Audio amp shorted. Which component failed?

[king.oslo]

Hello there,

The diagrams below are taken from this schematic: Full schematic. Today, I was adjusting the output amplifiers. Unfortunately the test probe slipped off a test pin and shorted a few of the test pins highlighted in this diagram:



I am pretty sure it was pin 2 and 3 which were shorted for a moment  , (clearly connecting the -45V rail to ground). Immediately after, I noticed that other power rails were low and two massive input resistors in the fuse board (R601 and R602) were starting to smoke so I immediately pulled the power cord. Heres the board with the resistors:


This is what I've found so far:
> I made a few other measurements, and I've noticed that right now, got 0.5 ohm between pin 2 and 3 marked in the top-most diagram.
> I didn't blow the two fuses on the fuse board, they measure .2 ohm and the smoking resistors seems to have the correct value. Since the resistors are smoking, the -45V must be lower than -24V, which is what the relay RY601 bypassing the resistors on the fuse board is rated for. So I assume something is shorted.

Using this information, are you able to suggest what needs replacement or which other measurements it is wise to make?

I have some extra material, here's the full schematic and this is the service manual for it. Hoping to get some help 

Thank you for your time.

Kind regards,
Marius

EDIT: I measure .2 ohms between the collector and the emitter of Q354, could it be shot? How would we explain that happening?

[Kevman]

If I'm reading the schematic right (probably not), it looks like Pin 2 is connected to the speaker output on the channel and pin 3 is the -45v main power rail. Ouch. Am I right?

Were there speakers connected at the time? I THINK there's a DC detector driving Relay 301 ("muting"), which would have triggered. It should have made a loud pop followed by Relay 301 opening in an attempt to keep your speakers from going up in smoke. Which they would have if it didn't open. Check your speakers...

Relay RY601 appears to be a soft-start and sure looks like it runs off of the -45v rail for some reason. When you shorted it, the rail may have sagged enough for it to open, causing the full amplifier power to go through the two resistors that got hot on you.

You might have taken out the negative power transistors Q354 or Q363 causing one to fail short and that's why you read .5 ohm? I'd check them and replace the soft start resistors.


Shorting pins 1 and 3 would not have likely caused any sort of significant damage as it would just bypass the transistor balancing resistor R391, which I think the transistor would have survived for a while, so that's not what happened.

[king.oslo]

Thanks for your reply,

Luckily, I had disconnected the speakers.  Regarding Q354 or Q363, how come you think that they have failed? According to the manual, the 2SC1986 was chosen. I thought about it, but perhaps I've mistaken: when i shorted pin 2 and 3, that means i effectively lowered the potential across the transistors? Is that true? And that isn't harmful is it? Is there something I've misunderstood?

It be great to have some idea what is shorted before I start stabbing the pcb with the soldering iron!

Thanks

[MagicSmoker]

I would check Q353 and Q362 because shorting TP2 to TP1 or TP3 would result in -45V to suddenly appear at their collectors.

[king.oslo]

I would check Q353 and Q362 because shorting TP2 to TP1 or TP3 would result in -45V to suddenly appear at their collectors.

Thank you for your suggestion, perhaps I will do that. But I'm measuring .5 ohms between pin 2 and 3. Suppose Q353 or Q362 failed, do you suggest that this contributed to the short between pin 2 to pin 3? Thanks

I measure .2 ohms between test collector and emitter of Q364. Could we explain that the accident may have caused it to fail?

EDIT: I got rid of Q364 because it was shot. But I still have a short between my -45V and 45V

[deadlylover]

I think all 4 of the output transistors are suspect, they have a Vceo of -+80V and they've possibly been exposed to -+90V. We know the Q353 and Q362 has a good chance of failing from directly putting -45V at their collectors, but after you took the probes away, if they've failed short then +45V may appear at the collectors of Q354 Q363, exceeding their Vceo as well.

You can try buzzing out the diode junctions of the trannies to see if it would give you any more info, but that's a bit iffy to do in-circuit, can't rely on it that much.

[king.oslo]

Thank you deadlylover

Turns out you and kevman were correct, Q354 was shot, the other shorts I had were transistors which had been burnt out while I was testing them disconnected from their heat sink

I wasnt able to find the original parts, 2SA771 or 2SC1986 at farnell or digikey. Do you reckon I MJE15032 and MJE15033 would be suitable substitutes, or are you able to recommend others which are better for my purpose?



Thanks,

Marius

[deadlylover]

I'd imagine it'll be fine. 

Hope it all goes well, that's a lovely looking amp you have there.

[MagicSmoker]

I would check Q353 and Q362 because shorting TP2 to TP1 or TP3 would result in -45V to suddenly appear at their collectors.

Thank you for your suggestion, perhaps I will do that. But I'm measuring .5 ohms between pin 2 and 3. Suppose Q353 or Q362 failed, do you suggest that this contributed to the short between pin 2 to pin 3? Thanks

I measure .2 ohms between test collector and emitter of Q364. Could we explain that the accident may have caused it to fail?

I suppose I should have wrote, "I would also check..." to make it more clear that all of the output transistors should be suspected in such a situation.

As for how everything failed, a reasonable chain of events would be that shorting TP2 to TP1/3 effectively bypassed the negative rail output Q's, putting -45V at the collectors of the positive rail output Q's, which caused them to fail short. When you removed the short from across the test points that put +45V at the collectors of the negative rail output Q's, causing them to fail short as well.

Which is why it is a good idea to insulate the probe tips with heatshrink tubing or electrical tape, like so:

[king.oslo]

I'd imagine it'll be fine. 

Hope it all goes well, that's a lovely looking amp you have there.

Me too, and thank you for your compliment.

I suppose I should have wrote, "I would also check..." to make it more clear that all of the output transistors should be suspected in such a situation.

As for how everything failed, a reasonable chain of events would be that shorting TP2 to TP1/3 effectively bypassed the negative rail output Q's, putting -45V at the collectors of the positive rail output Q's, which caused them to fail short. When you removed the short from across the test points that put +45V at the collectors of the negative rail output Q's, causing them to fail short as well.

Which is why it is a good idea to insulate the probe tips with heatshrink tubing or electrical tape, like so:

Thanks for your suggested chain of events, sounds very reasonable I think. The probe insulation tip is great, thank you

[king.oslo]

Installed new transistors in both channels, but there seems to be at least one more problem: When I set the DC bias current to adjust to 16.5mV across test pin 1 and 3, as I increase past 10mV, all of a sudden the bias voltage shoots up to 250mV (~ 750mA) (see the photograph)! I can turn the potentiometer further, and the bias current will increase further to about 300 mV ( ~ 1A). If I try to turn the pot back down, it rapidly decreases 0mV (0mA).

This happens in the channel where I didn't get the short a few days ago. This time I dont really have a clue what is wrong, so fixing it depends on your suggestions. What are your thoughts?





Thank you for your time. Kind regards,
Marius

[dom0]

Could be as simple as a bad pot.

[king.oslo]

Could be as simple as a bad pot.

Have I understood this correctly; if i set the RT352 dc bias pot to about the middle position, its safe to drive the IC351 pin 2 with a low impedance 0 to -8.5V power supply overriding the pot setting?

And second, is it correct that IC351 does not have a feedback, so that the chip is most likely faulty if all the voltages at all it's pins are nominal?

Thanks.

[dom0]

I dunno, just because it goes to a wiper of a pot doesn't necessarily mean that it is a high impedance input that can be driven from a voltage source. We don't have a schematic of that part. You can try, but set the current limit really low.

There should be some kind of feedback, either current or thermal (although almost all amplifiers use largely inferior and often borderline unstable thermal feedback), to stabilize the AP of the output stage. Is Q357 on the heat sink?

[king.oslo]

Hello again,

No Q357 is not on the heatsink.

Say I adjust the dc bias current to 30mA. If I probe any of the output transistors pins with a multimeter probe, sometimes that is sufficient to trigger a dc bias current of 0.75A! In fact I can toggle the dc bias current between 30mA and 0.75A by tapping the transistor leads with a multimeter probe. Tapping the leads of any of the other transistors or diodes or IC301 is also sufficient to set it off. Any ideas?

Thanks.M