[Amps] Thoriated tungsten lifetime when idling tspa

[John T. M. Lyles]

Tom, You are correct, for Eimac its normally stated as limit the 
current to 2 x the operating current on turnon. I was guessing 
conservatively. I checked and the same thing applies with the larger 
tubes by CPI/Eimac today. For RCA/Burle tubes, it is stated in the 
preambles to refer to the individual datasheets, and on them, i see 
nothing about maximum filament current on startup, only 
recommendations to ramp up slowly.  For E2V (Marconi, EEV) the number 
is all over the map, from 1. 5 x to 6 x, depending on the tube.

Another factor in very large tubes, with close spaced control grids 
made of pyrolytic graphite, is to ramp up slowly and limit inrush due 
to the changes for the filament basket to expand in width, before it 
expands in Z direction (length), which is under tension with springs. 
In a short time (< minute) the expansion of filament is taken up in 
the Z direction by the tensioner. Before this happens, though, there 
is some chance for a G-K short, esp if the grid bias is already on. 
The pyrolytic graphite grid is very stable and will remain 
dimensionally in shape as the nearby filament is changing as it warms 
up.
I was told this by a tube engineer at Thales in France, explaining 
their ramp up requirements. For a 250 kW tetrode of theirs, I have a 
ramp that takes about 8 minutes until its ready to run. Very slow.

73
John
K5PRO

>>Eimac typically says 150% of normal current, I believe, for TT
>>filaments
>
>They say twice for most tubes we deal with. The exact text says 
>limiting current to two times the operating value helps reduce 
>damage (movement) from repeated filament thermal cycles.
>
>>I have never heard that story that running filaments alone without HV
>>will shorten emission life.
>
>I've never seen a warning like that either.
>
>As a side note, when I was working closely with Eimac Salt Lake in 
>the 80's, I specifically asked the engineers about problems like 
>inrush.
>
>Buzz Miklos, WA4GPM, was head engineer for the power grid group at 
>Salt Lake. Buzz said they clamped a 3-500Z directly to the terminals 
>of a 200 amp transformer and despite hundreds of cycles they could 
>never measure any damage.
>
>The thing  everyone has to remember is reducing long term inrush 
>current stops uneven heating where one area of the filament heats 
>much hotter than normal and then settles back to a normal 
>temperature. The hot vs. cold stresses are still there and are a 
>matter of how often the temperature cycles between extremes.
>
>There are TWO ways to extend the life.
>
>1.) You don't turn the thing off and on needlessly.
>2.) You do a filament circuit that limits extended warm-up current 
>to twice the rated current or less.
>
>Neither one excuses or eliminates the other problem, and the 
>guideline is two times for most Eimac power grid tubes likely to be 
>found in a Ham shack...even 3 by 3's.
>
>73 Tom