Socket Weld Gap

By jon20013 (*****)

Date 11-15-2006 14:07

I don't think many (none that I personally know of) codes address a maximum gap for socket welds. I have seen it written in certain company procedures that the maximum should be equivilent to the wall thickness of the material. This seems to make pretty good sense to me.

By Shane Feder (****)

Date 11-15-2006 14:27

Hello all,
I agree with jon on the maximum gap being flexible.
We have minimum gaps allowable which I have always thought was to counter any weld shrinkage/stress problems.
The socket weld is going to be a fillet weld so as long as the gap is greater than the minimum required you are safe.
Off the top of my head I am not 100% sure that B31.3 doesn't give maximums and minimums.
Looks like I am jumping in the codes first thing in the morning
Regards,
Shane

By anuss Date 11-15-2006 14:34

I have, in the past, gone with thickness being the max allowed gap, guess I would like to be assured on this one. :-)
Thanks!

By RANDER

Date 11-15-2006 14:50 Edited 11-15-2006 15:22

Agreed, B31.3 states the minimum welding dimensions in para. 328.5.2 and fig. 328.5.2.c but no max. dimensions for gap are given.

Interpretation: 6-02
Subject : ANSI/ASME 83 1.3-1 984 Edition, with the ANSi/ASME B31.3a-1984 Addenda,
Para. 31 1.2.4 and Figs. 327.4.28 and C; Socket Welds, Gap Dimension
Date issued: December 14, 1987
File: 83 1-87-01 3
Question (1): In ANSI/ASME B31.3, para. 311.2.4(b)(2) states, "Weld dimensions shall not be less
than those shown in Figs. 327.4.2B and C." In Figs. 327.4.2B and C, there is a dimension that reads
"'1/16 in. approx. before welding." Do the above words mean that the gap should be no larger than
1/16 in.?
Reply (1 ): No.
Question (2): If the response to Question (1) is no, how large a gap would be acceptable?
Reply (2): The '''1/16 in." is not a "welding dimension." It is intended as an "approximate"
dimension for nominal clearance.

(Edited to add Interpretation)

By SA Inspector Date 01-11-2007 09:24

I am sure ASME has reasons for not being too prescriptive over the min and maximum gap sizes, before and after welding however to control this from a QC/NDT point of view becomes difficult when the approximate gap of 1.5mm is stipulated before welding. This leaves room for the contractor /fabricator to argue the point that there was a 1.5mm gap before welding but has now closed due to welding shrinkage. This cannot be proven via the only practical method ie. RT.
I believe an after welding gap should be stipulated as minimum 1mm. All the fitter needs to do is push the pipe in all the way to the socket shoulder, scribe the pipe, withdraw it 2mm (or an amount this contractor has proved sufficient for that pipe size/welding procedure), tack and weld. Any deviation to the requirement of 1mm gap is cause for rejection - simple.

Residual weld stresses in a socket joint through no gap are an uneccessary evil, when added to other forces like internal pressure, misalignment of piping, vibration, impacts and other unforseen loadings can and have resulted in failures.

By Shane Feder (****)

Date 01-11-2007 09:36

SA Inspector,
When I worked for SGS in New Zealand as a Radiographer we were regularly hired by the major oil companies to radiograph socket joints to prove that there was a gap.This was not a code requirement but a customer requirement (I presume to keep the contractor "on his toes")
Regards,
Shane

By Jim Hughes (***)

Date 01-15-2007 18:54

Hello everyone, I know I'm coming into this late but G.A.L does offer a water salable Gap-O-Let. The only Problem is G.A.L cannot give you a detail sheet to find out what's in it. The job I'm on has a halogen and chloride stipulation in the contract so we can't use them because we can't be sure what's in the material that the Gap-O-Let is made out of.

Thanks
Jim Hughes

By jarsanb (***)

Date 11-15-2006 14:22

A good fix to this problem is the gap-a-let spacer (G.a.l. gage co.). Required in some procedures. It helps to keep the branching member straight as well. But to answer your question, hopefully someone else here has some concrete info.

By jon20013 (*****)

Date 11-15-2006 14:32

Yes, I agree with the gap-o-let method however I have also had customers come back and tell me I could not use these on carbon steel systems because the gap-o-lets being made from stainless steel would then set up a galvanic corrosion couplant. After some review, I have to say I agree with what my customers have said. Seems like GAL should make these things out of both alloys to avoid this issue. (They may well do this now, it's been a long time since I've looked into them).

By anuss Date 11-15-2006 14:40

Gap-o-let spacers aren't a bad idea and have been used by some of the guys. When not used though I have been seeing isues (in a couple of fits) with the gaps being in upwards 3/8" to possibly 1/2". I'm ready to throw the hammer down :-)

By Shane Feder (****)

Date 11-15-2006 15:23

Anuss,
As everyone has mentioned as long as your welders are not going under the minimum allowable gap you are fine.
One way we found acceptable was if you slipped the pipe all the way into the socket, tipped it over as far as the socket would allow and then tacked it on the high side when you tipped the pipe back up to level you had your minimum gap.
Regards,
Shane

By CECW7611 Date 11-22-2006 14:07

Part of the regidity of the socket weld joint depends on the maximum amount of insertion into the joint given the allowable gap for expansion. The deeper insertion, the more rigid the joint.

By pipewelder_1999 (****)

Date 11-22-2006 15:34

Could you elaborate on this ? I find it hard to believe thet the weld metal will yield enough to allow what pipe is inserted into socket . I am not saying it is NOT SO. I just can't picture it.

Help me out.

Thanks

Gerald

By Sourdough (****)

Date 01-14-2007 22:56

Might want to rethink your handle on this forum. Guys like me could get ahold of that one and runnnnnn. . . . . . .

ANUSS??