Stop Cutting Coax to “Fix” SWR

A length of coax is not transparent whenever the load differs from the line’s characteristic impedance. It moves the complex impedance around the Smith chart. That can make a transmitter, tuner or meter behave differently while the antenna at the far end remains exactly the same load.

This is why “cut the coax until the SWR is good” is such a seductive workshop ritual. The result can be repeatable at one frequency, yet the explanation is usually wrong. The correct question is not whether the display changed. It is what quantity changed, at which reference plane, and at what cost?

The Load Stays at the Far End

At the antenna terminals, the reflection coefficient is set by the antenna-system impedance Z_L and the line impedance Z₀:

Adding a uniform line does not alter Z_L. On an ideal lossless line, the magnitude |Γ| and therefore SWR are constant along the line, while the phase of Γ changes with electrical length. The complex impedance seen at the input changes even though the antenna-plane SWR does not.

That distinction matters. A transmitter responds to the impedance at its own connector. A tuner responds to the impedance at its input and output planes. An inline meter samples at its own location. None of those readings is automatically the antenna-terminal impedance.

What the Smith Chart Shows

Plot a mismatched load on a Smith chart and travel toward the generator along a lossless line. The point rotates around a circle of constant |Γ|. Cutting or adding coax chooses a different point on that circle. Resistance and reactance at the radio end can change dramatically; the distance from the chart centre does not.

That is the visual reason a particular cable length can bring an input impedance inside a tuner’s range or away from a transmitter’s worst foldback region. The cable has transformed the impedance presented upstream. It has not made the far-end antenna resonant, efficient or well matched.

Rohde & Schwarz’s Smith-chart guidance emphasizes that every plotted impedance is normalized to a stated reference impedance. Keysight’s transmission-line examples likewise express movement in wavelengths toward the generator. The chart does not lose track of either the reference impedance or the measurement plane; neither should we.

Real Coax Can Make the SWR Look Better

A real cable has conductor and dielectric loss. The forward wave is attenuated on its trip to the load, and the reflected wave is attenuated again on its return. For a uniform line with propagation constant γ = α + jβ and physical length ℓ:

The input-plane SWR therefore moves toward 1:1 as attenuation increases. This is not an antenna improvement. It is evidence that the reflected sample has suffered a round trip through loss. The same cable also reduces the forward power arriving at the antenna, and mismatch changes the voltage, current and dissipation distribution along the line.

There is no universal promise that a short HF cable with a certain SWR loses less than some fixed number of decibels. Cable type, frequency, length, temperature, connectors, load impedance and source re-reflection all matter. Manufacturer matched-loss data are a starting point; a characterized two-port network or a suitable transmission-line model is better when the number matters.

When Cable Length Is a Real Design Variable

“Never cut coax to length” would be just as wrong as cutting it blindly. Electrical length is deliberately used in RF engineering:

• Quarter-wave transformers: a designed line with the required characteristic impedance can transform a specified load at its design frequency.
• Half-wave repeaters: an ideal half-wave line repeats the load impedance at its input, subject to loss and the actual velocity factor.
• Stubs: open or shorted sections provide frequency-dependent susceptance or a deliberate transmission zero.
• Phasing lines: controlled amplitude and phase feed arrays, hybrids and combining networks.
• Tuner-range management: a selected feed-line length can avoid extreme input voltage, current or impedance that a particular tuner cannot safely handle.
• Measurement fixtures: a known line can move a calibration or observation plane when its propagation is characterized.

Each use begins with a target impedance, frequency range, Z₀, propagation constant, power level and uncertainty. It is not an after-the-fact search for the cable length that makes one meter prettier.

Quarter Wave Does Not Mean Broadband Magic

For an ideal lossless quarter-wave section terminated in a real resistance, the familiar special case is Z_in = Z₀²/Z_L. The line impedance must be chosen for the two resistances being transformed. A random section of 50-ohm coax does not become a universal matching network merely because it is a quarter wavelength long.

Electrical length depends on frequency and the cable’s propagation velocity. Velocity factor is not a decorative catalogue number: construction, dielectric and manufacturing tolerance affect it. Connectors, adapters, bends and load reactance also move the result. A quarter-wave solution is therefore inherently frequency-sensitive, and a complex or changing antenna load needs the full transmission-line calculation rather than the resistance-only shortcut.

The Tuner Matches Its Own Input Plane

A shack tuner can present the transmitter with an acceptable load. That is valuable: it may prevent foldback and let the transmitter deliver power. It does not remove the standing-wave pattern or dissipation in the coax between tuner and antenna.

Whether the best match belongs at the feedpoint or in the shack is an engineering trade. A remote or feedpoint network can reduce high-SWR coax loss, but must survive the local voltage, current, weather and common-mode environment. A low-loss open-wire line can intentionally carry a large mismatch to a suitable tuner with far less dissipation than a small coaxial cable. The answer comes from the whole system, not an antenna-name rule.

Tuner loss is not one fixed fraction of a decibel. It changes with topology, component Q, transformation ratio, frequency, power, adjustment and load. Record the tuner state and temperature when comparing cable choices.

Keep Differential Transformation and Common Mode Separate

The equations above describe the intended differential transmission-line mode. Current on the outside of a coax shield is a different path. If the antenna’s return path is undefined, the cable exterior, mast, bonding and station wiring can become part of the radiating system. Moving or cutting the cable can then change the actual antenna, not merely rotate an impedance.

A common-mode choke is placed where the intended common-mode boundary belongs after the antenna’s return path has been defined. That position is not determined by a slogan such as “one quarter wavelength from the feedpoint.” Measure exterior current and check whether moving the cable, touching the connector area or changing the choke alters the trace. Instability is evidence that more than the intended differential line is involved.

A Measurement Sequence That Separates the Effects

• Choose the plane. Decide whether you need the antenna terminals, cable input, tuner input or transmitter connector.
• Measure the load safely. Use a calibrated VNA or analyzer at low power, with the reference plane established at the point being claimed.
• Characterize the cable. Record type, physical length, loss, velocity factor, connectors and temperature; use measured S-parameters when accuracy matters.
• Translate rather than guess. De-embed the characterized line or use the full complex transmission-line equation.
• Check common mode. Measure exterior current and repeat with a deliberately defined return path and choke boundary.
• Check power limits. Calculate or measure line voltage, current and heating under mismatch. A tuner accepting the impedance does not certify the cable.
• Verify over the operating band. A fortunate point at one frequency is not a broadband solution.

Do not connect or disconnect test equipment while RF power is applied. Elevated feedpoints, open conductors, wet weather and station bonding add shock, RF-burn, fall and lightning hazards. Use de-energized low-power measurements and equipment rated for the intended plane.

Primary and Authoritative Sources

• Keysight, Network Analysis—transmission-line theory, reflection coefficient, SWR, S-parameters, impedance and Smith-chart interpretation.
• Keysight, Understanding the Fundamental Principles of Vector Network Analysis—travelling waves, S-parameters, calibration and reference-plane measurement.
• Keysight, RF Design Software Learning Kit—worked Smith-chart examples using transmission-line electrical length toward the generator.
• Rohde & Schwarz, RF Port Impedance Verification—complex impedance, normalized Smith-chart display and calibration at the measurement plane.
• Rohde & Schwarz, Measurements on Balanced Components with Vector Network Analyzers—the formal relationship among reflection coefficient, impedance and SWR.

Practical Conclusion

If cutting coax changes the shack reading, believe the reading—but identify its plane. You have changed the impedance presented to the instrument and, on a lossy line, the magnitude of the reflection that returns to it. You have not proved that the antenna accepted more power or radiated it more effectively.

Cut coax because an electrical design calls for a transformer, stub, phase shift, safe tuner load or practical route. Do not cut it until an unexplained SWR number looks friendly. The Smith chart will show the difference, provided the reference plane and loss stay attached to the trace.

Questions, antenna-factor records or height trials to share? Contact RF.Guru.