NEC-LIST: RE: Small antennas

Prof. Grimes,
 
Thank you for your reply and the further clarification. In the antenna
community I think the emphasis is slightly different from what you said, and
it is that Q is accepted as a way of measuring bandwidth. The classic
"small" antenna problem is VLF transmission where extremely large but
electrically small antennas provide very small bandwidths, a Q of 100 being
typical for an antenna supported on many large masts. The end user needs
bandwidth, for obvious reasons (but not at the price of reduced efficiency).
 
A practical antenna has to be driveable by a single source, so when there
are two or more ports there will usually be a power-splitter network and
there is then a measureable bandwidth at the single input port, although I
understand that is a complication that would be rather unwelcome in an
already complicated analysis. Where the elements are fed by separate
synchronised sources, the issue remains how well each element is matched to
its source when the frequency is changed.
 
You mentioned a phased array. In designing phased arrays the problem of
mutual coupling between elements is very real, and the difficult trick is to
maintain a good input match under all conditions of scan angle. This is a
case where the dipoles are usually parallel, and they couple to one another
so that the input impedance of each dipole depends strongly on what is going
on in the other dipoles. I do not think something similar occurs between an
x-directed magnetic dipole and an x- or y-directed electric dipole (both
antennas assumed at the origin) because the fields distributions do not
allow it, either in the near field region nor the far field. I was
therefore questioning the operation of your 2-element antenna because
without mutual coupling I could not see the input impedance of either
element being affected by the current in the other element, and if the input
impedance does not change, the Q (in one sense) does not change either, even
though the stored field energy might change. I understand there is some
necessary asymmetry in your NEC model but it would be a pity if the coupling
arises only as a result of the asymmetry. I would welcome any clarification
you have on this point.
 
Kind regards,
 
Alan Boswell

-----Original Message-----
From: Dale M Grimes [mailto:dmg6_at_psu.edu]
Sent: 25 January 2003 16:48
To: alan.boswell_at_baesystems.com
Subject: RE: Small antennas

Dr. Boswell,

    These answers are based upon my interpretation of your points. If I
have misunderstood please let me know.
     I believe your first paragraph implies bandwidth is a perfectly good
method of measuring Q that is widely accepted in the antenna community, and
we don't use it. Why not? If our arguments are correct, why not show them
in that way?
        In response, there are two general methods of determining Q of an
oscillating object: bandwidth and stored energy. In EM one commonly
measures bandwidth then solves for Q using the Q-bandwidth relationship of a
simple RLC circuit (a lossy harmonic oscillator.) Your "how to" description
is exactly correct and, for an antenna driven by a single terminal pair, the
most convenient thing to do. Although we would have liked to do just that,
implementation of our conceptualization requires driving the system with a
minimum of two and a maximum of four terminal pairs. The frequency
dependence of the input impedances of unlike modes (TM and TE) are quite
different and like modes are ninety degrees out of phase. How does one
combine four such terminals to obtain a single bandwidth measurement? In
our limited circumstance we saw no way. We did see a way to measure Q via
energy storage; of course, the RLC circuit antenna model shows the
calculated Qs are the same. Therefore it was a practical matter, not an
arbitrary choice, to use the returned energy definition.
        Although Collin's Q analysis and hence those of Harrington and Fante
are energy-based, not in the original paper nor any subsequent paper have I
seen any mention of the underlying assumption, critical for their results,
that ALL stored energy returns to the source upon modulation changes.
Simply put orthogonality of the fields is a woefully insufficient argument;
orthogonality deals only with the total values. Consider a phased array of
dipoles as an example. The power and energy fields of each unit of the
array are orthogonal but the nonlinear field interactions produce different
results depending upon the relative orientations and phases of the source
dipoles. The obvious result is directed power flow. It is easy to show, as
we do in our book, that near energy field configurations are dependent upon
the driving orientations and phases. Therefore there is NO a priori
assurance that the modulation response of shifted local energy fields is
independent of the driving conditions. We are the first, and only, persons
I know who attempted to use such a measurement to measure these effects, and
we find quite the opposite: In some cases the energy returns to the source
and in others it becomes outbound.
        I'm uncertain of the full point of your second paragraph but I agree
with all your words. If I do understand it, note that analysis of this
configuration of superimposed, non-interacting radiating elements is much
easier than physical implementation. The implementation shown in the MOT
paper is the best we could do within our time and budgetary constraints. (I
don't mean to imply we could or could not do better with more assets.)
However the fact is implementation constraints coupled with driving
constraints to force us to a non-ideal implementation that produced the
near-field coupling you describe. Again, if you think we don't agree with
either paragraph we haven't made our point clear and I would appreciate your
letting me know.

Best regards,

Dale Grimes

***************************************************************************

This email and any attachments are confidential to the intended
recipient and may also be privileged. If you are not the intended
recipient please delete it from your system and notify the sender.
You should not copy it or use it for any purpose nor disclose or
distribute its contents to any other person.

***************************************************************************

-- 
The NEC-List mailing list <nec-list_at_gweep.ca>
http://www.gweep.ca/mailman/listinfo.cgi/nec-list