PC board

>From Ben Hadj:

>I'm using NEC2 to calculate the Electricfield radiated by a PCB card.
>in the output file, the POWER BUDGET is :
.................

Dick Adler pointed out (by phone) a potential problem in the circuit
board input file that I had not noticed. That is the use of EX5... for
voltage sources. This "biconical" voltage source, at best, is only
accurate for very thin wires. For thin wires it may give a more
localized voltage source than EX0... For thicker wires the field
distribution cannot be more localized than several times the wire
radius (due to the thin-wire approximation), and the accuracy goes bad
rapidly with increasing radius or decreasing (segment length)/radius
ratio. A source model that works for thick wires, and is related to the
bicone (EX5) is the coaxial-line source used by Popovic, et al. ("Analysis
and Synthesis of Wire Antennas", Research Studies Press). The
coaxial-line source is difficult to use with thin wires, and is not in
NEC.

The EX5... source probably should never be used without conducting
numerical experiments to check its accuracy in the model. In this model
with resistive loads it is particularly bad for computing P_in - P_loss,
since the load is always a "voltage source proportional to current"
using the applied-field (EX0) source model. Both voltage source and
loads should use the same applied-field model to get an accurate
difference.

Jerry Burke
LLNL
Received on Wed Apr 17 1996 - 00:02:00 EDT