[AMPS] chokes and Xfmrs

[John Lyles]

I have spec'd a number of transformers for plate, screen, bias and filament
service in amplifiers, and found good and bad sources.

For instance, Basler Electric in East St. Louis, MO, is a fine source for
plate power. Expensive, yes. Gates/Harris uses them a lot up in Quincy.
Stangenes in CA makes the best isolated transformers for filament power,
where you want 50 Kv of isolation between windings. (Use - klystron
filament, for instance). NWL in New Jersey seems to make fairly good plate
iron, but it is costly and big. SNC in Oshkosh, Wisconsin makes fine
transfomers but has sort of gotten out of the custom business and makes a
lot of wall wart supplies. American Magnetics in Carterville, IL makes some
very nice inexpensive iron for lower voltages at high I. Recently I got
them to make my TH555A filament iron, rated at 15 VAC at 300 Amps or so.
Aydin in CA is out of business now, but the principals moved into a shop
called NEWS Magnetics, in San Francisco. They are back into making larger
oil filled tranformer rectifier systems for TV xmtrs and particle
accelerators. Electo Impulse of CA is out of business. And then theres
Peter Dahl. He makes nice stuff, but it is costly and sometimes you don't
need hypersil wound cores and can live with EI stamped steel. I've had him
make a few replacement units for a KV or so output, where the original
oil-filled UTC units were no longer made.

There are numerous others that keep knocking at the door for work, such as
Neeltran in the Northeast US. And I am Carl has found another one. The
point is, be sure that your vendor really understands the temperature
ratings, and will supply the copper and iron needed to keep the temp rise
down. A lot of low bidders will cut on this, and their units run 140 deg C.
With high temp insulation, they work, but don't expect much headroom for
the day you have to crank 'er up.  Bigger, better insulation, heavier, more
copper, more iron, all translate into cooler units, in general.

On the subject of chokes: I built a resonant choke PS with 4500 Volts DC,
at an Ampere. It didn't fail, the choke was merely open frame, varnished.
It was made by SNC, in 1983.
i am not sure the production quantity of them that were made.

Last summer I had Carl Seivers, the OM at SNC who really knew transformers
(now retired), design a saturable reactor. This thing uses a programmable
Lambda DC supply, about 30 VDC at 1.5 Amps max, to supply the bias current
in one winding. The other windings are in series (there are two)  with the
480 VAC to my American Magnetics filament pig. In this manner, I program a
ramp up of the filament, over 8 minutes per the Thomson tube spec, without
moving parts like variacs. And SCRs are not used. This means it fails safe,
off, if the bias drops. Instead of slamming on if an SCR shorts on, and
causing a hot grid to filament short in the $43K tetrode. It also ramps
down the filament when you command the amplifier off, and runs the blower
for 15 minutes more to cool the bottom of the tube. The SNC sat reactor
really HUMS when it ramps up, mainly in the middle of the curve, before it
is fully saturated. This is normal. They used to be used a lot for theatre
lighting, before AC phasing SCR supplies became the vogue. The solution,
isolated mounting on rubber isopads. Seems to work.

John
K5PRO

--
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