HF Antenna in the Attic
I made a new antenna back in August and it's time to write about it. I began the project because I was not very pleased with my antenna on the balcony. It had some parts close to the ground (and stuff made of metal), some parts inconveniently high up (barely reachable with a ladder), was not cut to the right length, and had a feedpoint outdoors. I wanted something more properly constructed, with convenient access, and the possibility of future modifications. The solution was to put an antenna in the attic.
Figure 1 shows a sketch of the new attic antenna, which is an end-fed halfwave antenna (EFHW) for the 40m, 20m, 15m, and 10m bands.
The main radiating element is immediately under the ridge of the roof – as high up it can be while still indoors. The roof is only 12 meters long, so the end portions of the wire bend back down. I was very fortunate that there was a plastic conduit for the old TV antenna from the attic south end down to a outlet right next to my shack bench. This is a perfect feedpoint location for an end-fed antenna.
The "feedpoint box" in the drawing contains two things: a common mode choke (a.k.a. "balun") and a 49:1 impedance transformer (a.k.a. "unun"). The schematic can be found in figure 2 and figure 3 shows what it looked like before I hot glued down the toroids. The box is a cheap food box that I mostly used to keep moisture out and make it easy to open and inspect.
The common mode choke and the impedance transformer are interesting topics on their own, so I will write about those in separate posts.
I had a 25 meter spool of 1.5 mm² houshold wire. I cut a 22 meter main part and a 2.5 meter counterpoise from it and marked every five meters and every one meter with some tape. I put the long part up in the ridge of the roof – there was a small gap at the end of the beams where I could snake it though.
After I had built the feedpoint box, I invited a friend over for an antenna trimming night. We used a NanoVNA H4 and a laptop to measure the SWR of the antenna over the HF range.
The main wire was far too long at the beginning and the resonsances were too low in frequency (as expected). We ended up cutting and measuring at these lengths:
- 22.0 m – the starting length
- 21.5 m
- 21.0 m
- 20.5 m
- 20.25 m
- 20.125 m
- 20.0 m
- 19.75 m
- 19.5 m
- 19.4 m
- 19.3 m
- 19.2 m
- 19.1 m
- 19.0 m
- 18.9 m
- 18.8 m
- 18.7 m – final length
I was aiming for good SWR in the 40 meter band, but I also kept an eye on the 20, 15, and 10 meter bands. I was surprised that the resonance for 40 meters was slightly lower in frequency comparing to the ones of 20, 15 and 10 meters. I don't know why this happens, but I have heard Callum McCormick M0MCX (the DX Commander guy) mention "the end effect". Maybe that has something to do with it? I haven't researched this yet. Another mystery is why there seems to be some kind of resonance at 80 meters. Maybe it acts as a quaterwave antenna there?
I had to choose between good SWR on the 40 meter band or on the other bands. I chose 40 meters because it was the only band I had a working transceiver for at the time. Plus, I figured I could always add a short extension wire in the future, if needed. The final SWR plot can be seen in figure 5. There is a Touchstone file of the measurements as well.
Lastly, I ran an RG-58 coax down the plastic conduit which used to hold the TV antenna coax. I had to crimp one end up in the heat and dust of the attic (figure 6 gives you an idea), but it went pretty smoothly.
Down by the shack desk, I removed the old TV antenna outlet, put a lid in its place, and ran the coax along the skirting to the shack desk. All this is hidden behind a bookshelf. Neat, if I may say so!
I tested the antenna briefly when I finished it using my monoband BITX40 transceiver. It seemed to work equally well in receive. I did not call CQ that many times because I quickly got into my next project: building the multiband QMX transceiver kit from QRP Labs. I built the QMX in the autumn and started using FT8 on it in the winter. At first I only used the 40 meter band, but later on I also worked 20 meters, 30 meters and 60 meters. I've used a Z-match tuner (my own clone of the EMTECH ZM-2) in the shack, which worked susprisingly well for the 60 and 30 meter bands which the antenna was not made for.
Overall, I'm very happy with the antenna. The biggest win is that I now have a coax at my desk permanently. No more rolling out the coax across the floor and out the balcony door! I still have many things left to do. It's February now, and I should probably measure the SWR again and see if it has changed. I think it has, because I get lower SWR on 20 meters than 40 meters now. I should also try to call CQ on SSB again. FT8 (and also FT4) has just been so much fun! I don't have much leisure at the moment, so the digital modes have really given me a chance to make contacts in the time I have.