Faraday Strip Radials: 23% of What, Measured Where?

Chameleon Antenna describes its CHA FSR as two 20-foot conductive strips laid in an X. Its public product page reports a median 23–28% increase in relative field strength in initial Northern Alabama tests, compared with the four 12.5-foot wire radials supplied with two named vertical systems. That is the claim I am examining. I am not arguing that a strip cannot work. I am asking what the published number proves.

Start With the Quantity

Electric-field strength is an amplitude, normally expressed in volts per metre. Power and power density are different quantities. The two decibel conversions are:

If 1.23 or 1.28 is the electric-field ratio in a far-field plane wave, power density scales with E². The corresponding power-density ratios are about 1.51 and 1.64, which still produce the same 1.80 dB and 2.14 dB results. We must not apply 10 log to an amplitude percentage or 20 log to a power percentage.

Nor should we translate either result casually into “one-third of an S-unit.” S-meter response is receiver-, band-, mode- and calibration-dependent; fading and noise add further variability. A repeatable 2 dB field change is not nothing, but an S-meter slogan does not tell us whether it matters to a particular link.

The Claim Is a Comparison, Not an Efficiency Rating

The manufacturer identifies a useful comparison: its two-strip system versus four 12.5-foot wire radials. The public page does not publish enough of the initial field-test protocol to turn that comparison into a general antenna-efficiency number. At minimum, a reproducible result needs the band, radiator and matching configuration, accepted feedpoint power, feedline and choke boundary, radial orientation, soil and moisture, probe distance, bearing, height, polarisation, sequence, repetitions, raw observations and uncertainty.

A field-strength difference at one position can come from a change in total radiated power, elevation or azimuth pattern, common-mode current, matching loss, radiator current, or coupling to the measurement setup. It cannot by itself separate those causes. An input SWR change cannot separate them either.

The claim may be perfectly repeatable for the named Chameleon systems and test sites. It still does not establish that strip radials universally add 23–28%, or that the completed antenna has a particular absolute efficiency.

A Wide Strip Changes More Than Resistance

Width can matter. In Rudy Severns' primary study of copper-foil conductors, a finite-element model at 14 MHz showed conductor loss falling as 0.008-inch copper foil was widened from 0.125 inch to 1 inch. The improvement did not scale as simply as DC cross-sectional area because RF current was not distributed uniformly across the strip.

That result supports a bounded statement: conductor width can reduce AC loss in a declared material, thickness, frequency and current geometry. It does not prove that every conductive fabric has lower loss than every wire, or that lower strip resistance becomes an equal reduction in soil loss for the completed antenna.

A ground-contact strip can also change distributed capacitance, soil current density, radial resonance, feedpoint impedance and current division. Its fabric, plating, seams, grommets and contact resistance matter. Laying two strips in an X creates four physical arms, but that does not make them electrically equivalent to four independent wires—or to eight, sixteen or thirty-two wires. Measure the current instead of counting shapes.

What Rudy Severns Actually Measured

The N6LF experiments are often reduced to “more short radials are better.” Rudy's data say something much more useful and much less universal.

In Experimental Determination of Ground System Performance for HF Verticals, Part 2, one test used a 34-foot aluminium vertical at 7.2 MHz with four insulated surface radials varied from 33 feet to 18 feet. Feedpoint impedance, relative transmission and radial-current division were measured, and a common-mode choke isolated the antenna from the feedline and ground. Shortening those four radials towards roughly 23 feet improved the measured result in that installation because the sparse soil-loaded screen had a loss-producing resonance.

In the follow-on comparison, Rudy re-resonated an adjustable vertical as 4, 8, 16 and 32 radial fields were changed between 33-foot and 21-foot wires. The shortening advantage diminished as count increased. With 32 radials, the longer set was 0.12 dB ahead in that particular experiment. That is the opposite of a universal “shorter always wins” rule.

In Part 4, a fixed-wire comparison on a top-loaded one-eighth-wave vertical at 7.2 MHz found nearly the same relative signal from sixteen 33-foot radials and thirty-two 17-foot radials. The feedpoint impedances differed, and the radiator was lengthened to restore resonance for the shorter set. More shorter wires can therefore be an effective allocation of a fixed conductor budget in a particular installation. They are not automatically superior.

Elevated and Surface Radials Are Different Systems

Another N6LF result is routinely detached from its conditions. Part 3 of the 2009 series compared a 64-radial surface system with four carefully balanced radials elevated 48 inches in one controlled 7.2 MHz installation. Their normalised transmission results differed by only 0.1 dB.

That experiment does not establish a fixed 0.05–0.1λ height or prove that four elevated radials always equal a dense surface field. In his 2012 elevated-ground-system study, Rudy stressed the sensitivity of a four-radial fan to asymmetry, soil and nearby conductors. He recommended 10–12 or more elevated radials for robustness in that design context.

Surface radials couple strongly to the ground. Elevated radials are coupled antenna elements whose current balance, resonance and geometry can reshape the pattern. A count taken from one cannot be assigned to the other.

There Is No Fixed Loss Table by Radial Count

Four, eight, sixteen or thirty-two wires do not have an intrinsic loss in decibels. The result depends on radiator current, frequency, radial length and distribution, soil, matching network, conductor loss, feedline common mode and the measurement reference plane.

Radiation efficiency is a power ratio:

To calculate that ratio, we must know or estimate the radiated power and every relevant loss term. Feedpoint resistance is not automatically “ideal radiation resistance plus ground loss,” because the radiation resistance and current reference can change with the installed ground system. A low SWR says nothing by itself about η.

A Fair Strip-versus-Wire Test

1. Define the comparison. Equal conductor mass, equal footprint, equal deployment time, equal cost and equal radial count are different engineering questions.
2. Hold the radiator fixed. Record radiator length, height, loading, matching network, feedpoint and nearby conductors.
3. Control accepted power. Measure at a declared reference plane and correct for feedline and matching loss.
4. Control the return boundary. Characterise the choke and map current on the coax exterior before and after it.
5. Record the radial system. Include material, resistance, strip or wire dimensions, orientation, contact condition, soil and moisture.
6. Map current. Compare branch current and phase where practical; an input match can hide unequal or unintended return paths.
7. Measure remote field consistently. Use the same calibrated probe, distance, height, bearing and polarisation; test more than one bearing if pattern change is possible.
8. Use A/B/A restoration. Reinstall the baseline and confirm that time, weather, soil or placement did not manufacture the improvement.
9. Publish uncertainty and raw ratios. A median without sample count, spread or measurement uncertainty is not enough to judge repeatability.

Portable Value Is Not Only a Decibel

A radial system can earn its place because it deploys quickly, stays organised, is visible around people, survives repeated use or recreates the same geometry at every site. Those are legitimate system requirements. Weight, packed volume, drying, repair and trip management are legitimate costs.

Keep those mechanical advantages separate from RF proof. “Convenient and repeatable” can be the deciding result without inflating a conditional field comparison into a universal efficiency claim.

Primary Evidence

• Chameleon Antenna — CHA FSR product page: manufacturer wording, comparator, strip layout and reported 23–28% relative field-strength result.
• NIST Special Publication 250-32: decibel relationships for power and equal-impedance voltage ratios.
• ITU-R SM.329-8: field-strength and power-flux-density units and relationships.
• ARRL — Test Procedures to Measure BPL Interference: field-strength measurement geometry, probes and S-meter limitations.
• Rudy Severns, N6LF — Resistance of Foil Conductors for Antennas: frequency-, width-, material- and current-distribution boundaries for foil loss.
• Rudy Severns, N6LF — Ground System Performance for HF Verticals, Part 2: sparse surface radial length, count, resonance and current measurements.
• Rudy Severns, N6LF — Ground System Performance for HF Verticals, Part 4: radial-count and fixed-wire comparisons.
• ITU-R P.527-6 — Electrical Characteristics of the Surface of the Earth: frequency-, material-, moisture- and temperature-dependent ground properties.

Practical Conclusion

A reported 1.8–2.1 dB field-amplitude difference would be measurable and potentially useful. It is not “nothing.” But the engineering question remains: relative to what, at which frequency, with which radiator, return path, accepted power, soil, probe position and uncertainty?

I will not pretend that a strip cannot outperform four short wires. I will also not turn that comparison into “strips are more efficient” or answer it with the opposite folklore that more short wire always wins. Width, length, count, material, geometry and soil all move the installed currents. Put the baseline and the measurement method beside the percentage; then the number can finally speak.

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