truth and untruth.pdf

The Truth and Untruth about

Electrically Small Antennas

The word peerless caught in the throat of Jerry

Sevick, W2FMI who had written a book

entitled “Transmission Line Transformers”, 2 nd

Edition published by the ARRL in 1990. Jerry

considered that his transformer type, a bifilar wound

choke type balun on a toroidal core, was superior ---

and perhaps only he knew how to design them.

John S. (Jack) Belrose, VE2CV, VY9CRC

The title of my talk suggests that there are controversies

concerning the performance of electrically small

antennas, and indeed there are concerning in particular

two very different types of antennas: the so-called

crossed field antenna (CFA) and a follow on version the

EH antenna, and compact transmitting loop antennas.

This resulted in a controversy that went on for a year or

two. Jerry wanted QST to publish a rebuttal article. I

never did see his draft manuscript, which was never

published by the ARRL. But his views were published

under the collective title “Baluns Revisited” in

Communications Quarterly in 1992, four articles, and in

CQ Magazine in 1994.

But there is a wider range of controversy: concerning

the basic concepts of transfer of power generated by a

high frequency tuned power amplifier to the

propagation media. I have been involved in this debate

for about 15-years, and so I will begin with a brief

discussion of this topic.

There is nothing wrong with bifilar wound choke

baluns, so called transmission line baluns, but provided

the VSWR is not too high --- in some cases for example

at MF this type is the preferred balun. For example see

the Figure below .

But first concerning controversies there is also a

controversy concerning the use or non-use of baluns,

and which balun to use. Since many antenna types are

balanced, and present day HFTPA’s are unbalanced,

and, since the preferred method of feeding antennas is

to use coaxial cable, rather than 2-conductor balanced

transmission line, a balun should, often must be used in

the overall antenna system design.

I have during the past 50-years published about eighty

papers, technical articles and technical correspondence

on antennas (and an equal number concerned with radio

propagation). The basic theme of my work has been

concerned with providing an understanding of the

characteristics of various antennas, and in particular

addressing the topic: performance of antennas in their

operating environments. Most of what I have written

has not generated controversy, at least prior to about

1991.

For the amateur in radio the W2DU ferrite

bead over coax is a good balun ( Figure ) --- it

exhibits slightly lower loss ( Figure ).

CONTROVERSY BEGINS--- WITH JERRY

SEVICK, W2FMI CONCERNING BALUNS

In 1991 QST published an article I wrote entitled

“Transforming the Balun”. The QST Technical

Editor added the byline:

“In this QST breakthrough, W2DU’s peerlers 1:1

current balun design serves as the basis for excellent

ferrite-bead-choke current baluns capable of 4:1 and

9:1 impedance transformation”

-----------------------------------------------------------------

Presented at the QCWA 2004 International Convention,

Amateur Radio Technical Session, Friday, October 15,

held at the Lord Elgin Hotel, Ottawa, ON, Canada

Walter Maxwell, W2DU, Tom Rauch, W8JI, John

Fakan, KB8MU and I corresponded with WB for

several years, in an attempt to publish a QST article that

addressed our differing views --- but with no success.

In 1997 we gave up and published an extended article

in Communications Quarterly Fall 1997 issue,

expressing our view that for maximum power transfer

the HFTPA should be conjugately matched, and that the

effective output impedance of the amplifier under such

conditions was equal to the load impedance. We

further stated that this impedance was non-dissipative.

We presented the results of seven experiments, and one

of the experiments was the set-up initially exactly like

that used by Bruene, that supported our view (not his).

The following issue of Communications Quarterly,

Spring 1998 published Bruene’s rebuttal, a detailed

rebuttal claiming that every one of our experiments

were flawed, flawed, flawed.

Let me begin at the beginning by quoting from the

IEEE Dictionary:

“Resistance is also defined as the real part of an

impedance. Because the impedance of a network

deals with energy transfer, it has nothing to do with

where the energy came from, or what became of it.

The real part of the impedance of a network

therefore does not dissipate energy of itself. Only

that portion of the real part of an impedance that is

in fact a dissipative resistance will dissipate energy”.

Jerry send me several of his best baluns, which

unfortunately we destroyed during testing --- blue flame

arcing and smoke when subject to high VSWR and high

power (1 kilowatt). The W2DU ferrite bead over coax

balun did not destruct, excepting loss and heating

increases with increasing VSWR.

If we measure Vout/Iout , measured at the output

terminals of a properly adjusted PA, tuned for

maximum design power transfer, which in the practical

case is an antenna system, we can infer an impedance

Zout = Vout/Iout, which is equal to the load

impedance. But to measure Zout we have, as my long

time colleague Jim Wait, now deceased, told me, we

have to dissipate power. In other words operate the

power amplifier into a 50 ohm resistive load. To

measure Zout change the load a little bit, 5-10 percent,

and observe the change in Vout and Iout (rms values).

We have to make a small change in the dissipative load

resistance, because we do not want to change the

operating characteristics of the PA tube(s).

Whatever type of balun is used a current balun should

be used, and the balun should be on the input (tuned)

side of ther ASTU --- see my QST article in the

October 2004 issue of QST.

A CONTINUING SAGA

But the most controversial topic, I have folders and

folders full of correspondence, arose from the article by

Warren Bruene, W5OLY entitled “RF Power

Amplifier and the Conjugate Match”, published in

QST November 1991 issue. Warren wanted to discredit

Walter Maxwell, W2DU’s book “Reflections-

Transmission Lines and Antennas”, published by the

ARRL in 1990.

This means we have to accurately measure very small

changes in current and voltage. I show in the next

Figure a comparison between what we (Walter

Maxwell) measured compared with the controversial

curve presently by Bruene.

Bruene showed measurements (a curious set of

measurements) that in his view showed that the output

source impedance (referred to by him to be the source

resistance) of a tuned RF power amplifier was 5-times

the load impedance. This of course is wrong, but

Bruene has stuck to his guns to the present date. He

still thinks Walt Maxwell and I are wrong!!

comparable to a mobile whip --- for example a 1.7 m

diameter loop at 1.8 MHz and at 3.75 MHz, and that

traditional formula for radiation resistance, developed

about 60 years ago was correct.

And present day simulation agreed with experiment.

Mike Underhill, G3LHZ for whatever reason disagrees

with most of what has been written on compact loops,

beginning with his attention grabbing paper “Magnetic

Loop or Small Folded Dipole”, published in an IEE

Conference Proceedings in 1997. And since that date

he has written five or six papers/articles, each more

controversial than earlier papers. In an article entitled

“The Truth about Loops”, published in the RSGB

International Antenna Collection, 2003 he states:

“the very low efficiencies (the few percent for the

very small loop size that I spoke about above),

predicted by simulation and existing theory are,

shown (by his measurements (??)) to be quite

frankly wrong by up to 1000 times (30 dB). How

can such measurements have been overlooked for so

long? It is a bit of a mystery and arguably a bit of a

scandal.”

Now let me return to transfer of power to an antenna

system. Zout inferred from Eout and Iout is a

nondissipative impedance --- we do not dissipate power

there --- this is the impedance associated with the

generation of power. We want to transfer that power to

the antenna system. Therefore the antenna system must

be tuned to provide a conjugate match to the PA. The

feeder coax is associated with an impedance ---

typically Zo = 50 ohms. Zo is the characteristic

impedance of the transmission line --- it also is a

nondissipative impedance.

G3LHZ considers the efficiency of such small loops

to be 80-90 percent, not a few percent.

Pat Hawker, G3VA, in his Technical Topics column in

RadCom, December 2002, I have contributed his TT

Column, discusses this difference of opinion, and he

challenged the antenna establishment to comment. My

response was a 2-part article published in the June/July

2004 issues of RadCom.

The matched antenna system presents an impedance

looking in Zas = Rr + Rloss. For an efficient antenna

system Rr (the radiation resistance) >> Rloss. Rr is a

non dissipative resistance, since power is not dissipated

in this resistance. Rr is an impedance associated with

the power that is transferred to the propagation

medium.

But the controversy has not ended. In the

August/September issues of RadCom G3LHZ has

published an even more controversial article entitled

“New truths about small tuned loops in a real

environment”.

Now before I run out of time let me discuss the topic of

this talk, The Truth and Untruth about Electrically

Small Antennas.

But my loops were in their operating environment.

COMPACT LOOPS

Mike seems to have gone completely bananas, but since

he writes under the title of a Professor in the School of

Electronics and Physical Sciences, University of

Surrey, I suppose there are some who believe that he

knows what he is talking about??

I have operations used, evaluated by experimental

measurement and by simulation (numerical modelling)

and written on small compact transmitting loops,

dating back to the mid-eighties --- in the amateur

literature let me refer to my November 1993 QST

article “An Up-date on Compact Transmitting Loops”.

He claims that his inferred intrinsic efficiencies, 80-90

percent, inferred not from measured field strengths but

from Q-factor based on measured VSWR, to be

confirmed by his proposed extensions of EM theory ---

Maxwell’s EM theory is not quite right, according to

Underhill,

I considered that the performance of such antennas

(perimeter/wavelength as small as 0.03 to 0.06) was

about what one would expect, a few percent,

and the Somerfeld-Norton ground wave propagation

theory needs revision. He also disagrees with the Chu-

Wheeler Q criterion. And field strength measurements

over ground need to re-evaluated.

I could go on --- but I will stop there. My

recommendation is that we should stop reading what

G3LHZ has written --- since this will avoid further

confusion.

I cannot believe that such nonsense is published. We

do read nonsense published in some amateur literature,

but papers published by the IEE??

Let me tell you what I measure, and what I infer from

our numerical simulation studies ---- Figures.

I have numerically modeled various loop antenna using

NEC-4D, as a simple 1-turn loop, tuning capacitor at

the top (open squares), source on conductor, circles at

the bottom of the loop.

Since the loop is a high_Q inductive reactance antenna,

it can be tuned by means of a capacitor, and power

couped into it by means of a small auxiliary loop, the

size of which is adjusted so as to realized a 50-ohm

input impedance.

The Figure below shows the vertical radiation pattern

for a vertical 3.4m diameter loop, frequency 3.75 MHz,

base height 2m, compared with a half-wave dipole at

10m. Compare this figure with measured NVIS

performance, see below.

The measured bandwidths of AMA loops, VSWR < 2:1

is shown below.

The following Figure shows measured performance for

reception of a near vertical incidence skywave signal

(D2 for the dipole, and A2 for the loop). Notice that the

narrow bandwidth of the loop results in improved

reception of the monitored signal, in spite of the fact

that the received signal is about a S-unit less than for

the dipole.

The Figure below shows the vertical radiation patterns

for horizontal loops (diameters 1.7m and 0,8m),

frequency 14.15 MHz, compared with a horizontal

dipole, antenna heights10m. A part of the gain

difference, loops compared with dipole, is that the

dipole has directional gain, the horizontal loop has an

omni-directional pattern.

In the figure below we show measured gain (dBi) for

a commercial loop (vacuum variable capacitor used

to tune the loop), for a NVIS path (100 km length),

compared with theory (numerical modeling using

NEC).

THE CROSSED FIELD ANTENNA

Professor Maurice Hately, G3HAT, Brian Stuart, and

Fathi Kabbary, a student of Maurice’s have dreamed up

a super controversial antenna. Not only has this

antenna type, and the new EM theory developed by

them to explain how the antenna works, confused the

amateurs in radio, but this antenna type has been

patented, and as well presented to a learned audiences,

the IEE in an Antennas and Propagation forum, and the

IEEE Broadcast Technology Society, but it has

attracted wide attention and is being sold.

For the amateur in radio, he wants to know the

space wave gain for distance communication links.

The Figure below shows the calculated gain in free

space for three AMA Loops.

The original CFA concept was a cylinder over a disk,

both fed, and fed in phase quadrature. The disc was

said to generate an H-Field, and the cylinder an E-field,

and these fields generated an out-going Pointing vector.

E/H, by adjusting the power fed, was said to be 377

ohms in the near field, which is what E/H is in the far

field --- perfect coupling to the propagation medium.

The intrinsic impedance of free space is 377 ohms.

Plik z chomika:

Inne pliki z tego folderu:

Inne foldery tego chomika: