Re: Primary Heating

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

Original poster: "Paul Nicholson by way of Terry Fritz <twftesla-at-qwest-dot-net>" <paul-at-abelian.demon.co.uk>

Gary Lau wrote:

> I wonder if a conical primary, where the inner turns are less under
> the influence of the outer turn's magnetic field, can be shown to
> have lower proximity-effect losses?

Plots for a 45 degree primary with uniform current are in
 http://www.abelian.demon.co.uk/tmp/cone-field1.gif
 http://www.abelian.demon.co.uk/tmp/cone-field2.gif
Secondary carries no current - there for illustration only.

I made the grid artifacts go away by the simple expedient of
integrating each turn in 2000 steps instead of 20.

The inner turn field is still clearly higher than the rest of the coil,
but with this kind of graph it's hard to compare quantitatively.
We need firm numbers that we can put side by side.  I suppose if we
computed the loss according to squared field derivatives, then it
would give us some sort of basis for comparison, even though the
actual calculated loss would not be very accurate for these tubes. 
Guess that's another function into acmi next weekend.

> Since the losses necessary to heat up a significant length of copper
> tubing probably are significant,

Could somebody estimate it?  Run some 60Hz through the inner few turns
and bring them up to roughly the same temperature as seen when
operating.  Note the 60Hz power required.

> would the losses in a primary segment (just the innermost 2-3 turns,
> since Litz wire can't be tapped) constructed of Litz
> wire be significantly less?

I suppose so, but don't know how to calculate it.  Whether it makes
a worthwhile difference hinges on knowing what fraction of the bang
energy is being lost to this cause.  The above measurement would give
us a rough idea of whether this is worth persuing.
--
Paul Nicholson
--