Fan Vertical Mutual Coupling: The Tuning Trap

The recurring question is fair: if one wire is resonant and the others are not, why worry about their interaction? Because “not resonant” does not mean “electromagnetically absent.” Each branch has a self-impedance, and every nearby branch contributes mutual impedance. The current on one conductor therefore depends on the entire structure.

A well-designed fan vertical can work very well. Mutual coupling is not an automatic failure. It is also not automatically a minor SWR shift. The engineering question is whether the coupled current system still produces the loss, bandwidth and installed pattern we intended.

One Feedpoint Creates One Coupled Antenna

At the common feedpoint, the transmitter excites all connected conductors. The branch nearest its useful mode may carry the largest current, but the other branches present frequency-dependent impedances and support induced current. Their length, spacing, angle, conductor diameter, end geometry, supports and surroundings affect both magnitude and phase.

The radial system, soil and feed-line exterior complete the current path. They are not background scenery. Change one branch and the measured feed impedance can move because the branch currents changed, the return currents changed, or both.

That is why the picture of several independent quarter-wave antennas joined at one point is only a first approximation. Once assembled, it is one multiconductor electromagnetic structure.

Why Nearby Bands Demand More Attention

When two branches differ greatly in electrical length, one may dominate while the other presents a less troublesome reactive load. As the bands move closer, their physical lengths and modal behaviour become more alike. The same feed voltage can then support stronger induced currents, and adjustment of one branch can move several resonances.

The upper-HF group is a useful example. A 12-metre branch sits between 10 and 15 metres in frequency and length. A 17-metre branch sits between 15 and 20 metres. Adding a separate conductor for every one of those bands can create a dense cluster whose resonances remain tunable but whose currents are harder to separate.

This does not prove that a five-band fan vertical must fail. It explains why “one resonant wire per band” is not automatically the lowest-risk architecture. If a narrow adjacent band can be operated through a measured tuner/load region, omitting a branch may reduce the coupling and mechanical burden. The resulting antenna must still be checked on that tuner-assisted band; a match alone does not certify pattern or efficiency.

Staggered Wires for One Wide Band Are a Different Problem

Several slightly different wires can also be used deliberately to broaden one band. That is not equivalent to adding independent full-band resonators for several adjacent bands. The length offsets are smaller, all branches serve the same operating range, and the design objective is a controlled composite impedance and current response across that band.

The wide 10-metre amateur allocation is a case where that approach can be worth investigating. The branches still couple; coupling is part of the design rather than an accidental side effect. Their small physical differences may disturb the vertical-current geometry less than a cluster of conductors sized separately for 10, 12 and 15 metres. That remains an installed result to model and measure, not a general promise.

A Good SWR Dip Does Not Settle the Question

SWR reports the magnitude of reflection at a declared reference plane. It does not identify which conductor carries current, how much accepted power becomes heat, or where the radiated power goes. Feed-line attenuation, ground loss, conductor loss and matching loss can all make an impedance curve look easier.

Do not infer takeoff angle from electrical length alone after the antenna has become a coupled multibranch system. Elevation pattern follows current magnitude and phase over every radiator and return conductor, together with height, ground and nearby structures.

Spacing Helps, but No Fixed Gap Is Universal

Increasing physical separation or changing branch angles often reduces coupling. There is no installation-independent 20- or 30-centimetre rule. The relevant spacing is electrical as well as physical, and coupling depends on the full parallel length, orientation, conductor diameter, termination and environment.

Small construction variations do not reliably “prevent problematic resonance.” They can just as easily make one side different from the other, shift a branch into a stronger interaction, or introduce feed-line exterior current. Treat construction tolerance as a sensitivity case in the model and as a repeatability check in the prototype.

Reduce Complexity Before Trying to Tune It Away

When a dense fan is difficult to stabilise, I prefer to question the branch count before adding more corrections. The useful options include:

• Omit an adjacent-band branch when the measured antenna/tuner/feed-line system can operate that band acceptably.
• Increase branch separation where the mechanical design allows it, then retune the complete antenna.
• Change the branch arrangement so long parallel high-voltage regions do not sit unnecessarily close.
• Use switched radiators when independent elements matter more than feed simplicity.
• Use a deliberate same-band multi-resonator design when bandwidth, rather than extra band count, justifies the coupled branches.
• Accept a controlled tuner-assisted band only after measuring load range, loss, current, stress and pattern.

Loading coils and folded conductors can solve real space constraints. A coil’s magnetic field is not magically isolated from neighbouring conductors, and a universal 60–75% placement rule does not follow. Coil loss, placement, self-capacitance, current distribution and nearby coupling all enter the complete design.

A Measurement Programme That Can Decide

• Freeze the geometry. Record every radiator length, angle, separation, support, feed-line route, radial layout and nearby conductor.
• Calibrate at the antenna plane. Save complex impedance and S11 across all bands before and after each branch change.
• Map branch currents. Measure magnitude and, where practical, phase at repeatable positions near the base and farther along each conductor.
• Measure the feed-line exterior. A changed common-mode path can alter both the apparent match and the installed pattern.
• Model the complete structure. Include all branches, radial/ground return, feed conductor and important supports; compare current and three-dimensional pattern, not only impedance.
• Test loss and stress. Account for tuner, loading component, conductor and ground loss at a stated power, waveform and duty cycle.
• Compare A/B/A. Remove or reposition one branch, repeat the measurements, then restore it to expose drift or connector movement.
• Verify the field. Compare installed patterns or stable field/SNR observations at equal accepted power and stated uncertainty.

Primary and Authoritative Technical Sources

• NASA, Phased Array Antenna Handbook—mutual coupling represented through a multiport network and its effects on element impedance and pattern.
• NASA, Modeling and Simulation of Phased Array Antennas—why coupling and platform effects alter input impedance, realised gain and radiation pattern.
• Lawrence Livermore National Laboratory, Numerical Electromagnetic Code v5—complete wire, load, ground, current and pattern modelling.
• IEEE Std 149-2021, Recommended Practice for Antenna Measurements—impedance, pattern, gain, efficiency, test-site and uncertainty practice.
• IEEE Std 145-2025, Definitions of Terms for Antennas—consistent distinctions among impedance, pattern, directivity, gain and efficiency.
• Recommendation ITU-R BS.705-2—HF transmitting-antenna characteristics and pattern context.

Joeri’s Bottom Line

Mutual coupling in a fan vertical is neither a disaster by definition nor something I dismiss because the SWR dips can be moved back. If extra branches crowd adjacent bands, the analyser can tell us that we repaired the port while saying nothing about the currents and pattern we wanted.

Use only as many branches as earn their place. Treat same-band staggered resonators as a deliberate bandwidth design, not as proof that every adjacent-band wire is harmless. Then verify the whole antenna: impedance, branch and exterior currents, loss, stress and pattern.

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