 7 of IC7 outputs exactly what it's supposed to (i.e. a triangle wave at +/- 10v).

 

[Script]

I didn't follow all of your measurements above...

Did you (or the owner) check whether the synth shows any other malfunction or slight misbehaviour elsewhere ?

Really not more than a guess but could it be you are looking at a symptom, not the disease itself, like polluted ground reference maybe... ?

 

[Script]

Also in Service Manual further down, it describes how to check for correct PWM, DCO and Tuning. And looks like there is even a Program Setting for changing overall tuning...? :/

 

[Midphase]

I didn't follow all of your measurements above...

Did you (or the owner) check whether the synth shows any other malfunction or slight misbehaviour elsewhere ?

Really not more than a guess but could it be you are looking at a symptom, not the disease itself, like polluted ground reference maybe... ?

From what I can tell, everything else in the synth appears to work as expected. The Poly-61 isn't a terribly featured synth, when I went through several of the edit settings, it seemed to behave as expected. Arpeggiator works, program parameters change as expected. That's what makes this weird transposition issue so strange. I would expect any other serious issue to manifest itself elsewhere but it only seems to be affecting the overall tuning of the instrument.

Regarding the service manual sections on the tuning etc, I did go through the set up procedures outlined there and everything checks out. The main issue is that the entire keyboard tuning is transposed up about a 4th, and there is no way to bring it back down by enough range to get the synth to play in the proper key. When I disconnect Pin 3 from the Y-axis potentiometer, the issue seems to disappear, or at the very least become so slight that I can compensate for the deviation by adjusting the tuner trim setting. With the tuner and trim set at the 12 o/clock position C sounds roughly as C#, so I actually have to turn down the trim to 9 o'clock to get the C to sound in tune, but once I do everything is stable and in tune across the five octaves. The main issue is that by disconnecting Pin 3 from the Y-axis potentiometer, only one half of the joystick up/down function works. So I can add modulation by pushing it up, but pulling the joystick down doesn't have any effect obviously. It's a minor inconvenience in the grand scheme of things, but I would prefer to actually find and fix the issue rather than do a hacky workaround.

 

[Voyager10]

Ok...hold on....I feel like I'm doing this all wrong. If we are to try to figure out exactly what is happening, I feel like all of the measurements that I give you have to be with all pins of the pots connected, so effectively when the malfunction is present, otherwise it's all kinda moot.

Let's start again, this time with all pins of the X and Y pots on the joystick connected (i.e. with the pitch being wrong):

IC8 4558 dual opamp: p1 = .01v, p2 = .001v, p3 = .0001v, p4 = -15.4v, p5 = .0001, p6 = .0001, p7 = .0001, p8 = +15v

IC7 4558 dual opamp: p1 = square wave at 6.8hz with 13.4v RMS, p2 = 2.4v RMS and weird spikes on the scope, p3 = triangle/swaure wave at 6.8hz and 6.7v RMS, p4 = -15.5v, p5 = .0004, p6 = .001v, p7 = triangle at 6.7hz and 6.7v RMS, p8 = +15.3v

R33 is giving me -1.1V on the right side and +.3V on the left side

R34 is giving me .01V on the right side and +.3V on the left side

Thanks! So far these all look plausible to me.

R35 is giving me 13.6V on the right side and 14.8V on the left side

OK, so if the supply voltage is 15.3V this leaves about 0.5V across tuning pots VR4 and VR5, which suggests they have a combined resistance of about 830 ohms. Are they at mid-travel? You should be able to tune it lower by increasing their resistance. If they're at maximum resistance perhaps there's something wrong there?

R36 is giving me .3V on the right side and .001 on the left (ground?) side.

R37 is giving me 14.3V on the right side and 15.3V on the left

R22 is giving me .3v on the right side and about -.0001v on the left side

R27 is giving me .001V on the right side and .3v on the left

R27 is the 10K resistor between Q7 base / D5 and R36 etc, yes? The 'top' end should be 13.0V - is this just a measurement/transcription error?

D4 14.3v going in, 13.6v out, D5 13.6 in and 13.0 out.

Do any of these look like what they're supposed to be?

Apart from the minor notes above, yes.

The next stage is to measure the voltages on R31/R30/R29 (i.e the divider chain on Q7's collector), and also IC pins 12 and 14, and at each end of R14. (With the pitch bend joystick centred).

I would expect the voltage at the R31/R30 junction to match the IC pins and both ends of R14.

If this all looks OK, do you have a 'scope good for a few MHz to look around the oscillator itself? Perhaps that has a malfunction.

 

[Midphase]

OK, so if the supply voltage is 15.3V this leaves about 0.5V across tuning pots VR4 and VR5, which suggests they have a combined resistance of about 830 ohms. Are they at mid-travel? You should be able to tune it lower by increasing their resistance. If they're at maximum resistance perhaps there's something wrong there?

R27 is the 10K resistor between Q7 base / D5 and R36 etc, yes? The 'top' end should be 13.0V - is this just a measurement/transcription error?

The next stage is to measure the voltages on R31/R30/R29 (i.e the divider chain on Q7's collector), and also IC pins 12 and 14, and at each end of R14. (With the pitch bend joystick centred).

I would expect the voltage at the R31/R30 junction to match the IC pins and both ends of R14.

If this all looks OK, do you have a 'scope good for a few MHz to look around the oscillator itself? Perhaps that has a malfunction.

VR4 (noon position): Pin 1 = 15.3, Pin 2 = 15.3, Pin 3 = 15.4

VR5 trim (noon position): Pin 1 = 15.3, Pin 2 = 15.3, Pin 3 = 15

R27 I made a typo mistake, it's reading: 13.1V on the right side and .3V on the left.

R29: Right side = -15 on the dot, Left side = -9.6

R30: Right side = -6.3, Left side = -9.6

R31: Right side = -6.3, Left side = -1.3

R35: Right side = 13.7, Left side = 14.9

D2: Anode (right) side = -15 on the dot, Left side = -15.6 (this is a 4148 that I replaced the original diode with)

R14 = -5.7 at both ends (a bit weird considering it's a 1K resistor...but perhaps the Q2 transistor is compensating the voltage drop?).

IC6 (I assume that's the one you meant?) Pin 12 = -5.7 and pin 14 is also -5.7

Many thanks again, I am very appreciative for all of this help. Out of a couple of dozen synths I have now successfully repaired, this is by far the trickiest issue to pin down.

 

[Script]

This is the first DCO synth by Korg, I think.

The oscillator is Q5 and Q6 in the pic below, if I read it right.

C18 indeed looks a bit dirty(?) on the left side in the picture above?

 

[Voyager10]

The voltage look about right, there's 5.4V across R29, 3.3V on R30, 5.0V on R31, which is consistent with 0.6mA current through the chain, matching 1.2V on R35.

We might need to look at the oscillator next. The timing caps C18/C19 are the first thing, as @Script said. Maybe swap them for some cheap ceramic ones as a quick test, but the proper fix will need to polystyrene or similar with a low temperature coefficient.

Alternatively, if this has gone on for too long and you need to stop, maybe try increasing R35 enough to bring the tuning into range, say 3K3?

 

[Midphase]

The voltage look about right, there's 5.4V across R29, 3.3V on R30, 5.0V on R31, which is consistent with 0.6mA current through the chain, matching 1.2V on R35.

We might need to look at the oscillator next. The timing caps C18/C19 are the first thing, as @Script said. Maybe swap them for some cheap ceramic ones as a quick test, but the proper fix will need to polystyrene or similar with a low temperature coefficient.

Alternatively, if this has gone on for too long and you need to stop, maybe try increasing R35 enough to bring the tuning into range, say 3K3?

Yeah, those two polystyrene caps are a bit weird looking and have been nagging me a bit, although I suspect that brown residue might just be some adhesive residue that Korg used. I also have some film caps I could use in a jam.

Do you think if I replace the value of the Tune Trimpot to, say 10K instead of the current 1K that would also give me more range in the tuning adjustment to get to the right pitch?

The guy who owns this synth is trying to get his hands on another working Poly 61 that we might try swapping out the boards on and see if I can at least definitively narrow down which PCB is responsible for the pitch issue.

 

[Voyager10]

Yes, you could certainly make the trimpot a larger value, 5K or 10K. I'd be tempted to use a 10-turn one as it will be easier to set properly.