Loading Coils; was : Vincent antenna
 

Cecil Moore wrote:

With regard to your comment above, if the maximum amplitude and period
of a sinusoidal wave are both known, then given any instantaneous
amplitude and, knowing whether the slope is positive or negative, the
instantaneous phase can be readily determined.


Take I = K1*cos(x)*cos(wt), a standing-wave equation.
Let t be any fixed value. cos(x) is an amplitude value
and does NOT vary with time. Therefore, the phase of the
standing-wave signal is constant at any particular time
and does NOT depend upon position along the wire or coil.

The item residing inside the parentheses of a sinusoidal function is
in fact the 'phase' of that function. In the expression above, at any
given time, amplitude is determined by the independent variable,
position. Accordingly, for any given position and time there is a
unique amplitude and phase.

Anyone who understands the math would not dare show
his ignorance by asserting that the delay through a
100T coil is 3 ns on 4 MHz or that the measured phase
shift through a loading coil is somehow proportional
to the delay through the coil in a standing-wave antenna.

In the face of such a redoubtable accusation I'm somewhat reluctant to
admit my view that a phase shift across a coil of this sort would in
fact be directly proportional to any propagation delay through that
coil.

73, ac6xg

Loading Coils; was : Vincent antenna
 

On Tue, 04 Dec 2007 19:16:01 GMT, Cecil Moore
wrote:

I also measured ~12-13 ns delay
through 50 turns of the same coil stock that Tom
was using when he measured a 3 ns delay through
a 100 turn coil. That 12-13 ns delay is within
15% of the Corum equation predictions.

Using what equipment? And with what load?

Loading Coils; was : Vincent antenna
 

On Dec 4, 3:34 pm, Ian White GM3SEK wrote:
....
The Boyer paper that I referenced yesterday shows exactly how the model
of an antenna as a reflective unterminated transmission-line handles
inductive loading.


For those who may be interested, I have that article in PDF format,
along with some relevant pages from one of the references that help
with coming up with numerical results.

Cheers,
Tom

Loading Coils; was : Vincent antenna
 

Cecil Moore wrote:
Tom Donaly wrote:
It's hardly surprising that Cecil thinks there's no phase information in
a standing wave, since he leaves it out on purpose. "Cos(x)*Cos(wt)" is
just flat wrong. It's supposed to be "Cos(x + d/2)*e^(i(wt + d/2))."
"d" is the phase difference between a wave traveling in the forward
direction and an equal amplitude wave traveling in the opposite
direction. This is pretty poor shooting for a guy who claims a
degree in symbol slinging.

I copied the equations from "Optics", by Hecht, page
289 in the 4th edition.

Unfortunately, it is apparent that you will sacrifice
your technical ethics to try to discredit me. Everything
I have written is referenced to a source at zero degrees.
Your extra terms do absolutely nothing except obfuscate
the concepts. One can only assume that obfuscation is your
ulterior motive.

Here's what Gene Fuller had to say about this subject:

Regarding the cos(kz)*cos(wt) term in a standing wave:

Gene Fuller, W4SZ wrote:
In a standing wave antenna problem, such as the one you describe,
there is no remaining phase information. Any specific phase
characteristics of the traveling waves died out when the startup
transients died out.

Phase is gone. Kaput. Vanished. Cannot be recovered. Never to be seen
again.

Why don't you two get back to us after you thrash out the
details upon which you disagree?

You're right, I was wrong. It's Cos(kx + d/2). However, there is phase
information on a standing wave, and you know it. If you don't have the
mathematical facility to see how it works, that's o.k., but don't try to
claim you know something about waves if you can't even do the simple
math it requires to describe how it works. Parroting an equation from
a book without understanding what you are parroting doesn't add a thing
to your argument except an admission of ignorance.
73,
Tom Donaly, KA6RUH

Loading Coils; was : Vincent antenna
 

On Dec 4, 10:23 am, Richard Clark wrote:
On Mon, 3 Dec 2007 18:29:24 -0800 (PST), K7ITM wrote:
I can make the antenna conductor be the outside of a piece of
coaxial cable, and use the coaxial inside as a shorted stub which
reflects a pretty good (fairly high Q) inductive reactance back to a
particular point such as a quarter of the antenna length back from
each end, where the stub connects across a gap in the outer
conductor. Can I use such an inductive reactance to tune the
antenna? Will there then be a difference in current at each end of
the gap across which that reactance connects?

Hi Tom,

Interesting proposition. I like it.

73's
Richard Clark, KB7QHC

Hi Richard,

Note that it's also possible to make the stub in the form of a helical
resonator operated below resonance--that is, a loading coil that's
shielded by the tubular conductor whose outside surface is the
antenna. A problem with using plain coax is that the length is
prohibitive. For example, if you make an 80-foot long dipole from
RG-213-size coax, you find that you need about 550 ohms reactance at
points a quarter of the total length in from the ends, to get it to
resonate at 3.9MHz. But using a shorted stub of 50 ohm line requires
about 85 electrical degrees of line. Even with solid polyethylene
dielectric, that's 39 feet of line. Ooops. We only have 20 feet to
work with. Lengthen the antenna to, say, 120 feet, and the required
reactance drops to a low enough value to be practical to do with a
shorted stub co-axial with the antenna wire, but at that point, why
bother? You'd only have to add a few feet of wire to get the antenna
to resonate without inductive loading.

Mostly I find value in thinking about things like this because they
make more clear what's really important: it's primarily the inductive
reactance that tunes the antenna; the parasitic capacitance from a
loading coil to the outside world, which is what causes it to behave
like a helical delay line, is of much lower importance in determining
the antenna tuning. In a long antenna that's capacitively loaded, the
capacitors can have negligible parasitic series inductance and shunt
capacitance to the outside world, but they still strongly affect the
antenna loading.

Of course, the closer to the end of the antenna you put a large
loading coil, the more effect its capacitance will have. In the
limit, you can dispense with the coil and just add a capacitive hat
after all. Even modest size conductive balls on the ends of a thin-
wire dipole will have a significant effect on the resonance.

Cheers,
Tom

Loading Coils; was : Vincent antenna
 

Cecil Moore wrote:

Take I = K1*cos(x)*cos(wt), a standing-wave equation.
Let t be any fixed value. cos(x) is an amplitude value
and does NOT vary with time. Therefore, the phase of the
standing-wave signal is constant at any particular time
and does NOT depend upon position along the wire or coil.

Now take I = K2*cos(x+wt), a traveling-wave equation.
Let t be any fixed value. The length dimension 'x'
has an effect on phase, i.e. the phase of of the
signal indeed does depend upon BOTH position AND time.


Cecil,

I know what you are trying to say, but you got the message screwed up.
If 't' is fixed, then the equations are essentially the same with regard
to 'x'. That is typical; a snapshot in time does not say much about the
wave behavior.

73,
Gene
W4SZ

Loading Coils; was : Vincent antenna
 

Cecil Moore wrote:


Here's what Gene Fuller had to say about this subject:


Cecil,

There you go again; quoting a questionable source.

73,
Gene
W4SZ

Loading Coils; was : Vincent antenna
 

Richard Clark wrote:

Hi Roy,

The complaint is:
Wire 3 segment length too short. L = .01914 m; recommended min. =
.07495 m.
and so on for 800+ lines.

Attempting to find the Src Data results in:
Wires 3 and 10 contact improperly or are too close.
Wires 3 and 11 parallel and contacting.
Wire 3 end 2 contacts improperly or is too close to wire 12.
and so on...

However, on close examination
Pilot Error
The wire is too thick (I noticed this in modeling Tom's coil at the
Corum calculator and hadn't done the correction yet in EZNEC). The
geometry still complains, but it doesn't inhibit processing.

Thanx anyway.

Thanks for the additional info. EZNEC's checks are often not fully
understood, so I'll briefly describe what they do.

EZNEC provides two types of checks, segmentation (formerly called
guideline) and geometry. The first is advisory and tells when segment
lengths or segment length/wire diameter ratios are outside NEC
recommendations. As I mentioned in an earlier posting, warnings about
too-short segments can often be disregarded without significant
consequence, and this warning is often seen when a helix is created. The
geometry check looks for situations such as touching or overlapping
wires which can cause serious calculation errors. It takes finite wire
diameter into account, so sometimes the cause of a reported error
requires a bit of thought to determine, as it did here. EZNEC won't do
calculations if any geometry error is present because of the high
probability of the result being erroneous. The geometry check was
introduced in EZNEC v. 4.0; the segmentation check has been present for
a much longer time. NEC-4 has a check similar to but less complete than
EZNEC's geometry check, but no equivalent to the segmentation check.
NEC-2 has neither.

Roy Lewallen, W7EL

Loading Coils; was : Vincent antenna
 

Gene Fuller wrote:
Cecil Moore wrote:

Take I = K1*cos(x)*cos(wt), a standing-wave equation.
Let t be any fixed value. cos(x) is an amplitude value
and does NOT vary with time. Therefore, the phase of the
standing-wave signal is constant at any particular time
and does NOT depend upon position along the wire or coil.

Now take I = K2*cos(x+wt), a traveling-wave equation.
Let t be any fixed value. The length dimension 'x'
has an effect on phase, i.e. the phase of of the
signal indeed does depend upon BOTH position AND time.


Cecil,

I know what you are trying to say, but you got the message screwed up.
If 't' is fixed, then the equations are essentially the same with regard
to 'x'. That is typical; a snapshot in time does not say much about the
wave behavior.

73,
Gene
W4SZ

It's generally cos(kx), but maybe Cecil is using a wave where k = 1,
that is, the wavelength is 2*Pi.
73,
Tom Donaly, KA6RUH

Loading Coils; was : Vincent antenna
 

Ian White GM3SEK wrote:
The mathematics is clear enough, but it provides no justification
whatever for your conceptual leap to "a completely different KIND of
current" (my emphasis). You are only doing that to justify the different
kind of behavior that your model demands for a loading inductance - in
other words, you are trying to patch one error by adding a second error.

I suggest you study up on your math because what you said
above is just not true. cos(x)*cos(wt) *IS* obviously different
from cos(x+wt). Here is a graph of the difference between
standing-wave current and traveling-wave current. Please
study it until you comprehend the differences.

http://www.w5dxp.com/travstnd.gif
--
73, Cecil http://www.w5dxp.com