BS 2633, carbon steel branches and PWHT 2

[andyenergy]

I'm looking for some informed opinion on why this might be...

BS 2633 (Arc welding of ferritic pipework carrying fluids) details PWHT requirements for various materials depending on thickness.

For carbon steels, <0.25%C, there is no requirement for PWHT provided the thickness is less than 35mm. There are other caveats to this, but these are not relevant to this query.

For branch welds, there is specific clause that states that PWHT is required where:

The thickness of the thicker part is over 35 mm, or

the distance as required by design from the toe of a branch weld to the toe of the adjacent branch or attachment is less than 4 times the thickness of the main pipe.

Can anyone suggest why this is so and the presence of this adjacent branch requires PWHT? I think I know why, but I'd be intertested in the opinion of others.

Thanks

Andy

 

[metengr]

andyenergy;
Most of the time I don't get a chance to see other Codes and Standards in relation to my part time work with ASME Codes and Standards. However, in relation to your query, IMO the need for PWHT in this case for branch welds has more to do with reducing harmful residual stresses from welding the branch connection to the main pipe run for the given conditions.

 

[andyenergy]

Thanks metengr.

That's exactly what I thought too. Most of time branches that I see are separated quite well, but in this case someone had picked up this clause and i had to do a bit of head scratching yesterday.

Thanks again.

 

[unclesyd]

Adding a little:
Years ago prior to the advent of FEA and good strain gauges one of the pipe fitting companies accomplished a lot of testing of different fittings and welded branch connections using brittle lacquer. The assemblies were hydrotested to stress levels that were approaching the yield point of the materials. From the photos of the test results it was quite evident that the closer the toe to toe was the greater the stresses shown in the lacquer. The assumption, later verified by strain gauges, was that the residual stresses in these areas were extremely high.

The amount of stress in this area is evident if you ever try to make two welded branch connections close together on anything less that heavy wall materials. To mediate these stresses, along with some welds and resultant deformation we use a lot of extruded connections mainly on Ti and SS.

 

[andyenergy]

unclesyd -

Thanks! Always good to get a more complete story. very useful indeed.