Wire Antennas on Sloped Terrain: Level, Follow the Ground or Aim the Pattern?
Wire Antennas on Sloped Terrain: Level, Follow the Ground or Aim the Pattern?
A sloping site creates more than one idea of “level.” The wire can be horizontal to the geometric horizon, parallel to the local ground or deliberately tilted toward a path. The useful choice comes from electrical height, current distribution, terrain and the installed pattern—not from one rule for every wire antenna.
When I install a dipole, loop or half-square across uneven ground, I first separate a simple carpentry question from the electromagnetic one. A spirit level can tell me whether the top wire is horizontal. It cannot tell me the realised takeoff angle, the current balance or what the hillside does to the path.
Horizon-level is a useful geometric baseline, not a pattern guarantee. Keeping a designed horizontal section horizontal can preserve its intended shape, but the changing clearance above a slope changes ground coupling. Following the slope keeps clearance more uniform while tilting the antenna. Both choices must be analysed as installed systems.
Define Which “Level” You Mean
Four reference frames are easily mixed together:
- Geometric horizontal is perpendicular to gravity and can be set with a surveyed level.
- Local ground-parallel follows the slope immediately below the wire.
- Electrical height is the wire's clearance above the relevant ground in wavelengths, and it can vary along the conductor.
- Terrain or path horizon is the elevation angle of the visible obstruction profile in a chosen azimuth. A ridge, valley or downhill opening can make it very different from geometric zero elevation.
A wire can be geometrically horizontal yet have one end much farther above the soil. It can be parallel to a hillside yet tilted relative to the distant path. It can also clear the local ground while still looking into a ridge. Saying only “level” leaves the most important reference frame unstated.
Free-Space Pattern Is Not the Realised Site Pattern
In free space, the radiation pattern is defined relative to the antenna's own geometry and coordinate system. An outdoor HF installation is not in free space. The direct field combines with fields reflected by lossy ground, and the complete result is affected by soil parameters, wire height, terrain, vegetation, support structures, feedline current and nearby conductors.
Terrain is therefore not merely a cosmetic or second-order detail. ITU-R material for HF antenna planning treats ground, the surrounding environment and topography as contributors to the practical radiation pattern. ITU-R diffraction methods likewise show why an irregular path horizon can change field strength beyond what a flat-ground elevation plot predicts.
The ground below the wire and the terrain farther along the wanted azimuth play different roles. Local ground strongly affects image-current interaction and loss; more distant terrain changes the path geometry and can introduce reflection or diffraction. A useful model declares both rather than reducing the site to one slope angle.
Compare Three Defensible Layouts
| Layout | What it preserves | What it changes |
|---|---|---|
| Horizontal to the horizon | The intended geometric orientation of a nominal horizontal section | Clearance, ground coupling and support loads can differ along the wire |
| Parallel to the slope | More nearly constant clearance above the local ground | The whole current structure is tilted relative to gravity and the path horizon |
| Deliberately aimed or stepped | A site-specific mechanical or pattern objective | Geometry, current phase, coupling and stress must be modelled and measured explicitly |
None of these rows is the universal winner. A modest tilt may make little practical difference in one electrically low installation and a material difference in a higher or electrically longer one. The same physical slope also represents a different fraction of a wavelength on each band, so a multiband wire needs a band-by-band answer.
If the purpose is a particular long-distance path, “downhill” is still not a substitute for an azimuth-and-elevation requirement. Survey the site, include the terrain profile and compare the realised pattern toward the wanted azimuth. A tilt that helps one lobe can move a null or reduce response elsewhere.
A Half-Square on a Slope
A half-square makes the conflict easy to see: two nominally vertical legs are connected by a top wire. Keeping the top wire horizontal preserves that part of the intended geometry, but the lower ends can have very different ground clearances. Following the slope can make the clearances more alike, but it tilts the top conductor and changes the endpoints of both legs.
I would keep the electrical conductor dimensions and feed arrangement deliberate, then change the support geometry rather than trimming one radiating leg merely because the ground rises beneath it. Equal physical leg lengths do not guarantee equal installed currents, but casually shortening one leg introduces another asymmetry before the original one has been measured.
The familiar half-square description is a starting mode, not proof of a fixed low-angle pattern. Current phase and magnitude depend on frequency, conductor dimensions, feedpoint, ground, height, bends and nearby structures. A half-square also does not make the return-current question disappear. The feed system needs two source terminals, and any current that reaches the outside of the feedline can become part of the radiating structure.
Dipoles, Loops and Travelling-Wave Wires Differ
A centre-fed dipole over a slope can acquire unequal coupling on its two sides even when the wire is horizontal. A loop has a closed conductor, but its feedline and surroundings can still disturb the nominal current symmetry. An electrically long wire can develop several lobes and nulls, making a small geometric change more consequential in some directions.
Rhombics, V-beams and Beverages should not be treated as casual exceptions to a dipole rule. They are travelling-wave or directional systems with their own conductor height, termination, orientation and ground-interaction requirements. Following terrain may be mechanically unavoidable or electrically intentional, but the result belongs to that particular design and path—not to a universal “always follow the hill” rule.
Keep Transformation and Common-Mode Control Separate
Sloping terrain and unequal surroundings can convert differential current into common mode, but a transformer label does not reveal the installed current path. Choose impedance transformation from the measured complex load at the transformer's intended terminals and verify its loss, voltage, current and thermal range. Then examine common-mode current as a separate circuit problem.
A choke can be useful where the intended return structure ends or where the feedline must cross an RF boundary. Its required impedance and position depend on the installed conductors and frequencies. “One at the feed and one at the shack” is not a measurement. On an end-fed system, choking before a declared return path has been provided can change the effective antenna rather than merely clean it up.
This is why I first draw every plausible return path: intentional counterpoise or radial conductors, coax exterior, mast, control cable, station bonding and capacitance to the surroundings. A clamp-on RF current probe and controlled cable-routing changes can then show which of them is participating.
Measure the Installed Antenna, Not Just SWR
Start with surveyed coordinates for the feedpoint, supports, bends and wire ends. Record conductor length, sag, height above local ground, slope profile, soil condition, nearby conductors, feedline route and the azimuth of the wanted paths. An NEC-family model can represent wires, networks, transmission lines and homogeneous ground; complex terrain may require another method or an explicit terrain treatment. In either case, demonstrate convergence and state what the model omits.
At the antenna, measure resistance and reactance at a declared reference plane. A low SWR at the transmitter can result from feedline transformation and loss and says little by itself about current symmetry, efficiency or pattern. Repeat current measurements on both intended conductors and around the complete feedline. Move only one variable at a time.
Pattern claims deserve pattern evidence. Use a calibrated range where practical or repeatable field measurements around the installation. For path comparisons, an A/B/A test with the same receiver, time window, bandwidth and propagation controls is more informative than a single signal report. Separate observed field strength from the reason inferred for it.
Mechanical and Electrical Safety Set the Boundary
A horizontal span across a hillside can put unusual tension on the higher support and can increase clearance at one end while reducing it at another. Allow for sag, wind, ice, tree movement and the fall zone of every support. Use fittings and insulators rated for the actual load; do not use an electrical conductor as a structural guy unless the complete assembly was designed for it.
Never infer a safe clearance from antenna performance. Keep the antenna, feedline, supports and tools away from overhead power lines according to the network operator's requirements. Electrical flashover can occur without direct contact. De-energise and secure the station while work is in progress.
RF exposure also belongs to the installed system. Evaluate the accessible field with actual power, duty cycle, frequency, antenna pattern, feedline current and occupied areas. A slope can place people or buildings closer to one part of the radiator than a flat-site drawing suggests.
Primary technical references
- IEEE 145-2025 — standard definitions for antennas and antenna systems
- IEEE 149-2021 — recommended practice for antenna measurements
- ITU-R BS.705-2 — HF antenna characteristics, ground and site effects
- ITU-R P.526-16 — propagation by diffraction over irregular terrain and obstacles
- Lawrence Livermore National Laboratory — NEC wire, current and pattern modelling
- Keysight — Network Analyzer Basics, impedance and reference planes
- Roy W. Lewallen, W7EL — Baluns: What They Do and How They Do It
- UK Health and Safety Executive — overhead-line clearance and flashover hazards
- ICNIRP — radiofrequency exposure guidelines, 100 kHz to 300 GHz
Begin with a clean geometry, then let the installed evidence decide. I normally compare a horizon-level baseline with any slope-following alternative. The decision comes from electrical height, current balance, realised pattern, loss, path horizon and safe mechanics—not from making the wire look level in a photograph.
Mini-FAQ
- Should a wire antenna always be level to the horizon? No. Horizon-level is a useful geometric baseline, but electrical height, ground coupling, terrain, current distribution and the wanted path determine the installed result.
- Is constant height above a slope automatically better? No. Following the ground can make clearance more uniform while tilting the complete antenna; compare its realised pattern and loss with a horizontal installation.
- Does tilting a wire downhill improve DX? Not necessarily. Downhill tilt can move lobes and nulls, but terrain, ground and current phase decide whether it helps a particular azimuth and elevation angle.
- Does a half-square on a slope need radials? The name does not answer that. Define both feed terminals and the intended return path, then measure current on the feedline, supports and any ground or counterpoise conductors.
- How many common-mode chokes should I install? There is no universal count. Select choke impedance and placement from measured common-mode current and the boundary of the intended return system on every operating band.
- Can a low SWR prove that the slope has not changed the pattern? No. SWR describes the match at one reference plane; pattern, current balance, loss and common-mode current need their own measurements or a validated installed model.