Hi Ken,
I think Single Resonance SSTC is still a viable option for big sparks, but
a very important consideration is the secondary "impedance". Lowering the
secondary impedance will raise the system power capability (when faced with
a plasma load), which many folks find counter-intuitive. Yes, dropping
Lsec and raising Csec will both help deliver more power to those sparks,
assuming a constant coupling and primary inductance and drive voltage.
Simulations suggest that going to Csec of hundreds of pF (much bigger than
any toroid might provide) can make a *very* powerful SSTC, provided you can
drive it. The reason this works is because the secondary "loaded Q" is
increased, so the result is more voltage gain, because there is simply more
energy stored in the secondary. This effect will not make sense in
simulation unless you include some sort of spark load (series R and C to
ground).
I'll be interested to see the results of your setup, its got a nice big
toroid so that should help.
Steve
On Fri, Nov 8, 2013 at 1:04 PM, Ken Herrick <kchdlh@xxxxxxxxx> wrote:
Greetings-
Single resonance still holds an appeal for me--in part since Tesla
himself, with the benefit of solid-state technology, would surely have
favored it. He didn't, after all, want sparks: he wanted to transmit
power, not into wasteful sparks but out to the populace at large.
Cockamamie idea, of course, and he and his investors lost a bundle, but he
was still a genius.
So here's what I'm working on--after some years' hiatus. To produce
sparks, of course, not radiate power.
Based upon my original notion, which worked but in hardware I made
hopelessly complicated: 8, 1000 uF/450V capacitors arrayed together with 8
power MOSFETs in a ~12" diameter "ring" arrangement incorporating the
equivalent of a 6-turn primary coil. The capacitors to be charged to
full-wave-rectified and doubled mains voltage. This I've built and,
driving it temporarily with 2 signal generators (to provide for
pulse-bursts) and so far only at ~40 V charge, I find that it appears to
work as simulated. The scheme is for the MOSFETs to connect 4 of the
capacitors in series with the coil during each 1/2 cycle, at the
secondary's Fr. That yields, in simulation, ~240A p-p primary current at
full capacitor voltage of ~300. I've devised also a simple
constant-current capacitor-charge circuit so that I won't pop a circuit
breaker trying to charge 8000 uF (plus another 2000 for the doubler) right
off the mains from a cold start.
The 12" x ~39" secondary coil I'll use is left over from my prior
attempts, along with its 6 x 24" Landergren toroid. I have another,
taller, coil as well.
But in simulation I also found much that I wish I'd found out before: 1.
All capacitors may be charged directly in parallel, with the inclusion
only of a single 10 mH isolation inductor between the groups of 4. 2. All
MOSFET sources may be (and are) tied together. 3. Each 4 drains may be
(and are) tied together. And 4. All 8 MOSFETs may be (and are) driven from
a single (D44H8/D45H8 H-bridge) source. I've built the H-bridge and it
seems to do the job. Now I have to build the l.v. signal-processing part,
which takes secondary-return current and amplifies and gates it to provide
the MOSFET drive. I've so far simulated that successfully; it requires
only 3-4 CMOS DIP ICs + the usual small parts--plus a l.v. power supply, of
course, which I also have left-over.
So circumstances allowing (I'm 85), I'll be making sparks again before too
long. Since I won't use a breakout-point, the sparks very charmingly will
dance all around the toroid (as they did before at 20/second or so), making
lots of noise and ozone.
Ken Herrick
_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla