No. 51: The Tri-Moxon Switched Vertical Array 
Site and budget restrictions often rule out a rotatable Yagi as out 10-meter antenna. However, the basic dipole is bi-directional, denying us the full horizon. A simple vertical often lacks some of the gain we want. Suppose that we could come up with a vertical array with gain and front-to-back ratio and cover the entire horizon with a switch instead of a rotator. If we set three individual small antennas vertically, we can cover the horizon. If we select the right antennas, we can use a single support post about 35' tall and set the antenna in the form of a Y, switching to the unit that covers the relevant 1/3 of the field. Also, selecting the right antenna will let us make a Y that is less than 10' in radius.
The right antenna is a vertically oriented wire Moxon rectangle. Using AWG #14 wire, we can string 3 of them in Y-formation, keeping each one 4' from the post. That distance will reduce interaction to acceptable levels. Each rectangle will cover 28-29 MHz with less than 1.7:1 SWR using 50-Ohm coax as the feedline. Fig. 1 shows the general outline of the system from the top and the side. We shall look at supporting the antenna before we finish.
The right-most part of the sketch shows the general outline of each of the 3 Moxon rectangles. We have examined this array in past columns. As well, the design of Moxon rectangles has been reduced to a set of true equations that will reliably yield the dimensions of the antenna for any reasonable element diameter (using wire or aluminum tubing) for any design frequency from 3 to over 300 MHz. See my web site (..) for more detailed information on the antenna and links to downloading small design programs.
SP is the 48" spacing of the reflector from the post, and the 3 antennas are at 120° angles to each other. The dimensions are the same for all 3 rectangles. They will work for AWG #12 wire as well as AWG #14.
Dimension Feet Inches A 12.63 151.5 B 1.90 22.8 C 0.35 4.2 D 2.36 28.3 E 4.61 55.3
Each vertical rectangle covers about 125° of the horizon. So by switching from one to the next, we can cover the full horizon with only minimal decreases in gain at the overlap points. See Fig. 2 for overlaid azimuth patterns.
The front-to-back ratio is about 13 dB, in the range of a 2-element Yagi. The gain at the 11° elevation angle of maximum gain is about 6.6 dBi. The system has been designed for a top height of about 34-35', which places the lowest point of each rectangle just above 22'. The height selected is not accidental. At this height, each rectangle produces a broad vertical pattern suitable for both short- and long-range paths, with peaks at 11° and at 34°. Fig. 3 shows a typical elevation pattern.
Any triple array of beams will interact. The advantage of using the tri-Moxon design is that the interaction is minimal due to the high front-to-back ratio of each rectangle. However, since they are not purely back-to-back, but at angles, we can only minimize the interaction. The Moxon allows a reduction in the needed spacing from each other relative to other antennas that we might similarly arrange--for example, three 2-element Yagis. The spacing shown allows an adequate front-to-back ratio and a smooth 50-Ohm SWR curve. Fig. 4 shows the SWR curve for a coaxial cable feed system. Wide spacing would further reduce interactions, but would take up more backyard space and complicate the support system.
Fig. 1 carries the suggestion of supporting the entire set of 2 wire Moxon rectangles from a central support post. Fig. 5 shows one way to achieve this goal.
First, all support elements must be non-conductive. A dead tree or a telephone pole that extends 35' above ground makes good central support posts. The sketch shows 1 of 3 rectangles supported by top and bottom horizontal "limbs." The limbs can be PVC or similar material capable of self-support for about 8-9 feet away from the central post. You can drill through the central post, and vertically displace each rectangle by a few inches so that the support "limbs" do not meet in the middle of the post. Set the top and bottom horizontal supports a little above and a little below the ends of the rectangle.
Use synthetic twine or thin rope to keep the rectangle under tension between the horizontal supports. This technique keeps the wires from resting on the PVC or other material, which might change a dimension along the rectangle ends. As well, loops allow room for small adjustments to the rectangle dimensions to arrive at the most perfect passband. Although the sketch shows vertical corner supports, placing them at an angle will hold the corners taut. With a piece of twine filling the gap, you can hold the rectangle in perfect shape, even if PVC or similar limbs sag a bit along their length.
Feed each Moxon rectangle with a 50-Ohm cable. You have two main choices for feeding the array. First, you can bring coax lines from each one all the way to the shack and switch them indoors. Second, you can obtain a remote coax switch and place it on the support post as shown in Fig. 5. Modeling shows that you get slightly better performance if the unused rectangles show an open circuit. So if the lines from the feedpoints to the remote switch are either 1/4 wavelength or 3/4 wavelength, the unused ones should be shorted at the switch end to yield an open circuit at the actual feedpoint. If the lines are 1/2 wavelength long to the switch box, the switch ends should be open to show the same condition at the feedpoint. However, the difference between the 2 conditions does not create enough difference to be called critical. If the lines go all the way to the shack, use the simplest switching feasible.
The Tri-Moxon vertical array is not an answer to all situations. But it is a workable system using light wire individual antennas set to cover the entire horizon with gain, a reasonable front-to-back ratio, a good range of elevation angles, and direct 50-Ohm feed. It requires only 1 central support and standard construction techniques for the rest of the support structure. As well, it saves the cost of a tower and rotator. Hence, the array is worth adding to our collection of 10-meter antenna ideas.
See the Moxon Rectangles and Online Calculator page for more information.
Updated 04-18-2006. © L. B. Cebik, W4RNL. Data may be used for personal purposes, but may not be reproduced for publication in print or any other medium without permission of the author.
Go to An-Ten-Ten-nas Page
Return to Amateur Radio Page