Two Inverted-L End-Fed Wires for a 160/80/40 m Station

If I had space for two low-band end-fed inverted-L wires, I would treat them as complementary installed antennas, not guaranteed DX machines. The longer wire targets the 160/80 m relationship. The shorter wire targets the 80/40 m relationship. Their shared 80 m coverage can be valuable because different routes, heights and current distributions may favour different directions or elevation angles.

The Two-Wire Strategy

The first antenna is intended to work near a half-wave current distribution on 160 m and near a full-wave distribution on 80 m. The second applies the same idea one octave higher: near a half wave on 80 m and near a full wave on 40 m.

Those descriptions are starting regions, not cutting dimensions. Physical length differs from free-space wavelength because conductor diameter, insulation, bends, height, soil, trees, buildings, the return path and the matching assembly all affect the installed electrical length. Trim from measured complex impedance and the pattern objective; do not copy a length and assume the current distribution followed.

The overlap on 80 m is not wasted. It offers a way to compare two geometries or preserve coverage when one pattern has a null in an inconvenient direction. It also creates a coupling problem that must be included in the design.

Follow the Current Through the Bend

An inverted-L is a shape, not one electrical antenna type. A radial-fed quarter-wave inverted-L and an end-fed wire operating near a half-wave region can look similar in the garden while carrying very different currents.

In the simplified end-fed half-wave picture, current is low and voltage is high near the wire end, with a current-rich region farther along the conductor. Real current at the feed assembly is not zero because the circuit includes transformer capacitance, the intentional return conductor, coax exterior, bonds and surrounding structures. The vertical leg contributes strongly only when a useful part of the installed current actually flows there.

At the next harmonic region, the same wire supports more than one current-rich section with phase changing along the conductor. Bending those sections into vertical, horizontal and sloping portions changes how their fields combine. On 80 m for the long wire and 40 m for the shorter wire, expect a more structured azimuth and elevation pattern than the lower-band pattern. Do not call either antenna omnidirectional without a complete model or measurement.

Height Must Be Expressed Electrically

A physical height has a different electrical meaning on 160, 80 and 40 m. The same horizontal section may be electrically very low on 160 m and substantially higher on 40 m. Ground-reflected fields then combine differently with the direct field, moving elevation maxima and nulls.

A tall vertical section can support useful low-elevation radiation when it carries substantial current, but a vertical-looking feed end is not enough. The remainder of the wire, bend location, return conductor and coax exterior can reinforce, redirect or partially cancel the field. Terrain can matter as much as the last few metres of mast height.

The in-force ITU-R BS.705 recommendation treats ground conductivity, topography and nearby structures as practical radiation-pattern variables. Describe a low-angle advantage only for a declared installation and frequency, with realised gain at the relevant elevation angle—not from geometry alone.

Two Antennas Also Mean Mutual Coupling

These wires are not independent when installed near each other. Both can carry induced current, and both are intended to operate on 80 m. Spacing, crossing angle, parallel runs, height, termination state, feedline routing and connection to the station all affect coupling.

An unused antenna does not disappear. Leaving its feedpoint open, shorted, connected to a transformer, grounded through switching hardware or terminated in a receiver produces different induced-current and re-radiation conditions. Coupling can shift impedance, move current maxima, alter pattern and place unexpected voltage at the inactive port.

Model both wires, both return paths, both feedlines to their choke boundaries and the actual inactive-port state. At the station, use switching whose isolation, voltage, current, transient and failure behaviour has been verified for the intended power. If simultaneous transmit and receive is contemplated, that becomes a separate coupled-power and receiver-protection problem; antenna separation alone is not a protection specification.

Choose the Match From the Installed Load

A high-ratio transformer number is not part of the antenna’s identity. Measure the complex feedpoint impedance over each intended band at a declared reference plane, then choose a topology and transformation range that give acceptable wanted-mode transmission under those loads.

The completed transformer must be checked for insertion loss, magnetising behaviour, leakage, winding capacitance, voltage distribution, flux, temperature and insulation stress at the intended power, waveform and duty cycle. A good SWR trace does not establish low transformer loss or adequate thermal margin.

Keysight’s VNA guidance treats calibration as establishing the reference plane. Use the same principle for trimming and comparison: say whether impedance is measured at the wire terminals, transformer input, choke, line end or transmitter. Feedline electrical length transforms impedance, while line attenuation can make the transmitter-end SWR look better than the antenna-terminal mismatch.

Give Return Current a Deliberate Route

An end-fed wire needs a return conductor. Depending on the architecture, that may be a separate branch or counterpoise, a deliberately defined section of coax exterior, a radial or ground system, or a combination of conductors and displacement-current paths. If it is not designed, the coax, mast, station wiring, protective earth and nearby structures can become part of the antenna.

Do not use a ground rod as a universal substitute for an RF return network. Protective-earth bonding, lightning bonding, static discharge and RF current control serve different purposes and must follow the rules that apply at the site. Soil loss, conductor loss and current on unintended paths can change without producing an alarming SWR.

If a coax-exterior section is intentionally used as a return branch, declare its physical route and the common-mode boundary. A separate choke can raise impedance beyond that boundary, but its useful complex impedance, resonance, voltage, heating and installed current reduction must be verified by band. Roy Lewallen’s W7EL experiments remain a useful reminder that feedline current depends on the complete symmetry and termination, and that a current-balancing device can materially change the measured system.

Pattern, Efficiency and Match Are Different Results

A low SWR shows reflection magnitude at one plane. It does not measure radiation efficiency, elevation pattern, azimuth pattern, common-mode current or received noise. Accepted power can still be dissipated in transformer, choke, feedline, conductors, joints, vegetation and soil.

Likewise, hearing a DX station is not a calibrated antenna comparison. Propagation, transmit power at the far end, fading, receiver settings and local noise can dominate one observation. The useful comparison is rapid and repeated A/B/A switching with the same receiver chain, recorded bandwidth and gain settings, enough samples to reveal fading, and an antenna switch whose loss and isolation are known.

IEEE 149 treats pattern measurement as its own disciplined task. When absolute far-field work is impractical, combine a converged full-geometry model with current mapping, component loss measurements and controlled relative field or received-signal comparisons. LLNL’s Numerical Electromagnetics Code can include wires, conducting surfaces, loads, networks, transmission lines and homogeneous ground and can report segment currents and radiation patterns.

A Measurement Plan for the Pair

Before deciding that the pair improves the station, record:

• the full wire routes, bends, heights, conductor construction and surrounding structures;
• the intentional return conductors, coax routes, choke positions and all bonds;
• the inactive-port and switching state used on every band;
• complex impedance at common calibrated planes, both separately and with both antennas installed;
• transformer, choke, feedline and switch loss plus temperature at operating stress;
• exterior-current maps on both feedlines and station-connected cables;
• model convergence, ground parameters, segment currents and realised patterns; and
• repeatable A/B/A field or received-signal comparisons for the directions and path angles that matter.

This evidence can reveal a genuine station advantage: perhaps the long wire is better at one 80 m elevation angle while the shorter one fills a direction or null. It can also show that coupling makes the pair worse than either isolated design. Both are useful engineering results.

Respect Voltage, Exposure and Mechanical Limits

End regions and matching components can develop hazardous RF voltage. Keep wires, insulators, transformers and counterpoises out of reach; use strain relief and supports designed for wind and ice; and de-energise before adjustment. Evaluate RF exposure for the actual power, frequency, duty cycle, pattern and accessible area under the national rules that apply.

ICNIRP’s RF guidelines cover 100 kHz to 300 GHz, but they do not replace local amateur-radio procedures or an installation-specific assessment. A successful SWR sweep is not a safety certificate.

Questions, antenna-factor records or height trials to share? Contact RF.Guru.