Differential triggered gap

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

Original poster: "Daniel Barrett by way of Terry Fritz <twftesla-at-qwest-dot-net>" <dbarrett1-at-austin.rr-dot-com>

    Hi List!
    I have been trying to bring up an 8 inch by 36 inch DC powered coil
designed around a dual-MOT power supply and triggered gap and have had
limited success. I have an idea about the spark gap and wanted some feedback
from some of you gurus ;)
    About the design:
    Secondary is resonant at 140kHz with topload.
    The gap trigger is a microcontroller driving the usual
capacitive-discharge ignition coil circuit. I'm running this thing at about
5 BPS until I can convince myself that the power supply won't crater. I have
burned through about 250 1N4007's this week :)
    The power supply is 2 MOTs, one modified to float the ground connection.
These drive seperate pos and neg doublers, with approx +/- 10kV out. My tank
cap is a center-tapped 45nF geek-cap array (3 string of 5 per leg). This is
charged through a couple of 18mH inductors. Since the power supply is
bipolar, it seems to make sense to me to keep the primary circuit as
symmetrical and bipolar as possible. Because the power supply is reasonably
well decoupled from the tank via theses charging inductors, Im connecting
the gap as follows:

                                                   GAP
+10kvdc-------[18mH]----------O O-----------
                                       |               ^                   |
                                       -               |                   |
                                       90nF         trig               |
                                       -                                  >
                                        |                                  >
Tesla coil
power_gnd----------------                                  > Primary
                                        |                                  >
                                        -                                 >
                                      90nF                             |
                                        -                                 |
                                        |                                  |
-10kvdc-------[18mH]--------------------------

    Gee. Someone needs to write an ORCAD-to-ASCII schematic converter ;)

    Ok, this works, but I'm having a lot of trouble finding a gap spacing
that (a) fires reliably on each trigger, and (b) doesn't power arc or fire
muliple times per trigger. This is proving to be difficult. So now for the
question: Would it be advantageous to build a tripple-gap as described
below? The idea is that the center gap fires (via the trigger circuit),
which raises the potential of the entire primary coil to +some_high_voltage,
which should cause the bottom gap to over-volt and fire, pulling the primary
to -10kv, which overvolts the top gap. Now the main discharge happens.
That's my theory anyway.

                                                   GAP #1
+10kvdc-------[18mH]----------O O-----------
                                       |                                  |
                                       -                                 |
                                       90nF                           |
                                       -
>TC primary
                                        |                                  >
power_gnd----------------                                  > ------O
O--[igncoil]
                                        |                                  >
GAP #3
                                        -                                 >
                                      90nF                             |
                                        -                                 |
                                        |             GAP #2        |
-10kvdc-------[18mH]-----------O O----------

    Does this sound reasonable? Would this help increase the margin between
reliable firing and spontaneous firing? Any comments?
Thanks!
db