Amp debug help needed

[Quatro]

A client says his amp is smoking. He brings it to me and i look under the hood for thermal distress. No burning, scorching, etc.

Test power tubes. They test fine. Test rectifier (5y3 ) tube. No heater action at all. Install known working rectifier and plug into light bulb limiter. Popping and crackling from spk. Arcing/sparks form inside bell housing of OT on the plate side. Thoughts?

Thanks

[thelonious]

I'm a relative n00b at this stuff - you might have done this already - but first thing I would do is take out the tubes and work over that OT from top to bottom with a DVOM. Maybe primary getting shorted to core? I think there's an article on Geofex with a bunch of OT tests like that.

[Quatro]

Thanks!

Without any tests, I'd be shocked if the OT was not toast. But how did it get toasted?

[tubeswell]

Possibly worn-out output tubes, or someone forgetting to hook the speaker up.

[PRR]

OT probably toast.

But why? No-load, or bad power tubes. ("Test fine" is a partial answer; I had a tube that tested fine under the low stress of a generic tube-tester but shorted-out when fully hot in the amp.)

[Quatro]

i did some brief testing. Of course all testing had a load. Also I tried i known working set of power tubes. No change with fresh tubes. Any more ideas? I will do a an OT test, but my hopes are not high.

Maybe the question at hand is: did the OT take out the rectifier or did the rectifier take out the OT? Or something else?

[Ed_Chambley]

Which came first, the chicken or the egg.  Do you have any bad filter caps or anything downstream from the rectifier that is toast?  The OT does not connect directly to the rectifier unless something shorts.  At least that is what I am thinking.  But if the OT shorts across from the primary the tubes will take the heat.  Least this is what I have experienced.

[tubeswell]

i did some brief testing. Of course all testing had a load. Also I tried i known working set of power tubes. No change with fresh tubes.

If the OT is already toasted, then a new set of tubes won't fix that. What we were saying was that what probably took the OT out was either bad tubes or no load (because you initially asked what could've caused it).

To test the OT, you need to see whether the windings are shorted (either pr to sec, or one of either to the core). If you can't tell with a R-meter, then you can test it by putting a low VAC onto the sec, and measuring the VAC on the Pr. This will tell you whether the Pr is giving the 'correct' 'Pr:Sec VAC ratio. If it doesn't, the OT is toast.

[stratele52]

OT probably toast.

But why? No-load, or bad power tubes. ("Test fine" is a partial answer; I had a tube that tested fine under the low stress of a generic tube-tester but shorted-out when fully hot in the amp.)

What appened if power tubes get shorted ? Lots of current in output transformer primary .....Too much current and it could burn.
Follow tubeswell method with small AC voltage you could easely test the transformer.

[DummyLoad]

Popping and crackling from spk. Arcing/sparks form inside bell housing of OT on the plate side.

no need to test. if you saw and heard OT arcing then it's toast.

does the amp have a B+ fuse? if yes, then why didn't it pop? if no, consider installing one.

not that it matters, you don't mention make/model...

respectfully. 

--DL

[Quatro]

I haven't been completely clear in some of my answers and questions. Let me try to fix it.

But why? No-load, or bad power tubes.

If the OT is already toasted, then a new set of tubes won't fix that. What we were saying was that what probably took the OT out was either bad tubes or no load (because you initially asked what could've caused it).

i understood what prr was saying. I was trying to say that in testing I have a load hooked up (didn't forget) and that I have tried another set of tubes. So I believe that eliminates tubes as the culprit. My guess is running the amp with no load caused the issue. Is this consistent with the dead rectifier heaters? They must have been carrying a lot of current as they are dead and there are bits of white ash type stuff clinking around in there

[tubeswell]

If the rectifier winding is dead, then I wouldn't trust the PT (either)

[PRR]

Possible script: client sets-up, forgets to connect speaker, hears nothing, turns up and beats the strings "to get something". Extreme signal with no load kicks-up HIGH voltage in the OT. Paper insulation breaks-down. Winding shorts to core which is grounded. Now the PT and rectifier feels a dead short to ground, something burns-up.

Enter Quatro.

You apply load, replace dead rectifier, get arcing inside OT. Unless you peek inside and find a cockroach carcass, NO actual damage, then the logical assumption is that the OT is arced to death. It "could" be re-wound, but custom labor and both-ways shipping of heavy iron usually makes a new mass-produced OT the better option.

With new OT installed, bring it up without power tubes, with a load, with a meter on the B+, and preferably a 45 Watt lamp-limiter. It should come up without drama, and with a somewhat-high B+. Put in one output tube, should be no drama, hair-high B+, and it should "play" though not in full glory. Try the other output tube. Try it all together and beat it for a while LOUD and soft.

See if it is possible to fail-safe the load connection. You don't mention make/model, but any cathode-biased tube amp can be shorted without problem. A shorting jack protects against empty amp-jack. Nothing really protects against the other end laying on the floor behind the speaker. You could add a permanent load resistor. Perhaps 5X the resistance of the tap it connects to at 1/5 the amp's rated power.

[Quatro]

Wow! Great post!

A shorting jack protects against empty amp-jack.

You could add a permanent load resistor. Perhaps 5X the resistance of the tap it connects to at 1/5 the amp's rated power.

Is there a way to do either of these with multiple taps, each with their own separate jack (no impedance switch or knob)?

[sluckey]

Is there a way to do either of these with multiple taps, each with there own separate jack (no impedance switch or knob)?

You can't use multiple shorting jacks, but you can use a single resistor connected to the highest Ω tap.