~44kV primary tests

[FutureT-at-aol-dot-com (by way of Terry Fritz <twftesla-at-uswest-dot-net>)]

Hello coilers,

As we know, a higher tank voltage can reduce spark gap losses, so I
did a quick test and doubled my tank voltage.  Preliminary results
show an 8% improvement in overall efficiency.

Originally I was running with two 7.2kV potential xfmers in series
but I only turned up the variac enough to give a tank voltage of 
about 22kVpeak using resonant charging at 120 bps with the 
sync gap.  Cap is 0.0147uF.

In the new arrangement, I'm using two 14.4kV potential xfrmers in
series, and again, I turn up the variac only enough to give about
44kVpeak on the capacitor.  Cap is ~0.0035uF.  My #12 stranded
pvc close wound primary is holding up well.  I'm using 28 turns now
vs. 14 before.  Luckily this primary had 40 turns to start with, so
I didn't need to add any.  Also my sync gap is working well at
the higher voltage, but I did have to adjust my electrodes a
different way to prevent arc-overs,  HO  HO  !!

I figure my gap losses were originally about 38% based on a 
crude test I did, so the new high voltage setup should reduce
those losses by about 1/2, to about 19%.  I'm now using 600
watts to obtain 42" sparks, vs. about 650 watts before for the
same spark length using the 4.2" by 23" secondary with 1500
turns of #28 wire.  My VA measurement is very similar to the
wattage measurement...showing a good power factor here.

In considering another issue:
In previous tests using the 6.5" by 24" secondary wound with
#28 wire, I obtained 42" sparks using 570 watts.  I had thought
that the performance of the narrow and wide secondaries were
about the same.... but it appears that they are not.  Either that,
or I have a shorted turn in the narrow secondary.  That secondary
has 7 burned spots from previous higher power abusive tests.
I know from past experience that a shorted turn in the secondary
can reduce the spark length by about 10%.  Most of my comparisons
between the narrow and wide secondaries were made at higher
powers at which it was more difficult to measure the spark lengths.

Continuing with the secondary wideness issue:
In any case, if the performance of the wide secondary is better, this
would make sense because the higher inductance would reduce
the gap losses, by forcing more primary turns to be used.  It would
seem that any negative effect of the wider secondary, due to an
increased secondary Cself, may be small compared with the 
benefits of the larger primary surge impedance.  That is my thinking
at present.  I cannot easily go back and test the wide secondary,
because that is the one that I rewound with thicker wire to test
the relative merits of thick vs. thin wire.  (I lost 10% spark length
using the thicker wire.)

I should have more results on these issues eventually.

John Freau