Jim Lux wrote:
Cecil Moore wrote:
John Smith wrote:
I don't believe the above is correct.

Proof that NEC cannot model everything is at:

http://www.w5dxp.com/SUPRGAIN.EZ

Would you believe a vertical with
24 dBi omnidirectional gain?

Got a regular NEC deck or text description (can't read the EZ-NEC file..)

Although NEC can't model everything (for example, it can't model a patch
antenna containing dielectric), this model is not at all proof of that.
All it shows is that models can be carelessly or intentionally
constructed in such a way as to cause NEC to malfunction. An experienced
EZNEC or NEC modeler would immediately run an Average Gain test and seen
that the reported gain is about 16.7 dB too high due to numerical
problems. If desired, this can be subtracted from the reported gain to
give the gain that's much closer to reality.

A resistance, the jagged schematic symbol that we use in basic circuit
analysis, can be and often is used as a simple model of a resistor. But
anyone who has ever done any electronic design or analysis quickly
discovers that this model is adequately accurate only under some limited
group of conditions -- for example, it's a lousy model of a leaded
resistor at 10 GHz --, and any calculations made using it will be far
from measurement of a real resistor (assuming the measurements are
capably made). Like the resistance and all other models, NEC also has
limitations and boundaries over which it's valid. And it can be
misapplied to produce just as egregiously bad results as using a
resistance to model a leaded resistor at 10 GHz. In this case, the model
violates at least one NEC rule by spacing two 40-foot #14 wires 1/2 wire
diameter (0.032") apart. (NEC guidelines specify that parallel wires
should be at least several wire diameters apart.) It would be a great
challenge to actually construct this antenna. But, even at that, the NEC
results are probably not bad once the average gain correction is made.

Here's the model in .NEC format for those who don't have EZNEC. I've
changed the current source to a voltage source which simplifies the
model without making any difference in results.

CM 40m triangular loop
CE
GW 1,31,.003048,0.,3.048,.003048,0.,12.192,8.138E-4
GW 2,53,0.,0.,12.192,0.,16.4592,12.192,8.138E-4
GW 3,61,0.,16.4592,12.192,0.,0.,3.048,8.138E-4
GW 4,31,0.,0.,3.048,0.,0.,12.192,8.138E-4
GE 1
LD 5,1,0,0,5.7471E+7,1.
LD 5,2,0,0,5.7471E+7,1.
LD 5,3,0,0,5.7471E+7,1.
LD 5,4,0,0,5.7471E+7,1.
FR 0,1,0,0,7.
GN 2,0,0,0,13.,.005
EX 0,1,16,0,1.414214,0.
RP 0,181,1,1000,90.,0.,-1.,0.,0.
EN

Roy Lewallen, W7EL