Pascale ...
I faced the same problem modelling a CDDA (Wullenweber)
antenna - 64 doublets driven at different phases to achieve beam
steering.
My solution was to write a program in Turbo Pascal (could be
some other language) to create the NEC input files. Most of the
program just copies the ASCII records for the card images, but the EX
card write statement refers to variables which contain previously
calculated real and imaginary components of the drive voltages.
E.g., ...
writeln(f,'EX 0 1 6 0',e1.real:10:2,e1.imag:10:2);
writeln(f,'EX 0 1 6 0',e2.real:10:2,e2.imag:10:2);
writeln(f,'EX 0 1 6 0',e3.real:10:2,e3.imag:10:2);
etc ...
I later used the same technique for a CDDA modeled with
transmission line delays rather than forcing the phases. The line
lengths were read from an external file to create a NEC input file for
a given disposition of delay lines. The file creation/modification,
the NEC run and a process to extract the 'goodness' of the resulting
pattern was put in an optimising loop that updated the delay line data
to converge on the delays for best antenna pattern across the band.
All done in DOS, running unattended for a few weeks!!!
A similar process allows me to write multiple NEC input files
with different geometry (eg dipole length or height), or different
segmentation with frequency across the band when I run multiple NEC
jobs - anything up to 3000 NEC runs in a single batch - It works well
and saves a lot of typing!!!
Cheeeeeeeerz
Alan.
**************************************
Alan Nott BEE, CEng, MIEE
Senior Technical Specialist Engineer
Electromagnetics Cell
Software and Systems Engineering Unit
Army Engineering Agency
Tel + 61 3 9319 5889
Fax + 61 3 9319 5909
email: alan.nott_at_aea.sptcomd.defence.gov.au
Received on Fri Jun 30 2000 - 04:26:15 EDT
This archive was generated by hypermail 2.2.0 : Sat Oct 02 2010 - 00:10:40 EDT