[ARC5] Power Tube Life Question

[Kenneth G. Gordon]

On 16 Mar 2010 at 17:43, J. Forster wrote:

> The amp uses 4 type 838 tubes, push-pull and parallel connected, with
> directly heated cathodes (flaments) presumably Thoriated Tungsten,
> although the RCA data sheet does not say specifically.

Actually, both of the RCA data sheets I have on this tube, 
and the Amperex version, all state quite clearly that the 
filaments are indeed  thoriated-tungsten.
 
> Since the tubes are somewhat precious, I'd like to run them at reduced
> filament voltage (maybe 5-10% less) to extend their life. However, I'm
> concerned about loss of Gm and/or emission from prolonged operation
> with the reduced voltage.

According to what I have seen, Gm and/or emission are not 
effected at all. Furthermore, it appears that when the tube 
is severely "starved" of filament voltage, the distortion in an 
audio amp drops drastically. In one case, from about 10% 
to 0.03%.

Furthermore, according to Eimac, each 3% decrease in 
filament voltage from the "normal" voltage in a thoriated-
tungsten filamented tube results in a doubling of tube life. 
Conversely, each increase of 3% results in cutting the tube 
life by 50%, or 1/2.

I.e., if you reduce the filament voltage by 3% you double 
the tube life: if you reduce it an additional 3%, you double it 
again, etc.

You may wish to get and read Eimac's latest version of 
their "Care and Feeding of Power Grid Tubes". It is 
available on the web in a 6-part PDF and it explains this 
and much more of interest.

Just do a search of "Eimac Care and Feeding of Power 
Grid Tubes" and it will show up in multiple places.

Eimac suggests that in the interests of conserving TT tube 
life, the operator should reduce the filament voltage to a 
point at which the plate current just begins to drop. Then 
the tube(s) should be run in that condition for at least 24 
hours to determine if the tube(s) are stable at that voltage. 
If not, raise it a tad, and repeat the test until you arrive at a 
filament voltage that maintains needed emission 
(determined by the constant plate current).

Now keep in mind that these methods ONLY work for NEW 
tubes, or ones which are known to be good. Weak tubes 
will show plate current reduction under practically any level 
of reduction of filament voltage.

Also, it appears that you have to DRASTICALLY reduce the 
filament voltage in order to make them suffer from "cathode 
poisoning". In your case, I think that would never, ever, be 
a problem and I would not worry about it in the least.

Lastly, I think you should know that there are at least three 
methods of "reactivating" flat thoriated-tungsten filamented 
tubes. I have personally used all three methods and can 
attest to their working extremely well in certain cases, and 
not at all in others, the results depending on several 
factors, the most important of which is the ruggedness of 
the filament.

I am writing an article for Electric Radio Magazine on this 
subject and it will be published some time this year.

Ken Gordon W7EKB