[Prev][Next][Index][Thread]
Re: Inductive Kick Effects - was- Re: cap firing voltage s
Hi John, all,
> Original Poster: FutureT-at-aol-dot-com
>
<snip>
> > Obviously, the neon transformer is storing an overwhelming amount of
> > energy. At only 80 mA across each of it's giant 3700H inductances, a
> > secondary winding stores 1/2 x L x I^2 = 11.84 Joules per side!! I never
> > considered this before. The computer programs just crunched this factor
> > with everything else but I was oblivious to this giant energy being stored.
> > The neon is acting much like a current source do to the giant inductance
> > and will push current into the cap as it pleases causing the cap voltage to
> > skyrocket!! It is possible to get up to 100+ kV if one really tries! Of
> > course, the smaller the primary cap, the more voltage the inductor will
> > push it to...
>
> > This is apparently the mechanism behind the LTR coil's ability to
> charge
> > larger cap values than one would normally expect. I knew from computer
> > simulations that the effect was there but I never really understood why
> > before. *:-)
>
> Can static gaps be set to such a spacing that this effect will occur?
> This might explain why Gary got good results with static gaps and
> LTR's, and I generally didn't. I'll have to look at that again.
Static gaps have a fixed firing voltage but rotaries don't as they
will fire any time the electrodes are close enough for whatever
voltage is present. Backing the phase of the sync gap off will
moderate the effect but decrease output at the same time. In such a
case, there is a strong argument for multiply higher BPS if
overvolting is to be minimized. Of course, one might consider using a
remote mechanism which allows a static gap to be opened after the
cycle has been initiated to take advantage of this but if the gap
"misses" for some reason, the kick will be lost and the Q of the
charging circuit will again dictate the maximum voltage rise present.
In all cases it seems to me that the transformer is facing utter
peril.
Malcolm
<snip>