Vertical-Antenna Radials: Return Current, Ground Loss and Pattern

“Radials have two jobs” is a useful teaching shortcut. The more accurate RF picture has two practical jobs but three boundaries: current closure at the feedpoint, loss in the local soil beneath and around the antenna, and reflection from the wider earth and terrain. Confusing those boundaries is where many vertical-antenna myths begin.

Two Jobs, Three Electromagnetic Boundaries

The first two are the familiar “two jobs.” The third matters because a finite radial field does not end the electromagnetic problem at its outer wire tips. Beyond it lies real soil, terrain, seawater, buildings and other conductors. Those distant boundaries still affect the elevation pattern.

Current Must Close—but Not Through One Imaginary Wire Loop

A base-fed vertical monopole cannot be analysed as an isolated one-terminal conductor. Current delivered to the vertical is accompanied by current and displacement current in the rest of the electromagnetic structure. The radial fan may carry most of the deliberate conductor current, but soil, mast, coax exterior, station wiring and nearby metal can also participate.

Calling the radials “the other half of the antenna” can be helpful, provided it is not taken literally. A four-wire elevated ground plane, a dense buried radial field and a small conductive mat have very different current distributions. All can complete the feed structure; none is automatically an equipotential infinite plane.

RF Counterpoise Is Not Protective Earth

A DC ground rod can be essential in a safety or lightning system while being a poor standalone HF return. Conversely, a good elevated RF counterpoise is not a protective earth. Design and document the RF, mains-protection and lightning-current paths separately, then bond them as required by the complete installation design.

Where the Watts Go

At a compatible current reference, a simple loss model is:

The formula is useful only when the terms are referred to the same current or impedance plane. A matching network can transform resistance, and feedpoint resistance can contain radiation, conductor loss, soil loss and energy coupled into unwanted conductors. Subtracting a textbook radiation resistance from a measured input resistance is not, by itself, a reliable ground-loss measurement.

Radials reduce loss mainly by providing a lower-impedance conductor path and by intercepting fields that would otherwise drive lossy current in soil. The improvement depends on where the fields are strongest. That is why wire close to the feedpoint is valuable and why adding ever more distant metal eventually produces diminishing returns.

On-Ground and Buried Radials Are Loaded Transmission Structures

An on-ground radial is coupled to lossy dielectric soil. Its current does not follow the same distribution as an isolated free-space wire, and its useful length is not fixed by one fraction of free-space wavelength. Soil conductivity, permittivity, moisture, insulation, burial depth, wire radius, radial count and the vertical itself all matter.

Rudy Severns, N6LF, demonstrated this experimentally. In his QEX ground-system experiments, Part 2, four 33 ft on-ground radials at 7.2 MHz were substantially soil-loaded. In that particular very-good-soil test, shortening them toward 23 ft improved normalised S21 by about 2.8 dB. Eight radials performed better than four across the tested lengths. The lesson is not “23 ft is ideal”; it is that length, count and soil interact.

“Many short radials always beat a few long ones” is therefore too strong. The defensible statement is that, for a fixed amount of wire, there is an optimum trade between current density near the base and useful outer radius—and that optimum moves with band and ground.

Elevated Radials Are a Different Antenna

Elevated radials can keep much of the return current out of lossy soil. With a small number of wires, however, symmetry and current balance become important because the radials radiate and their individual currents contribute to pattern. Height, droop, nearby objects and feedline routing can change both impedance and azimuth pattern.

They do not have to be “resonant by themselves.” They must form the intended tuned antenna system. The relevant targets are complete-system reactance, balanced radial currents, acceptable feed impedance, low loss and the required pattern.

In N6LF’s Part 3 experiments, four elevated 17.5 ft radials with loading inductors were only about 0.36 dB below four elevated 35 ft radials in one 7.2 MHz test. That is valuable evidence for that geometry—not a universal recipe for half-length radials.

Multiband verticals make the interaction harder. A wire that is helpful on one band may be reactive or poorly placed on another. N6LF’s complete ground-system article series shows why radial count, length, height and frequency should be treated as measured design variables rather than folklore.

Mesh, Sheet, Conductive Cloth and Strips

A conductor does not become an RF ground plane because marketing calls it Faraday cloth. Judge any sheet or mesh by:

• sheet resistance in ohms per square at the frequencies of interest;
• contact resistance and the current-spreading connection at the feedpoint;
• outer dimensions relative to wavelength;
• mesh pitch or gaps relative to the local current distribution;
• corrosion, folds, coating, moisture and mechanical durability; and
• the soil and structure immediately below it.

A continuous low-resistance sheet has fewer gaps than strips of the same outer dimensions. Strips can cover a greater radius for the same material area or mass. Neither statement proves which arrangement wins for a particular portable station.

A 1.5 m × 1.5 m mat is only about 0.05 wavelength square on 40 m. It can still be a useful local current-spreading electrode, especially compared with no intentional return, but it should not be declared equivalent to 16, 32 or 64 wires without a controlled comparison. A fabric with high sheet or contact resistance may lose to copper wires; a well-bonded conductive sheet may beat a sparse short fan close to the feedpoint. “Cloth is useless” and “cloth replaces radials” are both unjustified absolutes.

Percentage Field Change and Decibels

Field strength is an amplitude quantity, so the conversion is:

The arithmetic is exact. The experiment may not be. A 23–28% change is meaningful only when power at the reference plane, antenna tune, feedline current, receiver linearity, geometry, polarisation, propagation and uncertainty are controlled. Repeat an A/B/A sequence and report the spread. Do not turn one daytime signal report into an efficiency number.

Radials and Common-Mode Current

If the intentional radial system presents a high or asymmetric impedance, the outside of the coax may become an attractive additional return path. Adding or improving radials often reduces that current—but not automatically. Current divides according to the complex impedance and coupling of every available path.

A common-mode choke raises the impedance of the coax-exterior path. It does not manufacture a missing return system or absorb “reflected power.” After a choke is installed, current must redistribute through the radials, mast, soil and other conductors. That can change feedpoint impedance, SWR and pattern.

Measure outside-coax current at several positions. A standing-wave minimum at one clamp location can hide substantial current elsewhere. On multiband antennas, repeat the measurement on every operating band and check choke impedance and heating under the intended power and duty cycle.

Why Low SWR Cannot Grade a Radial Field

A lossy return can make the feedpoint resistance look more convenient. A 50 Ω input may contain useful radiation resistance, unwanted radiation from the feedline, conductor loss, matching loss and soil loss. The transmitter sees their sum after any impedance transformation.

That means:

• low SWR does not prove high efficiency;
• a sharper resonance does not automatically mean more radiation;
• a changed SWR after adding a choke is not proof that the choke is bad; and
• adding radials can move the feed impedance away from 50 Ω while increasing radiated field.

A Fair Portable Comparison

There is no defensible universal instruction to “add six more wires” or “always deploy 8–16 radials.” Turn the choice into a small field experiment:

1. Define the constraint. Equal mass, equal deployed radius, equal setup time and equal cost are different comparisons.
2. Keep the radiator fixed. Do not retune length or move the feedpoint between radial tests unless the retuning is explicitly part of the comparison.
3. Measure impedance at the antenna plane. Record complex Z, not only SWR at the radio.
4. Measure accepted power. Account for feedline and matching loss.
5. Clamp the coax. Record common-mode current at several locations.
6. Compare fields. Use stable geometry, multiple bearings where possible and repeated A/B/A measurements.
7. Check temperature. Look for heating at fabric contacts, loading coils, matching components and chokes.
8. Record the ground. Moisture and site conductivity can change the ranking.

A useful matrix might compare a conductive mat, four long wires, eight medium wires and sixteen short wires while holding deployed mass or setup time constant. The best answer can change between 40 m and 10 m, and between dry sand and wet grass.

Model the Complete System

NEC models can separate hypotheses before field testing, but the model must include the conductors that carry current. Represent the vertical, every radial or sheet approximation, mast, coax exterior and any important nearby metal. Use an appropriate real-ground model, test segmentation and run sensitivity cases for soil and connection resistance.

Do not model a finite radial field as infinite perfect ground and then claim the result for the installation. For far-field elevation work, the local radial structure and wider earth must both be represented at a level appropriate to the question.

Claims Worth Keeping—and Claims to Retire

The Practical Verdict

Radials do two vital practical jobs: they provide an intentional feedpoint return and they reduce local soil loss by carrying and redistributing near-field current. Those jobs are coupled, not independent.

The wider earth and terrain still shape the far-field pattern. A finite radial fan, mesh or cloth patch does not automatically become the infinite reflecting plane of the textbook monopole. That distinction lets us compare radial systems honestly without dismissing useful portable counterpoises or overselling them.

Choose radial number, length, height and material for the band, soil, deployment constraint and target pattern. Then measure complex feed impedance, accepted power, coax-exterior current, repeatable field strength and heating. The winning system is the one that performs in that complete installation—not the one with the strongest slogan.

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