Can I feed a folded dipole constructed of hollow tubing by threading coax through the element to create both an impedance transformation and balun? I have fed coax through the tubing of the dipole from a point opposite the feed point and soldered directly to the open ends of the dipole from inside the tubing. SWR seems ok with 50 ohm cable but a I being fooled into thinking all is ok? A bit hard to explain, the coax is inside the element for 1/2 wavelength.

  • 1
    $\begingroup$ Thanks for joining us. Please report the observed SWR when coax is connected directly to the antenna terminals. $\endgroup$
    – Brian K1LI
    Commented Jun 15, 2019 at 20:49
  • 1
    $\begingroup$ When I connect the coax to the folded dipole terminals directly I get an SWR of around 8:1 . When I run the coax inside the tubing of the element for 1/2 wave before connection I get 1.2:1. I am wondering if this is actually working as it should because it seems "Too" good! I mean I can get a great SWR by just sticking the coax in a bucket of water but it doesn't radiate very well! $\endgroup$
    – R Johnson
    Commented Jun 16, 2019 at 8:25
  • $\begingroup$ A center-fed half-wave folded dipole has 4 times the impedance of a regular dipole, thus the 4:1 balun in Brian's answer. Where are you measuring the impedance or VSWR when you get that 8:1 reading? Also, a drawing of your antenna would help us a lot. $\endgroup$ Commented Jun 16, 2019 at 16:55

3 Answers 3


Running the coax through the folded dipole is one good way of feeding it.

Advantages over the more traditional 1/2 wave coax balun:

  • No balun required - unbalanced coax all the way to the feedpoint, with guaranteed symmetrical excitation
  • Easier to waterproof the feed, as the connections are inside, and you can heatshrink over the feedpoint
  • Only one side of the dipole touches anything, the other just hangs open


  • Matching network may be less broadband than the 1/2 wave feed
  • Manufacturing process is a little fiddly

To answer your questions directly:

  1. Yes you can feed a folded dipole by passing the coax through the tube

  2. It doesn't really create any impedance transformation
    (but it does take care of the balancing that's otherwise required)

  3. Your measured impedance seems suspiciously good - as you describe it I would expect you to find a 4:1 VSWR.

Let me draw a few diagrams

1) This is the case I believe you are describing:
coax feed no matching
The coax is hidden entirely inside the pipe, so it has the minimum effect on the patterns, and no balun is required to feed the balanced antenna from the unbalanced input. The coax should see the full 200-300 ohm impedance of the folded dipole, so a bad match without further work.

2) One of these probably describes what you did when you say
"When I connect the coax to the folded dipole terminals directly I get an SWR of around 8:1":
don't do this
and definitely not this
You shouldn't really connect the coax like this, it's not even worth measuring. The effect of the feedline is so profound, especially on the higher impedance of the folded dipole, that you won't learn anything about the dipole impedance.

3) I recommend something like this:
working folded dipole coax feeed
With the coax run through the pipe, the required balun action is taken care of.
Some impedance transformation is required. In this drawing I've shown a piece of 100 Ohm cable, 1/4 wave long, when I built some of these I used a small PCB across the feed gap, which makes the connections more robust and takes care of the 100 Ohm line. You could also use an actual 4:1 Unun transformer, or even an LC network, though that will be quite narrow band and spoil the dipole.

You can see a photo of a commercial dipole fed "through the pipe" here (not mine!).

  • 1
    $\begingroup$ Yes your diagrams are exactly what I have done. I think the 1/4 wave matching section is what I require! Thanks for that ! $\endgroup$
    – R Johnson
    Commented Jun 18, 2019 at 0:21
  • $\begingroup$ I wonder if I can fabricate the matching section from commercially available coax There is some 92 ohm cable around or will just length adjustment of somthing like 75 ohm cable be an option? $\endgroup$
    – R Johnson
    Commented Jun 18, 2019 at 0:23
  • 2
    $\begingroup$ Great answer. I'll add a small, but important point... the folded dipole will force balance despite an unbalanced connection. See: "Buxton, Carey G., Warren L. Stutzman, Randall R. Nealy and Aaron M. Orndorff (2001). The folded dipole: A self-balancing antenna." This, of course, doesn't negate the requirement to dress the feed away as your drawings help us understand quite well, but the notion a folded dipole is self-balancing is rare knowledge. I've tested it too... It's pretty cool. $\endgroup$
    – JSH
    Commented Jun 19, 2019 at 19:31
  • 1
    $\begingroup$ Stutzman is a big name! The paper shows a simulation of a folded dipole, with unbalanced feedline 1/4 wave long at right angles to the dipole. It shows currents on each half of the dipole, and they are remarkably similar. Currents in the feedline are balanced. Compared to a regular dipole with the same coax feed, where currents in the two halves are very different, and the feedline currents are unbalanced. It works remarkably well. I still wouldn't drape the cable down over the dipole. The effect is more narrowband than the dipole's impedance bandwidth. $\endgroup$
    – tomnexus
    Commented Jun 20, 2019 at 20:13
  • 1
    $\begingroup$ @Mike Waters "That implies that it automatically cancels common-mode current on the outside of the coax feedline, and the resulting radiation from it. Can you confirm that?' To a reasonable degree I have simulated and measured to confirm the result. There are two Japanese references concerning this in the link as well. $\endgroup$
    – JSH
    Commented Jun 21, 2019 at 21:50

It seems as though your configuration might represent something like the $\frac{\lambda}{2}$ 4:1 coax balun shown in figure 5 on page 18-6 of the ARRL Antenna Book:

enter image description here

I've been unable to see the equivalence, but running your coax feedline through the tubing does result in two coaxial lines:

  1. the center conductor and the inside of the shield of the feedline
  2. the outside of the shield of the feedline and the inside of the antenna tubing

This means that the feedline jacket and the air-space between the feedline jacket and the inside of the antenna tubing comprise the "dielectric" of the second coaxial line. We can't compute the impedance of this second coaxial line without knowing its dimensions and the composition of the jacket material, but it seems as though it would be difficult to control this impedance unless the fit was fairly snug. Since you observe a strong effect on the SWR at the input to the feedling, it's possible that you've hit on a serendipitous combination of variables that produces your desired results, one of which could even be the feedline length back to the SWR meter.

As long as the resulting signal strength is what you expect, it sounds like you have what you need. Nice work!

  • $\begingroup$ I have made 40m folded dipoles using exactly that balun. The dipole was made from 300 ohm TV twinlead, and the 4:1 balun was made from 75 ohm coax. It also worked on 15m. I suggest that you add that the velocity factor of the coax needs to be taken into account when measuring. $\endgroup$ Commented Jun 20, 2019 at 19:53

Here is the way I feed my half-square antenna. The choke is necessary.

enter image description here

  • $\begingroup$ Would you, please, explain the function of the 17ft aluminum tubing? $\endgroup$
    – Brian K1LI
    Commented Jun 15, 2019 at 20:48
  • $\begingroup$ @Brian The 17 ft aluminum tubing is one of the vertical radiators for a 20m half-square. The coax is normally routed at right angles to the other conductors and connected at the top of the radiator. My coax goes up through the radiator and connects at the top while using the radiator as a support. This configuration is somewhat similar to the question being asked. $\endgroup$ Commented Jun 15, 2019 at 21:38
  • $\begingroup$ Thanks - I see it now. Feeding the half-square at one of the corners gives a good match to 50$\Omega$ coax. $\endgroup$
    – Brian K1LI
    Commented Jun 15, 2019 at 23:52
  • $\begingroup$ I like that idea. It would be a useful way to get the feed point higher off the ground and may reduce the angle of radiation. I was just wondering how I was going to incorporate a vertical HF antenna in my "re'arrange frenzy"! Now you have given me an idea!! Thanks!! $\endgroup$
    – R Johnson
    Commented Jun 16, 2019 at 8:33

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .