How often do wind turbine blades fail?

The figure most often quoted for the blade failure rate comes from a specialist renewable energy underwriter, which in 2014 announced a study of its own claims and estimated 700,000 blades in operation globally and approximately 3,800 incidents of blade failure a year, each costing in the order of USD 1 million to resolve (Modern Power Systems, September 2014). A peer-reviewed review of blade failure mechanisms was still quoting the same numbers in 2022 (Mishnaevsky, Materials, 2022).

The estimate is a decade old, drawn from one insurer's claims, and the fleet has since grown and moved to longer blades. The same review quotes a survey of service companies in which leading edge erosion and lightning were the most frequently observed damage mechanisms, a measure of damage rather than of blades coming down. The useful number is the rate for one blade type, one plant and one production period, and it is established by inspecting the population.

What causes wind turbine blade failure?

A blade is a hollow composite beam: two sandwich shells bonded to shear webs, with the bending load carried by spar caps of unidirectional glass or carbon laminate, and the whole bolted to the hub through inserts or T-bolts in the root. Failures concentrate at the tip and leading edge, at transitions such as ply drops and the root, and along the adhesive bond lines (Mishnaevsky, 2022). The main blade failure causes follow from that geometry.

Root and insert failure

The root is a bolted joint carrying the whole bending moment. Insert failure, where inserts pull out of the laminate or the laminate around them cracks, can separate a blade in one event; the signature is a fracture close to the hub with the blade largely intact. Contributing factors are lost or never-set bolt pretension, adhesive voids around the inserts, moisture at the root face and fatigue cracking in the root transition.

Spar cap failure and shear web failure

Spar cap failure is usually a compression failure of the suction-side laminate where fibre misalignment, a ply drop or a manufacturing defect has reduced the strength the design assumed. Once the cap fails, the section loses its stiffness and the blade folds; the shells and web then fail as a consequence, not a cause, and shear web debond or buckling ends the same way. Blade collapse under extreme load is usually buckling, a panel or section becoming unstable before the material reaches its strength, and more often a consequence than a cause. The initiating blade fracture is found by reading fracture surfaces, not the extent of the damage.

Trailing edge failure and bond line failure

The trailing edge carries edgewise loads through a long adhesive bond that is thick, variable in thickness and hard to inspect inside the narrow cavity. Trailing edge failure typically starts as a bond line crack or void near the maximum chord, grows under edgewise fatigue and can progress to a transverse crack through the shell, often visible on drone imagery long before it is structural.

Leading edge erosion that becomes structural damage

Wind tunnel tests on an eroded wind turbine aerofoil measured a drag increase of 6 to 500% from light to heavy erosion, and estimated that an 80% drag increase from relatively small erosion could cost around 5% of annual energy production, rising towards 25% for heavy erosion with pits, gouges and delamination (Sareen, Sapre and Selig, Wind Energy, 2014). Once the laminate is exposed, water enters the bond line and the shell, and the damage stops being aerodynamic; leading edge protection only delays that point.

Lightning

Lightning is among the most common causes of blade damage and, when the protection system does not carry the current, a cause of structural failure. A study of 304 cases of direct lightning attachment on blades in the United States found most damage concentrated at the tip (Garolera et al., IEEE Transactions on Power Delivery, 2016). The system as built is checked against IEC 61400-24.

Transport and installation damage

Handling damage from yards, transport and lifting is usually local to the shell and trailing edge, and is sometimes repaired on site without entering the record; what is not found becomes a crack initiation site years later. The May 2024 blade incident at Dogger Bank A was publicly stated to be isolated to a single blade (SSE Renewables, 9 May 2024) and later attributed by the turbine manufacturer to an installation error at sea (offshoreWIND.biz, 25 July 2024); a second blade failure at the same site followed in late August 2024 (Fishing News, 5 September 2024).

Manufacturing defects: wrinkles, dry fibres and delamination

The manufacturing process for a large blade is manual where it matters: dry fabric laid by hand into a mould, resin infused under vacuum, shells and webs bonded with adhesive. Out-of-plane wrinkles form where thick unidirectional stacks pass over a change in geometry or a ply drop, and reduce compressive strength and fatigue life, most of all in a spar cap. Dry fibres and porosity form where the resin front does not reach; delamination and bond line voids where adhesive is missing, too thin or badly cured. The turbine manufacturer publicly attributed the 13 July 2024 blade failure at Vineyard Wind 1 to a manufacturing deviation, insufficient bonding that it said its quality assurance programme should have identified, and announced the reinspection of about 150 blades from the same factory (Vineyard Gazette, 24 July 2024).

What are the warning signs before a blade fails?

Most structural failures are preceded by damage that could have been seen, heard or measured:

  • transverse cracks over the spar cap, trailing edge cracks that grow between inspections, and coating cracks in a line along the spar cap
  • lightning receptor damage, or a strike log with no corresponding inspection
  • edgewise vibration, a changed power curve, unexplained pitch loads or rotor imbalance in the SCADA data, including imbalance from ice accretion
  • water in the root, loss of bolt tension, and repairs in the log without a documented method

Monitoring systems help in proportion to what they measure: drivetrain condition monitoring rarely sees the blade; blade load sensors and root strain gauges can register the natural-frequency shift of stiffness loss, acoustic sensors crack growth, and imbalance detection mass or aerodynamic imbalance; drone or internal crawler inspection can find the defect itself. We do not sell or recommend monitoring products; our interest is in what data exists in the days before a failure.

How is a blade failure investigation run?

A blade failure investigation reconstructs the sequence of events from the evidence and states what that evidence supports, what it suggests and what it cannot determine. That is the purpose of a root cause analysis wind turbine blade owners and insurers can act on. Blade failures usually have one initiating cause and several contributing factors, and the investigation separates them from each other and from the consequential damage.

The first days decide how much can be established later. Evidence to secure:

  • photographs and drone video of the blade and debris field before anything is moved
  • SCADA data around the event: wind, rotor speed, pitch, power, alarms and turbine state
  • lightning detection records and the strike log
  • inspection, repair and maintenance records for the blade and its neighbours
  • serial number, plant, mould, production date and any non-conformance reports
  • the failed structure, or at least the fracture surfaces, kept dry and unhandled

Fractography and laminate cross-sections

Fracture surfaces record direction and mode: fibre breakage indicates tension, kink bands compression, hackle patterns in the resin point back towards the origin, and arrest lines in the adhesive show cyclic growth before final fracture. Cross-sections near the origin show fibre volume fraction, wrinkles, porosity, dry fibres, adhesive thickness and voids, and show against the design basis and production records whether the blade was built as designed. Where the manufacturer runs the laboratory work, the owner's or insurer's expert should be present, because that is where the mechanism is established.

Load data and the design basis

The SCADA record at the timestamp shows whether the turbine was operating, idling or parked, at what wind speed and with what alarms. A blade that fails well inside its design envelope has a strength problem, not a load problem; a failure in a documented extreme event, or in a control state the design did not anticipate, raises questions about site conditions, control and the design load cases. The inspection record matters for the same reason: a defect graded as cosmetic three years running is a finding in its own right.

Isolated event or serial defect?

An isolated event is usually settled as a single claim. A serial defect is an exposure across a fleet. The distinction rests on the signature: the same location, laminate detail, load path and manufacturing step, and usually the same plant and production window. When the initiating defect is one the process produces, the presumption should be that it is a serial defect wind turbine blade populations from the same plant, mould set and period will share, because they share a process history.

Wrinkles are the clearest example. A wrinkle caused by the layup method over a particular ply drop, or by the mould geometry at a transition, will be in every blade built the same way until the method changes. Establishing that requires the failed blade's cross-sections, the production records and an inspection of a sample of the population by a method that can see the defect: for a wrinkle buried in a spar cap, ultrasonic testing rather than a visual survey.

What do insurers, owners and lenders need from the investigation?

Each party needs the same technical answer for a different decision. An insurer needs to know whether the cause is one the policy responds to, whether a defect exclusion applies, and whether there is a recovery against the manufacturer or an installer. An owner needs to know whether the remaining blades can operate and what the warranty position is. A lender needs an independent technical position before restart or the release of funds. All three need conclusions with a stated confidence level and a plain-language summary they can act on.

The manufacturer's root cause analysis is usually the first document on the table, and it is written by the party whose product failed. An independent blade expert instructed early can attend the forensic inspection, witness the laboratory work and argue the technical points while the evidence is still available. In a blade failure insurance claim or a dispute, that is the difference between accepting a conclusion and testing it.

Frequently asked questions

How common is wind turbine blade failure?

The most quoted estimate, published by a renewable energy underwriter in 2014, put it at roughly 3,800 incidents a year across an estimated 700,000 blades. It is a decade old and drawn from one insurer's claims, so it is best treated as an order of magnitude.

What is the difference between blade damage and blade failure?

Damage is a defect that has not yet changed how the blade carries load: erosion, coating cracks, a lightning burn at a receptor, a small delamination. Failure is a loss of structural function: a spar cap fracture, an open trailing edge, a root that has let go, a blade collapse.

What is a serial defect in wind turbine blades?

A defect that arises from the design, the materials or the manufacturing process and is therefore present in more than one blade, usually across a plant and a production period. It is established by matching the failure signature to the process that produced it and inspecting a sample of the population.

Do we need an independent blade expert if the manufacturer is already investigating?

The manufacturer's investigation is necessary and rarely sufficient. Its authors have data nobody else has, and a commercial interest in the outcome. An independent expert who witnesses the laboratory analysis and reviews the reasoning gives the owner, insurer or lender a conclusion they can rely on.

How Apex Wind can help

We are an independent blade engineering consultancy in Denmark. We carry out blade failure investigation and root cause analysis for insurers, loss adjusters, owners and lenders, from the forensic inspection on site to the serial defect assessment across the fleet. We provide expert witness and dispute support, including attendance at the manufacturer's laboratory analysis and review of the other party's root cause analysis. We sell no repairs, hardware or monitoring products and take no referral fees. If a blade has failed, or you have reason to think others may follow, contact us early, while the evidence is still available.