30 - 60 MHz Fan Dipole

Hung antenna image

Don't rush right out and build one. I've always been sort of a fan of the fan dipole, I've built a few, but never one with this many wires before. A fan dipole has several dipoles connected together at a common feed point. The idea is that the ones that aren't resonant at a particular frequency will present a high impedance at the feedpoint so they won't affect the others much. I've found that the proximity does have an effect though: building one you often have to adjust existing ones when you add another.

I wanted to cover 30 - 60 MHz and have horizontal polarization with a pattern that favored the sky. So using the 468/f formula I cut one for each 2 MHZ plus another that's cut for the Jove frequency of 20.15 MHz. So there are 16 dipoles hanging from the Jove dipole. I took apart some old network wire and cut the 16 dipoles from it, then put pieces of heat-shrink tubing around the whole thing to contain the mess.

I wanted to get away from electrical noise sources in the house, so I bought a cheap 500 foot roll of RG6 and hung this from some little trees about 200 feet from the house, oriented so the house is off one end.

It's quiet. Very quiet. For about the first 18 hours after I hung it I wondered if it was working at all. I did an ohmeter check on the end of the RG6 for shorts, that was OK. The next day I went out with a cliplead and clipped it across the antenna, then came back in and measured: not open either. I've heard a couple of QSOs on 10 meters, a few on 15 meters, various WWV frequencies, CBers, AM (MW) broadcast stations, lots of 20 meter activity. I haven't heard anything on 6 meters other than catching a few repeater IDs, but that's 6 meters. For quietness it's a dream antenna, but I don't have a lot of confidence in using it for radio astronomy.

So I went out with my MFJ-269 antenna analyzer and 3 sheets of paper and sat in the lawn chair and took measurements. I disconnected the RG6 going to the house and connected a 6 foot piece of RG6 to reach the lawnchair for this.

Antenna Analyzer plot

I took readings every 500 KHz from 14 to 70 MHz, just in the basic mode the analyzer powers up in. This isn't a very good way to evaluate a receive antenna probably but it was a convenient first attempt. I think maybe if I worked at it enough I could set up a makeshift test range connecting a receiver through RG6 from about 100 feet away and using the MFJ-269 as a signal generator. But again, the same problem surfaces of what to use for a "flat" antenna on the signal generator so its characteristics aren't going to affect the test. I need to build a flat isotropic antenna. And I really should take measurements from a helicopter or balloon overhead. Right.

Still, the analyzer didn't read an infinite SWR at any frequency I tried. And I swept the frequency around since every antenna has it quirks and strange resonances. At 100 MHz I got a 1.1 SWR with Rs 46 ohms and Xs 3 ohms, at 137.2 MHz I got 1.2 with Rs 42 ohms and Xs 7 ohms. Good VHF WEFAX antenna? Not what I was trying to do. It's a 2.0 SWR at 144 MHz too.

Image of wire

I built it in the basement. Here one side is finished with heat shrink and everything but I haven't started containing the second side.

The top (Jove) dipole is made from galvanized steel electric fence wire and supports everything. The gray cylinder is a piece of 2 inch PVC conduit (sunlight resistant) that I drilled holes in to mount things on. The steel wire goes through holes on opposite sides here to support everything. When the fence wire is new it can be soldered to fairly well with a 100 watt gun so I solder on pigtails of stranded copper wire to make connections with.

The wires from the network wire (16 of them) are soldered together and another pigtail attached. Then I ran them into one hole and out another and tied a knot in them to hold them in place.

These pictures are strange sizes and shapes because I had to play games with cropping, scaling, using lower quality to keep the file sizes down.

Here's the first end of one of the individual wires I've hit working out from the center. I cut many pieces of 1/4 inch heat shrink tubing and some 1/8, and I threaded the wires through them. Here are 3, with the clothespin marking the end of the wire. It took me parts of 4 days to build this.

Once I got everything about right on a side I lit my propane torch and walked along shrinking the heat shrink to hold everything in place.

image of

Here's the completed antenna coiled up on a stool ready for taking outdoors. The yellow toroid is from an old computer power supply for a choke. I connected my pigtails to a piece of RG-59 with copper shield and center conductor so it can be soldered to. From here on it's the cheap RG6 which has aluminum sheild and steel center, so only F fittings can be used on it.

Really though, I bought a 500 foot spool of RG6 at Home Depot for $32 so I can't complain. I've always thought that when you want to buy something cheap you go where the biggest market is, and this is used for cable TV and satellite TV wiring. Don't get the indoor (white) version.

I'm off. There's my camera tripod with collapsible fishing rod, a slingshot for shooting the fishline/string/rope over tree limbs, string, a spool of polyester rope and propane torch for cutting the rope without fraying. I don't walk so well anymore so I planned ahead and didn't have to make any unexpected trips for things I forgot. Loaded
cart image
antenna in air It was really kind of overkill to use the slingshot in these little trees, but it worked. These are "weed" trees that grew up in what used to be a raspberry patch so they aren't very tall yet. I can reach the antenna standing on the ground, but only barely, so it's up about 7 feet above ground. The center of the antenna is behind a tree branch here. The picture at the top of the page shows it.

Here's the antenna from the north. There's a very good take-off angle to the south, so the first things I heard on it were 10 meter QSOs in Spanish from South America. This lawn chair is where I sat to take antenna analyzer readings. The RG-59 on the antenna plus a 6 foot piece of RG6 was just the right length.

Antenna from
the north

Trying to model this before I built it was a complete failure. I've seen the same thing happen once before when I tried to model a multi-turn loop with closely-spaced wires. I think something gets "confused" and it didn't just predict bad performance, it looked like it was getting overflows. I tried both MMANA-GAL and NEC2. The files are here if you want to see. My antenna analyzer measurements are here. Why 30 - 60 MHz? That's one range I can sweep with my Icom IC-7000 without hitting relay clicks.

How does it work? Not very well, I think. This is with the IC-7000's internal preamp on in order to get much of anything. As you can see by the black vertical lines it was interrupted a few times. I finally figured out the white vertical lines like at the center: it's intermod. They only happen with the preamp on or on the big antenna, so I knew it was some sort of overload. The blue vertical strip at the left is without the preamp.

Notice the white horizontal bands. I think those are the resonant frequencies of the individual dipoles. Most of the values are under 1000 coming from the soundcard, but there are some peaks like in the 6 meter signals that go across (53.350 MHz especially) that are 5000 - 7200. In some ranges it's a confusing palette because the lower values like 0x40 - 0x50 look too much like the upper 0xEA - 0xFF. 1,591,830 data points here, kiddos, and it's not even a full 24 hours.

weak signal
palette image

The horizontal banding persisted with every antenna until a couple of months later I was doing a frequency survey of the 45 - 50 MHz range and I discovered this:

45 - 50 MHz

Apparently it's the radio, my IC-7000. Not much I can do about it.

AB1JX   /   toys   /   Radio Jove