Hydrotest Question - Closure Weld using a WN flange 3

[Krausen]

A number of my coworkers are in disagreement with my understanding of B31.3 hydrotest requirements for needing to hydrotest every component in a new piping system (with exception of the closure weld). A situation has come up where a new B16.5 WN flange is being used at a tie point where the buttweld of the flange's weld neck will serve as the closure weld to be exempted from hydrotest using 100% RT inspection of the weld. The current plan is to not hydrotest this new B16.5 WN flange before welding it in, leaving the flange face component untested before going into service (since B16.5 flanges are not pressure tested by the manufacturers). My understanding is this WN flange needs to be hydrotested first before welding in the weld neck as the closure weld in order to pressure test the flange face. Curious if the forum's interpretation is in line with mine on this?

 

[DekDee]

Hope this helps - you are correct

Interpretation: 17-28
Subject: ASME B31.3- i999 Edition, Para. 345.2.3(c), Special Provisions for Testing of Closure Welds
Date Issued: November 16, 1999
File: B31-99-025
Question (I): Does ASME B31.3-1999 Edition, para. 345.2.3(c) require that the described examinations
be performed for closure welds connecting piping systems or components that have been successfully tested
in accordance with para. 345?
Reply (I): No. The weld may be leak tested in accordance with para. 345.1 instead.
Question (2): Can weld connecting a flange to a piping system be considered a closure weld?
Reply (2): Yes, but if the weld is to be examined in accordance with para. 345.2.3(c), the flange, as well
as the piping system must have been successfully tested.

 

[Dhurjati Sen]

In addition to above:

If the above piping is for low pressure non hazardous service, you can skip the flange hydro-test and check its tightness at the time of commissioning. However, you need to take a 'Deviation' for the above. Do not go for this unless you have to.

DHURJATI SEN
Kolkata, India

 

[Krausen]

Just as a follow up here, I presented this B31.3 Interpretation 17-28 to my coworkers & they all still disagree that the B16.5 WN flange needs to be hydrotested prior to tying it in the system & performing required NDE at the new WN closure weld. They are claiming when this interpretation states " the flange, as well as the piping system must have been successfully tested" that this allows them to do an in-service leak test on the new flange, since the interpretation doesn't explicity state "successfully hydrotested" or "successfully tested per 345.4.2". They're also claiming this is how they've always treated this situation at the plant & have never seen anyone else do otherwise or take issue with it.

 

[LittleInch]

Can you describe this a bit better with a sketch or drawing?

My current interpretation is that you're cutting into an existing line, welding an flange and then will tie in some new pipe connection to that flange? for reasons not given, you can't hydro the new flange connection / weld and existing line.

As you have an open end and a flange face, can't you plug the existing line, attach a blind flange and then hydro that small section? Solves all this.

But I think you are right.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Krausen]

LittleInch - Attached is a sketch showing the concept here. We could use a Carber plug to test the field weld from the open end of new flange face, but this would only test the weld itself, but not the entire WN flange component. But at this stage the plan is to only NDE the closure weld of the new non-hydroed WN flange. I'm not familiar with any type of test plug that would allow you to insert from then open flange end and then install a blind flange to in order to hydrotest this small section through the blind flange. But if there is such a thing, that would solve it.

 

[1503-44]

No, that is not allowed.

Reply (2): Yes, but if the weld is to be examined in accordance with para. 345.2.3(c), the flange, as well as the piping system must have been successfully tested.

The tense of the reply makes your colleague's interpretation impossible.
"If the weld is TO BE examined" is obviously sometime in the future.
"Must have been successfully tested" is obviously in the past.
Those events do not occur at the same time, yet that is what would be required to accomplish both successful testing and examination of the weld, as interpreted by your colleagues.
It is also obvious that examination and testing are two separate and different activities. Why use two terms for the same activity, yet your colleague's interpretation makes them the same. What is their definition of examination and their definition of testing.

Regardless of what they have been doing for the last 100yrs, this is what you do.
What has to happen here is that the Flange must be bolted to another Flange on one section of piping. The exposed weld neck is welded to a short pup and then a cap is welded to the pup.
That section of pipe, including spool, spool flanges, new Flange, pup and cap will be hydrotested.
The other section of piping must also have been hydrotested.
Now the pup and cap are cut off, normally somewhere in the pup and the golden welding of the two sections at that cut is completed.
The golden weld is then examined for fitness.

Just move TP to the right to include a pup and that is solved. No plugs required.

Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[LittleInch]

One of these?

https://www.cw-estgroup.com/Product...lation-Plugs/High-Lift-Flange-Weld-Test-Plugs

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Krausen]

Just move TP to the right to include a pup and that is solved. No plugs required.

1503-44 - I agree, and this is exactly what I proposed to do in the IFC drawing that started this discussion with my coworkers to begin with. Thanks for confirming.

 

[Krausen]

One of these?

https://www.cw-estgroup.com/Products/Hydrostatic-T...

Much obliged LittleInch. This would solve this issue too. First I've seen of these types of test plugs.

 

[1503-44]

100%

Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[LittleInch]

TBH it was new to me as well, but they look like the real deal.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[RJW000]

No

 

[Krausen]

No

Ah, so the plot thickens...

 

[1503-44]

DekDee's post
Interpretation: 17-28
*********************************************************************
"Question (I): Does ASME B31.3-1999 Edition, para. 345.2.3(c) require that the described examinations be performed for closure welds connecting piping systems or components that have been successfully tested in accordance with para. 345?
Reply (I): No. The weld may be leak tested in accordance with para. 345.1 instead."

If you do not do the prescribed examinations, you must do the leak test.
If you do not test, you must do the prescribed examinations.

"Question (2): Can weld connecting a flange to a piping system be considered a closure weld?
Reply (2): Yes, but if the weld is to be examined in accordance with para. 345.2.3(c), the flange, as well as the piping system is specifically stated must have been successfully tested."

Yes the weld can be considered a closure weld. If you perform the required examinations on the closure weld without leak testing the weld, the flange and piping system still must have been successfully leak tested prior to the examinations.

***********************************************
RJW000 post
"...is the B16.5 flange required to be successfully leak tested in accordance with 345?
No. Also See 23-14"

A flange leak test would normally show leaks at the weld, or the face. Here the weld will be the closure weld, so it is exempt from leak testing requirements. The flange face is usually bolted to a blank or blind for isolation purposes during testing as discussed in 345.2.3(b), such blind or blank will be removed so as to complete the connection to pipe or vessel. If it is being used to isolate it is redundent to require the faces and gasket to be leak free during the test. This is the kind of installation they and 345.2.3(b) are talking about. The question as asked did not state exactly how the flange was to be used in the system, so presumably the answer would follow the line subject of the code, in an isolation setting flange to blank or blind. Neither does the interpretation strike Interpretation: 17-28. Doesn't mention it at all.
They also only discuss JOINTS (weld or flange face), not saying that clause 345.4, requiring every point in the piping system be hydrostatically tested is not applicable to points within the flange between the weld and the face, which must be tested hydrostatically according to 345.4, either prior to now, or during this supposed leak/hydrostatic testing opportunity you are trying to avoid with a "golden weld".

Plus, we know hydrostatically testing every point in the piping system is the right thing to do. It is the stated intent of the code. If it can be tested, and we know it is easily possible with a pup and an isolation blind exactly as the code has mentioned, we must test it.

They don't write this stuff for us to play lawyer and find some weasel words to convince ourselves we should go around the requirements of the rest of the words. Just read the paragraphs as written, assume the most logical safe result is what they want and then follow as much as you can to get there and it will be a lot easier.






Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[RJW000]

326.3 states the design, materials, fabrication, assembly, examination, and testing requirements of this code are not applicable to components listed in table 326.1 i.e. a B16.5 flange.

 

[1503-44]

So your considered engineering position is that, if you made a section of pipe from 2 B16.5 and 1 MSS-SP44 flanges, 4x B16.9 90deg ells of wrought steel butt weld fittings and a tee all connected to a relief valve via 500 feet of B36,10 weld seamless wrought steel pipe, you'd just weld it all up, drop it in place without any further testing and go on your merry way?

Those reference documents cover every component in a process system. Please tell me what parts of a process piping system you would test. Must be something. They're out there testing every day somewhere.

I read 326.1 as the requirements of B31.3 shall not be applied to the manufacturing of those components. The manufacturing and testing of components is covered in those respective reference documents.
The assembly and testing of those components into a process piping system is not covered by those reference documents. That's covered by B31.3.

Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[RJW000]

The welds would need to be tested.

 

[1503-44]

And ... every point in the system. They can't all be golden closure welds. Closure generally means one.

Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[KevinNZ]

I think the point 1503 is making. A B16.5 flange with no welds and not connected anything does not need hydro testing. We don't test replacement blind flanges before we bolt them on, do we?
Not sure if the ASME codes are actually letting us do this. (but B31.1 does have outs from hydro testing)