Vintage Synth Repair

Hi,

I'm trying to restore a Korg Polysix that has a dead voice. I replaced the memory battery which was in pretty good condition (no leak).

Before I check the schematics, does anybody knows which part tend to fail in the voice circuit.
Thank you!

Nicolas

Hi Nicolas,
I have found the SSM2044 filter chips to be problematic. I put tiny 
numbered labels on the chips and then swap them around to confirm the 
diagnosis. With the voice LED's it is easy to tell which voice is 
triggered. The envelope chips can sometimes be iffy as well but 
switching the VCA to gate mode will give you output if this is the case.
John.

Hi,

Thank you for replying.
I swapped the SSM2044 chip and the other chip on a IC socket but that did not fix the problem.

When I turn the EG Intensity knob full up I can hear sound from the defective voice at the same level as the other voices. I tried to find a trimpot for the EG Intensity but I coudn't find one. I guess there is a defective part in the circuit of that defective voice. I replaced the electrolytic capacitors but that did not help.

Any other ideas?

Thank you!

 Nicolas

 

--- In vintagesynthrepair@yahoogroups.com, John Henson <synthnerd@...> wrote:

>
> Hi Nicolas,
> I have found the SSM2044 filter chips to be problematic. I put tiny 
> numbered labels on the chips and then swap them around to confirm the 
> diagnosis. With the voice LED's it is easy to tell which voice is 
> triggered. The envelope chips can sometimes be iffy as well but 
> switching the VCA to gate mode will give you output if this is the case.
> John.
>

Hi,

Thank you for replying.
I swapped the SSM2044 chip and the other chip on a IC socket but that did not fix the problem.

When I turn the EG Intensity knob full up I can hear sound from the defective voice at the same level as the other voices. I tried to find a trimpot for the EG Intensity but I coudn't find one. I guess there is a defective part in the circuit of that defective voice. I replaced the electrolytic capacitors but that did not help.

Any other ideas?

Thank you!

 Nicolas

 

--- In vintagesynthrepair@yahoogroups.com, John Henson <synthnerd@...> wrote:

>
> Hi Nicolas,
> I have found the SSM2044 filter chips to be problematic. I put tiny 
> numbered labels on the chips and then swap them around to confirm the 
> diagnosis. With the voice LED's it is easy to tell which voice is 
> triggered. The envelope chips can sometimes be iffy as well but 
> switching the VCA to gate mode will give you output if this is the case.
> John.
>

Hi,

After investigating a little bit more on this Polysix, I found out I have a second dead voice.
I can get some sound from these voices but it depends of the VCF settings. The KBD track knob doesn't have any effect on the defective voices.

There was a little bit of corrosion on IC23 (14051) so I replaced it but that did not fix the problem.
When I check the outputs of this chip with an oscilloscope, I can see that the KBD Track knob has no effect on the outputs of the dead voices. 
 
I also replaced IC12 (14024) and IC20 (4558) but that did not help.

Any Polysix guru out there could help me?

Thanks !

Nicolas



--- In vintagesynthrepair@yahoogroups.com, "Nicolas" <nico678@...> wrote:

>
> Hi,
> 
> Thank you for replying.
> I swapped the SSM2044 chip and the other chip on a IC socket but that did not fix the problem.
> 
> When I turn the EG Intensity knob full up I can hear sound from the defective voice at the same level as the other voices. I tried to find a trimpot for the EG Intensity but I coudn't find one. I guess there is a defective part in the circuit of that defective voice. I replaced the electrolytic capacitors but that did not help.
> 
> Any other ideas?
> 
> Thank you!
> 
>  Nicolas
> 
>  
> 
> --- In vintagesynthrepair@yahoogroups.com, John Henson <synthnerd@> wrote:
> >
> > Hi Nicolas,
> > I have found the SSM2044 filter chips to be problematic. I put tiny 
> > numbered labels on the chips and then swap them around to confirm the 
> > diagnosis. With the voice LED's it is easy to tell which voice is 
> > triggered. The envelope chips can sometimes be iffy as well but 
> > switching the VCA to gate mode will give you output if this is the case.
> > John.
> >
>

Hi Nicolas,
I am not claiming to be a guru but I had a difficult Polysix here last 
weekend which couldn't read the Pot's correctly and did all manner of 
strange things as a consequence. Battery acid damage took out one of the 
multiplexers on KLM367, the CPU card.
Don't forget, most of the Sample and Holds for the Polysix are on the 
CPU board which is the one most likely to be damaged, and the keyboard 
filter tracking output voltage is on this (IC12 pin 7). Page 7 of the 
service manual is unusually clear for an old Korg manual. All the 
programmable Pot's outputs can be measured from the op-amp buffers on 
the CPU board (IC's 7 to 16, pins 1 and 7 on each) and they mostly swing 
from -5V to +5V, resonance is the exception swinging from 0V to 10V. It 
is probably worth checking that all these are correct before heading 
back to the voice card (KLM366).
Corrosion on IC23 is a little worrying, leaky batteries should not 
normally get that far, although the effects of leakage entirely depend 
on the physical attitude at which the keyboard was stored, as in was it 
on it's back, stored vertically either side down, upside down etc.
I had one here years ago where somebody replaced the bad battery with a 
cluster of Zinc Chloride batteries in a holder fixed to the right of the 
voice board, then stored it vertically pitch wheels to the ground and 
guess what, the NEW batteries leaked all over the Voice board.
I never could make this work properly, tuning was always bad, I had to 
buy a second-hand replacement board to make this machine work at quite 
an expense.
Do check all the Pot multiplexers, if all is good at least you know that 
setting up the basic test patch in manual mode is going to be effective, 
and will guide you more accurately towards the real problems.

Regards,
John.

www.retroactivesynth.com

If single voices are affected the problem never resides with the CPU board.
Problems with the CPU board affect all voices identically.

So we have various parameters that affect Cutoff. Most notably of course
cutoff/cutoff mod (this CV contains als the LFO modulation if applicable);
then we have the external filter control. Those two affect all voices
globally.
Then we have key track and EG intensity; those affect individual voices
(since the effect they have on a single voice depends also on the pressed
key and on the point the envelope has reached).

So if key track has no result for certain voices I would see if the
keyboard CV reaches this part of the circuit. This enters on pin 3 of IC
29. Check also the buffers between IC 29 and IC 25. Also, the buffers
behind IC 23 could also affect Cutoff in general.

Hi,

Thank you for replying.

I checked the ICs that you mentioned and they seems to be working fine.

What I don't understand is that the KBD Track signal enters at the input pin 3 of IC23 multiplexer and should come out at each output but it's just not there for 2 outputs.

Like I said, I already replace the buffer chip IC21 for those voices and that did not help.

Any other ideas?
Thank you!


Nicolas

--- In vintagesynthrepair@yahoogroups.com, Malte Rogacki <gacki@...> wrote:

>
> If single voices are affected the problem never resides with the CPU board.
> Problems with the CPU board affect all voices identically.
> 
> So we have various parameters that affect Cutoff. Most notably of course
> cutoff/cutoff mod (this CV contains als the LFO modulation if applicable);
> then we have the external filter control. Those two affect all voices
> globally.
> Then we have key track and EG intensity; those affect individual voices
> (since the effect they have on a single voice depends also on the pressed
> key and on the point the envelope has reached).
> 
> So if key track has no result for certain voices I would see if the
> keyboard CV reaches this part of the circuit. This enters on pin 3 of IC
> 29. Check also the buffers between IC 29 and IC 25. Also, the buffers
> behind IC 23 could also affect Cutoff in general.
>

Is the signal actually entering IC23? Have you checked with a scope if the
CV for all six voices is there?

What enters pin3 of IC23 is not just the keyboard track - it's the whole
sum of all Cutoff-affecting parameters.
I find it not unlikely that other parameters are to blame here. A likely
candidate: Envelopes.

Yes, the signal is entering IC23 and I've got CV for all the voices.
The envelopes work well on all the voice well on all the voices.



--- In vintagesynthrepair@yahoogroups.com, Malte Rogacki <gacki@...> wrote:

>
> Is the signal actually entering IC23? Have you checked with a scope if the
> CV for all six voices is there?
> 
> What enters pin3 of IC23 is not just the keyboard track - it's the whole
> sum of all Cutoff-affecting parameters.
> I find it not unlikely that other parameters are to blame here. A likely
> candidate: Envelopes.
>

This calls for a more thorough debugging.

Set the front panel controls as follows:

- Cutoff 0
- Resonance 0
- EG Intensity 0 (center)
- Key Track 0

Now check with a scope the situation at pin 3 of IC23. You should get a
proper DC voltage of +5V. If you move the Cutoff control to max the voltage
should change to -5V. Same should apply to the outputs of IC23.

Next we check the multiplexing of the keyboard CV. For this we monitor pin3
of IC25. Put the synth in Unison mode and press the highest note on the
keyboard. You should get a reading on the scope that looks like DC with a
single periodic 0v "dip".
The keyboard CVs for the six channels start immediately after this "dip".
You can check if they correspond to the keys pressed.

Now we set Key Track to Full, Cutoff to zero, again use Unison mode and
press the highest key. Check at pin3 of IC23. You should basically get a
"mirrored" reading from the one at IC25 - a weak DC signal with a single 5V
"peak". Again the channels would start immediately after this peak. Check
if the signals there make sense. By playing different notes you can check
if the voltages on the input (pin 3) correspond to the voltages on the
output of IC23 or if - for example - the voltage for channel 1 shows up on
the wrong output pin.

Hi,

You're a genius.
I followed you're instructions and I found out there was no signal on pin 1 of IC25 but there was a signal on pin 1 IC29
I replaced the 072 opamp chip in between those 2 chips an now everything is back to normal.

Thank you so much for your help!

Regards,

Nicolas

--- In vintagesynthrepair@yahoogroups.com, Malte Rogacki <gacki@...> wrote:

>
> This calls for a more thorough debugging.
> 
> Set the front panel controls as follows:
> 
> - Cutoff 0
> - Resonance 0
> - EG Intensity 0 (center)
> - Key Track 0
> 
> Now check with a scope the situation at pin 3 of IC23. You should get a
> proper DC voltage of +5V. If you move the Cutoff control to max the voltage
> should change to -5V. Same should apply to the outputs of IC23.
> 
> Next we check the multiplexing of the keyboard CV. For this we monitor pin3
> of IC25. Put the synth in Unison mode and press the highest note on the
> keyboard. You should get a reading on the scope that looks like DC with a
> single periodic 0v "dip".
> The keyboard CVs for the six channels start immediately after this "dip".
> You can check if they correspond to the keys pressed.
> 
> Now we set Key Track to Full, Cutoff to zero, again use Unison mode and
> press the highest key. Check at pin3 of IC23. You should basically get a
> "mirrored" reading from the one at IC25 - a weak DC signal with a single 5V
> "peak". Again the channels would start immediately after this peak. Check
> if the signals there make sense. By playing different notes you can check
> if the voltages on the input (pin 3) correspond to the voltages on the
> output of IC23 or if - for example - the voltage for channel 1 shows up on
> the wrong output pin.
>