04 Apr Design for an end-fed Windom antenna
By Steven P. (VK3TR)
Perhaps my favorite antenna design is the end-fed half wave. I’ve spent many hours designing and testing variations on the standard EFHW design, trying to better understand it and eke out some subtle benefit.
Recently though, I started down a less traveled path, wondering if it would be possible to combine the best parts of the flower-pot antenna design with a classic Windom / OCFD design? The idea being an antenna that’s physically end-fed, but electrically off-centre fed. Then it’d only need a 4:1 balun to match, while also working on multiple bands.
Of course, there’s nothing new in this hobby and it turns out I’ve re-invented something from a decade ago. Just when I thought I’d had an original idea!
For those who’ve not made its acquaintance, a flower-pot antenna is an interesting type of dipole antenna. It’s distinguishing feature is that it’s made entirely from coax cable. To achieve this the coax shield is stripped from a quarter wave length (the yellow, top radiating element in the diagram), and a coax coil choke is located a quarter wave length lower (the vertical grey, lower radiating element in the diagram).
The antenna is physically end-fed, but electrically it’s a standard centre-fed dipole with a nice 50-ish Ohm impedance. This works because the feed signal is contained within the coax centre conductor and inner surface of the coax shield. When the signal reaches the end of the shield (at the centre of the dipole) it then continues along the outer surface of the coax shield towards the choke, and along the continuation of the coax centre conductor.
Voilà it’s a standard half-wave dipole! As such it is a single band, even more than usual due to the frequency-dependent nature of the coil choke that terminates the lower radiating element.
In comparison, I expect everyone is familiar with the common, garden variety Windom antenna, sometimes called an off-centre fed dipole (OCFD). This design takes advantage of the fact that the current high point of multiple amateur HF bands align near to the 30% point along a half wave dipole. This gives them all a similar feed point impedance, allowing multi-band operation using a simple 4:1 balun.
My grand plan was to combine the physically end-fed design of the flower-pot antenna (dual purposing the coax shield) with the off-centre fed design of the Windom antenna. In theory this would yield a multi-band end-fed dipole, needing only a 4:1 matching balun.
Understanding that this had probably been tried before I tried searching for prior art but was initially unsuccessful. Obviously, this meant either I was a genius for thinking of something original, or it was such a bad idea that no one had ever wasted their time describing it (the more likely scenario!).
Asking online though I was soon informed that I had been anticipated by about a decade, with the design being christened the “City Windom” in 2011. Click the image for some history.
This was excellent news! Time to get building…
The curliest problem was how to design the 4:1 balun that sits between the coax feedline/radiator and the continuation of the antenna. And how to connect one of the outputs back to the coax so the signal could radiate from its outer surface?!
I’d previously built a good performing 4:1 balun using a single L15 ferrite from Jaycar, and having one on hand decided to repeat that here. I used two runs of eight bifilar turns of 1mm enamel wire, encased in heat shrink to give it some durability.
The coax centre and shield connect to the balun’s inputs, nothing unusual there. Of the balun’s two outputs, one (from the shield) returns back to the coax shield at the input, while the other connects to the other element.
I thought that I’d need a substantial ferrite choke since this needs to electrically terminate the radiating element, not just cope with some stray common mode currents. Certainly, an air wound coax choke wouldn’t do, not if I wanted this to be multi-band anyway.
I spent some time measuring a few different choke designs, but for this first test decided to use just a simple 12 turns on a FT240-43 ferrite core. I’ve got some ferrites on order and will probably revisit this part of the design.
For the moment though, it’s simple and elegant, and it worked much better than I had expected.
So I took myself over to a local park, threw a line over a tree branch and setup my new end-fed Windom as an inverted-V (~40 metres of wire, 80M HW length).
My initial testing is very positive. I’d not even taken a radio with me, just the antenna analyser, so can’t report any contacts or signal reports, but it has pronounced dips in all the expected places. Pretty impressed actually!
Next step is to plan some more time to get on the air with it and see how it really performs 🙂
PS/ I’ve added a diagram giving some brief guidance for construction. Please add a comment if you build your own and let me know how it goes on the air!