Is That Repeater Antenna Too Close to the Roof?
Is That Repeater Antenna Too Close to the Roof?
A Diamond X50, roughly 1.5 m of roof clearance, and a repeater with coverage problems. Before buying more gain, ask where the installed antenna is actually sending—and receiving—the signal.
RF.Guru working definition: Common-mode current is the non-cancelling phasor-sum current in a specified set of conductors, evaluated at a defined cross-section and using a declared current-direction convention. In the intended differential transmission-line mode, the outgoing and return currents are equal and opposite, so their phasor sum is zero. When they do not cancel, the remaining current must close through another reference or return path—such as the outside of a coax shield, a mast, equipment chassis, station wiring, nearby structures, earth, the operator, or distributed coupling through the environment.
This broader working definition is especially useful in practical antenna systems. On transmit, non-cancelling current on the outside of the coax can make the feedline and connected structures part of the radiating antenna system unless that path is intentional, clearly defined and properly controlled—for example by providing the required return path and placing a suitable common-mode choke at the correct boundary.
A repeater with coverage problems is often blamed on the transmitter, receiver, duplexer, coax or antenna gain. Sometimes that is correct. The photograph that prompted this article showed a Diamond X50 mounted roughly 1.5 m above a roof. It raised another question: could the installed radiation pattern be putting the useful signal somewhere other than the streets and stations the repeater is meant to serve?
The height was an estimate from the photograph, not a surveyed clearance. It cannot tell us the roof construction, hidden metalwork or actual coverage pattern. But it gives us a concrete installation problem to think through. On 70 cm, both that separation and the complete antenna are several wavelengths long. A roof is not just scenery around the antenna.
My point is not that the X50 is a bad antenna, or that a J-pole is automatically a better one. It is that the pattern you buy has to survive the place where you install it. If the useful lobe misses nearby users, extra peak gain is no consolation. A broader-pattern antenna can be the better engineering choice even when its catalogue number is smaller.
Choose coverage, not a gain contest. For this rooftop question, compare the antenna's response toward the wanted users, its interaction with the roof and mast, and the control of feeder current. Those are reasons to change an installation—not a blanket verdict that 1.5 m is always too low.
The X50 in the Photograph Is Not a Simple Whip
Diamond's current manufacturer pages identify the X50/X50N as a 144/430 MHz dual-band base antenna. The English page lists a 1.7 m overall length, approximately 19 cm radials, 4.5 dB gain at 144 MHz, 7.2 dB at 430 MHz, less than 1.5 VSWR, 50 Ω impedance, 200 W FM total input and a 60 m/s wind rating. The construction is described as 6/8-wave C-load at 144 MHz and three 5/8-wave C-loaded sections at 430 MHz. Diamond's Japanese page says the published gain figures are absolute gain, conventionally written dBi.
| Published item | What it supports | What it does not support |
|---|---|---|
| 144/430 MHz and 50 Ω | The intended bands and nominal port impedance | The installed feed-point impedance at every frequency and mounting geometry |
| 4.5/7.2 dBi | Manufacturer absolute-gain figures under its test/model conditions | Installed gain, close-in coverage or a site-specific elevation pattern |
| 1.7 m construction | Mechanical overall length | Array phase centre, required roof clearance or the location of every current maximum |
| 200 W FM total | A manufacturer input-power specification | Regulatory authorization, site exposure compliance, feeder/connector/duplexer rating or acceptable desense |
| 60 m/s wind rating | A manufacturer antenna rating | Approval of the mast, roof attachment, corrosion state, wind/ice load or building structure |
The important distinction for this case is the multi-section arrangement on 70 cm. The sections combine their fields to shape the elevation response; the 1.7 m housing is not a single uniform wire. The cited manufacturer pages do not give an elevation plot or beamwidth from which to locate this site's users. The gain figure therefore starts the question—it does not answer it.
The Wavelength Arithmetic Is Easy; the Boundary Is Not
λ = c/f
1.5 m ≈ 2.15λ at 430 MHz
1.5 m ≈ 0.73λ at 145 MHz
The arithmetic is correct when 1.5 m is the actual separation being discussed. But “height above the roof” needs a defined reference: roof plane to antenna base, lower radiator, radials, lowest current-carrying section or an inferred phase centre? A sloped metal roof, parapet, lift housing, railing, solar array and cable tray also cannot be reduced to one plane without checking the geometry.
The antenna's 1.7 m physical length is about 2.44 wavelengths at 430 MHz, yet that does not make it a single uniform 2.44λ radiator. It contains loaded sections. Overall length, electrical section length, element current and array phase centre are different quantities.
Clearance should be a measured performance requirement, not folklore. Define an allowed installed-pattern change, coverage margin, return loss, feeder current, intermodulation/desense, exposure boundary and structural load. Then determine the minimum geometry that satisfies all of them.
Why Roof, Mast and Nearby Hardware Matter
ITU-R BS.1195-1 documents the general mechanism clearly: supporting structures can behave as parasitic elements or secondary reflectors, asymmetrical mounting can skew a pattern, and platforms, stays, ladders, feeders and other antennas can influence it. The recommendation also notes that nearby structures modify radiation patterns according to their distance and physical characteristics, and that measured and calculated levels can differ especially near nulls and sidelobes.
At VHF and UHF, a roof can therefore change more than one thing:
- conductive surfaces can reflect and scatter fields;
- the mast, brackets, radials and roof edge can couple to the antenna;
- the feeder's outside surface can become part of the external current path;
- nearby antennas and metalwork can distort azimuth and elevation response;
- buildings, terrain and clutter can create diffraction and multipath far beyond the roof itself.
The installed pattern is the result. “The roof is a ground plane” and “the roof is a reflector” may both be useful approximations in a particular model, but neither phrase predicts the complete pattern without dimensions, materials, frequency and excitation.
The Best Lobe Is the One That Reaches the Users
A collinear does not create energy. Its sections combine fields so that more of the available radiation goes into selected elevation angles. A strong lobe near the horizon can be exactly what a flat-terrain distance service needs. A repeater above nearby streets or a valley may need substantial response below that lobe as well.
That is the useful comparison: at equal power delivered to the antenna, a lower-peak-gain design can give a stronger signal toward a user if it has more gain at that user's actual elevation and azimuth. It is not winning through mysterious efficiency. It is putting more of its radiation in the direction that matters. The same passive antenna pattern also matters on receive; receiver noise and desense remain separate parts of the uplink budget.
The direct geometric depression angle to a receiving point is:
θ = arctan[(hantenna − hreceiver)/dhorizontal]
For an antenna 30 m above its users, the straight-line geometry gives:
| Horizontal distance | Angle below the horizontal |
|---|---|
| 500 m | About 3.4° |
| 1 km | About 1.7° |
| 5 km | About 0.34° |
| 10 km | About 0.17° |
Thirty metres is an illustrative height difference, not a measurement of the photographed site. These angles locate users on an elevation plot; they do not predict path loss. Nor do they establish that the X50 has a null at any of those angles. Terrain, buildings, diffraction, polarization and multipath still belong in the link budget.
For a high site serving nearby users, I would therefore look at a broader elevation pattern, appropriate electrical downtilt or an array designed with null filling before simply buying more peak gain. For distant low-angle coverage, retaining a collinear and giving it a cleaner mounting environment may be the better answer. Both are deliberate engineering choices. ITU-R BS.1195-1 treats service-area nulls and beam tilt explicitly; it is not a recommendation that every repeater should use the same antenna.
Give Each Antenna a Job, Not a Universal Ranking
| Candidate | Why I would consider it | What the installation must preserve |
|---|---|---|
| Quarter-wave ground plane | A simple radiator and usually broad elevation response make a useful local-coverage candidate and reference installation. | A real radial/return system; a long active coax shield is not a substitute for it. Roof and mast coupling still matter. |
| J-pole | A half-wave radiator with a matching section can provide useful broad-pattern coverage without a separate quarter-wave radial set. | Control current on the feeder and mast, and keep the matching section clear of unwanted coupling. The name alone guarantees neither balance nor roof tolerance. |
| Half-wave, sleeve or folded dipole | A relatively simple, characterized pattern is often more useful than an unexplained high-gain claim. | The intended feed transition, orientation and mounting clearance; nearby metal can still reshape its response. |
| 5/8-wave vertical | Its shaped low-angle response can suit a more distant service region. | The specified feed and counterpoise geometry, plus clearance and feeder-current control that retain that response. |
| Multi-section collinear or engineered dipole array | Concentrated radiation, or deliberately designed downtilt/null filling, can put useful gain into the required service area. | Element phasing, mounting and the installed elevation pattern. Peak gain is valuable only where the users are. |
A broad useful lobe gives more angular margin when users occupy a range of elevations. That is the defensible reason to consider a simpler antenna—not a law that quarter-waves need less roof clearance than every J-pole, 5/8-wave or collinear. A roof can strongly disturb a simple radiator too. The candidate's actual pattern and installation decide the result.
Keep the Feeder from Becoming an Extra Antenna
The roof can affect both the fields around the radiator and the external current path through the mast, brackets and feeder. These are related but different mechanisms. A monopole's radial system provides its intended return; scattering from the building is not the same thing. Current on the outside of the coax can add another radiating conductor whose route changes the result.
For a rooftop installation, my practical aim is to make the intended antenna—not a convenient length of coax and roof hardware—set the coverage pattern. Plan the return path and feed-line isolation from the start. A suitable choke or an effective isolated feed arrangement can reduce that unwanted dependence. It cannot remove a reflection from the roof or replace missing radials, and one choke recipe is not suitable for every VHF/UHF antenna.
Use a calibrated clamp-on RF current probe, or a repeatable comparative probe, at several positions. Keep transmitter power, frequency and cable geometry fixed. If current or the field pattern changes materially when a known common-mode impedance is added, that is evidence of feed-line participation. Select the choke from measured complex common-mode impedance, frequency, voltage/current, temperature and environmental requirements—not from a generic dB label.
A Two-Ray Formula Is a Teaching Model
A direct ray plus one reflected ray can illustrate why field strength changes with height:
E ≈ Edirect + ΓEreflectede−jΔφ
Here Γ is a complex reflection coefficient and Δφ includes the path-length phase difference. The model requires a defined polarization, incidence angle and material, and normally assumes a sufficiently large, locally flat surface with transmitter and receiver in the applicable propagation region. A real rooftop with edges, penetrations, clutter and coupled conductors is not an infinite plane.
At 430 MHz, a quarter wavelength is only about 17 cm. A safely engineered height step of roughly 17–35 cm is therefore an electrically meaningful change, not merely cosmetic. If weak zones move after such a controlled change, geometry-dependent coupling or multipath becomes a stronger suspect. It still does not identify one specular roof reflection as the sole cause. Keep cable routing controlled and follow the access, structural and transmit-lockout precautions below; this is not an invitation to move an operating rooftop antenna by hand.
Diagnose the Repeater Before Changing the Antenna
- Separate uplink and downlink. A downlink-only complaint points toward transmitter, feeder or radiation-path problems; uplink-only trouble can indicate receiver sensitivity, local noise, duplexer loss or desense. Failure in both directions broadens the search.
- Verify power at declared planes. Measure transmitter power, reflected power and feeder/connector/duplexer losses with calibrated instruments. State whether power is at the transmitter, duplexer port or antenna connector.
- Test receiver health. Record sensitivity, local noise floor, duplexer isolation, transmitter leakage, intermodulation and desense under on-site operating conditions.
- Inspect the RF path. Sweep return loss or complex impedance at the antenna reference plane where practical. Check connectors, water ingress, cable damage, bonding and passive intermodulation risks.
- Map the service area. Use repeatable routes or fixed points, calibrated or at least stable equipment, both directions, documented antenna/power settings and comparable propagation conditions.
- Measure feed-line current. Probe several positions and repeat after one controlled configuration change.
- Use a reference antenna. A temporary, characterized installation at a safe approved position can separate antenna/site behaviour from repeater hardware.
- Model exact geometry. Include roof material, dimensions, parapets, mast, brackets, radials, feeder, nearby antennas and structures. Treat the model as a hypothesis until an installed measurement agrees.
- Change one variable. Height, lateral position, antenna, feeder route or common-mode impedance should be changed separately and logged.
Define Acceptance Before the Test
A useful site acceptance plan records:
- required uplink and downlink margin over the specified service area;
- maximum acceptable azimuth/elevation pattern deviation or defined field-strength minima;
- transmitter power, receiver sensitivity, duplexer insertion loss/isolation and desense limits;
- antenna-port return loss and feeder outside-surface-current limits;
- RF exposure, access-control, structural, wind/ice, weatherproofing, bonding and lightning requirements;
- measurement planes, instruments, calibration dates, uncertainties and environmental conditions.
“SWR is good” is not an acceptance result. A well-matched system can have a distorted pattern, feeder radiation, high site noise, duplexer trouble or poor coverage. Conversely, a small impedance change after moving the antenna does not prove that coverage improved.
RF, Structural and Lightning Safety Are Part of the Design
Do not perform rooftop experiments as casual antenna swaps. Inhibit and lock out transmission before touching the installation. Control roof access, use qualified fall protection, maintain electrical and RF clearances, and have the mast, attachments, corrosion state, wind/ice load and building structure checked by competent people.
RF exposure must be assessed under the rules that apply at the site, including simultaneous transmitters, antenna patterns, accessible areas and reflections. ITU-T K.52 (2024) provides guidance for compliance assessment of telecommunication installations, including rooftop situations; ICNIRP's 2020 RF guidelines cover human exposure from 100 kHz to 300 GHz. These references do not replace national law or a site-specific assessment.
Bonding, surge protection and lightning protection require a coordinated building system. Diamond notes that the antenna's DC-ground design is not protection against a direct lightning strike. Do not treat a DC continuity reading, a coaxial arrestor or a ferrite choke as a complete lightning-protection design.
Back to the X50 Above the Roof
The photograph does not convict the X50 or prove that 1.5 m is inadequate. It does expose the right design question: is a multi-section vertical, in that roof-and-mast environment, delivering its useful pattern toward the repeater's users?
If local users fall outside the strong part of the installed lobe, a broader-pattern antenna or deliberate downtilt can improve the service without increasing transmitter power. If the selected pattern already suits the service area, better placement and controlled feeder current can help preserve it. If the receiver is being desensitized or the feeder is lossy, repair that fault first. These are different remedies for different mechanisms.
I would choose a lower catalogue-gain antenna with dependable coverage of the required streets over a higher-gain antenna that misses them. The goal is not to win the specification sheet. It is to build a predictable repeater: a useful installed pattern, a healthy RF chain, controlled return paths and a mechanically and electrically safe site.
Primary Technical Sources
- Diamond Antenna: current X-series base-antenna specifications
- Diamond Antenna Japan: X50 construction, absolute-gain convention and DC-ground limitation
- ITU-R BS.1195-1: transmitting-antenna characteristics at VHF and UHF
- ITU-T K.52 (2024): guidance on EMF exposure compliance
- ICNIRP: 2020 radiofrequency exposure guidelines
Mini-FAQ
- Is 1.5 m above a roof enough for a repeater antenna? There is no universal number. Define the reference geometry, model the complete installation and verify installed coverage, feed-line current, exposure and structural limits.
- Is the Diamond X50 simply a 5/8-wave vertical? No. Diamond currently describes it as a 6/8-wave C-loaded design at 144 MHz and three 5/8-wave C-loaded sections at 430 MHz.
- Does low SWR prove that the roof is not affecting the pattern? No. Input match and radiation pattern are different measurements. A well-matched antenna can still have a distorted installed pattern.
- Must every rooftop vertical have a common-mode choke? No universal rule establishes that. Measure outside-shield current and the installed pattern, then use a qualified choke when the evidence and system requirements justify it.
- Does a coverage change after raising the antenna prove roof reflection? No. It implicates a geometry-dependent effect, which can include coupling, scattering, feeder current, obstruction, diffraction and multipath.
- Can a higher-gain vertical reduce close-in coverage? It can if the installed elevation pattern puts less energy toward the close-in service region, but the gain number alone cannot predict that outcome.
- What should be checked first when repeater coverage is poor? Separate uplink from downlink, then verify the transmitter, receiver, duplexer, feeder, antenna port, local noise and desense before attributing the problem to roof clearance.