So just to clarify, it seems the Silver version of this amp doesn’t have a zobel network - which I think is ok because I’m pretty sure the speakers I’m using have compensation in the crossover. But I learned something today!
All the passive resistors measure ok, as do the diodes. It could be leaky transistors.
But while researching replacements I found this interesting thread and will look on the other boards as well: https://audiokarma.org/forums/index.php?threads/repalacement-for-2sc1775-dc-offset.186869/
It won’t be tonight though as I’ve had a few red wines and don’t’ want to play with electricity….
But while researching replacements I found this interesting thread and will look on the other boards as well: https://audiokarma.org/forums/index.php?threads/repalacement-for-2sc1775-dc-offset.186869/
It won’t be tonight though as I’ve had a few red wines and don’t’ want to play with electricity….
ok, I didn’t plug anything in, but I did do a few more checks. Interesting that R453 measures as 30 ohms in circuit. Pulled it out and it’s correct at 100 ohms. I’m guessing that means there’s something wrong with Q423. R457, R463, R447, R443 all measure correctly in circuit.
There may not be anything wrong with Q423. Probably there's a transistor junction shunting the resistance measurement--- you've already confirmed 100 ohm out-of-circuit. Be sure to measure R453 with both ohmmeter polarities--- if the reading differ, there's a transistor junction shunting the measurement. Both Q423 and Q419 are tied to R453, so in may not be possible to find a polarity that doesn't cause junction conduction, and thus no 100 ohm in-circuit. Compare with behavior at R454 for reassurance.
I'm traveling, so my replies may be erratic. Good luck.
I'm traveling, so my replies may be erratic. Good luck.
Confirmed Q423 was faulty. Not open, not short, but giving low readings when measured with a multimeter diode tester and also reading values when it shouldn't.
I had replaced the SA1111/SC2591 pair on the right channel way back at the start because they had blown, I just did the ones on the left channel recently with the other pair that I had bought at the time, but I wasn't sure if they were genuine. I would love to know if there was a modern replacement that will work. For now the originals have gone back in as they measured ok.
Haven't hooked up again to test yet, I ran into some issues with a stripped thread on the earth lead and then broke the mounting tab off the heat sink...time to go to bed!
I had replaced the SA1111/SC2591 pair on the right channel way back at the start because they had blown, I just did the ones on the left channel recently with the other pair that I had bought at the time, but I wasn't sure if they were genuine. I would love to know if there was a modern replacement that will work. For now the originals have gone back in as they measured ok.
Haven't hooked up again to test yet, I ran into some issues with a stripped thread on the earth lead and then broke the mounting tab off the heat sink...time to go to bed!
have replaced Q423 and the problem still exists, not unexpected as it’s the other channel that seems to have the issue.
Bump…based on the above it seems the replacement Q424/426 combo that I installed are probably not quite right, so I need to see if I can source some genuine replacements (2SC2591/2SA1111).
Does it matter much that the good channel is not balanced? ie the voltage across R468/470 is quite different.
Does it matter much that the good channel is not balanced? ie the voltage across R468/470 is quite different.
Would you detail the voltages you're seeing across R468,R470,R472,R474? Be sure your measuring with no load at output terminals.
Further up in the post, whilst I was setting the bias. I’ve since increased to a total of 33mv, but one channel is balanced (the bad one) and the good one isn’t.
“One channel is definitely not as balanced. When I was setting bias initially, left was 11.5+11.5, while right was 10.5+12.5. Across R406 is 5mV on left and ~20mV on right. The 2SA1111/2SC2591 pairs have been replaced on both sides.”
The right channel now has the original transistors back in, while the left are new but I I think not genuine. My next step would be to swap them over.
“One channel is definitely not as balanced. When I was setting bias initially, left was 11.5+11.5, while right was 10.5+12.5. Across R406 is 5mV on left and ~20mV on right. The 2SA1111/2SC2591 pairs have been replaced on both sides.”
The right channel now has the original transistors back in, while the left are new but I I think not genuine. My next step would be to swap them over.
An update - I swapped the 2591/1111 pairs over and adjusted bias to ~30mV total. (somewhere in between the two different views of what it should be).
The imbalance is still there, left is 15.3/15.2 and right is 16.4/13.4.
However after 30 minutes of playing it hasn’t dropped out at all, so I’m a bit stumped now.
The imbalance is still there, left is 15.3/15.2 and right is 16.4/13.4.
However after 30 minutes of playing it hasn’t dropped out at all, so I’m a bit stumped now.
Tough to troubleshoot if it's behaving. You may have to wait unit it acts up again.
I'll offer some thoughts.
I don't understand your notation--- does 16.4/13.4 mean 16.4mV across R473 and 13.4mV across R471 ?
The design monitors only one of the two output pairs, eg. R473 and R471 are monitored, but R469 and R467 are ignored.
The voltage between the emitters of Q431 and Q427 should match Q433 and Q429. Be suspicious if you find dramatic differences.
With power off, try to measure and tabulate resistance of all four emitter resistors at the leads. Ideally they'll be equal, but differences may contribute to the different voltages you're observing. Try reflowing all these solder joints to ensure low (reliable) resistance to emitters and output. Heat in these areas can be a stressor of these solder joints.
I'll offer some thoughts.
I don't understand your notation--- does 16.4/13.4 mean 16.4mV across R473 and 13.4mV across R471 ?
The design monitors only one of the two output pairs, eg. R473 and R471 are monitored, but R469 and R467 are ignored.
The voltage between the emitters of Q431 and Q427 should match Q433 and Q429. Be suspicious if you find dramatic differences.
With power off, try to measure and tabulate resistance of all four emitter resistors at the leads. Ideally they'll be equal, but differences may contribute to the different voltages you're observing. Try reflowing all these solder joints to ensure low (reliable) resistance to emitters and output. Heat in these areas can be a stressor of these solder joints.
My meter doesn’t have enough resolution to go lower, the resistors all measure as 0.4 Ohms
LCh
Q428-432 is 30.7v
Q430-434 is 32.0v
RCh
Q427-431 is 25.0v
Q429-433 is 22.0v
I know the right channel is not balanced but I can’t figure out why.
In any case, it seems to have stopped cutting out since I swapped the drivers between channels so I’ll cut my losses and put it back into service, while keeping an eye out for some transistors.