EFHW Verticals for the Upper HF Bands
EFHW Verticals for the Upper HF Bands
An end-fed half-wave can be stood vertically and it can make excellent contacts. The useful question is not whether the name sounds efficient, but what the complete installation does with current, loss and pattern on 10, 12 and 15 metres.
On the upper HF bands, a half-wave wire is short enough to stand vertically from many portable and domestic supports. That makes the EFHW tempting: one support, a feedpoint near the operator and no long elevated feed line. Convenience is real. So are the transformer, return-path and installation questions that the simple sketch leaves out.
My usual single-band choice is still a quarter-wave vertical with a deliberate radial or counterpoise system. It normally asks less of the matching network and makes the return structure easier to identify. That is a design preference, not a claim that every EFHW vertical is inefficient or that a quarter-wave wins without a competent ground system.
The Current Minimum Is Not a Verdict
A thin half-wave conductor has a current maximum near its centre and current minima near its ends. Feeding near one end therefore presents a high impedance and high RF voltage. It does not follow that the antenna radiates poorly merely because current is low at the feedpoint. Radiation comes from the current distribution along the complete conductor, not from the current at one convenient observation point.
Standing that conductor vertically places its strongest current region roughly a quarter wavelength above the lower end. That changes its interaction with ground and nearby objects, but feedpoint height alone does not determine efficiency or take-off angle. Wire length and diameter, height, soil, supports, transformer, return conductor, coax exterior and surrounding metal all participate.
The useful correction: a voltage maximum at the feedpoint explains the demanding impedance and insulation problem. It does not, by itself, prove low radiation efficiency.
The Transformer Must Be Measured with the Load It Will See
An EFHW feed system usually uses a high impedance transformation. A nominal winding ratio does not establish either the antenna-side impedance or the loss. The practical network contains finite magnetising impedance, leakage inductance, winding resistance, inter-winding capacitance, core loss and lead or enclosure coupling. Their effects change with frequency, load, construction, drive level and temperature.
That is why fixed loss figures for 15, 12 or 10 metres cannot be assigned from “49:1” or “64:1” alone. Measure the completed transformer with representative complex loads at the intended reference planes. Record insertion loss, return loss, temperature rise and repeatability across the required power and duty cycle. A smooth station-end SWR trace can coexist with feed-line or transformer loss.
The Missing Terminal Still Exists
An end-fed antenna still needs a second RF branch. Depending on the installation, that branch can include a deliberate counterpoise, a defined length of coax exterior, the mast, bonding conductors, transformer capacitance and displacement current through the surroundings. If it is not designed, the antenna selects one for you.
A common-mode choke defines a boundary; it does not create the missing return path. Place it according to the intended antenna-side conductor length and the measured exterior-current distribution on every operating band. A multiband arrangement that behaves calmly on 15 metres can put the same cable section near a very different current or voltage condition on 12 or 10 metres.
What the Quarter-Wave Changes
A quarter-wave vertical is not “transformer-free perfection.” Its lower terminal needs a radial, counterpoise or ground system. Loss in that system can dominate the result. With a competent return structure, however, its feed impedance is often close enough to a practical transmission-line impedance that the matching task is modest. Its high-current region is also near the base, where the radial junction is explicit and accessible.
The EFHW trades that low feed impedance for a longer radiator and a high-ratio feed network. The trade can be sensible when the support, cable route or multiband requirement favours it. For a dedicated upper-HF vertical, the quarter-wave often wins my engineering shortlist because it is easier to measure, easier to decouple and less dependent on a broadband high-impedance transformer.
| Question | End-fed half-wave vertical | Quarter-wave vertical |
|---|---|---|
| Radiator length | About one half wavelength before end effects and tuning | About one quarter wavelength before end effects and tuning |
| Feed region | High voltage, low current and high complex impedance | High current and usually a much lower complex impedance |
| Return structure | Counterpoise, coax exterior and surrounding capacitance must be bounded | Radial, counterpoise or ground system must be bounded |
| Matching burden | Usually a high-ratio broadband transformation | Often modest, but installation dependent |
| Pattern | Must be modelled or measured with every participating conductor | Must be modelled or measured with its complete return system |
Band Numbers Set Scale, Not Performance
A free-space half wavelength is approximately 7.1 m at 21.2 MHz, 6.0 m at 24.9 MHz and 5.2 m at 28.5 MHz. A quarter wavelength is half those values. Real conductors tune shorter or occasionally differently because diameter, insulation, loading, height and nearby material change electrical length. These figures are planning scale, not cut charts.
On 10 metres, fractional bandwidth can expose transformer and matching limits more clearly than on a narrower band. On 12 and 15 metres, nearby conductors or additional band elements can alter current distribution through mutual coupling. None of that can be converted into a universal efficiency table or one take-off angle. The installed geometry and ground decide the pattern.
Measure the Comparison Fairly
- Fix the purpose. Decide whether you are comparing low-angle field, total efficiency, match bandwidth, common-mode current, convenience or all of them.
- Declare the reference plane. Calibrate the vector network analyser at the same connector, or characterise and de-embed the unchanged feed line and adapters.
- Record complex impedance. Save resistance, reactance and complex S11 across the entire band; SWR alone hides the direction of the error.
- Account for feed-network loss. Characterise the EFHW transformer and any quarter-wave matching network with representative loads, power and temperature.
- Map exterior current. Probe the coax, counterpoise, mast and accessible bonds at repeatable positions on each band.
- Keep geometry comparable. Record tip height, current-maximum height, cable route, radial layout and surroundings. Change one variable and return to the starting state in an A/B/A trial.
- Compare field, not anecdotes. Use stable remote receivers or a controlled field-strength path, alternate rapidly enough to limit propagation change, and report uncertainty.
- Model every conductor that carries current. Include real ground, the return structure and feed-line exterior when explaining pattern or take-off angle.
RF and electrical safety: the end-fed feedpoint and wire tip can carry high RF voltage. De-energise before adjustment, maintain clearances and strain relief, and do not alter protective or lightning bonding to improve an RF measurement.
Primary and Authoritative Technical Sources
- IEEE Std 145-2025, Standard for Definitions of Terms for Antennas—the terminology boundary for impedance, efficiency, gain and radiation pattern.
- IEEE Std 149-2021, Recommended Practice for Antenna Measurements—measurement conditions and uncertainty for passive antennas.
- ITU-R BS.705-2, HF transmitting and receiving antenna characteristics and diagrams—ground and environmental effects in practical HF pattern work.
- Lawrence Livermore National Laboratory, Numerical Electromagnetic Code v5—full-geometry modelling of wires, surfaces, networks, transmission lines, ground, current and pattern.
- Fair-Rite, Technical Information: Use of Ferrites in Broadband Transformers—the frequency-dependent equivalent circuit and loss mechanisms of practical ferrite transformers.
- Keysight, Specifying Calibration Standards and Kits for Vector Network Analysers—calibration, vector error correction and a fixed measurement reference plane.
- ARRL, Grounding—the return-system distinction between end-fed, quarter-wave and complete balanced antenna structures.
My Bottom Line
An EFHW vertical on 10, 12 or 15 metres is not disqualified by the current minimum at its lower end. It is a legitimate radiator whose high-impedance transformer and return path must earn their place through measurement.
For a dedicated single-band vertical, I usually start with the quarter-wave and a deliberate radial system. It is simpler to bound and leaves fewer places for loss or uncontrolled current to hide. If the EFHW solves a real support or multiband problem, use it—but compare the complete antennas at the same site, not idealised names on a page.
Mini-FAQ
- Is an EFHW vertical inefficient because current is low at the feedpoint? No. The complete current distribution radiates; the low feedpoint current chiefly creates a high-impedance, high-voltage feeding problem.
- Does a 49:1 transformer have a fixed loss on each band? No. Loss depends on its complete construction, frequency, complex load, drive, temperature and measurement planes.
- Does an EFHW work without a return path? No. A deliberate counterpoise, coax exterior, mast, capacitance and surrounding conductors can all form the second RF branch.
- Is a quarter-wave vertical always more efficient? No. Its radial or ground-system loss can dominate. With a competent return system it often has the simpler matching and common-mode problem.
- Can SWR identify the better DX antenna? No. SWR describes a match at one plane; it does not separately reveal feed-network loss, radiation efficiency, exterior current or pattern.
- How should the two designs be compared? Use fixed geometry and reference planes, characterise network loss, map exterior current and run controlled A/B/A field comparisons with stated uncertainty.