ML20207M189

SER Accepting Relief Request for Authorization of Alternative Reactor Pressure Vessel Exam for Circumferential Weld

ML20207M189
Person / Time
Site:	Hatch
Issue date:	03/11/1999
From:	NRC (Affiliation Not Assigned)
To:
Shared Package
ML20207M184	List: ML20207M177 ML20207M189
References
NUDOCS 9903190012
Download: ML20207M189 (15)

Text

_

~

Moog

• T" 4

UNITED STATES

{

j NUCLEAR REGULATORY COMMISSION

'2 WASHINGTON, D.C. 20555-0001 SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION REQUEST FOR AUTHORIZATION Of ALTERNAT(VE REACTOR PRESSURE VESSEL EXAMINATIONS FOR CIRCUMFERENTIAl WELDS SOUTHERN NUCLEAR OPERATING COMPANY. INC.. ET AL.

EDWIN 1. HATCH NUCLEAR PLANT. UNIT 1 DOCKET NO. 50-321

1.0 INTRODUCTION

By letters dated December 2,1998, and January 19 and February 5,.1999, Southen. Nuclear Operating Company, Inc. (SNC, the licensee) submitted a request seeking approval for an alternative examination program to the inservice inspection examination requirements for circumferential welds h the Edwin I. Hatch Nuclear Plant, Unit 1 (Hatch-1). Title 10 of the Code of Federal Reaulations (10 CFR) Section 50.55a(g)(6)(ii)(A)(2) requires licensees to augment their inspection programs by implementing once, as part of the inservice inspection 4.erval (ISI) that was in effect on September 8,1992, examinations of reactor pressure vessel (RPV) shell welds, as specified in item B1.10 of Examination Category B-A, " Pressure Retaining Welds in ithe] Reactor Vessel," to Table IWB-2500 in Subsection IWB of the 1989 Edition to Section XI of the American Society of Mechanical Engineers (ASME) Boiler and Pressure Vessel Code (ASME Code). SNC is seeking relief from these requirements pursuant to 10 CFR 50.55a(g)(6)(ii)(A)(S), which allows licensees to propose alternatives to the augmented inspection requirements when the requirements are determined to be impractical, if the alternative provides an acceptable level of quality and safety in lieu of implementing the requirements, as provided in 10 CFR 50.55a(a)(3)(i).

On January 19,1999, SNC supplemented its submittal of December 2,1998, and requested that the staff approve two additional alternatives to the augmented inspection requirements specified in 10 CFR 50.55a(g)(6)(ii)(A). SNC is specifically seeking approval to: (1) use a GERIS 2000 ultrasonic examination system (GERIS 2000) as the system for performing the volumetric examinations of the longitudinal welds in the Hatch-1 RPV, and (2) achieve less than 90 percent coverage when the volumetric examinations of the longitudinal welds are conducted.

SNC's basis for relief and the staff's evaluation of SNC's attemative programs for the volumetric examinations of the circumferential and longitudinal welds in the Hatch-1 RPV is provided in Section 10 of tinis Safety Evaluation (SE).

9903190012 990311 F

PDR ADOCK 05000321 Enclosure P

PDR e

l 1

' l

2.0 BACKGROUND

INFORMATION "BWR VESSEL AND INTERNALS PROJECTS. BWR REACTOR VESSEL SHELL WELD INSPECTION RECOMMENDATIONS (TECHNICAL, REPORT BWRViP-05)"

By !stter dated September 28,1995,' the Boiling Water Reactor Vessel and Internals Project (BWRVIP), a technical committee of the Boiling Water Reactor Owners Group (BWROG),

submitted the proprietary report, "BWR Vessel and Internals Project, BWR Reactor Vessel

{

Shell Weld Inspection Recommendations (BWRVlP-05)," which proposed to reduce the scope of inspection of the BWR RPV we!ds from essentially 100 percen! of all RPV shell welds to 50 percent of the longitudinal welds and 0 percent of the circumferential welds. By letter dated October 29,1996, the BWRVIP modified the recommendations in BWRVIP-05 by increasing the recommended examination coverage for longitudinal welds to 100 percent.

On May 12,1997, the NRC staff and members of the BWRVIP met with the Commission to discuss the NRC staff's review of the BWRVIP-05 report. In accordance with guidance provided by the Commission in Staff Requirements Memorandum M970512B, dated j

May 30,1997, the staff initiated a broader, risk-informed review of the BWRVIP-05 proposal, and issued a final safety evaluation (FSE) related to the review of BWRVIP-05 on July 30, 1998. In the FSE the staff generically approved the reduction in hv,pection of circumferential reactor vessel welds. The staff provided the Commission with its methods and acceptance criteria for considering both partial and permanent requests for relief of the augmented reactor vessel examinations required by 10 CFR 50.55a(g)(6)(ii)(A) in SECY-98-219.

In Gencric Letter (GL) 98-05, dated November 10,1998, the staff informed licensees owning i

BWR designs that the NRC staff had completed its review of BWRVIP-05. In the GL, the staff also informed BWR licensees that they could request periodic or permanent (i.e., for the remaining term of operation under the existing,initiallicense) relief from complying with the augmented inservice inspection requirements in 10 CFR 50.55a(g) for the volumetric examination of circumferential RPV welds if the following two criteria were met:

1. If, at the expiration of the license for the plant, the circumferential welds in the vessel are shown to satisfy the limiting cc,nditional failure probability for circumferential welds in the staff's July 30,1998, FSE; and

2. If it is demonstrated that the licensee for a facility has implemented operator training and j

has established procedures that limit the frequency of cold overpressure events to the degree specified in the staff's July 30,1998, FSE.

In the GL, the staff also informed BWR licensees that they would still need to perform their required inspections of " essentially 100 percent" of all longitudina! RPV welds.

l l

1 As supplemented by letters dated June 24,1996; October 29,1996; May 16,1997; June 4,1997; June 13, 1997; and December 18,1997.

F -

l l 3.0 EVALUATION OF THE PROPOSED ALTERNATIVES TO THE AUGMENTED i

INSPECTION REQUIREMENTS FOR LONGITUDINAL AND CIRCUMFERENTIAL WELDS IN THE HATCH-1 RPV 3.1 Reouest for Relief SNC is requesting relief from performing augmented volumetric examinations of the longitudinal l

and circumferential shell welds in the Hatch-1 RPV.

3.2 Acolicable Reauirements Pursuant to the requirements of 10 CFR 50.55a(g)(4), ASME Code Class 1,2 anh 3 components must meet the requirements, except the design and access provisions and the preservice examination requirements, set forth in the Section XI, " Rules for Inservice Inspection i

of Nuclear Power Plant Components," to the ASME Code (Section XI) to the extent practical within the limitations of design, geometry, and materials of construction of the components.2 The regulations require that all inservice examinations and system pressure tests conducted during the first 10-year interval and subsequent intervals on ASME Code Class 1,2. and 3 components must comply with the requirements in the latest edition and addenda of Section XI incorporated by reference in 10 CFR 50.55a(b) on the date 12 months prior to the start of the 10-year interval. 71.c applicable edition of Section XI for Hatch-1 during the current 10-year ISI intervalis the 1960 % tion, as modified through the Winter 1981 Addenda of the edition.

Section 10 CFR 50.55a(g)(6)(ii)(A)(2) requires that all licensees to " augment their reactor i

vessel examination by implementing once, as part of the ISI intervalin effect on September 8, 1992, the examination requirements for reactor vessel shell welds specified in item B1.10, Examination Category B-A, ' Pressure Retaining Welds in Reactor Vessel,' Table IWB-2500-1 to Section XI...." The section requires licensees to implement augmented examinations of essentially 100 percent of the RPV shell welds. ASME Code item B1.10 covers requirements for examinations of RPV circumferential shell welds (Examination item B1.11) and longitudinal shell welds (Examination item B1.12). Section 50.55a(g)(6)(ii)(A)(2) defines " essentially 100%"

examination as covering 90 percent or more of the examination volume of each weld. The schedule for implementation of the augmented inspection is dependent upon the number of months remaining in the 10-year ISI interval that was in effect on September 8,1992.

Paragraph IWA-2232 to Article IWA-2000 of Section XI requires that the requirements of Mandatory Appendix 1, " Ultrasonic Examinations," be applied when ultrasonic examinations are being used as the method for performing volumetric examinations of ASME Code Class 1,2, and 3 components. Mandatory Appendix I to Section XI requires that " ultrasonic examinations of vessel welds greater than 2 inches in thickness shall be conducted in accordance with the provisions of Article 4 of Section V, ' Nondestructive Examination,'" to the ASME Code. These requirements are applicable to the ultrasonic examinations proposed for the Hatch-1 RPV welds, because the sections used to fabricate the Hatch-1 RPV are all thicker than 2 inches in cross section.

2 Except for design and access provisions and preservice inspection requirements.

l L

m j

E

' i 3.3

. Basis for Relief I

Section 10 CFR 50.55a(c)(6)(ii)(A)5 allows licensees who are unable to completely satisfy the augmented RPV shell weld examination requirements to submit information to the Commission to support such a determination and to propose alternatives to the examination requirements j

that would provide an acceptable level of quality and safety in lieu of complying with the requirements.

Section 50.55a(a)(3)(i) indicates that altematives to the requirement in 10 CFR 50.55a(g) are justified when the proposed alternative provides an acceptable level of quality and safety in lieu l

of complying with the requirements.

3.4 Proposed Alternatives I

1 Pursuant to 10 CFR 50.55a(g)(6)(ii)(A)(S), SNC proposes the following three alternative programs to the augmented examination requirements of 10 CFR 50.55a(g)(6)(ii)(A)(2), and the inservice inspection requirements of Section XI:

. SNC's first alternative proposes use of a probabilistic failure (risk) analysis as the basis for justifying a permanent deferral of the required augmented volumetric examinations of the circumferential shell welds in the Hatch-1 RPV.

2. SNC's second alternative proposes use of the GERIS-2000 ultrasonic examination system as the method for performing the required a'ugmented volumetric examinations of the longitudinal shell welds in the Hatch 1 RPV. The techniques and procedures for i

calibrating the GERIS-2000 system do not entirely comply with requirements of j

Mandatory Appendix i to Section XI and with the requirements of Article 4 of Section V,

" Nondestructive Examination," to the ASME Code.

i

3. SNC's third alternative proposes that, due to the obstructions of components internal and external to the R'V, it is impractical to achieve 90 percent inspection coverage of the longitudinal shell welds, and requests that the NRC accept the proposal to achieve less than 90 percent inspection coverage when the ultrasonic testing (UT) e>:arninations of the longitudinal shell welds are conducted at Hatch-1.

3.5 Evaluation SNC will be performing an examination of the reactor vessel longitudinal shell welds to the maximum extent practical from the inner diameter, within the constrants of vessel internal restrictior, s. It should be noted that the current examination plan is designed to provide longitudinal weld coverage; however, incidental coverage at the longitudinal shell weld-to-circumferential shell weld junctures will result in an estimated inspection coverage of 2-3 percent of the intersecting circumferential welds.

L

. 3.5.1 Evaluation of SNC's Alternative Proaram for Volumetric Examinations of the Circumferential Shell Welds in the Hatch-1 RPV 3.5.1 -1 Meetina the Probabilistic Fracture Mechanics Analysis Criteria for Circumferential Weld Evaluations Technical Report BWRVIP-05 provides the technical basis for permanently deferring the avgmented inspections of circumferential welds in BWR RPVs. In the report, the BWRVIP concluded that the probabilities of failure for BWR RPV circumferential shell welds are orders of magnitude lower than that of the longitudinal shell welds. To assess the BWRVIP safety assessment, the NRC conducted an inder endent risk-informed, probabilistic fracture mechanics assessment (PFMA) of the an ilysis presented in the BWRVIP-05 document.' In its assessment, the staff conservatively calet 'ated the probability that an RPV shell weld would catastrophically fail during the licensed operating term for a BWR nuclear plant. In the assessment, the NRC used the FAVOR Code to perform the PFMA. The staff calculated the final failure probability for an RPV shell weld as the product of frequency for the critical (limiting) transient event and the conditional failure probability for the weld using the limiting conditions from that event.

For the analysis, the staff identified that a cold overpressure event in a foreign reactor was the limiting pressure and temperature event for BWR RPVs. By its calculations, the staff estimated that the probability for the occurrence of the limiting overpressurization transient was 1 x 10' per reactor year. The staff then determined the conditional prc'oabilities of failure for longitudinal and circumferential welds in Chicago Bridge and Iron Works (CB&l), ABB-Combustion Engineering (CE), and Babcock and Wilcox (B&W) fabricated vessels using the pressures and temperatures from the limiting event. The conditional failure probabilities for vessel welds were calculated as a function of a nil ductility reference temperature (Mean RTuor value) for the welds.*

Tablo 2.6-4 of the staff's PFMA identifies the conditional failure probabilities for the bounding reference cases for longitudinal and circumferential welds in CB&l, CE, and B&W fabricated vessels. The materials and neutron radiation parameters used by the staff in calculating the conditional probability failures for the referonce cases were also identified in Table 2.6-4 of the staff's PFMA. According to Table 2.6-4, B&W fabricated vessels were determined to have the highest conditional probability of failure for circumferentially oriented flaws (8.17 x 10 per 4

l reactor year ). For circumferentially oriented flaws in CE fabricated vessels, the conditional 4

probabilities of failure were slightly lower (6.34 x 10 per reactor year using data in the CE l

Owners Group (CEOG) Task Report and 2.81 x 10 per year using data from BWRVIP-05).

4 The corresponding Mean RTuor value used to calculate the conditional probability of failure for 3 The staff's PFMA of BWRVIP-05 is documented in a letter dated June 28,1998, to Mr. Carl Terry, Chairman of the BWRVIP.

4 The kg parameters in the analysis for calculating the Mean RTno7 values are the initial RTwor value for the weld, the end-of-license mean neutron fluence, the mean chemistry (percent copper and nickel) of the welds. The methods for calculating the Mean RTcvalues aie consistent with the methods in Regulatory Guide 1.99, Revision 2.

. the CEOG reference case was 98.1*F. Using this data, the staff calculated the best-estimate failure probability for CE fabricated circumferentially welds to be 6.34 x 10* per reactor year.5 The staff considers that when the adjusted reference temperature (RTuor) value for ari RPV shell weld is less than the Mean RTuor value for its correspond limiting weld reference case study (as specified in Table 2.6-4 of the PFMA), the shell weld has less embrittlement than the corresponding weld in the case study, and therefore has a conditional probability of failure less than or equal to that calculated for the reference case study. In reviewing SNC's assessment, the staff confirmed that the chemistry factors, 4RTuor values, margin terms, and RTwor values were calculated in accordance with the guidelines of Regulatory Guide 1.99, Revision 2. and that the copper and nickel contents listed for the circumferential welds were consistent with the values listed in the CEOG Task Report CE-NPSD 1039, Revision 2. Table 3.5.1-1 (attached to this SE) illustrates that the RTuor values for the circumferential welds in the Hatch-1 RPV are less than the Mean RTuor value (98.1"F) for circumferential welds from the CEOG reference case. Since the RTuor values for the circumferential welds are bounded by the corresponding Mean RTuor values for the CEOG reference case, the staff concludes that SNC has provided sufficient assurar$ce that the degree of projected embrittlement of the circumferential welds in the Hatch-1 RPV are also bounded by that assessed for the CEOG reference case. The staff therefore concludes that the probability of failure for the circumferential welds in the Hatch-1 RPV should be less than that calculated by the staff (6.34 x 10 per reactor year) for the 4

corresponding CEOG reference case. Based on this analysis, the staff concludes that the assessment of the circumferential welds in the Hatch-1 RPV is consistent with the staff's analysis in SECY-98-219.

3.5.1 -2 Meetina the Operational Considerations Criteria for Circumferential Weld Evaluations During the review of the BWRVIP-05 report,"BWR Reactor Pressure Vessel Shell Weld Inspection Recommendations," the staff identified nondesign-basis events, which should have been considered in the BWRVIP-05 report in particular, the potential for and consequences of cold overpressure transients should be considered. The licensee assessed the systems that could lead to a cold overpressurization of the Hatch-1 RPV. These included the high pressure core injection (HPCI), reactor core isolation cooling (RCIC), normal feedwater supply, core spray (CS), residual heat removal (RHR), control rod drive (CRD) and reactor water cleanup systems (RWCU).

The HPCI and RCIC pumps are steam driven and do not function dt. ag cold shutdown. The reactor feedwater pumps, which supply normal feedwater, are the high pressure makeup system during normal operations. The reactor feedwater pumps are also steam driven and therefore, cannot be operated during cold shutduwn. Although not addressed in the licensee's submittal, the staff notes that there are no automatic starts associated with the standby liquid control system (SLCS). Operator initiation of the SLCS should not occur during shutdown.

However, the SLCS injection rate of approximately 40 gallons per minute (gpm) would allow operators sufficient time to control reactor pressure if manual initiation occurred during shutdown.

5 Th;s value is the product of the conditional probability of failure for the CEOG reference case (6.34E-5 per reactor year) and the estimated frequency for the limiting event (1 E-3 per reactor year).

s-

. The CS and RHR systems are low pressure emergency core cooling systems (ECCS) with shutoff heads below 400 pounds per square inch gauge (psig). If either cne of these systems were manually or inadvertently initiated during cold shutdown, the resulting reactor pressure and temperature would be below the pressure-temperature limits. The CRD and RWCU systems use a feed and bleed process to control RPV level and pressure during normal cold shutdown conditions. Plant procedures are in place to respond to any unexpected or unexplained rise in reactor water level, which could result from spurious actuation of an injection system. The procedural actions include preventing condensate pump injection, securing ECCS system injection, tripping CRD pumps, terminating all other injection sources, and : $wering RPV level via the RWCU system.

In all cases, the operators are trained in methods of controlling water level within specified limits in addition to responding to abnormal water level conditions during shutdown. The licensee also stated that procedurai controls for reactor temperature, level, and pressure are an integral part of operator training. Plant-specific procedures have been established to provide guidance to the operators regarding compliance with the Technical Specification pressure-temperature limits. On the basis of the pressure limits of the operating systems, operator training, and established plant-specific procedures, the licensee determined that a nondesign-basis cold overpressure transient is unlikely to occur.

The staff agrees with the licensee that the information provided on the Hatch-1 high pressure injection systems, operator training, and plant specific procedures provides a sufficient basis to support approval of the alternative examination request for circumferential shell welos in the Hatch-1 RPV. The staff concludes that a nondesign-basis cold overpressure transient is unlikely to occur at Hatch-1.

3.5.2 Evaluation of SNC's Alternative Prcorams for Volumetric Examinations of the Lonaitudinal Shell Welds in the Hatch-1 RPV SNC proposes to use the General Electric (GE) Company's GERIS-2000 system as the system for performing remote controlled, automated UT examinatlons of the longitudinal welds in the Hatch-1 RPV. GERIS-2000 uses an alternative procedure for qualifying and performing UT examinations rehtive to that required by Section XI and recommended by Re0ulatory Gu~ide (RG) 1.150. The examination procedure is dependent on the echo-dynamic motion and tip diffraction characteristics of the flaw instead of the amplitude characteristics required by the Code. The areas where the procedure deviates from Code and the RG are explained in of SNC's submittal of January 19,1999. An examination performed with the procedure will result h the examination vclume being interrogated by the same straight and angle beam search units as an ASME Section V, Article 4 procedure. Any areas of limited access would be common to the Article 4 procedure. An additional examination by the 70* refracted longitudinal search units, not required by Article 4, is also performed.

GE demonstrated the qualification of the procedure at the Performance Demonstration Initiative (PDI). The qualification process was performed in accordance with PDI Section No. 61-02, which is consistent in accordance with the Appendix Vill requirements of the 1992 Edition of Section XI, inclusive of the 1993 Addenda for the edition. Appendix Vill was developed to ensure the effectiveness of UT examinaticns within the nuclear industry by means of a rigorous, l

item-specific, performance demonstration. The performance demonstration was conducted on l

an RPV mockup containing flaws of various sizes and locations. The demonstration

l established the capability of equipment, procedures, and personnel to find flaws that could be detrimental to the integrity of the RPV. Although Appendix Vill is not currently required, the qualification of equipment, procedures, and personnel to Appendix Vill criteria demonstrates examination and evaluation techniques that equaled or surpassed both the requirements specified in Paragraph IWA-2232, " Ultrasonic Examination," to Section XI, and the i

recommendations found in RG 1.150. The staff therefore concludes that the GERIS-2000 ultrasonic system is acceptable to use as the equipment for conducting the volumetric examinations of the longitudinal welds in the Hatch-1 RPV.

SNC has completed volumetric examinations of longitudinal welds C-1-A, C-1-B, C-1-C, C-2-A, C-2-B, and C-2-C from the outer surface (OD) of the Hatch-1 RPV during the Cycle 17 refueling outage (RFO) for the plant. Inspection coverages from the volumetric examinations of the C-1-A, C-1-B, and C-1-C longitudinal welds were all in excess of 90 percent and therefore in compliance with the coverage criteria stated in the augmented inspection requirements of 10 CFR 50.55a(g)(6)(ii)(A). Weld coverages of the examination on welds C-2-A, C-2-B, and C-2-C ranged from 75.0 percent - 77.7 percent of the weld lengths. SNC has scheduled adManal volumetric examinations of longitudinal welds C-2-A, C-2-B, C-2-C, C-3-A, C-3-B, C-3-C, C-4-A, and C-4-C for the Cycle 18 RFO. SNC estimates that the combined coverage of volumetric examinations (from both the OD and inner surface (ID) examinations of the vessel) of welds C-2-B and C-2-C will range from 98.0 percent - 100.0 percent; SNC therefore estimates that the coverage for welds C-2-B and C-2-C will satisfy the 90 percent coverage criteria in the augmented inspection requirements of the rule, and that no relief will need to be submitted relative the C-2-8 and C-2-C welds. Table 3.5.2-1 (attached to this evaluation) gives a summary of the examination coverages that were achieved duiing Cycle 17 RFO and also the estimates of the examination coverages that will be achieved during the Cycle 18 RFO.

The Cycle 18 RFO examinations are scheduled to be initiated from the inside surface of the Hatch-1 RPV. SNC has stated that the combined volumetric examination estimates for longitudinal welds C-2-A, C-3-A, C-3-B, C-3-C, C-4-A, C-4-B, and C-4-C will result in less than the 90 percent coverage required by the augmented inspection rule. SNC therefora seeks approval (pursuant to 10 CFR 50.55a(g)(6)(ii)(A):5)) of the reduced examination coverage estimates for the welds. The following bases summarize the hardships for these welds:

C-2-A weld is completely blocked from ID access by the steam dryer rod at the 0* azimuthal location, and partially from the OD by the insulation support ring.

Access to longitudinal welds in the C-3 and C-4 shell courses is completely obstructed from the OD direction by the bioshield for the vessel. In addition, dose rates at contact are high (1.5-1.7 rem /hr range) and create additional hardship. The configuration of the bioshield in relation to the vesselis such that only manual examination can be performed from the OD, and lead shielding cannot be effectively used to reduce the exposure level to examination personnel.

C-3-A, C-3-B, C-3-C, U-4-A, and C-4-C welds are partially restricted from the ID by small intenials such as jet pump brackets, shroud support plate gussets, and/or by instrumentation or fluid system nozzles. The C-4-8 is completely obstructed from the ID by the core shroud restraint rod at the 135* azimuthal location and by the shroud support plate.

Repositioning of the shroud restraint rod to allow access to the GERIS 2000 equipment would reqcire special equipment; and while the rastraint design provided a contingency for

.g.

t removal, repositioning or removal of the restraint rod was never anticipated (or recommended by GE) after finalinstallation of the rod.

Since the C-2-A longitudinal weld is entirely obstructed from the ID, SNC is requesting that the staff accept the actual reduced coverage of the weld (77.2 percent) that was schieved from the OD inspection the 90 percent coverage was conducted during the Cycle 17 RFO. The staff concludes that the coverage is not significantly below the 90 percent coverage required by the rule, and the achieved coverage is sufficient to provide a reasonable estimate of the degree of cracking in the C-2-A longitudinal weld. The staff therefore concludes that SNC's requent is acceptable based on the hardships created by obstructions to weld's access.

SNC is also requesting that the staff accept the reduced coverage estimates for the ultrasonic examinations that are scheduled for the longitudinal welds in the C-3 and C-4 RPV shells; SNC's request for relief is based on the hardships created by the obstructions to the welds' access and by the dose rate levels for the shells. The longitudina! welds ln the C-3 and C-4 RPV shells are of particular importance because portions of the welds are located in the beltline portion of the vessel where neutron flux leve!s may be sufficient to enhance the degree of embrittlement in the welds over time. SNC projects that total examination coverages for the C-3-A, C-3-B, C-3-C, C-4-A, and C-4-C weids (which equals the coverage estimated for the ID examinations) will be in the 52 percent to 77 percent range, with an average projected coverage of 71.4 percent. The projected examination coverages include a 14 percent uncertainty in the estimates. The staff concludes that th3 average projected coverage is not significantly below that required by the rule, and the projected coverages for the C-3-A, C-3-B, C-3-C, C-4-A, and C-4-C welds will be sufficient to provide reasonable estimate of the degree of cracking, if any, in the welds. The staff therafore concludes SNC's request is acceptable based on the hardships created by obstructions to the welds' access.

SNC is also requesting that the staff accept the reduced coverage (0 percent) for the C-4-B longitudinal weld. SNC has emphasized that the current state-of-the-art for ultrasonic equipment makes it impractical to access weld C-4-B for examinations from the ID. SNC has stated that the C-4-P is completely obstructed to access from the ID by one of the core shroud tie rods (restraint rods), which were installed internally as a modification to the cere shroud ciructure, and that the weld cannot be volumetrically inspected without removing the obstructing tie rod from the modified shroud design. Although GE, the designer of the tie rod design, did provide a contingency for the rud's removal, SNC has stated that it was never the intent of GE to remove and reinstall the tie rods on a regular basis. SNC has also indicated that the weld is completely obstructed to access from the OD by the RPV bioshield.

By letter dated February 5,1999, at the request of the staff, SNC submitted a bounding analysis to show that the adjusted reference temperatures for the longitudinal welds in the C-4 RPV shell were bounded by the adjusted reference temperatures calculated for longitudinal welds in the C-3 shell. The staff uses adjusted reference temperatures as one of the many

. mmeters for estimating the degree of embrittlement in ferritic RPV materials. The staff's methodology for calculating adjusted reference temperatures is provided in NRC RG 1.99, Revision 2 (May 1988). Table 3.5.2 2 (attached to this SE) provides the 'results of the bounding

. adjusted reference temperature analysis. Since the adjusted reference temperatures for the longitudinal welds in the C-4 shell (including longitudinal weld C-4-B) are bounded by the adjusted reference temperatures for the longitudinal welds in the C-3 RPV shell, the staff concludes that the degree of embrittlement for longitudinal weld C-4-B would be less

. than that of the longitudinal welds in the C-3 shell and equal to that of the other longitudinal welds in the C-4 RPV shell. Even if no inspection of the C-4-B weld is performed, SNC estimates that the projected examinations of remaining longitudinal welds in the C-3 and C-4 shells will cover an average weld volume of 65 percent.

The staff concurs that because the tie rods were installed to provide an alternative load bearing capability for the shroud in lieu of the circumferential shroud welds, the tie rods should only be dismantled from the shroud design as a contingency or if a repair of the tie rod design is necessary. The staff also concurs that it will be difficult to inspect the C-4-B weld using the.

current state-of-the-art equipment for ultrasonic inspections without dismantling the obstructing i

tie rod from the core shroud design. Since the proposed examinations will result in an average

]

of 65 percent of the volume of the longitudinal bettine welds, and since the C-4-B longitudinal weld has less embrittlement than the longitudinal welds in the C-3 shell and is obstructed from access to the GERIS-2000 ultrasonic technology, the staff concludes that the C-4-B weld need not be inspected at this time. However, pursuant to 10 CFR 50.55a(g)(6)(ii)(A), the staff w;ll still require a volumetric inspection of C-4-B weld, if SNC determines it is necessary to remove the obstructing tie rod from the shroud design or if ultrasonic inspection technolo;;y is improved to the point where a qualified inspection of the C-4-B weld could be accomplished without necessitating removal of the tie rod.

j 3.5.3 Assessment of Neutron Fluence Estimates Used in Adiusted Reference Temperature Assessment The staff has determined that the fluence values provided by SNC for the purpose of this relief request were consistent with those previously approved for the Hatch-1 extended power uprate.

The fluence methodology used by GE for determining the Hatch-1 fluences was examined by

)

the staff during the extended power uprate review and during the course of this review. While the GE methodology may not conform to the guidance proposed by the NRC staff in Draft Regulatory Guide DG-1053, " Calculational and Dosimetry Methods for Determining Pressure Vessel Neutron Fluence," the staff has found that the GE methodology provides an acceptable assessment of the embrittlement of the Hatch-1 RPV as applied to the establishment of thn Hatch-1 pressure-temperature limits and for the purpose of this review.

4.0 CONCLUSION

S

]

4.1 Conclusions Recardina SNC's Alternative Proarams Prosed for the Circumferential Welds in the Hatch-1 RPV The staff has determined that SNC has proposed a reasonable alternative to performing the augmented inspections of the circumferential welds in the Hatch-1 RPV. The staff has also determined that, pursuant to 10 CFR 50.55a(g)(6)(ii)(A)(5), SNC's PFMA, when taken in

]

conjunction with SNC's operator training and procedures for limiting the frequency of a cold overpressure transient, provic;es an acceptable alternative for permaner.tly deferring the augrr,ented inspections of the circumferential welds required by 10 CFR 50.55a(g)(6)(ii)(A).

The staff has also determinF pursuant to 10 CFR 50.55a(a)(3)(i), that the altemative program provides an acceptable level t ? quality and safety relative to assuring the structural integrity of the circumfervntial welds. and, therefore, the PFMA, training, and procedures may be used as a basis for pennanently deferring the augmented inspections required by 10 CFR 50.55a(g)(6)(ii)(A). The staff therefore concludes, pursuant to 10 CFR 50.55a(g)(6)(ii)(A)(5L 1

l1 that the augmented inspections of the circumferential shell welds in the Hatch-1 RPV may be permanently deferred for the remaining term of operation under the existing, initial operating license.

4.2 Conclusions Reaardina SNC's Alternative Proarams Prooosed for the Lonaitudinal Welds in the Hatch-1 RPV l

The etaff has determined, pursuant to 10 CFR 50.55a(g)(6)(ii)(A)(5), that SNC has proposed

~ reasonable alternatives to complying with the augmented inspection requirements for inspections of the longitudinal welds in the Hatch-1 RPV. The etaff has determined that the following alternatives provide an acceptable ' level of quality and safety and are acceptable

' pursuant to 10 CFR 50.55a(a)(3)(i):

'a The reduced examination coverage (77.2 percent), which resulted from the previous volumetric examination from the OD of the C-2-A :ongitudinal weld in 1997, was sufficient

' to determine the extent of cracking in the weld.

)

The proposed inspections of the C-3-A, C-3-B, C-3-C, C-4-A, and C-4-C longitudinal welds, when coupled with the bounding adjusted reference temperature analysis and the hardship argument (obstructed access argument) for the inspection of the C-4-B weld, provide a reasonable basis for estimating the degree on cracking in the C-4-B weld and for deferring the inspections of the C-4-8 weld.-

The reduced estimated coverages for the inspections of the C " A, C-3-B, C-3-C, C-4-A, and C-4-C longitudinal welds will be sufficient to determine the udent of cracking in the welds and to provide assurance of the structural integrity of the welds during the life of the plant.

Use of the GERIS-2000 ultrasonic system is acceptable for conducting the volumetric examinations of the longitudinal welds in the Hatch-1 RPV.

Therefore, in relation to the volumetric examinations proposed for the longitudinal welds and

_ pursuant to 10 CFR 50.55a(g)(6)(ii)(A)(5), the staff concludes that:

1.

The previous augmented incpection of the C-2-A weld provides sufficient coverage and that no further augmented examination of the C-2-A weld is necessary; 2.

SNC may defer the ID examinations of the C-4-B weld until such time that SNC determines that it is necessary to remove the obstructing tie rod from the snroud design or if ultrasonic inspection technology is improved to the point where a qualified inspection of the C-4-B weld could be accomplished without necessitating removal of the tie rod.

3. ~ SNC's estimated coverages (wl::ch include a

• 4 percent uncertainty in the estimates) for the volumetric examinations of the C-3-A, C-3-B, C-3-C, C-4-A, and C-4-C longitudinal welds will be sufficient to determine the degree of cracking in the welds and provide a sufficient basis for complying with the augmented inspection requirements of 10 CFR

~

- 50.55a(g)(6)(ii)(A).

I

s.

l.

12-l

4. ; SNC may use the GERIS-2000 ultrasonic examination system as the technology for performing the volumetric examinations of the C-3-A, C-3-B, C-3-C, C-4-A, and C-4-C i

L

' longitudinal welds.

Attachment:

Tables 3.5.1-1, 3.5.2-1, and 3.5.2-2 '

1 Principal Contributors: J. Medoff.

K. Kavanagh j

l Date: March 11, 1999

]

l

)

i 4

.e

i Table 3.5.1-1 Comparison of SNC Probabilistic Fracture Mechanics Assessment (PFMA) to the Staff's PFMA for the Limiting CEOG Case Study for Circumferential Welds s

COMPARATIVE INPUT NRC's LIMITING PARAMETERS AT 32 EFPY PFMA l

FOR THE CIRCUMFERENTIAL ANALYSIS PARAMETER WELDS IN THE Hatch-1 l

DESCRIPTION REACTOR VESSEL

}

Weld Heat Weld Heat CE Circum.

No.90099 No.33A277 Welds "

I Initial RTuor, 'F

-10

-32 0.0 2

Fluence, n/cm 1.32x10'8 1.32x10'8 2.0 x 10's Fluence Factor 0.475 0.475 0.569 Cu Content (Wt. %)

0.197' O.258 '

O.0.183 Ni Content (Wt. %)

0.060 '

O.165' O.704 Chemistry Factor 91 126 172.2 aRTuor,

43.2 59.8 98.1 Margin Term 43.2 56.0 0.0 RTuoy value 76 65.8 98.1 - Mean RTyn, value 1.

Latest copper and nickel contents for the circumferential welds in the beltline of the Hatch-1 RPV provided in

. Combustion Engineering Topical Report CE-NPSD 1039. Revision 2, and confirmed in SNC's submittal of July 28,1998 (SNC response to the staff's requect for additional information regarding GL 92-01, Revision 1, Supplement 1.)

2.

Combustion Engineering Owners Group (Task 1039) Reference Case for Circumferontial Welds.

l1 l:

Table 3.5.2-1 Volumetric Examination Coverage Table for Longitudinal Welds in the Hatch-1 Reactor Pressure Vessel ISI Welti

% Coverage

% Coverage Estimated % of Total Identification

' of ID Surface of OD Surface Beltline Covered

% Coverage C-1 A NA' 99.6 NA 99.6 2

C-1 -B NA1 90.7 "

NA 90.7 C-1 -C NA' 93.3 :

NA 93.3 C-2-A 0.0 77.7 8 NA 77.7 C-2-B 73.2 '

75.0" NA 98.0 82' C-2-C 73.2 '

77.2 NA 100.0 3.4 8

8 2

84 8

C-3-A 52.6 0

50 52.6 4 C-3-B 80.2 '

O 70 80.2 a.4 6

C-3-C 80.2 'A 0

70 80.2 '

8 C-4-A 72.1 '

• 0 100 72.1 'A C-4-B 0

0 0

0 l

C-4-C 72.1 aa 0

100 72.1 '

Footnotes:(NA = not applicable) 1.

Emminations from the ID not required since the coverage since the coverage from the OD or combination of coverage from inspections from the ID and OD will cover more than 90% of the weld length.

2.

Volumetric examinations were performed from the RPV exterior surface during the fall 1997 refueling outage.

Percentage reported is the actual coverage achieved.

3.

Estimated coverage based on access evaluation performed by the GE Electric Company for the volumetric i

examinations to be conducted during the Spring 1999 refueling outage.

4.

Estimated coverage to include uncertainties of

• 4% to account for uncertainties in the estimates on the amount of obstructions caused by components or lack of access capability.

d i

7 F

i Table 3.5.2-2 Bounding Adjusted Reference Temperature (RTuor) Analysis for Longitudinal Welds in the Hatch-1 RPV i

COMPARATIVE BOUNDING ANALYSIS INPUT PARAMETERS AT 32 EFPY FOR THE LONGITUDINAL BELTLINE PARAMETER WELDS IN THE HATCH-1 DESCRIPTION REACTOR VESSEL Weld Heat Weld Heat Weld Heat No.13253 No.1P2809 No.1P2815

- Weld Nos.

- Weld Nos.

- Weld Nos.

C-4-A,B&C C-3-A&C C-3-A,B,&C Initial RTwor, 'F

-50.0

-50.0

-50.0 Fluence, n/cm' 1.32x10'8 1.94x10 e 1.94x10 s i

Fluence Factor 0.475 0.562 0.562 0.221' O.220' O.316 '

Cu Content (Wt. %)

i Ni Content (Wt. %)

0.732' O.735' O.724 '

Chemistry Factor 189.0 189.3 218.6 aRTuor 89.8 106.3 123.1 Margin Term 56.0 56.0 56.0 RTuny value 95.8 112.4 128.9 1.

Latest copper and nickel contents for the circumferential welds in the beltline of the Hatch-1 RPV provided in Combustion Engineering Topical Report CE-NPSD 1039, Revision 2, and confirmed in SNC's submittal of July 28,1998 (SNC response to the staff's request for additional information regarding GL 92-01, Revision 1, Supplement 1.)

.4