THE CLASSICAL COAX-FED WINDOM
[ Classical Coax-Fed, Off-Center-Fed Dipole ]
THE GENERAL IDEA:
When the original Windom antenna was developed around 1923, the idea was to take a half wavelength of wire, and feed it at a point along the wire that would have [near] equal impedance on all amateur bands at the time (i.e. 80/40/20/10m). On a half wavelength piece of wire, there is only one place where that occurs and it is at 1/3 of the length (on the fundamental frequency).
Keep in mind, that this only really happens in theory. In space they would all be equal.
In the real world we have to consider the influence of the ground below the antenna.
From antenna basics, we know that a standard dipole has an impedance of about 70 Ohms when it is about a half wavelength above normal ground. As we lower it, the impedance drops. The same thing happens with the Windom.
In the case of the dipole, we were only focusing on one band.
In the case of the Windom, we are focusing on several bands, and of course as we raise or lower the antenna's height above ground, its electrical height is different from band to band.
As a result, depending on height, we find that the impedance is closer to about 235 Ohms on most bands for heights under 50 ft. (which is what most people have) and increases to about 270 Ohms for heights above 70 ft.
For most people, a 4:1 balun is going to provide a better match than a 6:1 balun. If the antenna is really high, perhaps a 6:1 balun would be better. In any case the design and quality of the balun is just as important as the impedance ratio of the balun.
BEST PRACTICE: Originally I thought we should use a 4:1 Guanella Balun (current balun) with 2 toroids. Using 2 toroids is important because there are 2 transformers in the 4:1 Guenella and each transformer should be wound on a seperate core from the other transformer.
However that idea has a couple of problems:
- Most dual-core 4:1 Guanella (current) baluns are wound onto toroid cores with insufficient permeability and too few turns of transmission line. Thus their Common Mode Impedance (CMI) is insufficient for the job.
- Further, my intensive tests in 2015/16/17 on an 80m OCFD at 40 ft. height proved beyond a doubt that even the best 4:1 Guanella balun (alone) is insufficient. And additional RF choke connected directly to the balun was required to remove all residual Common Mode Current (CMC).
Note: My earlier tests (2013) of a 40m OCFD found a 'good' dual-core 4:1 Guanella balun wound on #43 ferrite to be sufficient ON THAT ANTENNA.
EVOLUTION: In an effort to find a lighter weight solution than the dual-core 4:1 followed by the single-core choke (total 3 heavy cores), I developed a new "Hybrid Balun" using just 2 cores. This consisted of a 4:1 Ruthroff (voltage) ((YES, 'VOLTAGE')) balun, followed by a 1:1 Guanella (choke) balun.
BEWARE: The internet is full of instructions for building a 4:1 Guanella balun on a single Toroid. I STRONGLY ADVISE NOT TO DO THAT.
As I said above, you should use two toroids for this balun, wrapping each transformer on its own core. SEE: http://www.dj0ip.de/balun-stuff/bad-baluns/
I have pictures of some of my own baluns here: BALUN PICTURES
Note: you will see a 4:1 Guanella on that page which uses a single toroid. I built this so that I could compare it to my other baluns using dual cores. I do NOT recommend a single core 4:1 balun.
More Bands . . .
In 1927, 15m was not yet available to ham radio operators.
The WARC bands came more than 50 years later.
Today we have 4 more bands than hams had in the 1920s.
It is quite possible to modify the Windom to give us a decent SWR on some of these new bands.
HOWEVER, we do this at the expense of the original 4 bands.
Specifically, we must accept a slightly worse SWR on some of the old bands in order to obtain use of some of the newer bands.
THE WINDOM IS A COMPROMISE:
The windom is a compromise antenna. We have to choose between fewer bands, but overall better SWR, or more bands with overall slightly worse SWR.
THERE ARE TRICKS:
Tricks such as inserting a coil, inserting a capacitor, or both, at strategic points along one of the legs will improve the situation. But this adds complexity and weight to an otherwise simple multi-band antenna.
For more information on this, see ON4AA's web page HERE
OR you can simply add a second, separate radiator to the normal Windom. Simply take a half wavelength of wire for the 15m band, cut it in a 1/3-to-2/3 ratio and solder it in parallel to the other wires of the Windom. If you can rotate it 90 degrees or so, fine, but if not, then suspend it from the main radiator such that the ends of the 2nd radiator hang about one or two meters below the main radiator. This adds 15m to the antenna. I have done this myself back in the 1970s and had friends who were doing it 10 years earlier than that.
Another idea I had but have not tried yet is to cut the 2nd radiator to 10.55 MHz. This is twice as long as the one described just above. It will work fine on 15m and will also work on 30m, albeit with somewhat higher SWR, yet still low enough SWR that you can use a matchbox and expect good results. As I said, I HAVE NOT TRIED THIS YET.
AND THERE ARE SIMPLE COMPROMISES:
By proper careful positioning of the feedpoint, it is possible to build a multi-band Windom that does a good job on most of the bands. However there are also some drawbacks (i.e. trouble with common mode current which MUST be addressed). I will cover these points in more detail on my web page describing these antennas. See: 80m OCFD
For the heavy theory behind this antenna, I will point you once again to others who are smarter than me. I will share my experience with you but leave the theory to the real experts.
Here are some good sources to inform yourself about this antenna:
THE CHIT CHAT:
If you are seriously interested in this antenna, I recommend you join the Windom and OCFD Group on Yahoo!. You'll find hundreds of people there who have built and used these antennas.
As with any group, it has highly experienced members as well as some who haven't a clue, yet all of them will share their experience with you.
JOIN THE GROUP: https://groups.io/g/ocfd