Types of NDT methods: RT, UT, PT, MT, VT and ECT.
Non-destructive testing (NDT) is the examination of a material, weld or component for discontinuities without damaging it or affecting its future use. Six methods cover the overwhelming majority of industrial work: visual (VT), liquid penetrant (PT), magnetic particle (MT), ultrasonic (UT), radiographic (RT) and eddy current (ECT). None of them finds everything. Choosing the wrong one — or accepting a report from a method that physically cannot detect the defect you are worried about — is one of the most common and least visible quality failures on a fabrication package.
The six standard methods split into surface techniques (VT, PT, MT, ECT) and volumetric techniques (RT, UT). Surface methods find what breaks or nearly breaks the surface; volumetric methods look through the section. Specifying a surface method against an internal defect concern is a wasted examination.
Method selection is driven by the referencing code, the material, the section thickness, the joint geometry and the defect type of concern. Acceptance criteria never come from the NDT method itself — they come from the construction code or the project specification.
- 01VT, PT, MT and ECT are surface or near-surface methods; RT and UT are volumetric.
- 02MT only works on ferromagnetic materials; PT covers non-porous materials including stainless and aluminium.
- 03RT images volumetric defects well and leaves a permanent record; UT is better at planar defects such as lack of fusion and cracks.
- 04Personnel must be qualified and certified to a written practice such as ASNT SNT-TC-1A, ANSI/ASNT CP-189 or ISO 9712.
- 05Acceptance criteria come from the construction code (ASME, AWS, API) or the specification — not from the NDT procedure.
The six methods at a glance
Start here when a specification names a method and you need to know what it will and will not tell you.
| Method | Detects | Materials | Key limitation |
|---|---|---|---|
| VT — Visual Testing | Surface-breaking defects, profile, undercut, overlap, misalignment | All | Surface only; depends heavily on access, lighting and inspector skill |
| PT — Liquid Penetrant | Surface-breaking discontinuities: cracks, porosity, laps | Non-porous metals and some plastics | Only open-to-surface defects; surface must be clean and non-porous |
| MT — Magnetic Particle | Surface and slightly subsurface discontinuities | Ferromagnetic only (carbon and low-alloy steel) | Not usable on austenitic stainless, aluminium, copper; orientation-sensitive |
| UT — Ultrasonic | Internal planar defects: lack of fusion, cracks, laminations; also thickness | Most metals; difficult on coarse-grained austenitics | Needs skilled operator, couplant, calibration blocks; no permanent image without encoded data |
| RT — Radiographic | Internal volumetric defects: porosity, slag, incomplete penetration | Most materials within thickness range | Poor at tight planar defects parallel to the beam; radiation safety and access constraints |
| ECT — Eddy Current | Surface and near-surface cracks, tube wall loss, conductivity/sorting | Electrically conductive materials | Shallow penetration; results affected by permeability, geometry and lift-off |
Method by method
What actually happens on the shop floor, and the conditions under which each result can be trusted.
- Step VTSurface | Performed on every weld
Visual testing
Direct or remote examination of the surface against defined acceptance criteria, using weld gauges, measuring tools, mirrors, borescopes and adequate illumination (commonly a minimum of 100 fc / 1000 lux at the surface for weld examination). VT is the first and cheapest examination and catches most rejectable conditions on structural work — undersized fillets, undercut, overlap, arc strikes, craters and poor profile.
- Step PTSurface | Non-magnetic materials
Liquid penetrant testing
A coloured or fluorescent penetrant is applied, allowed to dwell so capillary action draws it into open discontinuities, then removed from the surface before a developer blots it back out to form a visible indication. Requires a clean, non-porous surface and disciplined dwell and development times. Works on austenitic stainless, aluminium and other non-magnetic materials where MT cannot be used.
- Step MTSurface & near-surface | Carbon steel
Magnetic particle testing
The part is magnetised (yoke, prods or coil) and ferrous particles — dry or wet fluorescent — are applied. Flux leakage at a discontinuity holds the particles, forming an indication. Discontinuities are best detected perpendicular to the magnetic field, so examinations are performed in two directions approximately 90 degrees apart. Faster and more tolerant of surface condition than PT, but limited to ferromagnetic materials.
- Step UTVolumetric | Planar defects
Ultrasonic testing
High-frequency sound is coupled into the material and reflections from discontinuities and back walls are timed and amplified. Straight-beam UT measures thickness and finds laminations; angle-beam shear-wave UT examines welds. Phased array (PAUT) steers multiple elements electronically to produce sectional imaging with encoded, reviewable data. Calibration blocks, scan plans and technician competence determine whether the result means anything.
- Step RTVolumetric | Permanent record
Radiographic testing
X-ray or gamma radiation passes through the weld onto film or a digital detector, producing an image where thickness and density changes appear as density differences. Excellent for porosity, slag inclusions and incomplete penetration, and it produces a permanent, independently reviewable record. Tight cracks and lack of fusion oriented across the beam may go undetected, and radiation safety controls constrain the work area.
- Step ECTSurface & near-surface | Conductive materials
Eddy current testing
An alternating current coil induces eddy currents in a conductive part; discontinuities disturb the current flow and change the coil impedance. Used for surface crack detection through thin coatings, heat-exchanger tube inspection, weld toe examination and material or heat-treat sorting. Penetration is shallow and results are sensitive to geometry, permeability and probe lift-off.
When specifications require each method
Method choice is rarely open — the construction code and the project specification usually decide it for you.
Pressure equipment (ASME VIII, B31.1, B31.3)
RT or UT on butt welds according to the joint efficiency and service category claimed, with PT or MT on fillet, attachment and root passes where specified. Examination method and extent follow the code case selected at design, not site convenience.
Structural steel (AWS D1.1, CSA W59)
VT on all welds; UT commonly on complete-joint-penetration groove welds, RT where specified; MT on fatigue-critical details. Cyclically loaded structures attract more onerous criteria than statically loaded.
Pipeline girth welds (API 1104)
RT or automated UT on production girth welds at defined frequency, with acceptance criteria either workmanship-based or by engineering critical assessment where the specification allows.
In-service inspection (API 510 / 570 / 653)
UT thickness surveys for corrosion monitoring, plus MT/PT on welds and ECT on exchanger tubing at intervals driven by the inspection plan and corrosion rate.
Castings and forgings
RT and UT for internal soundness, MT or PT for surface integrity, with severity levels referenced to ASTM reference radiographs where applicable.
Stainless and non-ferrous fabrication
PT replaces MT because the material is non-magnetic; UT on austenitic welds requires procedures qualified for coarse-grain attenuation.
Where acceptance criteria actually come from
The NDT procedure describes how the examination is performed. What counts as rejectable comes from somewhere else.
ASME Section V (and equivalents such as EN ISO standards) describes how each method is carried out. The construction code — ASME Section VIII, B31.3, AWS D1.1, API 1104 — states what is acceptable. The project specification can tighten those limits but should never loosen them. When a report cites only the method standard and no acceptance standard, the result has no verdict attached to it.
Typical criteria are expressed as limits on indication length, area, cumulative length within a weld length, distance between indications, and absolute prohibitions on cracks, lack of fusion and incomplete penetration. Rounded indications are generally treated more leniently than elongated or planar ones because of their lower stress-concentration effect.
Where NDE extent belongs
Examination method, extent and acceptance criteria should be fixed inside the Inspection and Test Plan before fabrication starts — not negotiated after the welds are made.
Why oversight sits apart from the testing
Where the party performing the NDT also judges the results, marginal indications have a way of becoming acceptable ones. Independent oversight separates the two: procedure and personnel verification before mobilisation, witnessing of the examination on the day, and independent review of radiographs, UT data and reports before they enter the quality record.
That review is also where record quality is fixed. Reports that cannot be traced back to a weld number on the weld map are indefensible at handover, regardless of how good the examination was.
Independent welding and NDT inspection
CWI-led weld inspection with ASNT-qualified oversight of RT, UT, PT, MT and PAUT — procedure and personnel verification, witnessing, and independent interpretation.
Questions we get on this topic
What are the main types of NDT?
The six standard industrial methods are visual testing (VT), liquid penetrant testing (PT), magnetic particle testing (MT), ultrasonic testing (UT), radiographic testing (RT) and eddy current testing (ECT). VT, PT, MT and ECT examine the surface or near-surface; UT and RT examine through the volume of the material.
What is the difference between RT and UT?
Radiography produces an image of density differences through the section, so it is strong on volumetric defects such as porosity and slag and leaves a permanent record. Ultrasonics reflects sound off discontinuities, so it is stronger on planar defects such as lack of fusion and cracks, and it is preferred on thicker sections and where radiation safety or access is a constraint. Many specifications require both on critical joints.
When should PT be used instead of MT?
Magnetic particle testing only works on ferromagnetic materials such as carbon and low-alloy steel. On austenitic stainless steel, aluminium, copper alloys and other non-magnetic materials, liquid penetrant is used instead. Where both are possible, MT is faster, more tolerant of surface condition and can detect slightly subsurface discontinuities.
What qualification do NDT technicians need?
Personnel are qualified and certified to a written practice — most commonly ASNT SNT-TC-1A or ANSI/ASNT CP-189 in North America, or ISO 9712 elsewhere — at Level I, II or III for each method. Level II interprets and reports; Level III writes and approves procedures. Certification files and current vision examinations should be verified before mobilisation.
Who sets NDT acceptance criteria?
The construction code or the project specification, not the NDT method standard. ASME Section V describes how examinations are performed; ASME Section VIII, B31.3, AWS D1.1 or API 1104 state what is acceptable. A report that names no acceptance standard cannot support an accept or reject decision.
How much NDT does a project need?
The extent is set by the referencing code, the service category and the project specification, and it should be recorded in the Inspection and Test Plan before fabrication starts. Codes set minimum examination for certain joint types; owners commonly increase it on critical or high-consequence welds.
Need independent NDT oversight on your fabrication?
Send us the code, the weld scope and the shop location — we'll confirm the right method coverage, certification levels and mobilisation window within one business day.