v0.1 (beta)

Verification Documents

Linearization & Campbell (IEA 22 MW)

UI walkthrough — Campbell diagram + mode shapes from a linearization run.

This manual reproduces the linearization / Campbell-diagram result for the IEA-22-280-RWT-Onshore reference turbine using nothing but the TurbineX web app — the same click-through format as the steady-state manual.

Download the full manual — IEA 22 MW linearization (PDF)

Step 1 — Configure the turbine for linearization

Linearizing a 280 m rotor is where OpenFAST cold-start convergence bites. On the production bundle (gravity on, -6° shaft tilt, flexible tower) the model cold-starts into a tower strike around t = 37 s before it reaches the linearization time. The linearization-ready configuration removes the three production-physics terms that fight the eigenanalysis — the same changes the retired IEA-22-280-RWT-VT3 bundle carried:

SettingProductionLinearization-readySet in the UI via
Gravity9.810Advanced Settings → Disable gravity
Shaft tilt-6°Advanced Settings → shaft tilt = 0
Tower DOFsflexiblerigidAdvanced Settings → Rigid support structure

There is no separate VT3 turbine in the Library any more — you recreate it inside the study, in the Advanced Settings → Solver / module overrides panel of the Run Simulation tab, together with the Linearize toggle. Turning Linearize on fills in the rest of the recipe automatically (a single linearization near the end of the run, the generator DOF locked, controller modes zeroed).

Step 2 — Run the case in the app

Linearization is a Single Simulation analysis — one operating point per run. Create a Single Simulation study, pick IEA-22-280-RWT-Onshore on the Turbine tab, set Steady wind on the Run Simulation tab, apply the Advanced Settings above, and submit. Repeat for each wind speed in the operating grid (this manual uses 4, 6, 8, 10, 11.5 m/s); a Campbell diagram across wind speeds is assembled by stepping through those single-sim jobs.

When a job completes, open the Results tab. Because the job was run with Linearize on, two extra analysis tabs appear:

  • Stability — the Campbell diagram: mode frequency vs wind speed, coloured by damping, with the 1P/3P/6P/9P per-rev excitation lines overlaid.
  • Mode Shapes — a 3D viewer that animates each eigenmode of the rotor, with a mode picker (frequency + damping per mode) and camera presets.

Expected result

  • The below-rated jobs complete and expose the Stability and Mode Shapes tabs — no cold-start divergence, no tower strike.
  • The Campbell diagram shows blade families flat with rotor speed. Tower modes are absent by construction — the support is rigid.
  • Every resolved mode is positively damped — no point crosses into the negative (unstable) band — and no mode sits on a per-rev excitation line inside the operating envelope.
  • The Mode Shapes viewer animates recognisable blade bending shapes.

On the reference run the four below-rated points (4–10 m/s) converge cleanly; the rated 11.5 m/s point does not — with gravity off, the ElastoDyn modal blade exceeds OpenFAST’s small-angle assumption at rated thrust and diverges near t ≈ 1 s (the Job Monitor shows the batch as Mixed (4/5)). Reaching rated would need the geometrically-exact BeamDyn blade. The platform surfaces this limit rather than hiding it.

A note on the mode labels

This manual verifies the linearization workflow — that the run produces a stable Campbell diagram and animatable mode shapes. Treat the mode picker’s family labels as indicative only: the auto-classifier can mislabel (a ~3.2 Hz mode is tagged “1st flap” when the first flapwise mode is ~0.4 Hz). The robust, trustworthy result is qualitative: every resolved mode is positively damped across the below-rated range.