NEC-LIST: CFA CEM Modelling --- 2

CFA-CEM Modelling --- 2
________________________

I have revised my model for the original Egyptian monopole type CFA:
smaller grid size (0.00047-wavelengths), so the disk looks more
circular; a square hole in the centre of the disk (the photograph for
the antenna shows a circular hole in the centre (about 0.3 m ?)); and
the right diameter (4 m).

This changes the antenna's impedance significantly, but the gain is
not much changed, *excepting* I had to change the relative phase of
the feed wrt my ealier coarser grid model for the disk (here cylinder
-45 degrees; disk 45 degrees). Again, if we change the phases ( 45
degrees and - 45 degrees respectively) we calculate a gain of - 99.99
dBi.

The antenna's impedance at 1161 kHz follows:

        Z (cylinder) = 187 - j 6

        Z (disk) = - 159 - j 11,880

and Gain = - 18.6 dBi.

Colin Davis [1993] measured - 23 dBd for a VHF dipole version.

This looks like a pretty impossible antenna to feed (according to
NEC-4D); and the gremlin turning the crank to provide an antenna with
gain has gone to lunch.

I am not sure what to expect, since clearly this is a most interesting
complex antenna, however comment follows.

1) The cylinder which is (looks like) a very short monopole seems to
be resonant; with a high input impedance (!!!).

2) If you can succeed in feeding the disk with same current, its
impedance is largely a capacity, which seems reasonable --- excepting
its input resistance is negative, and so you have take power from this
element.

3) Do not trash my results because a radiation resistance is negative.
We frequently find a negative radiation resistance associated with one
(or more) towers in a closely-coupled multi-tower MF broadcast array,
if we force currents feeding the towers to have a particular current
magnitude and phase [c.f. Belrose, in The Handbook of Antenna Design,
Volume 2 (editors Rudge, Milne, Olver and Knight), published by the
IEE, 1983, pp. 614-616].

Finally, concerning impedance, Hatley, Kabby and Khattab [1992] show a
SWR = 1:1 for a 50-ohm source, for a CFA with *no* matching (Figure
8), and, 30 kW of power feeding each antenna element (???). Figure 3
of that paper shows no connection to ground (unless the phasing unit
is grounded to the elevated counterpoise mat). There must be currents
flowing everywhere: on the wire connecting the elevated mat to ground;
and on the outside surface of the coaxial cable feeding the antenna;
perhaps on the quarter wave tower previously used, unless it was
detuned.

There must clearly be something the authors are not telling us. Or
NEC-4D is giving us crazy results --- and I do not believe that.
Making no change to the model, but changing the phase of the currents
leads to different results, in particular when the phases are zero (or
the same) the impedances seen by the sources is about what we expect.

I will continue to be interested in this complex antenna, but I do not
think I will not rush out and buy or build one.

John S. Belrose, VE2CV
30 March 1999

_____________________________________________
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 Tue Mar 30 1999 - 20:21:28 EST