I have a 20m long dipole that starts at 2m above the ground and ends at perhaps 20m above the ground. It was -10 °C here tonight (that's +14 °F), currently around zero (32 °F) and my antenna is out of tune because of this (I noticed it before with slightly warmer weather). There are also small icicles everywhere on it, of perhaps 1-2mm long.

My antenna tuner already was close to its limits when tuning for 20m and now its completely out of range. The obvious solution is to use a smaller capacitor to shift the range. Is that the best option, or can I better do something else to deal with the cold weather?

From the transceiver to the tuner is coax, from the tuner to the antenna ladder (4m inside, 5m outside). The antenna is a 20m dipole, so two sides of 10m.

  • 1
    $\begingroup$ Transmit with a high enough power to melt the ice? :) $\endgroup$ – Phil Frost - W8II Jan 7 '17 at 13:15
  • $\begingroup$ Is the antenna fed with coax or ladder line? What kind of tuner are you currently using? how long is the antenna itself? $\endgroup$ – Hamsterdave Jan 7 '17 at 13:47
  • $\begingroup$ @CamilStaps When you add more information, please edit it into the question (as I just did) rather than only putting it in a comment. $\endgroup$ – Kevin Reid AG6YO Jan 7 '17 at 16:01
  • $\begingroup$ @KevinReidAG6YO right, I knew that. Apologies, thanks. $\endgroup$ – user885 Jan 7 '17 at 16:10
  • $\begingroup$ By 20m dipole, do you mean a dipole for the 20 meter band (14 MHz)? Or do you mean the dipole is 20 meters long, and as such would be resonant around the 40 meter band (7 MHz)? $\endgroup$ – Phil Frost - W8II Jan 8 '17 at 15:34

You say your dipole's total length is 20 meters, and you are using it on the 20 meter band. When the dipole's length is equal to some multiple of the wavelength (1λ, 2λ, 3λ, ...) the dipole is antiresonant, and the feedpoint impedance is at a maximum. Being the worst possible match, it will require maximum range from your tuner.

If the total length of the dipole is half a wavelength, or some odd multiple thereof (0.5λ, 1.5λ, 2.5λ, ...) the antenna is resonant. Feedpoint impedance is at a minimum, around 73 ohms. This is a good match and you don't even need a tuner, although you may need to take the antenna up and down a few times to adjust the length initially.

You can explain this effect by thinking of one side of the dipole as a quarter-wave transformer. The end of the dipole must be a high impedance (where can the current go?) So consequently the feedpoint, a quarter-wavelength away, will see a low impedance. With the dipole a total wavelength long, each side is half a wavelength long. You can think of this as two quarter-wavelength transformers together, in other words, no impedance transformation at all. So the feedpoint sees the high impedance at the end of the dipole.

I suspect if you make your dipole any other length, you'll be able to tune it. And if you operate primarily on the 20 meter band, I'd suggest making the antenna a half wavelength at that frequency. You don't gain anything by making it longer, you'll lower SWR on the feedline and associated losses, and your tuner won't need as much range.

| improve this answer | |
  • $\begingroup$ I had never expected this. I thought that this antenna is optimal for 40m, and therefore also for 20m, 10m, 5m (if it would exist), etc. Do you by any chance have an article which explains this phenomenon in more detail? As I don't have a lot of freedom in my antenna setup, I have now adapted the tuner's range and it seemed to work fine today (though I couldn't test a lot and it has been less cold). I used to have a simple 10m longwire, in about the same position but the maximum point lower. Would you think that would give better performance on 20m? $\endgroup$ – user885 Jan 8 '17 at 22:33
  • $\begingroup$ antenna-theory.com/antennas/dipole.php is a good reference for dipoles of arbitrary length. If you have a low-loss feedline, then there probably isn't a huge difference in performance moving to a naturally resonant, shorter dipole, given your space restrictions. However it's more difficult to get a short wire to tune for a long wavelength than the other way around, so if you want to retain 40 meters keep it long. $\endgroup$ – Phil Frost - W8II Jan 9 '17 at 17:39
  • $\begingroup$ I edited to add a brief explanation as well. $\endgroup$ – Phil Frost - W8II Jan 9 '17 at 18:07

The overall antenna length being very close to a perfect 1 wavelength is the linchpin of the issue, I suspect. It is not surprising that it can be easily moved outside of the tuning capability of your tuner, as the impedance peak near resonance can be quite sharp (depending on the Q of the antenna system), allowing the impedance to change by hundreds or even thousands of ohms over a bandwidth of just a few percent of the signal's frequency. It is quite possible to see a swing from 10:1 SWR at the tuner (which is well within the range of most decent tuners) at 14.350MHz to 100:1 at 14.450MHz in a well engineered antenna with good quality feed line.

I would wager that if you were to do an impedance sweep of the antenna system during warm weather, you'd find that the actual impedance maximum (indicating a perfectly resonant 1 wavelength) would be just a bit above 14.350MHz, probably not by more than a few tens of kHz. Water, ice, and snow on the antenna can contribute a small amount of electrical loading, effectively decreasing the resonant frequency, in this case dragging the impedance maximum in to the ham band. This is common with insulated antenna wire and 300 or 450 ohm "window" type balanced feed line.

In addition, the larger dielectric losses contributed by the liquid around the antenna compared to air can alter the resistive component of the complex impedance, and asymmetrical dielectric losses on the two sides of the system, especially on the balanced line, can contribute to significant imbalance and common mode current, causing the feedline to act as part of the antenna, which will further alter impedance and cause unpredictable changes to SWR.

Since one end of the dipole is quite low to the ground, just 1/10 of a wavelength or so, coupling to the ground and nearby objects is probably relatively significant. If the recent cold weather also came with a new snow fall, that may further contribute to a shift in resonant frequency.

The point being, you are likely seeing the antenna's resonant frequency shift from somewhere above 14.350MHz to somewhere very near the 20 meter band due to loading effects of water, ice, and snow. It's also possible that the cold is causing mechanical changes to the system, such as sagging that brings some part of the antenna or feedline close to an object it can couple to. You may also be experiencing an increase in common mode current on the feed line, which will result in inaccurate SWR measurements. The BLT tuner should have good common mode isolation with that isolated balun, so I wouldn't be quite as inclined to suspect the SWR meter here.

Exactly how to remedy the problem depends on quite a few variables, not least of all how much flexibility you have with the antenna location and construction. If the balanced line is "window" type line, sheltering it to prevent any rain or ice from accumulating on it (dust can cause problems too), or moving it up away from the ground to avoid problems from snow accumulating around it may help. A better bet would be to ensure that it is kept away from any object it could couple with by a distance of at least the distance between the two conductors, but ideally 2 or 3 times that. Replacing the window line with true ladder line can minimize problems with dielectric losses when wet or covered in ice, and would probably be the optimal solution.

Getting the low end of the antenna another couple of meters above the ground may also significantly improve the situation. As it is currently installed, there is almost certainly some degree of imbalance on the balanced line due to the asymmetrical interaction with the ground between the two ends of the dipole, which will make the balanced line all the more likely to interact negatively with objects around it.

Another option would be to slightly lengthen or shorten the antenna, moving the 1 wavelength impedance maximum away from the 20 meter band. A total length of around 25 meters is often a good compromise, as it will retain good characteristics on 40m, while still allowing it to tune both 20 meters and 30 meters with a reasonably good tuner.

| improve this answer | |
  • $\begingroup$ Thanks a lot, I understand it a bit better. Unfortunately, I don't have a lot of freedom with my antenna setup, so it's difficult to change the length or move it further up. Once it gets a bit warmer I will put an isolation around the window line in the hope it will improve. (Un)fortunately, it doesn't get this cold often any more here, so I don't have a lot of opportunity to test. For now, I changed the tuner's range by disabling one of the variable capacitor's sections and it seemed to tune fine this morning (though I couldn't test a lot today and it has been much less cold, around 0 C). $\endgroup$ – user885 Jan 8 '17 at 22:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy