Weld Procedure Qualification (PQR): Mechanical Testing and ISO 15614-1

Weld procedure qualification under ISO 15614-1 is the formal process by which a manufacturer demonstrates that a proposed welding procedure will consistently produce welds with the mechanical properties and soundness required by the design standard. This article explains every stage of the qualification sequence — from drafting the preliminary WPS through to issuing the qualified WPS — with particular focus on essential variables, test weld geometry, required mechanical tests, acceptance criteria, and ranges of approval.

The Qualification Framework: WPS, pWPS, and WPQR

ISO 15614-1:2017 (superseding the 2004 edition) specifies the requirements for the qualification of welding procedures for metallic materials by testing. It forms part of a broader family of standards: ISO 15607 defines the general framework, ISO 15608 provides the material grouping scheme, and ISO 15610 through 15614 address specific qualification routes. For steel and nickel alloys welded by arc and gas processes, ISO 15614-1 is the primary document.

The three documents that underpin any qualification are distinct in purpose:

• pWPS (preliminary Welding Procedure Specification) — the proposed parameter set used to deposit the test weld. It covers all welding conditions but is not yet qualified for production use.
• WPQR (Welding Procedure Qualification Record) — the record of actual parameters measured during the test weld, together with all NDE and destructive test results. Once accepted, this document becomes the evidence of qualification.
• WPS (Welding Procedure Specification) — the production instruction issued to welders, derived from the WPQR, specifying parameters within the qualified ranges of approval.

The standard distinguishes between arc welding processes covered under Part 1 (SMAW, GMAW/FCAW, GTAW, SAW, PAW, ESW/EGW) and other processes addressed in Parts 2 through 13 of the ISO 15614 series.

ISO/TR 15608 Material Grouping

The material group system is central to ISO 15614-1 because it determines the transferability of qualification — a WPQR performed on a material within one group may or may not qualify welding of materials in adjacent groups. The principal groups relevant to ferrous materials are:

A qualification on Group 1.2 material does not automatically qualify welding of Group 1.3; a qualification on Group 2 material qualifies Groups 1 and 2 but not Group 3. Table 4 of ISO 15614-1 gives the full transferability matrix.

Essential Variables and Their Significance

An essential variable is any parameter whose change outside defined limits alters the metallurgical character of the weld or HAZ in a manner that cannot be predicted from the existing test data. ISO 15614-1 Clause 8 lists the essential variables applicable to each welding process. A change in an essential variable beyond its range of approval voids the existing WPQR and requires a new test weld to be qualified.

Process-Independent Essential Variables

Heat Input Calculation

Heat input is one of the most consequential essential variables because it governs HAZ grain growth, peak temperature, and cooling rate through the martensite start temperature, all of which determine HAZ toughness and hardness. The arc energy Q_arc and net (corrected) heat input Q_net are:

Q_arc = (U × I) / v        [kJ/mm]

where:
  U = arc voltage (V)
  I = welding current (A)
  v = travel speed (mm/s)

Q_net = Q_arc × η

Thermal efficiency factors η:
  SMAW   : η = 0.80
  GMAW   : η = 0.80
  GTAW   : η = 0.60
  SAW    : η = 1.00
  FCAW-G : η = 0.80

Test Weld Assembly and Execution

Test Piece Dimensions

The test weld geometry must be representative of the intended production joint. ISO 15614-1 Clause 7.3 specifies minimum test piece dimensions:

• Butt weld in plate: minimum length 350 mm (or 150 mm either side of weld); width such that full specimens can be extracted after weld quality checks remove end zones of 25 mm each.
• Butt weld in pipe: full circumference or at minimum a quarter-circumference arc, ensuring at least one face-bend, one root-bend, one tensile, and one macro specimen can be extracted.
• Fillet weld (Clause 8.3 qualification): T-joint assembly minimum 300 mm long; specimen extraction for macro and hardness only.

Welding Conditions

The test weld must be performed in the same manner as intended for production. Conditions that must be recorded and reproduced within the essential variable tolerances include:

• Joint preparation geometry (included angle, root face, root gap) within ±2° and ±0.5 mm of nominal
• Base material grade, heat, and batch reference (for traceability)
• Filler metal trade name, classification, and batch certificate
• Preheat temperature measured by contact thermocouple or Tempilstik at least 75 mm from the joint line
• Interpass temperature measured with calibrated contact pyrometer immediately before each pass deposition
• Travel speed measured or derived from weld length divided by arc time per pass
• Current, voltage, and polarity recorded per pass

Non-Destructive Examination of the Test Weld

Before destructive specimens are cut, the test weld must pass NDE to ISO 5817:2014 Level B acceptance criteria (Level C for specific fillet weld qualifications). The required NDE methods are:

NDE is typically completed at least 16 hours after welding for ferritic steels susceptible to hydrogen cracking (Groups 2, 3, 4, 5, 6), and 24 hours for hydrogen-sensitive steels welded with high hydrogen potential processes. Phased array UT (PAUT) per ISO 13588 is an acceptable alternative to conventional UT provided the technique is validated for the weld geometry.

See the hydrogen-induced cracking article for the metallurgical basis of delayed hydrogen cracking and its relevance to NDE timing.

Destructive Test Requirements and Specimen Extraction

Once NDE is passed, a cutting plan is prepared. End zones (25 mm each end) are discarded. The destructive test specimens are extracted in the sequence shown in Figure 2. Table 1 of ISO 15614-1 specifies the required number of each specimen type as a function of test piece thickness.

Transverse Tensile Test

Two transverse tensile specimens are required. Specimens are machined to remove the weld cap flush with the plate surface; the weld root reinforcement is also removed. The gauge length spans the full weld width plus HAZ. Tests are conducted to ISO 4136 at ambient temperature (23 ±5 °C).

Acceptance criterion: Tensile strength R_m of the specimen must be ≥ the minimum specified R_m of the parent material. If the weld metal has a specified strength exceeding the parent material (overmatching filler), fracture in the weld metal is acceptable at any strength. Fracture at the fusion boundary below the parent material minimum is a failure.

Bend Test

Bend tests assess ductility and detect planar flaws in the weld cross-section. For t < 12 mm, two face bends and two root bends are required; for t ≥ 12 mm, four side bends replace face and root bends (ISO 5173). The mandrel diameter is specified in ISO 15614-1 Table 4a as a function of elongation A and thickness t:

Mandrel diameter d = 100t / A − t

where:
  t = specimen thickness (mm)
  A = minimum elongation (%) from parent material specification

For A ≥ 20%  (e.g., structural steels): d ≈ 4t typical
For A < 20%  (e.g., high-strength steel): d is larger per formula

Acceptance criterion: No single imperfection exceeding 3 mm in any direction, measured after bending, visible on the outer surface of the bent specimen. Total cumulative length of cracks <6 mm. Corner cracks up to 6 mm are not cause for rejection unless they originate from a visible slag inclusion or lack of fusion.

Macro-Section Examination

One macro-section cut transversely through the weld and prepared to a 1µm finish, etched with 2–5% Nital (for steels), and examined at magnification 1× to 10×. The section must reveal:

• Full fusion to both sidewalls and inter-run boundaries
• Complete root penetration or root fusion (for one-sided welds)
• Absence of cracks in weld metal, HAZ, and parent metal
• Absence of inter-run porosity exceeding ISO 5817 Level B limits
• Correct pass sequence and number of runs consistent with the pWPS

Macro-section acceptance is to ISO 5817 Level B. The macro also provides confirmation of the actual throat thickness for fillet welds and the actual weld bead geometry for correlation with the hardness traverse.

Hardness Survey (HV10)

A Vickers hardness survey (10 kg load, HV10) is performed on the macro section per ISO 9015-1. Three traverse lines are specified:

• Top traverse: approximately 2 mm below the weld cap surface
• Middle traverse: at mid-thickness (for t ≥ 20 mm)
• Root traverse: approximately 2 mm above the root

Each traverse covers parent metal (at least 4 mm from fusion boundary), HAZ at 0.5 mm, 1 mm, 2 mm, 5 mm from the fusion line, weld centre-line, and the mirror positions on the opposite HAZ and parent metal.

Acceptance criterion: Maximum 380 HV10 in HAZ for steels in Groups 1–4, in the absence of a specific material standard requirement. Groups 5 and 6 (Cr-Mo and P91/P92) require PWHT, and hardness limits post-PWHT are set by the relevant material standard (typically ≤250–265 HV for P91 after PWHT at 745–760 °C).

The hardness testing methods article provides a comprehensive background on the Vickers test principle, indenter geometry, and conversion factors relevant to this application.

Charpy V-Notch Impact Testing

Charpy testing is required when the base material standard mandates impact properties, or when the design code specifies a minimum absorbed energy requirement. ISO 15614-1 specifies testing to ISO 148-1 at the temperature required by the material specification or design code.

Notch positions and specimen orientations:

Acceptance criterion: Mean of three specimens ≥ the minimum specified energy. No individual value below 70% of the minimum (i.e., ≥0.7 × minimum); one value may be between 70% and 100% of the minimum provided the mean is met. Specimens are transverse to the weld direction and positioned to ensure the notch root samples the intended microstructural zone.

For the metallurgical interpretation of HAZ toughness, see the Charpy impact testing guide and the HAZ microstructure article covering coarse-grained HAZ grain growth and its effect on impact energy.

Ranges of Approval

The range of approval defines the production welding conditions that are legitimately covered by a given WPQR. ISO 15614-1 Clauses 8 and 9 define ranges of approval for each essential variable. Key ranges for commonly encountered variables are:

Thickness Range for Multi-Process Qualifications

When a test weld is deposited using two processes (e.g., GTAW root + SMAW fill), each process is qualified only for the thickness it deposited in the test. If the GTAW root run deposited 4 mm and the SMAW fill deposited 16 mm, the GTAW process is qualified for root applications up to 2 × 4 = 8 mm, while the SMAW is qualified for fill passes 0.5 × 16 = 8 mm to 2 × 16 = 32 mm. This distinction is critical for complex joint configurations on heavy wall piping.

Post-Weld Heat Treatment (PWHT) and Its Role in Qualification

PWHT is an essential variable. A WPQR qualified with PWHT does not cover production welding without PWHT, and vice versa. This is because PWHT fundamentally changes the microstructure of both the weld metal and HAZ — tempering martensite, precipitating carbides, reducing residual stress, and altering hardness and toughness profiles. For hydrogen-susceptible steels (Groups 3, 4, 5, 6), PWHT is not optional and must be part of the qualification record.

The PWHT parameters recorded on the WPQR — temperature range, holding time, heating and cooling rates — define the qualification. Changes in temperature of more than ±15 °C or holding time of more than −25% require re-qualification. This tight control is particularly important for P91 (Group 6) steels, where the tempered martensite microstructure and creep rupture properties are critically sensitive to PWHT conditions.

Refer to the quenching and tempering article for the metallurgical principles of tempering as they apply to PWHT of alloy steels, and to the martensite formation article for the microstructural basis of HAZ hardness.

Common Qualification Failures and Corrective Actions

Transverse Tensile Failure at the Fusion Boundary

The most common tensile failure mode in procedure qualification is fracture at the fusion boundary with strength below the parent material minimum. Root causes include: lack of fusion arising from incorrect arc gap or travel speed; excessive dilution of low-strength filler with high-strength parent metal at the root; or weld metal hydrogen embrittlement in inadequately preheated joints. Corrective action typically involves revising the pWPS to increase preheat temperature, reduce travel speed, or change to a higher-strength matching filler.

Excessive HAZ Hardness

HAZ hardness exceeding 380 HV10 in carbon and low-alloy steels (Groups 1–4) indicates an as-deposited martensite-dominated HAZ microstructure. This arises from rapid cooling through the martensite transformation range — a consequence of low heat input, low preheat, or high carbon equivalent material. Carbon equivalent (IIW formula) predicts susceptibility:

CE(IIW) = %C + %Mn/6 + (%Cr + %Mo + %V)/5 + (%Ni + %Cu)/15

CE < 0.40 : low susceptibility, preheat generally not required
CE 0.40–0.50 : moderate susceptibility, preheat 50–100°C typical
CE > 0.50 : high susceptibility, preheat >100°C, potentially PWHT

Corrective actions: increase preheat, increase heat input within the allowable range, add PWHT to the procedure. See hydrogen-induced cracking for the relationship between HAZ hardness and cold cracking risk.

Charpy Impact Failure in the Coarse-Grained HAZ

The coarse-grained HAZ (CGHAZ) consistently presents the lowest Charpy values in the qualification test. The CGHAZ undergoes grain coarsening above approximately 1100 °C; in multi-pass welds, local reheating can produce a partially retransformed zone between two adjacent beads where a mixture of MA constituents (martensite-austenite islands) forms, further degrading toughness. Corrective actions include: reducing peak heat input to limit grain growth; using fine-grain normalised parent material; specifying controlled interpass temperature to promote fine-grained microstructure between passes.

For a full treatment of the microstructural zones in the HAZ and their properties, see the HAZ microstructure article.

Bend Test Cracks from Lack of Fusion

Lack of fusion defects aligned parallel to the weld axis, invisible to RT in some orientations, are readily detected by bending. A crack >3 mm originating at a fusion boundary or an entrapped slag stringer constitutes failure. Corrective pWPS changes: increase arc energy, widen joint preparation angle (reduce sidewall angle), improve interpass cleaning.

Qualification Under EN 15614-1 vs. ASME Section IX

ISO 15614-1 and ASME BPVC Section IX are the two most widely encountered weld procedure qualification standards globally. While their underlying philosophy is similar — qualify the procedure by testing, define essential variables, specify ranges of approval — there are significant differences that prevent direct cross-acceptance:

Many fabricators operating across both European and North American markets maintain dual qualification sets — one per ISO 15614-1, one per ASME Section IX — for the same welding procedure. Some third-party inspection bodies accept cross-qualification where the more conservative test program covers both standards’ requirements, but this must be verified case by case with the relevant notified body or authorised inspection agency.

Industrial Applications and Standards Context

ISO 15614-1 is invoked by — or directly incorporated into — a wide range of product standards, design codes, and customer requirements across pressure equipment, structural steel, offshore, pipeline, and nuclear sectors:

Understanding the requirements of the specific design standard — not merely ISO 15614-1 itself — is essential before commencing qualification. Design codes frequently impose additional essential variables, lower hardness limits, stricter impact requirements, or mandatory PWHT that are not required by ISO 15614-1 in isolation.

Recommended References

Disclosure: MetallurgyZone participates in the Amazon Associates programme. If you purchase through these links, we may earn a small commission at no extra cost to you. This helps support free technical content on this site.

Further Reading