Other SOTA sites: SOTAwatch | SOTA Home | Database | Video | Photos | Shop | Mapping | FAQs | Facebook | Contact SOTA

Interesting new SOTA Vertical Antenna

, i think these tuned cp ideas are worthy of experimentation

I look forward to your findings
John ve3ips

When I first started in SOTA, I made myself some resonant end-fed half waves and for the counterpoise, I experimented with lengths by having the antenna analyser attached. I was surprised to find that counterpoise elements of just an 1/8 wavlength looked good on the analyser. I saw no reason why 1/8 wave should work and I reverted back to longer lengths when actually operating. Then I bought myself a linked dipole and a squid pole and the difference was so great I never went back to the end-feds. Note - the end-feds were just being thrown on the top of hedges or into tree branches, had they been taken vertically up the squid pole, I’m sure they would have worked a lot better.


Hi Ed,

I see your point, but I think it is unfair to compare an EFHW thrown over a few bushes with a center-fed dipole mounted on a squid-pole. With my 3- and 5-band EFHWs on a 6m mast, I get excellent results - often 17 - 20 dB around Europe, and occasionally NA contacts.

The thing about the counterpoise is IMO that in you case, i.e. on an end-fed antenna, the feedpoint has a very high impedance; thus, an imperfect counterpoise will be just okay (often just the braid of the coax). But with my proposed Up-and-outer design, the feedpoint impedance will be in the 30 - 70 Ohms region, thus the counterpoise will be essential for effective radiation.

AFAICS, a linked dipole and an end-fed differ just by the position of the feedpoint (and that traps or links will be symmetrical in a dipole and just one per band on an end-fed).

73 de Martin, DK3IT

Hi Martin,

That’s why I said, I think they would have worked a lot better on a pole.

My point was the counterpoise length of 1/8 Wavelength appearing to be good by the Antenna Analyser - might be worth trying if you have the time to see if you find the same.

73 Ed.

Hi Ed,
actually, I can already tell you that it is very different on this vertical / up-and-outer, because I tested in with the previous versions, like this one

Electrically, it is equivalent to the new design; the difference is that the loading coil is an air coil in the first version and wound on a toroid in the new design.

When tuning the first version, even 10 cm +/- make a difference. So 1/8 will not work.

What still surprises me, though, is that the resonant length for the counterpoise is typically significantly less than 1/4 lambda.

This is also equivalent to the counterpoise lengths recommended for the ATX-1080 antenna - they indicate a length of

counterpoise_length = 54 / frequency_in_MHz

But I do not understand where this formula comes from.

73 de Martin, DK3IT

My understanding, based on the statements by G. Janzen in his publication “Monopol- und Vertikalantennen” is as follows:

Monopole antennas (usually mounted in a vertical manner and therefore simply called Vertical antennas) are, as the name implies, only Half-Antennas. For these monopoles to radiate, a maximum of current must be able to build up at the base - which requires an electrically equivalent counterpart.
This electrical counterpart was/is called Gegengewicht by the German-speaking antenna experts (in English: counterweight or Counterpoise).

This Gegengewicht/counterpoise can then be realized either exclusively through the earth (Marconi antenna) or by metallic plates/roofs or radially arranged wires, the so-called Radials.

So, radial(s) is understood as a possible realization of the electrical counterpoise.

IMHO, with a little generosity, the two terms can be used quite synonymously, isn’t it?

Hi Heinz,

I believe the term “counterpoise” is an attempt to distinguish between tuned radials (normally quarter wave) and one or more much shorter conductors thrown into a design because they make an antenna show a usable impedance when without them, it doesn’t.

One of the articles I found in an archive of articles by LB Cebik W4RNL pondered the introduction of the term “counterpoise”. He eventually concluded that the term had no established meaning in antenna theory and declined to use it. He was a purist but perhaps his puzzlement over the term indicates a lack of theoretical validity.

But now, some years later, I suppose it is taken to mean lots of different things. Like many things in amateur radio, some are well supported by communications theory and some are not.

Andrew vk1da/vk2uh

You probably refer to these contributions by L. B. Cebik:


Indeed, there are actually about as many opinions as there are experts. Too bad that a joint meeting with all these experts could unfortunately not take place, hi.

BTW, I myself am not an expert on monopole antennas, only a few 100 out of over 28’000 SOTA QSOs were made using such antennas.

By the way, Cebik’s original Web page is long gone, but the Web Archive preserves a copy of at least a part of it:


It is incomplete, however.

73 de Martin, DK3IT


Yes that was the article I dimly remembred. I did not remember that he was saying that “counterpoise” had that specific meaning, but that is not how it is used today.

His conclusions are what stuck in my head. That the term had been misused and corrupted steadily through the years.

As you say it would be good if the experts could form a joint view of all these matters and put some fallacies and myths to rest. but that is not going to happen…

Thanks for that link.

73 Andrew VK1DA/VK2UH

I have a resonant efhw for 20m it’s a UK commercial one. I put it as vertical as possible on a 10m pole. So it’s slightly sloping in fact. I’ve had a s2s with Australia on it from g/tw003 a year or two ago. It has no dedicated counterpoise wire attached.
So I like it.

Hi Martin,

The “Upper and Outer” has a venerable pedigree - an internet search and search of the reflector will turn up quite a bit of interesting info :slight_smile:

Shortening of the required length of the element run close to the ground is a well known phenomenon, ascribed to the increased capacity to earth experienced by it. It has been suggested that a way to adjust the length of this element is to monitor the current in it, and adjust for equal current in both legs.

This is equivalent to .18 of a wavelength, or 72% of a (physical) quarter wave which is comparable with the shortening quoted in other sources, though I’d have thought this would be a very ball-park figure, subject to the vagaries of a host of variables.

I suspect the formula may have been derived empirically?

As Cebik points out, many antenna configurations can be resolved into an off-centre fed arrangement - it could be that Ed found a serendipitous sweet spot which provided a good match with the radial at that length?

Aren’t antennas fascinating :slight_smile:

73 de Paul G4MD

1 Like

Here’s an anecdote about the QRPguys antenna. I made the kit Saturday night. My family wanted to hike yesterday so I packed the radio, figuring I’d test it out over at a Michigan State park. I definitely like the smaller footprint of the vertical over my normal 60’ of EF wire. I immediately found two problems though: the 40m torroid wasn’t working properly (I am guessing I didn’t get the enamel off the wire) and I didn’t pack a fresh battery for the KX2. I called cq twice and the power dropped to 5w.

I made a couple contacts on 30m, way up into Alberta Canada. When I switched to 20m I got quite a surprise. In the span of 20 min I worked Romania, Italy, and Belgium along with a few more distant Canadian stations. I’ve only been a ham for ~6mo, and before yesterday I’d made only two DX contacts. So this was one heck of an exciting trip to the park for sure. The last time I used code outdoors it saved my bacon on Mount Guyot. This time I talked to the other side of the world on 5W. I feel like the code gods are trying to tell me something.

Of course, it was probably the conditions yesterday, not the antenna per se. I’m going to straighten out the 40m torroid and give it another shot this weekend (hopefully on a couple of summits).


Yesterday there was the ARRL HF DX contest on, so that will have helped get some activity on the bands. Well done with the DX contacts - it’s quite something to work those distances on QRP!

I’ve just received the wire for the driven element and the four radials so it’s time I get busy and build the kit as well!! Mine will be (I hope) a 20m, 40m, 60m version as I don’t operate CW hence 30m outside of IARU Region 3 is of no interest.

73 Ed.

A further thought. It seems clear from Cebik’s article that in his assessment of the historical use of “counterpoise” he recognised that the term “counterpoise” applies to an arrangement of a number of wires of non-specific lengths, placed close enough to the ground that they create a substantial capacitive coupling into the earth. This function assumes the ground is actually conductive and would require a decent network of wires, so would not be provided by a single short wire, elevated or not. And also, it would not work on a rock, which many mountains effectively are. Rocks have poor conductivity. So the single wire sticking out of one end of the impedance transformer or ATU for an ostensibly end-fed antenna, is not a counterpoise, it is one end of an off centre fed antenna and should be treated as such.

Andrew VK1DA/VK2UH


I built up my version of the kit as per the directions. In my instance, the antenna works great on 20 and 40m with the switches in their proper place, however for 30m I found the antenna works better with the 40 vs 30/20m switch still set to 40m. I’m not sure if this means on of my torroids is not wound properly or needs adjusted, however using it in this manor worked just fine for me.

The one modification I did make after assembly was the use of loctite threadlocker blue (non-permanent) compound to get the two nuts and lock-washer more solidly locked in place. This prevents the attachment point from falling apart in the field. I have always had issues with the qrpguys bolt, nut and wingnut attachment method without the use of said loctite. I emailed them and suggested they add this to the directions.

I will be giving this antenna a shot this year for my next activation or two. It is lighter than my pacific antenna trapped dipole, which is currently my favorite SOTA antenna. It will also allow me to use less lossy coax then my current 25ft of RG-174. This is actually quite significant!

I compared a 10ft length of RG-8X to 25ft of RG-174, at 14.030, with an swr of 1.5 to 1 using 15 watts (my typical operating conditions). I ignored the bnc insertion loss since both will have this.

The 25ft Length of RG-174:
0.80 db of loss
12.48 watts reach the antenna
83.2% transmission line efficiency.

The 10ft length of RG-8X:
0.11 db loss
14.60 watts reach the antenna
97.3% watts reach the antenna
Much better…

I have plenty of RG-8X jumpers around in the sub 10ft length variety.

Because I’m initially using 26 AWG stranded wire for both the radiating element and radials, adding at least 4 more radials should take up very little space or weight so I plan to do so. For longer term activations, or during activations where having the best signal possible matters (S2S days, contests, etc) I may bring up to 16 or even 32 radials (most sources agree 16 radials is really the starting place for on-the-ground verticals) and replace the vertical radiator with a 20 or 18 AWG wire (to decrease loss in the radiator).

You might like to try DXW174 from DX Wire:

The losses at 7Mhz are only 62% of those of normal RG174. DXW174 has a solid copper core, which reduces the so-called skin effect at low frequencies. It’s cheap too.

Hi Evan,
Is it possible you have the two torroids the wrong way around? Basically the combinations made with the switches should be:

  1. straight through - 20 metres
  2. both torroids in series - 40 metres
  3. only the 30m torroid selected.
    the other combination, which I think you are saying works best, if wired correctly would select the other torroid - the one that adds on to the 30m one to make it into a 40 metre inductance. If you have the two toroids reversed on the board, this “odd” switch setting is actually selecting the 30m toroid.

I’ve just finished winding my torroids - in my case both are identical with 45 turns making 11.5 microhenrys. So in my set up I have

  1. Straight through - 20m
  2. One Torroid selected - 40m
  3. both torroids selected - 60m.
    It just so happens that 23 microhenrys is what is required for 60 metres.

I still waiting on the on-off-on switch that I ordered and will have the antenna ready for that when it arrives. I wont be using the slide switches.

I like the way that the QRP-Guys have the bolt head above the wing-nut so that it cant be lost. I have tightened the two nuts with the locking washer tightly into the PCB so I don’t expect it to come lose but if it does loctite is a good idea. I’ve also soldered hook connectors onto the end of my driven element and radials (all 4 connected together) wires, so that I can hook and unhook them rather than relying on screwing down onto the wire.

I’m using RG-58AU coax (as I have it), it’s probably not as good as RG-8X but better than RG-174.

In my opinion, … in a portable situation the advantage of multiple radials drops off after 4 - yes there’s an improvement but (again in my opinion) not worth the extra effort. More justifyable I think, is making the radials the resonant length for the bands you want to work on, rather than the randon 10 ft. length.

Good luck with the antenna. I won’t be using mine until after the big S2S event as I simply don’t know how well it will work. Better to stick with known entities on such an event.

73 Ed.

Bought the kit from qrpguys today. :grinning: