Have a Mk2 in from a customer and its got me pulling my hair out! The service manual is about the most arcane document I've ever seen for a start... but I digress....
It is used almost all day every day apparently and left running continuously. The customer reported that when returning to the TT he found it had stopped rotating and noticed that the PSU was much warmer than normal.
When delivered to me and set up well lo and behold it worked perfectly... for about 10 mins then it simply stopped.
Seems to be that worst of all faults, probably an intermittent IC or transistor which is neither S/C or O/C.
Following the fault finding chart got me nowhere. Everything checks out as it should when going through the "platter fails to start" flow chart. This flow chart is incredibly misleading as many of the voltages it asks you to check in fact vary widely depending on platter position! It says A1 to 3 should be 23V DC but in fact they are the drives to the motor and go from around say 3V to 28V ish as the platter completes a rotation pushed by hand. The PSU, drive heatsink and even motor all get warm as they are sitting there with a load on them but "stalled". Yes the brake is not stuck on and works as it should!
If you manually push the platter it seems "to want to go" and sometimes will rotate a few turns on its own.
The PSU is fine and recently recapped and all voltages correct.
I presume there should be a permanent frequency presented to the phase comparator/control board for comparison with the FG signal but there is nothing but DC levels on its edge connector. The manual seems designed to protect Matshusitas IP rather than aid fault finding and I cannot find any reference to eg "the reference signal from the oscillator divider board enters the control board on pin E11..."
Also, what is the input to the drive board? It's not clear if there should be an AC signal which is the drive to the motor after being split into 3 phases or or if the drive board takes a DC level and speed is proportional to this?
"a2" looks like an input but the fault flow chart says this should simply be 22V DC, which it is. Is there a simple way of feeding a signal from a signal generator to the drive board as a mock drive signal just to see if it eliminates this board? Or alternatively a point which can be either grounded or have a specific DC level applied to do the same thing?
I've spent about 7 hours on it so far with no results other than everything checked so far looking fine other than a suspicion that I should be seeing waveforms on the PLL/control board but there are non. It doesn't help that the manual shows loads of waveforms which should be there WHEN THE PLATTER IS REVOLVING but as it's a closed loop system it's a fault finding nightmare! I repair hi fi for a living but it's decades since I last took on a DD TT and I took a chance in accepting the job that it would likely be something obvious like a faulty PSU or S/C output transistor etc but it was not to be and in fact it seems to have one of those faults from hell.... can't win 'em all eh...
Edit: The crystal oscillator works fine, a square wave which changes in frequency when 33, 45, 78 are selected is present at "P2". The 50KHz oscillator on the drive board is working and all the "feedback phases" from the "sensing/commutating coils" give a DC signal proportional to platter position as they should via their respective rectifiers etc. It just sits there pulling current but not commutating and very much "still life".
It is used almost all day every day apparently and left running continuously. The customer reported that when returning to the TT he found it had stopped rotating and noticed that the PSU was much warmer than normal.
When delivered to me and set up well lo and behold it worked perfectly... for about 10 mins then it simply stopped.
Seems to be that worst of all faults, probably an intermittent IC or transistor which is neither S/C or O/C.
Following the fault finding chart got me nowhere. Everything checks out as it should when going through the "platter fails to start" flow chart. This flow chart is incredibly misleading as many of the voltages it asks you to check in fact vary widely depending on platter position! It says A1 to 3 should be 23V DC but in fact they are the drives to the motor and go from around say 3V to 28V ish as the platter completes a rotation pushed by hand. The PSU, drive heatsink and even motor all get warm as they are sitting there with a load on them but "stalled". Yes the brake is not stuck on and works as it should!
If you manually push the platter it seems "to want to go" and sometimes will rotate a few turns on its own.
The PSU is fine and recently recapped and all voltages correct.
I presume there should be a permanent frequency presented to the phase comparator/control board for comparison with the FG signal but there is nothing but DC levels on its edge connector. The manual seems designed to protect Matshusitas IP rather than aid fault finding and I cannot find any reference to eg "the reference signal from the oscillator divider board enters the control board on pin E11..."
Also, what is the input to the drive board? It's not clear if there should be an AC signal which is the drive to the motor after being split into 3 phases or or if the drive board takes a DC level and speed is proportional to this?
"a2" looks like an input but the fault flow chart says this should simply be 22V DC, which it is. Is there a simple way of feeding a signal from a signal generator to the drive board as a mock drive signal just to see if it eliminates this board? Or alternatively a point which can be either grounded or have a specific DC level applied to do the same thing?
I've spent about 7 hours on it so far with no results other than everything checked so far looking fine other than a suspicion that I should be seeing waveforms on the PLL/control board but there are non. It doesn't help that the manual shows loads of waveforms which should be there WHEN THE PLATTER IS REVOLVING but as it's a closed loop system it's a fault finding nightmare! I repair hi fi for a living but it's decades since I last took on a DD TT and I took a chance in accepting the job that it would likely be something obvious like a faulty PSU or S/C output transistor etc but it was not to be and in fact it seems to have one of those faults from hell.... can't win 'em all eh...
Edit: The crystal oscillator works fine, a square wave which changes in frequency when 33, 45, 78 are selected is present at "P2". The 50KHz oscillator on the drive board is working and all the "feedback phases" from the "sensing/commutating coils" give a DC signal proportional to platter position as they should via their respective rectifiers etc. It just sits there pulling current but not commutating and very much "still life".
Last edited:
Have you checked ripple on the PSU voltages. Ripple will effect how the TT operates. An intermittent fault that happens after a time is a classic symptom of dry joints. I also used a hot air gun and spray freeze in my early days but later on before I retired we used a Fluke thermal camera to view PCB's for components that were operating too hot.
After that I would start at the Logic PCB. All the waveforms listed in the manual for the Logic board should be there with the TT stopped and on 33.
Has anyone been there before you, I had one Logic board where one idiot had replaced all the IC's and installed sockets I spent 2hrs and gave up and replaced the PCB. I suspect there were multiple via's missing. There was oscillation on some of signals (can't remember which ones now).
Good luck
After that I would start at the Logic PCB. All the waveforms listed in the manual for the Logic board should be there with the TT stopped and on 33.
Has anyone been there before you, I had one Logic board where one idiot had replaced all the IC's and installed sockets I spent 2hrs and gave up and replaced the PCB. I suspect there were multiple via's missing. There was oscillation on some of signals (can't remember which ones now).
Good luck
Thanks for the reply.
Yes I've thoroughly checked the PSU and it has been recently recapped by someone who obviously knew what they were doing. It's apparently worked fine since the recap for several months.
I've ruled out "tappy" faults ie dry joints and yep I've been down the hairdryer and freezer spray path already.
My suspect No1 was indeed ancient TTL... and the thought of just changing out the lot had crossed my mind...
Knowing that the waveforms on the logic board are actually still present without the platter rotating is very useful! I was aware that some are and they seem OK but I'll look deeper here. Most of the manual says "waveforms with platter rotating at 33.33RPM"... which when the platter won't spin up rather begs the question of which if any should still be present without the platter rotating!
Do you know if there should be a permanent reference signal to the control/PLL board from the logic/divider board.
A PLL needs a reference signal but I'm seeing just DC levels on all the edge connector pins on the PLL board. Hoping there should be and a TTL on logic board acting as a buffer/driver to the PLL board has failed. It's a right PITA to fault find as it both a closed loop system and functionality is spread between the boards bouncing backwards and forwards between them it seems.
Edit: No one has been there before no. There's no evidence of the cover ever having been off before I got to looking at it. Apparently they were made until '87 IIRC and the electrolytics in it all look from around then or no older than maybe '84. I remember that type. All Panasonic/Matsushita of course. Could feasibly be an electrolytic but the suddenness of the failure works against, this whilst not ruling it out IMHO. I'll check ESR on them all tomorrow.
Yes I've thoroughly checked the PSU and it has been recently recapped by someone who obviously knew what they were doing. It's apparently worked fine since the recap for several months.
I've ruled out "tappy" faults ie dry joints and yep I've been down the hairdryer and freezer spray path already.
My suspect No1 was indeed ancient TTL... and the thought of just changing out the lot had crossed my mind...
Knowing that the waveforms on the logic board are actually still present without the platter rotating is very useful! I was aware that some are and they seem OK but I'll look deeper here. Most of the manual says "waveforms with platter rotating at 33.33RPM"... which when the platter won't spin up rather begs the question of which if any should still be present without the platter rotating!
Do you know if there should be a permanent reference signal to the control/PLL board from the logic/divider board.
A PLL needs a reference signal but I'm seeing just DC levels on all the edge connector pins on the PLL board. Hoping there should be and a TTL on logic board acting as a buffer/driver to the PLL board has failed. It's a right PITA to fault find as it both a closed loop system and functionality is spread between the boards bouncing backwards and forwards between them it seems.
Edit: No one has been there before no. There's no evidence of the cover ever having been off before I got to looking at it. Apparently they were made until '87 IIRC and the electrolytics in it all look from around then or no older than maybe '84. I remember that type. All Panasonic/Matsushita of course. Could feasibly be an electrolytic but the suddenness of the failure works against, this whilst not ruling it out IMHO. I'll check ESR on them all tomorrow.
Last edited:
Don't change all the TTL chips. In the dozens of SP10mk2's I've repaired I have had very few IC failures. IMO IC failures are most likely caused by the 5v being way too high because of cap failure in the PSU. I have been unable to find data sheets for the logic chips so absolute compatibility is unknown. On the faulty Logic PCB I have, swapping a 74HCT73 and 74LS73 completely changed the fault. One caused oscillation and the other didn't.
E11 and 12 on the control PCB should have the 19ms signal spinning or not. The sawtooth is generated from the FG and with the TT switched on spinning the platter by hand that waveform should be there. If the E11,12 waveforms aren't there I'd be starting at IC15 and working back, bearing in mind a short Tr on the Control PCB could pull them to ground.
E11 and 12 on the control PCB should have the 19ms signal spinning or not. The sawtooth is generated from the FG and with the TT switched on spinning the platter by hand that waveform should be there. If the E11,12 waveforms aren't there I'd be starting at IC15 and working back, bearing in mind a short Tr on the Control PCB could pull them to ground.
FG working and waveforms OK. Not yet gone back to it today but from memory of last look at it there were no waveforms on E11, 12.... hoping there is finally a trail to follow! Thanks.
Indeed there seems to loads of errors in the manual. I still can't see what the nature of the input, the stimulus if you like, to the drive board is. It simply isn't mentioned in the manual that I can see. As I said above the manual is simply awful! From my own deductions I'm thinking "a2" has to be the input but this is not mentioned anywhere in the manual. I'm presuming it should be driven by a 5.555Hz (at 33.33RPM) drive waveform but again no mention of this. If so then it is missing. I'm also assuming (again no mention in manual) that "B1" and "B2" are forward and reverse?
Simply having confirmation of the above assumptions or being put on the right track here would be very helpful!
Thanks. As hinted in an earlier post I wasn't sure if it took an external drive waveform or self commutated as you describe. So it should be possible to spin up the motor at some random speed even with malfunctioning logic or control boards by giving it a suitable current limited input voltage from a bench PSU? FWIW all the waveforms on the control board look about right as far as one can tell from spinning it by hand so probably not the drive board but if I can rule it out by say feeding it 10V DC via a 10K resistor and see it spin up to say 110RPM then that would be great.
I guess this is one of those repairs where you end up spending 30 hours on it and can only charge for a fraction of that but it is years since anything beat me
I guess this is one of those repairs where you end up spending 30 hours on it and can only charge for a fraction of that but it is years since anything beat me
@Shinken A universal controller will not work because every motor is different. All the DD TT's I've looked at use a synchronous motor, which means the rotor poles (magnet pairs) are dragged around by the rotating magnetic field generated by the stator they basically lock to the drive frequency. Most of the motors use different pole/slot counts which means they require a different drive frequency and have different speed feedback sensors.
There is development of an MCU controller to drive Technics SP10 motors currently underway.
There is development of an MCU controller to drive Technics SP10 motors currently underway.
I guess this is one of those repairs where you end up spending 30 hours on it and can only charge for a fraction of that but it is years since anything beat me
That sucks, but I guess it's a learning experience. I'm a retired Tech so spending a few hours is no skin off my nose.