I recently serviced a Technics SA-5760. New caps, diodes, transistors, etc where appropriate. I worked fine for a couple of weeks. Then the output relay wouldn't engage. It turned out that one of the few transistors I didn't replace (input differential pair 2SA798 (combined housing) on poweramp board) crapped out. I replaced those, the relay works once again, but now I'm faced with a right speaker output that clips on the positive side of the waveform at about 1,2 to 1,5 volts, depending on the temperature. Thinking it was a problem with the output transistors, I swapped them with the other channel, but the issue remained. Everything looks and measures fine. No odd or missing voltages as far as I can tell. In a move of desperation, I began swapping out components with the other channel, but after all transistors, caps and the NTC were swapped, the problem still remained. I hope someone with maybe expierence with this amp or similar can offer some advice. Any insight or guidance is more than welcome. I attached the service manual below.
The topic title refers to a "monster amp", the service manual is of an old timer stereo receiver?
Is this the output stage you are working on?
Is this the output stage you are working on?
Hahaha I thought these amps (and some types by other brands) were known as 'monsters', given their weight and size.
The image is indeed the section I'm working on. In this case it's the lower part (parts with even numbers) that's giving me a headache, but they are identical.
The image is indeed the section I'm working on. In this case it's the lower part (parts with even numbers) that's giving me a headache, but they are identical.
OK, I missed the weight... 🙂
It seems that you need to take first the DC voltages from key points and compare the broken channel to the good one - a set like this is pretty useful, especially when powering up using a variac and monitoring the voltage changes at that time:
A walk with a scope would also be useful.
Without that information it is kind of difficult to guess anything - even resistors can start malfunctioning (go open or intermittently open),
It seems that you need to take first the DC voltages from key points and compare the broken channel to the good one - a set like this is pretty useful, especially when powering up using a variac and monitoring the voltage changes at that time:
A walk with a scope would also be useful.
Without that information it is kind of difficult to guess anything - even resistors can start malfunctioning (go open or intermittently open),
Last edited:
Does the clipping onset/amplitude vary with 8 ohm load vs. no load? A defect in the current limit circuit might provoke clipping. Viewed with a scope, do you see any differences in waveforms vs. the working channel?
A troubleshooting experiment that occurs to me: bias the amp output into the problem zone--- eg. connect about 2M resistor between -40V rail and base of TR601 right (feedback) side. I'd tack in this mod first on the working channel and you should observe about +2V at amp emitter outputs. Then add same mod to the problem channel and compare DCV readings. Apply 8 ohm loads if that helps dramatize the differences.
Good luck!
A troubleshooting experiment that occurs to me: bias the amp output into the problem zone--- eg. connect about 2M resistor between -40V rail and base of TR601 right (feedback) side. I'd tack in this mod first on the working channel and you should observe about +2V at amp emitter outputs. Then add same mod to the problem channel and compare DCV readings. Apply 8 ohm loads if that helps dramatize the differences.
Good luck!
I would use a single DVM, measure all relevant points and write the values into a printed copy of the schematics.
Afterwords turn-off the power and study your schematics until you are convinced to have found the fault.
Afterwords turn-off the power and study your schematics until you are convinced to have found the fault.
So, I did as suggested and measured voltages on various points in the amp. The problem with this amp is the fact the board is very difficult to access when in place, so I had to resort to various time consuming ways to get all the info. As it turned out, it was a single resistor (R645) that failed open. Onze replaced, all was good.
