The practical answer
Magnetic particle inspection is commonly useful for surface and near-surface discontinuities in suitable ferromagnetic materials. Dye penetrant is useful for surface-breaking discontinuities in suitable non-porous materials. Ultrasonic testing can support internal flaw detection, thickness measurement and other specialised examinations depending on technique and geometry.
Those descriptions are only a starting point. Surface condition, coating, geometry, access, orientation, thickness and operator technique can matter as much as the nominal capability of the method.
The engineering team should define what the inspection result will be used for: finding a crack, sizing damage, verifying a repair, screening a fleet or supporting remaining-life assessment.
When this guide is useful
- A visible indication needs confirmation and extent mapping.
- A weld repair needs pre-repair and post-repair examination planning.
- Wall loss or internal damage is suspected.
- A fabrication or maintenance scope says 'NDT required' but does not define what problem the inspection must solve.
- Treating one method as universally superior.
- Selecting a method without considering material suitability or access.
- Using a clean inspection result as proof that the component is structurally adequate for service.
- Copying an examination percentage from another job without matching the design and fabrication basis.
A practical NDT selection framework
Start with the suspected defect and required decision, then work back to the method and coverage.
Describe the suspected defect
Surface crack, near-surface crack, internal flaw, wall thinning, lack of fusion, lamination or another mechanism?
Confirm material and geometry
Material type, thickness, weld geometry, curvature, surface condition and access determine which methods are practical.
Define the inspection purpose
Screening, locating, sizing, repair verification or in-service assessment require different confidence and coverage.
Choose method and coverage together
A capable method used over the wrong area is still a poor inspection plan. Define where and how much to examine.
Link results to an engineering action
Set what happens if indications are found, uncertain or absent so the inspection supports a real decision.
Detailed engineering examples
These cases show how the framework changes a real engineering decision. They are deliberately written around uncertainty, options and evidence rather than one universal answer.
Weld toe crack on a heavy mounting bracket
A fine line is visible at a weld toe on a carbon steel bracket. The team needs to confirm whether it is a crack and map its extent before repair design.
What makes the decision difficult
- The suspected defect is surface-breaking and close to a weld geometry change.
- The steel surface is painted and contaminated by service conditions.
- The repair scope depends on knowing the crack endpoints and whether similar indications exist nearby.
How the guide should be applied
- Start with visual examination and suitable surface preparation.
- Discuss a surface-crack method appropriate to ferromagnetic steel and the local geometry.
- Expand the inspection area beyond the visible line based on the expected stress field and repeated-detail pattern.
- Use additional methods only if the defect mechanism or geometry suggests subsurface damage that the surface method cannot address.
The method choice is driven by a likely surface-breaking crack in steel and the need to map extent before repair. The result then feeds into the engineering repair assessment.
Possible corrosion under insulation on a pressure line
Insulation is removed locally and external corrosion is found. The site needs to determine whether the damage is isolated or part of broader wall loss.
What makes the decision difficult
- The engineering decision depends on remaining wall thickness and affected length.
- Only part of the insulation can be removed during the initial inspection window.
- Spot measurements may miss localised damage patterns.
How the guide should be applied
- Define the degradation mechanism and likely locations before selecting inspection points.
- Discuss UT-based thickness examination and coverage strategy rather than treating the problem as a crack search.
- Use results to decide whether the inspection area needs expansion and whether remaining-life assessment is justified.
- Record locations and repeatable measurement points for future trending where appropriate.
The useful inspection plan focuses on wall-loss extent and measurement quality, not on choosing a familiar crack-detection method by habit.
Surface-breaking indication on stainless steel equipment
A stainless component shows a fine linear indication near a fabricated attachment. The surface can be cleaned and accessed directly.
What makes the decision difficult
- The material is not suitable for magnetic particle inspection.
- The suspected defect is still surface-breaking.
- Surface preparation and cleanliness are critical to reliable examination.
How the guide should be applied
- Confirm material and surface condition first.
- Discuss a penetrant-based surface examination with the inspector and define preparation requirements.
- Consider whether the geometry or service mechanism also creates a need for additional internal examination.
- Use the result within the broader engineering assessment rather than as a stand-alone acceptance decision.
Material suitability changes the method conversation, but the same principle remains: choose the method to match the defect type and engineering purpose.
Common traps
Starting with 'we always use MPI' before defining the defect mechanism can miss the actual inspection need.
A suitable method applied only to the visible defect may miss the broader damage field.
Inspection data are less useful if no one has decided what indications mean for repair or continued service.
Every method has limitations. A clean result only supports the questions the method and coverage were capable of answering.
Evidence to gather before deciding
Suspected defect mechanism and location
Material and thickness
Weld or component geometry
Surface condition and coating
Access and isolation constraints
Decision the result must support