EFHWs Below 20 m: Popular, Easy—and Often the Wrong Compromise
EFHWs Below 20 m: Popular, Easy—and Often the Wrong Compromise
Below the 20 m band in frequency means longer wavelengths: 40, 80 and 160 m. These are the bands I am defending—not writing off—when I ask an EFHW to have a clear design purpose instead of an all-band sales pitch.
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.
The low bands deserve more than a wire, a transformer box and a promise that everything from 80 to 10 metres will take care of itself. My preference is straightforward: if 160, 80 or 40 m matters, give it a purpose-built monoband or bounded dual-band design. Keep the convenience of end feeding, but stop making the lowest bands pay for a long list of higher-band SWR dips.
Mark, the Ham Florida Man, takes up this argument in EFHW Antennas Below 20 Meters are Almost Worthless. From about 1:16 he reads this RF.Guru article, written by me, Joeri Van Dooren, ON6URE. His opening is about 80 and 40 m; later he explicitly makes room for properly designed 80/40 m and 160/80 m inverted-L systems. That distinction matters: the target is the casual all-band promise, not the existence of useful low-band EFHWs.
The question he reads is the heart of my argument: “How much of your RF actually gets radiated?” Making a contact proves a path existed. It does not reveal the losses between the transmitter and the field, or show whether a different antenna would put more useful energy in the required direction.
Keep the low bands; narrow the promise. A half-wave/full-wave band pair gives an end-fed wire a sensible job. Match and rate the assembly for that job, provide its return path, and choose its shape for the contacts you want. That is a stronger starting point than treating a low SWR on many bands as proof of one efficient broadband antenna.
The Low-Band Designs I Mean
A wire near a half wavelength on one band can be near a full wavelength on the next octave-related band. That gives us a useful starting point—not a guarantee that every band edge lines up exactly, or that the end impedance and pattern stay the same. Wire diameter, insulation, bends, nearby objects and the feed system all matter.
| Intended job | Wire-mode starting point | Why I would choose it |
|---|---|---|
| 160/80 m | Approximately half-wave on 160 m, full-wave on 80 m | Concentrate the matching and installation on the two longest HF/MF amateur bands in this discussion. |
| 80/40 m | Approximately half-wave on 80 m, full-wave on 40 m | A useful two-band low-HF brief without also demanding 10 m behaviour from the same assembly. |
| 40/20 m | Approximately half-wave on 40 m, full-wave on 20 m | Keep the same bounded design idea at a more manageable physical length. |
| 40 m or 20 m monoband | A half-wave-like wire on the chosen band | Make one operating band the priority when its matching and pattern matter more than the number of bands on the label. |
For examples of those intended band groupings, see RF.Guru's EFHW16080, EFHW8040, EFHW4020, EFHW40 monoband and EFHW20 monoband. The engineering choice is the band plan; the corresponding product information supplies the model-specific installation and operating limits.
These are not the only possible useful EFHW modes. Longer wires can support further harmonic-region resonances, and a carefully engineered multiband system can use them. My argument is that those extra opportunities should not dictate an unnecessarily compromised low-band design.
Why the Transformer Cannot Be an Afterthought
Between 3.5 and 29.7 MHz, an 80–10 m assembly spans more than three octaves. A 40–10 m assembly spanning 7 to 29.7 MHz still covers more than two. That is a very different task from a monoband design or a neighbouring band pair, especially with a high and changing antenna impedance.
Fair-Rite's broadband-transformer guidance explains the competing ends of the problem: sufficient shunt inductance is needed at the low end, while leakage inductance and winding capacitance limit the high end. Adding turns can help the former and worsen the latter. The paper concerns low-power transformers; it is not a power rating for an EFHW assembly.
That is why I favour a narrower brief. Core material, core size, winding arrangement, impedance ratio and compensation can be chosen around the bands that matter. Neither “43” nor “52” is a certificate of efficient 80–10 m operation; neither is a universal declaration of failure. The completed assembly still has to tolerate its actual voltage, current, load and duty cycle without unacceptable loss or heating.
Nor would I prescribe 49:1, 68:1 or 70:1 from the band name alone. Those are impedance-ratio labels, not measurements of the installed antenna. The same nominal band can present a different complex load when the wire is lowered, bent or brought close to a building. A band-specific design earns its advantage by solving that defined problem, not by printing a different number on the box.
A Shunt Capacitor Is Compensation, Not Free Coverage
A capacitor can cancel part of a network's reactance and genuinely improve the input match. That is real circuit behaviour, not fake physics. But an improved SWR does not establish that the transformer is low-loss, that the capacitor has comfortable RF current and voltage margins, or that the wire radiates usefully on every labelled band.
Calling every compensated EFHW a dummy load would be as careless as calling every SWR dip proof of efficiency. The useful criticism is more specific: do not spend the low-band design margin simply to make the higher-band input trace look attractive. If compensation serves a declared band and load range, it is engineering. If it substitutes for a loss and stress budget, it is a sales shortcut.
Why an Inverted-L Can Be the Practical Low-Band Choice
On 160 or 80 m, “just put the whole wire high in the air” is not much of a garden installation plan. An inverted-L uses available vertical support and then turns the remaining wire into the horizontal space. That is a real mechanical advantage, and it is why this form belongs in the low-band discussion.
Its pattern follows the current distributed over the complete wire and return system. An end-fed half-wave has low current near its high-impedance end, so a short upright section does not automatically behave like a strongly driven quarter-wave vertical. A bend does not automatically put the current maximum at the corner either. Vertical and horizontal current contributions, their phase, height and ground all decide the result.
I would choose the L when it makes a useful current-bearing geometry possible on the site. I would choose a flat or sloping installation when its support arrangement and pattern suit the intended contacts. A low horizontal wire can be useful for regional work; there is no universal half-wavelength-height switch below which all DX suddenly becomes impossible.
The Other Side of the Feedpoint Still Exists
W8JI's end-fed analysis shows why the matching network, wire and return structure have to be considered together. A large resistive part of the end impedance reduces terminal current for a given accepted power; it does not make the complementary current disappear.
Give that current a deliberate return structure and use an appropriate common-mode choke to restrict its continuation toward the station. Otherwise the coax exterior, mounting structure and station wiring can become part of the installed antenna. A universal “put the choke at 0.05 wavelength” recipe does not settle that problem. The position changes which conductors participate, and the choke's impedance is frequency-dependent.
A static-bleed path is a different function again. It is not automatically a suitable RF return or a lightning-protection system. ARRL's grounding guidance distinguishes RF control, electrical safety and lightning protection for exactly this reason. Keep high-voltage feed and wire ends inaccessible; do not make adjustments while transmitting.
Choose the Antenna for the Band You Actually Want
If 160/80 or 80/40 m is the job, I would start with a matching assembly and wire geometry designed for that pair. If 40 m is the priority, a dedicated 40 m design is an entirely respectable answer. You gain a more focused design problem and the freedom to spend the available height, conductor and component margin where you need them.
If broad multiband operation is the real requirement, a doublet with a suitable matching system and low-loss balanced line, or another intentionally designed end-fed system, may be the better installation choice. It brings its own requirements; it does not have to pretend that one high-ratio transformer can make every load equally comfortable.
On 20–10 m, where full-size wires and verticals fit more readily, I would also compare the EFHW with a dipole, a properly installed quarter-wave vertical or an EF-OCF arrangement. That is a separate upper-HF question, not a reason to discard a useful 160/80 m design. Compare actual feed losses and installed patterns, not antenna names.
That is the point behind the provocation: I am not asking you to abandon the low bands. I am asking you to stop treating them as a bonus feature on an all-band transformer. Give them their own engineering brief. The right mono- or dual-band EFHW can then be a purposeful antenna, rather than a compromise whose cost is hidden behind the word “works”.
Mini-FAQ
- What does below 20 m mean in this article? It means below the 20 m band in frequency: chiefly 40, 80 and 160 m, whose wavelengths are longer. It does not mean the 10–17 m bands.
- Are low-band EFHWs mostly useless? No. The criticism is of the all-band promise and the assumption that contacts or a low SWR prove efficiency. A properly designed monoband or bounded dual-band EFHW can be a useful low-band choice.
- Why favour a half-wave/full-wave band pair? It gives the wire a clear modal purpose on two neighbouring bands and narrows the matching assembly's design range. Neither the match nor the radiation pattern is automatically identical on both bands.
- Does a capacitor create fake resonance? No. A capacitor can provide real reactive compensation. What it cannot establish by itself is low loss, safe component stress or a useful radiation pattern across all advertised bands.
- Is 49:1, 68:1 or 70:1 the correct low-band ratio? There is no band-only answer. The correct matching network depends on the installed complex load, core, windings, operating frequency and RF stress; a nominal ratio alone cannot settle it.
- Must a low-band EFHW be an inverted-L? No. An inverted-L is a practical way to allocate limited height and horizontal space, but its current distribution and surrounding ground determine the pattern. A flat wire remains useful where its pattern suits the intended contacts.
- Should the choke always be 0.05 wavelength from the feedpoint? No. Choose a deliberate return structure and a choke boundary that suits the installation. A fixed spacing is not a universal substitute for controlling exterior feed-line current.