DC TSG - It works!

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

Original poster: "S & J Young by way of Terry Fritz <twftesla-at-qwest-dot-net>" <youngs-at-konnections-dot-net>

Coilers,

Today I bit the bullet and tried a variable triggered spark gap (VTSG) in a
filtered DC powered TC setup.  It actually worked quite well, which
surprised and pleased me greatly!  I will describe the VTSG I built, then
the interesting results.

DESCRIPTION:  My VTSG is a pair of 3/16 inch tungsten electrodes spaced
about 3/8 inch apart.  The electrodes are held in copper plumbing fittings
as described my recent posting.  The trigger electrode is a 3/4 inch id
washer centered around the mid point of the tungsten electrode gap.  The
washer has the inside edge sharpened to increase the e-field and improve
triggering performance.  I suppose this configuration is a field distortion
TSG.  The washer is actually a 1 inch to 3/4 inch "reducing washer" used for
reducing the size of a conduit opening in an electrical box.  I soldered it
in a slot in the end of a brass rod which inserts in the perpendicular
electrode holder.  The ring is held at mid-potential of the main gap by two
strings of resistors, and is driven through five .001 mF 10 KV caps in
series.

Allegedly this sharpened ring configuration should perform better than the
TSGs used by Marc and others which have a third tungsten electrode
perpendicular to the main electrodes.  The annular sharp edge is supposed to
induce multiple spark paths.  I viewed my gap in operation through a welding
glass, and it does appear that more than one arc occurs.  Whether I am
seeing parallel arcs or a series of single arcs jumping around, I can't
tell.  The main arc passes through the ring without connecting to it.  This
is a nice feature as you don't have to worry about the trigger electrode
eroding or melting.

I drive the trigger ring with a HEI coil driven by 12 volts switched by a
power transistor.  The power transistor is driven by a 555 with a fixed
pulse width of about 1.1 ms.  This width prevents the ignition coil from
saturating and allows high BPS.  This is driven by another 555 in a variable
oscillator configuration.  Actually, the oscillator is running at 1000 times
the BPS, and drives a divide by 1000 circuit.  This lets me monitor the
oscillator frequency with my Wavetec 27XT reading KHz instead of Hz.  This
provides more accurate readings of BPS.  Output BPS goes from about 1 to
1000 in two overlapping ranges.

The circuits are posted at Greg's Garage site under "Power supply ideas from
Steve Young".  http://hot-streamer-dot-com/greg/  These circuits show a
two-channel setup - just throw out the stuff for the second channel (was
originally for a SPDT type TSG which didn't work.)

I couple the output of a small Hoover canister vacuum cleaner to a half inch
PVC fitting spaced close to the gap.  The air blows through the gap at about
45 degrees to avoid toasting the trigger ring.

DC power is up to 10 KV at up to 200 ma with max DC power around 1.7 KW.  It
goes through a 48 Henry charging reactor (3 MOTs).

PERFORMANCE:  What I didn't know was if the TSG would quench with a filtered
DC supply.  My experience with a RSG was that as the gap dwell time got
longer, a point was reached where the gap would power arc - not a good
thing.  I was hoping a blast of air would prevent power arcing in the VTSG.
It turns out that given enough air velocity and volume, the VTSG works just
as well as a RSG - hurray!

Running the little Hoover canister vacuum at full bore provided the best
performance.  I could crank the power up almost all the way and the VTSG
performed just like the RSG.  Streamer length appeared to be just as long
with the VTSG.  But I don't have quite enough air flow to handle full power.

The effect at full power is dramatic and unusual - possibly a completely new
mode of operation for TCs?  The VTSG does power arc but immediately
extinguishes, then strikes a new power arc, extinguishes, behaving like a
relaxation oscillator.  The output streamers increase in brightness
substantially - there obviously is a lot more energy going into the coil as
this happens.  I suspect streamer length would be significantly longer
during this effect, but at the moment, I can't separate my twin coils more
than 4.5 feet.  No doubt this is very hard on the power supply and MMC and
not a good thing, but it is an amazing effect!

Another interesting effect which happens in the 300 to 500 BPS range is that
if the DC voltage to the VTSG is reduced some, the gap BPS jumps down an
octave (e.g. BPS is 200 instead of 400), firing every other cycle for some
reason.

Another plus is that with the VTSG, I can run at any BPS I want, even down
to 1 BPS if I choose.  With my RSG, I could not go below about 200 BPS
without power arcing.

The downside of this setup is that it takes a good deal of power to generate
a high enough air flow to quench the VTSG.  I am using about 700 watts into
the vacuum cleaner to quench roughly twice that much power in the gap.  700
watts is almost a horsepower, and most RSGs use motors much smaller.  So
perhaps one needs some strong magnets to help "blow out" the arc and reduce
the required air volume?

Overall, it is nice to have a little knob on a pot to vary the BPS and not
worry about high speed noisy RSGs.

Comments and suggested improvements are welcomed.

--Steve