- What it runs on
- Finite element stress and modal
- What it decides
- Peak stress at the notch
- Cost of being wrong
- A life prediction nobody can defend
What it checksCriterion
- 01Mesh convergence at the notchpeak stable under refinement
A peak that has not converged is a mesh result, not a stress result.
- 02Nonlinear step accepted earlyresidual below its own tolerance
An increment that stopped iterating before the equilibrium was found.
- 03Element behaviourno hourglass, no locking
Modes that carry no stiffness, and elements that carry too much.
These are criteria, not results. Nothing here has been measured — the product is not built.
What comes back
- The cause, and the line of solver output it came from
- The change, as a diff you can reverse
- Physics checks, measured against their limits
- A record somebody else can open on a link
Targets
- CalculiX (target)
- code_aster (target)
- Commercial FE adapters (planned)
Named as intent. No adapter for any of these exists today.
Standing
Not built. The loop is the same one Fluids runs — the work is a parser for solver output, a checker set for these quantities, and a fix whitelist that is safe for a structural model. None of it exists yet.