How much NDT a weld actually needs, and where the extent comes from
Last reviewed
Short answer
Under AS/NZS 1554.1 only visual scanning is applied to 100 per cent of welds. Everything else is a sample: Table 7.4 suggests visual examination of 5 to 25 per cent for GP welds and 10 to 50 per cent for SP, with radiography or ultrasonics at nil for GP and 0 to 10 per cent for SP, and volumetric examination of SP butt welds only applies where the principal requires it and specifies it on the drawings. For pressure equipment the extent is tabulated directly against construction class in AS 4037, running from no volumetric requirement for Class 2b and Class 3 up to 100 per cent for Class 1, 1h and 2ha. The actual contracted extent is a matter to be resolved between principal and fabricator before work starts.
Key numbers
- Visual scanning
- 100 per cent of welds, GP and SP (Clause 7.3)
- Visual examination
- GP 5 to 25 per cent, SP 10 to 50 per cent (Table 7.4)
- Magnetic particle or penetrant
- GP 0 to 2 per cent, SP 0 to 10 per cent (Table 7.4)
- Radiography or ultrasonics
- GP nil, SP 0 to 10 per cent (Table 7.4)
- AS 4037 Class 1, 1h, 2ha
- 100 per cent volumetric
- AS 4037 Class 2a
- 2 per cent for automatic SAW, 10 per cent for other processes
- AS 4037 Class 2hb
- 10 per cent for automatic SAW, 25 per cent for other processes
Extent is a function of three inputs
Asking "how much NDT does this weld need" without those three inputs cannot be answered. The extent falls out of the governing standard, the category or class assigned by the designer, and the material family. Change any one and the answer changes.
There is a fourth input that people forget: the contract. AS/NZS 1554.1 lists the type and extent of inspection, including the requirement for ultrasonic examination for transverse cracks, as a matter to be resolved between principal and fabricator before work commences.
- Governing standard: AS/NZS 1554.1 and 1554.5 for structural; AS 4037 for pressure equipment examination extents
- Weld category or construction class: SP, GP or FP for structural; Class 1 through 3 for pressure
- Material family: ferritic versus austenitic changes the surface method
Structural: two levels of visual, then a sample
The most misquoted line in Australian structural welding is "AS/NZS 1554.1 requires 100 per cent visual inspection". What the standard actually defines is two different visual activities, and only the first is applied to everything.
Visual scanning under Clause 7.3 covers 100 per cent of welds. Visual examination against Table 6.2.2 is a sample from Table 7.4. Volumetric examination of SP butt welds only applies where the principal requires it and specifies it on the drawings (Clause 6.2.1).
- Visual scanning (Clause 7.3)
- 100 per cent, GP and SP
- Visual examination to Table 6.2.2
- GP 5 to 25 per cent, SP 10 to 50 per cent
- Magnetic particle or liquid penetrant
- GP 0 to 2 per cent, SP 0 to 10 per cent
- Radiography or ultrasonics to Table 6.2.1
- GP nil, SP 0 to 10 per cent
- 100 per cent visual examination recommended
- SP welds complying with NZS 3404.1, part of a seismic-resisting system, or subject to fatigue loading
- AS/NZS 1554.6 stainless
- Penetrant preferred over magnetic particle; ultrasonic testing to ISO 22825 rather than AS 2207 alone
Sampling is a program, not a percentage
Where less than 100 per cent examination is required, Table 7.4 Note 2 describes a program the principal draws up and the design engineer approves. It is front-loaded and self-adjusting rather than a flat percentage across the job.
- Start with full testing of the first major component, or 5 per cent of welds, as appropriate
- Progressively reduce the frequency of testing as each test demonstrates compliance
- On non-compliance, test the next 5 per cent of welds or the next major component consecutively
- For radiographic spot examination, a failed spot triggers two more spots of the same length: if both pass, repair and re-radiograph only the original; if either fails, the entire weld is radiographed or replaced
Pressure equipment: AS 4037 construction classes
AS 4037 tabulates volumetric extent against construction class. Class 2b and Class 3 carry no volumetric requirement from the standard, but visual inspection remains at 100%. This is an orientation table, not a full pressure NDT syllabus.
- Class 1, 1h, 2ha
- 100% volumetric
- Class 2a
- 2% for automatic submerged arc welding, 10% for other processes (spot)
- Class 2hb
- 10% for automatic submerged arc welding, 25% for other processes
- Class 2b and Class 3
- Volumetric NDE not required by the standard. Visual inspection still 100%
Choosing the method: RT, UT, MT, PT or VT
Volumetric and surface methods answer different questions. Magnetic particle testing needs a ferromagnetic material, so it is not an option on austenitic stainless. Where both are available on ferritic material, AS/NZS 1554.1 prefers magnetic particle where convenient.
Clause 6.2.1 treats radiography and ultrasonics as alternatives selected by the principal, not as a thickness-gated hierarchy. What Clause 6.4.2 ties to thickness is the ultrasonic evaluation level: Level 2 to AS 2207 up to 50 mm, Level 3 above 50 mm.
- VT (visual): scanning on 100 per cent, examination on a sample
- RT (radiographic): volumetric, good on porosity and slag. AS 2177
- UT (ultrasonic): volumetric, strong on planar defects. AS 2207 for ferritic, ISO 22825 for austenitic and dissimilar welds
- MT (magnetic particle): surface and near surface, ferritic only
- PT (penetrant): surface breaking only, used where MT is unavailable. AS 2062
What the ForgeQM NDT checker returns
The checker takes the standard, weld category or construction class, joint and material, and returns the required extent with the clause it came from and a status of mandatory, conditional or not appropriate.
It does not invent a minimum thickness ban on ultrasonics for production examination. Under AS/NZS 1554.1, Clause 6.2.1 lets the principal choose RT or UT as appropriate, and Clause 6.4.2 sets evaluation level by thickness rather than forbidding UT on thin butts.
Acceptance criteria themselves are reported as the applicable clause rather than computed as an automated pass or fail. Interpretation of NDT results stays with a qualified inspector.
Frequently asked questions
Does AS/NZS 1554.1 require 100 per cent visual inspection?
It requires 100 per cent visual scanning, which is not the same thing. Scanning under Clause 7.3 confirms that every weld called for on the drawings exists and looks for gross defects, and it applies to all welds in both categories. Detailed visual examination against the Table 6.2.2 quality levels is a sample: Table 7.4 suggests 5 to 25 per cent for GP and 10 to 50 per cent for SP, rising to a recommended 100 per cent for SP welds that are fatigue loaded, seismic-resisting, or comply with NZS 3404.1.
Is ultrasonic testing allowed instead of radiography under AS/NZS 1554.1?
Yes. Clause 6.2.1 treats radiographic and ultrasonic examination as alternatives for Category SP butt welds, applied "as appropriate" where the principal requires them, so the standard does not gate ultrasonics behind a minimum thickness. What does change with thickness is the evaluation level: Clause 6.4.2 requires Level 2 evaluation to AS 2207 up to 50 mm and Level 3 above 50 mm. On austenitic stainless the ultrasonic procedure follows ISO 22825 because of the coarse, anisotropic grain structure.
Why is magnetic particle testing not used on stainless steel?
Austenitic stainless steel is not ferromagnetic, so it cannot hold the magnetic field that magnetic particle testing relies on to reveal leakage flux at a defect. Penetrant testing is used instead for surface breaking defects, which is what AS/NZS 1554.6 directs.
What does 10 per cent NDT actually mean in practice?
Under AS/NZS 1554.1 it means a program, not a flat percentage. Table 7.4 Note 2 says the principal draws up a testing program, approved by the design engineer, that begins with full testing of the first major component or 5 per cent of welds, then progressively reduces frequency as compliance is demonstrated, and escalates to the next 5 per cent or the next major component consecutively if testing shows non-compliance. Confirm the basis in the project ITP rather than inferring it from the percentage alone.
Sources
This guide is written against the published standards below. Standards are revised, and project specifications override general guidance, so confirm clause numbers against the edition your contract names before relying on them.
- AS/NZS 1554.1:2011 Structural steel welding, Part 1: Welding of steel structuresClause 6.2.1 methods of inspection of completed welds, Clause 6.3 radiographic examination and spot escalation, Clause 6.4 ultrasonic examination including the Clause 6.4.2 evaluation levels, Clause 7.3 visual scanning, Table 7.4 suggested extent of non-destructive examination and its notes.
- AS 4037 Pressure equipment, Examination and testingTable 7.1 volumetric extents by construction class, Table 7.2 crack sensitive material groups, Table 8.1 planar imperfection limits.
- AS/NZS 1554.5 Structural steel welding, Part 5: Welding of steel structures subject to high levels of fatigue loadingTable 7.4 inspection regime for category FP welds.
- AS 2207, AS 2177, AS 2062 and ISO 22825Ultrasonic testing of ferritic welds, radiographic testing, penetrant testing, and ultrasonic testing of austenitic and dissimilar welds respectively.
Related guides
- AS/NZS 1554 Weld Categories: SP vs GP vs FPSP and GP are weld categories in AS/NZS 1554.1, and FP belongs to AS/NZS 1554.5. What each category changes about qualification, workmanship, inspection extent and acceptance, and which part of the 1554 series applies to your material.
- MDR and ITP: Building a Fabrication Handover PackAn ITP plans the inspection, an MDR proves it happened. What a manufacturing data record contains, how it maps back to the ITP hold and witness points, and why compiling it at the end of a job is what makes handover painful.
- Welding Qualification Ranges ExplainedA qualification test on one plate qualifies a band of production work. How welder thickness, position, joint type, process and material group ranges are derived, where they differ from procedure ranges, and what triggers reapproval.
Stop checking this by hand
ForgeQM applies the AS/NZS 1554 series as checks while you write the record, not after. Qualification ranges are generated from the PQR, NDT extents come back with the clause they came from, and job data carries across every document through to the MDR.
ForgeQM is in a controlled rollout. Request a scoped trial or contact sales for an Organisation discussion.