XGH9, centered at Channel 25

Since I'm still trying to get to my goal of WCNY reception, I've developed a variation on John E. Davis's 'Extended Grey Hoverman' design, which trades bandwidth for gain.

In this design, the six squares and 9 pairs of reflectors get an effective gain above 18 dB at channel 24 and 25, with a reasonably gradual drop off on the lower side of channel 24. Now I just have to build this and test it in the real world.


XGH9-ch25.nec

Antenna 'GH6 Channel 25 A', 6 gauge copper elements and 3/8 inch aluminium reflectors, optimized for UHF channels 15 through 25:

Antenna 'GH6 Channel 25 D', 6 gauge copper elements and 1/4 inch aluminium reflectors, optimized for UHF channel 25:

Two different optimization strategies, two very different gain curves:

While it's not the original Gray Hoverman Antenna design on the web, Dr. John E. Davis (aka j3d) has done his fair share to advance the state of the art in this particular antenna design.

Through numerical methods, he has developed a series of models that specify optimized antennas for the post-transition UHF Spectrum, using various numbers of reflectors (The GH4, pictured here, uses four mirrored pairs of aluminum reflectors.)



He has also developed a few 'gapless' Gray Hoverman antennas, where the reflectors in the back of the antenna are unbroken. The one antenna I've physically built is a poorly executed version of one of these antennas, with 12 quarter inch diameter aluminum rods reflecting onto a pair of 6 gauge copper antenna elements.

Antenna Modelling

The first step down the path of antenna design I took was to get to know the NEC2 program (or since it's FORTRAN, 'codes'), and the free Windows GUI, 4NEC2.

NEC2 is a 'Method of Moments' simulation program developed under contract from the US Navy in the early 1980's which is now in the public domain, and runs on just about any modern computer. For my purposes, I use either the legacy NEC2 program, modified to calculated with double precision floating point instead of single, running on a Fedora Linux machine, or the great 4NEC2 front end on Windows.

4NEC2 is a labor of love from Arie Voors, and includes all sorts of niceties such as graphical modelers, visualizers, automated optimization tools, and model error checkers. Until I can be bothered to successfully procrastinate on other parts of my life to write a Linux front end that doesn't crash on me consistently, this program is almost always running on my windows machine.

Both of these options work on input files that are best described as archaic.
For further reading on these input files, consult the NEC2 Manual

Introduction

This is a blog for me to post musings on the continued development of antennas for the 2009 post-transition digital TV spectrum in the USA. The designs here are based on work done by others, at the Antenna Research & Development forum of digitalhome.ca, and are licensed under the GNU Public License, v 3.0 except where otherwise noted.