Choosing Between Near-Resonant and Non-Resonant End-Fed Wires
Choosing Between Near-Resonant and Non-Resonant End-Fed Wires
If the available supports let me put the important bands near useful antenna modes with manageable feedpoint loads, that is where I start. It gives the matching hardware less work to do and can simplify the station. An arbitrary wire length with a suitable tuner is a valid alternative when the site or frequency coverage demands it—but I would not discard a favourable wire layout just because a tuner can match something else.
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.
My preference is to solve what I can in the antenna geometry before asking a transformer and tuner to solve it electrically. On a few priority bands, a well-chosen mode and feed arrangement can leave a modest resistance conversion and little reactance to compensate. That is a useful engineering advantage: less matching burden, potentially fewer components in the power path, and simpler operation. The advantage comes from those actual loads—not from the word “resonant” on its own.
My starting choice: when the layout is adjustable and the required bands are known, choose useful modes with manageable resistance and reactance before choosing a wide-range matching system. Keep a non-resonant wire and feedpoint coupler for the situations where their flexibility solves a real site or coverage problem. Measure the finished system to confirm the advantage you designed for.
What “Near Resonance” Actually Says
At a declared antenna-side reference plane, write the load as ZA = RA + jXA. Resonance occurs where the reactance XA passes through zero. “Near resonance” describes operation close to that condition; it is not another name for any impedance a tuner can match. For this design choice, I want both little residual reactance and a resistance that the chosen network can handle comfortably. Those are separate requirements, and neither is a universal SWR threshold.
A resonant resistance can be far from 50 Ω. A matching network can also present 50 Ω to the transmitter while the wire is not resonant. Resonance, transmitter match, low SWR and radiation efficiency are four different results. Keeping them separate prevents a convenient match from becoming an imaginary efficiency certificate.
On a multiband wire, the useful resonances are modes of the complete installed structure—not simply whole-number multiples printed beside a wire length. Height, slope, bends, conductor diameter, insulation, the return branch, coax exterior, soil and nearby metal all move the measured impedance.
Why I Prefer to Reduce the Matching Burden
Suppose the useful wire layout brings the priority-band load into a region needing only modest resistance conversion and reactance compensation. We can then choose the network for that region, instead of demanding a very wide transformation range from it. If that arrangement also gives an acceptable load at the coax entrance, we avoid carrying a severe mismatch through a long lossy cable. Those are concrete ways to remove avoidable loss and electrical stress from the system.
What less reactance actually buys: consider a load R + jX, with positive R, accepting real power P. Its RMS current is I = √(P/R). An ideal series element cancelling X must support a reactive voltage of magnitude I|X|. At the same R and accepted power, reducing |X| reduces that compensating voltage—not the load current. Real components also have resistance and dielectric loss, so a simpler, less demanding match can be useful without treating reactance itself as heat.
Changing a wire also changes its resistance and current distribution, so that example isolates a mechanism; it is not a measurement of two antennas. A resonance at an extreme feedpoint resistance may still demand substantial transformation. I am looking for a useful mode and a workable feed arrangement, not the deepest analyser dip at any cost. Analog Devices' matching examples show why resistance conversion and compensation of the load's reactance must be considered together.
What a Deliberately Non-Resonant Wire Offers
A non-resonant end-fed wire deliberately accepts a complex load and asks a tuner, sometimes together with a broadband transformer, to convert it to the transmitter's required impedance. That can be an excellent answer when supports limit the available length, one wire must cover many frequencies or a remote coupler can sit directly at the antenna terminal.
The word “non-resonant” does not mean unpredictable or inefficient. A fixed geometry produces repeatable impedance and current distributions in a stable environment. It does mean that the matching system must have enough range and margin for the actual loads, and that its losses belong in the result.
A remote tuner at the antenna feedpoint can keep a long coax run near its design impedance. A shack tuner may instead leave high standing-wave voltage and current on the transmission line between tuner and antenna. Real feedline then dissipates more power than its matched-loss figure alone suggests. This is an architecture question, not a verdict against non-resonant wires.
Transformer Ratios Are Starting Hypotheses
A nominal 4:1 or 9:1 impedance ratio describes an ideal transformation. It does not tune away arbitrary reactance, and it does not guarantee that the resulting load lies inside a tuner or transmitter range. The correct ratio follows the measured load region, not the wire's marketing category.
A real UNUN adds magnetizing impedance, leakage inductance, winding capacitance, copper resistance and core loss. High load resistance may create difficult voltage; low resistance may create difficult current; large reactance may raise circulating energy. Frequency, power, modulation duty cycle, enclosure and cooling decide whether a network remains within its electrical and thermal limits.
My aim is therefore not to bolt a favourite ratio onto every wire. It is to choose the wire, return branch and feed arrangement so the important bands fall inside a sensible network operating window. A near-resonant wire can be easier on one band yet awkward on another, while a non-resonant wire can be gentle when its coupler sees a favourable load. Check the complete load sweep and representative complex loads before assigning power or efficiency limits.
Count Loss at the Same Reference Plane
Use the same input boundary when comparing two systems. A useful loss ledger includes:
- Matching network: component resistance, dielectric loss and circulating current or voltage under the installed load.
- Transformer: winding, core, leakage and stray-field loss under frequency, power and thermal conditions.
- Feedline: matched attenuation plus the additional dissipation created by the actual standing-wave distribution.
- Return network: intentional counterpoise, coax exterior, mast, soil and capacitive coupling to surrounding structures.
- Radiator: conductor, loading, joints and nearby-material loss before accepted power becomes radiation.
Radiation efficiency is radiated power divided by accepted power at a stated antenna boundary. System efficiency may include upstream transformer, tuner and feedline losses. Both can be useful, but they are not interchangeable. A low SWR at the radio reports neither value.
| Design question | Near-resonant starting point | Non-resonant starting point |
|---|---|---|
| Matching burden | May be modest on selected measured modes; no guarantee of 50 Ω. | Usually expects a wider transformation range; actual burden follows the complex load. |
| Feedline choice | Coax may be practical where the matched load and loss are acceptable. | A feedpoint coupler or low-loss line may prevent a severe mismatch from occupying a long lossy coax run. |
| Band coverage | Chosen modes may simplify operation on priority bands. | A suitable coupler may reach frequencies between or beyond those modes. |
| Opportunity to reduce loss | A favourable load can simplify the match and reduce avoidable network or mismatched-line loss; resonance alone does not establish radiation efficiency. | A suitable low-loss feedpoint coupler can provide flexibility without putting a severe mismatch on the long coax run. |
| Pattern | Follows the installed current distribution on each band. | Follows the same physics; electrical length and environment can create several lobes and nulls. |
| Best evidence | Complex load, network loss and stress, exterior-current map, accepted-power pattern and repeatable field comparison. | |
Every End-Fed Wire Has a Return Branch
Current leaving one terminal must return to the other. A drawing that shows only the main wire has not removed the return path; it has left that path to the coax exterior, a dedicated counterpoise, mast, station wiring, soil and distributed capacitance.
That return structure changes the impedance, common-mode current and radiation pattern. It can be intentional: a declared section of coax exterior may form an antenna branch between transformer and choke. If so, the choke belongs at the planned current boundary and needs adequate measured common-mode impedance on every operating band. Placing a choke automatically at the transformer can remove a needed return branch; placing it only at the shack can let the whole feedline and station participate.
Transformation and common-mode suppression remain separate functions. The UNUN handles the differential impedance conversion for which it was selected. A separately specified choke controls exterior current at the chosen boundary. Neither function proves the other.
Pattern Is Not Promised by the Matching Category
At HF, the installed current distribution, electrical length, height, orientation, bends, ground and nearby structures form the azimuth and elevation pattern. A wire that is near one resonance may have a simple current distribution on that band and several current maxima on a higher band. A non-resonant wire can do the same as its electrical length changes.
Therefore “near resonant” does not guarantee a repeatable DX lobe, and “non-resonant” does not mean that lobes move randomly. Both patterns can be modelled when the complete geometry and environment are known. Both should be checked on every intended band rather than inferred from SWR.
Adding a long return wire to obtain another band changes the antenna into a different two-branch structure. It may alter impedance and pattern usefully, but no fixed added length guarantees top-band coverage, NVIS performance or efficient full-HF operation. Model and measure the resulting system as a new antenna.
Where Each Approach Earns Its Place
For a station with adjustable supports and a defined set of favourite bands, I would first spend the effort on a useful wire layout. If its installed loads suit a modest transformer or matching network and its patterns serve those paths, I gain something practical: less tuning range to provide, potentially fewer boxes, fewer connectors to weatherproof and fewer settings to manage. A tuner may still be needed; the objective is to give it an easier job, not to win a tuner-free badge.
If the supports dictate the wire length, or coverage between the useful modes matters more than simplifying a few bands, I would choose the non-resonant route with an appropriate feedpoint coupler and return structure. That is a deliberate trade: matching-system flexibility in exchange for freedom in wire length and frequency. The tuner is part of the design, not an embarrassing accessory. Nor does automatic tuning necessarily make band changes slow.
Convenience is also a valid design objective. Fast band changes, available supports, weather protection, access for maintenance and safe voltage clearance may decide the better antenna even when two systems have similar RF performance.
Confirm the Advantage in the Installed System
- Document the complete geometry: main wire, return branch, feedline route, choke, mast, bonds and nearby conductors.
- Measure the antenna-side complex impedance: record resistance and reactance across every intended band at a declared plane.
- Test the matching hardware: reproduce the expected loads and verify transformation, loss, voltage, current and temperature at representative power and duty cycle.
- Account for the line: use its length, frequency-dependent matched loss and actual load rather than a generic cable claim.
- Map exterior current: confirm which return branch is intentional and where the choke establishes the boundary.
- Check radiation evidence: use a complete-geometry model or controlled accepted-power A/B/B/A measurements, and restore the baseline when propagation changes.
My choice: use the available geometry to get the priority bands near useful modes with manageable loads, then fit the least demanding matching arrangement that serves them. That is why I prefer this starting point to an arbitrary long wire when the site gives me the choice. It can remove avoidable matching and feedline burden while making the station simpler. When the site or coverage makes that choice impractical, a non-resonant wire with a well-chosen feedpoint coupler is the purposeful alternative—not a failure to build a “proper” antenna.
Primary technical references
- IEEE Std 145-2025 — Standard for Definitions of Terms for Antennas
- Recommendation ITU-R BS.705-2 — HF transmitting and receiving antenna characteristics and diagrams
- Keysight — Impedance Measurement Handbook, sixth edition
- Analog Devices — RF Impedance Matching: Calculations and Simulations
- Icom AH-730 — Official tuner range and antenna-condition specifications
- Recommendation ITU-T K.136 — RF electromagnetic fields along telecommunications conductors
Mini-FAQ
- Does resonance mean that an antenna is efficient? No. Resonance says that reactance is zero at a declared reference plane. Conductor, ground, transformer, tuner and feedline losses still decide efficiency.
- Does a near-resonant wire work without a tuner? Not necessarily. Resonant resistance may be far from 50 Ω, and the installed geometry can move both resistance and reactance.
- Is a 9:1 UNUN always correct for a non-resonant wire? No. The ratio must follow the measured complex load and the verified range, loss and stress of the complete matching system.
- Is a non-resonant wire inevitably lossier? No. A suitable low-loss feedpoint coupler and return structure can perform very well. Loss must be measured for the installed architecture.
- Should the common-mode choke always sit at the transformer? No. It belongs at the intended end of the antenna-side exterior-current branch, with sufficient measured impedance on each band.
- Which type is better for multiband operation? When the layout is adjustable and the priority bands are known, I start with useful modes and manageable loads to simplify matching. When the site fixes the wire or wider frequency coverage matters, a non-resonant wire with a suitable feedpoint coupler is the practical alternative. Check loss, stress, return currents and patterns in either installation.