UT of Welds when UT is not specified on drawings

Date 09-12-2012 16:40

Just curious....are the same welders involved with both UT flagged welds and Backgouge flagged welds? Just wondering what was done different on the backgouged welds to cause the trouble passing UT inspection.

By Bill M

Date 09-13-2012 19:27

Good question John. I will have to ask
I am just getting involved with this one...(long story)

By jwright650 (*****)

Date 09-13-2012 20:02

I'm going out on a limb and guess that they were not the same welders.

I hate it when these things go South and there is resistance to get the joints cleaned up and sent on their way. I'm sure their customer doesn't want CJPs with reflectors, so why the resistance to repair them?(spelled out in contract or not, just make it right)
People mess up every now and again and miss something during the backgouging....it happens, just fix it and send it on. OK, I'm done ranting.

By welderbrent (*****)

Date 09-12-2012 20:07 Edited 09-12-2012 20:09

Bill,

To take the customer's side as "Devil's Advocate" a little bit here...

Without having a copy of your plans to look at the first thing I would suggest is a thourough examination of the GSN as well as the Job Specs. Make sure of how all those items are called out in any of the Engineer's Notes. There may be something there that makes the intent clear, items that often get missed by the fabrication shops.

Second, If the WPS's as well as drawings are all done according to D1.1 and A2.4 then the question would be if the joint in question was a Pre-Qualified weld joint? If so, then you must compare with Details in Clause 3 that are Pre-Qualified. For PJP's there should be notation as to groove depth and weld depth. But, as none of the PJP Pre-Qualified joints show 'Backgouging' as a method of welding from the 2nd side, it would be more properly interpreted to be the intent that it fits the description of being a detail of a Clause 3 CJP with "Backgouging" to complete the weld from the second side. All of the welds with backgouging that are Pre-Qualified are in the 3.4 CJP section. Is there a specific Detail, such as TC-U9a-GF, called out? If so, that would cause one to consider what that detail specs out in ALL of it's various aspects beginning with the fact that it is a Pre-Qualified CJP Groove Welded Joint Detail. Rather answers your question right there.

Third, why would one call out a joint with backgouging without intending for it to be CJP? I would just start with an edge that had a bevel on it that would allow a partial penetration weld if I did not WANT CJP. Cleaner, faster, easier.

Now, I don't think one should jump to the conclusion that we must define "gross nonconformance to the code". One must consider though rather an RFI would have been in order had there been any question at all as to the interpretation/intent from the engineer. But if all involved at the Fabrication Shop simply assumed PJP but with Backgouging that made it come close to CJP then while it was an error in interpretation it is not gross nonconformance. There are prescribed methods for remediation of such conditions. May mean repairs made at your cost. May mean determining why every one of those joints has problems at the root. It would seem some of them should have passed if they were even close to being backgouged to CJP.

The customer, through his engineer, along with testing personnel and welding supervisors will have to come up with a plan to procede.

Have a Great Day, Brent

By pipewelder_1999 (****)

Date 09-13-2012 03:38

I would start with a thorough review of the project purchase order, the drawings and specifications referenced by that purchase order, and verify that there was not a requirement for UT of all CJP welds. Within those documents there should be guidance for the order of precedence for conflicts (Drawings over specs, specs over drawings etc.). What is contained in those documents is what you sold.

Anything above and beyond that is in my opinion, outside the scope of the original contract and should be fought out by those "contract guys" however keep in mind that the customer has a valid concern and the perception that may surround any attempts to just follow the contract may hurt things in the future. Their lack of interest in what is contained in para 6.6.5 could make things difficult.

There is reference to backgouging and verification of the root face removal by MT in table 6.2 and 6.3. Is there any type of inspection done on the backgouged joints prior to depositing the second side? Even the NON UT joints. Just a curiosity on my part.

I have been on both sides of the "additional NDE" and some weird specs calling for all groove welds to be CJP and all CJP welds to be UT 100% and UT only specified on SOME welds on drawings. Point- Sometimes detailers make some assumptions about how the fabricator will interpret welding symbols. And of course sometimes they have no clue what they are drawing. Worst part of a weld symbol is "TYP".

I have been jumped on by fabricators for sliding a camera down a column to take a picture of a splice in a plate that closed in one side of the beam. I asked them if it was fitup to D1.1 and the WPS. They said yes. I showed them the pictures. The joint was designed and detailed as a CJP Single Vee Groove Weld on a backing strip. There was no backing strip. No Root opening, and no evidence of any welding near the root. They said taking pictures was not in the "spec". I forwarded the pictures to the end users with my findings and let them work it out. I just report what I see. It meets the code/spec/etc or it doesn't. These columns supported a large amount of air pollution control equipment at a power plant and these plates boxed in the flanges of the columns. Though the camera inspection wasn't part of the project, the splices were all cut out and rewelded and a UT step was added.

Point is, if the welds need the level of soundness provided by UT to either 6.2 or 6.3, then they should be inspected to provide that and they should have been clearly detailed or specified. Who pays for it ? Who knows but make sure that its isn't paid for by someone killed or injured.

Understand, I have also seen the extreme where flat panel stiffener splices on ductwork required 100% UT to table 6.2. Extremely silly for ductwork. But I reviewed each ut report, and visually inspected each weld. Found quite a few splices with UT Acceptance documentation, UT OK written on the joint, and insulation installed prior to my inspection. I forced the insulation and lagging down an guess what, visible 1/8" deep unfused area on the underside of the accepted UT . Hmmmm.

If welds are specified to be backgouged, that is understood to be to "sound metal" . If the welders are not doing so, it could be because of lack of training or the inability to know what they are looking for. Sometimes depth only does not get rid of the root face(s) or areas on incomplete fusion at a bevel. Subsequent grinding or lack thereof can mask this area.

I may not have helped your issue but you did just ask for "comments" :)

Have a nice day

Gerald Austin
Iuka MS

By eekpod

Date 09-13-2012 10:20

You need to read and referance the specifications and contract drawings for the project. In there it will or should say which welds require UT inspection. It is common on our projects that all CJP require UT, not some here and some there, but it could happen.
Then expect that it was a detailing error, and if you have the chance to ask the detailer don't be surprised if they missed it or didn't realize.

By jwright650 (*****)

Date 09-17-2012 16:07

Hi Bill,
I ran across something...not sure if it applies to your situation, but may help if it does....

AISC manual of steel construction 14ed page 16.1-177 "Min requirements for inspection"
paragraph 5B CJP Groove Weld NDT....spells out NDT required for structures in certain risk categories

By 803056

Date 09-16-2012 18:58

We are being asked to offer an opinion based on limited information. We do not have access to the project specification or the contract drawings. We do not have access to the shop drawings. In order to draw a conclusion there are certain assumptions that are necessary to fill in the blanks.

It is pretty clear that a specific class of welds completed by the contractor failed to meet the code and the customer’s expectations.

Under the auspices of AWS D1.1-2010 it is the Engineer’s responsibility to provide the fabricator with the information necessary to construct a structure that will meet the service requirements of the Owner. Clause 1.4.1 assigns the Engineer, representing the Owner, the responsibility to develop the project specification and structural drawings to provide the requirements needed to ensure the structure will meet the needs of the customer and the applicable codes. The Engineer is given the authority and latitude to modify the code when and where necessary to ensure the end product is safe, sound, and meets the utility needs of the Owner. In the same clause there are a number of items that must be addressed by the Engineer. They include, but are not limited to:
(1) Code requirements that are applicable only when specified by the Engineer,
(2) All additional NDT that is not specifically addressed by the code,
(3) Verification inspection when required by the Engineer, and
(4) All additional requirements that are not specifically addressed by the code.

Likewise the code assigns responsibility to the contractor. Those responsibilities include, but are not limited to, performing the work in conformance with all the requirements of the code and contract documents.

AWS D1.1 does not grant the contractor the right to produce nonconforming welds when additional NDT is not specified. Nor is the contractor granted the right to produce nonconforming welds when verification inspection is not specified. The basic presumption of the code is that all welds comply with the code whether the Owner provides verification inspection, volumetric NDT or not. It is the contractor's responsibility to take whatever steps are necessary to ensure all the welds comply with the code. It is the contractor’s obligation to produce materials that comply with the contract documents and the code. Failure to do so places the contractor in a position that is nearly indefensible. The arguement "I would have done a better job if I had known it was going to be inspected" is never a strong defensive position.

When a back gouge operation is specified, it is presumed that the welder excavates all unsound metal to the depth necessary to reach sound metal. Inexperienced welders often make the mistaken presumption that they can leave traces of the unfused root intact because the “next bead will burn it out”. Subsequent NDT usually proves the assumption to be incorrect. In this case it is clear that the welders knew what procedures were necessary to produce welds that could pass UT. However, it is clear the same level of care was not taken when the welders knew the weld was not subject to UT. In other words, the welders knowingly produced welds that would not meet the soundness requirements of AWS D1.1

The fact that the welds identified as being subject to UT passing the NDT points to a situation where the contractor intentionally defrauded the Owner by not properly back gouging the double groove welds. The fact that the contractor was able to produce welds that met the soundness requirements of AWS D1.1 shows the contactor had the ability to deposit sound welds. The fact that the welds not identified as being subject to UT were not welded with the same care and with the same technique as those welds subject to UT indicate the contractor willfully and knowingly produced substandard welds.

If the failure rate of all the welds was about equal, I would not support the position that the contractor knowingly produced substandard welds. However, in this case the Owner is correct in demanding all welds be tested and he is justified in demanding the contractor absorb the cost of making any necessary repairs. I would also support the Owner’s position that the contractor should pay for the added cost of examining all the welds with UT.

Best regards - Al

By Bill M

Date 09-18-2012 13:29

PLease comment on what is the intent of this paragraph from AWS D1.1?

"If NDT other than visual inspection is not specified in the original contract agreement but is subsequently requested by the Owner, the Contractor shall perform any requested testing or shall allow for any testing to be performed in conformance with 6.14. The Owner shall be responsible for all associated costs including handling, surface preparation, NDT, and repair of discontinuities other than those found in 6.9, whichever is applicable, at rates mutually agreed upon between the Owner and Contractor. However, if such testing should disclose an attempt to defraud or gross nonconformance to this code, the repair work shall be done at the Contractors expense.”

By Bill M

Date 09-18-2012 14:16

So why do you think there are two accept / reject tables, two separate levels of weld quality requirements?

Based on extending your logic, every full penetration weld then should pass table 6.3. And when a job requires compliance to table 6.2, The welder whose UT welds fail 6.3 due to root defects but the same weld defects pass 6.2 requirements, must have defrauded the customer because he did not adequately backgouge and provide the customer with a pure full penetration weld?

By 803056

Date 09-18-2012 15:01 Edited 09-19-2012 12:30

You provided limited information and asked for an opinion. You got what you asked for.

I apologize if I struck a raw nerve Bill. I have been accused of practicing snow shovel diplomacy before. It served a purpose; it got your undivided attention.

You have a problem and it can cost your employer a lot of money to correct the welds in question. I have been in similar situations, so I can empathize with your plight. The trick is to come up with a solution that is equable to both parties.

You stated that the UT acceptance criteria was Table 6.3 for cyclically loaded nontubular connections. The criteria is very stringent, so it is a safe assumption the Engineers has concerns and considers the connections to be "critical" to the integrity of the structure. It is also a safe assumption that the connections that were not tagged for UT examination were not as critical to the integrity of the structure, but never the less, there are certain expectations on the part of the Owner and the Engineer regarding the soundness of all the welds.

While I assume you took some level of care in preparing the joints that were not subject to UT, I have seen situations where the contractor took the attitude that anything goes if the connections where not UT'd. Back gouging was approached with no regard to whether sound welds were deposited. Hopefully, that does not describe the attitude of the welders or management in your facility.

The fact is the Owner has identified a problem. The Owner is convinced the work is substandard and you have to come up with a solution that will not bankrupt the company.

What I would propose is that the Engineer did not flag certain joints for UT because they are not critical. They need only pass visual inspection. However, they do have to meet all the visual criteria with regards to fit-up, preheat, interpass cleaning, completeness of back gouging, the back gouges are required to comply with a prequalified U-groove or J-groove before welding from the second side, etc. The welders have to employ tabs at the ends of the grooves to ensure sound welds, etc. The contractor has an obligation to ensure those requirements are met. There are quality control measures that are the contractor's responsibility. As a Verification Inspector I typically ask if the contractor has inspection reports available as objective evidence proper quality control measures are in place and the contractor verified those measures were fully implemented; all welds checked to verify the fit up was proper, preheat and interpass temperature were verified, the completed welds were inspected. If there are no inspection records completed by the contractor there is a serious problem. The owner has a reasonable case that until the contractor proves the welds are sound; there is no reason to assume they are sound.

My next step is to visually examine the welds. Did the contractor use tabs at the ends of the grooves? No? Then it is reasonable to grind the ends of the grooves flush with the adjacent members to check the welds for evidence of subsurface incomplete joint penetration, incomplete fusion, and slag inclusions. If there is any evidence of the defects listed, there is cause to employ PT or MT to assess the seriousness of the problem. How far into the joint does the defect extend into the joint? There are two means of determining the extent of the problem: (1) excavate the groove to sound metal or (2) employ UT as an expedient means of assessing the extent of the problem.

I would suggest that it is prudent for the contractor to suggest to the Engineer and the Owner that the contractor will pay for the joints in question to be tested using UT. The contractor should also offer to repair defective welds if the Engineer and the Owner agree to use Table 6.2 as the basis of acceptance or rejection. I would not be so fast to suggest this approach if the contractor had in process and final inspection reports. I would not suggest this if the contractor used tabs and the ends of the grooves exhibited none of the defects listed in the preceding paragraph. I operate under the assumption the contractor took the necessary steps needed to ensure the welds complied with the design drawings and the code. I would default to UT only if it has been demonstrated the contractor did not meet the visual criteria I have outlined. If the welds fail visual examination, it is the contractor's responsibility to demonstrate the welds are sound and comply with D1.1.

If the Engineers agrees that Table 6.2 is appropriate and the connections meet that criteria the problem is solved. The cost is minimal and the Engineer and Owner are satisfied they have a sound structure. If the welds are found to be defective, take it on the chin and fix them. Public relations are an important part of doing business and growing a business. A disgruntled customer has lasting repercussions.

The information provided in the inquiry is very sparse. There are many unanswered questions because the information is not provided in the inquiry. Did the contractor provide in-process inspection? Did the contractor provide any fit-up inspection? Did the contractor verify the groove details complied with the tolerances of figures 3.4? Did the contractor verify the minimum preheat requirements were met? Did the contractor provide any inspections of the back gouged grooves? Did the contractor check to see that the welders were properly cleaning between weld passes? We have no idea of what level of quality control, if any, was provided by the contractor's quality control/quality assurance personnel.

Many people that have not been involved in fabrication and have not performed verification inspections would expect that all of the QC functions mentioned in the preceding paragraph are automatically provided. I find that many contractors provide little if any QC/QA unless there is a verification inspector camped on premises during fabrication. My experience has been that even though the contractor is an AISC certified fabricator, the minimal level of QC is provided only when it is enforced by the Owner. I have volumes of photographs of defective welds that were clearly the result of the contractor not taking the basic steps required to produce acceptable welds. I have encountered several projects were the contractor told the welders interpass slag removal was not required because the FCAW electrode was approved for "multipass". I have had several projects were the welders were using GMAW short circuiting transfer when the WPS specified spray mode transfer with a different shielding gas than that being used. I have many photographs were the contractor provided no back gouging because as he put it, "No one ever inspects those joints."What steps did this contractor have in place to ensure the welds met even the basic requirements of AWS D1.1? We would like to know the whole story, not bits and pieces. Reference was made to Table 6.2, but it was never stated that the "bad" welds passed UT to those lesser requirements.

Rarely is a weld found to be "defective" because of one unacceptable discontinuity. A short story can usually be written about all the reasons why a welded connection was deemed defective. So, I ask, what was the whole story on this project?

It is safe to say that the Engineer is the individual that will have to play the part of Solomon. He will have to weigh the facts and make a determination whether the "defective" welds are "good enough" or whether they will have to be repaired. If an equitable solution is not agreed to, the lawyers and the courts can make the decision, but it will cost more than it will to make the necessary repairs and the contractor involved will suffer more than a sullied reputation when the dust settles.

Al