Gentleman,

The point of my post was not to point out the obvious
fact that lumped circuit analysis has some limitations
when used in the context of antenna loading coils. The
debate (at least the one I am familiar with), was whether
or not the current magnitude across an antenna loading
coil varied as the current would vary in a linear section
of antenna having same physical length as the loading
coil, or whether the current magnitude would vary as the
current would vary in a linear section of antenna have
the same physical length as the section of antenna that
the loading coil replaced.

In either case, distributed effects not accounted for
in simple lumped element models are recognized to be
at work. For the former scenario to be true, the current
retardation through the loading coil is presumed to be
roughly equal to that observed in a linear section having
the same physical length as the loading coil. In this case
the retardation would be Tau = length physical/Vp. This
scenario recognizes that distributed effects are at work
(hence the small, but finite current taper), but suggests
that the dominant factor responsible for the loading of
the antenna is the phase shift between the inductor
current and the voltage across it.

The latter case also suggests that distributed effects are
at work, but to a much greater degree than in the former.
In this case, the loading of the antenna is presumed to
be the result of the large current retardation introduced by
the loading coil. In this case, the retardation is presumed
to be Tau = length effective/Vp or Tau = length replaced/Vp.
In this scenario, the effect of the phase shift between the
loading coil current and the voltage across its terminals
seems to be considered incidental and is largely ignored.

The point of my loaded transmission line example was
to show that under either set of assumptions, the
loading coil will produce the desired result. That is to
say that it will load the physically short structure (in the
case of my example, a transmission line) thus bringing
it into so-called resonance. Thus the fact that the loading
coil produces the desired result (e.g. input impedance
match) can't be pointed to as proof that one physical
mechanism is dominate and the other is not. The
transmission line stub loading network doesn't have to
behave the same way as the lumped inductor loading
coil to produce the same desired result (e.g. input
impedance match, resonance, or whatever you want to
call it).

What I am getting at, is that both camps may be
wrong. The answer may lie somewhere in between
these two extremes (e.g. taper equivalent to physical
length vs taper equivalent to electrical length), but this
isn't attractive because its ambiguous and doesn't make
for nice diagrams that can be placed on websites, in
textbooks, or in antenna handbooks (not to mention
all of the accompanying self-righteous chest beating).

73 de Mike, W4EF.................................

P.S. for those of you who have already heard all this
please accept my apologies as I missed out on last
months debate.


"Richard Harrison" wrote in message
...
Richard Clark wrote:
"I thought this was dead long ago."

So did I. This recent posting is a repetition for me, but sometimes
repetition is needed for those who weren`t there in whole or in part for
the earlier postings.

I don`t expect anyone to accept a statement without proof from me that
ordinary circuit analysis does not apply to antennas, but from 3 E.E.
Sc. D.`s who were at the time they made the statement giving their very
best for victory in WW-2, I would expect some serious consideration and
at least a first assumption that the opinion is correct.

Best regards, Richard Harrison, KB5WZI

Michael Tope wrote:
What I am getting at, is that both camps may be
wrong. The answer may lie somewhere in between
these two extremes ...

As I understood it, there is an extreme on only one side. One side
says the current through a loading coil doesn't change. The other
side says that the current through a loading coil does change.
You can look at the decrease in the feedpoint impedance of a loaded
antenna Vs a wire antenna and prove that the coil doesn't exactly
replace that length of antenna. The coil is a more efficient inductor
and less efficient radiator than the wire it replaces which results
in a higher net current at the feedpoint. To the best of my knowledge,
no one has said there is an exact 1:1 correspondence between the coil
and the wire it replaces. The correspondence is only approximate.
--
73, Cecil http://www.qsl.net/w5dxp



-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 100,000 Newsgroups - 19 Different Servers! =-----

Cecil Moore wrote in message ...
Michael Tope wrote:
What I am getting at, is that both camps may be
wrong. The answer may lie somewhere in between
these two extremes ...

As I understood it, there is an extreme on only one side. One side
says the current through a loading coil doesn't change. The other
side says that the current through a loading coil does change.

The current through the coil is not the issue as far as my "camp" is
concerned.
I can see where the current could taper across the coil in certain
setups.
The issue as far as I'm concerned is: does this taper drastically
cause error in modeling compared to lumped elements? I don't think it
does to any great degree, and others data, including Richard Clarks,
and also W4RNL, seem to concur. Or at least as far as I can see. The
taper of the current through the coil is of no great concern to me.
The claim that this variation of current across the coil causes
drastic modeling error is what I have problems with. To me, it's
trying to explain a problem that doesn't really exist, with something
that really doesn't matter that much as far as that problem is
concerned. No one yet has shown any examples of large modeling errors
that is due to this tapering of current. And THATS what the real issue
is. Or at least as Yuri tells it. MK

Mark Keith wrote:
The current through the coil is not the issue as far as my "camp" is
concerned.

Apparently it isn't now, but it was quite an issue for a while there.
Initially it seemed the only correct point of view was the one which
held that loading coils behave strictly as lumped inductances. Remember
that?

The issue as far as I'm concerned is: does this taper drastically
cause error in modeling compared to lumped elements?

I think the answer is essentially, no.

For me the issue was always whether current can be unequal at opposite
ends of an inductor. I find the fact that it can to be very
interesting, and I wanted to understand just how it could be so. I
guess I'm just not willing to accept the notion that just because
fundamentals such as these may be inconsequential to how well an antenna
is modeled, that they are also inconsequential to a thorough
understanding of how it works.

73, Jim AC6XG

On Thu, 04 Dec 2003 15:11:29 -0800, Jim Kelley
wrote:
I find the fact that it can to be very
interesting, and I wanted to understand just how it could be so.

Hi Jim,

It is simply that Kirchhoff's laws have been corrupted in discussion.
The Kirchhoff law of current relates to the flow into and out of "a
closed surface" or a point (where any number of components' common
leads come together) and not to the components themselves (as they
have been incorrectly injected as argument). The corruption is found
in that the current law has been expressed in the language of
Kirchhoff's voltage law by nearly EVERY correspondent.

EZNEC treats loads as lumps, lumps are the metaphor for the "closed
surface" or a point. EZNEC conforms to Kirchhoff's current law, but
not the physical reality simply because in nature a load cannot
exhibit its characteristic within a point (there are no infinitesimal
capacitors or inductors). Hence a protocol was offered to decimate
the inductor and spread its characteristic across the apparent
physical space to achieve the same, virtual response of a true
inductor immersed in reality.

The result of the protocol exhibits roughly the same characteristics
offered by ON4UN's drawings (which are also approximations
themselves).

73's
Richard Clark, KB7QHC

Jim Kelley wrote:

Mark Keith wrote:
The issue as far as I'm concerned is: does this taper drastically
cause error in modeling compared to lumped elements?

I think the answer is essentially, no.

So you haven't tried to model an antenna with a 180 degree phase-
reversing coil, have you? :-)
--
73, Cecil http://www.qsl.net/w5dxp



-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 100,000 Newsgroups - 19 Different Servers! =-----

Cecil, W5DXP wrote:
"So you haven`t tried to model an antenna with a 180 degree
phase-reversing coil, have you?"

Kraus` Figure 23-21(b) has phase-reversing coils used as traps. "Here
the elements present a high impedance to the coil which may be resonated
without an external capacitance due to its distributed capacitance."

Kraus` trap is a self parallel resonant circuit, not just another
inductance.

However, a center-tapped coil can make an excellent phase inverter as in
the vacuum tube type Detroit Electric Company (Delco) Buick car radios
of the late 1930`s and early 1940`s. This radio had great sound despite
limited frequency response inevitable with choke and transformer
coupling.

Best regards, Richard Harrison, KB5WZI

Richard Harrison wrote:
Kraus` trap is a self parallel resonant circuit, not just another
inductance.

Point is that all real-world coils possess distributed capacitance
and distributed resistance as well as inductance. There is capacitance
but no capacitor in Kraus' trap. If one replaces the "phase-reversing
coil" with a phase-reversing stub, EZNEC gives the correct current
distribution.
--
73, Cecil http://www.qsl.net/w5dxp



-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 100,000 Newsgroups - 19 Different Servers! =-----

Mark Keith wrote:
The claim that this variation of current across the coil causes
drastic modeling error is what I have problems with.

Try modeling a 180 degree phase shift coil using EZNEC. (I have a
180 degree phase shift coil in my 70cm mobile antenna.) I guarantee
you will see drastic modeling errors for such an antenna.
--
73, Cecil http://www.qsl.net/w5dxp



-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 100,000 Newsgroups - 19 Different Servers! =-----

Mike, W4EF wrote:
"What I am getting at is that both camps may be wrong."

One of the arguments is that current into one end of a loading coil
equals current out of the other end of the coil. That is not required of
an antenna loading coil in the middle of an antenna. Recall the diagram
of a center loaded short vertical whip from ON4UN`s Fig 9-22 that Yuri
Blanarovich posted early in the dispute. 45-degrees of the 90-degree
total antenna length is replaced by the loading coil. Current tapers
cosinusoidally from 1A at the drivepoint to 0A at the tip.

Cosine of 22.5-degrees = 0.924
Cosine of 67.5-degrees = 0.383

Roy sarcastically referred to "Yuri`s Cosine law". Yuri is right.
Current into the bottom of the coil is 0.924 A, and into the top of the
coil it is 0.383 A. Roy disappeared from the argument.
Yuri seems to have tired of the dispute too.

On page 86, King, Mimno, and Wing say:
"It is fundamentally incorrect to treat a centerdriven antenna as though
it were the bent-open ends of a two-wire line."

This is true for a whip as a continuation of a coax line too. The
antenna should radiate and the line should not. The difference between
an antenna and a transmission line is fundamental. Consider the
equivalent circuit of the balanced line. It is made from distributed
series-connected inductors with distributed capacitors shunted across
the inductor junctions. The two line conductors are closely coupled and
enforce balance in the line. The close equal and opposite currents
discourage radiation from the line.

Attach a non-radiating balanced load across the feedline. The currents
into both terminals of the load must be the same. There is much looser
coupling between the two sides of a dipole than between the wires of a
transmission line.

In a transmission line feeding a mismatched load, the reflected energy
"sees" Zo as does the incident energy traveling the line. Zo is enforced
in both directions by the inductance and capacitance distributed
uniformly in the line.

Due to energy escape in an antenna, incident and reflected energy can
"see" differing impedances on either end of a loading coil. The coil
doesn`t enjoy the type of enforced balanced feed imposed by a balanced
transmission. The feed at its ends is asymmetrical.

Best regards, Richard Harrison, KB5WZI