Use of Gap-a-Lets in Socket Weld Piping Fit-up 2

[Stevebrown45]

I am looking into the use of Gap-a-Lets for socket weld fit-up for piping on US Navy ships. I am looking for anyone currently using them, or any reasons NOT to use the.
Any help would be grealtly appreciated, thanks,

 

[gtaw]

I believe the customer would have to give their approval before using them.

There is a requirement in NAVSEA TP27 to use scribe lines to ensure the proper gap between the bottom of the socket and the end of the pipe. I mention NAVSEA TP27 because it is the governing fabrication document for much of the piping aboard Navy ships.

Best regards - Al

 

[Stevebrown45]

I concur gtaw, but this request is to find information relating to a cost improvment, so NAVSEA will be the final approval athority. Prior to starting my own test program, I want to get input from the community so I can have some level of expectation.

Thanks.

 

[pennpiper]

It is called Gap-O-Let
Look here for a discussion:

http://pipingdesigners.com/phpBB3/viewtopic.php?f=1&t=2814&hilit=Socket+weld+gap

For the record I do not recommend it.

prognosis: Lead or Lag

 

[Stevebrown45]

Looking at the documentation and the samples I see "Gap-A-Let" manufactured by G.A.L. Gage Co.

I appreciate the link to the comments from 2006, but I need objective data not "what if". The study I am doing is for the shop where snow, rain, etc. are not an issue.

For the record, the "gap" is only there to allow for thermal expansion so the weld is not stressed and possibly crack. As long as the insert allows for thermal expansion of the pipe/fitting during welding whether or not use of the Gap-A-Let constitutes elimination of the "gap" is irrelevant.

I will post the results of the testing to this forum, but was hoping for a little objective evidence of whether or not they work.

 

[DekDee]

Steve,
Have never used them so have no idea if they work or not but going from the tried and tested "scribe" method as noted by Al to actually purchasing something additional to do the same job will not be a cost improvement.
IMHO the only true way of confirming a specified minimum gap is with the scribe method - how can you confirm there is actually a "gap-a-let" (and therefore a confirmed gap) installed once the pipe is tacked in ?
I welded piping socket welds for 15 years and have been inspecting them for another 15 years so have a bit of experience with them.
Not sure of your actual role in the military (are you from a welding background or an engineer ?) so if you are fully aware of the steps I have listed I apologise.
1 Install pipe so it is roughly level and bottomed out in the socket.
2 Scribe around the circumference of the pipe.
3 Tilt the pipe away from you as far as it will go.
4 Place a tack on the side nearest to you.
5 Pull the pipe back towards you until it is level.
6 Tack the side furthest from you.
7 Level the pipe in the opposite direction.
8 Place tacks on opposite sides.
9 Continue tacking pipe/sockets until spool is complete.
10 Call QC to check scribe marks - if all acceptable, weld them all out.

Hope that helps,
Cheers,
DD

 

[moltenmetal]

The function of a gap-o-let is merely to provide something to check for (presence/absence) in the unnecessary radiograph of the joint. Otherwise, you end up in an argument about whether the code requires a gap before welding or post welding, which from a code perspective is a settled question. A gap before welding is required- a gap post-welding is NOT required, hence you cannot use RT post welding as a valid means of NDE unless you use the gap-a-lets. Instead you must rely on VT during fit-up.

 

[Stevebrown45]

Thank you DekDee. I am actually a Piping Designer, with Navy Submarine experience and have been in the submarine overhaul/new construction piping and mechanical field for over 40 years. You are quite correct with your summation of using and verifying the required gap, but what I am suggesting here is a change to the paradigm. If the welder/Inspector verifies installation of the gap-a-let and the joint is say, marked in a certain way to indicate its use, then use of the gap-a-let could quite possibly reduce the amount of time currently required to fit-up a socket welded joint. When you make thousands of joints in one product, time is money, and a lot of it.
And thank you moltenmetal for your observation concerning the use of RT to validate a socket welded joint, where RT is not commonly used to NDT a socket weld joint.
We have to trust our Welders and Inspectors to “do the right thing”.
Commenter’s, please help me stay on track with objective quality evidence as to the value in using the Gap-a-Lets. Has anyone out there witnessed a systemic failure after using Gap-A-Lets? Has anyone done a cost comparison in using/not using Gap-A-Lets for piping system installation? Etc.
Thanks to the entire community for your input; it all has value.

 

[moltenmetal]

I've never witnessed a failure resulting from using a gap-o-let, nor have I witnessed the cracking that is supposed to be inevitable if the code-required pre-welding gap is NOT maintained. But it is a code requirement and comparatively easy to ensure by proper VT and supervision of the fitters and welders, so it's not a big problem. What bothers me is calling for unnecessary and misleading RT that cannot inspect what the code is asking for.

 

[DekDee]

Steve,
There will be no time saving in using the Gap-A-Let - the pipe still has to be levelled whichever way you choose to obtain the gap.

moltenmetal,
I don't know how many Quality managers I have called idiots (in my head) who continue to demand RT of socket welds after welding even when I have printed the relevant pages fom the codes and highlighted "before" welding.

Cheers,
DD

 

[weldstan]

Gap-A-Lets have been and are being used effectively to provide the minimum gap. They have been in use for decades. The scribe and back-off 1/16" practice is also effective. After the weld is made, confirmation of the use of either cannot be done visually. Confirmation must be done during fit-up or by X-Ray after welding. Confirmation of a 1/16" (minimum by Owner/Engineer Spec) gap by X-Ray is a waste of money and has never been required under Code.

Of far more concern is the minimum insertion distance. I have seen many a leak in socket welds where the insertion was less than or equal to 1/8".

 

[Stevebrown45]

Who is using Gap-A-Lets? Who has hard data? Whether or not a joint leaks is a result of the weld, not the gap; although I suppose an excessive gap could cause a joint to erode over time and fail.

 

[weldstan]

Stevebrown45,
Construction companies for which I have worked during the pasts 41 years have used Gap-A-Lets on a number of mulibillion dollar petrochemical and power projects, especially where random 10% x-ray was required.

Inadequately penetrated socket welds failed generally in short cycle fatigue or thermal fatigue.

 

[jte]

There are lots of experiences out there. There is little question that the code (B31.3) requires no gap AFTER welding. But this is not settled even within the code committee. From Process Piping: The Complete Guide to ASME B31.3, 2nd Edition by Chuck Becht IV (who has very good Code Committee qualifications):

A second issue that has caused considerable controversy is the 1.5mm (1/16”) approximate gap before welding indicated on the figure [12.7 in the book, B31.3 Fig. 328.5.2C]. This is a requirement for a gap before welding, so that weld shrinkage will be less likely to cause small cracks in the root of the fillet weld. Whether or not such cracks cause problems is questionable, and fatigue testing has shown that socket welds that are welded after jamming the pipe into the socket have longer fatigue lives than ones welded with a gap. There is no requirement for a gap after welding, and weld shrinkage can close a gap that was present prior to welding. [underline added]

Some owners require random radiographic examination to ensure proper socket welding practice. One of the items checked for is the presence of a gap. An argument for doing this is that it is not possible to determine if there was a gap prior to welding unless there is a gap shown by radiography. If desired, the requirement that there be a gap after welding should be specified as an additional requirement of the engineering design. It is not a Code requirement.​

That's the perspective from one committee member, as presented in a published book. I won't pretend to be an expert on this particular subject; figured I'd throw it out for consideration.

jt