Modeling a shorted stub in NEC via the TL facility requires the
addition to the model of a distant wire (a short, 1 segment thin wire
will do--well out of range of having any affect upon the model proper)
as a terminator for the TL specified. A shorted stub wire should have
high conductance values. This process is automated in some commercial
implementations of NEC. Viewing the resulting NEC file will give a
good indication of the technique involved.
However, all TLs should observe the restrictions--the key one of which
is that they have equal currents on each side of the segment
containing the TL entry. Otherwise, the line will not accurately
reflect reality. For example, the center of a 1/2 wl element works
well, but half-way out the element toward the tip, TL lines become
inaccurate with respect to the length required for a given Zo.
For impedances above about 300 (depending upon wire size), as Jack
Belrose has noted, one may construct accurate shorted stubs using
modeled physical wires. For an application where the physical
dimensions are more significant than the precise line impedance (such
as the quad reflector problem noted), direct construction is the
shortest route to the solution. Where some sense of the precise
characteristic impedance of the line is required (for bare wire
lines), it may pay to model the line and determine its impedance.
There are small differences between the standard short equations for
calculating lines from desired Zo values and vice versa and the
results returned by physical models of the lines. For many
applications, these differences are too small to make a difference,
but for other problems, they may be important. When combined with an
antenna structure, be careful of unequal diameter wires at angular
junctions in NEC, although MININEC is generally accurate relative to
such junctions.
There are some notes that may be of use to those first using
transmission lines in either the TL or physical form in the antenna
modeling series at my web site (url below). I hope this is useful.
Regards,
L. B. Cebik
L. B. Cebik, W4RNL /\ /\ * / / / Tel: (423) 938-6335
1434 High Mesa Drive / \/ \/\ ----/\---
Knoxville, Tennessee /\ \ \ \ / / || / http://www.cebik.com
37938-4443 USA / \ \ \ \ || e-mail: cebik_at_utk.edu
Received on Thu Feb 17 2000 - 10:02:59 EST
This archive was generated by hypermail 2.2.0 : Sat Oct 02 2010 - 00:10:40 EDT