eHam.net - Amateur Radio (Ham Radio) Community

Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net



[Articles Home]  [Add Article]  

Mobile Loading Coil Measurements

Cecil A. Moore (W5DXP) on November 16, 2011
View comments about this article!

75m Mobile Texas Bugcatcher Loading Coil Measurements

Cecil Moore, W5DXP, Rev. 1.3, 11/2/2011

Introduction

The purpose of this article is to present some actual laboratory measurements made on a 75m mobile Texas Bugcatcher loading coil and explain the relationship between the measured phase delay, calculated time delay, and the number of real-world electrical degrees occupied by the loading coil. A number of conclusions will be drawn as this paper proceeds.

Test Setup

The Texas Bugcatcher 75m loading coil has 26 turns of #14 solid wire at four turns per inch with a coil diameter of six inches. W5DXP loaned his 75m Bugcatcher Loading Coil to Louisiana Tech University where EE graduate students ran some simple measurements. The current magnitude and phase at each end of the coil was measured with Tektronix CT2 current probes at 4 MHz under various resistive loads. Here is the test setup (based on an EZNEC graphic).

Measured Phase Shift Results

Keeping the source frequency constant at 4 MHz, for four different load resistances, 50 ohms, 1200 ohms, 1930 ohms, and 3170 ohms, the phase shift between the current into the coil and the current out of the coil was measured.

The following graph shows the phase shift results.

The question immediately arises: Why does the phase shift through a loading coil change with changing load when the frequency is constant?

The answer is that the total current phase shift is the result of the superposition of forward and reflected current. If the current is 100% standing wave current, the phase shift through the loading coil is zero degrees. In free space at 4 MHz, an EM wave travels at 1.44 degrees/nanosecond.[1] But for the loading coil, only if the current is 100% traveling wave current is the current time delay through the loading coil directly proportional to the current phase shift through the coil. An earlier article presents the mathematical difference between standing waves and traveling waves.[2]

What we are looking for is the valid calculated delay through the loading coil. In this paper, we will present the boundary conditions necessary to ensure that the calculated delay through the coil is valid which occurs only when reflected energy is absent.

Conclusion Number One: For any constant frequency, the ratio of the time delay to the phase shift through an air-cored 75m bugcatcher coil is NOT a constant and depends upon the amount of reflected energy present in the measurement setup. Only in a test setup free of reflected energy will the ratio of degrees of signal to phase delay equal 1.44 degrees/nanosecond (at 4 MHz).

How can we guarantee that there is no reflected energy in the coil current measurement circuit? - By terminating the loading coil with its characteristic impedance, just as we would do for a transmission line. If we want to eliminate reflections on a transmission line, we terminate (load) the line with a resistance equal to the characteristic impedance (Z0) of the line. If we load 50 ohm coax with a 50 ohm resistor, we eliminate reflections on the coax and for a low-loss line shorter than 90 degrees, the magnitude of the current-in will almost equal the magnitude of the current-out. The Hamwaves web page contains an inductance calculator that will provide an estimate for the characteristic impedance (Z0) of the loading coil.[3]

Note that if we did not know the Z0 of a piece of coax, we could vary the load resistance until the current magnitude into the coax was equal to the current magnitude out of the coax and that value of load resistance would be a close approximation to the Z0 of the coax. We can do the same thing for the 75m mobile Texas Bugcatcher coil.

For a high-Q coil, that traveling-wave-only condition is very close to being true when the current magnitude into the loading coil is equal to the current magnitude out of the loading coil. There is a small error due to I2R losses in and radiation from the coil but, like lossless transmission lines, we will consider that error to be small enough to ignore during this conceptual discussion. In fact, these measurements indicate that a 75m loading coil can be modeled as a transmission line with a high characteristic impedance (Z0=~1930 ohms) and a very low velocity factor (VF=~0.0193).

Current-In vs Current-Out Measurements

The following graph indicates what happens to the Current-Out/Current-In ratio for the loading coil for the four load resistances used for the measurements.

The point at which the Current-Out/Current-In ratio equals 1.0, i.e. Current-Out=Current-In, indicates the approximate value of the characteristic impedance (Z0) of the coil which is approximately 1930 ohms. From Fig. 2, the phase shift through the coil when the load resistance is 1930 ohms is 41 degrees. This is the only phase shift that will yield valid results in the calculation of the actual delay through the coil and the number of electrical degrees occupied by the coil in a 75m mobile installation.

Note that at 4 MHz the ratio of degrees/nanoseconds = 360deg/250ns = 1.44 deg/ns.[1] Therefore, a phase shift of 41 degrees means a current time delay of 41/1.44 = 28.47 ns (but only for traveling waves).

Conclusion Number Two: Any 75m air-core loading coil must necessarily be in the ballpark of 30-45 electrical degrees long which falls under the classification of a distributed network, NOT a lumped circuit.

What Happens If We Choose a 50 Ohm Load for Our Coil Measurements?

With a 50 ohm load, the total current is primarily standing wave current with an SWR of 1930/50 = ~40:1. The phase shift through the coil measures 4.3 degrees on the 75m mobile Texas Bugcatcher coil. To calculate the apparent "delay", we divide the phase shift by 1.44 deg/ns and get a "delay" of 3 nanoseconds. Of course, 3 ns is not the actual real-world delay through the coil and is an erroneous calculation based on a false concept which is: Under all conditions, the time delay through the coil is related to the phase shift through the coil by some constant multiplier - an obviously false statement as can be inferred from Fig. 1.[4]

Notes and References

[1] In free space, for any fixed frequency, the ratio of (degrees of signal passing a certain point) to (the time it takes for that to happen) is a constant. For HF frequencies, simply divide the 360 degrees (in one cycle) by the number of nanoseconds in a cycle. For instance, 360 degrees divided by the 250 ns in one 4 MHz cycle equals 1.44 deg/ns.

[2] Current Through a 75m Bugcatcher Loading Coil

[3] Analysis of an 80m Base-Loaded Mobile Antenna

[4] That same erroneous 3 ns delay based on that same false concept still exists on a well known ham radio web page as of the date of this article. Exactly the same errors in delay measurements exist there as would happen if we mistakenly took only the measurement data from the Texas Bugcatcher coil using a 50 ohm load. Inductor Current Time Delay

Member Comments:
This article has expired. No more comments may be added.
 
Mobile Loading Coil Measurements  
by AC7ZN on November 16, 2011 Mail this to a friend!
Hi Cecil,

Interesting article. I have no reason to doubt your measurements and conclusions but wonder: what did you do to assure there were no common-mode currents on your source, probes and load during your measurements?

73,

Glenn AC7ZN
 
RE: Mobile Loading Coil Measurements  
by W5DXP on November 16, 2011 Mail this to a friend!
I was not present for the measurements but will ask the question of the supervising professor at Louisiana Tech. My assumption is that, like I^2*R losses in and radiation from the coil, any common mode current was considered to be small compared to the differential currents.
--
73, Cecil, w5dxp.com
 
Mobile Loading Coil Measurements  
by AC7ZN on November 16, 2011 Mail this to a friend!
Thanks for a prompt reply, Cecil. You may be right, but I am worrying about the magnetic field from that large coil inducing currents on nearby conductors, which might be significant when impedances over 1000 Ohms are involved. I imagine there is a somewhat lo-z loop around the coil through the instrument AC grounds and cable shields (depending on their coupling), but maybe it is a mountain from a molehill. An easy way to tell would be to see if the measurements change when cables are moved.


73, Glenn AC7ZN


 
Mobile Loading Coil Measurements  
by KB4QAA on November 16, 2011 Mail this to a friend!
OK, I'll play the rube...

-What practical value is this to me and my mobile mount?

Given that in the real world, we don't have a constant load, but rather use a wander lead or taps to choose the appropriate inductance; essentially backwards of the lab method used?

Bill
 
Something is fishy  
by KASSY on November 16, 2011 Mail this to a friend!
I'm not an engineer, but I diddle some with circuits. I'm sure there's some field and radiation stuff involved in this that I don't know about.

But there's a simple circuit part here that looks wrong.

==The current into one leg of the source must be equal to and opposite the current out of the other leg of the source.

==The current into one leg of the load must be equal to and opposite the current out of the other leg of the load.

===The load and the source are connected to each other, and nothing else, at the line called 'gnd' twice.

====So, the current going into and/or out of the load and/or source must be of the same magnitude and equal or opposite in phase.

And...the source out and load in currents are exactly the same currents at the ends of this coil.

So, I conclude measurement error, since you basically just proved Kirchoff wrong.

But I'm not a circut sage, so I could be wrong, too.

-k
 
RE: Mobile Loading Coil Measurements  
by W5DXP on November 16, 2011 Mail this to a friend!
> KB4QAA wrote: -What practical value is this to me and my mobile mount?

We won't know that answer until we give up on the concept that a 75m air-core loading coil can possibly have a delay of 3 ns.
--
73, Cecil, w5dxp.com
 
RE: Something is fishy  
by W5DXP on November 16, 2011 Mail this to a friend!
> KASSY wrote: So, I conclude measurement error, since you basically just proved Kirchoff wrong.

This is a common conclusion by someone seduced by the lumped circuit model. Dr. Corum explained it better than I ever could:

http://www.classictesla.com/download/corum_lumped_failure.pdf

http://hamwaves.com/antennas/inductance/corum.pdf

Since a 75m air-core mobile loading coil is a distributed network instead of a lumped inductor, the problem you described disappears and the lumped circuit model is invalid for the purpose of analysis. The sum of the currents into and out of a junction is still zero, but the currents some distance away from the junction can be any value without violating Kirchhoff's laws.
--
73, Cecil, w5dxp.com
 
RE: Something is fishy  
by W9AC on November 16, 2011 Mail this to a friend!
> So, I conclude measurement error, since you basically just proved Kirchoff wrong.

Go back and study the current mechanism through the coil. You can't treat the analysis of a standing wave inductor the same as is learned in basic AC theory.

We've all seen the RF current distribution curve of a base-fed 1/4-wave vertical. The base current is always shown greatest at the base, then tapering to the end of the radiator.

Consider a vertical radiator as an infinite series of inductors and distributed parallel C from each node of those inductors to the ground plane. If the vertical is made up of an infinite inductance series, and if each inductance node's In/Out current is supposed to be equal as generally demanded by Kirchhoff in lossless inductors, then why is the current distribution along the radiator tapered? Why, Kirchhoff is being violated an infinite series of times along the radiator -- but only if you incorrectly apply basic AC theory where one current exists. The general Kirchhoff concept is the wrong model for the application, unless you consider all currents.

 
RE: Mobile Loading Coil Measurements  
by KB4QAA on November 16, 2011 Mail this to a friend!
> KB4QAA wrote: -What practical value is this to me and >my mobile mount?

<We won't know that answer until we give up on the <concept that a 75m air-core loading coil can possibly <have a delay of 3 ns.
<--
<73, Cecil, w5dxp.com

So this article is aimed at showing how clever you are and how dim someone else is with no useful information for average hams?

I'm not impressed with power games played out in public forums.
 
Mobile Loading Coil Measurements  
by ON4AA on November 17, 2011 Mail this to a friend!
This is Serge, ON4AA of hamwaves.com
Interesting experiment Cecile. Out of curiosity, I immediately entered the coil values in my inductance calculator to compare with your lab results:
D=154.028mm, N=26, l=165.1mm, d=1.628mm, f=4MHz
I was shocked to see the calculated characteristic impedance (3795.5ohm) to be 200% off with yours (1930ohm). The calculator certainly is not the holy truth, but such a discrepancy would be a first.
However, I rather suspect that there is something wrong with your unconventional method for determining Zc.
Nowhere I read that you also matched your source to Zc.
Anyhow, at the end of the day the measured inductance should shed some light on this situation. If came to accept that inductances are really transmission lines, than L = Zc/omega*tan(beta*l) should yield the inductance. I got 72.9uH.
 
Mobile Loading Coil Measurements  
by ON4AA on November 17, 2011 Mail this to a friend!
As for the frequency dependency, I got this alternative explanation:

A coil can be best seen as a helical waveguide with a kind of helical surface wave propagating along it. The phase propagation velocity of such a helical waveguide is dispersive, meaning it is different for different frequencies. (This is not the case with ordinary transmission lines like coax or open wire.) Lower frequencies propagate slower along a coil. The actual phase velocity at a specific frequency for a specific wave mode is obtained by solving a transcendental eigenvalue equation involving modified Bessel functions of the first (In) and second kind (Kn) for, respectively, the inside and the outside of the helix.
 
RE: Mobile Loading Coil Measurements  
by W5DXP on November 17, 2011 Mail this to a friend!
> KB4QAA wrote: So this article is aimed at showing how clever you are and how dim someone else is with no useful information for average hams?

No, that's not what I said. We need answers to the following questions which will yield useful information for average hams:

1. What is the relationship between loading coil Q and coil losses vs the delay through the loading coil and the Z0 of the loading coil?

2. What is the relationship between mobile antenna gain and the delay through the loading coil?

3. What is the relationship between loading coil efficiency and the Z0 of the coil?

We will never know the answers to those questions as long as we mistakenly believe that the delay through a 75m air-core loading coil is 3 ns or as long as we keep trying to use the lumped-circuit model on a 0.11 wavelength long 75m mobile loading coil.
--
73, Cecil, w5dxp.com
 
RE: Mobile Loading Coil Measurements  
by W5DXP on November 17, 2011 Mail this to a friend!
> ON4AA wrote: I was shocked to see the calculated characteristic impedance (3795.5ohm) to be 200% off with yours (1930ohm).

Yes, I noticed the same thing but was not surprised. Another data point is that EZNEC seems to predict a Z0 of 2571 ohms which is pretty close to the square root of the sum of the squares between these measurements and your calculator results. Obviously, the closer the coil is located to a ground plane, the lower the measured Z0. It doesn't surprise me that a large coil Z0 measured close to a ground plane could be half of the Z0 calculated for free space. In addition, the previous owner of this particular coil spray painted the coil flat black and I know nothing about the chemical composition of the paint.

> However, I rather suspect that there is something wrong with your unconventional method for determining Zc.

Note that Serge is using Zc for the characteristic impedance of the coil while I am using Z0 for that same value.

The method is not unconventional for transmission lines so why would it be unconventional for loading coils that exhibit a characteristic impedance and a velocity factor akin to a transmission line with a high Z0 and a low VF? I pointed out in the article that the method assumed zero losses in the coil which, of course, introduces an error just as it does for transmission lines. I do not know the magnitude of the error because I do not know the losses in the coil.

One of the main points is that using a 50 ohm load on a Z0=2000~4000 ohm coil cannot yield the delay through the coil because the current is mostly standing wave current with a phase shift near zero over each 1/2WL.

> Nowhere I read that you also matched your source to Zc.

I looked at Thevenin equivalent sources with varying source impedances. The source impedance doesn't seem to matter regarding the ratio of the current magnitudes and relative current phases.

> As for the frequency dependency, ...

Note that 4 MHz was the only frequency involved in the measurements. It is the phase shift between the ends of the coil that is dependent upon the load resistance. The variable phase shift is caused by the difference in magnitude between the traveling wave and the standing wave in the measurement circuit.

Total Current = [Imax1*cos(kx)]sin(wt) + Imax2*sin(kx+wt)

The first term is the standing wave current. The second term is the traveling wave current. sin(wt) doesn't change phase with position kx over each 1/2WL. sin(kx+wt) is the only phase shift that matters in the measurement of delay. If Imax1 is an appreciable percentage of Imax2, then the total phase shift cannot be used to measure a valid time delay through the coil.

P.S. Your excellent inductance calculator added another dimension to the loading coil argument as did Dr. Corum's papers.

Another ham engineer has shown interest in duplicating the measurements presented here so please stand by for those results.
--
73, Cecil, w5dxp.com
 
RE: Something is fishy  
by W5DXP on November 17, 2011 Mail this to a friend!
> W9AC wrote: If the vertical is made up of an infinite inductance series, and if each inductance node's In/Out current is supposed to be equal as generally demanded by Kirchhoff in lossless inductors, then why is the current distribution along the radiator tapered?

Here's a graphic that should help:

http://www.w5dxp.com/coilphsh.GIF

There are zero losses in the coil. The forward current magnitude into the coil is equal to the forward current magnitude out of the coil. The reflected current magnitude into the coil is equal to the reflected current magnitude out of the coil. Therefore, Kirchhoff requirements are fulfilled.

The coil is 45 degrees long (1/8 wavelength). Each current component phase is shifted by 45 degrees while the component magnitudes remain constant. The magnitude of the total net current at one end of the coil is 70.7% of the magnitude at the other end of the coil, i.e. a 29.3% current "droop". That magnitude difference is caused, not by losses - nor by radiation - nor by capacitance to ground - nor by displacement current - but has ONLY ONE CAUSE - the traveling wave phase shift through the coil plus superposition of the forward and reverse traveling waves.

Moral: Lumped-circuit thinking will not (and cannot) explain the 29.3% current "droop" in the above ideal (lossless) loading coil.
--
73, Cecil, w5dxp.com
 
RE: Mobile Loading Coil Measurements  
by N3OX on November 17, 2011 Mail this to a friend!
ON4AA writes:

=========
"I was shocked to see the calculated characteristic impedance (3795.5ohm) to be 200% off with yours (1930ohm). The calculator certainly is not the holy truth, but such a discrepancy would be a first. "
=========

Also beta*length should be 33 degrees by the predictions. You say the discrepancy would be a first... do you happen to have some experimental tests of the predictions about the propagation velocity of the waves or know who else has done them? Do those show better agreement?

It seems to me that checking many successive predicted self resonances by feeding one end of the coil and having the other end out in free air is the easiest test. I tried this with a small coil once but I think that particular test was possibly excessively confounded by radiation. There is also possibly the problem of end-effects when the coil is open-ended... at least my comparisons of well-converged self-resonant helices in EZNEC with the model predictions. I don't remember now if I had really considered the role of radiation there, but I think I was testing the smallest coils I could at relatively low frequencies.

I intend on re-testing the series of predicted self-resonances at some point using a coil that would show several HF and or very low VHF self-resonances at a foot or so of physical length but most of the coils I have at the moment are wound on PVC forms or are not sufficiently uniform in pitch to be particularly good for detailed tests of the models.

The sequence of coil self-resonances is a rather non-trivial prediction of the Corums' sheath helix model and it's possibly easier to test than most.

73,
Dan
 
RE: Mobile Loading Coil Measurements  
by N3OX on November 17, 2011 Mail this to a friend!
KB4QAA writes:

==========
"-What practical value is this to me and my mobile mount?
"
==========

A model of how this stuff should work is encoded in ON4AA's calculator

http://hamwaves.com/antennas/inductance.html
(the model is derived in Ref 1 there)

Provided the predictions about wave propagation on coils are accurate, (which I think they are, but the strongest evidence I've seen so far is strange) the model should be useful for several things.

It should tell you the self resonances of different coils that have about the same inductance. In the context of mobile antennas you do probably do not want to operate a coil near or especially above any self resonance.

There are some situations in which the coil self resonances would be useful. You know those little phasing coils in the middle of VHF/UHF mobile antennas? The model in ON4AA's calculator should tell you how to design those. It is in principle possible to design a coil where the current goes IN to both ends at a given instant of time, giving you a way to reverse phase.
You can do that with transmission line stubs too but that's physically more awkward. I suppose the model can be used to design self-resonant coils to be used as traps. That's potentially useful because it's just wire. Things that are just wire are easy to homebrew.

Going out on a limb, I think this could be useful for some narrow-band phase shift networks IF the losses are low enough.

====

I would simply encourage anyone who has a need for any of the above things to figure out how to use ON4AA's calculator to predict the necessary stuff and give it a try in their application.

73,
Dan
 
RE: Mobile Loading Coil Measurements  
by W5DXP on November 17, 2011 Mail this to a friend!
More info: Considering the I^2*R losses in and radiation from the coil to be negligible is based on EZNEC modeling. Using #16 copper wire for the coil, the source power was 25.74 watts and the load power was 25.72 watts indicating that I^2*R losses plus radiation "losses" from the coil amounted to only 0.08% of the total power. The 75m Bugcatcher Coil EZNEC file can be downloaded from:

www.w5dxp.com/bugctchI.EZ
--
73, Cecil, w5dxp.com
 
RE: Mobile Loading Coil Measurements  
by WD5GWY on November 17, 2011 Mail this to a friend!
<quote>KB4QAA wrote:

So this article is aimed at showing how clever you are and how dim someone else is with no useful information for average hams?

I'm not impressed with power games played out in public forums.</quote>
Power games?
What makes you think it's not useful to average hams? If it can inspire someone to do some reading and research to see what is being discussed and LEARN how it can be useful to them, then this discussion in an open forum is the IDEAL place for it.
Part of Amateur Radio is about learning. That is one of the reasons Amateur Radio exists in the first place.
That you don't understand what is being talked about or do not find it useful, doesn't mean that it should not be here where, others who are interested in learning, can have the opportunity to read it and even learn from it.
If it's not of interest to you, don't bother reading it! It's really that simple.
james
WD5GWY
 
RE: Mobile Loading Coil Measurements  
by W5DXP on November 18, 2011 Mail this to a friend!
Instead of responding here, W8JI has chosen to respond on his web page to the [4] reference in my article. When I (hurriedly) wrote Conclusion Number 2, I was thinking in the context of 75m mobile antennas and I inadvertently omitted a few words that have been corrected in Rev. 1.6 on my web page at:

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

Conclusion Number 2 now states: "Any 75m air-core mobile loading coil used with an 8-12 foot mobile antenna must necessarily be in the ballpark of ~30~45 electrical degrees long which falls under the classification of a distributed network, NOT a lumped circuit. Since the VF (velocity factor) of the coil is 0.0193, it is impossible for the 6.5 inch long loading coil to exhibit a 3 ns delay at 4 MHz."

When I made the statement, I was talking in context about 75m MOBILE antennas using a loading coil. The 8~12 foot height limit would mean that the coil must necessarily occupy IN THE BALLPARK OF ~30~45 DEGREES.

I apologize for the inadvertent omission of the context "mobile" and with its inclusion, I stand by my ballpark statement. I want to thank Tom, W8JI for pointing out my typo which indeed did make the original statement false.

As far as W8JI's second objection, I have changed my article to read: "Here is a block diagram of the test setup (based on an EZNEC graphic)."

W8JI says, "... some people just like to argue for the sport of arguing, even where there is nothing useful to be learned or gained."

The Hamwaves inductance calculator, EZNEC results, and Louisiana Tech measurements all disagree with the 3 ns technical information on W8JI's web page. How can W8JI be certain that "there is nothing useful to be learned or gained from a technical discussion"? Wouldn't technical correctness be useful and something learned and gained? I would invite Tom to join in a purely technical discussion here (hopefully free of his "some people" innuendos).
--
73, Cecil, w5dxp.com
 
RE: Mobile Loading Coil Measurements  
by ON4AA on November 18, 2011 Mail this to a friend!
Like many others, I no longer like the tone of this discussion. Neither do I believe that much can be learned or taught here, not for the newbie nor for the more experienced. This is starting to resemble a platonic cave where some chained individuals are discussing the shadows on the wall. Only by annihilation of the ego will truth prevail... Anyhow, I will not leave this cave without leaving a trail for those who would like to be initiated to the understanding that inductors really are helical transmission lines. Lumped circuit models are merely an approximation of Maxwell field theory.

A good point to start reading is the site of David Knight, G3YNH: http://www.g3ynh.info/zdocs/magnetics/index.html

He also developed a more valid way to measure the self-resonant frequencies of a coil: http://www.g3ynh.info/zdocs/magnetics/appendix/self_res/self-res.pdf

Then of course, there is this classical must-read paper: Corum K. L. and Corum J. F., "RF coils, helical resonators and voltage magnification by coherent spatial modes," Microwave Review, IEEE, Vol. 7, No. 2, Sep. 2001, pp. 36-45

Finally, reading the FAQ and source code of your servant's inductance calculator should also be helpful, see: http://hamwaves.com/antennas/inductance.html
 
RE: Mobile Loading Coil Measurements  
by K3GAU on November 18, 2011 Mail this to a friend!
GM All,

Interesting discussion. I would like to point out one little thing about the EZNEC model of the coil. It only uses 8 lines per turn of the coil to represent the coil. With just that many lines per turn the circumference (wire length) of an actual coil would be longer than the sum of the lengths of the 8 wires making up the model effecting the accuracy of the modeled results when compared to an actual physical coil.

I forget what the difference is in percent for 8 wires but at one point some years ago I did calculate the difference and could see it on measurements it with a network analyzer. To help with accuracy you can either increase the number of lines per turn or increase slightly the diameter of the modeled coil. There are formulas available to calculate the difference in cord length verses arc length.
 
RE: Mobile Loading Coil Measurements  
by W5DXP on November 18, 2011 Mail this to a friend!
> ON4AA wrote: I no longer like the tone of this discussion ... Finally, reading the FAQ and source code of your servant's inductance calculator should also be helpful, see: http://hamwaves.com/antennas/inductance.html <

Hi Serge, I don't like the tone either but I didn't originate the tone. The basic question is: Do you, or anyone else besides W8JI, have any technical evidence to support a 3 ns traveling wave delay through a 100T, 10" long, 2" diameter, 100 uh loading coil? Your own inductance calculator indicates a delay of ~22 ns at 4 MHz, seven times the W8JI "measurements". (I apologize for getting emotional about technical facts vs old wives' tales and myths. This argument is almost a decade old and wearing thin.)
--
73, Cecil, w5dxp.com
 
RE: Mobile Loading Coil Measurements  
by W5DXP on November 18, 2011 Mail this to a friend!
> K3GAU wrote: I forget what the difference is in percent for 8 wires but at one point some years ago I did calculate the difference and could see it on measurements it with a network analyzer. <

Molehills and mountains - actually, I am more worried about concepts here than with accuracy. We can draw some conceptual conclusions without being extremely accurate. Even given the inaccuracies, the EZNEC model indicates that a 3 ns delay through a 75m Texas Bugcatcher loading coil is impossible.
--
73, Cecil, w5dxp.com
 
RE: Mobile Loading Coil Measurements  
by W9AC on November 18, 2011 Mail this to a friend!
For any technical discussion that invokes passion about the subject matter, it's almost necessary to officiate the event with a referee with opinions evaluated by qualified peers.

This is nothing new. The tone of this thread is mild when compared to the jabs thrown back & forth between Lee De Forest and Edwin Armstrong. (See Proceedings of the IRE, Volume 3, pp. 239-247). Or, between Ellery Stone and M.B. West over the classic "Power Factor Debate" at the first annual ARRL Convention held in Chicago during September, 1921. You want to see some bitter, published rivalry among the inventive geniuses of the day? Read some of that material.

The IRE had such a forum in the early days for helping to resolve disputes. There was no Internet back then - and no efficient way to disseminate ideas except through a publisher who functioned as referee. What I like about the old IRE Proceedings is that at the end of each article, the discussion is opened up among peers. Sometimes common ground was found. Sometimes not. But at least it fostered a spirit to advance more measurements and even more discussion about a topic - even if the topic became heated. It's no different than taking a position during on a high school debate team, and when necessary, cross-examining the opposing party for the truth. Lawyers use interrogative technique every day to distil the truth from lies and fiction and yet, that "tone" is allowed to survive.

I take each and every technically-based ham website with a "grain of salt." I don't care who it is. Unless the material has been scrutinized by a qualified committee, it may as well be patently false material being offered for its truth.

So, despite the fact we have no referee, the discussion in and of itself still provides through-provoking ideas. If we're not too lazy, we can then reach out and find the truth in peer-reviewed academic literature for ourselves.

Paul, W9AC
 
RE: Mobile Loading Coil Measurements  
by W8JI on November 18, 2011 Mail this to a friend!
Cecil,

This will be my only reply to you. It is my belief you intentionally alter what other people say just to foster arguments. I have no time for that immature nonsense.

This is a direct quote from you. You said, just a few posts earlier:

"How can W8JI be certain that "there is nothing useful to be learned or gained from a technical discussion"? "

What I actually said was this:

"It's my belief that some people just like to argue for the sport of arguing, even where there is nothing useful to be learned or gained. To enjoy the sport of arguing, they sometimes find it necessary to alter what other people say. "

So we can all see, once again, you somehow feel it necessary to make something up and post it, claiming I said something I did not say and presenting it as a quote from me.

It is my belief that you do this just to bait or draw people into arguments, or to make people look bad. It is very dishonest to say someone said something, presenting it as a quote, when they actually never said it.

That is a terrible habit.

73 Tom
 
RE: Mobile Loading Coil Measurements  
by W5DXP on November 18, 2011 Mail this to a friend!
> W9AC wrote: So, despite the fact we have no referee, the discussion in and of itself still provides thought-provoking ideas. <

Paul, I'm all for a referee - want the job? :)

The question is: Is it possible for a 100T, 10" long, 2"dia 75m mobile loading coil to exhibit a 4 MHz EM traveling-wave delay of 3 ns? If anyone has such evidence, now would be a good time to present it.
--
73, Cecil, w5dxp.com
 
RE: Mobile Loading Coil Measurements  
by W9AC on November 18, 2011 Mail this to a friend!
Paul, I'm all for a referee - want the job? :)

Would love it, except I'm not qualified! Being an "engineer turned lawyer," I doubt folks would like my format.

Perhaps solving the issue through the "Socratic Method?" :-)

Paul, W9AC
 
RE: Mobile Loading Coil Measurements  
by W5DXP on November 18, 2011 Mail this to a friend!
> W8JI wrote: This is a direct quote from you. You said, just a few posts earlier:

"How can W8JI be certain that "there is nothing useful to be learned or gained from a technical discussion"? "

What I actually said was this:

"It's my belief that some people just like to argue for the sport of arguing, even where there is nothing useful to be learned or gained. To enjoy the sport of arguing, they sometimes find it necessary to alter what other people say. " <

Tom, once again you concentrate on nick-picking a typo by an old man instead of addressing the technical facts. (I apologize for my palsy/macular-degeneration reading/typing problems.) I simply put the close-quote sign in the wrong place and, as you know, once you hit the "post" button, things cannot be changed. What I meant to post was: How can W8JI be certain that "there is nothing useful to be learned or gained" from a technical discussion?

Please accept my humble apologies for my age-related reading/typing problems.
--
73, Cecil, w5dxp.com
 
RE: Mobile Loading Coil Measurements  
by K3GAU on November 18, 2011 Mail this to a friend!
Cecil,

After all this is cussed and discussed
it all boils down to a couple of questions.

1. What is the 'take away' from this discussion suppose to be?

2. Given that we accept your explanation of things, can you give us any guidelines, rules, equations, graphs, etc. that will help us to design and/or build the most efficent mobile antennas possible given a certain set of constraints??

If this discussion does not advance the 'state of the art', then it is purely academic and not overly useful to most folks.

Have a good day.
 
RE: Mobile Loading Coil Measurements  
by W5DXP on November 18, 2011 Mail this to a friend!
> K3GAU wrote: After all this is cussed and discussed
it all boils down to a couple of questions. What is the 'take away' from this discussion supposed to be? <

1. A 75m mobile air-core loading coil is NOT a "small loading inductor" so forget anything based on that lumped-circuit concept. Comprehend that a distributed network analysis MUST be used for these large air-core loading coils to ascertain what is happening in the real world because *a lumped-circuit model gives wrong answers*. Then answer the following questions:

2. What is the relationship between the velocity factor of the loading coil and the radiation efficiency of the antenna system?

3. What is the relationship between the characteristic impedance of the loading coil and the radiation efficiency of the antenna system?

4. What is the relationship between the electrical length of the loading coil and the radiation efficiency of the antenna system?

There are no doubt many other questions to be asked and answered. And there is the satisfaction of knowing that yet one more old-wives'-tale-myth within ham radio has been laid to rest.
--
73, Cecil, w5dxp.com



 
RE: Mobile Loading Coil Measurements  
by W8JI on November 18, 2011 Mail this to a friend!
Cecil Moore,

You clearly attributed something I never said to me. It was not a misplaced quote, it is entirely creating a new statement. None of the words you attributed to me inside quotes were my words.

This is not unusual for you. You do the very same thing with technical content. For example, here is a direct quote from the page you link to. This is, unedited, what I wrote:

"Current at BOTH ends of inductor, when there is no shunting capacitance present to increase phase shift, lags voltage at the source-end by a value that depends on system termination impedance and the inductor's reactance!"

You change the termination impedance, change the test fixture, change the coil, change lead lengths, get a different result, and want me to admit a calibrated measurement I made is wrong, because with all those radical changes, if you even really had those measurements made, the results changed!

I clearly said, if stray capacitance is changed, or termination is changed, the delay changes!

I can change the delay hundreds of percent just by placing my hand near the coil, and move current all over the place.

Neither your alleged test setup by these unknown "students", nor my test setup, is anything remotely like the termination or environment offered by a short mobile whip antenna, where the coil is in the air and terminated by a whip antenna far from any groundplane or test leads.

I state the system changes, and never imply it is a representation of a mobile antenna. You say your test represents every loading coil in the world on 75 meters! Logically, I don't think anyone in the world would think a resistor terminating a coil on a bench is the same as the same coil in the air terminated in a whip antenna.

I would never make an all encompassing statement that every 75 meter loading coil in the world has the same characteristics on a bench over a groundplane terminated in a resistor, as it would have in open air on a vehicle terminated in a whip antenna. I don't think many people with common sense would buy into your universal rule, that all 75 meter loading coils are the same.

Everyone in the world who has ever had a mobile antenna on lower bands knows any overhead wire, dozens of feet above the tip detunes the antenna, often so badly the signal virtually disappears.

For some reason you have taken on the role of the Energizer Bunny, beating a drum that you can measure and document what happens in the antenna by throwing a resistor on the end of a loading coil in a room on a bench, and that your supposed students can magically show what happens in an antenna system on a test bench.

Give everyone a break.

What this is really about is you are promoting you standing wave analysis method, that actually disagrees or deviates worse than a lumped model does, as the only way to do things.

I don't particularly care if you have some uncontrollable fascination with your reflected waves, so long as you stop writing ridiculous text, placing quotes around it, and telling people in public it is something I said. Let's not play sad old nearly blind man, and claim the practice of making up whole long strings of words and quoting them in someone else's name, is caused by your failing old eyes.

Please stop making up long strings of words, publicly posting it inside quotes, and saying it is from me or that it is something I said. We all know that cannot from poor vision.

There is no reason to engage anyone who makes up statements, and attributes them to others in public.

Please, stop making up statements and posting in public that they came from me. That is all I ask.
 
RE: Mobile Loading Coil Measurements  
by W5DXP on November 18, 2011 Mail this to a friend!
> W8JI wrote: None of the words you attributed to me inside quotes were my words. <

I am going to ignore your false statements and limit my response to the one technical subject about which you seem to be confused and that is the relationship between measured phase shift and calculated time delay.

> I clearly said, if ... termination is changed, the delay changes! <

Well Tom, you are just wrong about that. The forward traveling-wave delay through a fixed coil configuration at a fixed frequency is constant with changing load just as it is for transmission lines.

FOR YOU TO BE CORRECT, THE FORWARD-TRAVELING WAVE WOULD HAVE TO KNOW WHAT THE LOAD IS BEFORE IT ENCOUNTERS THE LOAD, so please explain how that is possible.

It is the phase shift between the ends of the coil that changes with load, not the delay time. The phase shift you measured is accurate. The conclusion about delay time based on the phase shift is absolutely incorrect. Phase shift and delay time are unrelated unless reflections are eliminated. A phase shift of 4.3 degrees at 4 MHz equates to a 3 ns delay only for a traveling wave. A phase shift of 4.3 degrees does NOT equate to a 3 ns delay when reflections are present. You don't seem to comprehend that fact of physics.

Every RF engineer will tell you that a 0.1 wavelength transmission line requires a distributed network analysis because a lumped circuit analysis fails to yield the correct results. That is just as true of a 0.1 WL loading coil as it is for a 0.1 WL transmission line.

Consider a 0.1 wavelength piece of 450 ohm line on 4 MHz. It obviously has a 25 ns delay. If you terminate that line with a 10 ohm resistor and "measure" the delay like you did with your loading coil, guess what delay value you will get? Yep, 3 ns. Does that mean that signals are traveling through the 0.1 WL feedline at faster than the speed of light? Of course not.
--
73, Cecil, w5dxp.com
 
60-meter band breaking news!  
by N4QA on November 19, 2011 Mail this to a friend!
19 Nov 2011

CW comes to 60 meters ! ...other modes too...

Thanks to Cory, WA3UVV, for posting this on QRP-L this morning:
http://www.fcc.gov/document/amateur-radio-service-5-mhz

72,
Bill, N4QA
 
RE: 60-meter band breaking news!  
by N3OX on November 19, 2011 Mail this to a friend!
K3GAU writes:

==========
"After all this is cussed and discussed
it all boils down to a couple of questions.

1. What is the 'take away' from this discussion suppose to be? "
==========

W8JI sux, man. Didn't you get the memo?


========
"2. Given that we accept your explanation of things, can you give us any guidelines, rules, equations, graphs, etc. that will help us to design and/or build the most efficent mobile antennas possible given a certain set of constraints??"
========

I think there are some things that could be helped by quantitative prediction of the transmission line behavior of coils. I think that it is probably correct that some loading coils in use in ham radio need this treatment, and certainly helical antennas do. What I question is how four sparsely spaced data points that badly disagree with the relevant theoretical models are supposed to advance our understanding of that.

I think ON4AA's calculator is actually quantitatively correct. I think it BECAUSE of W8JI's group delay measurement on his inductor time delay page.

========
"If this discussion does not advance the 'state of the art', then it is purely academic and not overly useful to most folks."
========

It's worse than that. I think the confrontational aspects of this discussion have actively hindered advancement of hams' understanding of loading coils. Listen to ON4AA on this: the topic is interesting but there is little to be learned here, in this forum discussion or in the historical forum discussions, except by following trails away from them. Read the references on ON4AA's page. If you're really curious set up some experiments.

In the early days of electromagnetic theory, if you wanted to check the standing wave pattern on a transmission line, you would just excite it with a strong field and bring a neon or other glow tube near it. That tube would light up at the field maxima.

Coils with big ham radio loading coil dimensions are predicted to show a variation of fields with a wavelengths of a few inches to a few yards over HF frequencies. ON4AA's calculator can be used to predict this wavelength. It is not completely trivial to set up several feet to several yards of nice, uniform coil stock to demonstrate the standing waves. But wouldn't that be a nice demonstration that the model was correct?

I personally think the model is correct but from a practical perspective you likely want to use ON4AA's calculator to design coils that have the LONGEST wavelength possible. That is, you want the coils that are closest to lumped as possible and show the least variation in current along the coil.

The engineering advice in terms of short monopoles is to AVOID any significant delay as much as possible. The predictions do suggest that 80m loading coils for mobile antennas will have some inevitable variation from slow waves on the coil but I am confident that goal is to have as little variation as possible. The model IS potentially useful for this, but I doubt it is useful enough to make it worth it in most cases.

For designing HELICAL short monopoles the transmission line behavior is potentially critical because it is possible to get phase reversals of the current on a long coil, which causes parts of the antenna to radiate out of phase from other parts in a short distance!

But unfortunately the presence of a dielectric stick on which the antenna is wound is not yet handled by ON4AA's calculator. And one way to avoid building a crappy helical antenna is to avoid building a helical antenna at all.

From an engineering perspective for short antennas, my advice would be "don't use antennas that fundamentally require transmission line models for coils in their design"

Of course understanding the transmission line behavior is the only way to always always avoid that, but most sound and successful empirically derived practical advice about inductors in mobile antennas will push you away from "electrically long" coils.

73,
Dan







 
RE: 60-meter band breaking news!  
by W5DXP on November 19, 2011 Mail this to a friend!
> N3OX wrote: From an engineering perspective for short antennas, my advice would be "don't use antennas that fundamentally require transmission line models for coils in their design" <

Since the Hamwaves inductance calculator, EZNEC, the Louisiana Tech measurements, and the Dr. Corum papers all agree that the average 75m air-core mobile loading coil is approximately 0.1 wavelength long, it is difficult to avoid analyzing such a coil using a distributed network model. Note the phase shift through a 0.1 wavelength long loading coil is 36 degrees.

Why isn't a lumped-circuit analysis valid for such a coil? Dr. Corum answers that question in his papers:

http://hamwaves.com/antennas/inductance/corum.pdf

http://www.classictesla.com/download/corum_lumped_failure.pdf

Dr. Corum - "Lumped element representations for coils require that the current is uniformly distributed along the coil - no wave interference and no standing waves can be present on lumped elements."

Dr. Corum - "... lumped-element circuit theory inherently employs the cosmological presupposition that the speed of light is infinite" Therefore, 3 ns "measurements" for 75m loading coils may seem reasonable to some people.

The application of the distributed network model to 75m loading coils indicates an electrical length of approximately 0.1 wavelength. An accurate analysis of an electrical 0.1 wavelength device is not possible with lumped circuit theory since such a device is not recognized by that theory.

One other thing to remember. If there exists a disagreement between a lumped-circuit analysis and a valid distributed network analysis, the distributed network analysis wins every time because it is closer to Maxwell's equations.
--
73, Cecil, w5dxp.com



 
RE: 60-meter band breaking news!  
by KG6WLS on November 19, 2011 Mail this to a friend!
Wow!! It's been awhile since I stopped by in here. Nothing has changed much. Well, back to the radio and off the computer. Carry on, gents!!
 
RE: 60-meter band breaking news!  
by W5DXP on November 20, 2011 Mail this to a friend!
Here's a "Socratic Method" question for anyone who believes that a 75m mobile Texas Bugcatcher coil is a "small loading inductor" for which a lumped-circuit analysis will yield valid results.

From the 50 ohm load measurements in Fig. 1, the current at the source end of the coil is 140 ma and the current at the load end is 220 ma, 57% higher than at the source end.

Using the lumped-circuit model which presumes equal current magnitude/phase throughout the inductor and a coil delay of zero, from what energy source does that 57% of additional current come?
--
Using the distributed network model, it's simple and easy to explain. A standing wave exists along the length of the inductor. The standing wave envelope is known to be a sinusoidal function of (kx) where (x) is the coil length parameter and (k) is the constant that converts (x) to radians or degrees (with no shunting capacitance required).

Often, the simple and easy explanation is the correct one.
--
73, Cecil, w5dxp.com
 
RE: Loading coils  
by W8JI on November 20, 2011 Mail this to a friend!
RE:N3OX on November 19, 2011
"I think there are some things that could be helped by quantitative prediction of the transmission line behavior of coils. I think that it is probably correct that some loading coils in use in ham radio need this treatment, and certainly helical antennas do. What I question is how four sparsely spaced data points that badly disagree with the relevant theoretical models are supposed to advance our understanding of that."

As do I. IMO one of the largest setbacks to the average person understanding what the loading coil does, and how to optimize it, come from articles like this one.

Clearly a loading coil totally stretched out into a line behaves 100% like a transmission line. That's how dipoles, Marconi antennas, and almost all other antennas develop the current distribution they have.

At the other end of the scale a compact inductor with very tight flux coupling, and minimal displacement current to the outside, emulates a two terminal device with no measurable or observable transmission line characteristics.

The worse the loading system becomes, the more it behaves like a transmission line. No magic there, and that is something I have always said.

N3OX:
"I think ON4AA's calculator is actually quantitatively correct. I think it BECAUSE of W8JI's group delay measurement on his inductor time delay page."

That calculator appears to agree with my measurements. Oddly, Cecil uses it to dispute my measurements.

Someone said:
"If this discussion does not advance the 'state of the art', then it is purely academic and not overly useful to most folks."
========

N3OX said:
"It's worse than that. I think the confrontational aspects of this discussion have actively hindered advancement of hams' understanding of loading coils."

I agree. I have no use for anyone who appears to only want to argue endlessly, while intentionally misquoting people just to bait them out. All that does is confuse people and prevent any technical discussion.

N3OX:
" Listen to ON4AA on this: the topic is interesting but there is little to be learned here, in this forum discussion or in the historical forum discussions, except by following trails away from them. Read the references on ON4AA's page. If you're really curious set up some experiments."

Good advice.

N3OX:
"I personally think the model is correct but from a practical perspective you likely want to use ON4AA's calculator to design coils that have the LONGEST wavelength possible. That is, you want the coils that are closest to lumped as possible and show the least variation in current along the coil."

That's correct. The only gross errors I see are Cecil's statements that a certain delay is universally applied to all cases, and an inductor behaves like a characteristic transmission line for all external conditions and all inductor sizes.

A transmission line almost always makes a very poor loading system. It is always worse than a lumped component behavior by better system design.

N3OX says:
"The engineering advice in terms of short monopoles is to AVOID any significant delay as much as possible. The predictions do suggest that 80m loading coils for mobile antennas will have some inevitable variation from slow waves on the coil but I am confident that goal is to have as little variation as possible. The model IS potentially useful for this, but I doubt it is useful enough to make it worth it in most cases."

That's right. When the model applies, the system is poor. Follow the widely understood rules, minimizing displacement currents that rob the upper areas of the antenna of current, and the system gets better.

N3OX:
"For designing HELICAL short monopoles the transmission line behavior is potentially critical because it is possible to get phase reversals of the current on a long coil, which causes parts of the antenna to radiate out of phase from other parts in a short distance!"

I measured some CB antennas like that. They intentionally wound extra wire into the antenna to improve FS, but actually added out-of-phase sections!

N3OX:
"But unfortunately the presence of a dielectric stick on which the antenna is wound is not yet handled by ON4AA's calculator. And one way to avoid building a crappy helical antenna is to avoid building a helical antenna at all."

I can't think of a case where a helical antenna is a good electrical performance idea for loading an antenna.

The more the loading coil acts like a transmission line of any appreciable length, the worse the system is.

N3OX:
"From an engineering perspective for short antennas, my advice would be "don't use antennas that fundamentally require transmission line models for coils in their design"

Of course understanding the transmission line behavior is the only way to always always avoid that, but most sound and successful empirically derived practical advice about inductors in mobile antennas will push you away from "electrically long" coils. "

That's right. I can build a delay line. TV set manufacturers did that to bring color and video in phase because of delays through the more complex color processing circuits compared to the more direct video systems.

Delay coils, that behaved like transmission lines, were long small diameter coils wound over a ground plane, or parallel to a groundplane conductor. They behaved like a transmission line and very little like a lumped component.

As we make a loading inductor longer and smaller diameter, reducing mutual coupling from turn to turn, and as we get it close to a groundplane like the sheet metal of a vehicle, we make it more and more like a transmission line. Doing bad things makes it electrically appear more and more like a transmission line, or delay line in a color TV. :-)

The very best systems I have measured have the least of these errors, and behave more as a lumped component.

IMO, the biggest single mistake anyone can make is to think a loading inductor always acts like a transmission line, or like a missing number of electrical degrees. Such thoughts can easily be proven wrong.

This article, unfortunately, does a good job of setting back understanding of loading systems.

Thank you, Dan, for bringing some rational common sense into this topic.

73 Tom
 
RE: Loading coils  
by W5DXP on November 20, 2011 Mail this to a friend!
> W8JI wrote: IMO one of the largest setbacks to the average person understanding what the loading coil does, and how to optimize it, come from articles like this one. <

Your or my opinions simply do not matter unless they are technically correct. So far, you have neglected to answer my technical questions - one wonders why.

> W8JI wrote: At the other end of the scale a compact inductor with very tight flux coupling, and minimal displacement current to the outside, emulates a two terminal device with no measurable or observable transmission line characteristics. <

That's true but a 75m Texas Bugcatcher coil or your 100T, 100uh test coil doesn't come anywhere close to satisfying those requirements. All hams need to recognize the point at which they must discard the simplistic lumped-circuit model and adopt the distributed-network model. IMO, it is obvious that you do not know where that point is.

> W8JI wrote: That (Hamwaves) calculator appears to agree with my measurements. Oddly, Cecil uses it to dispute my measurements. <

A false statement! I don't dispute your measurements. In fact, I have previously said that the Louisiana Tech measurements using a 50 ohm load *duplicated your measurements*. Please stop misquoting me. It is your conclusions about your measurements that are incorrect.

That the Hamwaves calculator agrees with your group/phase shifts/delays is not important because the group/phase shifts/delays are irrelevant in standing wave systems and do not represent energy transfer through the length of the coil. In fact, a pure standing wave with a group delay of zero, doesn't transfer any energy. It is impossible to transfer EM energy through your test coil in 3 ns. If you don't believe that, try transferring just one bit of information through it in 3 ns.

Where the Hamwaves calculator disagrees with your measurements is in beta, the axial propagation factor in radians/meter, related to the VF of the coil, and easily converted to degrees per inch by multiplying by 1.455. The beta for your coil is 2.122 rad/m. Multiplying by 1.455 yields 3.088 degrees per inch. Since the coil is 10 inches long, the degrees occupied by your coil at 4 MHz is 30.88 degrees which gives a traveling wave delay through your coil of 30.88/1.44 = 21.4 ns, not the 3 ns group delay that you measured. Thus, the Hamwaves calculator disagrees with your 3ns conclusions, not with your 4.3 degree group phase shift (delay). The Hamwaves calculator indicates that your coil has a velocity factor (VF) of 0.04. Transferring EM energy through a coil with a VF of 0.04 in 3 ns is impossible and violates the speed of light limit.

Maybe you should refresh your memory on Group/Phase Shifts/Delays in a standing wave environment.

http://www.microwaves101.com/encyclopedia/waveguidemath.cfm

> W8JI wrote: That's correct. The only gross errors I see are Cecil's statements that a certain delay is universally applied to all cases, and an inductor behaves like a characteristic transmission line for all external conditions and all inductor sizes. <

There you go again, Tom, deliberately misquoting me. I stand by the following statement. If you have any technical evidence to the contrary, now would be a good time to present it. Please note that I am talking only about 75m air-core mobile loading coils, not "all cases or all inductor sizes universally applied". If you are technically correct, why do you need to engage in unethical nonsense? Why not just take the time to engage me and prove me wrong? I would even agree to an objective proof reader referee for our discussions. Here's what I said:

"Any 75m air-core mobile loading coil used with an 8-12 foot mobile antenna must necessarily be in the ballpark of ~30~45 electrical degrees long which falls under the classification of a distributed network, NOT a lumped circuit. Since the VF (velocity factor) of the coil is 0.0193, it is impossible for the 6.5 inch long Texas Bugcatcher loading coil to exhibit a 3 ns delay at 4 MHz."

> W8JI wrote: IMO, the biggest single mistake anyone can make is to think a loading inductor always acts like a transmission line, or like a missing number of electrical degrees. Such thoughts can easily be proven wrong. <

Again, you are implying something I never said. A 75m air-core mobile loading coil certainly does act like a transmission line and contributes a number of missing electrical degrees to a 75m mobile antenna system. It is impossible to prove that wrong because it is technically correct.

Tom, once more I invite you to engage in a purely technical rational discussion free from the falsehoods like you posted in this thread. If you will engage me on a technical level, and stop all the unethical false accusations, I think I can convince you that I am right about 75m mobile air-core loading coils. A good start on your part would be to respond to the purely technical questions in my previous postings above.

OTOH, if you are right and I am wrong, seems you would enjoy engaging in and winning the argument. If you don't have anything to hide, how about engaging in a purely rational technical discussion for a change? Just remember if you use the lumped-circuit model correctly and I use the distributed-network model correctly and our outcomes differ, the distributed network model will win and I'm betting that will be the case.
--
73, Cecil, w5dxp.com
 
RE: 60-meter band breaking news!  
by W5DXP on November 20, 2011 Mail this to a friend!
> N3OX wrote: What I question is how four sparsely spaced data points that badly disagree with the relevant theoretical models are supposed to advance our understanding of that. <

Dan, why do you think they "badly disagree with the relevant theoretical models"? I have processed them through the distributed network model and although not extremely accurate, they agree with the distributed network model. If you are saying that they disagree with the lumped-circuit model, then I agree completely because the lumped-circuit model is incapable of handling a device that is 10% of a wavelength long.

For instance, with a Z0 of 1930, the reflection coefficients from 50 ohms, and 1200 ohms will be positive so the current at the load side of the coil has a higher amplitude than the current at the source side of the coil. The reflection coefficient inflection point is when the load equals 1930 ohms and there are zero reflections. Above a 1930 ohm load, the reflection coefficient is negative and the current on the source side is higher than the current on the load side, just a predicted by the reflection model. We can argue about accuracy but I don't see any qualitative conceptual disagreement at all.
--
73, Cecil, w5dxp.com
 
RE: Mobile Loading Coil Measurements  
by K8KAS on November 20, 2011 Mail this to a friend!
What a bunch of BS, what does Joe Ham get out of this posting, how helpful is this to the performance of a loading coil and the factors that effect the performance of the antenna? I guess it is to prove what an academic Cecil is or was. 3OX will like this one for sure...73 Denny K8KAS
 
RE: Mobile Loading Coil Measurements  
by WB4JZY on November 20, 2011 Mail this to a friend!
"What a bunch of BS, what does Joe Ham get out of this posting"................

if your joe ham is not intellectually curious, nothing.
he should then move along and not criticize those trying to learn something.
 
RE: Mobile Loading Coil Measurements  
by W5DXP on November 20, 2011 Mail this to a friend!
> K8KAS wrote: What a bunch of BS, ...

Someone suggested a solution for that problem earlier. If you don't like it, ignore it, forget it, and find a more pleasurable activity in which to engage.

The number of degrees occupied by a loading coil is important because it might have an effect on performance.
--
73, Cecil, w5dxp.com
 
RE: Mobile Loading Coil Measurements  
by N3OX on November 20, 2011 Mail this to a friend!
K8KAS writes:

=======
"3OX will like this one for sure..."
=======

Nope.
 
RE: Loading coils  
by N3OX on November 20, 2011 Mail this to a friend!
W8JI writes:

=========
"That calculator appears to agree with my measurements. Oddly, Cecil uses it to dispute my measurements.
"
=========

I think he uses it to dispute your interpretation and I do not necessarily disagree with him. I think it is possible that your coil, in isolation, when terminated in about 4700 ohms, would act as a delay line at 4MHz about 26 degrees long. That is the prediction of ON4AA's calculator, and that calculator predicts the self-resonances of your coil well. It uses a model to do that that should apply well for a lot of frequencies below self resonance. Of course how you set up "in isolation and terminated" is tough, but "in isolation" really just depends on the extent of the fields around the coil that are responsible for the delay line behavior compared to the perturbing stuff.

At some point that specific model will break down because the way the derivation is set up won't work for really really short coils. I don't really know what "too short" is but I think maybe L/D = 5 like your coil is NOT too short.

The idea of terminating the coil on the bench in the characteristic impedance predicted by ON4AA's calculator and doing experiments on it has merit.

W5DXP writes:

=========
"Using the distributed network model, it's simple and easy to explain. A standing wave exists along the length of the inductor. The standing wave envelope is known to be a sinusoidal function of (kx) "
=========

But does that standing wave envelope exist on your particular coil in your particular test conditions or not? If your inferred characteristic impedance and electrical length were closer to the predictions of the Corums' model and ON4AA's additions I wouldn't be so skeptical... because it would be hard to explain those SPECIFICS any other way.

The thing is that your measured results and their quantitative disagreement with the Corums model and with EZNEC make it more likely that alternative explanations are correct, like the coil MUST be coated with dielectric or the coil MUST be near the ground plane for the transmission line effects to be observed on your coil. It could also be that the coil fields themselves cause the delay, but that's supported better by quantitative agreement or direct measurement of the field pattern along the whole length, not just at the ends.
 
RE: Loading coils  
by W8JI on November 21, 2011 Mail this to a friend!
Hi Dan,

N3OX:
"I think he uses it to dispute your interpretation and I do not necessarily disagree with him."

If he intends that, he should say that. To me, he sounds like he rewrites everything everyone else says just to argue that every system in the work can be analyzed as a transmission line. He even tries to do tank systems, and non-linear sources in amplifiers, as transmission lines with reflected waves.

I'm not sure what motivates someone to over-complicate the analysis of the system, especially when their data does not mesh with the real world and they make statements that cannot be supported, like "every 75 meter loading coil acts like this length of transmission line", when the test setup shows it does not.

My only point was people who think, at radio frequencies, xx feet of wire coiled in a coil behaves like the same xx feet of wire stretched out, or that a loading coil acts like the missing electrical degrees of an antenna, as thinking the wrong way.

If I failed to make that point, I am interested in where I failed. If you see something that does not help with that point, let me know.

N3OX:
" I think it is possible that your coil, in isolation, when terminated in about 4700 ohms, would act as a delay line at 4MHz about 26 degrees long. "

So? I miss your point. I don't believe I said anything one way or the other about that. I said the behavior changes with termination and layout changes.

N3OX:
"That is the prediction of ON4AA's calculator, and that calculator predicts the self-resonances of your coil well. It uses a model to do that that should apply well for a lot of frequencies below self resonance. Of course how you set up "in isolation and terminated" is tough, but "in isolation" really just depends on the extent of the fields around the coil that are responsible for the delay line behavior compared to the perturbing stuff."

That's reasonable. Every distributed system or component has some transmission line effects. Electrically short mobile antennas are a particularly difficult thing, because reactive impedances are so terribly high. The distributed capacitance along a loading inductor can approach or exceed the termination impedance of the inductor, in which case displacement currents can rob the upper area of the antenna from current.

The system to generally avoid is the system where the loading inductor is behaving like a transmission line. That's why we generally keep loading coils compact, and away from sheet metal.

Explaining things like this would be useful and helpful to other Hams, and would be interesting.

N3OX:
"At some point that specific model will break down because the way the derivation is set up won't work for really really short coils. I don't really know what "too short" is but I think maybe L/D = 5 like your coil is NOT too short."

I'm sure there could be a set of approximations developed that might be useful, helpful, and interesting.

I measured no delay or current taper at all in compact inductors.

Have fun with the Texas troll. :-)

73 Tom
 
RE: Loading coils  
by W5DXP on November 21, 2011 Mail this to a friend!
>> W5DXP writes:
=========
"Using the distributed network model, it's simple and easy to explain. A standing wave exists along the length of the inductor. The standing wave envelope is known to be a sinusoidal function of (kx) "
=========

Thanks Dan, seems you have been drafted as the referee.:)

> N3OX wrote: But does that standing wave envelope exist on your particular coil in your particular test conditions or not? <

If the coil is 41 degrees long at 4 MHz, we would certainly expect the standing wave envelope to develop.

For the 50 ohm load, the current at the source end was 140 ma and the current at the load end was 220 ma. Is it likely that extra 80 ma of current at the load is created out of thin air? Isn't it more likely that the forward current and reflected current at the 50 ohm load are in phase and superpose to 220 ma because the current reflection coefficient is positive? Isn't it more likely that the traveling wave phase shift through the coil causes the forward wave and reflected wave to be some number of degrees out of phase on the source end of the coil so that they superpose to 140 ma. Here is a conceptual graphic that illustrates a current magnitude droop purely from forward and reverse current superposition. The system is completely lossless, i.e. no displacement current, no radiation, no I^2*R losses, etc. All that exists are equal amplitudes of forward and reflected current.

http://www.w5dxp.com/coilphsh.GIF

That same pattern occurs along transmission lines with standing waves and on standing wave antennas. Why is it hard to accept that it can happen in a very large loading coil installed in a standing wave antenna?

> N3OX wrote: ... like the coil MUST be coated with dielectric or the coil MUST be near the ground plane for the transmission line effects to be observed on your coil. <

The coil was spray painted with flat black paint by the previous owner and I don't know the composition of the paint. I also don't know how close to ground the coil was for the tests. Perhaps those things explain the quantitative differences with the Hamwaves calculator. At the present time, I am more interested in the qualitative differences.

The number of degrees occupied by the Texas Bugcatcher coil at 4 MHz varies as follows:

1. W8JI's measurement method using a 50 ohm load = 4.3 degrees of *group delay*.

2. EZNEC simulation with a 2571 ohm load = 42 degrees of traveling wave delay.

3. Louisiana Tech method with a 1930 ohm load = 41 degrees of traveling wave delay.

4. Hamwaves calculator = 33.4 degrees calculated from the axial propagation factor.

Even the lowest Hamwaves value is 7.7 times the group delay measured by w8ji.

I've asked the following question before and got no answer: Given a 1/4WL helical monopole, it is obvious that a standing wave exists on that standing wave antenna and the standing wave current follows the classic cosine distribution.

If we cut the helical in half, we could use 1/2 of the helical as a loading coil with a straight length of conductor on top. Does that straight length of conductor on top prohibit the lower helical half of the antenna from exhibiting standing waves?

You can see where such a discussion would lead. Exactly when does part of a helical monopole abruptly stop acting like a transmission line and start acting like a lumped inductor? IMO, it just doesn't happen for 75m mobile air-core loading coils. Dr. Corum says the lumped-circuit model ceases to be valid for phase and delay questions at an electrical length of 15 degrees. According to the Hamwaves calculator, EZNEC, and the Louisiana Tech measurements, a 75m air-core mobile loading coil is 2~3 times longer than 15 degrees.

Thanks again, Dan, for adding a level head to the discussion.
--
73, Cecil, w5dxp.com
 
RE: Loading coils  
by N3OX on November 21, 2011 Mail this to a friend!
W5DXP writes:

======
"If the coil is 41 degrees long at 4 MHz, we would certainly expect the standing wave envelope to develop.
"
======

I agree, because there's no option :)

If the coil is a 41 degree transmission line it will act like a 41 degree transmission line.

======
"For the 50 ohm load, the current at the source end was 140 ma and the current at the load end was 220 ma. Is it likely that extra 80 ma of current at the load is created out of thin air?"
======

Nope, but when inductive reactance gets high, a small stray capacitance can become quite important.. like maybe the capacitance of the current probes to the wire is important. The electric field associated with that would be concentrated between the wire and the probe body... i.e. it is a "lumped" capacitor.

There are lumped circuits that can explain a higher current out than is going in. If you had showed turn by turn on that coil that the current was a standing wave, there would be less ambiguity.

Then you'd have to grapple with the contentions that the coil will ONLY show delay line behavior near a ground plane, but that's just an effect of the tendency of this discussion attracting people who want to argue six things at once.


======
"Exactly when does part of a helical monopole abruptly stop acting like a transmission line and start acting like a lumped inductor?"
======

It doesn't happen abruptly. But it MUST happen.

A one turn coil is not a transmission line. I doubt that a uniform transmission line model is much use to a coil with length to diameter ratio of, say, 0.5.

I don't know for an almost square coil like yours.

But these are just hypotheses based on intuition, so the things I think are probably rather *inaccurate,* because this is a tricky subject.

The way to tell when the uniform transmission line model breaks down is to directly compare the fields of the simplified transmission line model of coils of different lengths against a fully three dimensional complete numerical simulation. But that requires someone to write it themselves or use incredibly expensive commercial software like HFSS.

Someone could take a stab at it in EZNEC for a big coil of many segments per turn, I suppose. At some point the shortness of the coil will invalidate the assumptions of the form of the solution to Maxwell's equations used to develop the transmission line model.

W8JI writes:

======
"I measured no delay or current taper at all in compact inductors. "
======

If ON4AA's calculator and the Corums' model derived from Maxwell's equations are correct, the delay issue I should be explained by standing waves on the coil. Standing waves = bad delay line.

As far as current taper, I suspect you haven't built an antenna that requires 100uH to resonate on 75m ;)

W8JI writes:

======
"The distributed capacitance along a loading inductor can approach or exceed the termination impedance of the inductor, in which case displacement currents can rob the upper area of the antenna from current. "
======

That's exactly what the Corums' model is trying to help describe quantitatively. It's offering a self-consistent model for the "distributed capacitance" of a coil with arbitrarily tight wire spacing and small diameter in terms of a wavelength.


 
RE: Loading coils  
by W8JI on November 21, 2011 Mail this to a friend!
By the way, if we want to talk about totally false statements or conclusions, here is one from the article:

Conclusion Number Two: Any 75m air-core loading coil must necessarily be in the ballpark of 30-45 electrical degrees long which falls under the classification of a distributed network, NOT a lumped circuit.


What a silly thing to say! Every 75 meter air core loading coil is 30-45 degrees long. That disagrees with everything factual.



 
RE: Loading coils  
by W5DXP on November 21, 2011 Mail this to a friend!
> W8JI wrote: To me, he sounds like he rewrites everything everyone else says just to argue that every system in the work can be analyzed as a transmission line. <

Again, I am going to ignore the ad hominem attacks, the misquotes, and the gross exaggerations of what I have said. Here's what I have said:

I said that for any RF circuit for which a lumped-circuit analysis yields valid results, a distributed network analysis will yield the same valid results. When a lumped-circuit analysis yields different results from a valid distributed-network analysis, the lumped-circuit results are wrong, i.e. the lumped-circuit model is proven inadequate for that particular analysis and the distributed network model is closer to Maxwell's equations.

Dr. Corum says for phase and delay purposes, the lumped-circuit model fails around an electrical length of 15 degrees. A 75m mobile air-core loading coil has an electrical length 2~3 times that 15 degree value and is indeed in the ballpark of being 10% of a wavelength.

I am not saying that every system needs a distributed network analysis although such an analysis would be entirely valid. What I am saying is that a distributed network analysis will yield valid results on almost all systems. I am saying that loading coils, like the 75m Texas Bugcatcher coil, that are electrically longer than 15 degrees need to be analyzed as distributed networks because the lumped-model fails. I simply agree with Dr. Corum on that point.

You failed to recognize the standing waves in your measurement setup because the lumped-circuit model doesn't allow standing waves. You failed to obtain an accurate delay through your test coil because the lumped-circuit model doesn't allow delays through coils. It is your lumped-circuit concrete mindset that blinds you to the technical truth.

A distributed network analysis does indeed work for everything - it is just messier than the lumped-circuit shortcut method. BUT WHEN THERE IS ANY DISAGREEMENT BETWEEN A LUMPED-CIRCUIT ANALYSIS AND A VALID DISTRIBUTED NETWORK ANALYSIS, THE DISTRIBUTED NETWORK ANALYSIS WINS EVERY TIME BECAUSE IT IS CLOSER TO MAXWELL'S EQUATIONS. That is what you are missing - the distributed network analysis has contradicted your lumped-circuit results for 75m air-core mobile loading coils for the better part of ten years now. A 75m air-core mobile loading coil is simply too physically large to be treated as a lumped-inductor. Your assertions to the contrary are opinionated rationalizations, not science.

> W8JI wrote: The system to generally avoid is the system where the loading inductor is behaving like a transmission line. <

THEN WE MUST AVOID ALL 75M MOBILE AIR-CORE LOADING COIL SYSTEMS BECAUSE THOSE INDUCTORS, BEING 30~45 DEGREES LONG, INDEED DO BEHAVE LIKE TRANSMISSION LINES. The reason that we have no choice is that, since we require a 90 degree antenna and the whole rest of the 75m mobile antenna (no top hat) is only 12 degrees long, we are forced to supply the additional 88 degrees with the delay through the coil plus the impedance discontinuity phase shifts at the bottom and top of the loading coil.

> W8JI wrote: I measured no delay or current taper at all in compact inductors. <

But you must realize that since EM energy cannot exceed the speed of light, there has to exist delays and current tapers even in the smallest of RF systems and that it is impossible for EM energy to move from the source to the load in zero time. That those real-world delays are often negligible allows one to use the lumped-circuit model with reasonable results. It is when the real-world delays are not negligible that we need to switch models and you don't seem to comprehend where the switch point is. The result is your 3 ns faster than light delay through a loading coil whose calculated velocity factor is 0.04 giving the 10 inch coil an electrical length of 20.8 feet. Tom, if the coil indeed does have a VF=0.04, you are asserting speeds of seven times the speed of light. Please try pushing one bit of information through that 100 uh coil in 3 ns and get back to us. Hint: Group delay has nothing to do with how fast EM energy is traveling in a system with standing waves. After all, a pure standing wave has a group delay of zero. Does that mean energy is traveling through the standing wave at faster then the speed of light?

A 75M MOBILE AIR-CORE LOADING COIL IS *NOT* A COMPACT INDUCTOR. Ask anyone who has ever been hit over the head with one of those humongous 75m Texas Bugcatcher coils.:)

Tom, so far, you haven't answered any of my technical questions. Here's another technical question for you (and others).

You say your 100turn, 2in dia, test coil is a lumped inductor incapable of supporting standing waves during your 4 MHz testing.

The Hamwaves inductance calculator indicates that a 290 turn coil of the *same coil stock* would make a 4 MHz helical resonant 1/4WL monopole which is a standing wave antenna known to be electrically 90 degrees long. It has a standing wave current loop at the feedpoint, a standing wave current node at the open end, and a cosine standing wave envelope 90 degrees long.

Are you really trying to get us to believe that if you cut off 34% of that resonant 1/4WL coil, the resulting 100 turns of coil, when installed in a 4 MHz standing wave antenna, is devoid of standing waves and incapable of exhibiting transmission line effects???

If we chop off 34% of a 90 degree long transmission line, the result is a transmission line that is 31 degrees long. Hey, wait a minute. The Hamwaves calculator says that your 100 turn coil is 33 degrees long!!!

Why wouldn't the coil still be ~34% of 90 degrees at 4 MHz as the Hamwaves calculator indicates? Why does the coil support standing waves in a helical antenna but does not support standing waves in an ordinary 75m mobile antenna with stinger?

Can you or anyone else give a concise, direct, technical answer to those simple questions?
--
73, Cecil, w5dxp.com
 
RE: Loading coils  
by W5DXP on November 21, 2011 Mail this to a friend!
> W8JI wrote: By the way, if we want to talk about totally false statements or conclusions, here is one from the article:

Conclusion Number Two: Any 75m [MOBILE] air-core loading coil must necessarily be in the ballpark of 30-45 electrical degrees long which falls under the classification of a distributed network, NOT a lumped circuit.

What a silly thing to say! Every 75 meter air core loading coil is 30-45 degrees long. That disagrees with everything factual. <

Tom, IMO, it is downright unethical for you to keep harping on a one word-omission typo time after time when you already know it has been corrected in an earlier posting. I have, once again, corrected the typo in your above quote. It was a simple one-word omission typo. I was not talking about all 75m antennas. I was talking only about 75m MOBILE antennas with an 8-12 foot length limit.

I don't know of any way to correct typos in eHam articles but I have corrected that typo on a later version of this article at:

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

As I told you in an earlier posting, here's what I intended to say:

Conclusion Number Two: Any 75m air-core mobile loading coil used with an 8-12 foot mobile antenna must necessarily be in the ballpark of ~30~45 electrical degrees long which falls under the classification of a distributed network, NOT a lumped circuit.

If you want to make fun of a statement, make fun of the actual typo-corrected statement, not harp, time after time, on a past history typo about which I can do nothing.
--
73, Cecil, w5dxp.com

 
RE: Loading coils  
by W8JI on November 21, 2011 Mail this to a friend!
Hi Dan,

W8JI writes:
======
"I measured no delay or current taper at all in compact inductors. "
======
N3OX:
"If ON4AA's calculator and the Corums' model derived from Maxwell's equations are correct, the delay issue I(t) should be explained by standing waves on the coil. Standing waves = bad delay line."

But a lumped component model with distributed capacitance to the outside world shows exactly the same thing because, of course, that is what a transmission line is. The lumped component model works perfectly in tank circuits also, and seems to be more accurate (in all the cases I have looked at) for predicting behavior over wide bandwidths.

The transmission line model needs modified for every large frequency change.

N3OX:
"As far as current taper, I suspect you haven't built an antenna that requires 100uH to resonate on 75m ;)"

No, I have not. The point of my writing was to provide something useful, not an academic exercise in how a transmission line can be used in some cases.

It cannot be a transmission line without displacement currents and series inductance. This is why for many years people have used lumped component calculations in tank circuits, and why they work for all frequencies where the parasitic reactances are a small part of the overall reactances.

For example HF pi network tank systems with a loaded Q of around 12 have about 135-degree phase shifts for most common impedance ratios. We COULD model them as transmission lines, but that only works on one frequency. It does not work at all for harmonic suppression or behavior at lower frequencies, or if termination changes.

There is a point where things work better, and are more easily understood and explained, by using a different model.

W8JI writes:
======
"The distributed capacitance along a loading inductor can approach or exceed the termination impedance of the inductor, in which case displacement currents can rob the upper area of the antenna from current. "
======

N3OX:
"That's exactly what the Corums' model is trying to help describe quantitatively. It's offering a self-consistent model for the "distributed capacitance" of a coil with arbitrarily tight wire spacing and small diameter in terms of a wavelength."

There has been a great deal of work on this in the past. WP Czerwinski wrote an article in Electronics in August 1960 on this, expressing an optimum design process graphically. There was a very long study about this many years ago by someone working on a doctorate degree, but I can't lay my hands on it.

The transmission line model is, however, one of the very last and least important things on my to-do list. I'd rate the usefulness, for most people working with HF and lower systems, as nearly zero. I have limited time for things with nearly zero importance or utility.

Now if I was working with helical resonators or radiators, it would be a different matter. In that case I probably would defer or refer to John Kraus or someone with a broader view and better research as a technical reference, and not someone who has measured one inductor at four data points some unknown way.

As it is for me, I have not found a flaw using lumped models and common sense in the real world and have no interest is a system that works only at specific points or under a narrow range of conditions.

I like to do things the easiest way that works best.

73,
Tom
 
RE: Loading coils  
by W5DXP on November 21, 2011 Mail this to a friend!
> N3OX wrote: If you had showed turn by turn on that coil that the current was a standing wave, there would be less ambiguity. Someone could take a stab at it in EZNEC for a big coil of many segments per turn, <

Someone already has done that using EZNEC. :) Here is a turn by turn current report for a load of 50 ohms and a load of 2571 ohms. The phase shift graphs are even more interesting if you are interested.

http://www.w5dxp.com/BugI.GIF
--
73, Cecil, w5dxp.com
 
RE: Loading coils  
by W5DXP on November 21, 2011 Mail this to a friend!
> W8JI wrote: But a lumped component model with distributed capacitance to the outside world shows exactly the same thing ... <

This is, of course, a false statement or else the distributed network model would have never been required. In particular, the lumped component model is incapable of predicting the correct phase of the currents. I have just posted an EZNEC simulation of a 75m Texas Bugcatcher coil at:

http://www.w5dxp.com/BugI.GIF

It's pretty obvious that difference in the coil loaded with a 50 ohm load vs the coil loaded with its Z0 value is a standing wave. What does your lumped distributed capacitance have to say about standing waves?
--
73, Cecil, w5dxp.com
 
Mobile Loading Coil Measurements  
by WB1AIW on November 23, 2011 Mail this to a friend!
Hi Cecil,

What is missing is a detailed description of the setup used to obtain this data.

What kind of resistors were used for the load?
(Power rating, voltage rating, type (wound/carbon/metal film) actual manufacturer and part numbers would be great!)

What was used for the signal source?
What was the source power output for each load configuration?
What were the input and output voltages for each test?
Was any phase (voltage-to-current) data taken at the source end for each load? (They are in-phase at the far end, obviously).
How was the coil supported?
How far was it from any grounded objects (test equipment, etc.)?
How was the oscilloscope set up?
Do you have any photos of the setup?

Any information you could provide would be appreciated.

Thank,
Lin
WB1AIW
 
RE: Mobile Loading Coil Measurements  
by W5DXP on November 24, 2011 Mail this to a friend!
Hi Lin, I wasn't there for the measurements but I will forward your questions to the EE professor at Louisiana Tech.

In the mean time, I have succeeded in generating an EZNEC file that simulates the measured characteristic impedance of the Texas Bugcatcher coil and I have adjusted the EZNEC source current to match those in the measurements. Those results are quite interesting and I will be including those results in Rev. 2.0 of this article located on my web page at:

http://www.w5dxp.com/coilmeas.htm
--
73, Cecil, w5dxp.com
 
RE: Mobile Loading Coil Measurements  
by W5DXP on November 24, 2011 Mail this to a friend!
Here are the results of the EZNEC simulation. The EZNEC coil height above the ground plane was adjusted until the current-in equaled the current-out when the load resistor was 1930 ohms, just like the measurements. Then the current-in was adjusted for each load resistor to match the measurement data. The current-out was the unknown variable. Here are the results.

Load Resistor
(1)50 ohms, (2)1200 ohms, (3)1930 ohms, (4)3170 ohms

Measured Current-Out/Current-In Ratio
(1)1.57, (2)1.26, (3)1.0, (4)0.746

EZNEC Current-Out/Current-In Ratio
(1)1.58, (2)1.258, (3)1.0, (4)0.704

The above results of the measurements vs the results of the simulation could hardly be any closer. From the graphic at:

http://www.w5dxp.com/BugI.GIF

We can certainly observe the effects of the standing wave inside the loading coil.

This tends to confirm what I suspected earlier - that the Z0 of the coil is dependent upon the location of the ground plane which doesn't exist for the Hamwaves calculator.
--
73, Cecil, w5dxp.com
 
NOT ? RE: Mobile Loading Coil Measurements  
by KQ6XA on November 26, 2011 Mail this to a friend!
Is the title "Mobile Loading Coil Measurements" wrong or deceptive?

It begs the question...
IS THIS AN ARTICLE ABOUT A COMPUTER MODEL?

If so, then the title should be something like:
"Analysis of a Mobile Loading Coil Computer Model"

GIGO!

If you actually measured the coil... such as one would normally measure a coil using an RF Network Analyzer (or Q meter or other RF test equipment) ...then the title might be correct "Mobile Loading Coil Measurements". If that's the case, then please disregard my exclamation.

Regards,
Bonnie Crystal KQ6XA

... at the RF test bench
 
RE: NOT ? RE: Mobile Loading Coil Measurements  
by KQ6XA on November 26, 2011 Mail this to a friend!
Current probes?

"Various resistance source and loads"?

Why?
 
RE: NOT ? RE: Mobile Loading Coil Measurements  
by KQ6XA on November 26, 2011 Mail this to a friend!
Why not a Q measurement?
 
RE: NOT ? RE: Mobile Loading Coil Measurements  
by KQ6XA on November 26, 2011 Mail this to a friend!
Some of the important things are:

Q of the coil
Self resonant frequency of the coil
 
RE: NOT ? RE: Mobile Loading Coil Measurements  
by KQ6XA on November 26, 2011 Mail this to a friend!
One other thing on mobile loading coils:

If you test them with a ground plane under them, the Q and other qualities of the coil will be quite different than without a ground plane!
 
RE: NOT ? RE: Mobile Loading Coil Measurements  
by W5DXP on November 26, 2011 Mail this to a friend!
> KQ6XA wrote: If you actually measured the coil... such as one would normally measure a coil using an RF Network Analyzer (or Q meter or other RF test equipment) ...then the title might be correct "Mobile Loading Coil Measurements". If that's the case, then please disregard my exclamation. <

I said in the article that the measurements were the result of bench testing by some Louisiana Tech graduate students. The EZNEC stuff is just to show that EZNEC agrees with the measurements. When measurements and simulations agree, they are usually both valid.
--
73, Cecil, w5dxp.com
 
RE: NOT ? RE: Mobile Loading Coil Measurements  
by W5DXP on November 26, 2011 Mail this to a friend!
> KQ6XA wrote: "Various resistance source and loads"? Why? <

"Source" should not be in the above quote. I did not say anything about source resistance in the article. What I said was "various resistive loads".

"Various resistive loads" was to demonstrate that the 75m mobile Texas Bugcatcher has a similar response to a transmission line. As I said in the article, if one wants to ascertain the characteristic impedance and electrical length of a piece of transmission line (shorter than 90 degrees) one can vary the load resistance and monitor the current at each end. For low-loss transmission lines, when the current magnitudes are equal at each end of the test line, the load resistor is approximately the Z0 of the line and the current phase shift is the traveling wave delay through the line in degrees.

> Why not a Q measurement? <

The Q of a Texas Bugcatcher coil has been measured before. Unless the Q falls very low, it doesn't much affect the conceptual qualitative results.

> Self resonant frequency of the coil <

The Hamwaves calculator says 8.2 MHz. My own measurements made on a mag mount indicate that when the loading coil is close to a ground plane, the self-resonant frequency decreases which is an expected result. The main concern is to keep a loading coil away from its self resonant frequency. The test results indicate that the loading coil occupies ~40 degrees of an antenna while the stinger may occupy ~12 degrees. That leaves another ~38 degrees to be effected by the impedance discontinuity at the coil to stinger junction. Here is a graphical Smith Chart presentation of how a 75m mobile air-core loading coil works.

http://www.w5dxp.com/mobant.JPG

Note that this is not the Texas Bugcatcher coil but is instead the 100T coil that W8JI used for his "measurements". The graphic is part of a series of two articles previously posted to eHam.net and reproduced on my web page at:

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

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

> If you test them with a ground plane under them, the Q and other qualities of the coil will be quite different than without a ground plane! <

Yes, that true. Each test setup has its own characteristics and results. Those particular conditions have been duplicated in an EZNEC simulation available at:

http://www.w5dxp.com/bugi.EZ

Again, this article exercise was meant as more of a conceptual qualitative presentation than any highly accurate quantitative presentation. The thing to realize is that the traveling wave delay through a 75m mobile air-core loading coil used on 75m is at least 7-10 times greater than 3 ns under any conditions likely to be encountered in real-world mobile operation.
--
73, Cecil, w5dxp.com
 
RE: NOT ? RE: Mobile Loading Coil Measurements  
by W8JI on November 27, 2011 Mail this to a friend!
"When measurements and simulations agree, they are usually both valid."

Correctly this would be:

When a large number of measurements and simulations agree, they are usually valid.

When it is one model and one set of measurements and they agree, it only means they are both right or wrong.

One model, and one set of measurements, nothing like the load the coil sees in the real world, was used to form an "always happens" rule.

At least with my one measurement, I said it can change with different loads.....which it does.

It is the silly "always happens" stuff, making hard rules without dozens or hundreds of varying measurements, that almost always will get us in trouble.

and....What does this matter to anyone, and how is useful or helpful?
 
RE: Mobile Loading Coil Measurements  
by W8JI on November 27, 2011 Mail this to a friend!
Here are the two rules from the article, where each conclusive statement calls the other conclusion a liar:

"Conclusion Number One: For any constant frequency, the ratio of the time delay to the phase shift through an air-cored 75m bugcatcher coil is NOT a constant and depends upon the amount of reflected energy present in the measurement setup. Only in a test setup free of reflected energy will the ratio of degrees of signal to phase delay equal 1.44 degrees/nanosecond (at 4 MHz)."

Rule number 1 says the time delay will vary, all the way to almost zero, for a given coil as load impedance changes. (To say nothing about other variables!)

.....and after being taught delay (that few of us care about) varies and is not constant as load varies, we have this second rule:

"Conclusion Number Two: Any 75m air-core loading coil must necessarily be in the ballpark of 30-45 electrical degrees long which falls under the classification of a distributed network, NOT a lumped circuit."

So now conclusions set a MUST "ballpark" limit of 30-45 degrees for every antenna and every loading coil in the world for 75 meters, yet rule one by the author taught us all that delay varies down to near zero for a given coil with only a termination change.

NONE of these terminations was anything even remotely like the actual antenna.

The whole thing puzzles me, and leaves me with more questions than answers.

We have one article have two conclusions expressed as "always" rules, each rule calling the other rule a liar, and then have these rules based on tests of one coil in one fixture with four data points, with NONE of the terminations even close to the actually environment, a representation of every 75M loading coil in the entire world.

I haven't leaned a thing, except the delay (that I don't care much about) varies or stays the same in all cases, depending on which particular "conclusion" I chose to listen to from one article.

 
RE: Mobile Loading Coil Measurements  
by W5DXP on November 27, 2011 Mail this to a friend!
> W8JI wrote: "Conclusion Number One: For any constant frequency, the ratio of the time delay to the phase shift through an air-cored 75m bugcatcher coil is NOT a constant and depends upon the amount of reflected energy present in the measurement setup. Only in a test setup free of reflected energy will the ratio of degrees of signal to phase delay equal 1.44 degrees/nanosecond (at 4 MHz)."

Rule number 1 says the time delay will vary, all the way to almost zero, for a given coil as load impedance changes. (To say nothing about other variables!) <

No, Rule number 1 says the time delay will not (much) vary at all for a fixed frequency. It is the phase shift that is varying, not the time delay. The traveling wave time delay does not (much) vary for fixed configurations at fixed frequencies. It is the phase shift that varies all the way to almost zero and it is indeed zero for a pure standing wave.

Using your logic, you would say that standing wave energy travels through a coil in zero time when just the opposite is true. A pure standing wave delivers zero net energy when its phase shift is zero.

You have yet to learn that the phase shift is unrelated to the traveling wave time delay which is the delay experienced by the energy that flows through the coil from one end to the coil to the other end. Until you correct your technical misconceptions, you will continue to misunderstand.

THE TRAVELING WAVE TIME DELAY THROUGH A COIL HAS NOTHING TO DO WITH THE PHASE SHIFT AT THE ENDS OF THE COIL AS LONG AS REFLECTIONS ARE PRESENT! Hint: Reflections are always present in standing-wave antennas. The terms "phase delay" and "group delay" do NOT indicate how fast energy is being delivered through a loading coil. Traveling wave delay is what indicates how fast energy is being delivered through a loading coil.

> W8JI [again unethically quoted a typo in the original article after being informed twice already that it was a typo]: "Conclusion Number Two: Any 75m [mobile] air-core loading coil must necessarily be in the ballpark of 30-45 electrical degrees long which falls under the classification of a distributed network, NOT a lumped circuit." <

Here's what I said in a previous posting:

[begin previous posting quote]Tom, IMO, it is downright unethical for you to keep harping on a one word-omission typo time after time when you already know it has been corrected in an earlier posting. I have, once again, corrected the typo in your above quote. It was a simple one-word omission typo. I was not talking about all 75m antennas. I was talking only about 75m MOBILE antennas with an 8-12 foot length limit.

I don't know of any way to correct typos in eHam articles but I have corrected that typo on a later version of this article at:

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

As I told you in an earlier posting, here's what I intended to say:

Conclusion Number Two: Any 75m air-core mobile loading coil used with an 8-12 foot mobile antenna must necessarily be in the ballpark of ~30~45 electrical degrees long which falls under the classification of a distributed network, NOT a lumped circuit.
[end quote]

So as you can see, there is no contradiction between Conclusion 1 and Conclusion 2 and I stand by both statements once the typo is corrected. Just so you don't "misunderstand" yet once again, here is a summary:

1. THE DIFFERENCE IN CURRENT PHASE AT THE ENDS OF A LOADING COIL HAS NOTHING TO DO WITH THE TRAVELING WAVE DELAY THROUGH THE COIL. You seem to think otherwise and that is the source of your technical blunder. When standing waves are present, the phase of the current bears no resemblance to the position x along the coil. Here's why:

Standing Wave Current Envelope MAGNITUDE is proportional to position x along the coil, i.e. cos(kx). Pure standing wave current phase, relative to position x, is constant for each 180 degrees of antenna, coil, wire, or transmission line. A 75m mobile antenna is almost always a standing-wave antenna.

2. In a 75m 8'-12' mobile antenna, the air-core loading coil must necessarily occupy ~30~45 degrees of the antenna, i.e. the forward and reflected traveling waves on the standing wave mobile antenna must necessarily take much, much more time than 3ns to travel through any humongous 75m air-core loading coil. It is impossible for anything to travel through (not around) your 100T test coil in 3 ns. Your measured phase shifts are mostly just standing and not delivering (much) energy.
--
Complaint mode on: Again, I resent the fact that you feel compelled to misrepresent what I have said and repeatedly harp on a long ago corrected typo as the only way to try to win a technical argument. Why not just take it technical point by technical point in a purely technical and polite discussion?
--
73, Cecil, w5dxp.com
 
RE: NOT ? RE: Mobile Loading Coil Measurements  
by W5DXP on November 27, 2011 Mail this to a friend!
> W8JI wrote: At least with my one measurement, I said it can change with different loads.....which it does. <

Of course, the phase shifts at the ends of a coil change with load just as they do for a transmission line but THE DELAY TIME THROUGH THE TRANSMISSION LINE DOESN'T CHANGE WITH CHANGING PHASE. Exactly the same thing happens with a physically large air-core loading coil. Your changing phase measurements have nothing to do with the time it takes to deliver energy through the coil. IT IS IMPOSSIBLE TO DELIVER ENERGY THROUGH A COIL FASTER THAN A TRAVELING WAVE CAN MOVE THROUGH THE COIL. Why don't we calmly discuss that one technical subject?

>...What does this matter to anyone, and how is useful or helpful? <

Can you (or anyone else) answer the following technical questions?

1. What is the relationship between antenna efficiency and the number of degrees occupied by the loading coil?

2. What is the relationship between antenna efficiency and the characteristic impedance of the loading coil?

The answer to those questions should matter to every serious mobile ham operator and could prove to be very useful and helpful. But we will never know the answer to those questions until we give up on the concept of 75m Texas Bugcatcher loading coils being lumped inductors.

Here's an easy experiment for you (or anyone else). Put your 100T, 100uh coil in series with a 10k ohm resistor and measure it at 4 MHz. Lumped circuit theory tells us that the series impedance should measure 10k+j2513 ohms. But the measurement will actually show a capacitive impedance because of transmission line effects, i.e. Rload > Z0 with 31 degrees of coil.
--
73, Cecil, w5dxp.com
 
RE: NOT ? RE: Mobile Loading Coil Measurements  
by W5DXP on November 28, 2011 Mail this to a friend!
> W5DXP wrote: Here's an easy experiment for you (or anyone else). Put your 100T, 100uh coil in series with a 10k ohm resistor and measure it at 4 MHz. Lumped circuit theory tells us that the series impedance should measure 10k+j2513 ohms. But the measurement will actually show a capacitive impedance because of transmission line effects, i.e. Rload > Z0 with 31 degrees of coil. <

Simulations using uSmith indicate that the measured impedance should be in the ballpark of 1670-j740 ohms, nowhere near the lumped circuit value of 10k+j2513 ohms. Exactly how does 10k ohms get transformed to 1670-j740 ohms? Transmission line effects, of course, in an environment where Rload/Z0 = 2.5:1 or higher.
--
73, Cecil, w5dxp.com
 
RE: NOT ? RE: Mobile Loading Coil Measurements  
by W5DXP on December 2, 2011 Mail this to a friend!
The important conclusion to take away from this discussion is the fact that, according to Dr. Corum's paper, any component that is electrically longer than ~15 electrical degrees needs to be analyzed using the distributed network (transmission-line-reflection) model because the lumped-circuit model will not yield valid results for phase shifts and degrees of delay.

This is true for wave guides, wires, transmission lines, and 75m air-core mobile loading coils. EZNEC simulations and loading coil measurements are consistent in agreeing that large 75m air-core loading coils experience transmission line effects to such an extent that a lumped-circuit analysis is invalid in predicting the delay through the coil.
--
73, Cecil, w5dxp.com
 
RE: NOT ? RE: Mobile Loading Coil Measurements  
by W5DXP on December 5, 2011 Mail this to a friend!
The Louisiana Tech EE professor has sent me some pictures of the Texas Bugcatcher loading coil test setup which will be included in updates to my article on my web page at:

http://www.w5dxp.com/coilmeas.htm
--
73, Cecil, w5dxp.com
 
RE: NOT ? RE: Mobile Loading Coil Measurements  
by KQ6XA on December 5, 2011 Mail this to a friend!
I viewed the image of the bench test setup of the coil.
I noticed some clip leads in the background. How much do the clip leads enter into the test?
It seems that clip leads are an unknown... both the connection and added inductance, or capacitance to ground and other things.
http://www.w5dxp.com/coilmeas.htm

Bonnie KQ6XA
 
RE: NOT ? RE: Mobile Loading Coil Measurements  
by W5DXP on December 6, 2011 Mail this to a friend!
> KQ6XA wrote: I noticed some clip leads in the background. How much do the clip leads enter into the test? <

The clip leads on the left of the coil are associated with the source which was a Kenwood 940. We are not concerned with the time it takes to transfer energy from the source to the coil, i.e. the clip leads have very little to do with the length of time it takes to transfer energy from one end of the coil to the other which is what we are trying to measure.

> It seems that clip leads are an unknown... both the connection and added inductance, or capacitance to ground and other things. <

The phase difference measurements between the ends of the coil are what they are given that particular test configuration which was unchanged during the four measurements except for the load resistance. The result with a 50 ohm load agrees with W8JI's results. What W8JI failed to recognize is the effect of the standing wave current on the measured phase (group delay) probably because standing waves are completely ignored by the lumped-circuit model.

The group delay does indeed vary with changing load. But that variable group delay has nothing to do with the length of time it takes for energy to be transferred from one end of the coil to the other end. Only when a single traveling wave is present, devoid of reflected energy, does the variable group delay equal the actual length of time it takes for energy to be transferred through the coil which is relatively constant with constant frequency and changing load.

Again, the measurements were not an attempt to nail down the exact delay through a large loading coil or to determine an exact characteristic impedance. They were merely designed to indicate a range of values for the energy transfer delay none of which are anywhere near 3 ns. It is impossible to transfer RF energy from one end of a 75m mobile air-core loading coil to the other end in 3 ns. Given the ~0.02~0.04 velocity factor of the coil, 3 ns violates the speed of light limit.

Years ago, I wrote an article that explains mathematically why standing wave current is so different from traveling wave current.

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

Again, if the current in the coil was a pure standing wave, the group delay between the ends of the coil would be zero! That does not mean that energy is traveling through the coil in zero time. In fact, it means just the opposite, i.e. a zero group delay through the coil means that there is ZERO NET ENERGY FLOW through the coil. Pure standing waves do not deliver any energy. Only traveling waves deliver energy.

Please stand by for Rev. 2.0 of my article at:

http://www.w5dxp.com/coilmeas.htm
 
Mobile Loading Coil Measurements  
by WB1AIW on December 12, 2011 Mail this to a friend!
Hi Cecil,

I have a few points to make about your article.


1.
"If the current is 100% standing wave current, the phase shift through the loading coil is zero degrees."

I think what you're trying to say is if there is a complete reflection at the other end, the delay is 0 ns. A complete reflection means either the other end is shorted to ground or open. By the way, shorting to ground gives us an easy way to measure the inductive reactance at that frequency with a oscilloscope; divide the voltage by the current.


2.
"In free space at 4 MHz, an EM wave travels at 1.44 degrees/nanosecond."

Actually, 1.44 degrees/ns is just another way of stating a frequency of 4 MHz; it is not a velocity, therefore, it is the same for any medium; free space, water, glass, coax, etc.


3.
"The point at which the Current-Out/Current-In ratio equals 1.0, i.e. Current-Out=Current-In, indicates the approximate value of the characteristic impedance (Z0) of the coil"

Approximate? It is EXACTLY the Z0!


4.
I decided to try to verify your statement that an loading element that presents a certain inductance at a certain frequency will always have the same delay when terminated into it's characteristic impedance:

To check to see if the delay will always be the same regardless of "coil" parameters, I decided to use a 50 ohm coax section to "match"/resonate a 796 pF cap (which mimics a whip in my humble model world) at 4 MHz. Using the data for RG-58 coax, I found the length required to create the correct inductive reactance was 244" long, thus the delay was 31 ns (Vf = 66 percent).

I then did the same with 450 ladder line (Wireman's 551, which is actually about 405 ohms), and found I needed only 52". The delay in this case was only 4.9 ns! (Vf = 90.2 percent).

From this, I conclude that if you wound a "bugcatcher" coil with different wire and/or with different dimensions or added a ferrite core (rx only perhaps), and trimmed each to present the same inductive reactance in all cases, the delay through the each of these is going to be different, and could be by a significant amount. Neither is there anything mathematically that I can find that forces the delay to be the same.

Maybe I've missed the boat completely, but I just can't verify your claim.

Lin
WB1AIW
I apologize in advance if there are any spelling or grammatical errors above.
 
RE: Mobile Loading Coil Measurements  
by W5DXP on December 13, 2011 Mail this to a friend!
WB1AIW wrote: (everything marked with '>')

> W5DXP previously wrote: (everything marked with '>>')
>> 1."If the current is 100% standing wave current, the phase shift through the loading coil is zero degrees." <<

> I think what you're trying to say is if there is a complete reflection at the other end, the delay is 0 ns. <

If the phase shift through the coil is zero degrees, then the group delay is 0 ns but that doesn't mean that EM energy is being transferred from one end of the coil to the other end in 0 ns which is obviously a violation of the speed of light. The speed at which RF energy is transferred through a coil is no less and no more than the speed of light in the medium and is equal to the speed of a traveling wave. A pure standing wave doesn't transfer any energy.

>> 2. "In free space at 4 MHz, an EM wave travels at 1.44 degrees/nanosecond." <<

> Actually, 1.44 degrees/ns is just another way of stating a frequency of 4 MHz; it is not a velocity, therefore, it is the same for any medium; free space, water, glass, coax, etc.<

You're right of course but that is not the concept I meant to convey. What I was comparing was a traveling wave in free space vs a standing wave in a transmission line. My statement was correct but probably misleading. I will rephrase the statement in Rev. 2.0. Maybe: At 4 MHz, an EM traveling wave travels at 1.44 degrees/nanosecond.

>> 3. "The point at which the Current-Out/Current-In ratio equals 1.0, i.e. Current-Out=Current-In, indicates the approximate value of the characteristic impedance (Z0) of the coil"

> Approximate? It is EXACTLY the Z0!

Only for lossless lines. If the line has losses, the Current-Out/Current-In ratio will equal to something less than 1.0. For instance, take a look at the difference in the current at both ends of a 100 foot long piece of RG-58 at 440 MHz feeding a 50 ohm dummy load. Iout/Iin = ~0.22 in a perfectly matched system.

> 4. I decided to try to verify your statement that an loading element that presents a certain inductance at a certain frequency will always have the same delay when terminated into it's characteristic impedance: <

Whoa Lin, I *NEVER* said that! I used the word "inductance" only one time in my entire article and that was to reference the Hamwaves *inductance* calculator. So your above assertion is a misunderstanding of what I said. I said that ONE certain fixed individual coil at a fixed frequency will always have close to the same traveling wave delay with varying loads. Two coils would have to have identical VFs and Z0s in order to possess the same delay and the same is true for transmission lines.

> Neither is there anything mathematically that I can find that forces the delay to be the same. Maybe I've missed the boat completely, but I just can't verify your claim. <

You missed the boat because I never made that claim. Of course, the delay through the coil and the number of degrees occupied by the coil can vary with different physical configurations, i.e. with different velocity factors and characteristic impedances. Please note that two coils with the same inductance and different form factors could have different delays at the same frequency just as two different transmission lines can.
--
73, Cecil, w5dxp.com
 
Mobile Loading Coil Measurements  
by KE4ZHN on December 14, 2011 Mail this to a friend!
eHam..one of the only amateur radio sites I know of where hams try to rewrite the laws of physics on a daily basis...and then argue about it till they get a massive coronary when real engineering data proves them wrong.
 
RE: Mobile Loading Coil Measurements  
by W5DXP on December 14, 2011 Mail this to a friend!
> KE4ZHN wrote: eHam..one of the only amateur radio sites I know of where hams try to rewrite the laws of physics on a daily basis...and then argue about it till they get a massive coronary when real engineering data proves them wrong. <

Let's put this particular problem in more specific terms. Ignoring quantum effects, Maxwell's equations are considered to be part of the "laws of physics". Those equations contain second order partial differential equations. Here's what Dr. Corum had to say about the lumped circuit model in his IEEE white paper reproduced on the Hamwaves web page:

http://hamwaves.com/antennas/inductance/corum.pdf

"Consequently, lumped circuit theory does not (and cannot) accurately embody a world of second order partial differential equations in space and time."

If someone wants to delve into the laws of physics, the above web page contains plenty.

Some people consider the simplified lumped-circuit model to be the "laws of physics" when it is only a very simplified/limited math model. It is up to the user of any math model to recognize when the math model is no longer applicable to a real-world problem, i.e. when the model itself is violating the actual laws of physics.

Nobody would use the lumped-circuit model on a transmission line that is 30 degrees long. So why does anyone insist on using the lumped-circuit model on a 75m mobile loading coil that is 30 degrees long?
--
73, Cecil, w5dxp.com
 
RE: Mobile Loading Coil Measurements  
by WB1AIW on December 16, 2011 Mail this to a friend!
Hi Cecil,

I do believe your coil is long/big enough that you need to consider the displacement current, and hence, the "transmission line" behavior if you want to understand why the current can be different at each end. You made a remark, however, that isn't possible, in connection with your figure http://www.w5dxp.com/coilphsh. First, you certainly can have more or less current going out at the far end than at the near end of the coil. But you say it is solely due to reflections, not displacement currents, etc. The trouble is, we know that, according to Ampere's law, this isn't possible. Whenever the charge density of an object changes, there is a displacement current. Period. By having more or less current flowing in than out at any instant in time, the charge density of the coil is changing, therefore a displacement current, equal to the difference between the in-going and out-going current, is flowing.

I made a model of a non-ideal coil and it behaved similar to what your measurements revealed with changing load. It made me realize that the current magnitude out of the top (far end) can be significantly less than that going in, if the coil is long. The displacement current is going to lower the already low input impedance of my mobile antenna, and it won't be contributing to my radiated signal. If it is going to have significant displacement currents, than I might as well put it at the very top and add a large capacitor hat above that for the remainder to flow in to. All the current (the coil displacement current and flow-through current into the hat) must therefore go through the radiating portion of the antenna system.
(Let's not split hairs about the fact that the slender conductor going to the loading coil and hat is also subject to some displacement current.) And by the way, all the coil current going into the capacitor hat is also leaving the hat as, you guessed it, displacement current (hence the term "capacitor hat"). But you probably already understand that, sorry, I digress.

You stated:
"If the phase shift through the coil is zero degrees, then the group delay is 0 ns but that doesn't mean that EM energy is being transferred from one end of the coil to the other end in 0 ns which is obviously a violation of the speed of light. The speed at which RF energy is transferred through a coil is no less and no more than the speed of light in the medium and is equal to the speed of a traveling wave. A pure standing wave doesn't transfer any energy."

But we know that the (group) delay through the coil is a constant regardless of load, so long as the system remains static (that is, the coil and it's environment remain stationary and the frequency is not changed) and obviously cannot be zero. So if the phase shift through the coil is zero degrees, then the group delay is still whatever it is with any other load. It takes x amount of time to propagate to the end, and, if there is any reflection, that reflected energy takes x amount of time to propagate back to the source end. You said it yourself, that the signal can't know what the load is until it gets there.
A pure standing wave isn't transferring any energy... because all the energy is completely reflected back to the source end; it's the only condition that can form a pure standing wave (SWR=infinity).
Do you agree with this?

Next:
"You're right of course but that is not the concept I meant to convey. What I was comparing was a traveling wave in free space vs. a standing wave in a transmission line. My statement was correct but probably misleading. I will rephrase the statement in Rev. 2.0. Maybe: At 4 MHz, an EM traveling wave travels at 1.44 degrees/nanosecond."

I'm not sure what you mean here. This is what stumps me, 1.44 degrees/ns is a frequency and has nothing to do with traveling/propagating/standing/etc. What is it that you mean by "travels" at 1.44 degrees/ns? Traveling implies motion along a length, but 1.44 degrees/ns has no "length" associated with it. How are you relating degrees/ns to unit length?


"Only for lossless lines. If the line has losses…"

Of course, as you said in the paragraph I was referring to. That response seemed a little argumentative to me, since I was responding to you assuming your conditions and I'll quote them here: "There is a small error due to I2R losses in and radiation from the coil but, like lossless transmission lines, we will consider that error to be small enough to ignore during this conceptual discussion."

“You missed the boat because I never made that claim.”

Yes, I really did miss the boat the first time through your article, sorry about that.


"Conclusion Number Two: Any 75m air-core mobile loading coil used with an 8-12 foot mobile antenna must necessarily be in the ballpark of ~30~45 electrical degrees long which falls under the classification of a distributed network, NOT a lumped circuit."

If it's being used to bring the whip into resonance, where at each point along its length the currents are in phase anyway, can you then just treat it as a lumped element? At that point, isn't it acting like a lumped element? I think it would be fine to do so for a first order antenna design. If you wanted make it as efficient as possible, then, yes, it would be nice to be able to model a coil accurately to see how much of the input current is being "lost" through displacement current and have the ability to try different geometries to limit that loss. I'm with you on that.

Your coil, like any other transmission line, has an inductance per unit length, Ls, and a capacitance per unit length, Cs, and therefore has a propagation velocity of 1/(SQRT(Ls*Cs)) and a characteristic impedance of SQRT(Ls/Cs) at a given frequency. It is really that simple. I say at a given frequency because there is also winding capacitance (turn to turn) to deal with. I believe it distorts the normal transmission line behavior in that its transmission line properties could change significantly with changing frequency, depending on its geometry/physical dimensions.

Well, I hope some of my input here will be informative to you. I have certainly learned something. I wish you and Tom could work together on this, if he is at all interested in doing so, as he probably has the equipment to measure these parameters. And it's fine if he isn't interested. What is important to one person isn't necessarily important or interesting to another.

I expect you’ll disagree with a lot of what I’m saying right off the bat and tell me so, but instead, please take a while to digest it. I’ll read what you have to say and if I find I am wrong, I’ll let you know.

Best,
Lin
WB1AIW
 
RE: Mobile Loading Coil Measurements  
by W5DXP on December 17, 2011 Mail this to a friend!
> WB1AIW wrote: First, you certainly can have more or less current going out at the far end than at the near end of the coil. But you say it is solely due to reflections, not displacement currents, etc. The trouble is, we know that, according to Ampere's law, this isn't possible. Whenever the charge density of an object changes, there is a displacement current. Period. <

Lin, because of limited time for responses at the moment, I am going to take things one subject at a time.

Consider a 45 degree long lossless transmission line with a forward current that is equal in magnitude at both ends of the line and a reverse current that is equal in magnitude at both ends of the line. Don't those conditions satisfy Ampere's law no matter what the magnitude of the total current at each end of the line?

If the forward current is 1.0 amp at zero degrees and the reverse current is 1.0 amp at zero degrees at one end of the line, won't the total current be 2.0 amps at zero degrees?

At the other end of the line, if the forward current is 1.0 amp at -45 degrees and the reverse current is 1.0 amp at +45 degrees, won't the total current be 1.414 amps at zero degrees?

Does 2.0 amps at one end of the line and 1.414 amps at the other end of the line really violate Ampere's law? If it does, I am ready and willing to demonstrate such a result on my work bench.

Please tell us exactly why displacement current is required and why the 2.0 amp vs 1.414 amp result violates Ampere's law.
--
73, Cecil, w5dxp.com
 
RE: Mobile Loading Coil Measurements  
by WB1AIW on December 21, 2011 Mail this to a friend!
Hi Cecil,

You question:

“Please tell us exactly why displacement current is required and why the 2.0 amp vs 1.414 amp result violates Ampere's law.”

Please, take a moment right now, to carefully and slowly, re-read the third sentence of my last reply. Please don’t go any further here until you do.

I’ll repeat that sentence here:

“First, you certainly CAN have more or less current going out at the far end than at the near end of the coil.”

I emphasized the word “CAN”, because it seems as though you thought I wrote "CAN’T". Hmm, that’s frustrating.

Anyway...


Let me say a little something about displacement current. That first paragraph I wrote doesn't explain what it is, and it seems you’re not entirely sure what it is either.

I don’t mean for the following to sound like a 1st grade teacher, but we might have a wide audience that is also interested, so please bear with me.

First, wherever a capacitor is subjected to a time-varying signal, there is displacement current across its plates. When charge is forced into one plate, an equal amount of charge on the other plate is “displaced” from or pushed from it, due to the changing electric field between them. Remember, like charges repel.

Though displacement current doesn’t actually consist of a flow of charge, it is considered a current in electrodynamics in order to mathematically complete the circuit between the plates of the capacitor. It is very handy.

The way you analyzed the coil, using a second wave traveling the opposite direction of the first, is perfectly fine and it yields the correct answers. It shows you the current at any point you want along the conductor. But remember that the voltage is varying too! In general, when current at one end of a conductor is different than that of the other, charge is accumulating (or being depleted), causing a change in VOLTAGE and therefore, a changing electric field between it and everything around it. Since the definition of displacement current is “a time-varying electric field”, one can correctly say that the difference between the current in and the current out is equal to the displacement current flowing to nearby conductors, such as the earth, for a wire over earth; a ground plane, for a wire over a ground plane; the shield, for a coaxial cable; the other conductor in twin-lead transmission line;
the car body and nearby ground, for a car-mounted mobile antenna; etc.

Anything that is accumulating or being depleted of charge, has a changing voltage. A changing voltage means a changing electric field, which IS displacement current. And it is equal to the rate of that accumulating/depleting charge. It is just a consequence of the fact that the accumulating charge is pushing/pulling on the charge in the objects around it.

You probably can completely ignore it without any problem whatsoever in your analyses. No need to mention it at all. You already know that not all of your input current into the coil is making it into the whip, and that is enough.

On my part, I just wanted to correct you, when you did mention it, that is all.



Your question was:

“Please tell us exactly why displacement current is required and why the 2.0 amp vs 1.414 amp result violates Ampere's law.”

Displacement current is just a consequence. A current of 2.0 amp vs 1.414 amp doesn’t violate Ampere’s law.

Displacement current is just a consequence of the fact that the currents in and out are different. In electrodynamics, its inclusion IS required as one part of how to explain the affects that charges have on another, via electric and magnetic fields.


Happy Holidays!

Lin,
WB1AIW
 
RE: Mobile Loading Coil Measurements  
by W5DXP on December 21, 2011 Mail this to a friend!
> WB1AIW wrote: I don’t mean for the following to sound like a 1st grade teacher, but we might have a wide audience that is also interested, so please bear with me. First, wherever a capacitor is subjected to a time-varying signal, there is displacement current across its plates. <

Lin, I don't want to sound like a teacher either but we need to move from 1st grade level to the college physics level. RF energy is actually transferred across the capacitor plates by photons propagating directly between plates. In fact, all EM energy transfer involves photons.

When Maxwell invented the concept of displacement current, the photon had not been discovered. If Maxwell had understood the QED relationship between electrons and photons, he would never have invented that magical displacement current. When the photon was discovered and the QED relationship between electrons and photons became understood, the concept of displacement current became obsolete even though the math model continues to yield valid results. To try to introduce displacement current to explain the current magnitude envelope of a standing wave is technical nonsense. The current envelope of a standing wave is a sinusoidal function even in a perfectly lossless situation.

All RF energy transfer is photonic in nature and travels at the speed of light in a medium which is much, much faster than electrons can travel. In fact, most of the electrons on your capacitor plates never make it off the plates. In the Wikipedia example on electron drift velocity a 3 amp DC current moves an electron 0.29 mm in one second. You can divide that 0.29 mm by 10,000,000 to estimate approximately how far an electron moves when being driven by a 10 MHz signal. It can be considered to simply be oscillating in place while absorbing and emitting photons that travel directly from one capacitor plate to another.

The need to introduce the concept of displacement current has not existed for about a century and is presently understood and explained by the completely normal propagation and radiation of photons of which Maxwell was unfortunately ignorant.

If we look at the current on an ideal unterminated transmission line, we can find a point where the net current is zero and a maximum current point 1/4WL away. That does not violate Ampere's current law. The forward current magnitudes at those two points are equal and the reflected current magnitudes at those two points are equal, i.e. there is zero displacement current, i.e. no loss of photons from the ideal transmission line. Why is it so hard to understand that if Ifor-Iref=0, then Ifor+Iref=2*Ifor 1/4WL away???

Here is an article I wrote years ago that uses EZNEC to attempt to explain the importance of using phasor math to superpose the forward and reflected waves.

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

"Apparently, the primary factor in the net current at the bottom and top of a loading coil is the superposition of the forward and reflected currents, not the losses in the coil, not the radiation from the coil, and not even the local environment of the coil. The effects of these last factors appear to be secondary compared to the superposition of the forward and reflected currents."
--
73, Cecil, w5dxp.com

 
RE: Mobile Loading Coil Measurements  
by W5DXP on December 21, 2011 Mail this to a friend!
> WB1AIW wrote: You already know that not all of your input current into the coil is making it into the whip, and that is enough. <

That is a popular misconception based on the lumped circuit model. In a base loaded 75m mobile antenna, more than 90% of the forward current is making it all the way to the tip end of the whip and being reflected. After all, it is a "standing wave" antenna with an SWR on the antenna greater than 20:1.

Again, consider a 1/4WL (90 degree) LOSSLESS transmission line, i.e. no radiation and no displacement current. If the current at the base is 1.0 amp, the current at the 60 degree point will be 0.5 amp. There is zero displacement current. The decrease is 100% the result of the forward current and reflected current being 120 degrees out of phase at that point. Just as much current is flowing from the tip of the whip back to the feedpoint as is flowing from the feedpoint to the tip of the whip. Zero energy is lost from the system.

At the feedpoint:
Ifor=0.5a at 0 deg
Iref=0.5a at 0 deg
Phasor sum Itot = 1.0a at 0 deg

At the 60 degree point
Ifor=0.5a at -60 deg
Iref=0.5a at +60 deg
Phasor sum Itot = 0.5a at 0 deg

At the tip of the whip
Ifor=0.5a at -90 deg
Iref=0.5a at +90 deg
Phasor sum Itot = 0 a
--
73, Cecil, w5dxp.com

 
RE: Mobile Loading Coil Measurements  
by WB1AIW on December 21, 2011 Mail this to a friend!
Cecil,

You have some concepts correct and some woefully wrong. Unfortunately, you've closed your mind to any new learning or clarification of misunderstood concepts. I was truly hoping to exchange technical information with you for the benefit of both of us. I found your article and links to others' work interesting and intriguing, thanks for that.

For me, the discussion is over.

Photons? Really?

Neither of my two college textbooks on electromagnetic theory ever mention the word "photon" in any context relating to being a replacement for displacement current. Not does that word ever appear in the index of either.

But they do "mention" displacement current and the equations describing it and how those equations make up part of the backbone of electromagnetic theory.

Go figure!


HAPPY HOLIDAYS!

Lin
WB1AIW
 
RE: Mobile Loading Coil Measurements  
by W5DXP on December 22, 2011 Mail this to a friend!
> WB1AIW wrote: Unfortunately, you've closed your mind to any new learning or clarification of misunderstood concepts. For me, the discussion is over. <

Lin, I'm sorry you feel that way but that doesn't seem like a good reason for terminating a technical discussion. Gentlemen can disagree without being disagreeable.

Let's set aside our differences and discuss one narrow subject. We don't even have to mention photons or displacement current to have that discussion.

Let's limit our discussion to a 1/4WL open-circuit lossless stub, an example of which one will find in any textbook on transmission lines. Hopefully, we can agree 100% on that classic subject.
--
73, Cecil, w5dxp.com
 
RE: Mobile Loading Coil Measurements  
by KU6X on December 22, 2011 Mail this to a friend!
Hello Cecil. In your Figure 1, if you place your coil inside a conductive box, with wires protruding
through holes on either side (generator and load on the outside), will the current ratio on each side of
the coil measured on the inside be the same as the ratio measured on the outside of the box?

73, John, KU6X
 
RE: Mobile Loading Coil Measurements  
by W5DXP on December 22, 2011 Mail this to a friend!
> KU6X wrote: ... will the current ratio on each side of
the coil measured on the inside be the same as the ratio measured on the outside of the box? <

John, it seems to me that, given the "inside point" and the "outside point" are only an inch apart, i.e. less than 1% of a wavelength apart, my WAG is that the two ratios would be very close in value but not exactly the same. Is it a trick question?
--
73, Cecil, w5dxp.com
 
RE: Mobile Loading Coil Measurements  
by KU6X on December 22, 2011 Mail this to a friend!
Hi again Cecil. No trick questions here. It seems to me that the experiment could be better replicated if the
environment was better defined by putting a box around it. I guess the next questions are 1) can a box be made (large enough?) so that the currents into and out of the coil would show a similar difference to what was measured in your experiment and 2) if so, how would the size of the box make a difference?
I'm trying to understand how "displacement currents" aren't involved in this experiment.

73, John
 
RE: Mobile Loading Coil Measurements  
by W5DXP on December 23, 2011 Mail this to a friend!
> KU6X wrote: ... how would the size of the box make a difference? <

The conductive box would change the characteristic impedance and velocity factor of the coil which would change the value of the results. It would not change the concepts involved. We would just need to find the correct Z0 value that causes reflected waves to be negligible. In general, the closer the coil is to any conductor, the lower the characteristic impedance. That probably explains the difference in the free space results from the Hamwaves inductance calculator vs the results obtained from real-world measurements and EZNEC simulations.

> KU6X wrote: I'm trying to understand how "displacement currents" aren't involved in this experiment. <

Like you, I am going to put "displacement currents" in parentheses to identify that concept and say something more about it later.

I didn't mean to imply that "displacement currents" are not involved. What I meant to say is that everything else besides the primary effect of phasor superposition of the forward and reflected waves seems to a be a secondary effect, i.e. a relatively small percentage of the total effects that cause "current droop" through a coil. If we assume there are no "displacement currents", the results are approximately the same.

Again, let's assume a perfectly ideal 1/4WL (90 deg) open-circuit stub in free space with zero losses of any kind including zero "displacement current" and assume 100 watts of forward power in the 50 ohm stub.

Since the stub is lossless, the reflected power from the open-circuit end will also be 100 watts. The "feedpoint" impedance looking into the stub will be zero ohms, i.e. 0+j0 ohms. From the power, knowing that Z0=50 ohms, we can calculate the component voltage and current values.

At the feedpoint end:
Forward Current: Ifor=1.414a at 0 deg
Reflected Current: Iref=1.414a at 0 deg
Total NET Current: Itot=2.828a at 0 deg

At the open-circuit end:
Forward Current: Ifor=1.414a at -90 deg
Reflected Current: Iref=1.414a at +90 deg
Total NET current: Itot=0a at 0 deg

Halfway (45 deg) from the ends:
Forward Current: Ifor=1.414a at -45 deg
Reflected Current: Iref=1.414a at +45 deg
Total NET Current: Itot=2.0a at 0 deg

The current distribution is virtually identical to a 1/4WL thin-wire monopole. Here's what Kraus said about a 1/2WL thin-wire dipole: "It is generally assumed that the current distribution of an infinitesimally thin antenna is sinusoidal and that the phase is constant over a 1/2WL interval, ..."

And indeed, the current at the halfway (45 deg) point is 2.828cos(45)=2.0a at 0 deg and the current phase is zero degrees all along the stub. Standing wave current envelopes on standing wave antennas are pretty close to being sinusoidal on efficient antennas.

The current droops from 2.828a at the feedpoint to 2.0a at the halfway point in a completely lossless 1/4WL stub system. In the ideal system, that droop is 100% caused by phasor superposition of the forward and reflected waves and nothing else. It does not violate Ampere's current laws.
--
If you would like to continue this discussion, I will illustrate how large air-core loading coils function even if "displacement current" is absent in an ideal system.
--

On the subject of "displacement current": I didn't say that "displacement current" doesn't exist, just that it is a misnomer. "Displacement current" is actually just ordinary photon flow. Some physicists call it "photon current" which is also somewhat of a misnomer. They should probably call it "photon-equivalent current" because it is based on the H-fields of the photons and not associated with charged carrier particles.

Steady-state DC doesn't require photons and DC theory was adapted to AC before photons were known to exist. Hence, the magical idea that electrons rush from one capacitor plate to another at light speeds through some sneak path. In the real world, we now reject that myth and know that electrons involved in RF energy transfer move hardly at all and tend to oscillate in place while absorbing and emitting photons, the actual transporters of the energy. The "displacement current" math model still works but does NOT represent electrically charged particles moving at the speed of light, i.e. the AC fields measured by Maxwell were simply clouds of photons each transferring ExM energy/unit-time.

Often, "displacement current" would have to be traveling faster than the speed of light around its sneak path to equal the measured RF energy transfer delay through a capacitor. Of course, that's not a problem for the lumped-circuit model which presumes that everything happens faster than the speed of light. :)
--
73, Cecil, w5dxp.com

 
Email Subscription
You are not subscribed to discussions on this article.

Subscribe!
My Subscriptions
Subscriptions Help

Related News & Articles
The 'Wadetenna' for HF Pedestrian Mobile


Other Antennas Articles
Using Yagis Backwards
The 'Wadetenna' for HF Pedestrian Mobile