Last Modified: January 25, 2012
Contents: Basics; Matching Input Impedance; Reactance vs. SWR; Inductive Matching; UNUN Matching; Capacitive Matching; Odds & Ends;
Unadulterated fact: If your HF installation doesn't require any matching (input SWR under 1.6:1), you either need a better antenna, or a better mounting scheme, or both!
If the aforementioned is the case, you should read these articles; Antennas, Commercial, Antenna Mounts, Antenna Efficiency, and Grounds, RF & DC.
Matching a mobile antenna to the requisite 50 ohms is a requirement for several reasons. For example, modern solid state radios are designed to reduce their output power when the input SWR reaches ≈2:1. Some will handle a little more, some a little less. The SWR doesn't have to be flat, so anything below 1.6:1 is close enough.
One very important point needs to be mentioned before proceeding. If you're using a remotely controlled HF mobile antenna like a Scorpion, the motor leads (and reed switch leads if used), must be properly choked. If they're not, you'll have terminal problems (that's no pun!). Do yourself a favor, and read this article: Antenna Controllers.
If you don't use the reed switch, here's a hint or two. Attach a lug to the leads and connect them to the antenna's mast. This saves you from having to choke them off with a ferrite core. What ever you do, don't ground them! And don't let them float (hanging in the wind). Doing otherwise will affect the input impedance of the antenna, making it nearly impossible to match.
Further, improper choking of the motor leads will also affect the impedance of your antenna. Therefore, the motor leads should be disconnected at the antenna before attempting to adjust any matching method. Once matched, if reconnecting the motor leads changes the matching point and/or SWR (no matter how small the change is), it is a good indication that the motor lead choke impedance is too low. Again, read this article: Antenna Controllers. As pointed out in the article, the chokes supplied by every antenna manufacturer, with two exceptions, are totally inadequate for the purpose.
By the way, antenna manufacturers often tell their customers to cut their coax feed lines to a specific length in order to get a good match. All this does is mask the problem, by moving the SWR node to a different position along the feed line. While this may appear to fix the problem, it doesn't fool most automatic controllers. The truth is, if the antenna is properly matched, it doesn't make any difference how long the feed line is.
In the following sections, it is necessary to know the exact resonant point (X=Ø) of the antenna we're trying to match. This fact alone, should not infer that exact resonance is a requirement; it isn't! Rather, in this case, it is only a means of arriving at the end point (i.e.: match). Once the matching is complete, whatever the SWR is (assuming it is under ≈1.6:1) is irrelevant. It should also be mentioned, that some transceivers can generate excessive IMD levels the FCC mandated ones when the SWR is over 1.8:1 or so—yet another reason to properly match antennas.
Properly matching the antenna impedance to the line impedance is important, especially if you're using an automatic antenna controller. And, to repeat, if your HF antenna doesn't require matching to provide a low SWR (with the possible exception of 10 and 12 meters), then you need a better antenna, a better mounting position, or both! Further, proper matching reduces the chances of RF flowing on the outside of the coax feed line (common mode currents).
Literally thousands of articles have been written about antenna impedance matching. Whatever method you choose to achieve a match is fine, as long as the antenna element is DC grounded! There are two reasons for this.
First, DC grounding helps control static discharges from the antenna, thus reducing some of the received hash we all put up with. Secondly it is a safety issue. If the antenna were to come into contact with a live overhead power line, DC grounding will help prevent damage to your equipment and perhaps to you as well. There are several ways to accomplish DC grounding, and an RF match at the same time. These are discussed below.
Some neophyte amateurs are under the assumption that a DC grounded antenna (element) won't work, but this is not the case. Just because we DC ground the antenna element, doesn't mean it is RF grounded too. They also assume that grounding the antenna's mounting structure will assure a low SWR and/or is a replacement for an adequate ground plane. Neither of these assumptions are true.
There is another related issue which needs to be mentioned at this point, and that is lightning safety. Contrary to popular belief, inductive shunt matching, does not increase the likelihood of lightning damage, should a strike occur. If anything, it reduces the possibility.
Due in part to popular press, way too much emphasis is placed on achieving a low SWR. Adding insult to injury, the methodology most amateurs use to check (or set) their SWR is incorrect. A fact which will soon become glaringly evident.
At resonance, the input impedance of a decent-quality, correctly-mounted, HF mobile antenna will be about 25 ohms. By definition, the resonance point is where the reactive component equals zero (X=Ø, or +Øj if you prefer). Since the requisite impedance of our feed line is 50 ohms, the resulting SWR would be 2:1, as read at the radio end of our coax run. However, if you adjust the antenna to a frequency lower than the true resonant point, the indicated SWR will decrease, perhaps to 1.5:1. For this reason, you should use an antenna analyzer with a reactance readout, when adjusting any antenna matching coil.
Again, you should look for the lowest reactance (X=Ø), not the lowest SWR when adjusting any matching device. Once the matching device is properly adjusted, the minimum SWR point on your transceiver (or external SWR meter) will be very close to the actual resonance point of the antenna.
Just to make sure this point is as clear as possible... With respect to the input impedance of an antenna (mobile or otherwise) that is other than 50 ohms, when the frequency if moved away from the true resonance point, the resistive component increases faster than the reactive component. In other words, the apparent SWR decreases, however, the antenna is no longer in exact resonance!
You can demonstrate this for yourself by adjusting your antenna analyzer to the lowest reactance (X=Ø, or as close as you can get to it), and noting the SWR. Then, adjust the analyzer's frequency until the SWR is at its lowest, and note the reactance. It will mimic the chart shown upper left (the reactance is shown in red, and the SWR in blue).
Again, to make sure this point is as clear as possible... When we're dealing with an antenna whose input impedance is other than 50 ohms, the lowest SWR is not the resonant point! Since the primary use of an antenna analyzer in a mobile scenario is to adjust the requisite matching coil, knowing this fact becomes imperative.
The two photos (a MFJ-259B) show a 40 meter resonant antenna before (left), and after (right) proper matching. It should be noted the right photo shows a few ohms of reactance, due to inaccuracies of the meter.
The reactance readout on an antenna analyzer may not be exactly zero at its lowest obtainable setting. This is due to several factors, not the least of which is the basic accuracy of the instrument in question. It may also be due to a nearby broadcast transmitter, and here's how to check. With the MFJ-259B connected to your antenna, push the mode switch until the frequency counter displays. If the SWR meter significantly deflects, you probably have BCI. MFJ does sell an optional BCI filter unit for the 259B which eliminates the problem.
Make note of the frequency shown on the 259B. This is indicative of what you'll see while you're in the process of of adjusting an antenna matching coil. It is included here, because (as noted above) adjusting a remotely tuned HF antenna's matching coil is the prime use for an antenna analyzer in a mobile scenario.
Inductive matching does DC ground the antenna. If you're planning on using a remotely tuned antenna like a Scorpion or one of the many other remotely controlled designs, and an automatic antenna controller like the Better RF unit, then inductive matching is your only choice if you're seeking, true, fully automatic operation.
Several important caveats needs to be inserted here. Inductive matching works by borrowing a small amount of capacitive reactance from the antenna (by tuning the antenna slightly above the actual transmitting frequency). This borrowed capacitance, and the shunt matching coil's inductance, form a highpass, LC network which transforms the antenna's low impedance (typically 25 ohms or so) to that of the 50 ohm feed line. Installed and adjusted properly, shunt matching will provide a decent match (<1.6:1) over several octaves. Enclosing the coil, even in plastic, will affect the frequency versus reactance of the coil, effectively reducing its bandwidth. Further, the coil must be as clear of surrounding metal as possible. For example, factory supplied shunt coils are often mounted against the antenna's mounting bracket. For best results, these coils should be relocated. You should also avoid any commercial units which surround the mast. Lastly, some commercial matching coils short out a portion of the coil, to achieve a match. As with antenna loading coils, short-tapping reduces the effective Q of the coil, which further increases matching losses. Obviously then, open air, shunt matching coils provide the best match, and least loss of any other matching methodology.
At left is a photo of the MFJ-908 L match unit. Building one rather than buying one won't save you any money, but you might learn how they work if you do. The ARRL Antenna Handbook is a good place to start.
However, you typically don't need a switched inductor like the aforementioned MFJ unit (see Odds & Ends below). Instead a simple inductor, like the one shown in the right photo (or the upper right pictorial), will suffice. One end is attached to the antenna feed, and the other end is connected to ground. The ground end of the coil should be collocated with the coax shield ground.
The coil at right has 9 turns, is 1 inch inside diameter, and wound with #14 Thermalese® (enameled) wire. The coil's form factor should be kept close to 1:1 (length to diameter). Long skinny coils do not work nearly as well. You can also use building wire, but it is a little harder to work with. In actual use, the turns are spaced a little further apart to adjust the inductance. The coil needs to be about 1 uH, but in the real world, the value may be between .5 uH and 1.5 uH depending on the actual input impedance at resonance.
Several commercial versions of the screwdriver antenna have machined-in matching coils, which is fixed in value, and therefore cannot provide an ideal match over the entire resonant frequency span of the antenna. In fact, it is often suggested that a specific length of coax be used to feed the antenna, in order to provide a low SWR. Doing so only masks the real problem; an incorrectly sized (impedance value) shunt coil. Although these antennas can be modified to use an adjustable coil, it's best to avoid the buying the problem in the first place.
There is a specific procedure which must be followed if proper adjustment of a shunt matching coil is to be achieved. To alleviate problems, I have moved (and expanded) the procedure to this article, Antenna Coil Adjustment.
You can also use an UNUN (UNbalanced to UNbalanced) RF transformer. Like the LC network above, it provides a DC ground, and the requisite impedance match. The overall system losses are low, so the UNUN can be mounted near the radio rather than at the antenna. An MFJ-907 unit shown at left.
Keep in mind, if you use a remotely controlled antenna, you will have to change taps between bands. In most cases, you can use one tap for 80 and 40, another for 20 and 17, and straight through for 12 and 10. If you don't like the idea of changing taps, then make one of the aforementioned inductors.
If you want to built your own UNUN, the schematic is at right (click to enlarge). An F114-67 ferrite core, a few feet of #14 enameled copper wire (enough for 9, bifilar turns), a small box to mount it in, and you're home free. I prefer to cover the core with 3M #27 glass tape as it makes winding easier, but isn't required.
I don't recommend capacitive matching for two reasons. First, the method doesn't DC ground the antenna. Secondly, as the frequency increases, the reactance (in ohms) decreases, which means you have to use a different value capacitor for each band, and sometimes within a band.
If you're already using capacitive matching and don't want to bother changing it, at least add a 10K resistor across the antenna terminals (high power will require several resistors in series). While this will not protect you from live overhead wires, it will help tame the static discharges. A correctly sized RF choke will to the same.
By the way, MFJ offers two sizes of capacitive match boxes; one rated at 300 watts, the other one 600 watts. Both models are stressed at half their power rating.
Antennas with extended coverage down to 160 meters, may require an MFJ-908 L or similar switched inductor. The reason is, a 160 meter loading coil is about 5 times larger (in inductance) than an 80 meter loading coil. All else being equal, the coil losses will more than double, and therefore the input impedance will be close to 50 ohms (no matching needed). There's a hidden factor at work here too. A 160 through 10 meter remotely controlled antenna, will exhibit significantly more coil loss at any frequency, than an 80 through 10 meter antenna. Potential purchasers should be aware of this fact.
As an alternative, you can make your own switched shunt inductor. The one pictured at right is from Myron Schaffer, WVØΗ. The coil is similar to the one described above. The switch shorts out the bottom 4 turns, reducing the inductance from 2 uH, to .7 uH. This method costs less, and is just as effective as a purchased one, albeit requiring a bit of tweaking to get the correct inductances.
I cover this in detail in my coupler article, but it needs repeating here. Either an internal or external auto coupler may be used to match a mobile antenna's input impedance to 50 ohms. And, they can be used to extend the bandwidth of a monoband antenna. However, using one with a remotely tuned antenna presents some operational problems. If you use one, keep the following in mind.
The antenna in question should be adjusted to resonance (lowest SWR is close enough) before the auto coupler is turned on. This is especially important with external couplers with their greater matching range. Under the right circumstances, failing to tune the antenna close to the operating frequency can cause the RF voltage to be high enough to arc over most base insulators, and might even exceed the ratings of the coax between the coupler and the antenna.
As mentioned above, another good reason to DC ground a mobile antenna is for lightning protection. It won't necessarily protect your radio from damage, but does offer a level of personal protection. Don't laugh, I've been hit several times! The last time was May 25, 2007. The corona ball now has a new hole, the the whip nearly burned in two, and the matching coil was damaged. The radio was on at the time, and suffered no damage.