Thursday, April 13, 2017

Antenna Alchemy - End Fed Half Wave and The Way of the Goat

Sota Mountain Goat is an award for accumulating 1000 activator points from mountain summits (see below). At least in Colorado, many of the winter activations are harrowing experiences in cold weather, deep snow, and high wind. Since it's not always practical to "hang around" at the summit the speed to set up and tear down can be critically important. No one wants to invest extreme energy and time climbing a mountain for an activation a descending with nothing to show for it. So, quick deployment and effectiveness are the watchwords for the kit.

When I first started SOTA activations, I looked around at what other, wiser, Goats were using in their kit and tried to copy them. I favored the lighter weight alternatives - no heavy radios, batteries or antennas for me. Aside from the transceiver, and mountaineering gear, the antenna stands out as the most important element of the system. I noticed that most of the Mountain Goats were using the End Fed Half Wave, or EFHW for short. While I had never previously used one, I thought I knew just about everything I needed to know about them. After all, isn't it just another flavor of vertical. I was wrong. What I discovered shocked and intrigued me, as I think it will you, too.

At first, I was a bit confused about the difference between a random length end-fed, and the EFHW. The terms seem to be used interchangeably. Some of the articles I read had charts that showed lengths to avoid when cutting so called "random length" wires. Turns out that if you randomly cut your random length wire to one of the "forbidden" lengths, it won't match, even with a tuner, and the SWR losses will be high. So, I initially avoided them. Only later did I learn that those forbidden lengths happen to be the the half wavelengths or HW part of the EFHW, and these these lengths are where some magic happens.

The Sota Goats don't want to dilly dally at the summit. They often need to set up, work their quota and get down. So, they don't have time for the one thing that we are all told we absolutely need in order to make a vertical work efficiently - radials. They need to extend a pole, string a wire, attach it to the radio and get on the air.  The only way to do that, and put out a decent signal too, is with the EFHW - an antenna with a forbidden length.

The problem with the EFHW, the reason that it is "forbidden", is that it has an outrageously high radiation resistance at resonance - on the order of 2500 ohms. This is not a number that is friendly to any radio and it is outside of the range of typical matching devices. However, it is the high radiation resistance that is the secret of its success. Tame it and it will serve you well.

Why is the high impedance that we've been told to avoid actually an advantage? To answer that, let's examine the characteristics of a cousin of the EFHW that we are all familiar with, the quarter wave vertical. The radiation resistance of the quarter wave vertical is about 36 ohms. In a typical installation (4 or so radials) the ground resistance can be substantial - let's say 42 ohms for the sake of illustration. So the antenna presents shows a decent match to 50 ohm coax of 36 plus 42 ohms = 78 ohms, a good match to 50 ohm coax. The efficiency of this antenna is calculated by dividing the radiation resistance by the radiation resistance plus the ground loss... Rr/(Rr + Rg) =  59%. About 40% of the power is lost in the ground system. The situation gets much worse, however, if only one or two radials are used. The ground loss will rise to perhaps 80 ohms or greater and the efficiency will drop to 25% or so. In other words, the 5 watt QRP radio lugged to the top of the mountain will radiate around 1 watt with one radial attached to a quarter wave vertical.

The situation doesn't improve with the so-called end fed random wire. Yes, the typical impedance of this antenna can be around 500 ohms at certain frequencies. However, while impedance (Z) is comprised of both resistance and reactance (R and X), only the radiation resistance is a factor in antenna efficiency (ignoring SWR losses for the moment), and unfortunately the high impedance of the end fed random wire is mostly comprised of reactance. The radiation resistance isn't typically any higher than that of a quarter wave vertical. Large numbers of radials are required to make this antenna efficient, just as it is with the quarter wave vertical.

Now let's look at the EFHW. At resonance, the radiation resistance is approximately 2500 ohms (or higher). So, even if no radials are employed with just a short 3 foot counterpoise and a stupidly high ground resistance of 100 ohms the efficiency will be 2500/2600, or 96%. Wow!

But there has to be a catch, right? Otherwise, everyone would be using these cheap and simple antennas instead of quarter wave verticals. Well, yes, but most of the limitations just aren't very relevant for QRP in the field.

Here are a few of the characteristics and limitations of the EFHW. There aren't any deal breakers on this list for SOTA work. The reason that this antenna isn't more popular in fixed stations has to do more with impedance matching issues than anything else - the popularity of coaxial cable for transmission lines and the lack of good high efficiency unun's made it impractical. Broadband unun's with high transformation ratios are still the stuff of experimenters - but the experiments have been promising. So, have at it. .
  • The EFHW can be built for multiple bands by installing traps or jumpers. I use an EFHW built for 40m with a jumper to shorten it for 30 meters. It works well on 40, 20, 10 with the jumper inserted, and on 30, 17/15 with the jumper open. SWR is 1:1 on the half wave frequencies, and matchable with the KX2 internal tuner on the multiples. 
  • SWR can be affected by touching the equipment
  • Ground losses increase off the resonant frequency - cut it precisely for your SOTA freq.
  • The high radiation resistance also applies to multiples of the half wave, ie. one cut for 40m will have also have high radiation resistance on 20 and 10, albeit with a higher reactive component in the impedance.
  • the impedance of the EFHW is too high for most matching devices (tuners)
  • the antenna requires a 50 to 1 unun impedance transformer to match 50 ohm sources. It may be necessary to add a capacitor across the source for a match on the higher bands.
  • A counterpoise with an ideal length of .05 wavelengths is recommended. I use a 10 ft length of rg-174 to the unun. 
  • High impedance transformations can be tricky and inefficient. Don't assume your unun is lossless. Care should be taken in the selection of the roroid to avoid losses >2 dB.
  • I built six unun's on different toroid cores with high transformation ratios and all had limited bandwidth. The best one had about 15MHz bandwidth measured with my SARK 110. Packtenna and MyAntenna devices outperform my homemade unun's. 
  • Be aware of power limitations and do not overdrive unun's to saturation. Ferrite unun's can be permanently damaged by excessive power. 
  • Commercial 50:1 unun's in various power ratings are only available from a few suppliers. I recommend the Pactenna 50:1 for SOTA and the MyAntenna for base station use. 
In summary, the EFHW is the ideal antenna for QRP (and some QRO) situations where a fast deployment is required and radials are not practical. It has been analyzed theoretically and field tested extensively. It is so effective that it has won over many Mountain Goats who started out thinking otherwise. When you absolutely, positively have to make at least four QSO's after hiking 5 miles to the top of a mountain, the EFHW is a good way to go.