Arc Strikes-Continuing Discussion

By jon20013 (*****)

Date 05-02-2007 16:35

js55, with all due respect to EPRI, I say hogwash. Failures in service due to arcstrikes? Hard to say, although I'm sure we've all seen John Wright's photos of bend testing arc strike plates, very little to debate there. Certainly the process causing the arc strike would have some affect also, i.e., copper inclusions. Speaking strickly from a mechanical perspective, I cannot find it in my mind to simply ignore arc strikes. Whether or not they contribute to failures over the lifetime of an assembly is, as you mention, open ot a hotbed of discussion. There's my $0.02. ;-)

By swnorris (****)

Date 05-02-2007 16:50

Related article. Although it mentions EPRI: "Arc strikes and associated blemishes are acceptable provided no cracking is visually detected", we will continue to follow D1.1. I agree with jon20013 that arc strikes cannot be ignored.

http://www.steelstructures.com/StlInspNews/NEWS%20arc%20strikes.htm

By js55

Date 05-02-2007 17:34

Tell us what you really think Jon. LMAO!!!
My post was not meant to disparage a workmanship criteria. If it was, I wouldn't have created a QC Manual here at my plant with standards beyond that required by AWS or
AISC utilizing the exact same idea.
This post was meant to create a discussion and awareness of what can often be the extreme difference between workmanship criteria as opposed to engineering criteria, and metallurgical theory as opposed to engineering and service reality.
We all inherit patterns of thinking that must constantly be reevaluated in the face of new facts.
I will continue to reject arc strikes, but will no longer have so much heartburn with the possibility of missing a few.
I am annoyed at myself since I know I have this document and the significance of parts of it was apparently lost upon me the first time I read it.

By jon20013 (*****)

Date 05-02-2007 19:18

js55, you found my week spot, lol... I tend to be rather spontaneous at times but will always say whats on my mind even when it bites me in the behind... yeah, this fitness for duty thing has been around a good long while... I remember way back when Naval Joining Center (among a few others) were creating artificial undercut (which actually can be beneficial when done properly) and then the studies on how porosity can actually "strengthen" welds... I apologize if I sound less than forward thinking on some of these issues... lol! I still say hogwash to the arcstrikes and porosity issues even with the greatest respect for EPRI...

By js55

Date 05-02-2007 19:29

Jon,
No apology necessary. And I don't believe its a question of a lack of forward thinking. I mean, do we really want to accept ugly even if its engineeringly (how's that for a creative adjective) sound? No.
It came as a revelation to me even though all the facts had been staring me in the face for years. Even though I have read that EPRI document.
I think its a question of framing the discussions, the debates, and yes, even code requirements.

By GRoberts (***)

Date 05-03-2007 00:57

If you think grinding and inspecting arc-strikes is hogwash, you would have a hard time with some of the pipeline customers I have encountered. One such customer prohibited all arc strikes. If an arc strike outside the joint was found, the whole joint area including the arc strike was to be cut out and re-welded. If a grinding mark was made outside the joint, it was assumed that someone was trying to cover up an arc-stike and the whole joint and grind are was cut out. I don't mind so much when customers only want arc strikes ground and inspected.

By jon20013 (*****)

Date 05-03-2007 02:19

Greg, I think I'm aware of the criteria you speak (navy nuke?)... if it isn't I would likely turn my "hogwash" into B.S.... lol!!! Enhanced dangers of failure tragedy may be cause for greater measures of protection....

By GRoberts (***)

Date 05-05-2007 05:45

Nope, just commercial oilfield pipeline work.

By jon20013 (*****)

Date 05-05-2007 11:51

Holy crap!!!! It sounds like straight out of the nuclear navy....

By Mwccwi

Date 05-03-2007 02:21

Read this the origin is very reputable http://www.boulder.nist.gov/div853/Publication%20files/NIST_IIW_report_Siewert_2001.pdf Page 6 of 13 has an interesting diagram. :)

By js55

Date 05-03-2007 13:58

I did a quick scan of the article. I will read it in more detail later becasue it certianly seems important.
I would comment however upon what I read of arc strikes, for this context.
It seems their conclusion was solidly and justifiably logical, but still speculative, as is most of the workmanship standards associated with D1.1 inspection criteria, in that they anticipate a stonger inclination to cracking, and greater risk of stress corrosion. But still no hard evidence of the risk in which they are concerned.
Also, the HAZ of the tested weld bead was hitting HV's of over 300 at the start and over 400 near the stop. Not out of line with the arc strike.
The other thing is, they mention greater risk of stress corrosion. But that would depend upon the thickness of the material. The corrosive medium is going to be on the ID. Whereas the HAZ of the arc strike is going to be on the OD (unless of course you are perfomring a backweld), and it will probably not penetrate through the thickness very far. So what is it in the atmosphere that is going to contribute to corrosion?
This is very similar to other alloys and applications where we have conerns for OD issues that really aren't related to service viability since it will never come into contact with the corrosive medium, unless, for example, we're talking about SS at a desalination plant on the coast where there are real OD corrosion issues.
And this is really the point. Instead of a knee jerk response to arc strikes, I don't believe it hurts to take a real engineering viewpoint.
Having referenced the EPRI report, it should be kept in mind that their results would certainly not speak for phemomena beyond the alloys they tested.
But this is essentially the same argument I make. Making too general a statement form things that from an engineering standpoint should be specific.

By Mwccwi

Date 05-03-2007 19:00

Here's another article - maybe not exactly what your after but seems convincing enough for most
http://www.resnapshot.com/COR0900.htm

By js55

Date 05-03-2007 20:08

Mwccwi,
Not sure of the meaning of 'convincing enough for most'. My point has never been that changes in microstructure, as the article mentioned, do not take place, my point has never been that something doesn't need to be done, even if only from a workmanship standpoint, as I made clear in earlier posts, and my point has never been that the possibilty of cracks does not exist. So I myself am convinced of, something. And perhaps we agree more than what some posts seem to imply.
But I would ask the question in the context of the article posted: since the cracks in the article were discovered at 60X magnification (and at 600X mag the deepest crack seems to have a rounded tip which may indicate that the ductile BM has arrested the crack), why is it AWS itself does not require greater examination of arc strikes than, and I quote "ground to a smooth contour and checked to ensure soundness". No definition of what checked to soundness might be. And in the commentary to 5.29 says, and I quote, "MAY result in hardening or LOCALIZED cracking, and MAY serve as POTENTIAL sites for initiating fracture.(capitals mine)". Can "checked to ensure soundness" really verify that cracks do not exist? Or is AWS implicitly concuring with EPRI to a certain extent?
Look close enough you will find something. There has never been a weld made in a manufacturing environment that didn't have flaws or discontinuities in it. In fact, the very concept of weld denotes discontinuity.
The question is, are those flaws or discontinuities responsible for failures. EPRI found that in many instances the answer is no. I think to a certain extent AWS implicitly concurs. Otherwise with every arc strike they would perhaps work to find a way of examining arc strikes with much more thoroughness than "checked to ensure soundness".
If indeed my point is invalid then perhaps AWS D1.1 needs to add considerable robustness to their arc strike requirements. Because the language appears to be grossly inadequate.
My point is, perhaps not. I think maybe they got it right. Minimalist wisdom. Workmanship as a minimum standard approach which engineering facts can justify.

By Lawrence (*****)

Date 05-03-2007 20:30 Edited 05-04-2007 11:53

This is a good thread...... a good idea to keep talking about this.

I think where the rubber meets the road is cost.

Requiring a safeguard mechanisim such as a surface temper etch after every arc strike removal will be costly... and the higher the carbon content the greater risk that grinding itself will also provide positive surface temper etch indications.

Grinding arc strikes sufficiently to remove the hard spot may also reduce material or cross sectional thickness to such an extent that the item is outside of engineering limits. (As evidenced by the severe restrictions of our cross country pipeliners)

A reality that must be recognized is that code committees are indeed composed of experts from industry... those experts often factor the bottom line of their employers into the decision making process when it comes to tollerances....... Not a bad thing, the bottom line is important too.. Just something to be recognized.

After examining the *code welds* in modern buildings, bridges and parking structures destroyed by the Loma Preata, LA and Kobe earthquakes, FEMA and others did indeed dictate more robust low hydrogen electrode regulations. Those welded structures may not have had cracks, but they did not perform to design either... It would be hard to sift through rubble to discover if Arc Strikes were a factor too but we sure do know they change the quality of the surfaces they are struck on.

By downhandonly (***)

Date 05-03-2007 02:58

I just came off of a job like that. just imagine how hard it is to avoid arc strikes when you have three bead hands, two hot passers, two hot fillers, and two cappers all taking turns on the same joint. there were a few cut outs but the inspectors let us use sanding discs on the minor ones as long as they were near the bevel.

By JTMcC

Date 05-03-2007 04:35

GRoberts hit it right on the head, arc burns are a cutout on a pipeline. Cutouts are very, very bad. Welders get fired. It's just not even debatable in that world.

JTMcC.

By 803056

Date 05-03-2007 02:40

It's been a long day, but I will read the article tomorrow!

Al

By DaveBoyer (*****)

Date 05-03-2007 05:40

I will mention again in greater detail an incident from the auto frame plant: On one of the line stations there were arc welding operations going on while riveting operations were being done. The hydraulic rivet yokes started breaking much more frequently than normal. To make a long story short the reason for the increased breakage was traced to an arc impingement in the tensile area of the yoke caused by a faulty grounding strap on the fixture. The welding current was finding ground by arcing against the yoke which was grounded by its supporting hardware. The yokes were AISI S5 tool steel, a fully hardenable oil hardened material. The arc heats a small area above the critical point, and the cold mass of the surrounding material quenches the heated material, leaving untempered martinsite, and the arc crater causes a stress riser in the brittle hard area. The combination of the 2 greatly lowers fatigue resistance. Now this may be an extreme example, being tool steel, but the posibility does exist in ANY heat treatable material.

By bozaktwo1 (***)

Date 05-05-2007 15:28

I'll keep this short and in the realm of my experience.

Would you, as a customer, buy an item of raw stock in any shape with any visible discontinuities present? Then why would you buy off on an ark strike? In my experience (submarine and surface ship hull work, MIL-STD 1688/1689) any arc strike is to be removed by grinding, then refilled. If the strike is outside the original HAZ of the joint, I had to treat it as a new joint or a repair attempt.

Now, my personal belief on most visual acceptance criteria is zero defects. In practice, I try to follow that, commiserate with the application I am performing the procedure on (i.e. farm equipment vs. submarine hulls). If I am looking at it from the point of view of a customer, I want it pretty, and I want it to work. If it's a high risk of failure item, I want it safe.

By ssbn727 (*****)

Date 05-06-2007 05:01

I concurr Bozak!

The USN is especially serious when it comes to arc strikes anywhere in or outside of the "Boat" or as everyone else call them - ships!!! I wish I had a nickel for every arc strike I repaired on submarine hulls or inside either within the pressure hull or in the aft or foward main ballast tanks... Then there's the arc strikes that hold pipe hangers or any other weldments that the Navy inspectors wo'nt sign off on so in go the mirror welders like myself that make these type of repairs in some of the darndest places... Cost overruns anyone?

Unfortunately, some of the folks that worked @ EB were classic cases of the wrong person for the job and some of them were basically NUTS!!! I mean, if they got suspended for missing time or insubordination, these poor excuses for human beings would literally commit sabotage within the confines of these submarines!!! The end result was costly repairs to total rip outs of parts, components, piping to complete trim ballast bottles that were situated in between certain hull frames in various compartments within the subs!!!

If they caught these scum or found evidence that they were implicated in doing these childish acts of intentional destruction of government property then, they were handled appropriately and prosecuted to the fullest extent of the law!!! This type of sabatoge was one of the main reasons why the Los Angeles and Trident/Ohio class - attack & FBM's took so long and cost so much to make asides from the fact that the technology incorporated into these boats at the time of their construction were at that time -state of the art and the prototypes just took longer to complete -PERIOD!!!

Now do'nt get me wrong folks!!! Most of the personnel that worked in hte yards @ EB were very patriotic and focused in their work there... However, all it takes is few isolated incidents where intentional acts of revenge or outright stupidity to delay and increase the costs of such a sophisticated compilation/integration of very complex systems and subsystems that make up the components and parts of a nuclear powered submarine!!!

In summary, after the loss of the "Thresher" and the "Scorpion" submarines of the USN, Rear Admiral Hyman Rickover instituted the USN's SUBSAFE program of quality control that is very strict as Jon mentioned... In fact many civilian code governing bodies have adapted many of the programs strict controls...
That is why most of the ASME nuclear code standards originate from similar standards used by the USN's SUBSAFE standards... Walk into any commercial Nuke plant in the US and most of the folks working there are ex-navy nukes!!! Did anyone ever wonder the reasons why??? Figure it out!!!

Respectfully,
Henry

By 803056

Date 05-06-2007 16:44

The differences in opinions between the practitioners and the academics are legendary. In all fields, in all applications, the subject does not matter; the academics will perform experiments that they hope will mimic "real" world conditions. Sometimes they are successful, sometimes they aren't, the results are published never the less and we, the practitioners, all sit back, aghast at their conclusions.

I remember attending a symposium on bolted connections. The speaker, an academic, stated that it doesn't matter if the bolt is properly tightened, as long as the bolt does not fall out of the hole, the final load capacity of the connection is unaffected.

Statements like that cause my hair to stand up on end. However, if you think about it, the final failure mode is that of the bolts in bearing. That is, if the bolted connection is classified as slip critical, when overloaded or if not properly pretensioned, the connection will slip and the final failure mode will be as a simple pin connection with the bolts in bearing. So, whether the bolts are properly pretensioned or snug tight isn't going to affect the ultimate load carrying capacity of the connection. I assume the assumption is true if the loading is a one time event and not cyclic.

All well and good in the laboratory where workmanship issues rarely arise, but not so in the real world where codes are applicable and aesthetics do have to be considered to satisfy a paying customer. Remember, "He who holds the gold is king."

Best regards - Al

By darren

Date 05-06-2007 19:16

some of the third party inspectors that go over our final product will let some obvious flaws go through the door but stop the advancement on the TINIEST of arc strikes because for whatever reason the purchaser of the product has it in their mind that the entire thing will fail if there is an arc strike. and in the end it doesn't have to make sense we just have to get paid, and its the customer who is always right.
after reading the thread there are obvious places an arc strike would matter but i am sure it is a lot less then standards would dictate. have a friend in russia right now and they butt all their pipe together weld with 7018 and if it leaks weld over the leak. he says its crude but seems to be working. so somewhere between the too much control sometimes exercised in the first world countries and the total lack of control in second and third world countries would probably be the best.
but i am just a simple welder and have to rely on the experience of the industry as a whole to keep me safe and keep the products that i work on effective and safe for the end consumer. so no arc srikes is no arc strikes.
i just wish there was a way to impress people in the industry to be more careful and not let the arc strikes happen in the first place, because repairing one is not so bad but repairing hundreds from careless tacking drives a guy nuts. (a lit rod is not a flashlight to be used to find the next tacking spot while your helmet is down)
darren

By 803056

Date 05-06-2007 19:49

"What do you mean it's not a flashlight? How else am I going to find the groove? Have you ever worn a hood? You can't see nuthin until you strike an arc on somthin!"

How many times have you heard that?

Best regards - Al

By JTMcC

Date 05-06-2007 22:16

darren says:

i just wish there was a way to impress people in the industry to be more careful and not let the arc strikes happen in the first place, because repairing one is not so bad but repairing hundreds from careless tacking drives a guy nuts.

I says:

There is and it's been effectivly employed for decades in cross country pipeline construction. Arc strikes are a cut out, repeat offenders are fired. It's a simple system.

JTMcC.

By dbigkahunna (****)

Date 05-07-2007 13:11

Dot 192 and 195 give guidance on arc burns. They do not prohibit it. They must be removed by grinding and testing with ammonium pyrophosphate to ensure removal.
However, there was a guy years back who ran a training school for pipeline welding inspectors who had a section of 16 inch 0.250 wall 5L X80 pipe with an arc burn on it. It was feathered with a sanding disk. The AB could not be seen. Spray it with FDP and hit it with the black light and it looked like you hit it with a hammer. A dwell time of 5 minutes and look at the area inside the pipe and the cracks had gone all the way through the pipe. The pipeline industry has learned arc burns bad. There was a pipeline company based out of Chicago who required the contractor to remove a pipe diameter either side of the arc burn and charged them for the pipe! You did not see AB on their pipe. Arc burns are a workmanship issue. They happen once in a while. But if the inspector or QC is seeing a lot of them, it is because the welders dont care! And if the welders dont care, it is because the company and management they work for does not care about craftsmanship.

By 803056

Date 05-07-2007 13:42

Nothing drives a lesson home like a demonstration or an example of the topic being discussed.

I would love to see the pipe.

I have performed PT and MT on some arc strikes. On occation I've seen some small surface cracks, but none that had any appreciable depth. Seeing is believing and anytime you can show a real world example, it takes the wind out of the nay sayer's sails.

Best regards - Al

By js55

Date 05-07-2007 16:04

Many of the arguments in this thread seem a bit messy to me. There seems to be a consensus for a real concern for arc strikes and the catastrophic damage they can cause. Yet, at the same time an implicit assumtion that the AWS requirements are acceptable. Can visual examination really reveal the extent of cracking?
I would have to argue that an acceptance of AWS's requirements for arc strike evaluation is not consistent with a maximum concern for catastrophic failure due to arc strikes. I would say that the quoted pipeline or military specs ARE consistent with this concern. And I would conclude that I think the truth lies in the middle somewhere and that concern is engineeringly justified based upon material and application. In other words, sound engineering judgment. In other words, consistent with AWS (do we really want D1.1 to break out each possible application and material seperately?) and EPRI (should EPRI be concerned with military applications?). Do we want arc strike requirements on non bearing columns or beams, mini storages, or plant air and water lines to be consistent with main steams, submarines, and natural gas lines?

By dmilesdot (**)

Date 05-07-2007 18:27

Could it be that the article from EPRI was written with in-service inspection in mind? Typically in service inspection is only concerned with I.D. connected flaws.

By js55

Date 05-07-2007 19:36

David,
I would, honestly, have to leave that possibility open. Since, my sources for what the research actually says is not my own reading, unfortunately, (because I have as yet not found the D#&$ thing in my library, although I know I have it somewhere), but quotes from other knowledgeable individuals. Though until proven otherwise I doubt it, considering the title of the document is "Visual Weld Acceptance Criteria", with no mention of Section 11. And in small print at the top of the title page it states as topics "Nuclear Power Plant, Construction, and Inspection." Nothing about Inservice Inspection. AND, in my opinion seems to be very consistent with D1.1's approach where grinding and visual inspection is acceptable. And since the idea was origianlly expressed to me by a gentlemen who sits on D1.1 it is clear that at least some of the committee members have read and assimilated the document and accept its conclusions. Perhaps more than a few.
But if so, what an interesting tangent. Though I know this wasn't your intent. For, if a flaw is considered irrelevent in accessing the ID in a pressure boundary application such a piping, is that flaw relevent at all? Doesn't every compromise of a pressure boundary have to access the ID at some point? Even if it is a sudden catastrophic failure. Would EPRI, under a study for ID emphasized Section 11, as you state, conclude that arc strikes are irrelevent to inservice inspection if catastrophic failure were possible?

By CWI555

Date 05-07-2007 21:13

I believe Section III article G-2000 will shead some light on the subject.

As I recall, Section XI will kick back to III or, the acceptance criteria be be stated in linear or non linear criteria or, as a factor of stress K.
I do not believe Section XI directly mentions arc strike, but rather one of the three derived values above.
Site procedures will many times reference arc strike as a fracture mechanics derived criteria (stress K)
I may be proven wrong, but I don't think epri would say Accept arc strike cart blauch, but rather as a factor of evaluated stresses one way or another.

By 803056

Date 05-07-2007 21:19

One consideration that hasn't been mentioned (at least I didn't see it) is the condition of loading, i.e., is the load static as in the structural frame of a building, or cyclic as in a bridge or piping system that undergoes thermal cycles or variations in pressure?

An arc strike adjacent to a weld that is subjected to static loads is not going to be a major concern. If the load is static and the arc strike is in an area subjected to compressive loads, it is even less concern.

The same can not be said for an arc strike on the exterior of a pipe. The vast majority of piping systems are pressurized, thus the OD of the pipe is subject to tensile hoop stresses. Most piping systems in a power plant or chemical plant are going to undergo fluctuations in pressure or temperature or both. A bridge structure or supporting structural framing for a crane is going to be subjected to cyclic loads. The piping system, bridge structure, or the crane system are the "worst" case for arc strikes located in areas subjected to cyclic tensile stresses. Anything resembling a notch type discontinuity is going to affect the service life of the structure. The notch could be undercut, overlap, changes in geometry, and yes, metallurgical notches associated with arc strikes and the HAZ of the welds.

This sounds like a good project for someone with access to a fatigue testing machine. Even if there are papers that already have looked at the problem, it would still be a good exercise.

I do a demonstration for my courses with E70S-X filler metal for GTAW. I cut 1/8 inch diameter wire in to 18 inch lengths. One third of the group is left round. The second third of the group is filed with a flat in the center of the length on one side about 1/32 (or less) inch in depth (no sharp corners mind you, blend the flat). The last few rods are nicked in the center of their length with the corner of a rectangular mill file so the notch is the same depth as the rods that were filed with a flat. 1/3 of the class get rods that are round, 1/3 of the class get rods that have a flat, and the lucky few get the rods with the notch. Each group grasps their pieces of filler metal at the ends and bends them back and forth until they break. The round rods provide a real workout. Make sure everyone tracks how many times their piece is bent back and forth. The results are surprising even to those that already recognize the notched rods will fail first. It drives the point home of how much influence a notch or change in geometry has when subjected to cyclic loads.

Best regards - Al

By dbigkahunna (****)

Date 05-07-2007 21:36

I have never seen a crack on low grade pipe such as Gr-B or X 42. Where the problem shows up is in X60 and greater pipe. The same with plate materials. The low grade mild carbon steel is a very forgiving material. The higher grades are not.

By 803056

Date 05-07-2007 21:59

Good point.
Al

By DaveBoyer (*****)

Date 05-08-2007 03:06

Yes, This is the whole point, in extremely ductile materials there isn't much notch sensitivety, if the ductile material is low enough carbon that it won't harden from heat & quench, the arc strike is insignificant.

By js55

Date 05-08-2007 14:43

Somebody is just going to have to get their hands on one of these EPRI docs and determine what materials were the basis of study. One thing I think we can assume is that it will be a piping material. I can't find mine and this is just too good a subject matter to let it go.

By Tommyjoking (****)

Date 05-10-2007 08:09

arc strikes are not permitted in any aircraft work.......simply I presume for the possibility of stress cracks forming in the parent material....sure sucks ass when im doing aluminum

By bozaktwo1 (***)

Date 05-11-2007 00:20

First Al, let me say that I laughed long and hard at your previous post in this thread. Too true!

Your biggest mistakes are the ones that you remember best. Mine was an accidental strike on an insert made from HY80 over 1" thick. On grinding, I figured I could just feather it in, and be done. Oh no, never so simple as that. It was over 1/2" deep! It wasn't in the s/s boundary, however it was a level1 component and I had to dish it and fill it completely, grind it smooth and mt the entire weldment. Just because of one split second of stupidity.

By cremx

Date 12-15-2007 18:29

We have discussed with a contractor and the owner the acceptance of the method to inspect and remove the arc strikes on 9% nickel plate for an LNG tank; the procedure to inspect and repair is LP after the grinding, however our concern is the metallurgical damage rather than the microcracks; we believe that the inspection done is more than the AWS requirement

By metal_monger Date 12-16-2007 01:55

this is refrenced by d1.1