Hi all.
After a stupid mistake I've made (wrong power connector) i found a shorted power supply in my old korg m1 (before any further tests).
Disconnected the power supply from the main-board, took it out and started testing it. Replaced the fuse and after some trial- error (some small shorted components) the first problem came down to a totally shorted q1. Replaced it with a 2sc2335 i had around (good equivalent of the original 2sc3570 according to thus guide https://alltransist…). I’m not certain if the transistor is in perfect shape.
All seemed ok, but i realized that the secondary was giving double figures. 10v instead of 5v on the 5v rail, and +-24v on the +-12v rail.
The transformer is not shorted (but other than that i can’t tell details about primaries - secondaries ratios since none is given).
I guess the problem lies before or just around the 4 pins of transformer T1.
Has anybody any specific ideas what the problem can be and which components could cause such a problem.
I send a link to the m1 service manual from my google drive. The pcb i’m fixing is klm 1267,1268 (it’s 1268) on page 11 of 28.
https://drive.googl…
Thanks in advance
Comments
Korg M1 PSU
You are not allowing access to your Google Drive for us to look at your copy of the service manual. No matter, I found it elsewhere on the net.
I would assume that the feedback path through the opto-isolator is not making it all the way back to Q1 to control the switching current. Try measuring the d.c. volts across the LED diode of PC1, which should be in the region of 1.2V to 1.5V. If it's much less than this, measure the volts between the anode of PC1 and the anode of IC1. If this is near +10V, check the volts across the pot VR1 and the resistors R24 and R27. If VR1 were open circuit, the feedback loop would be permanently demanding higher output, which is what you report.
Another explanation is that whatever took out Q1 also damaged Q2. If you can't find a problem with the feedback circuit, I would try replacing Q2.
Thank you once more Boswell…
Thank you once more Boswell for the input
The link problem is sorted thanks for mentioning it.
Actually i replaced q2 as well as part of the initial repair. Of course it might be damaged once again during testing.
Pc1 is an ic (sharp pc 817) so i can’t be very certain what pins are the “led” ones. I have to check the datasheet pinout.
I will do the tests and report back.
So here are the results of…
So here are the results of my testings.
Voltage across pc1 led =1.6v
voltage between anode pc1 and anode ic1 (pin 2) =5v
voltage between r23 (typo on the manual it’s r25 on the board) on the junction ic1 (pin1) and vr1= 0v
voltage between the other pin of vr1 and r24 (it’s not r27) =0v.
The trimmer doesn’t alter the overal voltage more than +-20mv.
do you think vr1 is gone? It seems like an easy explanation.
But the resistance on the trimmer is good. 850ohm on a very clean 1k trimmer.
ps. R10, R11 and q1 (the untested transistor) seem like they are getting realiy hot. Especially the resistors. Near burn condition.
Something a bit odd going on…
Something a bit odd going on. If the volts between the anodes of PC1 Led and IC1is +5V, and the nominal 5+V output rail is +10V, it means you have 5V across R23, a 22 Ohm resistor. That implies a current of around 220mA through R22, but where can it be going? The PC1 Led would burn out if it were going through there, and the other route has the VR1 pot plus KOhm resistors in series.
I'm not surprised if the components in the Q1 area are heating up as they are trying to respond to the demand for ever more output voltage.
I would double-check the voltages around IC1/PC1 (Led only), and see if there really is 5V across R23. If R23 has gone high in value (e.g. to a few KOhm), then all is explained.
Justto clarify. What is…
Justto clarify. What is written as r23 on the manual schematic is r25 on the board and what is written as r25 is r23.
New measurements
Voltge between anodes the same as before.
voltage across the 22ohm resistor is -5.4v (i guess polarity doesn’t matter) and the 3.6k resistor in line has 1.8k resistance across it. Off line 3.6k normal. The 22 ohm resistor online stays precisely on 22 ohm.
I wonder why are we measuring voltage between ic1 A and pc1 led anode, when they are not connected.
Ic1 K (cathode i guess) is connected with pc1 led cathode. Am i wrong?
Just a reminder
If you remember, on the initial post i mentioned that the same problem (double voltage values) appears on the other output rail too (instead of +- 12v it gives +-24v).
Am i wrong to believe that focusing on the secondary ( or just the 5v part of it). will not lead to the solution?
If the manual had typed the voltages on the transformer primary and secondary pins ( or around the transistors) things would be clearer.
Other rail voltages
The voltage reference and feedback are on the 5V output rail. The +/-12V rail has no reference of its own, so if the 5V rail doubles, the +/-12V will double as well.
I'll take another look at the schematic when I next have an hour or so free...
Thanks for your time and…
Thanks for your time and insight Boswell.
The voltages across the…
The voltages across the transformer are not fixed as they would be in a mains transformer. Instead, the transformer is in a feedback loop, and (in normal operation) its winding voltages will vary depending on load and other factors. In this case, the loop is not working, so the control voltages are jammed hard against one end, resulting in double the normal output volts.
It looks as though the d.c. output side is probably OK, but what it's trying to tell the mains side is not being received correctly or not acted upon. The feedback is delivered via PC1, the opto coupler, which is a PC817. These parts are cheap and readily available, so that would probably be my first thing to try changing. The transistor side of it is intimately connected with T1, the power transistor that I believe you changed.
It's more difficult working on the mains side of things because, even with the mains cable unplugged, there can still be high voltages stored on the reservoir capacitors. You must discharge these before working on the components on this side of the transformer.
Thank you very much for the…
Thank you very much for the advice and insight Boswell.
I was about to change all 3 transistors and optocouplers and the large schotky diode w.
Now i will only replace only the optocoupler and the power transistor (since the one i placed was not guarranteed working). Apparently a 2sc3310 is a good replacement forbthe original - obsolete- one.
Will let you know.
Bad news
Unfortunately no change.
Replaced the pc817 optocoupler. The ic1 transistor , the q1 and q2 transistors but nothing changes.
Haven’t replaced the diodes (that skq30 maybe?) and the resistors. The resistors measure ok offline. But i wonder if they are still ok on their power ratings since they are getting really hot and they’ve ‘ve been nearly burnt.
The small signal diodes look good and the ordinary power diodes also looks good.
The other thing is that some resistors measure half inline ( could this be a clue for the location of the problem , which could lead to double voltage on the outputs). Even one of the small sigbal diodes give double readings inline. Offline all measure good. All around q1.
I wonder what else could i do. I mean it’s only diodes, resistors and capacitors the other things i haven’t replaced.
I'm not sure what you mean…
I'm not sure what you mean by measuring components "online" and "offline". Are you talking about being in circuit vs. when removed from the circuit? You can't get an accurate resistance/capacitance figure for a component unless at least one end of it is disconnected from all other wires, tracks and components on the board.
I'll do some more looking at the schematic.
Thanks Boswell. Yes. That…
Thanks Boswell.
Yes. That’s exactly what i mean. In circuit and out of it . I know that components have to be removed to give accurate results, that’s why i always remove them. But just for the sake of it i also measure them on board.
In reply to I'm not sure what you mean… by Boswell
Boswell wrote:I'll do some…
I don’t want to be annoying but did you have the chance to take another look?
Finally the power supply…
Finally the power supply issue is sorted. Weirdly enough it was needing load. 5ohm 25w did the job. The strange thing is that before the initial accident, it was giving good values without load. Can’t tell what.
The accident had destroyed opto trabsistor side, resistors, transistors. All on the primary side.
Apparently the synth has some issues sround the dac to start with (maybe some comparators) and maybe some noisy ram chips.
Thanks for the input and the help Boswell.
I'm glad you seem to have…
I'm glad you seem to have got it fixed, although I'm a bit suspicious of it failing to regulate on no load.