RE: NEC-LIST: CFA modeling results

>Jack,
>
>If I read correctly regarding your comments on phasing,
>you are saying (from NEC simulation results) that CFA
>works somewhat when disc leads cylinder by 90 degree
>and not at all at other phase difference. What if two
>feeds are in phase?
>
>My 2nd interpretation is that disc has to lead the cylinder
>by 90 degree only in those 3 specific ways otherwise
>(such as -20 and +70) it won't work. This would be
>strange unless there is something serving as phase reference.
>
>Regards, Feng

Feng,

I have perhaps said too much. Let me recap.

To function like a CFA the cylinder and the disc must be fed in
quadrature, quadrature feed having the right phase relation. I have
in general used minus/and plus 45-degrees, for the cylinder/and disc,
since the Hatley et.al. (ICAP'91 paper) was fed in such a balanced
arrangement.

But one can of course employ any other phasing, such as -20/+70
degrees as you suggest --- or in fact 0/90 degrees.

But reversing the signs reduces the gain to zero.

The ICAP'91 paper suggests that the phasing is very very very
critical. Figire 4 in that paper shows that the output power
decreases by 3 dB for phase differences of minus and plus 0.1 degrees
(7 dB for minus and plus 0.15 degrees) from 90-degrees.

This figure is ridiculous. For real antennas one cannot measure or
maintain phase differences to this accuracy.

For my model I can change the phase by minus and plus tens of degrees
before anything appreciable starts to happen --- but as the phase
difference approaches zero, gain and impedance change abruptly. For
0/0 degrees the source impedances are closely identical to the self
impedances of the cylinder and disc parts of the antenna system alone.

For -90/90 (or 90/-90) degrees, the real part of the impedance for the
cylinder (or disc) becomes vanishing small.

The Hately ICAP'91 paper shows phases of 45/-45 degrees, opposite to
that for my numerical model --- but the CFA is a complex antenna
system and what one gets is model dependant.

For my original wire grid model I did not have radial symmetry for the
wire grid disc, and (curiously) the phases had to be 45/-45 degrees,
and the direction of the circulating power is reversed from my present
model.

All my current results are for a disc having radial symmetry, see

<http://members.home.net/propagation/cfa.html>

All this confirms what I have been saying for some time: The CFA is a
very very poor radiator (according to my modelling), but it is an
antenna of some academic interest.

At least I find it so. As I said before:

The CFA has source impedances of 150-200 ohms (plus and minus), but
the antenna behaves like that for an antenna system having radiation
resistances equal to fractions of an ohm.

The resistive part of the impedances for the CFA is only one
transmitter doing work on the other. The one that sees a positive
resistance may be happy, but the other is having to eat all of that
power plus any that it generates. This is no different than the
reactive power that surges out of and into a single source. (But
Hatley, Stewart and Kabbary say that the near induction fields are
very small. So this is another point on which we disagree.)
(Continuing) It may be a worse problem, since it goes into somewhere
different than where it came out of. In the NEC model with no loss it
is implied that the power absorbed by the source that sees negative
input resistance is recycled, but that might be hard to accomplish in
the real world. The V/I at one source cannot be simply interpreted as
an impedance when there are other sources operating. That is why in
the message that that Jerry Burke sent to nec-list he used radiated
power/I**2 as a sort of "radiation resistance" for each source.

Regards, Jack

_____________________________________________
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 Thu Jul 15 1999 - 16:11:47 EDT