Rich Measures wrote:
>If you lived a bit closer, Ian, I have three, intact, filament/grid
>shorted 3-500Zs that we could disassemble and inspect. My best guess is
>that we would find bent filament helices in three of them.
To further things come to mind.
One was the message from Reid Brandon at Eimac, posted a few weeks ago
by W0UN. Here's the currently relevant part:
>Tungsten wire undergoes a physical change when raised to its
>operating temperature and returned back to room temperature.
>In tubes with thoriated tungsten filaments, the end result is that
>over many on/off cycles the shape of the filament begins to become
>distorted, ever so slightly but depending on how many cycles, the
>spacing between the filament and grid (which must be very close to
>achieve good electrical characteristics) may decrease, and eventually
>a short may occur.... Reducing the filament inrush current
>is no solution, tungsten must still go through a large excursion in
>temperature.
That's the first point. The second one is mentioned in RB's message and
then taken up by someone else:
>Reducing filament inrush current is important for the reduction in
>the magnetic field between filament wires, and allows the spring
>tension that exists across the filament structure to gently keep
>the filament aligned (special springs are used for this purpose
>in many transmitting tubes.)
Someone else recalled in an e-mail (posted by permission) that:
>EIMAC once put a 3-500Z base structure in a bell jar, evacuated it
>and observed what happened to the filament during inrush conditions.
>Most of the time, when operated at rated voltage using a small
>filament transformer (suitable for a single tube) nothing happened.
>However they did notice once or twice the filament "twitched" a bit.
>Nothing serious, However when a large capacity transformer was used,
>if they happened to turn it on at the peak of the cycle, the
>filament was observed to physically move a small amount.
This correspondent guessed that if the tube originally had a bowed
filament (this sometimes happens due to poor control of filament tension
or whatever) the tube would be susceptible to filament movement that
might short it to the grid during turn-on, especially if filament
voltage was high.
Me again: If we combine those two observations, there seems a distinct
possibility that the filament could gradually migrate towards the grid
over the course of time (especially if the filament tension wasn't quite
correct, which wouldn't show up in the normal factory acceptance test).
We already know that 3-500Z filaments can't tolerate much sideways
force, which is why they're only allowed to be operated vertically.
When the filament eventually moved so far that it touched the grid, it
would connect the grid to a voltage tap somewhere along the heater
circuit, which could indeed produce the burnt-out grid choke and the
rest of the fireworks.
That seems consistent with Dr Rich's most recent set of patients, one of
which had the short, but the other was on its way but not quite there.
Still trying to fit *all* the facts together: this theory would imply
that the grid choke in the second tube should *not* be burnt out.
BTW, the second correspondent was not Buzz Miklos ;-)
73 from Ian G3SEK Editor, 'The VHF/UHF DX Book'
'In Practice' columnist for RadCom (RSGB)
www.ifwtech.demon.co.uk/g3sek
--
FAQ on WWW: http://www.contesting.com/ampfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|