AF0D's Website Amateur Radio and More

Antennas are a lot of fun to build, provided you have an inkling about how to do it. While it can be exactly as easy as just throwing up a long piece of wire (it's called a LongWire antenna, and can work quite well), generally, better results are achieved when one puts in a little analysis first.

The first thing that a ham used to do, back before VHF and UHF became popular, was to put up a dipole. Nowadays, a ham's first homebrew antenna is often a J-Pole.

Either way, it helps to be able to predict what it might do before going to all the hassle of actually making one. That's where antenna modeling software comes in.

Lawrence Livermore Labs wrote a program called NEC, which stands for Numerical Electromagnetics Code. It calculates currents on the wires in an antenna, and from that calculates a far-field power gain. It comes in two variants, NEC2 and NEC4. NEC4 is newer, costs money, and does a few things better than NEC2. For most amateurs, NEC2 is fine.

Anyway, all of the gain plots that I have on this site were produced using NEC2, run on Linux or FreeBSD. Since the output is not suitable for direct display in a plot, I wrote a little perl script to massage the data and then output it in a format suitable for gnuplot. The perl script, "plot.pl", can be found here.

On the scales used

You'll find two different plots used on all my pages; both the ARRL standard plot and an Effective Radiated Power (ERP) plot.

The ARRL plot is great for looking at half-power beamwidth and front-to-back ratio; but for looking at the true pattern, or for comparing antennas for gain, it isn't too good. It seems to really place too much emphasis on side lobes, a result of the odd scaling that it does.

An ERP plot simply uses the equation for Decibels:
$$

dbi=log


P2


P1

Db = log(P₂/P₁)

and solves for P₂. P₁ is the input power, which is up to me to choose. Since I like simplicity, I choose 1, and the equation becomes quite easy.