Power tube failure modes

[xtian]

Hoping you can confirm what I'm thinking...

Mesa Boogie Bass 400+ on the bench, with one DOZEN 6L6GCs in push-pull. On one socket, the grid and screen resistors are burned open. But when I pulled the tube and measured pin-to-pin resistances, there are no shorts.

Can I assume the failure happens only at high voltage, and that elements inside the envelope did not fuse together, so it reads normally at low voltage?

[JazzGuitarGimp]

Your assertion sounds plausible, but I am no expert. Have you looked at both sides of the socket as well as the base of the tube for carbon tracks? Also, if you have a tube tester, checking for shorts and gas might shed some light on what's going on.

[Stevem]

If that's a gassy tube then it may not take even a signal load being developed in the output stage for it to arc, but yet not burn open and truly physical short out sections of the tube.
Better for tubes, circuit parts and scokets to arc/ burn up then the big OT like that amp has to developed shorted windings!

[xtian]

Lou, I have a Knight tube tester that tests shorts, emission, and gas. This tube passes all tests.

[R.G.]

(1) has this happened once, or more than once?
(2) if more than once, is it always the same socket?

Remember:
One time is an accident.
Twice is a coincidence.
Three times is enemy action.

[xtian]

Thanks, RG! This is the first time I've seen the amp. The tube sockets are ceramic, easy to access, and are free from carbon, though there was smoke residue on the ceramic from the burning resistors, which cleaned up with alcohol.