[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [TCML] RE: musing on lists ( Wireless Transmission Theory)
William Beaty wrote:
I hope I'll be corrected if I'm wrong, but I sincerely doubt I am about
On Sun, 10 Feb 2008, Scott Stephens wrote:
William Beaty wrote:
I give the coil a 90deg kink, can I stop the waves and reflect them back?
Sharp bends in wire affect the impedance, so yes the impedance
discontinuity will cause reflected power.
What if it doesn't?
Impedance is the combination of capacitive and inductive reactance.
Capacitive reactance is a result of capacitance - the amount of charge
displaced for a given voltage (Q=CV). Inductive reactance results from
the amount of magnetic flux a given magneto-motive-force (measured in
amp-turns-meters) creates. Bending a piece of wire into a loop or spiral
concentrates flux, because there is more amps flowing around the same
area, which displaces more flux.
That's the principle behind the lovely flux-compression generator EMP
bomb. Just like pulling the plates of a charged capacitor apart requires
mechanical force which does work, resulting in an increased voltage.
Yes, the capacitance decreases BUT since E=C x V^2, the energy increase
is exponential where the capacitance decrease merely linear.
When you dilate a copper pipe with a circulating current with
explosives, again the inductance decreases linearly BUT since E= L x I^2
the current increases exponentially.
And I've also read microwave designs should avoid acute transitions.
I usually don't think in terms of wavelength, but rather time-domain and
what the impulse-response would be. Although, through the magic of a
Fourier/inverse Fourier transform you can go back & forth. Some radar is
pulse-echo, other kinds are FM chirps.
A sharp discontiuity would be << wavelength, so it
might be invisible to the travelling wave.
Wouldn't you get crappy magnetic coupling from the primary to most of
Or in other words, if a TC has a long narrow secondary,
First why not use the usual equations to calculate the distributed
inductance and capacitance, then use the free LTSpice to model the two
legs as as transmission lines coupled with a transition impedance? I
think that, using a model of the spark-excited primary LC tank, could
give you a very accurate approximation and allow you to play with large
variations much faster than you could build something. Although it
wouldn't tell you that putting an acute kink will have a concentrated
electric field, resulting in corona or breakout there.
and we cut it in half to form an inverted "L"
shape, will the TC stop operating? (Or perhaps only operate at 2x
frequency?) I suspect that a chain of reasoning might not give the same
results as an actual experiment.
Anyways, just taking a SWAG I'd say the impedance transition would be
really trivial in this application. In radar applications you want to
know about really small impedance discontinuities, especially ones
headed your way at high mach-numbers.
That's a good point. But is it practical? Why don't they start using
high-frequency AC in power transmission systems? They use 600 cycle in
aircraft so the transformers are smaller. I can see them even boosting
that to 6 KHz now that good ferrite is cheaper. But as the frequency
goes up, the skin effect starts becoming significant, and the losses
(A kink in the "electricity hose.") Or, can I connect three together to
form a "Y" junction and have the waves divide evenly?
Google hybrid power combiners, transmission-line transformers, Wilkinson
There is no shield conductor. This is one wire. It's like a G-line. Of
course we could couple the signal back to coax cable, then use
conventional devices. But if this is a waveguide for, say, 50KHz,
wouldn't it be interesting to try to develop some waveguide devices which
can connect directly to plastic tubes wrapped with wire?
Yep. I think LTSpice has some demos you can watch the power slosh back &
forth between the resonant sections. I like the mechanical depictions -
Somewhere I've seen one for multiple rather than two sections, but that
applet at least depicts what goes on with the TC's primary & secondary.
And about that "Y" junction secondary: if the two branches are of slightly
unequal length, won't the TC generate a slow beat-note? Perhaps the
corona discharges would squeal with audio difference frequency.
I think we've chatted about microwave plasma before. I've trapped the
plasma in jars and melted the jars into white-hot goo, and posted the
reference to an article about how to trigger it with a spark and keep it
suspended in a glass chimney. But as the capacitive reactance becomes
relatively low over a few megahertz, an awful lot of power is required
to make high voltage.
I can't help but suspect that if I knew how to make the right shape
waveguide, and stick it in my microwave, and ionize the plasma, I could
spew out hundreds of kilovolts if not megavolts of accelerated electrons.
A microwave *IS* a CW Tesla coil. Well, it's a resonant RLC device being
driven by a sine wave. And it's "secondary" is being driven at a high
overtone, with lots of nodes in a 3D array. If we remove everything from
inside the oven (including the glass dish,) then e-fields approach
air-breakdown because of Resonant Rise. Under these conditions a
sharpened rod of metal or carbon will give off a corona flame resembling a
blow-torch. If a large plastic bag full of argon is placed in the oven, a
3D network of fine sparks results (and then the bag melts.) Complete
description at http://amasci.com/weird/microwave/voltage2.html#storm
No, what I was getting at was making a traveling-wave electron
accelerator. Take a fluorescent light's glass tube and put concentric
rings of wire at certain intervals along its length. The microwaves
excite these resonant rings, and electrons are accelerated.
ooh, lookie what I googled "traveling wave accelerator" :-D
Perhaps its time for some Radiation Therapy...
(pg15) Inject 50KV at the cathode, electrons are injected into a
waveguide at .4C, electrons increase to near C in the first 30cm
Magnetic field required :-(
(pg29) 10 MEV requires a 3 MW Magnetron. Sounds formidable, but I think
that means pulsing the Maggie with 100KV for a millisecond, which sounds
a bit brutal. Common oven maggies would probably arc over around unless
pulsed for microseconds. You couldn't use massive magnets anyways, so
you want to pulse the electromagnet around the waveguide too. And 100
amps sounds about right, but that pulse would need to be in
milliseconds. You'd have to wrap some wire turns around the magnetron
magnet too if the voltage is going to be bumped up because its tuning is
fixed mechanically, and the voltage and magnetic field must match and
Anyways, although it sound intriguing and plausible (since I'm so
ignorant about accelerator design and physics) I suspect a dense plasma
rather than a vacuum in the waveguide would dampen things down. Yet
still, several kilowatts of power would be pushing a very thin, hot gas
and its electrons apart, rather than back'n forth.
Tesla mailing list