You put up an antenna, connect the feed line, and check the SWR. The needle climbs. Three to one. Five to one. Maybe higher.
“Well, that antenna isn’t going to work!”
Hold on—don’t take it down just yet.
An excellent radiator can have a high SWR. That reading tells you how well the antenna’s impedance matches the feed line. By itself, it doesn’t tell you how efficiently the antenna turns accepted power into a signal.
A common mistake among new hams is treating a high SWR reading as proof of a poor antenna. It’s an easy assumption to make. A low reading is reassuring, and a high reading deserves attention. But SWR is only part of the story. The feed line, the tuner, and the antenna’s own losses determine how much of your transmitter’s power actually reaches the air.
So what happens when a 500 watt radio feeds an antenna with five to one SWR—and you still make great contacts? And how could a meter on that feed line show 900 watts forward?
Let’s follow the power and see where it goes.
The station we are explaining
Our example uses a radio delivering 500 watts into a correctly adjusted shack tuner, approximately 100 feet of Messi & Paoloni Hyperflex 10 Sahara coax, and operation near 21 MHz on the 15 meter band. We assume 5:1 SWR at the antenna connection. To focus on the feed line, we initially neglect tuner loss, antenna loss, connectors, and common mode effects.
The 5:1 value is a case assumption, not a measured specification for a particular antenna. A multiband wire antenna can present a difficult impedance on a particular band and still be a useful radiator. The actual impedance, radiation pattern, and losses depend on its design and installation.
A match at the radio and a mismatch on the coax
The tuner uses inductors and capacitors to transform the impedance presented by the feed line into the load the radio requires. The radio can therefore see 1:1 SWR while the coax beyond the tuner still carries standing waves. A shack tuner does not change the antenna feed point impedance or eliminate that mismatch. [1, 2]
That is why the location of an SWR or wattmeter matters. Before the matched tuner, it sees the radio side of the system. After the tuner, it sees the forward and reflected waves on the mismatched feed line.
What a good contact establishes
A good contact shows that enough signal reached the other station under the conditions at that time. Antenna pattern, propagation, receiving conditions, and radiated power all play a part. A contact alone cannot establish system efficiency, but high SWR alone cannot establish that the antenna is poor either.
A perfect match can simply make heat
Here is a useful reality check: an ideal shielded 50 ohm dummy load presents a 1:1 match to 50 ohm coax. It accepts the transmitter’s power and deliberately dissipates it as heat. That makes it useful test equipment, not a useful transmitting antenna. Real dummy loads have finite bandwidth and small imperfections; the comparison is conceptual.
The lesson works both ways. A low SWR does not prove that a load radiates efficiently, and a high SWR does not prove that an antenna is a poor radiator. Matching, radiation efficiency, feed system losses, and radiation pattern answer different questions.
Supporting example ARRL QST dummy load review September 2021
Why a 500 watt radio can produce a 900 watt forward reading
Nine hundred watts on the meter from a 500 watt radio? That certainly gets your attention! The explanation starts with two directions of travel, not an extra source of power.
First imagine a lossless tuner and lossless coax. At 5:1 SWR, the reflected power is four ninths of forward power. A directional meter after the tuner would indicate 900 watts forward and 400 watts reflected. The antenna accepts the difference: 500 watts. These are simultaneous steady state wave powers. [3]
| Measurement location | Forward | Reflected | Net delivered |
|---|---|---|---|
| Before matched tuner | 500 W | 0 W | 500 W |
| After tuner on lossless coax | 900 W | 400 W | 500 W |
What is that 900 watts really? It is the power associated with the wave traveling toward the antenna. It is a real directional wave measurement, but it is not 900 watts of fresh transmitter output or 900 watts being radiated.
The forward wave includes the effect of energy participating in reflections. The returning wave is accounted for separately. Subtract what is coming back from what is heading out to obtain net power flow. The tuner has not become an amplifier.
The pattern stands still while energy keeps moving
The forward and reflected waves overlap everywhere along the coax. Their voltages reinforce at some positions and partly cancel at others. This produces fixed locations of voltage peaks and minima. At 5:1 SWR, the maximum voltage amplitude is five times the minimum. It is not a five to one ratio of forward watts to reflected watts. [3]
Reading the forward and reflected power meters
Here is what a directional wattmeter would show at two different locations in our ideal example. The radio supplies 500 watts into the matched tuner. The antenna has 5:1 SWR. For this picture only, the tuner and coax have no loss, and the instruments are ideal. [1, 3]
The 900 watt forward indication is the power associated with the wave heading toward the antenna at that point. The 400 watt reflected indication is the power associated with the wave returning. Both waves exist at the same time. Their difference is the 500 watts of net power flowing toward and accepted by the antenna.
The tuner has not created another 400 watts of transmitter output. Reflections contribute to the forward wave, and a directional meter reports that wave separately from the returning wave. Read the pair together: forward minus reflected equals net delivered power.
A cross needle meter illustration
The separate forward and reflected readings can share one meter face. This Palstar inspired illustration makes the two needles and their intersection easier to recognize. The numerical example remains 900 watts forward, 400 watts reflected, and 500 watts net delivered on an ideal lossless line with 5:1 SWR.
This depicts conceptual sensing AFTER the matching network. It is not a claim that a stock AT2K built in meter would show these readings while its radio side is matched to 1:1. At the matched radio side, the ideal readings remain 500 watts forward and zero reflected.
Visual reference Palstar AT2K product and cross needle meter
A real meter may combine these readings on a cross needle face or select them with a switch. Its sensor must be on the coax side of the tuner to show this mismatched line example. On the next page, cable loss changes the readings to approximately 764 watts forward and 264 watts reflected at the coax entrance.
Putting the actual cable into the example
The manufacturer lists Hyperflex 10 Sahara attenuation at 21 MHz as 1.8 dB per 100 meters at 20 degrees Celsius. One hundred feet is 30.48 meters, giving approximately 0.549 dB of matched cable loss. The catalog feet column rounds this to 0.5 dB. We use the meter based value for the calculation. [4]
With 500 watts supplied through an ideal matched tuner and 5:1 SWR at the antenna, the estimate becomes about 374 watts accepted by the antenna and 126 watts dissipated in the coax. Under our assumption of negligible antenna loss, approximately 374 watts is radiated.
| Measurement location | Forward | Reflected | Net toward antenna |
|---|---|---|---|
| Radio before tuner | 500 W | 0 W | 500 W |
| Coax entrance after tuner | 764 W | 264 W | 500 W |
| Antenna connection | 673 W | 299 W | 374 W |
Values are rounded. Both wave components attenuate as they travel. The 299 watt reflected wave at the antenna arrives back at the tuner as about 264 watts. Likewise, the 764 watt forward wave at the coax entrance arrives at the antenna as about 673 watts.
The forward meter tells us what is heading toward the antenna at that location. Forward minus reflected tells us the net power continuing through that location. At the antenna connection, that difference is the power the antenna accepts.
About three quarters of the radio output still reaches the antenna. That is useful transmitting power, although the cable loss is worth understanding. Real tuner loss and antenna loss would reduce the radiated result further.
Following the power through the coax
This drawing puts the real cable example in one place. The blue arrow follows the forward wave toward the antenna; the amber arrow follows the reflected wave back toward the tuner. The two waves are present at the same time. At each measurement location, subtract reflected power from forward power to find the net flow.
Three readings that tell different parts of the story
At the radio side, the ideal matched tuner gives 1:1 SWR and accepts the radio’s 500 watts. At the coax entrance, 764 watts forward minus 264 watts reflected gives the same 500 watts net. At the antenna, 673 watts forward minus 299 watts reflected leaves about 374 watts accepted.
The 126 watt difference heats the coax along its length. Of that loss, about 91 watts is the reduction of the forward wave from 764 to 673 watts, and about 35 watts is the reduction of the returning wave from 299 to 264 watts. These describe the directional power losses; they are not extra losses to subtract again.
The lossy cable also reduces the reflected wave reaching the shack, so SWR at the coax entrance is about 3.85:1 even though it is 5:1 at the antenna. The tuner then transforms the line input impedance so the radio side reads 1:1. A lower SWR reading caused by cable attenuation does not mean the antenna match improved.
The math behind the readings
These equations use the usual low loss coax model with a real nominal characteristic impedance, sinusoidal steady state, and a lossless tuner adjusted to match the radio. SWR is specified at the antenna. A lossy line shows a lower SWR at its shack end; here that is approximately 3.85:1 before matching. [3]
S = antenna end SWR = 5
|Gamma| = (S − 1) / (S + 1) = 4 / 6 = 2 / 3
r = reflected power / forward power = |Gamma|² = 4 / 9
L = matched one way cable loss = 1.8 × 30.48 / 100 = 0.54864 dB
a = one way power transmission factor = 10^(−L / 10) = 0.881325
P = net power entering coax = 500 W
Solving for the forward and reflected powers
Let F be forward power at the coax entrance. At the antenna, forward power is aF and reflected power is raF. On returning to the coax entrance, reflected power is ra²F. Net input is therefore P = F − ra²F.
F = P / (1 − ra²) = 763.609 W
Reflected at coax entrance = ra²F = 263.609 W
Forward at antenna = aF = 672.988 W
Reflected at antenna = raF = 299.106 W
Power accepted by antenna = aF(1 − r) = 373.882 W
Equivalently Pantenna = P × a(1 − r) / (1 − ra²)
Coax heat = 500 − 373.882 = 126.118 W
Cable delivery efficiency = 373.882 / 500 = 74.8 percent
Total coax loss with mismatch = −10 log10(373.882 / 500) = 1.26 dB
As a cross check, with a lossless line a = 1, so F = 900 watts, reflected power = 400 watts, and accepted power = 500 watts. With a matched antenna r = 0, accepted power is simply aP = 440.662 watts.
Do not treat 44.4 percent reflected as 44.4 percent of radio output wasted. That percentage refers to the forward wave at the antenna connection. Reflection and dissipation are different physical processes.
The takeaway: before you sentence that antenna to the trash, check the evidence! About 374 watts still reaches it in this example. With efficient radiation and suitable propagation, that can support contacts around the world. A high SWR reading alone is no verdict on performance.
A documented doublet example
A commercial example is the DX Engineering 135 foot multiband doublet. The manufacturer sells it as a nonresonant multiband dipole for 80 through 10 meters with a wide range tuner and supplies 300 ohm ladder line. This establishes a real product using the doublet approach; it does not establish a measured radiation efficiency percentage. [5]
For a separate numerical example, L. B. Cebik W4RNL analyzes a 44 foot doublet at 66 feet above ground in Antennas from the Ground Up, article 34. His published table gives the following results on 15 meters. These describe modeled performance, not field measurements, and do not describe the DX Engineering product. [6]
| Quantity | Value |
|---|---|
| Frequency | 21.2 MHz |
| Feed point impedance | 4160 + j155 ohms |
| Maximum antenna gain | 9.0 dBi |
| SWR relative to 450 ohm line | Approximately 9.3:1 calculated here |
The antenna can have a large impedance mismatch and still produce useful radiation in its main lobes. The peak gain includes the effects of the modeled installation and direction. It is not a radiation efficiency percentage, an all direction gain, or a complete transmitter to air efficiency figure.
Calculating the mismatch
For a real line impedance Z0 and antenna impedance Z = R + jX, the magnitude of the reflection coefficient is:
|Gamma| = square root of [((R − Z0)² + X²) / ((R + Z0)² + X²)]
Using R = 4160 ohms, X = 155 ohms, and Z0 = 450 ohms gives SWR = (1 + |Gamma|) / (1 − |Gamma|) = 9.257, or about 9.3:1.
This SWR is relative to 450 ohm line at the antenna connection. It must not be confused with SWR relative to 50 ohms or a reading before a matched tuner. Feed line and tuner losses still have to be evaluated separately.
For the post: A published model of a multiband doublet shows that a severe mismatch can coexist with useful radiation performance. The commercial antenna illustrates the design family; the published model illustrates the physics. Neither supplies a measured efficiency percentage for the commercial product.
What a doublet antenna is
In this discussion, a doublet is a center fed wire antenna with two equal radiating halves. One feed line conductor connects to the left half; the other connects to the right half. The two halves are separated mechanically by a center insulator. The antenna is used on several bands through a suitable tuner, rather than being trimmed for a low SWR on every band. [7]
The doublet is an old favorite with plenty of life left in it. With suitable dimensions, low loss balanced line, and a capable tuner, it can make an excellent multiband HF antenna. Simple wire still has plenty to say! [7, 9]
Here is why ladder line earns its place: properly installed ladder line can have very low loss on HF even with high SWR. For a strongly mismatched multiband doublet, it can lose substantially less power than a comparable length of typical coax. High SWR still increases loss; the advantage is starting with a very low loss line. Results depend on frequency, line, length, mismatch, and installation. [2, 7]
Follow the signal path
The radio feeds the tuner through a short coax connection. A tuner designed for balanced line feeds the two conductors of the window line. Those conductors run side by side up to the center of the antenna. The crosspieces between them are insulation, not electrical connections. At the top, each conductor connects to its own radiating wire half.
Window line holds the conductors apart with perforated insulating material; open wire ladder line uses spaced insulating crosspieces. Both are balanced two conductor feed lines. The drawing uses a simplified ladder pattern so the two separate conductors are easy to see.
Keep the feed line suspended and clear
Do not lay the line on the ground, drape it over metal, or let its sag bring it within one foot of the earth in this illustration. Route it away from the antenna at approximately a right angle, support it with insulators, and maintain separation from tree trunks and other conductors.
For its 450 ohm ladder line, DX Engineering specifies at least one foot from parallel earth or large metal surfaces and supports every 25 feet or less. Follow the actual line’s instructions. One foot is routing guidance, not a personnel or electrical safety clearance. [8]
The beginner version
Picture energy traveling toward the antenna, with some returning because the antenna and coax do not quite match. In the tuned system, reflections contribute to energy traveling toward the antenna again. This all happens together, very quickly, while the antenna keeps accepting power and sending a signal into the air.
The coax is not perfect, so some energy warms the cable along the way. Think cable warmer, rather than ground warmer. Reflected power has not automatically disappeared into heat; the cable’s actual losses determine how much is lost.
In our example, the radio supplies 500 watts. About 125 watts warms the coax, and about 375 watts reaches the antenna. Assuming the antenna radiates that accepted power efficiently, there is still plenty of signal available to make contacts.
The forward meter shows power in the wave heading toward the antenna. Subtract the reflected reading at that same location to find the net power going forward.
So don’t let a lively SWR needle send a good antenna to the trash! In this example, about 374 watts still reaches the antenna, ready to become a signal under our efficient-radiator assumption. With suitable propagation, that can mean contacts around the world. Check the match and the losses before declaring the antenna a dud. Even our Bluetick knows: a big SWR number is no reason to bury a good antenna!
References for the explanation
The calculations above apply established transmission line relationships to the stated case. The sources below are manufacturer data, ARRL technical guidance, and university textbook material. They are not presented as a separate peer reviewed experimental validation of this installation.
[1] ARRL More About Antenna Tuners
Explains impedance transformation and why a tuner beside the radio leaves standing waves on the feed line beyond it.
[2] Edward J Farmer AA6ZM Let’s Talk Transmission Lines QST June 1997
Discusses radiation resistance versus loss resistance, reflections, tuner placement, and feed line loss under mismatch.
[3] MIT Electromagnetic Fields and Energy Section 14 6
Textbook treatment of reflection coefficient representation, standing wave ratio, and transmission line power relationships.
[4] Messi and Paoloni Hyperflex 10 Sahara Catalog printed page 14
Manufacturer attenuation table at 20 degrees Celsius. At 21 MHz: 1.8 dB per 100 meters, or 0.5 dB per 100 feet after rounding. Accessed October 19, 2026.
[5] DX Engineering 135 foot multiband doublet product specifications
Commercial construction and intended feed arrangement. No measured radiation efficiency percentage is supplied on the product page.
[6] L B Cebik W4RNL 1 Wire 7 Bands 2 Directions Article 34
Also titled The 44 foot Doublet as a 40 to 10 Meter Antenna. Updated November 6, 2003. Archived copy of the author’s analysis. The 44 foot doublet table supplies impedance and modeled gain; the SWR above is calculated from that impedance.
[7] ARRL Random Length Multiband Dipoles
Explains the two equal wire halves, center feed, ladder line, and tuner arrangement.
[8] DX Engineering 450 Ohm Ladder Line installation guidance
Specifies suspended installation, insulating supports every 25 feet or less, and at least one foot from earth or large metal surfaces when run parallel to them.
[9] ARRL Back to Basics QST August 1995
Earlier ARRL guidance describes the simple multiband wire dipole with ladder line and a balanced output tuner, documenting the longstanding approach.