The Ground Plane Grows Up
May 1954 Radio & Television News

May 1954 Radio & TV News

[Table of Contents] Wax nostalgic about and learn from the history of early electronics. See articles from Radio & Television News, published 1919-1959. All copyrights hereby acknowledged.

Prior to the advent of personal computers and handy-dandy antenna design software like EZNEC, determining the effects of varying parameters - element spacing, angles and length, ground plane distance and extent, feedpoint impedance, the presence of conductive structures, etc. - it was necessary to make a series of often complex mathematical calculations and ultimately perform real-world measurement. Huge amounts of time would be invested in the design and verification process. It has been know for a long time that the distance an antenna sits above a ground plane has a significant effect on the radiation pattern - particularly the vertical pattern. The information provided in this 1954 Radio & TV News magazine article undoubtedly required many hours to assimilate and required someone (author William Harrison) with a lot of knowledge in the science/art of antennas. While some empirical testing is still needed for critical applications, in most cases these days the results of computer simulations suffice for determining the viability of a particular antenna design and installation. Nearly every month Joel Hallas (W1ZR) in his "The Doctor Is In" QST column publishes at least one antenna analysis in response to a reader's question that includes graphs produced by EZNEC simulations.

The Ground Plane Grows Up

By William H. Harrison, W6ULD

A new slant on a ground plane antenna that increases efficiency on higher frequencies.

During the past few years a great deal of interest has been centered on the quarter-wave ground-plane antenna. As many know, the important advantages of this type of antenna are low-angle radiation, ease of feeding, small space requirements, and relatively low cost. The purpose of this article is to show how it is possible to further improve the antenna by doing nothing more than going straight up a few more feet.

We will first consider the angle of radiation. Most transmissions on 20 meters take place at angles between 6 and 17 degrees, 15 meters between 4 and 14 degrees, and 10 meters at angles below 10 degrees. Anything above these angles is useful only for short skip contacts. How many of you have had difficulty working those DX stations but find that you have obtained excellent reports during short skip sessions? The reason is that the antenna is radiating a good portion of the power at the higher angles. The quarter-wave vertical antenna radiation pattern follows quite closely a cosine curve from the horizon and concentrates much of the energy at the low angles; however, at angles over 20 degrees there is still a great deal of energy radiated that is of little value. The antenna to be described is non-directional and tends to concentrate practically all of the radiation below 20 degrees. The vertical plane pattern is quite similar to that of a well designed rhombic. As the height of a quarter-wave antenna is increased, the energy radiated becomes more concentrated at the lower angles. This continues with increased height to 5/8 wavelength (0.625λ), as may be seen in Fig. 2. If the length is further increased, some of the energy in the low angle lobe begins to form in a lobe at a much higher angle, and power at the desired angle is lost. With the 5/8 wave radiator a small high-angle lobe is present, however it is a relatively small portion of the total radiated energy. Information taken from the "FCC Standards of Good Engineering Practice" indicates that maximum radiation along the horizontal takes place with a 5/8 wave radiator.

Let us now consider feeding the 5/8 wave radiator. The quarter-wave vertical antenna may be fed directly with 52-ohm coax because it has a base resistance of 40 to 50 ohms and a fairly low reactance. Measurements made on several such quarter-wave antennas indicate that they can be fed directly and will maintain a standing wave ratio on 52-ohm coax of about 1.5. As the height is increased the base resistance of the antenna increases to a point around 0.42 wavelength where it reaches a maximum and then decreases with a further increase in height. At a height of 5/8 wavelength the base resistance is again near 52 ohms, which makes it simple to feed. The resistance and reactance curves shown in Fig. 3 indicate the impedance for various heights of radiators and points of actual measurement.

The 5/8 wave antenna on 20 meters is approximately 44 feet and measurements were taken on either side of this value. It was found that the base impedance is slightly higher than indicated in various texts possibly because of a difference in base capacities or antenna cross-section dimensions. The height was increased to 47 feet where the resistive component is 50 ohms, making this a desirable height. The measured reactance is low at this height (60 ohms capacitive) which can be tuned out by a small series coil as indicated in Fig. 6. If the coil is not used the system will still work favorably as the measured standing wave ratio is only 1.5 to 1. By using the little coil, in series with the coax feeding the base of the antenna, the v.s.w.r. was reduced to approximately 1.05 to 1.

When working an antenna against ground, the ground system is very important. Broadcast stations are required to install a ground system of at least 120 radials 1/4 wavelength or greater. The main reason is to provide maximum radiation at the ground level. Without the extensive ground system much of the energy radiated at the extremely low angles is lost in absorption. In this case low angles are meant to represent those below 5 degrees. While we as hams are not particularly interested in such low radiation angles, yet it is necessary to work the antenna system against ground, and the better the ground system the less energy will be lost. At our former home in Tempe, Arizona, I installed a ground system in the back yard before the grass was sown. It consisted of 120 radials 35 feet long, using #16 galvanized iron wire.

The wires were buried 1 to 2 inches. This task wore out two hoes and the neighbors' curiosity. The coax line was buried at the same time making a neat installation with only the vertical antenna and mount showing above the ground. Two sets of guy wires were used. I would like to mention that I have found it desirable to use two egg insulators at the point where each guy is attached to the tower as well as breaking up the guys every tenth wavelength with another egg insulator. This gives assurance the guys are not connected to the tower electrically which would de-tune the system. The ground system used with my present antenna consists of 16 radials ranging from 25 to 45 feet running out in all directions from the base of the tower. One may plant as many as the wife will permit; however, make certain to use at least 4 radials.

Information available in various texts verifies the fact that added low angle radiation is obtained with the 5/8 wave radiator; however it was desired to construct a miniature 5/8 wave radiator as well as a quarter-wave vertical antenna system and actually make radiation measurements. If the frequency is increased to 1000 mc. the radiators become 7.4 and 2.96 inches respectively, making them easy to mount and rotate so that a radiation pattern may be obtained. The two antennas (individually) were originally mounted in a ground-plane which consisted of sheet iron approximately a yard square. The whole assembly was then rotated and patterns taken to determine the directivity of the antennas in the vertical plane. Due to the small size of the ground-plane the results did not resemble the desired patterns. The major lobes were forced up away from the ground-plane and minor lobes developed off the back.

The data we are interested in is based on an infinite ground-plane. It was, however, possible to use this same ground-plane to make patterns of horizontal antennas (mounted parallel to the ground-plane) because the voltage vector cannot be supported on the ground-plane and the radiation is driven up away from the ground, as indicated by the patterns shown in Figs. 7C and 7D for horizontal antenna mounted a quarter wavelength and a half wavelength above ground respectively. The final patterns for the 5/8 and 1/4 wave vertical antennas, shown in Figs. 7 A and 7B, were taken by using duplicate elements working against one another instead of against the ground-plane, that is, a complete dipole was used (see Fig. 5) so that the pattern would not be dependent upon a ground-plane of finite size. The image antenna in this case is not a mere image but the real thing and radiates equally well, thus duplicating the pattern of the top half of the antenna below the imaginary horizon. The results are identical to the published patterns of broadcast engineering firms, except of course in duplicate.

It would be interesting to make a study of the effects of the size of a ground-plane with respect to the antenna height and also the effect of tapering the ground-plane wires to form different sizes of conic ground-planes as are used by many hams. All of these features affect the angle of radiation, and of course the height of the system above the actual ground will also make some difference. In the case of 10 and 15 meters, due to the small height, it might be desirable to place the antenna in a ground-plane located above ground (depending on the individual circumstances) so that the antenna system will be in the clear.

Now we get down to the facts of how much improvement can be expected. The relative merits of the two antennas may be compared by comparing their field intensities at the desired radiation angles. This information is available in the FCC publication mentioned previously. Based on 1 kilowatt input to each antenna, the quarter-wave element produces a field of approximately 195 millivolts per meter at one mile, while the 5/8 wave radiator under the same conditions provides about 275 millivolts per meter. (See Fig. 4.) This is the field produced at zero degrees. The difference in the two fields is reduced as the angle is increased, as can be seen by the patterns in Figs. 7A and 7B. They become equal in magnitude at about 20 degrees but, as mentioned earlier, long distance communications on 20 meters takes place at an average angle of 12 degrees, 9 degrees on 15 meters, and about 5 degrees on 10 meters.

Based on the field strengths just given, we get a power gain of (275/195)² = 2.0 (power gain is proportional to the square of the voltage ratio). However, because the antenna length was increased slightly beyond 5/8 wavelength (47 feet) to improve the impedance match to the coax line, the field strength is reduced to approximately 263 millivolts per meter. This value gives a gain over a quarter-wave vertical of 1.8 on the horizon and at a vertical angle of 10 degrees, 1.6. In terms of db, the two gains are respectively 2.56 db and 2 db.

The 5/8 wave radiator has another advantage of producing maximum radiation at a height of 3/8 wavelength above the ground which reduces loss due to absorption of energy by house, garage, trees, etc. Maximum radiation from an antenna takes place at the current maximums, and in this case we have a current maximum located at a considerable height; i.e., one-quarter wavelength from the top. On 20 meters, this amounts to a height of 30 feet.

If you have the opportunity of comparing the vertical 5/8 wave antenna with another antenna, do so with a station which is located a considerable distance. I recall an instance several years ago when a friend informed me that his ground-plane on 40 meters was not as good as his little low horizontal job. He had made checks with a station located about 60 or 70 miles away. Further checks were made from my location under similar conditions and what he had said proved to be true. After studying the matter it was seen that the ground wave in either case has long since been attenuated so ground-wave communications between the two stations was not possible. Communications therefore were via sky wave, which in the case of the ground-plane was not possible because of the low radiation angles causing the signal to skip right over the receiving station. The energy from the horizontal, however, was concentrated at very high angles (see Fig. 7C) so that some of the energy was hitting the ionosphere at the proper angle and returning to the receiving station. Comparing the same two antennas at 2000 miles, the ground-plane was found to be far superior to the horizontal.

I have found that a thirty-foot telescoping pole, normally used to support TV antennas, makes an excellent bottom portion of the 5/8 wave radiator for twenty meters. This pole is available for approximately fifteen dollars. Several lengths of thin wall steel conduit that telescope inside the smallest section of the 30 footer can be added to bring the total length to 47 feet. The complete assembly is light (40 pounds) and can be put up by two people, first figuring what the length of the guy wires should be and installing two of them while the antenna is on the ground. While one fellow holds the base down the other fellow can walk it up into position. Once in position the base is secured so that the base man is free to pull up on the third guy.

The coil is made with an inside diameter of 1/2 inch, #12 wire, wound 8 turns per inch. For the 20-meter band, the antenna should be 47 feet high and the coil should consist of 12 turns. A coil of 10 turns should be used with a 31-foot antenna for the 15-meter band. For the 10 meter band, use a coil of 8 turns with a 23-foot antenna.

The 47-foot vertical was used during the field day contest producing good reports from the east coast while running only 30 watts. A 579X report was received from Australia. The 5/8 wave radiator is nothing new but definitely something that has been overlooked by the hams. It adds that extra punch you've been looking for at a very reasonable cost.

Appreciation is extended to Microwave Engineering Company for granting permission to photograph and use their equipment to make the pattern measurements.

Posted April 23, 2020