AWS D1.1 Endorsement study materials
AWS D1.1 Endorsement study resources

Film-Clean but Not Fusion-Clean

AWS D1.1 Endorsement · Put your knowledge to work in context.

All case studies

Film-Clean but Not Fusion-Clean

A fabricator is completing 1 in thick flange-to-web CJP groove welds in a transfer girder, welded by FCAW under a qualified WPS. The contract documents and the EOR's notes require 100% ultrasonic examination of the CJP welds. On Friday the fabricator's radiographic subcontractor shot film on nine of eleven welds, reported them acceptable, and shipped two members. On Monday the owner's UT technician, certified to Level II, reports reflectors in the girder's first weld that he classifies as lack-of-fusion type. That same weld was the one reported acceptable by radiography.

1. Is the radiographic examination an acceptable substitute for the specified ultrasonic examination? Give the reasoning, not just a verdict.

Read the worked answer

Two separate reasons say no.

Contractual/code-administrative. The examination method, extent, and acceptance criteria are set by the contract documents and the EOR's requirements. Substituting a different method is a deviation that requires written approval from the EOR (and typically the owner) before it is performed, not a unilateral choice by the fabricator's subcontractor. An unapproved method change is a documentation nonconformance on its own.

Technical. In a T-joint configuration, radiation is directed through the flange thickness toward the groove, and lack of fusion lying parallel to the fusion face - the classic planar defect in this joint - is poorly oriented for detection; the radiation path does not reliably open the defect to the film. A film-clean result on this geometry is weak evidence at best. Ultrasonic examination, by contrast, is chosen precisely because it responds to planar, fusion-face discontinuities.

So the radiographic report cannot clear the weld, and the owner's UT result is the governing examination.

2. How should the inspector handle the conflicting results and what is the correct disposition path for the affected weld?

Read the worked answer

The specified method governs. Steps:

  1. Document both results, the dates, the personnel, and the fact that RT was performed without approval.
  2. Treat the weld as containing rejectable indications until shown otherwise by the specified method; do not average or "split the difference" between the two reports.
  3. Require excavation of the indicated region to sound metal, using a method that confirms removal (MT or PT of the excavated area, or re-UT), then repair welding under a qualified WPS with qualified welders, then re-examination by UT.
  4. Extend examination: if one weld in a group shows lack of fusion, the inspector should ask the EOR whether the sampling should be expanded to the remaining welds and to the two shipped members, since the same welder, WPS, and technique produced them.
  5. Escalate the shipped-member question immediately; the inspector's report should not wait for the shop repair to be finished.

Where the acceptance criteria for the specific indication are borderline or loading-dependent, that judgment belongs to the EOR - but the inspector must put the question in writing rather than accept by default.

3. What must the inspector verify about the UT technician and the examination itself before relying on that report?

Read the worked answer

A UT report is only as good as the chain behind it. Verify:

  • Personnel certification. The technician should be certified to Level II in ultrasonic examination under a recognized practice (SNT-TC-1A-type written practice), with the certificate and the employer's written practice available; certification in a different method is not the same thing.
  • Procedure and technique. A written UT procedure appropriate to the joint geometry and thickness, with the scanning angles and coverage that can actually reach the fusion faces from the accessible surface.
  • Equipment and calibration. Current calibration records for the instrument, probes, and reference blocks, and calibration checks performed at the site on the day.
  • Report content. Weld identification traceable to the member, location and depth of indications, amplitude/length classification, the acceptance criteria applied, and the examiner's signature and date.

If any of these is missing, the inspector should not silently accept the result; the correct move is to require the documentation before the examination is credited - the same standard that was applied to reject the RT substitution.