Antenna Claims: How to Separate Match, Efficiency, Gain and Evidence

SWR, transformer loss, pattern shape, RBN spots and successful contacts are all useful observations. None can silently substitute for a different antenna metric.

The Claim Is the Unit of Review

An antenna can make contacts, be convenient and still be described with the wrong physics. Review the claim without turning the review into a judgement about its builder.

First rewrite the sentence as a measurable proposition:

• “efficient” becomes radiation or total efficiency at stated frequencies;
• “4 dB gain” becomes realised gain relative to a named reference, direction, polarisation and elevation;
• “directional” becomes a pattern or front-to-back ratio under stated geometry;
• “no radials” becomes a documented return-current path;
• “broadband” becomes an impedance or realised-gain bandwidth at stated limits; and
• “field proven” becomes a repeatable test protocol with uncertainty.

SWR and Mismatch Efficiency: The Exact 96%

For a real reference impedance, the magnitude of the reflection coefficient is:

|Γ| = (SWR − 1) / (SWR + 1)

The mismatch efficiency at that reference plane is:

η_m = 1 − |Γ|²

At 1.5:1 SWR, |Γ| = 0.2 and η_m = 0.96. So “96%” is not invented: 96% of the incident power is accepted at that plane, and the corresponding one-way mismatch loss is about 0.18 dB.

The error is renaming mismatch efficiency as radiation efficiency. A matched resistor can accept nearly all incident power and convert nearly all of it to heat.

Reference plane matters. If the analyser is at the shack end of a lossy feedline, line attenuation can make the measured SWR look better while consuming power before it reaches the antenna.

Radiation Efficiency, Total Efficiency and Gain

A NIST antenna-efficiency paper distinguishes radiation efficiency from total efficiency: radiation efficiency is radiated power divided by net power accepted at the antenna port; total efficiency also includes mismatch.

The useful relationships are:

ITU-R BS.705-1 makes the central distinction: directivity is a property of pattern shape, while gain also accounts for antenna efficiency.

A Component Efficiency Is One Factor in the System

A measured transformer loss of 0.3 dB corresponds to a power ratio of about 93.3% under the stated fixture, impedance and calibration. That can be excellent component performance. It is not automatically 93.3% antenna-system efficiency.

A simplified cascade is:

η_system = η_transformer × η_feedline × η_mismatch × η_radiation

Each factor must refer to compatible ports and conditions. Transformer loss can also change with impedance, common-mode current, voltage, frequency, temperature and drive level, so a single matched two-port result is not a universal power rating.

“No Radials” Still Needs a Return-Current Map

An unbalanced radiator does not operate without a return path. The return may use explicit radial wires, a counterpoise, a vehicle body, a metal roof, mast, coax exterior, capacitive coupling to earth, station wiring or some combination.

“Radial” and “counterpoise” can describe different geometries and design intentions, so it is too broad to say that every current-carrying return conductor is literally a radial. The honest requirement is simpler: identify it, include it in the model and disclose any installation dependence.

If changing a return conductor changes impedance, external feedline current, loss or pattern, it is an active part of the antenna system. Calling it a tuning tail does not remove that role.

Two Radials Can Create Directivity—But Gain Still Needs Loss

An asymmetric return system can redistribute radiation and therefore create real directivity. This is not merely a visual artefact. If the total radiated power were unchanged, more radiation in one solid angle would require less elsewhere.

The unproven leap is from an asymmetric normalised pattern to an absolute forward-gain claim. Ground loss, conductor loss, return-current loss and feedline participation may reduce total radiated power. A normalised polar plot hides that scale.

Front-to-Back Ratio Is Not Forward Gain

Front-to-back ratio compares radiation in two directions. It can improve because the rear field fell, because the forward field rose, or both. It contains no absolute scale.

A 20 dB front-to-back ratio could describe a strong forward lobe and ordinary rear field, or a mediocre forward field and a very deep rear null. Report forward gain and front-to-back ratio separately, with the angles and polarisation used.

NEC Produces the Answer to the Model You Built

NEC is an electromagnetic solver, not an automatic representation of a park. Lawrence Livermore National Laboratory's NEC-5 can calculate currents, impedance, patterns and near fields for wires and conducting surfaces above ground.

Its validation manual separates two error classes:

• numerical-model error from segmentation, mesh, junction treatment and solver limitations; and
• physical-model error from simplifying or omitting parts of the real structure and environment.

For a portable antenna, publish the model file and state wire diameter, segmentation, source model, loads, real-ground parameters, height, radial layout, feedline exterior, choke impedance and nearby conductors. Run convergence and sensitivity checks instead of quoting a one-decimal gain from one geometry.

From Polar Plot to Picnic Table

A model can validly predict what a defined geometry does. The question is whether the field deployment reproduces that geometry.

Coax routing, radio and battery leads, operator distance, wet soil, metal furniture, fences and vehicles can change current distribution. Do not demand that every small object be modelled; identify the variables likely to carry significant current or scatter a significant field, then test sensitivity.

A good engineering statement is: “Within these assumptions, the model predicts this behaviour.” A stronger field claim needs field evidence.

RBN and WSPR Can Support Comparisons—With Experimental Design

The Reverse Beacon Network is a network of receiving stations that reports what it hears and offers tools for quantitative comparisons as heard by a chosen reverse beacon on a chosen band and time.

That makes RBN more than anecdote, but not an antenna range. Propagation, transmit power, timing, frequency, receiver noise, AGC/calibration, polarisation and geometry remain variables.

A useful on-air A/B test should:

• switch antennas rapidly enough that propagation is approximately common;
• use the same transmitter power and feed reference plane;
• compare reports from the same receiving stations;
• repeat many cycles, bands, times and path directions;
• avoid cherry-picking only successful spots; and
• report the distribution and uncertainty, not only the best result.

Such a test can estimate relative field performance on sampled paths. It still does not directly deliver absolute efficiency or a complete three-dimensional pattern.

Contacts Prove a Link Closed

A completed QSO proves that, under those conditions, the link budget was sufficient. It sets a lower bound on usable performance for that path and mode. It does not isolate antenna gain, efficiency or pattern from propagation, receiver sensitivity, noise, coding and operator technique.

This is not an insult to practical operating. “Easy to deploy and made the required contacts” can be the most important result for a portable antenna. It is simply a different claim from “4 dB gain” or “90% radiation efficiency.”

Shape Does Not Supply an Unstated Reference

Diamond, rhombic, folded, spiral and fractal describe geometry. Geometry controls current distribution, impedance, bandwidth and pattern, but the name does not import the performance of a larger antenna that looks similar.

A true gain claim still requires a reference and loss accounting. On receive, effective aperture and gain are linked by A_e = λ²G/(4π). A compact structure may be useful, but visual resemblance is not a gain measurement.

An Evidence Ladder for Antenna Claims

1. State the smallest defensible claim. “Usable SWR,” “modelled directivity,” “relative RBN improvement” and “measured realised gain” are different.
2. Name the reference. Is gain dBi, dBd, relative to a quarter-wave vertical, or relative to the previous deployment?
3. Draw the complete current path. Include return conductors, coax exterior, choke and nearby conductive structure.
4. Publish model inputs. Include ground, loads, geometry and convergence evidence.
5. Measure components separately. Transformer, feedline and choke results help close the loss budget.
6. Perform controlled A/B tests. Change one variable at a time and repeat enough to estimate scatter.
7. Report contrary results. Band edges, null directions and failed configurations define the real operating envelope.

Better Claims Are More Useful

The Practical Verdict

The opposite of antenna hype is not cynicism. It is metrology.

A useful antenna does not need every metric to be exceptional. A compact system may trade efficiency for deployment speed; two radials may create useful directivity; an RBN test may reveal a repeatable relative advantage. Those are worthwhile results when the measurement and boundary conditions match the words.

Ask what was measured, where it was measured, what reference was used and what alternative explanation remains. That is how a working antenna becomes credible engineering.

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