Resonant versus Non-resonant Antennas

One of the things I hear is that, especially for QRP operating, resonant antennas are greatly preferred.

Some of this seems to stem from some misconceptions centered around low SWR, resonance, radiation resistance, antenna efficiency, feedline losses, and so on.

So maybe it’s worth a little exploring around those issues to see if that exploring sheds some light on things.

What, exactly, does ‘resonant’ mean?

An antenna is resonant when it is a multiple of a quarter wavelength long.

More usefully from an antenna analyzer point of view, an antenna is resonant when its impedance is purely resistive, with no reactance.

It’s worth noting that an antenna can be resonant, and still present a high SWR to a 50 ohm transmission line or transceiver.

In other words, resonant does not imply low SWR.

Beyond that, observe that an antenna cannot be resonant across a range of frequencies - that is, you can’t have an antenna resonant across the entire 20m band.

Radiation efficiency and resonance

The radiation efficiency of a wire doesn’t depend on resonance. The radiation depends on current. If RF current is flowing, the wire is radiating. The actual resistance of the wire matters - better conductors will have higher current and radiate more - but resonance doesn’t matter.

Resonance is all about making it simple to match the impedance of the antenna to the impedance of the transmitter.

In other words, we know that the impedance of a half-wave dipole at about 1/2λ above ground level will be, roughly, 73ish ohms, which is a pretty good match for the 50 ohm impedance of the coax most of us use. The upshot is that not a lot of energy is reflected back toward the transmitter at the junction between the coax feedline and the antenna itself.

So, why do we care about resonance, then?

Tweaking the length of a wire antenna to hit the resonant length means we’ve dialed out any reactance, which makes it easy to match the impedance of the antenna to the impedance of the feedline and transmitter.

And if we can pick an antenna design that has a natural impedance at the point of resonance that is 50 ohms, or a close match to 50 ohms, we don’t need an antenna tuner and our losses in the 50 ohm feedline will be low.

Where can losses appear

One reason lots of hams think non-resonant antennas are inefficient is that feedpoint impedance will be (by design) fairly high (around 450 ohms) and some sort of matching device is required to transform this down close to 50 ohms.

One common matching device is a simple toroid based transformer with a turns ratio of 3:1 which transforms the 450 ohm feedpoint impedance down to 50 ohms. And these transformers are sometimes quite lossy, although with careful selection of toroid geometry, winding pattern, and toroid mix, it’s possible to get a transformer which will give you 85-95% efficiency, or a loss of 0.2-0.7db. No one wants to turn power into heat, but it’s worth looking at this transformer loss relative to other losses in the entire transmitter/antenna system.

Coax losses

So let’s look at the losses we’re likely to see in our feedline, the likely cause of loss in most antenna systems.

We’ll use the 20 meter band for comparison. Comparisons for other frequencies are left as a trivial exercise for the reader.

Loss for 100 feet of RG8x is about 1.1 dB. Loss for 100 feet of RG-174 is about 3.0 dB.

So if you have 25 feet of feedline, the difference between using RG8x and RG-174 is .25*(3.0-1.1)= .475 dB, roughly the same scale as the losses in a well built 9:1 transformer.

More to the point, the feedpoint for a dipole is probably up pretty high in the air, and you need extra feedline to reach it - feedline you probably don’t need to reach the feedpoint at the end of a vertical or sloper configured non-resonant antenna.

So the question becomes “how much extra coax would match the loss of the end-fed transformer?”

If we figure 0.7db loss in the transformer, we’d need 63 feet of RG8x, or 23 feet of RG-174. So if you’re working in the field, using RG-174 or rg-316, the losses of the extra feedline to reach the feedpoint of an elevated resonant dipole will be on the same scale as the losses of a 9:1 transformer for an end-fed non-resonant antenna.