UT of Moment Connections

[pattimelt]

Another individual was talking about how in the course of his carreer (military & government connected) he had always had any welds involving moment connections UT. This got me thinking-I have a number of these connections in one of our structures, and had only done the VT. I have looked at the drawings that were issued for construction, and no notes from the P.E. AWS D1.8 is not a stand alone code, and it is not referenced in my scope. What say you all - is this a D1.1 requirement or is this something I can file as good to know for the next one?

 

[connectegr]

Shop or field welded CJP welds should all be UT tested. For flange welded moment connections the typical detail requires CJP welds.

http://www.FerrellEngineering.com

 

[pattimelt]

yes, but is it Code required? What chapter can I quote when I go back to the project manager? "Should be" is not going to make it so - if it is specified in D1.1 then I have a legitimate leg to stand on, not just an overcautious inspector.

 

[connectegr]

Most of our project specifications require 100% UT of CJP welds. I don't believe there is a specific "thou shall" in AWS D1.1. However the 2010 AISC Spec, Chapter N, has some additional requirements (p177).

In most cases a CJP weld is provided because the full strength of the joined member is required. Without UT and repair, discontinuities in the weld can significantly reduce the capacity. Discontinuities or poor fusion can propagate cracks which can reduce the expected strength by more than 70%. The is especially true in moment connection welds which are loaded transversely to the weld.

I am certainly against requiring CJP welds unless necessary. PJP welds and fillet welds are designed for an effective strength and include reductions for expected areas of incomplete fusion. However, if adequate visual inspection is not provided of fit-up, weld prep, workmanship, cleaning between passes, etc, UT testing may be necessary to verify adequate weld is provided.

My point is that this can be a dangerous area for short cuts.

http://www.FerrellEngineering.com

 

[gtaw]

You may want to investigate the applicable building code for additional requirements for CJP welds.

AWS D1.1 requires VT for structural welds as the base line extent of inspection. However, D1.1 is not a stand-alone document, other codes may apply depending on the end-use of the structure and they may invoke additional requirements.

I would quote chapter and verse, but different areas of the country may use different editions of the building code or a different building code.

Other considerations must be checked, such as the type of loading, static versus cyclic, seismic versus non-seismic loading, etc.

AWS D1.1 gives the Engineer wide latitude when making decisions on whether to invoke specific requirements or to allow exceptions to the requirements of D1.1. For instance, the Engineer can accept welders qualified to different welding standards or WPSs that were qualified to different standards. They can allow non-listed base metals to be included as prequalified and exempt from testing. They can even permit alternative procedures and acceptance criteria to be used for the purpose of inspection.


Best regards - Al

 

[Duwe6]

The Engineer-of-Record is expected to provide the minimum NDT requirements under the D1.x coces. Thus no 'automatic' UT of CJP welds, even splices of main beams.

A prudent E-of-R requires 100% UT of splices and at least 10% of the remainder of CJP welds.

But I've been on jobs where there was nothing in the spec or on the dwgs -- E-of-R ASSumed that an untested CJP weld [Visual only] would give 100% joint strenght, 100% of the time. Customer listened to reason, we ended up having 10% UT by the fab shop [their standard, to keep their welders 'honest'] , another 10% in the field.

 

[pattimelt]

Thanks to all I have sent drawings to the EoR and asked for him to review.