UL84 grid current cause

[PJL]

Working on a previously fried Ekco, I found the UL84 grid voltage creeping up steadily as the set wormed up. The Mullard tester did not show a fault but the Ekco has a 1M grid resistor so 1uA already causes a problem.

The original EL84 looks visibly perfect with no signs of darkening and a good getter. I tested 2 other well used UL84's, one is a little weak but does not show the fault in the set, the other stronger one shows the same rising grid voltage.

I have had problems before but that was a PCB set on which I performed major surgery to lift the grid connection from the PCB and change resistor values to reduce the voltage and current.

There seem to be several possibilities:
1. The valve socket is leaking once it gets warm
2. The valves have internal leakage (common in UL41's) but one looks near perfect
3. The valves have lost some vacuum
4. The valves are suffering secondary emissions due to previous thermal runaway

All of these are heat related and are difficult to reproduce outside of the set and as the socket is riveted I am not keen on replacing it just to test a theory.

It would be useful to have a definitive means of determining grid leak cause as this is a common problem that destroys radios.

[lesmw0sec]

Presumably you tested or replaced the grid coupling cap?

[murphyv310]

Hi
Never had a leaky UL84 but that's not to say it's impossible. I have had leaky bases and a lot of them but I've literally done hundreds of vintage repairs. I've had leaky PCBs too. Why a low valve doesn't show the problem could be it's running cooler and the base is not beginning to leak but that's purely a guess.
I'd change the base and eliminate that in the first instance, I'm sure I mentioned that previously or perhaps on another person's thread.