You can just model it with loads over a reasonable range and pick the
value that gives the smallest SWR. If it takes a long time to run
the model, you can enter the load with a NT command. Then you can
try different loads without the code repeating the matrix fill and
factoring.
If it still takes a long time or you want to get a very good match,
you can model the transmission line with a 1 V source and
short-circuit termination, and then with 1 V at the termination and
short-circuit at the actual source point. Then you can can
characterize the TL between source and termination points as a
two-port network with short-circuit admittance parameters. Save all
of the segment currents to files for each of the two excitations.
Then you can write a code separate from NEC to determine the voltages
at the source and termination points for each load value. A load is
just a voltage source with voltage proportional to current. After
you determine the voltages for a given load value you can add
together the currents for 1 V at the source and 1V at the termination
multiplied by the actual voltages to get the current along the TL.
You need to write a routine to estimate the SWR along the TL, maybe
averaging a difference approximation of the second derivative of the
magnitude of current over a section of the TL. Then you can couple
with an optimization routine to minimize the SWR. That should give a
nearly perfect match if you allow complex loads. If you have
multiple terminations on your multi-conductor TL you would need more
that a 2-port network model.
Jerry Burke
LLNL
-- The NEC-List mailing list NEC-List_at_robomod.net http://www.robomod.net/mailman/listinfo/nec-listReceived on Tue Oct 06 2009 - 19:54:55 EDT
This archive was generated by hypermail 2.2.0 : Sat Oct 02 2010 - 00:10:46 EDT