When it comes to seals, as noted, glass to metal is an art. At best, the
glass is not a perfect bond to the metal. We are talking clearances
that are so small, a molecule of gas may take months or years to work
it's way from one end of the seal to the other and those molecules are
always moving.
OTOH Ceramic to metal can easily be automated using a precision pulse to
heat the surfaces. IIRC the two surfaces are coated with a mix of flux
ans silver solder. The surfaces are placed into intimate contact with
each other and a pulse of heat melts the solder into a molecular contact
with each surface and bonds them together producing a gas tight seal.
This many forms of this process are readily adapted to the batch
process. Problems often come from starting up the production line and
getting all the settings correct.
Higher prices are due in part to the higher precision and placement of
parts. as well as the tiny and fragile wires used in the grids and
screens. I wonder why we don't see Pentodes in production any more.
IIRC they are just a Tetrode with a grounded screen added to eliminate
secondary emission.
73
Roger (K8RI)
On 1/26/2015 12:21 AM, John Lyles wrote:
Ceramic/metal construction is usually very vacuum tight, with
exceptions due to mfr mistakes. Somewhere I have a paper from
Eitel-McCoulough extolling the virtues of their new "CX" tubes back then.
Glass to metal seals are the earliest type, learning first from the
Edison bulb ideas to Mr. Deforest, with improvements by Mr.
Housekeeper at Westinghouse and a long line of GE improvements.
Sealing metal to molten glass edge is an art, and requires good glass
blowers/turner.
I don't believe that vacuum seal leaks are due to glass 'sagging' over
time, but just due to longevity of those seals themselves. Corrosion
of the metal is one factor. Temperature extremes between the metal and
glass is another (from overloading).
The major tube manufacturers like Eimac/CPI, RCA/Photonis/Burle and
Thales/Thomson have dropped all glass tubes from production in the
past decade for reasons that it was getting very difficult and costly
to by the glass blank cylinders from companies like Shott Glass of
Germany. I know one scientific user in UK that uses a big glass
pulsed-megawatt triode, they had to by a 20 year supply of the glass
for a major tube manufacturer to guarantee a supply of the tubes.
Where I work, we needed a source for 120 KV switch tubes, small planar
triodes not much large than ham sized. We requested glass, like Eimac
used to make in Salt Lake City, for reasons of improved HV standoff
(when in oil), not with RF of course. They did make them for us in
Palo Alto, once we had need for more than 50 of them.
For ceramic/metal, there has been many decades for improvement. The
biggest US-produced VHF gg triode still in production is the the RCA
7835, which has pressed seals of metal to ceramic. These tubes come
with 2 l/sec vacion appendage pumps connected to them. No
metalization, just tremendous mechanical force keeps those tubes
tight. Eimac makes the 4CM2500KG tetrode, their largest for RF, with
metallized ceramic seals. Thales does same for their big tetrodes for
500 kW and up, like TH558, TH576, and Th628. Metallized ceramic, when
done right, is very good for vacuum tightness. A mechanism allows for
flexure of the two components - which are closely matched thermally,
where RF current is carried through good conductors inside. The
metallization allows for the seal to be brazed tight.
BTW, its amazing to see those big tubes being lit during their
production process, where they are run very hot. Thoriated Tungsten
runs about 1900 Kelvin. I saw a 7835 in Pennsylvania at the factory
where the ceramic was bright pink, almost white. It looked sort of
transparent. The tube has a 30 kW filament,no kidding, 5 Volts DC at
6800 Amps.
I love ceramic metal tubes. As others have said, they aren't as
beautiful as glass, but a lot more rugged.
73
John
K5PRO
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
---
This email has been checked for viruses by Avast antivirus software.
http://www.avast.com
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|