Wrought Carbon Elbows

[mizzoueng]

I have an A234-WPB wrought carbon elbow on a high pressure high temp steam line. The elbow has inclusions in the base metal, not the welds. I am attempting to determine what the Go/No-Go acceptance criteria are per code, but there seems to be no real code covering this.

Anyone have any experience or advice in this area?

 

[Supe]

The individual codes don't cover raw fittings, the SA spec does. For inspection and NDE criteria, SA-234 will refer you to SA-960/SA-960M for acceptance criteria. It's pretty clear cut regarding surface defects. Additional types of NDE other than visual criteria are also given, but are only supplemental, and aren't applicable unless specified in the purchasing agreement.

 

[mizzoueng]

Yeah, I saw the surface defects criteria, but these are inclusions inside the base metal. So below the surface. That is why it is so tricky.

I have the original contract (1978) but it has no mention of acceptance criteria. I am beginning to think I need to just use the criteria laid out in B31.1 for welds on these inclusions.

 

[Supe]

How were the inclusions discovered? You say they're subsurface, so were they picked up on RT, UT, or... ? What sort of dimensions are you looking at for these inclusions? If it's definitively in the base metal (and not in the weld end prep), and the defect meets the SA spec requirements and does not violate min wall, then you're a bit tied outside of a judgment call.

 

[mizzoueng]

We first found them while performing UT thickness (D-meter) readings. The meter used is a point -type, so it was insanely hard to map out the exact size of the inclusions. I know the general area where one (or a cluster of many) is located.

The thickness readings at the points where the inclusions (voids, non-metallic, etc) are violate min wall, but using UT Shear Wave we know there is metal underneath these spots because we could "bounce" past them. Readings around the spots are about 1/4" thicker than min wall, so we know that the surface-to-surface dimensions meet min wall, but how deep the inclusions are is the mystery and the problem. If they are 1/16" deep I believe we are okay, one of my co-workers found something in ASME Section III, but it was for materials over 2" thick. I am going to see if I cannot find the same for mine today.

Either way, I believe I am outside jurisdiction on all points. everything I have read to this points indicates that the criteria should have been set up by the client (or the clients engineer) prior to fabrication. There are no records of soundness or criteria on file, so more than likely they just said "make it sound" and did not give them criteria. I doubt the manufacturer actually inspected the fitting either. If they did they thought it was okay when they shipped it out, but no inspection papers came with it (or at least they are not on file).

 

[Supe]

If it falls under NBEP (non-boiler external piping) for B31.1, then you're probably going to have some difficulty finding anything on them after 30 years as that documentation wouldn't be required for turnover to the client like it would for ASME I/BEP. Section III Nuclear won't do you much good. May as well just go by the weld reject criteria as you've mentioned.

 

[mizzoueng]

So its a lamination in the elbow. Again, not in the weld. So now this is a whole other ball of wax. The piping falls under B31.1 BEP as there is no stop valve in the line between the superheater outlet header and the turbine.

This is so far above my pay grade............

 

[ChitownPE]

Laminations in pipe base material that are stacked circumferentially are typically non-injurious. There is nothing in the Code that will provide guidance for this. If the elbow was accepted for use during construction, this means there was no supplemental requirement for checking for inclusions. Second point, I have come across this before where our older steam lines have stacked inclusions based on thickness testing. Bottom line, these steam lines have been in service for many years with no problems.

 

[Supe]

Laminations are a bit of a different story than other forms of defects. If it was porosity or something that could have been linked radially through the wall of the fitting, you have a greater likelihood of failure. When laminations are stacked, the defects are oriented opposite to the hoop stress that the fitting sees, which reduces risk of through-wall cracking as I believe Chitown was alluding to. If by some chance there was a trunion or some other attachment welded to it, I might look at the situation a bit differently though.

Of course I feel obligated to reiterate that I'm neither a pipe engineer nor a PE, but I did stay at a Holiday Inn Express last night, and unfortunately get to deal with a lot of this crapola being a welding guy.

On a side note, we manufacture our own fittings in house, but specifying these sorts of details in the contract stage is becoming increasingly important due to the crap coming in from China. Stuff like this is usually the easier stuff to pick up. Where you get into the real fun is when you start receiving fraudulent CMTR's and bogus postweld heat treat reports...

 

[mizzoueng]

Well, lots has happened on this forefront. We did extensive ultrasonic examinations and found that all the discontinuities (notice the big word?) were NOT laminations. We were able to blow the signal (ha!) through the area and pick it up on the back reflection (signal off the ID).

What does this mean? it means we have 29 locations in a 22" elbow that are just locations where the metal was poorly formed. It could be sand, dirt, low carbon areas, silica, anything. It is NOT a void though, voids would bounce back right away, these did not.

We also found that they were all circumferential and none were overlapping.

They are all subsurface, all at relatively the same depth (0.300" out of 0.680") and are all the same thickness (0.025").

What does all of this mean? It means the fitting was made of crappy steel and that the combination of high heat forging and a possible high heat post weld treatment caused these "pockets" to open up and bubble inside. Does this effect the soundness or the pipe? Probably not, but there is no way to know until it fails in service.

So we are replacing it. With 7 days left, we are JUST NOW deciding to replace it........ arty-smiley-048:

 

[Supe]

Just be glad it's only .680" wall. They can knock that out in 2 shifts. Then you get to pray that it RT's good!