audio-talk

I think its more likely the cross connecting of the feedback is the reason for the fault. If the resistors were metal film, I have seen them fail many times at far lower than their dissipation because of a surge of current. I wonder of you created a multivibrator type of circuit with the cross connections.

Yes possible.

So, I checked everything out and the casualties are the two large Arcol wire wound resistors and one half of the output transformer on one channel which has gone open circuit.

I'm pretty sure the blow up occurred when the feedback was connected the correct way round as the amp was working with no feedback squeal when it failed. Beyond a failed half of the output transformer primary there's no convincing explanation for the resistor blow up-or indeed the transformer failure

To repeat myself:

Mike H wrote: ↑Fri Dec 08, 2017 9:34 pm Sounds like a 300B went short-circuit and took out both the OPT and the rectifier resistors.

Must have been a biasing failure of one of the output valves, so it went full on resulting in a big current surge.

NB: according to its data sheet, a GZ37 can pass 1.5 Amps peak. And a 300B is not too far behind, IIRC.


-

Hi Mike,
Unfortunately there is no evidence of a 300 fault. There was no change in the music output before the bang, eg the usual crackle or pop you get with 300b failure, nor any sign of redplating or other sign of failure. All four valves still work perfectly. It seems to point more to an output transformer fault that caused excess current before the transformer failed open circuit than anything else
David

If the dropper resistors are as you say 820r, they would have been dissipating 8-9w so they would have been working hard. If they were metal cased types, were they fixed to the casework?

And further to that, if the TX did short between primary and case, then they would be asked to pass 350ma or so, which would be handled by the rectifier as it was up to temperature, but would have asked the resistor to dissipate 100W or so, so a failed TX could explain what you have seen.

Hi Nick,
The resistors were Arcol metal-cased wirewound attached to the chassis and rated at 50 watts. So the 100 watts current seems consistent with failure and the failed transformer.
David

Those metal cased resistors are de-rated a lot of they are not on an actual heat sink, so I would think just the 8W of normal operation would leave them running at a fair temperature.

David wrote: ↑Mon Dec 18, 2017 1:53 pm Unfortunately there is no evidence of a 300 fault. There was no change in the music output before the bang, eg the usual crackle or pop you get with 300b failure, nor any sign of redplating or other sign of failure. All four valves still work perfectly. It seems to point more to an output transformer fault that caused excess current before the transformer failed open circuit than anything else

Just FYI, when mine went it wasn't a 300B at fault either, a cathode bypass cap went short-circuit so valve ended up with zero grid bias. (The cap that went was a polystyrene film.)


-

That's a very good point! I'll check those Black Gates too.......