3.2 Surface vs Subsurface and Method Selection Logic

Key Takeaways

  • Choose a method from material, access, expected discontinuity class, and production versus in-service constraints, then weigh safety and cost.
  • An alternating-current yoke concentrates flux at the surface and is the usual weld-toe fatigue tool; prod direct current is chosen when a ferromagnetic casting needs a deeper circular field.
  • Solvent-removable penetrant (Method C) is the portable field-weld method; water-washable penetrant (Method A) fits production lines with rinse stations.
  • Orientation governs detectability: magnetic field across the crack, ultrasonic beam near-normal to the reflector, radiographic density change along an open path through the plane.
Last updated: August 2026

Method selection is the specific-exam skill that sits on top of capability. The general exam asks whether a method can work. The specific exam asks whether a procedure chose the right method, technique, and current or penetrant family for this product, this access, and this expected discontinuity.

Why selection is a decision, not a habit

A shop that always does magnetic particle testing on every metal, or always shoots radiography on every weld, will fail Applications/Techniques items. Build the decision in this order:

  1. What is the material?
  2. What is the access (one side, two sides, local, or full circumference)?
  3. What discontinuity class is expected?
  4. Is this production or in-service?
  5. What do safety, cost, speed, and the governing procedure allow?

Industry procedures written to ASME Section V, AWS D1.1, ASTM E1444, ASTM E1417, ASTM E94, and ASTM E164 do not pick methods at random. They encode this logic.

Material first

Material classTypical examplesMethods that can be primaryImmediate rejects as primary
FerromagneticCarbon steel, most low-alloy steels, some 400-series stainless, cast ironsVisual, penetrant, magnetic particle, radiographic, and ultrasonic testingNone of the five is automatically excluded
Nonmagnetic metalAustenitic stainless (304/316), aluminum, copper alloys, titanium, nickel alloysVisual, penetrant, radiographic, and ultrasonic testingMagnetic particle testing — no usable flux leakage
NonmetalMany plastics, ceramics, composites (procedure-dependent)Visual testing; penetrant testing if the material is not attacked and is nonporous; radiography; ultrasound with the right probe and couplantMagnetic particle testing; penetrant testing on porous or solvent-sensitive surfaces

If the stem says Type 316L stainless fillet, magnetic particle testing is already wrong. If it says painted carbon-steel girder, in-service toe crack, magnetic particle or penetrant testing after local coating removal is on the table. Eddy current testing can find surface cracks in conductive metals in industry, but it is not an exam method in this ASNT NDT Level II program.

Access: one side versus two sides

Two-side access means you can place a source on one side and a detector or film on the other (radiography), or you can put a yoke or prods across the area, or you can scan from complementary surfaces. One-side access means you can reach only the near surface: the inside of a closed vessel is not available, a backing bar is still in place, or insulation was stripped only on the outside diameter.

  • Radiographic testing in the conventional sense needs a path from source through the volume to the detector. A closed box with no film-side access is a poor radiographic candidate.
  • Ultrasonic testing thickness scans and many weld scans are one-side methods. That is why ultrasound is the default for in-service remaining thickness and for many butt welds when only the cap side is reachable.
  • Penetrant and magnetic particle testing need access to the examination surface of interest. A root crack that is not open to the outside diameter will not appear on an outside-diameter penetrant test. An inside-diameter-initiated fatigue crack needs inside access or a volumetric method (ultrasound).
  • Visual testing needs line of sight, or a remote visual tool, plus lighting.

Expected discontinuity drives the method

Expected conditionWhy it formsFirst-line methodsWeak or wrong first-line choices
Fatigue crack at a weld toe or threadCyclic load, surface initiationVisual testing plus magnetic particle testing (ferromagnetic) or penetrant testing (any metal); ultrasound to size depthRadiography — tight, often unfavorably oriented
Porosity in weld metal or a castingTrapped gasRadiography (best map); ultrasound if large; visual or penetrant testing if surface-breakingMagnetic particle testing unless the pores break a ferromagnetic surface
Lamination in platePipe or inclusions rolled flat, parallel to the surfaceStraight-beam ultrasonic testingRadiography (plane usually parallel to the film); penetrant or magnetic particle testing unless the edge is open
UndercutWelding technique, toe grooveVisual testing is primary; penetrant or magnetic particle testing confirmUltrasound and radiography are the wrong tools for a surface groove you can see
Corrosion / erosion wall lossService environment or flowUltrasonic thickness; visual testing if accessible; radiographic profile in some geometriesPenetrant or magnetic particle testing find pits or cracks, not remaining ligament

Production versus in-service

Production (shop) examinations favor throughput, permanent records, and process feedback. A fabrication shop may run water-washable penetrant Method A on a line of fittings, shoot radiography on every complete-penetration groove weld required by the construction code, and use magnetic-particle yokes on carbon-steel fillet welds before paint.

In-service examinations favor portability, one-side access, minimal insulation removal, and the discontinuities that grow in service — fatigue, corrosion, erosion, and hydrogen cracks. Field crews reach for alternating-current yokes, solvent-removable penetrant kits, and ultrasonic thickness or shear-wave kits. Radiation exclusion zones and film processing are harder in an operating unit.

Safety, cost, and speed

  • Radiographic testing brings ionizing radiation, barriers, and (for gamma) source control. It is slow to set up and expensive per shot, but it produces a permanent volumetric record of many voids.
  • Ultrasonic testing has no radiation zone. It is fast for thickness and, in skilled hands, fast for welds. It is more operator-dependent and usually has no film-like picture unless the scan is encoded.
  • Magnetic particle and penetrant testing are relatively cheap and fast on local areas. Penetrant chemicals have flammability and health controls; magnetic particle testing has electrical and, with prods, arc-burn hazards.
  • Visual testing is the cheapest first look and is required, formally or informally, before most other methods.

Orientation matters

Detectability is directional. Memorize three orientation rules.

Magnetic field versus crack. Leakage is greatest when the magnetic field crosses the discontinuity (field roughly perpendicular to the crack length). A field parallel to a tight crack may produce no useful particle pattern. That is why procedures require two field directions about 90° apart, or a technique that produces a vector field. ASTM E1444 and ASME Section V, Article 7 both hammer this point.

Radiographic beam versus planar flaw. A useful film indication needs a path of reduced absorption. The favorable orientation is a beam that travels along an open crack plane. The unfavorable orientation is a beam that strikes the face of a tight plane and crosses only the crack opening. Tight lack of fusion along a bevel often sits in that unfavorable geometry. Image-quality indicators prove technique, not that every crack was imaged.

Ultrasonic beam versus reflector. Echo amplitude is highest when the beam is near-normal to the reflector. Straight beam (0°) finds laminations and backwall loss parallel to the scanning surface. Angle-beam shear wave is chosen so the central ray hits weld fusion faces and the root land. A 70° probe on a thin weld and a 45° or 60° probe on a thicker weld are geometry choices, not superstition. ASTM E164 and ASME Section V, Article 4 describe that calibration and scanning logic.

Procedure flavor the specific exam likes

Why a procedure picks a yoke alternating current on a weld toe versus prod direct current on a casting

An alternating-current (AC) electromagnetic yoke produces a longitudinal field between the poles and, because of the magnetic skin effect, concentrates flux at the surface. It needs no electrical contact with the part, is one-person portable, and matches in-service weld-toe fatigue cracks. Prod magnetization drives high amperage into the part through contact prods and produces a circular field. Direct current (DC) or rectified current penetrates deeper than alternating current, so prods plus direct current (or half-wave direct current) are chosen on castings when the concern is near-surface shrinkage cracks or other subsurface conditions in a ferromagnetic casting. The cost is arc-burn risk, usually two-person handling, and the need to space prods and set amperage (commonly discussed in ASTM E709 / E1444 as a function of prod spacing). Do not put prods on a finished pressure boundary or a thin aerospace part unless the procedure explicitly allows it.

Why penetrant Method C on a field weld versus Method A in a line

ASTM E1417 and ASME Section V, Article 6 classify penetrant methods by excess-penetrant removal. Method A is water-washable: a rinse station, controlled water temperature and pressure, high throughput, good for production lots and relatively rough surfaces. Method C is solvent-removable: a spray can of cleaner, a cloth, and a local work area. It is the field-weld and repair-weld method when water would contaminate a process unit, when only a few inches of weld were ground, or when the crew is on a scaffold. Method C is slower and is easy to over-remove (wiping penetrant out of the discontinuity). Method A is the wrong mental picture for a single painted structural connection in a running plant.

Why ultrasonic angle-beam on a butt weld versus radiography

A complete-penetration butt weld hides its most serious planar defects — lack of fusion on the bevel, incomplete penetration at the land, and cracks — on planes that radiography often under-calls. Angle-beam ultrasonic testing places a shear wave so those planes are reflectors. It needs one-side access, no radiation boundary, and a calibration block. Radiographic testing remains excellent when the construction code wants a permanent image of volumetric conditions (porosity, slag) and two-side access plus a radiation program are available. Many construction codes allow either radiography or ultrasound for groove welds; the specific exam wants you to know why a procedure picked one. If the stem emphasizes planar fusion-face defects, limited access, or an operating unit, ultrasound is the reasoned choice. If it emphasizes rounded inclusions, a shop bay with a vault, and a film record, radiography is the reasoned choice.

Putting a decision together

Work a full stem the way the specific exam writes it.

Carbon-steel pressure-vessel long seam, 1 inch (25 mm) thick, post-weld heat treated, in the shop, two-side access, construction code allows radiography or ultrasound, concern is both slag and incomplete fusion. Both radiography and ultrasound can be justified. Radiography will map slag; ultrasound will be stronger on incomplete fusion. A procedure may require both, or ultrasound with a supplemental visual and surface method on the cap and root.

Austenitic stainless field weld, outside diameter only, looking for service-induced chloride cracks at the toe. Magnetic particle testing is out (nonmagnetic). Radiography is weak on tight surface cracks and needs two-side access. The logical pair is visual testing plus penetrant Method C after local insulation and coating removal; ultrasound can supplement for depth if a crack is found.

That decision table — material, access, discontinuity, production versus service, orientation, then safety and cost — is the Level II skill this chapter exists to build.

Test Your Knowledge

A procedure for an in-service carbon-steel pipe weld calls for an alternating-current electromagnetic yoke at the toes, not prod magnetization. Why is that the usual choice?

A
B
C
D
Test Your Knowledge

A field weld on a painted carbon-steel structural connection must be examined locally after a small repair, with no water available. Which penetrant approach matches that constraint?

A
B
C
D
Test Your Knowledge

A 1-inch carbon-steel butt weld must be examined for incomplete fusion at the bevel faces. Two-side access for a source and film is poor, and the expected flaw is planar. Which method-and-technique pair is the most logical primary choice?

A
B
C
D