Recently a Major North American ERW Pipe Manufacturer had a spate of weld line cracks that did not fit into any familiar or normal terminology. In fact, both the terms oxidized and un-oxidized were used to describe this indication. The UT inspections show a condition very similar to the “mill cracks” other Line Pipe Manufacturers are famous for when describing atypical indications and who can never tell an Inspector what causes them, but just tries to send them back through the system. The Metallurgist, a damn bright guy, told me that he had tried to upgrade conditions to be in line with other Manufacturers. Naturally, as you know, most Inspectors have an almost photographic memory and are copious note and mental data collectors. I was determined to quietly observe, discreetly take photos, and ideate through this, sometimes maybe a little too wrapped up in the investigation. The Mill’s end millers are pretty good for normal machine shop work, but not heavy duty enough or clean enough for the application. I still prefer the old method of using knives used by a lot of other mills at the trim line. After much observation, consulting with friends in the Precision Manufacturing & other Manufacturing Sectors, and studying the photos, I realized what we probably all know. The end miller lacks the capabilities to completely remove the entire chip, and the “wiper” technique used is lacking. Chips remain clinging to the edge of the skelp and either turns down akin to a lap or turns up and is melded into the weld. Photos show these conditions. My concern is the down turned chips and the laminar indication developed that show up on UT. So, I ask is the potential for failure while in service enhanced by this or does it not matter (especially in an area where a near UT hit is found).
Quite a few of these are rejectable indications per N5 notch and 1’8″ hole.
Should I be concerned because the UT operator also gets a lot of close hits, proves some up with MT, and lets some pass?
Any thoughts on this?
I am writing a paper on this, discrete as it should be to specifics, and plan on presenting it for publication
And the collaboration between my Mentors: Miles Free, Chris Schoen, Allen Book, John Bowen, and myself.
The first one looked like a black weld from shorting across the vee — just what a chip or whisker would do if it contacted both sides of the vee away from the crotch.
Of course you should be concerned, but you’re probably limited by the customer spec, and API 5L doesn’t cover near misses. If you can get them to also look at some near misses with UST in static mode, and search for max amplitude (same as prove-up for exceeding threshold), it would provide more confidence.
Black is bad and I don’t care what they say. Generally, that indicated a fusion problem of some nature, especially in the ERW field. Naturally that’s why it is frowned upon in the Pressure Vessel field. Pipelines still love it due to price. But what is a pipe, except the most prevalent pressure vessel in use around the world. The oxidation which can be a cause of the blackened areas also leads to the poor fusion in the welded areas. Any type of resistance welding application MUST have the proper “LASH” which is my way of saying the 4 things controlled by welders and amounts to the success of a good weld. Length, angle, speed and heat. Length of the arc, angle of the electrode, speed of travel and heat on the machine. These variables are critical in the ERW process. If the wheels are not prepared properly (length and angle) the weld will suck, if the process is not set to procedure (speed and heat) the process will suck. With ERW, those factors must be controlled especially tight. I used to make the statement that I could teach a welder to make an automated weld with enough stalks of bananas but that really was a joke. PQR’s are critical for any welding process and just looking at those pictures with no history, it appears that somebody did not follow procedure very well.
Give me some more details and I can go further but from the photos, it appears to be a bad fusion problem.
Boiler and Pressure Vessel systems, Nuke and some pipeline companies get extremely concerned about the “leftovers” that you are picking up with UT. Most Pipeline Companies could give a rat’s ass if the line pipe meets API specs and rightfully so since the hoop stresses are covered by the strength requirements in the specs. API can really screw it up when they do not have the pipe you order and can substitute a stronger spec since it meets or exceeds the original order. If they are out of X60, they can legally substitute X70. Really screws with the electrode groupings on a PQR/WPS.
If the piping system is subject to cyclic loading, I believe the concern should go well beyond the Norm for pipelines as appears to be your case. I did a 2 year stint with an EP out in West Texas and saw a lot of what you are working on but only because the metallurgist on the project at the time went into enough detail and concern to do some testing and find out why the mechanisms failed and do a root cause. Most of the piping showed the same discontinuities you are seeing. Neat research but costly if not a necessity. A lot of their failures were due to soil type which caused accelerated corrosion as well as erosion due to water being in the lines and the hilly terrain out there. After a period of time the water would build up and create a blockage effect in the line and wash back and forth causing the erosion internally. When it reached the discontinuities in the ERW welds, everything speeded up and failure occurred prematurely. And in some cases, catastrophically.
I believe to some extent, the mills are at fault due to the end results of getting the pipe out the door. All the reasons you cited and the fact that ERW sucks as a welding process unless the parameters are tightly controlled. There is limited research in these areas to support our concerns and you may be on to something really big. Most likely, the lack of research is due to the lack of funding to support the time involved. The military does some in the area of explosion welding which creates the same type discontinuity in a lot of their cladding applications when viewed under magnification. IE: aluminum clad to carbon steel. When magnified, it appears like small cat claws in the surface of the carbon steel. Failure normally occurs at the interface of these discontinuities. DUH!!!!!!!!!!!!!!!! Classic stress risers.
Solutions to Edge Oxidation
Solutions are either mechanically remove the oxide, use a chemical etch such as deoxidine, or use nitrogen as the assist gas. Each of these has a price. The first method is labor, the second is labor and chemicals, and the third is cost and possible loss in productivity. When cutting with nitrogen, there is a point where the feed-rate becomes slower than that with oxygen and the amount of gas used is dramatically more.
©Donald Crusan 11-Nov-14