[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: A double resonance solid state Tesla coil



Original poster: "Jan Wagner by way of Terry Fritz <teslalist-at-qwest-dot-net>" <jwagner-at-cc.hut.fi>

Hi,

On Tue, 31 Dec 2002, Tesla list wrote:
 > Original poster: "jimmy hynes by way of Terry Fritz 
<teslalist-at-qwest-dot-net>" <chunkyboy86-at-yahoo-dot-com>
 >
 > First, I did put pads under all the parts, but it turned out the ones I
 > bought from digikey were non-insolating!

ok! good you were aware of the problem :o)  Btw, I found the wiring
style a bit scary... OTOH, as you used heavy cabling, maybe the wiring
inductance really is low enough (especially at soft-switching).
At least, when something blows up it's easier to replace ;-)

 > I don't know if you saw my original post, but my intent is to transfer a
 > large energy packet into the secondary in a short period of time to
 > simulate a conventional spark gap coil.

Yeah, 've read it.

 > Here's how I chose the frequency and coupling design parameters.
 >
 > 1) The low 70khz frequency keeps the primary current per capacitor
 > reasonable. With lower freqencies you have to run higher capacitance, lower
 > voltage capacitors, so you parrallel them instead of series them.

Sounds reasonable.

 > I'm using
 > motor run caps, don't yet know how much current they can stand. I'm
 > guessing they are going to blow up on me.

I'd imagine that too... Motor run capacitors are intended for 50/60Hz
operation, not RF range <=> 100kHz, nor high current.

It would be much better if you used FKP1 or MKP10 type polypropylene
capacitors. These are rated for awesome pulse currents and high frequency
operation.

http://www.wima.de/navig/menue/prsubnav/forhcr/forhcr.htm

And, they aren't an expensive rarity. I'm sure you can get them from
digikey too, if your corner electr. components store doesn't have any.


 > I didn't want to go too low in
 > frequency because I wanted to stay in the range of existing coils.
 > More primary capacitance of course means more energy is available to
 > transfer.

True. But at the same time it reduces the input surge impedance
Z_tank=Sqrt(L/C) which means you get a higher initial peak current, no
prob for IGBTs. If i understood you correctly then that's what you're
after...

Right now, with the 500nF and 10.3uH primary tank, its input impedance
(with the secondary in place) peaks to just 0.3 Ohms.
   http://www.hut.fi/~jwagner/tesla/temp/

(might be erroneus tho... wasn't the TC sec actually parallel resonant
as seen from the primary? or then i remember wrong...)

 > 2) Surprisingly, lower coupling results in more energy transfer because it
 > allows time (cycles) to pump into the primary before the secondary sucks it
 > down. With a coupling coefficient of 0.1, the energy buildup in the
 > secondary still takes <200usecs.

Yeah :)

 > It sounds like you ran a steady state (frequency domain) sim.

Yup, correct.

 > Your numbers
 > match, but my emphasis is on the transient. I will shut off the primary
 > drive at the first secondary peak and repeat whole the process (break rate)
 > at any desired interval.

Ah, ok.
I thought it would be running CW of some sorts. :)

 > the current in the capacitor doesn't go up to 2800
 > amps because it only has one beat to build up on. I just need to watch the
 > power dissipation in the primary caps, switches, electrolytic rectifying
 > caps etc.

I'm not so sure if this first notch "solid state quenching" works - if the
tank isn't driven by the IGBTs, then the secondary will drive the tank
via the freewheeling diodes. I.e. I don't think that the primary is an
open circuit to the secondary after "quenching". Or?
But if the TC breaks out at the first notch then this won't be a problem
either...

Ok all in all an interesting SSTC approach!! :o) Would be nice if you
could keep the tesla list posted once in a while! :)

good luck!!

sincerely,

  - Jan

--
*************************************************
  high voltage at http://www.hut.fi/~jwagner/tesla
  Jan OH2GHR