The mechanical expansion of a cold filament is the function that weakens
them. I suppose this means molecular level... Some examples:
As Jim Thompson said, thoriated tungsten cold R is much lower than hot.
So the inrush current, unless limited, will cause rapid movement of the
filament. Eimac and the others give recommended peak current limit in
the datasheets. Usually for big tubes it is no more than 2 x the nominal
operating current, sometimes less. A transformer without current
limiting may exceed this on turnon with a cold filament. Step start or
ramp up is a way to current limit open-loop and as the filament heats,
resistance stabilizes at the operating level. When a filament basket
expands rapidly, fractures can form along grains in the surface of the
carburized layer on thoriated tungsten. Repeating this action time after
time can eventually cause defects and other mechanical destruction on a
filament. This may reduce the active emitting area, just like ion
bombardment (poisoning) can do. This translates to low output, low gain,
low efficiency, symptoms of a weak tube due to low cathode emission.
Another worse problem, especially on big filament baskets of KW ratings
and up, is the longitudinal and transverse expansion different and some
tubes have spring mechanisms ("pantographs") to maintain tension in the
length direction (up and down). If heated too rapidly, the transverse
expansion may exceed tolerences and cause a grid to filament to touch,
shorting out momentarily. If heated slowly, the longitudinal expansion
is controlled, excess taken up by the mechanisms, and there is less risk
of touching.
Cathode type types (oxide coating) still have heaters albeit at lower
temperature. They can be damaged similarly by too rapid a turn on.
Emission of the oxide layer is facilitated by the indirect heating, so
naturally it takes time for this to start happening in a cold tube.
In all cases there are tubes which might survive many more hits of high
starting current than others. Manufacturers apply statistical analysis
to determine what is safe for a large family of tubes.
Once the price of a tube exceeds a few thousand dollars, it behooves the
user to follow the guidance and slow heat because the cost of failure is
high, and warranties will not cover a transmitter user whose circuitry
or operator disregards their recommended operating steps.
> Message: 1
> Date: Fri, 15 Jun 2012 03:59:12 +0000
> From: Charles Harpole<k4vud@hotmail.com>
> Subject: [Amps] filament instant heating query
> to check on a supposition............
> It appears possible that the sudden application of voltage to a tube filament
> causes degradation of the filament.
> What seems to happen? Is it possible that the very rapid heating of a
> filament could cause so much quick molecular action (rapid increased bouncing
> around of the molecules) that the shape of the filament could be changed? A
> deformed filament could then be too close to other components to cause an
> arc, perhaps?
> Or is the action on the atomic level...i.e., does heat change the actions of
> electrons in an atom?
> But, maybe the filament rapid heating works just fine, and all of the "slow
> start" "inrush protection" circuits are a waste?
> Another inquiring mind wants to know.
> 73,
> Charles Harpole
> k4vud@hotmail.com
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|