Analysis Step

Says how many steps the analysis takes and how long each one is — the arguments of the OpenSees analyze command.

🔧 Grasshopper component

Analysis StepAlpaca4d ▸ 07_Analysis

Inputs

NumIncrInteger1Number of steps to run.

DtNumber(none)Time step, in s. Transient analysis only.

DtMinNumber(none)Smallest step the solver may fall back to when a step fails.

DtMaxNumber(none)Largest step it may grow to.

JdInteger(none)Iterations aimed for at each step, used to decide whether to grow or shrink Dt.

Outputs

AnalysisStepAnalysisStepFor Analysis Settings.

📈 When to use it

Use it when you are running anything through Run Analysis. Every analysis needs one.

Do not use it when you only want modes → Natural Vibration runs its own eigenvalue problem and takes no steps.

💡 Choosing the values

Static, linear. NumIncr = 1 and nothing else. One step applies the whole load.

Static, non-linear. Raise NumIncr and lower the Integrator's Lambda so their product reaches the load factor you want. Twenty to a hundred steps is normal for a pushover; more steps mean a smoother capacity curve and a better chance of converging.

Transient. NumIncr × Dt is the duration of the analysis. Dt must resolve both the excitation and the response:

• Small enough to sample the highest mode you care about — twenty points per period.
• Small enough to sample the input signal. A ground motion recorded at 0.005 s should not be run at 0.05 s.
• For CentralDifference, small enough to stay stable, which is usually much smaller than either of the above.

Adaptive stepping. Setting DtMin, DtMax and Jd lets the solver shrink the step when it struggles and grow it again afterwards. Worth it for long non-linear runs; leave them empty otherwise.

🔗 Relation to OpenSees

set alpacaAnalyzeOk [analyze $numIncr $dt $dtMin $dtMax $jd]

The result is captured so that Run Analysis can tell a non-convergence from a successful run — analyze returns a negative number on failure but does not raise a Tcl error.