[Amps] General question on filament life

[Larry]

Maybe I've got my list "bounce" problem fixed.....


Around 1970 I put together an amp with a pair of Penta Labs used 
4-400A's. Around 1988 or so one of the tubes failed due to a filament to 
cathode short. Both tubes were replaced with used Eimac tubes. Still 
using those two tubes. Around 1988 I built up a pair of used Eimac 
4-400A's in a 40 meter monobander. Around 1999 one of those tubes failed 
due to filament to cathode short. It was replaced with another used 
Eimac tube. Those tubes were in use until a few months ago when that amp 
deck was removed from service.

Both these amps are/were turned on to make and QSO and turned off when 
finished. Some days they may never get turned on, but some days they 
each may see 10 to 12 power cycles. After two failures in 58 tube/years 
I decided to do something about that high failure rate and installed 
step start circuit for the filaments. No failures since then and I still 
turn the things on and off all the time. I have had similar results with 
the 813 amps I use. Always cycle the power, never have had an 813 fail.

Seems to me there are tubes that can handle this sort of treatment IN 
AMATEUR SERVICE and tubes that have a problem. Maybe it's a better idea 
to use a tube that matches your lifestyle rather than alter your 
lifestyle to fit a tube that is to delicate to survive without "special 
care".

For an extreme case take a look at the RCA 6949. The MAXIMUM ALLOWABLE 
inrush surge current is spec'ed at only 165% of the nominal operating 
current!

You guys that like big jugs (toobes, not boobs) and water cooling might 
like this 6949. It requires a minimum of 80 GALLONS PER MINUTE of 
DISTILLED water through the anode cooler, 7 GPM  through the beam 
forming cylinder, and 1 GPM through the socket grid connection!

Back to filament life, when I was still working there was a RF heating 
unit buried deep in a classified area. I only knew about it because a 
pair of 3CX1000's would show up on my buddies desk about once a year. He 
said they would go bad because the filaments were left running all the 
time and anode power was only applied when it was needed. I got a pair 
and checked them and sure enough, NO EMISSION! Seems I've seen this 
written up by Eimac or some other manufacturer. Running thoriated 
tungsten filaments without anode current shortens the life time faster 
than on-off cycling. Any comments on this failure mode?

73,

Larry - W7IUV
DN07dg
http://w7iuv.com