Ultrasonic Weld Testing vs. Radiographic Weld Testing

[RobsVette]

Hey guys, I question came up were I work today that I would like to try and get some better information on.

Here is the issue, a spec for a project we are working on requires Ultrasonic Testing of all pipe welds being performed in accordance with B31.9 Building Services Piping. The systems are basic chilled water, condenser water, hot water, etc... All pipe is ERW carbon steel.

My questions are as follows;
1) I had always thought that an Ultrasonic Test was the best method of performing NDE on a welded joint. It shows the defects to the operator the best. However, since there is no film produced, it is up to the operator to determine the pass or fail. Here in NYC everyone likes to have an Xray done because it gets you a film that can be reviewed by anyone.

Can you guys provide me with any other reasons to perform a UT test vs an RT test and what are advantages and dis-advantages of both?

2) Granted that B31.9 Building Services piping is very lienient as far as NDT requirements are concerned (it actually doesnt require it at all but has a suggested minimum quality level,) However since all welds are to be UT tested to B31.9 standards, I still feel there is a considerable difference between a weld that will not be tested and the minimum quality level of B31.9.

What do you guys think of this? This is a major project with as many as 10,000 welds being performed for the entire building, so this could become a cost issue for us.

Please provide any insight on quality levels for welding to B31.9.

If you need further information or clarification please let me know and thanks in advance for any help.

Rob

 

[salmon2]

I don't know b31.9 but has worked with erw tube for fairlr long time. UT testing is best for erw tube as it a online process so it is very fast. Rejection criteria is set to certain level depending the grade or whatever quality level you want, so it won't produce any film for an operator's judgement. It is an automatic pass/fail inspect.

Actually it is usually a full body UT, not only weld sealm.

 

[blacksmith37]

Has the horse already left the barn ? Did you have 3rd party witness the mill UT (assumed) of the weld seam ? Not to mention % RT of the fittings.
RT , UT, PT, EMI, have different strengths and weaknesses. I prefer RT of field welds; It is not nearly as sensitive to "artifact" orientation as is UT.Of course there are many books on the subject.
Being a cremudgeon, I suspect the objective is to have a certification in a file that the welds are good.

 

[RobsVette]

Hey guys, sorry for the mis-leading of the statement. The concern is not about the seams on the ERW pipe. The concern is on the field welds made on site. The requirement is that all welds be UT tested, even basic chilled and condenser water system for water cooled chillers.

Blacksmith, what do you mean when you say Artifact orientation? I agree with you on RT of field welds. I think that is the best way to go. Also then, what would be the strengths of a UT vs. an RT?

 

[EdStainless]

I dislike field RT since there is rarely clear indication of the actual exposure/sensitivity. And it takes two shots of a joint to show the full thing. But in this case it might be the best option.

UT can be just as bad. For it you need references, sample joints with known artificial defects in them. Sloppy welds can swamp the defect signal.
It would be tough to see imposing a signal to noise ratio requirement on field welds in steel.

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

 

[blacksmith37]

I was suggesting that a QA program should start with pipe/ fitting manufacture. Until you have determined the size of an "artifact" (relative to specification) you don't know if it is a defect or an artifact.
At this point I think examination (RT) of a % of each weldors work is what can be done.
Weld quality generally increases significantly if the weldor knows there will be NDE (even if no NDE is done).
Old guy story: Contractor says the owner cannot do NDE after the job is done because there was no prior notification, (ie. if they knew there was NDE, they would have done a better job).

 

[Ron]

Both have their limitations. If the wall thickness of the pipe is less than 5/16, you can have issues with UT. UT is also very single operator interpretive....RT can be reviewed by anyone. RT's limitation is that it only detects defects perpendicular to the film plane. Lamellar defects cannot be easily determined with RT.

Further, RT can be done on small diameter pipes in a single shot, but requires multiple exposures for larger diameters.

 

[metengr]

Time of flight diffraction (TOFD) ultrasonic testing using phased array is the best approach for most field and new construction welds for piping above 2". Why? Because this is the most sensitive NDT technique and seems to be gaining momentum each year. It is recognized by Codes and Standards and provides a volumetric record of each weld.

Most are now using this method during construction for baseline inspections and compare it to future in-service inspections.

 

[HgTX]

Phased array easily provides a volumetric record for specific locations, to show the kind of indication you're looking at. Depending on the application (I haven't seen a setup that would work for pipe but I'm no expert) there is also a tracker wheel type setup that lets one continuously record the inspection and provide a full record that can be reviewed later to see if the operator missed something, which is really where RT provides warm fuzzies.

Hg

Eng-Tips policies: faq731-376

 

[brimmer]

You can record the scans with UT now (digital) depending on what equipment your NDT vendor has, if your worry is you won't ever have a record other than a UT report. There is also digital RT now.