FFS Level 3 Crack Analysis Shell Seam CS-102
Contents
- 1. Executive Summary
- 2. Project Overview
- 3. FEA Methodology
- 4. Boundary Conditions & Loading
- 5. Load Case Analysis
- 6. Detailed Results
- 7. FAD Assessment - Level 3 Calculations
- 8. Model Validation
- 9. Lessons Learned
- 10. What Could Have Gone Wrong
- 11. Recommendations
- 12. Limitations & Assumptions
- 13. Conclusion
Figure 1: Shell with Weld Seam CS-102 and Crack Locations
1. Executive Summary
For Decision Makers
Two cracks were detected at shell weld seam CS-102 during phased array ultrasonic testing (PAUT) of equipment Tag-16-CC-00-401. The vessel operates at 15.926 MPa internal pressure and 260 deg C – severe service conditions that demanded the most rigorous fitness-for-service assessment available. The critical question for the asset owner: can this equipment continue to operate safely with these cracks, or is an emergency shutdown required?
We performed API 579-1/ASME FFS-1 2021 Level 3 assessment – the highest tier – using elastic-plastic FEA with J-integral computation and Failure Assessment Diagram (FAD) evaluation.
- FFS Level 3 per API 579-1/ASME FFS-1 2021 – the most rigorous assessment tier
- Elastic-plastic FEA with multilinear SA-516 Gr 70 material model
- Both cracks fall in the acceptable zone on the FAD
- Crack 1: Kr = 0.00119, Lr = 1.056 – ACCEPTABLE
- Crack 2: Kr = 0.001106, Lr = 1.012 – ACCEPTABLE
- Cracks are non-growing under current operating conditions
Business Impact: This assessment saved the client from an unnecessary emergency shutdown, avoiding millions in lost production while providing documented assurance of safe continued operation. A monitoring and re-inspection plan was recommended to track crack stability over time.
2. Project Overview
Design Parameters
3. FEA Methodology
The Level 3 assessment follows the API 579-1 procedure for elastic-plastic fracture mechanics analysis. This is the most rigorous tier of FFS assessment, requiring explicit crack modelling, J-integral computation, and FAD evaluation. The methodology involves 12 sequential steps:
- Step 1: Classify all loads as primary, secondary, or residual. For this equipment, all loads are primary (pressure-driven)
- Step 2: Construct elastic-plastic FE model with multilinear stress-strain curve for SA-516 Gr 70 at 260 deg C
- Step 3: Model cracks explicitly using dimensions from PAUT inspection data
- Step 4: Generate spider mesh at crack tips for path-independent J-integral evaluation
- Step 5: Apply loads incrementally (10 steps from 0.1 to 1.0) and compute J-integral at each step
- Step 6: Convert J to equivalent stress intensity factor K_J
- Step 7: Determine elastic J from initial slope of J vs applied stress plot
- Step 8: Compute Kr (vertical FAD coordinate) at each step
- Step 9: Compute Lr (horizontal FAD coordinate) using reference stress solution
- Step 10: Plot Kr vs Lr on FAD and evaluate against the envelope curve
Mesh Quality
Figure 2: Crack modelling from PAUT data – Crack 1 at CS-102
Figure 3: Spider mesh at crack tip for J-integral computation
Figure 4: FE Mesh – SOLID186 with crack tip refinement
4. Boundary Conditions & Loading
The FE model includes the shell section containing weld seam CS-102 with both cracks explicitly modelled. Boundary and loading conditions are applied as follows:
- Internal pressure: 15.926 MPa applied on all internal wetted surfaces
- Shell cutout thrust: Pressure-induced thrust forces applied at shell cutout boundaries, acting outward to balance internal pressure end loads
- Operating weight: Full operating weight of the equipment applied at shell cutout in the downward direction
- Gravity: Standard Earth gravity applied to entire model
- Fixed boundary: All DOFs constrained at shell cutout to simulate the equipment’s stability and support conditions
The loading is applied incrementally in 10 equal steps (time 0.1 to 1.0) to enable step-by-step J-integral computation and FAD curve construction.
Figure 5: Boundary conditions – pressure, thrust, fixed support
5. Load Case Analysis
Load Case 1: Design Operating Condition
Full design pressure with operating weight and gravity. This is the primary load case for the crack assessment:
- Internal pressure: 15.926 MPa applied incrementally (10 steps)
- Shell cutout thrust forces (pressure balance)
- Operating weight in downward direction
- Standard Earth gravity
The incremental loading allows J-integral computation at each step, enabling construction of the complete K_J vs pressure relationship needed for FAD evaluation. At each step, both the J-integral and stress intensity factor (SIF) are computed at both crack tips.
6. Detailed Results
The elastic-plastic analysis ran 10 incremental load steps. At each step, equivalent stress and stress intensity factor (SIF) were computed at both crack tips.
Equivalent Stress at Each Step
Stress fields computed at 10 load steps (0.1 to 1.0 of design pressure). The stress distribution shows progressive plastic zone development around the crack tips as load increases.
Figure 6: Equivalent stress at shell – crack region
Figure 7: Von Mises stress at full design load
Stress Intensity Factors
SIF computed along the crack front at each load step for both cracks. The K_J values are derived from J-integral using the relationship K_J = sqrt(J * E / (1 – v^2)). The K_J vs pressure plot provides the elastic slope needed for FAD coordinate calculation.
7. FAD Assessment - Level 3 Calculations
The Failure Assessment Diagram (FAD) is the definitive tool for Level 3 crack assessment. It plots two coordinates:
- Kr (vertical axis): ratio of applied stress intensity to material fracture toughness – measures proximity to brittle fracture
- Lr (horizontal axis): ratio of reference stress to yield strength – measures proximity to plastic collapse
If the operating point (Kr, Lr) falls inside the FAD envelope curve, the crack is acceptable. If it falls outside, the crack is unacceptable and remedial action is required.
FAD Results
Both operating points fall well within the acceptable region of the FAD. The extremely low Kr values (approximately 0.001) indicate that the fracture driving force is negligible relative to material toughness, the cracks are dominated by ductile behavior rather than brittle fracture risk.
Figure 8: FAD – Crack 1 operating point in acceptable zone
Figure 9: FAD – Crack 2 operating point in acceptable zone
8. Model Validation
Hoop Stress Validation
FEA hoop stress compared against ASME VIII-2 Clause 4.3.10.2 hand calculation:
Force Convergence
The elastic-plastic solution converged under applied loads per ASME VIII-2 Clause 5.2.4.4 Step 5. Force convergence was confirmed at all 10 load steps, verifying that the equipment is structurally stable under the full design loading. Non-convergence would indicate that the structure cannot sustain the applied loads, a fundamentally different and more serious failure mode.
Figure 8: FAD – Crack 1 operating point in acceptable zone
Figure 9: FAD – Crack 2 operating point in acceptable zone
9. Lessons Learned
Spider Mesh Is Non-Negotiable for J-Integral
J-integral computation requires concentric ring elements (spider mesh) around the crack tip. Standard tetrahedral or hexahedral meshing produces wildly inaccurate J values because the path-independence of the J-integral relies on properly formed integration contours. The spider pattern ensures these contours are correctly defined, it is the foundation of the entire Level 3 assessment.
Multilinear Material Model Captures Reality
Linear elastic analysis cannot capture the plastic deformation at crack tips that occurs under high-pressure service at elevated temperatures. The multilinear stress-strain curve for SA-516 Gr 70 at 260 deg C enables accurate J-integral computation under elastic-plastic conditions. Level 3 explicitly requires this material model fidelity, linear elastic J values would be incorrect.
Low Kr Is Reassuring - It Means Ductile Behavior
Both cracks show Kr approximately 0.001, meaning the fracture driving force is negligible relative to material toughness. The cracks are dominated by the Lr (plastic collapse) axis rather than the Kr (fracture) axis. This indicates ductile behavior, which is the preferred failure mode for pressure equipment because it provides warning before failure and is more predictable than brittle fracture.
10. What Could Have Gone Wrong
Using Level 1/2 Instead of Level 3
Lower assessment levels use simplified screening criteria that can be overly conservative for complex crack geometries. A Level 1 assessment might have triggered an unnecessary repair or shutdown when these cracks are actually stable and acceptable. Level 3’s rigorous fracture mechanics approach provides the definitive answer, potentially saving millions in avoided downtime.
Elastic-Only FEA
Linear elastic FEA would overestimate crack tip stresses because it cannot capture the plastic redistribution that occurs at crack tips. The J-integral from elastic analysis alone would give incorrect FAD coordinates, potentially classifying acceptable cracks as unacceptable and triggering unnecessary remedial action.
Incorrect Crack Dimensions from NDT
The entire Level 3 assessment depends on accurate crack dimensions from NDT data. Underestimating crack depth or length would produce unconservative results, the FAD point would be closer to the origin than reality. PAUT (Phased Array Ultrasonic Testing) data was used as input, providing the most reliable crack sizing accuracy available in non-destructive inspection.
11. Recommendations
For Repair Planning
- No immediate repair required – both cracks are acceptable per API 579-1 Level 3
- If future inspections show crack growth, re-assess with updated dimensions before planning repair
- If repair is eventually needed, grind-and-weld repair is recommended with full PWHT and re-inspection
For Monitoring
- Implement a crack monitoring program with PAUT inspections at regular intervals
- Recommended initial re-inspection interval: 2 years (or next scheduled turnaround, whichever comes first)
- Compare crack dimensions at each inspection against the FFS assessment assumptions – any growth triggers re-assessment
For Re-Assessment Timeline
- Re-assess if crack dimensions change by more than 10% from current PAUT measurements
- Re-assess if operating conditions change (higher pressure, higher temperature, or different cycling)
- Document all inspection results for future FFS re-assessments – trending data improves confidence in continued operation
12. Limitations & Assumptions
- Crack dimensions are based on PAUT inspection data. If actual crack sizes differ from PAUT measurements, the FAD assessment results may change. Conservative sizing is recommended for safety.
- The multilinear material model uses ASME Section II Part D properties for SA-516 Gr 70 at 260 deg C. Actual material properties (from MTR) may differ – using code minimum values provides conservative results.
- All loads are classified as primary (pressure-driven). Secondary and residual stresses (e.g., weld residual stress) are not explicitly modelled. API 579-1 provides guidance for including these if needed.
- The analysis assumes static loading. If the equipment is subject to cyclic pressure or thermal loading, a fatigue crack growth assessment may be required in addition to the static FAD evaluation.
- Crack interaction effects between the two cracks are not explicitly modelled. If the cracks are close enough to interact, a combined assessment would be needed.
13. Conclusion
- Both cracks fall within the acceptable region of the Failure Assessment Diagram
- Cracks are non-growing under current operating conditions (15.926 MPa, 260 deg C)
- Kr is approximately 0.001 – negligible fracture driving force, ductile behavior confirmed
- FEA validated against hand calculation within 0.03%
- Elastic-plastic solution converged – equipment structurally stable under full design loads
- Regular PAUT monitoring recommended with 2-year re-inspection interval
Status: Equipment Tag-16-CC-00-401 is cleared for continued service with the existing cracks at shell seam CS-102. The cracks are stable, non-growing, and well within the acceptable zone of the API 579-1 Level 3 FAD assessment. A monitoring and re-inspection program is recommended to ensure ongoing safe operation.
Download the Full Technical Case Study
The full technical report includes:
- API 579 Level 3 crack assessment with J-integral and FAD evaluation
- Detailed stress intensity (KJ) and crack behavior analysis
- Elastic-plastic FEA methodology with multilinear material modeling
- Load case definition and incremental loading approach
- Stress contour plots and crack tip stress distribution
- Validation including hoop stress comparison and convergence checks
- Code references including API 579-1 / ASME FFS-1 and ASME Section VIII