ML20345A198

NCP 2020-008 Submittal VEGP Withdrawal of NCV with Attachment

ML20345A198
Person / Time
Issue date:	12/10/2020
From:	Marissa Bailey NRC/RGN-II
To:
References
NCP 2020-008
Download: ML20345A198 (30)

Text

B NRC FORM 757 U.S. NUCLEAR REGULATORY COMMISSION

[06-2019)

NRC MD 10.156 NONONCURRENCEPROCESS !Go to Instructions I COVERPAGE ML20345A198 -

The U.S. Nuclear Regulatory Commission (NRC) strives to establish and maintain an environment that encourages all employees to promptly raise concerns and differing views without fear of reprisal and to promote methods for raising concerns that will enhance a strong safety culture and support the agency's mission.

Employees are expected to discuss their views and concerns with their immediate supervisors on a regular, ongoing basis. If informal discussions do not resolve concerns, employees have various mechanisms for expressing and having their concerns and differing views heard and considered by management.

Management Directive, MD 10.158, "NRC Non-Concurrence Process," describes the Non-Concurrence Process (NCP).

The NCP allows employees to document their differing views and concerns early in the decisionmaking process, have them responded to (if requested), and include them with proposed documents moving through the management approval chain to support the decisionmaking process.

NRC Form 757, "Non-Concurrence Process," is used to document the process.

Section A of the form includes the personal opinions, views, and concerns of a non-concurring NRC employee.

Section B of the form includes the personal opinions and views of the non-concurring employee's immediate supervisor.

Section C of the form includes the agency's evaluation of the concerns and the agency's final position and outcome.

NOTE: Content in Sections A and B reflects personal opinions and views and does not represent the official agency's position of the issues, nor official rationale for the agency decision. Section C includes the agency's official position on the facts, issues, and rationale for the final decision.

1. If the process was discontinued, please indicate the reason (and skip to #3):

D Non-concurring employee(s) requested that the process be discontinued D Subject document was withdrawn

2. At the completion of the process, the non-concurring employee(s):

D Concurred

[Z] Continued to non-concur D Agreed with some of the changes to the subject document, but continued to non-concur

3. For record keeping purposes:

D This record is non-public and for official use only

[Z] This record has been reviewed and approved for public dissemination NRC FORM 757 (06-2019) Use ADAMS Template NRC-006 (ML063120159) Page 1 of8

NRC FORM 757 U.S. NUCLEAR REGULATORY COMMISSION 1. NCP Tracking Number

[06-2019) 2020-008 NRC MD 10.156 NON-CONCURRENCE PROCESS (Continued) Date 12/10/2020 Section A - To Be Completed By Non-Concurring Employee

2. Title of Subject Document 3. ADAMS Accession Number Withdrawal of NCV 05200025_2016004-02 FINAL N/A (document not issued yet)

4. Document Signer 5. Document Signer's Phone Number (Enter 10 numeri c digits)

Nicole Coovert (404) 997-4510

6. Title of Document Signer 7. Office (Choose from the drop down list or fill in)

Branch Chief RII

[3

8. Name of Non-Concurring Employee(s) 9. Employee's Telephone Number (Enter 10 numeric digits)

Anthony Ponko (404) 997-4509

10. Title of Non-Concurring Employee 11. Office (Choose from the drop down list or fill in)

B Senior Construction Inspector RII 12.

Document Author Document Contributor Document Reviewer [Z] On Concurrence

13. Name of Non-Concurring Employee's Supervisor 14. Office (Choose from the drop down list or fill in)

Nicole Coovert RII

[3 15 . Title of Non-Concurring Employee's Supervisor 16. Supervisor's Telephone Number (Enter 10 numeric digits)

Branch Chief (404) 997-4510 17.

[Z] I would like my non-concurrence considered and would like a written evaluation in Section Band c.

I would like my non-concurrence considered, but a written evaluation in Sec tions Band C is not necessary.

18. When the process is complete, I would like management to determine whether public release of the NCP Form (with or without redactions) is appropriate (Select "No" if you would like the NCP Form to be non-public):

[Z] Ye s No 19 . Reasons for the Non-Concurrence, Potential Impact on Mission, and the Proposed Alternatives See attachment A.

20. Signature and Date of Non-Concurring Employee Anthony Panko Digitally signed by Anthony Ponko Date: 2020 09.24 13:45:38 -04'00' NRC FORM 757 (06-2019) Use ADAMS Template NRC-006 (ML063120159) Page 2 of 8

NRC FORM 757 U.S. NUCLEAR REGULATORY COMMISSION 1. NCP Tracking Number (06-2019) 2020-008 NRC MD 10 156 NON-CONCURRENCE PROCESS (Continued) Date 12/10/2020 Section B - To Be Completed By Non-Concurring Employee's Supervisor

2. Title of Subject Document 3. ADAMS Accession Number Withdrawal of NCV 05200025_2016004-02_FINAL N/A (document not issued yet)

4. Name of Non-Concurring Employee's Supervisor 5. Office (Choose from the drop down list or fill in)

Nicole Coovert RII

[3

6. Title of Non-Concurring Employee's Supervisor 7. Supervisor's Telephone Number (Enter 10 numeric digits)

Branch Chief (404) 997-4510

8. Comments for the NCP Reviewer to Consider The NRC values differing views and alternative approaches and want to ensure a non-concurrence is heard, understood, and considered. As a supervisor, I appreciate the effort and detail Mr. Ponko has put into the non concurrence of the non-cited violation (NCV)05200025_2016004-02 withdrawal letter.

For additional information, a timeline is listed below regarding the NRC inspection reports and the inquiries/

interpretations for the AWS D1.4-98, Structural Welding Code --Reinforcing Steel Code Committee.

• In January 2017, a code inquiry was submitted to the Code Committee for ANSI AWS D1.4-98, Structural Welding Code --Reinforcing Steel. In April 2017, the Code Committee completed the interpretation under AWS Log#

D1.4-98-102 regarding qualification limits of Table 6.2. The specific inquiry questions and Committee interpretations/responses are listed below.

Inquiry Question 1: If a WPS is qualified per 6.2.4.1 and in accordance with Table 6.2, does qualification using one of the "Production type of joint" listed, qualify for all the corresponding "Joints Qualified per Figure" shown for that row? (E.g., Testing with "Direct Butt [6.5(A))" qualifies the WPS for all direct butt joints shown in figure 3.2)?

Question 2: Does essential variable#11 of Table 6.1 prohibit a change in groove type where such allowances are permitted or required by Table 6.2? (E.g., Figure 6.5(C) flare-bevel groove is required to qualify Figure 3.4 flare vee groove.)

Interpretation: Question: 1: Yes Question: 2: No

• On February 13, 2017, inspection report "VOGTLE ELECTRIC GENERATING PLANT UNITS 3 AND 4 - NRC INTEGRATED INSPECTION REPORTS 05200025/2016004, 05200026/2016004" (ML17044A539) was issued.

This report contained the Unit 3 NCV 05200025/2016004-02, "Failure to qualify WPS in accordance with applicable code," which documented an ITAAC finding of very low safety significance (Green). The associated NCV was for 10 CFR Part 50, Appendix B, Criterion IX, "Control of Special Processes" for SNC's failure through their contractor WEC to adequately implement measures to assure that special processes, including welding, are accomplished in accordance with applicable codes. The Unit 3 ITAAC affected is 3.3.00.02a.i.b (Index# 761).

The NRC originally exited on January 11, 2017 with an Unresolved Item (URI). On January 20, 2017, the NRC re-exited with an NCV.

• On June 6, 2017, SNC issued Engineering & Design Coordination Report (E&DCR) APP-CR01-GEF-850264 Rev.

0, to document the Design Authority evaluation of their usage of AWS D1.4-98. The closure of the E&DCR included the Code Committee interpretation D1.4-98-I02-CD1-BM.

NRC FORM 757 (06-2019) Use ADAMS Template NRC-006 (ML063120159) Page 3 of 8

NRC FORM 757 U.S. NUCLEAR REGULATORY COMMISSION 1. NCP Tracking Number (06-2019) 2020-008 NRC MD 10 156 NON-CONCURRENCE PROCESS (Continued) Date 12/10/2020

• On January 31, 2018, NRC issued "ERRATA FOR VOGTLE UNIT 3 COMBINED LICENSE, VOGTLE UNIT 4 COMBINED LICENSE NRC INTEGRATED REPORTS 05200025/2016004, 05200026/2016004" (ML18032A756) for the original inspection report. As a result of a routine internal NRC audit performed in accordance with Inspection Manual Chapter (IMC) 2505, "Periodic Assessment of Construction Inspection Program Results," the Office of New Reactors (NRO) identified where additional wording was needed to be included in the report section documenting the Construction Significance Determination for the Green NCV 2016004-002, in order to meet the requirements of IMC 0613, "Power Reactor Construction Inspection Reports." Specifically, the last sentence in italics was added to the report for the Green NCV 2016004-002.

"The inspectors concluded the finding was associated with the Construction/ Installation Cornerstone. The inspectors evaluated the finding in accordance with IMC 2519, Appendix A, "AP 1000 Construction Significance Determination Process," and determined the finding was of very low safety significance (Green) because the finding was associated with a structure assigned to the intermediate column of the AP1000 Construction Significance Determination Matrix, and there is reasonable assurance that the structure can meet applicable design functions."

• A second code inquiry was submitted to the Code Committee for ANSI AWS D1.4, Structural Welding Code -

Reinforcing Steel. In May 2020, the Code Committee completed the interpretation under AWS Log# D1.4-18-101 regarding qualification limits of Tables 8.2 and 8.3. It is noted that the licensee is not committed to AWS D1.4-2018, however, in this code interpretation, Inquiry 3 reiterated statements for the 1998 inquiry. Specifically, D1.4-18-101, "Use of Essential Variables for WPS Qualification," Inquiry 3 and Committee interpretations/

responses are listed below:

Inquiry Question 3: Must the test assemblies and the production joint being qualified in D1.4/D1.4M:2018 conform to one another within the limits of the essential variables?

Interpretation 3: See Inquiry and Response to Question 2 and answer of the D1.4-98-102 interpretation.

In August 2020, the NRC generated a withdrawal letter for Unit 3 NCV 05200025/2016004-02. The basis for the withdraw letter was informed by the AWS interpretations that provided clarification for welding procedure specification (WPS) qualifications with respect to essential variables for joint type and groove type and the licensee's evaluation for the code applicability documented in E&DCR APP-CR01-GEF-850264, Revision 0, "Rebar WPS Qualification." As a result, the NRC determined that the violation was not valid and subsequently determined that SNC was in compliance with the requirements of AWS D1.4-98. On September 2, 2020, the letter was sent to applicable NRC staff for concurrence.

9. Signature and Date of Non-Concurring Employee's Supervisor Nicole C. Coovert Digitally signed by Nicole C. Coovert Date: 2020.11.04 10:26:41 -05'00' NRC FORM 757 (06-2019) Use ADAMS Template NRC-006 (ML063120159) Page 4 of 8

NRC FORM 757 U.S. NUCLEAR REGULATORY COMMISSION 1. NCP Tracking Number (06-2019) 2020-008 NRCMD 10156 NON-CONCURRENCE PROCESS (Continued) Date 12/10/2020 Section C - To Be Completed By NCP Coordinator

2. Title of Subject Document 3. ADAMS Accession Number Withdrawal of NCV 05200025_2016004-02_FINAL N/A (document not issued yet)

4. Name of NCP Coordinator 5. Office (Choose from the drop down list or fill in)

Brendan Collins RII

[3

6. Title of NCP Coordinator 7. Coordinator's Te lephone Number (Enter 10 numeric digits)

Senior Reactor Inspector (404) 997-4420

8. Agreed Upon Summary of Issues The non-concurring employee contends that the finding issued in the Vogtle 3 & 4 2016004 report remains valid, and should not be withdrawn on the basis of Code Interpretation AWS D1.4-98-102 and/or D1.4-18-101.

Background

• The welded splices used in the construction of Seismic Category I structures are critical structural connections and are, therefore, important to the structural integrity of said structures.

• Vogtle Unit 3 UFSAR Section 3.8.4.4.1, ACI 349-01 Section 12.14.3.2, AWS D1.4-98 Sections 6.1.2.3, 6.2.1.4 and Table 6.1 establish the requirements pertinent to this discussion for qualification of a Welding Procedure Specification (WPS) for welding these splices.

• Violation 2016004-02 was issued for the licensee's failure to adhere to these requirements. Specifically, the licensee failed to use a qualified WPS to weld splices used in several Seismic Category I structures. The violation stated that the licensee made use of a WPS that allowed single-V, single-bevel, double-bevel and double-V groove welds, but the supporting Procedure Qualification Records (PQRs) only justified double-V groove welds contrary to the limitations of AWS D1.4-98 Table 6.1 Item 11.

• Further consideration of the violation has been given by NRC management on the basis of Code Interpretations AWS D1 .4-98-102 and D1.4-18-101, and it is unclear to the non-concurring employee how these interpretations can be used to conclude that the licensee's approach was allowed. Specifically...

o The 1998 interpretation issued by the Code results in an apparent contradiction between Table 6.1 and Table 6.2 in the 1998 Edition of the Code.

o The 2018 interpretation resulted in an addendum to the 2018 Edition which revised the essential variable for groove type to joint type. This interpretation, however, is not retroactive to AWS D1.4-1998. As a result, the apparent contradiction described above still exists in the 1998 Edition.

o As such, there is a non-compliance that must be addressed by the licensee in their CAP.

• The licensee completed actions in their Corrective Action Program and during the NRC's closeout review of the issue (ITAAC), the inspectors identified two more issues ...

o The corrective actions that had been taken by the licensee were determined to be inadequate, and as such, a Criterion XVI violation existed o A second non-compliance existed pertaining to AWS D1.4-98 Table 6.1 (Item 10, position), wherein -

NRC FORM 757 (06-2019) Use ADAMS Template NRC-006 (ML063120159) Page 5 of 8

NRC FORM 757 U.S. NUCLEAR REGULATORY COMMISSION 1. NCP Tracking Number (06-2019) 2020-008 NRCMD 10156 NON-CONCURRENCE PROCESS (Continued) Date 12/10/2020 s1milarrotFie original issue - tne licensee used a VVl-'S fnat was not qualified 6y ffie supporting PQRs Summary of Issues

• Ultimately, the background information leads to three issues which need to be addressed...

o First non-compliance of AWS D1.4-98 Table 6.1 (Item 11, groove type) (NCV 5200025/2016004-02) o Criterion XVI violation for inadequate corrective actions associated with first non-compliance o Second non-compliance of AWS D1.4-98 Table 6.1 (Item 10, position)

9. Evaluation of Non-Concurrence and Rationale for Decision As the Non-Concurrence Process (NCP) Approver, I would like to start by stating that the contributions of all parties involved with this NCP were highly valued and served to underscore the importance of the process itself. It is necessary for us to continue to embrace the idea that every voice matters and healthy discourse can only improve our ability to continue to protect the public and the environment.

It is also important to recognize that after such discussions, there will be times when complete consensus is unattainable, and a decision must be made. That is the case here, so the remainder of this commentary is intended to explain the decision for these issues. For clarity, I will address each of the three bullets from the Summary of Issues, one at a time.

The decisions reflected in the commentary below are the result of a detailed review of Section A and multiple one-on one and group discussions betvveen me, the non-concurring employee, and several U.S. Nuclear Regulatory Commission (NRC) technical experts. These technical experts were agency senior materials engineers, structural engineers and Code experts (including the NRC's Senior Level Advisor for Structural Mechanics).

In summary, based on this review and the discussions, I have determined that it is appropriate to withdraw non-cited violation (NCV)05200025/2016004-02, "Failure to qualify WPS in accordance with applicable code." I have also determined that NRC staff has no basis to conclude that a second non-compliance exists pertaining to American Welding Society (AWS) D1.4-98 Table 6.1 (Item 10, position).

Bullet #1: First non-compliance of AWS D1.4-98 Table 6.1 (Item 11, groove type) (NCV 5200025/2016004-02)

The concern hinges on qualification of the welds. Qualification of welds is addressed by the American Welding Society (AWS) standards. Common terms include: (1) a welding procedure specification (WPS) is a written procedure that is used by the welder to make production welds; (2) a procedure qualification record (PQR) is a written record used to confirm/test that the WPS can produce acceptable welds.

A great deal of information was exchanged regarding this issue, both verbally and in writing, but ultimately the resolution hinged upon a single point: whether AWS D1.4-98, Table 6.2 provides an allowance for meeting the requirements of AWS D1.4-98, Table 6.1, Item 11, for qualifying different groove types.

After hearing the non-concurring employee's and the NRC technical experts' perspectives, I have concluded that Table 6.2 does indeed represent such an allowance. AWS D1.4-98, Table 6.2 indicates that WPS qualification of direct butt joints includes the joints indicated in Figures 3.1, 3.2, and 3.5(D). Figure 3.2 provides examples of groove types that encompass those that the inspectors observed being used at Vogtle Units 3 and 4. Thus, the WPS for these groove types is qualified even if the groove types were not used in the PQRs that supported WPS qualification (that is, the licensee's PQR for butt joints qualifies the WPS for the other groove types used at Vogtle 3 and 4, even NRC FORM 757 (06-2019) Use ADAMS Template NRC-006 (ML063120159) Page 6 of 8

NRC FORM 757 U.S. NUCLEAR REGULATORY COMMISSION 1. NCP Tracking Number (06-2019) 2020-008 NRCMD 10156 NON-CONCURRENCE PROCESS (Continued) Date 12/10/2020 if ffie offier groove types were not explicitly parrof fne PQR). NRC tecflnical experts concluaeatflat ffi1s interpretation of Table 6.2 is consistent with standard industry practice and provides sufficient confidence in the quality of the welds produced in accordance with the WPS. The AWS Code interpretation D1.4-98-102 also supports this view, and AWS Code Interpretation D1.4-18-101 further clarifies that Table 6.1, Item 11 was never intended to be as limiting as it could be interpreted. Moreover, this position is consistent with other Codes, such as AWS D1.1 which has a similar essential variable (variable 32 in Table 4.5) but has explicit wording concerning other grooves qualified as follows: "A change in groove type (e.g., single V-to double V). Qualification of any complete joint penetration groove weld qualifies for any groove detail complying with the requirements of 3.12 or 3.13."

Further, no evidence could be provided to support the position that groove type should be an essential variable from a technical perspective. The 2020 amendment D1.4/D1.4M:2018-AMD1 to AWS D1.4-18 directly supports this conclusion by removing groove type from the essential variable table. If groove type is not an essential variable, then a change in groove type does not have the potential to reduce the quality of the welded product in the field. That being the case here, the issue has no safety impact to the components being constructed at Vogtle 3 and 4.

After considering all these facts, I have concluded that it is still appropriate to withdraw NCV 5200025/2016004-02, and we will proceed with the appropriate documentation to do so.

Bullet #2: Criterion XVI violation for inadequate corrective actions associated with first non-compliance Once the decision for the first issue was made, it simplified the decision for this second issue considerably: if no violation exists regarding groove type, then no corrective action is required to be taken to address the non compliance.

Nonetheless, I have determined that the licensee's actions to pursue more information pertinent to the issue (submitting the AWS Code Inquiries that led to the Interpretations mentioned previously) were appropriate and assisted in the NRC's ability to come to the appropriate regulatory position. Similarly, I agree with the licensee's interpretation of the issue (provided in their CAP documentation), which stated that the 1998 AWS Interpretation justified their actions.

Therefore, there is no Criterion XVI violation for inadequate corrective actions associated with the original NCV. This aspect will also be documented in the withdrawal letter.

Bullet #3: Second non-compliance of AWS D1.4-98 Table 6.1 (Item 10, position)

There were two reasons this second potential non-compliance was included in the Summary of Issues.

First, because if the original NCV was not withdrawn, this issue would potentially represent a second instance of the same violation (failure to comply with AWS D1.4-98 Table 6.1). The NRC does not typically "stack" violations as a result of extent of condition reviews, and as such, we would have likely amended the original NCV to explain that a second instance had been found during the follow-up inspection.

Second, we included the second potential non-compliance to proactively track the issue, such that we could make sure it was processed appropriately.

Based upon thorough discussions with the NRC technical experts on the potential second non-compliance, I have determined that there is no violation associated with the licensee's failure to follow AWS D1.4-98 Table 6.1 (Item 10, position), based on the commentary that follows.

NRC FORM 757 (06-2019) Use ADAMS Template NRC-006 (ML063120159) Page 7 of 8

NRC FORM 757 U.S. NUCLEAR REGULATORY COMMISSION 1. NCP Tracking Number (06-2019) 2020-008 NRCMD 10156 NON-CONCURRENCE PROCESS (Continued) Date 12/10/2020 1 ne tocal point ot me issue Is as stated in me Affact ment roecf1on A, t-'Ut- p.l:H::l:

AWS D 1.4-98, 6.2.3, states, in part, that a WPS shall be required for each production welding position. WPS 8.1.1S03 allows all positions. However, PQ907, PQ908, PQ909, PQ910, PQ911, and PQ912 were all completed in either the vertical (3G) or overhead (4G) positions. No PQRs were performed for the flat (1G) or horizontal (2G) positions.

The non-concurring employee discusses AWS D1.4-98, Table 6.1, Item 10, which states that "A change in position not qualified per 6.2.4" is an essential variable. The NRC technical expert group discussed this aspect and determined that "6.2.4" is likely a typographic error, because 6.2.4 imposes no limitations on position in terms of qualification. It is their belief that Item 10 should refer to either paragraph 6.2.3 or 6.3.4, which does impose such limitations. In the technical discussions, the technical expert group looked at this from both of the following perspectives:

1. Looking specifically at paragraph 6.2.3: the requirement imposed by that paragraph is that "a WPS shall be required for each production welding position." As stated above, a WPS is a written procedure that is used by the welder to make production welds and a PQR is a written record used to confirm/test that the parameter of the WPS can produce acceptable welds. For each production weld associated with this NCP, there was a WPS, so the requirement was met. Paragraph 6.2.3 does not require a PQR for each position covered by a WPS, and as explained below, qualification in the overhead and vertical positions is technically adequate to qualify all positions.

2. Looking at the broader scope of how positions are qualified: this is covered in A\NS D1.4-98 paragraphs 6.3.4.1.(3) and (4). Although these paragraphs discuss welder qualification, the same principles should logically also apply to WPS qualification, and it is reasonable to infer that AWS D1.4-98, Table 6.1, Item 10 was meant to invoke these provisions for WPS qualification to avoid repetition. For the production welds associated with this NCP, the positions listed in the WPS-supporting PQRs (vertical, overhead) do qualify the WPS for all positions, so that requirement is met as well. It is noted that the use of the vertical and overhead positions to qualify the WPS for all positions is consistent with other codes such as AWS D1.1 and ASME Code,Section XI, and therefore paragraph 6.3.4 is likely the correct reference in AWS D1.4-98, Table 6.1, Item 10. This conclusion is further validated in AWS D1.4-2011, which revised Table 6.1, Item 10 to reference paragraph 6.3.4 in lieu of 6.2.4.

Therefore, AWS D 1.4-98 does not require PQRs to be performed in the flat (1G) or horizontal (2G) positions if PQRs were performed in the vertical and overhead positions, and the WPS associated with this NCP satisfies AWS D 1.4-98 as it relates to welding position.

10. Signature and Date of NCP Coordinator Brendan C. Collins Digitally signed by Brendan C. Collins Date: 2020.12.08 15:00:12-05'00'

11. Signature and Date of NCP Approver Marissa G. Bailey Digitally signed by Marissa G. Bailey Date: 2020.12.10 08:13:12-05'00' NRC FORM 757 (06-2019) Use ADAMS Template NRC-006 (ML063120159) Page 8 of 8

I respectfully non-concur on the withdrawal of non-cited violation (NCV)05200025/2016004-02, "Failure to qualify WPS in accordance with applicable code," contained in NRC Integrated Inspection Reports 05200025/2016004, 05200026/2016004, issued February 13, 2017, Accession No. ML17044A539, and in Errata for NRC Integrated Inspection Reports 05200025/2016004, 05200026/2016004, issued in January 31, 2018, Accession No. ML18032A756.

The welded reinforcing bar splices used in the construction of Seismic Category I structures 1 of VEGP Units 3 & 4 have not been installed using an appropriately qualified Welding Procedure Specification (WPS).

The welded splices in question are critical structural connections that have been used extensively in risk significant structures of VEGP Units 3 & 4. As such, these splices are important to the structural integrity of these structures and must perform satisfactorily during design basis events.

The nature of special processes such as welding, in general, along with the importance to safety and extensive use of welding reinforcing bar splices in VEGP Units 3 & 4, places great emphasis on the quality assurance measures implemented to provide confidence that the splices will meet performance and quality requirements.2 WPS qualification is a quality assurance measure that provides confidence that a sound weld can be made by a qualified welder. The WPS is used to communicate the parameters by which a particular weld is to be made. These parameters include the essential variables.3 Qualification of the WPS provides evidence that welds having the required mechanical properties or soundness can be achieved in test coupons by a qualified welder using the WPS.

Changes beyond the limitations of the essential variables require requalification of the WPS.

Without adequate WPS qualification, it may not be possible to conclude that VEGP Units 3 & 4 ITAAC 3.3.00.02a.i.a (760), 3.3.00.02a.i.b (761), 3.3.00.02a.i.c (7 62), and 3.3.00.02a.i.d (763) have been met.

The Seismic Category I reinforced concrete structures, in which welded reinforcing bar splices have been installed in VEGP Units 3 & 4, are designed and constructed in accordance with American Concrete Institute, "Code Requirements for Nuclear Safety Related Concrete Structures (ACI 349-01)." 4 ACI 349-01 section 12.14.3.2 requires that all welding conform to American Welding Society, "Structural Welding Code - Reinforcing Steel," (AWS D1.4-98).

1 Structures desig nated as Seismic Category I must be designed to remain functional if the Safe Shutdown Earthquake Occurs. The pertinent requirements of Appendix B to 10 CFR SO apply to all activities affecting the safety-related functions of Seismic Category I Structures, Systems, or Components, Regulatory Guide 1.29, "Seismic Design Classification, Revision 4, March 2007.

2 The introduction to 10 CFR Part SO Appendix B states that "quality assurance" comprises all those planned and systematic actions necessary to provide adequate confidence that a structure, system, or component will perform satisfactorily in service. Quality assurance includes quality control, which comprises those quality assurance actions related to the physical characteristics of a material, structure, component, or system which provide a means to control the quality of the material, structure, component, or system to predetermined requirements.

3 AWS A3.0M/A3.0:2020, "Standard Welding Terms and Definitions," defines an essential variable used in WPS qualification as "a process parameter deemed critical to the creation of a test coupon acceptable to the applicable VEGP Unit 3 & 4 UFSAR, Section 3.8.4 (ML20181A496) standard."

4

NCV 05200025/2016004-02 was based upon a reasonable review and understanding of the requirements of AWS D1.4-98 for WPS qualification by NRC inspectors. The basis for the NCV is supported by the plain language of the code and industry standard practices for the use of essential variables in WPS qualification.

NCV 05200025/2016004-02 is being withdrawn based on code interpretations D1.4-98-102 and D1.4-18-101. However, only code interpretation D1.4-98-102 applies to AWS D1.4-98, the welding code used for WPS qualification at VEGP Units 3 & 4. This code interpretation is ambiguous and does not provide a sound basis for concluding that NCV 05200025/2016004-02 was dispositioned in error and should be withdrawn. The other code interpretation referenced in the withdrawal letter - D1.4-18-101 - applies to AWS D1.4/D1.4M:2018, not AWS D1.4-98.

Moreover, neither of these code interpretations have been reviewed or endorsed by the NRC for use on nuclear safety-related projects to my knowledge.

AWS D1.4-98 and AWS D1.4/01.4M:2018 are two versions of the AWS D1.4 code that were published 20 years apart. These two versions of AWS D1.4 differ in some requirements, including the essential variables for WPS qualification.5 These differences do not support extension of code interpretation AWS D1.4/D1.4M:2018 backwards to AWS D1.4-98.

Code interpretation AWS D1.4:D1.4:2018 revises the essential variable for groove type to joint type. This is a substantive change to a code provision in AWS D1.4/D1.4M:2018 that is not retroactive to AWS D1.4-98. These changes are addressed in an associated amendment of AWS D1.4/D1.4M :2018. 6 This amendment further revises the requirements of the test assemblies used for WPS qualification of reinforcing steel direct butt splices to bring them into alignment with the change in essential variable from groove type to joint type. In effect, code interpretation D1.4-18-101 widens the differences in essential variables for WPS qualification between AWS D1.4-98 and AWS D1.4/D1.4M:2018.

On the surface, code interpretation D1.4-98-102 would appear to support withdrawal of NCV 05200025/2016004-02. However, doing so would require the essential variable for groove type to either have special status, be in error, or represent something completely different than it does in other AWS codes and the welding industry in general. None of these conclusions are supported by the plain language of AWS D1.4-98. Code interpretations are intended to clarify existing requirements, not upend them. Code interpretation D1.4-98-102 should be in harmony with the existing AWS D1.4-98 provisions for WPS qualification, not in conflict.

Setting aside the fact that code interpretation D1.4-18-101 is not applicable to AWS D1.4-98, this interpretation undermines the use of code interpretation D1.4-98-102 as a basis for the withdrawal NCV 05200025/2016004-02. Code interpretation D1.4-18-101 reaffirms that position is an essential variable for WPS qualification. The use of this essential variable in WPS qualification is analogous to that of groove type. Therefore, a reasonable conclusion is that both these essential variables must be accounted for during WPS qualification and changes beyond the limitations of either essential variable would require WPS requalification. Reading code interpretation D1.4-98-102 in this light calls into question its use as a basis for withdrawal of NCV 05200025/2016004-02. Additionally, the magnitude of the revisions required to change the 5 AWS Dl.4-98 essential variable 12 places further limitations on changes to the geometry of the groove used for WPS qualification. This essential variable involves a change in the groove shape or decrease in groove angle, decrease in root opening, or an increase in the root face . It was removed from Table 6.1 in AWS Dl.4/D.14M :2011.

6 AWS Dl.4/D1.4M:2018-AMD1

essential variable from groove type to joint type in code interpretation D1.4-18-101 and associated amendment AWS D1.4/D1.4M :2018-AMD1 indicate that the designation of groove type as an essential variable in AWS D1.4-98 was by design, the term means what it does in other AWS codes and the welding industry in general, and this essential variable was intended to be used in a similar manner to other essential variables in WPS qualification.

Neither code interpretation D1.4-98-102 nor D1.4-18-102 revised the requirements that essential variables be controlled during the WPS qualification process, or that changes beyond the essential variable limitations require WPS requalification. These provisions are fundamental to WPS qualification in the AWS D1 group of standards, as well as, other industry standards that address WPS qualification. The existing AWS D1.4-98 provisions for WPS qualification should not be discarded or subordinated by either of these code interpretations.

At the time NCV 05200025/2016004-02 was identified, welded reinforcing bar splices were being used in a limited manner in VEGP Units 3 & 4. However, the use of welded reinforcing bar splices or reinforcing bars welded to structural steel plates has increased as construction has progressed. Welded reinforcing bar splices or reinforcing bars welded to structural steel plates have been observed by NRC inspectors in the locations identified below. This list is not all inclusive, but only represents that areas in which I have direct knowledge where these types of connections have been installed.

1. Units 3 & 4- radial hoops around main steam generator penetration through Wall 11 (ITAAC 3.3.00.02a.i.c [762] Critical Section)

2. Units 3 & 4- Top reinforcing bars in operating deck inside containment (ITAAC 3.3.00.02a.i.a [760])

3. Units 3 & 4- Tensile bars in Wall 1 and 4 corbels supporting crane over spent fuel pool and cask pits (ITAAC 3.3.00.02a.i.d [763], Wall 1 is Critical Section)

4. Unit 3- Isolated areas of RC Shield Wall (ITAAC 3.3.00.02a.i.b [761])

5. Unit 4- Layer 6 Bars in Wall 11 at main steam and feedwater penetrations through Wall 11 (ITAAC 3.3.00.02a.i.c [762] Critical Section)

6. Unit 4- Horizontal Bars in Walls M and P at 117'-6" and 135'-3" (ITAAC 3.3.00.02a.i.c

[762])

7. Unit 4- Floor slab between Walls M, P, 11 and Shield Building cylindrical wall at elevation 135'-3" (ITAAC 3.3.00.02a.i.c [762], Critical Section)

When NCV 05200025/2016004-02 was dispositioned, the NRC had not yet identified that welded reinforcing bar splices were not approved for use in the construction of VEGP Units 3 &

4. 7 As a result, NCV 05200025/2016004-02 documents a noncompliance with AWS D1.4-98 which is adopted by reference in ACI 349-01. NRC guidance on the use of welded reinforcing splices, however, is that they meet the requirements of American Society of Mechanical Engineers (ASME) Boiler and Pressure Vessel (BPV) Code Section Ill, Division 2 (ASME Division 2). 8 This guidance, to my knowledge, is based on the importance to safety of welded reinforcing bar splices used in Seismic Category I structures and a reasonable concern with 7 Welded reinforcing bar splices were not approved as an alternative to lap splices in the APl000 Design Control Document (DCD) and VEGP Units 3 & 4 Updated Final Safety Analysis Report (UFSAR). As a result, the use of welded reinforcing bar splices in the construction of VEGP Units 3 was determined to be a violation of regulatory requirements. This violation is documented in Vogtle Electric Generating Plant, Unit 3 - NRC Inspection Report 05200025/2019-010 (M L20197A240).

8 This guidance is provided in Regulatory Position 4.2.2 of Regulatory Guide 1.142, Revision 3; Sections 3.8.3.6 and 3.8.4.6 of Regulatory Guide 1.70, Revision 3; and Sections 3.8.3.1.6.B and 3.8.4.1.6.B of NUREG-0800, Revision 4.

achieving adequate performance and quality requirements. As a result, the corrective actions taken to address the violations documented in NRC Inspection Report 05200025/2019-010 should address the deviations from this guidance, as well as, the quality assurance measures required to assure that the splices will meet performance and quality requirements, including WPS qualification. 9 Since NCV 05200025/2016004-02 was associated with an ITAAC, the NRC initiated an inspection on September 16, 2019 to close the NCV. During this inspection, the corrective actions taken by Southern Nuclear Company (SNC) to address the noncompliance were carefully reviewed and due consideration given to the views of SNC and the materials they provided to support their position. However, it was concluded that SNC had failed to adequately implement measures to assure that conditions adverse to quality were promptly corrected.

When an NCV is identified, there is an expectation that the licensee will place the performance deficiency into their corrective action program (CAP) and take actions to restore compliance.

SNC did not contest the NCV and placed it in their CAP. Unfortunately, SNC has yet to take adequate corrective actions to close NCV 05200025/2016004-02 to my knowledge.

SNC relied on code interpretation D1.4-98-102 to conclude that the WPS being used for the installation of welded reinforcing bar splices in VEG P Units 3 & 4 did not need to be requalified.

A code committee, however, should not be considered the ultimate authority for determining compliance with a license issued by the NRC. Nor should one be expected to act as a binding arbitrator to settle disagreements between the NRC and licensees about the corrective actions necessary to address violations of regulatory requirements. As a result of the inspection to close NCV 05200025/2016004-02, a more-than-minor violation of Criterion XVI, "Corrective Action" was initially proposed. Subsequently, SNC committed to take further corrective actions and the violation was held in abeyance pending review of those corrective actions.

During the inspection to close NCV 05200025/2016004-02, it was determined that in addition to not complying with the essential variable for groove type, the WPS being used for the installation of welded reinforcing bar splices in VEGP Units 3 & 4 also did not comply with the essential variable for position. As this noncompliance was similar to the example documented in NCV 05200025/2016004-02, the intent was to discuss it in the proposed Criterion XVI violation described earlier or, subsequently, to verify that it was addressed in the belated corrective actions taken to close the original NCV. However, the inspection to close NCV 05200025/2016004-02 was suspended without any further enforcement action being taken, or adequate corrective actions being completed by SNC to my knowledge. As a result, even if NCV 05200025/2016004-01 is withdrawn, the WPS being used for the installation of welded 9 SNC committed to Regulatory Guide 1.70, Revision 3 in the COL as documented in VEGP 3&4 UFSAR Appendix lA.

Regulatory Guide 1.70, Revision 3, Section 3.8.4.6, "Materials, Quality Control, and Special Construction Techniques", states that the materials, quality control programs, and any new or special construction techniques should be addressed as outlined in Section 3.8.3.6. This section is relevant for the VEGP 3&4 Shield and Auxiliary Buildings - ITAAC 3.3.00.02a.i.b (761), 3.3.00.02a.i.c (762), and 3.3.00.02a.i.d (763), respectively. SNC did not take exception to this guidance in the VEGP 3&4 COL as documented in VEGP 3&4 UFSAR Appendix lA . Regulatory Guide 1.70, Revision 3, Section 3.8.3.6.4 states, in part, that If welding of reinforcing bars is proposed, the extent to which the design complies with the ASME Code, Section Ill, Division 2 should be provided in the UFSAR and any exceptions taken should be identified and justified. This section is relevant for the VEGP 3&4 Containment Internal Structures and by the above reference the Shield and Auxiliary Buildings as well - lTAAC 3.3.00.02a.i.a (760),

3.3.00.02a.i.b (761), 3.3.00.02a.i.c (762), and 3.3.00.02a.i.d (763), respectively. SNC did not take exception to this guidance in the VEGP 3&4 COL.

reinforcing bar splices in VEGP Units 3 & 4 would still not conform to the A WS D1 .4-98 requirements for WPS qualification.

NCV 05200025/2016004-02 should remain open until it can be verified that appropriate corrective actions have been taken to address the noncompliance. A dditionally, SNC's failure to take appropriate corrective actions in a timely manner should be dispositioned as an ITAAC finding and associated more-than-minor violation of the Code of Federal Regulations (10 CFR)

Part 50, Appendix B, Criterion XVI, "Correction Action" for SNC's failure to adequately implement measures to assure that conditions adverse to quality were promptly corrected. The significance of this violation should be dependent on whether SNC can requalify the WPS to bound the welded reinforcing bar splices installed in VEGP Units 3 & 4 within the limitations of the essential variables of AWS D1.4-98, or provide evidence demonstrating that the Structures, Systems, and Components (SSCs) that rely on welded reinforcing bar splices will be able to adequately perform their safety functions. SNC's corrective actions should address deviations from the essential variable for position in addition to groove type which was provided as an example in NCV 05200025/2016004-02.

Background

(Attached to the back of this non-concurrence for reference is a brief description of reinforced concrete cons truction, the different types of reinforcing bar splices us ed, and the quality assurance measures for welded reinforcing bar splices)

NCV 05200025/2016004-02, "Failure to qualify WPS in accordance with applicable code" During the week of November 28th, 2016, the inspectors determined that the complete joint penetration welded reinforcing steel splices in the Unit 3 shield building reinforced concrete cylindrical wall did not meet the commitments established by the VEGP Unit 3 UFSAR.

Specifically, the splices were not qualified in accordance with A WS D1.4- 98, "Structural Welding Code - Reinforcing Steel."

The inspectors observed welded CJP direct butt splices of #11 vertical bars in VEGP Unit 3 shield building cylindrical reinforced concrete wall between azimuths 350 and 53 degrees at approximately elevation 105'-0". The observed production splices appeared to have been installed using a single-bevel in the upper bar with split pipe backing.

The inspectors noted that VEGP U3 UFSAR Section 3.4.4.4.1, "Seismic Category I Structures,"

states, in part, that the design and analysis procedures for the seismic Category I concrete structures are in accordance with the Code Requirements for Nuclear Safety Related Concrete Structures (ACI 349-01).

ACI 349-01, Sections 3.8.2 and 12.14.3.2 require, in part, that, welded splices conform to AWS D1 .4-98, "Structural Welding Code - Reinforcing Steel".

A WS D1.4-98 subsection 6.2.1.4 requires WPS requalification for any changes beyond the essential variable limitations of AWS D1.4-98 Table 6.1.

A WS D1.4-98, Table 6.1 essential variable 11 is defined as "a change in groove type (e.g.,

flareV to flare-bevel groove)."

The inspectors determined that WECTEC WPS Number WPS8-1.1 S03, Revision 6 allows the use of groove types not qualified by the supporting procedure qualification records (PQRs): all the PQRs referenced by WPS8-1.1S03 (PQ109, PQ109-1, PQ110, and PQ110-1) indicate the groove type as double-V-groove. No PQRs are referenced for single-V-groove welds, single bevel-groove welds, or double-bevel-groove welds, although, these groove types are allowed by WPS8-1.1 S03. The type of groove, however, is an essential variable as indicated in AWS D1.4-98 Table 6.1. Changes to essential variables beyond the limitations of AWS D1.4-98 Table 6.1 require WPS requalification in accordance with AWS D1 .4-98 subsection 6.2.1.4. As a result, the inspectors concluded that the CJP direct butt splices that were observed to be installed using single-bevel groove welds were not qualified in accordance with A\NS D1 .4-98 and, therefore, the welded reinforcing steel splices were of indeterminate quality.

The inspectors determined that the failure to adequately implement measures to assure that WPSs are qualified in accordance with applicable codes was contrary to the requirements of 10 CFR Part 50, Appendix B, Criterion IX, and was a performance deficiency.

The finding was considered more-than-minor because the performance deficiency represented a substantive failure to adequately implement a QA measure that rendered the quality of an SSC indeterminate. The finding is also similar to IMC 0613, Power Reactor Construction Inspection Report, Appendix E, example 6 which indicates, in part, that a WPS qualification issue is not minor if it is related to a change in an essential variable, and the WPS was required to be re-qualified.

The inspectors determined that the finding represented an ITAAC finding because it was material to the acceptance criteria of VEGP Unit 3 ITAAC 761, in that, if left uncorrected, the licensee may not have been able to demonstrate that the acceptance criteria of this ITAAC was met. The acceptance criteria of this ITAAC require that all deviations between the as-built structures and the approved designs be reconciled to verify that the as-built structures will withstand the design basis loads without a loss of structural integrity or other safety-related functions. The inspectors determined that the failure to adequately implement measures to assure that that special processes, including welding, are accomplished in accordance with applicable codes may have resulted in a deviation from the approved design that would not have been reconciled by the licensee.

The inspectors concluded the finding was associated with the Construction/ Installation Cornerstone. The inspectors evaluated the finding in accordance with IMC 2519, "Construction Significance Determination Process," and determined the finding was of very low safety significance (Green) because the finding affected a portion in the intermediate column of the risk importance table.

The inspectors reviewed the finding for a possible cross-cutting aspect in accordance with IMC 0613 Appendix F, "Construction Cross-Cutting Areas and Aspects," and determined the finding has a cross-cutting aspect in the Human Performance area because the licensee did not recognize that the WPS was not qualified in accordance with AWS D1.4-98. [H.9].

During this inspection, SNC questioned the NRC inspection team's interpretation of the AWS D1.4-98 requirements for WPS qualification. The NRC inspectors considered SNC's position and reviewed material provided to them by SNC. However, the inspectors did not find SNC's position to be consistent with the requirements of AWS D1.4-98. SNC requested that in lieu of an NCV, an Unresolved Item (URI) be documented until the AWS code committee could be consulted through a request for code interpretation. The NRC, however, did not believe this

was a prudent course of action in that the time required, content, and acceptability of a code interpretation were unknown. As a result, NCV 05200025/2016004-02 was dispositioned and documented in NRC Integrated Inspection Reports 05200025/2016004, 05200026/2016004, issued February 13, 2017, Accession No. ML17044A539. SNC did not contest the NCV and placed it in their corrective action program.

NRC Inspection to Close NCV 05200025/2016004-02 Since NCV 05200025/2016004-02 was associated with an ITAAC, an inspection was initiated on September 16, 2019 to close the NCV. As a result of this inspection, a more-than-minor violation of Criterion XVI, "Corrective Action" was initially proposed. Subsequently, SNC committed to take further corrective actions and the violation was held in abeyance pending review of those corrective actions.

During the inspection to close NCV 05200025/2016004-02, it was determined that in addition to not complying with the essential variable for groove type, the WPS being used for the installation of welded reinforcing bar splices in VEGP Units 3 & 4 also did not comply with the essential variable for position. As this noncompliance was similar to the example documented in NCV 05200025/2016 004-02, the intent was to discuss it in the proposed Criterion XVI violation described earlier or, subsequently, to verify that it was addressed in the belated corrective actions taken to close the original NCV. However, the inspection to close NCV 05200025/2016004-02 was suspended without any further enforcement action being taken, or adequate corrective actions being completed by SNC to my knowledge. As a result, even if NCV 05200025/2016004-01 is withdrawn, the WPS being used for the installation of welded reinforcing bar splices in VEGP Units 3 & 4 would still not conform to the AWS D1 .4-98 requirements for WPS qualification.

During this inspection, the inspectors determined that the conditions adverse to quality noted in a previous NRG-identified Green NCV 05200025/2016004-02, "Failure to qualify WPS in accordance with applicable code," had not been corrected. Specifically, WPS8-1.1S03 had not been requalified in accordance with AWS D1.4-98.

The inspectors reviewed SNC corrective action documents and determined that SNC relied on a code interpretation provided by AWS to conclude that WPS8-1.1-S03 was qualified in accordance with AWS D1.4-98, there was not a performance deficiency associated with NCV 05200025/2016-02, and no corrective actions were warranted. Specifically, CAPAL 100436639 states, in part, "Based on AWS Interpretation D1 .4-98-102, this issue 'meets requirements.' No further corrective action is necessary." However, NCV 05200025/2016004-02 documents a green ITAAC finding.

The inspectors reviewed the NRC Enforcement Policy and noted that in accordance with Section 2.3.2, the licensee must restore compliance (or demonstrate objective evidence of plans to restore compliance) within a reasonable period of time (i.e., in a timeframe commensurate with the significance of the violation) after an NCV is identified. Furthermore, it is stated in Section 2.3.2 that licensees are not required to provide written responses to NCVs; however, they may provide a written response if they disagree with the NRC's description of the NCV or dispute the validity of the NCV.

The inspectors determined that SNC did not provide a written response to the NRC disagreeing with the NRC's description of the NCV or disputing the validity of the NCV. Moreover, the inspectors determined that the code interpretation that SNC is relying on to conclude that no

corrective actions are required has never been formally reviewed or endorsed by the NRC.

Because SNC did not provide a written response to the NCV disagreeing with N RC's description of the NRC or disputing the validity of the NCV, and the code interpretation SNC is relying on to conclude that no corrective actions are warranted has not been endorsed by the NRC, the inspectors concluded that the condition adverse to quality identified in the NCV should have been corrected within a timely manner.

The inspectors determined that SNC initially took measures to address the condition adverse to quality identified in NCV 05200025/2016004-002. These actions included processing a design change in E&DCR APP-CR01-GEF-850264, "Rebar WPS Qualification" and performing additional PQRs to address conditions identified in N&D SV3-CR01-GNR-000695, "CA22 Rebar Welding Performed to GWS-8 Att. 104 Fig. 3." SNC ultimately concluded, however, that these actions were unnecessary as documented in CAPAL 100436639.

The inspectors reviewed the corrective actions taken by SNC to address NCV 05200025/2016004-002 and concluded that these measures were inadequate to restore compliance because of the following:

1. SNC did not provide a written response disagreeing with NRC's description of the NCV or disputing the validity of the NCV.

2. The code interpretation SNC relied to conclude that no corrective actions were necessary has never been formally reviewed or endorsed by the NRC.

3. WPS8-1.1S03 has not been requalified in accordance with AWS D1.4-98 as described in NCV 05200025/2016004-002.

4. The design change documented in E&DCR APP-CR01-GEF-850264 does not conform with ACI 349-01 as required by the VEGP Units 3&4 UFSAR. Specifically, ACI 349-01, Section 12.14.3.2, states, in part, that except as provided in this code, all welding shall conform to "Structural Welding Code - Reinforcing Steel" (ANSI/AWS D1.4). However, the design change documented in E&DCR APP CR01-GEF-850264 deviates from AWS D1.4-98 requirements for WPS qualification as indicated in NCV 05200025/2016004-002.

5. The additional PQRs performed to address conditions identified in N&D SV3-CR01-GNR-000695 are not adequate for requalification of WPS8-1.1 S03. Specifically,

a. AWS D1.4-98, 6.2.1.1 states, in part, that the largest bar size to be production welded shall be used for qualification. WPS 8-1.1S03 indicates that it is applicable for bar sizes #14 and below. However, the additional PQRS (PQ907, PQ908, PQ909, PQ910, PQ911, and PQ912) were all performed with #11 Bars.

b. AWS D1.4-98, 6.2.1.2 states, in part, that WPS qualification shall be performed using a steel that has a carbon equivalent (C.E.) at least equal to the highest C. E. to be encountered in production. WPS 8-1.1S03 allows a maximum carbon equivalent of 0.55. The carbon equivalent of the bars used in PQ907, PQ908, PQ909, PQ910, PQ911, and PQ912 does not appear to be recorded on the PQRs. These additional PQRs were all performed with A706 steel. The maximum C. E. allowed for A706 is 0.55. Lacking documented evidence that the bars used in the additional PQRs have a C.E.

of 0.55, it is not possible to conclude that the limit on C.E. of WPS8-1.1S03 is supported by the additional PQRs.

c. AWS D.1-98, 6.2.3, states, in part, that a WPS shall be required for each production welding position. WPS 8.1.1S03 allows all positions. However, PQ907, PQ908, PQ909, PQ910, PQ911, and PQ912 were all completed in

either the vertical (3G) or overhead (4G) positions. No PQRs were performed for the flat (1G) or horizontal (2G) positions.

AWS D1 .4 Code Interpretations ANSI/AWS D1.4 is a structural welding code that applies to the welding of reinforcing steel to reinforcing steel and the welding of reinforcing steel to carbon or low-alloy structural steel. 10 This code was first published in 1979 and has been periodically updated with the 2018 version being the most recent. The 1998 version is currently being used in the construction of VEGP Units 3

& 4.

The focus of AWS D1.4 is not nuclear safety-related construction. It is a code with a broad range of applicability that may be used for both nuclear and non-nuclear construction. The NRC has never formally endorsed AWS D1.4-98 for the welding of reinforcing bars in nuclear safetyrelated construction to my knowledge. However, it may have been approved for use in some facilities.

As stated previously, NRC guidance on the use of welded reinforcing bar splices is that they conform to ASME Division 2. The requirements for WPS qualification are addressed in the 1998 version of ASME Division 2 which is contemporary with AWS D1.4-98 . This version of the ASME code did not reference AWS D1.4, in any manner, for the welding of reinforcing bar splices. The requirements for WPS qualification of reinforcing bar CJP direct butt joints are provided in ASME Division 2-1998, Appendix XI, "Qualifications for Arc Welding Reinforcing Bars." 11 Similar to other codes, AWS has provisions for requesting interpretations of the welding codes published by the organization. These interpretations are typically used to clarify existing requirements or address errors. They are not intended to alter existing requirements. Code revisions follow a separate process from that used for interpretations.

Interpretations of AWS D1.4 are made by the AWS D1 Committee on Structural Welding. This committee is responsible for creating and revising codes and standards on structural welding. It publishes a group of standards that address different structural welding applications. This group of standards includes AWS D1.4. The main AWS D1 committee has a number of subcommittees which oversee specific standards. The AWS D1I Subcommittee on Reinforcing Steel is responsible for the development and maintenance of AWS D1.4. Requests for interpretations of AWS D1.4 are reviewed and responses to the requests developed and approved by the AWS D1 I subcommittee. These responses are then reviewed and balloted by the main AWS D1 committee.

The AWS D1I Subcommittee on Reinforcing Steel is primarily composed of industry representatives. A review of the subcommittee members listed in the most recent edition of the standard -AWS D1.4/D1.4M:2018 - indicates that, excluding the Secretary of the Subcommittee who is an employee of AWS, 5 of the 12 subcommittee members appear to represent companies that have or have been involved in the construction of VEGP Units 3 & 4.

10 The connection of a reinforcing bar to another reinforcing bar with a welded joint is typically referred to as a welded reinforcing bar splice while the latter type of connection is typically encountered when reinforcing steel bars are used for the anchorage of structural steel embedment or anchorage plates.

11 Beginning with the 2013 edition of ASME Division 2, AWS D1.4 was referenced in Appendix VIII for WPS and welder qualification. Specifically, AWS D1.4-2011 is referenced.

On most nuclear code committees, such as those responsible for nuclear standards prepared by ASME, AISC, 12 or ACI, industry representation would be balanced in part by NRC involvement. Additionally, the primary codes used in the design and construction of nuclear power plants are typically reviewed and endorsed by the NRC. However, the NRC does not have a representative on either the main AWS D1 committee or the AWS D1I Subcommittee on Reinforcing Steel, nor has AWS D1.4-98 been formally endorsed by the NRC to my knowledge.

In fact, there does not appear to be a single representative of a federal agency on either the current main AWS D1 committee or the AWS D1I subcommittee.

The NRC does not by default accept or endorse code interpretations for use in nuclear safety related construction to my knowledge. 13 In general, any code interpretations should be carefully reviewed and evaluated by the NRC for their reasonableness and suitability for use on nuclear projects prior to their acceptance or endorsement. That is especially so where a high percentage of the subcommittee members have a relationship to a specific project on which the interpretation is proposed to be used and the NRC lacks involvement in the development of that interpretation.

Subjecting code interpretations to a thorough review, prior to allowing their use on a nuclear project, is a prudent measure that aligns with the standard NRC practice of not endorsing code interpretations by default. Such a review of code interpretation D1.4-98-102 raises questions about the intent of this interpretation and its relationship to long-standing code requirements that have a well-established history of use. Far from clarifying matters, code interpretation D1.4 102 creates unnecessary confusion and could be read to upend existing code requirements associated with WPS qualification. Code interpretation D1.4-18-101, on the other hand, appears to substantially revise an existing code provision outside of the standard process. 14 AWS Code Interpretation D1.4-98-102 Code interpretation D1 .4-98-102 was requested by the Chief Engineer, Welding Technology &

Services, WECTEC Global Project Services, Inc. which was a contractor at VEGP Units 3 & 4 at the time.

The code interpretation is as follows:

Inquiry: Question 1: If a WPS is qualified per 6.2.4.1 and in accordance with Table 6.2, does qualification using one of the "Production type of joint" listed, qualify for all the corresponding "Joints Qualified per Figure" shown for that row? (E.g., Testing with "Direct Butt [6.5(A)]" qualifies the WPS for all direct butt joints shown in figure 3.2)?

12 American Institute of Steel Construction. This organization publishes AISC N690, "American National Standard Specification for the Design, Fabrication, and Erection of Steel Safety-Related Structures for Nuclear Facilities."

13 Do so, would potentially subordinate the NRC to a non-governmental body and result in an inappropriate delegation of responsibility for nuclear safety.

14 AWS Dl.4/Dl.4M:2018 is designated as an American National Standard. As such, code revisions should follow the process described in American National Standards Institute (ANSI), "Due process requirements for American Nationa I Standards."

n Question 2: Does essential variable #11 of Table 6.1 prohibit a change in groove type where such allowances are permitted or required by Table 6.2? (E.g., Figure 6.S(C) flare-bevel groove is required to qualify Figure 3.4 flare-vee groove.)

Response: Question 1. Yes Question 2. No The responses to code interpretation D1.4-98-102 imply that qualification using a single direct butt joint - presumably two test assemblies configured as one of the joints depicted in Figure 3.2 - qualifies all of the joints of Figure 3.2. There does not appear, however, to be any basis for this reading in AWS D1.4-98.

Direct butt joints made with CJP groove welds are addressed in Table 3.1 and Figure 3.2 of AWS D1.4-98. The appropriate groove weld types for bars with either a horizontal or vertical orientation are provided in Table 3.1. The corresponding joints are graphically depicted in Figure 3.2. The position of the weld for each joint is determined from Figure 6.3.

Both the responses of code interpretation D1.4-98-102 refer to the requirements of Table 6.2.

This table addresses the configuration and number of specimens required to qualify a WPS for a specific joint type. It does not provide relief from any code provisions but must be used in conjunction with all other applicable provisions.

The test assembly referenced in Table 6.2 for the qualification of CJP direct butt joints is that depicted in Figure 6.S(A). The test assembly provided in Figure 6.S(A), however, is only a graphical representation of a CJP direct butt joint. No single all-encompassing test assembly is provided in Figure 6.S(A). The essential variables prevent such a test assembly from being defined for CJP direct butt splices. In effect, Figure 6.S(A) defines not one, but a group of test assemblies. The size of, and variation within, the group depends on the specific joints being qualified. Recognizing these facts about the test assembly provided in Figure 6.S(A) is key to understanding Table 6.2 and this code interpretation. In practice, the test assemblies used for WPS qualification of a CJP direct butt joint will conform to one or more of the joints depicted in Figure 3.2.

The text accompanying Figure 6.S(A) clearly states that "the groove used shall conform to that being qualified." As a result, Table 6.2 by its reference to Figure 6.S(A) does not allow qualification of a CJP direct butt joint that differs in groove type from the test assembly. Doing so would be inconsistent with the essential variable for groove type. As a result, the test assembly defined in Figure 6.S(A) varies depending on the CJP joint or joints being qualified, and multiple test assemblies are required to qualify all the joints of Figure 3.2.

AWS D1 .4-98 subsection 6.2.1.4 requires WPS requalification for changes exceeding the essential variables limitations of Table 6.1. The joints depicted in Figure 3.2 do not share a common position or groove type which are defined as essential variables 10 and 11, respectively, in Table 6.1. These two essential variables should be dealt with in an analogous manner for WPS qualification. If Table 6.2 allowed a single CJP direct butt joint to qualify all the joints of Figure 3.2, position and groove type would not be essential variables and their listing in Table 6.1 an error. The same case could be made for essential variables 12 and 13. 15 15 Essen tial variable 12 involves a change in the groove shape or decrease in gr oove an gle, decrease in root open in g, or an increase in the root face. This essential variable was removed from Table 6.1 in AWS Dl.4/D.14M :2011. Essen tial variable 13 involves the omission of backin g material.

The joint depicted in Figure 3.2(C) is a single-v-groove weld with split pipe backing. This joint could not be used to qualify any of the other joints depicted in Figure 3.2 because none of them are made with backing material. Doing so would exceed the limitations of essential variable 13.

Likewise, because the joints in Figure 3.2 do not all share a common position or groove type, none of the joints depicted in Figure 3.2 could be used to qualify all the joints in the figure without exceeding the limitations of these essential variables.

The minimum number of test assemblies is provided in Table 6.2. The actual number required to qualify a WPS depends on the configuration and details of the joints being qualified. 16 None of the CJP direct butt joints of Figure 3.2 are prequalified. 17 As a result, all the joints shown in Figure 3.2 need to be individually qualified by test. The only exception would be if the differences between the joints did not exceed the essential variable limitations of Table 6.1. In this case, a single set of test assemblies could be used to qualify multiple joints. 18 The status of position and groove type as essential variables is reflected in multiple areas of AWS D1.4-98:

1. AWS D1.4-98 Table 6.1 defines position as an essential variable . 19

2. AWS D1.4-98 subsection 6.2.3 states that "[a] WPS shall be required for each production welding position." This requirement is consistent with essential variable 10.

The test positions for WPS qualification are provided in Figure 6.1.

3. AWS D1.4-98 Table 6.1 defines groove type as an essential variable.

4. AWS D1.4-98 Table 6.1 Essential Variable 12 involves a change in the groove shape or decrease in groove angle, decrease in root opening, or an increase in the root face. This essential variable places further restrictions on changes in groove geometry.

5. The text accompanying Figure 6.5 states that "the groove used shall conform to that being qualified." This statement is consistent with essential variables 11 and 12 and reinforces the requirement that the groove type and details match between the test assemblies and the production joint being qualified. As a result, Table 6.2 does not allow qualification of a CJP direct butt joint that differs in groove type from the test assembly.

The use of essential variables for WPS qualification is not novel to the AWS D1.4 code.

Essential variables, or qualification variables as they are sometimes known, have been used in many other codes. 20 While the specific variables may differ, the role of essential variables in the 16 It is also dependent on whether the tensile test specimens can be reused for the macroetch tests. Based upon the number of tensile and macroetch tests alone, a total of 4 specimens may be required.

17 Subsection 6.1.2.1 requires WPS qualification by testing for all joint types except those made with fillet welds. 18 For example, a set of two test assemblies configured using a double-bevel groove type could be used to qualify a direct butt splice between bars of a different diameter (Figure 3.1), the same diameter [Figure 3.2(0)], or a T-joint

[Figure 3.5(0)] provided the position, groove type, and groove configuration are within the limitations of essential variables 10, 11, and 12, respectively.

19 The reference to 6.2.4 in Table 6.1 of AWS 01.4-98 appears to be an inadvertent error. This section was revised and renumbered between AWS 01.4-92 and AWS 01.4-98. The change does not appear to have been carried over to AWS 0.14-98 when the code was revised. AWS 01.4-92 subsection 6.2.4 requires that welds be welded in the position for which the procedure is to be qualified.

20 AWS B2.1, AWS 01.1, AWS 01.2, AWS 01.5, AWS 01.6; ASME BPV Code Section IX, ASME Division 2 (ACI 359);

etc.

WPS qualification process is well established. In standard practice, deviating from the essential variables requires WPS requalification.

The AWS 01.4-98 requirements for WPS qualification are consistent with those of the contemporary ASME standard addressing welded reinforcing bar splices.21 The ASME code also defines position and groove type as essential variables and requires WPS requalification if an essential variable is changed. 22 No table similar to Table 6.2 is provided in the ASME standard.

The merit of defining groove and position as essential variables can be debated. However, they have been designated as essential variables in AWS 01.4-98 and must be appropriately addressed during WPS qualification and documented in the WPS.23 As defined, these variables are "essential" to WPS qualification and they should be treated equally to any other essential variable.

Concerning Question 2 of the code interpretation, the specific example provided does not apply to CJP direct butt joints. As discussed earlier, Table 6.2 does not allow or require a test assembly to be used for the qualification of a CJP direct butt splice that deviates from the essential variables. In addition, the failure to assign the joints in Figure 3.4 to an appropriate row of Table 6.2 based on groove type appears to be an inadvertent error in the code which was corrected in AWS 01.4/01.4M:2018.

Essential variable 11 in Table 6.1 is defined as "[a] change in groove type (e.g., flare-Vto flare bevel groove)." As a result, the joints depicted in Figure 3.3 appear to have been assigned to different rows in Table 6.2 to provide consistency in the groove type between the test assemblies and the joints being qualified: 3.3(A) and 3.3(8) assigned to the third row since they are made with flare-bevel groove welds similar to 6.5(C); and 3.3(C) assigned to the fourth row since it is made with flare-V groove welds similar to 6.5(0). Following this logic, it appears reasonable to conclude that the joints depicted in Figure 3.4 would be treated in a similar manner: 3.4(8) assigned to the third row of Table 6.2 since it is made with flare-bevel groove welds similar to 6.5(C); and 3.4(A) assigned to the fourth row since it is made with flare-V groove welds similar to 6.5(0). However, both the joints depicted in Figure 3.4 are assigned to the third row of Table 6.2 even though they differ in groove type. It does not seem logical that the joints depicted in Figure 3.3 would be assigned to a row of Table 6.2 based on groove type, but those of Figure 3.4 vvould not. Additionally, listing 3.4(A) in the third row of Table 6.2 conflicts with subsection 6.2.1.4.

Table 6.2 and Figure 6.5 were revised between AWS 01.4-92 and AWS 01.4-98. In the earlier edition, it was up to the organization performing the qualification to select the appropriate test assemblies to qualify flare-bevel and flare-Vgroove welds, taking into consideration the essential variables. The inclusion of 3.4(A) in the third row of Table 6.2 appears to be an inadvertent oversight resulting from the revisions between AWS 01.4-92 and AWS 01.4-98.

This oversight appears to have been corrected in AWS 01.4/01.4M:2018.

21 ASME Division 2 -1998 (ACI 359-98) 22 The example given in ASME for the "groove type" essential variable is a change from a single-Vee to a double Vee.

23 Position and groove type are designated as essential variables in the following editions of AWS D1.4: 1979, 1992, 1998, 2005, and 2011.

Concerning essential variable 10, it is important to note that the subsection addressing position of test welds was revised and renumbered from subsection 6.2.4 to 6.2.3 betvveen AWS D1.4-92 and AWS D1.4-98. However, the referenced subsection in AWS D1.4-98 Table 6.1 essential variable 10 does not appear to have been updated to reflect the renumbering. 24 This error also appears to have been corrected in AWS D1.4/D1.4M :2018.

Non-concurring Employee's Requests for AWS D1.4-98 Code Interpretations AWS interpretation D1.4-98-102 could be read to contradict the plain language of the code and upend existing code requirements and industry standard practices for the use of essential variables in WPS qualification. While, SNC did not request any action be taken, regional staff and management, at the time, believed it would be prudent to seek clarification from AWS when this interpretation was provided to us by SNC. However, it was determined at the time, that as a matter of policy, the N RC does not typically initiate requests for code interpretations. As a result, I personally submitted three requests for interpretation of AWS D1.4-98 in an attempt to resolve the confusion AWS created with code interpretation D1.4-98-102.

The requests for interpretation of AWS D1.4-98 vvere submitted with clear background information supporting the proposed responses. The information provided was consistent with the earlier discussion about code interpretation D1.4-98-102. One would think AWS would want to resolve the confusion created by code interpretation D1.4-98-102, and that it would be a simple matter to do so, by responding to the requests for interpretation outlined below. They are not complicated. These requests for interpretation were submitted in March 2018, however, I have yet to receive a response from AWS. These requests for interpretation are provided below as they appear to have some bearing on code interpretation D1.4-18-101.

Inquiry #1

Subject:

Use of essential variables for WPS qualification Code Edition: AWS D1.4-98 Provision: Table 6.1 Question #1: Subsection 6.2.1.4 requires WPS requalification for changes exceeding the essential variables limitations of Table 6.1. The response to question 1 of code interpretation AWS D1.4-98-102 implies that qualification using a single direct butt joint - presumably two test assemblies configured as one of the joints depicted in Figure 3.2 - qualifies all of the joints of Figure 3.2. The joints depicted in Figure 3.2, however, do not share a common position or groove type which are defined as essential variables 10 and 11, respectively, in Table 6.1. Are position and groove type erroneously listed as essential variables in Table 6.1?

24 It also should be noted that the reference in Table 6.1, essential variable 10 was revised to subsection 6.3.4 in the errata to AWS Dl.4/Dl.4M:2005. However, subsection 6.3.4 is not applicable to WPS qualification. It is associated with welder qualification. The correct reference in Table 6.1, essential variable 10 should be subsection 6.2.3. This can be verified by reviewing editions of AWS D1.4 previous to 1998. SNC appears to have erroneously used the errata to AWS D1.4/D1.4M:2005 for WPS qualification.

Question #2: Must the test assemblies and the production joint being qualified conform to one another within the limits of the essential variables?

Proposed Response:

Question #1: No Question #2: Yes Inquiry #2

Subject:

Alignment between test assemblies and joints being qualified.

Code Edition: AWS D1.4-98 Provision: Table 6.2 Question: Should the joint depicted in Figure 3.4(A) be assigned to the fourth row of AWS D1.4-98 Table 6.2 since it is made with flare-V groove welds similar to 6.5(0)?

Propose Response: Yes Inquiry #3

Subject:

Correct referenced subsection Code Edition: AWS D1.4-98 Provision: Table 6.1 Question: Should the correct reference in Table 6.1 essential variable 10 be 6.2.3?

Proposed Response: Yes AWS Code Interpretation D1.4-18-101 and AWS Code Amendment D1.4/D1 .4M:2018-AMD1 Code interpretation D1.4-18-101 and associated amendment D1.4/D1.4M:2018-AMD1 were not requested by any party external to AWS to my knowledge. My understanding is that both the code interpretation and associated amendment were developed in response to the requests for code interpretation that I submitted to AWS.25 Those requests, however, were for an interpretation of AWS D1.4-98 not AWS D1.4/D1.4M:2018. They were intended to clarify the use of the essential variables of AWS D1.4-98 in the WPS qualification process.

Instead of responding to the requests for interpretation that I submitted, the AWS D1 I subcommittee appears to have taken it upon themselves to provide an interpretation of AWS 25 Email correspondence from Secretary, D1.41 Subcommittee on Reinforcing Steel, dated December 12, 2019.

D1 .4/D1 .4M:2018. It is unknown why this action was believed necessary and prioritized over responding to the requests for interpretation of AWS D.14-98 that I submitted.

Code interpretation D1.4-18-101 and code amendment D1 .4/D1 .4M :2018 do not address the issues that require clarification and prompted the requests for code interpretation that I submitted or resolve the confusion the AWS subcommittee created with code interpretation D1 .4-98-102.

The inquires and responses are as follows:

Inquiry 1: Is position an essential variable in D1 .4/D1 .4M: 2018 Table 8.2?

Interpretation 1: Yes Inquiry 2: Is it the intent of D1.4/01.4M: 2018 Table 8.2, item 11 "A change in groove type (e.g. flare-Vto flare-bevel groove)" to actually refer as "A change in type of joint (e.g. direct butt to indirect butt) of D1 .4/D1 .4M:

2018 Table 8.3?"

Interpretation 2: Yes Inquiry 3: Must the test assemblies and the production joint being qualified in D1 .4/D1 .4M:2018 conform to one another within the limits of the essential variables?

Interpretation 3: See Inquiry and Response to Question 2 and answer of the D1.4-98-102 interpretation.

Inquiry 4: Do direct butt welds (Figure 8.5A) in D1 .4/01 .4M:2018 Table 8.3 qualify single V groove, double V groove, single bevel and double bevel groove as shown in D1 .4/D1 .4M:2018 Figure 5.2?

Interpretation 4: Yes The first inquiry and response confirm that position is an essential variable. As discussed earlier, the use of this essential variable in WPS qualification is analogous to that of groove type. Since the joints depicted in Figure 3.2 do not share a common position or groove type, it does not appear reasonable to conclude that the response to Question 1 of code interpretation D1 .4-98-102 allows a single test assembly to qualify all the joints depicted in Figure 3.2 without regard to these two, or any other, essential variables. Doing so, would not align with the purpose of essential variables in the WPS qualification process or conform to AWS D1.4-98 subsection 6.2.1.4 which requires WPS requalification for any changes beyond the essential variable limitations of Table 6.1.

The second inquiry and response revise essential variable 11 from groove type to joint type.

This change is not applicable to AWS D1 .4-98. To implement this change, the essential variable in Table 8.2 is changed from groove type to joint type, the example provided in Table 8.2 of a change in groove type is deleted, and the text accompanying the text assemblies shown in Figures 8.5(A) and 8.5(8) which require the groove used to conform to that being qualified is deleted. These are substantive changes to a code provision.

In responding to Inquiry 3, AWS strays from their standard practice of responding to a request for code interpretation with a simple yes or no answer. It is also unusual that an interpretation of an earlier revision of a standard would be referenced, especially as the code requirements and numbering of the relevant code sections has changed between revisions, adding unnecessary confusion. When reviewing this inquiry and response, it is important to keep in mind that the essential variable for groove type was revised to joint type in the preceding inquiry and response of the code interpretation. As a result, the inquiry and response to Question 2 of code interpretation AWS D1 .4-98-102 is not relevant to AWS D1.4/D1.4M :2018 as it references the older essential variable 11. Moreover, the lap joints depicted in Figure 3.4 that are cited in the example provided in Question 2 of code interpretation D1.4-98-102 are assigned to different rows in Table 8.2 of AWS D1 .4/01 .4M:2018 than in Table 6.2 of AWS D1.4-98. Far from clarifying the requirements for WPS qualification, the response to Inquiry 3 of code interpretation D1.4-18-102 only muddies the water even more. Regardless, it is not applicable to AWS D1.4-98.

As with the third inquiry and response, it is important to remember, when reading the fourth inquiry and response, that the essential variable for groove type was revised to joint type in the second inquiry and response of code interpretation D1.4-18-101. As a result, the groove types referenced in the fourth inquiry and response are no longer essential variables. As a result, they do not need to be accounted for in the WPS qualification process of AWS D1.4/D1.4M:20018 as they do in AWS D1.4-98.

Essential variables must be controlled during the WPS qualification process. Changes beyond the essential variable limitations require WPS requalification. This is a requirement of AWS D1 .4-98 and AWS D1.4/01.4M :2018 subsections 6.2.1.4 and 8.2.1.4, respectively. The code committee did not revise this requirement in either code interpretation D1.4-98-102 or D1 .4 101. It is fundamental to the definition of essential variables and their use in the WPS qualification process of the AWS D1 group of standards, as well as, other industry standards that address WPS qualification.

Inspections, Tests, Analyses, and Acceptance Criteria (ITAAC)

Noncomplying reinforcing bar splices were first observed in the reinforced concrete portions of the Unit 3 cylindrical shield wall which fall under ITAAC 3.00.02a.i.b (761). Since these initial observations, welded reinforcing bar splices also have been observed in other Seismic Category I structures as well. As a result, VEGP Units 3 & 4 ITAAC 3.3.00.02a.i.a (760),

3.3.00.02a.i.b (761), and 3.3.00.02a.i.d (763) may also be affected.

These deviations are material to the acceptance criteria of the ITAAC listed above, in that, if left uncorrected, the licensee may not be able to demonstrate that the acceptance criteria of these ITAAC have been met. The acceptance criteria of these ITAAC require that all deviations between the as-built structures and the approved designs be reconciled to verify that the as-built structures will withstand the design basis loads without a loss of structural integrity or other safety-related functions. The failure to adequately implement measures to assure that special processes, including welding, are accomplished in accordance with applicable codes may result in deviations from the approved design that could not be reconciled by the licensee.

Conclusions/Recommendations The welded reinforcing bar splices used in the construction of Seismic Category I structures of VEGP Units 3 & 4 have not been installed using an appropriately qualified Welding Procedure Specification (\NPS).

Neither code interpretation D1 .4-98-102 nor D1.4-18-101 provides a sound basis for concluding that the WPS being used for the installation of welded reinforcing bar splices in VEGP Units 3 &

4 complies with AWS D1.4-98, or that NCV 05200025/2016004-02 was dispositioned in error and should be withdrawn.

NCV 05200025/2016004-02 should remain open until it can be verified that appropriate corrective actions have been taken to address the noncompliance. Additionally, SNC's failure to take appropriate corrective actions in a timely manner should be dispositioned as an ITAAC finding and associated more-than-minor violation of the Code of Federal Regulations (10 CFR)

Part 50, Appendix B, Criterion XVI, "Correction Action" for SNC's failure to adequately implement measures to assure that conditions adverse to quality were promptly corrected. The significance of this violation should be dependent on whether SNC can requalify the WPS to bound the welded reinforcing bar splices installed in VEGP Units 3 & 4 within the limitations of the essential variables of AWS D1.4-98, or provide evidence demonstrating that the Structures, Systems, and Components (SSCs) that rely on welded reinforcing bar splices will be able to adequately perform their safety functions. SNC's corrective actions should address deviations from the essential variable for position in addition to groove type which was provided as an example in NCV 05200025/2016004-02.

Reinforced Concrete Construction Concrete is strong in compression but weak in te nsion. When subjected to loads that create tension, cracks will occur in a concrete structure if the tensile strength of the concrete is e xce e de d. These cracks can lead to a sudden catastrophic failure due to the inability of the concrete to resist the tensile forces.

In a reinforced concre te structure , deformed steel bars, typically referred to as reinforcing bars, are embedded in the concrete in such a way that the tensile forces can be resisted by the bars after the concrete cracks. 26 The tensile forces are the result of internal stresses that develop in the concrete and are associated with axial, she ar, and torsion forces and bending mome nts. In some instances, the reinforcing bars may also be used to supplement the concrete in resisting compression force s. The reinforcing bars resemble circular rods with raised ridges to lock the bar to the surrounding concrete.

The reinforcing bars are essential components of a reinforced concrete structure that are necessary to resist the loads imposed on the structure and maintain structural inte grity. Ideally, continuous re inforcing bars would be installe d. Howe ve r, this is generally not feasible given the standard manufactured lengths of the bars, structural details, and construction considerations.

As a result, the reinforcing bars are typically spliced together at various locations within the structure . The design function of the splice is to join two reinforcing bars together so that the assembly behaves as one continuous reinforcing bar, typically, having the same diameter as the spliced bars but a longer length.

It should be clear from the above description that reinforcing bar splices are key components in reinforced concrete construction. The splices are used to provide continuity for the transfer of forces within reinforced concrete structure s. They are a potential "weak link" in the structural system and must be installed correctly to ensure acceptable pe rformance . Inadequately installed reinforcing bar splices can lead to structural failure.

Reinforcing bars may be spliced together with lap splice s, mechanical connections, 27 or welded splices. 28 Regardless of the type of splice use d, the splices are typically designed to develop at le ast 125% of the yield strength of the bar being spliced. 29 These performance requirements are intended to provide ductility in the structure which is necessary to ensure adequate structural behavior. 30 26 James G. MacGregor, Reinforced Concrete Mechanics & Design, 2nd ed., Prentice-Hall, Englewood Cliffs, N.J.,

1992, p. 1 27 "Manual of Standard Practice," Concrete Reinforcing Steel Institute (CRSI), 28th Edition, 2nd printing, 2011 defines a mechanical splice in the glossary as a complete assembly of an end-bearing sleeve, a coupler, or a coupling sleeve, and possibly additional material or parts to accomplish the splicing of reinforcing bars; ACI 439-07 has a similar definition and indicates that a "mechanical splice" was formerly called "mechanical connection" in ACI 318-95 and earlier editions.

28 The CRSI Manual of Standard Practice defines a welded splice in the glossary as a means of joining two reinforcing bars by electric arc welding. Bars may be lapped, butted, or joined with splice plates or angles.

29 The performance requirement that mechanical and welded splices develop at least 1.25Fy of the bar aligns with ACI 349-01, subsection 21.2.5 which requires the ratio of the actual ultimate tensile strength to the actual tensile yield strength to be not less than 1.25 for reinforcement in members resisting earthquake-induced forces. Later editions of ACI 349 require that mechanical and welded splices develop the full specified tensile strength of the bar.

30 Ductility is the ability of a material to deform beyond the elastic range without fracture.

Lap splices are typically the most frequently encountered type of reinforcing bar splice. In a lap splice, two bars are laid adjacent to one another with the end of each bar overlapping the other for a calculated distance known as the lap splice length. The force is transferred from one bar to another through the concrete between the bars. For large diameter bars, the lap splice length can be many feet. While prevalent in reinforced concrete construction, lap splices are not allowed for very large bar sizes, increase reinforcement congestion, require adequate space to be installed, and have other limitations that may make them impractical or impossible to install in some instances. In these situations, a mechanical or welded splice may be installed.

A mechanical connection or splice typically makes use of a sleeve that overlaps each bar end to provide a connection between two bars. The sleeve and ends of the bars may be threaded much like a nut and bolt, or the sleeve filled with metal or grout to form a connection. There are many different systems of mechanical connections. Hovvever, all of them rely on a mechanical connection between the bars to transfer forces. In a welded splice, on the other hand, the bars are fused together either directly or indirectly by electric arc welding.

Of the three types of reinforcing bar splices, welded splices require the most stringent controls over the splicing process and the highest level of skill to install correctly. Welding is also typically the least cost-effective method of splicing reinforcing bars. Given these factors, welded splices have generally been used sparingly in the general construction industry. Recommended practice is to minimize their use if possible. 31 This recommendation is partly due to concerns about consistently producing welded reinforcing bar splices that meet the specified performance and quality requirements. If used, however, careful consideration should be given to the quality assurance measures implemented to ensure that welded reinforcing bar splices are installed correctly and will perform adequately.

The main type of welded splice being installed in VEGP Units 3 & 4 is a direct butt joint. 32 In this type of splice, the ends or the bars are beveled at an angle such that a groove is formed when the bars are butted together end to end. The groove between bars is filled with weld material to fuse the two bars together. The weld joining the bars together is referred to as a complete joint penetration (CJP) groove weld. The weld has a cross-section area equal to that of the bars joined. By design, the weld has the same strength as the bars and can be highly stressed.

Quality Assurance Measures for Welded Reinforcing Bar Splices Welding is a special process. ASME NQA-1-1994 defines a special process as "a process, the results of which are highly dependent on the control of the process or the skill of the operators, or both, and in which the specified quality cannot be readily determined by inspection or test of the product."

Given the definition above, it should be clear that to repeatedly produce quality welds, the process must be controlled, and the welders possess the requisite skill. Also, the quality of the weld and its ability to meet its design function cannot be readily determined from a visual 31 "Manual of Standard Practice," Concrete Reinforcing Steel Institute (CRSI), 28th Edition, 2nd printing, 2011, p 1-4 and Flathua, W.J., "Dynamic Tests of Large Reinforcing Bar Splices," U.S. Army Engineer Waterways Experiment Station, Technical Report N-71-2, Vicksburg, MS, April 1971 p. 141.

32 Also referred to as a direct butt splice when two reinforcing bars are spliced together.

examination of the weld alone. As a result, advanced methods are typically used to verify the mechanical properties and quality of production welds in nuclear safety-related construction.

Many variables can affect the soundness or quality of a weld. Some of these can be controlled such as the filler material, groove type, position, travel speed, etc. Those variables which affect the mechanical properties or soundness of the welds are typically defined as qualification or essential variables.

Welding procedure specifications (WPS) are used to communicate the parameters by which a particular weld is to be made. These parameters include the essential variables. Qualification of the WPS provides evidence that welds having the required mechanical properties or soundness can be achieved in test coupons by a qualified welder using the WPS. Changes beyond the limitations of the essential variables require requalification of the WPS.

Welder performance qualification tests are intended to document the ability of a welder to make sound welds by following a WPS. These tests are typically performed in a controlled environment. The ability of welders to repeatedly produce quality welds varies depending on skill level, experience, and other factors. Moreover, the satisfactory completion of a qualification test does not guarantee that a welder can produce sound welds for all conditions that might be encountered during production welding. Welded reinforcing bar splices in particular require welders to acquire and retain specific skills necessary to do quality work. 33 Production welds are often made in conditions that differ from those of the tests used for WPS or welder qualification. For example, access may be limited, or environmental conditions may be different. As a result, the soundness and quality of production welds must be verified through test and inspection.

The type of inspections conducted to verify weld quality often correlate to the type of weld and associated stress. Typically, all welds are visually examined, and, if warranted, examinations using more advanced nondestructive examination (NOE) methods are performed on a sampling basis. Highly stressed vvelds that are used in critical applications typically receive a greater level of inspection than lower stressed welds used in secondary applications. Welds in which the available types of advanced inspection methods are limited due to feasibility, such as partial joint penetration (PJP) or fillet welds, are typically designed to lower allowable stresses or higher factors of safety, depending on the design method, to compensate in some manner for uncertainty in the quality of the weld. 34 CJP welds used in critical applications typically warrant more than visual examination. 35 33 "The Procedure Handbook of Arc Welding," The James F. Lincoln Arc Welding Foundation, 14th Edition, 2000, p 13.4-4.

34 P artial joint penetration welds have a throat dimension less than the thickness of the materials joined while fillet welds are applied to the surface or edges of the materials joined. These types of welds are designed with a Safety Factor of 2 per AWS D1.4/D1.4M :2018 Table 4.1.

35 This statement is true whether the weld is used in a mechanical system, a structural steel beam, or a reinforcing bar splice. The weld, after all, does not know where it is used or what code of record governs the design and construction of the system in which it is installed. CJP welds used in risk significant Seismic Category I structures designed to ACI 349 are not by nature any less important to safety than CJP welds used in other components that are designed to other codes. The importance of the welds to structural integrity and the risk significance of the systems in which they are installed need to be considered.

These inspections are in addition to visual examination and are important because visual examination only looks at the surface of the weld. Visual examination cannot identify any subsurface defects such as lack of fusion, porosity or slag inclusions that could impact the strength of the weld, nor can it reliably identify microscopic surface cracks or defects. The only methods capable of inspecting the full weld cross sections are ultrasonic (UT) or radiographic examination (RT). All welds subject to ultrasonic or radiographic examination must first pass visual examination. Typically, all welds are visually examined, and ultrasonic or radiographic examination is conducted on a sampling basis. If defects are found using NOE methods other than visual, a greater number of welds may be inspected using these methods to determine if there is a larger problem with the welding procedure or welder. Welds that do not meet quality requirements are typically repaired.

Since the mechanical properties of welds cannot be determined through non-destructive examination alone, some destructive testing of CJP welded reinforcing bar splices has also typically been performed. This testing may be conducted on production joints cut directly from in-place reinforcement or removable "sister joints" made adjacent to production joints and under the same conditions. If the tested joints do not meet strength requirements, additional joints are tested, and the cause of the problem investigated.

For structures having a high risk to public health and safety in the event of damage or failure, it is common practice to specify that CJP welded reinforcing bar splices be subject to either destructive testing, radiographic examination, or a combination of both, in addition to visual examination, to verify that the required performance and quality requirements have been met.

NRC guidance on the use of welded reinforcing bar splices aligns with this practice.