I’m impossibly behind in my posting of DXing sessions and AM station recordings, so I’m not sure how I’m going to take care of that. But, I hope to make several posts over the next month or two on antenna testing. To start this off, I did some formal pattern testing of my 6-foot tuned loop antenna this morning. Ideally, one should do this at high noon, but the testing was 2 to 2.5 hours after local sunrise, and all of the skywave signals seemed to be long gone by that point.
My tuned loop antennas have a short board at the bottom for mounting the tuning capacitor, so I put my hiking compass on the other end of the board, and the whole thing is mounted on a camera tripod. Starting at true north (confirmed at night using Polaris), I noted the strength of several stations on my software-defined radio (SDR) display every 10 degrees around the compass. I repeated the measurement of the first point at the end to make sure there wasn’t a significant change in the incoming signal during the test. The actual values are rather meaningless, and it’s the relative values that are useful. The nulls of loop antennas can be very sharp, so I added data points at the two exact nulls in the pattern. The relative angles are probably good to about 1-2 degrees and the angles could be systematically off by about 2 degrees.
The antenna was set up in my dining room, in its standard “winter” indoor location. Thus, I consider this a “real world” test. The nulls might be deeper outside away from indoor electrical interference. I dictate the ~40 or so measurements into a voice recorder and it takes about 6-8 minutes to do a “scan”.
The following plots should be mostly self-explanatory, but a few notes. First, ignore the scale on the x and y axes, and forgive the somewhat pixelated and slightly tinted image; I’ll take care of that in future plots. (At least you can click for a larger version of the image.) The rings depict the labeled number of dB below the peak value. The thick line is the direction of the station, with the dashed line being 180 degrees opposite. That should be the axis of the peaks of the double-lobed pattern. I started and ended at true north, and you can see that in two cases, the starting and ending measurements differed by 1 dB, a minor difference. I did dot-to-dot plots, but at some point I would like to fit patterns to the data. Although KFNX is a full-power station during the day, its directional pattern only puts about 2.5 kW of effective radiated power in this direction and was the weakest signal of the four.
As far as results, first note that the patterns seem to be skewed by some 10-15 degrees for my two hyper-local stations. I wonder if this might be a terrain effect from the nearby mountains. KAZM is also skewed (in the opposite direction) by about 5-10 degrees. KFNX seems to be about dead on. You can see that there was at most a 1 dB difference between the peaks of the two lobes, but greater differences between the nulls. The following summarizes the peak-to-null signal ratios for both the stronger and weaker nulls.
1490-KYCA: 17dB and 27dB
1450-KNOT: 15dB and 27dB
1100-KFNX: 19dB and 26dB
780-KAZM: 23dB and 26dB
These differences may be due to local effects (particularly since the worst cases were the local stations where the pattern angles were skewed) and/or interference partially filling in one or both of the nulls. Also, the loop has to be tilted slightly so that it clears the tripod legs when it is rotated, which also may skew the nulling to one side or the other. But, it’s noteworthy that the best null was statistically identical for all four stations. Although this is the first time I’ve mapped out entire patterns, the magnitude of the nulls matches previous values for both daytime groundwave signals and strong nighttime skywave signals, so I’m pleased with these results. I should probably try an outdoor test to see if I can get the nulls to be a little stronger.