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

Re: TSG Configuration



Original poster: "B2 by way of Terry Fritz <twftesla-at-qwest-dot-net>" <bensonbd-at-erols-dot-com>

Hi Greg, All,
    See Below for reply:

>The trigger electrode appeared to be
> inside an insulated sheath, with the business end of
> it brought flush with a hole in the face of the main
> electrode. This suggests that the trigger spark would
> only need jump a very short distance to the edge of
> the hole in the main electrode, and this somehow
> triggers the main gap. Is this field distortion, or
> some other mechanism such as ion flooding? How come
> home brew spark gaps and commercially made spark gaps
> are so different from each other? The physical
> arrangement doesn't even seem related.
> 
> Greg
> http://hot-streamer-dot-com/greg

    The type of gap you are describing sounds like a trigatron.  The
trigatron mechanism has been a mystery for quite some time(1).  A study in
1989 (2) showed that the breakdown actually occured by first forming a
streamer between the trigatron pin and the opposite electrode with
subsequent breakdown to the main electrode holding the pin.

    The trigatron is one of the earlier gap designs.  It's main drawback is
erosion of the trigger pin over time.  With a fresh trigger pin it can be
triggered down to about 20% of the full gap holdoff voltage.  A 10 KV
trigatron should be able to be triggered by a 2 KV pulse.  This voltage
increases as the pin ages.  

     T. E. Broadbent (7) decided to try to find a new type of switch by a
sort of geometrical exfoliation of the trigatron concept as it was
understood before (1).  He first tried a three ball spark gap.  He found
that triggering was reliable if the balls were spaced at less than half of
their diameters (uniform field).  Next, he replaced the middle trigger ball
with a flat plate.  The same relation held for the plate.  Then he decided
to put a hole in the plate to let the UV through.  He found that a flat
plate trigger electode, with a hole in it equal to one half of a ball
diameter, and the electrode balls spaced at less than one half of their
diameter gave the best triggering.  This type of spark gap is called the
swinging cascade.  As the midplane breaks down (cascades) first to one main
electrode and then to the other, the midplane voltage swings from the
trigger pulse voltage (say -10KV) to the HV electrode voltage (say +10 KV)
and then back to almost zero as the switch breakdown process completes.

     Maxwell (and others) took this design several steps further(3)(4).
They found that sharpening the inner edge of the hole caused such a
profound field distortion from the trigger pulse that multistreamering took
place.  This caused multiple spark channels to form simultaneously.  This,
in turn, dramatically reduced the spark gap resistance.  They also made one
more change.  They added a trigatron like preionization pin within the hole
in the plate.  This was to reduce the time jitter between switches.  Time
jitter becomes important for EMP generators where up to 80 switches have to
breakdown within nanoseconds.

    As switches have to handle bigger loads the electrodes become bigger.
As the electrodes become bigger the switch becomes easier to trigger.

    Homemaid spark gaps tend to be limited to commonly available materials
whereas commercially produced spark gaps are usually made for uses where
cost and materials availability is not an issue.  The sealed gap trigatron
spark gaps are disigned from a materials cost perspective.  A swinging
cascade design would involve more material costs in addition to extra
vacuum sealing operations which may not be cost effective for
manufacturers.  Trigatrons are fairly easy to make from old Volkswagon
spark plugs (long centers) and some nonresistor lawnmower spark plugs.  The
swinging cascade is much lower maintenance and has less of a problen due to
erosion (less sensitive).  Brass doorknobs and brass plates are readily
available from hardware stores and Small Parts Inc. respectively.  Getting
everything lined up just right requires about the same ingenuity as a
rotary spark gap.  The Tesla coil spark gap will evolve to take advantage
of these designs.

    There are many other neat designs such as the V over M switch (5)(6)
which is sort of an involute of the swinging cascade switch.  They both
take advantage of the fact that a plate parallel to an equipotential line
and with the same potential (bias) "disappears" electrically from the view
of both electrodes.  There are cutting edge designs such as the
"Pseudospark" and "BLT" (backlit thyratron) which replace the trigatron pin
with a stream of ions delivered across different pressure regimes.  All of
these could be suitably reconfigured for Tesla coil use but the cost is a
major limiting factor.

    The US Patent office (8) gave 3397 hits for "spark gap".  There might
be a few that I have missed!

(1) Single- and Double-Mode Gas Trigatron: Main-Gap and Trigger-Gap
Intractions, By P. H. Ron, K. Nanu, S. T. Iyengar, V. K. Rohatgi, Journal
of Physics D, Volume 21, November 1988, Pages 1738-1744.
(2) Triggering in Trigatron Spark Gaps: A Fundamental Study, By P. F.
Williams and F. E. Peterkin, Journal of Applied Physics, Volume 66, Number
9, November 1989, Pages 4163-4175.
(3) Dr. Y. G. Chen (Maxwell Laboratories now General Atomics) personal
discussion.  Dr. Chen now works for Hipotronics in Taiwan.
(4) Numerical and Experimental Design of Three-Electrode Spark Gap for
Synthetic Test Circuits, By P. Osmokrovic, N. Arsic, Z. Lazarevic, D.
Kusic, IEEE Transactions on Power Delivery, Volume 9, Number 3, July 1994,
Pages 1444-1450.
(5) The Use of Triggered Multichannel Gas Spark Gaps to Produce Nanosecond
Risetimes, By I. Smith and V. Carboni, Proceedings of the Defense Nuclear
Agency Advanced Pulsed Power Conference-31 July-3 August 1990, DASIAC
TR-90-006, Technical Report, Pages 6-57 to 6-66.
(6) Low-jitter, low-Inductance, Electrically Triggered Spark Gap, By L. L.
Small, D. C. D. McKen, and A. A. Offenberger, Review of Scientific
instruments, Volume 55, Number 7, July 1984, Pages 1084-1089.
(7) T. E. Broadbent, The Characteristics of the Trigatron Spark Gap at Very
High Voltages, Proceedings of the IEE (London), 107C, Volume (?) 213, 1960.
(8) US Patent office on line: http://164.195.100.11/netahtml/search-bool.html

Cheers,

B2