How about AM with a 3cx1500 ?

[John K5PRO]

The 8877/3CX1500A does not use BeO on the ceramic. It is Al2O3, alumina. As a matter of fact, Eimac used BeO for the conduction cooled bottles but most other are alumina insulators.

On all of the high power cavity amplifiers that I have designed for broadcast and scientific for 30 years, I have used Dwyer diaphram type pressure switches measuring with hose connected to a nipple at the enclosure. This is measured there the air hose from blowers connects, so that if it falls off, pressure drops and turns off RF, HV and filament (and screen). The only time it can be faked is if the outlet about tube was blocked and pressure would actually rise. I don't use temperature monitors on the tube, since, with correct air, the tube will be under 200 deg C at all points. I measure the air pressure vs flow from the blower, compared with the tube datasheet, and adjust for elevation above sea level. Then, run the amplifier at high continuous power while Tempilaq indicator paint has been applied in small dots around the ceramic seals. Open it up and remove the tube and observe. With these precautions during design, haven't lost a tube due to cooling.

With the big tubes, UPS becomes impractical. So we just live with the fact that it will loose air and water at least several times a summer from storms. I have 20 kW filament power supplies for each tube. 19VDC at 950 amps.

[Opcom]

John you always get to play with the big stuff! Your lab is in the high desert, isn't it?

[flintstone mop]

Just as an aside thought about high power operations. What if.....the transmitter is controlled by a computer. Filament and high voltage. And the dang computer shuts the filaments off before the HV? The malfunction was either operator error or the computer was not set-up/sequenced properly?
Fred

[K1JJ]

Just as an aside thought about high power operations. What if.....the transmitter is controlled by a computer. Filament and high voltage. And the dang computer shuts the filaments off before the HV? The malfunction was either operator error or the computer was not set-up/sequenced properly?
Fred

Not sure I understand...how would that hurt the tube?  If the filament is off, basically the tube is dead.  There should be little to no current flow in any tube element nor should there be any heat generated.  ??

T

[W3RSW]

Indirectly heated cathode in an oxide emission tube called on to draw beaucoup electrons when originally red hot but with ever declining heat from turned off filament can't be too good.

This includes your beloved 3cx350's too. 


Wonder if Eimac or others other tested real tubes, both thoriated tungsten and the oxide cathode types with this type of fault.?

[K1JJ]

Oh, I see.....  we are still trying to pull power right after the fil power drops out on an indirectly heated cathode.  

That would require yet another fail-safe safety feature - one that Tom Vu does not have and never thought about... [sigh]

I wonder if one or two instances is enuff to hurt a tube...

T

[W3RSW]

The cathode has to be hot enough to emit properly to prevent back bombardment. Might get by once or so, but...

I'll have to think about it.

Unfortunately most of this type are the more expensive variety of bottle.
Hey ThomasATron,  try it with your stash of the older and cheaper series. You like to play with such.

[AB2EZ]

If the filament is hot enough for electrons to escape from the filament's surface, and if the indirectly heated cathode is not hot enough for electrons to escape from its surface... and if the screen and/or plate voltages are on... then electrons emitted from the surface of the filament will be accelerated toward the cathode (by the electric field between the filament and the screen and/or plate). They would bombard the cathode, and damage it.

Once the cathode is hot enough to emit electrons, then a cloud of electrons will form between the cathode and the filament... that shields the filament from fields that would accelerate electrons emitted from the filament toward the cathode. In fact, this cloud of negatively charged electrons between the cathode and the filament would repel electrons emitted from the surface of the filament... and cause them to drift back to the filament.

This is, I guess, why you are supposed to wait a few minutes, after turning on the filament supply... before applying the screen and/or plate voltages... when using a tube with an indirectly heated cathode.

In the scenario outlined by Fred, I would imagine that the filament would cool down much more quickly than the cathode... Thus avoiding the bombardment of the cathode by electrons emitted from the filament.

Stu