Socket Weld Requirements per B31.3 1

[marklobo]

We regularly employ socket weld-type joints in fabricated piping components, and attempt to comply with ASME B31.3-1999, paragraph 328.5.2, and specifically, Figure 328.5.2C. I have trouble with the use of the word "approximate" in any specification, and in this case, the word causes lots of discussion with regards to the "approximate" 1.5mm gap "specified". Now, using the two words in the same sentence creates an oxymoron, but I digress. Is the gap before welding required to limit the induced stress in the weld after cooling, or for some other function?

OK, now, if the pipe and socket have been fabricated with an interference fit, does that defeat the purpose inherently, or is that dependent on the fit class?

 

[kstaylor]

Answers to your questions:

1) The gap at the bottom of the socket is to assure that you don't crack the fillet weld during welding if the pipe and socketweld fitting expand at different rates (e.g. the pipe will heat up faster with it being usually thinner than the fitting). The word "approximate" is used as it is only necessary to provide a gap large enough to accommodate the differential thermal expansion and really doesn't serve any other purpose after the weld has been completed.

2) Socketweld fittings manufactured to ASME B16.11 have an inside diameter of the socket that is larger than the nominal OD of pipe made to ANSI B36.10. In over 20 years of working in the piping industry I have yet to see an interferrence fit between pipe and socketweld fittings. You might want to see if you have a piece of tubing instead of pipe (tubing is sized on actual OD dimensions) or some pipe made to metric pipe sizes.

 

[marklobo]

Sure, B16.11 sockets are loose in order to accomodate the wide OD tolerance of pipe. For some of our components, we fabricate our own sockets and turn pipe OD so that linear alignment is held. My question then is: is this OK or does it defeat the purpose of the "gap"?