Why “Antenna Loss Calculators” Aren’t That Useful
Why “Antenna Loss Calculators” Aren’t That Useful
A calculator can solve its model perfectly and still answer the wrong station question. The result becomes useful only after the cable, load, source and measurement planes are defined.
RF.Guru working definition: Common-mode current is the non-cancelling phasor-sum current in a specified set of conductors, evaluated at a defined cross-section and using a declared current-direction convention. In the intended differential transmission-line mode, the outgoing and return currents are equal and opposite, so their phasor sum is zero. When they do not cancel, the remaining current must close through another reference or return path—such as the outside of a coax shield, a mast, equipment chassis, station wiring, nearby structures, earth, the operator, or distributed coupling through the environment.
This broader working definition is especially useful in practical antenna systems. On transmit, non-cancelling current on the outside of the coax can make the feedline and connected structures part of the radiating antenna system unless that path is intentional, clearly defined and properly controlled—for example by providing the required return path and placing a suitable common-mode choke at the correct boundary.
Enter a cable type, length, frequency, power and SWR. A web page returns “2.1 dB loss,” “62 W at the antenna” and perhaps an ERP figure with two decimal places. That feels like measurement. It is actually a calculation built on whatever definitions and assumptions the page accepts.
I am not against calculators. I use models constantly. I am against letting a precise display hide an incomplete problem statement. The question is not, “Can the software multiply numbers?” The question is, which power, at which plane, through which line, into which load, under which conditions?
The KV5R Calculator Is a Useful Case Study
The KV5R coax-loss calculator is worth discussing because its own notes are unusually candid. The page says that its JavaScript came from AC6LA’s 2001 implementation of the ARRL “additional loss due to SWR” formula, that AC6LA later withdrew that code in favour of more rigorous methods, and that the provenance of some K1 and K2 cable constants is not known.
Those statements do not make the page fraudulent or useless. They tell us exactly how to use it: as a quick estimate whose inputs, line model and approximation must be checked before the output is promoted to evidence.
Short version: if a calculator’s own author tells you to verify the cable data and calculation, believe him. A planning estimate is not an installed-system measurement.
AC6LA’s later article, “Additional Loss Due to SWR” Is in Quotes for a Reason, explains why the simple ARRL-style expression is not generally exact. His TLDetails program instead works with the transmission-line equation, complex impedance and a distributed line model. That history is the narrative here: convenient SWR arithmetic is not the same thing as solving the actual lossy two-port.
“Loss” Is Several Different Quantities
A single result labelled loss may combine quantities that answer different questions.
| Quantity | What it means | What it does not prove |
|---|---|---|
| Matched-line attenuation | The reduction through a specified line when its ports are terminated in the stated reference impedance, under stated frequency and environmental conditions. | Loss with an arbitrary complex antenna load, connector defects, common mode or a tuner elsewhere in the system. |
| Mismatch | Reflection caused by a load differing from the reference impedance. At one plane, its magnitude can be expressed as reflection coefficient, return loss or SWR. | Heat dissipated in the cable or radiation efficiency of the antenna. |
| Line dissipation | Real power converted to heat in conductor and dielectric loss between two declared planes. | Total station loss unless connectors, adapters, matching networks and unintended modes are included. |
| Insertion or transducer loss | A source-to-load power ratio under defined source, load and reference impedances. | A topology-independent property of the cable when the term is used without those definitions. |
| Radiation efficiency | Radiated power divided by accepted power at the antenna terminal, with the boundary declared. | Something that follows from SWR, matched cable attenuation or one gain number. |
| Realized gain or ERP | A directional far-field result that includes the stated mismatch and uses an isotropic or dipole reference as defined. | A single installation-independent number for an antenna mounted at arbitrary height and ground. |
Reflection is not dissipation. A reflected wave can travel back toward the source, be attenuated in the line, be transformed by a matching network or be reflected again. Loss occurs where real power becomes heat or leaves the intended circuit by another mechanism. That is why “SWR loss” is a dangerous label when its definition is not shown.
SWR Tells You Magnitude, Not the Complete Load
At one reference plane, SWR determines the magnitude of the reflection coefficient:
|Γ| = (SWR − 1) / (SWR + 1)
Γ = (Z − Z0) / (Z + Z0)
The first expression gives |Γ|. It does not give the phase of Γ, so it does not identify the complex impedance Z = R + jX. Many different loads sit on the same constant-SWR circle.
In a low-loss approximation with a nearly real characteristic impedance, load-end SWR magnitude and known line attenuation can support a useful estimate of additional cable dissipation. That is the limited case behind many simple calculators. A rigorous lossy-line solution, however, uses the line’s complex propagation constant and characteristic impedance together with the complex termination. It also states what power is held constant and which source and load planes define the result.
There is no free gain hidden in a reactive load. If a program reports a negative “additional SWR loss” relative to one chosen baseline, that is a warning to inspect the definition, source match and plane—not evidence that reactance made a passive cable amplify power. Compare available, incident, accepted and delivered power consistently.
The Reference Plane Changes the Number
Most operators know the SWR displayed by the transceiver or an instrument in the shack. A calculator may ask for SWR at the antenna end. Those are not interchangeable on a lossy line.
The line transforms impedance with electrical length, while attenuation reduces the returning wave. As a result, SWR measured at the transmitter end can look lower than load-end SWR. A tuner beside the radio may present 50 Ω to the transmitter while the line between tuner and antenna still carries a large standing wave. A tuner at the antenna can instead change the line termination itself.
That does not make either measurement wrong. It means every impedance, SWR and power value needs a plane:
- transmitter output;
- input and output of the tuner;
- station end of the feed line;
- antenna end of the feed line; and
- antenna terminal after any transformer or matching network.
A vector network analyser can move a reference plane mathematically only when the intervening network is characterised well enough to embed or de-embed it. The Rohde & Schwarz de-embedding guide makes the general measurement point clear: fixtures and lead-ins must be characterised before their effects can be removed.
A Cable Family Name Is Not a Measurement
“RG-8X,” “RG-58” or even a commercial family name does not describe every installed assembly. Conductor dimensions and plating, dielectric, shield construction, manufacturing tolerance, connector choice, temperature, bend history, water ingress and ageing all affect the line.
Use the exact manufacturer’s data where possible, including the conditions attached to it. The Times Microwave LMR-400 data sheet, for example, explicitly states that its attenuation table is typical and referenced to VSWR 1.0. Its power table has separate ambient-temperature, conductor-temperature, altitude and dry-air conditions. That is engineering: the number arrives with a boundary.
Connectors and adapters belong to the assembly. So do workmanship and weather seals. A calculator populated with generic catalogue constants cannot diagnose a wet connector, crushed foam dielectric, damaged braid or badly installed termination.
The Differential-Line Model Is Not the Whole Antenna
A normal coaxial line carries the wanted differential mode between the centre conductor and the inner surface of the shield. Exterior-shield current is a separate mode referenced to the surrounding installation. If the antenna, transformer, mast, shack wiring or environment excites that exterior path, the feed line can contribute to radiation, pickup and pattern change.
A conventional coax-loss calculator models the intended transmission-line mode. It does not know the exterior current distribution, the choke’s complex impedance, the mast bond or the path through the station. That does not invalidate its differential-mode calculation; it limits the system conclusion you may draw from it.
Measure exterior current separately, at several positions and on every operating band. Change one boundary at a time—choke position, feed-line route or bonding condition—and repeat the measurement. A single current reading near the radio cannot map the whole common-mode loop.
ERP Needs a Pattern, Not a Guess
Some calculators multiply estimated delivered power by an entered antenna-gain value and label the result ERP. That is only defensible when the gain definition and direction are known, the antenna boundary is declared and all losses up to that boundary are included.
Installed gain varies with frequency, azimuth, elevation, height, ground, nearby structures and unintended feed-line current. ERP is referenced to a half-wave dipole; EIRP is referenced to an isotropic radiator. Mixing the reference, using a free-space peak-gain number for a low installed antenna, or ignoring pattern direction can create an impressive but meaningless output.
Use the Calculator for the Job It Can Do
A simple calculator is useful for screening choices:
- comparing the matched attenuation of two specified cables at one frequency;
- checking whether a very long or small-diameter run is an obvious problem;
- building a first-pass link or thermal budget near matched conditions; and
- testing sensitivity to cable length, attenuation tolerance and SWR assumptions.
It is not enough by itself to answer:
- how much heat the actual mismatched assembly dissipates;
- whether a tuner has reduced loss on the line;
- how much accepted power the antenna radiates;
- whether the feed-line exterior is part of the radiating structure; or
- what ERP exists in a useful direction.
A Measurement Chain That Survives Reality
When the answer matters, I use this sequence:
- Name the measurand. Matched attenuation, total cable dissipation, power delivered to the antenna terminal, radiation efficiency and realized gain are different jobs.
- Draw the planes. Mark the transmitter, tuner, cable ends, transformer and antenna terminal.
- Identify the exact assembly. Record cable manufacturer and type, length, connectors, adapters, temperature, age and installation condition.
- Measure complex data. Obtain R + jX or calibrated S-parameters over the band, not only one SWR value.
- Characterise the line. Use trusted manufacturer data or measure the assembly; include uncertainty and calibration residuals.
- Cascade the real networks. Include tuner, transformer, adapters and line as defined two-ports where appropriate.
- Check the unintended mode. Map exterior current and repeat after controlled routing or choking changes.
- Validate the claimed outcome. Temperature supports dissipation questions; a suitable field method supports gain or pattern; a controlled receiver test supports SNR.
Then run sensitivity cases. Change attenuation within its tolerance, vary the measured load, include connector uncertainty and compare warm and cold conditions. Reporting a credible range is often more useful than printing “1.73 dB.”
Keep the Convenience, Drop the False Certainty
The calculator is not the villain. The category error is. Matched cable attenuation is useful. SWR is useful. Complex impedance is useful. A two-port model is useful. None of them alone is radiation efficiency or installed ERP.
Use a simple web result to decide what deserves a better model or measurement. When the conclusion becomes expensive, thermal, safety-related or technically important, carry the complex load and reference planes through the complete system. Precision belongs in the evidence, not merely in the number of digits on the screen.
Primary Technical References
- KV5R: Coax Loss Calculator and Its Published Method Notes
- AC6LA: “Additional Loss Due to SWR” Is in Quotes for a Reason
- AC6LA: TLDetails Distributed Transmission-Line Model
- AC6LA: Transmission-Line Mathematics and Measurement Workbooks
- Times Microwave Systems: LMR-400 Data Sheet and Declared Conditions
- Rohde & Schwarz: Accurate Test-Fixture Characterisation and De-embedding
Mini-FAQ
- Is high SWR the same as cable loss? — No. SWR describes reflection magnitude at a stated plane. Cable loss is real power dissipated in the line, and mismatch can change that dissipation.
- Why is load-end SWR different from shack SWR? — The line transforms impedance, and attenuation reduces the returning wave. A lossy line can therefore make the transmitter-end SWR look lower.
- Does a tuner beside the radio reduce loss on the feed line? — Not automatically. It matches the transmitter to the line input but does not change the standing wave already present between the tuner and antenna.
- Why does complex impedance matter? — SWR gives only the magnitude of the reflection coefficient. Complex impedance also supplies phase, which a rigorous lossy-line calculation needs.
- Can I trust a manufacturer’s attenuation figure? — Use it within its stated frequency, termination, temperature and installation conditions. The finished cable assembly and its condition still need verification.
- Can a coax-loss calculator predict ERP? — Only with a defensible delivered-power result and installed directional gain using the correct reference. One generic antenna-gain entry is not enough.