Carbon Fiber Drone Propellers: The Cheapest Part, The Tightest Constraint

A carbon fibre propeller is the cheapest part on a drone that changes what the drone can do. It sets endurance, payload, control response and acoustic signature — and in 2026 the thing standing in the way of making more of them is not fibre supply. It is the mould, the cure cycle and the qualification file behind every blade.

Four blades per aircraft, counted in the millions

Every multirotor carries at least four propellers — six on a hexacopter, eight on an octocopter — and the industry convention is “typically four or more per aircraft.” Multiply that into the volumes now moving through the drone economy and the propeller stops looking like a consumable accessory. Ukraine’s defence industry, which the country’s National Security and Defence Council said in January 2026 had capacity for more than 8 million FPV drones a year, and which President Volodymyr Zelensky said in July 2026 was actually producing 10 million a year, implies an annual propeller requirement of about 40 million units (Pheonix Research calculation: 10,000,000 airframes × 4 rotors per quadcopter).

Most of that volume is not carbon. FPV airframes overwhelmingly fly injection-moulded polymer blades, and the reason is price: as Ukrainian industry has found, domestically made components run 10–20% more expensive than imports, and Chinese parts are cheaper still. Carbon fibre propellers are the premium end of the same part — the end where endurance, payload and noise are worth paying for.

Why carbon wins the specification

The mechanical case is straightforward and old. A propeller converts motor torque into thrust, and the blade has to hold its designed pitch while doing it. A compliant blade twists under load, shedding thrust precisely when the aircraft is heaviest or climbing hardest; a stiff blade does not. Carbon fibre-reinforced polymer offers that stiffness at a fraction of the mass of aluminium or glass-reinforced polymer, and mass at the blade tip is the worst place to carry it — every extra gram costs rotational inertia, motor torque and control bandwidth.

The acoustic case is newer and, commercially, worth more. Joby Aviation, which began manufacturing propeller blades at its Dayton, Ohio facility in late 2025, describes the blades as “central to its low acoustic profile” — a decade of design work compressed into the component that shapes whether the aircraft is acceptable over a neighbourhood. Each of its propellers carries five blades and each aircraft carries 30, a blade count that turns one aircraft sale into a repeating order book.

Which is why a part that used to be bought off a hobby shelf is now treated as a conforming, inspected component. Joby’s Dayton line is being built to produce up to 15,000 blades a year — enough for about 500 aircraft sets (Pheonix Research calculation: 15,000 blades ÷ 30 blades per aircraft) — and its blades must meet FAA inspection and certification requirements before installation on flight test aircraft.

The constraint is the cure, not the fibre

Ask what limits propeller output and the instinctive answer is carbon fibre supply. The evidence points elsewhere. When Mejzlik Propellers, the Brno-based specialist making CFRP propellers since 1989, wanted to move from prototypes to production rates, the project it ran for roughly three years was not a materials programme. It was a process programme: a prepreg compression moulding line, online in January 2024, that cut cure time to 7–10 minutes per propeller against 8–10 hours for wet layup or oven cure — a compression of roughly 64× on midpoint values (Pheonix Research calculation: 540 minutes ÷ 8.5 minutes = 63.5).

The catalogue is 150 off-the-shelf patterns from 0.3 to 2 metres in diameter, with custom parts to 3 metres. What changed was what customers asked for. “Two years ago, nobody needed several hundred props per month,” its team told CompositesWorld. “Now, we have customers who need a few hundred per month or even per week.” Orders that begin at four or eight propellers for testing now ladder into hundreds a month.

That is the real shape of this supply chain. Fibre is available; qualified moulds, press time and inspection capacity are not. Every new blade design is a new tool, a new cure recipe and a new qualification file, and none of that scales by buying more tow.

Three demand curves, only one of which pays carbon prices

The demand pulling on this capacity is not one market. It is three, moving at very different speeds.

The United States is the slow, rich curve. The FAA’s Aerospace Forecast for fiscal 2026–2046 puts the commercial small UAS fleet — aircraft above 0.5 lb and up to 55 lb — at approximately 1.1 million at the end of 2025, growing to 1.5 million by the end of 2030, a 36.4% increase, or about 6.4% a year (Pheonix Research calculation: 1.1 million to 1.5 million over five years). The recreational fleet is projected to peak over the next five years, from 1.55 million units to roughly 1.62 million by 2030, at a 0.9% CAGR. These are the buyers who can pay aerospace-grade prices.

Ukraine is the fast, poor curve. Capacity above 8 million FPV drones a year as of January 2026; 10 million produced annually according to the president in July, against a stated target of 20 million; and an assessment from NATO’s deputy military commander, Air Chief Marshal Johnny Stringer, putting Ukrainian output “well north of 5 million” for 2026 against 5,000 drones in 2022. Ukraine’s defence ministry said in October 2026 that its procurement agency had contracted for 5.5 times as many drones as in the same period of 2025 and taken delivery of almost 3.7 times as many. This curve buys volume, not specification.

The third curve is institutional: the U.S. Department of War’s Replicator initiative, announced in August 2023 and aimed in its first phase at fielding thousands of autonomous systems across multiple domains, with Replicator 2 now pointed at counter-small-UAS capability and Joint Interagency Task Force 401 making its first purchase — two DroneHunter F700 systems — in January 2026. Small in unit terms; decisive in what it certifies as acceptable.

One curve counts aircraft in millions. One counts them in hundreds of thousands. One buys carbon.

What the fibre producers’ numbers actually show

The materials side corroborates the split. Toray’s carbon fibre composite materials segment closed the year to March 2026 with revenue of ¥300.1 billion and core operating income of ¥17.6 billion — down 21.7% year on year — with the company pointing to price correction and structural reform in industrial applications, an adjustment phase in pressure vessels and delayed recovery in wind turbine blade applications. Inside the same segment, aerospace sales rose from ¥104.9 billion to ¥118.4 billion, a 12.9% increase (Pheonix Research calculation: ¥118.4bn ÷ ¥104.9bn − 1).

Hexcel tells the same story from the other end of the chain: sales of USD 1,894 million in 2025, essentially flat on 2024, fourth-quarter sales of USD 491 million, with commercial aerospace in the quarter up 7.6% to USD 299.5 million, and 2026 guidance of USD 2.0–2.1 billion, roughly 8% growth at the midpoint. The company’s own framing is that when commercial aircraft OEMs hit their production targets across all programmes, it generates approximately USD 500 million of incremental annual revenue — 26.4% of its 2025 sales (Pheonix Research calculation: 500 ÷ 1,894).

Both numbers point the same way. Aerospace-grade capacity is being absorbed by an aviation rate ramp that is already contracted. A drone or eVTOL blade programme buying qualified prepreg is bidding for the same fibre as a narrowbody, at a tiny fraction of the tonnage. The scarce resource is not carbon fibre. It is carbon fibre that arrives with a qualification file attached.

The rules stop at the electronics

Then there is the regulatory line, which is drawn somewhere unexpected. On 22 December 2025 the FCC added uncrewed aircraft systems and UAS critical components produced in a foreign country to its Covered List, on a national-security determination that foreign-produced UAS equipment poses unacceptable risk. The critical components it enumerates are data transmission devices, communications systems, flight controllers, ground control stations and controllers, navigation systems, sensors and cameras, batteries and battery-management systems, and motors — a list that is expressly not exhaustive, but is almost entirely electronic and electromechanical. Structural composite parts, propellers included, are not named.

Exemptions run until 1 January 2028 for equipment on the Defense Contract Management Agency’s Blue UAS Cleared List and for products qualifying as domestic end products under the Buy American standard; the first conditional approvals arrived in March 2026. China’s response came on 5 August 2026, when the Ministry of Commerce subjected US-bound dual-use drone components to case-by-case export review and withdrew the licence facilitation that had let approved exporters ship against a standing permission.

The pattern matters more than any single measure. Regulation is aimed at the electronics, the radios and the motors. The composite part is left to whoever can qualify a mould.

Ukraine’s experience shows how that plays out under pressure. In January 2026, Brave Technics began producing drone propellers domestically by injection moulding — polycarbonate copies of an existing pattern and composite blades in glass-filled polyamide — because availability beat performance. Ukrainian manufacturers still depend heavily on Chinese components, and Ukrainian officials acknowledge they will not mass-produce fully China-free drones because it is much cheaper not to. For the propeller specifically, the substitute for carbon fibre is not a better composite. It is injected plastic.

What Pheonix Research covers here

Pheonix Research does not publish a market called “carbon fibre drone propellers,” and any figure presented as one should be interrogated for what sits inside it — propellers only, or the whole propulsion train; UAV only, or eVTOL blades folded in; units, airframes or value. What the graph does carry is the structural context this component sits inside: the Aerospace & Defense industry, its Defense Systems segment, and the upstream Chemicals & Materials industry where the fibre and prepreg are made. The propeller sits at the intersection of all three, which is exactly why it is easy to miss and hard to substitute.

What to watch

Two things will decide whether carbon fibre propellers stay a specialist line or become a distinct, priced supply chain.

The first is cure-cycle throughput. Pricing power in this component will sit with whoever owns qualified tooling and press time, not whoever owns fibre contracts — and the mould qualification clock, not the materials order, sets how fast a new customer ramps. The second is the eVTOL blade layer. A single aircraft programme building 30 blades per airframe turns a component counted in hundreds into one counted in thousands, and Joby’s 15,000-blade target is a demand layer that did not exist when today’s UAV propeller supply base was tooled. If both hold, the cheapest part on the aircraft ends up being the one that decides who gets to build the aircraft at all.

Frequently Asked Questions

Why are drone propellers made from carbon fibre?

Because stiffness at low mass is the whole specification. A carbon fibre-reinforced polymer blade holds its designed pitch under load instead of twisting and shedding thrust, and it removes weight from the blade tip, where extra mass costs rotational inertia, motor torque and control response. The same properties shape acoustic signature, which is why eVTOL programmes treat blades as a core design component.

Are all drone propellers carbon fibre?

No. Most FPV and consumer drones fly injection-moulded polymer blades, including glass-filled polyamide, because they are far cheaper. Carbon fibre is the premium end of the part, used where endurance, payload capacity or noise performance justify the price — commercially operated UAVs, defence platforms and eVTOL aircraft.

What limits carbon fibre propeller production?

Not fibre supply. The constraint is qualified tooling, press time and inspection capacity. Prepreg compression moulding cut cure time to 7–10 minutes per propeller from 8–10 hours for wet layup or oven cure, but every new blade design still requires a new mould, a new cure recipe and a new qualification file.

How many propellers does one drone need?

A quadcopter needs four, a hexacopter six and an octocopter eight — the industry convention is four or more per aircraft. At the 10 million drones a year Ukraine has said it produces, that implies an annual propeller requirement of about 40 million units (Pheonix Research calculation: 10,000,000 airframes × 4 rotors).