Nondestructive Examination Coordination: Methods, Timing, and Acceptance
Choosing and sequencing methods
Nondestructive examination (NDE) confirms weld soundness without destroying the joint. The common methods are visual, penetrant, magnetic particle, ultrasonic, and radiographic testing. Each has strengths and blind spots: surface methods detect surface-breaking defects; volumetric methods detect internal ones; ultrasonic is sensitive to planar defects and requires access and surface condition; radiography is sensitive to volumetric defects and depends on density differences and orientation. Candidates should know that no single method finds everything, which is why specifications often combine visual with a volumetric method.
Timing and coordination
The examination must be performed at the correct stage. Visual examination follows cleaning and precedes or accompanies other methods; penetrant and magnetic particle testing are typically done after the weld has cooled and been cleaned; volumetric examination is usually performed after completion and before any pressure test or coating. Traps include examining before the weld has cooled sufficiently, coating before examination, or performing magnetic particle testing on austenitic stainless steel, which is generally nonmagnetic.
Personnel, procedures, and acceptance
The examiner must be qualified to the level required for the method and technique, and the examination must follow a written procedure. Acceptance criteria come from the applicable code or contract, not from the examiner's judgement. A frequent question trap is confusing a rejection with a repair instruction: the examiner reports indications against criteria; the disposition and repair method belong to the quality system and the welding procedure.
Rule of thumb
Match the method to the defect type and location, confirm the timing against the inspection plan, and confirm that the examiner, procedure, and acceptance criteria are all specified before examination begins.
Worked example: a specification requires internal examination of a butt weld in carbon steel. Ultrasonic testing is selected, but the surface is too rough for reliable coupling. The correct response is to condition the surface or select an alternative method with agreement, not to proceed and report a result of doubtful validity.