[TCML] high frequency wave propagation along secondary coils

[Steve Ward <steve.ward@xxxxxxxxx>]

Hi List,

Im hoping to get some minds to check my thinking here on a problem I'm
encountering on a job.  Lets get down to what im thinking about:

During a ground discharge, the voltage at the topload has been
measured to collapse within 100nS (worst case) and 250nS (more
typical).  This measurement was just a simple scope probe in the air
to look at the changing E-field.  So, what happens during this event
is a wave front propagates down from the top of the coil.  I suspect
this wavefront eventually "crashes" into the base of the coil,
applying some over-voltage to the bottom end of the coil due to the
fact that its a very abrupt change in impedance (in terms of TL
theory, its like a short).  Sort of like waves in water, the presence
of land causes the wavefront to steepen as they come to shore, rather
than simply reflecting off a wall.

Now, the capacitance from end to end of the secondary (including the
capacitance from the topload to the coil) should essentially provide
dispersion, making this wave-front not so steep, but i dont have the
time or mental capacity to work out how big of an effect this is.  So
I tried my hand at getting a rough estimate of what distance this
wavefront might exist over.  So to figure out this length, i figured
out the propagation speed assuming the secondary capacitance was
evenly distributed along the coil, and using the coil's self
inductance.  The Cself (from java TC) is 12pF (12.75x46" winding), and
the Lsec is ~306mH.  If i use:

VF = 1/(c*sqrt(LC))

i get a propagation constant of .001739c (really SLOW!), or 521853m/s.
 For a 100nS long wavefront, this should take up a length of just
5.2cm (about 2").  Of course, dispersion probably makes this distance
considerably longer, but as i said, i have no idea how much it helps.
By the way, i used to design high-speed kicker magnets, so I'm fairly
familiar with transmissions lines and their non-ideal behaviors
(though its been awhile).

So, my question:  Does my line of thinking make sense?  Could ground
strikes responsible for damaging secondary coils right near the bottom
of the winding?  Has the TSSP project looked into what happens during
ground arc events (step inputs at the top end of the coil)?

Anyway, the "problem" is that ive got a coil here that developed a
shorted turn right at the bottom 3/4" of the winding, which is
something ive never had happen before.  There is a generous 2"
clearance between primary and secondary so i dont think it was a
flashover that caused it.  Im basically trying to rule out if its a
fundamental problem (due to this HV wavefront theory) or simply
defective materials.

One idea i think is worth exploring would be space winding the bottom
portion of the coil.  This would provide possibly a softer transition
in impedance, and would also allow for insulation between turns making
it much more robust to turns shorting out.

Other than this single failure, the other machines (7 other identical
ones) have been quite well behaved, though we did have flashovers from
the strike rail to the lower portion of the secondary coil (maybe also
related to my suggested HV wavefront theory).  I should mention, we
had to use spherical toploads (33" diameter) instead of conventional
toroids, i suspect this doesnt help matters.

TCML archive searches havent been very helpful on this topic (maybe im
picking the wrong words).  This cant possibly be the first time
someone has asked about such phenomena.

Thanks,

Steve Ward
_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla