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

Re: Sold state IGBT disruptive coil spark gap idea

Original poster: FutureT@xxxxxxx In a message dated 5/1/06 12:33:15 PM Eastern Daylight Time, tesla@xxxxxxxxxx writes:

example, my little 4.5" coil is only 895 turns with 24 awg. D.C.
would likely add two or three hundred more turns and use a smaller
gauge wire. This may be a phenomenon for high L coils. I guess only
D.C. could say if he's done the same testing with lower turn coils.


I also use 1200 to 1700 turns on my coils, but I haven't seen
the effect.  I do usually run at lower breakrates, but at times
I've run at higher breakrates.  D.C. mentioned something about
the need for large toroids to get some of his effects, so that may
be a factor.

>I agree that coils which use a smaller bang size with higher
>bps should be less likely to form racing sparks for a given
>streamer length.  Also tighter coupling should be safely
>achievable.  It should be possible to get longer sparks
>from a given secondary coil and a given toroid size using
>small bang/higher bps also.  This is because smaller
>bang/higher bps systems develop spark length more via streamer
>"growth due to fast repetition", than via bang size.  However
>the spark length for a given input power tends to drop as the
>bps rises past some rate.

Yes, when multiple breaks per half cycle are in process, gap losses
in the form of heat I think are the main reason. Sparklength also
drops when the cap is too large for the tranny to keep up. Large
coils get to big sparks very fast due to the current available to the
streamer to keep the streamer tip hot as it incinerates it's way
through the surrounding atmosphere. (by "hot", I simply mean at
required potential). Smaller coils are easier to physically see
leader growth, but I don't know that coilers actually observe it.

I often observed streamer growth on videotape from bang to bang.
I don't have the equipment to see intra-bang growth.


occurs immediately, but there is also a heating gap which is stealing
energy at the same time. Because of the gaps direct influence here, I
suspect many coilers actually observe a decrease in sparklength (if
they are simply turning the switch on and trying to observe leader growth).

>Perhaps you could add a feature to your excellent JavaTC
>program which would predict the k point where racing sparks
>would occur for a given coil, power level, bang size, bps,
>toroid size, etc.  It could get tricky, especially since racing
>spark causes may not yet be fully understood.  Still some
>empirical info might be able to be utilized with multi-dependent
>factors, etc.  The study of such things might lead to new
>insights or discoveries in that area.
Yes I could, but there is nothing to add.
We have a lot of theory and no real test data. Even empirically, I
think we are shooting in the dark. Granted, there is window we
already know of, but that window in the case of coupling is massive
in my mind. Some 2-coil systems can run relatively high k (0.25) and
others just can't for whatever reason (and maybe can't get beyond
0.1). I agree it would be fantastic to predict this, but I think it
would take a TSSP type effort to accomplish and gentlemen like Paul,
Gerry, etc. who have that special ability to coordinate such a task.

I too have great confidence in their abilities.