The antenna and feeder writing on this site, collected in one place. It is the most useful radio work I have done, because it is the part of a station that changes what you can hear rather than how loud you can say it.

The setting matters. The station is at Marybank on the Isle of Lewis, 50 metres above sea level with open Atlantic to the west. That means salt in everything, a prevailing wind that will find any weakness in a mast, and a horizon so low that a mediocre antenna can still work very well — particularly to the west, where there is nothing but water between here and Newfoundland.

Start here: the feeder is not the boring bit

Almost every “my station is deaf” problem I have met has been a feeder problem, not an antenna problem. Loss in coax is not a small tax on your signal — on 2 metres it is the difference between a working station and a decorative one, and the losses add up on both transmit and receive. Why feeder matters is the place to begin, and the cable coax loss table gives the per-metre figures for the common types so you can work out what your own run is costing you before you spend anything on an antenna.

Antenna first, amplifier second is the companion argument: put the effort and the money into the radiator and the feed line, because an amplifier at the wrong end of a lossy feeder amplifies the loss.

Baluns, ununs and matching

If you are feeding a wire antenna, some form of transformer is usually doing quiet, essential work. Baluns and ununs explains the difference, and there are build guides at each of the useful ratios:

Wire antennas

Wire is where I would start, and where I keep coming back to. Three builds on the site cover most of what a single-operator station needs:

NVIS deserves its own explanation, because it is the one HF technique that genuinely suits a station surrounded by water and hills. The NVIS antenna covers why a low horizontal radiator can reach the next few hundred miles when nothing else will.

Arrays, stacking and the pursuit of gain

Once a single antenna works, the temptation is to add more. Antenna stacking and baying covers what you gain and what it costs in complexity, and four square arrays for DXing is the more serious treatment for anyone chasing low-angle radiation. Both are worth reading before you buy a second mast.

The theory that changes what you do

Antenna work only makes sense against propagation, because the band conditions decide whether your best effort is heard at all. Propagation in amateur radio is the long piece — HF to microwave, day and night, summer and winter — and the ionosphere explains the layer doing most of the work and what would happen if it were not there. Understanding sporadic-E covers the mode that produces surprising 6 m and 2 m openings with no warning at all.

Measuring instead of guessing

Two instruments will save you more time than any amount of forum reading. A multimeter finds the broken connection, the shorted plug and the dead battery, and it is the first thing to reach for when something stops working. A spectrum analyser tells you what is actually on the air rather than what you assume: the tinySA review covers the cheap one I use. For antennas specifically, the NanoVNA review covers the £50 vector analyser that tells you whether an antenna is actually resonant — before you start blaming the feeder.

Before you climb anything

The conditions here are unforgiving in a way that a sheltered inland site never is. Wind loading on a wet mast is the calculation people skip, and a south-westerly at 60 mph will remove an antenna that has survived two calm years. Guys and fixings that look generous on the mainland look marginal on a Hebridean roof. If the mast is going on a house, and the house is older than the antenna, get someone else to check the fixing points.

If you are starting from nothing and want an order to read this in: feeder, then the dipole, then propagation. That is enough to build a station that works, and everything else on this page is refinement.