where expertise comes together - since 1996

Web's Largest Portal of Nondestructive Testing (NDT)
Open Access Database (Conference Proceedings, Articles, News), Exhibition, Forum, Network

All Forum Boards
Technical Discussions >
AUT for weld joint PIPE to FITTING
Career Discussions
Job Offers
Job Seeks
Classified Ads
About NDT.net
Articles & News

3603 views
03:41 Jun-20-2009

Dao Huu Xuan

Other,
Vietnam,
Joined Feb 2007
18
AUT for weld joint PIPE to FITTING

I have doubts with some specialist proclaim that AUT (PE and/or TOFD) is able to completely replaced for RT of piping fabrication. That, they have procedures/techniques to scan all piping joint configuration.

With Wellhead platform or integrated modul, the piping system often include more than 70%
of joints that connected to fitting (pipe to flange, pipe to bow, bow to flange, ...)

Which I suspect comes from the mentioned joints above is scanned from one side.
Anyone who have experience on this, please share a discussion.

Thank you,


    
 
15:48 Jun-20-2009

Nigel Armstrong

Engineering, - Specialist services
United Kingdom,
Joined Oct 2000
1094
Re: AUT for weld joint PIPE to FITTING In Reply to Dao Huu Xuan at 03:41 Jun-20-2009 (Opening).

Dao Huu Xuan

No NDT system addresses all issues, even with a simple pipe-pipe configuration, each method has a Probability of Detection less than 100% and each system has its own foibles. The promary issue should be "criticality" or "risk" rating - lethal service, high pressure system in a populated environment will require very careful planning from initial design philosophy through ensuring the philosopphy is avted on in alla spects of the project.

But I agree with you, if the project is very high criticality with probable loss of life in case of failure then no one system should be relied upon completely but complementary inspections, say RT, Phased Array, TOFD and pulse-echo should be instigated.

However if the consequences of failure are mild then a one-sided inspection may be sufficient.


    
 
23:11 Jun-20-2009
Mike
Re: AUT for weld joint PIPE to FITTING In Reply to Dao Huu Xuan at 03:41 Jun-20-2009 (Opening).

Dao Huu Xuan
You are correct to have doubts in a specialist that proclaims that AUT (PE and/or ToFD) will replace RT examination of piping fabrication. There are many piping/fitting configurations that preclude 100% coverage of a weld with AUT. Some examples include:
1) Piping to Flange welds - the area on the flange side of the weld is usually insufficiant for the placement of the probe (PE/PAUT/ToFD) and thus the fusion face of the flange side of the weld is not examined.
2) Piping to Elbow welds - the intrados of the the elbow on small diameter elbows prevent flat placement of the probe and prevent the transmission of ultrasound into the weld. The sides of the trunk weld of a Tee to piping have similar problems.
3) Elbow to Flange welds - compound the issues stated above

Unless the weld is ground flush, to attempt a one sided UT inspection of a weld (far fusion face examined in 3rd leg) is foolish because of reflections and mode converted signals from the weld cap.

Adding the issue of consequences of failure, I believe, only confuses the issue. If you are going to take the time and effort to examine a weld for quality, you might as well examine it throughly.

Regards,


    
 
14:45 Jun-21-2009

Ed Ginzel

R & D, -
Materials Research Institute,
Canada,
Joined Nov 1998
1211
Re: AUT for weld joint PIPE to FITTING In Reply to Dao Huu Xuan at 03:41 Jun-20-2009 (Opening).

Huu Xuan
I must concur with Nigel. But as the inspection company you may not have the final say in the details of what methods are used. This is usually an engineering and cost decision.

Pipe to flange or fitting is problematic for many NDT methods (not just TOFD, AUT). Manual UT and radiography also come up short on these configurations. More than once I have designed a technique for these configurations where I requested the fabricator provide a welded demonstration piece that I could machine targets in. Samples would arrive with assurances and radiographs showing no indications. When we machined the calibration targets and set up the scanning apparatus it was noted that in addition to the calibration targets long (>100mm) nonfusion flaws were present well over the calibration sensitivity.

I agree that the flaws on only one side were detected as a result of the restricted access; however, the ones on the side where access was not restricted were easily detected...by the UT technique. None of the nonfusion flaws were detected by the radiography!

Nigel's points of using multiple complementary NDT methods can be expanded on. If the owner is truly concerned about potential failure of due to flaws then they should make provision (mechanical and financial) to facilitate all methods to detect them. This would include removing the weld cap reinforcement to allow better UT access. As well, the removal of the weld cap would improve the ability to interpret RT results and the RT should be done using maximum sensitivity with X-ray and Class 1 flim from multiple directions (including parallel to the weld bevel).

ASME has acknowledged that conditions are not always ideal for weld examination by UT and includes a clause to identify what to do:
T-472.2 Restricted Access Welds. Welds that
be fully examined from two directions using the
beam technique (e.g., corner and tee joints) shall
be examined, if possible, with a straight beam technique.
These areas of restricted access shall be noted
examination report.


    
 
19:58 Jun-22-2009

Roger Duwe

NDT Inspector, API-510, 570, 653
MISTRAS,
USA,
Joined Jan 2009
148
Re: AUT for weld joint PIPE to FITTING In Reply to Ed Ginzel at 14:45 Jun-21-2009 .

Ed has mentioned the reason that UT needs to replace RT. RT misses more Lack-of-Fusion than it finds. Fine, tight L-o-F [essentially a braze] does not present itself on RT film, but gives a HUGE reflection on UT. The reason that RT is still in such widespread use seems to be tradition.

Back to Dao's concern: Bigbore elbows can be UT'd from the ell side with small 'ducers. Smallbore buttweld ells are a problem. I can get about the 1/3 of the heel area of the ell side of the weld, remainder is inaccessible from that side of the weld. One way to compensate for this coverage [remember -- the weld was well scanned from the pipe side] is multiple angles. And Phased Array gives you multiple angles when 'Sectoral' mode.

The question now becomes: Do you want 100% RT coverage that can't/won't find Lack-of-Fusion, or 100% UT from 1 side & fair to full coverage from the other [and finds ALL fusion defects very well]?

If lack of RT film is a problem, go with 'real' PA and have the contractor submit a linearly encoded scan of each weld, plus the software to display it. Then you have a color-coded image of the interior of your weld that you and your Level III can review. This scan file is electronic, so multiple reviews from folks on different continents is easy.


    
 


© NDT.net - The Web's Largest Portal of Nondestructive Testing (NDT) ISSN 1435-4934

Open Access Database, |Conference Proceedings| |Articles| |News| |Exhibition| |Forum| |Professional Network|