Half wave dipole antennas are generally considered the reference point for all antennas in ham radio, especially on HF. When fed from the center, a dipole makes for an easy impedance match to 50 ohm coax. When fed off-center at an appropriate location (typically the 1/3 point) and fed with a 4:1 balun, the dipole becomes a solid multi-band antenna. Feeding a half wave antenna from the end, however, presents additional challenges because the impedance is in the thousands of ohms. In spite of this, end feeding antennas can be an incredibly convenient configuration because you only need one support (like a tree) and you can easily place your operating position at or very near to the feedpoint of the antenna. This has led to this antenna design to being very popular with portable operators and others who want an antenna that is easy to erect quickly.
While researching this type of antenna I found a couple of blogs (here and here) that have a lot of good information regarding end fed half wave antenna designs. These designs rely on the principles used by the PAR Endfedz which consist of an impedance transformer between the antenna and transmitter as well as a capacitor across the feedpoint. Based on this design I made an impedance transformer using a FT140-43 ferrite toroid (this size toroid is overkill for a QRP application) with 27 turns on the secondary and 3 turns on the primary (24AWG enamel wire). This is then wired such that the start of both the secondary and primary are connected to the coax connection shield. The other side of the primary is connected to the coax center pin. The remaining secondary connection is the attachment point for the antenna. A 150pF is then wired across the coaxial connection. I used a 1000V mica capacitor since very high voltages are present at the feedpoint.
The matchbox was constructed using a 3.25″ x 2.125″ x 1.5″ ABS plastic box and 8-32 stainless steel hardware. To provide strain relief I epoxied a piece of 1/8″ acrylic to the back of the matchbox enclosure. I also made a strain loop at the end of the antenna wire for attachment to the acrylic sheet using an S hook. This allows the acrylic to carry the load of the antenna, not the antenna connection point. I also used pieces of acrylic for the end insulators since it is the perfect material to weave small wire through and lock it in place.
I wanted to experiment with the effectiveness of this matchbox with different antenna designs. I also wanted to test the antennas in a typical field installation configuration; in this case they were erected as a sloper with one end in a tree about 25 feet in the air and the feedpoint about 5 feet off the ground.
40/20 Meter Half-Wave
This antenna is a full size 40 meter half-wave with a tuning stub in the center to adjust the resonance of the antenna as a 20 meter full-wave. The tuning of this antenna was very straightforward; I simply tuned the main element for the center of the 40 meter band and then adjusted the 20 meter stub for the center of the 20 meter band. With the 26AWG stranded copperweld wire that I used the antenna ended up being about 62 feet long with a 2 foot long stub in the center. This antenna exhibits great bandwidth and easily covered both bands with under 2:1 SWR.
40/30 Meter Loaded Half-Wave
This antenna is a full size 30 meter half-wave with a loading coil/choke and tuning stub at the end of the antenna to provide resonance on 40 meters as well. The loading coil/choke consists of 55 turns of 24AWG enamel wire on a piece of 3/4″ PVC pipe. This coil is approximately 47uH of inductance, which should have an impedance of almost 3000 Ohms at 10MHz. The purpose of the coil is to choke off the current flow and electrically shorten the antenna on the 30 meter band while providing the necessary inductance to resonate the full antenna on the 40 meter band since it is shorter than a full half-wave on that band.
Tuning this antenna required a fair amount of trial and error because the 30 meter element and tuning stub length interact and affect the resonance on both bands. I initially trimmed the main element without the loading coil and had a good match with 42.5 feet of wire. After attaching the loading coil and several feet of tuning stub I found that the antenna appeared to be too short for 30 meter resonance and too long for 40 meter resonance. Eventually after several trimmings I found that a stub length of about 3 feet resulted in the 30 and 40 meter resonances tracking each other when I adjusted the length of the main element. I then added wire to the main element until I achieved a good match on both bands, in this case a main element of 48 feet works well. 30 meters is a narrow band and this antenna easily covers the entire band with under 2:1 SWR. Because of the loading coil, this antenna does not exhibit particularly high bandwidth on 40 meters, however, the purpose of this antenna is for QRP digital operation which does not involve a lot of tuning around, so it was trimmed to provide the best match at the low end of 40 meters and should have plenty of bandwidth for PSK and JT65 operation.
After my successful experiments with the QRP matchbox I wanted to build a more robust version for higher power applications. This requires the use of thicker gauge wire and a larger toroid to handle the higher currents and and more powerful magnetic fields. In this case I used 18AWG enamel wire wound on a FT240-43 ferrite toroid, which should easily handle 100 watts of power. A 27:3 turns ratio was used again as well as the same 150pF 1000V mica capacitor. I mounted the completed toroid in a 4″ x 4″ x 2″ NEMA 4X box and mounted it to a DX Engineering Balun Bracket. For the antenna connection I used 10-32 stainless steel hardware.
80/40 Meter Loaded Half-Wave
This antenna is constructed similarly to the 40/30 meter version described above. This time, however, the loading coil consists of 67 turns of 20AWG enamel wire on a piece of 1″ PVC pipe. This coil has an inductance of about 66uH which is required to achieve an appropriate amount of current choking at 7MHz. For strain relief I used 1/4″ Lexan sheet to make the connection points for the antenna and coil wires. I then epoxied the Lexan to the PVC coil form and bolted the connections using 10-32 stainless steel hardware. The coil was sealed using two coats of polyurethane.
I found this antenna to be easier to tune than the 40/30 meter version. This is most likely due to the larger difference in frequency ratio between the 80 and 40 meter bands vs the 40 and 30 meter bands. After trimming I found that a main element length of about 67 feet gave a good match across the 40 meter band. As anticipated this antenna has a limited 2:1 SWR bandwidth on the 80 meter band (about 90KHz). I decided to construct a way around this by adding a tuning stub to the end of the 80 meter section to allow for adjustment of which portion of 80 meters I wanted to operate in. Since the primary usage of this antenna would be for field deployment and emergency communications I would most likely need to be able to use it in the digital portion of the band (3.583MHz or so) as well as the higher end of the band (3.983-3.99MHz) where the ACS nets in my area take place. After some experimentation I found that the antenna was resonant in the voice section I wanted with an 80 meter stub 9 feet in length. I then made an additional 3 foot length of wire that I can add to the end using Anderson Powerpoles that shifts the resonance of the antenna to the digital portion of the band. This allows me to easily change the section of the 80 meter band I want to use by simply adding or removing this small section of wire. This additional wire has a very minimal effect on the 40 meter resonance of the antenna (around 10KHz) and does not prevent the antenna from achieving an SWR of under 2:1 across the entire band whether it is installed or not.
To assess the end fed’s performance I did a side by side comparison with my 80 meter loop skywire. I setup the end fed as a sloper with the loaded end supported by a tree about 25 feet in the air and the feedpoint about 3 feet off the ground. I then observed the signal strength when listening to the local ACS net on 80 meters. I found that I could copy everyone easily with the end fed half-wave, however, they were generally 1 or 2 S-units weaker than with my loop. I also used the End Fed Half-Wave during Winter Field Day and was able to easily make contacts on both 40 and 80 meters using both SSB and PSK31. With the winder this antenna weighs 4lbs, not bad considering the weight and bulk added by the matchbox. Overall I think this antenna is a very solid semi-compromise antenna for field use and will definitely be part of my Go Kit going forward.