Why Do Wind Turbines Stop in Strong Wind?
Updated 8 October 20267 min readWind & Hydropower
Because the energy is not the problem — the loads are. Power in the wind rises with the cube of speed, so a gale offers far more than a turbine's drivetrain can convert or its structure can carry. Above rated speed the machine deliberately spills the surplus, and above cut-out speed it stops taking any at all and protects itself.
Key takeaways
- Available power scales with v³ while structural loads scale roughly with v², so both outrun the machine long before the strongest winds arrive.
- The flat top of a power curve is not saturation. It is the controller actively throwing energy away to hold output at the generator's rating.
- Pitching the blades reduces their angle of attack, which is how a modern turbine sheds power smoothly and continuously.
- Cut-out protects the structure from loads and fatigue, not the generator from overload — that job was already done at rated speed.
- The energy given up above cut-out is small, because the strongest winds occur for very few hours a year.
On this page
A power curve has four regions
Everything about this question is visible in the shape of a turbine's power curve, which has four distinct regions and only one of them follows the physics of the wind.
Below cut-in, torque is too small to overcome friction and the machine produces nothing. Between cut-in and rated, output follows the cube law and the controller works to hold the rotor near its best tip-speed ratio. At rated speed, output reaches the generator's rating and stays there. Above cut-out, the machine shuts down.
Typical numbers fix the shape in mind. Cut-in is typically around 3–4 m/s (7–9 mph), rated speed typically 12–15 m/s (27–34 mph), and cut-out typically about 25 m/s (~55 mph) — figures typical of published utility-scale turbine power curves rather than any standard. Large offshore machines, low-wind-site designs and older stall-controlled turbines all sit elsewhere, so read them as a shape, not a specification.
The dashed line in that figure is the point of this article. The energy above rated speed is genuinely there — and the machine is choosing not to take it.
| Region | What the rotor does | What limits it |
|---|---|---|
| Below cut-in (typically below ~3–4 m/s) | Idles, produces no useful torque | Friction and the minimum torque to turn the drivetrain |
| Cut-in to rated (~3–4 up to ~12–15 m/s) | Tracks the optimum tip-speed ratio, output follows v³ | Aerodynamics — how much the rotor can extract |
| At rated (from ~12–15 m/s to cut-out) | Holds output flat, sheds the surplus by pitching | Generator, converter and drivetrain ratings |
| Above cut-out (typically above ~25 m/s) | Feathers and stops generating | Structural loads and fatigue, not electrical rating |
Speeds are typical of published utility-scale turbine power curves and vary by model — they are design choices published per machine. The behaviour of each region is common to modern variable-speed machines.
Why output flattens at rated speed
A generator has a rating: a current it can carry, a converter it can feed and a drivetrain that can transmit torque, all sized for a design point. Above that, more aerodynamic power is available than the machine can convert.
Sizing the electrical system for storm winds would mean paying in mass and cost for equipment that spends nearly all its life far below its capability. So the rating is chosen for the winds that actually produce energy, and the surplus above it is spilled, as a solar inverter clips surplus energy.
That is a genuine choice, not a limitation of the rotor. The rotor could take more — the aerodynamic ceiling is the Betz limit, which is nowhere near reached at these speeds — but the machinery behind it cannot use what it would collect.
How a blade throws power away
Spilling power sounds crude. In a modern machine it is precise, continuous and quiet, and it happens by changing one angle.
A blade is an aerofoil, and its lift depends on the angle of attack between the oncoming flow and the blade's chord line. Rotate the blade about its own long axis — pitching it — and that angle changes, so the lift changes with it.
Pitch control rotates the blades toward the wind — feathering them — reducing the angle of attack so the flow stays attached and lift falls smoothly. The controller trims this continuously, gust by gust, to hold output flat.
Stall control, the older approach, fixes the blades and relies on the angle of attack growing with wind speed until the flow separates from the upper surface and lift collapses on its own. It needs no moving parts in the hub, but the transition is abrupt and cannot be tuned.
Both approaches leave the rotor spinning while refusing the extra energy. What changes is how gracefully. Water turbines face the same problem and solve it outside the runner: Pelton and Turgo turbines swing a deflector into the jet, so surplus water bypasses the buckets instead of driving the runner faster.
Cut-out is about loads, not power
Here is the distinction that makes sense of the whole question. Once output is held flat at rated power, the electrical system is safe at any higher wind speed. The machine still stops. Why?
Because thrust — the force pushing the whole rotor downwind — keeps growing. It scales roughly with the square of wind speed, and it does not care that the generator is comfortable. That force bends the blades toward the tower, applies a moment at the blade roots, and pushes the tower over about its base.
Two further effects compound it. Turbulence intensity rises in strong winds, so the loads are not merely large but rapidly varying — and rapidly varying loads are what cause fatigue, the accumulation of damage that decides a structure's life. And gusts arrive faster than a pitch system can fully answer, so the design must tolerate excursions above whatever the average wind is doing. Machines in the sea meet the same arithmetic in a harsher form, which is why survival loads rather than average output dominate the design of wave and tidal energy devices.
Shutting down removes nearly all of it. A feathered rotor presents very little area to the flow and generates almost no torque, so both the steady thrust and the fluctuating component collapse.
What "stopped" actually means
A turbine that has shut down is rarely locked solid. Several states exist, and they escalate.
- Feathered and idling is the usual storm state: blades turned edge-on, minimal torque, rotor free to drift slowly. Letting it turn keeps bearings lubricated and avoids loading one blade position continuously.
- Parked applies the brake and holds the rotor, with yaw still tracking the wind so the machine faces it correctly.
- Locked pins the rotor mechanically and is a maintenance state, not a weather one.
Restarting is deliberately hysteretic: the machine waits for the wind to fall meaningfully below cut-out and to stay there, rather than switching back on the moment a gust subsides. Cycling in and out of service in a storm would impose exactly the transient loads the shutdown was meant to avoid.
The energy you are giving up
It is natural to assume that refusing a gale costs a fortune in lost generation. The wind speed distribution says otherwise.
Energy is the product of how much power a wind speed delivers and how many hours a year it blows. Strong winds score highly on the first and very poorly on the second, so the area under the product curve is concentrated in the middle of the range. Extending the design envelope to capture the tail means carrying heavier structure through every hour of the machine's life — the same trade that settles blade count in why wind turbines have three blades, and one that offshore machines weigh differently because their transmission and access costs are so different.
So a still rotor in a storm is not a fault, and it is not waste. It is a machine declining energy it was never built to carry, on the grounds that it would rather be there next year.
Frequently asked questions
Doesn't stopping in high wind waste a lot of energy?
Surprisingly little. Very strong winds occur for a small number of hours each year, so although each of those hours is rich in energy, their combined contribution is modest. Building a machine to harvest them would cost mass and load capability across its whole life for a small annual gain.
What is the turbine doing when it has stopped?
Usually idling rather than standing still. The blades are pitched edge-on to the wind — feathered — so they generate almost no torque, and the rotor turns slowly or not at all with the brake released. Yaw keeps the machine oriented so it presents the least load to the wind.
What is the difference between cut-in, rated and cut-out speed?
Cut-in is the lowest wind speed that produces useful torque, typically around 3-4 m/s (7-9 mph). Rated speed is where output reaches the generator's rating and is then held flat, typically 12-15 m/s (27-34 mph). Cut-out is where the machine shuts down to protect itself, typically about 25 m/s (~55 mph). All three are design choices published for each model and vary with it.
Could a turbine be destroyed by a storm if it did not stop?
The risk is not a single dramatic failure so much as loads and fatigue beyond what the structure was designed to accumulate. Thrust on the rotor, bending at the blade roots and moments at the tower base all grow steeply with wind speed, and they act on a structure intended to last decades.
Do all turbines stop at the same wind speed?
No. Cut-out speed is a design decision that depends on the class of site the machine is built for, and some newer machines reduce output gradually through very high winds instead of shutting down abruptly, which keeps a little generation available and softens the transition.
Sources
Named organisations whose published material underpins this article. Where no link is given, the source is named rather than linked.
- National Renewable Energy Laboratory (NREL)Wind turbine control strategies, loads and power performance research.
- International Energy Agency (IEA)Technology overviews for wind power.
- Turbine manufacturer power curves and data sheetsCut-in, rated and cut-out speeds and the control strategy are published per machine and vary with the design class.
Editorial Team
Research, drafting and review
Articles are drafted from primary engineering and physics references with AI-assisted tools, then reviewed and fact-checked line by line by a human editor before publication. We publish explanations, not recommendations: no products, no pricing, no country-specific rules, and no invented author personas.
Last reviewed 8 October 2026. How we research and review