I've built a ground plane antenna with a 19.5" vertical and 4 20" radials. Currently, at 147.135 MHz my SWR is 3:1, clearly not ideal (and even bad for the transmitter!).

I'm reluctant to start snipping radials because I really don't know if the problem is that they are too long or too short. How can I figure this out?

My thought is that I could measure SWR at 142 MHz and again at 149 MHz and that this should tell me, based on the difference, if they are too long or two short or if, possibly, simply changing the radial down angle will help, but I have no idea how to interpret the differences.

  • 2
    $\begingroup$ Common-mode currents on the feedline are a common problem with verticals and will cause the impedance to be much different than what's predicted by theory. Might be worth checking. $\endgroup$ – Phil Frost - W8II Aug 14 '15 at 16:10
  • $\begingroup$ Hmmm... I'm not sure I follow you. I'm not doing anything in theory once the antenna has been built. I'm actually measuring it...? $\endgroup$ – David Hoelzer Aug 14 '15 at 19:35
  • 1
    $\begingroup$ If you have common-mode currents on the feedline, then the feedline is part of the antenna. So you can calculate the theoretical length of a ground plane vertical required to tune at your desired frequency, but if what you really have is a ground plane vertical with this extra wire coming out the other end (the feedline), all that theory about verticals isn't really accurate. $\endgroup$ – Phil Frost - W8II Aug 15 '15 at 13:24

My trick is instead of cutting, rather extend the whip first, and see if the SWR gets better or worse. Extend it 1/2" with a bit of wire wrapped around it and sticking out. If the SWR gets better, then you made it too short in the first place. If it gets worse, you can start trimming.

Remember you should only measure SWR when you are away from the antenna - anyone within 1 or 2 m of the antenna will change the SWR and confuse you as you tune.

You need to bend the radials down at 45 degrees to be able to match it properly. The length of the radials isn't as critical as the length of the whip, rather just bend them up and down.

Lots of small steps is the key, not one big jump. Enjoy the antenna tuning!

| improve this answer | |
  • $\begingroup$ Thanks @tomnexus! Great suggestion. I have them at 45 degrees or so now, but perhaps bringing them up a bit can be revealing as well. $\endgroup$ – David Hoelzer Aug 14 '15 at 10:28
  • $\begingroup$ A small piece of aluminum foil may also serve as an effective extension of an antenna. In fact, my current 2m antenna uses an 18" long strip of aluminum foil as the extension to a piece of wire in a BNC connecor to be the primary radiator. $\endgroup$ – user2943160 Jun 6 '16 at 2:01

If you take SWR measurements at a range of frequencies and plot them (or just watch the SWR reading as you change frequency), you should get a graph with one or more valleys of low SWR.

Ideally, if you find the ratio between the valley's actual position and where you want it to be, and trim or rebuild-bigger the antenna according to that ratio (or rather, the reciprocal of it — note that physical lengths go with wavelength, not frequency!), then the resulting antenna will be spot on.

In practice, it won't be exact because you're not building an exactly scaled antenna — your feed-point components and your wire diameter will be staying constant. But you can aim for a little bit too big, then trim it down until the SWR stops dropping.

The above procedure will get you minimum SWR for that antenna design. But remember that you can have a resonant antenna that still has the wrong resistance (i.e. not 50 Ω, assuming you're using normal equipment and coax). Resizing the antenna will not change the resistance — you have to change the shape or add a matching network.

The simplest way to figure this out is to use an antenna analyzer that can read out the resistance value. Find that minimum SWR point, then look at the resistance (or the magnitude of the impedance — which is the same here).

You can also tell from the shape of the SWR graph. A matched antenna will have a very sharp valley that hits 1:1 SWR. One with a different resistance will have a broader valley that does not ever reach 1:1.

| improve this answer | |
  • $\begingroup$ Yes.... but how can I tell based on what I have now, using the SWR at a range of frequencies, whether the elements are too short or too long? $\endgroup$ – David Hoelzer Aug 13 '15 at 18:52
  • $\begingroup$ @DavidHoelzer That's the first two paragraphs — find the valley, take the ratio between what you have and what you want, and resize. $\endgroup$ – Kevin Reid AG6YO Aug 13 '15 at 19:01
  • $\begingroup$ Mmm... I doubt I'm going to find a "valley" since I'm using it on a 2 meter transceiver with a 9 MHz transmission range. :) More likely I'm going to find a slope. Let me ask it another way... If the slope rises toward 149 MHz from 142MHz would that indicate that the radials are too long or too short? $\endgroup$ – David Hoelzer Aug 13 '15 at 20:00
  • $\begingroup$ Ah, I see. If the SWR increases as you go up in frequency, the valley is lower in frequency, so your antenna is too big. (Note that it won't necessarily be a simple curve with one valley — but unless you have an antenna of totally the wrong size or a deliberate multiband antenna, that won't likely happen here.) Also, I'm not familiar specifically with the design of ground plane antennas, but I think the vertical element matters a lot more than the radials. In any case, you should certainly be checking/adjusting its length as well. $\endgroup$ – Kevin Reid AG6YO Aug 13 '15 at 21:16
  • $\begingroup$ Thank you! The vertical is 19.5", so it should be very close. The radials aren't "fixed" at the moment.. Currently they're just hooked on, but I'm planning to solder or tack weld them permanently when it's close. If SWR indicates it's too large, though, from what I've read, the first thing to do is start snipping the ends off of the radials for this type of antenna. Thank you! $\endgroup$ – David Hoelzer Aug 13 '15 at 21:28

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.