Fender 212R power amp blown

[J M Fahey]

Don't worry Little Enzo !! (Pat pat in your head).
These good guys at thew Forum will help you repair it !!
Please stop crying .
In no time, say in 300 posts, you'll have it as good as new !!!
he he he he

[lowell]

Ok so bandaid after bandaid they came to this pa. Seems they could trade in the bandaid parts for 2 more output transistors and keep costs pretty close.

[lowell]

Ha!! Another one just came in for repair. Seems muted as there's no hum or hiss or sound whatsoever. Gonna see if i can tackle this on my own thanks to all the help yall have given me on the first one.

[J M Fahey]

Yes, now you are on your own.
At least, you already have the 300 page manual "How to repair FM212 amps".

[Enzo]

I'm gonna have to read through this thread again to see what I am doing...

[lowell]

Ditto.

[MarkusBass]

I have a Fender Bassman 250 chassis in here now. SUffered a drink spill, and has.... DC on the output. The circuit? Same damned power amp circuit as this.

I don't have Bassman 250 schematic but if you check either Bassman 200 or 400, you can see that 212R is just simplified version of a higher power bass amp. So the design is not stupid - they just took a higher power amp and used it as a 100W amp. And the solutions used there (e.g. double bias circuit) are typical for such amps. Just compare it with e.g. Ampeg. So this is not a product of a crazy designer - they just wanted to save money by taking an existing design.

Mark

[MarkusBass]

Ok right on. I wasn't sure if this was an acceptable finished fix. I trust your expertise and it's good to know that it is indeed a good fix for a bad design.

Lowell,

Don't you see that, after disconnecting diodes and capacitors, what is left is fully symmetrical circuit (and even with those components is was also symmetrical)? If so, there is no reason for the DC offset. Since I'm tired with this amp and the way you try to fix it, I simulated the output stage in SPICE software. And I got only 15 mV on the speaker (possibly due to differences of transistors' parameters). Trying to fix the amp by adding a resistor paralel to R91 is like painting with a white paint a guy who suffers from jaundice. He will be white but will he be cured?
The amp still has some problem and you have to find it instead of doing hocus-pocus.

I wish we could figure out why the pa is not symmetrical. Must be that the negative side is "on" more than the positive side right?

Yes, you are right (if the voltage really is negative on the speaker).

It's just difficult to narrow it down I guess.

No, it's not. You have output stage that should be symmetrical. If you still don't know what is wrong with the amp, I would start from desoldering R103 (to disconnect the circuit from the D34-37 diodes). If this doesn't help, I would desolder the output transistors (and possibly driver transistors). Then you would have this only double bias circuit left, which cannot be unsymmetrical.

It seems to me that the diode string D34-D37 might be the cause. I don't quite understand how it functions though. I can't see how D37 could ever be turned on, or D35 for that matter. D34 drops 5.1v between R101 and D35.

Are you sure about this voltage on D34? You said that you replaced all these diodes and nothing has changed. Aren't they soldered reversed? To test it you can desolder R103 (or just look at the cathode symbol on each diode and on the board). In my opinion 5.1V od D34 is incorrect. The diodes should not conduct and there should not be 5V on them. Looks like at least D34 is soldered reversed.

On your place I would also measure voltages on Q14 (Vce), Q15(Vce), D30 and D31 - just to be sure that this part of the circuit is OK. You could also provide voltages on D34, D35, D36 and D37.
What voltage meter are you using? I'm asking because some of the voltages you specified are not reliable - they just don't follow Kirchoff's law. Is it an old low-input impedance meter?

Mark

[MarkusBass]

As you correctly noticed, D34 is forward biased (so it does not "Zener", and D35 is reverse biased, blocking any current through that string anyway).

I'm sorry but your guess here is completely wrong. It's exactly the oposite - at least in correctly working amp. D34 is not forward biased simply due to the fact that it is in series with D35 which is reversed biased. If D35 is reversed biased, there is not current flowing through it. So there is also no current flowing through D34 and there cannot be 5.1V on it. Of course this assumes that D34 is soldered in correct direction.

The designers are not nuts (I never even implied that) , so probably at some point during the output waveform swing , at a certain power level, say -6V into the speaker, that pair of diodes gets reversed polarity and from there on it gets forward biased, maybe triggering Q16 making it clamp somewhat the waveform ... or maybe it's an add-on to make the short circuit protection ( that's what Q16 is "officially" there for) follow the second breakdown curve of the output transistors somewhat closer, don't know.
If it's a waveform bending trick, it's useless because it's within the negative feedback loop, which will flatten whatever it does.

Again, I'm sorry this explanation is also wrong.
Here is how it works:
If you have a signal with positive amplitude (not negative) let's say x Volts, over certain amplitude diodes D34 and D35 (through R103) start conducting. The voltage drop on them is 0.6+5.1V. So x - 5.7V is passed to the base of Q16, which is a short circuit protection. Q16 also gets voltage from the R109 resistor through R108/R107 divider. It means that this voltage is compared with the voltage on R109 (so it is floating - that's why I wrote x Volts - it depends on the signal on the speaker).

Mark

[Steve Conner]

I simulated the output stage in SPICE software. And I got only 15 mV on the speaker

Well, this thread got me so curious that I was going to do that this afternoon. But you beat me to it Words of caution though:

1. Simulated op-amps don't have DC offsets. The op-amps are DC coupled to the circuit, so their DC offsets will affect (possibly dominate, even) the output offset.

2. Likewise, simulated transistors of a given part number all have the same beta, and all sit at the same temperature, so you can't easily simulate the other main cause of DC offset, which is transistor mismatch caused by parameter spread and thermal gradients. I don't think SPICE even allows you to run one transistor at a different temperature to the others.

3. Therefore, I expect a simulation would give a MUCH better DC offset than the real thing, unless you specifically took effects 1 and 2 into account, say by adding voltage sources of a few mV in strategic places to simulate the worst-case offset of a TL07x, or tweaking the transistor models so that the two devices in the front end have different betas.

I agree that the circuit is basically well designed. It makes sense that it's just a Bassman 200 with fewer output devices, and it seems better than those Ampeg MOSFET things that explode all the time. Maybe it was one of the designs that Fender inherited when they bought SWR?

As a high-powered bass amp, it should have a well-controlled DC offset, so I'm sure the 2 volts is not by design.

Here's a final thought. If one of the protection transistors blew short, then that would surely cause a large DC offset, because it would stop one half of the output stage from turning on.

[MarkusBass]

Well, this thread got me so curious that I was going to do that this afternoon. But you beat me to it Words of caution though:
1. Simulated op-amps don't have DC offsets. The op-amps are DC coupled to the circuit, so their DC offsets will affect (possibly dominate, even) the output offset.

Steve, thanks for remarks but as I said, I simulated only the output stage of this amp (starting from resistor string R91....and all transistors and diodes to the right on the schematic). So op amps were not included. I think that currently Lowell has also the output stage disconnected from the input stage (opamps and Q9 - Q13 transistors). So I'm simulating what he exactly has at the moment.

2. Likewise, simulated transistors of a given part number all have the same beta, and all sit at the same temperature, so you can't easily simulate the other main cause of DC offset, which is transistor mismatch caused by parameter spread and thermal gradients. I don't think SPICE even allows you to run one transistor at a different temperature to the others.

There were different beta for npn and pnp transistors used (and I used different transistors than on the schematic - I don't have models for all of them). Appart from this, we have here mostly emitter followers where beta does not count that much. Also I don't think that the temperature is important here - Lowell has "mayor" problems - were are not talking about few milivolts differences.

I agree that the circuit is basically well designed. It makes sense that it's just a Bassman 200 with fewer output devices,

I fully agree. It just seems too complex for 100W output power.

and it seems better than those Ampeg MOSFET things that explode all the time.

My favorite amp is Ampeg SVT3Pro. I agree that the MOSFETs sometimes explode but it does not happen that frequently (at least in the amps I get). And the main reason for the MOSFET failures is poor cooling and parts not correctly selected during production process. The differences in bias currents for those MOSFETs quite often differs by 200-1000%. And they all should be in a range of 20-30%. I assume that they are getting incorrectly selected MOSFETs from the manufacturer. Another problem is very poor cooling. If the fan fails, the amp will also fail.

As a high-powered bass amp, it should have a well-controlled DC offset, so I'm sure the 2 volts is not by design.

Thanks for supporting me - I also think so. It's clear that this is widely used power amp and such problems are simply not possible.

Mark

[Steve Conner]

I think that currently Lowell has also the output stage disconnected from the input stage (opamps and Q9 - Q13 transistors)

Testing the output stage on its own could well be the CAUSE of the measured DC offset. No part of any transistor power amp is expected to work on its own without a DC offset: the DC only comes under control when the feedback loop is closed. Usually, if the feedback loop is opened anywhere the amp will smash into one rail or the other: if he's only getting 2 volts he is lucky. I don't believe the 15mV simulation result.

God knows, I had enough trouble getting my new hybrid topology (which is basically an output stage on its own, driven by a Fender Champ OT) to work without a huge DC offset.

(But how does he know it "sounds great" if the output stage is disconnected? This thread makes less sense by the minute.)

Having said that, if I play along with you guys, assume that the simulation is right, and the 2v offset is in the output stage tested by itself, and caused by a genuine fault, I'd suspect one of the bootstrap capacitors C55, C56 of being leaky or shorted.

[MarkusBass]

Testing the output stage on its own could well be the CAUSE of the measured DC offset.

The offset was exactly the same when the input stage was connected to to output stage.

No part of any transistor power amp is expected to work on its own without a DC offset: the DC only comes under control when the feedback loop is closed. Usually, if the feedback loop is opened anywhere the amp will smash into one rail or the other: if he's only getting 2 volts he is lucky. I don't believe the 15mV simulation result.

This is very special circuit: the output stage is fully symmetrical and it has gain < 1. If the circuit is fully symmetrical, why would positive or negative side conduct stronger than the other (appart from few milivolts caused by differences in transistors parameters)?

(But how does he know it "sounds great" if the output stage is disconnected? This thread makes less sense by the minute.)

You haven't read all posts. The amp "sounded great" when everything was connected. Now the input stage (Q12, Q13) is disconnected, diodes D28, D29 are disconnected and C55,C56 capacitors are also disconnected (at least to my latest knowledge).

Mark

[lowell]

"(But how does he know it "sounds great" if the output stage is disconnected? This thread makes less sense by the minute.)"

Because I reassembled everything, and left the 33k across R91. This thread is so long it can be confusing to follow.

"Having said that, if I play along with you guys, assume that the simulation is right, and the 2v offset is in the output stage tested by itself, and caused by a genuine fault, I'd suspect one of the bootstrap capacitors C55, C56 of being leaky or shorted. "

Again, I already disconnected those as I came to the same hypothesis. They were not the issue.

Mark,
I reassembled the amp and returned it to its owner. It sounded fine and the offset was at a minimum. No I never found the REAL issue. But I had another one just come in so I hope to really figure this one out.

I learned a TON from this thread and this amp. I know how to see if a transistor is on/off. I know that 0v over a component means there is no current through it. I better understand SOA, bias transistors, emitter follower pa's and how they "follow" the voltage (thanks Enzo), muting circuitry etc...

[MarkusBass]

Mark,
I reassembled the amp and returned it to its owner. It sounded fine and the offset was at a minimum. No I never found the REAL issue. But I had another one just come in so I hope to really figure this one out.

It's a pitty because I think that you were very close. Definitely this 5.1V on D34 was not correct. Wasn't it soldered reversed? The diode without a signal shouldn't conduct. Have you left this 33k resistor parallel to R91?
Let's see what problem you and Enzo have with the other amps. And how quickly you are going to fix the amp.

Mark