On Tue, 13 Jun 2006 09:51:43 -0700, Richard Clark
wrote:


As this discussion has been largely coaxial based, the outer
conductive surface of the coax is the primary imbalance to a dipole
through its "third wire" connection at the dipole feed point. It
appears as one of three wires to the abstract source established at
that feed point and it presents an ad-hoc Z load. The value of this
load is rarely determined, except when one deliberately attempts to
make their feed line 1/2 wave long (or some multiple). Of course,
that means 1/2 wave for the velocity factor of the outside conductive
path of the coax. This is often accomplished through cut and try
rather than modeling or measuring currents, but these too would be
good first pass approximations.

Richard,

There is a focus on evaluating baluns as a passive component in a
bench test jig designed to characterise them by a simple equivalent
circuit to show their imperfections in certain scenarios (eg impedance
range, frequency range, common mode impedance with different balance
points on the load and floating loads). Collectively, we seem to have
done that to death. To date, I don't think anyone has discovered a
practical balun that is close to ideal in all applications.

Baluns (being any device that is designed to facilitate the transition
from un-unbalanced load (might be perfectly balanced, might not be) to
an unbalanced transceiver (one terminal approximately grounded) seem
to be usually employed to:
- reduce common mode RF currents entering the shack where they may
disrupt operation of equipment, and more recently may be a health
concern;
- reduce the feedline's participation in radiation or reception
(pattern distortion, unwanted noise pickup, EMC / proximity to other
equipment).

It seems to me that NEC models are a worthwhile tool in developing
insight into the effects that occur. I think NEC modelling of a coax
fed, centre fed dipole at various heights as a centre fed conductor
with generator at the centre of two wires, and the "third wire" to
ground through some loss resistance is quite revealing. (The coax
example is not to suggest that this phenomena is peculiar to coax feed
alone.)

You mention measuring currents. It seems that discussion on common
mode currents has spawned a new market for clip on RF current probes.
I even see suggestion that common mode current be continuously
monitored in much the same way as VSWR is monitored at a spot adjacent
to the tx.

NEC modelling reveals that the currents on the "third wire" (common
mode current) varies with position (no surprise there), and that
depending on the topology, can be insignificantly low at some points
while it is significantly high at other points. Measurement of common
mode current at just a single point does not necessarily provide
enough information to detect or assess a common mode current problem.
Spot measurement is a superficial approach.

So if there are maxima and minima in the common mode standing wave
current on the feedline, the influence of chokes on current in all
parts of the antenna system (hence pattern) and loss of chokes will
depend on where they are located.

That is not to suggest that it is all too complicated. I think there
are good reasons to routinely deploy baluns of appropriate type an
location, but they aren't a cut and dried idealised solution and
further work may be required to identify and rectify residual
problems.

Owen
--