Cecil Moore wrote:
Richard Harrison wrote:
Distributed network theory is newer than lumped network theory but both
have been around plenty long enough to be well established.

Neumann didn't prove that the current through a coil is uniform.
He *assumed* it to be true as a simplification, so he could factor
out the current term from under the integral sign in one of Maxwell's
equations. Now Neumann's mathematical shortcut is being reported
here as a fact of physics. That's how well established it is.

At the most basic level, Cecil does not understand what scientific
theories are all about.

That was not a "shortcut" or approximation. It was a rigorous test for
the limiting case where the coil has no other properties except pure
inductance, so no electromagnetic radiation is taking place. In that
limiting case, Maxwell's equations MUST join up with conventional
circuit theory... and indeed they DO.

In that limiting case, the current at the two terminals of a pure
inductance must be the same in both magnitude and phase. By "current" we
mean the simple, straightforward movement of charge. If you count the
electrons in and out at the two terminals, there can be no difference in
either magnitude or phase because that would require electrons to be
stored or lost from somewhere - which inductance cannot do. Kirchhoff's
current law recognises the logic of this.

This is how inductance always works in every type of non-radiating
circuit, both in theory and in real life.


When developing a new theory, it is normal, standard required practice
to test it for simplified, limiting cases that we already understand.
The new theory MUST work for all these test cases; it MUST connect
seamlessly with everything we already know.

At his point, some heckler pipes up: "Ah, but what about Einstein?"
Thank you, sir - the perfect example to prove my point! If Einstein's
equations of relativity are tested for the limiting case where
velocities are very low, they connect seamlessly into Newton's laws of
motion. If they hadn't, Einstein would have thrown them out and gone
back to think again.

Another point: antennas are the home territory of classical physics,
where everything is consistent with everything else. Classical physics
is plain physical reality, verified every day, a billion times over.
All of our validated knowledge joins together completely seamlessly -
and that's how we test anything new. In the home territory of classical
physics, reality is hard, sharp and clear. It allows no gaps and no
excuses. "What don't fit, ain't true."

Trained scientists accept the discipline of that, and turn it into a
useful tool. Engineers aren't always trained to think that way, but the
best are of a mind to do it anyway. They're creating new ideas all the
time, and the ones that "don't fit" or "don't make sense" are simply
abandoned (sometimes they're abandoned before they even make it into
fully conscious thought). One way or another, lots of ideas get thrown
away; no big deal, it's just part of the process of *having* ideas.

So...

When Cecil's theory is tested for the simple limiting case of pure
inductance, it MUST join up seamlessly with conventional circuit theory.
If it requires anything that "don't fit", such as a phase shift in
current through a pure inductance, or special kinds of "current" that
are different from the simple, straightforward movement of charge
(electrons), then the theory fails.



--
73 from Ian GM3SEK 'In Practice' columnist for RadCom (RSGB)
http://www.ifwtech.co.uk/g3sek