NEC-LIST: Trouble with dipoles

Tiffany,

A dipole does not resonate at a half wavelength, but at a somewhat
lower frequency, depending on the wire radius. You did not say what
method you were using to model a thin wire in FDTD. I have gotten
good agreement with NEC using the transmission line equation method
described by Holland (IEEE Trans. EMC, May 1981) and Taflove's method,
building in the 1/r field behavior. To get good agreement over the
first three resonances, I used 39 segments in the NEC model and 38
cells in FDTD, since the transmission line equations give an effective
extra half cell length on each end. I excited the center three
segments in both models with 1/3 volts, since the wider source gap
makes the effective gap capacitance (which may differ between the
models) less important. I think if you just set a row of cells to
high conductivity in FDTD the effective wire diameter is somewhat less
than a cell width.

Jerry Burke
LLNL
Received on Fri Oct 30 1998 - 10:57:34 EST