ASME VIII Division 2 Design by analysis of welded joints

[JonLister]

Dear forum members,

I am trying to qualify a pressure containing clamp (see first attached image) using Design by Analysis in accordance with ASME VIII Division 2, and I am particularly interested in how the welded regions are treated by the code with regards to joint efficiency.

According to Part 5.1.1.2(b) of ASME VIII Division 2, welds must have a geometry that is listed in Part 4 (which this geometry is not, see second attached image), or they should be assessed in accordance with Part 5, however no specific considerations are specified in the code for how to qualify the weld. My suggested method for analysing the weld would be as follows (note that only visual inspection of the welds will take place):

Using Table UW-12 of ASME VIII Division 1 I would classify the upper weld as a single-welded butt joint without the use of a backing strip, resulting in a joint efficiency factor of 0.6. The lower weld I would classify as equivalent to a double full fillet lap joint, resulting in a joint efficiency of 0.55.

I would then run the FEA, using a factored down allowable stress of 0.6S and 0.55S respectively in the weld regions, for the plastic collapse, local failure and ratcheting check criteria (I am making separate justifications for not running a buckling and fatigue check).

My proposed method would be to use an elastic model to sasisfy the local failure and the ratcheting check, and a limit load analysis to perform the check for plastic collapse. However, since this one is a highly optimised design with low margins I consider I will probably need to switch to elastic plastic methods for the local stress check to get this one to pass.

Any thoughts on this proposed methodology would be very welcome. Best regards, Jonathan

 

[JonLister]

I forgot to add that, since the welds will only be visually inspected I am proposing to use the allowable stresses that would normally be used for an ASME VIII Division 1 analysis (i.e. lower allowables) - image of clamp referenced in first post on this thread included - see attached.

 

[TGS4]

In order to perform an evaluation according to ASME Section VIII, Division 2, Part 5, the components are required to satisfy the inspection requirements of Table 7.1. Specifically, examination groups 1a, 1b, 2a, and 2b are permitted for Part 5. Table 7.2 indicates the extent of inspection required for those inspection groups.

It is not permitted to modify the allowable stress basis to account for non-unity weld joint efficiency. There is no provision for this (and there will not be in the future).

Essentially, if you can't/won't do 100% examination (as described in Table 7.2) to ensure a weld joint efficiency of 1.0, then you can't do DBA.

 

[JonLister]

Dear TGS4,
Many thanks for the response. Apologies, I think I got my wires crossed a bit before, I am trying to qualify some regions of a component that fall outside the scope of Division 1, using Article U-2(g) of ASME VIII Division 1. I think I got a bit carried away in trying to qualify the whole component (including the welds) using design by analysis as well. Considering your response my suggestion for a more suitable approach would be as follows:

Use hand calculations to determine acceptability of the welded regions, while incorporating relevant welded joint efficiencies from Division 1.

Use design by analysis from Division 2 to determine acceptability in the non-welded regions of the clamp (using allowable stresses from Division 1). For the plastic collapse failure criteria I would be using limit load analysis, and I wouldn't expect failure at the welded regions. For the local failure criteria and ratcheting check I would only consider regions at a suitable distance away from the welds. As I mentioned I will not be assessing for buckling or fatigue.

I had hoped to be able to validate the whole component using design by analysis, but I see your point that due to the weld geometry and lack of volumetric inspection this would not be a reasonable approach.

Again, thank you for your answer.

Best regards, Jonathan

 

[TGS4]

For demonstrating Protection Against Local Failure, I would absolutely try my best to model the weld - it is in these types of locations where high triaxiality is found. But don't use the elastic method, as ASME PTB-1 states

Two issues that are apparent is the use of an elastic stress basis for a local criterion and the stress category that is used with this criterion. It is not apparent how pseudo elastic stresses, i.e. elastically calculated stresses that exceed the yield strength can be used to evaluate a local fracture strain of a ductile material with strain hardening. In addition, the type of stress used in the criterion (i.e. linearized or average values verse stress at a point) and stress category (i.e. primary, secondary and peak) needs to be resolved. Since local failure is the failure mode being evaluated, the type of stress and stress category used in VIII-3 would appear to more correct. For ductile materials, a local criterion based on elastic analysis may not meaningful and the elastic-plastic method that follows is recommended for all applications.

Since you're going to need to perform an elastic-plastic analysis for Local Failure, then you might as well use elastic-plastic for the Plastic Collapse, as well. I would recommend most engineers away from the Limit Load method.

And include the welds in your ratcheting check, too. The only place where the weld joint efficiency comes into the equation is in plastic collapse, anyways (reference the handling of weld joint efficiencies in API 579-1/ASME FFS-1 (2016)).

Since you're designing to VIII-1, using VIII-1 allowable stresses, make sure to use the 2017 Edition of VIII-2. Table 5.5 is now expressed in terms of beta (which is the margin against the ultimate strength).

 

[JonLister]

Many thanks for the response, you've given me a lot to think about, I'm going to run the analysis as you've suggested. Best regards, Jonathan