Loading Coils
Home Up Loading Coils Beta Match Propagation Razor Beams CQ Cover OK3BU story TS-870 mod XJ3ZZ/1 DXped


About Me
Photo Gallery
TopBand Records

Current distribution in the Antenna Loading Coils.

Yuri Blanarovich, K3BU, VE3BMV, VE1BY

A recent article on eHam.net by Alan Applegate, K0BG "In Search of 'The Perfect Mobile Antenna'" on Aug. 5, 2003 and posted comments, created some controversy and heated exchange between myself and the "well respected engineer" (according to Aaron, NN6O) Tom Rauch, W8JI. This article had some flaws and was far from approaching "perfect" mobile antenna. To which W8JI stated:

"By using a flawed and seriously over-simplified model, the results are totally misleading when applied to conventional antennas. Repeating misleading information in article after article does NOT make it correct. It certainly does not make our community more skilled or better informed about how things work."


"While I appreciate all your (K0BG) efforts, it is important that readers and writers fully understand why and how something works before reaching conclusions. Otherwise this all just wastes bandwidth, and people learn incorrect information. The goal of E-Ham and Internet should be to INCREASE technical skills through mass peer review and learning, not to repeat misinformation. "

In view of the above and in order to "INCREASE technical skills through mass peer review and learning"  I have summarized in my posting the most important items contributing to high efficiency of mobile or shortened antennas. W8JI commented that I was  wrong about the current distribution in the antenna loading coils, where I stated:

"4. Loading coils. Should be mounted as high as possible in the antenna mast to increase the current radiating portion of the whip. Wire or tubing should be rated to carry the power without melting the coil. Q of the coil is not that important, coil in this situation is the part of the radiating element, most of the current is at the base of the coil and surprisingly Q or form factor is not THAT important as measured and verified experimentally by W9UCW. Loading coils at the base or autotuners are the worst. Bugcatchers, single coils with no shorting, spider mounts for multiple coils are very efficient."

To which W8JI replied:

"The idea current is high in only the start of a coil is not correct.
Model an antenna with EZnec, and look at the load. Model a coil in any software, and look at current. Read any textbook, even beginner's textbooks, and see what they say. Measure a real antenna yourself!


You are like to call names, insult people, and argue rather than take the time to learn basic electronics. This is in any book, including the ARRL Handbook. If you look at HOW an inductor works, the current flowing in one terminal ALWAYS equals the current flowing out the other terminal. "

(The rest of the exchange can be seen at eHam.net following the K0BG article at http://www.eham.net/articles/5998)

This was quite an accusation and a challenge to reality and my knowledge. This was not the first time that W8JI "challenged" me and was proved wrong. Knowing what was the reality (uneven current distribution, coil gets hotter at the bottom) and what Barry Boothe, W9UCW measured, I was curious what was the source of W8JI misinformation. I ordered 19th edition of ARRL Antenna Book and followed chain of references that led to information on page 16-7 and Fig. 9 and 10.

Looking at those two pictures, we can see that the current across the radiator was "linearized" to be a nice straight line instead of actual cosine curve. Also, Fig. 10 is missing the important cross-hatched area, the current across the coil is shown as a "nice" linear current over h2 and coil apparently has zero physical length. This passage in the Antenna Book is written by Bruce Brown, W6TWW "Optimum Design of Short Coil-Loaded High-Frequency Mobile Antennas" first published in ARRL Antenna Compendium, Volume 1, page 108.

Going back to that reference we can see on page 109 Fig.1 a current distribution on 1/4 wave radiator.

This is a true representation and shows the last 30 degrees "linearized", which is OK for simplification, but it introduces an error, which could magnify in precise calculations and modeling. 

This is expanded in Fig. 2,

The base coil is omitted, which "simplifies" or distorts the picture of real current distribution in shortened radiator with base loading coil. As we will see later, if the coil was shown, 1 A current applied at the bottom, and current measured at the top of the coil, authors would have seen the drop across the coil and current at the bottom of the radiator (top of coil) would not be 1 A, but more like shown by the shaded area in Fig. 3.

 Then we see in Fig. 3 coil inserted in the middle of 30 deg. radiator. Coil has "zero" physical length and current distribution across the coil is shown as constant and as W8JI claims. The implication is that coil magically widens the area under the current curve over the top 15 deg. Here is the cross-hatched area that is missing in the ARRL Antenna Book, which is really what is happening in the coil loaded radiator. This figure implies that current across the coil is constant and actually makes short radiator work better than the "naked" one, without the coil (proportional to the areas under the current curves). In reality, the picture should show current across the coil coming from the bottom right corner of the shaded area to the top current curve at 15 deg. or bottom of the coil.

Bruce, W6TWW, states:

"Therefore, the current exiting the top of the coil is the same as that entering the bottom of the coil. (This is true for conventional coils. However, radiation from long skinny coils allows coil current to decrease, as in helically wound antennas.) This is easily verified by installing RF ammeters immediately above and below the loading coil in a test antenna. Thus, the coil forces a much higher current into the top section than would flow in the equivalent part of a full 90-degree-high-antenna."

So here is qualifier that in long skinny coils, as in helically wound antennas, radiation allows coil current to decrease. The problem seems to be that in one case the current decreases across the coil (helical), but in "regular" loading coil that is not allowed, which is false. (Where was the measurement, verification?)

Is this really true or is it based on a previous reference? Lets follow the trail to the referenced article (by W6TWW) in 1953 QST, p. 30 by J. Belrose, VE3BLW (now VE2CV) "Short Antennas for Mobile Operation" and we see the origin of the "constant" current across the coil and the "linearized" current distribution..


Curves with 1, 2, 3 show various current distributions from 1 for no coil, to 3 for coil that brings the antenna to resonance. Jack in his calculations assumes that the current across the coil is constant and that seems to perpetuate all the way to the latest edition of ARRL Antenna Book. So much for the "theory". What is the reality? I repeatedly asked W8JI to measure the current in typical mobile coil loaded antenna, like in Hustler 80 m resonator. His reply was that he measured thousands of coils and he found constant current. He would not reply to this one case that represents a typical situation and is the subject of this dispute.


What is the truth?

It all started with discussion on the TopBand Reflector (see the archives on eHam.net for May 2003) with thread "160 m Mobile Antenna Suggestions"


Barry, W9UCW pointed out his findings based on real life measurements of decreasing current across the loading coils. This agreed with my "unscientific" experience, when I fried the loading coil with 600W into Hustler resonator, melting heat-shrink tubing and wire at the bottom of the coil. There was an exchange of arguments on the subject of current in the loading coils, with W8JI insisting on constant current across the coil. Unfortunately, the TopBand reflector moderator terminated W9UCW responses, so W8JI had his famous "right" last word.

Tom, W8JI on his web page http://www.w8ji.com/mobile_and_loaded_antenna.htm states:

"The modeled current distribution for 1-ampere applied at the base (in 1-foot intervals) is:

1ft= 1.0031 
2 ft= 1.0091
3ft= 1.0178
4ft= 1.0318
5ft= 1.0175
6ft= .97512
7ft= .92984
8ft = .89522

Measuring the current into and out of the loading coil with a small thermocouple RF meter, I detect no difference This is in close agreement with the model. " 

and "conclusion"...

"Clearly there is no basis to the claim current is high only in the first few turns of an inductor, or that current tapers in relationship to "electrical degrees". The most accurate way to state the effect would be to say: "When the loading coil is short and the capacitance of the antenna beyond the coil is reasonable (in this case 3000 ohms Xc or less), there is an immeasurable reduction in current in the coil."

First, there is a problem in his modeling with current increasing from the base towards the coil. That should be the flag telling him that 2 + 2 is not sometimes 4.04. Second, EZNEC has no provision for incorporating physical length of coil. It just considers LC parameters. Roy Lewallen, W7EL, author of EZNEC and Richard Clark, KB7QHC recommend workarounds to replace the coil with cylinder of similar size or breaking the coil to number of physical segments with appropriate inductances. W8JI "findings and measurements" hardly reflect the reality. As someone said, one measurement is better than thousands of theories. The question is, how was W8JI measuring the current, and getting more current than it was applied at the base?


Barry, W9UCW among other arguments and explanations wrote:

"In our measurements, we used long and short coils and the current taper was
was almost identical if the topmast capacitance was held the same. If the
"make up" was above the coil, there was slightly less taper down of current,
due to the larger capacitance above. It would also resonate lower in freq.
It appears to us that the current decrease in the coil has most to do with
the section of the quarterwave element that it effectively replaces. The
actual decrease in our tests was always a little more than the decrease
calculated for the "replaced" section, no matter what coil was used. I hope
that answers your questions."


"I think your position (W8JI) is clear....that under the condx described, current
reduction in a loading coil can't, won't, never did, never will happen. My
position is that it always does, and I've measured it. Neither of these
hypotheses will  go far to satisfy the real objectives of our study."

Barry, W9UCW was kind to provide proof in the form of some measured data and photographs showing the antenna setup, loading coil and RF ammeters installed at the top and bottom of the loading coil, which could be reversed:

"Here are some actual measurements of current below and above loading coils.
 92" mast, using a HI-Q coil (openwound airdux, 2 1/2"d) with small thermocouple type meters mounted on the insulated coil support. First for 40m, moving the coil in the mast from base to center to top (with hat) and reresonating.
                                    Base    --100ma below & 66ma above
                                    Center --100ma below & 45ma above
                                    Top     --100ma below & 37ma above
 Then, same test but for 30m
                                    Base     --100ma below & 75ma above
                                    Center --100ma below & 60ma above
                                    Top     --100ma below & 52ma above
On a long, skinny 160 resonator with 25pf of top hat and whip, mounted on an 8' mast, I read 100ma below and 65ma above the coil.
Because of the constant claim that this must be due to the fact that the coil is so big compared to a wavelength, I measured the in and out current on a TOROIDAL loading coil used on a 20m mobile antenna. It was a 78" base mast (including spring and mount) with a 38" top whip (including 12" of alum. tubing for adjustment).
                                     Below   --100ma & Above  --79ma
When I moved the coil to the top of the mast and made a horizontal "X" top hat to resonate it back on the same freq, I got
                                      Below  --100ma & Above  --47ma
So, It happens even in a totally shielded loading coil with miniscule power going thru it!  Kirchoff has no laws about current being the same on both ends of inductors. His current law is about one POINT in a circuit and his voltage law is about a closed loop."

... and some significant difference W9UCW in field strength measured between the base and center loading coil:

"The actual difference in signal strength between top and base loading of a 9' antenna is about 16 db (measured) on 75m, but Tom calculates 8db on 160. That's because he assumes the same current in the coil. Actually it's worse on 160 than 75."


Barry's  pictures are worth a thousand words:

W9UCW's setup with radial field (60), base loaded vertical with RF thermocouple ammeters inserted at the top and bottom of the coil.


Here is the coil in center loaded radiator, 100 mA meters at both ends of the coil. The bottom one is showing full deflection (with power adjusted to) - 100 mA while at the same time the top ammeter is showing 45 mA as described above. The meters were mounted that way so that they could do a test and then just turn the coil assembly upside down and do another test to make sure results were the same and that no anomalies crept in. Results were always identical.


The reality.

So how does the real distribution of current in loaded antennas look? The answer can be found in the John Devoldere's "Bible" - "ON4UN's Low Band DXing", 3rd Edition, on page 9-34:

When I pointed out this reference to W8JI, his response was:

"I just looked at that, and you are right. John is incorrect, and I'll bring it to his attention. Thanks for pointing that out."

This is not the first time that W8JI is wrong. His typical modus operandi is first to attack and ridicule the opponent, then the exchange of arguments ensues. When he realizes he is wrong, rather than admitting, he clouds the issue with his "arguments". After staying quiet for a while, he then emerges, pretending to be the expert on the subject with corresponding postings on his web page, without giving credit to the originator. Normally this is called plagiarism. 

The Internet is a great place to publish ideas, good and wrong. In the spirit of Tom's posting on the eHam.net's purpose, I had to react to his disinformation by presenting the facts, especially when it happened more than once. 

Why is this important? Technical subjects have their laws and rules. Perpetuating wrong information doesn't serve anybody. As we can see in this example, something that was "established" 50 years ago, perpetuated through "peer reviewed" books to this day, can cause problems and wrong conclusions.

In summary:
The current in a typical loading coil in the shortened antennas drops across the coil roughly corresponding to the segment of the radiator it replaces. 

... and that's the way it IS, hardly W8JI's  -

"...there is an immeasurable reduction in current in the coil."

I hope this will help to better understand the loaded antennas, to incorporate the effect into the modeling software and to develop more efficient shortened antenna systems.  

Below is the result of plotting current in the G5RV antenna using inductors in the form of loading stubs as done by W5DXP in Eznec. It can be seen that the current entering the stub is greater than current exiting the stub. When simple inductance in Eznec is inserted in place of the stubs, the current erroneously is shown as the same at the both ends of the inductor. 


Comments from the REC.RADIO.AMATEUR.ANTENNA News Group