colinear representation in NEC
 

Tom Donaly wrote:
Jim Lux wrote:
. . .
No it doesn't do prop delay. It does a steady state model. The TL is
just another two port that gets dumped into a giant matrix which is
solved as a system of linear equations. Think of TL as a special case
of NT.

What kind of two port does NEC use, Jim? What is "just another two port?"
73,
Tom Donaly, KA6RUH

NEC network objects are two port networks described by a set of Y
parameters -- see the NEC-2 documentation on the NT "card". When a
transmission line is specified via a TL "card", the Y parameters
appropriate for the specified line length and Z0 are calculated for a
standard Y parameter network which is then used in the model. In the
code, this is done in the NETWK subroutine between line labels 16 and 17.

EZNEC v. 5.0 allows the user to specify a skin-effect proportional
transmission line loss. It accomplishes this internally by appropriately
modifying the network Y parameters.

Roy Lewallen, W7EL

colinear representation in NEC
 

Cecil Moore wrote:
Jim Kelley wrote:
Cecil Moore wrote:

Could it be that a monopole is a "STANDING WAVE ANTENNA"?

The supposition is true, so the intended implication must be that only
standing wave current can be measured on monopole antennas. And Roy
therefore would have to have measured standing wave current (whatever
that is).

I must decline to agree. :-)

About 90% of the total current on an open-ended 1/4WL
monopole is standing wave current with close to unchanging
phase. That's why a 1/4WL monopole is called a "standing
wave antenna".

Presumably they're not called that because of 'standing current'. A
standing wave is just the stationary pattern that results from the
interference of waves. It doesn't really have a 'life' of it's own.

This is such a simple concept - I don't see the problem
in understanding that a wave with the following equation
doesn't change phase with position (x).

Cecil - as the equation is written, the phase term IS position. The
phase of the sine function changes with x.

The phase is the
same over 90 degrees of length no matter what fixed x and
fixed t are chosen. EZNEC supports that fact of physics.
Here's the standing wave equation from "Optics", by Hecht:

The phase of a time varying function changes with time except in the
special case of a 'standing wave' function, where it changes with position.

E(x,t) = 2E01*sin(kx)*cos(wt) quoting "Optics", by Hecht:

"[Standing wave phase] "doesn't rotate at all, and the resultant
wave it represents doesn't progress through space - its a standing
wave."

Right. He could (and should) have gone on to say that standing waves
don't really *DO* anything at all.

Another interesting thing about the standing wave equation
is that the sign of (wt) can be reversed, i.e. standing waves
don't move in either direction - they just stand there. EM
waves cannot stand still so "EM standing wave" is an oxymoron.

Quoting one of my college textbooks, "Electrical
Communication", by Albert:

"Such a plot of voltage is usually referred to as a
*voltage standing wave* or as a *stationary wave*.
Neither of these terms is particularly descriptive
of the phenomenon. A plot of effective values of
voltage, appearing as in Fig. 6(e), *is not a wave*
in the usual sense. However, the term "standing wave"
is in widespread use."

From "College Physics", by Bueche and Hecht:

"These ... patterns are called *standing waves*, as
compared to the propagating waves considered above.
They might better not be called waves at all, since
they do not transport energy and momentum."

Right. All of which deepens the mystery of why you would continue to
insist on claiming that Roy measured standing wave current.

One can use EZNEC's VERT1.EZ to view the essentially
unchanging phase on a standing wave monopole. Just look
at the difference in phase between the feedpoint and a
point 45 degrees up the antenna. In 45 degrees of antenna,
the current phase changes by 3.65 degrees. That is the
current Roy used to measure phase shift through a coil
in order to support w8ji's 3 nS delay "measurements".

I still need you to explain what standing wave _current_ is, and, just
as importantly how it phase shifts by *TRAVELING* through a coil.

73, ac6xg

colinear representation in NEC
 

Tom Donaly wrote:
What's the Z0 of a loading coil, Cecil?

Z0 and VF depend upon the geometry of the coil
*and the frequency*. A 75m Texas Bugcatcher coil
has a Z0 of ~3800 ohms and a VF of ~0.02. The
coil that w8ji used for his 3 nS "measurements"
has a Z0 of ~5300 ohms and a VF of ~0.033. I've
generated an EXCEL file that does the calculations:

http://www.w5dxp.com/CoilZ0VF.xls

I've also got a web page that explains why the
current phase in a standing-wave antenna cannot
be used to measure delay.

http://www.w5dxp.com/current2.htm

I have done the suggested bench experiments myself
and the results are nowhere near w8ji's results.
When traveling wave current is used instead of
standing wave current, the delay is obvious on a
dual-trace O'scope.

This is nothing new. It is based on the information
in the IEEE paper which someone presented years ago:

http://www.ttr.com/TELSIKS2001-MASTER-1.pdf
--
73, Cecil, IEEE, OOTC, http://www.w5dxp.com
"Government 'help' to business is just as disastrous as government
persecution..." Ayn Rand

colinear representation in NEC
 

Roy Lewallen wrote in
:
....

NEC network objects are two port networks described by a set of Y
parameters -- see the NEC-2 documentation on the NT "card". When a
transmission line is specified via a TL "card", the Y parameters
appropriate for the specified line length and Z0 are calculated for a
standard Y parameter network which is then used in the model. In the
code, this is done in the NETWK subroutine between line labels 16 and
17.

EZNEC v. 5.0 allows the user to specify a skin-effect proportional
transmission line loss. It accomplishes this internally by
appropriately modifying the network Y parameters.

I suspected as much from model behaviour.

My question is then, can the 'stub' in figure a) be replaced by a TL
element for a valid model of a)?

Owen

colinear representation in NEC
 

Jim Kelley wrote:
[A standing wave] doesn't really have a 'life' of it's own.

My point exactly, Jim. We may be closer than you think.

Cecil - as the equation is written, the phase term IS position. The
phase of the sine function changes with x.

My point exactly, Jim. We may be closer than you think.
Tom and Roy did NOT use position to determine the phase.
That is the entire point of my posting.

The phase of a time varying function changes with time except in the
special case of a 'standing wave' function, where it changes with position.

Again, my point exactly - something that (apparently)
neither w8ji or w7el wants to admit.

Right. He could (and should) have gone on to say that standing waves
don't really *DO* anything at all.

My point exactly! Now try to tell it to w8ji and w7el
who used primarily standing wave current to prove their
points.

I still need you to explain what standing wave _current_ is, and, just
as importantly how it phase shifts by *TRAVELING* through a coil.

My point exactly, Jim. We may be closer than you think.
I am the one who is saying that it doesn't phase shift
while traveling through a wire or a coil. W8JI and W7EL
apparently think that it does phase shift through a coil
and can therefore be used to measure the delay through
a coil. I am the one who disagrees with that concept.

You and I are on the same side, Jim. This reminds me of
the time someone else realized the gurus were wrong and
simply stopped posting in order to save guru face and
to avoid proving them wrong.
--
73, Cecil, IEEE, OOTC, http://www.w5dxp.com
"Government 'help' to business is just as disastrous as government
persecution..." Ayn Rand

colinear representation in NEC
 

Owen Duffy wrote:
Roy Lewallen wrote in
:
...
NEC network objects are two port networks described by a set of Y
parameters -- see the NEC-2 documentation on the NT "card". When a
transmission line is specified via a TL "card", the Y parameters
appropriate for the specified line length and Z0 are calculated for a
standard Y parameter network which is then used in the model. In the
code, this is done in the NETWK subroutine between line labels 16 and
17. . .

I suspected as much from model behaviour.

My question is then, can the 'stub' in figure a) be replaced by a TL
element for a valid model of a)?

Owen

No, it can't. The NEC two-port network object can't have any common mode
current -- the current out of one terminal of a port is always exactly
equal to the current into the other terminal of that port regardless of
external connections, which means that common mode current is zero by
definition. The wire stub, on the other hand, couples to external fields
which can cause common mode current on the wires.

The only time you can substitute a transmission line (network) object
for a wire transmission line is when the transmission line is carrying
no common mode current. An example would be a transmission line
connected to the center of a symmetrical dipole and positioned
symmetrically with respect to the dipole so it gets equal coupling from
both legs.

A coaxial line can be modeled as a combination of a transmission line
(network) object and a wire, the former carrying the differential mode
current and the latter the common mode current, as described in the
EZNEC manual. This is possible because the two components are physically
separated on a coax line. However, I don't know of any way to do the
equivalent thing with a parallel-wire line because the two components
aren't physically separated as they are on coax.

Roy Lewallen, W7EL

colinear representation in NEC
 

Cecil Moore wrote:
Jim Kelley wrote:
[A standing wave] doesn't really have a 'life' of it's own.

My point exactly, Jim. We may be closer than you think.

Cecil - as the equation is written, the phase term IS position. The
phase of the sine function changes with x.

My point exactly, Jim. We may be closer than you think.
Tom and Roy did NOT use position to determine the phase.
That is the entire point of my posting.

The phase of a time varying function changes with time except in the
special case of a 'standing wave' function, where it changes with
position.

Again, my point exactly - something that (apparently)
neither w8ji or w7el wants to admit.

Right. He could (and should) have gone on to say that standing waves
don't really *DO* anything at all.

My point exactly! Now try to tell it to w8ji and w7el
who used primarily standing wave current to prove their
points.

I still need you to explain what standing wave _current_ is, and, just
as importantly how it phase shifts by *TRAVELING* through a coil.

My point exactly, Jim. We may be closer than you think.
I am the one who is saying that it doesn't phase shift
while traveling through a wire or a coil. W8JI and W7EL
apparently think that it does phase shift through a coil
and can therefore be used to measure the delay through
a coil. I am the one who disagrees with that concept.

You and I are on the same side, Jim.

That's true more than you know, which is why I can't figure out why you
keep claiming that someone has measured standing wave current phase
shifts (whatever they are).

73, ac6xg

colinear representation in NEC
 

Jim Kelley wrote:
That's true more than you know, which is why I can't figure out why you
keep claiming that someone has measured standing wave current phase
shifts (whatever they are).

I previously published the currents in a 20m dipole
with 21 segments. Here they are again. Can you comprehend
why there is only a maximum phase shift of 4.54 degrees
in 180 degrees of antenna? That is NOT the characteristic
of traveling wave current. That is the characteristic of
primarily standing wave current which has constant phase.

Roy reported that the phase shift across a loading coil wasn't
measurable which is a true statement because the standing wave
current indeed doesn't change phase across a coil or through a
wire. But he then used that same evidence to support w8ji's
ridiculous 3 nS delay through a 75m mobile loading coil when
there is no relationship between standing wave current phase
and the delay through a loading coil. Traveling wave current
must be used to measure the delay through a loading coil,
something I have been saying for years.

EZNEC+ ver. 4.0

20m dipole 3/23/2009 8:43:06 PM

--------------- CURRENT DATA ---------------

Frequency = 14.2 MHz

Wire No. 1:
Segment Conn Magnitude (A.) Phase (Deg.)
1 Open .09631 -4.54
2 .2561 -4.25
3 .39868 -3.93
4 .5289 -3.60
5 .64603 -3.25
6 .74868 -2.86
7 .83539 -2.43
8 .90483 -1.94
9 .95592 -1.38
10 .98795 -0.68
11 feedpoint 1 0.00
12 .98795 -0.68
13 .95592 -1.38
14 .90483 -1.94
15 .83539 -2.43
16 .74868 -2.86
17 .64602 -3.25
18 .5289 -3.60
19 .39868 -3.93
20 .2561 -4.25
21 Open .09631 -4.54
--
73, Cecil, IEEE, OOTC, http://www.w5dxp.com
"Government 'help' to business is just as disastrous as government
persecution..." Ayn Rand

colinear representation in NEC
 

Roy Lewallen wrote in
treetonline:

Owen Duffy wrote:
....
My question is then, can the 'stub' in figure a) be replaced by a TL
element for a valid model of a)?

Owen

No, it can't. The NEC two-port network object can't have any common
mode
....

Thanks Roy.

Is NEC capable of modelling the configuration shown at
http://www.vk1od.net/lost/King-22.3b.png (which is the same type of
problem as my figure b)?

My attempts to model b) by modelling a plain conductor and inserting a
load in the segment where the open end of the coax stub would otherwise
be, does not result in an in-phase current distribution.

King discusses the coaxial stub and suggests that the conductors need to
be significantly large in diameter, and the stub length would be less
than a quarter wave for in-phase radiator currents.

Owen

colinear representation in NEC
 

Owen Duffy wrote:

Thanks Roy.

Is NEC capable of modelling the configuration shown at
http://www.vk1od.net/lost/King-22.3b.png (which is the same type of
problem as my figure b)?

My attempts to model b) by modelling a plain conductor and inserting a
load in the segment where the open end of the coax stub would otherwise
be, does not result in an in-phase current distribution.

King discusses the coaxial stub and suggests that the conductors need to
be significantly large in diameter, and the stub length would be less
than a quarter wave for in-phase radiator currents.

Owen

Your description of the model is correct. Technically, the wire
representing the outside of the coax (which in the model is located
where the coax line is) should be the diameter of the shield, as we've
discussed before. The stepped wire diameter error of NEC-2 might,
however, result in less accurate results by doing this than by leaving
the diameter the same as the other wires. Experiments with your earlier
b) model showed that the transmission line object characteristics have
almost no effect on the wire currents when it's inserted at a point of
very low current, and that it doesn't result in in-phase current
distribution. That will be true here also at frequencies where the
current is very low near the open end of the stub.

Roy Lewallen, W7EL