The W5QJR Antenna
….a revolutionary concept!
SADLY – The Original Article had a load of links to Images that were not backed up and are therefor now lost when the site went down, if anyone has a copy of the CD or just the missing images I would be very grateful – MD0MDI
The Concept
In the modern world, many hams are able to erect a tower to support a beam on the higher bands, but don’t have the room for a good conventional low band antenna. Consequently, a few random wires are strung and loaded with a tuner and the resulting radiation pattern of that antenna is what it is.
Any antenna design should begin (not end) with the desired radiation pattern. Figure 1 presents the curve of an ideal antenna pattern shape for rag chewing on the low bands. This is a curve of normalized gain of an antenna that produces equal strength to all stations out to 1,000 miles. It is based on the standard space loss equation, which provides signal loss versus range. This is approximately only, because it does not consider D layer absorption as a function of angle, but is adequate for this article.
Missing – Figure 1
The search for an antenna that fits the desired radiation pattern resulted in one that produced the patterns displayed in Figures 2 thru 9.
Missing – Figures 2 thru 9
Figures 10 and 11 are presented as standards of comparison. All elevation patterns include two plots, which are 90 azimuth degrees relative to each other. The azimuth patterns are included to show non-directional characteristics of the antenna. All patterns are run over average ground on the MININEC3 program. The W5QJR Antenna has less high angle gain than a dipole by roughly a.5 S-units, but actually has more gain than a 1/4 wavelength vertical at low angles over average ground.
Missing – Figures 10 and 11
The antenna is sketched in Figure 12. Since it is difficult to describe the antenna in a few words, it needed a name. I hope some of you will suggest a more appropriate one. The configuration and performance are completely different than a sloper or any other standard antenna.
Missing – Figure 12
Design
The design process began with a desire for a good 160 meter antenna in the confines of the back yard. I had an existing 60-foot tower (1/8 wavelength at 2 MHz) that is grounded at the base for lightning protection. The concept of the sloping wire to make the tower resonate was an inspiration!
If the length of the wire is varied, the antenna can be made to resonate at any desired frequency. By setting the angle of the wire relative to the tower and selecting the location of the feedpoint, an excellent match to 50 ohms can be achieved.
Tests with the computer indicate that the antenna is very tolerant. For example, with un-insulated guy wires attached to the tower and grounded, the antenna parameters changed very little. The beam above the tower had little effect. This led to the conclusion that multiple antennas for various bands could be attached to the same tower. However, the antennas interact to change both the feed impedance and the pattern. The interaction is demonstrated in Figures 13, 14 and 15. This interaction occurred even though the 40 meter wire pointed east, 80 pointed south and the 160 meter wire pointed west. Although the 160 meter antenna is almost unaffected, there is a lot of interaction between the 40 meter antenna and the others, as would be expected. This is due to undesired current flow when all antennas use a common conductor, in this case the tower.
Missing – Figures 13, 14 and 15
The computer simulation indicated a good set of radials will significantly improve performance, but a good lightning ground is enough to give some performance. (This has been proven by experiment—the radials are really worth the effort, even if there are several only a few feet long at the base of the tower and a few longer ones.)
Why It Works
What makes this antenna work? The total length of the tower and wire is approximately 0.32 wavelengths for resonance. It is longer than 1/4 wavelength to achieve resonance, since the wire folds back on itself which reduces the inductance of the wire. The total length of the antenna varies from 0.32 to 0.38 wavelengths depending on the angle of the wire and other conductors in the area.
The currents in the tower section are essentially constant throughout its height, and the current in the wire is maximum near the tower. The wire contributes the horizontal component of the radiation pattern, which fills in the null that would occur if there were only vertical tower currents. The relative amounts of vertical and horizontal components can be adjusted depending on the height at which the wire is attached to the tower and the slope of the wire.
The First Try
After the computer told me all the above, I constructed an 80 meter antenna. Believe it or not—the darn thing didn’t work! All of the measurements indicated I was barking up the wrong tower. That really seemed a shame because the computer said it was an outstanding antenna. It was determined that the problem was due to feeding the antenna at a location other than the base, and not using a perfect balun. Even a minor unbalance in the balun allowed the coax feedline to interact with the antenna. If the tower were insulated and fed at the bottom, multiple wires attached to the tower would create a good multi-based antenna with a feed point impedance of about 10 ohms. It should be possible to run an insulated wire beside the tower and wires out from that, allowing a base feed. (If you try this, report the results so we can publish them.) This new design offered a lot of promise and I was not willing to settle for anything less than a 50-ohm feed point impedance.
The Fix
To make a long story short, a new feed was devised that used the coax as part of the radiation element. This allowed the coax to run to the feed point of the antenna without introducing spurious effects due to the feedline. Figure 16 shows the coax braid connected to ground and the remaining coax insulated from the tower. From the ground, the coax is 1/8 wavelength up the tower and a little more than 1/16 wavelength out. Apiece of wire slightly longer than 1/16 wavelength connected to the center conductor of the coax completes the antenna. If you compare the dimensions in Figure 16 with those in Figure 12, the difference is due to the large effective diameter of the tower. A large diameter conductor requires more length to achieve resonance. If the coax is near the tower, mutual coupling causes the large effective diameter of the tower to be a part of the antenna, requiring additional wire length for resonance.
Missing – Figures 16 and 12
With the New Feed
With the New Feed arrangement, the antenna really did what the computer said it should, and it has a 50-ohm feed point impedance at resonance. It really works great! On the air reports verified the pattern shape of the antenna, with excellent reports at various distances. I got good reports with a ground rod only, but after radials were installed and the antenna re-tuned and matched, the antenna gave excellent performance. It was also noted that the 3-dB bandwidth of the antenna was about 200 kHz on 75 meters with the ground rod and 130 kHz with radials. The computer predicted a bandwidth of 120 kHz. The difference in bandwidth can be directly attributed to ground loss, since bandwidth = KL/R. For a fixed amount of inductance at a given frequency, the only factor that varies the bandwidth is resistance. Since the feed point is 50 ohms, the change in radials made a change in loss resistance. Therefore, the bandwidth is directly related to efficiency.
Feed is Important
This new feed arrangement was important for other reasons. The antenna currents for multiple antennas on the same tower are somewhat de-coupled with this arrangement, which reduces the interaction. Another advantage is that there are no bad connections in the antenna such as bolted tower sections. The feed gives an excellent 50-ohm match. Figure 18 represents the feed point impedance as a function of the feed location. Note that the feed point can be located over a wide range and still achieve a low VSWR. A relay box at the base of the tower allows several antennas to be fed from a single feedline.
Missing – Figure 18
Easy to Support
The selected feed arrangement opened up other possibilities: You don’t even need a tower—just hang the coax from any support that is approximately 1/8 wavelength high at the desired frequency. This could be the peak of a house or a wooden pole, or a convenient tree. To determine this height, divide 123 by the frequency in MHz. For example, for 4 MHz, the height should be close to 31 feet. This is not a critical value.
Construction
The following is presented to assist in the construction, tuning and matching of this antenna. Let me suggest you build a prototype first. Then when it is what you want, build one that will be a permanent installation.
Buy a piece of RG-58 coax (because it is cheaper than big coax) and cut it to an electrical 1/2 wavelength at the desired operating frequency. This is readily done with an Antenna Noise Bridge or comparable piece of test equipment. The 1/2 wavelength coax will minimize confusion when you make impedance measurements of the antenna. These pre-cut pieces of coax should be a permanent part of the antenna test kit for any antenna eXperimenter.
Lay out your ground radials. If you build the antenna and add radials later, the resonant frequency and impedance will change frequency and impedance will change causing you to have to do the tuning and matching all over again. Next, hang a large diameter pulley at a height of 1/6 wavelength above ground.
At the end of the coax that will become the antenna feed point, connect the coax braid to an insulator. Connect a piece of wire 1/16 wavelength long (plus some extra for tuning) to the other hole of the insulator, then connect the center conductor of the coax to the wire. The insulator acts as a strain relief for the center of the coax. Add another insulator to the end of the wire, and an extension wire or rope for the antenna support.
Mark the coax with visible tape at a distance of 1/16 wavelength from the antenna end of the coax. Run the coax through the pulley to the tape marker. Remove insulation and connect a ground wire to the coax braid at the ground of the antenna. Stretch the top of the antenna out and tie it to the desired anchor. The preferred angle of the wire is 45 degrees, but is not critical.
Now you are ready to begin the tuning and matching process. Measure the antenna resonant frequency. Change the wire length to make the antenna resonant at the desired frequency. Next, measure the impedance at resonance. If it is not 50 ohms, move the coax feed point closer to the tower to lower the impedance, or further to increase the impedance. Each time you change the coax length, change the wire length an equal value to maintain resonance. To change the coax length, slack off on the antenna tie and change the ground connection on the coax braid. Now you know why I suggested a pulley at the top. The pulley also provides the necessary bend radius for the coax. Coax can be ruined by a sharp bend over a period of time.
After a little eXperimentation, you will have the desired feed point impedance at the desired frequency. Now you can hook up the rig and get on the air. If you are happy, take the antenna down and make all connections permanent. For use with transmitter powers much over 100 watts, replace the RG-58 coax with a larger coax such as RG-8xx.
A Critical Point
Experimental results with this antenna indicate only one problem. The antenna has a narrow bandwidth (about the same as a dipole). Spread wires from the feed point out should effectively increase the diameter of the conductor. The coax could also be encased in a wire cage to increase the effective diameter. A large diameter conductor reduces the Q by reducing the inductance, which increases the bandwidth.
Summary
This article was written to present a new antenna design of mine in 1989, to include information related to the design process, and to provide information to allow construction of this type of antenna for your use. Operational results have verified the computer simulation and indicate that another station with a properly erected and tuned dipole (very rare antennas) will beat you at close range, but only a very small amount. For longer range, the competing station must switch to a vertical. You will beat him (her) there, and you did not have to switch antennas! If you are a net control station, you have the optimum antenna to communicate with all net members. If you like DX, this is a very good antenna. The low radiation angle is believed to be due to a distribution of ground currents, as compared to concentrated ground currents from a conventional vertical antenna.
I have a lot of various types of antenna on 40, 80 and 160 meters, but none of them have offered the solid performance this one does. The only improvement desired for this antenna is more bandwidth. (but, see update below about this)
The antenna is very simple, very small and inexpensive because it does not take a significant support structure. This leads to the thought of several W5QJR Antennas in a directive array. Some day I will move to my farm in Georgia where there is plenty of room and I will report on such a system. I hope some of you will beat me and report your results.
Fascinating Phased!
As a teaser, let me inform those of you that have an interest in phased arrays, some preliminary results are fascinating. If you have two of the antennas on the same frequency on the same tower, but the wires are on opposite sides of the tower, they act like two antennas 1/4 wavelength apart. The computer produced the pattern in Figure 17 by feeding them 90 degrees apart. Be forewarned that the impedances are a little on the wild side due to mutual coupling.
Missing – Figure 17
My friend Bert Bittner, K0WQN, deserves credit for suggesting the coax feed configuration and the phased array possibilities. Without his contributions, this antenna would not have become a practical reality.
Although this antenna was designed for use on the lower ham bands, the design is applicable to any frequency. A small, high-performance, inexpensive antenna—a single pole with a directional pattern—who was it that said “all worthwhile developments had already been accomplished?”
Part 2 represents an update on the W5QJR Antenna
Construction
The following is presented to assist in the construction, tuning and matching of this antenna. Let me suggest you build a prototype first. Then when it is what you want, build one that will be a permanent installation.
Buy a piece of RG-58 coax (because it is cheaper than big coax) and cut it to an electrical 1/2 wavelength at the desired operating frequency. This is readily done with an Antenna Noise Bridge or comparable piece of test equipment. The 1/2 wavelength coax will minimize confusion when you make impedance measurements of the antenna. These pre-cut pieces of coax should be a permanent part of the antenna test kit for any antenna eXperimenter.
Lay out your ground radials. If you build the antenna and add radials later, the resonant frequency and impedance will change frequency and impedance will change causing you to have to do the tuning and matching all over again. Next, hang a large diameter pulley at a height of 1/6 wavelength above ground.
At the end of the coax that will become the antenna feed point, connect the coax braid to an insulator. Connect a piece of wire 1/16 wavelength long (plus some extra for tuning) to the other hole of the insulator, then connect the center conductor of the coax to the wire. The insulator acts as a strain relief for the center of the coax. Add another insulator to the end of the wire, and an extension wire or rope for the antenna support.
Mark the coax with visible tape at a distance of 1/16 wavelength from the antenna end of the coax. Run the coax through the pulley to the tape marker. Remove insulation and connect a ground wire to the coax braid at the ground of the antenna. Stretch the top of the antenna out and tie it to the desired anchor. The preferred angle of the wire is 45 degrees, but is not critical.
Now you are ready to begin the tuning and matching process. Measure the antenna resonant frequency. Change the wire length to make the antenna resonant at the desired frequency. Next, measure the impedance at resonance. If it is not 50 ohms, move the coax feed point closer to the tower to lower the impedance, or further to increase the impedance. Each time you change the coax length, change the wire length an equal value to maintain resonance. To change the coax length, slack off on the antenna tie and change the ground connection on the coax braid. Now you know why I suggested a pulley at the top. The pulley also provides the necessary bend radius for the coax. Coax can be ruined by a sharp bend over a period of time.
After a little eXperimentation, you will have the desired feed point impedance at the desired frequency. Now you can hook up the rig and get on the air. If you are happy, take the antenna down and make all connections permanent. For use with transmitter powers much over 100 watts, replace the RG-58 coax with a larger coax such as RG-8xx.
A Critical Point
Experimental results with this antenna indicate only one problem. The antenna has a narrow bandwidth (about the same as a dipole). Spread wires from the feed point out should effectively increase the diameter of the conductor. The coax could also be encased in a wire cage to increase the effective diameter. A large diameter conductor reduces the Q by reducing the inductance, which increases the bandwidth.
Summary
This article was written to present a new antenna design of mine in 1989, to include information related to the design process, and to provide information to allow construction of this type of antenna for your use. Operational results have verified the computer simulation and indicate that another station with a properly erected and tuned dipole (very rare antennas) will beat you at close range, but only a very small amount. For longer range, the competing station must switch to a vertical. You will beat him (her) there, and you did not have to switch antennas! If you are a net control station, you have the optimum antenna to communicate with all net members. If you like DX, this is a very good antenna. The low radiation angle is believed to be due to a distribution of ground currents, as compared to concentrated ground currents from a conventional vertical antenna.
I have a lot of various types of antenna on 40, 80 and 160 meters, but none of them have offered the solid performance this one does. The only improvement desired for this antenna is more bandwidth. (but, see update below about this)
The antenna is very simple, very small and inexpensive because it does not take a significant support structure. This leads to the thought of several W5QJR Antennas in a directive array. Some day I will move to my farm in Georgia where there is plenty of room and I will report on such a system. I hope some of you will beat me and report your results.
Fascinating Phased!
As a teaser, let me inform those of you that have an interest in phased arrays, some preliminary results are fascinating. If you have two of the antennas on the same frequency on the same tower, but the wires are on opposite sides of the tower, they act like two antennas 1/4 wavelength apart. The computer produced the pattern in Figure 17 by feeding them 90 degrees apart. Be forewarned that the impedances are a little on the wild side due to mutual coupling.
My friend Bert Bittner, K0WQN, deserves credit for suggesting the coax feed configuration and the phased array possibilities. Without his contributions, this antenna would not have become a practical reality.
Although this antenna was designed for use on the lower ham bands, the design is applicable to any frequency. A small, high-performance, inexpensive antenna—a single pole with a directional pattern—who was it that said “all worthwhile developments had already been accomplished?”
Part 3 an update on this most interesting antenna!
In an effort to achieve more instantaneous bandwidth from the antenna, fan wires were added as shown in Figure A. This has the effect of adding capacity and reducing the inductance, thereby increasing the bandwidth.
The basic antenna for 75 meters is in Figure B. That antenna has a calculated 3-dB bandwidth of 110 kHz. When the fanned wires were added, the matching needed to be changed to retain a 50-ohm match. The difference in the coax length is obvious from a comparison of the two figures.
The measured values for 3-dB bandwidth and VSWR are plotted in Figure C. The RF source was a TS-440S and the internal VSWR meter was used. The 3-dB bandwidth was measured with a field strength meter.
Figures D and E show a couple more sketches of the tower configuration.
The most curious aspect of this antenna is a direct correlation between the 3-dB bandwidth and the 2.0:1 VSWR bandwidth—they are both the same. Theory says the 2:1 VSWR bandwidth should be much less than the 3-dB bandwidth. When I figure why the VSWR bandwidth is much more broader than predicted, more data will be provided. In the meantime, any comments from readers will be appreciated.
Well, there you have it up to this point. Hope you enjoyed this article about a very special concept. Mostly, we hope you will eXperiment with the idea and let us know how well it worked for you. If the information is followed closely, and you avoid some of the pitfalls, you should realize the exciting success that comes with this concept. If you still have questions, send those in too. We welcome any responses to the W5QJR Antenna! The concept is being patented, and any efforts to market this antenna for commercial sale is prohibited. Otherwise, enjoy!
Originally posted on the AntennaX Online Magazine by Ted Hart, W5QJR
Last Updated : 28th April 2024