J7.1 Visual Welding Inspection Documentation
Visual inspection by a qualified quality control and quality assurance inspector(s) prior to, during, and after welding is emphasized as the primary method used to evaluate the conformance of welded joints to the applicable quality requirements. Joints are examined prior to the commencement of welding to check fit-up, preparation of bevels, gaps, alignment, and other variables. During welding, adherence to the welding procedure specifications (WPS) is maintained. After the joint is welded, it is then visually inspected to the requirements of AWS D1.1/D1.1M and the requirements of this provision.
Specification Section N5.4 Commentary on welding inspection contains extensive discussion regarding the observation of welding operations, including the determination of suitable intervals for performing such inspections. Welds in the seismic force-resisting system (SFRS) should be considered for higher levels of observation, compared to welds not in the SFRS and addressed by Specification Chapter N. Welds designated demand critical within the SFRS should be considered as warranting higher levels of observation, compared to other welds not designated demand critical within the SFRS.
J7.2 NDT of Welded Joints
The use of nondestructive testing methods as required by this section is recommended to verify the soundness of welds that are subject to tensile loads as a part of the SFRS, or to verify that certain critical elements do not contain significant notches that could cause failure. Ultrasonic testing (UT) is capable of detecting embedded flaws in groove welds in all standard welded joint configurations. UT is not suitable for inspecting most fillet welds and partial-joint-penetration (PJP) groove welds. Magnetic particle testing (MT) is capable of detecting flaws on or near the surface of all types of welds and is used for the surface examination of critical groove welds. The use of penetrant testing (PT) is not recommended for general weld inspection but may be used for crack detection in specific locations, such as weld access holes, or for the location of crack tips for cracks detected visually.
J7.2a CJP Groove Weld NDT
UT is used to detect embedded flaws in groove welds, but it may not necessarily be capable of detecting surface or near-surface flaws. MT is used to detect flaws on or near the surface of these welds. Because visual inspection is also implemented for all complete-joint-penetration (CJP) groove welds, thus detecting the most serious surface defects, MT is performed at a rate of 25%.
J7.2b PJP Groove Weld NDT
When UT is specified for PJP groove welds, the Specification requires that UT technicians be qualified in accordance with AWS D1.8/D1.8M (AWS, 2021) using weld joint mock-ups incorporating PJP groove welds. This training is to be performed and documented in accordance with ANSI/ASNT SNT-TC-1A (ASNT, 2020a) or Standard for Qualification and Certification of Nondestructive Testing Personnel (ANSI/ ASNT CP-189) (ASNT, 2020b).
It is difficult to properly interpret UT signals generated from the root areas of PJP groove welds. The geometry of the unfused root lends itself to mode conversions of the UT beam that renders the evaluation of this area extremely difficult and often leads to nonrelevant indications. Signals from the root region are considered “false indications” and are not to be used as the basis for rejection of the PJP groove weld.
Except for the column splice welds and the column to base plate welds listed in this section, all PJP groove welds are exempt from NDT requirements. The Specification applies a 0.6 reduction factor to the available strength of PJP groove welds subjected to tension in lieu of UT inspection. However, the prescriptive column splice detail in Sections E3.6g.2 and E3.6g.4—as well as column-to-base plate welds utilizing PJP groove welds permitted for intermediate moment frames, special moment frames, and special truss moment frames—will subject the welds to demands in excess of what is permitted by the Specification, and the consequence of failure of these welds, designated as demand critical, would be essentially identical whether designated as a CJP or PJP groove weld.
To provide confidence that the design intent of the joint is met without performing UT, the following process may be an acceptable method for verification.
- (1) Prior to welding, joint preparation should be visually inspected. The depth of bevel should be measured and documented.
- (2) After the root pass is deposited, it should be cleaned, and visual testing (VT) should be performed. Supplemental light may be necessary for VT of the root pass in deeper bevels or in low light conditions. The supplemental light should not be only pointed directly at the face of the root weld. The light should also be placed at an approximate angle of 30° to provide a shadow of deficiencies that may be present. The light should be directed at both toes of the root weld.
- (3) After VT acceptance, MT should be performed on the root pass in accordance with AWS D1.8/D1.8M, Annex G. For indications open to the surface, MT may be supplemented by penetrant testing (PT). Indications determined to be cracks or incomplete fusion should be unacceptable, regardless of length.
- (4) Each subsequent weld layer should be examined by VT. At agreed upon depths of the weld, VT and MT should also be performed like for the root pass. After completion of the weld, VT and MT should be performed in accordance with AWS D1.8/D1.8M, Annex G. For indications open to the surface, MT may be supplemented by penetrant testing. Indications determined to be cracks or incomplete fusion should be unacceptable, regardless of length.
J7.2c Base Metal NDT for Lamellar Tearing and Laminations
Lamellar tearing is the separation and tearing of base metal along planes parallel to a rolled surface of a member. The tearing is the result of decohesion of “weak planes,” usually associated with elongated “stringer” type inclusions, from the shrinkage of large weld metal deposits under conditions of high restraint, applying stress in the through-thickness direction of the base metal.
Lamellar tears rarely occur when the weld size is less than approximately 3/4 to 1 in. (19 to 25 mm). Typically, inclusions located deeper from the surface than t/4 do not contribute to lamellar tearing susceptibility.
An appropriate criterion for laminations in SFRS connections does not exist in current standards. Although AWS D1.1/D1.1M, Table 8.2, criteria have been written and are applicable to weld metal, not base metal, the use of Table 8.2 criteria has been deliberately selected as conservative acceptance criteria for laminations in these applications, immediately adjacent to and behind the weld.
J7.2d Beam Cope and Weld Access Hole NDT
The stress flow near and around weld access holes is very complex, and the stress levels are very high. Notches serve as sources of stress concentrations, locally amplifying this stress level, which can lead to cracking. The surface of the weld access hole must be smooth and free from significant surface defects. Both PT and MT are capable of detecting unacceptable surface cracks.
J7.2e Reduced Beam Section Repair NDT
Because plastic straining and hinging, and potentially buckling, take place in the thermally cut area of the reduced beam section, the area must be free of significant notches and cracks that would serve as sources of stress concentrations and crack initiation sites. Inadvertent notches from thermal cutting, if sharp, may not be completely removed if relying solely upon visual inspection. If a welded repair is made, NDT is performed to verify that no surface or subsurface cracks have been caused by the repair.
J7.2f Weld Tab Removal Sites
Weld tabs are removed because they serve as locations for the starting and stopping of welds and are, therefore, likely to contain a number of weld discontinuities. So that no significant discontinuities present in the tab extend undetected into the finished weld itself, MT is performed at the exposed face of the weld tab. Any weld end discontinuities would be present at the surface of the joint, and therefore would be more detrimental to performance than an embedded discontinuity.
J7.2h Reduction of Percentage of Magnetic Particle Testing
A completed weld is defined as a weld with a defined stop and start; refer to Specification Section N5.5d for a description of how CJP welds over 3 ft (1 m) in length are considered when determining the rejection rate. For a column splice with two CJP flange welds and a CJP web weld, three welds are involved.