On Fri, 9 Oct 2009, Matt wrote: > I am experimenting with some 3 Ghz antennas. I have a 2 way power > divider feeding two 120 degree sector antennas rated at 15 dBi each. I > have a 11 dBi omni that I am comparing it too. It seems the omni is > getting better performance then the array is. I am seeing more than a > 4 dB difference, loss. My goal is to get more down tilt in my antenna > pattern, as the sectors offer a mechanical down tilt. > > Am I doing something wrong?
Quite a few things, but that's beside the point. 1) If you're on 2.4GHz and working with 3GHz antennas, you may experience excessive downtilt due to operating below the design frequency of the antenna. On the other hand, if you're far enough away from the design frequency, the antenna may radiate upwards instead. 2) You have 240-degrees of circular coverage, leaving a 120-degree gap. 3) .5W (+30 dBm) /2 = +27 dBm per antenna. +27 + 15dB = +42 dBm between the -3dB points. Are you sure you're in compliance with Part 15 or Part 97? However, your received signals will have an uncorrected additional loss of -3dB, regardless of which antenna it is received on. 4) You're going to get more noise from sources you're now looking at better. Since you were previously comparing to an 11dBi omni, this isn't a point of consideration. Your problem isn't new. Ten years ago, when I got into IT working at a small WISP, our vendor advised us to avoid using the 12dB antennas, and to stick with the 6dB antennas. The reasons detailed are exactly what you detailed -- the signal was unable to reach users under the antenna within a reasonable distance and put the signal on the horizon where it did no good. Most of the WiFi gain antennas have a commercial VHF/UHF antenna for comparison. One may plot a site survey based at 2.4GHz and visualize the coverage of terrain one would see at VHF frequencies with impractical amounts of gain (you've got to have a lot of tower real estate to pull off a DB-228*2 [or DB-2216 as I like to call it]). Most WiFi suppliers stock a 3dB, 6dB, 9dB, and 12dB antenna. The commercial equivalents would be a DB-222, DB-224, DB-228, DB-2216 respectively. The 222 and 2216 are fictional antennas made by separating or joining the 224 and 228. > I am feeding the antennas with LMR400 and feed is short as possible. > Each feed from the divider is 15 inches long to the antenna. The > insertion loss in the divider is <0.4 dBi according to the spec sheet. As long as everything is equal on this level, you won't have to figure anything else out. -- Kris Kirby, KE4AHR Disinformation Analyst
