tuning an antenna with an antenna analyzer
last updated 22 December 2024.
Antenna analyzers (and other similar devices) have become much common over the last 20 years or so. They provide a lot more data than just SWR, and I’ve seen several different methods suggested for how to use them to tune an antenna.
Let’s assume you are tuning a simple dipole or inverted vee. There is just one adjustment – the total wire length, although it is often easier to think of the length of a side (assuming it is fed in the middle).
The purpose of tuning the antenna is to get it as close as possible to design load impedance of the transmitter, so that it can deliver maximum power. We have a convenient measurement that tells us just what we need to know: SWR. (Specifically, SWR referenced to the design load impedance.) So, the best parameter to use is SWR. (If you don’t have a fancy antenna analyzer, you would use a similar process with a common SWR meter).
Let’s say that the dipole is for 40m. I’d start by plotting the SWR across a range that includes the desired frequency band, plus enough extra on either side to accommodate the expected accuracy of the initial wire length. (Remember, the traditional formula is only an estimate: the actual length depends on many other factors.) So in this case, I’d probably plot the SWR from 6 to 8 MHz. Here is where the analyzer is most useful: if I were using a transmitter and an SWR meter, I could only measure the SWR across the ham band. That often will give me an idea of whether the antenna needs to be longer or shorter, but will take a few more iterations to get to the same result.
Now look at the results on the display. You should see the SWR is high at each side of the display, and low somewhere near the middle. Then you adjust the wire length (equally at each end) to move the point of minimum SWR to your desired frequency: shortening the antenna shifts the curve higher in frequency, and lengthening it moves it lower in frequency.
I generally start with my wires a bit long, and leave a foot or two hanging down from where it is tied to the end insulator that I can snip back as needed. Folding the ends back (as shown in this video) will also work, and gives you the opportunity to recover if you shorten it too much. (Just be aware that the two methods end up with different wire lengths: that’s not a problem, but if you cut off the folded wire, the antenna may be too short.)
So, let’s say that my SWR curve is minimum about 6.5 MHz. According to my dipole length tables I need to shorten the antenna by about 14cm ( 5 1/2 inches ) to shift the resonant by 100 kHz, so, if my target is 7.100 MHz, that would be about 84cm ( 33 inches ) at each end.
But I’m not going to do that all at once, especially for an inverted vee with the ends near the ground. That’s because the height of the ends above ground is one of the factors that affects the resonant frequency, and, with the dipole end ropes tied off to the same point, shortening the wire will change the spacing to ground. So I might start with 40-50 cm ( 15-20 ) inches at each end to start with. (The simple solution to cutting the same amount off of each end is to measure or estimate the desired amount at one end, cut it off, and use that cut-off piece to measure how much to cut off the other end. Unless you are folding back the ends, of course.)
Then repeat your SWR sweep across frequency. As the antenna gets closer to the desired frequency, you can use a narrower “range” setting to get better resolution. A couple tries, and you should get close enough. Note that you don’t need to be precise: try to get as much of the band as you can below 1.5 : 1 or so. Sometimes plotting the SWR across the whole band gives you such a flat plot that it is difficult to tell where the minimum is. Don’t worry – if the SWR is low across the band, the antenna doesn’t need any further adjustment.
What if the SWR is still too high at the lowest point? Sometimes that happens, because the feedpoint impedance depends on the height above ground, the slope of the wires in an inverted vee, and other factors that changing the wire length doesn’t correct for. Changing the wire length affects the reactance at the feedpoint, but doesn’t change the resistance significantly. In the case of a dipole or inverted vee, try varying the height of the center and/or the ends and see if that helps.
Some other types of antennas do provide additional adjustments that can be used to improve the SWR at the minimum point of the curve. For example, with a full-wave loop antenna, changing the height-to-width ratio can change the feedpoint impedance over a useful range: https://practicalantennas.com/theory/loop/full-wave/, to account for the feedpoint resistance at a particular height above ground.
what about all those other numbers on the display?
So, if SWR is the best way to tune the antenna, why are there all those other numbers on the display? They come in handy for other types of measurements, or sometimes for trouble-shooting. Let’s look at the values often provided on various analyzers.
R and X
are the resistance and reactance components of the complex impedance that the device is measuring. The impedance is R + jX where “j” is equal to the square root of -1 (or “i” in mathematics, but engineers use “j” to avoid confusion with “i” for current). Both are in ohms, but you can’t add one to the other to get a single number, because impedance is a complex value. When X = 0 at the feedpoint of an antenna, we call that “resonant”. If R = 50 at the same time, then you get a perfect match (well, close enough, anyway. The characteristic impedance of practical coax cable isn’t exactly 50 + j0 ohms.)
You will often see recommendations to adjust your antenna to get X = 0. That sounds good, but that can lead you astray, unless you are measuring right at the antenna feedpoint. The reason is that, then the SWR is not 1 : 1, the impedance (both R and X) gets transformed along the coax cable. So the setting that gives X = 0 for one coax length won’t be the same as that for a different length. (There are ways to work around this.) Even when the antenna is resonant (X = 0 at the feedpoint), it may register positive or negative reactance some distance down the cable. What we really need in this case is a measurement that tells you how far from 50 + j0 ohms the measured impedance actually is: we call that measurement “SWR”.
C and L
are the value of Capacitance or Inductance that has the same reactance as the impedance being measured. This is handy if you want to measure an unknown coil or capacitor (especially the setting of a variable capacitor, in order to replace it with a fixed capacitor once you have found the right setting). Except for that specific application, I’d use the X value instead. For example, if you want to calculate the size of base-loading coil that you need to resonate a short vertical antenna, you can measure the impedance (right at the feedpoint, not through a random length of cable). In that case, if X = -300 ohms, then you need a coil with a reactance of +300 ohms at the operating frequency to tune it to resonance. It is much easier to use X in this case than C or L.
|Z|, or sometimes just Z, as a single value
If there was a way to remove this from the display, that’s the first thing I would do. IGNORE IT.
This is the magnitude of impedance, not the actual impedance, and is rarely useful in RF work. The problem is that too many people assume that |Z| = 50 ohms means the SWR is 1 : 1, but in practice it could also be infinite SWR, or anywhere in between. This causes more trouble than it is worth for many hams.
RL
Return Loss is the measurement that professional engineers tend to use instead of SWR. it’s like measuring temperature in Fahrenheit or Centigrade: the numbers are different, but they exactly correspond to each other. Return Loss gives better precision when dealing with low values of SWR, and SWR gives better precision at higher values. A return loss of 0 dB is an infinite SWR. Lower SWR will correspond to larger values of Return Loss (which should be positive, although that is a common mistake many people, including some manufacturers, make).
Phase
Technically |Z| (the magnitude of impedance) and the Phase Angle represent the impedance in a polar coordinate system, while R and X represent it in rectangular coordinates. There may be a rare occasion in antenna engineering when a particular problem is easier to solve in polar coordinates, but I haven’t encountered one yet, especially since this is centered on the origin rather than the characteristic impedance of the feedline.
I generally use SWR for tuning antennas, and R and X for various other calculations, like calculating matching networks. And mostly ignore the others.