Also, R53:
40ohm
200ohm
C26 to ground:
0.0
62.8
Q8:
0,0,0
61.1, 60.8, 61.8
Q10:
0,0,0
62.5, 63.4, 62
Apparently the bad channel doesn't turn off everything left off Q17.
Also, R55 was disconnected a long time ago.
40ohm
200ohm
C26 to ground:
0.0
62.8
Q8:
0,0,0
61.1, 60.8, 61.8
Q10:
0,0,0
62.5, 63.4, 62
Apparently the bad channel doesn't turn off everything left off Q17.
Also, R55 was disconnected a long time ago.
Last edited:
Hi Peter,
I'm sorry I'm so slow to respond. Life intervenes...
I annotated the partial schematics with your measured voltages and will post them for comparison, tomorrow. (Late here.) An interesting phenomenon present in both working and failed units is the behavior of the emitter voltages at Q11 and Q9. In all four cases, the base-emitter junctions are reversed biased. The only explanation I can guess is that bias currents are so small the meter load currents to ground are sufficient to generate reverse biasing.
I suggest connecting the reference lead of the meter to the positive rail and probing the base and emitter voltages of Q11 to see if Q11's b-e junction then shows expected forward bias. If Mute were released in the good channel, I think this phenomenon would disappear.
The other glaring phenomenon in the problem channel is the roughly 60V on the collector of Q11. This implies 0.3W dissipation in both R73 and R48. Assuming the center ground path is intact, these resistors should be getting HOT! Are they? And it they are hot, where is the ~5mA+5mA current coming from? And how does this square with the small currents seemingly implied in the previous paragraph?
I think the problem lurks in this area.
TTYL
I'm sorry I'm so slow to respond. Life intervenes...
My apologies, as I intended to type left of Q37 rather than Q17, but didn't notice the error. In Mute, the balanced differential input stages are unbiased by virtue of having their current sources disabled. Later stages pass minimal current when the amp is Muted.
I annotated the partial schematics with your measured voltages and will post them for comparison, tomorrow. (Late here.) An interesting phenomenon present in both working and failed units is the behavior of the emitter voltages at Q11 and Q9. In all four cases, the base-emitter junctions are reversed biased. The only explanation I can guess is that bias currents are so small the meter load currents to ground are sufficient to generate reverse biasing.
I suggest connecting the reference lead of the meter to the positive rail and probing the base and emitter voltages of Q11 to see if Q11's b-e junction then shows expected forward bias. If Mute were released in the good channel, I think this phenomenon would disappear.
The other glaring phenomenon in the problem channel is the roughly 60V on the collector of Q11. This implies 0.3W dissipation in both R73 and R48. Assuming the center ground path is intact, these resistors should be getting HOT! Are they? And it they are hot, where is the ~5mA+5mA current coming from? And how does this square with the small currents seemingly implied in the previous paragraph?
I think the problem lurks in this area.
TTYL
Last edited:
Yes, R48/R73 are slightly discolored but I haven't noticed them getting too hot. They look similar on the good amp, so I'll replace them with higher wattage resistors.
Perhaps they have failed. They should be hot to touch. Together they would be drawing about 10mA from the positive rail. What voltage drop do you see across R23? Compare with the good channel.
Q11, positive probe on positive rail, probing with negative:
87.6, 87.8, 87.5.
Same with mute on or off.
87.6, 87.8, 87.5.
Same with mute on or off.
R23:
2v
4.5v
Across R48:
61v
Across R73:
64v
... and yes, these are warm. These are 0v on the good amp.
Another oddball, across these diodes:
(Good amp on top)
D14:
83
142(!)
D14
83
18(also !)
Diode string from Q24-R14:
-86.6/-87
-85.4/-86.4
-85.2/-85.5
From Q37-R14:
86.5/84.3
85.9/83.8
85.2/83.2
2v
4.5v
Across R48:
61v
Across R73:
64v
... and yes, these are warm. These are 0v on the good amp.
Another oddball, across these diodes:
(Good amp on top)
D14:
83
142(!)
D14
83
18(also !)
Diode string from Q24-R14:
-86.6/-87
-85.4/-86.4
-85.2/-85.5
From Q37-R14:
86.5/84.3
85.9/83.8
85.2/83.2
At first glance, these similar voltages seem to confirm my theory that the reverse-biased b-e junctions were the result of meter load resistance currents to ground. But on closer inspection they made no sense, as I would anticipate only a few volts showing on the rail referenced voltmeter, but you find nearly the entire supply voltage. This is inconsistent with Q11 voltages in post 301; and the Q11 voltages seem inconsistent with there being 60V at R73 re ground.
I suggest printing a copy of the sub-circuit PDF in post 300 and recording the observed voltages on the schematic. Because the most recent measurements were the most mysterious, I suggest again taking the readings with positive meter probe on the positive rail, eg at any convenient point tied to Q37 collector. Having done that, I would connect meter to ground and probe at same points and confirm sensical, consistent readings.
Last edited: