Strange link dipole shenanigans?


In the true spirit of this page, i installed a poorly built antenna on the weekend, and I don’t know why?

I got the courage to cut my unicorn blood DX wire, and began the process of making a link dipole. Followed the guides form all the normal places, VK3ZPF and VK3PF and the Sotabeams Calculator among many. I built up the antenna, followed the instructions and started with the 20m bit first and its miles off, I adjusted and trimmed and checked and no luck.

VSWR no better than 1.7:1, and more of a worry, impedance is ~75ohms.

sark_001 sark_004

I made the links with anderson power pole quick connectors and the insulator is two pieces of light core flute stuck together for strength.

 IMG_3403 IMG_3406 IMG_3409 IMG_3404

I connected the longer bits of wire and have built a wonderful 40m dipole, at 7090Khz, its 1.03:1 and impedance is ~50ohms. on my Sark. But on 20m? Whats going on? (the pics above show you the sark results)

I used the home made 1:1 current balun I made the other week, this measures fine and appears to work well in other usage. but on this new antenna, at 14Mhz, no luck?

Things I need to do myself to self resolve this include direct connecting the coax to the wire (remove my Balun). Try another location, i was in the park, but away from “stuff” no more cluttered than any sota summit? try different wire?

So, does anyone have any tips, ideas of pointers to what Ive done wrong and what my issue may be? Sure I can tune around the problem with my Elecraft T1, but thats not the idea of building a resonate antenna.


9 thoughts on “Strange link dipole shenanigans?

  1. HI Andrew,

    I am no antenna expert, but here are some thoughts that might be worth examining further with one of the antenna modelling packages.

    On 7 MHz, the inverted V ends are relatively close to the ground, with the antenna centre at low height with respect to wavelength. The capacitance between the antenna and ground will affect the impedance, lowering the theoretical 72 ohm impedance of a dipole in free space to the value of 50 ohms.

    On 14 MHz, the wire ends are higher above ground, so the capacitance effects are lower, therefore having less effect on the antenna impedance? Also, whilst still “low” in terms of wavelengths, the feed is higher above ground on 20 m than 40 m.


    Peter VK3PF

    PS: I think I know what you mean when you say “7090Khz”. What you have written is actually NOT defined. It should be written as “7090 kHz”. The SI system says that there there is a space between the numbers and the units. The unit for frequency is Hz for Hertz (not hz). The SI symbol K says Kelvin – the SI unit for temperature. For multipliers above 1, the SI prefix is usually a letter in capitals, except for the multiplier for x 1000, which is in lower case (because of the Kelvin using K).

    Lots of people misuse (I think abuse) the SI system as they have not learnt the system properly. This is not aided by many who are simply not worried about the real language they should be using. Do not start me about American usage! They simply have little idea at all of anything that originated outside of the USA. (Tongue firmly in cheek 🙂 ).
    PS added as an attempt to educate, not to be critical.

  2. Andrew!

    Here is a great reference for the ‘rest of us’:

    I have to admit: I usually count on the feed point impedance being around 200 Ohms (theoretical!)

    However, in reviewing my personal notes I rediscovered that this is a rule of thumb for the creation of Off Centre Fed Dipoles (OCF). I use OCFs for both home and portable operations; implementing a 4:1 BalUn at the feed point and feeding with 50-ish Ohm nominal impedance coaxial cable.

    It is apparent that you might (kinda) expect a feed point impedance of around 72-ish Ohms in a normal dipole that you are building; however you are deploying it rather close to the ground.

    Without repeating all the science explained in the link provided, I will just direct your attention to the tables embedded in the page: they explain some theoretical Heights A.G.L. that might make things work better (or not!?).

    Bear in mind: since you are mounting your antenna well below the…what shall we call it?…”benchmark”(?) height of 40 feet (12.19 metres) , you are creating an NVIS antenna more than a DX one. Generally speaking.

    One last thing: at the end of the SOTABeams link you provided, there is a diagram of what appears to be a variable impedance matching device. Clearly, this is intended to be part of the system being outlined by the author. Are you using one?


    • Thanks. So 70 ohms with my setup. Height above ground. Etc appears to be on the money. I have the elect aft T1 automatic atu to tune the thing. Just wanted it better to start with?

      • Yes, of course we all want our antennas to be the most efficient straight off the bat. Well, except for those who have been using “poorly built” ones for decades LOL.

        Truth is, we will almost always have to use some sort of impedance matching gear in order to keep power amplifiers happy and reflected power to a minimum. At home, one might be able to keep a fixed dipole at the correct height in order to have a 50 Ohm feed point AND resonance AND decent efficiency AND decent radiation pattern. Maybe.

        Matching BalUns are always slightly lossy. But the good news is that the guy on the other end is likely using one also; so nobody has the upper hand 😉

        –… …–

  3. Andrew,

    I vote for Peter’s explanation of the 70 ohm impedance on the 20m dipole. I also agree that the approx 50 ohm impedance on the 40m dipole would be related to the increased capacitance to ground for the longer legs closer to ground.

    I wouldn’t be concerned by the 1.7:1 SWR as it will generate insignificant power loss in your configuration (perhaps 0.15 dB). I’d recommend the following article to get a good perspective on SWR and its effect on system losses.

    I use a benchmark of about 2.5:1 for acceptable SWR in a portable operation and this is not so much for the loss involved, but rather to avoid transmitter damage from a less than ideal load for the PA. With a tuner, even a 10:1 SWR on the line would likely lead to less than 1/2 S point difference in transmitted signal (assuming 10m of RG-174).



Leave a Reply