Antenna Tuners for Compromise Antennas: What a 1:1 Match Does Not Tell You
What a tuner actually fixes on a restricted-space antenna: where to mount it, how much power the match really costs, and which units suit an indoor or stealth wire.
This article contains affiliate links. I may earn a small commission at no extra cost to you.
A tuner does not tune your antenna. It lies to your radio about what the antenna looks like, and it does that by putting a matching network between the two. The antenna radiates exactly what it radiated before you bought the tuner.
That distinction matters more in a restricted space than anywhere else, because compromise antennas run far off resonance most of the time and the tuner is doing heavy work on every band. Where you put it and how hard you make it work decide whether you are giving up a tenth of a dB or half your transmit power. The SWR meter reads 1.0 either way.
The match happens where the tuner is, and nowhere past it
This is the single most useful thing to understand about tuners, and it explains most of the disappointment people report.
A tuner in the shack presents 50 ohms to the transmitter looking forward. Looking backward from the antenna, nothing changed. The coax between the tuner and the antenna still carries the full mismatch, still has standing waves on it, and still burns the extra loss that mismatch causes. All the tuner did was stop the transmitter from folding back its output power.
A tuner at the feed point removes that problem entirely, because there is no coax between the match and the antenna. The coax from the shack to the tuner runs at 1:1 and behaves like plain coax.
So the question is not "do I need a tuner." It is "how much does my feedline lose while mismatched," and that has a number.
How much the mismatch actually costs on your coax
Loss under mismatch is the matched loss of the run plus an additional term that depends on SWR. Approximate additional loss, in dB, on top of the matched loss:
| Matched loss of the run | 3:1 SWR | 5:1 SWR | 10:1 SWR | 20:1 SWR |
|---|---|---|---|---|
| 0.5 dB | +0.2 | +0.5 | +1.0 | +1.9 |
| 1.0 dB | +0.4 | +0.9 | +1.8 | +3.2 |
| 2.0 dB | +0.8 | +1.7 | +3.3 | +5.5 |
Now put your own run against it. Fifty feet of RG-58 has roughly 0.9 dB of matched loss at 14 MHz and about 1.3 dB at 28 MHz. Fifty feet of RG-8X is closer to 0.7 and 1.0. Fifty feet of LMR-400 is about 0.35 and 0.5.
Read the table with those numbers and the picture resolves fast. A short run of decent coax at 5:1 costs you well under a dB, which is nothing you will hear. The same run at 15:1 on 10 meters, which is entirely normal for a 29-foot random wire, costs several dB and is worth solving. A long run of RG-58 to an attic wire at high SWR is the case where a shack tuner quietly eats most of your signal while showing you a perfect match.
The practical rule I use: if the mismatch on the feedline is under about 4:1, a shack tuner is fine and a remote tuner is not worth the money or the weatherproofing. Above roughly 8:1, put the match at the antenna. Between those, run the numbers against your actual cable and frequency.
Feedline choice interacts with all of this, and the run itself is usually the constrained part of a stealth install. That is covered separately in the coax routing article.
The loss inside the tuner is real and invisible
The other place power disappears is the tuner itself, and no meter on the front panel will show it to you.
A matching network shifts impedance using an inductor and capacitors. The inductor has finite Q. When the load is close to 50 ohms the network barely works and the loss is negligible. When the load is 5 ohms or 4000 ohms, the network has to circulate large currents or support large voltages inside itself to bridge that gap, and the resistive losses in the coil scale with it. A tuner asked to match a genuinely awful load can dissipate 20 to 50 percent of your power as heat in the inductor. That is 1 to 3 dB, gone, with a 1.0 SWR on the display.
There is a free diagnostic for it. Transmit a steady carrier at your normal power for thirty seconds, then put your hand on the tuner case near the inductor. Warm means you are paying. Hot means you are paying a lot, and you should change the antenna length or the band plan rather than buy a bigger tuner.
Voltage is the other consequence. A high-impedance feed point puts real voltage across the tuner's capacitors. At 100 watts into 3000 ohms the RMS voltage is about 550 volts, with peaks near 800. That is where T-network tuners arc, and it is why a wire cut close to a half-wave multiple on your favorite band is a bad idea even though it is resonant on paper.
Your rig's internal tuner is narrower than you think
Most 100-watt transceivers with a built-in tuner match roughly 3:1, sometimes a little more on some bands. That covers a resonant dipole drifting off band edge. It does not cover a random wire, a shortened loaded vertical away from its design frequency, or an attic doublet on a band it was not cut for.
If you bought an IC-7300 or an FT-891 expecting the internal tuner to handle a stealth wire, it will not, and that is a specification limit rather than a fault. You need an external unit. QRP rigs are the same story at lower power.
Ladder line is the honest exception
Open-wire and window line have very low matched loss, on the order of 0.1 dB per hundred feet at HF. Plug that into the table above and even a 20:1 mismatch costs a fraction of a dB. This is why the classic doublet fed with ladder line into a balanced tuner is genuinely efficient on every band despite wild SWR.
It is also why it rarely survives contact with a small space. Ladder line needs several inches of clearance from metal, drywall screws, gutters, and anything conductive, and its balance degrades badly when routed through a window frame or taped along a wall. In an apartment or a tight attic you usually cannot give it the clearance it needs, and an unbalanced ladder line feed pours common-mode current into the shack.
If you have a clear vertical drop and a way to keep the line in free air, a doublet with a balanced tuner is the best-performing option on this page. If you are threading a feedline through a window sash, it is not a candidate.
The random wire case, done correctly
The most common restricted-space setup is a single wire, an unun at the base, and a tuner. Three details determine whether it works.
Pick a length that is not near a half-wave multiple. A wire close to an electrical half wave on your band presents thousands of ohms, which is where tuner loss and arcing live. Lengths that avoid half-wave resonance across 40 through 10 meters include roughly 29, 35.5, and 41 feet. Those numbers are not magic. They are chosen to sit away from the high-impedance points on every band at once.
Use a 9:1 unun, not a 4:1 or a 1:1. A non-resonant end-fed wire typically presents several hundred to a couple thousand ohms. A 9:1 unun drops that into a range most tuners handle without straining, which is the whole point of the exercise.
Give it a counterpoise. An end-fed wire is half an antenna. The other half is whatever the unun's ground side can find, and if you do not provide it, that path is your coax shield, your desk, and your mains wiring. One wire about a quarter wave long on your lowest band, or several shorter ones, gives the current a defined return. This is the same failure mode covered in the RF grounding and counterpoise article, and it is the number one reason a random wire installation feels hot in the shack.
What to buy
| Unit | Power | Type | Mounts | Best fit |
|---|---|---|---|---|
| Elecraft T1 | 0.5 to 20 W | Auto L-network | At the rig, pocket size | QRP, portable, apartment operating |
| LDG Z-100Plus | 0.1 to 125 W | Auto switched-L | At the rig | 100 W station on a coax-fed wire |
| LDG RT-100 with RC-100 | 100 W | Auto, remote head | Outdoors at feed point | Long feedline to a stealth wire, any rig |
| Icom AH-4 | 120 W | Auto, rig-controlled | Outdoors at feed point | Icom owners running a wire, cleanest integration |
| CG Antenna CG-3000 | 200 W | Auto, weatherproof | Outdoors at feed point | Non-Icom rig needing a feed-point match |
| Palstar AT2K | 2 kW | Manual roller inductor | At the rig | Efficiency-first shack tuner, high Q coil |
| Emtech ZM-2 | 15 W | Manual balanced Z-match | At the rig | Ladder-line doublet at QRP |
Notes worth having before you spend money.
The Elecraft T1 runs on an internal 9V battery and weighs under six ounces, which makes it the default for anyone operating from a balcony or a table rather than a permanent shack. It will not handle 100 watts, so do not try.
The LDG Z-100Plus covers 1.8 through 54 MHz and stores a couple thousand match memories, which matters because a compromise antenna gets retuned constantly as you move around a band. Manual tuners are more efficient at their best settings and considerably more annoying when you retune six times an hour.
Remote units are the real upgrade for stealth installs. The Icom AH-4 is controlled over a four-wire cable from a compatible Icom rig and matches a 7-meter or longer wire on 40 through 6 meters, or a 2.5-meter wire on 7 MHz and up. The CG Antenna CG-3000 does the same job for any transceiver at up to 200 watts and senses RF to trigger a tune rather than needing a control line. Both need weatherproofing and a counterpoise at the mount point, and both remove the feedline-loss problem entirely.
MFJ tuners still dominate the used market and are worth buying secondhand, but MFJ ceased on-site production in May 2024, so treat anything listed as new with suspicion. The same caution applies to the rest of the used gear market, covered in the used HF rig buying guide. DX Engineering carries current stock from the manufacturers still shipping.
What I would do
Measure before you buy. Put an analyzer at the shack end of your existing feedline and write down the SWR on each band you care about. Then compute the matched loss of your actual run at those frequencies and read the table above.
If everything is under 4:1, buy an inexpensive auto tuner for the rig and spend the difference on the antenna or on lowering your noise floor, which is usually where the real gain is hiding in a small space. That is covered in the RFI hunting article.
If you are seeing 10:1 or worse on more than one band, which is normal for a random wire, put an auto tuner at the feed point and run coax to it. You will get back two to four dB that a shack tuner was burning on the feedline, and you will stop heating the coil in your shack.
Either way, add a counterpoise and a 1:1 current choke at the feed point before you conclude the tuner is the problem. A tuner that cannot find a match on a wire that should be easy is nearly always looking at an antenna with no defined return path, not at a tuner that is too small.