Amp cuts out at loud volume

[mattmacdonald]

Hi gang. I've been lurking here for a while, and now I have a problem that I need a little help with.
I just build a clone of a BF Princeton, but without the tremolo and reverb. I've used the old power/output transformer and Mallory Cap cans from an old Lifo Cobra amp. It seems to work alright at low/moderate volume, but when I crank the thing up to about 7 and hit a loud chord the amp goes dead (no sound or sometimes quiet farty sound) for about 5-10 seconds. I've checked the circuit and everything seems to be wired up ok, and I've done some tube swapping to rule out bad tubes.....any ideas?

[Tubenit]

oscillations

Sounds like parasitic oscillation which can cut the sound completely off. Are the tubes redplating by any chance?

Have you tried other tubes?

This could be a lead dressing or layout problem also.

With respect, Tubenit

[pdf64]

Using old electrolytic caps is a bad idea, they are a limited life consumable which will wear out. They might be ok for awhile, on their last legs, or be the cause of your problem. Need to replace them with new known good caps to rule them out as the cause.
Or it might be worst ever case of blocking distortion, try a search on this. 10s might be a bit much though unless you've (accidentally) used high value coupling caps.
Or failing insulation on the OT primary.
Have you tried it with different speakers? I've had a couple which would occasionally go open circuit on big signals but then recover themsevles. I traced the problem to a break in the voice coil leads where they run up the cone a little under the black cement, before they join to the flexible tinsel leads to the terminals.
Oscillation which cut out a big audio signal would be unusual, but never say never.
Once you've got good caps and confirmed a different speaker exhibits same fault, it would be useful to monitor and report the current through the OT (current meter in series with OT centre tap) during the various conditions (no signal, small signal, big signal, cut out). Be careful, isolate from wall socket and drain caps before sticking fingers in. Careful not to take voltage readings with your meter still in current mode. Peter.

[mattmacdonald]

It turns out I had a pull-down resistor on the wrong side of the .022 coupling cap between the tone stack recovery and phase inverter. Problem solved!

[pila]

Hate it when that happens!