Re:OLTC update - Poor seconadry Q

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

Original poster: "Greg Leyh by way of Terry Fritz <twftesla-at-qwest-dot-net>" <lod-at-pacbell-dot-net>

Hi John,

That's an interesting equation... I'd like to
see it with the non-whole number roots.
Did you arrive at this equation empirically
by matching it to data points?

The 120" secondary out at Jim Heagy's place
in the shipyard beats the equation by about
15%, and Electrum falls short of the equation,
also by about 15%.  This could be explained by
the different engineering approaches of the two
coils.  Electrum is conservatively designed, and
has a few 'artistic compromises' in the engineering
solution as well.  The shipyard coil enjoys a more
optimal top load capacitance and primary Q, and is
run 'balls out', in terms of design conservatism.

Are there many data points that fall in a nice
distribution around this curve?  It would be
interesting to find how far out this curve
could be reasonably extrapolated.

Inserting 240BPS and 3.6MW into the equation
yields 402ft... which is more than twice the
spark length performance required per tower.
On the other hand it's probably good to have
some reserve capacity, especially when it's
not practical to field rework many of the
critical coil parameters.


>Original poster: <FutureT-at-aol-dot-com>
>
>[snip]
>The more complex formula follows, but I must admit I don't
>trust it all that much, and as I mentioned, it would benefit from
>the use of non-whole number roots.
>
>Spark length inches =
>     [(3.9*16th.root watts)*sqrt watts] / 4th.root BPS
>
>(for 120 - 480 bps)
>(watts are wallplug watts)
>
>I would estimate that the lengths predicted by my formula are
>far above the average spark length, especially for smaller coils.
>For example, many 12/30 NST powered coils give only about
>an 18" spark or so, which falls far short of the 42" that my
>formula predicts, and which my old research coil achieved.
>Many small coils are limited by; poor spark gaps, too small
>toroids, too few primary and secondary turns, etc.  It would
>be really tough to beat the 42" spark length for a 12/30 NST
>(600 watts), by any significant amount I think.  Actually, I've
>never heard of anyone beating that result.
>
>At higher powers, many well designed coils meet or slightly
>exceed the formula's predicted spark lengths.
>
>John