To no one in particular:
I have little to add about ham/contesting amplifiers so sorry to disrupt
with this posting. I wanted to do an introduction, and then vanish into the
readership. (we'll see). My background is in the design of RF PA's, both SS
and thermionic. In early 1980s was transmitter designer for Broadcast
Electronics in Quincy, IL. Designed transmitters or portions (and cavity
amplifier PAs) for FM (100 MHz) using 8877, 4CX3500A, 4CX7500A, 4CX20,000A
(8990) - most of the hot new Eimac tubes of that decade. The 3500A was a
fantastic improvement over the 4CX5000A at VHF-the standard that other
tetrode rigs used. It was undoubtedly due to the shorter socket/stem and
the shorter ceramic I believe. I tried the 4CX3500A in place of 4CX5000A in
an identical amplifier, except for the socket spacing, filament xfmr, etc.
Gain, efficiency all better in that particular design. We got high 70
percent for class C operation. The SK350/370 socket was instrumental to
getting the tube stable and screen-inductance neutralized. It has very good
shielding between grid/cathode side and the screen. It used Kapton
polyimide film bypasses in the filament. We punched holes in the solid
socket, to let the cooling air through, without sacrificing the isolation
very much. This is now the SK350. I believe the SK370 has solid sides. The
SK300A was the HF socket, which you may be familiar with. Sorry to ramble
on sockets...
>From 1984 through 1991 I was at E.I. DuPont Co in Delaware. We had lots of
RF oscillators based on 3CW30,000H3 triodes, at 27-35 MHz. No modulation,
just raw RF voltage for dielectric heat processing.
The rigs were huge, and we did cool stuff to polymers and water. I picked
up a nice little LaRose 90 Mhz RF heater at a junk yard recently, uses an
Eimac 3CX2500 or 3000H3 I believe.
Since 1992, I have been at Los Alamos National Laboratory, where we use
3-500Z, 4CX1500B, 4CX3000A (use Econco rebuilds for these now), and pair of
4CW250,000B as plate modulator for our 3 MW pulsed 200 MHz systems. There
are four RF systems there. Each uses a Burle (former RCA) 7835 super power
triode in cathode driven service in a large cavity to drive a linear proton
accelerator (linac). Plus about forty-four 1 MW klystrons on line driving
side-coupled linac cavities. Presently am back on the HF stuff, designing a
2.8 and 5.6 MHz class A RF amplifier, that is push-pull, driving 500 Ohms
per side with 10 KV of peak RF. It's about 100KW peak per side. Luckily
it's only 5% duty factor so the average dissipation is less than 25 KW. It
can use the CPI/Eimac 4CW150,000E or the Thomson TH555A (uses pyrolytic
graphic grids). So I've been reading your postings with interest and
sometimes chuckles. There's some good advice buried in here. Problem is
finding it through the chaff. No matter.
We have acquired a number of premium components, such as Comet vacuum
variables, HEC (high energy) ceramic caps, homeade (the best!) plate
chokes, Multronics, Kintronix RF inductors. Surcom of Carlsbad, CA is
another good source of parts. And Surplus Sales of Nebraska. I'm sure this
is no news to you guys.
PS, If you find G10 glass/epoxy material overheating as support material in
RF (it does have RF loss which is certainly measurable, compared to
material such as PTFE - Teflon) try G7. It is a grade which uses a silicone
binder with less RF loss than G10. This may not be a concern at the lower
voltages needed in amateur/contesting rigs. It would eat your lunch in the
RF heating business, however.
I use SB220 at home, with some of AG6K's mods, and a few of my own. When I
lost diodes,
I had a set of Semikron (rectifier module samples) SKHM7. They worked
great, in the same place. Did the TR switching mods, put a step start relay
in, added a standby switch, and added a HV discharge relay and resistor, to
speed up bleed down. The parasitic supressors and resistor mods appear to
do their thing, as it had burned bandswitch when I got it, and arced while
tuning. Not anymore.
Part of the bandswitch problem was the input and output wafers had gotten
loose, and lost alignment, so it was possible to be on detent, but not
contacting the correct band, either in open circuit, or bridging two bands.
Ugly.
One of my other hobbies besides ham or Tesla Coils is collecting Diathermy
machines. Anyone got any?
John Lyles
former WB4PRO, now K5PRO
John Lyles
WB4PRO
--
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
|