Fatigue Analysis Software

ASME BPVC Section VIII Div 2 (2023) mandates a strain-based or stress-based fatigue analysis using the design-by-analysis approach, whereas Div 1 relies on conservative, rule-based allowable cycles from mandatory Annex 5–1 and excludes explicit weld-specific fatigue curves. Div 2 requires evaluation of local stresses—including structural stress at weld toes—using the hot-spot or structural stress method (per Annex 5.4), and explicitly incorporates weld geometry factors (e.g., Kₐ) to adjust fatigue strength. Div 2 also permits use of material-specific S–N data and mandates consideration of mean stress effects via Goodman or Gerber corrections—unlike Div 1’s simplified constant-amplitude, zero-mean assumption. Compliance must be demonstrated via fatigue usage factor ≤ 1.0 per UG-28 and Part 5.

The weld geometry factor (Kₐ) in ASME VIII Div 2 Annex 5.4 accounts for stress concentration amplification at weld details—distinct from nominal stress. It is applied multiplicatively to the fatigue strength coefficient (σ′_f) in the modified Basquin equation: σₐ = (σ′_f × Kₐ) × (2N_f)^b, where b is the fatigue exponent. Kₐ values (typically 0.4–0.9) are selected from Table 5.4.1 based on joint type, weld preparation, and post-weld treatment (e.g., 0.8 for as-welded CJP groove with good profile). Importantly, Kₐ is *not* applied to the applied stress range; it modifies only the material’s inherent fatigue resistance. Using Kₐ incorrectly—e.g., dividing stress instead of scaling strength—violates UG-28(c)(2) and invalidates compliance.

No—500 MPa is a representative default for normalized carbon steel (e.g., SA-516 Gr. 70) but is not universally applicable. ASME VIII Div 2 Annex 5.2.2 requires σ′_f to be derived from material-specific, condition-appropriate S–N test data per ASTM E466 or E606. For example, quenched & tempered steels like SA-543 may have σ′_f ≈ 650–750 MPa, while stainless steels (SA-240 304L) often fall near 420 MPa due to lower yield strength and notch sensitivity. Using an unverified default risks non-conservative life estimates—especially for high-cycle fatigue (>10⁶ cycles)—and violates UG-28(c)(1), which mandates traceable material property inputs. Always validate σ′_f against certified mill test reports or qualified welding procedure specifications (WPS) per QW-180.

The safety margin (%) is computed as [(N_allowable / N_actual) − 1] × 100, where N_allowable is the cycles permitted by ASME VIII Div 2 fatigue rules (i.e., the N_f solving σ_range = (σ′_f × Kₐ) × (2N_f)^b), and N_actual is the specified design life (e.g., 10⁵ cycles for a reactor vessel). This differs from stress-ratio margins—it reflects life reserve, not strength reserve. Per UG-28(d), compliance requires the fatigue usage factor U = n/N ≤ 1.0; thus, safety margin ≥ 0% implies U ≤ 1.0. Note: The software assumes constant-amplitude loading; for variable amplitude, rainflow cycle counting and linear damage accumulation (Miner’s rule) per Annex 5.5 must be performed separately.

No—the current implementation applies the base air-fatigue S–N relationship per Annex 5.2 and does not automatically derate for environment or temperature. ASME VIII Div 2 Annex 5.6 explicitly requires reduction of fatigue strength for corrosive service (e.g., seawater, H₂S) and temperatures above 370°C (700°F), typically via fatigue strength reduction factors (FSRFs) or adjusted Kₐ values. For instance, FSRF = 0.5–0.7 is common for sour service per NACE MR0175/ISO 15156. Users must manually scale σ′_f or apply a custom Kₐ before inputting values. Relying solely on default parameters without environmental derating violates UG-28(c)(3) and may lead to noncompliant designs—especially critical for offshore or chemical process equipment.

Accuracy depends critically on how well the input stress range reflects the true structural (hot-spot) stress at the weld toe—a challenge for partial-penetration T-joints, where stress gradients are steep and geometry-dependent. The software assumes a uniform, elastic stress range; however, ASME VIII Div 2 Annex 5.4.2 requires structural stress determination via 3D FEA with mesh-independent extrapolation or validated strain-gauge correlations. Using nominal or membrane-plus-bending stress here introduces errors up to 3× in predicted life. For such joints, always verify the input stress_range using FEA per IIW Recommendations and document the stress extraction method. Without this, even correct application of Kₐ and σ′_f yields non-compliant results per UG-28(a)(2).

Yes—ASME VIII Div 2 Table 5.4.1 explicitly assigns distinct Kₐ values for improved weld profiles: e.g., Kₐ = 0.9 for HFMI-treated welds vs. 0.7–0.8 for ground-as-welded, and 0.4–0.6 for as-welded without treatment. These values reflect measured improvements in fatigue strength due to compressive residual stresses and reduced stress concentration. However, the software does not auto-apply these—users must select the appropriate Kₐ based on documented post-weld treatment per QW-280 and supporting validation (e.g., IIW Doc. IIW-1823-15). Using Kₐ = 0.8 for an HFMI-treated joint would underestimate life and violate UG-28(c)(2), while overestimating Kₐ risks noncompliance. Always tie Kₐ selection to qualified procedures and inspection records.