On Mon, 11 May 2009 02:26:04 -0700 (PDT), wrote:

I should have been more explicit.

I took the "Axial Propagation Factor" (4.372 rad/m) figure which was
given by the HamWaves calculator and multiplied it by the coil length
(155mm) to find the effective electrical length of the coil (38.83
degrees). Then I took cos(38.83)=0.779 as the fall-off in current
across the coil.

Hi Steve,

I don't often drop into this side-thread as the topic had drifted into
a stagnated intellectual backwater.

On this and one prior posting by you:
On Sat, 9 May 2009 13:56:31 -0700 (PDT), wrote:
OK, I tried what you suggested. I put my loading coil midway up a 20ft
vertical wire in the EZNEC model. I reduced the number of turns to
lift the resonant frequency to 5.6MHz.

I note how little Corrum really has to offer when you had to take the
same:
effective electrical length of the coil (38.83 degrees)
and change it (to the same effective electrical length? I think not.)
to fit the same available wire, at the same specific frequency - only
at a different height along the available wire.

By my quick read on the stale crisis of current "fall-off" and proving
Corum by EZNEC; it seems quite apparent that EZNEC (the authority) is
driving the coil requirements which are then force fitted by Corum's
inappropriate application.

After all, Corum says nothing of:
1. Application;
2. Base loading;
3. Mid or Top loading;
4. Stinger selection;
and yet all solutions seem to derive from their math with the elegance
of an ad-hoc "missing degrees" provision (that is quickly discarded as
shown above when current becomes the focus).

Corum DOES say that the formula is only applicable for certain
constraints which I note are NEVER observed in the application nor the
breach. All of the commentary proceeds through equation (32) when
every argument is an instance of equation (31).

How much are you willing to accept of that paper (which is another way
of asking how much you are willing to discard)?

I will ask one ace-buster question that I expect no one will answer:
Show me the computation for M (= tau · a)
which would be appropriate for the NON-quarterwave resonance of the
coil in question at 3.85 MHz.

For extra credit:
1. What is the wave number, k for 3.85 MHz?
2. What is the phase velocity for the original (not changed) coil?
3. What is tau for the original (not changed) coil at 3.85 MHz?

Yes, this is intimidating to ask; but seeing there are so many
authorities on Corum; and that these considerations would have been
done by the authors themselves; then their solutions must reside
somewhere in notes or as marginalia for quick reporting (or could be
summoned up through running through the same math as before).

73's
Richard Clark, KB7QHC

Richard Clark wrote:
I note how little Corrum really has to offer when you had to take the
same:
effective electrical length of the coil (38.83 degrees)
and change it (to the same effective electrical length? I think not.)
to fit the same available wire, at the same specific frequency - only
at a different height along the available wire.

Richard, I explained that phenomenon in a posting last
week which you obviously didn't read. Please go back and
read my posting of 5-9-09 at 1:08pm to this thread.

It is also explained on my web page at:

http://www.w5dxp.com/shrtstub.htm
--
73, Cecil, IEEE, OOTC, http://www.w5dxp.com

Hi Richard,

I wont even attempt to answer the "intimidating" questions - they're
far too tough for me!

But just a couple of comments:

1) The change in coil size when I swapped from a base-loaded to a mid-
loaded model was nothing more than a convenience to reduce the total
number of segments and reduce the computation time. It was not borne
out of any electrical considerations, so please don't read anything
more than that into it. In retrospect it was a silly thing to do
because it has probably introduced a "red herring".

2) You suggest that the Corum method has little utility. However, the
inductance calculator based on the method appears to give usefully
accurate predictions of "equivalent lumped reactance" and SRF (jury
still out on that one). If that calculator was not available, it seems
to me that designing a coil for something like a mobile whip loading
application would require tedious iterations of the helix generator in
EZNEC.

73,
Steve G3TXQ



On May 11, 9:19*pm, Richard Clark wrote:

Hi Steve,

I don't often drop into this side-thread as the topic had drifted into
a stagnated intellectual backwater.

On this and one prior posting by you:

On Sat, 9 May 2009 13:56:31 -0700 (PDT), wrote:
OK, I tried what you suggested. I put my loading coil midway up a 20ft
vertical wire in the EZNEC model. I reduced the number of turns to
lift the resonant frequency to 5.6MHz.

I note how little Corrum really has to offer when you had to take the
same:effective electrical length of the coil (38.83 degrees)

and change it (to the same effective electrical length? *I think not.)
to fit the same available wire, at the same specific frequency - only
at a different height along the available wire.

By my quick read on the stale crisis of current "fall-off" and proving
Corum by EZNEC; it seems quite apparent that EZNEC (the authority) is
driving the coil requirements which are then force fitted by Corum's
inappropriate application.

After all, Corum says nothing of:
1. *Application;
2. *Base loading;
3. *Mid or Top loading;
4. *Stinger selection;
and yet all solutions seem to derive from their math with the elegance
of an ad-hoc "missing degrees" provision (that is quickly discarded as
shown above when current becomes the focus).

Corum DOES say that the formula is only applicable for certain
constraints which I note are NEVER observed in the application nor the
breach. *All of the commentary proceeds through equation (32) when
every argument is an instance of equation (31).

How much are you willing to accept of that paper (which is another way
of asking how much you are willing to discard)?

I will ask one ace-buster question that I expect no one will answer:
* * * * Show me the computation for M (= tau · a)
which would be appropriate for the NON-quarterwave resonance of the
coil in question at 3.85 MHz.

For extra credit:
1. *What is the wave number, k for 3.85 MHz?
2. *What is the phase velocity for the original (not changed) coil?
3. *What is tau for the original (not changed) coil at 3.85 MHz?

Yes, this is intimidating to ask; but seeing there are so many
authorities on Corum; and that these considerations would have been
done by the authors themselves; then their solutions must reside
somewhere in notes or as marginalia for quick reporting (or could be
summoned up through running through the same math as before).

73's
Richard Clark, KB7QHC

Tom,

One further puzzling featu

When I look at the EZNEC currents in the bottom 10ft of my 20ft mid-
loaded model there is *NO* current reduction from bottom to top: 1A at
the bottom and 0.99996A at the junction with the coil. So no evidence
of a cosine shape starting at the bottom.

Brain hurts - time for bed!

73,
Steve G3TXQ

wrote:
When I look at the EZNEC currents in the bottom 10ft of my 20ft mid-
loaded model there is *NO* current reduction from bottom to top: 1A at
the bottom and 0.99996A at the junction with the coil. So no evidence
of a cosine shape starting at the bottom.

In many center-loaded antennas, the current increases
from the feedpoint through the base section to the
bottom of the coil. Converting that non-cosine current
into an equivalent cosine current with the proper
phasing/delay may take some doing. I don't know of
anyone who has accomplished that feat so far. However,
it would be a very useful algorithm.

One way to do such would be to compare the current in
an ideal transmission line with the current through
the loading coil as reported by EZNEC.
--
73, Cecil, IEEE, OOTC, http://www.w5dxp.com

On Mon, 11 May 2009 14:09:29 -0700 (PDT), wrote:

Hi Richard,

I wont even attempt to answer the "intimidating" questions - they're
far too tough for me!

Hi Steve,

That's OK. Even the Corums didn't answer them and it accounts for the
rather thin material being leveraged into the new-age science we get
discussed here.

But just a couple of comments:

1) The change in coil size when I swapped from a base-loaded to a mid-
loaded model was nothing more than a convenience to reduce the total
number of segments and reduce the computation time. It was not borne
out of any electrical considerations, so please don't read anything
more than that into it. In retrospect it was a silly thing to do
because it has probably introduced a "red herring".

It is only a "red herring" if you were intent on mischief. If you
were, you are surrounded by accolytes better versed than you.

However, silly or otherwise, it doesn't answer the intent of the
question. The coil size doesn't change by segment count, but by wire
count, diameter, turns per inch, length. If any of those changed
along with number of segments, then you haven't really done anything
but compared two arbitrary designs to discover they don't match.

What profit in that? (and why did we branch the topic?)

Or they do!

What is to be learnt that this illustrates? (and why did we branch
the topic?)

2) You suggest that the Corum method has little utility. However, the
inductance calculator based on the method appears to give usefully
accurate predictions of "equivalent lumped reactance" and SRF (jury
still out on that one). If that calculator was not available, it seems
to me that designing a coil for something like a mobile whip loading
application would require tedious iterations of the helix generator in
EZNEC.

That is an objection, not a reason, and very far from a discipline
(Corum vs. the world).

What I asked is, if you use Corum (against its provisos) to obtain a
value (fully acknowledged to be erroneous when the provisos are met,
they aren't), for applications that contain considerations that the
Corums do not contemplate, AND you use another tool to validate the
answer - why is it that you wonder on the happenstance of correlation?
This puzzlement is enough to suggest Corum may bring grief
unexpectedly. Does this make the muddy prospects of its utility
clearer?

Still and all, this side-topic is still stuck at the gate to your
buying the farm. Skip the coyness by subscribing to Corum and let
Cecil introduce you to new vistas where these missing degrees will
suddenly emerge again. When that happens, all that is required is
that you suspend your doubt that if that coil at the base of a fixed
height antenna were moved, it would fulfill resonating that fixed
height antenna with the same number of Corum "electrical degrees" in
migration. If Corum "electrical degrees" have to be augmented with
appended theory (diluting the original's importance to elevate the
appendix, as it were); then I would ask again: Does this make the
muddy prospects of its utility clearer?

Yes, Steve, I can full well appreciate that you wouldn't necessarily
expect the same coil to resonate the same, short antenna every where
you happened to place that coil along its fixed length. But this
side-thread isn't going to get any traction until you go with Cecil's
flow (which will undoubtedly swirl into another stagnation at this
point of my observation).

73's
Richard Clark, KB7QHC

Richard Clark wrote:
When that happens, all that is required is
that you suspend your doubt that if that coil at the base of a fixed
height antenna were moved, it would fulfill resonating that fixed
height antenna with the same number of Corum "electrical degrees" in
migration.

Using standard stub theory and transmission lines,
I have shown how moving parts of dual-Z0 stubs from
one place to another requires a change in the length
of parts of the stub.

Why do you have such difficulty applying this standard
transmission line theory to loading coils? Could it be
that you are dismissing technical facts because you
are incapable of understanding them? If so, you have
lots of company down through history.
--
73, Cecil, IEEE, OOTC, http://www.w5dxp.com

Richard Clark wrote:

That's OK. Even the Corums didn't answer them and it accounts for the
rather thin material being leveraged into the new-age science we get
discussed here.


Giving rise to the phrase, "Lack of De-corum".

(cymbal crash)


- Just catching up here...... 73 de Mike N3LI -

On May 11, 2:09*pm, wrote:
Hi Richard,

I wont even attempt to answer the "intimidating" questions - they're
far too tough for me!

But just a couple of comments:

1) The change in coil size when I swapped from a base-loaded to a mid-
loaded model was nothing more than a convenience to reduce the total
number of segments and reduce the computation time. It was not borne
out of any electrical considerations, so please don't read anything
more than that into it. In retrospect it was a silly thing to do
because it has probably introduced a "red herring".

2) You suggest that the Corum method has little utility. However, the
inductance calculator based on the method appears to give usefully
accurate predictions of "equivalent lumped reactance" and SRF (jury
still out on that one). If that calculator was not available, it seems
to me that designing a coil for something like a mobile whip loading
application would require tedious iterations of the helix generator in
EZNEC.

73,
Steve G3TXQ

For what it's worth, I've been using a coil program for quite a few
years now that is able to calculate the performance of a coil based on
a helical transmission line model. It was developed out of travelling
wave tube theory. It turns out I discovered a bug in the program and
reported it to the author, who very kindly corrected it. I've come to
trust it to come up with answers that are very useful in an
engineering sense. I would not expect it to tell me inductance or
other parameters (e.g., first parallel self resonance and first series
self resonance) accurately enough to be used as a precision lab
standard, but that's not what I use the program for.

When I became aware of the HamWaves web page, I was curious about how
well its answers compared with the ones I'd become used to trusting.
They do differ a little, but again, for what I do with them, I trust
them both. Either one will provide results I can use to wind a coil
for a filter and know I won't have to much to adjust the coil to being
"right on." And in fact, I also found a very small bug (or at least
an anomaly or inconsistency) in the HamWaves calculation, and reported
that to Serge, who likewise very graciously acknowledged it and who I
believe corrected it.

So I'd strongly support your thought that the HamWaves calculator
provides useful results. Understand that they won't be perfect, but
also understand that you may have trouble making measurements accurate
enough to know how much they are in error. But for almost everything
I do with coils, what I care about is whether the filter or tank
circuit or antenna in which the coil is used actually works like I
want. My trust in these programs comes from being able to build a lot
of filters over the years that all work like I designed them to work,
with very little effort to tweak the coils I built per the programs'
predictions. I'll adjust my expectations if I ever find cases where
the programs lead me astray.

Cheers,
Tom

K7ITM wrote:
So I'd strongly support your thought that the HamWaves calculator
provides useful results.

So who are we to believe? W8JI's 3 nS delay measurements through
a large 75m loading coil, or the HamWaves 21.5 nS prediction?
--
73, Cecil, IEEE, OOTC, http://www.w5dxp.com