Inspection & QA/QC·5 June 2026·4 min read

Choosing the right NDT method for a welded joint

Non destructive testing (NDT) is how you demonstrate that a weld meets its acceptance criteria without damaging the product. Selecting the right method is not a matter of preference. The choice follows from the material, joint geometry, accessibility, the specified acceptance criteria and, above all, from the defects you expect and where they sit: at the surface or within the volume of the weld.

This article puts the five common methods side by side: visual testing (VT), penetrant testing (PT), magnetic particle testing (MT), ultrasonic testing (UT) and radiographic testing (RT). For each method you will find its typical use, strengths, limitations and the relevant standards, so engineers and QC staff can make a well founded choice.

Surface methods versus volumetric methods

The first split is straightforward. Surface methods (VT, PT, MT) detect defects at or just below the surface: cracks, undercut, open pores and lack of fusion breaking the surface. Volumetric methods (UT, RT) look through the weld and reveal internal defects: porosity, slag inclusions, internal lack of fusion and incomplete penetration.

The rule of thumb: every weld receives visual testing first. The execution class or application then determines whether additional surface or volumetric testing is required. EN 1090-2, for instance, links the percentage of supplementary NDT to the execution class (EXC1 to EXC4) and the joint type, while for pressure equipment under PED 2014/68/EU the extent of testing follows from the category and the conformity module.

VT: visual testing as the foundation

Visual testing to ISO 17637 is the fastest and cheapest method and, in practice, catches a large share of all rejectable defects. A trained eye immediately spots undercut, excessive penetration, spatter, shape imperfections and surface pores. Assessment is made against the quality levels of ISO 5817.

The advantages: immediate results, no special equipment, applicable to any material. The limitation: you only see what is visible at the surface, and the quality of the examination depends on competence, lighting and access. Our article on the visual weld inspection method covers the practical routine.

PT and MT: revealing surface defects

Penetrant testing (PT) uses a penetrating liquid that is drawn into surface breaking openings and produces a clearly visible indication once a developer is applied. PT works on almost any non porous material, including austenitic stainless steel and aluminium. Its drawback: only surface breaking defects are found, and the surface must be clean and dry.

Magnetic particle testing (MT) magnetises the workpiece and uses iron particles to reveal flux leakage. MT also finds defects just below the surface and is faster than PT, but it only works on ferromagnetic material, so it is not an option on stainless steel or aluminium. For both methods, personnel should be qualified in accordance with ISO 9712.

UT: ultrasonic testing for internal defects

Ultrasonic testing sends sound waves through the material and measures reflections from interfaces and defects. UT excels at finding planar, unfavourably oriented defects such as lack of fusion and cracks, precisely the defects that matter most mechanically. It works well from a few millimetres of wall thickness upwards and provides depth information.

There are limitations. Thin plate and coarse grained materials are difficult, the geometry must be scannable and interpretation requires an experienced, certified operator. Modern variants such as phased array (PAUT) and TOFD provide recorded, reviewable data and replace radiography in many projects, without radiation hazards or production stops.

RT: radiography as a recording volumetric method

Radiographic testing produces a projection image of the weld using X ray or gamma radiation. Volumetric defects such as porosity and slag inclusions show up clearly, and the film or digital image is a permanent record for the project dossier. That makes RT popular for pipelines and pressure equipment.

On the other hand, planar defects oriented unfavourably to the beam can be missed, radiation safety requires the work area to be cleared, and exposures are relatively slow and costly. Wall thickness, material, expected defects, contract requirements and logistics all play a part in the UT versus RT decision. Independent advice on this is part of our NDT coordination service.

Making the choice in practice

Work through these steps and record the outcome in the inspection and test plan:

Establish the requirements: standard, execution class or category, client specification and acceptance criteria.
Identify the expected defects per welding process and joint type: which defects, at the surface or internal.
Select the combination per joint: always VT, then PT or MT for the surface (MT on ferromagnetic steel, PT on stainless steel and aluminium) and UT or RT for the volume.
Check the practical constraints: access, wall thickness, surface condition, safety and schedule.
Record extent, timing and acceptance criteria per method and verify that NDT personnel are qualified to ISO 9712.

If you are unsure which defects to expect, read our article on common weld defects and their causes.

Frequently asked questions

No. The methods complement each other. Many critical defects, such as cracks and undercut, sit at the surface and are found with VT, PT or MT. Volumetric testing adds coverage for internal defects, it does not replace surface examination.

UT is preferred for thicker sections, for planar defects such as lack of fusion and cracks, and where radiation safety or continued production matters. RT is strong on thin walled work, volumetric defects and where a recorded image is required. Contract and standard often settle the choice.

NDT personnel should be qualified and certified to ISO 9712 in the relevant method and sector, typically level 2 or higher for performing and evaluating examinations. The manufacturer remains responsible for the correct extent and timing of testing.

That follows from the standard and the contract. EN 1090-2 links supplementary NDT percentages to the execution class and joint type, while for pressure equipment the extent follows from the PED category and conformity module. The inspection and test plan records this per project.