This simulator models a three-blade collective-pitch mechanism and its governing controller — the system that regulates rotor speed and limits power by rotating the blades toward feather as wind, actuator limits and generator loading change. Drive a gust, jam the actuator, or trip the generator offline and watch the controller, actuator and rotor dynamics respond in real time.
• A real-time 3D cutaway of the pitch system — pitching blades, pitch actuator, blade-root pitch bearing, pitch motor and reduction gear, blade-angle feedback encoder, and the pitch controller — with a toggleable enclosure, auto-rotate, expand and selectable components with callouts, plus a home view and a hide-labels toggle. • Governor & actuator controls: wind speed (3–30 m/s), an automatic pitch controller on/off toggle, manual pitch demand (0–85°) for use when automatic control is disabled, maximum pitch rate (0.5–12°/s), a jam-pitch-actuator fault checkbox, a generator-electrically-loaded toggle, and air density (0.9–1.4 kg/m³). • Six live metrics: rotor speed, actual blade pitch, demanded pitch, electrical output, generator shaft torque and power coefficient. • Time controls: pause/resume, advance 0.1 s, advance 1 s, and playback speed from 10x slow motion to real time to 10x and 60x faster, plus a reset-protection control for clearing a latched trip. • A Control response tab with two live charts (demanded vs. actual pitch angle and torque/power), the underlying model equations (the rotor inertia/torque balance, aerodynamic torque, the pitch-demand control law and its integral term, and the actuator rate limit), a stated model scope, and snapshot measurements. • An Experiments tab with five guided experiments (above-rated gust, slow actuator, actuator jam, load rejection, manual feather), a Model verification bench that runs the checks.js suite of deterministic checks against a fresh model instance, and a timestamped event log with a trial-report export. • A Learn & assess tab with guided lessons, a three-question knowledge-check quiz and a written scope-and-references statement linking to NREL technical background.
The rotor's angular acceleration follows a torque balance: aerodynamic torque accelerates the shaft, while generator torque, friction and (when tripped) emergency braking oppose it. The pitch controller compares rotor speed against a 2.5 rad/s target and computes a demanded pitch angle from a proportional term plus an integral term that accumulates speed error over time, clamped between 0° and 85°.
Crucially, the demanded angle and the actual blade angle are separate quantities in this model: a finite actuator slew rate (the configured maximum pitch rate) limits how fast the actual angle can follow the demand, so during a fast gust the two can diverge visibly. A jammed actuator fault goes further — it freezes the actual angle completely regardless of what the controller continues to demand, which is a qualitatively different failure from an ordinary rate-limited lag. High-wind cut-out and overspeed conditions latch a protective trip, request full feather and apply the modeled emergency brake; resetting protection clears the trip but does not repair a stuck actuator.
The Control response tab's demanded-vs-actual pitch chart is the key diagnostic: when the two traces track closely, the actuator is keeping up with the controller; when they separate, either the actuator rate is too slow for the disturbance or the actuator has jammed. The torque/power chart shows how aerodynamic and generator torque interact through the inertia balance during a gust or a load-rejection event.
The five built-in experiments isolate the model's key behaviors — an above-rated gust with automatic pitch engaged, the same gust with a slowed actuator, an actuator jam under high wind, sudden generator load rejection, and manual feathering with automatic control disabled — each with a one-line explanation of the expected result. This is a teaching model with one rigid rotor, collective pitch, an analytic Cp surface and an illustrative PI controller, not a tuned commercial turbine controller: tower flexibility, individual blade loads, drivetrain torsion and converter transients are omitted, and emergency braking is modeled as an ideal constant opposing torque. The Model verification bench runs the checks.js suite (actuator rate never exceeds its configured slew, a jammed actuator holds its actual angle, high-wind cut-out latches protection, removing generator load accelerates the rotor, and automatic pitch increases under above-rated wind) against a freshly constructed model, leaving your live experiment untouched.
Demanded pitch is the angle the controller calculates as needed to regulate rotor speed; actual pitch is the blade-root actuator's real position, which can only change at the configured maximum pitch rate. During a fast gust the two can separate noticeably — plotting both on the Control response chart is the way to distinguish a controller decision from a mechanical lag.
A slow actuator (a low maximum pitch rate) still eventually reaches the demanded angle, just with a lag. The jam-pitch-actuator fault instead freezes the actual blade angle completely, regardless of what the controller continues to demand — resetting protection clears a latched trip but does not repair a jam; you have to clear the fault itself before another trial.
With the generator-electrically-loaded toggle switched off, the opposing generator torque in the rotor's torque balance disappears, so aerodynamic torque is no longer balanced and the rotor accelerates. The pitch controller and protection logic respond to the resulting overspeed as they would to a real load-rejection event.
This is a teaching model with one rigid rotor, collective pitch, an analytic Cp surface and an illustrative PI controller — not a tuned commercial turbine controller. Tower flexibility, individual blade loads, drivetrain torsion and converter transients are omitted, and emergency braking is an ideal constant opposing torque rather than a modeled hydraulic or mechanical brake.