From model to measurement.

Antenna gain is a model input, not a diagnosis

For a dish, gain depends on electrical aperture: diameter relative to wavelength and the fraction of aperture used effectively. Doubling diameter at fixed frequency and efficiency adds about 6.02 dB. Doubling frequency has the same theoretical effect only if efficiency remains unchanged. Real feeds, illumination, blockage and surface accuracy can change that assumption. Compare a calculation with a source-verified specification at the same frequency; do not replace a measured product value with an ideal-aperture result.

Separate range geometry from measurement quality

The Fraunhofer distance describes a conventional phase-front criterion for a chosen antenna dimension. It does not describe the cleanliness of the measurement environment. An adequate distance with strong reflections, poor source illumination or polarization error can still yield inaccurate gain or pattern measurements. Record source antenna, alignment, reference planes and repeatability along with distance. A radiation pattern is frequency- and plane-specific; compare like cuts rather than assuming one graph represents every frequency.

Clearance is a path question

The first Fresnel radius varies along the link and reaches its maximum at the midpoint of a symmetric path. A 60% illustrative clearance target gives a useful screening value, not a completed path survey. Place that radius against actual terrain and obstacle heights, antenna heights and an appropriate curvature/refraction model. A clear visual line of sight does not by itself establish adequate Fresnel clearance.

Measure the entire RF jumper assembly

Cable attenuation scales with length, but an assembly also includes connectors, adapters and installation effects. Use the exact manufacturer/model curve and units. A quoted dB/100 m value must be divided by 100 before multiplying by metres. A measured full-assembly insertion loss should replace, rather than be added to, that assembly estimate. Measurements must use comparable frequency and reference planes. A lower calculated loss may improve link budget; it does not establish a particular data-rate improvement.

Compare power conventions before comparing antennas

The power reported by a radio may be per chain, combined, averaged or otherwise processed. Check the vendor definition before using it as received signal power. Under an identical-antenna assumption, unexplained loss is split equally between the two estimated gains. With a known TX reference, all unexplained loss is assigned to the estimated RX gain. Neither operation isolates antenna manufacturing quality. Change one installation variable at a time and keep a measurement log.

Installation checks

Confirm configured TX power at the actual modulation and bandwidth.

Confirm polarization and chain connections at both ends.

Inspect connector seating, adapters and return-loss behavior.

Inspect weather seals and moisture without assigning an unmeasured loss value.

Check mounting rigidity and pointing stability under wind.

Compare repeated measurements to identify variation consistent with multipath or interference.

A surplus is a reason to review assumptions, not proof of extra antenna gain.