NDT Methods for Welds
and Pressure Equipment —
A Practical Guide

Non-destructive testing (NDT) — also called non-destructive examination (NDE) — is the family of techniques used to evaluate the integrity of materials, welds and components without rendering them unserviceable. For mechanical engineers specifying pressure vessels, process pipework, structural steelwork or fabricated assemblies, NDT is not simply a quality assurance step imposed by the inspection department. It is a design decision: the choice of NDT method, the extent of examination, and the acceptance criteria directly determine what defects the finished product may contain and what confidence you have in its integrity over its service life.

This article covers the five methods that account for the overwhelming majority of weld and pressure equipment examination — visual testing (VT), radiographic testing (RT), ultrasonic testing (UT) including phased array (PAUT), magnetic particle testing (MT), and liquid penetrant testing (PT) — together with the standards that govern their application, their comparative strengths and limitations, and practical guidance on specifying examination scope in engineering documents.

Visual Testing — VT

Visual testing is the foundation of all weld examination and is required as a minimum under every major fabrication code. EN ISO 17637 governs VT of fusion welds; ASME Section V Article 9 covers visual examination under the ASME framework. VT detects surface-breaking discontinuities: undercut, overlap, surface porosity, cracks, incomplete fill, incorrect weld profile, and dimensional deviations from the weld drawing. It requires no consumables and minimal equipment, but it requires a properly trained examiner — an eye that knows what a good weld looks like and what deviation from that profile might indicate about the subsurface condition.

VT is performed at every stage of fabrication: before welding (joint fit-up and cleanliness), during welding (interpass temperature compliance, bead profile, slag removal) and after welding (final profile, surface condition, dimensional checks). In standards terms, VT is a prerequisite for all other NDT methods — you do not RT or UT a weld that has not first passed VT, because volumetric methods will not reliably detect surface-breaking conditions as clearly as direct visual examination, and a weld with surface defects visible to the naked eye needs remediation before further examination is justified.

Radiographic Testing — RT

RT uses penetrating radiation — X-rays from an X-ray tube or gamma rays from a radioactive isotope such as Ir-192, Se-75 or Co-60 — to produce an image (on film or a digital detector) of the internal structure of a weld. Dense material absorbs more radiation; voids, porosity, slag inclusions and cracks absorb less. The resulting image shows these as darker areas against the lighter parent-metal background. RT is a volumetric method: it produces a two-dimensional projection through the full weld thickness and detects buried as well as surface discontinuities.

The governing standards are EN ISO 17636-1 (film radiography) and EN ISO 17636-2 (computed and digital radiography) in the European framework; ASME Section V Article 2 in the ASME framework. Both define image quality indicators (IQIs or penetrameters) that must be placed on the weld to verify that the film or detector has sufficient sensitivity to detect a defined defect size. Class B radiographs (the higher quality class in EN ISO 17636) require more controlled technique and are specified for structural and pressure-critical applications where Class A minimum quality is insufficient.

RT's principal limitation is its sensitivity to planar defects that lie parallel or near-parallel to the radiation beam. A tightly closed crack that is favourably oriented for the beam may produce no discernible indication. It also involves ionising radiation, requiring radiation protection controls, exclusion zones and regulatory notification in most jurisdictions — RT on a live industrial site requires a competent radiation protection supervisor and written working rules. Computed radiography (CR) and direct digital radiography (DR) have largely replaced film for new installations due to improved dynamic range, digital archiving, and faster turnaround, though film remains specified on some legacy projects and for applications where digital equipment access is restricted.

Ultrasonic Testing — UT and Phased Array UT

Ultrasonic testing uses high-frequency sound waves — typically 2–10 MHz for weld examination — pulsed into the material from a transducer. Sound reflects from internal interfaces: grain boundaries, weld fusion lines, and discontinuities. The time-of-flight from the transducer to the reflector and back, combined with the material's acoustic velocity, gives the depth and position of any reflector. UT is significantly more sensitive to planar defects than RT — a lack of fusion at the weld sidewall, which RT may miss entirely, is a strong reflector for UT because it presents a flat surface perpendicular to the sound beam.

Conventional UT uses a single fixed-angle transducer and is governed by EN ISO 17640 (manual UT of fusion welds in ferritic steels) and ASME Section V Article 4. The examiner manually scans the transducer across the heat-affected zone and records any indications above a defined amplitude threshold. Sensitivity is set against a reference calibration block with machined side-drilled holes or notches of defined dimensions.

Phased Array Ultrasonic Testing (PAUT) uses an array of piezoelectric elements that can be excited in controlled time sequences to steer and focus the sound beam electronically, generating a swept scan through multiple angles from a single pass. This produces a cross-sectional image (S-scan) that shows the full weld volume and provides significantly better probability of detection, better defect sizing, and a visual record of the examination. PAUT is governed by EN ISO 13588 and is increasingly specified as the preferred method for safety-critical pressure vessel and pipeline welds where ASME B31.3 and nuclear QA programmes apply. Time-of-flight diffraction (TOFD) — a further UT variant — is used alongside PAUT for highest-confidence weld inspection, particularly for production welds in pipelines and subsea structures.

Magnetic Particle Testing — MT

MT detects surface and near-surface discontinuities in ferromagnetic materials by magnetising the component and applying fine magnetic particles (wet ink or dry powder) to the surface. Where a discontinuity interrupts the magnetic field, a leakage field forms at the surface and the particles are attracted to it, forming a visible indication. MT can detect cracks, laps and seams at or just below the surface — typically to a depth of 2–3 mm — with excellent sensitivity for surface-breaking defects that are perpendicular to the applied field direction.

Governing standards: EN ISO 9934-1 (general principles), EN ISO 9934-2 (detection media), EN ISO 9934-3 (equipment); ASME Section V Article 7. MT is limited to ferromagnetic materials — carbon steel, low-alloy steel, and ferritic stainless steels. It cannot be used on austenitic stainless steel, aluminium, copper, or non-magnetic alloys. The method requires at least two magnetisation directions to ensure detection of discontinuities oriented in any direction. Fluorescent MT (using UV light and fluorescent ink) is specified where maximum sensitivity is required, such as in aerospace NDT or on safety-critical lifting equipment subject to LOLER inspection.

Liquid Penetrant Testing — PT

PT detects surface-breaking discontinuities in any solid, non-porous material by applying a coloured or fluorescent penetrant liquid to the surface, allowing it to enter open defects by capillary action, removing the excess penetrant from the surface, and then applying a developer that draws the trapped penetrant back out of the defect to produce a visible indication. PT is the only surface-breaking NDT method applicable to non-magnetic materials including austenitic stainless steel, aluminium, titanium and nickel alloys.

Governing standards: EN ISO 3452-1 (general principles), EN ISO 3452-2 (testing of penetrant materials); ASME Section V Article 6. PT is highly sensitive to fine surface cracks but gives no depth information — the indication shows that a crack is present and roughly how long it is at the surface, but tells you nothing about depth. Fluorescent penetrant (Type I) is more sensitive than visible dye (Type II) and is specified where maximum probability of detection is required. PT is sensitive to surface cleanliness: contamination, paint, oxide scale or corrosion products covering a crack will prevent penetrant entry and produce a false-negative. Surfaces must be clean, dry, and free of any coating over the area to be examined.

Weld Quality Acceptance Levels

NDT results are only meaningful when evaluated against defined acceptance criteria. The two principal frameworks for weld quality acceptance are EN ISO 5817 (fusion welds in steel, nickel, titanium and their alloys — quality levels) and ASME Section VIII Division 1 with Appendix 4 and 12 for radiographic and UT acceptance standards for pressure vessels.

EN ISO 5817 defines three quality levels:

• Level B — Stringent. The highest quality level, specified for safety-critical applications where consequences of failure are severe. Tight limits on all imperfection types including undercut, overlap, porosity, and profile.
• Level C — Intermediate. The standard level for pressure vessels, process pipework and structural applications under most European fabrication codes. EN 13445 (pressure vessels) and EN 1993-1-8 (structural joints) typically require Level C or better.
• Level D — Moderate. For non-critical applications where visual appearance and gross structural integrity are the primary requirements. Not appropriate for pressure service.

Extent of Examination — Spot, Sample and 100%

The required examination coverage is determined by the applicable fabrication code and the consequence of failure:

• 100% examination — every weld joint examined by the specified method. Required for Category 1 pressure vessels under the PED, for High Pressure service under ASME B31.3 Appendix K, for Level 1 nuclear fabrication, and for primary structural welds in offshore structures.
• Random / spot examination — a defined percentage of weld joints selected at random. ASME B31.3 Normal Fluid Service requires 5% spot RT or UT for quality factor E = 0.85 to apply. A higher quality factor (E = 1.0) is achieved by 100% examination.
• Production test pieces — representative test welds made alongside the production weld by the same welder, with the same parameters, examined destructively. Used where geometry prevents direct examination of the production joint.

Personnel Qualification

All NDT examinations must be performed by personnel certified to the appropriate level in the relevant method under EN ISO 9712 (European scheme) or ASNT SNT-TC-1A / CP-189 (North American scheme). In the UK, PCN (Personnel Certification in NDT) through BINDT is the recognised scheme for industrial NDT. Level 1 personnel perform examinations under supervision; Level 2 personnel are qualified to set up, perform, and interpret examinations; Level 3 personnel have full technical responsibility and can approve procedures and certify Level 1 and 2 staff. Inspection certificates and examination records must identify the examiner's name, certification number, certificate expiry date, and the specific technique used.

Summary

VT is the baseline requirement for all fabrication inspection. RT is the standard volumetric method for butt welds in plate and pipe where access allows and radiation controls are manageable. UT and PAUT are preferred where planar defects are the primary concern, access is limited, or a permanent digital record is required. MT is the go-to surface-breaking method for ferromagnetic materials. PT covers surface-breaking examination on non-magnetic materials and non-ferrous alloys. Specifying the right method for the application — and stating clearly in engineering documents the required technique, coverage, and acceptance standard — is as much a mechanical engineering responsibility as sizing the weld itself.