May 1953 QST
Maybe I suffer from cranial rectumitis at the moment, but I'm having a hard time with a statement made about coaxial feedline impedance, to wit, "102-ohm line (52-ohm lines in series)." I must be missing something because I don't understand how placing two 52-ohm transmission cables in series results in twice the impedance. Aside from that, author John Avery presents an interesting article on multi-impedance dipole antennas. Empirical data is presented on how the feedpoint impedance of a dipole varies with distance above the ground. His results are very close to theoretical values which assumes non-sagging elements, perfectly linear alignment, a perfectly conductive ground, etc. He then extended his investigation into 2-wire (4x impedance) and 3-wire (9x impedance) folded dipole antennas.
Closer Matching at Various Antenna Heights
By John D. Avery,* W1IYI
While looking at a chart in the ARRL Handbook that shows the impedance of a half-wave antenna at various heights above ground, I began to wonder if a wire at my QTH would behave "like the book says." My location is at a lake shore, where water and wet ground are always present, so any measurements could be made over a period of time without running the possibility of a significant change in the electrical ground. ("Electrical" ground and the "surface" ground do not coincide - the electrical ground is usually some feet below the surface, depending upon the characteristics of the soil and the radio frequency being used.)
The measurements were made with a single wire a half wavelength long, split in the center so that a 52- or 72-ohm coaxial line could be connected. Two 60-foot towers were used to support the antenna, and a standing-wave bridge was available for use in the coaxial line. The procedure that was followed was quite elementary - with a given coaxial line connected to the antenna, the antenna was raised a few feet at a time until the minimum s.w.r. was indicated.
Starting with 52-ohm coax and the antenna, one foot off the ground, the s.w.r. was rather high, but as the antenna was raised the s.w.r. dropped and was quite close to unity at around 35 or 40 feet. Substituting 72-ohm coax for the lower-impedance line, the s.w.r. was higher. As the antenna was raised (now fed with 72-ohm line) the s.w.r. was dropping as the maximum height of 60 feet was reached.
It is safe to say that practically everyone ignores the effect of "height above ground" on the impedance of an antenna. This doesn't make any practical difference in many cases, but it can where you use a "flat" line and no means for adjusting the match. This article tells how W1IYI didn't ignore the height factor, and how it led to some interesting results and a slightly different concept in antenna design.
Having run out of height at my place, I managed to prevail upon a ham in Rhode Island who had higher supports to test with a 102-ohm line (52-ohm lines in series), and he found the minimum s.w.r. to fall at around 90 feet.
The diagram in Fig. 1 shows part of the Handbook graph that started this whole thing, with the three experimentally-determined points shown as small circles. Since they don't fall too far off the curve, they seem to prove that "the book is right."
This got me thinking about what might be happening to folded dipoles at various heights above ground. Since a two-wire folded dipole shows a four-times step-up in impedance (and a three-wire dipole a nine-times step-up) I added these values to the chart, on the left-hand side. A little study of this chart shows that, for low antenna heights (low in wavelengths) such as one runs into on 75 meters, the first choice of antenna and feed line might not always be the best. For example, a two-wire dipole only 25 feet off the ground should match better with 72-ohm coaxial line than with 300-ohm ribbon. A 3-wire dipole 35 feet above the ground offers a better match for 300-ohm line than does the more conventional 2-wire folded dipole. These statements are based on "electrical" ground, of course, a somewhat variable plane in most cases. Nine times out of ten it can only be found by experiment.
The Multi-Impedance Dipole
In an attempt to make better use of this (to me) newfound knowledge, a simple spreader system was devised that would permit changing quickly between two- and three-wire folded dipoles and a single-wire dipole. As shown in Fig. 2, the basic three wires can be used in these three ways. The spreaders were made from soft pine turned down to size and then boiled in hot paraffin. Fig. 3 shows some of the construction details. A center connecting block of 1/4 inch Lucite was built to take terminals and a coaxial connector, for quick changing of the various feed lines. It is apparent from Fig. 2 that changing from one antenna to another only requires changing a few jumpers, and perhaps disconnecting one feed line and connecting another.
With a system like this, it is not too difficult a task to find the best combination of line and antenna for the particular height you have available. You will probably want to use the maximum available height for the antenna, so it isn't suggested that you run the antenna up and down for a perfect match, although you may find the experiment interesting, as I did. For any length of antenna there is one frequency at which the s.w.r. is a minimum, and the s.w.r. will increase slowly as the frequency is changed in either direction. However, I noticed that the 2- and 3-wire folded dipoles, and the 3-wire dipole of Fig. 2A, seemed to be "broad" in this respect and not at all critical.
Although it has been pointed out many times before, it is worth repeating here that you need suitable coupling at the transmitter for each type of line. If you use an antenna coupler, small 2- or 3-turn links are adequate for coupling between coupler and transmitter, but larger links may be required if no coupler is used. I use plug-in links and a variable condenser in series with one side of the line. The largest plug-in link is 12 turns.
Right now I am using the antenna of Fig. 2A, fed with 52-ohm coaxial line. Local and DX results on 75 have been very encouraging.
This is all that is required for a 20-meter antenna in which several different values of impedance can be obtained. An antenna for a lower-frequency band would require only more wire and more intermediate spacers.
Posted April 28, 2016