three phase coiling

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: "Jim Lux" <jimlux-at-earthlink-dot-net> 

I've been running some interesting simulations, in respect of having 3
basically identical (but slightly different) coils running simultaneously,
near each other, to see if you could get triangular or Wye sparks.

The first cut is to try and force the coils to run 120 deg out of phase with
each other (RF phase, not AC charging).  I worked out some LC networks and
such that can do this (you basically create something that looks like a
phase shift oscillator network), but I suspect that as the sparks grow, the
loading will change the frequency enough to perturb the situation.

As a first cut, say you've got a secondary C of around 70 pF total, and you
form a 8 foot streamer: the streamer's going to add 5-10 pF to the system,
pulling the resonant frequency a few percent.

The other thing is that the capacitive coupling between the toploads will
also have a significant effect. That is, if the top load is, say, 40-50 pF
to ground, then it's probably 10-30 pF to a topload a similar distance away
horizontally.

It then occurred to me that if the three coils are "close" in frequency, but
not exactly the same, and if they're reasonably high Q, that as they ring
down, the phase between any pair will go through all values, including 180
deg, where the voltage difference is highest.  In fact, the pathological
case is when they are all precisely in phase at the "bang", because by the
time the phase difference has drifted through 180, most of the energy has
been dissipated.

I suspect that any sort of spark gap system will have enough jitter in it
that precisely synchronized firing (we're talking tenths of microseconds,
here) is unlikely to occur.  In fact, one can help things along a bit by
providing some loose coupling with a phase shift between the primaries, and
running a triggered/rotary gap  on one primary and static gaps on the
others.  As the first coil fires, the voltage change in it will raise or
lower the voltage on the other coil (depending on if it's leading or lagging
coupling), and cause it to fire a bit early or late, making the RF out of
phase.

Comments?