NDT reports in a data book

Non-destructive testing happens on the shop floor in minutes. But in a data book, it is not the test that counts — it is the report. Years after handover, when someone asks whether weld 142 was radiographed and passed, the only thing that answers is a signed, traceable NDT report filed against that weld. If the report is missing, ambiguous, or signed by an uncertified technician, the proof fails — regardless of how good the weld was.

This is why the NDT chapter is one of the most heavily scrutinised parts of any Manufacturing Record Book. A reviewer is not re-doing the examination; they are checking that every required examination was done, recorded correctly, and accepted against the right standard.

Which method applies to a given joint is set by the code and the client specification — often a combination, such as 100% VT plus a percentage of RT or UT on butt welds, and MT or PT on fillet welds.

Visual testing (VT) is the foundation — every weld is visually examined for profile, undercut, misalignment and visible defects, and the result is recorded. Liquid penetrant testing (PT) draws a dyed penetrant into surface-breaking defects, then a developer pulls it back out to make cracks and porosity visible; it works on any non-porous material. Magnetic particle testing (MT) magnetises a ferromagnetic part and applies iron particles that gather at flux leakage from surface and near-surface defects. MT is faster than PT on suitable materials but only works on ferromagnetic steels.

Radiographic testing (RT) passes X-rays or gamma rays through the weld onto film or a digital detector, producing a permanent image in which internal porosity, slag and lack of fusion appear as density changes. Its strength is the permanent, reviewable record; its limitations are access (you need both sides) and safety (radiation control).

Ultrasonic testing (UT) sends high-frequency sound into the material and reads the echoes to locate and size internal flaws. Modern phased-array UT (PAUT) and time-of-flight diffraction (ToFD) increasingly replace RT on thick sections because they need access from only one side and avoid radiation. UT excels at detecting planar defects like cracks and lack of fusion that RT can miss.

Beyond the five core methods, the NDT chapter often carries supporting examinations: positive material identification (PMI) to confirm the alloy of installed material, hardness testing (especially after PWHT), ferrite measurement on duplex and stainless welds, and the post-weld heat treatment (PWHT) charts that prove the heat-treatment cycle met the procedure. These are not always called "NDT" strictly, but they live alongside it as part of the examination evidence.

A report that will survive review identifies, at minimum:

• the item or weld number examined, traceable to the weld map;
• the method and written procedure reference;
• the equipment, consumables and technique (e.g. film type, probe angle, penetrant batch);
• the extent of examination (spot, 10%, 100%);
• the acceptance standard and level applied;
• the result — accept or reject, with defect type, size and location for any indication;
• the technician's name, certification level and signature, and the date.

A common review failure is a report that records a result but omits the acceptance standard, or is signed by a technician whose certificate for that method is missing from the book.

The acceptance criteria come from the construction code, not the NDT method. For pressure equipment that is typically ASME BPVC Section VIII with Section V for the methods, or EN 13445 / ISO 5817 in Europe; for process piping, ASME B31.3. The same indication can be acceptable under one code and rejectable under another, so the report must state which standard and quality level were applied.

Personnel are certified by method and level under either ISO 9712 (central third-party certification) or SNT-TC-1A (employer-based, ASNT). Level II is the usual minimum for interpreting and signing results. The certificates themselves belong in the documentation package.

NDT does not stand alone — it is the verification layer over the welding records. A reviewer takes the weld map and, for every weld number, confirms there is an NDT report with an acceptable result, to the correct procedure, by a certified technician. Because the weld map also links each weld to a welder, a recurring defect can be traced back to a specific welder or procedure. That chain — materials to welds to NDT — is the backbone of traceability, and it is exactly what a third-party inspector follows when accepting a Manufacturing Data Book.