Discussion about the Korg PolySix synthesizer group photo

Yahoo Groups archive

Discussion about the Korg PolySix synthesizer

Index last updated: 2026-04-28 23:40 UTC

Thread

Trying to Restore a P6.

Trying to Restore a P6.

2016-08-17 by Jim Wright

Hi,

New member here who has been struggling for years to get a P6 working that was not stored well, and had the typical battery leakage.

It's in good shape, without extensive damage to the KLM 367 board, but all of my attempts to get it working correctly have failed.

Symptoms are:

1. Does not power up properly with A and 1 buttons lit.

2. It is possible to get sound from it, but adjusting any parameters from the MG, EG ,VCO, VCF, Effects or VGA sections causes random Bank, Programmer, Write, and Manual LEDs to light, which produces patches, noises, and sometimes nothing. I can find no pattern to this, except that D4 usually seems to produce a patch of some kind.

3. The Arpeggiator and Key Assign modes seem to work just fine.

I recently got the mod wheels, and tuning functions working again when I found problems with the traces associated with CN05 pin 12, and its connection to R10 and other components.

Also pulled VRs 2, 3, 4, 5, 6, and 7, and found more corrosion under them, but fixing that did not seem to change it's behavior much.

PSU voltages check out fine and are adjustable.

I have replaced IC 31, and checked all traces to it, as well as IC30 (though I did not replace IC30), and most other ICs on the board. All bad traces I can find were repaired.

Battery was replaced with coin cell holder, as per the Old Crow site.

Based on reading here, I have repeatedly checked all data, and "P" lines for continuity and have not found any problems or shorts.

Everything connected to IC 22 has been checked several times compared to the schematic, and I have noted the errors found in the schematic, as listed on this group.

I would also add that it appears that R83 should switch positions with VR6 on the schematic.

Board has been cleaned and brushed with alcohol to remove any corrosion found.

Multi-pin connectors to the other boards have been tested for continuity and found to be fine.


My questions would be:

Can anybody recommend a good way to test ICs 18 and 19? It seems most of the wrong-working controls are connected to these two ICs in some way. They seem to have proper power and ground connections.

Same goes for IC 30. Do the symptoms make it suspect? Should I replace it?

; I've seen folks here recommend the D/A calibration for people with similar problems, but I can't get it to step 2 of the Reset adjustment, where Bank A and number 1 LEDs are lit.

I'm happy to hear any suggestions of other tests to try, or places to look for a solution. Would love to get this working, and start using it. I DO have an oscilloscope, but am not well versed on it's use, though I am happy to learn.

Thanks in advance for keeping this group going, and any responses.

Jim


Sent from Outlook

Re: [PolySix] Trying to Restore a P6.

2016-08-17 by Bob Grieb

Welcome.   In manual mode, the voltage at each output of the two 4051's 18 and 19 should be adjustable
by changing the pot on the front panel that corresponds to it.   But that assumes the CPU is able to read 
the pot, which requires the signals that control the muxes on the panel to be connected and working,
and also the DAC and its comparator must be working.  Getting the DAC calibration working involves
much less code and hardware than normal operation, which is why I sometimes suggest it.  If you can't 
get that to work, then I would fix it before worrying about 18 and 19.   The scope is a really powerful
tool.   I would suggest that you use this as an opportunity to learn more about using it.  One technique 
that can work with digital circuitry is just probing lots of digital signals and looking for floats (trace eaten)
or intermediate voltage levels (could indicate a short).  You could also look at the DAC output and see if
it's trying to convert the front panel controls.   I uploaded some photos a while back that may help.

Bob

--------------------------------------------
On Wed, 8/17/16, Jim Wright tchgtrst@hotmail.com [PolySix] <PolySix@yahoogroups.com> wrote:
Show quoted textHide quoted text
 Subject: [PolySix] Trying to Restore a P6.
 To: "PolySix@yahoogroups.com" <PolySix@yahoogroups.com>
 Date: Wednesday, August 17, 2016, 3:43 PM
 
 
  
 
 
 
   
 
 
     
       
       
       
 
 
 
 Hi,
    New member here who has been struggling for years to
 get a P6 working that was not stored well, and had the
 typical battery leakage.
 It's in good shape, without extensive damage to the
 KLM 367 board, but all of my attempts to get it working
 correctly have failed.
     Symptoms are:
 1. Does not power up properly with A and 1
 buttons lit.
  2.  It is possible to get sound from it, but adjusting
 any parameters from the MG, EG ,VCO, VCF, Effects or VGA
 sections causes random Bank, Programmer, Write, and Manual
 LEDs to light, which produces patches, noises, and
 sometimes nothing. I can find no
  pattern to this, except that D4 usually seems to produce a
 patch of some kind.
 3. The Arpeggiator and Key Assign modes seem to work just
 fine.
     
    I recently got the mod wheels, and tuning functions
 working again when I found problems with the traces
 associated with CN05 pin 12, and its connection to R10 and
 other components.
    Also pulled VRs 2, 3, 4, 5, 6, and 7, and found more
 corrosion under them, but fixing that did not seem to change
 it's behavior much. 
 PSU voltages check out fine and are adjustable.
   I have replaced IC 31, and checked all traces to it,
 as well as IC30 (though I did not replace IC30), and most
 other ICs on the board. All bad traces I can find were
 repaired.
 Battery was replaced with coin cell holder, as per the
 Old Crow site.
 Based on reading here, I have repeatedly checked all
 data, and "P" lines for continuity and have not
 found any problems or shorts.
 Everything connected to IC 22 has been checked several
 times compared to the schematic, and I have noted the errors
 found in the schematic, as listed on this group.
 I would also add that it appears that R83 should switch
 positions with VR6 on the schematic.
 Board has been cleaned and brushed with alcohol to remove
 any corrosion found.
 Multi-pin connectors to the other boards have been tested
 for continuity and found to be fine.
 
 
 
 My questions would be: 
      Can anybody recommend a good way to test ICs 18
 and 19? It seems most of the wrong-working controls are
 connected to these two ICs in some way. They seem to have
 proper power and ground connections. 
    Same goes for IC 30. Do the symptoms make it
 suspect? Should I replace it? 
    I've seen folks here recommend the D/A
 calibration for people with similar problems, but I
 can't get it to step 2 of the Reset adjustment, where
 Bank A and number 1 LEDs are lit.
    I'm happy to hear any suggestions of other tests
 to try, or places to look for a solution. Would love to get
 this working, and start using it. I DO have an oscilloscope,
 but am not well versed on it's use, though I am happy to
 learn.
 Thanks in advance for keeping this group going, and any
 responses.
 Jim
 
 
 
 
 Sent from Outlook
 
 
 
 
 
 
 
 
     
      
 
     
     
 
 
 
 #yiv0166812404 #yiv0166812404 --

Re: Trying to Restore a P6.

2016-08-20 by 7zlwncpbis4x22jwvkd4oc5hzu2njcsjeup42is6@yahoo.com

OK...So stop poking around it with a multi-meter out of circuit, and start poking at it with the scope in circuit and active.
Are the images to which you refer the amber-traced scope images found in the Photos section? They don't seem to be associated with any messages. Is there any other explanation besides the captions? Are these the way the signals should look at these points?
And it seems you are saying that I have bigger fish to fry than ICs 18 and 19, since my P6 can't
get thru the initial stages of calibration.
Am I correct to assume that IC 33 is the DAC, and IC 27 is it's comparator? I'm guessing this because VR6 and 7 connect to this pair, and adjust the DA functions.
And if that guess is correct, is the output of the DAC pin 1 on IC 27, which is unlabeled on my schematic, but connects to TP2?
My scope is out and dusted off, but I'm not quite sure what I'm looking for. Digital circuits are a new experience for me, and I haven't fully grasped this schematic.
I'm assuming that since I seem to have checked all traces with a multi-meter, and have found no shorts or any more broken traces, that my problem is a blown IC. None of them get overly warm in operation, though.
Thanks for your time.

Re: [PolySix] Trying to Restore a P6.

2016-08-20 by Patrick Belcher

Start simple, when you edit a sound, does it recall on boot? Memory failures, can't remember ic no, are common cause of weirdness. Test functionality before electronic troibleshooting, it's the biggest pointer in my experience
Show quoted textHide quoted text

On Saturday, 20 August 2016, 7zlwncpbis4x22jwvkd4oc5hzu2njcsjeup42is6@yahoo.com [PolySix] <PolySix@yahoogroups.com> wrote:

OK...So stop poking around it with a multi-meter out of circuit, and start poking at it with the scope in circuit and active.

Are the images to which you refer the amber-traced scope images found in the Photos section? They don't seem to be associated with any messages. Is there any other explanation besides the captions? Are these the way the signals should look at these points?
And it seems you are saying that I have bigger fish to fry than ICs 18 and 19, since my P6 can9;t
get thru the initial stages of calibration.
Am I correct to assume that IC 33 is the DAC, and IC 27 is it's comparator? I'm guessing this because VR6 and 7 connect to this pair, and adjust the DA functions.
And if that guess is correct, is the output of the DAC pin 1 on IC 27, which is unlabeled on my schematic, but connects to TP2?
My scope is out and dusted off, but I'm not quite sure what I'm looking for. Digital circuits are a new experience for me, and I haven't fully grasped this schematic.
I'm assuming that since I seem to have checked all traces with a multi-meter, and have found no shorts or any more broken traces, that my problem is a blown IC. None of them get overly warm in operation, though.
Thanks for your time.

Re: Trying to Restore a P6.

2016-08-20 by 7zlwncpbis4x22jwvkd4oc5hzu2njcsjeup42is6@yahoo.com

Hi Patrick,
It does not recall on boot.
Trying to edit any sound causes weirdness. The only way to get a working sound is by punching random buttons, or turning knobs until one happens by. You also get noises that sound the same on all keys, and sometimes just constant static and/or noise.
For example, turning the VCA Attenuator switch will cause it to go through several different settings, some which have sounds and some not. It might also light up the Manual or Write buttons. The likelihood of getting an actual patch is about 10%. I see no repeating pattern.
The bank and program buttons do not work reliably.
It will not go into Manual mode, and does not power up the same each time. It seldom powers up with A-1 lit, as I believe it is supposed to do (please correct me if I'm wrong), and when it does, changing almost any parameter will cause randomness.
After putting a scope on it, I notice the same sine shape on all test points in the KLM-367 (including ground), except TP2, which has the same amplitude sine, but with smaller square wave distortions.
After looking more closely, I see that the signal remains a pure sine until it leaves IC 27 via pin 1. It comes out of IC 33 pin 4 pure and enters IC 27 clean on pins 2, 6, and 7, as it should according to the schematic.
Should I be suspicious? Is any of that normal or does it indicate problems with IC 27 or associated components?
Thanks again to all for help. I'm determined to fix this, but have little experience with digital circuits.

Re: [PolySix] Trying to Restore a P6.

2016-08-21 by Patrick Belcher

Ok, you've tested the psu I understand, all things considered time cost and effort wise, a careful klm367 rebuild may be the way. Jed and Andrew didn't embark on that project without good reason, and plenty of us are grateful benefactors. Tough synths to troubleshoot!
Show quoted textHide quoted text

On Sunday, 21 August 2016, 7zlwncpbis4x22jwvkd4oc5hzu2njcsjeup42is6@yahoo.com [PolySix] <PolySix@yahoogroups.com> wrote:

Hi Patrick,

It does not recall on boot.
Trying to edit any sound causes weirdness. The only way to get a working sound is by punching random buttons, or turning knobs until one happens by. You also get noises that sound the same on all keys, and sometimes just constant static and/or noise.
For example, turning the VCA Attenuator switch will cause it to go through several different settings, some which have sounds and some not. It might also light up the Manual or Write buttons. The likelihood of getting an actual patch is about 10%. I see no repeating pattern.
The bank and program buttons do not work reliably.
It will not go into Manual mode, and does not power up the same each time. It seldom powers up with A-1 lit, as I believe it is supposed to do (please correct me if I'm wrong), and when it does, changing almost any parameter will cause randomness.
After putting a scope on it, I notice the same sine shape on all test points in the KLM-367 (including ground), except TP2, which has the same amplitude sine, but with smaller square wave distortions.
After looking more closely, I see that the signal remains a pure sine until it leaves IC 27 via pin 1. It comes out of IC 33 pin 4 pure and enters IC 27 clean on pins 2, 6, and 7, as it should according to the schematic.
Should I be suspicious? Is any of that normal or does it indicate problems with IC 27 or associated components?
Thanks again to all for help. I'm determined to fix this, but have little experience with digital circuits.

Re: Trying to Restore a P6.

2016-08-21 by 7zlwncpbis4x22jwvkd4oc5hzu2njcsjeup42is6@yahoo.com

--9nlWnTS9ocNmxdhsK3FKVLWiM6a5jwOaA8HvrLV
Content-Type:  text/plain;  charset=utf-8 
Content-Transfer-Encoding: base64




--9nlWnTS9ocNmxdhsK3FKVLWiM6a5jwOaA8HvrLV
Content-Type:  text/html;  charset=utf-8 
Content-Transfer-Encoding: base64




--9nlWnTS9ocNmxdhsK3FKVLWiM6a5jwOaA8HvrLV--

Re: Trying to Restore a P6.

2016-08-21 by 7zlwncpbis4x22jwvkd4oc5hzu2njcsjeup42is6@yahoo.com

It may come to that, if they still have boards. I have checked all the connections on this one many times, and can't find any other problems.
I did find the photos and files that Bob was referring to, and will spend some time getting familiar with the scope.

Re: Trying to Restore a P6.

2016-08-21 by 7zlwncpbis4x22jwvkd4oc5hzu2njcsjeup42is6@yahoo.com

OK...Please disregard my earlier statements about what I was seeing with the scope. I was not using it properly.
Now that I am, I can see what I think Bob was talking about. With the scope attached to TP2, I can see traces moving when I move the front panel controls that are causing the problems. I also see the patterns changing when Bank and Program buttons are pushed, as well as these patterns changing as I move the front panel controls, and at the same time can hear/see the machine lose it's patch, and the pattern changes.
However, patches do seem to repeat the same pattern, when they are pressed, even if that pattern produces no sound.
I will spend some time today looking more closely at Bob's images and text, and see what I can figure out.

Re: Trying to Restore a P6.

2016-08-24 by 7zlwncpbis4x22jwvkd4oc5hzu2njcsjeup42is6@yahoo.com

After looking at Bob's files, and checking P20-22, I find that almost everything seems to be in order.
However, P23 is fine until it enters IC23, but when it exits via pins 4 and 10, I get two differing waveforms. One looks much like Bob's example of the P23 trace, basically a square wave, but the other is not the same.
I understand that the original P23 wave is being modulated by the wave coming from P16, but shouldn't the outputs on 4 and 10 look the same?
I can see the clock clearly at TP4, and as mentioned above, can see the pattern change on TP2 as I move front panel controls.
Any tips, clues, or suggestions are appreciated.
Thanks.

Re: [PolySix] Re: Trying to Restore a P6.

2016-08-24 by Bob Grieb

The signals at pins 4 and 10 should both be nice low or high level digital 
signals.   They are both logical combinations of P16 and P23, but are different
signals, so only one would be low at a time, to enable either IC18 or 19.

--------------------------------------------
On Tue, 8/23/16, 7zlwncpbis4x22jwvkd4oc5hzu2njcsjeup42is6@yahoo.com [PolySix] <PolySix@yahoogroups.com> wrote:
Show quoted textHide quoted text
 Subject: [PolySix] Re: Trying to Restore a P6.
 To: PolySix@yahoogroups.com
 Date: Tuesday, August 23, 2016, 10:39 PM
 
 
  
 
 
 
   
 
 
     
       
       
       After looking at Bob's files, and checking
 P20-22, I find that almost everything seems to be in
 order.However, P23 is fine until it enters IC23,
 but when it exits via pins 4 and 10, I get two differing
 waveforms. One looks much like Bob's example of the P23
 trace, basically a square wave, but the other is not the
 same. I understand that the original P23 wave is
 being modulated by the wave coming from P16, but
 shouldn't the outputs on 4 and 10 look the
 same?    I can see the clock clearly at TP4,
 and as mentioned above, can see the pattern change on TP2 as
 I move front panel controls.   Any tips, clues,
 or suggestions are
 appreciated. Thanks.

Re: [PolySix] Re: Trying to Restore a P6.

2016-08-24 by Bob Grieb

BTW, a square wave is a digital signal that is low half of the time
and high half of the time.  A digital signal that is either low or high is 
not a square wave.   It's just a nice clean digital signal.    

Bob


--------------------------------------------
On Tue, 8/23/16, 7zlwncpbis4x22jwvkd4oc5hzu2njcsjeup42is6@yahoo.com [PolySix] <PolySix@yahoogroups.com> wrote:
Show quoted textHide quoted text
 Subject: [PolySix] Re: Trying to Restore a P6.
 To: PolySix@yahoogroups.com
 Date: Tuesday, August 23, 2016, 10:39 PM
 
 
  
 
 
 
   
 
 
     
       
       
       After looking at Bob's files, and checking
 P20-22, I find that almost everything seems to be in
 order.However, P23 is fine until it enters IC23,
 but when it exits via pins 4 and 10, I get two differing
 waveforms. One looks much like Bob's example of the P23
 trace, basically a square wave, but the other is not the
 same. I understand that the original P23 wave is
 being modulated by the wave coming from P16, but
 shouldn't the outputs on 4 and 10 look the
 same?    I can see the clock clearly at TP4,
 and as mentioned above, can see the pattern change on TP2 as
 I move front panel controls.   Any tips, clues,
 or suggestions are
 appreciated. Thanks.   
 
     
      
 
     
     
 
 
 
 #yiv9088296080 #yiv9088296080 --

Re: Trying to Restore a P6.

2016-08-24 by 7zlwncpbis4x22jwvkd4oc5hzu2njcsjeup42is6@yahoo.com

I feel like I'm on to something here.
The signal coming from pin 4, IC 23 is smeared on my scope, especially when it is compared to the pin 10 signal. Not an inverse or any relation whatsoever.
Not only that, but I looked at the signals coming from the other side of IC 19, and I found that I could see the voltage change on all connections from IC 19 to such things as MG Level, MG Decay, KBD track, etc. The signal goes up about 70% on all of them, and then falls back to the lowest voltage. MG Freq, was the only one that didn't fall back to the lowest voltage.
When I looked at that side of IC 18....nothing on all parameters.
Did the corrosion get to IC 23? There are no external components in that part of the circuit to account for the smearing.
Thanks for the pointers, Bob. I'm glad I got out my old analog scope, and am learning from it.

Re: [PolySix] Re: Trying to Restore a P6.

2016-08-24 by Bob Grieb

One thing to mention...   When you see a signal that looks wrong at the output of
a logic gate, the first thing to do is look at the inputs to that gate.   A floating or 
incorrect input can cause the output to be wrong.   Also, if the gate is being 
loaded down, that can affect its output.   To check for a load issue, with the synth 
powered off, measure from pin 4 to ground, and also from pin 5 to ground with a 
multimeter in the ohms position.   Keep the polarity of the leads the same for 
both measurements.  The readings should be similar, since both feed the INH inputs
of 4051 chips. 

IC23 contains four gates.   Of course, if one is bad, then the whole chip would be 
replaced.  Just saying that each of the four gates has two inputs and one output 
and they act independently.   
  
--------------------------------------------
On Wed, 8/24/16, 7zlwncpbis4x22jwvkd4oc5hzu2njcsjeup42is6@yahoo.com [PolySix] <PolySix@yahoogroups.com> wrote:
Show quoted textHide quoted text
 Subject: [PolySix] Re: Trying to Restore a P6.
 To: PolySix@yahoogroups.com
 Date: Wednesday, August 24, 2016, 4:10 AM
 
       
       I feel like I'm on to something
 here.The signal coming from pin 4, IC 23 is smeared
 on my scope, especially when it is compared to the pin 10
 signal. Not an inverse or any relation
 whatsoever.     Not only that, but I looked at
 the signals coming from the other side of IC 19, and I found
 that I could see the voltage change on all connections from
 IC 19 to such things as MG Level, MG Decay, KBD track, etc.
 The signal goes up about 70% on all of them, and then falls
 back to the lowest voltage. MG Freq, was the only one that
 didn't fall back to the lowest voltage. 
  When I looked at that side of IC 18....nothing on all
 parameters.   Did the corrosion get to IC 23?
 There are no external components in that part of the circuit
 to account for the smearing.Thanks for the
 pointers, Bob. I'm glad I got out my old analog scope,
 and am learning from it.

Re: [PolySix] Re: Trying to Restore a P6.

2016-08-24 by Bob Grieb

Oops.   I meant pins 4 and 10, not 4 and 5.
Sorry.


--------------------------------------------
On Wed, 8/24/16, Bob Grieb bobgrieb@yahoo.com [PolySix] <PolySix@yahoogroups.com> wrote:
Show quoted textHide quoted text
 Subject: Re: [PolySix] Re: Trying to Restore a P6.
 To: PolySix@yahoogroups.com
 Date: Wednesday, August 24, 2016, 9:48
       
       
       One thing to mention...   When you see a signal
 that looks wrong at the output of
 
 a logic gate, the first thing to do is look at the inputs to
 that gate.   A floating or 
 
 incorrect input can cause the output to be wrong.   Also, if
 the gate is being 
 
 loaded down, that can affect its output.   To check for a
 load issue, with the synth 
 
 powered off, measure from pin 4 to ground, and also from pin
 5 to ground with a 
 
 multimeter in the ohms position.   Keep the polarity of the
 leads the same for 
 
 both measurements.  The readings should be similar, since
 both feed the INH inputs
 
 of 4051 chips. 
 
 IC23 contains four gates.   Of course, if one is bad, then
 the whole chip would be 
 
 replaced.  Just saying that each of the four gates has two
 inputs and one output 
 
 and they act independently.

Re: Trying to Restore a P6.

2016-08-24 by 7zlwncpbis4x22jwvkd4oc5hzu2njcsjeup42is6@yahoo.com

The first thing I checked on IC 23 was the inputs, and they checked fine, but will check them again before purchasing a new IC. And the two outputs come within 2 thousandths of an ohm resistance of each other.
Do I need a replacement with the exact suffix letters? I see 14011 BCPs on E-bay, whereas the chip in the P6 has only BP as the suffix.
If any other points I should check occur to you, please let me know.
Thanks again for the pointers. Very excited that I may have a handle on this.

Re: [PolySix] Re: Trying to Restore a P6.

2016-08-24 by Bob Grieb

Different manufacturers used different numbering schemes.  
C might mean "consumer" temperature range.   Or it might not.
You can check each mfr's data sheet to see exactly what the letters
mean for that mfr.
With the 4000 series chips, the important thing is whether or not
you want a B (buffered) part.   The original chips did not have a B
at the end, and were not "buffered" so they had less drive capability.
After some years, most parts were available in B versions, and those
are the parts you normally find.   Also, Motorola added a "1" in front, 
so another company's CD4008 might be their 14008, just to add to the 
confusion.   To make things even worse, later on, some of the 4000-series
parts were made in 5V-only versions with numbers like 74HC4017.
Many of the 4000-series parts are still available brand new, BTW, 
from Digikey and Mouser, so you don't have to get them from ebay.

To minimize the wear and tear on the pcb, the preferred technique
for removing a chip is to cut the leads right at the body with small
"nippy cutters" or diagonals.   Then remove the body and desolder 
each lead by itself while pulling it out with tweezers.   Then clean 
both sides of the pcb with isopropyl, install an IC socket, and the 
new chip.   Best to use a socket as these old pc bds can't take a lot
of de-soldering without damage.  If the new chips fails, you want to 
be able to replace it without messing with the pcb traces again.
There may be special cases where you wouldn't want to use an IC
socket, like some kind of high-frequency chip, but most chips in an 
analog synth should be fine in a socket.


--------------------------------------------
On Wed, 8/24/16, 7zlwncpbis4x22jwvkd4oc5hzu2njcsjeup42is6@yahoo.com [PolySix] <PolySix@yahoogroups.com> wrote:
Show quoted textHide quoted text
 Subject: [PolySix] Re: Trying to Restore a P6.
 To: PolySix@yahoogroups.com
 Date: Wednesday, August 24, 2016, 3:06 PM
 
       The first thing I checked on IC 23 was the inputs,
 and they checked fine, but will check them again before
 purchasing a new IC. And the two outputs come within 2
 thousandths of an ohm resistance of each
 other.Do I need
 a replacement with the exact suffix letters? I see 14011
 BCPs on E-bay, whereas the chip in the P6 has only BP as the
 suffix.
 If any other
 points I should check occur to you, please let me
 know.Thanks again for the pointers. Very excited
 that I may have a handle on this.

Re: Trying to Restore a P6.

2016-08-24 by 7zlwncpbis4x22jwvkd4oc5hzu2njcsjeup42is6@yahoo.com

Digikey showed them all as obsolete, but Mouser had plenty, and some are on their way.
I may already have a 14-pin machined socket, and am on familiar territory with replacing ICs, as well as having a passable temp-controlled iron.
Mainly a hobbyist, I've been able to repair such things as tube and solid state guitar amp circuits, outboard studio gear, as well as rebuild a few mixing desks and their power supplies. I've done mic mods and built tube mics from scratch, but this thing has been a real lesson for me.
I truly appreciate your help, and have learned a lot in several areas, which is why I do this. Your photos and text helped me understand the circuit better.
Let's hope this fixes most of my problems. tho I know I have a bit of keyboard contact cleaning to do. That can wait until I'm sure the rest is working.
Will continue to poke and familiarize myself with the scope.
Much thanks.

Re: Trying to Restore a P6.

2016-08-30 by 7zlwncpbis4x22jwvkd4oc5hzu2njcsjeup42is6@yahoo.com

So...new IC 23 installed. Trace looks like it should now at pin 6, IC 18, the INH input, and I can see the outputs reacting as they should on all but two parameters.
However, the symptoms remain pretty much the same.
Which leads me to the two outputs that are not working correctly. Pin 1 on IC 18 has the exact same voltage as the +5 power rail input into the IC (pin 16, +4.98V). This voltage shows up on pins 1, 2, and 3 of IC 10 also. This is supposed to be the ATTACK control. My guess is that IC 18 has been corroded, but it seems I should replace IC 10 also. 4558s are cheap.
And pin 1 on IC 14 does not show any response from the MG Delay control. I see the signal at it's output on IC 19 (pin 1) and it's input to IC 14 (pin 2), but all I see on pin 1 is a constant +14.51 V. Can't quite figure out where that is coming from, but perhaps I don't understand the circuit. I see a wavering voltage on pin 3, which I imagine is correct due to the Delay function, and there is a negative voltage on the far side of D12, as it appears there should be.
Again, 4558s are cheap. Should I just replace all 3 ICs? Looks like Digi-key still has some 16 DIP 14051 ICs.

One other thing I notice on the working control outputs (muxes?). Most of them get to a certain voltage (somewhere around 70% of their potential), and then fall back to zero with an audible click, which also seems to change all settings. This can be seen on the trace at TP2 also. It seems like this indicates a short somewhere once a certain voltage is reached, but there are two or three noticeable exceptions (MG SP, RESO, and REL) which do not do this.

All of which makes me wonder if I should just be starting with a new circuit board, as maybe this one is just too corroded. but I would like to try to continue fixing this one, and any help is appreciated.
Thanks.

Re: Trying to Restore a P6.

2016-08-30 by tlule@gmx.de

Man, what a nightmare you have there. And you seem to be not the only one.


I was wondering about the following point.


It should be 'quite easy' to set up a jig that replaces the programmer CPU, and-or the assigner CPU with a modern CPU. It just needs enough pins to look at each of the 8048 pins in a row. (e.g. STM32) , plus LCD interface, or serial interface.


The CPU would run a series of 'sanity checks'. If you made that open source, everybody could contribute further program parts to execute. some parts could ask the user to turn a pot, others ask for keypresses. Each part would then report for good-intermittent-missing-stuck or whatever contact it finds on that pin.


That would make troubleshooting a piece of cake compared to the hunting of the snark you are up to.


what do you think guys out there?



Re: [PolySix] Re: Trying to Restore a P6.

2016-08-30 by Bob Grieb

An interesting idea, and one that I have actually been thinking about over the past few months
after reading posts about debugging KLM-367's.    I was considering making something 
that would plug in in place of the 8048, and allow the user to step through different tests.
I was thinking to test the LEDs first, so that they could be used to display results once they 
were working.   But so far I haven't done any actual work on this.  Just an idea.

Another idea would be to make something that would have a USB interface, and control it
from a PC, so that it could be a more general purpose 8048 circuit tester.

Bob


--------------------------------------------
On Tue, 8/30/16, tlule@gmx.de [PolySix] <PolySix@yahoogroups.com> wrote:
Show quoted textHide quoted text
 Subject: [PolySix] Re: Trying to Restore a P6.
 To: PolySix@yahoogroups.com
 Date: Tuesday, August 30, 2016, 2:06 PM
 
 
  
 
 
 
   
 
 
     
       
       
       Man, what a nightmare you have there. And
 you seem to be not the only one.
 I was wondering about the following
 point. 
 It should be
 'quite easy' to set up a jig that replaces the
 programmer CPU, and-or the assigner CPU with a modern CPU.
 It just needs enough pins to look at each of the 8048 pins
 in a row.  (e.g. STM32) , plus LCD interface, or serial
 interface.
 The CPU would run a
 series of 'sanity checks'. If you made that open
 source, everybody could contribute further program parts to
 execute. some parts could ask the user to turn a pot, others
 ask for keypresses. Each part would then report for
 good-intermittent-missing-stuck or whatever contact it finds
 on that pin. 
 That would make
 troubleshooting a piece of cake compared to the hunting of
 the snark you are up to. 
 what
 do you think guys out there?
 
 
 
     
      
 
     
     
 
 
 
 #yiv5784458769 #yiv5784458769 --

Re: [PolySix] Re: Trying to Restore a P6.

2016-08-30 by Bob Grieb

Sounds like you are making progress, maybe slower than you had 
hoped.   IC10 pins 1,2,3 are configured as a "voltage follower".
This is a buffer with gain of 1.   That means whatever you have at
the input should appear at the output, as you are seeing.

There are several simple rules you can use for analyzing op amp
circuits:

  1) If the op amp is in a feedback circuit, the output will try to
make the two input voltages equal.   (This makes the operation of
the voltage follower rather obvious)  

  2) The input current to op amps is very low, and can be considered
to be 0.   In other words, current flows in the components around 
the op amp, but not into the op amp inputs.   This is just for the inputs,
of course.   The output of an op amp must source or sink current to drive
its load.

Using these two rules and ohm's law, you can figure out lots of op amp circuits.

Note that in some cases, like IC14 pins 1,2, and 3, the op amp is not being used
in a feedback circuit.   The 4.7M resistor R80 provides hysteresis, and is typical
for a "comparator" circuit.   In this kind of circuit, if the "+" input is more positive
than the "-" input, the output of the op amp will go as high as it can.   For many op amps
this is maybe 1-2 volts below the positive supply.   This is what you are seeing.
If the - input is more positive than the + input, then the output of the op amp will go 
as low as it can, approaching its negative supply rail.  If the output of the op amp 
behaves this way as the inputs are varied, then it's probably OK.

   Bob


--------------------------------------------
On Tue, 8/30/16, 7zlwncpbis4x22jwvkd4oc5hzu2njcsjeup42is6@yahoo.com [PolySix] <PolySix@yahoogroups.com> wrote:
Show quoted textHide quoted text
 Subject: [PolySix] Re: Trying to Restore a P6.
 To: PolySix@yahoogroups.com
 Date: Tuesday, August 30, 2016, 12:30 PM
       
       So...new IC 23 installed. Trace looks like it
 should now at pin 6, IC 18, the INH input, and I can see the
 outputs reacting as they should on all but two
 parameters. However, the symptoms remain pretty
 much the same.Which leads me to the two outputs
 that are not working correctly. Pin 1 on IC 18 has the exact
 same voltage as the +5 power rail input into the IC (pin 16,
 +4.98V). This voltage shows up on pins 1, 2, and 3 of IC 10
 also. This is supposed to be the ATTACK control. My guess is
 that IC 18 has been corroded, but it seems I should replace
 IC 10 also. 4558s are cheap.And pin 1 on IC 14
 does not show any response from the MG Delay control. I see
 the signal at it's output on IC 19 (pin 1) and it's
 input to IC 14 (pin 2), but all I see on pin 1 is a constant
 +14.51 V. Can't quite figure out where that is coming
 from, but perhaps I don't understand the circuit. I see
 a wavering voltage on pin 3, which I imagine is correct due
 to the Delay function, and there is a negative voltage on
 the far side of D12, as it appears there should
 be.Again, 4558s are cheap. Should I just replace
 all 3 ICs? Looks like Digi-key still has some 16 DIP 14051
 ICs. 
 One other thing
 I notice on the working control outputs (muxes?). Most of
 them get to a certain voltage (somewhere around 70% of their
 potential), and then fall back to zero with an audible
 click, which also seems to change all settings. This can be
 seen on the trace at TP2 also. It seems like this indicates
 a short somewhere once a certain voltage is reached, but
 there are two or three noticeable exceptions (MG SP, RESO,
 and REL) which do not do this.
 All of which makes me wonder if I
 should just be starting with a new circuit board, as maybe
 this one is just too corroded. but I would like to try to
 continue fixing this one, and any help is
 appreciated.Thanks.

Re: Trying to Restore a P6.

2016-08-30 by 7zlwncpbis4x22jwvkd4oc5hzu2njcsjeup42is6@yahoo.com

Thanks again for your patience and guidance on this, Bob. I'm going to get this working right eventually.
I see no voltage change on pin 1 of IC 14, and pin 3 exhibits odd behavior, in which putting a probe on it seems to make the voltage rise or fall to settle around 12.8V. The voltage at pin 2 rarely rises above 3V, due to the fall-off in voltage I see on some of the controls, so perhaps I'm seeing the V at pin one due to the fact that the - input is always less than the + input.
Also, I'm seeing a constant 4.97V on pins 3 and 4 of IC17, which seems odd for it's outputs considering it's function, and what is going into it. I bought some extra 4011s tho, so not a problem to swap out, but please clue me in if I am misunderstanding the circuit.
And, I will wait to see how IC10 reacts after installing a new IC18, Those connections check out continuity-wise, and I get what you are saying about feedback circuits.
First I want to pull the board and make sure I didn't do something stupid on the bottom while correcting bad traces, that might have caused the 5V line to get crossed.

Re: [PolySix] Re: Trying to Restore a P6.

2016-08-31 by Bob Grieb

IC17 pins 3 and 4 look like part of the tape out circuit.
Are those really the ones you are looking at?

Also, one thing I forgot to mention.   The voltage at the output of
the 4051 should track the panel potentiometer, I think, although there may be a scaling
there, for instance the pot may be set up to swing from 0 to 10 or 0-15 volts, 
and the corresponding output of the 4051 may only swing from 0-5V for instance.
If you are seeing
a big step when you hit a particular spot in the pot's rotation, then that may 
mean that one bit of the DAC is not working.   For instance, say the DAC value is 
changing from 0011_1111 to 0100_0000 and the D6 bit doesn't work.   Instead of
seeing a small step up, you would see the value jump back to 0 when you hit that
point.   Not sure exactly what you are seeing, but I would replace the two 4051's 
first, then try to determine if the whole range of the DAC is working by slowly 
turning one of the pots and watching the corresponding output of the 4051 on the scope.

BTW, the charging resistor R86 is 2.2M, so if you are using the x1 position of your scope
probe (1M ohm) then probing that point would lower the voltage.  Is the "ST" signal on 
the left end of C35 switching?  In general, you would want to use the X10 scope probe 
position if you can, since that has a 10M input resistance and will load the circuit a lot
less.


--------------------------------------------
On Tue, 8/30/16, 7zlwncpbis4x22jwvkd4oc5hzu2njcsjeup42is6@yahoo.com [PolySix] <PolySix@yahoogroups.com> wrote:
Show quoted textHide quoted text
 Subject: [PolySix] Re: Trying to Restore a P6.
 To: PolySix@yahoogroups.com
 Date: Tuesday, August 30, 2016, 7:34 PM
       
              Thanks again for your patience and guidance on
 this, Bob. I'm going to get this working right
 eventually.    I see no voltage change on pin 1
 of IC 14, and pin 3 exhibits odd behavior, in which putting
 a probe on it seems to make the voltage rise or fall to
 settle around 12.8V. The voltage at pin 2 rarely rises above
 3V, due to the fall-off in voltage I see on some of the
 controls, so perhaps I'm seeing the V at pin one due to
 the fact that the - input is always less than the +
 input.    Also, I'm seeing a constant 4.97V
 on pins 3 and 4 of IC17, which seems odd for it's
 outputs considering it's function, and what is going
 into it. I bought some extra 4011s tho, so not a problem to
 swap out, but please clue me in if I am misunderstanding the
 circuit.   And, I will wait to see how IC10
 reacts after installing a new IC18, Those connections check
 out continuity-wise, and I get what you are saying about
 feedback circuits.    First I want to pull the
 board and make sure I didn't do something stupid on the
 bottom while correcting bad traces, that might have caused
 the 5V line to get crossed.

Re: Trying to Restore a P6.

2016-08-31 by 7zlwncpbis4x22jwvkd4oc5hzu2njcsjeup42is6@yahoo.com

Yes IC17, It just seemed strange to get a constant 5V on those outputs when it was part of the tape circuit.
And I have seen the pots track on all the 4051 outputs EXCEPT pin 1 on IC18, That sits at a constant 4.9V
But I will replace both 18 and 19 and use machined sockets.
And yes, using the x10 setting on the probe.
Understood about the scaling.
What you describe about missing bits sounds like what is happening. This is what seems to set the synth off, and scramble all the data in a patch. You can watch the program and bank buttons change when that happens, as well as the write and manual buttons, and see it on the scope from TP2 as well.
And I'm loathe to admit I have not been able to figure out what "ST" is or does, but I'd be glad to check it if you can tell me what switches it.
Thx.

Re: [PolySix] Re: Trying to Restore a P6.

2016-08-31 by Bob Grieb

A non-working DAC bit will not change LEDs or switch settings.
It would only interfere with the DAC's ability to output all voltage levels.
That's analog voltages.   The switches and LED's are digital.  0 or 1.
The DAC is not involved with reading or writing to them.   The DAC is 
used to set the voltage levels on the outputs of the 4051's.  So if there
is a problem with the DAC swinging over the full range, then it will 
affect those chips as well.  I thought that was what you were describing,
but now I am not sure.

Op amps can fail in such a way that the input could always be at a constant
voltage, instead of being a high impedance input.  So your 4051 output that
is stuck at 5V could be held there by a failed op amp, or more likely (IMO), the 
4051 is bad.   

ST is fed from a register chip on the KLM-366 board.   I think it should be changing.
Did you look at it?   

The 8048 reads the switches and saves their values in RAM when you save a patch.
In manual mode, it just reads them in and feeds them out again to the appropriate
places in the circuit.  When you are using a patch instead, the values are read from 
RAM and fed out instead of the current front panel switch states.

If all of the settings change at once, are you in manual mode at that time, or are you 
using the patch setting from RAM?

Were you thinking that the tape circuitry would be active when you are not saving anything 
to tape?   Or maybe you tried to save patches to tape and found that those signals
did not change?




--------------------------------------------
On Tue, 8/30/16, 7zlwncpbis4x22jwvkd4oc5hzu2njcsjeup42is6@yahoo.com [PolySix] <PolySix@yahoogroups.com> wrote:
Show quoted textHide quoted text
 Subject: [PolySix] Re: Trying to Restore a P6.
 To: PolySix@yahoogroups.com
 Date: Tuesday, August 30, 2016, 9:19 PM
 
 
       
       Yes IC17, It just seemed strange to get a constant
 5V on those outputs when it was part of the tape
 circuit.And I have seen the pots track on all the
 4051 outputs EXCEPT pin 1 on IC18, That sits at a constant
 4.9VBut I will replace both 18 and 19 and use
 machined sockets.And yes, using the x10 setting
 on the probe. Understood about the
 scaling.
 What you describe
 about missing bits sounds like what is happening. This is
 what seems to set the synth off, and scramble all the data
 in a patch. You can watch the program and bank buttons
 change when that happens, as well as the write and manual
 buttons, and see it on the scope from TP2 as
 well.And I'm loathe to admit I have not been
 able to figure out what "ST" is or does, but
 I'd be glad to check it if you can tell me what switches
 it.Thx.

Re: Trying to Restore a P6.

2016-08-31 by 7zlwncpbis4x22jwvkd4oc5hzu2njcsjeup42is6@yahoo.com

I have a constant voltage of just under +5 on the ST side of C35, and -5 on the side toward IC14.

The synth won't go into manual mode, tho the switch does light on occasion. Any attempt to adjust a pot or switch will change settings in a seemingly random manner, causing a new patch to come up, or more often, noise.The program and bank lights do not always come with a patch, and sometimes when they do light, there is no sound.

Apologies for the side track with IC17. I just noticed those voltage while poking around, and was puzzled. However, as the synth goes thru it's random routine, the Write button does sometimes light or even blink.

And as always, I appreciate the overview you are so kind to include. Will need to reread in the morning.

Re: Trying to Restore a P6.

2016-09-01 by chipaudette@yahoo.com

Regarding a modern CPU to replace the 8048, I replaced my Polysix's Key Assigner with an Arduino Mega. Here's a quick overview...


I've shared my code for the Key Assigner, though there are some caveats: (1) it's super ugly code, and (2) it assumes that you've already replaced your keybed with a MIDI keybed. With that said, the code is here:

Chip

Re: [PolySix] Re: Trying to Restore a P6.

2016-09-01 by Bob Grieb

Yes, I am familiar with your project.   Cool stuff!   I developed a PIC
replacement for the key assigner on the Poly6, which plugs into the same
socket and replicates the fcn of the original code, but adds MIDI note on and 
off capability.   Replacing those old slow Intel CPU's with a PIC or an AVR chip
is sort of like "shooting fish in a barrel" (haven't heard that expression in a while!).

Bob


--------------------------------------------
On Thu, 9/1/16, chipaudette@yahoo.com [PolySix] <PolySix@yahoogroups.com> wrote:
Show quoted textHide quoted text
 Subject: [PolySix] Re: Trying to Restore a P6.
 To: PolySix@yahoogroups.com
 Date: Thursday, September 1, 2016, 4:10 PM
       
       Regarding a modern CPU to replace the 8048, I
 replaced my Polysix's Key Assigner with an Arduino Mega.
  Here's a quick overview...
 http://synthhacker.blogspot.com/2013/03/polysix-aftertouch-and-portamento-demo.html
 I've shared my code for the Key
 Assigner, though there are some caveats: (1) it's super
 ugly code, and (2) it assumes that you've already
 replaced your keybed with a MIDI keybed.  With that said,
 the code is here:
 https://github.com/chipaudette/ArduinoPolySix  Chip

Re: [PolySix] Re: Trying to Restore a P6.

2016-09-02 by chipaudette@yahoo.com

Your drop-in replacement for the Key Assigner is way more useful for the broader community than mine...since mine has the requirement to also replace the keybed. In my case, the whole project was motivated by replacing the keybed, so it all worked out well. But, for the typical P6 modder, your PIC-based drop-in replacement sounds better!

Chip

Re: Trying to Restore a P6.

2016-09-03 by 7zlwncpbis4x22jwvkd4oc5hzu2njcsjeup42is6@yahoo.com

OK...Have replaced both 4051s, and am now seeing the proper ins and outs from IC18, as well as IC 10.
IC 19 seems to be behaving itself also, and for the moment, I'll assume things are right with IC14 pin 1.
Discouragingly, the behavior of the synth is still essentially the same. The only difference seems to be that the Bank D Program 4 lights are permanently on now, no matter what else lights up. It will still not complete the first calibration reset step, and will not enter manual mode.

It seems as thought the signals for controlling parameters are being confused for program change signals.

Here's what I am noticing regarding the odd behavior of voltages coming from the control pots.They seem to swing from -5 to +5, but most of the pots fall back to -5 with an audible click as soon as they pass the threshold to plus values. This then scrambles the current "patch", and changes it to another, or to noise. They can be coaxed to +5 with some control twisting, and even sometimes "repair" themselves and act normally, but only until another pot is moved and "fails". A few seem to work normally all the time. This behavior can be seen on IC 19 and 18, as well as the outputs and inputs of the ICs they feed.

Something connected to ground that shouldn't be?

I guess I'm dealing with a bigger corrosion problem than is obvious from the apparent condition of the circuit board, and maybe I should just start rebuilding on one of the replacement boards, but am glad to keep digging into this one, if anyone can give me a direction to go.

At this point, I'm curious about what I should be seeing at IC33 pin 4, or on pin 1 of IC27. Right now, it shows nothing there on my scope.

As always, I'm grateful for the help, and feel I am getting somewhere, but seem to be creating more questions than answers.

Glad to see this post stirring other discussions within the group.






Re: Trying to Restore a P6.

2017-05-03 by 7zlwncpbis4x22jwvkd4oc5hzu2njcsjeup42is6@yahoo.com

OK. Progress. Got around to replacing the RAM chip (IC32) and this seemed to be the source of my basic problem with restoring this P6.
Now all front panel controls seem to be working properly, whereas previously, any attempt to rotate a pot caused random results.
It is also possible to get into Manual mode for the first time. As a result, I was able to get thru step 8 of the adjustment procedure, and all Bank and Program switches seem to be working fine, except that once I leave Bank A, it will not return there unless I flip the Tape switch, or power down and back up.
Very few of the program locations have sounds in them, but I assume this is because I have replaced the RAM chip.
My biggest problem now is that output is distorted in an odd way, in which all sounds seem to have the same strange warbled distortion to them. This is a recent problem that has developed in the last year as I have been working on the synth.
Any suggestions about where to look to solve this problem would be appreciated. On the scope, the synth seems to be switching properly on ICs 18 and 19.
Thanks again for the help to all here, and I am very excited that the synth seems to be salvageable. It needs a good keyboard contact cleaning, but now I know that will not be a waste of time.

Re: Trying to Restore a P6.

2017-05-04 by 7zlwncpbis4x22jwvkd4oc5hzu2njcsjeup42is6@yahoo.com

What I am seeing when I put the scope on the output test point on the KLM-366B board is ghost images of the square or triangle wave the synth is producing. The faint ghost signals precede the stronger image of the output.
I am at a loss. Time to replace the DAC chip? Anybody have any similar experience?

Re: [PolySix] Re: Trying to Restore a P6.

2017-05-05 by Florian Anwander

Hello

The "ghost images" story tells me, that you do not know how to use an 
oscilloscope correctly, so...
 > Time to replace the DAC chip?
... to be honest: It is time to replace the technician by a more skilled 
one.

Florian

Re: [PolySix] Re: Trying to Restore a P6.

2017-05-05 by Bob Grieb

I am afraid I must agree with Florian on both points.

The ghost images are most likely improper triggering of the oscilloscope.

And when I saw your comment about the DAC, after you described the symptoms,
I was very surprised.

Probably it's time to get some professional help with the repair.

RE: [PolySix] Re: Trying to Restore a P6.

2017-05-05 by Sylvester Leontiadis

Don’t give up your repair.

I am quite sure that anyone who knows will help you.

Check if you have done something wrong during checking and let us know.

From: PolySix@yahoogroups.com [mailto:PolySix@yahoogroups.com]
Sent: Friday, May 05, 2017 6:49 PM
To: PolySix@yahoogroups.com
Subject: Re: [PolySix] Re: Trying to Restore a P6.

Thanks guys. Sorry to have bothered you with my ineptitude.

RE: [PolySix] Re: Trying to Restore a P6.

2017-05-05 by jaydiplock@gmail.com

Yeah stick with it... poor form to criticise, we all have different levels of skill and experience with stuff like this, but we all share the passion for vintage synths. Good luck...

Move to quarantaine

This moves the raw source file on disk only. The archive index is not changed automatically, so you still need to run a manual refresh afterward.