ANTENNAS FROM THE GROUND UP
One of the themes of this series has been how we can get the most antenna into the lest space. That theme makes sense in light of the shrinking yards of modern urban and suburban life. However, there are amateurs who have room to spare. There are a number of interesting wire antennas suited to such hams. Perhaps the most interesting a simply a piece of wire that is many wavelengths long at some operating frequency.
Consider the simple antenna in Fig. 1. It consists of a relatively short vertical wire that is just long enough to reach from the ground to the much longer horizontal wire. We feed the antenna at the ground. However, this antenna is not an inverted-L--although if the horizontal wire were considerably shorter, it might become one. (We are for the moment ignoring that third wire with the resistor in it.) Instead, the wire--considering only the fed vertical section and the horizontal section--is a classical longwire antenna. As we make a wire many wavelengths long, the number of lobes increases, with the strongest lobes approaching the line of the wire (in contrast to the broadside pattern of a 1/2 wavelength dipole). Because the wire has finite conductivity, the pattern is not truly bidirectional, using the wire ends as a reference. Instead, the pattern leans toward the open end of the wire. This unterminated longwire antenna is highly usable. especially if we have two target communications areas that are 180 degrees apart and one is somewhat more important than the other.
The 2-legged version of this antenna is, like almost (but not quite all) of the antennas that we have explored, a standing wave antenna. The open end of the wire presents an open circuit, so the electrical energy arriving at the end is virtually totally reflected. The combination of incident and reflected waves creates a standing wave on the antenna.
Let's now add the final leg of the antenna in Fig. 1. In that leg, we place a large resistor. If we let the resistor equal the feedpoint impedance, there will be no reflected wave and hence no standing wave on the wire. Traditionally, we call such antennas traveling wave antennas.
One consequence of placing the resistor in the third leg is to change the shape of the pattern. Instead of a slightly unbalanced bi-directional pattern, we arrive at a very directional pattern. Fig. 2 shows the difference for wires that are 5 wavelengths long and 1 wavelength above ground. We lose about 1 dB gain for the termination, but gain 15 dB of front-to-back ratio. In both cases, note the presence of significant sidelobes. Since a wire antenna that is multiple wavelengths long has many lobes, they remain part of the pattern. Terminating the wire can suppress them to some degree, but it cannot eliminate them.
There are techniques for calculating the termination resistor for antennas, especially when the height is low--the Beverage antenna. However, for upper HF use, we generally place a 600-800-Ohm resistor and allow the feedpoint impedance to make gradual swings as we change operating frequency.
We can make a rudimentary comparison between a standing-wave and a traveling-wave antenna by examining the current magnitude swings down the long horizontal wire. In the case of the 5 wavelength version, we obtain the current patterns in Fig. 3. The left end of the graph represents the part of the long wire that is closest to the feedpoint.
For a standing-wave antenna, the current makes very large magnitude swings that approximate (but do not equal) a sine wave. In the traveling-wave antenna, the current magnitude is relatively constant. It would be even flatter if the wire itself had no loss and if the terminating resistor exactly equaled the feedpoint impedance. We can see another difference when we examine the swings in the current phase angle, as shown in Fig. 4.
The current goes through one complete phase-angle cycle with each wavelength. At "junctions" between wavelengths, the phase angle will swing very quickly from negative to positive. However, between those junctions, the terminated antenna shows a virtually linear rate of change from the most positive to the most negative phase angle. In contrast, the unterminated version shows variations in the rate of change that are associated with the near-sine wave behavior of the current magnitude.
The bottom line is that a terminated longwire antenna becomes a highly effective directional array. The pattern does not have the clean forward look of a Yagi array. However, unless the various sidelobes present a problem in terms of potential interfering segnals, the antenna will perform very well.
The longwire also has a couple of other advantages worth noting. First, it will operate over a 4:1 frequency range. Fig. 5 shows the 600-Ohm SWR curve for the 5 wavelength version of the antenna. The low SWR values across the range suggest that a single wide-band transformer (conventional or transmission-line) might provide a good match between the target 600-Ohm feedpoint impedance and a 50-Ohm feedline. (Although transmission line transformers have become very wide-spread for matching impedances, we only lose a couple per cent of efficiency with a well-designed conventional transformer.)
The gain and horizontal beamwidth of the antenna will, of course, vary with the operating frequency. At 1/2 of the design frequency, the sample antenna is 1/2 the length and 1/2 the height. So we can expect a lower gain and a wider beamwidth. In contrast, at twice the frequency, the antenna is twice the length (10 wavelengths instead of 5) and twice the height (2 wavelengths rather than 1). The result is a higher gain, a narrower beamwidth, and a lower elevation angle of maximum radiation.
For maximum gain and the narrowest beamwidth, try for the longest wire that you can support. To get an idea of the differences that length can make to the longwire, see Fig. 6.
As may be clear from the patterns, the longwire just gets started toward high levels of directivity as it passes the 5 wavelength mark. If we wished to use a longwire antenna for the 40- through 10-meter bands, we might make it about 5 wavelengths at 40 meters or about 700'. Such a length will be 20 wavelengths on 10 meters. One of the secondary advantages of the longwire is that we do not need to be very precise in cutting the wire length. We shall not likely notice any operational differences between a 10 wavelength antenna that requires the erection of a new support post or a 9.5 wavelength longwire that ends near the limb of an existing tree.
At 10 wavelengths, the beamwidth of the array is about 30 degrees. Assuming that we have supports and a good supply of wire to go along with our acrage, we can place a longwire antenna in the direction of each of our target communications areas. Then, a simple switch will allow us to change antennas to the one pointed at a particular target. We need no rotator. As well, we do not have to make the wire and support investment all at once. We can add new target areas as the supports and wire become available.
There are two disadvantages to the terminated long wire antenna. First is the terminating resistor. If we only desire to receive, we can make up the resistor from low-wattage carbon resistors, strung in series-parallel combinations to approximate the target value. However, transmitting changes the problem. We need a non-inductive resistor capable of dissipating about 1/2 the power supplied to the antenna. Such resistors tend to be very expensive, although some occasionally appear on the surplus market.
A second challenge presented by the terminated longwire antenna is the matching for the feedpoint. If we only need a single band, we can use a standard L-network. Since the antenna feedpoint is inherently unbalanced, an automatic tuner would work for multi-band operation. However, this option would be expensive if we decide to put into place several of these antennas in order to have more communications targets. The most general solution is to build or buy a wide-range transformer capable of matching 600 to 50 Ohms (a 12:1 impedance transformation). For the design of transmission line transformers that would fit the need (either singly or in a combination of 2 of them), see the writings of Jerry Sevick, W2FMI.
The longwire antenna is attractive due to its seeming simplicity. However, the challenge of supports, terminating resistor, and impedance transformation make it an antenna that requires a good bit of pre-decision thought before committing the family farm to a system of them. In addition, the antenna has relatively low gain that depends upon the wire length as measured in wavelengths. Between lengths of 5 and 10 wavelengths, the forward gain over average ground varies between 7.5 and 9.5 dBi. Hence, the chief reason for using the terminated longwire is directivity. To obtain more gain and good directivity from a wire array, we shall have to examine some other designs.
Updated 11-07-2003. © 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 series index page
Return to Amateur Radio Page