I decided to expand my horizons a bit, so to speak, and bought an Antennacraft FM6 6-element yagi antenna cut for the US FM band (87.5-107.9 MHz). This is the time of year when sporadic E propagation (“E-skip”) starts up and I also find the concept of meteor scatter fascinating.
I haven’t had the antenna too long, just long enough to see what stations I can get with a good directional antenna under normal conditions. I’m lucky for DXing in that I live in a smallish community isolated from big cities. Thus, both AM and FM broadcast bands aren’t full of locals. In particular, I have a lot of “clean” frequencies on FM, something that can’t be said for people in large urban areas. This is helped out by having 2000 feet of terrain in between me and the nearest big city (Phoenix), so even with a good antenna, I don’t get much from that direction. But, E-skip and meteor scatter signals bounce off the upper atmosphere, so they come in above just about any terrain unless you live at the bottom of a steep mountain valley.
Anyway, I did my first overnight recording looking for meteor scatter. I used 90.5 MHz, for which the nearest station is over 100 miles away and the nearest station using more than 1 kW of power is nearly 175 miles away. Since this is an empty frequency away from other stations, I was able to crank up the gain of my RTL-SDR USB dongle to help with weak signals. I actually had the antenna set up indoors at the northeast corner of the house, so I pointed the antenna northeast. Coincidentally, the Eta Aquarid meteor shower peaked a few days ago, but it has a very broad peak of about a week, so was still moderately strong. I recorded audio from 1:30-6:00am MST (8:30-13:00UT) and was rewarded with a lot of pings! Unfortunately, meteor scatter usually only lasts for a second or two, and often less, as the super-heated air along the path of the meteoroid loses electrons to create the “electrical mirror” for the radio signal, then rapidly gets the electrons back or the electron cloud disperses. A particularly large and bright meteor can produce many seconds of enhancement, sometimes long enough and at the right time to identify an FM station hundreds of miles away! Unfortunately, none of the audio allowed me to identify the stations (I think there were several different stations over the morning), but I was very pleased with the results.
Here is a screenshot of Audacity after processing the audio (there seems to be an art to this, that I’m not quite proficient at yet). All of the spikes, including the weak ones, are audible, and there are many more audible peaks that are not visible when zoomed all the way out to display all 4.5 hours of the recording. (I processed the file as a whole, so the strongest peak set the maximum for everything.)
Here are a few audio samples, starting with the largest spike at 3:20 into the recording (4:50am MST/1150UT):
Here is a super-short ping that lasted only a tenth of a second, but I believe it is real (4:10/5:50am/1240UT):
Finally, here is one that doesn’t show up in the big plot, but was fairly long lasting (0:10/1:40am/0840UT):