Roy Lewallen wrote:
Tom Donaly wrote:

There aren't many people who would support a lumped-element analysis on
this newsgroup. Most people know the limitations of using network theory
in these circumstances. The technical arguments against Cecil's approach
were offered a long time ago. This latest is just a flareup that will
soon die down. You shouldn't be confused. The transmission line model of
antennas is well accepted and hoary with age, particularly for
bi-conical antennas (see Schelkunoff). There are a couple of other
types of models with equal validity. If you really want to know the
physical
score, though, you have to get an electromagnetics text that discusses
the integral equations that govern antenna behavior. Pay particular
attention to the parts that explain why numerical methods like EZNEC
have to be used for solutions rather than the symbolic math most
people would expect and want.

73,
Tom Donaly, KA6RUH

I did and do support lumped element analysis for a very small toroidal
loading inductor, and extensively posted the reasons why in this
newsgroup about six years ago ("Current in antenna coils controversy",
2003). Cecil and Yuri were arguing that the coil would replace some
number of "degrees of antenna" and its current therefore would have a
substantial phase difference between input and output ends. I made and
posted careful measurements to support my statement, after which Cecil
invented his "standing wave current" and went off in various directions.

Roy Lewallen, W7EL

I should have specified a large, solenoidal loading coil such as Cecil
is so fond of using. Cecil has since eschewed his "degrees of antenna"
position, but, for some reason, he keeps claiming your tests on the
small solenoid were wrong.
73,
Tom Donaly, KA6RUH

Owen Duffy wrote:
"Tom Donaly" wrote in
:

...
not be easy. Finally, a modest question: if you have EZNEC, why would
you be wasting time with something inferior? The gold standard is the
gold standard.

NEC (in whatever form) might give a good estimate of the inductance of some
helices, although many practical inductors require smaller segment lengths
that would normally be advised for NEC models, but it does not provide a
good loss estimate in many interesting cases.

My post entitled was about that
topic, and apart from Jim's suggestion of a sensitivity analysis, there was
no solution to evaluating the effective resistance of an inductor of
closely spaced turns (so proximity effect is signficant) made from a
braided round conductor and with a thin dielectric jacket. All three of
these factors are, as I understand it, not modelled in NEC-2.

Owen

So, what does model all the factors you mentioned, and how well does the
Cecil-Corum method do in this regard?
73,
Tom Donaly, KA6RUH

On Apr 23, 2:32*pm, Roy Lewallen wrote:
Jim Lux wrote:
Roy Lewallen wrote:

Let's see how well the principles involved are understood.

What is the delay through a physically very small toroidal coil with
the same inductance as the solenoidal coil? Why?
As in a coil wound on a toroidal magnetic core? or a air cored solenoid
bent in a circle?

I'll say one wound on a magnetic core, simply to keep the size small,
the coupling tight, and the field confined. I don't, however, care how
long a piece of wire it's wound with.

Roy Lewallen, W7EL

There are some other configurations that I personally think are
interesting to ponder. You might never actually build one this way,
but you'll probably gain some insights considering it: an antenna,
say a nominally 1/4 wave vertical for 40 meters made from 4" diameter
aluminum tube (irrigation pipe) twenty feet long, resonated with a
loading coil placed inside the tube across a gap of two or three
inches in the pipe. Capacitance from the coil is almost entirely to
the pipe in this case, not to the world outside the pipe, so the
effect is capacitance in parallel with the coil, not as in a
transmission line where the capacitance is to ground. That's a
different situation than one where a coil with a diameter much larger
than the antenna conductor is used, where the coil has significant
capacitance to the outside world (e.g. to ground).

It's also worth considering that the charge distribution on an antenna
is dynamic, so it's probably not a good idea to try to analyze the
antenna as if there was the same capacitance to ground from the coil
as there would be if the charge distribution on the antenna wire were
static (that is, the DC case, or at a frequency that's a tiny fraction
of the lowest natural resonance of the antenna system).

Analyzing exactly how even a simple wire antenna works in detail is
far from trivial, and when you add in a coil that has significant
physical size, it further complicates things. If you use a simplified
model, it can be useful to gain insights into what's going on, but
don't expect the details to be correct. Be wary about gaining
insights that aren't actually true.

Cheers,
Tom

On Apr 23, 3:22*pm, Cecil Moore wrote:
Art Unwin wrote:
You based your proof of a magnetic wave in a vacuum but it is an
accelerating charge
which obviously must have mass, that is radiation ala the particle.

The accelerating charges are slow-moving electrons.
The RF current moves at the speed of light in the
(conductive) medium. Therefore, the RF current is
associated with photons emitted by the electrons.
Photons have zero rest mass and zero electric charge.
Photons are the particles associated with RF waves.
--
73, Cecil, IEEE, OOTC, *http://www.w5dxp.com

You are quoting the aproach of the bookswhere it is acknowledged that
radiationn cannot fully be explained
For me I am quoting an alternative that does provide the explanation.
If current moves at the speed of light within the Universe ( the
speed of light relative to Earth is slower which creates delay
compared to the former. Insertion of Plank's constant I believe is
a metric that represents the ratio of that delay)
it imparts the same speed to a static particle when impacted, where
the acceleration is determined by Newton's law u.t +f.t sq/2. Since
the particle is static the "u.t" portion equals zero and f.t.sq/2 is
the acceleration from zero to that of the speed of current of the
particle, which is a measure of the expended kinetic energy that
creates the initial format of radiation. I state again without mass
there can be no acceleration.Period
Regards
Art

Roy Lewallen wrote:


Regards,
Steve G3TXQ

Yes, you would see this in the real world. EZNEC does a very good job of
modeling a wire antenna with a loading coil, provided that you model the
coil as a wire helix rather than lumped "load", and you can trust the
results. As I've implied, a lumped load is quite a good model for a
physically small, essentially non-radiating loading coil like a toroid
on a magnetic core.

Roy Lewallen, W7EL

which makes perfect sense...
NEC is a MoM code and is ideally suited to calculating the current
induced in one wire by the currents in other wires. I should think it
would do an excellent job modeling a air core solenoid, especially if
the wire diameter is small compared to the spacing between turns,etc,
assuming that you don't get into numerical precision problems.

"Art Unwin" wrote in message
...
I state again

and again, and again, adding bafflegab and gobbledygook with every
iteration... I love it art, how much deeper can you go with this?

Roy Lewallen wrote:
Jim Lux wrote:
Tom Donaly wrote:
Finally, a modest question: if you have EZNEC, why would you
be wasting time with something inferior? The gold standard is the gold
standard.
Perhaps more the silver or electrum standard.
EZNEC doesn't do dielectric loading, for instance. (unless you get the
Nec4 engine from Roy)

All program types, including the demo, of EZNEC v. 4.0 and later do
dielectric loading similar to NEC-4. (The method came from sources other
than NEC-4.) Like the NEC-4 implementation, it's of limited accuracy and
usefulness -- it's really good only for thin wire insulation of moderate
permittivity.

I stand corrected. Thanks.


And, it's a MoM code, so things not well represented by collections of
wires aren't necessarily modeled well.

Absolutely true. And it can't handle things like patch antennas or
antennas printed on a PCB.

NEC does OK at microstrip patches with air dielectric (or foam with very
low permittivity). I've used it to model an array of 9 patches and the
port to port coupling calculated by NEC and measured by a VNA were
pretty close (within measurement uncertainty).

It's pretty darn slow at this, though (lots and lots of wires in each
patch), I used lumped loads for the matching network model (capacitive
probe feed)

Tom Donaly wrote:
Jim Lux wrote:
Tom Donaly wrote:
Finally, a modest question: if you have EZNEC, why would you
be wasting time with something inferior? The gold standard is the gold
standard.
Perhaps more the silver or electrum standard.
EZNEC doesn't do dielectric loading, for instance. (unless you get the
Nec4 engine from Roy)
And, it's a MoM code, so things not well represented by collections of
wires aren't necessarily modeled well.

Nothing is perfect, but which is better, EZNEC or the Cecil-Corum method
of modeling antennas?

Depends on what your modeling needs are. NEC and it's ilk are more
generalized, but take more computational effort. The Corums have an
analytical approximation that is reasonably good for a certain class of
configurations, although I have to say that for the original Corum
application of Tesla Coils, a lumped approximation gets you almost as
close, at much less work, considering the usual construction tolerances
in a tesla coil. Modern Tesla Coil modeling is typically done with
either a lumped model or a FEM code that assumes it's axially symmetric
and often an assumed voltage distribution. The assumed distribution the
result of a combination of more detailed analytical modeling and some
experimental measurements on real coils, and speeds up the computation
drastically, while not adversely affecting the accuracy of the results
(that is, the changes are less than a few percent, comparable to
construction tolerances on these things).



73,
Tom Donaly, KA6RUH

Tom Donaly wrote:
Nothing is perfect, but which is better, EZNEC or the Cecil-Corum method
of modeling antennas?

EZNEC and the Cecil-Corum method are in agreement.
How would you measure the delay through a wire
or through a coil using the following current
reported by EZNEC through a 90 degree monopole?

EZNEC+ ver. 4.0
thin-wire 1/4WL vertical 4/23/2009 8:01:44 PM
--------------- CURRENT DATA ---------------
Frequency = 7.29 MHz
Wire No. 1:
Segment Conn Magnitude (A.) Phase (Deg.)
1 Ground 1 0.00
2 .97651 -0.42
3 .93005 -0.83
4 .86159 -1.19
5 .77258 -1.50
6 .66485 -1.78
7 .54059 -2.04
8 .40213 -2.28
9 .25161 -2.50
10 Open .08883 -2.71

Exactly how does one use a current that changes phase
by 2.71 degrees in 90 degrees of antenna to obtain
the delay through anything?

Your silence on this subject speaks volumes.
--
73, Cecil, IEEE, OOTC, http://www.w5dxp.com

Tom Donaly wrote:
I should have specified a large, solenoidal loading coil such as Cecil
is so fond of using. Cecil has since eschewed his "degrees of antenna"
position, but, for some reason, he keeps claiming your tests on the
small solenoid were wrong.

Yes, I have fine-tuned my concepts over the past 5 years.
What rational person would not adjust their concepts to
match the technical evidence? (It's a rhetorical question.
We all know who refuses to do that.)

Roy's tests were wrong in the sense that they were meaningless
no matter how accurate the readings. Quoting my web page:

"All of the reported conclusions based on loading coil
measurements using the current on standing-wave antennas
are conceptually flawed."
--
73, Cecil, IEEE, OOTC, http://www.w5dxp.com