Your First 160-Meter Antenna
The following notes rest on a small set of assumptions.
1. You want to get on 160 meters for the first time (or perhaps, for the first time in a long time).
2. You want to set up the simplest possible effective antenna using all wire construction. In fact, all of the antennas will be made from AWG #14 or AWG #12 wire. 2-mm (0.0787") diameter wire falls right between these sizes, so all of the data will use that value. However, nothing much changes by reducing the diameter to AWG #14 (0.0641") or increasing the diameter to AWG #12 (0.0808").
3. You do not have unlimited vertical space for your antenna. In these notes, the limit will be about 70'. In fact, I shall use 21 m (68.9') as the standard top height for all antennas.
I have set these limits so that we can compare the performance of a collection of relatively simple antennas.
For all comparisons, we shall use average ground with a conductivity of 0.005 S/m and a relative permittivity (dielectric constant) of 13. For vertical antennas especially, you should expect lesser performance from worse ground and better performance from better ground--but not radically worse or better. Horizontal antennas are less affected by ground quality, but the top height is so low (about 1/8 wavelength) that the ground will influence performance much more than for the antennas you place 1 wavelength above ground for the upper HF region.
160-meter antennas are naturally much larger (longer, taller) than antennas for the other HF amateur bands. Therefore, be prepared to spend a little more money for quality wire and insulators to durably bear the antenna weight. Copperweld is desirable. Supporting structures--whether natural or constructed--need to be stronger and taller than the average sorts of things that populate a backyard. How you handle the support structures I shall leave to you, since every yard is different, as are the locally available materials and the construction skills at hand.
With those qualifications, let's get started in our work, starting with some vertical antennas.
160-Meter Wire Verticals
We shall begin with an antenna that violates the upper height limit of our task: the full-size 1/4 wavelength vertical monopole. A wire version of this antenna needs to be about 39 m (128') tall. The convenience of the vertical monopole is that we can feed it at the base--at or near ground level. The inconvenience is that we must install radials. The radials should be about 1/4 wavelength long and placed as symmetrically as the yard space allows. To see how many radials we might need, I modeled the vertical using 4, 16, and 64 1/4 wavelength radials, each 6" (0.15 m) below the surface. Fig. 1 shows the outlines of the 3 models.
The following table shows the anticipated results, assuming that the vicinity of the antenna is not filled with RF-eating ground clutter. Conductive objects--even semi-conducting trees and shrubs--can distort antenna patterns and absorb some RF energy, so keeping the antenna area as clean as possible is important to getting the most out of any vertical antenna.
1/4-Wavelength Vertical Monopole with Variable Radial Systems Average Ground No. of Maximum TO Angle Feedpoint Z Radials Gain dBi degrees R +/- jX Ohms 4 -0.72 23 57 + j 1 16 0.48 23 44 - j 8 64 1.14 23 37 - j12
Note that we gain about 1.2 dB by increasing the radial field from 4 to 16 wires, with another increase of about 0.7 dB by raising the count to 64. Fig. 2 shows the relative radiation pattern strengths. The radiation plot also shows that a vertical antenna is best for lower-angle long-distance skip signals, but almost unusable for NVIS (Near Vertical Incidence Skywave) very short distance communications. Many vertical users also find a vertical less noisy that a horizontal antenna in terms of QRN from lightning, but more susceptible to local man-made noise sources. As well, as we increase the number of radials, the impedance decreases, indicating a reduction in energy lost to the ground.
As the impedance decreases due either to the number of radials or ground quality, a number of operators use a simple means of obtaining a good match for coaxial cable. By making the vertical longer, they increase the resistive component of the impedance and the reactance moves from being slightly capacitive to being more definitely inductive. Adding a series capacitor at the feedpoint between the cable center conductor and the feedpoint itself allows them to compensate for the reactance, leaving a nearly perfect match for the 50-Ohm cable. A fixed capacitor may work if you have a specific operating frequency, but a remotely tuned variable is necessary for obtaining a low SWR over a wider operating bandwidth. Since we want the antenna to be at least slightly inductively reactive at all operating frequencies, setting up the antenna for the low edge of the band is the usual practice.
The full-size vertical monopole is useful as a reference for comparing other vertically polarized antenna candidates. With that data, we can see what we gain or lose from each one. We shall look at 2 candidates, each no more than 70' tall.
The Tee-Vertical: If we must limit the height to a certain level--70' in our case--but still desire a perfectly circular pattern, we need to create a shorter vertical antenna. Many vertical users opt for inductively loading the vertical either at its base or higher up on the wire. However, inductive loading has two disadvantages. First, the inductor always has a series resistance that reduces the radiated energy. Second, inductive loading reduces the feedpoint impedance faster, the closer the inductor is to the feedpoint.
One of the simplest and most efficient ways to shorten a vertical monopole is to create a hat at the top. The usual vision of a hat consists of several hat wires radiating from the top of the vertical wire. However, we actually need only 2 wires to effect a hat. (The more wires that we have in a symmetrical arrangement, the shorter that each must be to set the antenna at resonance. However, any wires not in the same line as the supports for the top of the vertical section require additional supports.)
We shall look at 3 versions of a Tee-vertical: with 4, 16, and 64 radials. Fig. 3 shows the relative complexity of each version. The vertical wire is 21 m (68.9'), and each leg of the Tee is 11.6 m (38') long.
For the same set of conditions used to model the full-size vertical monopole, the shortened Tee-vertical shows the following performance values.
Shortened Tee-Vertical Monopole with Variable Radial Systems Average Ground No. of Maximum TO Angle Feedpoint Z Radials Gain dBi degrees R +/- jX Ohms 4 -1.45 25 42 + j 2 16 0.20 25 29 - j 6 64 1.11 25 23 - j11
The fewer the radials, the more that Tee-vertical performance lags behind the performance of the full-size vertical monopole. With 64 radials, there is almost no difference in performance with respect to gain. The Shorter vertical section of the Tee version does show a 2-degree increase in the TO angle. As well, the impedance at the feedpoint is only about 70% of the value for the full-size vertical. Fig. 4 shows the relative radiation patterns. We do not need azimuth patterns because, like the full size vertical, the Tee-vertical provides virtually a perfect circle of radiation (assuming that there are no nearby objects to distort that pattern).
The Tee-vertical is amenable to the use of lengthening techniques to raise the feedpoint impedance with a series capacitor to compensate for the inductive reactance. Lengthening the Tee legs (in equal amounts to preserve symmetry) saves you the trouble of increasing the height. However, you will need more horizontal space for the increased Tee-top. If you use a series capacitor at the base of the antenna, I recommend a double waterproofing case system, along with regular preventive maintenance. As well, be sure that you use a beefy capacitor able to handle the high current level at a 50-Ohm impedance.
The 1/4-Wavelength Inverted-L: A second alternative for our 70' height limitation is the inverted-L. As shown in Fig. 5, the L does not worry about symmetry, but simply uses a horizontal extension of the vertical wire to reach resonance on 160 meters. Because the top is not symmetrical, the horizontal wire radiates. However, the current is lower in the horizontal part of the antenna and the pattern is not seriously distorted on 160 meters. In the model for 1.85 MHz over average ground, the horizontal wire is 19 m (62.3') for the same vertical wire used in the Tee-vertical.
The performance of the inverted-L is not significantly different from the Tee, as shown by the following performance figures.
Shortened Tee-Vertical Monopole with Variable Radial Systems Average Ground No. of Maximum TO Angle Feedpoint Z Radials Gain dBi degrees R +/- jX Ohms 4 -1.53 26 43 + j 3 16 0.08 26 30 - j 6 64 0.98 26 24 - j11
Due to the small horizontal component of the radiation patterns, the elevation angle has increase by another degree. However, the impedance values are almost identical to the corresponding values for the Tee-vertical. Fig. 6 shows the elevation and the azimuth patterns for the inverted-L. Note that the presence of a non-symmetrical horizontal section does not allow the pattern overhead to go to nearly zero, although the level is not strong enough for effective NVIS communications. The azimuth pattern shows a slight push in the direction of the top section of the L. However, the differential is not large enough to be noticed during operation.
The 160-m 1/4 wavelength inverted-L has another advantage. With a wide-range tuner at the feedpoint (perhaps one of the remote tuners on today's market), the antenna is usable for general communications on virtually all of the amateur bands. Above 160-meters, the radial system acts like a good RF ground between the operating position and the antenna base, since the antenna is 1/2 wavelength or longer on all bands above 160 meters. If you choose to use a remote tuner for such an inverted-L system, add another layer of water-proofing as an additional guard against weather penetration of the tuner and the connection. For further information on multi-band use of the inverted-L, see "Straightening Out the Inverted-L."
There is one temptation to avoid with the 160-meter 1/4 wavelength inverted-L. Many operators obtain rather poor results because they place the vertical section of the antenna too close to a natural or man-made support. The vertical section needs as much clearance from other objects as the corresponding part of the full-size and the Tee verticals.
160-Meter Wire Horizontal Antennas
We have examined the main candidates for vertical wire antennas, although there are manmy variations on the basic designs that we have used as examples. We should also look at some horizontal basic wire antennas. Any horizontal antenna will be severely limited by the 70' height restriction that we placed on the exercise. 70' is only about 1/8 wavelength above ground, a height that is even below optimum for NVIS operation--although it will work quite well in this service. One advantage of the horizontal wire is that it does not require any radials. A second advantage--at least for our work--is that horizontal wires do not change performance characteristics very much as we change ground quality. Therefore, the use of average ground provides a good indication of operation over any soil type. Finally, there are only 2 important horizontal variations that are possible within our height restriction: linear wires and closed horizontal loops.
The 1/2-Wavelength Dipole: There is no magic about the 1/2 wavelength dipole except that at resonance, it is a reasonably good match for coaxial cable. If we wish to use parallel feedline and a tuner, we can be less critical about the exact length without changing the pattern in any detectable way. Fig. 7 shows the details of our model set-up. The wire is 78 m (256') long.
Since we have only a single model with which to deal, our performance table is simplified.
1/2-Wavelength Horizontal Dipole 70' above Average Ground
Maximum TO Angle Feedpoint Z
Gain dBi degrees R +/- jX Ohms
6.72 90 49 + j 0
Note that the horizontal wire provides the strongest radiation (and receiving sensitivity) straight up. Fig. 8 compares the elevation pattern of the dipole with the elevation pattern for the inverted-L with 16 radials. The horizontal wire is superior for NVIS service, but inferior for long-range, low-angle service. The horizontal wire is likely to be more susceptible to lightning noise, but less susceptible to man-made noises. The patterns for the two types of antennas cross at about the 23-degree elevation mark.
Despite the low height of the dipole when registered as a fraction of a wavelength, the azimuth pattern at almost any elevation angle is still bi-directional and broadside to the wire. Fig. 9 shows the azimuth pattern at a lower angle (25 degrees elevation). Radiation (and reception) off the ends of the wire is about 8-dB or about 1.5 S-units weaker than broadside to the wire.
Linear wires with open ends can build considerable levels of static charge unless we take measures to bleed it off as it develops. One technique is to place either a high-value resistor or an RF choke across the antenna feedpoint, ensuring that one side is connected to the coax braid--and the coax braid is well grounded. Inserting a transmission-line transformer type of balun at the feedpoint will defeat this measure by physically isolating the feedpoints from the cable braid. However, using a W2DU-type ferrite-bead choke as the balun will allow the bleed-off component to do its work.
The 2-Wavelength Horizontal Loop: A closed loop antenna is more immune to static charge build-up, but has some special requirements. To understand why the heading specifies a 2 wavelength circumference for the loop horizontal antenna, we should proceed a step at a time. Let's begin with a simple square loop, like the one shown in Fig. 10. Our initial exercise will place the loop in free space and vary the circumference from 1.0 to 2.5 wavelengths.
The following table lists the free-space performance values for the loop. The column marked "Horizontal Gain" lists the gain in the plane of the loop. The column labeled "Vertical Gain" shows the gain broadside to the face of the loop.
Free-Space Performance of Horizontal Loops of Various Sizes Circumference Horizontal Vertical Feedpoint Z WL Gain dBi Gain dBi R +/- jX Ohms 1.0 0.09 3.27 124 + j 17 1.5 1.49 2.97 5300 - j 4700 2.0 3.07 0.18 300 + j 240 2.5 2.06 1.09 2600 - j 2700
The 1 wavelength loop is most useful in parasitic beams called quads, where the individual loops are set up vertically to take advantage of the stronger radiation broadside to the plane of the loop. However, when we place the loop horizontally over ground, the radiation from the edge of the loop--the plane of most interest--is much weaker. As the table shows, the edge, in-plane, or "horizontal" radiation is strongest when the loop is about 2 wavelengths in circumference. For our test model, that length is about 340 m (1115'). Since the loop is not resonant, we shall need parallel transmission line and a tuner. Hence, the exact length is not at all critical. Any total circumference around 1100' will work fine.
Fig. 11 compares the elevation and azimuth patterns for 1 and 2 wavelength loops 70' above average ground. Note that due to the low height, even the 2 wavelength loop has a relatively high TO angle. However, the 2 wavelength radiation strength (and reception sensitivity) at lower angles is considerably greater than the 1 wavelength loop. The advantage at lower angles appears clearly in the azimuth patterns on the right. The "tilt of the pattern follows the placement of the feedpoint, shown in Fig. 10. Note that the 2 wavelength loop does not produce a circular--or even an oval--pattern. Rather, it has four wide major lobes. The following table completes the equivalent data for all of the loop sizes that we tested in free-space. Note that the impedance reports change relative to the free-space values--as a function of the low height of the antennas above ground. The resistive component is lower, while the reactive component is more inductive.
Performance of Horizontal Loops of Various Sizes 70' above Average Ground Circumference Maximum TO Angle Feedpoint Z WL Gain dBi degrees R +/- jX Ohms 1.0 7.38 90 100 + j 100 1.5 6.65 90 2600 - j 5200 2.0 5.65 50 200 + j 380 2.5 6.02 53 1400 - j 3300
The pattern shapes and TO angles for a horizontal loop change as we change the shape of the loop. They also change if we move the feedpoint, say, from a corner to the middle of a side. As samples of the sort of changes that we might encounter with relatively symmetrical simple structures, I modeled triangular, square, and hexagonal loops, feeding each structure both at a corner and in the middle of a side. The following table summarizes the results. It adds a column listing the maximum gain at a "standard" 30-degree elevation angle, since the TO angle is considerably higher in most cases and varies from case to case.
Performance of 2-Wavelength Horizontal 70' above Average Ground Loop and Maximum TO Angle Gain at 30-deg Feedpoint Z Feed positionL Gain dBi degrees dBi R +/- jX Ohms Triangle-Corner 6.05 54 3.56 135 + j 315 Triangle-Side 5.99 58 3.18 225 + j 300 Square-Corner 4.92 55 1.24 75 + j 220 Square-Side 5.65 50 3.51 200 + j 380 Hexagon-Corner 5.65 53 2.75 140 + j 320 Hexagob-Side 5.57 54 2.45 145 + j 320
The wires of a 2 wavelength loop interact with each other to produce distinctive patterns for each combination of overall shape and feedpoint placement. Fig. 12 shows the azimuth patterns for the two triangles, with plots taken at the TO angle at at a standard 30-degree elevation angle. The insets show the loop outline and the feedpoint placement relative to the pattern for each version of the triangle. In all of the plots of 2 wavelength horizontal loops, the feedpoint will be at the top or 0-degree azimuth direction.
The two triangle patterns are similar, although there is a small displacement of the pattern toward the long-wire side and away from the triangle point. More significant is the fact that in both cases, the pattern is significantly stronger (by about 3 dB) along a line from the feedpoint through the center than from side to side. Otherwise, there is not much to choose between the two versions of the triangle.
The patterns in Fig. 13 confirm what the data in the table suggest: the feedpoint position makes a much more important difference to performance with a square loop than with any other form. With a corner feed, we obtain nearly circular patterns, but at lower strength. With a side-feed, we obtain more gain, but the patterns take on the 4-lobe shape. The lower the elevation angle, the more distinct that the lobes become. Whether the pattern shape and gain provide an advantage may depend on the possibilities for laying out the antenna relative to desired communication targets.
As we make the loop more circular, the exact shape and feedpoint make less difference to performance. The hexagon patterns appear in Fig. 14. Neither the pattern shape nor the gain change very much as we re-orient the loop and the feedpoint. As well, the corner-fed and side-fed versions of the loop exhibit feedpoint impedance values that are much closer together than for either the triangle or the square.
The most desirable version of a 2 wavelength horizontal loop would be a circle. However, the realities of antenna construction will not only require simpler forms, but as well, they may dictate somewhat irregular shapes. Nonetheless, virtually any horizontal loop will provide very reasonable performance. In addition, unlike a dipole, they will provide a null overhead, much like the nulls of vertical antennas. Therefore, if NVIS operation is the goal, you much either create a 1 wavelength loop or a dipole. For operation in the 20-30-degree elevation range, the 2 wavelength loop will usually provide as much or more gain than a wire vertical. Fig. 15 compares the elevation patterns of the corner-fed hex loop and the inverted-L with 16 radials. The maximum gain limits of the loop are similar to those of the dipole at the same 70' height, but the pattern is nearly circular rather than being bi-directional.
Both the 160-meter dipole and the 2 wavelength loop are useful as multi-band antennas if we feed them with parallel transmission line and employ an antenna tuner to achieve a match with the transceiver. A number of other items at this site address the kinds of patterns that we can expect from a 250+' doublet and from horizontal loops (HOHPLs) of various shapes across the HF region.
Conclusion
We have surveyed some of the simplest antennas used on 160 meters. They are simple in principle, but require a lot of wire, whether used in the element or in radials. Insulation on the wire makes virtually no difference to performance. As noted early on, element wire should be strong, and copperweld is desirable. However, radials may use virtually any wire available. If a sale on wire allows you to add more radials to a vertical system, then it is worth the price. However, exposed elements require good strength or additional supports. As well, use good non-conductive insulators wherever an elevated wire terminates or changes direction. Do not lay a wire directly over a tree limb or wood support. High voltage has been known to gradually sever limbs or to set dry limbs ablaze. Suspend an insulator below the support and run the wire through the insulator. Likewise, use a strain relief fixture for any connection between the element and parallel transmission line.
We have not examined a number of excellent antenna systems, such as phased or parasitic verticals. 160-meter wire Yagis and LPDAs are also possible. These are advanced projects, and our mission was to set out and compare some basic antennas. However, eventually, you will wish to purchase a copy of ON4UN's book on Low-Band DXing. It is possibly the best collection of 160-meter (and 80- and 40-meter) antenna ideas available.
Updated 6-7-2005. © 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.
Return to Index