ML20137H581

Forwards Response to Request for Addl Info Re First 10-Yr ISI

ML20137H581
Person / Time
Site:	Braidwood
Issue date:	03/25/1997
From:	Tulon T COMMONWEALTH EDISON CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
TAC-M92120, TAC-M92121, TAC-M97134, TAC-M97135, NUDOCS 9704020288
Download: ML20137H581 (6)

Text

-

Commonwealth filium Company Braidwood Generating Station

• Route of , llox 81 Bracesille. IL 60407-9619 Tel H t 5 45&2801 March 25,1997 United States Nuclear Regulatory Commission Attention: Document Control Desk Washington, D.C. 20555-0001

Subject:

Braidwood Nuc! car Power Station, Units 1 and 2, Response to Request for Additional Infonnation Regarding First 10-year ISI, Facility Operating Licenses NPF-72 and NPF-77, NRC Docket Nos. 50-456 and 50-457

References:

1. Robert A. Capra (USNRC) letter to D. L. Farrar (Comed), Safety Evaluation of the Inservice Inspection Program Relief Requests Nos. NR-20, NR-21, NR-22 for Braidwood Station Units I and 2 (TAC Nos. M 92120, M92121), dated September 1, 1995.

~

2. H. G. Stanley (Comed) letter to Document Control Desk (USNRC), Resision 4 of the Insenice Inspection Plan for i Braidwood Station Units I and 2 dated October 8,1996.

3. George F. Dick (USRNRC) letter to Irene M. Johnson (Comed),

Request for Additional Information Regarding First 10-Year

- Insenice Inpsection Program - Braidwood Station (TAC Nos. M97134 and M97135), dated February 11,1997.

4. T. J. Tulon (Comed) letter to Document Control Desk (USNRC), Response to Request for Additional Information Regarding First 10-Year ISI Braidwood Station Units 1 and 2, dated February 25,1997.

5. George F. Dick (USRNRC) letter to Irene M. Johnson (Comed),

Request for Additional h1 formation Regarding First 10-Year U~,p n

p]- Insenice Inpsection Program - Braidwood Station (TAC Nos. M97134 and M97135), dated March 17,1997, Commonwealth Edison Company's (Comed's) Braidwood Nuclear Power Station, Units 1 and 2 (Braidwood),

performs inservice inspections (ISI)in accordance with Section XI of the 1983 Edition of the American Society of Mechanical Engineers Boiler and Pressure Vessel Code through the Summer 1983 Addenda (ASME Code), as required by Title 10, Code of Federal Regulations, Part 50, Section 55a, Paragraph f, Subparagraph 3110 CFR 9704020208 PDR 970325 -

G ADOCK 05000456 },

pyg gipulipp,nspp5 A t!nicom Company

l '

1 -

i 1 * , ,

50.55a(f)(3)], except where alternatives have been authorized or relief has been requested and granted by the

United States Nuclear Regulatory Commission (USNRC).

The information requesting in Reference 5 is provided in Attachment 1.

4 j Comed respectfully requests USNRC continued expedited review and approval of Relief Request NR 29, such that .

]

these requirements can be implemented during the upcoming Unit I and Unit 2,1997 outages. These outages represent the last outages in the first ISI interval for Braidwood Station. Please address any conunents or questions regarding this matter to Ms. Patricia A. Boyle, at (815)458-2801 extension 2519.

. Sincerely,

]

T' hy J. Tulon N

, tion Manager j raidwood Nuclear Generating Station .

l

, Attachment l

} TJT/fb/97nrcolo. doc i cc: A. B. Beach, Regional Administrator - RIII

' G. F. Dick, Jr., Braidwood Project Manager - NRR C. Phillips, Senior Resident Inspector - Braidwood

. Office of Nuclear Facility Safety - IDNS I Michac! T. Anderson, INEL Research Center i

i i

J i

i i

s

ATTACllMENT 1 NRC RAI Item #2.1:

It is the NRC staff's understanding that the PDI test specimens used for vessel performance demonstration qualification are designed for examination from two sides.

Use of existing PDI test specimens for qualification of procedures and personnel when performing one direction scans perpendicular and parallel to the weld may not provide adequate assurance of flaw detection. The staff believes that certain adverse conditions that would challenge an examiner's skills when examinations are performed from one side only may not be represented by the PDI test specimens. For instance, certain actual field conditions such as off-normal flaw orientations at the weld root, incomplete fusion, flaw geometrics that may mask other flaws, and rough cladded surfaces that cause beam redirection may result in flaws going undetected when scanning from one side only.

Provide a discussion that addresses the staffs concerns. Include a basis for concluding that one-sided examinations provide equivalent flaw detection as the Code-required two-sided examinations.

Response

The PDI specimens were designed to follow all applicable Appendix Vill code requirements and to incorporate features representative of field conditions that are important to examination. PDI conducted a comprehensive survey of plants to determine weld geometries, material thicknesses, surface conditions and other relevant geometric or material conditions. This survey information was a primary input into specimen design.

The following discussion addresses the particular concerns of the RAl.

Off-Normal Flaw Orientations and Lack of Fusion:

In the RAI, the NRC Staff expresses concern that off-normal flaw orientations would make flaw detection difficult when examination is performed from only one side or direction. While it is tme that severely tilted planar flaws would be diflicult to detect from one direction, this is not a realistic condition. Off-normal flaws are expected to be i introduced during the manufacturing process. Inservice flaws are expected to initiate and propagate in the direction normal to the vessel surface.

Underclad cracks that may be present in the Reactor Pressure Vessel (RPV) are known to be oriented perpendicular to the vessel surface (Reference 1). Manufacturing lack of fusion flaws that may have formed on the weld preparation faces are likewise expected to be primarily perpendicular to the vessel surface because of the steep (5*-

7*) of the RPV weld preparations. Flaws at the weld root interior to the plate are conunon (Reference 2) and are primarily volumetric slag or small lack of fusion defects. These flaws that may be present in the interior of the plate thickness have little safety significance and, accordingly, the allowable flaw sizes for this region are relatively large. Underclad cracks, lack of fusion, and slag defects are all present in PDI test specimens.

Some possibility exists for lack of fusion between weld beads, sometimes called inter run lack of fusion, which could have off-normal orientation. Such manufacturing flaws are expected to be limited in size by the weld bead thickness, but could be oriented in a variety of ways. The flaw would follow the bead contour and exhibit a cup-shaped geometry (Reference 3). Thus, such a flaw that may be located near the inner surface of the vessel is expected to present some flank or facet that would be detected by either the 70 or the 45* beam angles used for the  ;

examination of this region. Flaws of this type could potentially be present in the interior of the plate thickness and  !

might present a detection problem for both single-sided or two-sided examination. However, these interior flaws are oflittle safety significance as discussed above.  !

l i

1 l

Baseline inspections were performed using the applicable ASME Section V and XI rules. Pre-existing manufacturing ofr-normal flaws that exceed ASME Code recording criteria would have been detected and documented. The baseline inspections also used the requirements of Reg. Guide 1.150 for recording at 20%

distance amplitude curve (DAC) for the inner (1/4)T. The probability is very low that significant original manufacturing off-normal flaws exist that have not been detected.

Detection of planar flaws is optimum when the incident sound beam is normal to the face of the flaw where maximum energy is reflected back to the transducer. As the flaw is rotated either forward or backward along the axis of the beam the reflected energy will decrease due to directivity pattern of the reflection to a point where none of the reflected energy is reficctea back to the transducer. However, planar flaws also exhibit difTraction patterns at the edges (tips) of the flaw. These diffracted signals radiate in all directions which allow them to be detected over a much larger range of flaw orientations than possible when just using the specular reflection from the flaw.

As an illustration consider an embedded planar flaw that is perpendicular to the vessel surfaces. The F<amatome procedure detw.ed these flaws in the PDI samples with 45' angle beams. For these type of flaws, the plane of the flaw is at a 45' angle to the beam axis such that the beam is reflected away from the transducer at 90 to the incident beam. Flaw detection, in this case, is limited to low energy reflections ? hat return to the transducer due to the rough surface of the flaw and the tip diffracted signals from the flaw tips. The tip difTracted signals are usually stronger than the low level reflected energy from the roughness of the flaw. The Framatome procedure is optimized to oetect these signals from off-normal oriented flaws as well as from service-induced flaws, as proven by the performance demonstration on the PDI samples.

Flaw Geometrice that mar Mask Other Flaws:

Tiaws that may be large enough to mask or shadow other flaws are expected to be a low probability of occurrence.

When using the FTl PD1 qualified procedure in the field at Braidwood, this would not be a concern. The FTl PDI

}

qualificJ procedure conducts detection scans from one direction, but requires sizing scans in two directions. When conducting sizing scans,100% of the code required volume (i.e., ID to OD and a distance of 1/2 thickness off each side of the weld)in the area containing the flaw is reeanned from both directions using the same transducers that are used for detection, except a smaller scan index (0.2" compared to 0.5") is used. Thus, a flaw that may be masked by another flaw geometry would be detected during the sizing scans.

l Rouch Claddinc: )

l Rough cladding certainly is an important variable affecting flaw detection reliability. Appendix VIII and PDI require that specimens contain cladding processes and conditions generally representative of actual field conditions.

l The PDI specimens contain typical cladding surface variations including ground areas and depressions severe  ;

enough to cause beam re-direction and affect flaw detection. Thus, single-direction examination procedures must '

be capable of coping with this cladding condition in order to qualify at PDI.

1 The Braidwood vessels were built by Babcock and Wilcox. Fabrication requirements were established to assure a suitable surface finish of the clad for future ISI from the vessel ID. Surface preparation of the clad consisted of belt grinding of the manually deposited back-clad areas of the weld to bM it with the automatically clad shcIl sections. This generally resulted in a uniform surface suitable for oamination. This finishing process is considered to be represented by the clad surfaces of the PDI specimens.

2

~

i

Conclusion:

The vessel inspections at Braidwood are required to be performed per the ISI Codes of record, which are the ASME ]

Section V and XI 1983 Edition, through Summer 1983 Addenda. When implementing these codes, the following generally applies (per Section V):

1) Scanning for reflectors oriented parallel to the weld, the adjacent base metal in the examination volume must be completely scanned by two angle beams, but need not be completely scanned by both angle beams i from both directions. The weld material requires scanning to be done in two directions 180* apart from each other. Where the ultrasonic beams are directed essentially normal to the plane of the weld, such as when the examination is performed from the flange face for the sheli to flange weld, beam angles i ruflicient to provide complete coverage of the weld from one direction shall be acceptable.

2) Scanning for reflectors transverse to the weld, scanning shall be done on the weld and base material in two disections 180 to each other to the extent possible.

3) Straight beam scanning for planar reflectors shall be performed on the entire volume of weld and adjacent base material to the extent required by the referencing Code section.

In lieu of the above scanning requirements, Braidwood proposes to use an examination procedure, equipment and personnel qualified by performance demonstration in accordance with the PDI program for implementation of ASME Section XI, Appendix Vill. The FTI PD1 qualified procedure requires scanning at higher sensitivity than that required by the ISI Code of record. Additionally, for flaw detection, the FTI PDI qualified procedure not only j looks for sound reflecting from the flaw face but also looks for tip diffracted signals. By implementing enhanced i UT techniques, the FTI PDI qualified procedure has demonstrated the ability to detect the range of flaws presented in the PD1 specimens while scanning with the angle beams from one direction perpendicular to the weld and one direction parallel to the weld. )

Examination of RPV welds using procedures qualified for one directional detection scans of the examination l volume meets Appendix VIII, ASME Section XI Code requirements as modified by the PDI Program and aru expected to prmide adequate flaw detection capability. The FTl procedure has been demonstrated and qualified by PDI. The PDI qualification process cha:lenges inspection procedures and examiners by presenting realistic  ;

conditions that may affect inspection performance for either single or two sided examinations. The newer ASME j Codes have provided for use of qualind procedures as an alternative to all older code requirements of Section XI '

and V. Qualification provides documented evidence of UT procedures effectiveness. The prosed draft Generic Letter, " Effectiveness of Ultrasonic Testing Systems in Insersice Inspection Programs" (61 $ ?. REG. 69120),

concurs in that it states that UT procedures qualified under the PD1 program provide an acceptable level of safety.

NRC RAI Item #2.2:

Verify shat all Code requirements, except far those specifically described in the licensee's February 25, 1997, response, will be met.

Response

A review of the procedure that was qualified in accordance with Appendix VIII of the 1992 Edition of ASME Section XI, through 1993 Addenda as modified by the PDI Program Description Document, Revision I was j performed against the requirements of ASME Section XI 1983 Edition, through 1983 Summer Addenda. Code i differences hav~ %en identified in the February 25,1997, submittal and the remaining applicable requirements of ASME Section XI 1983 Edition, through 1983 Sununer Addenda will be implemented.

3

l l

i f NRC RAI Item #2.3:

Verify that the subject examinations will be performed by implementing the same essential variables as those used for the PDI qualification.

I Response j The examinations will be performed in accordance with the vendor's qualified PDI procedure. All essential variables that are necessary per their qualified PD1 procedure will be required during the Braidwood Unit 1 and 2 l inspections.

NRC RAI Item #2.4:

1s it the intent of the subject reactor pressure vessel examinations to satisfy the current interval Section XI

requirements as well as the augmented reactor pressure vessel examinations required by 10 CFR g 50.55a(g)(6)(ii)(A)?

Response

Yes, the subject RPV examinations are used to satisfy the current interval Section XI requirements as well as the 2

augmented reactor pressure vessel examinations required by 10 CFR 50.55a(g)(6)(ii)(A). Braidwood will comiuct

, the subject RPV examinations during the AIR 06 and A2R06 outages. The extent of Category B-A, item Bl.10 welds are in accordance with those of ASME Section XI 1989 Edition. The augmented RPV NDE procedures used for these examinations are consistent with the applicable rules of the ISI Code of record. For the Section XI and Augmented RPV examinations, Braidwood is proposing through NR-29 to substitute the qualified PDI examination techniques for those applicable UT techniques specified by the ISI Code of record.

c

References:

1. Launary, J.P., Results of EDF/Framatome Underclad Crack Detection Methods. EPRI NP-2841. January.

1983

2. Ammirato, F. et al., Oualification ofInsenice Examination of the Yankee Rowe Reactor Pressure Vessel.

EPRI TR-101761. December.1992.

3. Chapman, 0.3.V, and Simonen, F., A Simulation Modelfor Estimating Probabilities of Defects in ll' elds, work performed for the USNRC Contract DE-AC06-76RLO 1830, S. Malik, Program Monitor.

4 I

._