5.8 GHz Antenna Gain and Beamwidth: What the Numbers Change in Practice

The number on the antenna is a trade, not a score

Two 5.8 GHz antennas sit side by side in the same shop listing. One is sold as 2 dBi, the other as 9 dBi. The obvious inference — the 9 dBi one flies further — is the one that costs people the most flights, because it is true only while the aircraft stays inside a cone that the headline number never mentions.

A 5.8 GHz antenna does not amplify anything. It is a passive shape that takes the power arriving from the transmitter and decides where that power goes. Gain is how much more of it goes in one direction than a perfect sphere would send; beamwidth is how wide that direction is. They are the same property described from two ends, and every antenna you can buy sits somewhere on that trade. This is the starting point for the wider FPV antenna selection and installation problem: the antenna that wins is the one whose beam still covers your aircraft at the attitudes you actually fly, not the one with the biggest number on the box.

Gain means directivity, not extra power

Antenna gain is a ratio, not a quantity of energy. It compares the power density an antenna produces in its strongest direction with what an isotropic radiator — a theoretical point source radiating equally in every direction — would produce from the same input (antenna gain). Because it is a comparison, it is always quoted against a reference, and the reference is written into the unit:

• dBi — referenced to an isotropic radiator in free space.
• dBd — referenced to a half-wave dipole, which itself has about 2.15 dBi of gain. A dipole-referenced figure therefore reads about 2 dB lower than the same antenna in dBi.
• dBic — referenced to a circularly polarized isotropic source. A linear antenna radiates equal power in both senses of rotation, so measuring it against a circular reference splits that power between the two senses and the same physical antenna reads about 3 dB lower in dBic than in dBi. Treat it as a different scale, not a translation you can eyeball.

Mixing those units up is the most common spec-sheet error in FPV, and it is worse than it sounds, because the number that actually matters to the link is not gain at all — it is EIRP, the effective isotropic radiated power: transmitter output in dBm, plus antenna gain in dBi, minus everything lost between them in dB (EIRP). Those are three different units in one sentence, and they do not cancel each other out.

Worked once, with numbers picked only to show the arithmetic: a transmitter set to 200 mW is about 23 dBm. Add a 9 dBi antenna, subtract 1.5 dB of feed and connector loss, and the link sees roughly 30.5 dBm EIRP — about 1.1 W radiated in the peak direction. Set the same transmitter to 25 mW (14 dBm) behind the same antenna and the link sees about 21.5 dBm. The antenna did the same amount of work in both cases; what changed was the power handed to it. That is why antenna gain and transmitter power, heat and power supply are two levers on one budget, and why a bigger antenna on a low power setting can be the cheaper way to reach the same place — with a beam that is much fussier about where you point it.

Beamwidth is the other half of the same bargain

Beamwidth is normally quoted as the 3 dB beamwidth: the angle between the two directions where radiated power has fallen to half its peak value (beamwidth). Push more of the pattern into one direction and the pattern that is left over has to get narrower — the total power is fixed, so the solid angle it is spread over shrinks. Double the gain in linear terms and, broadly, you halve the solid angle the beam covers. There is no antenna that is both very high gain and very wide — not at any point in the market, from any manufacturer.

That is why gain without beamwidth is half a specification. The table below gives the ranges you will typically meet in FPV hardware. The numbers are indicative only — the values that matter are the ones on the datasheet of the exact part in your hand, and even those deserve the scrutiny in the next section.

Notice that the row with the most gain also has the least margin for error, and that the two circular rows are quoted in a unit that is not comparable to the linear ones. A reader who sorts that table by the middle column alone will buy the helical, mount it on a tripod, and then discover that the launch is the easy part.

What a narrow beam does to the picture when the aircraft banks

A beamwidth figure describes a static pattern in free space. The aircraft does not sit still inside it. Three things move the geometry every second of a flight:

Attitude. A patch with 60° of vertical beam, aimed at the horizon, loses a quad that climbs to 45° above it and keeps going. The beam did not change; the target left it. An omni handles the same climb without complaint and then has nothing to offer when the aircraft is 2 km out on the deck.

The airframe itself. Carbon plate, a battery pack and the pilot's own body all sit between the antennas for part of every flight. A narrow beam pointed at a known direction has no path around an obstruction; a wide pattern has more paths, and circular polarization lets the link use whichever reflection arrives with the right sense of rotation.

Polarization. A circular signal keeps its sense through a bank or a roll; a linear one rotates with the airframe and fades as it crosses the receiver's plane. This is why circular antennas dominate FPV video at 5.8 GHz despite giving up gain to do it (circular polarization). Antenna selection for the two ends is a pairing problem, covered in the frequency bands and link engineering cluster — a high-gain antenna on one end does not compensate for a polarization mismatch on the other.

A narrow beam buys reach on one heading and spends it everywhere else. This one is committed to the valley in front of it, and effectively blind to everything behind.

The 6 dB rule, and why it is optimistic

Free-space path loss grows with the square of distance: every doubling of range costs about 6 dB (ITU-R P.525). Invert that and you get the rule of thumb everyone quotes — 6 dB of extra gain, or 6 dB of extra transmitter power, buys you double the distance.

The arithmetic is right and the conclusion is usually wrong, for three reasons.

First, the 6 dB rule assumes the antenna pattern does not change. Going from a 2 dBi omni to a 9 dBi patch does not hand you 7 dB in every direction — it hands you 7 dB along the boresight and takes away coverage everywhere else. At the moment the aircraft turns, the budget you were counting on is pointing at empty sky.

Second, free space is the optimistic case. Over ground, with a ridge or a treeline between you, the loss grows faster than the square of distance, and the Fresnel zone — the ellipsoid of air the signal actually needs — gets clipped by terrain long before the path looks blocked. A high-gain antenna with a narrow vertical beam is the worst possible tool for flying low behind terrain: its vertical pattern has little margin, so a few degrees of nose-up attitude or a ridge a few hundred metres away takes the aircraft out of the main lobe entirely.

Third, the receiver is half the budget. Gain applied at the transmitter raises the signal at the receiver, but if the receiver is already at its sensitivity floor, or the link is digital and sitting on a cliff edge, the last few dB behave nothing like the first few. The link budget has to be balanced at both ends; a huge number on one side is a hint that the other side is the constraint.

Why the 9 dBi antenna does not automatically win the flight

Put the trade in flight terms. You are at the far end of your normal range, the aircraft is a dot, and the picture starts to break. Four different things can be happening, and only one of them is fixed by more gain:

1. Not enough signal in the beam. Genuine gain shortfall — the aircraft is inside the pattern and the link is simply weak. More gain helps here, and only here.
2. Signal outside the beam. The aircraft banked, climbed, or the antenna drifted on its mount. More gain makes this worse, because the pattern's edges get steeper.
3. Loss before the antenna. A long, thin pigtail, a worn connector, or an SMA/RP-SMA mismatch quietly removes the dB you paid for. Gain added downstream of a lossy feed is gain you never see at the radiator. Check the connector and cable before buying a bigger antenna.
4. Interference, multipath or a polarization fault. The signal is strong and the picture is still bad. This is a fault-tracing problem, not a gain problem, and adding directivity does not remove multipath; it changes the failure mode — instead of a gradual fade the link drops out when the aircraft leaves the narrow main lobe.

The uncomfortable practical consequence: a high-gain antenna is a ground station decision, not an airframe decision. On the aircraft it costs weight, drag and crash damage, and the beamwidth you give up is the one that was covering your turns.

How to read a 5.8 GHz antenna datasheet

Most of the argument above collapses if the spec sheet is read properly. Six things to check, in order:

• Which reference the gain uses. dBi, dBd or dBic. If two candidates are quoted in different units, convert before comparing, but convert like with like: the 3 dB gap described above applies to a linear antenna measured against a circular reference, and a circular antenna's own dBic rating is roughly its dBi rating, not 3 dB below it.
• Beamwidth in both planes. A single "beamwidth" number usually hides an asymmetric pattern: wide horizontally, narrow vertically. The narrow plane is the one that decides whether a climb or a dive breaks the link.
• Polarization and axial ratio. A circular antenna with a poor axial ratio is radiating an ellipse, not a circle, and behaves like a partially linear antenna when the aircraft banks. The sense — left or right hand — must match the other end exactly.
• Bandwidth against the channels you actually use. FPV video at 5.8 GHz spans a wider stretch than the four or five channels most people fly, and some transmitters extend down toward 4.9 GHz. An antenna tuned for the top of the band is a poor match at the bottom. Check the manufacturer's stated range against your channel plan, and check the transmitter's own table for what it supports.
• Connector type and gender. SMA, RP-SMA, MMCX and U.FL are not interchangeable, and a gender mismatch is a five-second error with a five-minute consequence. Read it off the product table rather than from a photo. For reference, MASA's own product tables list an SMA 50 Ω connector on the 4.9–6.0 GHz 5 W transmitter and an MMCX connector on the 2.5 W unit — which is exactly the kind of difference that has to be confirmed per model, not assumed across a range.
• What was measured with what. Gain quoted for a bare antenna, with no cable, on a ground plane you do not have, is an optimistic number. VSWR matters too: a poorly matched antenna reflects power back into the transmitter instead of radiating it.

A field procedure that separates two antennas in one session

You cannot settle this on paper once the candidates are within a few dB of each other. Settle it on the ground, with one variable at a time:

1. Fix everything that is not the antenna. Same channel, same power setting, same mounting height, same battery. If you change the antenna and the power setting in the same session, you have learned nothing.
2. Change one end at a time. Transmitter antenna, then receiver antenna. Swapping both at once hides which end did the work.
3. Fly the same line twice. Out and back on a fixed heading at a fixed altitude, then the same line with a deliberate worst-case attitude — banked hard, climbing overhead, and turning away behind you. The second pass is where narrow beams lose.
4. Log where it breaks, not how far it goes. "Clean to the far treeline, broke on the turn at the ridge" is a usable result. "Went 2 km" is not, because it says nothing about the flight you will actually have.
5. Repeat at the far end of your normal range. Antennas that are indistinguishable at 300 m separate cleanly at 2 km, and that is the distance you bought them for.
6. Write the result down. Gain, beamwidth, polarization, connector, channel and the break point. Six months later that note is the only thing standing between you and buying the same wrong antenna again.

If the test result is ambiguous, the antenna is not your constraint — the scenario and link design is, and the next thing to change is geometry: height, route, or where the ground station stands, not the antenna.

What to carry away

Gain and beamwidth are one number wearing two hats. A 5.8 GHz antenna with more gain is not a better antenna; it is a more committed one — committed to a direction, to an attitude, and to a mounting position that keeps the aircraft inside the pattern. The useful question is never "which has more gain", but "how much of my flight does this pattern actually cover, and what does it stop covering".

Read the datasheet for the reference unit, both beamwidth planes, the polarization sense and the connector. Then decide the ground station's pattern from the geometry of the flight you fly, not from the number in the product title. For the pairing and mounting decisions that follow from this, start from the antenna selection pillar, and for the question of how much power is worth pushing through that antenna once it is chosen, see transmitter power, heat and power supply.

A last note on the regulatory side: radiated power is limited by the rules where you fly, and those limits are written in EIRP, not in transmitter output. The FCC's part 15 rules for this band (47 CFR 15.247) are one example of how a gain increase interacts with a power limit; local rules differ, and none of them are settled by the antenna's spec sheet. Check the rules that apply to your operating location, and treat antenna gain as part of the emission, because that is how the regulator treats it.