×
INTELLIGENT WORK FORUMS
FOR ENGINEERING PROFESSIONALS

Log In

Come Join Us!

Are you an
Engineering professional?
Join Eng-Tips Forums!
  • Talk With Other Members
  • Be Notified Of Responses
    To Your Posts
  • Keyword Search
  • One-Click Access To Your
    Favorite Forums
  • Automated Signatures
    On Your Posts
  • Best Of All, It's Free!
  • Students Click Here

*Eng-Tips's functionality depends on members receiving e-mail. By joining you are opting in to receive e-mail.

Posting Guidelines

Promoting, selling, recruiting, coursework and thesis posting is forbidden.

Students Click Here

Jobs

Socket Welding
8

Socket Welding

Socket Welding

(OP)
I would like to know if a space on a socket weld for pipe is required between the pipe and the shoulder of the socket it self

RE: Socket Welding

2
klkweld-

It's good practice since if you bottom out the pipe in the socket then when the piping is put into service and heats up it will try to expand but be restrained by the socket. Failure of the fillet weld is one possible outcome.

If you're working to B31.3, take a look at Fig. 328.5.2B which calls out a 1/16" approx. gap.

There's a product called a gap-o-let which I couldn't find on google, but we use it often in our socket welded construction. The other option is to RT the finished socket welds to confirm the gap.

jt

RE: Socket Welding

2
When bottomed out, cooling during welding produces contractive forces which can press against the socket fitting. This in turn produces a shear load across the weld, which when great enough promotes hot cracking of the weld. Similarly, if the material is hardenable and a suitable martensitic microstructure is obtained, hydrogen induced cracking may also occur.

RE: Socket Welding

I f you check ASME b31.3 the information is 1/16" gap after welding.

RE: Socket Welding

The B31.3 requirement is an approximate 1/16 inch gap before welding.  There is no requirement for the gap dimension after welding.

RE: Socket Welding

bvi,
Please advise the acceptable criteria for gap after welding. We may confirm gap after welding by RT. However, there is doubful for the accepted criteria.

RE: Socket Welding

There is no requiremnts in Code. This is an engineering decision.  Essentially any gap is acceptable unless it is too large >>>1/16"  

RE: Socket Welding

How about stick in the pipe fully & scribe a line around it 1" above the shoulder, then pullout 1/16"& weld.  = easy to inspect. - that was the standard procedure in Turbine Land

RE: Socket Welding

I must apologize it ie after welding.

RE: Socket Welding

I've seen Boilermakers stick pieces of light cardboard inside the socket joint to keep the gap, and it just burns away during the welding process

RE: Socket Welding

stanweld,
The inspectors used to comment to accept gap not over 3 mm from RT result in my previous project. Please comment.

RE: Socket Welding

Inspectors often use subjective reasoning for acceptance of workmanship. I have seen numerous leaks in service when the protruding pipe into the socket was less than 1/4".     

RE: Socket Welding

stanweld,

There MAY be some piping Codes that do not REQUIRE (and quantify) the subject gap but I can assure you that the ASME B31.1 and B31.3 Codes ABSOLUTELY DO REQUIRE a 1.5 mm (1/16 inch) gap (see, e.g., B31.3 Figure 328.5.2C).

Regards, John.

RE: Socket Welding

JohnBreen
Please refer to B31.3 Fig 328.5.2C "1.5 mm (1/16in.) approx. gap before welding". Also refer to B31.1, FIG. 127.4.4(B)"Approx. 1/16" (2.0 mm) gap before welding".

APPROXIMATELY is not exactly! Before welding is not after welding when measured by viewing a radiograph! As previously stated weld shrinkage will generally cause the approximate 1/16" gap prior to welding to be reduced. There is ABSOLUTELY NO REQUIREMENT to meet the 1/16" gap after welding.      

RE: Socket Welding

stanweld,

When we (at B31, MDC) put those rules in the Codes the "approximate" in the gap was a round-up, based upon our calculation of how much expansion would result from the welding process (heating the weldment).  We based that upon the expansion coefficient of stainless steel.  So, what we intended was that when at ambient temperature, before the welding process, (i.e., before the components were heated to welding temperature) we would have a big enough gap such that the end of the pipe would not "bottom out" at the  base of the socket when the weldment was AT welding temperature.  Of course upon cool down after the weld was made, the gap would be approximately the original 1/16 inch.  It is implicit in the rule that upon fit-up the gap would be approximately 1/16 inch (cold) and after the weldment had cooled (after welding) the gap would return to approximately 1/16 inch.

If you think it is something that needs the full Committees to interpret, I would be glad to sponsor an inquiry to B31.1 and B31.3.  I would ask them to add words that unambiguously specify that the gap should be approximately 1/16 inch after the weldment has cooled.

Regards, John.

RE: Socket Welding

2
John,

It's interesting to hear the background for how this requirement came to be part of B31.  I've welded many socket welds for Navy nuclear and non-nuclear applications, and it was also a MIL-STD / NAVSEA requirement to have a 1/16" gap before welding.  For those standards, we had to qualify separately for socket and groove welds.  The socket welder test coupons were required to be examined by PT, then cut into quarter sections for macroetch examination.  Many times the NDE shop would try to hold us to the 1/16" minimum gap AFTER welding.  If we set the gap at 1/16", the weld shrinkage would always close the gap to almost zero.  To get 1/16" after welding, we would have to set it at 3/32"-1/8" before welding.

I don't see the logic in requiring any gap to be present after welding, except for disimilar welds where thermal expansion could cause the pipe to grow at a different rate than the socket fitting.

RE: Socket Welding

John
If it is truly the intent of B31.1 and B31.3 that the 1/16" minimum gap is required and is to be demonstrated after welding, then amend the Code accordingly. Require RT of 5% (or more) of the socket welds to confirm this workmanship standard; noting that there is no current requirement to do so. As I have stated and Welder 4556 has noted, weld shrinkage will most often reduce the original gap; it will not return to the orignal gap prior to welding.  

RE: Socket Welding

Gentlemen,

That IS very interesting.  Maybe 1/16 inch after welding is not enough (?).

Perhaps the best approach is to sumbit inquiries to B31.1 and B31.3 and have them provide a formal interpretation.  I will send inquiries to these Committees and see if we can get a fast interpretation or Code Case.  By the way, anyone who has a question can submit an inquiry.  If you look at B31.3 Appendix Z, it provides guidance for submitting inquiries.  I will post updates on this thread if I can.

Regards, John.

RE: Socket Welding

The 1/16 inch is before welding, as stanweld states, and there is no code requirement for a gap after welding.  Weld shrinkage will tend to close the gap.  If you need to see a gap for quality control purposes after welding, you would need to specify that as an additional requirement of the engineering design.  There are existing B31.3 interpretations on this issue, and they confirm the gap is before welding, and that there are no requirements for after welding.

See also the discussion in section 12.12 of the book, Process Piping, by Becht.

RE: Socket Welding

So what is it about words like approximate or recommended that make people commonly jump to a mandatory shall?????

I promise to vote negative on any B31.3 ballot that takes the word approximate out of the gap requirement. There is no need to make more out of this requirement than what is stated.

As for 3/32" its approximately close to 1/16"!

Instead of worrying about the gap, what about the filler metal, or undercuts at the toe of the welds, or leg height and geometry?



Regards,
XHPIPE

RE: Socket Welding

XHPIPE;
Being involved with Codes and Standards committees I agree, and a star for your negative vote along with mine for B31.1! Leave the wording as is.

RE: Socket Welding

metengr/XHPIPE
Had I a vote, I would vote negative with you. There is no reason to change the decades long wording. Unfortunately, a number of Owner/Engineers do not appreciate the present wording and the reasons for it and have required RT of socket welds to demonstrate conformance.  

RE: Socket Welding

Hello all,

This has been an interesting discussion.

bvi,

Please save me some time and tell me the numbers for the existing interpretations having to do with the socket weld gap.  I am now very curious and I would like to educate myself on the full committees' reading.

I understand the current Codes wording RECOMMENDS an APPROXIMATE gap of 1/16 inch BEFORE WELDING (that is plainly written).  There is nothing in either Code that requires any "after welded" gap.  No need for further discussion there.

OK, summing it up.  The answer to the original question by klkweld is simply "B31.1 and B31.3 RECOMMEND an APPROXIMATE gap of 1/16 inch.  Please refer to B31.1 Figure 127.4.4(B) and to B31.3 Figure 328.5.2C".

Contrary to what I THOUGHT I was reading, I can now see that nobody in this thread objects to the APPROXIMATE gap of 1/16 inch BEFORE WELDING.  I see that the point is that no matter what gap is used there can be no guarantee as to what the resulting after weld gap will be.  Since the subject has generated previous inquiries and interpretations there is likely no need to go there again.  No words to change, no negative votes required.

I find the experience of Welder4956... "If we set the gap at 1/16", the weld shrinkage would always close the gap to almost zero",... to be really interesting.  It makes me wonder how many times, when the Gap-A-Let device has been used; it caused insufficient room for weld shrinkage.  I also benefited from stanweld's experience regarding leakage when the pipe was inserted less than 1/4 inch.  Since the range of NPS's for socket welds is all the way up to NPS 4, there is a need for experienced judgment in deciding the optimum pre-weld gap.

Thanks to all for the discussion.

Regards, John.

RE: Socket Welding

Why are you adding words to the code.  The word "Recommended" does not exist in the code with regard to this subject.  It is a minimum requirement of an approximate gap.  So it is a requirement, not a recommendation, but the specified dimension of the gap is approximate.  

John, see interpretation 16-06.

RE: Socket Welding


  indeed this discussion reelly interesting
  
  The question raised in our forum is construction concern. So, based on ASME/ANSI B31.3, because i'm working with , it is clear that  1/16 inch is requiered. in general we have to ovoid any contact steel - steel .
since we are talking about socket weld, the branch connection is greater than 2 inchs , can you use tee in stead ?

RE: Socket Welding

Adding a little information:
The last project using socket welds that I was associated with was what in some circles could have been considered a nightmare.
The project had a total of 3268 welds of which 2013 were socket welds.  These welds not only had to meet the applicable codes but were also required to meet our piping standards.  Our standards called for a 1/16" gap after welding and minimum engagement of 2t (pipe) using 2 passes.  

This was a turnkey project until the pressure testing phase started.  The initial test was a vacuum  test with a pressure decay of less than 1 # hr.  They were unable to pull a vacuum on completed section.  This attracted our group’s attention and a cursory visual inspection of the completed socket welds prompted a scope change on the project to where we would do the weld inspection and testing.    Our normal inspection requirements for piping in this service (vaporized heating fluid at 700? F) was VT, 100% RT, followed  with a He leak test at 100 psig had been waved based on assurances by the contractor of their expertise and QC that convinced project management to go with a 5% RT program to save money.  Our initial over site was not what it should have been.

Here is our inspection summary:

Of the 2013 Socket Welds, mainly 1" or 1.5" CS the following defects  were found:

    101 welds that had no gap or excessive gap.  The sockets were being used as a crutch.  In fact some pipe didn’t make it to the fitting.
75 welds that had Cap problems taken out by VT.
52 just bad welds, mainly porosity.
21 weld with fusion problems.

The second half of this project with an almost equal number of welds (socket and grove) had less than a 1% weld defect rate of all welds.   

We went to court over our inspection program and after 7 years,just recently we won.    
Something  good did come out of this in that several welders who worked on the project have told me they now know how to fit and weld out socket fittings now.  The negative side is that some have told me on later jobs they were told to stab it and one pass it as it’s only a socket weld.    

RE: Socket Welding

To me, that's just pipefitting that's sloppy at best, and incompetent at worst. The number of guys out there who cannot work end to centre or centre to centre on their fitting work is large & growing. Many cannot even do 45* offsets. And rolling offsets? Get ready for a lot of trail and error and a pile of scrap pipe.

Red Flag This Post

Please let us know here why this post is inappropriate. Reasons such as off-topic, duplicates, flames, illegal, vulgar, or students posting their homework.

Red Flag Submitted

Thank you for helping keep Eng-Tips Forums free from inappropriate posts.
The Eng-Tips staff will check this out and take appropriate action.

Reply To This Thread

Posting in the Eng-Tips forums is a member-only feature.

Click Here to join Eng-Tips and talk with other members!


Resources