Re: NEC-LIST: CFA

Brian,

Grand Bingeman has forwarded the message you wrote to him promoting
the CFA. As you say the CFA has so far not received a good press.

I copy below two statements you made, and my response.

First point:
-------------

"In all of this, one must also get the ratio of E/H to match space
impedance for maximum power transfer, so you can appreciate that 90
phase and little power is actually a space impedance matching
problem."

The idea of *impedance matching* to the intrinsic impedance of free
space has been with us, and been kicked around, for as long as I can
remember. Perhaps you did not see my posting on the CFA, see below.

In fact it is for this reason that I have been from the outset
skeptical about the CFA, and I remain skeptable.

Second Point
-------------

"Now, in terms of *conventional theory*, well the CFA is fundamentally
different and standard antenna formulae and techniques cannot be
invoked when E and H exist close to the structure in time
phase. Conventional theory cannot accomodate this fact and thus should
not be applied. I suppose it is obvious from another
viewpoint. Conventional monopoles have one signal feed to the antenna
structure and an earth. The GP CFA has two signal feeds to two
structures and an earth. There is no way conventional theory of
monopole structures can be transferred directly without alteration to
another type of antenna."

You keep referring to conventional theory, and you further imply that
conventional theory based on Maxwell's equations cannot be used to
predict performance of the CFA.

You are wrong!!! NEC could indeed predict performance of the CFA ---
the only difficulty (for me) would be in creating the model. We would
have to model the plates as a wire grid, and the large (presumably
closed ??) cyclinder as a wire grid --- but having done so we could
then numerically model the antenna.

Maurice Hatley starts out with Maxwell's equations, as a basis for
explaining how the antenna works. So where does Maxwell go wrong, and
we have to invoke some new (divine) theory??

I have been modelling antennas for years, antennas which employ
multiple sources. There is no problem with multiple sources, the only
caution is to consider carefully the phasing, as you say for the CFA
--- since wires are used to model the antenna we have to be careful
when we model different parts of the antenna, that end one of a wire
on which we have placed a source has the right physical arrangement
with respect to end one of a second wire on which we place another
source.

I have modelled many multiple tower MF antennas, in which each tower
is fed by a different current, in amplitude and phase; and multiple
tuned VLF antennas in which a number of umbrella top loaded towers are
fed with equal currents, in phase and amplitude, to simulate a point
source (since the wavelength is so long compared with tower
separation). There is no problem in modelling such antennas.

Conventional theory can rigorously predict performance of the CFA ---
the only unconventional aspect is the physical arrangement of the
antenna --- comprising plates (circular plates) and cylinders, which
are fed separately from a basic source.

And, finally, let me quote Colin Davis (see reference [4] below):

"To sum up, the field strength measurements made at the receiver site
showed that the CFA radiated signal levels were consistently 23 dB
below those made using a dipole (or worse), clearly showing that the
CFA was not operating efficiently. The required feed phase angle of
90-degrees was arranged for both the forward wave signals and the
resultant signal at the plate terminals. In each case (he tried
impedance matching with different baluns), no signal improvement or
dip in VSWR was observed when sweeping the phase angle around the
90-degree point. The results demonstrated that the phase angle was
unimportant to the operation of the antenna --- at least under these
conditions --- contrary to Hately's suggestions."

Where did Colin go wrong??

That is why I keep saying that the inventors of the CFA are witholding
something, providing it is true that the performance of antenna in use
in Egypt and Cyprus (the second successful (??) instalation is news
for me) is so superior. Originally it was claimed that the CFA in
Egypt produced field strengths comparable to that for the previously
used quarter wave monopole --- now it seems that improvements(??) have
been made and the CFA is 6dB better than a quarter wave!!!

What are we to believe??

Regards, Jack

ANNEX
------

The Cross-Field Antenna
Should we go for it or let it Rest in Peace?
_____________________________________________

An antenna is sometimes described as a coupling device, coupling
electromagnetic energy to space, and following on with this line of
thinking, antenna amateurs/specialists have wondered whether there
were methods of achieving this process in more efficient ways than
provided by use of monopoles, dipoles and loops. The performance of
electrically small antennas is a subject that has been of interest for
me for more than 50-years.

For example, can we devise an antenna system that will radiate only a
true Zenneck-wave, and thus provide extended ground-wave coverage,
without the interference effects of a skywave? The answer to this
query is no. Since a spacewave arriving tangentially at the surface
of the earth will generate a ground wave, a ground wave will most
certainly generate a spacewave. And, concerning more efficient
coupling to the propagation media, most ideas that have so far emerged
are hoaxes. In one, where one considers an antenna as an opened out
transmission line, it is claimed that the characteristic impedance of
the antenna should be made equal to the intrinsic impedance of free
space (377 ohms). This is nonsense. In another, it is claimed that
the antenna's radiation resistance should be made equal to the
intrinsic impedance of free space. This latter statement has even
less meaning, since the antenna's radiation resistance for a specific
antenna depends on where you decide to reference it on the antenna
structure.

Continuing, in 1989 (see Electronics + Wireless World, March, July and
November, 1989, and December 1990) and in following years, Maurice C.
Hatley, GM3HAT describes a cross-field antenna (CFA), in which
quadrature E and H fields are separately generated. A follow on paper
[1] offered an explanation of its properties in terms of "Poynting
vector synthesis". In the words of the inventor, reversing the form
of Maxwell's equations led to the "realization and development of this
revolutionary new antenna system". Quite small versions of it have
(apparently) been fabricated and tested, cf. reference [1], to
demonstrate its ability to efficiently couple EM energy to space ---
but, notwithstanding, in my view this antenna also fits the category
of a hoax, since attempting to generate E X H in the antenna itself,
and claiming that such an antenna more effectively couples to the
space surrounding the antenna is about a fruitless as the other ideas
discussed above.

Proving beyond any doubt that a theory is completely wrong is very
difficult [c.f 2, and inventor's reply 3]. But it is possible to
conduct experiments as scientifically and impartially as possible, and
to draw firm and well founded conclusions from them under the test
conditions [4] --- but even here we have inconsistency --- since the
result obtained depends on who is doing the experiment, the inventor
or someone else.

The Hatley, Kabbary and Khattrab paper claims that the field strengths
and, service area for a small (1.6 percent of a wavelength high) MF
CFA in Egypt is identical to that provided by a 75 m quarter
wavelength vertical.

The Egyptian antenna is a ground-plane (GP) type CFA on the roof of
the transmitter building, and after listening to the presentation of
that paper, I questioned whether the radiation realized was in fact
generated by the GP CFA, or by currents flowing on the outside surface
of the coaxial cable feeding the antenna. I also wondered whether the
previously used quarter wave vertical antenna might indirectly
(induced currents and re-radiation) be a part of the antenna system.

At a more recent IEE ICAP Meeting (Edinburgh, April 1997), I was
presenting a paper on elevated radials, I met Hatley and his Egyptian
colleague (Khattrab), or should I say they cornered me; and I learned
that the Broadcaster was happily still using the antenna --- or should
I say the authors still believed in the results of their experiment
--- in retrospect I am not sure now which is which.

Maurice told me that he was developing HF versions of this antenna for
use by radio amateurs, for mobile communications, and he would send me
details on it. I told him I would be glad to measure the radiation
efficiency of his amateur version --- but, no correspondence or offer
of an antenna has to date been received.

In 1992 Colin Davis, University of Surrey [4], conducted a study of a
VHF dipole version of the CFA, his model was a 50-percent scaling of
Hatley's original antenna. His work, carried out to investigate
whether the CFA does operate as an efficient radiator, casts doubts
about it, which seem overwhelming.

Despite his best efforts to make it work, the best Davis could
achieve, with respect to gain was - 23 dBd ["The great tragedy of
science --- the slaying of a beautiful hypothesis by an ugly fact",
T.H. Huxley].

The article by Davis starts off by a consideration of the impedance of
the antenna, since power cannot be successfully coupled to the antenna
unless its impedance is known. Little information is available in the
literature to indicate what value this might be. It is interesting to
note that a discussion with Hately [Davis-Hately, private
communications], suggested that both input impedances of the CFA
antenna (E-plate and D-Plate parts of the antenna) should be around
300 ohms --- clearly referencing performance to the intrinsic
impedance of free space.

Smith [5] has considered input impedance. He considers that the
E-plates should have the characteristic impedance of an electrically
small dipole; and the D-plates are nothing more than a capacity, the
D-plates having a similar capacitive reactance to the E-plates. Smith
also concludes that there is nothing magical about the cross-field
antenna, since it will encounter the same efficiency-bandwidth
limitation, as for any electrically small antenna [6].

And, Smith makes reference to my article [7] which reveals that the
hula-hoop antenna (earlier called the DDRR antenna) is nothing more
than a resonate inverted-L --- and so the efficiency of this antenna
becomes vanishing small as the height (above the ground plane) of the
horizontal part of the hula-hoop becomes small.

So, in conclusion, is the debate concluded --- should the CFA RIP?

73, John S. (Jack) Belrose, PhD. Cantab, VE2CV
john.belrose_at_crc.ca
10 February 1999

References

1. M.C. Hately, F.M. Kabbary and Khattab, "An Operational MF Antenna
using Poynting Vector Synthesis", Proceedings of 7TH International
Conference on Antennas and Propagation, Part 2, Conference Publication
No. 333, April 1991, pp. 645-648.

2. K. Donaldson, "Misguided CFA", Electronics World+Wireless World,
October 1992, pp. 837-838.

3. M.C. Hately, "CFA -- no tricks", Electronics World+Wireless World,
December 1992, p. 1007.

4. Colin Davis, "CFA --- RIP?", Electronics World + Wireless World,
May 1993, pp. 405-407.

5. M.S. Smith, "Conventional Explanation for 'Crossed Field Antenna'",
Electronic Letters, 28, 13th February, 1992, pp. 360-361.

6. H.A. Wheeler, "Fundamental limitations of small antennas",
Proc. IRE, 35, 1947, pp. 1479-1484.

7. J.S. Belrose, "Transmission line low profile antennas", QST,
December 1975, pp. 19-25.

8. J.M. Boyer, "The Hula-hoop antenna: a coming trend?", Electronics,
11, January, 1963, pp. 44-46.

____________________________________________
John S. (Jack) Belrose, PhD Cantab, VE2CV
Senior Radioscientist
Radio Sciences Branch
Communications Research Centre
PO Box 11490 Stn. H
OTTAWA ON K2H 8S2
CANADA
TEL 613-998-2779
FAX 613-998-4077
e-mail <john.belrose_at_crc.ca>
_____________________________________________
Received on Wed Feb 17 1999 - 02:49:44 EST