Tom Donaly wrote:
It doesn't matter what it lacks or doesn't lack. There is no experiment
that anyone can do that will satisfy Cecil that he's wrong, but there
are an infinity of vague, disconnected references and twisted
interpretations that he will seize upon to prove he's right.

Tom, you seem to me to be emotionally about eight years
old and incapable of anything except ad hominem attacks.
Technically speaking, please put up or shut up.
--
73, Cecil http://www.w5dxp.com

"AI4QJ" wrote in
:

I ran W8JI through the corum calculator. I got a super-well tuned
calculation of several coil parameters that I assume to be correct for
now, and surprise, phase angle is still close to 90 degrees (what else
could it be?). The beta axial propagation factor is 2.122 rad/m. I am
trying to figure out the significance of that; it looks like a
velocity factor correction relative to a traveling wave in a vacuum???

You have to view the helix as a transmission line.

The phase change along the transmission line representation of the helix
is 2.1 rad/m, so along 0.254m, there is a phase change of 0.53 rad or 30°
in waves travelling along the line.

To help get your mind into gear, solve this problem:

You have 4.124m of RG58C/U, the far end is s/c, what is the apparent
inductance at 4MHz looking into the line.

Did you get 1.1uH?

What is the electrical length of the line?

What is Beta?

Now apply the same thinking to the helix transmission line equivalent.

Owen

"AI4QJ" wrote in
:


"Owen Duffy" wrote in message
...
"AI4QJ" wrote in
:

I ran W8JI through the corum calculator. I got a super-well tuned
calculation of several coil parameters that I assume to be correct
for now, and surprise, phase angle is still close to 90 degrees
(what else could it be?). The beta axial propagation factor is 2.122
rad/m. I am trying to figure out the significance of that; it looks
like a velocity factor correction relative to a traveling wave in a
vacuum???

You have to view the helix as a transmission line.

The phase change along the transmission line representation of the
helix is 2.1 rad/m, so along 0.254m, there is a phase change of 0.53
rad or 30° in waves travelling along the line.

To help get your mind into gear, solve this problem:

You have 4.124m of RG58C/U, the far end is s/c, what is the apparent
inductance at 4MHz looking into the line.

Did you get 1.1uH?

I got 1.025uH using 250nH/m.

Well, it is about the right answer, but could be for the wrong reasons.

The line is 30° in length, and Zo is 50, so Xl is about 50*tan(30), then
work out L from there. Is that how you did it?



What is the electrical length of the line?

Electrical length at 4 MHz = 49.5 m assuming polyethylene dielectric.

VF=0.66, so the electrical length is 4.124/0.66m or 6.248m. That is
6.248/75 wl, or 30° (which you needed above).


What is Beta?

Beta = 0.08 rad/m for 75m

Beta=2*pi/75 rad/m or .083 rad/m... so I will pay 0.08.



Now apply the same thinking to the helix transmission line
equivalent.

Thanks Owen.

Owen Duffy wrote:
You have to view the helix as a transmission line.

At a fixed frequency, it has L and distributed C and
some losses, just like a transmission line. A coil
can be conceptually replaced with a transmission line
of the equivalent Z0 and VF if one is mindful of the
boundary conditions for the model.
--
73, Cecil http://www.w5dxp.com

Owen Duffy wrote:
'You have to view the helix as a transmission line."

B. Whitfield Griffith, Jr. treats "Transmission Lines for Special
Impedances" starting on page 293 of "Radio-Elewctronic Transmission
Fundamentals". On page 295 he gives Zo = sq. rt. L/C. Also, velocity of
phase propagation is given as: v = 1/sq.rt.LC.

Best regards, Richard Harrison, KB5WZI

Owen Duffy wrote in
:

....
You have to view the helix as a transmission line.

I mightn't have spelled this out fully.

A 15m straight length of wire could be viewed as a transmission line.

Coiling the wire into the form that Tom measured changes the transmission
line characteristics. Corum's paper and his references deal with
development of parameters for the new transmission line.

Corum argues that solving this new transmission line is a better method
of determining the inductance of the coil that other methods. That
probably understates it a bit, because he represents that his method is a
better way of estimating self resonance, better than the notion of a
lumped parasitic capacitance to deal with distributed capacitance
(another transmission line attribute).

Estimating inductance has been a challenge over a long time.

Corum's method with some extensions is incorporated in the online
calculator to which I gave a link. I have used this calculator to a
limited extent, and it has reconciled well with measured values. My
observations don't prove that it is accurate, but they haven't proved it
grossly inaccurate, and so far, it seems a good estimator for the types
of problems I have solved.

I must admit that I haven't tried solving inductance using David Knights
approach, it may well work and looks to be a degree simpler, but it
doesn't seem to deal with the distributed capacitance /self resonance
issue. One of my to-do jobs is to knock up a calculator and compare both
solutions to some measured values... which is something of a problem for
me because I mainly depend on others for quality measurements.

This issue has held up further work on bootstrap coax traps
(http://www.vk1od.net/coaxtrap/index.htm) . After a lot of work modelling
one of my correspondent's traps, he didn't want the analysis published
because it disagreed with his QEX article. A later QEX article supports
my model. Ah, that's ham radio!

I see people quoting the late Reggie's tools, but unfortunately he did
not expose the underlying algorithms to most of them, so they are a bit
of an unknown quantity in that respect.

I digressed a bit there, but estimating inductance remains a great
interest for RF practitioners.

Owen

"AI4QJ" wrote in
:


"Owen Duffy" wrote in message
...

You have 4.124m of RG58C/U, the far end is s/c, what is the
apparent inductance at 4MHz looking into the line.

Did you get 1.1uH?

I got 1.025uH using 250nH/m.

Well, it is about the right answer, but could be for the wrong
reasons.

Well, I used a manufacturers' spec.. The same manufacturer I think
used 53 Ohms for Zo. I like your method better although mfr's spec. is
probably accurate over a very wide frequency band.

RG58C/U would usually be 50 ohms Zo.

You cannot use the inductance/metre figure in the general case. If you
approached it that way, though your answer is close, the method is
inaccurate for all but very short lines. Compare the results of both
methods on 16.496m s/c stub of the same line at 4MHz, your method gives
4uH, bit the correct answer is that it is not inductive, it looks like a
capacitive reactance around 86 ohms.

The next dimension is the loss or equivalent series R of the impedance
looking into the stub. The maths is a bit uglier, and is implemented in
the calculator at http://www.vk1od.net/tl/tllc.php . Playing with the
calculator will reveal why transmission line elements don't necessarily
make good reactors.

This concepts above will be explained in any decent transmission line
text. Some of these concepts are the reason you are having difficulty in
understanding Corum's transmission line representation of the coil.

Owen

Owen Duffy wrote in
:

....
Simply, I doubt that Tom's experiment, as far as described, was likely
to reveal the value of Beta, except were resonance was observed (which
implies 90° one way phase change along the equivalent line). The test
setup was unlikely to reveal the true undisturbed resonance of the
helix alone, there would be better configurations.

In thinking a little more about this, and thinking aloud, there is no
reason to think that the blip on Tom's delay measurement at around 16MHz
was for a mode of resonance where the one way phase change would be 90°, it
may well have been the next higher mode.

Owen