Poulsen help

["Mike Nolley" <nolleym-at-willamette.edu> (by way of Terry Fritz <twftesla-at-uswest-dot-net>)]

	In the last week I've been tweaking the Poulsen arc TC quite a
bit.  However I'm afraid I haven't been successful at producing longer
sparks. I am still hovering at around 5 inches max, as I said, about
the same length and quality as a large flyback type oscillator.  I tried
lowering the supply voltage to 1000 volts and reducing the ballast
resistance.  The results here were interesting:  instead of the 5-10k of
resistance I normally need to prevent the gap from "looping" (blowing out
into an arc), I only needed 500 ohms.  This decreased the arc length while
increasing its current substantially.  Decreasing the distance between the
blow out magnets didn't seem to help.  Also, it seems that the only way
the coil will put out anything significant is if it is operated in bipolar
mode.  The classic configuration results in <1 inch sparks.  Perhaps I'm
not actually operating at the fundamental frequency. . . It could be that
I'm actually getting the secondary to resonate at a partial.  Could anyone
give me some pointers?  Here are the specs again:

2kv supply voltage -at- 1-4A limited by a 10K resistor
.0165 uf capacitance (5 1600V 3.3nf MMC capacitors in parallel)
22 turn primary, tapped every 1.5 turns usually resonates at around 13-15
turns       
800 turn secondary

	I'm thinking that the only way I'm going to get longer sparks is
by increasing the voltage two or three times, and increasing the
resistance to 40K or so.  Even if it doesn't work it will be
interesting:  since it isn't very often that one gets to operate CW at
8kv. . .   Thanks in advance for any help.
		--Mike