Can you Disposition Laminations in Wrought A234 WPB Elbows?

[Bambie]

Phased Array inspection of a wrought A234 WPB, 18"nps sch 40, 90 degree, LR elbow has revealed laminations and inclusions.

The majority are located mid thickness along the cheeks of the elbow, not the extrados or intrados.

The majority lie parallel to the surface and do not break through either the OD or ID surfaces, however, some transition vertically through half the wall thickness.

Is there a sure-fire way of confirming that these are not cracks?

Perhaps someone familiar with A234 WPB wroughting techniques could comment on whether laminations with vertical transitions would form in the cheeks of large elbows.

 

[metengr]

The phased array UT is an accurate volumetric examination method and requires experienced NDT personnel to interpret. With that said, it seems like you have your answer - lamination and inclusions. To go further would, in my opinion, require either a second PA UT by another NDT person, conventional shear wave UT or a core sample to confirm indications by metallographic examination.

 

[Bambie]

metengr,

The NDT person labelled the vertical transitions 'indications' and did not rule out 'cracks'.

Rather than challenge the interpretaion of these electronic smudges, I was trying for the mechanistic approach; if wroughting an elbow can produce radial orientations of axial laminations in the cheek material, then it is likely they are not cracks.

 

[metengr]

I would investigate shear wave UT because this is fairly reliable in evaluating laminations. Yes, radially stepped laminations can occur in wrought elbows at random locations.

 

[Bambie]

metengr,

Would you happen to have a source for that last comment?

 

[metengr]

Yes. the best source is first hand experience.

 

[racookpe1978]

Yes, NDE do "like" to call "things that they see" as cracks. See, if they did that, then you could blame NDE on having to cut out and replace "cracks" ... If they are "indications" - implying of course that thay are "indications of problems in the pipe found by NDE" - then YOU (the customer or the PM) have get to choose to cut out the problem area and fix it. Or grind out the pin point holes and fix them. Or grind out the RT (XRAY) flaw and fix it.

And, in a way, I can understand their reluctance to claim something has a flaw in it. Technically, they are "only indications" - in your case, indications typical of a lamination in the elbow wall halfway through the wall.

Are you really looking for an excuse NOT to grind out and investigate the probable lamination flaw tht large?

If you are reluctant, get an Xray of the elbow from both angles. Get a large diameter core drill and core out a sample all the way through. You'll have to reweld the hole, but it will be cheaper than replacing the entire elbow.

Get 6 more elbows from the same supplier (or, better yet) from the stockpile of not-yet-installed elbows and measure them inside, outside, and with RT through the walls.

 

[Bambie]

racookpe1978,

One excuse is that this elbow has been in service for 30 years carrying demineralized water at 264 degF with no sign of wall loss.

The other excuse is the normal operating pressure of 60 psig, requiring very little wall thickness for structural integrity.

The last excuse is that there are 12 other, identical elbows in service with approximately the same history.

Based on this history, would 'core drilling' the lamination and re-welding the hole pose less risk than monitoring?

Elbow replacement also poses the risk of losing 'self sprung' stress reduction in adjacent pipe and fittings.

 

[Sloss]

If you think these indications are new defects then you need to do an analysis to see if the system can still perform within design. However if they are new and growing then it's really a 'renew the elbows' situation.

It does look like these are Legacy defects, they have been there since day one. If so then I would Plan a monitor say every 6 months to see if the defects grow.

 

[weldstan]

Bambie,
One must ask why you were examining this elbow with PAUT? One would normally only examine for corrosion damage. Based on your operating conditions, why worry?

 

[Bambie]

Sloss,

In our plant, significant in-plane flexure of 6"nps, 90 degree elbows have produced hairline cracking along the inside surfaces of the cheeks.

These 18"nps elbows are on the inlet to an anchored pump with vertical pipe runs supported by variable springs and there is slight vertical movement that would produce torsion and out-of-plane bending.

We don't think these laminations are new and I can't correlate the in-service strains with a mid-thickness radial cracking mechanism in the cheeks.

weldstan,

Standard UT checks for FAC (flow assisted corrosion) noted significant wall thickness variation (because of lamination shielding), which needed PAUT to confirm.

 

[Sloss]

In my experience cracks tend to be surface breaking. We are not talking hydrogen embrittlement at this age.

You've had 30 years service so far. The Variable Spring arrangement should help smooth out the stresses and strains if they've been working and not seized half the time! (My guys even found some after 10 years that were never energised on an Offshore Platform.)

Laminations do mask the back wall for UT thickness checking but a decent NDT Op should know that and still give you accurate results. PAUT is only now becoming more common as it's expensive and requires special training etc., etc.

So unless you have changed the Operating Parameters of your system then it will probably continue to perform fine and you can do some monitoring to ensure the integrity.

 

[weldstan]

If the laminations preclude accurate sizing of the crack-like indication emanating from the lamination with UT, use RT to determine its length. If you cannot see the flaw with RT, don't worry about it.

It appears that you may have another issue. Namely that laminations prevent accurate measurement of corrosion loss. On the other hand you also indicated no corrosion loss, so I'm not sure that it is a real issue or not.

 

[Bambie]

weldstan,

Do you see a 'crack-like' indication emanating from the lamination in the attachment?

 

[weldstan]

I did not mean to imply that it was a crack. The flaw appears to emanate from the lamination and has some through thickness linearity.
It appears that you are uncertain as to whether it might be a crack and are concerned, should it be. I have simply given you another NDE method to allay or confirm your apparent concern.

 

[EdStainless]

Let's ask this another way.
If this was and internal lamination with cracks extending from it, given the service conditions what is the critical defect size?

The easier thing to do is to get an accurate picture and then come back in 6 mo or 12 mo and see if there has been any change.

All that any NDT can do is to indicate an anomaly. If you really want to know what it is then you have to do destructive testing. The proper NDT lingo is 'does the indication exceed the threshold established by the reference standard', because that is the only question that you can answer.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[metengr]

If you are that concerned, take a 1" diameter core sample and examine it using metallographic examination. Core samples can be easily removed especially being carbon steel and replaced with a threaded and seal welded RT plug. I really don't think it is that difficult and it will be definitive.

 

[racookpe1978]

65 psig distilled water at low temperature in carbon steel?

Yes. Core drill a 1-2 inch diameter plug. Tap the hole with NPT threads, then replace the plug with either a threaded plug, or a re-welded replacement circle. Or a threaded plug as above, then seal weld the plug.

Benchmark (engrave) each plug you are removing so when it falls out, you can tell which elbow it came from, which end was upstream and which was downstream, and from which side of each elbow.

 

[Bambie]

racookpe1978 & metengr,

Does ASME B31.1 offer guidance on core drilling elbows?

Would you invoke PCC-2 Section 2.1?

 

[metengr]

ASME B31.1 does not provide guidance for core drilling because this is a construction code and not a repair code. However, the Pipe Fabrication Institute does provide guidance for RT or inspection plugs;

http://www.pfi-institute.org/products.php