From: Chuck Simmons
Organization: You jest.
X-Mailer: Mozilla 4.61 [en] (X11; U; Linux 2.0.33 i586)
Subject: Re: Scary, high voltage power supplies. (7Kv, 50A, 50ns, 20Khz)
Date: Sat, 09 Nov 2002 21:40:40 GMT
NNTP-Posting-Date: Sat, 09 Nov 2002 13:40:40 PST
> I'm interested in making a copper-halide vapour laser.
> This is basically a tube with .01 bar of helium/neon in, some copper
> chloride (or bromide), which is heated to 400C, and large pulses of power
> This results in a bright green laser beam, with around 1% of the input
> energy, in an ideal case.
> I can get to 7Kv 50ma easily (microwave transformer, and a few diodes
> and caps).
> I'm less sure exactly how to go about pulsing it, though I am clear
> that I'll be doing it from another room, doubling up on all the
> switches and bypass resistors, and having voltmeters on all the caps.
> I assume there is no easy way of building a transformer capable of going
> from a 50ns 500A 700V pulse to 7Kv (I have a large number of BUZ80 800V, 4A
> FETs around).
I suddenly thought of your 20kHz prf (pulse repetition frequency). You
need to adjust that downward a lot for various reasons. The only way I
know of to support that prf is a hard tube modulator and the tubes will
cost an arm and a leg. The soft tubes, hydrogen thyratrons, can do about
5kHz prf at 7000 volts but that is pushing them to the limit imposed by
deionization time. I don't have figures for deionization time of
triggered air spark gaps but it is much longer than for hydrogen
thyratrons. Ignitrons are better than air gaps for that matter.
You also have power constraints. Consider 7000 volt 50 amp pulses at
20kHz. With practical switches (other than hard tube) you will have a
minimum pulse width of around 1 microsecond. This is 0.35 watts for 7000
volts and 50 amps averaged over 1 second. With a 20kHz prf, that is 7000
watts. I hope the microwave oven you took apart was made to cook whole
cows in under two minutes! You wanna back that down to maybe 1kHz to get
into reasonable power levels?
The last thing you want to do is get into hard tube modulators so start
looking at relaxing some of your specifications.
... The times have been,
That, when the brains were out,
the man would die. ... Macbeth
Chuck Simmons email@example.com