Tests: Thin wire vs. thick with fewer turns, etc.

Hello coilers,

Today I decided to wind a new secondary for my 42" spark TC.  I
re-used the same form, but rewound it with 660 turns of #20 magnet
wire, compared with the original 1500 turns of #28 magnet wire.  The
winding is 6.5" by 23.5" (same as before).

I then removed the old primary which used 22 turns (out of 44 total)
of #12 stranded close wound PVC insulated wire.  I installed a 1/4"
copper tubing primary (spaced 1/4") in it's place.  This new primary
is unusual in that the bottom 6 turns are coned, and the top 3.5
turns are flat.

All other components were left unchanged; 0.0147uF cap, toroids,
xfrmers, ballast, 120bps sync rotary gap, etc.  The new frequency
is about 200kHz, the old freq was about 90kHz.  The new secondary
is 17mH, the old one was 95mH.  The DC resistance is much lower,
as can be imagined.  The new primary is about 50uH, the old one
was about 200uH at the tap point.

I tested the new arrangement at 580 watts, and the sparks seemed
a little weaker than with the old set up and did not reach 42".  I 
wasn't sure if I had enough primary turns, so I installed a slightly
larger cap (.0164uF).  I was able to determine that it's tuned best at
about 1/8th turn from the end now.  After adjusting the ballast, sync
phase, coupling, etc., I was able to barely obtain 42" at 590 watts.

I still have to play around with the sync phase, and ballasting some
more, and I'm hoping I'll be able to at least equal the performance of 
the old set-up.  The sparks just look short and sluggish now although
they did hit the 42" with difficulty.  In any case, the thicker wire does
not seem to make the sparks brighter, hotter, whiter, or anything
like that, but this does not surprise me.  I will also try higher power 

In any case, I think it's pretty safe to say that there's not a great deal
of difference between; thick wire or thin, higher freq or lower, more
turns or less, thick primary or thin, etc.  Again, the main keys to
performance appear to be; power input, break-rate, and toroid size.
(A caveat; it is possible that some of the new parameters are cancelling
other ones, thereby completely hiding various good or bad features.)
Of course a larger TC should use thicker wire in general, etc.

In any case, the coil is still mostly obeying:

        spark length (inches) = 1.7*SQRT input power (watts)

My secondary is much heavier now, because of the thicker wire, but
it was easier to wind (in a way).  I should strengthen my variable
coupling post to handle the heavier weight. The primary looks a lot
nicer now, and is easier to tap.

The primary and secondary run pretty coolly, but the tank wiring (#12)
gets warm, so I may beef it up.  

When I did a similar comparison some time ago, I saw a great 
reduction in spark length when I used fewer turns on the secondary. 
I think this other secondary used about 550 turns of #24 pvc insulated
wire, and I used a coax primary that time.  I attribute the poor results
of that test to the use of the close wound coax primary.  I don't think
the secondary would make that much difference.  The spark length
dropped from 42" to 29" I seem to remember, in that test.

The work will continue.
John Freau