Saturday, January 7, 2017

New Antenna Expectations

During the several years I have written this blog I wrote quite a few articles comparing antenna performance versus height. Most often this was for 40 meters, which is a particular focus of mine. However I did not neglect the higher bands. Since vertical polarization is most often used on 80 and 160 the focus is less on height than it is on efficiency, especially reducing ground loss.

Now that the Trylon tower is up and decorated with yagis this is an excellent opportunity to look forward to what I can expect. First, let's look at the antennas I have put up:
  • Hy-gain Explorer 14: 3-element tri-band yagi on a short (4 meter) boom
  • Cushcraft XM240: 2-element loaded (short) yagi
This is a temporary arrangement for the winter and spring, doing the best I can with the one completed tower and the selection of antennas in my stockpile. The temporary antennas I am planning for 80 and 160 are not horizontally polarized and therefore are outside the scope of this article.

Explorer 14 Tri-bander

This was my primary high bands antenna for 20, 15 and 10 meters at my former QTH. Its height was a little over 14 meters on my light duty guyed tower. With the antenna now at the greater height of 23 meters its low angle performance should be enhanced. The question is: by how much? That is, does the additional 9 meters make a significant improvement?

A direct comparison isn't possible so I returned to a model of a similar antenna I developed back in 2014: a 3-element tri-band trap yagi. Although not the same as the real antenna I am using it is more than sufficiently similar to compare performance versus height.

The result is shown in the adjacent elevation plot from EZNEC for 15 meters. The primary trace is the antenna at a height of 14 meters over medium ground and the secondary one is with it at 23 meters. Not surprisingly the low angle performance is improved. Also not a surprise is the appearance of additional minor lobes, and nulls in between.

With respect to DX performance I am using a median elevation angle of 5°, which is where the cursor is located. The behaviour is similar on 20 and 10 meters. At that elevation angle the gain improvement at 23 meters height is in the range of 3.5 to 4.0 db. This is significant. You won't notice it when signals are strong but when the opening is marginal it will make a large difference. This is all the more important now that the solar cycle is dying, and is in any case important for contesting where the extra signal strength will add many QSOs to the log.

After a week of use through some difficult conditions, and without the possibility of a direct A-B comparison, I am confident that I am getting the predicted performance improvement. The best test may be the first contest I enter. That is likely to be NAQP CW later this month. A better test will be ARRL DX in February.

Cushcraft XM240

According to my previous modelling of many 40 meter antennas there ought to be about 9 to 10 db low angle gain (at 10° elevation) compared to the inverted vees I used at my Ottawa QTH. Here is a summary chart of several antennas from a previous article comparing height versus gain at 10° elevation.

My XM240 is roughly equivalent to the modelled 2-element short yagi. The specific model selected includes estimated gain reduction due to coil loss and element shortening. We also need to keep in mind that the gain bandwidth for a 2-element yagi is poor, which is why I bracket the range as 9 to 10 db. Modelling and careful estimation is the best I can do short of flying a field strength meter on a drone.

Early indications are that the antenna meets my expectations. There is little that the big guns can hear that I cannot, even if not at the same signal strength. That was far from true before. In particular I can hear Asians long path, work previously difficult paths such as UA9, and far more (and weak) European stations. However this does not mean I crack every pile up. That 10 db is with respect to my low inverted vee whereas many others have wire antennas better than that, plus 10 db or more courtesy of an amplifier.

Models and feedback from other users of the XM240 that it works on 17 meters with a modest SWR. This is useful since at least for the time being I have no 17 meter antenna. The measured SWR is a little under 2.5, which my rig happily loads without a tuner. The few stations I've called hear me fine. Of course a proper antenna would work far better but that's off in the future. It is not a priority since 17 is not a contest band.

One task I plan over the next week is to profile the F/B across 40 meters to get an idea of where the reflector is tuned. The SWR which is lowest at 7.125 MHz is not a indicator of where the gain and F/B curves fall. In a 2-element yagi it is the parasite and the parasite alone that determines the gain and F/B curves. Tuning the driven element only adjusts the match, provided its self resonance is in the near vicinity of that of the reflector.

There are reports on the internet that it can be dangerous to drive the tower plus XM240 (as a top hat) as a low band vertical. The explanation is that the Cushcraft matching network (balun) cannot withstand the high voltage between the balun input (capacitively coupled to the tower and boom) and the driven element when running high power. Since I do not intend to put radials on this tower in any case I will not risk the XM240 balun by shunt feeding the tower on 80 or 160.


The 6.5' (2 meter) spacing between the yagis is an absolute minimum to avoid deleterious interaction on 15 meters. Even better would be 3 meters. Unfortunately that would be unwise since the mast and tower wind survival would be dangerously poor (see below).

In initial testing there is no degradation to the 15 meter SWR/impedance. Modelling tells me that the most sensitive to interaction is F/B and other lobe distortion. Since it's winter and the solar flux is very low I have not yet had a chance to test whether there is a pattern problem. I plan to do that when I am able.

There is a potential for other interactions with wire antennas attached to the tower. My main concern is with 40 meters and 15 meters interaction with the XM240. I am less concerned with interactions with the Explorer 14 since it is higher up and therefore further from wire antennas. There will need to be at least one wire antenna for diversity in this winter's contests and to have something for 80 and 30 meters. I may do a model to predict interactions.


Within the first 72 hours of completion we were hit with snow, freezing rain and wind gusts of 80 kph. This brings the importance of survivability solidly to the fore. This is as applicable to a permanent installation, a short term one such as mine or for a Field Day weekend. All that changes is the probability of destructive weather events declines over shorter time periods.

Yet since anything can happy anytime it is a matter of how much risk you can tolerate versus your investment. In this respect I have made my choices for this initial phase of my new antenna farm.

There are several aspects to consider in the failure modes of this installation. Although it is temporary, perhaps only until May, I will not accept excess risk. However I deliberately increased the risk I will tolerate in comparison to a more permanent installation. I'll walk through those points and my thoughts. Not all would agree with my choices.

Mast: The mast is just the sort that professionals warn you against: water pipe. Specifically 19' (6 meters) of 1.5" (nominal) Schedule 40 galvanized water pipe. This is the same pipe that served as a mast for stacking VHF yagis above a TH6 in my previous stations from 1985 to 1992. Almost 12' projects above the tower although I am not using all of it, nor was that my intent. In its previous use 10' projected above the tower and I did use all of it.

For now I only need enough mast to minimize interactions between the tri-bander and the 40 meter yagi. For that I need no more than 8'. The additional length is intended to be used when the 40 meter yagi is moved to the guyed tower later this year and the mast is freed up for VHF, 6 meters in particular. If that had not been my plan I would have cut the mast shorter.

As built the XM240 is 1' above the top bearing and the Explorer 14 is 6.5' above that. Therefore only 7.5' of the available 12' is being used.  When plugged into AC8Y's mast stress spreadsheet (based on the ARRL Antenna Book) and choosing a conservative strength of 30,000 psi for the water pipe the wind survivability of the mast is ~115 kph. This is poor but acceptable for the next several months. I live in a 135 kph (85 mph) wind zone and I have only seen such winds twice in the past 3 decades.

The maximum wind speed that can be survived is improved by turning the yagis so that they are broadside to the wind since the booms have a smaller area than the elements for both of my antennas. That is, if you trust yourself to do so when the wind comes up, and you happen to be at home. Our worst winds tend to cluster in the summer and early fall so I feel the risk is acceptable for the next several months even if I fail to turn the antennas.

Tower: Trylon frequently updated the load limits for the Titan series over the years. It is not that the tower changed. Instead they appear to have reacted to how their customers are using their towers. Hams in particular tend to look at the wind load capacity in the most benign case and, unsurprisingly, come to grief. Then they blame Trylon, which is unfair. To gain a deeper understanding of what the tower can support I needed to delve into my archive of old Trylon engineering data that they no longer publish in their catalogues.

From those old sources the T400 can handle 240 of lateral wind load for an antenna mounted within 3' of the top plate. At 135 kph this translates to approximately 12 ft² of cylindrical surface. With ice the wind that can be survived can be greatly reduced due to the increased surface area of the antennas and tower. Do keep in mind that the capacity figures include the surface area of the tower so that ice on the tower also reduces the antenna load.

As for the mast there is a bit of a gamble involved since I am counting on the weather to keep my tower load within the limit for the next several months. The major concern is the multi-element Explorer 14 that is 7.5' above the top plate and therefore reduces the survivable load. Unfortunately a calculation is difficult to perform due to the many factors involved: gust and wind duration; presence of ice; construction quality; etc.

In my estimation the tower has a wind survivability similar to that of the mast.

Rotator: The bending stress of the wind acting on the antennas and mast are transferred to the tower and then to ground -- all earth bound buildings transfer load to ground. Rotators generally have a lower tolerance to radial loads than to axial loads. This is the case with the Tailtwister.

The main contributor to radial load is the Explorer 14 located approximately the same distance above the tower top as the rotator is below it. For the moment let's assume the that the tower top bearing is a perfect fulcrum (it isn't). Therefore the wind load on the yagi translates to an equal radial load on the rotator, but in the opposite direction (windward).

If we again assume 6 ft for the wind area, at 135 kph (85 mph) the wind force is 120 lb. The radial thrust is well within the capacity of the Tailtwister. Although the rotator will survive it may have difficulty turning and braking in a high wind. The solution is to not operate the rotator in a wind storm or to reduce the radial thrust. Actually doing both is wise.

The presence of a second thrust bearing on the second plate of the top section substantially reduces radial thrust on the rotator. It does so by transferring the radial thrust to the tower. The cost is that the tower takes additional bending stress since the load transfer occurs ~2 meters higher than when the second bearing is omitted. Although the proper engineering calculation is beyond my ability I suspect that the trade off is a sensible one.

One intermittent problem I am experiencing is that occasionally the rotator will not rotate. Sometimes if I keep the motor energized it eventually gets going and other times it does so on a later attempt. Thinking that it was a wiring issue I went up the tower to remove all flexing of the wiring near the connector and terminal strip. Now I suspect a failing phasing capacitor or something else. When I overhauled the rotator nothing appeared amiss with the motor or reduction drive.

Back to antennas?

Something odd happened over the past several months. This blog was supposed to contain articles about antennas perhaps more than any other. Yet this is the first one since early August, 5 months ago! Antennas had to take a back seat since I had little time for them during the process of buying and selling property, taking down and putting up towers, and of course dealing with the crush of tasks responsibilities that come from a change in residence.

Although there is much that is still going on behind the scenes to occupy my time I expect antennas to return to the fore. If not immediately, at least in the coming weeks and months. After all, the primary reason I moved was to have the freedom to do much more with antennas, and especially high performance antennas for contests and DXing.

No comments:

Post a Comment

All comments are moderated, and should appear within one day of submission.