End Fed Half Wave (EFHW) Antenna

 

 Ever since I saw a video of someone with this antenna operating portable on 20 Watts I have been compelled to build one. He was making contacts all over the place with it just slung over a couple of trees. This is a 1/2 wave resonant antenna for 40m, but should also be good at the harmonic frequencies too, making it somewhat multiband. There are lots of resources on line about these antennas and I recommend some research is done before attempting the build. The 49:1 unun in particular needs to be studied because it isn't straight forward. The 9:1 and 16:1 ununs and transformer types I have employed before will not suffice for this project. They will be too lossy and inefficient at this ratio with a resonant antenna.  Capacitors need to be employed to improve SWR and performance on the higher bands. (counter-acting the inductance at higher frequencies)

(Diagram of 49:1 unun)

I am only using low power so just one toroid core should suffice. It seems that the higher the power, the more cores you will need, and the transformers can get hot. I have seen people opt for the FT240-31 core, but after researching the differences between type 31 and type 43, I decided to go with the recommended type 43 core. The capacitor required is a 100pF high voltage type which I found to be expensive compared to the 220pF of similar spec. I acquired some 220pF 15kV ones. If I connect two in series the capacitance will be halved and the working voltage will be doubled resulting in a 110pF at 30kV rating (belt and braces). Also, if your wallet allows, the theory is that using two toroids makes the transformer more efficient, especially at lower frequencies. Your choice. Theory also dictates that a good ground is required, however, with the right length of coax, the coax screen can act as a counterpoise because the current flow will be minimal at resonance. Some experimentation is required here. It might be that a counterpoise wire will be required on the ground connection, anything from 1m to 10m long. A good ground plane system can also be used if available. My finished project can be seen below.

I am assuming it will be good up to 50 Watts, but if the SWR drifts at higher power, this could be a sign of over-heating.

The box will have to be at least 100mm square. Mine is about 115mm square. There is a terminal post on the left for the antenna radiator. A link was made to a terminal post on the right to accomodate a ground / counterpoise connection.The two capacitors can be seen connected at the SO-239 socket, with some hot glue as a make-shift insulator on the joint.  I have left two large holes at the bottom for cooling / ventilation. I need to cover these with some kind of gauze to keep nasties out. I initially intend to house the box indoors. Just needs the 20m radiator wire erecting and we should be good to go. I want to construct a pulley and weight arrangement at the far end of the wire to maintain tension and prevent breakage in high winds.

Results and Testing

I can just fit a 20m wire from near the top of the roof to the diagonal corner of the garden. There is about 3m of radiating wire in the loft making connection to the box and counterpoise. A post was erected in the corner with the pulley attached at the top.

A piece of nylon rope was threaded through the pulley and connected to the wire.

The weight, a plastic bottle with just enough water to hold the wire tight, was tied to the rope and tension was achieved.

The theory is that any extra tension on the rope from high winds, temperature expansion / contraction or perching birds will allow the weight to rise and fall accordingly, relieving the strain on the wire. I think the antenna would perform much better if higher from the ground. At the moment it stands at about 12 feet at its lowest, rising to about 25 feet. 

Results

After experimenting with various counterpoises, 2m, 3m and longer lengths, I settled on a 1m length of wire. I really didn't know where to check the SWR for adjusting, so I decided to check at 14MHz, which is a full wavelength. The reading was about 1.5 after snipping about 6 inches off the radiating wire. Further cutting of about 2 inches caused the SWR to rise a small amount, so I decided not to trim the wire any more. I also experimented with varying the capacitance value from 220pF to 110pF, 73pF and no capcaitor at all. Some interesting and weird results were obtained. Using 220pF flattened the SWR peaks, but raised the overall SWR over the bands. 73pF raised all the readings. So I settled with the design spec of 110pF. At one point I was also getting gradually rising and falling SWR values at the higher frequencies which I couldn't explain. I guess it was something to do with changing the capacitor values, or a dodgy solder connection. The analyser results pretty much reflect the results from the radio SWR readings. Obviously, looking at the graphs, the SWR will be lowest where the impedance is 50 ohms.

(Analyser results - 1.8MHz to 30MHz sweep)

Final SWR Readings

1.8MHz = 2.9

3.5MHz = >10 (useless)

  7MHz  = 1.9

10MHz  = 1.6

14MHz  = 2.1

18MHz  = >10 (useless)

21MHz  = 2.2

24MHz  = 3

28MHz  = 1.7

Conclusions

The final results I obtained are far from perfect with respect to SWR, but the antenna is very effective when used with the internal ATU. It was immediately apparent that the received signals sounded much better and clearer compared to my other antennas. It would also seem that the antenna works better at long distances rather than closer to home, judging by the signal reports. Contacts with USA for example, are much easier to get now, but I can still make contacts across the UK. I will have to do without the 3.5 and 18MHz bands for now until I come up with a solution. The problems must lie with the transformer build. I might try a different approach in the future. Generally very impressed with this antenna, once again proving that the simplest methods are usually the best. 

 

 

Update - September 2023

Just for the hell of it I decided to take apart the 49:1 unun and add another ferrite core to see if there would be any difference. Apart from a possible increase in power handling ability, its hard to see any real improvement with performance. No improvement with the higher SWR readings on the lower frequencies. However, I can now tune the 18Mhz band. I found a small improvement by disconnecting the counterpoise. I also noticed that after moving things around in the loft, there was a change in readings. This reinforces the fact that you shouldn't have any objects close to your antenna. My situation isn't ideal because part of my antenna is in the loft! On the positive side, all the frequency bands above 3.5MHz fall below the 3:1 threshold. This is still my best performing antenna. So my advice would be that if you are only running up to about 50 Watts, then I think you would be OK with just one ferrite. Above this, you probably need two. Below is the full band sweep from the antenna analyser.

Dummy Load 50 Ohm 50 Watts

I thought it would be useful to have a dummy load to take power measurements. Rather than buy one I decided to make my own (as usual!) I purchased 10 x 470 Ohm 5 Watt Metal Oxide Film resistors. They are low inductance which is very important at RF. My power is limited but I decided to make it rated for 50 Watts. 10 x 470 Ohm in parallel gives 47 Ohms nominally. Parallel connection means that the power rating becomes 10 x 5 = 50 Watts nominally. (I read one article online which stated 4 x 100 Watt resistors in series gave him 400 Watts. This is incorrect! A common mistake.) The main problem I find with most projects is the hardware, in particular, cases and enclosures. Mooching around, I found this sewing tin with a hinged lid. Using a metal enclosure ensures good screening and prevents any transmissions radiating from the unit. A quick inspection found that the SO-239 socket would just fit on the side, provided I cut the lid away to allow it to close. The metal is very thin, so care is needed when drilling the holes. Sharp drill bits and proper support whilst drilling is the key. When assembled I measured the DC resistance with a multimeter at 47 Ohms. I'm not sure how accurate my multimeter is.

                                        

I assembled the project in about 2 hours. The tin made for a neat compact unit. everything just fit inside really snug. Total cost was about £6.00. A lot cheaper than the commercial ones and it is rated at 50 Watts.

The resistors were all soldered onto 2 pieces of solid copper wire and covered with heat shrink sleeving. A bit of hot glue secured everything in place. The hot glue you can see on top is covering the centre feed wire to the SO-239 to prevent any accidental short circuits. The ground wire was attached via a small crimp-lug and bolted to the holding screw. On reflection, I could have put some sleeving over the wires. I've tested it up to 20 Watts on all HF frequencies and the SWR reading does not change from 1:1. If operating at higher powers, the tin could be filled with dry sifted sand to dissipate heat, and then sealed up.

Update:

I did fill it with sand and seal it up because I noticed that it got quite warm when testing up to 25 Watts. Hopefully, problem solved.

14MHz Coil Shortened Dipole

 After a multitude of antenna builds using other people's designs, I thought it was time I designed my own from scratch. I decided a single frequency would be best to start with so I chose 14MHz because of the amount of traffic on this band. I think it's so popular because the antenna size is not too large to construct especially in a confined area. I looked at trap dipoles and the like and the art of shortening using coils. My final design would be a 6 metre long dipole so that it could be formed into an Inverted V to join my nest of dipoles in the loft. The coils were calculated for a 20mm former and came out at 6.68uH - 25 turns of 1.13mm diameter insulated solid copper wire, at 14.15MHz centre frequency. I decided to centre load the elements.

Testing

14MHz = 1.1:1

Bonus: 50MHz = 1.7:1(ATU) 

I had to add about 100mm to each end to get the SWR down. This means the calculations, or the interpretation of them, was in error. The measurements were made to the centre of the coils which is probably incorrect. When tested at 50Mhz, it was found that the SWR was a useable 1.7:1. I suppose the antenna has a tuneable harmonic.

Contacts made:

Croatia -6dB    Poland -15dB    Germany -11dB    Belarus -12dB    France -12dB

Hungary -17dB    Croatia(2) -13dB    Hungary(2) -11dB  (All on FT8)

Spot Reports for 14.074MHz FT8:

Conclusions

The SWR readings achieved were to be expected. This is a single frequency antenna with a comparatively narrow bandwidth. The Inverted V angle was very large, probably greater than 120 degrees. If this antenna was erected outdoors, the performance would be much better. Loft antennas tend to perform a lot worse. It seems to be working very well though and I will keep this antenna on-line for using on a regular basis.