Dear All,

Can we use ASME B31.1 appendix D (Branch welded-on fitting) for latrolet SIF calculation or this is only valid for olet? If not, what software do you suggest for calculating SIF?

I would be disinclined to apply appendix D to angled branches. Stresses should be able to be run in either FE/Pipe or NozzlePro, and likewise, they can be used to compute SIFs.

Appendices D from ASME B31.3 & B31.1 Codes can exclusively be used for Orthogonal intersections.

ASME B31J/2017 formulas have been developed for the same application area.

So yes, it remains FEA approach as single valid approach at this time, if we need the SIFs for laterals.

BUT, have you modelled weldolets using FE Pipe? Have you seen how many parameters (angles, distances) you need to define the model?
Moreover, since weldolets (like latrolets)include a thick walled portion, Shell Elements are not quite recommended to get "accurate" results. And have you seen how complicated is to develop a 3D-Brick Element model in FE Pipe?

Currently, MSS SP-97 does not provide the dimensional parameters in detail. Therefore, the actual dimensions (excepting OAL and welding diameters) depend on Manufacturer.

The conclusion is: unless you have Manufacturer detailed drawing of the fitting, you are unable to develop a reliable FEA Model. Just assuming some values does not mean a sound approach.

Personally, whenever I have to model latrolet-type fittings, I use Bonney Forge Latrolet SIF values from CII database. Boney Forge still remains a Reference in Piping Fittings technology. And anyway, those SIFs are validated by tests and practice...

Regards,

I have not modeled dimensionally accurate latrolets into FE Pipe, and admit I would not recommend doing so. Even replicating the model from Bonney Forge's catalog data into Inventor is daunting because the listed data did not appear to match the offered diagram.

Since there is no standard latrolet design, I tend to go with the pipe+pipe+weld approach and accept over-reporting.

So long as the latrolet piping is significantly longer than the latrolet itself (plus angled distance from wall to center of main header), you can assume standard flexibility or infinite rigidity (through offsets) with comparable results.

And for what it's worth, I concluded that the catalog pictures depict a solid cylinder that's been inserted into a pencil sharpener at 2 different angles, and cylindrically bored from apex to apex. The ID at the end that's connected to the header has been chamfered at 90 degrees and smoothed out to some unlisted radius.

http://www.bonneyforge.com/products.php?pg=branch&subpg=latrolet
http://www.bonneyforge.com/specs/branch/latrolet.cfm