Re: Spheres vs Toroids

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

Original poster: Paul Nicholson <paul-at-abelian.demon.co.uk> 

Hi All,

Thanks for everybody's comments.

I'll continue this opportunity to plug the thesis that
the topload should be treated as the most important
component of the system, and that it should be chosen to
give a good 'throw' to the streamers. Then the rest
of the TC is designed to support the chosen topload in
that role.

As far as the issue of controlling the field gradient around
the top of the secondary is concerned, I think that's a
comparatively trivial matter.

If a topload chosen for best breakout field happens not to
give good control to the top of the secondary, you can always
fit a small corona-ring type toroid specifically for this,
and this accessory would not significantly alter the field
in the breakout region around the main toroid.

John Couture wrote:
 > Whether you start with the toroid or the TC you will still
 > be confronted with those difficult to coordinate E-fields
 > that were mostly left out in the past discussions.

Yes, and until we make that confrontation I don't think we'll
make much progress in putting up a theoretical basis for
optimum topload shape and size.

Spheres and toroids are the convention, but maybe some other
shape is best for putting out long streamers.  Maybe an
inverted cone with rounded edges, or some such...

I think it would be quite interesting to examine a wide
variety of topload shapes to see which offers the best
field for streamer formation.  For each shape we would

  a) compute the location of the breakout point - easy.
  b) compute the path that a streamer would most likely
     follow - harder but do-able.

(Remember, for now we can forget issues of coil, Fres,
  BPS, etc at this stage and just consider a static field.)

Note that we're not trying to say how far the streamers
would actually develop along that path in a given case -
the suggestion is merely to predict the path they are
likely to follow given that they have enough umph behind
them.  I need to talk more about 'umph' later...

We might find that neither sphere, spheroid, or toroid
is the best shape, and something more exotic would be
better.

I should say that finding such an optimum shape would not
likely make a vast improvement on current performance
of toroids, which may well already be around optimum.
The fact is, once you get a reasonable distance away
from the terminal, the field becomes much the same
regardless of what shape the terminal is.

But it can do no harm to explore theoretically what
shapes might give the best start to the streamers. We
already know from experience that toroids do better
than spheres in this respect, so there is definately
something in the shape to be optimised.

Oh, I seem to be talking myself into another of those
number crunching sessions...
--
Paul Nicholson
--