Fender 212R power amp blown

[MarkusBass]

Markus,
The vdrops across D21-23 are all different. However the vdrops across D25-27 are all apprx .5v.

D21: fluctuates from .5-.9v
D22: .1v
D23: .02v

D22 and 23 are identical and connected in series. If there is any current flowing, the voltages would be identical. And as you suspect the voltage drop should be ~0.5V each. Quick and dirty solution would be to disconnect D28 (but first think what role it plays in the circuit), measure D22-23 again and if the voltages are the same, replace them (maybe including Q9).
I personally would still make some measurements. You have current source there, diodes. You could easily check whether these current sources work correctly, it is very easy to check every diode. What about fusible resistors? Have you checked them?
I would also check voltage drops across R85 and 86. Are they similar?

Mark

[J M Fahey]

Dear Markus, let me make clear that I *crave* for technological advance, if it's real.
In this case, there is not such thing, only useless complication, that's what scratches me the wrong way.
The amplifier *still* is technologically a 1972 vintage amplifier , with a few, mostly cosmetic, add-ons .
The basic amplifier structure is made out of:
Q12, which is the heart of this amplifier, providing most of its open loop gain, (over 60dB), and setting its main phase response, distortion, etc. , providing such gain loaded by:
Q13, its active load. Later we'll add something extra about it.
It receives input signal from:
Q9 / Q10 : the input differential pair. It also provides DC balancing (others might call it an offset correction circuit, fine with me) but specially provides a very convenient point to inject feedback, which in this, as in all Lin type amps, is massive (in the order of 40 dB). This is an important detail.
The huge, rail to rail voltage swing available on Q12/Q13's collectors, is all the voltage that will be available on the load, not an iota more, because later stages provide only current gain, (some may call it active impedance matching), but *no* voltage gain, nil, zero.
Q18/Q20 and Q19/Q21 are complementary Darlington pairs, with high current gains (the product of the individual transistors current gains), which provide said load driving, but which provide neither voltage gain nor , more important, waveshaping which could turn this into a more "tubey" amplifier.
So far, as I said earlier, a Classic amp.
Let me state an analogy: if somebody tells me he has invented a new, improved car engine, be certain I will be very interested (being the nerdy type) and will want to see it.
Now if I see that it burns leaded gas, has a carburetor, spark plugs, a distributor, and a mechanically actuated "platinum", I couldn't care less if the output manifold is made out of Titanium, its valves of depleted uranium, the gas filter is made out of a graphite/epoxy composite, and the tachometer is digital and run by the latest PIC, it is not technologically different from my Chevy '57 engine.
My lunch break is over, tonight I will explain all the kludgy add-ons as I see them, but more specially *why* I find them useless.
Sometimes the tree hides the forest.
EDIT: let us not forget that our friend Lowell still has a dead amp in his hands, any suggestion which may help him, is welcome.

[lowell]

Guys I replaced the diodes D21-23 and still DC offset at speaker. I don't have suitable transistors for Q9 and Q12 in stock so I ordered them from BnD. I'll report back once I install those. And thanks again for the help thus far. MB I checked the fusible resistors and they test good.

[MarkusBass]

Guys I replaced the diodes D21-23 and still DC offset at speaker.

This was not what I suggested. I suggested to desolder D28 (one pin), measure voltages on D21-23 and only then if they are still bad, replace the diodes (possibly along with Q9).

MB I checked the fusible resistors and they test good.

I don't know what you mean by "test good". I usually measure resistance with amp switched off, and then switch on the amp and measure voltage drop on each resistor. Did you do it?
Lowell, in similar cases I usually print amp schematic (as big as possible e.g. A3 format) and measure voltages looking for the ones that are incorrect (and I put these voltages on the schematic). When I see a voltage that is definitely incorrect (like e.g. 0.02V on D23) I look further. In this case I would measure voltages on D28, on Q9 (base-emiter), on R85 (and compare it with R86), on Q8 (base-emiter), R82. It is still posssible that the incorrect voltages are caused by a problem in the output section (since the amp is DC coupled) but still such measurements could give you more details what possibly is wrong.
I would also measure voltage drops on every resistor in vertical branch (if this is a correct term ). I mean on: R83, R84, R87, R88 and R91. Speaking in computer language; you have to have some input to calculate the output. You cannot try to guess what is wrong - as you see it takes too much time. This is a problem for 1-2 days.

Mark

[lowell]

I appologize Markus. I took Enzo's suggestion of replacing the 10 cent diodes just to see. I will take your advice too and check that other stuff and report back. Thanks!

[MarkusBass]

I appologize Markus. I took Enzo's suggestion of replacing the 10 cent diodes just to see. I will take your advice too and check that other stuff and report back. Thanks!

Lowell, there is no need to apologize. Enzo's suggestions are always very practical and good. Especially if you have an IC like TL072 in a socket and it will take only 5 seconds to replace it. But if the replacement requires some soldering, I analyze first the probability that this part causes problem. If I'm "almost sure", I replace it. If not, I start thinking. It's just a matter of aproach (and maybe experience). In your case I would collect as much information as possible. Replacing components that did not needed replacement somehow influences the printed board. You cannot do it many times. That's why I avoid it. But if you have good soldering iron and skilled hands, go for it - I don't see a problem.

Mark

[lowell]

Alrighty then. Here goes.

Voltage drops across these read:
R83:0v
R84:44mv
R87:20mv
R88:0v
R91:32v
R89:60mv
R90:10mv
R104:4mv
R106:1.1v
R102:35mv
D28:42v
Q9be:0v
R85:.1mv
R86:.1mv
Q8be:.3mv
R82:.5mv

[MarkusBass]

OK, but life is not that easy. You tell me which of these voltages does not seem to be correct.
Additional hint; to the left of Q8 you have Q7, which is a thermal protection and muting circuit. I hope the amp is not muted while you make measurements.
Hint 2: previously you reported that on D21 there is at least 0.5V. So the voltage drop on R83 cannot be 0V. Even Fender must adhere Kirchoff low. So either you measurements are not correct, or R83 is not connected to D21, is it?

Mark

[Steve Conner]

Solid state troubleshooting is a specialised skill. If you run a repair shop and don't have that skill, is it fair to take solid-state amps for repair and expect us to fix them for you?

99% of solid state amp failures go like this: an output transistor blows and takes a random selection of other components out with it. If you don't replace all of the blown components, the next time you turn it on, it blows the outputs again plus another random selection of components. If you don't replace all of those, it does the same again... And again... Troubleshooting that over the Internet is like doing brain surgery with a pointy stick. If you don't understand the circuit, a valid approach is just to replace every component. And like Markus says, don't wreck the PCB.

[lowell]

Ok I'll get back about this on Monday. Steve yes I'm asking for a lot of information that you guys have all learned on your own and I'm very thankful for the shared knowledge. I'm also reading alot on the subject and trying to teach myself about solid state circuits. I'm not charging the amp owner anywhere near what this repair might cost if I were a pro in this arena. It's more of a learning experience for me and the ower hopefully gets his amp fixed too. I do mostly tube repairs/building, but solid state circuits are something I'd eventually like to wrap my brain around.

[lowell]

Oh and Markus, I don't think the amp is muted because I can hear the hum from the DC on the output and it also sound like static you hear when an opamp is bad.

[lowell]

I'll recheck R83 and the diode string below it tomorrow. As for now...

Not sure how much voltage should be dropped across these components. I can say this though:

There should be some voltage across R83
R88 should have what R90 has across it right?... 10mv Although the voltage across R90 could be wrong if there's an issue with Q12.
R104 and 102 should be the same right? I'd suspect that the reading on R102 is off. My reasoning is that it's connected to Q9 through D28 and it seems there is an issue with Q9, though I'm surmising.
Q9 must be off with a Vbe of 0v. That tells me that Q9's emitter is not getting the negative voltage that it should be getting through Q8. And Q8 seems off as well with a Vbe of .3mv. Is Q8 a ccs for the differential splitter?

My guess, and I'm not sure if my thinking is right, is that the front end differential splitter is off due to Q8 being shut off. If Q8 were functional then Q9 would have a more reasonable Vbe, assuming Q9 isn't bad.

You seem to be pointing me toward the protection circuitry. I'm really trying to see how it works but it's making my brain hurt. Where is the "mute" connected to?

Is RT2 a thermistor? I believe it's mounted on the heatsink to monitor temperature. Maybe its value increases as it gets hotter thus turning Q6 and Q7 on more and allowing some of that -16v to reach the base of Q8... shutting it off?

[MarkusBass]

I'm a little bit busy now so just a quick response. Yes, you are right that both 0V are suspected.
It seems that you need some information about current sources (yes, Q8 is a current source). Please read this: Current source - Wikipedia, the free encyclopedia - especially the description of Figure 3. Current source maintain a constant current but it requires some initial condition to work correctly - there must be a voltage drop between base and emitter of the transistor. And this is (should be) provided by D20. Which in turn is connected to the curcuit with RT2 (yes, this is a termistor), and to the muting circuit. It is very important to see how the muting circuit works. It is connected to J1 INPUT. And if there is no plug in J1 nor in J2, it is connected to ground at J4 POWER AMP IN. I suggest to check all these connection to understand whether the amp is muted or not. When you clarify this, I suggest that you make sure that Q8 is correctly working as a current source and then check why Q9 is not conducting. It may be blown or it still gets some incorrect voltages.
I will take a look at other information you provided at home in the evening.

Mark

I'll recheck R83 and the diode string below it tomorrow. As for now...

Not sure how much voltage should be dropped across these components. I can say this though:

There should be some voltage across R83
R88 should have what R90 has across it right?... 10mv Although the voltage across R90 could be wrong if there's an issue with Q12.
R104 and 102 should be the same right? I'd suspect that the reading on R102 is off. My reasoning is that it's connected to Q9 through D28 and it seems there is an issue with Q9, though I'm surmising.
Q9 must be off with a Vbe of 0v. That tells me that Q9's emitter is not getting the negative voltage that it should be getting through Q8. And Q8 seems off as well with a Vbe of .3mv. Is Q8 a ccs for the differential splitter?

My guess, and I'm not sure if my thinking is right, is that the front end differential splitter is off due to Q8 being shut off. If Q8 were functional then Q9 would have a more reasonable Vbe, assuming Q9 isn't bad.

You seem to be pointing me toward the protection circuitry. I'm really trying to see how it works but it's making my brain hurt. Where is the "mute" connected to?

Is RT2 a thermistor? I believe it's mounted on the heatsink to monitor temperature. Maybe its value increases as it gets hotter thus turning Q6 and Q7 on more and allowing some of that -16v to reach the base of Q8... shutting it off?

[MarkusBass]

Lowell,

I just looked at the schematic and you can see how a capability to read and understand a schematic is important. Please check again how the amp is muted. The MUTE wire from the protection circuit is connected (besides Q8) to the J1 INPUT jack (ring), to the J2 INPUT jack (ring) and to the J4 POWERAMP IN jack (ring). And there it is connected to the ground. So if you do not have any plug in either J1, J2 or J4, the amp is muted. To "un-mute" the amp you have to insert a plug into any of these jacks. I bet that you didn't do it so you measured muted amp and most probably you should repeat the measurements once again. As a result of this, Q8 is most probably not conducting, which could influence Q9 and diodes and resistor connected to collector of this transistor.
Just to recap: MUTE connected to the ground results in amp being muted, disconnected from ground means the amp is unmuted.
So you see that when I would first find out 0V on R83, I'd think that the transistor is not conducting, I'd check voltages on Q8 and found out the problem with muted amp. And I would do it because I understand how the amp work. Without it you can spend a month trying to fix the amp. To be able to do this on your own you have to learn to read schematics. You have here just a few typical circuits like: constant current source, differential amplifier and emitter follower. You can learn it in a week.
Of course there is an easy way: replace all semiconductors in the amp (that haven't been replaced yet). But I wouldn't go this way - you won't learn anything in this way. And when you get another amp with similar problems, you would need to do again the same.
Also remember what Yoda said when he was asked: "Why not do it the easy way, Master?" . He said: "Because it is easy."

Mark

[lowell]

Markus thank you. I'll report back later today.