Shunt Regulator - poof

[Metropolis488]

Hi,

Installed the shunt regulator into my happily operating Stereomour II last night.

On power-up, I heard a little but of a sizzle sound, followed by a smell that reminded me of burnt plastic.  Thought, to myself "crap".

Since there was no fire or smoke (that I could see), I decided to take voltage measurements - all were good (slightly high - I measured my mains at 126 volts, so expected).  IA and IB were 411v.  So, I thought, maybe there was just something like a dog hair on the board when I fired it up... ergo, I hooked up speakers and gave it a listen.

Couldn't really evaluate as it was late and I had to keep the volume super low, but the smell persisted... and that made me uncomfortable.  So, I removed the shunt regulator and returned the amp to stock configuration and it works fine.

There is still a faint burnt smell on the removed shunt board, but everything looks ok. 

Not sure what to do, but I certainly don't feel comfortable (safe) putting it back in.  Is there any way to trouble shoot this?

Thanks!
 

[Paul Joppa]

You did not mention the rest of the voltages, which will give a good indication of what's going on. But I understand your caution (and, as an engineer, I approve of it!). So here are two initial things to try, and a next step too:

1) measure the resistance of each individual resistor on the SR boards.

2) Inspect the board - both sides - for solder flux contamination, quality solder joints, and any signs of heat damage. You may want to use a magnifying glass - think Sherlock Holmes!

If that does not produce a solution, then post photos of each board - both sides, again.

I'll be out of easy reach for a week or two, so I might miss the resolution - good wishes! The team is nearly always successful.

[Metropolis488]

Alright - readings across all resistors looks good.  However, I get no readings across any of the legs of the 431 regulator on the B side or the PN2907 on the A side.

[Paul Birkeland]

Open readings on solid state parts are a good sign.  When they fail, two legs will usually short together.  Sometimes you'll get a nice pop when that happens and little chunks will go flying.

[Metropolis488]

Ok, then should I NOT be getting readings on the other redulator and transistor?

[Paul Birkeland]

It depends a little bit on what the reading is.  If you get a really low resistance, that's not a good thing.  If you get hundreds of thousands to millions of ohms, then not so much of an issue. 

[Metropolis488]

I’ll double check but pretty sure they were in the Megs.
So, I’ve double checked all my joints, tested all the resistors and caps and, bas d upon what you are telling me, the regulators and transistors should be fine (although I still detect a smoky, burnt smell from them).  Where does this leave me?  My gut is telling me I probably need to replace the regulators and transistors, but don’t know for sure.

[Paul Birkeland]

Fire it up and see what happens.  The fuse is there to protect you (though usually if there's something that wrong a transistor or LED will give up long before the fuse does).

[Metropolis488]

Well, my concern is that nothing has changed, ergo why expect a different outcome? Something is definitely wrong (there shouldn’t be a smokey smell) and I don’t want to mess up my perfectly working amp by reinstalling this. Are there any other checks I can perform?

[Metropolis488]

I still don't have a good path forward for this - do you guys sell individual transistors and regulators?  I think they are the problem, and don't want to have to purchase another shunt regulator kit.

[Paul Birkeland]

Have you actually tried using these yet?

[Metropolis488]

Yes, that’s what I outlined in my first post in this thread

[Paul Birkeland]

Without finding any shorts on any of the solid state components, it's entirely possible that the board will work when power is applied.  The poof/smell could've been a component lead temporarily shorting something out. 

Without actually knowing what problematic voltages you may or may not have, and knowing that there are no shorts on any of your transistors, you don't have enough information to even know if there's an issue to resolve.

[Doc B.]

Peebs beat me to the punch - You could just replace the parts and see if it solves the problem. But it's difficult for us to say with the info we have whether that is the correct solution, and reworking boards always poses a slight risk of damage to the traces. If you could post some good close up photos of the board top and bottom we might be able to spot something. If your voltages were good it's not too likely that anything off the board would be damaged even if something on the board failed. And not too likely you can get much further in diagnosis without trying it again and rechecking voltages now after you have smelled the smoke. The reality of bench repair is that once it has been established that everything is connected properly a tech sometimes has to just plug the thing in and see if anything goes poof to figure out what's going on.