[Prev][Next][Index][Thread]
Re: Any Very High Freq. TCs?
Tesla List wrote:
>
> Original Poster: "Malcolm Watts" <MALCOLM-at-directorate.wnp.ac.nz>
>
> Hi Steven,
>
> > Original Poster: Steven Ivy <adder_black_the-at-yahoo-dot-com>
> >
> > I have been looking for a while and I have yet to see any real mention
> > of the significance of the overall operating frequency of a TC other
> > than it is best to have the primary and secondary resonances tuned to
> > the same frequency. I was just wondering if the standard TC topology
> > is still useful at very high frequencys like at 1 Mhz or even much
> > higher? In principal I don't see any reason why the whole concept
> > should not scale nicely and produce a perfectly good high performance
> > design that would only need to be a small fraction of the size of the
> > ones we usually see. It would be a lot of fun to have half a million
> > volts at 1 GHz operating on my kitchen table : ) Is there some
> > particular flaw in this scaling idea other than the difficulty in
> > producing a spark gap capable of operating at these very high
> > frequencys.
>
> Conventional TC technology:
> There is a problem assuming one wants a physically small machine :
> how do you prevent 1/2MV from arcing over a secondary a few inches
> long?
> There is a second problem: to get any real energy into the
> primary, the cap value would have to be such that the primary
> inductance was vanishingly small.
>
> You could do something with waveguides but I'm not sure how you'd
> handle 1/2MV. However, there may be ways around these obstacles I've
> not thought of. Be interested to hear what others have to say.
>
> Malcolm
What you want is a quarter wave transmission line that is open at one
end and shorted at the other. Rather than lumped LC resonance like a TC,
you're using the line as an impedance transformer. Low impedance (at the
shorted end) where you excite the line, is transformed to high impedance
at the other (where you get high voltage at low current.
Actually, they use this technique in radars for TR switching. You put a
stub line in with a spark gap at the high impedance end. When the
transmitter pulse comes down the guide, the gap breaks down, absorbing
the energy.
Your real problem is going to be in getting the power into the line. A
spark just doesn't have all that much energy in it.
The breakdown question is just a matter of insulation ("all you gotta
do"). In high power waveguides they use pressurized gases to fill them,
and, they make sure that there aren't any impedance discontinuities to
create standing wave hot spots.
A 1 GHz waveguide is going to be pretty big (about a foot by 6", or
thereabouts). At 1 GHz, you are at a point where perhaps some sort of
stripline might work better. Or, build a coaxial line out of two
concentric copper pipes.