Tweet
Enzo,
It's alive - I hope
Hope this isn't too long-winded.
Been doing some reading. And looking at the schematic. This thing might act like a big servo,
but it sure looks like the innards of a big, discrete, op amp.
When I said the output looked like a half wave rectifier, it might not have been
exactly what was going on. I was using the scope on AC input, and what I was looking
at might have been a severely clipped output - on the negative side of the waveform.
Maybe not. Wish I had thought to look for an offset at the ouput with the scope
The problem is that now I can't get it to mess up. After last post I removed the unleaded/resoldered
D24, Q9, Q10 and a few other components. After that, I started watching the output with the scope on
DC. I saw the sine wave start drifting up once, but after a second or two, it was pulled back down.
Haven't seen it since.
The next day I replaced the lead-free on the entire power amp section with the good stuff.
It's still working with no fubars so I don't think I hurt it any.
It looks better - the lead-free looked dull and many joints were dull gray & grainy
maybe even whisker-y on some joints.
(Noticed that when I reflowed a few days earlier that it would look really good and then,
when it cooled enough, would skin over with what looked like a cold joint.)
During resoldering noticed that R145 was a little loose and the the trace was, while not broken,
partially detached and creased at the D58 pad connected to R145. Reinforced it with a piece
of copper hookup wire just in case.
BTW, the schematic calls for 200ohm 5W, for R144, R145. This amp has 330ohm 5W.
They look an identical type to the rest so I'm thinking this was a design revision.
Got out my old Active Devices textbook.
Was looking at the diff amp wrong. I thought it was fed at both inputs. Don't think so.
If I'm looking at it right, it is a single input at base of Q9.
If I am now seeing this right, U6-B is an input buffer and U6-A is the servoamp (or whatever the correct terms are).
Both speaker outputs are fed back to the junction of R77 and R69 and then to non-inverting input of U6-A
and the feedback loop diodes of U6-B.
Is there a cancellation at R77 & R69 and what's left is fed to U6-A and the base of Q10?
Is this where the DC offset is detected or controlled?
I was concerned about the waveform at Q10, but it's working fine, so does it matter?
Still seems to me that D24 was right in the middle of the problem...whatever it was.
And that +.5v shift in the output transistor bases...and just about everywhere else I checked...
is it just a coincedence that it's the same forward voltage drop as many of the diodes in this amp?
Hate it when something starts working again and I'm not sure why.
Doesn't make for confidence in future reliablity.
It's alive - I hope
Hope this isn't too long-winded.
Been doing some reading. And looking at the schematic. This thing might act like a big servo,
but it sure looks like the innards of a big, discrete, op amp.
When I said the output looked like a half wave rectifier, it might not have been
exactly what was going on. I was using the scope on AC input, and what I was looking
at might have been a severely clipped output - on the negative side of the waveform.
Maybe not. Wish I had thought to look for an offset at the ouput with the scope
The problem is that now I can't get it to mess up. After last post I removed the unleaded/resoldered
D24, Q9, Q10 and a few other components. After that, I started watching the output with the scope on
DC. I saw the sine wave start drifting up once, but after a second or two, it was pulled back down.
Haven't seen it since.
The next day I replaced the lead-free on the entire power amp section with the good stuff.
It's still working with no fubars so I don't think I hurt it any.
It looks better - the lead-free looked dull and many joints were dull gray & grainy
maybe even whisker-y on some joints.
(Noticed that when I reflowed a few days earlier that it would look really good and then,
when it cooled enough, would skin over with what looked like a cold joint.)
During resoldering noticed that R145 was a little loose and the the trace was, while not broken,
partially detached and creased at the D58 pad connected to R145. Reinforced it with a piece
of copper hookup wire just in case.
BTW, the schematic calls for 200ohm 5W, for R144, R145. This amp has 330ohm 5W.
They look an identical type to the rest so I'm thinking this was a design revision.
Got out my old Active Devices textbook.
Was looking at the diff amp wrong. I thought it was fed at both inputs. Don't think so.
If I'm looking at it right, it is a single input at base of Q9.
If I am now seeing this right, U6-B is an input buffer and U6-A is the servoamp (or whatever the correct terms are).
Both speaker outputs are fed back to the junction of R77 and R69 and then to non-inverting input of U6-A
and the feedback loop diodes of U6-B.
Is there a cancellation at R77 & R69 and what's left is fed to U6-A and the base of Q10?
Is this where the DC offset is detected or controlled?
I was concerned about the waveform at Q10, but it's working fine, so does it matter?
Still seems to me that D24 was right in the middle of the problem...whatever it was.
And that +.5v shift in the output transistor bases...and just about everywhere else I checked...
is it just a coincedence that it's the same forward voltage drop as many of the diodes in this amp?
Hate it when something starts working again and I'm not sure why.
Doesn't make for confidence in future reliablity.
Comment