[old_tv_nut]

ElectronicM:
Interesting that you've seen glass roundies go gassy (as opposed to cathode wear-out). I think this may be the first mention i have heard of it.

I don't have a good feel for 21AX's (or any metal cone tube) higher failure rate being due to temperature cycling. Anybody here really know? The leaky 21AX I donated for helium testing at ETF turned out to have physical damage in the metal to metal faceplate seal, which Scotty (the rebuilder in Iowa) did not suspect. He was more suspicious of the metal to glass seals, which were fine. (The damage was a tiny ding, the size of a pen point.) Granted, the leak was obviously much worse than it could have been prior to rebuilding, but that was after the extreme temperature cycling of the rebuilding process, not likely due to daily cycling in use.