The balun

The word ‘balun’ comes from ‘balanced to unbalanced’. These two images show a half-wave dipole broken at the centre and connected to a feeder cable. This, with many variations, is at the heart of the yagi aerial, and also of course the half-wave dipole aerial, as used for FM and DAB. In the upper image the dipole is connected directly to the cable, and in the lower one it’s connected via a balun. The balun is represented here by a simple transformer, but there are many different ways to make a balun, not all of them using a transformer.

The output from the dipole — the voltage across its terminals caused because the dipole is in an RF field — is balanced, meaning that it is exactly equal on both sides, and ‘floats’ relative to ground. If the terminals are connected directly to the coax inner and screen, the energy transfer to the feeder will be inefficient because the screen will partially radiate the signal. Down at the bottom end of the cable the available signal — the voltage between inner and screen — will be weaker than it should be.

The bottom end of the downlead’s screen is likely to be grounded, directly or indirectly, so the top end of the screen is also held at or near ground potential. If it’s connected directly to one half of the dipole the potential on that half can’t vary, or at least the downlead tries to prevent it. The dipole, which for its efficient function has to have balanced currents and voltages on its two halves, has now been rather forcibly unbalanced. This will affect the sensitivity and directional pattern of the whole aerial. This is why balun-less aerials are not so good at rejecting off-axis interference. And if you’ve ever pointed a balun-less aerial at a transmission mast that you can see and wondered why the front lobe of the aerial seems to be slightly to one side of its physical alignment, well now you know.

Another way of explaining this is to ask what happens if you put a finger on the dipole. The signal output drops of course. You are unbalancing the currents in the dipole so it can’t work properly.

The cable, at least where it’s near the aerial, is in the field created by the transmitter, so the screen will pick up signal directly. Some of this signal will pass up to the dipole where it will combine with the signal the dipole has picked up directly. The phase relationship between the two versions of the signal will be random, so they could combine constructively or destructively. In other words they might add together, or one might subtract from the other. This means the gain of the aerial (as measured at the bottom of the downlead) will have a random element that will vary with frequency.

Because signal picked up on the cable can come from any or every direction, if it reaches the dipole the directivity of the installation will be compromised. It will seem as if the aerial isn’t properly directional. A polar response pattern will clearly show the random pick-up superimposed on the aerial’s theoretical pattern, and if the cable is moved the random element will change. What’s more, any interference picked up on the screen of the cable further down — from adjacent mains wiring for instance — could travel up to the dipole and combine with the signal. It’s all very unsatisfactory and makes correct aerial alignment a bit hit and miss.

Here’s a practical example. In the 1970s (when I were a lad) I used to install TV aerials for rental companies. All that mattered was price. We used to do all sorts of ridiculous things to shave a few pennies off the cost of each job. Of course we used the cheapest possible aerials; flimsy little contract quality things that cost about 80p. There was no balun of course. In those analogue days ghosting was the big problem in many areas. I used to have a little TV set with a 6″ screen that I could take onto the roof, and I soon found that I could make the ghosting and the signal level come and go simply by moving the cable a few inches.

What all this amounts to is that without a balun the performance of a yagi has a lot of randomness and inefficiency thrown in. Gain and immunity to interference are compromised.

The balun shown in the lower image is a simple transformer. That sort of balun will usually have a tiny Polo-mint-shaped ferrite core with the two windings wrapped through it. There are many different balun designs though.

A balun can include impedance matching as well as balance/unbalance conversion, and most do. These convert from the dipole’s characteristic impedance to the 75Ω impedance of the coax cable. In the case of a transformer balun this is done by adjusting the number of turns on the windings. Normally though the aerial designer will get the dipole’s impedance fairly close to 75Ω anyway, usually by folding the dipole (which is necessary anyway to increase the bandwidth), and then build a balun that gives an almost 1:1 match and provides an exact match between dipole and feeder.

A dipole aerial for DAB will not have a broad enough response to cover the whole DAB band. If whole band coverage is needed the answer is to fold the dipole, because that increases its bandwidth. But a folded dipole with no parasitic elements has a characteristic impedance of about 300Ω, so a balun with 4:1 impedance conversion is used. There’s a picture of non-folded and folded DAB dipoles in the DAB section Basic TV and radio aerial guide.

Some UHF TV aerials have full wave dipoles and because these have very high impedance at the centre the balun has to include something like 5:1 matching.