Composite NDI: five pitfalls that turn a clean part into a program risk
Most composite NDI programs fail the same five ways. They're all avoidable with a coverage strategy tied back to the structural analysis.
Composite non-destructive inspection sits between structures and quality, and it's usually owned by neither. That's how a program ends up with an inspection plan that produces data nobody knows how to act on. A few recurring patterns:
1. Method chosen before defect type
The ultrasonic scanner is not always the right answer. Tap testing, thermography, and X-ray each catch different populations of defects. Pick the method against the failure modes the structural analysis actually cares about — porosity, delamination, disbond, impact damage — not against the equipment on hand.
2. Coverage strategy that ignores geometry
Every composite part has zones the inspection technique struggles with — sharp radii, ply drops, hidden interfaces. If the coverage plan doesn't call those zones out and specify a secondary method, the inspection report is quietly incomplete.
3. Acceptance criteria disconnected from the analysis
A defect map is only useful if the reader can decide whether the part flies. That requires acceptance criteria written by someone who understands the residual strength of the structure — not a generic table copied from a spec.
4. Repair engineering treated as paperwork
Scarf, doubler, or replace is a structural decision, not a shop-floor decision. When repair engineering is delegated to the depot without a margin justification, the same defects come back with the same repairs — and no one asks whether the root cause was ever addressed.
5. No feedback loop to design
Every recurring finding is design telling the program something. A functioning NDI program feeds anonymized trends back into the next design revision. Without that loop, the inspection function is an audit — not an engineering asset.