=| es 2 Ei 5 =The ARRL Antenna Compendium Editor Gerald L. Jerry) Hall, KITD Assistant Editors Jim Kearman, KRIS Steve Ford, WB8IMY Production Jean Wilson Joe Shea ‘Technical Illustrations Dianna Roy David Pingree, NINAS Cover Design ‘Sue Fagan Volume 3 About the Cover: Clockwise from top left: Beginning on page 79, Steve Powlishen, KIFO, takes aim at the EME polarization-mismatch problem with a 432- MHz rear-mount Yagi array. Th the calculated H-plane pattem for 4x 14-element Yagis stacke part ‘Three antenna experimenters collaborated on a 160-meter antenna for small lots, “The Square-Four Receiving Array.” It’s described in the article beginning on page 33. This view is the calculated azimuth radiation pattern ‘when the array is operated in a square configuration. (It can also operate as a diamond.) ‘The calculated three-dimensional pattern of a commercial antenna—the 20-meter Hy-Gain 205CA, 70 feet above real earth,Copyright © 1992 by ‘The American Radio Relay League Inc Copyright secured under the Pan- ‘American Convention ‘This work is publication No. 162 of the aagio Amateurs Library, published by the League. All rights reserved. No part of this work may be reproduced in any form ‘except by wntten permission of the, ables. Alloghts of ransiaon are reserved. Printed in USA ‘Queden reservados todos Jos derechos ISBN: 0-87259-401-7 First Edition ‘Second Printing, 1993Contents VERTICALS, GROUND PLANES AND SLOPER ANTENNAS The Triband Triangle Robert J. Zavrel, Jr, W7SX Making Tower-Mounted Half Slopers Work for You Duane R. Sanderson, WOTID The Double Cross Vertical Antenna Robert Wilson, AL7KK Ground Planes, Radial Systems, and Asymmetric Dipoles L. A. Moxon, GOXN Phased Arrays for the Low Bands Al Christman, KB8I The Square-Four Receiving Array Gary Nichols, KD9SV, John C. Goller, KOUWA, and Roy W. Lewallen, W7EL Phased Verticals with Continuous Phase Control Peter H. Anderson, KZ3K The High Profile Beam— A High Performance Vertical Antenna with No Radials Ed Suominen, NM7T MINIATURE, TRAP AND MULTIBAND DIPOLES A Miniature Ground-Independent Dipole for 40 through 10 Meters Jack Kuecken, KE2QI An 80/40/17-Meter Super-Irap Dipole* Albert C. Buxton, W8NX ‘Trap Dipoles with Specified Harmonic Operation* Albert C. Buxton, W8NX How to Design Off-Center-Fed Multiband Wire Antennas Using that Invisible Transformer in the Sky Frank Witt, AIH YAGI ANTENNAS The K4VX 10-Meter Elephant Gun Yagi Lew Gordon, K4VX Rear-Mount Yagi Arrays for 432-MHz EME: Solving the EME Polarization Problem Steve Powlishen, KIFO 12 15 19 28 33 37 42 50 54 58 66 76LOOP ANTENNAS Loop Antennas: The Facts, Not the Fiction 99 AI. Henk, G4XVF Fun With Small Loop Antennas on 80 Meters 108 James E. Taylor, W20ZH A QUAD ANTENNA 12-Meter Quad 114 Howard G. Hawkins, WB8IGU LOG PERIODIC ARRAYS The Log Periodic Loop Array (LPLA) Antenna* 115 Duane Allen, N6JPO ‘The K4EWG Log Periodic Array 118 Peter D. Rhodes, K4EWG A MOBILE ANTENNA | A Streamlined Mobile HF Antenna for Vehicles with Fiberglass Tops 124 Steve Cerwin, WASFRF A DIRECTION FINDING SYSTEM A Simple Seeker Direction Finder 126 Dave Geiser, WA2ANU CONTROLLED CURRENT DISTRIBUTION ANTENNAS The CCD Antenna— Improved, Ready-to-Use Construction Data 131 Harry Mills, W4FD, and Gene Brizendine, W4ATE Controlled-Current-Distribution Antenna Performance: By Analysis 134 Bill Shanney, KJ6GR OTHER ANTENNA TYPES The Cross Antenna 137 R. P. Haviland, W4MB. The Skeleton Discone 140 D. Wilson Cooke, WA4RHT The T-L DX Antenna 144 Robert Wilson, AL7KK COMPUTERIZED ANTENNA MODELING AND FEED-LINE CALCULATIONS A Beginner’s Guide to Using Computer Antenna Modeling Programs 148 L. B. Cebik, WARNL.Modeling HF Antennas with MININEC— Guidelines and Tips from a Code User’s Notebook 156 John S. Belrose, VEZCV Smith Chart Impedance Matching on your PC* 165 Lynn A. Gerig, WA9GFR GROUND EFFECTS Horizontal Antennas and the Compound Reflection Coefficient 175 Charles J. Michaels, W7XC TRANSMISSION LINES AND FEED METHODS Quadro-Line: The Ideal UHF Transmission Line 185 Karol Dillnberger, DL7SS Efficient Feed for Sky Hooks 188 Stan Gibilisco, WIGV IMPEDANCE MATCHING TECHNIQUES The Z-Match Coupler—Revisited and Revised 191 Charles A. Lofgren, W6JJZ Fixed Inductor Variable Impedance Matching Networks 196 Robert F. White, W6PY A Precision Tuner for Antenna Couplers 205 Jack Kuecken, KE2Q INSTALLATION A New Way to Tree a Wire 2u Stan Gibilisco, WIGV RECEIVER OVERLOAD Front-End Overload, A Worst Case Example 214 John Stanley, K4ERO MEASUREMENTS AND TEST EQUIPMENT A Remote Field-Strength Metering System 218 John Svoboda, W6MIT The Hybrid Junction Admittance Bridge 223 Wilfred N. Caron PROPAGATION VHE/UHE Ray ‘racing with Computer Graphics* 231 John H. Priedigkeit, WOZGN *Software to accompany this article is available for the IBM PC or compatible computers on a separate diskette.Foreword This book contains 40 previously unpublished articles on antennas, transmission lines, computer modeling and other related subjects. Some, like Steve Cerwin’s HF mobile antenna, are practical and straightforward. Others, like Frank Witt’s discussion of off-center-fed multiband wire antennas, stress the theoretical bases for more-complex types of antennas. Regardless of your level of expertise, you'll find much of interest in this book. There’s something for everyone in Volume 3 of The ARRL Antenna Compendium’ Have an antenna-related project in mind? We'd appreciate the opportunity to review it for possible publication in Volume 4. David Sumner, KIZZ Executive Vice President Newington, Connecticut December 1992 Please note: Throughout this book, notes and references appear at the end of each article,Diskette Availability Programs for five papers in this book are available in IBM and IBM- compatible format on an optional diskette offered by the ARRL. (The papers with accompanying programs are denoted with asterisks in the table of contents.) The diskette is copyrighted but not copy protected. ‘The program files have been supplied by authors of corresponding papers, and filenames have been assigned to resemble the paper titles. There are four BASIC programs and one spreadsheet file in Quattro and Lotus 1-2-3 WKI format. The diskette is made available as a convenience to the purchaser. The ARRL has verified that the programs run properly, but does not warrant program ‘operation or the results of any calculations. Diskettes are available in both 54-inch (ARRL Order No. 4033) and 3¥-inch (ARRL Order No. 4041) formats. viiThe Triband Triangle By Robert J. Zavrel, Js, W7SX 117 Locatelli Lane Scotts Valley, CA 95066 have been interested in phased-array vertical antennas for many years. ‘These antennas have a low radiation angle. The ability to steer these arrays ele trically makes them even more attractive. ‘Much of the experimental work has con centrated on the low-frequency bands, here practical Yagi antennas become abit ‘out of reach for most hams. At the peaks of sunspot cycles, phased verticals might be useful at the higher frequencies (14-30 MHz). The cost of towers, rotators, and triband trap Yagis can easily top several thousand dollars. Additional factors such as aesthetics, occupied real estate, and safety turn me off towers, Yagis and quads can be Set on roof-mounted tripods and Provide good results. Roof-mounted ‘ground-plane antennas produce good DX results for a fraction of the cost of a roof mounted Yagi or quad, My goal for this project was to take advantage of the economic, mechanical, and performance characteristics of a roof ‘mounted ground-plane antenna. T wanted to achieve multiband operation and some gain with my antenna system, while keep- ing cost and complexity down. Performance comparable 10 a roof- ‘mounted Yagi or quad should be possible at a substantial savings, Several design goals were defined: 1) A low-angle radiation pattern 2) Multiband operation (10, 15, and 20 meters) 3) Small size 4) Simple impedance matching 5) Low cost 6) Easy installation 7) Gain on all frequencies 8) Pattems steerable to cover 360° 9) Simple electrical steering. from the shack 10) Elimination of parasitic effects One of the striking features of articles ‘written about phased vertical arrays is that they usually offer only monoband opera- 8 tion.' Many of these designs require exten- sive impedance matching and phasing systems for single-band operation, Build ing multiband, multidirectional matching. and switching circuits could make it dif cult to achieve my goals of providing simple impedance matching, low cost and casy installation, Several of my goals seemed to indicate the possibility of using multiple phased ground-plane antennas. T decided 10 do some experimental work with these an- tenna systems. To achieve maximum H-plane gain, [chose a 4-A.antenna forthe highest operating frequency (28 MHz), This should add about 3 dB to any direc tional gain achieved on 10 meters. Multiband operation seemed to mandate -the-shack tuning of the array. T have an old Johnson Kilowatt Matchbox, so I de- cided to use that with the antenna system. By keeping the feed impedances high atthe antennas and choosing my feed-line lengths properly, { would obtain a proper ‘match with the tuner. For several months [ experimented with a single antenna at dif- ferent lengths and with various feed-line lengths. [also did extensive experiments with ground-radial systems. Each step in- volved on-the-air testing with DX stations. ‘System Configuration Multiple, identical ground-plane anten- nas seemed to work best, but [had todecide ‘What configuration to use. Should | use two, three, or four elements? Simple switching and impedance matching required either 0 ‘or 180° phasing. That is easily accom- plished for multiband operation. ‘Two elements would provide gain, but achieving multiband operation with a steerable pattern covering 360° becomes a difficult trade-off. Four clements at the corners of a square form an attractive arrangement. The “Four: square array” provides four switchable unidirectional patterns with excellent ‘teen engin, Fig 1—The Triband Triangle consists of three 20-foot-long elements, spaced 10 feet apart at the comers of an equilateral triangle. The racial system is shown in Fig 4 front-to-back ratios. The switching net- work for this array is somewhat complex, however, violating one of my design goals. ‘Three elements arranged at the corners of an equilateral triangle providea solution toall ten goals. Two of the three elements are selected and fed 180° out of phase, providing three switchable bidirectional patterns. Fig I shows how the antenna is, laid out. Each vertical element is 20 feet long, and the sides of the equilateral triangle are 10 feet. I chose a length slightly shorter than ¥% 2.at 28 MHz to avoid resonance at 21 Miz. The high self-resonant base imcuewent & oe eb ~~ a9 [mm > cototies | val [ne \ St] +4 in Shock ‘ Reloye ! > 1 a 4 Shown "OF | ! | 1 1 LAY 2 ELEMENT A 1 ea tH I at Contra Line | wa vt ¥ 3 n | 72 , ee ¢ 1 ese +] ak [HRS | | al 1 Fig 2—Three DPOT relays select the vertical pedance might make it difficul to feed the array at 21 MHz, so [used 320-foot length Techose 10-foot spacing to raise the feed- point impedance on 20 meters to an acceptable value while maintaining gain over the range from 14 to 28 MHz. One ‘might cal this array a “switchable ground plane W8JK,” but “Triband Triangle” is more succinct. The two active elements induce equal signals 180° out of phase in the unused third element. There is no cur- rent in the unused element, and no pattern distortion results Feed-Line Considerations 1 tried several feed-line configurations. Calculations from the Smith Chart, mutual impedances, and clement self-impedances together yielded relatively low common feed-point impedances. Open-wire line was desirable toreduce losses, but the char- acteristic impedances are quite high. Then I discovered 300-0 tubular TV twin-lead. Since the system is always fed out of phase, it can be treated as a balanced system. We can ignore coaxial cable, with its high losses. All transmission lines are 300-0 tubular cable, whieh costs about 5 cents per foot. In earlier tests I ied RG-8 coax and 600-02 open-wire Hine. Losses with the co- axial cable were high. With homemade open-wire line, the high characteristic im- pedance (600 2) made matching more difficult since the SWR was very high. Hams may want to pay more attention (0 300-0 tubular line! The optimum feed-line Tength to the transmitter depends on the characteristics ofthe tuner to be used in the shack, such as the range of impedances the tuner can match, ‘Switching and Phasing Con On a tip to a local surplus dealer, 1 secured some 120-V ac DPDT relays for $1.50 each (Struthers-Dunn type 425XBX). use three DPDT relay’ to switch between the three possible antenna combinations. Fig 2 shows a wiring diagram. A plastic ‘weatherproof box houses the relays, which, are mounted on a V-inch-thick piece of plywood inside the box. A big advantage of using a plastic box is that all the RF termi- nals can be machine screws and nuts ‘mounted through holes diilled in the box. Multiconductor control cable may be the ‘most expensive part of the array! Three toggle switches in the shack control the relays. Fig 3 shows a truth table for the switch settings. One word of caution; if you use @ common conductor for the relays it must be able to handle the total relay cur- rent. You may want to consider using 12-V relays if you don’t like the idea of 120-V lines running to your antenna, but that re- ‘quires heavier control cable and a power supply. In either case, remember to switch lerations sments and phasing to steer the Triband THangle. A common wire can be used to all three relays. The switches are SPST types that connect each relay col to the required control voltage supply. Fetey Box Leching Boun on Arey enon few OF Or hese Xow Fig 3 —The possible directions you can steer the array. These directions assume ‘elements B and C are aligned on an ‘east-west ine and element Ais due north of the center point between them. The truth table shows switch positions to ‘elect each desired pattem direction; don't care, ff all power to the relays before working, ‘on the box! Since phasing is always 180°, phasing switches are simply cross-wired relays.Fig 4 — A top view of the radial system sod with the Triband Triangle, The. radials should form a plane parallel to Jovel ground or tit downward from the array symmetrically, ‘This system is inherently broadbanded. Although my antenna works on 10, 15 and 20 meters, this array should also work on 24 and 18 MHz. The array should also. serve asa good SWL antenna, covering the bands between 14 and 28 MHz. A suitable antenna tuner in the shack is all that you need to tune the array to any frequency over the octave for transmitting. Antenna Elements ‘You can make the twenty-foot vertical elements from many different materials, ‘Aluminum eleetrieal conduit comes in 10- foot lengths, so you can use six of these, with couplings to thread them together. I used thin-wall tubing guyed with heavy- ‘duty nylon twine. (I found tubing sections that telescoped together in progressively smaller sizes from 11/4 inches in diameter at the bottom to V-ineh diameter at the top.) I anchored the guy lines using small hooks serewed into my wooden shake- shingle roof. Guy lines should be insulated ‘material. [paid $3 forall the twine required for the array. The hooks cost $5, Three ‘ceramic stand-off insulators serve as base ‘mounts for the 20-fo0t pipes. Small glass bottles will do nicely as base insulators if important ‘ean be quite high in this array, Ground System Any discussion of ground s) vertical antennas brings up work on ele vated verticals by Arch Doty (K8CFU) and Al Christman (KB8D."* Four slightly ele- vated radials are as effective as 120 buried radials. “Slightly elevated” can be 15 feet at 1.8 MHz, Many hams are experimenting 10 Fig 5 —ELNEC was used to compare the Triband Triangle (curve A) with a 20-foot-long vertical antenna (curve B) and a Vé-i. vertical (curve ) on 10 meters. This isthe ‘lovation plat or all three pattems. The graph is scaled to show the Triband Triangle ‘maximum gain at 0 dB, This 0-dB reference is actually 5.24 El. The elevation angles of ‘maximum response for the three antennas are 11, 18 and 26° with new ideas for radial systems. Previous “conventional wisdom” is collapsing tunder the weight of new ideas and models that seem to work quite well in practice. | tried several ground systems in my experiments. | tried no ground, different grounds for each antenna, and @ common ‘ground. By far the most effective and elec- trically stable approach is a common ground, No violation of conventional wis- ‘dom here! A rule of thumb scems to be that a more extensive radial system is better, ‘Some advantage in pattern symmetry seems to result when the radial lengths are equal to the vertical element height. Fig 4 shows ‘an overhead view of the radial system, Other Configurations This array could be extended to 20-foot spacing and 40-foot verticals for a 20, 30, and 40-meter array. Additional sealing should be possible for 40 and 80 or 80 and 160-meter operation. It may be possible to use the Triband Triangle in conjunction with clevated vertical techniques to pro- vide multiband Iow-irequency systems. ‘The variations seem endless, and this work may serve as a model for other low-band systems. ‘This array might work over L.S octaves. For example, three 33-foot verticals with 20-foot spacing could be used between 7 and 21 MHz. There are two areas of con- ‘cern that should be noted for such an array, The shorter element lengths at 7 MHz will lower the feed-point impedance, making it more difficult to match. At 4 MHz, the lengths approach 12 A, raising the input impedance to a high level. This may also ‘make it more dificult to match the imped- ance. Array balance may also be more difficult to achieve. These potential prob- Jems are certainly not insurmountable. Gain Calculations Antenna gain is often given relative to reference isotropic or dipole antennas. It seems more reasonable to use a single Vi-2 ‘ground-plane antenna as the reference for the Triband Triangle. Be careful about comparing the Triband Triangle gain fig- tures with horizontally polarized arrays such as Yagisand quads. Such comparisons may be misleading. Many factors influ ence E- and H-plane gain characteristies, ‘These include the Brewster angle, nearby ‘objects and antenna characteristics. Keep in mind that optimum azimuth gain with ‘Yagis and quads is achievable only with hhigh antennas.* That would violate my goals of small size, low cost and easy in- stallation. It is interesting to compare the Fig 5A graph with Yagi-antenna patterns from Chapter 5 of Yagi Antenna Design. Using a single 4-2 ground-plane an- tenna as a reference, the favored-direction gain is © 64 dB at 10 meters (3.4 dB phased- array gain + 3.0 dB gain from increased ‘element length) + 5.6 dB at 15 meters (3.6 dB phased- array gain + 2.0 dB gain from increased clement length) # 4.0 dB at 20 meters (3.8 dB phased- array gain + 0.2 dB gain from increased clement length) ‘The higher gain figures at the frequencies result from the longer electricallength of the elements, The phased-array szain figures are taken from The ARRL An- tena Book [Computer gain calculations indicate the figures given above may be somewhat optimistic —E2d.] Fig 5 shows clevation-plane radiation patterns on 10 meters, calculated using ELNEC, a method ‘of moments analysis program. This plot ‘compares the gain of a V2 vertical, a 20- foot vertical and the Triband Triangle. Impedance Matching ‘Take care to make the system symmet- rical. Antenna tuning should be consistent for all three directional settings. An exact match may be difficult to obtain, but a near-ideal compromise setting can be ound for all three direction settings corre- sponding to the phone and CW bands. 1 obtained an SWR of 1.5:1 orbetterover the desired bandwidths forall three directions. This permits switching directions without, requiring me to adjust the matching net- ‘work. OF course with some “tweaking” of the tuner, you can optimize the match to ‘squeeze out that last fraction of a dB. Follow these guidelines forthe best sys- tem symmetry. 1) The feed linesto each element should be identical. (Remember to count the inches inside the relay box!) 2) The three elements should be mounted as far away from other objects as possible (walls, tees, power lines, and the Tike). 3) The ground-radial system should be as symmetrical as possible. This means identical radial lengths laid out at identical angles around each element. 4) The vertical elements should be per- fectly vertical. Conclusions ‘This antenna is relatively easy to build. Construction requires a minimum of time, ‘money, and materials. The design is inher- ently broadband over at least one octave. It is advisable to erect a simple single-cle- ment ground-plane antenna to check your location for noise before considering the Triband Triangle. Vertical antennas are ‘more susceptible to noise pickup than hori- zontally polarized antennas. A disadvantage of this array is the lack of front-to-back rejection when receiving, al- though I measured 20-dB side nulls on ten meters.® The DX performance-per-dollar ratio of this antenna system is hard to beat if you have a location that i relatively free from electrical noise. This array will not ‘compare with high Yagis or quads, but will, compete admirably against the more com- mon roof-mounted Yagis and quads. Notes 'y, Hall, Ed, The ARRL Antenna Book, 15th ed (Newington: ARRL, 1988), Chapter 29, “Topical Bibliography on Antennas.” See the "Phased Array” opie, p 29-17, 7A. C. Doty, Je J. A, Frey and HL J. Mills, “Biticient Ground Systems for Verical Ar tennas,” OST, Feb 1983, pp 20-25, 34, Christman, “Elevated Vertical Antenna Sys- tems.” QS7, August 1988, pp 35-42. Ay, Lawson, Yagi Antenna Design (Newington ARRL, 1986), Chapter 5, “The Effects of Ground." See the. “Antenna Performance ‘Over Ground!” and “Best Height” sections, pp 5-610 5-14, 51, Hall, Ed, The ARRL Antenna Book, 16th ed 'ARRL, 1991), pp 8-6 and 8 in figures in previo editions are Accurate gain measurements can be made with the receiver AGC “off,” as long as the r= ceiver is operating ia its linear response region (usually signals less than $9) Simply measure the detected audio voltages and take the ratios. Other References J. Devoldere, ON4UN, Low Band DXing (New- ington: ARRL, 1988). "Making Tower-Mounted Half Slopers Work for You By Duane R. Sanderson, WéTID. 3735 SE Stanley Rd Tecumseh, KS 66542 leur Radio, like most things A in life, tends to follow trends. ‘The current trend in station equipment typically involves factory-built transceivers, amplifiers and antennas. To- day's amateur equipment has advanced to a high level of performance and sophisti- cation, In fact, we have nearly reached the point where average hams cannot create anything of similar quality on their own. ‘There are many amateurs, however, who still possess that inner desire to create at leasta portion of their station layout with their own two hands. With this idea in mind, the station an- enna system provides a great opportunity forcreative experimentation and discovery without the need for elaborate and expen sive test equipment. This is especially true for various types of wire antennas. Sooner or later, however, most amateurs feel the need to erect a tower. ‘The typical project will often include a ‘tower, a triband beam or monobanders, and additional antennas for VHF and other bands, Foceasionally work amateurs on 10, 15, or 20 meters who have towers and beams. When the discussion shifts to the topic of working the lower bands, Fusually hear responses such as, “I don't work the lower frequency bands, I don’t have room forlargerantennas,” or, “Yeah liried some slopers on my tower, but [ couldn't get a decent SWR. It was a waste of time,” Despite such a pessimistic assessment, T ean assure you that iis possible to create an all-band antenna system, on one tower, in a backyard situation. The result can be an antenna array that performs very well without adding a great deal of time and expense toa basie tower project. ‘The motivation to begin my own project ‘came one spring evening about a year ago, when a tornado devastated my QTH. It 12 totally destroyed my home and my anten- nas, creating the need to literally rebuild ‘everything from the ground up. Despite the bleak task that confronted me, [realized that I eould use the opportunity to correct Some of the shortcomings in my original antenna layout. The result is @ beam and half-sloper combination tha provides very satisying results to this day. The Tower Since I have been a ham for over forty years, enough time has passed to reduce my enthusiasm for climbing towers. There- fore, I chose a tower that permitted a ground-level work position, My require- ‘ments included 1) atilt-over type design approximately 50 feet high with hand-erank operation, 2) multi-level guying eapability, 3) a large poured-concrete base, and 4) a multiple ground-rod system with ‘cach rod driven deeply into the ext, ‘All of these requirements are fairly common considerations and were achieved with standard hardware construction tech- niques. Guy Wires and Overall Tower Resonance | deviated from the standard practice of breaking up guy wires with insulators. Ac- wally, I did not place insulators in any of the guy wires. All of the guys make direct ‘metal-to-metal contact at their attachment points with the tower and with the grounded guy anchors. Thisis an important step in lowering the natural resonant fre- quency of the tower to its lowest possible point. By enlarging the electrical size and ross section as much as possible, the tower, the beam and the guy wires combine to form a large vertical mass with its own resonant frequency. This is a crucial point to keep in mind when considering the use ‘of tower-mounted half slopers As you physically view any tower, you are looking at a metal structure projecting above the earth. The bottom is usually ‘grounded, making it the low-impedance end. As you progress up the tower, the impedance inereases significantly. If you are about to install a half sloper and you select an operating frequency that is eiose to the resonant frequency of the tower, you will quickly discover that you can't feed it! The impedance mismatch is so bad, the low coax impedance (52 ohms) can’t even begin to transfer power tothe tower/sloper ‘combination. The impedance on the tower isjust too high. Small wonder that so many amateurs have had so much difficulty with tower-mounted half slopers! It is a very frustrating experience to search endlessly forthe resonant frequency of the half sloper you have just put up, tunable to find a point where your pruning. will drop the SWR to an acceptable value. ‘The grim fact is, a high feed-point imped- ance creates a large SWR reading that masks any efforts (0 prune the sloper for a resonant frequency. Obviously, low tower resonance is a ‘must for successful operation of half sloper antennas. The benefits of this approach should become clear as we progress through the tune-up and usage of the slop- Other Advantages of Uninsulated Guy Wires, There are two other advantages of the uninsulated guy wire approach: It creates a large ground footprint which helps to dis- sipate direct lightning strikes, and it produces a large RF ground footprint Which enhances the ground return of any antenna connected to the tower.A Word of Cau The guy wires should be maintained in a reasonably taut condition. This insures that the metal-to-metal contact is constant, eliminating any noise that may result from loose connections. My installation is com- pletely noise-free in this regard. Antenna System Goals ‘The following criteria were established for the antennas on my tower: 1) the provision to operate on all HE bands from one tower, 2) coax feeds that do not require coils or traps, 3) survivability in severe weather and high wind conditions, “4) broadband operation without the need fora tuner, '5) good performance day and night, and 6) no additional towers or supporting masts required. This is a fairly demanding list! How- cover, with the suecessful installation of half slopers, all of the requirements were met. Past Experience Over the years, Ihave tried most of the HF antennas described in The ARRL Antenna Book. The sloper type antenna usually gives better performance than hori zontal dipoles and has demonstrated good DX ability as well ‘The shortcomings of a sloping dipole have been evident at my QTH in terms of wind and weather survival. The sloping pole has a hanging coax connected to its ‘center, Not only does the coax whip around in the wind, but it is also subject to the stresses of ice loading. The result has been, the frustration of watching the coax break ‘ff in the middle of a winter storm when it's not possible to do anything about it ‘until good weather returns! The half sloper does not have this problem. T would like to say at this point that a collection of half slopers is not the only antenna arrangement that will provide all- ‘band performance on a single tower. A ccenter-fed Zepp with a tuner would accom- plish the same for many amateurs. However, considering the six goals [listed earlier, the half sloper is the best choice. Putting It All Together Ahalf sloper is a Vi-A wire fed with 52- ‘ohm coax at the top end where it is at- tached, with an insulator, to the top of the tower. The braid of the coax is connected (0 the tower and the center conduetor is, connected to the sloping wire The length of any half sloper is com- puted using the standard quarter-wave Fig 1—The upper half ofthe tower with guy ines and sloper wires visible. antenna formula (¢n = 234/fyi) cut my slopers several inches long so that I could prune them down to the desired resonant frequency. This isa lot easier than solder ing small lengths of wire onto a short sloper! I used RG8X mini-foam coax to feed the slopers. While I prefer the smaller diameter and greater flexibility of RGBX, RGB is obviously the more durable coax. T used stainless-steel hose clamps to secure the coax braids to the tower. I sol- dered the center conductors of the coax directly to the sloper wires. Be sure to use. ‘a good quality moisture sealant on the ex- posed coax opening around the braid and center conductor insulation. Twelve-gauge solid copper house wiring was used for the sloper wires. All insulators were made from 12-inch schedule 40 PYC pipe. If you were to look down on my tower from a helicopter, it would look like a spoked wheel without a tie. The slopers ‘and guy wires fan out around the entire tower (see Fig 1). The horizontal separa- tion angles between adjacent slopers varies from 15° to 30°. The vertical slope of cach ‘wire is about 45° with the exception ofthe. 160-meter antenna, which is long enough to require a much flatter angle. With this ‘arrangement there is virtually no interac- tion between the slopers. There is one cexception—that I use to my advantage! 13It seems that most amateurs usually have to cope with the old problem of 75 versus 80 meters. fer all, everyone wants an antenna that will work both ends of that band. A tuner could be used to alleviate the problem, but my goal was tocreate alayout that would not depend on a tuner to operate properly. ‘While the half slopers on my tower were fairly broad-banded, I shared the same 75- ‘versus 80-meter dilemma. So, Iput up nvo slopers, one for 75 and one for 80. The ‘75-meter sloper resonated at 3800 kHz and. the 80-meter sloper resonated at 3580 kHz. Unfortunately, pruning either sloper had some effect on the other. tried positioning them on opposite sides ofthe tower, but the {interaction till existed, With alittle experi- mentation I found that pruning the 80-meter antenna produced a greater effect fon the 75-meter antenna than vice versa. ‘That was when I made a discovery that tumed out to be a real bonus. With the 75-meter sloper connected to the transceiver, I shorted the 80-meter coax inthe shack with a wire connected between the braid and the center conductor. As so, the 75-meter sloper jumped upward in resonant frequency to 3850 kHz. At the same time, the signals received from the ‘east suddenly increased about 5 dB on the § meter. ‘My 75-meter sloper is on the east side of the tower. The 80-meter sloper is on the west side of the tower. The shorted 80- meter sloper became a reflector for the ‘75-meter antenna, effectively creating a two element vertical beam! Another Bonus Believe it or not, you can feed two or ‘more half slopers with the same coax line. It’s not a complicated procedure. Just tie the top ends of the slopers to the same coax center conductor, As of this writing, TRave ‘not tried o feed more than two slopers with ‘one coax. Two slopers fed by one coax are best suited for a common feed when they are positioned next to each other. advise against trying to feed both the 75- and 80-meter slopers with the same ‘coax. I tried it and soon realized that I was. ‘going through an endless and unsuccessful attempt to unseramble the interaction. Interaction With Guy Wires ‘The guy wires appear to be neutral as far as any noticeable disturbance to nearby slopers. The guy wires at the top of the tower Slope downward at about 25°. A sloper trailing away from the tower at 45° hhas a considerable distance developing be- tween itself and the guy as the wires 14 progress downward. This permits a sloper to be positioned within inches of a guy at the top of the tower with no noticeable effect. The bottom ends of all my slopers are tied off past the end insulators to metal fence posts driven into the soil. The loca- tions of the metal posts were chosen to provide the direction, spacing and 45° slope angle mentioned above. Planning for the Worst, ‘The coaxial feed lines are taped to the tower legs at one-foot intervals all the way up the tower. This is my own standard for securing coax since I don’t want to see anything moving in a SO-mish wind. ‘Midwest ice storms can take down all but the strongest of antennas. Even so, ice-loaded sloping wires are mechanically ‘more durable and will survive better than horizontal wires, The strain load on the tower is generally downward and balanced with a reduced likelihood of tower damage from excessive sideload stress. Tune-Up ‘Without exception, every one of my half slopers was resonated (0 the spot in the band I wanted, and produced a 1:1 SWR. ‘Assmall amount of pruning was necessary ‘on most bands to locate the optimum reso- rant frequency. The only serious deviation from the Vi-A formula was on the 160- meter band. The 126-foot formula length proved to be too long for resonance at 1850 KHz, Since it was so near to the earth's surface, I suspected that the low fractional wavelength on the vertical position of this sloper was effectively lowering the reso- nant frequency. I had to prune off several feet to bring it on frequency. 'SWR Measurements All half slopers will produce a 1:1 SWR at their resonant frequencies. The band- widths vary, with the 160, 80 and 75-meter slopers covering at least half of the band. with a 1.5:1 SWR. The 40-meter sloper covers 200 kHz of the band with a 1.5:1 'SWR. The higher frequency slopers cover their entire bands with a 1.5:1 SWR. How Well Do They Work? How well do they work? Very well— both day and night! Half slopers are more or less vertically polarized, which is excellent for low-angle nighttime propagation. Some propagation ‘experts suggest that vertically polarized signals are tilted off the vertical plane when they travel near the earth’s surface. This it condition would help explain why all types ‘of slopers perform so well. There also ap- pears to be a very small amount of direc tivity in the direetion of the slope. Thave made comparisons at my station between an 80-meter full-wave horizontal loop and a sloper. Daytime performance slightly favors the loop. Atnight, the sloper ‘will be generally one to two $ units better. Band Coverage [As of this writing, Ihave erected tower- mounted half slopers for the following bands: 160, 80, 75, 40, 30 and 17 meters. "Tho 75- and 40-meter slopers are fed by the same coax, as are the 80- and 17-meter slopers. The 40-meter sloper has a low SWR on 15 meters and operates as a Yi-h antenna on that band. T usually use my triband beam on 15 meters, sothe choice is optional Thave left one band out: 12 meters. In ‘my case, 12 meters happens to bea band of low interest. However, it’s just a matter of ‘connecting one more wire (o the array and ‘choosing an existing sloper as its mate so that both can be fed from the same cable. I will probably connect itt the 80/17-meter pair, feeding three slopers with one coax. Interaction from the Tribander Beam Rotation of the tribander beam on the top of the tower does not have asi ‘effect on any of the slopers. I would expect a full-size 20-meter Yagi to produce some noticeable effects. Certainly a 40-meter ‘Yagi would eveate a large overhang of over- all tower mass that would influence the radiation angle significantly. Other half-sloper users have reported that changes in sloper directivity become vvery noticeable when big Yagis ae rotated, ‘The difference between the ends of the Yagi elements hanging over the slopers versus the broadside portions of the ele- ments in the same position appears to be the mechanism that varies performance. It ‘can be readily seen thatthe angle of rai tion would be modified in this case. Investigation of this phenomenon will probably bea future adventure at my QTH. Conclusion have been using half slopers in the windy and turbulent midwest for eight ‘years now. No other antenna has demon- strated their combined survivability and performance. It is a good feeling to look fut the window on the morning after abi storm and see that the sloper farm is still ‘one piece. Give these half slopers a try on your tower. If you can keep the tornadoes. way, you'll have the same satisfying view from your window too!The Double Cross Vertical Antenna By Robert Wilson, AL7KK Box 110955 ‘Anchorage, AK 99511 bennas have been part of my A business for years and T have placed my designs at large sae ions in a number of counties. (Money is no object with the big commercial ays) Even hough design antennas fora living, 1 sil ket ply wth them at home. Part of he fun of antenna constuction i pro- ducingahigh qalityfficentantennathat istruly low in cos One snowy Alskan affemoon Twas inspired to design a home-brew antenna project. Simplicity was the primary consid Eraton.Ithadtobesomething could build withusta couple of oa connector, afew Coat hanger and my trusty soldering on Talo had to be theoretically feasible and efficient Despite my best intentions, my fntennas havea tendency to grow Tike rab: tits Withina shor perodof time my ving foom was filed with paper and wie! ‘The Double Cross Design Where would antenna designers be ‘without computers? After a bit of work at ‘my keyboard I soon developed the antenna. As Texamined the design I noticed a unique property of the “X”: It could be ‘mounted on metal pole with almost no interaction. (The vertical center line be- tween elements was also a null line.) Stacking the antenna seemed natural, and after looking at the result (see Fig 1) the name “Double Cross” seemed natural too! ‘This antenna isa variation on a vertical ipole, Imagine two V-shaped wires, one opening upward and one downward. Now imagine that both V wires are connected 10 a coax cable at their apexes. The coax shield is soldered to one V and the coax center wire to the other V. By adjusting the angle of the V t0 70°, the antenna can be made to resonate at the desired frequency and that angle will give ‘The 2-meter experimental version of the Double Cross antenna made with no. 10 wire, coax, silicone glue and a 2x 4 board the best bandwidth. The angle does not have to be extremely precise. Angles from about 60 to 110 degrees do litle to chany the feed impedance. (The impedance re- ‘mains about 30 to 35 Qover these angles.) ‘The mid-line between the V wires is null, line. This permitsthe support pole and coax Feeder harness to be mounted in the center line area with no problems. ‘The angle of the V apex is set easily by adjusting the tip-to-tip spacing of the V clements. On paper and in actual practice it seams to be reasonably noncritical. For ‘example, atip-to-tip error of £59 seems 10 make little difference in the operation of the antenna, Italso occurred to me that grounding the topmost elements on the antenna would offer lightning protection for the receiver. ‘Turning the top dipole upside down and ‘mounting the hot element of the next lower dipole upward also seemed to be @ neat symmetrical method to balance the antenna currents, Feeding the dipoles 180° out of phase is all that i required for both protec- tion and balance. This can be easily accomplished by adjusting the length of the feed hamess (sce Fig 2). ‘Old hams told me that using two wires ina ¥ would not give a circular pattern because it was not balanced. This “old hham’s tale” is simply not true; the pattern of the Double Cross is only 0.5 dB out of, round, For the sake of a litle irregularity there isn't a compelling reason to increase the complexity by adding more wires. To prove that more complex designs were not necessary, I constructed three-dimen- sional version of the antenna and found it vwas indeed very difficult to handle, After a thorough examination of the calculations. and construction of the flat “X" antenna, I proved that a two-dimensional design was entirely sufficient 15=3 an Fig Double Cross antenna design, After 1 built the first coat-hanger “X"" antenna [also discovered that the measured bandwidth was better than expected. The clement diameter was increased to 4 inch fon the 2-meter theoretical model and this allowed the antenna to cover more than the full band, ‘Table 1 provides element lengths, tip: to-tip spacings, element diameters and Phasing-line information for most ham bands, Table 2 offers the same information for 1 to 30 MHz in arbitrarily numbered bands, Incidemually, using wire for the ele- ments is quite acceptable even though tubing is indicated. For ham operation a sufficient bandwidth can be obtained by using thin wire elements, A single stacked “X" dipole could be strung between two trees for low-fre- quency operation, Attempting the Double Cross stacking arrangement on 80 or 160 ‘meters would be unreasonable, bua single tunstacked “X” would perform as an excel- Tent wide-band vertical on these bands, It may be a good idea to employ a matching transformer between the 50-0 ‘coax and the antenna, Such an RF trans- former can be made by using ahigh quality, high frequency powdered core with a cross section of at least ¥2 by V4 inch for 150 ‘wats. For the HF bands the 50-Q primary should be 10 tums and the 30-0 secondary should be 8 turns. I prefer to use 18 gauge 16 Table 1 Double Cross Antenna Lengths for Amateur Bands All dimensions are int frequencies, MHz. uous Band Freq. Element Tip-ip Spacing 160 1,90 111.05 127.38 323.68 80 © 3.70 5703 65.41 166.22 75 3.90 $4.10 62.08 157.69 40 7.15 2951 3385 86.07 90 1013 2084 2390 60.74 2 1418 1489 1707 4339 17 1811 11.65 19.8 93.96 152123 994 «11.40 28.98 12 2493 848 © 9.71 2467 10 2850 740 849 21.58 6 5010 421 48a 12.28 2 148.00 145 1.68 4.21 3. “Bandwidth” shows the lower and upper SWR “ow 6 Coax Coax2 Diam. Bandwicth 25896 8545 304 1.72.1 191.64 4388 156 33-41 12489 4163 148 3543 6812 2271 081 64-79 48.11 1604 087 9.1—11.1 3496 1145 041 128-156 2690 897 092 163-199 2295 765 027 19.1239 1954 651 0.23 224-274 1709 570 020 25.7314 972 324 012 45.1854 334 1.11 0.04 191.4-160.6 a Table 2 Double Cross Antenna Lengths for Various SWL Bands Al frequencies, MHz u Bw ous Band Frog. Element Tip-tip Spacing 1 100 211.00 242.02 618.00 2 1.20 175.83 201.68 51250 3 1.44 146.53 168.07 427.08 4 473 121.97 139.69 355.49 S 207 101.99 118.92 297.10 6 249 8474 97.20 246.99 7 299 7057 80.94 205.69 8 358 5894 67.60 171.79 9 490 4907 56.28 143.02 10 516 4089 48.90 119.19 11 619 34.09 99,10 99.35 12743 2840 9257 82.77 19 892 2365 27.13 6a95 14 10.70 1972 2262 57.48 15° 1284 1643 1885 47.90 16 1541 1369 1871 39.91 17 1849 11.41 1309 9326 18 2219 951 1091 27.72 19 2662 793 9.09 23:10 20 91.95 660 757 19.25 rensions are in feet. “Bandwidth” shows the lower and upper 2:1-SWR “os 6 Coax 1 Coax2 Diam. Bandwictn 487.08 16238 5:77 091.1 40590 13580 481 1.11.8 93825 1275 401 19-16 28155 93.85. 394 16-19 23590 7843 279 19-23 19561 6520 292 22-27 162.90 5430 193 27-23 196.06 4535 161 32-39 11927 9776 134 39-47 9440 8147 1.12 46-57 7869 2623 093 56-88 6556 2185 078 67-82 5461 1820 065 80-98 4852 1517 054 96-118 9793 1264 045 11.614. 3161 1054 037 139-17.0 2634 878 031 166-203 2195 7.82 026 200-244 1830 6.10 022 240-293 1525 5.08 0.18 28,8-35.1 enn wire with Teflon insulation, wrapping the first winding and then interlacing the sec- ond. 1 also like to tie down the ends with fishing line and coat the transformer with heavy layer of clear silicone glue. A prop- erly constructed transformer should last up to 50 years—if it doesn’t take a direct lightning strike!Design Calculations A simple four-function calculator can be used (0 calculate the antenna dimen- 1) The lengths of each leg of the V elements (LI): — 64.2 meters ae 211 feet fate ine 2) The tip-to-tip distance (L2) of the ‘open endof the V element gives a 70° apex angle: L2= 1.147 x LI (results in either feet or ‘meters, according to the original L1 values) 3) The separation of two “X” is SA or L3: _ 187.0 meters _ 615 feet fre fun 4) LA represents the length of the 50-0 solid polyethylene dielectric coax (veloc: ity factor 0,66) from the upper dipole to the summing junction where upper and lower dipoles are connected: ements L3 A meters 0.66 _ fate 5) LS represents the length of 50-0 solid polyethylene dielectric coax from the ower dipole o the summing junction men tioned above, and the length of the two parallel 70-2 coaxial cables used for im- pedance matching: x ise 180.66 = 246 feet x0.66 fute 6) L6 represents the diameter of the V legs required for the indicated bandwidth, However, ordinary wire also works well for practical ham antennas futne 1,759 moters _ 69.2in. _ 5.77 L6 = ee = ute fume fiaste Double Cross Construction Construction of the first "X” antenna was accomplished with coat-hanger wire soldered to acoax connector. The coax line ‘was connected and the free end was pulled through a ¥2-inch support pipe for testing. ‘The results were excellent. The SWR was low and the bandwidth was exactly as cal- culated, ‘The next step was construction of a stacked 2-meter Double Cross antenna with a complete coaxial feed harness. This ‘was done using no. 10 Copperweld wire salvaged from an open-wire telephone sys- tem, It was serewed to a2 x 4 board and oe (pieces of 7091 co08) Fig 2—Double Gross coax hamess, glued in place with silicone adhesive. There is no question that this was @ mini ‘mom cost antenna! Once again, everything ‘worked fine and the SWR was 1.1:1 over the entire 2-meter band. ‘The stacking hamess shown in Fig 2 is designed for a 180° phase reversal. The shield of the %4-A section of 50-0 coax goes to the top V. This is length LA. The shield of the V4-A 50-Q coax section goes to the bottom V. This is length LS. ‘The two 50- phasing lines are sot- dered together at a common point. The impedance at this point becomes about 36 Q because there are two 50-2 lines in parallel. Matching is easy with a “Q” section, a Ved series matehing section made from 42-Q coax. Two pieces of solid polyethyl- ‘ene coax are cut to length LS and soldered in parallel (shield to shield and center to center) to make the "Q" section. Tused two pieces of 70-0-coax andachieved an excel lent match, but itis possible to substitute ‘one length of 95-0 coax if you need 10 improve the match further. Examples of 95-2 coax are RG-180B or RG-195A. Ale tematively, it is possible to build short pieces of 42-2 coax from brass hobby tub- ing. The inner conductor needs to be 0.217 inch or about 7/32 inch, and the outer €on- ductor's inner diameter should be 0.5 inch, However, I recommend starting with two parallel 70-Q coax cables first. ‘All harness connections should be kept shor. I prefer to connect all center conduc- {ors first, Make one last check against Fig, 2 to be sure that everything is correctly connected, Then smooth the joints and ‘wrap them tightly with Teflon plumber’s tape. Now connect the shields, taking care not to melt the polyethylene insulation. A wet cloth will quickly cool the joint after soldering. Let it cool for several minutes because the polyethylene core cools much more slowly than copper. The joint can 7then be dried and coated with siicone glue to make it waterproof. If you insist on a neater appearance, slide some shrink tub- ing over the joint before the silicone hardens and shrink it, Wipe off the extra silicone for a first-class job. ‘The final stepistousea VOM and make a resistance check of the antenna with the coax in place. The path from the center conductor of the coax to the shield should 18 exhibit an infinite resistance (open circuit. ‘The path from the center conductor to the {wo inside V elements should show a short. ‘The resistance from the coaxial shield to the outside V elements should also indicate a short. Route all coax straight down the middle linc ofthe antenna, Secure itin place so that, ‘wind, ice and time will not change its loca tion. ‘These Double Cross stacked dipoles have given me the extra low-angle gain necessary for improved 2-meter coverage. L keep thinking about how well a long dipole stack would perform on the UHF bands and how nicely a 20-meter Double Cross could work—if I could only gettwo tall tees to grow in my muskeg swamp!Ground Planes, Radial Systems, and Asymmetric Dipoles Hindhead, Surrey GU26 6S) England everal years have elapsed since 1 S frst drew attention tothe advan- tages of replacing 2/4 radials by ‘much shorter ones sharing a common load- ing inductance.' This greatly reduces the space occupied without adding signi cantly to the losses, and through being “shrunk” in this way a “ground plane an- tenna” (GPA) becomes ideally suitable as parasitic element for use in close-spaced beams (which can be fully assembled, tested, and used at or near ground level). ‘This also overcomes another problem: that of ensuring equal current amplitudes and correct phasing when resonant wires of high Q (and subject to finite tolerances) are connected in parallel Unfortunately this topic has remained clouded by controversy. It is obvious that a true ground plane could hardly continue functioning after “shrinking” in such fash- ion. Consider such a ground plane: a fat ‘metal sheet extending for a sufficient di tance from the antenna base, asin Fig A. ‘The antenna would then have a radiation resistance, R,, of 36 @ independent of its height above true ground (from which it ‘would be shielded by the ground plane). ‘When a ground plane is replaced by a set of 1/4 radial wires (Fig 1B), no such shield exists, We have an antenna with two poles. One (the radials, collectively) is con- structed ina way that prevents it from radiating. The otheris a short wire that (like any other such wire) produces a symmetri- cal cos @ radiation pattern that is not influenced by the presence of the radials. Since the radiating element is only half the length of a 2/2 dipole, it follows that the GPA .R, is equal to that of a dipole divided by 4,n012, otherthings being equal. There fore, it has a value of 18 @ (remote from. ground influences) plus a small correction. {0 recognize the 0.4-dB gain difference ” Fig 1—A comparison of ideal and real GPAS. At A, the ideal. A 2/4 radiator is located ‘above a sold, conductive plane that extends several wavelengths in all directions. The radiating elements not visible, electrically, from the shadow of the plane. B shows a typical GPA installation; the 1/4 radiator is iocated above several 1/4 wire or tubing radials. The antenna is electrically visible from all points on the ground. between short and 2,2 dipoles; this brings itup to about 19.5 ©. Asheight decreases, the free-space value of R fluctuates in the normal manner (as a consequence of ‘ground reflections) and rises to the usual value of 36 Q at zero height. This is shown in Fig 2, which was derived from mutual- impedance data (assuming perfect ground tnd treating the antenna with its image as. a collinear pair). Despite approximations and although the heights plotted are those of the feed point, this is almost identical with results obtained by Fletcher, \VK2BBF, on the basis of rigorous mathe ‘matical analysis? As Fletcher points out, ‘the height at which the impedance changes 10 its “elevated” value is remarkably low. Except at very low heights (where errors appear to cancel), the “center of gravity” of the current distribution more accurately represents the true height, and the graph has been drawn accordingly. (To obtain the ‘mutual resistances, 0.08 2 was subtracted from the heights used.) Ilustrating this, ‘mutual resistance is zero and Ris therefore ‘equal to its free-space value of 19.5 9 for separations of 0.7 and 1.2 2 between sources (corresponding to radial heights of 0.27 and 0.52 3). Recognition ofthe GPA as an asymmet- rical dipole is at odds with much “conven- tional wisdom.” It requires both convine- ing proof and answers (© a number of questions before the opportunities can be fully grasped, For example, it is well known that significant inductive loading 19ex} Ae oo a2 aa aa oe 0 acl Height (8) (equa to 0.00) Fig 2—Radiation resistance, R, versus feed-point height for a 4/4 GPA over Perfect ground. The calculation was performed for a pair of point sources Separated by 2hp, where ho isthe height ‘measured to the conter of current distribution. The marked points are those for which R should equal ts free-space value of 19.5 0 (see text), as derived by Yigorous mathematical analysis, Note the agreement between this chart an that in ett seriously degrades the efficiency and band- width of dipole antennas, so why should GPAs be exempt? What are the best meth- ods, the limitations, and the penalties? What are the ground influences and the implications for feeder systems? Are these ‘questions interrelated, of ean they be con- sidered independently? Are there implica- tions for other types of antennas? Armed with a few of the answers and a belie that the others were within easy reach, [et out in search of a complete set of guidelines, ‘encountering a number of setbacks and not 4 few surprises betore the main objectives ‘were achieved, Since the radials have no screening or reflecting properties and form no part of the radiating system, they can be reduced in length without any direct influence on efficiency. Nonetheless, we must take due account of losses in any loading compo- nents needed to maintain resonance, and the radials must not be hindered in dis- charging their proper functions: providing, the only return) path for the antenna cur- rents and anchoring the coaxial cable shield firmly at ground potential AAs stated in the text of Ref 1, initial experience with such arrangements. was encouraging (including their use as para sitic elements), though at that time a ‘umber of questions relating to (1) basic limitations and (2) the influence on band- ‘width had noc been fully resolved, Further, there appeared to be no reliable informa. tion on ground losses and none at all on how these might be affected by reducing the length of radials, 20 ° Fig 3—Loading of radials. Aand B show a full-size GPA antenna and a vertically oriented full-size dipole, respectively. C ‘and D show similar antennas that have been reduced in size by inductive loading. Shee Wire, Loaded Couing Loos so) Fig 4—An asymmetric horizontal dipole used to verify current distribution with the equivalent of very shor radials, This was the starting point for what eventually became a protracted investiga tion covering all aspects of asymmetric oles, including an extension of short- ‘adial principles. Most ofthe observations can be explained on the basis of equivalent circuits and relatively well-established be- havior of horizontal wires close to earth ‘There were three major topics, as follow. Radial lengths: Predictably, radial lengths ean be varied between extremely wide limits (From 4/4 to 2/20) with no significant effect on SWR or field strength, and only about 50% reduction of band width. Less obviously, the radial height below which losses from currents in the ground become appreciable is extremely low and independent of radial length There were no discernible losses in para sitic elements, Attention was also directed 10 the need for nonresonant radial wires, in order to avoid unequal current amplitudes, and phases. End-jed 222 dipoles: The results indi cated that half-wave dipoles ean be end fed with coaxial cable, without losses, feed- line radiation or bandwidth restrictions characteristic of other methods of end feed- ing. Bandwidth has proved to be readily caleulable, but other problems can arise and have been identified Resonance of loaded horizontal anten- ‘nas: Other aspects include the discovery of a close link between height and resonant frequency of loaded horizontal wiresat low heights. The link has been analyzed and used to determine the height above which ‘ground losses can be neglected. Radial Loading Fig 3A shows a so-called GPA in free space with only two radials, Fig 3B shows, 4/2 dipole with an R,of 73.2::ts radiation Pattern approximates a cos 8 curve. Sincea Percatage of 25 edit Length (1) Fig 5—Variation of X/Zo and AX/Zo with radial length, where Zo is the approxi characteristic impedance of a single r or the antenna, neatly all radiation from the antenna in Fig 3A comes from wire AB, the pattern also approximates cos 0; the field strength at ‘any point in space is the same for both antennas. Because the length is halved in Fig 3A, Ry is divided by 4, not by 2 as ‘commonly assumed. This implies an initial disadvantage in terms of bandwidth and efficiency, but there is one other very im- portant and rather less obvious difference ‘between the two antennas. Perhaps it can be best appreciated by trying to fit each antenna into a small space as illustrated in Figs 3C and D. For the dipole, Rr drops sharply to about 7 Q; whereas the GPA radiator AB) is unchanged, and R,-remains at some value between about 18 and 36 Q, depending on its height above ground. In principle, the shortening process can con- ‘inue until the resistance ofthe loading coil, becomes comparable with Rr, which could mean a radial length of about 0.015 2 at 14 MH, ‘Though this option is emphatically dis- couraged, itis less strange than it appears, and correct operation was nearly achieved with a single helically wound radial only slightly longer than 0.015 2. Adjustments, though, were much too critical. I suspect, that the coax shield was making an “un- authorized” contribution to the ground system. Lengths of 2/8 to M/12 are recom- mended, and considerably shorter radials (J16 10 2/20) have been used successfully as follows. ‘When a pair of 2/16 radials were substi- tuted for 2/4 radials, antenna current appeared to increase marginally. Nonethe- less, the signal induced in a nearby vertical antenna dropped by about 2 dB. What ‘caused this? The radial currents appeared ‘tobe more than half ofthe antenna current. Although there was no apparent current on © Fria = Percentge of Optimum (aur 4/4 Rods) hor aoe a0 G1 G1 GN OIG O18 G29 G22 G20 0.26 Radi Lenath (3) Fig 6—Varlation of bandwidth with number and length of radials. A set of four 1/4 racials is taken as standard, All conductors have the same diameter. the coax shield (because the power level was too low) such current was later found {0 be the cause of the trouble. Because the direction of shield current flow is opposite to that in the antenna, field strengths are no longer directly related to the current in the radiator, and the feed-point impedance is augmented by additional losses so that complete suppression of shield currenis is critical (as deseribed later). Prior to the shield-current discovery, a nonsinusoidal current distribution was thought to be a plausible explanation (in view of the very small size of one antenna pole). This was dificult to test because most of the vertical element was out of reach, After a precarious balancing act failed to produce a conclusive result, a horizontal model of the antenna was set up as shown in Fig 4. This restored confidence in short radials and eventually provided the pattern for more efficient methods of end feeding horizontal antennas. Placement ofthe feed pointisextremely critical, in proof of my contention clse- ‘where that asymmetric feeding of antennas. is viable if, bt only if, resonance is estab- Tished in both directions from the feed point (see Ref 1, p 42). Otherwise, the feeder inevitably becomes part of the radi- ating system, ‘Another experiment at 14 MHz. used ‘wo radials, with lengths varying from At to 4/16 while observing SWR, bandwidth and the current induced in another vertical antenna at a range of some SO meters (164 fi). There was little change apart from de- creases in bandwidth (in line with that shown Fig 6) and a just-perceptible drop in field strength at each end of the range. The drops were caused by radiation from the 24 radials at one extreme anc! losses inthe Toading coil at the other. Ii interesting to note that radiation from radials decreases very rapidly with their Iength. This results in part from their smaller “current x length” product, but the main cause is closer spacing of the antiphased sources. Estimation of Bandwidth ‘The Following calculations assume three orfive conductors aradiator, and two or four radials) of equal diameter. Imped- ances (and their variations with frequency) ‘were read from the outer seale of a Smith Chart and plotted in Fig 5 (which is valid for any value of characteristic impedance, Ze). For each 1% of detuning, the radiator generates a reactance 2nZ9/400 2 (to ‘which must be added & Xin forthe radials, and 0.01 X¢/n for the loading coil). From, the Smith Chart, we also find that when Re ig matched, the SWR rises to 2 when the reactance equals 0.7 R, so the percentage half-bandwidth is given by 07 xR AX, +001 X, 2p + fq) Based on a set of four 2/4 radials as standard, Fig 6 shows the extent to which bandwidth is degraded by decreasing the number and/or length of radials. Note that a set of four 2/10 radials yields the same bandwidth as a pair of /4 radials. For aZp ‘of 500 ©) and an R, of 36 @, the standard bandwidth is 5.4% and the worst case in- cluded in Fig 6 (a pair of 2/25 radials) is 24%» round Losses (Ohms) 5 %0 “ort © Gos O10 GIS GA O28 OS O35 OAT OMG OSD OSS OHO DES B70 nr Fig 7—Comparison between perfect ground (Ref 5) and measured results for typical {ground (Ref 4) showing the steep rise in feed-point resistance of horizontal aipoles at very low heights and the effect of a ground screen (mosh = 0.0003 4), curves 1. 2 and 3, respectively. Caloulations for dieloctric ground (curve 4, k= 5) show good agreement with curve 2, except as indicated (when h < 0.2 2). Curve 3 checks clesely with theory putting k= =, Curve 5 illusiratos the effect of wet weather. Note that the atference between curves 1 and 2 below h = 2/4 consists of ground losses. 1.9%, which is more than enough to cover the 20-m phone band. So, faced with the previously mentioned problems of critical adjustments and feed-line radiation, band~ width is probably the least ofone’s worries if the reduction of radial Iength is carried to its ultimate limit consisting probably of «8 pair of short whiskers loaded by a very large coi Ground Losses When a 2/2 horizontal antennais moved down towards ground froma height of W/4, Re at first closely follows the curve for perfect ground, but it departs (increasing) from the curve at about 0.17 2, At ground level, Rrreaches about twice the free-space value, as illustrated in Fig 7. That figure is based on Ref 3, which mentions good agree- ment between measurements at SO MHz and theory based on the assumption of di- clectric ground, but questions the validity of this assumption at lower frequencies (particularly below 7 MHz and at very low Incights), though the findings seem to be in ‘200d accord with Ref 4 and my own obser- vations. Thave been unable to find similarresults with respect to vertical antennas, Nonethe- less, two observations point strongly to the conclusion that there are no significant ground losses associated with vertical an tennas as such, only in ground connections tnd radial systems consisting in the main of horizontal wires, which may beanalyzed in the same way as horizontal antennas: 1) Failure to observe any ground loss when a GPA is excited parasitically. 2) A horizontal 3/2 dipole within about an inch ofthe ground was found to have an R, of 140.0, which would be expected to translate into a value of 35 Qf its center is used as a ground connection. That is what happened. The variation with height is plotted asthe lower curve in Fig 8. Com- paring this with the upper curve, we can see that the disappearance of ground loss with increasing height is much more rapid than with horizontal dipoles, In conjunction with Fig 11 (described later, this is con- vvincing evidence that ground effects (other than as described by Fig 2) disappear com- pletely at radial heights exceeding about 2116—a result entirely to be expected. ‘To explain ths. itis helpful to start by considering a 2/4 transmission line in free space, Fig 9.1 this line isunfolded to make a dipole, its characteristic impedance in- creases 10 a value given by? % soe) - 120 (Eq2) Except for the acquisition of Re, it re- mains otherwise unchanged. Like any other 1/4 resonator, it can be approximated by an inductive reactance, X., equal to Zo ‘and tuned by the capacitance between the ‘conductors. We can therefore construct the equivalent cireuit shown in Fig 10 (minus (C2 and C3 for the moment) AAs the line approaches ground level, it reaches point where the capacitance from each pole to ground provides an easier current path than that between the poles. Fig 8 Variation of ground losses with radial height. Curve Ais a crude estimate based on dividing the values of curve 2 in Fig 7 by four. Curve B was plotted trom rough measuremonts at 14 MHz (with various radial longths). The measured values were obtained by subtracting Pe (96.0 in this case) from the total rasistance at the feed point. Note the Convergence of the two curves towards a similar high value at zero height ‘The antenna starts to resemble a pair of series-connected unbalanced lines. This is where C2 and C3 take over from Cl. The impedance of each line (see Ref 5, p 174) given by mn In terms of the equivalent circuit, this is equal to X1/2. For a given conductor of radius “a” near ground level, Zp is depend- ent solely on height. (Remote from ground influences, Z» depends only on length.) AS an example, let us take a 20-m dipole for which 2/* 10m andassume, in anticipation of Fig 11, a height of | m. Typical values of “a” range from0.5 mm foran “invisible” wire dipole to 10mm for atypical rotatable dipole. Corresponding values of Xi, (given by the free-space formula) range from 1067 to 708 © and (from the low-height formula) 911 (0 635 Q. These last two figures are extremely height dependent. This has interesting practical conse- quences as demonstrated by the lower (experimental) curve in Fig II. That curve illustrates the above behavior and reveals ‘2 method 10 determine the presence of ground losses with no instrumentation other than a dip meter. As a dipole is short- ‘ened and resonance is restored by inductive loading, the required inductance is propor- tional to the Za ofthe dipole. Where ground. losses are present, Zo is height dependent and antenna resonant frequency varies with height. (The height-dependent formuta considers the return path through ground, If it ceases to apply, it is reasonable to ‘conclude the absence of ground losses.) Note from Fig 11 that nearly all fre- (43)© © © Fig 9—Derivatione of equivalent circuits of antennas. A typical feed line is shown at A. B shows the conventional representation of sucha line in terme of lumped constants. C shows the same line opened out to form a dipole. quency change in a 28-MHz 1/4 loaded horizontal wire takes place at heights below 1wo feet. Rough tests have indicated that this figure is proportional to wave- Jength but independent of radial length It is well known that the resonant fre- quency at very low heights is influenced by the dielectric properties of the ground, but this effect is very much smaller than the fone illustrated by the curves antenna height. An example is shown as a single point in Fig 11. You can see that, in the case of the unsymmetrical dipole (Fig 10B), reducing radial length must decrease C1 and increase Zo, The top curve in Fig 8 was a first at- tempt to relate the ground losses of radial systems to the relatively well-established behavior of horizontal dipoles. The dipoles ‘were “converted” into radials by using their centers as ground connections. From inspection and fora given efficiency, lower heightsare acceptable inthe case of radials, but this is pessimistic. It ignores the field cancellation that prevents radiation, and ‘can be expected to result in a considerable reduction of ground loss (except at zero height, as already indicated). This is dem- onstrated dramatically by the lower curve (which is based on measurements of feed- point impedance) even when allowance is, ‘made for inaccuracies. spired a search for a “more accurate” method of measurement. ‘This took the form of parasitic excitation from aremote antenna, observing the effect ‘on current from the insertion of known resistances. It was here that matters got really interesting: The expected height de- pendence was nonexistent. This was in line with observations dating from many years earlier and hitherto unexplained. It leads to the conclusion that for parasitic excitation of a GPA, there are no significant ground. losses even ata height of only an inch or so. In retrospect, this is less surprising: In ‘a base-fed element at ground level, the antenna current must return to the source via the ground, whereas the signal injection of a parasitic element is distributed along Inaceuracies its length Though intriguing, the practical value ofthis discovery is limited. From the evi ‘dence of Figs 8 and 11, the radial heights at which ground losses occur are very low. (They axe more suitable for a system of twip-wires than antennas compatible with dlomestie harmony!) It also appears that ground loss, particularly with short radials, should be readily avoidable with the help ‘ofa small earth mat that covers the area of aground subject to Fields generated by the radials, See Refs 3 and 4), In this context, an interesting point has ‘emerged that tends to reconeile the find- ings of myself and others: There is no significant advantage in increasing the ‘numberof radials from two to three or four, despite advice from some quarters (such as Ref 6) to use very large numbers (up to @ hundred or more at low heights). As height above a ground screen approaches zero, the radials can be perceived as merging into it. ‘There appears 10 be no reason why the screen itself should not be constructed as 2 23Soo ig (eotnued) Derivations of equivalent ecu of antennas, Ds te equivalent cao Fig 90 (see previous page, whch eitore from B only to the extent that Land Zo are large illustrate simpler versions of the equivalent circuit for the special case of inductance for which a impedance of the generator or, in the case of an antenna, the radiation resistance Fr. set of radial wires, provided that mesh-size requirements are met. Assuming radius of Nd and a mesh size of 0.01 A (whieh, from Ref 3, is sufficient to give a result almost identical with eurve3 of Fig) the required number of radials is 79. The maximum Useful mesh size is about 0.07 2, With less than 10 radials, no significant improve- ment can be expected. Nevertheless, Fig 8 shows a simpler solution, at least for the higher frequencies: Increase the height slightly (so the radials cease to function also as trip-wires). Demands for even greater heights are impossible to reconcile with well-established properties of wires close to ground (as already discussed). ‘They may be attributable to feed-line radia tion: Alternative feed possibilities include link coupling into loading coils with no direct connection. Though used success- fully, this merely alters the problem 24 strongly recommend that short vertical radiators at low height should be rearranged as symmetrical end- loaded dipoles. Very successful beams constructed on this basis have been de- scribed by me (Ref 1, p 193) and by GAGSF/ZSOBKW (Ref 7), who used the MININEC computer program to take the -auesswork out of the design. After adding. the experience recorded in these references to the observations with parasitic elements recorded above, this appears tobe a simple and effective way of eliminating ground losses, Other Unsymmetrical Dipoles ‘Unsymmetrical dipoles along the lines of Fig4, or possibly with two “radials,” are ‘one solution to the problem of end feeding. horizontal dipoles, half-wave verticals and the like with coax. The necessity to ensure resonance in both directions from the feed. , Cis smaller, and radiation resistance (not shown) has been acquirer E,FandG /4. The inner half of the line is replaced by a single 496 = Zo. The vottage ratio E/E is given in all three cases by Zo/R or «L/R, where Ais the internal point (so as to maintain the coax shield at ground potential) cannot be too strongly ‘emphasized. To this end, a radiator length ‘of 2/4, as thus far assumed, is a step in the right direction, since with the further as- ‘sumption of overall resonance the required condition should be fully met. Shortened radial procedures can be ap- plied to other radiator lengths, horizontal as well as vertical, subject to suitable tun- ing. Methods of loading short radiators have received extensive coverage, and the reader should find litle difficulty in relat- ing them to the use of shortened radials. ‘Arrangements such as in Fig 12 appear to be new, however, and could be of some importance as they extend the idea of “end- feeding with coax” to /2 elements (see Ref 8). Fig 13, prepared from a Smith Chart, shows how this can be achieved by moving inwards from one end a short distance, AA, to find a resistive impedance suitable forssn (2) (28mm) | _ Reonont Frequency Neovured sith Dip meter (i 20 ts) i : é ? A i (© Simnetsies toosng rrr 1 Aeyrmatis Lecing + ne “4 Opole olor (0) 1 tea th /he) - = hye em (35 In) VTL EE, Fig 11—Resonant frequency versus height of an inductively loaded 1/4 horizontal dipole. 1 eos, 4 \ die hat Geantesaee 7,2 “ ELIT? eam © Fg 10—Eagl cao Sail A ee Capea snare Somes eds Beri mh be taco ah erected | Sotgedane teats \ Pa ance niga a Seles tatoner ese be erat seasoacs te sien bine” Bocetd we BR ‘maithing tocoax with the help ofa suitable transformer, The relatively large value of series inductance is easy to tune out by @ small capacitor, as illustrated in Fig 12. This technique is readily adaptable for cther types of radiator such asthe inverted. GPA, Fig 14A, or “Maypole” beam le- ments, Fig 14B (see Ref 1, p 190, and Ref 9. For Maypoles, the use of short Win-

You might also like

• Amateur Radio Transceiver Performance Testing

  

  No ratings yet

  Amateur Radio Transceiver Performance Testing

  101 pages
• Absolute Beginner Guide NanoVNA v1 6

  

  100% (3)

  Absolute Beginner Guide NanoVNA v1 6

  75 pages
• ARRL Handbook For Radio Communications 1972 49th

  

  100% (2)

  ARRL Handbook For Radio Communications 1972 49th

  708 pages
• All About Vertical Antennas

  

  100% (9)

  All About Vertical Antennas

  99 pages
• Simple, Low Cost Wire Antennas For Radio Amateurs by William

  

  100% (3)

  Simple, Low Cost Wire Antennas For Radio Amateurs by William

  191 pages
• 4 5900066337668990814

  

  100% (2)

  4 5900066337668990814

  390 pages
• 1 4963228644928389236 PDF

  

  100% (5)

  1 4963228644928389236 PDF

  196 pages
• ARRL Handbook For Radio Communications 1981 58th

  

  100% (4)

  ARRL Handbook For Radio Communications 1981 58th

  644 pages
• A Guide To The NanoVNA

  

  100% (6)

  A Guide To The NanoVNA

  79 pages
• ARRL CourseInRadioFundamentals 1972

  

  100% (12)

  ARRL CourseInRadioFundamentals 1972

  187 pages
• ARRL More Wire Antenna Classics

  

  100% (4)

  ARRL More Wire Antenna Classics

  180 pages
• The ARRL Antenna Compendium Vol 6

  

  100% (4)

  The ARRL Antenna Compendium Vol 6

  243 pages
• The Arrl Antenna Compendium Vol 2 of 6

  

  No ratings yet

  The Arrl Antenna Compendium Vol 2 of 6

  213 pages
• The ARRL Antenna Compendium Vol 4

  

  100% (2)

  The ARRL Antenna Compendium Vol 4

  217 pages
• ARRL Antenna Compendium Vol. 3

  

  100% (2)

  ARRL Antenna Compendium Vol. 3

  238 pages
• W1FB Antenna Notebook

  

  100% (11)

  W1FB Antenna Notebook

  146 pages
• Practical Antennas

  

  100% (10)

  Practical Antennas

  59 pages
• The ARRL Antenna Compendium Vol 5

  

  No ratings yet

  The ARRL Antenna Compendium Vol 5

  203 pages
• ARRL Propagation

  

  100% (2)

  ARRL Propagation

  191 pages
• CQ 2017-12

  

  No ratings yet

  CQ 2017-12

  116 pages
• W1FB - Design Notebook

  

  100% (1)

  W1FB - Design Notebook

  200 pages
• ARRL The Care and Feeding of Trasmission Lines PDF

  

  100% (2)

  ARRL The Care and Feeding of Trasmission Lines PDF

  111 pages
• Arrl 1968 Radio Amateur Handbook

  

  No ratings yet

  Arrl 1968 Radio Amateur Handbook

  706 pages
• HF Antenna Cookbook PDF

  

  100% (6)

  HF Antenna Cookbook PDF

  20 pages
• ARRL Radio Amateur Satellite Handbook

  

  100% (4)

  ARRL Radio Amateur Satellite Handbook

  360 pages
• Radio Handbook 18 1970

  

  100% (6)

  Radio Handbook 18 1970

  897 pages
• QRP Classics - The Best QRP Projects From QST and The ARRL Handbook

  

  100% (9)

  QRP Classics - The Best QRP Projects From QST and The ARRL Handbook

  282 pages
• Understanding LF and HF Propagation PDF

  

  100% (2)

  Understanding LF and HF Propagation PDF

  16 pages
• ARRL Handbook For Radio Communications 1980 57th

  

  80% (5)

  ARRL Handbook For Radio Communications 1980 57th

  580 pages
• Antanna Book

  

  26% (23)

  Antanna Book

  11 pages
• Tactical HF Field Expedient Antenna Performance Volume 1

  

  No ratings yet

  Tactical HF Field Expedient Antenna Performance Volume 1

  78 pages
• ARRL Antenna Book 1981

  

  No ratings yet

  ARRL Antenna Book 1981

  337 pages
• Six Meters Ebook V3 PDF

  

  No ratings yet

  Six Meters Ebook V3 PDF

  49 pages
• Top 5 Wire Antennas

  

  100% (6)

  Top 5 Wire Antennas

  21 pages
• QRP Classics - The Best QRP Projects From QST and The ARRL Handbook - Text PDF

  

  100% (3)

  QRP Classics - The Best QRP Projects From QST and The ARRL Handbook - Text PDF

  282 pages
• Practical Antenna Design 2014 PDF

  

  100% (3)

  Practical Antenna Design 2014 PDF

  257 pages
• Radio Handbook 15 1959

  

  100% (3)

  Radio Handbook 15 1959

  809 pages
• W4RNL Antennas Made of Wires Vol 3

  

  100% (2)

  W4RNL Antennas Made of Wires Vol 3

  789 pages
• G QRP Club 1983 Fixed PDF

  

  No ratings yet

  G QRP Club 1983 Fixed PDF

  97 pages
• W4RNL Antennas Made of Wires Vol 1

  

  86% (7)

  W4RNL Antennas Made of Wires Vol 1

  535 pages
• K7QO - QRP - Laboratory PDF

  

  100% (2)

  K7QO - QRP - Laboratory PDF

  136 pages
• 25 Simple Amateur Band Aerials

  

  100% (3)

  25 Simple Amateur Band Aerials

  76 pages
• Facts About SWR and Loss PDF

  

  100% (1)

  Facts About SWR and Loss PDF

  62 pages
• Wire 2

  

  100% (2)

  Wire 2

  616 pages
• ARRL Handbook For Radio Communications 1964 41st

  

  100% (7)

  ARRL Handbook For Radio Communications 1964 41st

  722 pages
• FM and Repeaters For The Radio Amateur 3 Printingnbsped Compress

  

  100% (2)

  FM and Repeaters For The Radio Amateur 3 Printingnbsped Compress

  236 pages
• CQ Amateur Radio October 2020

  

  No ratings yet

  CQ Amateur Radio October 2020

  116 pages
• Mag Loop Design

  

  100% (2)

  Mag Loop Design

  49 pages
• Helix Antennas

  

  No ratings yet

  Helix Antennas

  48 pages
• Cubical Quad Vol 2

  

  100% (1)

  Cubical Quad Vol 2

  231 pages
• HF Antennas For All Locations - L. A. Moxon - 1982

  

  No ratings yet

  HF Antennas For All Locations - L. A. Moxon - 1982

  272 pages
• The ARRL Antenna Compendium Vol. 3 PDF

  

  No ratings yet

  The ARRL Antenna Compendium Vol. 3 PDF

  238 pages
• ARRL AC v3

  

  No ratings yet

  ARRL AC v3

  238 pages
• ARRL AC v2

  

  100% (1)

  ARRL AC v2

  213 pages
• WF1B Antenna Notebook

  

  100% (2)

  WF1B Antenna Notebook

  142 pages
• Joe Carr Receiving Antenna Handbook

  

  No ratings yet

  Joe Carr Receiving Antenna Handbook

  204 pages
• Weekend Antennas 3

  

  No ratings yet

  Weekend Antennas 3

  5 pages
• Antenna Handbook - 1937

  

  100% (2)

  Antenna Handbook - 1937

  66 pages
• 2 Meter Phased Vert Ant

  

  No ratings yet

  2 Meter Phased Vert Ant

  6 pages
• Introduction To Antennas: Dr. Bablu K. Ghosh

  

  No ratings yet

  Introduction To Antennas: Dr. Bablu K. Ghosh

  76 pages