Resonance Isn’t Your Radiation Pattern
Resonance Isn’t Your Radiation Pattern
An SWR dip can describe the impedance seen at one reference plane. It cannot show the three-dimensional field produced by the currents on the antenna, feed system and nearby conductors.
The myth I want to retire is that finding resonance tells us where the signal goes. It does not. Resonance is a port condition. Radiation pattern is a spatial field result. They are connected by the same electromagnetic system, but one number at the feedpoint does not determine a pattern in azimuth, elevation and polarization.
Feedpoint Resonance Is a Port Condition
At a declared reference plane and frequency, write the input impedance as Zin = Rin + jXin. The input is resonant when Xin = 0. That means the impedance at that plane is purely real. It does not mean 50 Ω, low loss, maximum gain or a desirable lobe direction.
The reference plane matters. A feed line transforms impedance, and a tuner can present zero reactance at the transmitter while the antenna terminal remains reactive. Even the phrase “the antenna is resonant” is incomplete until the frequency, feedpoint and measurement plane are stated.
Joeri’s short version: resonance answers “what is the input reactance here?” Pattern answers “how is the radiated field distributed over direction and polarization?” Do not ask the first measurement to answer the second question.
A Pattern Is a Three-Dimensional Function
A radiation pattern describes a field or power quantity as a function of direction under stated conditions. The current IEEE 145-2025 antenna terminology standard supplies the formal definitions. In practice, the result needs frequency, coordinate system, polarization, normalization and the reference quantity—field, power, directivity, gain or realized gain.
A familiar azimuth plot or elevation plot is only one cut through the 3D result. Two antennas can share one attractive cut and differ elsewhere. Null depth can also be polarization-specific: a co-polarized null does not prove that every cross-polarized component is absent.
ITU-T K.91 represents an antenna by a 3D function of elevation and azimuth and notes that numerical pattern results depend on antenna geometry and feeding arrangement. That is the right mental model: the full current-carrying structure produces the pattern.
Current Distribution Builds the Pattern
Every current element contributes a field with amplitude and phase. Those contributions add vectorially in space. Change conductor length, shape, orientation, spacing, loading, phase or return path and the current distribution can change. Once the current distribution changes, the lobes, nulls and polarization can change.
A matching component is not a steering knob by itself. If it only transforms impedance while leaving the intended conductor currents essentially unchanged, it need not materially reshape the pattern. But a matching or feeding change can alter the pattern when it changes current amplitude or phase, excites another mode, energizes the feed-line exterior or couples differently to the enclosure and nearby conductors.
Ideal Antenna Names Need Installed Conditions
| Example | Useful ideal picture | What must be declared in an installation |
|---|---|---|
| Half-wave dipole | A thin, straight, balanced half-wave in free space has current highest near its centre, approaching zero at open ends, with broadside maxima and nulls along the wire. | Actual electrical length, bends, conductor diameter, feed transition, balance, height, ground, nearby conductors and polarization. |
| Quarter-wave vertical | A straight quarter-wave monopole over an infinite perfect conducting plane is azimuthally symmetric. That symmetry belongs to the complete monopole-plus-return structure. | Radial or ground system, soil parameters, conductor loss, slope, mounting structure, feed-line exterior current and local obstructions. |
| Full-wave loop | A closed one-wavelength conductor supports a distributed current and is not described by the uniform-current pattern of an electrically small loop. | Shape, circumference, plane, height, feed location, polarization, imbalance, nearby objects and ground. |
| Shortened or loaded element | Loading can place an intended mode on a physically shorter structure. | Load position and loss, conductor current profile, radiation resistance, bandwidth, efficiency, support coupling and thermal limits. |
These ideal pictures are starting points, not gain promises. The in-force ITU-R BS.705-2 treats dipole height, vertical-monopole earth systems, ground conductivity, topography and nearby site structures as explicit pattern variables. Antenna type alone is not a complete model.
Feedpoint Position Does Not Vanish From OCF and EFHW Systems
Moving a feedpoint is not a direct beam-steering command. In the ideal limiting case, two excitations that create the same current distribution on the same conductors create the same far field. Real off-centre-fed and end-fed systems rarely let us ignore the feed structure so cleanly.
An off-centre feed changes the terminal impedance and can create unequal conductor currents at the transition. An EFHW feed sits near a current minimum and voltage maximum, but it still needs a complete RF return path. That return can include a counterpoise, coax exterior, mast, equipment chassis, station wiring and distributed capacitance. If any of those paths carries material current, it belongs in the geometry and can alter pattern and polarization.
A common-mode choke can help define the boundary, but its need, location and impedance are installation- and frequency-specific. The ARRL measure–install–retest workflow starts by measuring exterior-coax current over the operating frequencies. That is better evidence than assuming every feedpoint choke makes the feed line electrically disappear.
Height and Ground Need Polarization-Specific Analysis
For a horizontal HF dipole, changing height in wavelengths changes the phase relationship between its direct field and the ground-reflected field. The elevation lobes and nulls can move substantially. Ground reflection depends on polarization, incidence angle, conductivity, permittivity and terrain, so “below half a wavelength is high angle” is not a universal rule for every soil, orientation or antenna.
A vertical monopole is a different boundary-value problem. Its return system and soil loss matter, and moving the feedpoint upward is not equivalent to raising a horizontal dipole. A horizontal loop, vertical loop and sloping loop are different again. State antenna orientation and polarization before applying any height rule.
Height can be one of the strongest controls in a specific installation, but it is not automatically the best improvement. Raising an antenna can move it closer to a tower, roof, power line or another antenna; that coupling may matter more than the nominal height change.
Surroundings Can Join the Radiating Structure
Metal roofs, gutters, fences, masts, towers, cables and other antennas can carry induced current and reradiate. Dielectric objects and vegetation can detune, absorb energy or change coupling. Weather can change soil and vegetation properties. The direction and size of the resulting pattern change depend on the complete geometry; there is no universal “metal always hurts” or “rain always lowers gain” rule.
This explains a common field observation: moving a feed line or support can move both the impedance and a received-signal null. Resonance did not steer the beam. The effective current-carrying structure changed.
Keep Resonance, Match, Efficiency and Gain Separate
Resonance: input reactance is zero at a stated port, plane and frequency.
Match: the input impedance bears the intended relationship to the system reference impedance, normally judged through reflection coefficient or SWR at a stated plane.
Radiation efficiency: radiated power divided by power accepted by the antenna.
Directivity: how radiation intensity is distributed relative to its angular average.
Gain: directivity with antenna efficiency included.
Realized gain: gain with input mismatch included under the applicable definition.
Low SWR proves none of the last four quantities. A resistor can make an excellent match and a poor antenna. A mismatched antenna can retain a useful intrinsic pattern while delivering less accepted power from a particular transmitter/feed-line system. Pattern shape, gain and realized gain must therefore be named separately.
Resonance can still matter operationally. Large reactance can increase voltage, current and loss in a practical tuner or feed line. Changing wire length to move resonance can also change the current distribution. Those are real couplings, but they do not turn resonance into a pattern measurement.
Model the System That Carries Current
A useful model includes the conductors that can carry material current: radiator, radials or counterpoise, feed-line exterior where relevant, mast, loading, matching enclosure, nearby metal and a suitable ground model. Declare frequency, geometry, conductor properties, excitation, loads, segmentation, soil and requested polarization.
LLNL’s Numerical Electromagnetic Code can report model currents, radiation patterns and near fields for wire and conducting-surface structures with ground, loads, networks and transmission lines. A model still needs convergence checks and comparison with measurements; a detailed drawing does not guarantee correct material properties or boundary conditions.
Inspect currents before admiring the colored pattern. Unexpected current on a feed line, mast or “inactive” conductor is a clue that the modelled antenna is larger than the named radiator.
Measure a Pattern as a Pattern
A defensible pattern measurement uses a suitable antenna range or a valid near-field scan transformed to the far field. It records frequency, polarization, coordinate system, antenna orientation, cable routing, reference level, dynamic range, alignment, reflections and uncertainty. IEEE 149-2021 is the active recommended practice for antenna measurements.
Pattern uncertainty is not a footnote. NIST’s antenna-measurement work identifies theory, simulation and controlled changes to the measurement system as ways to estimate it. At HF, site reflections, range geometry, source stability, polarization alignment, cable current and background signals can dominate a casual test.
WSPR, Reverse Beacon Network reports and distant beacons can compare complete station performance when tests are rapidly alternated and propagation, time, power, receiver population and polarization are controlled as far as possible. They do not by themselves recover a calibrated 3D antenna pattern. Treat them as propagation-weighted field observations, not as an anechoic range.
Use an Order of Operations That Matches the Question
- Define the coverage objective. State frequency range, polarization and the azimuth/elevation regions that matter.
- Choose geometry and height. Include ground, terrain and nearby structures rather than naming only the radiator.
- Define the complete return path. Decide where differential current should end and measure exterior-feed-line current.
- Model currents and the 3D pattern. Inspect co- and cross-polarized results and more than one 2D cut.
- Design the match. Control loss and stress without pretending the matching network proves the pattern.
- Measure impedance at a declared plane. Record resonance and SWR for what they actually describe.
- Validate the field result. Use a calibrated pattern method where the claim needs pattern accuracy, or label controlled on-air comparisons as system observations.
Stop asking the SWR dip where the signal goes. Start with the current distribution on the complete installed structure, calculate its 3D field with ground and surroundings, then measure the quantity you intend to claim. Resonance remains useful—it simply has a different job.
Mini-FAQ
- Does resonance determine an antenna’s radiation pattern? No. Resonance means zero input reactance at a stated port, plane and frequency. Pattern is the directional and polarization-dependent field created by currents on the complete structure.
- Will two antennas resonant at the same frequency have the same pattern? No. They can have different geometry, current distribution, polarization, height, ground, surroundings and return paths while sharing the same resonant frequency.
- Does raising a dipole always produce a lower takeoff angle? No universal height rule covers every ground, terrain, polarization or geometry. Analyse the horizontal antenna at its height in wavelengths with the actual ground and nearby structures.
- Do OCF and EFHW feed positions change pattern? Feed position is not a steering control by itself, but it changes impedance and can change balance and return current. Any current on coax, counterpoise, mast or station wiring can alter the installed pattern.
- Does low SWR prove good gain or a useful pattern? No. SWR describes reflection at a reference plane. It does not measure efficiency, directivity, gain, realized gain or the location of lobes and nulls.
- Can WSPR or Reverse Beacon Network reports measure my pattern? Not by themselves. They combine antenna behavior with propagation, time, polarization, receiver locations and station calibration. Use them for controlled system comparisons, not as a calibrated 3D pattern.