Stress [MPa]
- Full series
- Monotonic convergence in the studied range
- Estimation window
- Monotonic; Richardson estimate available
- Estimation-window levels
- medium → fine → very-fine
- Observed order
- p = 2
- Estimated asymptotic value
- 10
- Finest mesh
- 10.0025
Multi-series analysis of result changes, trends, observed order and asymptote over successive FEA meshes.
Report an error, translation issue, or idea. The form will not change this page's inputs or result.
Compares observed quantities on successive meshes, distinguishes the full-series trend from the three-level estimation window, and reports refinement metadata.
Schema: mesh_convergence_analyzer v1
CSV uses commas, a dot decimal separator, and the required id, h and elements columns. Other columns become observed quantities.
For coarse, medium and fine values φ₁, φ₂ and φ₃, the observed order is p = ln| (φ₁ − φ₂) / (φ₂ − φ₃) | / ln(r).
The estimate uses φ∞ = φ₃ + (φ₃ − φ₂) / (rᵖ − 1). It is an asymptotic estimate, not an exact reference value.
The two refinement ratios must differ by no more than 10%. Within that tolerance, r is their arithmetic mean; outside it, no order or asymptote is reported.
The observed order and asymptotic estimate use the three finest, approximately geometrically refined levels. This result alone does not demonstrate mesh independence.
| ID | Characteristic mesh size h | Elements | Nodes | Degrees of freedom | Solve time [s] | Refinement ratio rₕ | Element-count multiplier | Node-count multiplier | Degree-of-freedom multiplier | Solve-time multiplier | Stress [MPa] | Displacement [mm] |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| coarse | 0.4 | 100 | 80 | 240 | 1.2 | — | — | — | — | — | 10.16 | 2.32 |
| medium | 0.2 | 400 | 300 | 900 | 3.8 | ×2 | ×4 | ×3.75 | ×3.75 | ×3.16667 | 10.04 | 2.08 |
| fine | 0.1 | 1,600 | 1,200 | 3,600 | 14.1 | ×2 | ×4 | ×4 | ×4 | ×3.71053 | 10.01 | 2.02 |
| very-fine | 0.05 | 6,400 | 4,700 | 14,100 | 52.8 | ×2 | ×4 | ×3.91667 | ×3.91667 | ×3.74468 | 10.0025 | 2.005 |