They Dared to Compare. . .
. . .Multi-Band HF Arrays and Vertical Antennas

L. B. Cebik, W4RNL

dare_t.gif (1004 bytes)he performance claims in advertisements for antennas usually defy comparison among different makers. Those about to make significant investments in antenna hardware wish that there were a set of independent tests directly comparing the most common brands and models before they spend their money. An early attempt at a comparison in 1982 was skimpy on test protocol details for the multi-band beams tested. No matter, anyone who has just spent a thousand dollars for a "great" antenna will howl in pain if he learns that it does not measure up to another antenna that he overlooked. Makers whose wares do not measure up will cry "Foul" and go to great lengths to try to invalidate the tests. Such is the battle ground for anyone who ventures into the arena of comparative antenna testing.

Into this no-man's land of controversy stepped Ward Silver, N0AX, and Steve Morris, K7LXC, a perfect combination of complex antenna installation, testing, and measurement expertise. From 1997 to 1999, they investigated multi-band parasitic arrays and published the first of their two books. Then they turned their attention to multi-band vertical antennas and the second book, which appeared in 2000.

fig1-1.gif (8637 bytes) fig1-2.gif (15457 bytes)

Fig. 1 shows the covers of the two volumes that report both their results and how they got them. The book format is 8.5 x 11 inches, with a binder that allows the book to lie flat while the reader moves back and forth through the material. At first site, the books may seem thin, at 88 and 63 pages, respectively for the beam and the vertical array tests. However, virtually every page is crammed with important information.

Some readers will be more interested in the results than in the test methods. Using a common test protocol for all HF horizontal beams, the intrepid duo reviewed the performance of the GEM quad, the Force-12 C-3 and C-31XR, the Mosley TA-33 and Pro-57B, the Cushcraft X-9, the Bencher Skyhawk, the KLM-34-XA, and the Hy-Gain TH-7DX and TH-11. These well-known names cover both relatively small and relatively large HF horizontal tri-band arrays. With another test protocol tailored to vertical antenna comparison, they tested the Cushcraft R8, the Butternut HF6V, the Hustler 6BTV, the Force-12 ZR3, and V-3, the Diamond CP-6, the MFJ-1798, and the GAP Titan.

Equally, if not more, important is the extensive discussion of the test protocol used in each of these studies. Prior to testing, the authors enlisted a large number of antenna experts to critique their proposed methods for testing the antennas and made extensive revisions and refinements before hoisting their first antenna into place. They did not have access to a rated antenna range or the absolute latest in sophisticated testing gear, although the equipment they did use goes well beyond what we can find in almost any shack. They based each of their protocols on three Rs: it must be reasonable in both set-up and evaluation; it must be repeatable to encourage other to make their own measurements; and it must be relative to a known standard. For the HF beams, the standard was a full-size dipole, and for the verticals, the standard was a 1/4-wavelength monopole. The protocols and the execution of the tests--including giving antenna makers a chance to reply and to have that reply included in the report--show that the authors adhered to an over-riding principle in their arduous testing: to be as fair as humanly possible.

Each of the main topics of the reports--the test results and the test protocol--appears in two separate sections. These are books that one is not likely to read in order. In each volume, there is a discursive section on the test results, complete with full discussions, tables, and interpretations of the outcomes of the tests, including apparent anomalies. As well, an appendix is devoted to the detailed test results for each of the subject antennas. Likewise, each book contains an extensive report on the development and implementation of its test protocol. A companion appendix provides a very extended outline of each element of the protocol and the standards used in testing. I recommend that you read the 2 parts of each topic together to obtain the most complete view, whether your interest is in the test results or in the means by which they were obtained.

There are distinct differences in the test protocols for tower-mounted HF multi-band beams and multi-band verticals. For example, the HF tests required a considerable distance between the antenna test site, which was the transmitting site, and the receiving site almost a mile away. Fig. 2 provides an outline of the test set-up, along with a map to show the path over which they took readings to obtain signal strength and pattern shape results.

fig2-1.gif (7766 bytes)
fig2-2.gif (7393 bytes)

However, testing could only occur once the authors--with help from numerous supportive friends--assembled each HF array and mounted it atop the test tower. Fig. 3 shows that managing the assembly and installation was no simple matter, which is where the tower and antenna installation experience of K7LXC became crucial.

fig.3.jpg (22890 bytes)

The test-site transmitting station, shown in Fig. 4, may appear deceptively simply, since it seems to consist of a transmitter and a Bird Wattmeter. However, each transmission required considerable calculation to ensure equal power levels at the antenna terminals, regardless of the SWR level on the line. At the receiving end, they used a spectrum analyzer to determine signal strength, with adjunct equipment preceding the analyzer to keep readings within its most accurate ranges.

fig.4.jpg (32255 bytes)

The authors calculated from their measurements the forward gain, the 180-degree front-to-back ratio, and the 50-Ohm SWR for each antenna at 5 points across each band (20, 15, 10 meters). As well, they calculated the pattern of each array through its full rotation. The appendix provides both tabular and graphical listings of the performance of each antenna. The advantage of the appendix is that it offers the data without comment or comparison, allowing the reader to reach his or her own conclusions. In contrast, the text provides much comparative and interpretive information, such as the average gain per band for each antenna and the gain per foot of boom length. Throughout, the testers used beams installed exactly as recommended by the manufacturer's literature. All gain values are referenced to the full-size dipole used at the site broadside to the receiving station.

The tests on multi-band vertical antennas involved somewhat less arduous installation labors--once the authors had created a ground radial field. Fig. 5 indicates--but does not show completely--the work involved in this preliminary step to all of the tests.

fig5.gif (36023 bytes)

Testing vertical antennas requires some rethinking of the test situation, since most of the commercial multi-band verticals are intended for ground mounting. Hence, the tests required a relatively clear field between the transmitting and receiving sites, along with a stable background at both ends to ensure that there would be no significant changes in reflected signals at either end. As Fig. 6 shows, these requirements only permitted a distance of about 500' between the transmitting and receiving antennas. Ideally, antenna tests use at least 10 wavelengths between transmitting and receiving locations. However, the reality of tests conducted by amateur radio operators for amateur radio operators is that rated antenna ranges are not available for use. Hence, the tests had to use the best possible available space. The receiving antenna was a non-resonant small shielded loop field probe to minimize the effects of unwanted signal coupling and other variables. Unlike the HF array tests, conducted on only 3 bands, the authors tested the verticals across 5 bands, ranging from 3.550 to 28.850 MHz. (Some of the test antennas only covered some of the 5 bands, and the non-harmonic bands--30, 17, and 12 meters--were specifically omitted from the tests.)

fig6.gif (9208 bytes)

In principle, the test set-up was similar to that used for the HF arrays, as shown in Fig. 7. The protocol for the verticals lacks certain aspects of the HF array protocol, for example, the need to check the directionalness of the pattern. The results listed in the appendix show gain in dBrv, that is dB of gain relative to the values obtained for 1/4-wavelength monopoles. (The 80-meter reference monopole was balloon supported.)

fig7.gif (7738 bytes)

In some ways, the vertical tests were simpler than the HF array tests. The assembly and installation of the antennas required no tower climbing--and re-climbing. However, the expanded band-coverage requirements required greater care to assure reasonable reliability of the results. As Fig. 8 reveals, N0AX required more than test equipment to accomplish the task. Snacks, drinks, a basket to carry cables, and even duck tape are necessary adjuncts to adequate antenna testing. As well, the authors tested not only multi-band 1/4-wavelength monopoles, but as well, antennas that were essentially 1/2-wavelength dipoles or used substitute means to achieve freedom from the need for a ground radial system.

fig.8.jpg (31027 bytes)

The differences in test protocols are reflected in the categories used to discuss and evaluate arrays. Of course, verticals do not have a front-to-back ratio, but the differences go beyond such obvious points. For example, the authors conclude the HF beam study by listing "high points" among the antennas: the big over-achiever, the loudest, the best front-to-back, and the most consistent across the three bands. A number of well-known multi-band HF beams fail to receive mention in any of the "high-point" listings.

In contrast, the authors found no clear winners in the comparisons among vertical antennas. The HF verticals fell into niches: the all-band suburban or rural antenna, the low-band (80/40) antenna used also as a back-up for the upper bands, the restrictive-covenant antenna requiring no ground screen, the apartment or townhouse antenna that permits no ground-level installation, and the portable/expedition antenna. Virtually every one of the tested verticals has an assigned niche or two.

Amateur radio is a communal enterprise, and it is doubtful that the antenna tests would have occurred were it not for the serendipitous combination of the authors' combined talents and experiences--with a lot of help from their friends. As well, commercial antenna designs and designations come and go, but solid test protocols are a foundation upon which future generations can build. We shall need a renewed maximum effort every few years like the ones that produced these books. Therefore, I recommend very close reading of the sections of the reports devoted to the test protocols. The protocol outlines in the appendix are very detailed, allowing the reader to envision and develop refinements for the future. However, be certain to read the sections that discuss the implementation of the test program. They will provide insight into the practical limitations and considerations that adapt a given protocol to real testing situations.

The volumes have a permanent place on my bookshelf more for what they teach about testing antennas than for the ranking of any of the antennas tested--at least until I am in a position to upgrade my antenna system. -30-

* * *H. Ward Silver, N0AX, and Steve Morris, K7LXC, HF Tribander Performance: Test Methods & Results, 2nd Ed. (Champion Radio Products, 1999), 88 pp., $17.00, and HF Vertical Performance: Test Methods & Results (Champion Radio Products, 2000), 63 pp., $17.00. Available from

Click for Biography of Author

~ antenneX March 2003 Online Issue #71 ~

Click to Print/Save