Measure the resistance of the pot from the wiper (middle leg) to whichever end leg is used. Set to maximum resistance. Or measure on ohms between B and E of the transistor and again turn to the end that gives highest resistance. You can do it in circuit.
Unless any transistor reads obviously short circuit then you need to unsolder to test fully.
Around 50millamps is a pretty safe value to set the current.
Unless any transistor reads obviously short circuit then you need to unsolder to test fully.
Around 50millamps is a pretty safe value to set the current.
Debugging strategy...
Thanks for your help. A few last questions, fefore I really get into the mud:
- How do I measure the short circuit? (Emitter Collector Resistance in "built in state"???)
After finding all dead shortet resistors, should I replace, adjust lowest Bias and power up???
I have to check Q420, Q424, Q428 and the mirror part. I also have to check Q432, Q436. I can leave out the mirror part, because Voltage on the power resistor collector is ok??? Can I guess the other trannies are ok???
Or do I need to unsolder Q420, Q424, Q428 and the mirror part for correct checking???
Thanks for your help. A few last questions, fefore I really get into the mud:
- How do I measure the short circuit? (Emitter Collector Resistance in "built in state"???)
After finding all dead shortet resistors, should I replace, adjust lowest Bias and power up???
I have to check Q420, Q424, Q428 and the mirror part. I also have to check Q432, Q436. I can leave out the mirror part, because Voltage on the power resistor collector is ok??? Can I guess the other trannies are ok???
Or do I need to unsolder Q420, Q424, Q428 and the mirror part for correct checking???
Most transistors fail short circuit between C and E so they just read as such whether in or out of circuit. You won't find any resistors short, they fail by normally going open circuit or high in value. You can't really guess at anything, you have to measure and test and be sure. If you are not sure unsolder them and test.
Always use a bulb tester to power up (100watt bulb in series with mains supply) as this prevents outputs failing if these is a fault.
Always use a bulb tester to power up (100watt bulb in series with mains supply) as this prevents outputs failing if these is a fault.
Debugging...
Thanks for your support. So the next to do ist to measure with the multimeter for shortcuts over CE. But first I will have to care for my girlfriend (Back from Mexico after 8(!) weeks...)
I'll post the results of my tests - I guess next weekend.
Dirk
Thanks for your support. So the next to do ist to measure with the multimeter for shortcuts over CE. But first I will have to care for my girlfriend (Back from Mexico after 8(!) weeks...)
I'll post the results of my tests - I guess next weekend.
Dirk
Hi,
I got time to continue with the work: I checked all driver and output transistors for shortcut. All seemed fine. Then I checked Q435 and Q431. Q435 is fine,
Q431 has a shortcut. They were previously replaced by BD912 - seems a good choice, because they have higher max ratings.
But does the unsymmetrical supply explain the high offset??? Next I would do now: Replace the transistor, power up, take a measurement - and see...
Some further good advise??? (I think, I don't have to readjust the bias - didn't touch anythink - correct???
Dirk
I got time to continue with the work: I checked all driver and output transistors for shortcut. All seemed fine. Then I checked Q435 and Q431. Q435 is fine,
Q431 has a shortcut. They were previously replaced by BD912 - seems a good choice, because they have higher max ratings.
But does the unsymmetrical supply explain the high offset??? Next I would do now: Replace the transistor, power up, take a measurement - and see...
Some further good advise??? (I think, I don't have to readjust the bias - didn't touch anythink - correct???
Dirk
The unsymetrical supply doesn't explain the offset so I think there is also a problem with the amp itself as well as the PSU.
Q431 and it's driver Q435. If you remove both these the amp should run on the "normal" lower supply voltage provided via D421 for that rail.
You need to fix the main fault on the amp. Check the main output transistors Q427 and 429 also Q419, Q421, Q423, Q425 and resistors R455, R459, R461 R463.
Q431 and it's driver Q435. If you remove both these the amp should run on the "normal" lower supply voltage provided via D421 for that rail.
You need to fix the main fault on the amp. Check the main output transistors Q427 and 429 also Q419, Q421, Q423, Q425 and resistors R455, R459, R461 R463.
Continuing...
Hi Mooly,
thanks again, learning a lot.
So I will remove the faulty Q431 to get Power supply voltages all right.
Q29, Q27 as well as Q23, Q25 survived the transistor "short test".
I didn't test Q19, Q21 because I thought, only driver and power usually fail. But I will do.
R455 should be easy to measure built in because of the cap.
Can I measure R459 built in???
Resistors R461 R463 are "anti oscillate - correct??? Can they really fail in this arrengement? If I think of the low value and the good current gain of the output transistors... But you're right: I have to test.
I'll continue on weekend.
Thanks, Dirk
Hi Mooly,
thanks again, learning a lot.
So I will remove the faulty Q431 to get Power supply voltages all right.
Q29, Q27 as well as Q23, Q25 survived the transistor "short test".
I didn't test Q19, Q21 because I thought, only driver and power usually fail. But I will do.
R455 should be easy to measure built in because of the cap.
Can I measure R459 built in???
Resistors R461 R463 are "anti oscillate - correct??? Can they really fail in this arrengement? If I think of the low value and the good current gain of the output transistors... But you're right: I have to test.
I'll continue on weekend.
Thanks, Dirk
You must remove both Q431 and 435. Although Q431 is short it isn't the reason for the offset voltage.
If all the other transistors are OK then Q19 and 21 probably are OK.
R455 ? that's the 1K. Again this will be OK if all the other transistors are good.
I think you are looking at the 10 ohm R457. The circuit isn't very well drawn.
R459 If it doesn't read 330 ohms in circuit then to be sure you have to remove.
R461, 463. They can fail if any of the transistors have failed.
Edit... the fact remains that the amp has an offset of 40 volts or so and so far you have found no faulty components.
What would help is for you to write down clearly all the voltages on the transistors. Firstly measured normally from ground and also with a reading taken directly across the base and emitter such as,
Q432
C=43V
B=1.48V
E=0.71V
B-E 0.77V
do that for them all in that channel. You might think taking B-E reading is pointless as it's just B minus E voltage. It isn't always that simple when making real measurements and it can prove a lot.
If all the other transistors are OK then Q19 and 21 probably are OK.
R455 ? that's the 1K. Again this will be OK if all the other transistors are good.
I think you are looking at the 10 ohm R457. The circuit isn't very well drawn.
R459 If it doesn't read 330 ohms in circuit then to be sure you have to remove.
R461, 463. They can fail if any of the transistors have failed.
Edit... the fact remains that the amp has an offset of 40 volts or so and so far you have found no faulty components.
What would help is for you to write down clearly all the voltages on the transistors. Firstly measured normally from ground and also with a reading taken directly across the base and emitter such as,
Q432
C=43V
B=1.48V
E=0.71V
B-E 0.77V
do that for them all in that channel. You might think taking B-E reading is pointless as it's just B minus E voltage. It isn't always that simple when making real measurements and it can prove a lot.
Last edited:
Further progress...
Hi Mooly,
more work done:
R459 330Ohm - ok
R455 1k Ohm - ok
R461, R463: Copper wires, no resistors. - ok
Q423, NPN, B/C/E, 69,9/75,3/45,8, should be conducting, but is not → defect
Q427, NPN, E/C/B, 45,1/75,1/45,8 open, ok???
Q417, NPN, B/C/E, 58,8/62,1/58,4, open, ok
Q429, PNP, E/C/B, 45,3/-39,6/46,0, open, ok???
Q425, PNP, B/C/E, 59/-39,6/46,1, open, ok
Q419, NPN, B/C/E, 62,4/74,8/69,8, open, ok
Q421, PNP, B/C/E, 58,5/-74,2/58,8, open, ok
My conclusion: Q423 got killed, therefore the feedback loop creates all the imbalance. Seems that I have to replace Q423...
Regs, Dirk
Hi Mooly,
more work done:
R459 330Ohm - ok
R455 1k Ohm - ok
R461, R463: Copper wires, no resistors. - ok
Q423, NPN, B/C/E, 69,9/75,3/45,8, should be conducting, but is not → defect
Q427, NPN, E/C/B, 45,1/75,1/45,8 open, ok???
Q417, NPN, B/C/E, 58,8/62,1/58,4, open, ok
Q429, PNP, E/C/B, 45,3/-39,6/46,0, open, ok???
Q425, PNP, B/C/E, 59/-39,6/46,1, open, ok
Q419, NPN, B/C/E, 62,4/74,8/69,8, open, ok
Q421, PNP, B/C/E, 58,5/-74,2/58,8, open, ok
My conclusion: Q423 got killed, therefore the feedback loop creates all the imbalance. Seems that I have to replace Q423...
Regs, Dirk
Q423 is faulty and so is Q425 with 13 volts across base and emitter.
Q419 appears OK, the 7 volts B-E is a reversed biased condition.
You certainly have to replace those two... if it were me I would look to replacing Q419 and 423 as well simply because we don't know the exact cause of failure. tbh the failure of the 2SD669 is "typical" of that type of Japanese device with it failing open B-E.
And you still have the rail problem, which is almost certainly Q431 and Q435.
I would look to just fixing the amp to begin with, then if you are going to keep it perhaps replace all the 2SD669 and 2SB649's on both channels.
Q419 appears OK, the 7 volts B-E is a reversed biased condition.
You certainly have to replace those two... if it were me I would look to replacing Q419 and 423 as well simply because we don't know the exact cause of failure. tbh the failure of the 2SD669 is "typical" of that type of Japanese device with it failing open B-E.
And you still have the rail problem, which is almost certainly Q431 and Q435.
I would look to just fixing the amp to begin with, then if you are going to keep it perhaps replace all the 2SD669 and 2SB649's on both channels.
Hi Mooly,
Why Q425??? It's PNP, B-E ca. +13V, so it must be open. It should only be closed, if NPN...
I think, I have to organize a big bunch of silicon...
Thanks, Dirk
Why Q425??? It's PNP, B-E ca. +13V, so it must be open. It should only be closed, if NPN...
I think, I have to organize a big bunch of silicon...
Thanks, Dirk
Just measured a couple of PNP power devices for curiosity and the reverse B-E breakdown voltage seems very imprecise. As low as 10 volts on one device to nearly 18 volts on another so yes, Q425 is probably OK. It just seemed high when I saw it but apparently not.
When talking about small signal devices running them with reverse bias like this can degrade the noise figure... that would never be an issue here... but if it were me I would look to change the drivers for a matched pair. MJE340 and 350 are pretty universally available.
When talking about small signal devices running them with reverse bias like this can degrade the noise figure... that would never be an issue here... but if it were me I would look to change the drivers for a matched pair. MJE340 and 350 are pretty universally available.
Hi, Thanks.
So you would replace the 2SD669 / 2SB649 with MJE340 and 350???
Just ordered the other parts from Reichelt / Ebay... Do you think MJE340/350 are more stable / have better distortion numbers???
After having fixed the stuff, I'll come back here with the results. ;-)))
So you would replace the 2SD669 / 2SB649 with MJE340 and 350???
Just ordered the other parts from Reichelt / Ebay... Do you think MJE340/350 are more stable / have better distortion numbers???
After having fixed the stuff, I'll come back here with the results. ;-)))
Remember to use the bulb on first power up.
The MJE's won't affect the distortion at all. They might have been around a while but they are perfectly respectable devices and used in a great many designs, both DIY and commercial.
The MJE's won't affect the distortion at all. They might have been around a while but they are perfectly respectable devices and used in a great many designs, both DIY and commercial.
NAD 7240PE up and running again...
Hi Mooly, thanks for your help.
I finally fixed it today. Only Q423 (NPN Driver 2SD669) was broken.The switched on higher voltage was only due to the high offset voltage (at least Q431, Q435 were tested ok, so I soldered them back.
I didn't listen to your good advice to use a bulb, because I seemed impossible to get it in the rails.
So. I still have to adjust the bias, the repaired channel seems to run a bit hot. (Unsolder the copper bridge, measure across the 1Ohm resistance, adjust, resolder bridge). Yes, I know that was stupid risk to switch on, I would not have done with expensive gear, but my girlfriend annoyed me to get finished or throw it away...
Unfortunately, a few problems remain:
1. There is a humming noise in the amp. It also stays there if I split Pre/Power section. Is this due to the fact that the chassis is open on the bench??? What is a possible reason? How can I trace this down???
2. The combined volume/balance pot seems to suffer crackling due to dust. It seems unobtainable. Which is a recommended product (contact spray that minimizes the problem...)
Thanks, Dirk
Hi Mooly, thanks for your help.
I finally fixed it today. Only Q423 (NPN Driver 2SD669) was broken.The switched on higher voltage was only due to the high offset voltage (at least Q431, Q435 were tested ok, so I soldered them back.
I didn't listen to your good advice to use a bulb, because I seemed impossible to get it in the rails.
So. I still have to adjust the bias, the repaired channel seems to run a bit hot. (Unsolder the copper bridge, measure across the 1Ohm resistance, adjust, resolder bridge). Yes, I know that was stupid risk to switch on, I would not have done with expensive gear, but my girlfriend annoyed me to get finished or throw it away...
Unfortunately, a few problems remain:
1. There is a humming noise in the amp. It also stays there if I split Pre/Power section. Is this due to the fact that the chassis is open on the bench??? What is a possible reason? How can I trace this down???
2. The combined volume/balance pot seems to suffer crackling due to dust. It seems unobtainable. Which is a recommended product (contact spray that minimizes the problem...)
Thanks, Dirk
Pleased you have fixed it...
The bulb by the way goes in the mains input, in series with the mains supply if ever you need to do that again, not in the amplifier as such.
Bias... you must adjust that correctly. If it's running hot then that suggests it's too high. And you are sure the PSU is OK and the amp is running on the "lower" voltage. If it were not then that would cause large increase in heat too. Check both positive and negative rails are correct.
Noise/hum. In the amp (you say) ? or is it from the speakers ? May well be that it's not all screwed together and grounded properly.
Noisy contols can sometimes be cleaned OK, sometimes not if it's due to wear. Some controls can be dismantled (used to do this a lot in the repair trade years ago) and the track and contacts cleaned and moved slightly so they run on a fresh piece of track.
Cleaning on its own is not always successful. What to use... dare I say try a little WD40 squirted into the control itself.
The bulb by the way goes in the mains input, in series with the mains supply if ever you need to do that again, not in the amplifier as such.
Bias... you must adjust that correctly. If it's running hot then that suggests it's too high. And you are sure the PSU is OK and the amp is running on the "lower" voltage. If it were not then that would cause large increase in heat too. Check both positive and negative rails are correct.
Noise/hum. In the amp (you say) ? or is it from the speakers ? May well be that it's not all screwed together and grounded properly.
Noisy contols can sometimes be cleaned OK, sometimes not if it's due to wear. Some controls can be dismantled (used to do this a lot in the repair trade years ago) and the track and contacts cleaned and moved slightly so they run on a fresh piece of track.
Cleaning on its own is not always successful. What to use... dare I say try a little WD40 squirted into the control itself.
Hi Mooly,
the fix was possible due to your enourmous support. Otherwise, it would have lasted much longer...
I never worked with bulbs, I usually removed the board and connected it to my Lab supply with current/voltage control... ;-)))
So another important lesson learnt!
So next I 'll do: Check all voltages, then adjust bias, next I'll try to mess around with the WD40... (Please don't kill me for this!!!)
I will then reassemble the hole stuff and give it a listen...
I'll come back with the results.
Dirk
the fix was possible due to your enourmous support. Otherwise, it would have lasted much longer...
I never worked with bulbs, I usually removed the board and connected it to my Lab supply with current/voltage control... ;-)))
So another important lesson learnt!
So next I 'll do: Check all voltages, then adjust bias, next I'll try to mess around with the WD40... (Please don't kill me for this!!!)
I will then reassemble the hole stuff and give it a listen...
I'll come back with the results.
Dirk
Stupidity of the day...
Hi Mooly,
today I removed the soldering bridge to measure the idle current. It was 143mv over 1Ohm, so it was much to high. Around 140 mA vs. 50 mA as you proposed. Unluckily, my measuring instrument slipped across, shortcutted something, blew Q413, Q19 with a lot off smell.
Now I have full offset voltage again, but further damage.
I think, I'm done with this amp ;-(((((((((((((
It's not worth the money to bother. I gained a lot of experience.
But for future repairs: How can I prevent to slip away while measuring? What do you use to measure without slipping away??? It was always a hot job - but today I kicked my ***...
Regs, Dirk
P.S.: Thanks again for your great help and guidance. I nearly succeeded... ;-(((
Hi Mooly,
today I removed the soldering bridge to measure the idle current. It was 143mv over 1Ohm, so it was much to high. Around 140 mA vs. 50 mA as you proposed. Unluckily, my measuring instrument slipped across, shortcutted something, blew Q413, Q19 with a lot off smell.
Now I have full offset voltage again, but further damage.
I think, I'm done with this amp ;-(((((((((((((
It's not worth the money to bother. I gained a lot of experience.
But for future repairs: How can I prevent to slip away while measuring? What do you use to measure without slipping away??? It was always a hot job - but today I kicked my ***...
Regs, Dirk
P.S.: Thanks again for your great help and guidance. I nearly succeeded... ;-(((
Oh dear... at least it's a learning experience as you say.
How to prevent slips. Down to practice, after a lifetime in the repair trade you get used to holding two DVM leads in one hand and a 'scope probe in the other 😀
You could solder wires to the test points I guess and run them to the meter.
Whatever has happened it's only going to be a few semis and resistors that have popped... put it on one side and come back to it again at a later date.
How to prevent slips. Down to practice, after a lifetime in the repair trade you get used to holding two DVM leads in one hand and a 'scope probe in the other 😀
You could solder wires to the test points I guess and run them to the meter.
Whatever has happened it's only going to be a few semis and resistors that have popped... put it on one side and come back to it again at a later date.
- Status
- Not open for further replies.