EFHW on 20–10 m: The Ground-Sensitivity Problem
EFHW on 20–10 m: The Ground-Sensitivity Problem
A multiband end-fed half-wave can look calm on its lowest band yet move when rain, nearby objects or the feed-line route changes on 20, 15 or 10 metres. The useful question is which part of the complete RF path moved.
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.
I have seen multiband EFHW installations that barely change on a lower band but become noticeably sensitive to weather, cable routing or a person near the wire on 20–10 m. That observation is worth keeping. The explanation is more interesting than “everything is a larger fraction of a wavelength”: harmonic current distribution, local electric and magnetic coupling, transformer parasitics, the return path and the instrument plane all change together.
One Wire, Several Electrical Antennas
A wire close to a half wavelength on its lowest operating band is several half wavelengths long on harmonic bands. Its current and voltage distributions therefore acquire additional maxima and minima. A gutter, branch, wall or support does not perturb every point equally: coupling near a voltage maximum is mainly sensitive to added capacitance and leakage, while coupling near a current maximum can change current sharing, conductor loss and the local magnetic field.
The installed distribution is not the ideal sinusoid from a sketch. Wire diameter, insulation, slope, bends, height, termination hardware, soil, nearby conductors and the low-side return path all affect electrical length and impedance. The Numerical Electromagnetics Code developed at Lawrence Livermore models precisely this kind of conductor-and-environment interaction; a useful model must include the real geometry rather than only a wire length.
The practical observation can be real without being universal. An upper harmonic can be more sensitive because a local object lands near an important voltage or current region, because a parasitic branch becomes electrically significant, or because the feed line and choke have moved to a different common-mode state. Frequency alone does not identify which one.
A Few Picofarads Need a Circuit and a Location
The reactance magnitude of a capacitor is:
|XC| = 1 / (2πfC)
For example, 5 pF has a reactance magnitude of about 2.24 kΩ at 14.2 MHz and 1.12 kΩ at 28.5 MHz. Those are useful scale checks, not predictions of a fixed feedpoint change. The result depends on whether that capacitance is shunt or series, where it couples along the wire, the local RF voltage, the transformer network, the return path and every other complex impedance in the system.
Rain can add surface leakage and change capacitance around insulation, ropes, leaves and enclosure surfaces. Wet ground can change permittivity and conductivity; ITU-R P.527 publishes frequency-dependent electrical properties for wet, medium-dry and very dry ground. Those changes can move resonance, loss or pattern on any band. Their size is installation-specific and must be measured rather than assigned a fixed number of ohms.
The Transformer Has Its Own Upper-Band Story
A turns ratio gives an ideal impedance ratio of the turns ratio squared. It does not establish one universal EFHW feed impedance, and it does not make either 49:1 or 64:1 correct for every wire and band. The antenna terminal presents a frequency-dependent complex load, while the realised transformer also contains finite magnetising impedance, core loss, leakage inductance, winding resistance, distributed capacitance, lead and connector reactance and the low-side return path.
At the upper bands, leakage inductance and winding capacitance can create resonances or anti-resonances and can change the transformation. A compensation capacitor is a legitimate network element when it cancels a defined part of the measured reactance over a declared range. It is neither a universal cure nor proof that a problem is being concealed.
Ferrite loss does not follow one monotonic “higher band means more loss” rule. TDK, Ferroxcube and Fair-Rite publish complex permeability and loss information that depends on material, core geometry, frequency, flux density and temperature. Conductor skin and proximity effects, dielectric loss and contact resistance add other terms. A single imaginary 50 Ω element placed beside a nominal multi-kilohm antenna load is not a valid efficiency model for the completed transformer.
| Observation | Plausible mechanisms | Measurement that separates them |
|---|---|---|
| Resonance moves after rain | Wet insulation or foliage, changed ground properties, enclosure leakage, altered return path | Repeat dry–wet–dry impedance sweeps at the same plane and cable route; record weather and surface condition |
| Upper-band SWR changes when coax is moved | Exterior common-mode standing wave, changed coupling, analyzer or station becoming part of the return | Mark the route; measure net cable current at several positions before and after one controlled route change |
| Transformer warms on one band | Core, winding, dielectric or contact loss; a difficult complex load; a parasitic resonance | Calibrated input/output loss or calorimetry, temperature at declared points and a representative-load sweep |
| A nearby object changes only one band | Object near a voltage/current maximum or one pattern lobe; band-specific common-mode state | Move only that object in A/B/A order and compare complex impedance, cable current and field observations |
The Coax Exterior Is Part of the Experiment
An end-fed radiator needs a complete RF circuit. The low-side current may use a deliberate counterpoise, the outside of the coax shield, a mast, station wiring, distributed capacitance to the surroundings, or a combination. The feed line is electrically longer on the upper bands, so its exterior current can pass through different standing-wave maxima and minima as frequency changes.
A common-mode choke adds finite complex impedance at one chosen boundary. It cannot make current vanish, and it may have its own parallel resonance, voltage stress and loss. Moving the choke changes the allowed return path, so the antenna input can change even when the radiator wire has not moved. Measure net coax current on both sides and at several marked positions; one clamp reading at one point cannot describe the whole exterior-mode distribution.
This is also why a different cable length can appear to “fix” an upper band. It may simply transform the exterior-mode impedance at the transformer or at the station. If the result depends on an undocumented cable length and route, it is not yet a controlled antenna configuration.
The Analyzer Plane Can Move the Answer
A one-port VNA reports the impedance at its calibration plane. A length of coax transforms the load seen there; connectors and fixtures add their own errors, and bending the cable can change the common-mode return. Keysight and NIST measurement guidance both treat calibration plane and de-embedding as part of the result, not as administrative detail.
Calibrate at the plane you intend to report, or characterize and de-embed the intervening fixture. Keep the analyzer, cable route, choke position and nearby conductors fixed. Remember that a handheld analyzer, a transceiver and a station earth arrangement can provide different low-side impedances. Agreement is meaningful only when those boundaries are declared.
A Multiband Test That Survives the Weather
- Freeze the geometry. Record wire route, height, bends, supports, transformer orientation, counterpoise, coax type and length, choke position and station bonding.
- Set the measurement plane. Calibrate at the transformer input or state the cable and de-embedding used to move the plane.
- Save complex baselines. On every intended band, store R + jX, S11, resonance and bandwidth rather than only minimum SWR.
- Map common-mode current. Use a calibrated RF current probe at repeatable marked positions on the coax, counterpoise and other accessible conductors.
- Characterize the transformer separately. Sweep representative complex loads, measure loss at calibrated planes and monitor core, winding and connector temperatures at equal accepted power and declared duty cycle.
- Change one variable in A/B/A order. Test one cable route, choke position, nearby object, height or moisture condition, then restore the baseline and verify that the result returns.
- Record the environment. Note rain state, soil and foliage condition, ambient temperature and who or what is near the antenna.
- Compare antennas fairly. If testing an off-centre-fed or centre-fed alternative, use the same band, time, site, receiver state and accepted power. Different architecture is not automatic proof of greater stability or efficiency.
RF safety still sets the boundary. An EFHW end and transformer high-side can carry hazardous RF voltage. Keep them inaccessible, never touch or adjust the system while transmitting, isolate test equipment before applying power and discharge stored charge before service.
My conclusion: when a multiband EFHW wanders on 20–10 m, do not blame wavelength, ground or ferrite by slogan. Preserve the observation, then identify which current distribution, parasitic branch, return path, weather condition or measurement plane actually changed.
Primary engineering references
- Lawrence Livermore National Laboratory — Numerical Electromagnetics Code capabilities for wires, insulation, conductors and finite ground
- ITU-R P.527-6 — Electrical characteristics of the surface of the Earth
- C. L. Ruthroff — Some Broad-Band Transformers
- Ferroxcube — Soft Ferrites and wideband-transformer equivalent circuits
- TDK Electronics — Ferrite general definitions and complex permeability
- TDK Electronics — Ferrite materials and power-loss conditions
- Fair-Rite — Ferrite materials, impedance and transformer guidance
- Keysight — Impedance Measurement Handbook
- Keysight — Fixture de-embedding and calibrated reference planes
- NIST — Moving calibrated measurement planes through characterized interfaces
- ARRL — RF grounding and complete antenna-current paths
- ARRL QEX — Asymmetric antenna loads and common-mode impedance
- IEEE EMC Society — Common- and differential-mode current on multiconductor structures
Mini-FAQ
- Why can an EFHW appear more sensitive on 20–10 m? Harmonic current and voltage distributions add more maxima and minima, while transformer parasitics, local coupling and the feed-line return path can enter different states. It is plausible, but not universal.
- Do a few picofarads always move the feed impedance by hundreds of ohms? No. Capacitance has a frequency-dependent reactance, but the input change also depends on its circuit connection, position, local RF voltage and the complete complex network.
- Does ferrite loss always rise on every higher band? No. Complex permeability and loss depend on material, core geometry, frequency, flux density and temperature; winding, dielectric, contact and parasitic losses must also be measured.
- Is 49:1 or 64:1 always the correct EFHW impedance ratio? No. The ideal ratio is only a starting point. The installed antenna presents a band-dependent complex load, and the real transformer and return path alter the realised input.
- Where should I measure an upper-band change? At a declared calibrated plane, with cable route and choke fixed. Save complex impedance and map net feed-line current at repeatable positions before and after one A/B/A change.
- Is an off-centre-fed antenna automatically more stable or efficient? No. Its feedpoint and current distribution differ, but transformer loss, imbalance, common mode, environment and installation still decide the measured result.