Full-Wave Antennas Aren't Bad

Someone recently sent me a YouTube video comparing half-wave and full-wave antennas. The demonstration itself was reasonable, and the presenter’s observations were largely correct—for the antennas being compared.

The problem came with the conclusion.

Both antennas were straight, centre-fed dipoles, yet the results were used to suggest that full-wave antennas in general are bad. That is a much broader claim than the experiment supports.

What the video really demonstrated is that a centre-fed, one-wavelength dipole can be awkward to feed, particularly when connected directly to a system designed around 50-ohm coax. That is useful information, but it tells us very little about full-wave loops, delta loops, quad elements, horizontal sky loops or the many other wavelength-sized contraptions successfully used on HF.

An antenna is not defined by wire length alone. Geometry, current distribution, feedpoint, height, orientation, ground conditions and intended purpose all matter. Saying that every antenna containing approximately one wavelength of wire must behave badly is rather like saying that a bicycle wheel and a clothesline must behave the same way because both contain wire.

Why the half-wave dipole is so convenient

The half-wave dipole deserves its excellent reputation.

When fed in the centre, its feedpoint lies near a current maximum and voltage minimum. Its impedance is consequently in a range that works reasonably well with common feedlines—typically around 50 to 75 ohms, although height, conductor diameter and nearby objects can shift it considerably.

It is simple, efficient and predictable. Its broad figure-of-eight pattern is useful for many forms of HF communication, and it can usually be connected to coax through a suitable current balun or common-mode choke.

The half-wave dipole is not popular because every longer antenna is inferior. It is popular because its electrical characteristics are exceptionally convenient. Rohde & Schwarz’s antenna guide illustrates how strongly dipole impedance and pattern depend on electrical length.

What happens with a full-wave dipole?

Double the length of the dipole and the situation changes.

At the centre of a one-wavelength dipole, the feedpoint is close to a current minimum and voltage maximum. The resulting impedance can be extremely high—often several thousand ohms in a practical installation.

Connecting that point directly to 50-ohm coax creates a severe mismatch. The transmitter may reduce power, additional loss may occur in the feedline, and very high RF voltages can appear at the feedpoint and inside the matching equipment.

Under those conditions, the antenna may perform poorly. But that does not prove that a wavelength of wire is inefficient. It proves that the feedpoint and feed system are poorly matched.

High impedance is not the same as poor radiation efficiency. With an appropriate matching network or low-loss balanced line and tuner, power can still be delivered to the antenna efficiently. The practical difficulties are impedance transformation, high voltage, insulation and feedline loss—not some fundamental inability to radiate.

The full-wave dipole also has a somewhat narrower broadside pattern and greater peak directivity than a half-wave dipole. It does not create free gain; it takes energy away from some directions and concentrates it in others. As the wire becomes still longer, additional lobes and nulls eventually develop. The underlying current distribution and pattern are documented in this University of Texas antenna analysis.

So even the full-wave dipole is not inherently bad. It is simply less convenient than the half-wave dipole for ordinary centre-fed, 50-ohm operation.

Bend the wire and everything changes

Now take roughly the same amount of wire and close it into a loop. Electrically, we have created a very different antenna.

A resonant one-wavelength loop normally has substantial current at its feedpoint and a much more manageable impedance. For common square, triangular and circular loops, the free-space feed resistance is often somewhere in the general region of 100 to 150 ohms. Shape, wire thickness, feed position, height and surrounding objects can move it well outside that range.

An exact geometric circumference of one free-space wavelength is not guaranteed to be resonant, either. A practical loop must be trimmed and measured in its final surroundings. Research on resonant loop geometries also shows that their impedance and modest directivity advantage depend on shape rather than wire length alone. See this Virginia Tech resonant-loop study.

A full-wave loop can be circular, square, rectangular, diamond-shaped or triangular. A square version is normally called a quad loop, while a triangular version is a delta loop.

At its fundamental frequency, a full-wave loop generally radiates broadside to its plane. Its free-space gain advantage over a half-wave dipole is modest—roughly a decibel for common shapes—but its geometry, feed impedance and installation options can make it far more useful in a particular situation.

The horizontal full-wave loop

Install a one-wavelength loop horizontally at a modest height and it can become an excellent regional HF antenna.

On 80 or 40 metres, most amateurs cannot place an antenna half a wavelength above ground. A low horizontal loop therefore tends to produce strong high-angle radiation. That is exactly what is wanted for near-vertical incidence skywave, or NVIS, communication.

For regional nets, emergency communications and reliable short-to-medium-distance contacts, high-angle radiation is a feature rather than a fault. A Naval Postgraduate School modelling study specifically examined the usefulness of a one-wavelength horizontal quad loop for NVIS and medium-range HF communication.

Raise the loop higher and lower-angle lobes become more significant. Use the same loop on higher bands, where its circumference becomes several wavelengths, and its azimuth pattern develops multiple lobes and nulls.

Those lobes can help if they point towards the desired stations. They can be frustrating if an important direction falls into a null. This is why describing a large loop as “omnidirectional on every band” is misleading.

A horizontal loop can also be used as a multiband antenna with low-loss balanced line and a suitable tuner. Feeding a badly mismatched multiband loop through a long run of ordinary coax is less attractive: the tuner may satisfy the transmitter while considerable power is still being lost in the coax.

A tuner can present a comfortable impedance to the transmitter. It cannot recover power already dissipated in a lossy, high-SWR feedline.

Horizontal loops are also frequently described as inherently quiet receiving antennas. Some installations are indeed quieter, but this is not guaranteed. The improvement may result from radiation pattern, polarization, good balance or reduced feedline pickup. A poorly balanced loop with common-mode current on its coax can collect plenty of household noise.

Vertical delta and quad loops

Turn the loop vertically and it becomes a useful DX antenna.

A vertical delta or quad loop radiates mainly broadside to its plane, so it can be aimed by choosing its orientation. Feedpoint placement also influences polarization. On a square or rectangular loop, feeding the centre of a horizontal side generally favours horizontal polarization, while feeding the centre of a vertical side favours vertical polarization. On a delta loop, bottom-centre feeding commonly favours horizontal polarization, while a lower-corner feed commonly favours vertical polarization.

These are useful design rules, not guarantees. Height, ground interaction, nearby conductors and feedline common-mode current can alter the result.

Unlike a ground-mounted quarter-wave vertical, a full-wave loop provides a complete conductive return path and does not require a radial system. That does not make the earth irrelevant, however. A loop placed close to lossy soil can still suffer altered impedance, pattern distortion and ground loss.

When installed sensibly, a vertical loop can produce useful low-angle radiation and good efficiency without loading coils or radials. On the higher HF bands, its physical dimensions are also reasonably manageable.

Full-wave loops as beam elements

Perhaps the clearest argument against dismissing full-wave antennas is the cubical quad.

The driven element of a quad beam is a full-wave loop. Add a slightly larger loop as a reflector, and perhaps smaller loops as directors, and the result is a directional antenna with useful forward gain and front-to-back performance.

Delta-loop beams apply the same principle using triangular elements. In both cases, the full-wave loop is not an unfortunate compromise—it is the fundamental building block of the antenna.

ARRL training material describes quad and delta-loop beams as arrays of approximately one-wavelength elements and notes that changing the feedpoint changes polarization. See the ARRL General Class study guide. ARRL also maintains a collection of practical HF loop antenna projects, including horizontal sky loops, sloping loops and phased delta loops.

The disadvantages are mainly mechanical. Full-wave elements require more wire, taller spreaders and more support than comparable dipole-based elements. Wind, ice and available turning radius may decide whether a quad is practical.

Folded dipoles and misleading labels

The folded dipole demonstrates why counting wire is not enough.

A conventional folded dipole has approximately one wavelength of total conductor because the wire travels out and then folds back. Yet its end-to-end span and radiation pattern are those of a half-wave dipole. With equal conductor diameters, its feed impedance is roughly four times that of an ordinary dipole, or close to 300 ohms in free space.

It contains about one wavelength of wire, but it is not electrically the same as either a straight full-wave dipole or a one-wavelength quad loop. MIT’s antenna notes cover the folded-dipole relationship.

Again, geometry and current distribution matter more than the amount of copper.

When antennas become longer still

HF history contains many successful antennas that are several wavelengths long.

Long wires, V-beams and rhombics deliberately use phase relationships along electrically long conductors to create directional, low-angle radiation. Large rhombics were once standard equipment at international HF communication stations.

Terminated travelling-wave versions can provide wide bandwidth and good front-to-back performance, although some transmitter power is deliberately dissipated in the terminating resistor. Unterminated versions avoid that particular loss but become more frequency-sensitive and develop standing waves.

These antennas require land and supports, but their existence makes one point very clearly: becoming longer than half a wavelength does not suddenly make an antenna defective. The U.S. Army HF antenna manual documents several such directional HF systems.

When full-wave antennas shine

A full-wave design can be an excellent choice when:

• A loop fits the available supports better than a straight wire.
• High-angle radiation is wanted for NVIS or regional coverage.
• A vertical loop can be oriented for low-angle DX.
• A quad or delta-loop array is wanted for directional gain.
• A low-loss balanced feed system and suitable tuner are available.
• The desired stations lie inside the antenna’s useful lobes.
• Full-size construction avoids the losses associated with loading coils and severe shortening.

A half-wave dipole is often the better choice when simplicity, light weight, easy coax feeding and a broad, predictable pattern are the main priorities.

Full-wave designs become poor choices when an inconvenient feedpoint impedance is ignored, when severe SWR is carried through lossy coax, when the antenna is installed at the wrong height for the intended path, or when multiple lobes and nulls are mistaken for uniform coverage.

Compare complete antenna systems

A fair antenna comparison must consider more than SWR or the signal reported by one receiving station.

The antennas should be compared at the same electrical height, with the same accepted transmitter power, compatible polarization and comparable feedline losses. The relevant elevation angles must also be considered. An NVIS antenna should not be condemned because it loses a low-angle DX comparison, just as a low-angle vertical should not be judged solely by a station 100 kilometres away.

A low SWR does not prove efficiency. Resonance does not guarantee a convenient impedance. A strong signal in one direction does not reveal the nulls elsewhere. And a tuner can protect the transmitter without eliminating loss in the feedline.

Put that wavelength of wire into the right shape, feed it correctly, install it at an appropriate height and use it for the job its pattern supports, and it can be an excellent HF antenna.

The wavelength is not the problem. The design—and the conclusion we draw from it—is what matters.

Takeaways you can trust

• “Full wave” describes electrical length, not antenna quality.
• A centre-fed full-wave dipole is difficult mainly because its centre is near a current minimum and voltage maximum.
• A full-wave loop has a different current distribution, feed impedance and broadside radiation pattern.
• Horizontal loops can favour regional, high-angle coverage; vertical loops can favour lower-angle DX.
• Quad and delta-loop beams use full-wave elements deliberately for directional performance.
• Feedline loss, height, polarization and pattern must be included in any honest comparison.

Want more technical RF content? Subscribe for new deep dives and lab notes.

Have a question or field observation? Contact RF.Guru.