Fender 212R power amp blown

[lowell]

Ok. Well I plugged into J1 and the DC offset at the output bus is gone and the amp works fine.

D21, 22, 23 all have .5v across them now.

Q9Vbe: .6v
Q9 Vc: 39.3v
Q9 Ve: 0v

I'm confused now. So this amp is designed to be muted with nothing plugged in to J1, J2, or J4. So say the amp is all assembled and connected to the speaker and nothing is plugged into those jacks... the amp is then muted but there is -4v on the speaker. ??

[MarkusBass]

Ok. Well I plugged into J1 and the DC offset at the output bus is gone and the amp works fine.

You see? So operations over the internet are possible .

D21, 22, 23 all have .5v across them now.

Much better than before.

Q9Vbe: .6v
Q9 Vc: 39.3v
Q9 Ve: 0v

Also better but here I would quickly check what is voltage on R82 (and calculate current), and measure voltages on R85 and R86 just to see what are the currents.

I'm confused now. So this amp is designed to be muted with nothing plugged in to J1, J2, or J4. So say the amp is all assembled and connected to the speaker and nothing is plugged into those jacks... the amp is then muted but there is -4v on the speaker. ??

Is this voltage with a speaker plugged in?
I would again look at the schematic. There is a DC negative feedback in the amp: R79 from the output, R77, R74 and R67 to inputs of U6. But when the amp is muted, the differential input (Q9 and Q10 - connected to U6) is not working. So the voltage on the output relays on the symmetry of the output stages of the amp. And it looks like the output stage is symmetrical except for Q12 and Q13, which are connected slightly differently. I would check (with the amp being muted) what are the voltages on bases of Q20 and Q21, what are the voltages on bases of Q18 and Q19, what are the voltages on anode of D28 and cathode of D29, and finaly what are the voltages on collectors of Q12 and Q13 (and possible their bases). This would give you some more information how the amp works.

Mark

[lowell]

Markus... ok seems we're getting somewhere. Thank you!

The DC offset voltage is with no load.

Yes Q12 and Q13... the point where they connect is also the point where the bias transistors meet. I'm guessing they play a part in the biasing but I'm not sure how. Especially odd to me is that Q12 is pnp and Q13 is npn being that they're opposite from the supplies that I'm used to seeing them with.

I'll report back tomorrow with voltages on bases of outputs, drivers, D28 and D29, and Q12 and Q13 collectors/bases.

[MarkusBass]

Markus... ok seems we're getting somewhere. Thank you!

Great, I'm happy if you are happy.

The DC offset voltage is with no load.

Hmm, I thought that you mentioned that the voltage is on the speaker. But it was rather on speaker jack without a speaker. So the information was not precise.

Yes Q12 and Q13... the point where they connect is also the point where the bias transistors meet. I'm guessing they play a part in the biasing but I'm not sure how. Especially odd to me is that Q12 is pnp and Q13 is npn being that they're opposite from the supplies that I'm used to seeing them with.

I don't think that this is odd. This is where relativity comes into play. Q13 does not know that it is connected to negative rail. D27 and R89 provide a positive voltage that causes that the transistor conducts. The same story is with Q12. It doesn't matter that this is connected to negative rail. The transistor does not know about it .
I hope you learned something. Especially about the "easy way". Even if you'd replace all semiconductors, the amp would still work in the same way. And the printed circuit would be in much worse state. So it makes sense not to go an easy way - Yoda was right .

Mark

[lowell]

So is the amp fixed? What about the DC offset if the owner doesn't have a guitar plugged into the input? Ok I see how Q12 and Q13 are biased now, but what is their purpose?

[Enzo]

Did we get an answer? Wuth no load you have 4v offset on the outs, but does that go away when the load is present?

[lowell]

Hmmm.... I'll check tomorrow. I thought solid state power amps preferred no load. So with that being their ideal state shouldn't there be 0v offset without a load? Maybe you're asking me to see if it goes away without a load because that will lead us to the issue.

[Enzo]

SS amps are happy as a clam without a load. At least most times. I don't recall one that mattered offhand. But especially considering this convoluted circuit, we owe it to ourselves to see if the offset is an artifact. Perhaps with the load on it, it settles down, and thus problem solved. No, I don't normally expect things to act that way, but never assume. And yes, if it does, that may be illuminating.

[lowell]

Ok thanks, I'll report back tomorrow.

[MarkusBass]

So is the amp fixed? What about the DC offset if the owner doesn't have a guitar plugged into the input? Ok I see how Q12 and Q13 are biased now, but what is their purpose?

Lowell,

You are asking us such questions but you haven't provided us with answers to our previous questions. Previously you wrote:

The DC offset voltage is with no load.

Does it mean that with a load there is DC offset, or there is not? There can be only one answer: yes or no. Which one is it in your case?
I also do not understand you worries about the amp owner that does not have a guitar plugged into the input. Do you mean that the owner of the amp is going to use the amp without a guitar plugged into the input, and possible without a speaker plugged into the speaker jack? I think that in this case the amp is going to work very well .
Regarding the purpose of Q12 and Q13; this is a stage between the input differential stage and the output stage. So it shouldn't be a surprise that the stage provides voltage and current gain for the output transistors.

Mark

[Enzo]

Mark, I THINK what he is concerned over is that with the inputs empty, a DC voltage will appear on the speaker, and potentially damage the speaker. I think.

And that is of course why we need to know if that DC offset is present with a load or without a load. That tells us if the situation even exosts for the speaker.

[MarkusBass]

Mark, I THINK what he is concerned over is that with the inputs empty, a DC voltage will appear on the speaker, and potentially damage the speaker. I think.

I think so too. So this is a case without a guitar but with a speaker.

And that is of course why we need to know if that DC offset is present with a load or without a load. That tells us if the situation even exists for the speaker.

Appart from measuring the offset directly (which does not seem to be a difficult task), the amp can be easily verified by plugging in a guitar (so the amp is un-muted and DC servo should work), disconnecting the speaker and measuring the offset on the output. If the offset is 0V, the amp seems to be fixed (DC servo works as expected). If it's not, Lowell has to look at it again.
Even if it is fixed, I would still check the amp with a resistive load, a genarator and oscilloscope (half output power for 10-20 minutes).

Mark

[R.G.]

No magic insights here, but a possible explanation of the problems with no load connected. This power amp is set up with the old trick of raising the overall output impedance to let the speaker's impedance curve affect overall gain, so the speaker's impedance curve causes more prominent frequency response changes, like in a tube amp.

The relevant components are those resistors in the speaker return path to the power supply, and the capcitor that leads away from it (R114, R115, C59). These provide a feedback signal that makes the power amp circuit - which is, after all, just a big power opamp - increase its output voltage a bit into a higher impedance load. In theory, if the parts are all good, having a DC open circuit should make no difference to the DC voltage on the output.

However, the DC conditions on the power amp are changed by any input offsets. This amp has plenty of places to goof that up. Negative feedback from the output goes (eventually) to Q10 base, involving R69, R77, R78, R79, R80, C44 and the whole mess around U6-A.

I think that U6-A is a DC servo to keep the output at 0Vdc; I don't know exactly, because I don't like servos and don't use them in my amps, so I've self-limited my knowledge about them. I'd have to so some actual thinking (always hard and painful!) to suss that out. However, it is *very* clear that DC offset issues with the U6-A circuit and failures of the various capacitors in the feedback path or with U6-B would cause DC offsets.

These are difficult circuits to debug, and take time. I personally, on suspicion alone, would note that TL072s and capacitors are cheap and arbitrarily replace U6 and the various capacitors in the circuits I mentioned, if I had the capacitors. If you replace DIP ICs, clip the leads off at the body, then lift each lead out with the tip of your soldering iron. You can then poke a whittled-down wooden toothpick into the holes as you melt the solder and clear the hole.

Having this happen with the inputs open is one of those things which whisper "check for RF oscillation..." in my ear, so I'd want to scope it at both the input and output of the power amp. It could be that oscillation from earlier in the amp is being converted to a DC offset in U6A or B and telling the power amp to be offset, which it dutifully does. Or the disaster which got the transistors also damaged U6, which is tied right to the outputs through the feedback network.

[lowell]

Ok so with a speaker connected and with the amp muted there is still -4v DC offset. That concerns me because what if the amp is "on" but nothing is plugged in to unmute it?

[R.G.]

Ok so with a speaker connected and with the amp muted there is still -4v DC offset. That concerns me because what if the amp is "on" but nothing is plugged in to unmute it?

4V into a typical 8 ohm speaker's 6 ohms resistance is 666ma, 2.6W continuous into the voice coil wire. That's not disastrous, but it's not good for the speaker either.

Does the -4V go away when something is plugged in?