50 Ω Coax: Matched Is Not “Balanced”
50 Ω Coax: Matched Is Not “Balanced”
A 50 Ω termination suppresses reflection on a 50 Ω line. It does not change the cable’s geometrical classification, and a non-50 Ω load does not destroy the equal-and-opposite current relationship of the intended TEM mode.
Characteristic impedance, geometrical balance and common-mode excitation describe different properties. Fifty ohms answers a matching question. “Balanced” describes conductor symmetry or a modal current relationship. Outside-shield current asks whether the complete installation has excited an additional external mode.
Keep the definitions separate: coax is geometrically unbalanced at every load, while its intended TEM mode has equal-and-opposite centre and inner-shield currents at every load; a 50 Ω match makes the load reflection coefficient zero.
Three Properties That Must Not Be Mixed
| Property | Question | What changes it? |
|---|---|---|
| Geometrical balance | Are both conductors equivalent relative to the environment? | Cable and installation geometry—not load match |
| Differential match | Does ZL equal the line’s Z0 at the load plane? | Load impedance, frequency and reference plane |
| External/common mode | Is there net longitudinal current on the cable as a whole? | Mode conversion, asymmetry and the external return path |
Coax is conventionally called unbalanced because its centre conductor and surrounding shield are not interchangeable relative to chassis and the outside world. That remains true when the line is perfectly matched.
Inside a uniform coaxial section, below the first higher-order-mode cutoff and away from discontinuities, the intended TEM mode is a two-conductor mode. At each cross-section:
Icentre(z) + Ishield,inner(z) = 0
That modal current equality also remains true when the line is mismatched. Match and current continuity are not the same concept.
What a 50 Ω Match Does
At the load plane, the voltage reflection coefficient is:
ΓL = (ZL − Z0)/(ZL + Z0)
If a 50 Ω line is terminated in 50 + j0 Ω, ΓL = 0. There is no load-reflected differential wave. Voltage and current then have constant amplitudes along an ideal uniform line, apart from propagation phase.
That is all “matched” means here. It does not:
- make the coax conductors geometrically symmetric;
- move the normal return current to the shield exterior;
- prove the antenna system has no common mode;
- prove the feedline has zero attenuation; or
- prove the antenna is resonant or efficient.
Matched but common mode is possible. A complete antenna system can present 50 + j0 Ω in differential mode while still driving substantial current on the coax exterior. An SWR bridge around the internal mode does not measure that third path directly.
What Differential Mismatch Does
When ZL differs from Z0, ΓL is non-zero and a reflected differential wave travels back toward the source. The forward and reflected waves combine into spatially varying voltage and current.
For a lossless line:
SWR = (1 + |Γ|)/(1 − |Γ|)
The magnitude |Γ| and SWR are constant along a uniform lossless line. The phase of Γ rotates with position, so the local R + jX changes. In a lossy line, the reflected wave is attenuated on its trip toward the source, and the SWR measured at the transmitter can be lower than the SWR at the antenna.
Crucially, both the forward and reflected waves belong to the internal TEM mode. Their centre-conductor current still has an equal-and-opposite inner-shield counterpart. A mismatch does not by itself create current on the shield exterior.
Reactance changes current phase relative to voltage. It does not create a phase disagreement between the current entering one terminal of an ideal two-terminal load and the current leaving the other. If the conductor-current sum is non-zero, another physical path exists.
How Outside-Shield Current Is Excited
The shield has an inner surface supporting the normal coax mode and an outer surface capable of supporting an external mode. In a practical antenna installation, energy can be converted into that mode when the complete structure offers an asymmetric third path.
Real shields also have finite transfer impedance and may contain braid apertures, seams or connector discontinuities, so internal and external fields are not perfectly isolated. That non-ideality can couple modes, but it still does not make differential reflection and outside-shield current the same quantity.
Examples include:
- connecting coax directly to a balanced antenna without sufficient common-mode impedance;
- unequal coupling of antenna elements to earth, a mast, roof or wiring;
- an end-fed system using uncontrolled coax as part of its return path;
- asymmetric feedline routing through the antenna near field;
- station bonds, control cables or mains wiring completing an external loop; and
- environmental fields driving the cable exterior on receive.
Changing antenna tuning may change the external-mode current because the complete current distribution and boundary conditions change. That does not demonstrate that the reactive component created common mode.
Current-probe test: clamp around the whole coax. The centre and inner-shield TEM currents cancel in the probe aperture, so the remaining net current corresponds mainly to outside-shield current. Measure at several positions because the external mode can form its own standing wave.
Coax Transforms Impedance, Not Reflection-Coefficient Magnitude
For a lossless line of length l:
Zin = Z0(ZL + jZ0 tan βl)/(Z0 + jZL tan βl)
This explains why an analyzer shows different R and X when cable length changes. The load reflection coefficient is being rotated to a new reference plane. On a uniform lossless line, its magnitude—and therefore SWR—does not improve.
Useful special cases include:
- Half wavelength: the load impedance repeats at the input, neglecting loss.
- Quarter wavelength: Zin = Z02/ZL for a real load.
- Purpose-designed transformer: a quarter-wave section matches two real resistances when its own Z0 is chosen as √(RSRL).
A random extra length of the same 50 Ω lossless coax cannot turn a non-50 Ω load into a true 50 Ω match at the source; it changes the impedance phase around the constant-|Γ| circle. Deliberate matching with transmission line requires the correct characteristic impedance, one or more stubs, or additional reactive elements.
Why an SWR Meter May Change When Coax Length Changes
An ideal, accurately calibrated SWR measurement on a uniform lossless line would report the same SWR at every plane. Real observations can differ because of:
- line attenuation reducing reflected-wave magnitude toward the source;
- meter directivity, calibration and sensitivity to complex impedance;
- connectors, adapters or damaged cable introducing discontinuities;
- source output impedance or a tuner interacting with the line;
- frequency-dependent loss and velocity factor; and
- common-mode current causing the instrument, jumper and operator environment to join the measurement.
“The SWR changed when I added coax” is therefore a diagnostic clue, not proof that coax has several different SWRs or that a particular length balanced the system.
Line Length and Antenna Resonance
Antenna feedpoint resonance is commonly defined by zero feedpoint reactance at a stated geometry and environment. Adding line does not move that property of the isolated antenna. It changes the impedance observed at the new measurement plane.
The combined input of antenna plus line may have zero reactance at a different frequency. A tuner may also find an easy match elsewhere. Calling either observation “the antenna resonance moved” hides the reference plane.
Always state:
- where the measurement plane is;
- whether cable delay and loss were de-embedded;
- the antenna geometry and environment; and
- whether outside-shield current was controlled.
What a Choke Changes—and What It Does Not
A common-mode choke adds ZCM = RCM + jXCM to the external path while ideally leaving the internal TEM mode almost unchanged. It does not absorb a differential phase error or remove the reflected wave caused by ZL ≠ Z0.
A feedpoint choke is appropriate where the antenna transition would otherwise excite the cable exterior. A station-entry choke may help if a residual external mode or receive-noise path remains. Two chokes are not a universal prescription: placement follows the current paths and measurement.
If adding a choke changes SWR, the likely explanation is that the feedline previously participated in the antenna or measurement. The new reading may be a better description of the intended antenna, even if the number is less flattering.
A Test That Separates Match from Mode Conversion
- Calibrate the analyzer to the antenna or load reference plane.
- Terminate the coax in a compact shielded 50 Ω load and measure S11 plus clamp current along the cable.
- Replace it with a known reactive two-terminal load without changing cable routing or nearby objects.
- Confirm the expected differential reflection and again measure net cable current at several positions.
- Introduce a deliberate asymmetry or external return path and observe the new clamp-current distribution.
- Add a characterized choke and repeat, recording both differential S11 and external current.
This experiment can show a large reflected differential wave with almost no external current, and external current with an excellent differential match. That is the cleanest demonstration that match and balance are independent coordinates.
| Property or action | Engineering interpretation |
|---|---|
| 50 Ω termination on a 50 Ω line | Sets ΓL to zero at the load plane; coax remains geometrically unbalanced. |
| Differential mismatch | Creates reflected TEM waves; each still has equal-and-opposite conductor currents. |
| Outside-shield current | Occupies an external path distinct from the centre-to-inner-shield TEM mode. |
| Extra uniform 50 Ω coax | Rotates Γ and changes local impedance; loss can reduce the reflected wave measured toward the source. |
| Tuner or feedline transformation | Changes impedance at its input reference plane; it does not redefine the isolated antenna’s feedpoint resonance. |
| Common-mode choke | Impedes the external mode; differential mismatch remains a separate problem. |
Bottom line: 50 Ω is a matching condition, not a balance switch. Coax remains geometrically unbalanced, its internal TEM currents remain equal and opposite with or without reflections, and outside-shield current appears only when the complete structure excites an external mode.
Primary technical references
- Keysight — Reflection Coefficient, VSWR and Standing Waves
- Keysight — Transmission-Line Input Impedance and Reflection Equations
- Rohde & Schwarz — Coaxial Transmission-Line Electrical Properties
- Rohde & Schwarz — VSWR and Reflection-Coefficient Relationships
- ARRL — Transmission Line for Windows
- W7EL / ARRL — Equal-and-Opposite Coax Currents and Imbalance Current
- NASA — Multiple Conductors, Shields and External Shield Current
Mini-FAQ
- Is coax balanced at exactly 50 Ω? No. It is matched at 50 + j0 Ω if Z0 is 50 Ω; its geometry remains unbalanced.
- Does high SWR make the centre and inner-shield currents unequal? No. Forward and reflected TEM waves both preserve the equal-and-opposite conductor-current relationship.
- Can a matched antenna have common-mode current? Yes. Differential match does not measure or prevent excitation of the external mode.
- Can adding coax improve a real meter reading? It can change the reading through loss, meter limitations, source interaction or common mode. It does not reduce |Γ| on an ideal uniform lossless line.
- Can a quarter-wave line match impedances? Yes, when its characteristic impedance is deliberately chosen for the source and load, or when it is part of a designed stub/network.
- Does a choke fix mismatch? No. It suppresses an external current path; a matching network addresses differential impedance.