Antenna Current Distribution: Impedance, Pattern and Efficiency

I keep returning to one practical question: where does the RF current actually flow? For a permanent station, I generally favour a near-resonant, moderate-impedance off-centre-fed arrangement when the site lets me put its current-rich span high and clear and provide a deliberate return. That is a design choice with reasons—not a claim that an EFHW cannot work well.

The Current Plot Is a Result, Not a Rule

A familiar half-wave-dipole sketch shows a broad current maximum near the centre and current approaching zero at the open ends. That is a valuable first model for a thin, straight, isolated, centre-fed conductor near its fundamental resonance. It is not a stencil that can be pasted onto every bent, loaded, multiband or environmentally coupled antenna.

The installed distribution is a complex quantity: current has both magnitude and phase. It is determined by electrical length, conductor diameter and loss, bends, sources, loading and matching elements, ground, nearby objects and every available return path. At harmonic operation, a long wire may carry several current lobes with phase reversals. Moving or loading one section can reshape the whole solution.

At an ideal open end, axial conduction current approaches zero while charge and electric field can be large. That region is not irrelevant: it can couple capacitively to the ground, foliage, roofs, gutters, support lines and other conductors. Those boundary conditions can change resonance, current distribution and loss even though the local conduction current is small.

Current and Voltage Maxima Do Not Name the Feedpoint Impedance

The input impedance is the complex ratio of voltage to current at a declared port and reference plane. Feeding an ideal half-wave wire near its centre commonly produces a lower resistance than feeding it near an end, but “centre”, “off-centre” and “end-fed” are geometry descriptions—not guaranteed impedance ratios.

End effects, wire thickness, height, ground, conductor loss, nearby structures and the feed arrangement all move the result. On another band, the same wire can support a different standing-wave pattern, so the feedpoint can land near a different combination of current and voltage. A fixed transformer ratio chosen from the antenna’s name can therefore be a poor substitute for measuring the installed complex load across the required band.

Every Feed Current Has a Return

Current cannot leave a source terminal without an equal circuital return. In a balanced dipole that return is intended to flow on the opposite arm. In an asymmetric or incompletely balanced installation, part of it may use an intentional counterpoise, radial system, coax exterior, mast, bonding conductor, soil, station wiring or capacitance to the surroundings.

That return path is part of the antenna system whether it was drawn or not. Current on the outside of a coaxial shield can radiate, receive local noise, alter the pattern and bring RF into the station. It also changes when the feedline is rerouted or the station is re-bonded. A choke can establish a useful common-mode boundary, but its impedance is frequency- and installation-dependent; the completed feed system must be measured. Do not use touching transmitting equipment as a test, and de-energise the system before changing connections or conductor positions.

This is also why a feedpoint-current comparison is incomplete unless the other conductors are inspected. Two installations may show similar current on the wire yet differ materially because one has a controlled return branch and the other recruits the feedline and building.

Loading and Matching Reshape the Distribution

Loading coils, traps, capacitive hats, folded conductors and matching networks do more than change an SWR curve. They introduce reactance, loss and discontinuities that redistribute current and voltage. A coil placed in a current-rich region may carry substantial RF current and dissipate meaningful power; a capacitive element near a voltage-rich region may face strong electric-field stress. Neither statement supplies a universal ranking—the component values, Q, position, current, voltage, duty cycle and cooling determine the result.

A tuner can transform the impedance at its own terminals without undoing loss elsewhere in the installation. Feedline attenuation under mismatch, transformer loss, conductor and connection loss, ground loss and common-mode dissipation all remain part of the accepted-power budget. Matching and radiation are related through the installed system, but they are not synonyms.

Height and Ground Change More Than Loss

Physical height becomes electrically meaningful only when stated in wavelengths. The same ten-metre support represents a very different electrical height on 80 metres than on 20 metres. Ground conductivity and permittivity, terrain, antenna orientation and complete current distribution then set the elevation pattern; no fixed height alone guarantees “NVIS” or “DX”.

A current-rich section close to lossy soil or conductive clutter deserves attention because both coupling and conductor current are present there. But it is too simple to say that this section alone controls radiation. Low-current, high-voltage ends can couple strongly to nearby objects, and any induced current on those objects can become part of the field and loss solution. The complete geometry must be solved or measured.

Ground also plays several different roles. A radial or counterpoise system can carry intentional RF return current. Soil can absorb field energy. Protective earthing and lightning bonding exist primarily for safety. Treating all of these as one generic “ground” hides the paths that the RF current actually takes.

Pattern, Gain and Efficiency Are Different Questions

The current distribution generates the radiation pattern, but pattern shape alone does not give efficiency. Directivity describes how radiation is concentrated in angle. Radiation efficiency compares radiated power with power accepted by the antenna system at a stated port. Gain combines directionality with efficiency under the applicable convention.

For peak-phasor current, a first-order distributed conductor-loss calculation can be written as:

Here, s is distance along a conductor, R′(s) its RF resistance per unit length and I(s) the peak-phasor current there. Sum the conductor contributions, then account separately for matching components, loading elements, connections and field absorption in soil or other lossy materials. Those losses can dominate a budget that looked excellent when only the radiator wire was considered.

Similarly, radiation resistance is referred to a port current. Moving the feedpoint changes that reference current and the input impedance; it does not, by itself, prove that the underlying structure radiates more or less efficiently. Compare designs at the same accepted power and with the same definition of the system boundary.

Why I Prefer Moderate-Impedance Feeding for a Fixed Station

An EFHW—an end-fed half-wave—feeds near a high-voltage, low-current end of the fundamental mode. An off-centre feed can instead use a more current-rich point and present a moderate impedance. In an end-fed off-centre-fed arrangement, or EF-OCF, consider the visible wire and the intended return conductor as the complete structure: the end of the visible wire is not necessarily the end of the current system.

That distinction offers a practical design route. Arrange the complete antenna to produce a manageable load, use the transformation that load needs, and control the return separately. I would rather start there for a permanent installation than accept a very high feed impedance and ask a large-ratio transformer to solve everything. The ratio must still follow the installed complex load on each band; “off-centre-fed” does not automatically mean 200 Ω or a 4:1 match.

A Smaller Transformation Has a Concrete Benefit

Take two ideal, purely resistive examples fed from 50 Ω. A 200 Ω load needs a 4:1 impedance transformation, equivalent to a 2:1 voltage ratio. A 2450 Ω load needs 49:1, equivalent to 7:1. These follow the ideal transformer relationship: impedance ratio is the square of voltage or turns ratio.

The moderate-impedance example needs much less terminal-voltage rise. That eases the high-side insulation and electric-field burden and gives the matching designer a less extreme transformation to implement. The corresponding load current is higher—about 0.71 A rather than 0.20 A—so conductor and connection losses still deserve attention. These are circuit calculations, not measured antenna or transformer performance.

Nor does “fewer turns” automatically mean lower core loss. Flux depends on voltage per turn, frequency and core geometry; magnetising inductance, leakage, winding capacitance, material loss and cooling still constrain a real transformer. The advantage is a less demanding transformation requirement and lower terminal voltage for the same accepted power—not a free efficiency certificate. Use that opportunity to build a good matching network, rather than assuming the number on the box has done the work.

Put the Current Where the Site Can Use It

Matching is only part of my preference. An off-centre-fed layout can let a particular site keep the useful current-rich span higher and clearer while making the feed and return practical to route. If an alternative leaves that span close to a roof, foliage or lossy ground, changing the arrangement can improve the field and reduce unwanted coupling. The benefit follows the actual geometry; moving the feed alone does not magically relocate all the current or guarantee a lower take-off angle.

I also want an intentional return branch and a defined boundary before the station feedline. W8JI’s end-fed current-path analysis is useful here: the return cannot simply be omitted from the antenna drawing. An EF-OCF still needs it, and a centre-fed wire can develop common mode in asymmetric surroundings. Transformation and common-mode suppression are different jobs. A well-designed EFHW with a deliberate counterpoise and effective choke can control its return too.

A good EFHW can beat a poor off-centre-fed installation. Give the EFHW a sound transformer, a controlled return and a high, clear current-rich section, and it may be the better antenna at that site. My fixed-station preference is for the moderate-impedance arrangement when it makes those same jobs easier and removes avoidable matching burden. For portable work, one convenient support, rapid deployment or an already excellent EFHW installation may outweigh that preference. On multiple bands, judge each useful mode; neither topology supplies one pattern or one feed impedance everywhere.

Compare the alternatives under the same frequency and site constraints, recording their different geometries rather than pretending those differences do not matter. Equal accepted power at the antenna ports compares radiation behaviour; equal power at the station connector also exposes feedline and matching losses. State which question the comparison answers.

Build a Current-Path Evidence Chain

LLNL’s Numerical Electromagnetics Code can model conductors, ground, loads, networks and transmission lines and report currents, near fields and radiation patterns. That makes it a useful engineering bridge between a current-path hypothesis and a field prediction. The model is only as complete as its geometry, source, material, junction, ground and feed assumptions.

• Declare the reference plane. State whether impedance and accepted power refer to the antenna port, transformer input, tuner output or transmitter connector.
• Include the return structure. Model or measure the second antenna terminal, radials or counterpoise, coax exterior, mast, bonding and nearby conductors that carry material current.
• Inspect magnitude and phase. Current magnitude identifies stress and loss candidates; relative phase is essential to pattern reinforcement and cancellation.
• Characterise matching parts separately. Measure insertion loss, heating and common-mode impedance across the actual complex-load and frequency range.
• Close the power budget. Compare incident, reflected, accepted, dissipated and radiated power using consistent conventions.
• Validate the field. Use a controlled range, calibrated field measurements or a repeatable multi-site/on-air method. Keep propagation, polarization, transmit power and time variation inside the uncertainty statement.
• Repeat after one change. A/B/A tests of height, feedline route, choke position or loading reveal whether the mechanism follows the prediction.

IEEE 149 treats radiation pattern as a fundamental antenna property and sets out measurement-facility and instrumentation practices. NIST’s antenna-measurement roadmap likewise emphasizes experimental validation, calibration and uncertainty, particularly where cables, fixtures and the environment can become part of the measurement. One low SWR reading is not that evidence chain.

Primary and Authoritative References

• Mini-Circuits — ideal transformer relationships and real loss mechanisms
• Tom Rauch, W8JI — end-fed wires and their complete return-current paths
• IEEE 145-2025 — IEEE Standard for Definitions of Terms for Antennas
• IEEE 149-2021 — IEEE Recommended Practice for Antenna Measurements
• Lawrence Livermore National Laboratory — Numerical Electromagnetics Code, version 5
• NISTIR 3989 — dipole-current approximations and measurement boundaries
• NIST-linked roadmap — Antenna Measurement Challenges and Opportunities
• Burke and Poggio — Numerical Electromagnetics Code, method-of-moments theory

Practical Conclusion

A current maximum is not a magic patch of wire, and a voltage maximum is not dead space. The antenna is the coherent current distribution over the complete installed structure, including the intended return and any conductor recruited by accident.

For a fixed station, my starting preference remains a near-resonant, moderate-impedance off-centre-fed arrangement when it fits the available supports and return path. It gives me a direct way to combine useful current placement, a smaller transformation requirement and a feedline that does not have to become an accidental antenna. Keep a well-performing EFHW when it solves the site better. The choice is about which arrangement delivers those engineering benefits in your installation—not about awarding a winner by name.

It is not where the wire is—it is where the current flows. Follow all of it, keep its phase, and judge the result by measured loss and field rather than by the antenna’s name or SWR alone.

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