20 m Verticals: EFHW or Quarter-Wave?
20 m Verticals: EFHW or Quarter-Wave?
A quarter-wave monopole usually gets the easier engineering job: high feed current, modest terminal impedance and an explicit return system. A vertical end-fed half-wave can work very well too—but only when its transformer, return path, ground clearance and feedline boundary are treated as parts of the antenna.
I keep seeing these two antennas compared as if the extra wire in the EFHW must buy extra performance, or as if the radials under the quarter-wave make it primitive. Neither shortcut survives contact with the complete current path. The useful question is not which label wins. It is which installation puts more accepted power into the required pattern, with less loss and less unwanted current on everything around it.
Start With Two Complete Antennas
Near 14 MHz, a free-space quarter wavelength is about 5.35 m and a half wavelength about 10.7 m. Those are scale markers, not cutting dimensions. Conductor diameter, insulation, end effects, bends, height, ground and nearby material all move the installed electrical length.
The quarter-wave version is a monopole. Its vertical conductor is only one side of the driven system; elevated radials, an on-ground radial network, a conductive structure or another engineered return completes it. The familiar resistance near 36 Ω belongs to a thin straight quarter-wave monopole over an ideal conducting plane. A real terminal presents R + jX, including radiation resistance, conductor and ground loss, coupling and whatever current escapes onto the feedline or support.
The EFHW version places the feed near one end of a conductor close to a half wavelength. That region has relatively low current and high voltage, so its terminal impedance is commonly far above 50 Ω and requires a matching network. “End-fed” does not mean one-terminal. Power can enter only through a closed electromagnetic circuit. The other side may be an intentional counterpoise, the exterior of the coax, a mast, capacitance to earth and nearby conductors—or an uncontrolled mixture of them.
That distinction explains why the quarter-wave is often more repeatable. Its return structure can be visible, measured and made symmetric. An EFHW can be equally deliberate, but the transformer-to-choke coax section, counterpoise, support and surroundings must be defined rather than dismissed as “nothing.”
Current Distribution Sets the Engineering Problem
A quarter-wave monopole has high current near its feed region and falling current toward its open end. Loss close to that high-current region—radial resistance, a poor bond, a thin loading conductor or ground loss—can therefore consume a meaningful share of accepted power. Give the current a low-loss, stable return and the antenna becomes straightforward to match and model.
A vertical half-wave has a current maximum nearer its centre and low current near both open ends. End feeding places the matching network in a high-voltage region. That does not make the antenna inherently inefficient, and the current at a driven end is not literally zero. It does make insulation, transformer voltage, parasitic capacitance and the physical environment around the box important. Wet vegetation, a wall, soil, a metal mast or a different coax route can alter the terminal impedance and the common-mode current.
Wire length alone does not decide gain. The far field is the vector sum of current over the whole installed structure, including unintended conductors. A half-wave vertical and a quarter-wave monopole can produce different elevation patterns, but soil, electrical height, return-system geometry, loss and common mode decide whether that difference helps the required path. A low SWR does not reveal any of those terms.
The Transformer Needs a Measured Job Description
An EFHW matching network transforms the measured high terminal impedance into the line’s operating region. A label such as 49:1 is an impedance ratio, not proof that every installed end impedance is 2,450 Ω, and not a loss specification. The impedance moves with frequency and installation; the transformer’s complex insertion loss moves with frequency, flux level, winding capacitance, leakage inductance, load, temperature and waveform.
Measure the antenna’s R + jX at a declared reference plane, then evaluate the matching network across that actual locus. At transmitting power, confirm temperature rise, insulation margin and repeatability after a long carrier or the applicable duty cycle. A small-signal VNA sweep is necessary evidence, but it does not establish the high-power thermal or voltage limit.
The quarter-wave may need no impedance transformer when its installed terminal falls inside the transmitter or tuner range. That removes one loss and stress mechanism, but it does not make the system lossless. Radial, soil, conductor, connection, choke and feedline losses remain in its accepted-power budget.
Ground Coupling Is Not a Verdict by Itself
Putting an EFHW matchbox near soil can increase capacitive coupling and detuning, especially at its high-voltage terminal. Raising it may help, but no universal six-metre or eight-metre threshold exists. The result depends on field distribution, soil conductivity and permittivity, counterpoise geometry, support material, transformer enclosure, feedline route and nearby objects.
The quarter-wave is also ground-sensitive. Elevated radials can create a defined return with relatively few conductors when their length, height and symmetry are controlled. Radials lying on or buried in earth behave differently and usually need a design based on total conductor length, soil and layout rather than an elevated-radial count copied into the ground. Real earth also changes the reflected field and therefore the elevation pattern of both antennas.
ITU-R P.527-4 supplies representative electrical characteristics for earth materials. Those values are modelling inputs, not a replacement for the actual site. Sweep credible conductivity and relative-permittivity ranges and include terrain and nearby conductors when they matter.
Control the Outside of the Coax
Inside a coaxial line, the wanted mode has equal and opposite current on the centre conductor and the shield’s inner surface. Current on the shield’s exterior is a separate common-mode path. In the EFHW, an intentional length of exterior shield can be part of the return conductor; in the monopole, unintended exterior current can become an extra asymmetric radial. The same cable can therefore be transmission line on the inside and antenna on the outside.
A choke defines a useful boundary only over the frequencies, voltages, currents and impedances for which it is qualified. Its placement matters because every conductor between the antenna terminal and the choke remains available to the antenna. Do not select it from a turns count or a “1:1” label. Measure complex common-mode impedance over the operating band, verify the thermal and voltage margins, then measure exterior current in the installed arrangement.
Move the feedline or choke in a one-variable A/B/A trial. If terminal impedance, outside-shield current or received field changes repeatably, the feedline was part of the radiating or loss system. That result is more useful than arguing whether the antenna is “really” end-fed.
Compare at Equal Accepted Power
Forward power at the transmitter is the wrong comparison boundary. The two installations may have different feedline attenuation, mismatch, matching-network loss and common-mode paths. Establish accepted power at the antenna-system reference plane, then compare radiation in the required directions.
| Quantity | Quarter-wave vertical | Vertical EFHW |
|---|---|---|
| Driven conductors | Vertical radiator plus an explicit ground or radial return | Near-half-wave radiator plus an explicit or distributed return path |
| Feed region | High current, usually modest impedance in a controlled installation | Low current, high voltage and commonly high impedance |
| Principal matching burden | Often small, but installation dependent | High-ratio transformation across the measured impedance locus |
| Losses to expose | Return system, soil, conductor, joints, choke and feedline | Transformer, insulation leakage, return path, common mode, conductor and feedline |
| Space trade | Shorter vertical conductor plus radial footprint | Longer vertical conductor plus controlled feed/return geometry |
| Pattern evidence | Reproducible model and installed field data at the same frequency, site and accepted power | |
For field comparison, switch rapidly between antennas at the same site. Use a stable distant source or controlled transmitter, identical receiver bandwidth and gain state, the same polarization definition, and repeated measurements over the azimuths and elevations that matter. One contact, one S-unit report or two observations through changing propagation cannot establish a gain difference.
For modelling, record wire coordinates and radius, insulation, segmentation and convergence, excitation, transformer/network model, feedline and mast, return conductors, ground model and loss. State whether the plotted result is directivity, gain or realised gain and where accepted power is defined. NTIA’s NEC documentation is a useful foundation, but a model cannot include the fence, feedline or damp support that was left out of the geometry.
Build for the Site You Actually Have
If a roughly 5.35 m radiator and a credible return system fit, the quarter-wave is a strong first choice. It is mechanically shorter, exposes its current path and can avoid a high-ratio transformer. Use the radial geometry the site supports, measure the terminal impedance and exterior feedline current, and add common-mode control where the measurement shows it is needed.
If the site supports a roughly 10.7 m vertical conductor but cannot accommodate the quarter-wave’s return footprint, the EFHW can be a rational choice. Define the counterpoise or radiating coax section, keep high-voltage regions clear of people and lossy material, qualify the transformer under the real load and power, and stop exterior current at a measured boundary.
Both antennas can expose hazardous RF voltage, current and fields. Keep conductors, radial ends, counterpoises, matching hardware and supports inaccessible during transmission, and apply the station’s lightning, bonding, structural and RF-exposure controls. A matching network is not a safety barrier.
Bottom line: on 20 m, a well-engineered quarter-wave monopole often gives the simpler and more predictable route to an efficient vertical. A vertical EFHW remains valid when its longer radiator suits the site and its transformer, return path, common mode and ground coupling are designed and measured. Compare complete systems—not wire names.
Primary and authoritative references
- IEEE 145-2025 — Standard definitions of terms for antennas
- ITU-R P.527-4 — Electrical characteristics of the surface of the Earth
- NTIA — Numerical Electromagnetic Code capabilities and limitations
- NIST/NPL — Antenna measurement challenges, feed effects and uncertainty
- Keysight — Impedance Measurement Handbook
- Keysight — Reference-plane extension and de-embedding
- IEEE EMC Society — Differential- and common-mode current
- ARRL — Grounding, bonding and antenna return-path guidance
Mini-FAQ
- Does a quarter-wave vertical always beat a vertical EFHW? No. The quarter-wave often has a simpler matching and return-path problem, but installed efficiency and pattern depend on the complete geometry, loss, ground, common mode and accepted power.
- Is an EFHW literally current-free at its feedpoint? No. A driven antenna must accept power. The end region has relatively low current and high voltage compared with the centre, so its terminal impedance is high rather than infinite.
- Is a 49:1 transformer automatically lossy? It always has finite loss, but the amount is not fixed by the ratio. Measure it across the actual complex load, frequency, power, temperature and duty cycle.
- How many radials does a 20 m quarter-wave need? There is no universal count. Elevated and ground-contact systems behave differently; choose geometry from the site, then verify terminal impedance, loss sensitivity and exterior feedline current.
- How high must an EFHW matchbox be? No universal height exists. Raise and position it to control high-voltage coupling and loss, then verify the result with impedance, common-mode-current and field measurements.
- What is the fairest field test? Compare rapidly at the same site, frequency and accepted power, using a stable path, fixed receiver settings, declared polarization and enough directions or repetitions to expose pattern and propagation variation.