Multiple Defects, Interaction & Coalescence
How neighbouring cracks, pits, metal-loss regions and mixed defect types are screened for interaction and conservatively combined when one-defect-at-a-time assessment is no longer valid.
Why Defect Interaction Matters
Two individually acceptable defects can be unacceptable when their stress fields or reduced ligaments interact. This is particularly important for aligned cracks, clustered pits, multiple wall-loss patches, flaws on opposite surfaces and crack-like defects emerging from dents or corrosion. The first question is therefore not whether each indication passes independently, but whether the indications should be treated as one effective defect.
Interaction Is Geometry and Mechanism Dependent
Spacing alone does not determine interaction. Crack length, depth, relative alignment, through-thickness position, stress direction and local geometry all matter. Two cracks separated along the crack plane may interact strongly, while defects separated in another direction may have little effect. Validated interaction rules from the governing assessment method should be used where available.
Conservative Recharacterisation
A common approach is to combine interacting defects into a larger bounding flaw or damaged region. This is deliberately conservative but transparent. The recharacterised defect should preserve the dimensions relevant to the failure mode—such as total projected crack length, deepest depth or reduced ligament—without creating an impossible geometry that changes the mechanism entirely.
Crack-to-Crack Interaction
Neighbouring cracks can amplify stress intensity and may coalesce as they grow. Surface cracks can also interact through the thickness with embedded or opposite-surface flaws. When simplified interaction rules are exceeded, detailed fracture-mechanics solutions or crack models may be needed. Sensitivity to assumed coalescence is often appropriate because inspection cannot always resolve ligament continuity between close indications.
Metal-Loss Interaction
Multiple local-thin areas may merge into an extended weakness when the remaining full-thickness ligaments are too narrow to carry load independently. For pressure boundaries, orientation relative to hoop and axial stress can change the significance of the combined region. A thickness map should be reviewed for connected load paths rather than reduced to a list of isolated minimum points.
Mixed Defect Types
Some of the most severe integrity cases involve mixed damage: a crack at the base of a corrosion pit, gouge within a dent, crack at a weld toe inside local wall loss, or fretting damage at a highly loaded contact. The blunt defect may increase local stress while the sharp flaw controls fracture. Treating the defects independently can miss that coupling.
Growth Can Create Future Interaction
Defects that are independent today may interact after fatigue growth, corrosion growth or local deformation. Remaining-life analysis should therefore consider not only current interaction but when evolving defects cross the interaction criterion. Coalescence can cause a step change in effective crack length and crack-driving force.
Detailed FEA and Fracture Models
Advanced modelling can resolve complex interactions, but verification burden increases rapidly. Mesh refinement, crack-front definition, contact between crack faces, local thickness geometry and material nonlinearity can all influence the result. A detailed model should be checked against the conservative combined-defect assessment so that any recovered margin has a clear physical explanation.
Inspection Implications
When several indications are close together, the NDT technique must be capable of resolving individual flaw boundaries and the ligament between them. If it cannot, the structural model should not assume that the ligament is sound. A conservative merged flaw may be more defensible until improved inspection provides evidence to separate the defects.
Assessment Workflow
- Map all relevant indications in one common coordinate system
- Screen for interaction using the governing geometry rules
- Recharacterise interacting indications conservatively
- Assess the combined defect against fracture, collapse and other relevant limits
- Evaluate whether future growth can cause currently separate defects to interact
- Use refined modelling only where it removes identifiable conservatism and can be verified
Decision Principle
The objective is not to maximise the number of independent defects that can be claimed. It is to represent the damaged region in a way that bounds the credible structural response while remaining traceable to inspection evidence.
If inspection cannot demonstrate that two close indications are structurally independent, the assessment should not assume independence by default.
Engineering judgement — governing sensitivities
For Multiple Defects, Interaction & Coalescence, the most useful review question is not simply whether the solver has produced a plausible contour or scalar result, but whether the model preserves how measured defect morphology is converted into an assessment geometry. Local depth alone can be misleading when defect length, spacing, ligament interaction and measurement resolution control collapse or crack-like behaviour. This is where apparently small modelling choices can change the engineering conclusion. The analyst should identify the variables that can move the governing response, separate physical uncertainty from deliberate conservatism, and show that the selected modelling fidelity is proportionate to the decision being supported. Where the response is close to an acceptance boundary, sensitivity cases should bracket credible changes rather than apply arbitrary percentage perturbations.
Verification evidence for the engineering record
A defensible Multiple Defects, Interaction & Coalescence assessment should leave an evidence trail that another engineer can independently interrogate. At minimum, review inspection-map processing, interaction criteria, minimum ligament, profile idealisation, measurement uncertainty and comparison of simplified screening with detailed local-shell/solid analysis for borderline cases. Numerical convergence should be demonstrated on the response quantity that drives the decision, not only on generic mesh or solver metrics. The report should distinguish verified numerical behaviour from validation against test or service evidence, record any extrapolation beyond the supporting data, and state which assumption would most likely change the conclusion. This turns the analysis from a plausible calculation into an auditable engineering substantiation.