SA-8100 protection circuit issue
- Thread starter KeithD
- Start date
petehall347
the brandy coffee man
yes you could try that . or try one resistor back in at a time . see if it narrows it one channel or both .
Yes, that's probably a good first step. That would narrow it down.
Turned on the amp with D5, R9, and R10 out. It has not dropped into protection after 35 minutes. I did a couple of checks on pins 2 and 4. Pin 2 was at around -3mV after the amp was turned on and at 35 minutes. Pin 4 was at -100mV initially, and fell to about -10mV at the 35 minute mark. Seems to me that pin 4 on the protection board -> pin 23 on the power amp board is giving high values. In the neutral potential adjustment procedure both 3 and 23 should be 0 +/-20mV, and I know I had them down to +/-1mV when I made the adjustments.
petehall347
the brandy coffee man
one or both dc offsets is spiking up for a brief time sending it into protect . try one resistor back in at a time .
could be a transistor . most likely diff pair . try narrow it down to one channel .
could be a transistor . most likely diff pair . try narrow it down to one channel .
petehall347
the brandy coffee man
leave D5 out for now . less complications then .
Put R9 back in, pulled R10 and monitored pin 2. Average voltage was -5mV. The unit dropped into protection and the voltage was -1.7V. So, it seems the channel on pin 2 is the problem. Pin 2 on the protection board connects to pin 3 on the power amp board. Q1 and Q3 are on that circuit, but I replaced those two. There are also several resistors and two capacitors. Where might I look next?
petehall347
the brandy coffee man
did you change offset trimmers ?
Yes, I did that after replacing Q3 and Q4 on the protection board (see post 246). I think the offset is set ok, since I get steady readings for long times around zero volts. But the voltage on pin 2 of the protection circuit really spiked sending the unit into protection. I keep coming back to the feeling that something is warming up over about 15 minutes and then momentarily discharging enough voltage to trip the protection.
petehall347
the brandy coffee man
looking like a bad connection somewhere . could be a transistor popping . are the pre outs disconnected ?
try a shorting plug into main in and see if problem persists .
freeze spray might find it . along with some gentle heat like a hair drier .
try a shorting plug into main in and see if problem persists .
freeze spray might find it . along with some gentle heat like a hair drier .
petehall347
the brandy coffee man
looks more like a transistor to me . i have heard if you have a scope of caps doing this also .
if you have a scope you might trace it by looking for noise in the circuits . a dmm might not be fast enough to see it . freeze spray and heating whilst monitoring output might find it quickly . intermittent faults are always the hard ones to find . its right on it or chase around the houses .
if you have a scope you might trace it by looking for noise in the circuits . a dmm might not be fast enough to see it . freeze spray and heating whilst monitoring output might find it quickly . intermittent faults are always the hard ones to find . its right on it or chase around the houses .
OK, one thing that might be easy to do is reflow the solder on all the transistors on the power amp board. By freeze spray I assume you mean the spray janitors use to get gum off the floors - that type of thing, just to cool down individual joints and see if that does anything? Same with heat?
Also, might not be a bad idea to replace all the electrolytic caps on the power amp board, assuming the parts are not too expensive. I'd have to identify all the replacement parts. I believe some have mentioned there is a list somewhere on AK?
Also, might not be a bad idea to replace all the electrolytic caps on the power amp board, assuming the parts are not too expensive. I'd have to identify all the replacement parts. I believe some have mentioned there is a list somewhere on AK?
Reconnected all resistors and diodes on the protection unit, and reflowed solder on all original transistors (ones I had not replaced).
It the unit did still drop into protection for a couple of seconds after about 10 minutes. I have a Fluke 287 dmm that is able to take continuous readings and report average/high/low readings. Decided to monitor pins 3 and 23 on the power amp board.
Pin 3 averaged 0mV +/-6mV
Pin 23 was much more noisy. Average 0mV +10/-43mV
Pin 23 is the out put for one channel with input pin 16. Monitored that pin: -0.035mV +0.167/-0.149mV
Pin 10 is the other input channel: -0.51mV +0.015/-0.122mV
Pin 16 is quite a bit more noisy than pin 10. Is it possible the power amp is just amplifying noise coming in on pin 16? I think next I'll need to monitor pin 16 when the unit goes into protection to see if there is a spike there.
It the unit did still drop into protection for a couple of seconds after about 10 minutes. I have a Fluke 287 dmm that is able to take continuous readings and report average/high/low readings. Decided to monitor pins 3 and 23 on the power amp board.
Pin 3 averaged 0mV +/-6mV
Pin 23 was much more noisy. Average 0mV +10/-43mV
Pin 23 is the out put for one channel with input pin 16. Monitored that pin: -0.035mV +0.167/-0.149mV
Pin 10 is the other input channel: -0.51mV +0.015/-0.122mV
Pin 16 is quite a bit more noisy than pin 10. Is it possible the power amp is just amplifying noise coming in on pin 16? I think next I'll need to monitor pin 16 when the unit goes into protection to see if there is a spike there.
petehall347
the brandy coffee man
you could try a shorting plug into main in rca . that should see if problem is further down the line or not .
I'm not familiar with that. What does it entail?
petehall347
the brandy coffee man
basically shorting inner to outer on a rca plug . old scrap lead cut down and twist wire together is good enough .. any mv at the input should then be gone .
OK, did that. After turning the amp on the reading on pin 10 was 0.0mV, pin 16 -1mV variable. For some reason there was still a small voltage on pin 16. Pin 3 was -12mV, pin 23 -19mV. Monitored pin 23. after about 10 minutes unit tripped into protection for 1 second, no change in voltage on pin 23. So, either pin 3 went out of line, or this points back to the protection circuit... 
It seems like we just can't pin down the issue. I will need to start checking circuits and individual transistors for noise with the scope. Do I check all three legs on the transistors or just E, C?
It seems like we just can't pin down the issue. I will need to start checking circuits and individual transistors for noise with the scope. Do I check all three legs on the transistors or just E, C?
