ML18038A862
| ML18038A862 | |
| Person / Time | |
|---|---|
| Site: | Browns Ferry  |
| Issue date: | 08/23/1994 |
| From: | Machon R TENNESSEE VALLEY AUTHORITY |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| GL-94-03, GL-94-3, NUDOCS 9408310152 | |
| Download: ML18038A862 (54) | |
Text
REGULATORY INFORMATION DISTRIBUTION SYSTEM (RIDS)
ACCESSION NBR:9408310152 DOC.DATE: 94/08/23 NOTARIZED: NO DOCKET g
FACIL:50-259 Browns Ferry Nuclear Power Station, Unit 1, Tennessee 05000259 50-260 Browns Ferry Nuclear Power Station, Unit 2, Tennessee 05000260 50-296 Browns Ferry Nuclear Power Station, Unit 3, Tennessee 05000296 p
AUTH.NAME AUTHOR AFFILIATION MACHON,R.D.
Tennessee Valley Authority RECIP.NAME RECIPIENT AFFILIATION R
Document Control Branch (Document Control Desk)
SUBJECT:
Provides response to GL 94-03, "IGSCC of Shrouds in BWRs.>>
Results of core shroud insps in BFN Unit 3 during June
& Jul 1994 indicate that severe core shroud cracking not occurring at BFN Unit 3.Units 1
& 2 insp schedules also provided.
DISTRIBUTION CODE:
A001D COPIES RECEIVED:LTR ENCL SIZE:
TITLE: OR Submittal: General Distribution NOTES:
0 RECIPIENT ID CODE/NAME PD2-4 TRXMBLE,D XNTERNAL ACRS NRR/DORS/ONDD NRR/DRPW NRR/DSSA/SRXB OC B
EG I 01 EXTERNAL: NOAC COPIES LTTR ENCL 1
1 2
2 6
6 1
1 1
1 1
1 1
0 1
1 1
1 RECIPXENT ID CODE/NAME PD2-4-PD WILLIAMS,J.
NRR/DE/EELB NRR/DRCH/HXCB NRR/DSSA/SPLB NUDOCS-ABSTRACT OGC/HDS3 NRC PDR COPIES LTTR ENCL 1
1 2
2 1
1 1
1 1
1 1
1 1
~ 0 1
1 D
C U
NOTE TO ALL"RIDS RECIPIENTS:
PLEASE HELP US TO REDUCE iVASTE!CONTACTTHE DOCUMENTCONTROL DESK, ROOM PI-37 (EXT. 504-2083 ) TO ELIMINATEYOUR NAMEFROM DISTRIBUTIONLISTS FOR DOCUMENTS YOU DON'T NEED!
TOTAL NUMBER OF COPIES REQUIRED:
LTTR 24 ENCL 22
K
~
4 l t g
Tennessee Vattey Authority, Post Oftice Box 2000, Decatur, Atabama 35609-2000 R. D. (Rick) Machon Vice President, Browns Feny Nuctear Ptant August 23, 1994 U.S. Nuclear Regulatory Commission ATTN:
Document Control Desk Washington, D.C.
20555 Gentlemen:
In the Matter of
)
Tennessee Valley Authority
)
Docket Nos.
50-259 50-260 50-296 BROWNS FERRY NUCLEAR PLANT (BFN) - UNITS 1I 2 AND 3
RESPONSE
TO NRC GENERIC LETTER (GL) 94 1NTERGRANULAR STRESS CORROSION CRACKING (IGSCC)
OF SHROUDS IN BOILING WATER REACTORS This letter provides TVA's response for BFN Units 1, 2,
and 3
to GL 94-03, dated July 25, 1994.
As discussed with NRC staff in a meeting on August 11,
- 1994, TVA has been fully aware of IGSCC concerns in boiling water reactor (BWR) internals and has been working closely with the BWR Owners Group (BWROG) to address this issue.
Additionally, TVA performed core shroud inspections in Unit 3 during June and July 1994.
The results of the Unit 3 inspections indicate that severe core shroud cracking is 'not occurring at BFN.
TVA believes that similar results will be found in the Unit 2 core shroud since it was constructed using similar materials and procedures, and both units had similar cycle 1 through 5
chemistry history.
TVA has analyzed the Unit 3 inspection results and determined
'that Unit 3 can safely be returned to service and operated for at least one cycle of operation without repairs.
- Also, TVA performed a plant-specific safety analysis for Unit 2 using extrapolated Unit 3 inspection results.
Based on the
- analysis, TVA determined that postulated core shroud cracking will not impact the safe operation of Unit 2 until inspections are conducted.
"RAP '")
9408310152 940823 PDR ADOCK 05000259 L
p PDR
U.
S. Nuclear Regulatory Commission Page 2
August 23, 1994 TVA's reply to the information requested in GL 94-03 is provided in Enclosures 1 and 2.
As described, TVA will conduct inspections, using the best available technology, of selected Unit 2 core shroud welds during the Unit 2 Cycle 7
refueling outage that is scheduled to begin October 1,
1994.
Similar inspections will be conducted in Unit 1 prior to restart.
Plans for inspecting other shroud welds and vessel internals will be based on future BWROG recommendations that are being coordination with NRC.
Enclosure 3 provides a summary of the results of the Unit 3 core shroud weld inspections conducted in June and July 1994.
As described, there is no indication that significant cracking exists in the core shroud welds.
A summary of TVA's safety analysis that supports restart and operation of the Unit 3 core shroud is also provided.
The analysis used conservative assumptions for such factors as crack growth and uncertainties in the amount of cracking identified.
A summary of the'ommitments contained in this letter is provided in Enclosure
- 4. If you have any question please telephone me at extension (205) 729-2636.
Sincerely, R.
D.
M on Site V' President Subscribed and sworn to before me on this ~~day of 1994
~
Notary Public My Commission Expires Enclosures cc:
See page 3
V II b
U.S. Nuclear Regulatory Commission Page 3
August 23, 1994 Enclosures cc (Enclosures):
Regional Administrator U.S. Nuclear Regulatory Commission Region II 101 Marietta Street, NW, Suite 2900.
Atlanta, Georgia 30323 Mr. Mark S. Lesser, Section Chief U.S. Nuclear Regulatory Commission Region II 101 Marietta Street, NW, Suite 2900 Atlanta, Georgia 30323 NRC Resident Inspector Browns Ferry Nuclear Plant Route 12, Box 637
- Athens, Alabama 35611 Mr. J.
F. Williams, Project Manager U.S. Nuclear Regulatory Commission One White Flint, North 11555 Rockville Pike Rockville, Maryland 20852
ENCLOSURE 1
TENNESSEE VALLEY AUTHORITY BROWNS FERRY NUCLEAR PLANT (BFN)
UNITS 1i 2 AND 3
Response
to Generic Letter (GL) 94-03, "Intergranular Stress Corrosion Cracking of Core Shrouds in Boiling Water Reactors" BACKGROUND Intergranular stress corrosion cracking (IGSCC) of boiling water reactor (BWR) core shrouds has been identified as a
technical issue of concern by NRC and the nuclear industry.
TVA has been fully aware of IGSCC concerns in BWR internals and has been working closely with General Electric (GE) and the BWR Owners Group (BWROG) to address this issue.
TVA has assigned representatives to the various BWROG management steering groups and subcommittees that were created to address the BWR internals IGSCC phenomenon.
BFN Units 2 and 3 are considered to be medium risk plants based on the criteria provided in the BWROG "BWR Core Shroud Evaluation" report transmitted to NRC on April 5, 1994 (GENE-523-148-1193).
This classification is based on the variables that, are considered to have an impact on the relative susceptibility to core shroud IGSCC (i.e., materials, fabrication techniques, water chemistry, and neutron fluence).
Table E1-1 provides information for BFN about these factors.
TVA developed plans for addressing IGSCC of the BFN core shrouds based on GE and BWROG recommendations.
These
, plans included inspection of 100< of the accessible areas on welds H1 through H7 (Figure E1-1) using the latest available non-destructive examination (NDE) technology.
The inspections were conducted in Unit 3 from June 13 to July 14, 1994.
Unit 2 inspections are scheduled to be conducted during the Unit 2 Cycle 7
(U2C7) refueling outage scheduled to begin 'on October 1,
1994.
Unit 1 inspections are scheduled to be completed prior to Unit 1 being returned to service.
On July 24,
- 1994, GL 94-03 was issued requesting that (1)
BWR licensees inspect their core shrouds no later than the next scheduled refueling outage, and perform an appropriate evaluation and/or repair based on the results of the inspection; and (2) perform a safety analysis for the operating units supporting continued operation until
inspections are performed.
In order to keep NRC apprised of TVA's plans for addressing IGSCC of the core shroud welds and to ensure that our approach to the issue is consistent with GL 94-03 requested
- actions, a meeting between TVA and NRC staff was held on August 11, 1994.
The results of the Unit 3 inspections indicate that severe cracking of the core shroud welds is not occurring at BFN.
TVA and GE have analyzed the inspection results and determined that Unit 3 can safely be restarted and operated for at least one cycle after restart.
TVA believes that postulated IGSCC cracking of the core shrouds will not impact safe facility operation until future inspections are conducted.
A summary of the Unit 3 core shroud inspection results and safety analysis is provided in Enclosure 3.
The detailed results are available on-site for review.
TVA performed a safety analysis to show that it is safe to operate Unit 2 for the remaining six weeks of the cycle until inspections are made.
The safety analysis is principally based on extrapolating the results of the Unit 3 inspection.
The Unit 3 inspection results can be extrapolated to Unit 2 since Units 2 and 3 were constructed using similar materials and procedures, and there were no significant differences in cycle 1 through 5
chemistry operating history.
The only significant difference between Units 2 and 3 is that Unit 2 has operated longer.
TVA and GE performed the analysis to ensure that the Unit 2 core shroud structural margins remain adequate to support operation until the U2C7 refueling outage begins.
TVA and GE's analysis used conservative flaw growth factors (5 x 10 'nches/hot-hour) for estimating the size of the cracks in Unit 2 since it has operated for approximately 37,000 more on-line hours.
The analysis indicates that postulated cracks will be less that the safety criteria established by GE and the BWROG (GENE-523-A107P-0794 and BWROG letters to NRC dated July 13 and 14, 1994).
As a result, TVA concludes that Unit 2 can operate safely until inspections are conducted during the U2C7 refueling outage.
E1-2
I
TABLE E1-1 BFN CORE SHROUD SUSCEPTIBILITY FACTORS CONSTRUCTION MATERIALS (Units 1, 2,
and 3)
Shell plate ASTM A240 Type 304 stainless
- steel, nominal 2" thickness Forged rings ASTM A182 Type F304 stainless steel Carbon content 0.03 0.064 Inconel 600 below weld H7 WELD FILLER MATERIALS (Units 1, 2,
and 3)
Inconel weld filler INC0182 Stainless steel weld filler ASTM Type E308 and ER-308 Unit 1 Arosta Type 304:
7 94 ferrite Unit 2 Arosta Type 304L:
6 7.84 ferrite, AISI Type 308:
12 -14< ferrite, AISI Type 304: 8.8 124 ferrite, and Jungo Type 304:
5.4% ferrite Unit 3 Arosta Type 304 electrode:
6.4 7.84 ferrite, Jungo Type 304: 5.44 ferrite, and AISI Type 308:
12 -14% ferrite FABRICATION TECHNIQUES (Units 1, 2, and 3)
Shell plate and forged rings ASME Section IX welding procedures CYCLE 1-5 REACTOR WATER CHEMISTRY (Conductivity)
Unit 1 0.384 pS/cm Unit 2 0.364 pS/cm Unit 3 0.303 pS/cm MAXIMUM FLUENCE Unit 1 4.4 x 10~0 n/cm~
(H4 weld only)
Unit 2 5.2 x 10 n/cm (H4 weld only)
Unit 3 3.5 x 10~0 n/cm~
(H4 weld only)
E1-3
FIGURE E1-1 BFN CORE SHROUD CONSTRUCTION DETAILS Shr oud Head Flange H-1 H2 Top Guide Suppor t Ring H3 I
Oi4ll Q.
i0 O
O Cl 0
H-4 H5 Core Plate Support Ring H-6 H-7 H-9 Shroud Suppor t Plate H-8 H-10 Shroud Support Leg H-11 E1-4
II'EPLY TO GL 94-03 REPORTING RE UIREMENTS Reporting Requirement 1(a)
A schedule for inspection of the core shroud.
Per requested licensee Action 1, the inspections are to be performed as follows:
"Inspect the core shrouds in their BWR plants no later than the next scheduled refueling outage" TVA Response Unit 1 the core shroud will be inspected before restart.
Unit 2 the core shroud will be inspected during the U2C7 refueling outage scheduled to begin October 1, 1994.
Unit 3 the core shroud inspection is complete.
The inspections were performed from June 13 to July 14, 1994.
A summary of the results is provided in Enclosure 3.
Reporting Requirement 1(b)
A safety analysis, including a plant-specific safety assessment, as appropriate, supporting continued operation of the facility until inspections are performed.
Per requested licensee Action 2, the safety analysis is to be performed as follows:
"Perform a safety analysis supporting continued operation of the facility until inspections are conducted.
The safety analysis should consider, but not be limited to the following factors:
a ~
b.
Details of the conditions that would influence the probability of the occurrence of cracking and rate of crack growth (e.g., material types and forms, water chemistry,
- fluence, carbon contents, welding materials and procedures).
A plant-specific assessment accounting for uncertainties in the amount of cracking, which should include but not be limited to, the following:
An assessment of the shroud response to the structural loading resulting from design basis events (e.g.,
steam line break, recirculation line break).
If asymmetric loads can affect the shroud response, these should also be considered.
2.
An assessment of the ability of plant safety features to perform their function considering the shroud response to structural loading (e.g., control rod insertion, Emergency Core Cooling System (ECCS) injection)."
TVA Response Unit 1 not applicable.
A safety analysis supporting continued operation is not required since Unit 1 is shutdown.
Unit 2 a summary of the Unit 2 safety analysis is provided in Enclosure 2.
Unit 3 not applicable.
A safety analysis supporting continued operation is not required since Unit 3 is shutdown.
Also, the core shroud inspection is complete.
Reporting Requirement 1(c) r A drawing or drawings of the core shroud configuration showing details of the core shroud geometry (e.g.,
support configurations for the lower core support plate and the top guide, weld locations and configurations).
TVA Response The Unit 1, 2,
and 3 core shroud configuration is shown in Figure E1-1.
Reporting Requirement 1(d)
A history of shroud inspections for the plant should be provided addressing
- date, scope, methods and results, if applicable.
TVA Response Inspections of the core internals were conducted per GE RICSIL 054 g "Core Support Shroud Crack Indications, Revision 1.
The inspections consisted of visual examinations of the accessible areas of the seam welds and
,heat affected zones on the inside and outside surfaces of the shroud.
The inspections were performed using Westinghouse Model ETV-1250 cameras; however, the welds were not brushed since this was not specified at that time in the RICSIL.
The inspections were conducted in Unit 1 in June 1992, Unit 2 in November
- 1990, and Unit 3 in November 1991.
The inspections did not reveal any indications of core shroud cracking.
Enclosure 3 provides additional information about recent inspections in Unit 3.
Reporting Requirement 2
No later than 3 months prior to performing the core shroud inspections (If the inspections are scheduled to begin in less than 3 months from the receipt of this letter, the licensee should contact their NRC project manager to establish a schedule for providing the following information):
a.
The inspection plan in item 3 of Re uested Actions.
b.
Plans for evaluation and/or repair of the core shroud based on the inspection results.
Item 3 of Re uested Actions states:
"Develop an inspection plan which addresses:
(a) all shroud welds (from support attachments to the vessel to the top of the shroud) and/or provides a
justification for elimination of particular welds from consideration; and (b) examination methods with appropriate consideration given to use of the best available technology and industry inspection experience (e.g.,
enhanced VT-1 visual inspections, optimized UT techniques).
Standard methods for inspection of core support structures as specified by the ASME Code,Section XI, have been shown to be inadequate for consistent detection of IGSCC in core shrouds."
TVA Response a.
TVA's inspection plan is to inspect 1004 of the accessible areas on welds Hl through H7.
The inspection has been completed in Unit 3.
The inspection will be performed in Unit 2 during the U2C7 refueling outage.
The inspection will be performed in Unit 1 prior to restart.
TVA expects the accessibility of the Unit 1 and 2 welds to be similar to Unit 3.
The tentative Unit 2 Core Shroud inspection schedule is provided as Figure E1-2.
Changes or updates to the schedule will be coordinated through the BFN Senior Resident Inspector and NRC Project Manager.
Subsequent core shroud examinations will be performed as described in GE Service Information Letter (SIL)-
572 (i.e., currently once per cycle if cracks are found or once every other cycle if no cracks are identified) or until repairs are made.
The need for inspecting the vertical welds and the welds on the shroud support plate and legs (welds H8 through H11) will be evaluated for possible inclusion
in future refueling outages based-on BWROG and GE commendations.
Currently there are no plans to inspect the Unit 2 vertical welds and welds H8 through Hll during the U2C7 outage based on the following:
1.
Cracking of the vertical welds presents no safety concern if the horizontal welds are intact.
2.
Ultrasonic testing equipment and test mock-up facilities for the vertical welds are not yet available.
3.
The H8 through H11 welds are factory welds.'.
The H8 through H11 welds are in an extremely low fluence area, significantly below the core plate.
5.
The IGSCC that has been identified in BWR core shrouds has occurred in 300 series stainless steel.
The material below weld H7 is Inconel 600.
6.
TVA has reviewed a 1991 videotape of the Unit 3 shroud support legs and did not identify any cracking.
7.
The H8 through H11 welds are in areas that have very limited access.
The ultrasonic inspection equipment and test mock-up is not yet available.
8.
Performing inspections of these welds at this time does not appear to be consistent with radiation exposure guidelines to maintain dose "As Low As Reasonably Achievable."
b.
Units 1 and -2 core shroud examinations will be conducted using the best available technology.
The examination methods planned for Unit 2 will be similar to those used in Unit 3.
TVA will primarily rely on the GE Smart-20002 ultrasonic system.
This is an automated scanning system that uses at least one 45'hear wave transducer and one 60'efracted longitudinal wave search unit.
1Preliminary review of fabrication data indicated that wclds Hg through Hl 1 werc included in thc shop fabrication and werc incorporated in the vcsscl post weld heat treatment.
This was discussed in the August 11, 1994, meeting with NRC. Additional information indicates that only attachment standoff portions ofwelds H9 and Hl 1 were included in thc vessel post weld heat treatment.
IVAindicated in thc August 11, 1994, meeting with NRC that this system was called the GERIS-2000 system. The actual name of thc system is 'Smart-2000'ystem
Suction cup scanners will be used where practical in areas of limited accessibility.
Enhanced video systems (e.g.,
Westinghouse 1250 camera) will be available for use as necessary;
- however, TVA intends to use the camera principally to look for obstructions and enable ultrasonic testing setup.
Qualified inspection personnel will perform the inspections.
c.
Units 1 and 2
TVA will perform a structural margin analysis using the results of the core shroud inspections.
On the basis of the structural margin
- analysis, TVA will determine if the plants can resume operation without. repair.
The analysis will be based on BWROG safety assessments and use of established estimates for determining the acceptability of
'ontinued operation (e.g., flaw growth rate).
Unit 3 TVA performed a structural margin analysis using the results of the core shroud inspection.
A summary of the analysis is provided in Enclosure 3.
On the basis of the analysis, TVA determined that Unit 3 can resume operation without repair.
The details of this assessment are available on-site for review.
TVA is working with the BWROG and EPRI to develop acceptable repair plans.
As the repair plans are finalized and determined to be acceptable by NRC, TVA will develop plant-specific repair procedures.
Reporting Requirement Within 30 days from the completion of the inspection, provide the results of the inspection.
TVA Response Unit 1 TVA will provide inspection results within 30 days of completing the core shroud inspection.
Unit 2 TVA will provide inspection results within 30 days of completing the core shroud inspection.
Unit 3 the inspection results are provided in Enclosure 3.
E1-9
Figure El-2 EARLY EARLY REN START FINISH DUR OCT NOV 100CT'H 21:N ttacrR4 8:S'1 12 11DCN4 13ico 120cr'H 0~SR 12 120CN4 13ioo 13acrR4 OiSR 12 13DCTR4 Shoo 13acrR4 16iSR 12 14DCN4 13 00 15OCN4 6 SR 18 lsOCTR4 13 00 16acrR4 6 SR 18 170CPH 7ica 17actR4 12isR 6
17OCTR4 13iaa 17actRt toisR 6
17acrR4 22isR 0
205EPR4 7soa 3OCPH 14<SR 80 tacN4 tt sR 0
2acN4 10 sR 0
3OCTRt tsioa 4OCN4 SssR 15 4DCTcH 21<00 SacrR4 siSR 10 SDCTR4 7 oo sacN4 to sR socrR4 21 00 6acN4 20 sR 6OCTRI 21 F 00 7OCN4 8is'1 12 70crR4 R m 7acrR4 16 SR 8
7DCTR4 17<00 8OCN4 4iSR 1
7DCTR4 17soo 160CN4 OiSR 200 DacTR4 R oo 8acrR4 20 SR 12 RDCTR4 7ioo RDCN4 toiSR 12 RDCN4 23:00 loaCT'H 10:SR 12 FZGE010681 U-2 SHROUD INSPECTION PREREQS C EQPT SETUP ST00010150 REACTOR AT HOT SHUTDOWN ( ALL RODS IN )
tvt ST00010300 RX AT COLD SHUTDOWN QFH00010460 DETENSION RPV STUDS PFH00010500 RNV RPV HEAD ll, INSTL FLANGE PROT IFH00010520 RHV STE01 DRYER
~FH00010600 FLOOD CAVITY/XFER HOIST SEP/ACHIEVE MTR CLARITY QFZGE010680 FLUSH RPV NOZZLES AS REQD TO SUPPORT UT EXAHS QFZGE010682 INSTALL U-2 SHROUD INSP EQUIPNENT QFZGE010695 PERFORN UT EXAH OF RPV SHROUD MELD H1A ST00010680 PERFORH U-2 CORE SHUFFLE QFZGE010616 PERFORH UT EXAH OF RPV SHROUD MELD H1B QFZGE010687 PERFORH UT EXAH OF RPV SHROUD WELD H2A QFZGE010698 PERFORt4 UT EXAH OF RPV SHROUD WELD H28 QFZGE010683 PERFORH UT EXAH OF RPV SHROUD MELD H3" QFZGE010684 PERFORH UT EXAM OF RPV SHROUD WELD H38 QFZGE010685 PERFORH Ul EXRH OF RPV SHROUD MELD H4A QFZGE010686 PERFORH UT EXAH OF RPV SHROUD MELD H48 QFZGE0106tl0 PERFORM UT EXAH OF RPV SHROUD WELD HSA QFZGE010614 PERFORH UT EXAM OF RPV SHROUD WELD HSB IFZGE010692 PERFORH UT EXAH OF RPV SHROUD MELD H6 gFZGE010693 PERFORH UT EXAH OF RPV SHROUD HELD H7 OFZGE0$ 0687 U 2 RPV SHROUD INSPECTION COHPLETE Pi t P t 22NIGqk 043'
~git p
p iy y i Oat Pate
>ENlela T
OO ~
C ti joiwV Project Start 1 jUNR3 Oiao ~~
Progress Oor Project Ftnlsh 3CO<CR4 23,5R
~
c>/p ttuesttvvtirtolt ectlvlty t c) Prlcavera steno, re.
Beet 1 of 1
Tennessee Valley Authority Browns Ferr y Nuclear Plant UNIT 2 CORE SHROUD INSPECTION
ENCLOSURE 2
TENNESSEE VALLEY AUTHORITY BROWNS FERRY NUCLEAR PLANT (BFN)
UNIT 2 BFN UNIT 2 PLANT-SPECIFIC SAFETY ASSESSMENT INTRODUCTION This assessment supports the continued operation of Browns Ferry Unit 2 until the Cycle 7 refueling outage, scheduled to begin October 1, 1994.
Browns Ferry Unit 2 is currently in the seventh cycle of power operation.
The assessment of the Unit 2 core shroud response has been formatted to follow the General Electric (GE)
"BWR Shroud Evaluation Cracking Generic Safety Assessment,"
Revision 1
(GENE-523-A107P-0794).
Since the GE assessment addressed all GE BWR model reactor vessels, and the BFN Unit 2 reactor vessel was supplied by GE, BFN is bounded by the
- inputs, assumptions and conclusions of GE.
This safety analysis considers the details of the conditions expected on the Unit 2 shroud based on factors that influence the probability of crack occurrence and rate of crack growth (e.g., material types and forms, water chemistry,
- fluence, carbon contents, welding materials and procedures).
The assessment is principally based on extrapolating the results of the Unit 3 core shroud inspection conducted in June and July 1994.
In performing the assessment, TVA accounted for uncertainties in the extent of cracking, anticipation of the shroud response under design basis conditions, and the ability of plant safety systems to function considering the shroud response.
SHROUD FUNCTION The shroud is an austenitic stainless steel cylindrical assembly that provides a partition between the core region and the downcomer annulus.
The shroud separates the upward flow of coolant through the core from the downward recirculation flow.
The shroud also provides, in conjunction with other components, a eoolable core geometry and a floodable region following a postulated recirculation line break.
The shroud is not a primary pressure boundary component.
The shroud was designed and fabricated to GE specifications.
The shroud construction materials and procedures are summarized in Enclosure 1,
Table E1-1.
The shroud cylinder consists of a series of shells and
- rings, connected by welding to form one structure.
The names used to designate the girth welds in the BFN shroud structure are defined as follows (See Figure El-1):
Hl:
H2:
H3:
Upper cylindrical weld attaching the shroud head flange to the uppermost shroud section.
This weld is above the Core Spray spargers (the emergency core cooling system (ECCS) injection headers).
Upper cylindrical weld attaching the uppermost shroud section to the top guide support ring.
This weld is below the Core Spray spargers.
Upper weld located on the bottom side of the top guide support ring.
H4:
Mid-plane weld located above the core plate.
H5:
H6:
The weld located just above the core plate, attaching the lower portion of the upper shroud section to the top of the core plate support ring.
The weld located just below the core plate, attaching the top of the lower shroud section to the bottom of the core plate support ring.
H7:
H8:
Lower cylindrical weld located below the core plate, attaching the shroud to the shroud support.
Lower cylindrical weld located below the core plate, attaching the shroud support ring to the shroud support plate.
H9:
Lower cylindrical weld located below the core plate, attaching the shroud support plate to the reactor vessel.
H10:
Lower weld located below the core plate, attaching the shroud support ring to the support leg.
H11:
Lower weld located below the core plate, attaching the support leg to the reactor vessel.
III.
SUSCEPTIBILITY ASSESSMENT There are numerous factors which contribute to the susceptibility of a particular shroud to stress corrosion cracking (SCC).
Inspection results from 23 core shrouds provides information on the extent of cracking which can be evaluated against some of the parameters which influence the development of SCC.
The BWROG used the industry inspection information to establish E2-2
susceptibility groupings based on quantitative factors such as material type and form, operating time and conductivity history.
The following discussions indicate how BFN Units 2 and 3
align with the industry data to determine the susceptibility of the Unit 2 shroud to SCC.
The discussion also shows the correlation between Units 2 and 3 so that it is reasonable to conclude that the Unit 3 inspection results may be extrapolated.
FABRICATION OF CORE SHROUDS Based on inspections performed to date, the shroud welds that have exhibited the greatest extent of circumferential cracking
(>200 inches of effective crack length) are all associated with ring to shell welds H1, H2, H3, and H5.
The various support rings present above and below the top guide and at the core plate were either fabricated from forgings or arcs cut from rolled plate and welded into a ring configuration followed by machining to size.
The fabrication process for the welded-plate rings can result in high residual stress due to the presence of a relatively deep layer of cold work on the short transverse "end grain" surface which is directly exposed to reactor coolant.
The potentially higher SCC susceptibility of the welded-plate rings has been shown in those plants using this type of ring fabrication process to have more extensive cracking present.
The BFN shroud rings are fabricated from ASTM A182 F304 forged stainless steel and the shroud shell is fabricated from ASTM A240 type 304 stainless steel plate.
The use of forged rings on the shroud renders them less susceptible to extensive (360')
cracking.
WATER CHEMISTRY HISTORY Testing performed for SCC has shown that SCC initiation and growth is strongly dependent on the electrochemical corrosion potential (ECP) on the surface of a component.
ECP depends on the level of oxidants, such as oxygen and peroxide, in the reactor water.
One variable of water chemistry that has been measured regularly during operation at all BWRs is conductivity.
Laboratory SCC tests and field experience with recirculation pipe cracking and shroud head bolt cracking have shown a
definite correlation of conductivity during the first 5 cycles of operation with initiation and growth of SCC.
The average reactor water conductivity for Units 2 and 3
for the first five fuel cycles was 0.36 pS/cm and 0.30 pS/cm, respectively.
This places Units 2 and 3 in the medium susceptibility category.
It should be noted that E2-3
during the last two operating cycles (C6 and C7), the average Unit 2 reactor water conductivity levels were approximately 0.1 pS/cm.
This may result in a reduced crack propagation rate during these later operating cycles.
SHROUD MATERIAL CARBON CONTENT Shroud cracking has been detected in core shroud material having a range of reported carbon contents from 0.023% to above 0.064.
From shroud inspection data, all incidents where cracking has exceeded 180'f circumference have occurred in ring plate material with carbon content equal to or greater than 0.064.
This is consistent with current levels of understanding of expected IGSCC susceptibility.
BFN shroud shells are fabricated from ASTM A240 type 304 stainless steel plate material.
The carbon content for this material has a range of 0.034 to 0.064 per the material specification.
This would indicate that cracking of the BFN shrouds would be no more extensive (i.e.,
greater than 180'f circumference) than other shrouds with higher carb'on content.
NEUTRON FLUENCE Core shroud cracking can occur by IGSCC in the absence of significant fluence as evidenced by industry information on the H1 and H7 welds where fluence, f, is very low (f<1.0 x 10" n/cm ).
- However, a fluence effect on cracking susceptibility (irradiation assisted stress corrosion cracking (IASCC) at f>5.0 x 10 n/cm ) or a synergistic interaction of fluence in already sensitized material (IGSCC at f > 1.0 x 10'/cm ) is expected and has.been verified by samples at other nuclear facilities.
The only significant flaws observed on the Unit 3 shroud were along the H5 weld.
This weld is approximately 25 inches below the active fuel, where the fluence is expected to be as much as three orders of magnitude less than the peak fluence (3.5 x 10 n/cm ).
The Unit 3 inspection results indicate that extensive cracking is not occurring in the peak fluence areas.
The Unit 2 peak fluence is only slightly higher (5.2 x 10~~
n/cm~).
The areas of significant fluence in Unit 2 are the same as for Unit 3; therefore, TVA expects that any cracking in'he high fluence areas will be acceptable.
OPERATING TIME The frequency and extent of core shroud weld SCC is expected to correlate with hot exposure time (i.e., time above 200').
In addition, fluence increases with E2-4
exposure time leading to increased susceptibility at some welds.
Industry experience indicates that cracking in excess of 180's unlikely until a plant accumulates 10 on-line years of operation.
The data also shows that, for those plants with forged rings, the extent of cracking does not exceed 180'n plants with operating times in excess of 10 years.
Unit 3 has operated for slightly more than 5 fuel cycles (approximately 45,600 hours0.00694 days <br />0.167 hours <br />9.920635e-4 weeks <br />2.283e-4 months <br /> or 5.2 years on-line).
At the end of Cycle 7 operation, Unit 2 will have operated for approximately 82,700 hours0.0081 days <br />0.194 hours <br />0.00116 weeks <br />2.6635e-4 months <br /> or 9.4 years on-line.
Therefore, TVA expects that any cracking will be acceptable.
CONCLUSIONS Based on the discussions of each of the specific Unit 2 parameters which may cause a shroud to be susceptible to
- cracking, Browns Ferry Unit 2 is considered to be in a low or medium risk category.
Additionally, the similarities between Units' and 3 provide reasonable evidence to conclude that the Unit 3 inspection results can be extrapolated to Unit 2.
STRUCTURAL MARGIN ASSESSMENT Industry testing has demonstrated that shrouds are made of ductile material with high toughness properties, even after accounting for any effects due to neutron fluence.
Calculations performed for over fifteen plants show that the typical allowable 360'ircumferential flaw size is approximately 954 of wall thickness.
However, since seismic and LOCA loads vary somewhat, and not all plants have been evaluated, the required minimum ligament is conservatively assumed to be 104 of the wall thickness.
The BWROG generic safety assessment structural margin conclusions included:
o All shrouds where extensive, particularly 360',
cracking has been found have had sufficient structural integrity to return to operation for at least one full fuel cycle.
o Computer model predictions, using PLEDGE code, indicate that significant crack growth may have occurred during early cycles of high conductivity, and that conductivity improvements and the nature of the residual stress profile would greatly curtail subsequent crack growth, although oxide wedging may be a contributor.
E2-5
EXTRAPOLATION OF UNIT 3 INSPECTION DATA TO UNIT 2 The BFN Unit 3 shroud has been inspected.
Indications of inside diameter initiated cracking warranting evaluation were detected only at the H5 weld.
Of the H5 weld length examined by UT, approximately 264 was found to be cracked.
The deepest indication at the H5 weld had a depth of 0.68 inches and a length of 1.63 inches.
The longest continuous indication has a maximum depth of 0.62 inches and a length of 32.2 inches.
The vessel shroud at these locations is a nominal two inches thick.
TVA, with the assistance of GE, used the inspection data from Unit 3 to postulate the flaw configuration for Unit 2.
The postulated flaw configurations are evaluated accounting for the differences between the two shrouds in terms of on-line hours.
Unit 2 will have approximately 37,100 more on-line hours than has been accumulated on Unit 3.
TVA's evaluation addresses the worst case weld, H5, and assumes Unit 2 initial flaw conditions at 45,000 hours0 days <br />0 hours <br />0 weeks <br />0 months <br /> of operation equal to the current extent and distribution of cracking in Unit 3.
Since a portion of the H5 weld is inaccessible for examination, TVA used two methods to distribute the postulated flaws in these regions;
- 1) the entire postulated flaw length is assumed to be adjacent to the nearest observed flaw, and
- 2) the entire postulated flaw length exists as a set of distributed flaws, whose length is based on the average length of flaws found in the examined areas.
In both cases, the total flaw length is assumed to start at 264 of the total circumference weld length. It is also assumed that no postulated indication in the uninspected areas exceeds the maximum depth observed in the inspected Unit 3 areas.
EVALUATION OF POSTULATED FLAWS FOR THE UNIT 2 SHROUD After an initial flaw condition was established, an additional increase in flaw depth of 0.3 inches was added to all flaws to account for the uncertainties in depth sizing by UT inspection (e.g.,
0.68"
+ 0.3" = 0.98" assumed depth).
With this uncertainty correction, the postulated initial Unit 2 flaw configuration (i.e., at 45,600 hours0.00694 days <br />0.167 hours <br />9.920635e-4 weeks <br />2.283e-4 months <br /> of operation) was assumed to grow at a rate of 5x10~ inches per hour for 37,100 hours0.00116 days <br />0.0278 hours <br />1.653439e-4 weeks <br />3.805e-5 months <br />.
Following projected flaw growth, ASME Code Section XI proximity rules were applied.
These rules were used since data was obtained using ultrasonic examination methods (UT).
The rules stipulate that any two adjacent flaws separated by less than twice the flaw depth, limited to twice the nominal wall thickness, are assumed to coalesce into a single flaw, with the length equal to the sum of E2-6
the individual flaw lengths plus the connecting ligament.
After flaw growth and coalescence of adjacent flaws, the lengths of all individual flaws were summed and the total compared to a previously determined maximum allowable flaw length.
Allowable flaw lengths for the BFN shrouds were established assuming only a visual weld inspection would be performed.
Therefore the acceptance criteria is based on the assumption that any cracks observed would be through-wall.
Assuming through-wall cracks, GE determined the limit load allowable flaw length for each weld.
The limit for the H5 weld is 416 inches.
The cumulative effective flaw length of 416 inches fully assures that ASME Code margins will be met.
A 90'indow limit of one-fourth the allowable length limit (i.e.,
104 inches) is also used to limit long continuous flaws on one side of the shroud.
Use of this acceptance criteria is very conservative for the BFN application since inspections were conducted using UT methods.
RESULTS Cumulative effective through-wall flaw lengths were estimated by calculating flaw growth using the two methods described above.
The results show that the estimated total effective flaw length for the Unit. 2 shroud at the end of Cycle 7 ranges from 245 to 348 inches with the maximum 90'indow flaw ranging from 93 to 113 inches.
The estimated total effective flaw length is less than the allowable cumulative through-wall flaw length of 416 inches.
The upper end estimate of the maximum 90'indow flaw exceeds the allowable of 104 inches by 9 inches.
However, since the estimated cumulative length flaw is less than the allowed cumulative through-wall flaw and the Unit 3 inspection shows a reasonable distribution of flaws around the circumference, adequate conservative margin is maintained.
The BWROG evaluation utilizes loads assuming a
simultaneous occurrence of the worst case pipe break for shroud loading and the safe shutdown earthquake.
The combination of these events has an extremely low probability of occurrence.
Furthermore, the allowable flaw length determination is based on the conservative assumption that the sum of all individual flaws acts as a
single flaw with loads applied to the remaining net section in the worst case orientation.
Consideration of a net section based on realistic flaw distribution would result in a structural margins much larger than the margin associated with a single through-wall flaw extending 348 inches around the circumference.
E2-7
CONCLUSIONS Based on the conservative methodologies and the allowable limit assuming through-wall flaws, TVA's evaluation shows that any Unit 2 flaws would be acceptable.
Therefore, there is reasonable assurance that operation of Unit 2 until the scheduled refueling outage in October 1994 is acceptable.
LIKELIHOOD OF UNIT 2 SHROUD DISPLACEMENT In order for shroud displacement to occur, several conditions must exist simultaneously.
- First, a 360',
.greater than 904 deep crack must exist in the shroud.
- Then, a
LOCA must occur which generates loads on the shroud comparable to the assumed design basis loads.
This section provides a qualitative discussion of the overall likelihood of this scenario.,
CRACKING POTENTIAL Based on the crack growth predictions and the crack susceptibility of the shroud material discussed
- above, the likelihood of unacceptable Unit 2 shroud cracking is considered to be extremely unlikely.
Therefore, the combination of factors which would result in potential shroud displacement is also considered extremely unlikely.
PROBABILITY OF A LOSS OF COOLANT ACCIDENT LOCA EVENT A LOCA involving an instantaneously severed steam line or recirculation pipe has never occurred in a BWR.
This represents about 400 reactor years of operation.
Considering that there are two to four pipes per reactor which could experience a LOCA, 800 to 1600 years of piping operation has occurred without a LOCA of this type.
- Thus, the frequency of this type of LOCA event is very low.
When the very severe nature of the design basis assumptions is considered, the probability of having a
severe LOCA is extremely low.
In recent work for the
- BWROG, GE estimated a conservative large pipe break failure probability of 7.5 x 10~ per reactor year.
NUREG/CR-4792 provides analytical estimates of a BWR large pipe double ended guillotine break as being of the order of j 0-12 The BFN Probabilistic Risk Assessment (PRA) estimates the probability of all main steam line breaks at 1.4 x 10~
occurrences/year, and 4.0 x 10~ occurrences/year for the recirculation line breaks.
Some accident scenarios for shroud weld cracking assume that a seismic event creates forces which induce the LOCA described above.
The E2-8
probabilities associated with this scenario are not well
- defined, but calculations at Browns Ferry estimate the frequency of a Safe Shutdown Earthquake (SSE) at 3.5 x 10~
per reactor operating year.
The likelihood that a LOCA in combination with a SSE would be much less that the individual probabilities identified above.
INSPECTION TIMING The amount of time remaining for the current Unit 2 Cycle 7 power operation is conservatively estimated to be approximately 42 days (August 25 to October 6).
Therefore, the probabilities listed in the previous
- section, which are on an events per year basis, would be reduced by the fraction of a year remaining until the outage (i.e.,
42 365 or 0.115).
Therefore, the likelihood of a LOCA occurring during the remaining time until the outage is considered to be extremely improbable.
CONCLUSIONS The possibility that a
LOCA combined with a 360'reater than 904 deep crack will occur is extremely remote.
It is extremely improbable that a
LOCA and/or a seismic event that will induce a LOCA will occur during the next six weeks.
- Also, TVA concludes that any cracking of the Unit 2 core shroud will be acceptable.
Therefore, there is a reasonable basis to conclude that the combination of events and circumstances that would cause the shroud to be displaced will not occur.
VI~
NUCLEAR SAFETY IMPACTS The "BWR Shroud Cracking Generic Safety Assessment" provides an evaluation of the effect of a postulated separation of the shroud assembly along a horizontal weld during normal operation.
The upward displacement is calculated such that sufficient flow is allowed outside the shroud region for the inside pressure to equalize to the weight of the lifted upper shroud region.
The evaluation shows that the maximum displacement for the various weld locations is not sufficient to disengage the top guide from the fuel channels.
The top guide maintains the top of the fuel assemblies spaced properly.
Therefore, the core arrangement, fuel bundle orientation, and control rod insertability would be maintained, allowing for safe operation and shutdown of the unit.
At weld locations where the 360'hrough-wall crack would result in significant flow through the gap, the condition would be detected by the plant operators using existing instrumentation available for monitoring reactor performance.
Depending on the location of the failed E2-9
- weld, an anomaly indication of thermal power, recirculation loop temperature, asymmetrical
- mismatch, total core flow, pressure across
- core, pump flow to core flow, and/or core power to core flow would allow detection.
After detecting'he
- anomaly, a normal shutdown would be initiated in accordance with. BFN procedures and the anomaly investigated.
Should the gap at the failed weld be insufficient to allow flow anomalies greater than instrument uncertainties, the magnitude of the gap flow would have no impact on plant operation, and would go undetected.
A crack in a weld below the core plate may result in a 1% non-conservative minimum critical power ratio (MPCR) calculation at intermediate power and/or flow conditions, but is not significant due to the MPCR margins at this condition.
Also, the crack growth leading to the 360'hrough-wall flaw is expected to be gradual, and proper water level would be maintained precluding a water level transient.
ANTICIPATED OPERATIONAL TRANSIENTS The "BWR Shroud Cracking Generic Safety Assessment" addresses anticipated operational transients (AOTs) that could increase core shroud loads above those experienced during normal operation.
Some AOTs do not result in an increased load on the shroud and would not increase the shroud upward displacement beyond that during normal operation.
Therefore, the consequences of those AOTs are not affected by the shroud position.
The AOTs that could create additional upward loads on the shroud welds and could result in complete weld separation during the event, or that result in displacement greater than that existing during normal operation if the 360 through-wall crack existed prior to the AOT are:
Pressure regulator failure-open Recirculation control failure increasing to maximum flow Inadvertent actuation of the automatic depressurization system (ADS).
GE's analysis of these events shows that no core safety margins (e.g.,
MCPR, reactor water level, reactor pressure) are violated during AOTs. If the crack existed reactor undetected prior to the AOT, or is created by the additional load during the AOT, it is possible it would continue to go undetected after the AOT.
- However, as discussed
- above, anomalies in the monitored core performance parameters during resumed normal operation that could have an impact on safe operation would be detected.
E2-10
DESIGN BASIS ACCIDENTS This assessment of the potential impact of design basis accidents (DBAs) is based on the "BWR Shroud Cracking Generic Safety Assessment."
This assessment addresses the potential impact that a postulated undetected crack would have on:
The ability to insert the control rods The ability to cool and maintain coverage of the core The performance of the emergency core cooling system (ECCS)
The ability of the standby liquid control system (SLCS) to operate Two DBAs and the Safe Shutdown Earthquake (SSE) are pertinent when evaluating the potential effects of a 360 through-wall crack.
As such, the applicable DBAs addressed in this evaluation are the Main Steam Line Break (MSLB) and Recirculation Line Break (RLB).
The MSLB creates the greatest lifting loads on the shroud head and lower shroud welds, and has the greater potential to cause shroud weld failure and shroud displacement.
The RLB actually reduces the lifting loads on the shroud head, but has the greatest potential to cause fuel damage if excessive flow is possible through cracks.
Shroud cracks impact the plant's response to the MSLB or RLB only to the extent additional ECCS flow is needed to compensate for the through-wall leakage.
The SLCS is assumed effective if injection is possible.
The ability to insert control rods is assured if the fuel remains properly arranged in the core, and the guide tubes and shroud remain aligned.
A eoolable core geometry is maintained if the fuel and vessel internals do not impede the normal flow of coolant to the fuel.
Adequate core coverage is maintained if the minimum 2/3 core coverage is possible.
The primary factor in determining the impact is the movement of the shroud during the DBA or SSE.
Main Steam Line Break The "BWR Shroud Cracking Generic Safety Assessment" addresses the impact of the additional load on the shroud head caused by the rapid depressurization of a MSLB inside primary containment.
The assessment shows that acceptable results are obtained for BWR/4s such as BFN.
The shroud is shown not to liftthe top guides above the fuel channels.
Thus, the ability to insert the control rods should not be impacted.
- However, even if the control rods were damaged, the SLCS would remain operational to shutdown the reactor.
E2-11
The high pressure difference across the shroud is only expected to last a few seconds, then the shroud would rest in place with only a tight gap at the 360'hrough-wall crack.
Even if the Core Spray (CS) header is contacted by the uplifting shroud and damages the spargers, the ability to inject water to the core, without the spray function, would be maintained.
Therefore, the ability to mitigate the MSLB with the ECCS and keep the core flooded above the top of active fuel (TAF) is not affected.
Additionally, the ECCS would function to provide long-term shutdown cooling capability.
BFN Emergency Operating Instructions (EOIs) address operator action for use of the SLCS to inject boron if the ability to insert control rods is affected.
Also, the EOIs address vessel injection using any available source, and is not limited to the use of only the ECCS.
Based on the above, engineered features and operator actions provide sufficient defense in depth to ensure mitigation of the consequences of the MSLB.
Recirculation Line Break The BWROG's "BWR Shroud Cracking Generic Safety Assessment" addresses the impact of the load on the shroud caused by the rapid depressurization and blowdown of a RLB.
The evaluation shows acceptable
- results, concluding upward forces on the shroud head are reduced during the event and any existing crack would be reduced to a tight gap with minimal flow.
The BWROG assessment shows that the shroud is not lifted during the event.
Therefore, the CS spargers are not
The tipping force caused by the blowdown through the recirculation line would be bounded by the restoring moment of the shroud weight and only minimal rotation would occur before returning to normal position.
The core geometry would be maintained and the ability to insert the control rods would be maintained.
Therefore, the ability to mitigate the consequences of the RLB is not affected.
If significant shroud movement was caused by the blowdown load, it is postulated that shroud bypass leakage could exceed the capability of the ECCS and prevent two-thirds core coverage, although the core spray function would be maintained.
However, the occurrence of significant shroud bypass leakage would require several occurrences, including a 360'hrough-wall crack, a RLB, and resulting loads from the RLB being large enough to cause a
significant opening in the core shroud.
All of these conditions are considered to be highly unlikely.
The BFN EOIs address vessel injection. using any available
- source, and is not limited to the use of only the ECCS.
The worst case scenario would require flooding of the primary containment to ensure core coverage.
The EOIs prescribe operator actions for primary containment flooding that would be used for this postulated scenario.
Therefore, engineered features and operator actions provide sufficient defense in depth.
This ensures that the consequences of an RLB in combination with the highly unlikely conditions and events leading to this scenario can be mitigated.
Safe Shutdown Earth uake The combination of a simultaneous MSLB or RLB and an SSE are not considered to be credible scenarios.
- However, as described in the "BWR Shroud Cracking Generic Safety Assessment,"
these combinations of events were assessed.
The MSLB in combination with an SSE presents the worst case load combination with regards to impact on shroud displacement.
The combination of loads during an MSLB and SSE can induce shroud displacements greater than that discussed above for an MSLB.
The maximum estimated vertical and lateral displacement during an MSLB with SSE would be less than one inch.
Any rotation or tipping of the shroud assembly due to lateral motion during a seismic event is also expected to be less than one inch.
These displacements will not significantly alter the conclusions reached earlier and discussed above.
-Under postulated simultaneous DBA with SSE events, the top guide will still maintain proper fuel assembly alignments and small lateral, momentary movement or tipping will not prevent control rod insertion.
Therefore, if the core shroud were cracked and a
DBA coupled with an SSE were to occur, the core shroud displacement would not be sufficient to affect the ability to achieve and maintain safe shutdown.
EMERGENCY OPERATOR ACTIONS BFN operators are trained to use the EOIs when plant conditions warrant their use.
The EOIs are symptomatic and provide a flowpath of operator actions based on observed symptoms.
The EOIs do not rely on the operators being able to diagnosis a particular event.
The EOIs address a wide range of events from expected plant excursions to beyond design basis events.
The EOIs do not depend on the successful performance assumed in the design basis
- analyses, therefore, they do not lose effectiveness if actual events vary from the design basis assumptions.
E2-13
Reliance on symptomatic procedures provides reasonable assurance of proper operator action should excessive shroud cracking occur, even though the operator may not realize the shroud crack is contributing to the observed symptoms.
PROBABILISTIC RISK PERSPECTIVE The probability of a DBA occurring during any reactor year is extremely low.
The Unit 2 reactor is scheduled to shutdown for a refueling outage no later than the first week of October 1994.
Therefore, the likelihood of any event occurring in the period between the submittal of this assessment and shutdown is approximately 0.115 (i.e.,
42 days
'365 days/year) times the probability per reactor year.
The BFN specific probabilities are:
EVENT ANNUAL PROBABILITY 6-WEEK PROBABILITY MSLB RLB SSE MSLB&SSE RLB&SSE 1.4 x 10 4.0 x 10 3.5 x 10s 4.9 x 10~
1.4 x 10 1.6 x 10~
4.6 x 10~
4.0 x 10 5.6 x 10>o 1.6 x 10 The individual probabilities associated with each of the events described above are extremely low.
Therefore, TVA considers that the risk associated with continued operation until inspections are performed is acceptable.
VII.
CONCLUSIONS The BFN design features, engineered safeguards,
- EOIs, and the nature of the postulated failure itself combine to show that continued operation of Unit 2 is acceptable.
The probability that the conditions and events occurring at some point in time simultaneously prior to the Based on the extremely low individual probabilities associated with each of the accidents combined with the highly unlikely probability that Unit 2 has a 360',
>904 deep crack, TVA considers that it is acceptable to continue operating until shroud inspections are performed during the Cycle 7 refueling outage.
E2-14
ENCLOSURE 3
TENNESSEE VALLEY AUTHORITY BROWNS FERRY NUCLEAR PLANT (BFN)
UNIT 3 BFN UNIT 3 REACTOR CORE SHROUD ZNSPECTZON RESULTS AND ANALYSIS INSPECTION
SUMMARY
TVA and GE inspected 100% of the accessible areas on Unit 3 core shroud welds Hl through H7.
The inspections were performed from June 13 to July 14, 1994, using the best available technology.
The examinations were conducted by qualified personnel in accordance with approved BFN procedures.
Inspection procedures were prepared to implement the applicable portions of ASME Section V and XI recommendations, the NRC/EPRI/BWROG Coordination Plan, and General Electric (GE) Service Information Letter (SIL) 572.
The full inspection report is available on-site for review.
Ultrasonic examinations (UT) were performed using the GE Smart-2000'ystem and suction cup scanners.
Enhanced visual examinations (VT-1) were performed using the Westinghouse Model ETV-1250 camera with the Twin-50's light and lens.
UT was performed on 1004 of the accessible areas on all seven welds.
The VT-1 visual inspections were performed in selected areas.
Visual examinations were performed on selected attachments between welds H3 and H5, and at predesignated ten-inch segments at 10'nd 190'n welds H6 and H7.
Personnel performing the examinations were certified to at least Level II status in accordance with SNT-TC-1A, 1984 Edition.
Additionally, personnel performing UT examinations were qualified through the EPRI NDE Center in accordance with the Coordination Plan for NRC/EPRI/BWROG Training and Qualification Activities of NDE Personnel.
Accessibility for the inspections was limited due to various equipment or internal structures that restrict access to the welds.
The proximity of various components such as guide pins, lifting lugs, core spray downcomers, shroud head locking lugs, jet pump riser braces, and jet pumps precluded further examination.
Figure E3-1 shows a
"roll-out" of the areas inspected.
The following provides a summary overview of the amount of each weld TVA examined:
TVA indicated in the August 11, 1994, meeting with NRC that this system was called the "GERIS-2000" system.
The actual name of the system is "Smart-2000" system
WELD NUMBER Hl H2 H3 H4 H5 H6 H7 CIRCUMFERENCE EXAMINED INCHES 284.16" 470.40" 544.81" 439.83 266.07" 28.00" 28.00 PERCENTAGE EXAMINED 41.14 68.14 83.74 67.64 40.94 4.44 4.44 II INSPECTION RESULTS There were no indications identified during the VT inspections.
During UT inspections, surface connected planar flaws and/or Intergranular Stress Corrosion Cracking (IGSCC)/Irradiation Assisted Stress Corrosion Cracking (IASCC) indications were found in three welds (H1, H4, and H5).
The flaws were dispersed around the circumference of each weld.
There were no through-wall cracks identified.
A summary of the BFN Unit 3 UT inspection results is provided below:
WELD NUMBER H1 H2 INSPECTION TYPE UT UT RESULTS 5 indications (3.93" total length)
No reportable indications FLAW TYPE IGSCC N/A H3 UT No reportable indications N/A H4 H5 H6 UT UT UT
& VT 1 indication (1.81" total length) 22 indications 74.58" total length No reportable indications Planar Planar (3)
IGSCC/IASCC (19)
N/A H7 UT
& VT No reportable indications N/A The Examination Summary Sheets for shroud welds H1 through H7 are provided on pages E3-4 through E3-13.
E3-2
FIGURE E3-1 UNIT 3 INSPECTION RESULTS SHADED AREAS INDICATE LOCATIONS NOT SCANNED Hl 3 0'50'40'30'20'10'00'90 280'70'60'50'40'30'20'10 200'90'80'70'60'50'40'30 120'10'00'0'0'0'0'0'0 30 c>>'0
l I
I I
I I
I I
I I
~I I
H2 H3 360'50 340 330'20 310 300'90'80'70'60'50'40 230 220 210'00'90 180 170'60'50'40'30 120'10'00'0'0'0'0'0'0'0'0'0' I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
H4 H5 H6 360 350'40'30 320'10'00'90'80'70'60'50'40'30'20'10'00 190'80'70 160'50'40'30'20 110'00'0'0'0 60 50 40'0'0'O
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I I
I H7 E3-3
.,,:':,::::;::;,.,"::::::,:;,',:::,F,:,'::.:,:":::::;:::;:."GE::Niicleai;:Eiiergy:
EXAMINATION
SUMMARY
SHEET
~5~9@
PROJECT:
B 0 UC PROCEDURE:
REV~ FRR ~f N/A SYSTEM'ELD NO.:
CONFIGURATION:
EXAMINER'.CONTI EXAMINER:
EXAMINER:
S 0
G LEVEL:~
LEVEL:~
LEVEL:ME WELDTYPE:
REV~ FRR;~
~l REV:~l FRR:~
~l QMT QPT IIUT QVT Q CIRCUMFERENTIAL Q LONGITUDINAL II OTHER DATASHEET NO.(S):
CALSHEET NO.(S):
During the examination of the referenced weld, five (5) indications associated with IGSCC/IASCC were recorded by the Smart 2000 system utilizing 45'hear wave and 60'efracted longitudinal wave search units. These indications have the following parameters:
Indication Number 1
2 3
4 5
Distance Fram Lo 84.0'161.3" 96.7'185.7" 149.7'287.3" 276.2'530.4" 277.3'532.4" Total Length
'36'/
.69"
.66'1.27"
.31'/
.60"
.41'/
.79"
.30'/
.58" Remaining Ligament 1.60" 1A4" 1.93" 1.84" 1.60" Thruwafl Dimension"
.56"
.07"
.16" Side of Weld Lower Lower Lower Lower Lower Type Reflector IGSCC/IASCC IGSCC/IASCC IGSCC/IASCC IGSCC/IASCC IGSCC/IASCC Search Unit 60'5'60'5'5'5'/60' Length sizing for the five (5) indications was determined by using 50% drop points with respect to each indication's maximum amplitude point.
- The throughwall dimension for each indication was determined with the tip diffraction technique using the absolute arrival time sizing method.
The 45'hear also recorded non-relevant indications from both sides of the weld; as well as beam redirect, inside surface geometry, inside and outside surface weld crown geometry, along with four (4) of the previously recorded indications (2, 3, 4, & 5) from the lower side ofthe weld.
The 60'L also recorded non-relevant indications from both sides of the weld, along with three (3) of the previously referenced Indications (1, 2, & 5) from the lower side of the weld.
This examination was also limited to "L"dimensions of 28'o 32; 35'o 39', 43'o 47', 50'o 54', 58'o 62, 65'o 69', 73'o 77', 80'o 84',
88'o 92', 95'o 99', 103'o 107', 110'o 114', 118'o 122', 126'o 130', 134'o 138', 141'o 145', 148'o 152', 156'o 160' 200'o 204',
208'o 212', 215'o 219', 223'o 227', 230'o 234', 238'o 242', 245'o 249, 253'o 257', 260'o 264', 268'o 272', 275'o 279', 283'o 287',
290'o 294', 298 to 302', 305'o 31 0', 313'o 31 7', 320'o 325', 328'o 332', and 335'o 340'rom vessel '0'ue to the proximityof the shroud head locking lugs and core spray downcomers.
Circumferential "L"dimensions for afl examination scans were recorded in angular units in lieu of linear units. The conversion factor for circumferential measurements is 1.92" per degree.
SUMMARY
B G
EVIEWED BY
~t LEVEL DATE 2C LEVEL DATE GEIND PE DENT EW UTILITYREVIEW BY E3-4 4r -Ao-t DATE DATE ANBREVIEWBY DATE PAGE:~OF'~
FORFIR4IREV E
a a L t
~
~
~
~
I
~
t I
~
I
~
~ ~ ~
~
~
~
I
~
~
~
~
~
~
~
~
~
~ '
~
~
~
t
~
~
~
~
~ '
~
~
~
~
~
~ ~
~ '
~,
~
~
~
~
~
~
~ '
~ ~ '
~
~
~
~
~
~
~ ~
~ '
~
~ ~
\\ '
t
~
~
~
~
~
~
t
~ '
~ ~ '
~
t t
~
~
~
I
~
~
~
~
~
~ '
~
~
~
t ~',
~ ~
~
~ '
~
~
~ '
~
~
~
~
k ~
L L k k L
a k L
~ ~ ~
~
)
~
~
k
~ )
~
~
aL k
L a ~
a%i ~ ~ ~
L
~
~
o
~
~ ) ~
~
)
~
~
~ I l I
~
~
)
~
~ I ~
~
~
P
~
~
~
~
~ I
~
~
~
~
~
~
~
~ ~,
~
~
P ~
~ '
~
~
~
~
~
~
~
~
~
~
~
~ '
~
~
~
~
~
~ '
~
~
~ I
~
~
~
~
~
~
~
~
~ ~
~
~
~ '
~
~
~
~
~ '
~
~
~ ~
~
S
~
~
~
+
~ ~
~
~
~
~
~
~
~
~
~ I
~ '
~
~
+
~
4
~
I ~:
~
~
~
~
~
~
~
~
~
~ ~',
~ ~
~'
~
~
. r..
~
~.
~
~
~
~ ~ ',
~
~
~ '
~
EXAMINATlON
SUMMARY
SHEET
'E:Nuclear Energy REPORT NO.:
Z PROJECT:
0 S
SYSTEM' OU S
WELD NO.:
UCL Il'PROCEDURE:
REV~ FRR:~
~l REV~ FRR:~
~1 CONFIGURATION'XAMINER' CON I
EXAMINER:
EXAMINER:
LEVEL~
LEVEL:~
LEVEL:~h WELDTYPE:
R~~I FllR:~
~f QMT QPT g UT Q CIRCUMFERENTIAL Q LONGITUDINAL II OTHER DATASHEET NO.(S):
CALSHEET NO.(S):
During the examination of the referenced weld, one (1) inside surface connected planar flaw indication was recorded by the Smart 2000 system utilizing 45'hear wave and 60'efracted longitudinal wave search units.
This indication has the following parameters:
Indication Number Distance From Lo Total Length' 253.4'/458.7" 1.0'/
1.81" Remaining Ligament 1.43" Thruwall Dlmension-Side ofWeld Type Reflector In Weld Planar Search Unit 45'60' Length sizing for this indication was determined by using 50% drop points with respect to each indication's maximum amplitude point.
- The throughwall dimension for this indication was determined with the tip diffraction technique using the absolute arrival time sizing method.
The 45'hear also recorded non-relevant indications, welding discontinuities, inside surface geometry, inside and outside surface weld crown geometry from both sides of the weld, along viith acoustic interface from the upper side of the weld, as well as the previously referenced inside surface connected planar flaw indication from the lower side of the weld.
The 60'L also recorded non-relevant indications, shear component, weld discontinuities and inside surface weld crovm geometry from both sides of the weld, along with acoustic interface from the upper side of the weld, as well as inside surface geometry and the previously referenced inside surface connected planar flaw indication from the lower side of the weld.
This examination was limited to "L"dimensions of54'o 161'nd 197'o 333'rom vessel'0'ue to the proximityof the guide pins, liflinglugs, and core spray downcomers.
Circumferential "L"dimensions for all examination scans were recorded in angular units in lieu of linear units. The conversion factor for circumferential measurements is 1.81" per degree.
SUMMARY
BY LEVEL DATE GE R IEWED BY LEVEL DATE GE INDEPENDENT REVIEW Y
UTILITYREVIEW BY E3-7 C "47-q DATE ANIIREVIEWBY DATE PAGE:~OF:~
SCRMVf40$tV 2
,.-..":i:':.'"'::::::::.:" 'l" i:::::::.::::;:GC.-'Nucle'ai'.Eri'eigji',.:':,
EXAMINATION
SUMMARY
SHEET BQ PROJECT:
0 SYSTEM'ELD NO.:
CONFIGURATION:
EXAMINER: A CONTI EXAMINER'XAMINER:
V S
LEVEL:~
LEVEL:~
LEVEL:~l PROCEDURE:
WELDTYPE:
REV:~ FRR: ~l N/A
~l REV~ FRR:~
~l REV:~l FRR:~
~f I-IMT OPT IIUT OVT 0 CIRCUMFERENTIAL ii DATASHEET NO.(S):
CAL SHEET NO.(S):
During the examination of the referenced weld, three (3) inside surface connected planar flaw indications and fifteen (15) indications associated viith IGSCC/IASCC were recorded by the Smart 2000 system utilizing45'hear wave and 60'efracted longitudinal wave search units.
These indications have the following parameters:
Indication Number 1
2 3
4 5
6 7
8 9
10 11 12 13 14 15 16 17 18 Distance From Lo 71.3'/1 29.1" 72.5'/131.2" 103.0'/1 86.4" 106.1'/1 92.0" 107.2'/1 94.0" 109.6'/1 98.4" 111.5'/201.8" 115.5'/209.1" 118.7'/214.9" 121.5'/219.9" 127.7'/231.1" 135.2'/244.7" 137.7'/249.2" 140.8'/254,9" 246.2'/445.6" 269.2'/487.3" 272.1'/492.5" 311.5'/563.8" Total Length'4'
.72"
.6' 1.09" 2.3'I 4.16"
.7' 1.27"
.9' 1.63" 1.6' 2.99" 2.7' 4.89" 1.6' 2.90" 1.4' 2.53" 1.1' 1.99"
.7' 1.27"
.5'
.91" 2.0' 3.62"
.4'
.72" 17.8' 32.22"
.6'/ 1.09"
.9' 1.63" 1.0' 1.81" Remaining Ligament 1.68" 1.85" 1.68" 1.78" 1.81" 1.79" 1.76" 1.69" 1.68" 1.80" 1.69" 1.7r 1.70" 1.81" 1.38" 1.64" 1.32" 1.47" Thruwall Dimension-
.3r
.15"
.3r
.22"
.19"
.21"
.24"
.31"
.3r
.20"
.31"
.28"
.30"
.19"
.62"
.36" 6S"
.53" Side ofWeld In Weld In Weld Upper Upper Upper Upper Upper Upper Upper Upper In Weld Upper Upper Upper Upper Upper Upper Upper Type Reflector Planar Planar IGSCC IGSCC IGSCC IGSCC IGSCC IGSCC IGSCC IGSCC Planar IGSCC IGSCC IGSCC IGSCC IGSCC IGSCC IGSCC Search Unit 45'/60'5'60'5'/60'5'60'5'/60'5'/60'5'/60'5' 60'5'60'5'60'5'60'5'/60'5'/60'5'/60'5'/60'5'5'/60'5'/60'ength sizing was determined by using 50% drop points with respect to each indication's maximum amplitude point.
- The throughwall dimension was determined viith the tip diffraction technique using the absolute arrival time sizing method.
The 45'hear also recorded non-relevant indications from both sides of the weld, along with acoustic interface, welding discontinuities, inside and outside surface weld crown geometry, as well as the eighteen (18) previously referenced inside surface connected Indications from the upper side of the weld.
The 60'L also recorded non-relevant indications from both sides of the weld, along with shear component and seventeen (17) of the previously referenced inside surface connected indications from the upper side of the weld.
This examination was limited to "L"dimensions of 66'o 75', 81 to 88', 96'o 143', and 246 to 330'rom vessel 'ty due to the proximityofjet pump riser braces, guide pins, liftinglugs, and core spray downcomers.
Circumferential "L"dimensions for all examination scans were recorded in angular units in lieu of linear units. The conversion factor for circumferential measurements is 1.81" per degree.
,.z.
LEVEL DA GE INDEPENDENT REVIEW BY UTIUTYREVIEW DATE DATE ANIIREVIEW BY DATE PAGE:~OF:~
FCFRF VF4I FCV 7 E3-8
I I EXAMINATlON
SUMMARY
SHEET GE Nuclear Energy REPORT NO.:
MKQ9 PROJECT:
BROWNS FERRY NUCLEAR PROCEDURE'EV~ FRR;~
SYSTEM:
WELD NO.'
CONFIGURATION:
EXAMINER'XAMINER:
EXAMINER:
LEVEL:III LEVEL'M~
LFVEL~~
WELD TYPE REV~ FRR;~
REV~ FRR:~
QMT OPT INUT QVT C3 CIRCUMFERENTIAL 0 LONGITUDINAL N OTHER DATASHEET NO.(S):
CAL SHEET NO.(S):
During the supplemental examination of the referenced weld, four (4) inside surface connected indications associated with IGSCC/IASCC were recorded by the Smart 2000 system utilizing 45'hear wave and 60'efracted longitudinal wave search units. These indications have the following parameters:
Indication Number Distance From Lo Total Length Remaining Ligament Thruwall Dimension-Side of Weld Type Reflector Search Unit 1
15.12'/27.37" 1.04'/1.88" 1.64" 0.36" Upper IGSCC 45'/60' 16.82'/30.44" 0.60'/1.09" 1.63" 0.37" Upper IGSCC 45'/60' 191.00'/345.71R 0.44'/0.80" 1.80" 0.20" Upper IGSCC 45' 194.26'/351.61" 1.86'/3.37" 1.66" 0.34" Upper IGSCC 45'/60'ength sizing was determined by using 50% drop points with respect to each indication's maximum amplitude point.
- The throughwall dimension was determined with the tip diffraction technique using the absolute arrival time sizing method.
The 45'hear also recorded non-relevant indications from both sides of the weld, along with beam redirect, inside and outside surface weld crown geometry and the four (4) previously referenced inside surface connected indications from the upper side of the weld.
The 60'L also recorded non-relevant indications from both sides of the weld, along with shear component and three (3) of the previously referenced inside surface connected indications from the upper side of the weld.
This supplemental examination was performed in selected areas previously inaccessible using the "OD Tracker" scanning device. The ultrasonic examination was performed from approximately 11'o 19'nd 191'o 199'tilizing the "Suction Cup" scanning device. Circumferential "L" dimensions for all examination scans were recorded In angular units in lieu of linear units. The conversion factor for circumferential measurements is 1.81" per degree.
Reference previous examination summary (Report No. R-S94-08) for complete examination summary.
SU GE REVIEWED BY LEVEL DATE LEVEI.
E3-9 DATE p-'~5'ATE ANIIREVIEW DATE PAGE~OF:~
FOIMVFEEREV.E
~
R
/
F EXAMINATION
SUMMARY
SHEET GE Nuclear Energy REPORT NO.:
HL PROJECT BROWNS FERRY NUCLEAR PROCEDURE'RV~ FRR:~
SYSTEM'ELD NO.:
I CONFIGURATION:
N RRV~ FRR:~
A RRV~ FRR:~
EXAMINER:
EXAMINER:
EXAMINER:
LEVEL:~
LEVEL:~lb LEVEL.~/~
WELDTYPE:
OMT OPT IIUT QVT Cl CIRCUMFERENTIAL Q LONGITUDINAL W OTHER DATASHEET NO.(S)
CAL SHEET NO.(S):~
During the ultrasonic examination of the above referenced weld, no surface connected planar flaws or any indications associated with IGSCC/IASCC were recorded by the Smart 2000 system utilizing45'hear wave and 60'efracted longitudinal wave search units.
The 45'hear and 60'L search units did record non-relevant indications from the lower side of the weld.
No examination was performed from the upper side of the weld due to the component configuration. The examination from the lower side of the weld was limited due to the proximity of the outside diameter filletweld. This examination was also limited to "L"dimensions of approximately 11'o 19'nd 191'o 199'rom Vessel R0" due to the proximity of guide pins, lifting lugs, core spray downcomers, jet pumps and their associated braces and restraint brackets.
Circumferential "LR dimensions for all examination scans were recorded in angular units in lieu of linear units. The conversion factor for circumferential measurements is 1.75" per degree.
SU MARY LEVEL DATE GE R IEWED BY LEVEL DATE GE QC REVIEW UTIU ND REVI E3-10 FATE
~~X DATE ANIIREVIEW DATE PAGE~OF:~
FORFT VT4F RCV. E
...,:::.;::::i.:::;:,';:-:;::;::::"::,"":;'"'-'":GE,hfucle'ai Eiie'igy:::::
EXAMINATION
SUMMARY
SHEET PROJECT:
0 UC SYSTEM:
WELD NO.:
1 CONFIGURATION:
PROCEDURE:
REV:~ FRR: ~f N/A
~l REV~ FRR:~
REV:~ FRR:~
EXAMINER: J BRIGGS EXAMINER; EXAMINER:
LEVELt~
LEVEL'~lb WELDTYPE:
QMT OPT QUT IIVT IICIRCUMFERENTIAL 0 LONGITUDINAL 0 OTHER DATASHEET NO.(S):
CALSHEET NO.(S):
During the visual examination of the referenced weld, no recordable indications were detected.
LEVEL DATE
SUMMARY
B~
LEVEL DATE GE INDEPENDEIIt)REVI BY UTILITYREVIEW8 d~a-~ /
DATE ANIIREVIEW BY DATE PAGE:~OF:~
FORM VF40 REV, t
AF Ii EXAMINATION
SUMMARY
SHEET GE Naclear. Energy REPORT NO.:
&1L PROJECT:
BROWNS FERRY NUCLEAR PROCEDURE'EV~ FRR:~
SYSTEM:
WELD NO.:
I CONFIGURATION'XAMINER:
EXAMINER:
EXAMINER'EVEL:~
LEVEL:MIB WELD TYPE REV~ FRR:~
REV~ FRR:~
DMT CIPT N UT QVT 0 CIRCUMFERENTIAL D>>
~
DATASHEET NO.(S):
CALSHEET NO.(S):
During the ultrasonic examination of the above referenced weld, no surface connected planar flaws or any indications associated with IGSCC/IASCC were recorded by the Smart 2000 system utilizing 45'hear wave and 60'efracted longitudinal wave search units.
The 45'hear did record non-relevent indications, beam redirect, inside surface geometry, and inside surface weld crown geometry from the lower side of the weld.
The 60'L search unit recorded non-relevant indications and shear component from the lower side of the weld.
No examination was performed from the upper side of the weld due to the component configuration. The examination from the lower side of the weld was limited due to the proximity of the outside diameter backing ring and weld build-up area.
This examination was also limited to "L" dimensions of approximately 11'o 19'nd 191'o 199'rom Vessel "0" due to the proximity of guide pins, lifting lugs, core spray downcomers, jet pumps and their associated braces and restraint brackets.
Circumferential "L"dimensions for all examination scans were recorded in angular units in lieu of linear units. The conversion factor for circumferential measurements is 1.75" per degree.
S MARYBY GE R EWED BY LEVEL DATE LEVEL DATE GE QC REVIEW UTI ITYNDER EW E3-12 DATE MZc DATE ANIIREVIEW DATE PAGE~OF:
EI FEIEI VTEE EEV. E
~~, ~
i
,,:,"',.,"::;::::,:I::::::F:::I::::::::::::-.:.:';:::::GENiiciea'r. Eri er'gy,':::
EXAMINATION
SUMMARY
SHEET PROJECT:
8 0 S
U PROCEDURE:
REV:~ FRR;~
N/A SYSTEM:
WELD NO.:
CONFIGURATION:
C G
REV~ FRR:~
~f REVE/ FRR:~
EXAMINER: J BRIGGS EXAMINER'XAMINER:
LEVEL:~
LEVEL:~$
LEVEL:~/~
WELDTYPE:
ClMT Cl PT g UT IIVT S CIRCUMFERENTIAL 0 LONGITUDINAL 0 OTHER DATASHEET NO.(S):
CALSHEET NO.(S):
During the visual examination of the referenced weld, no recordable indications were detected.
GE R IEWED LEVEL DATE
/i~
SUMMAR LEVEL DAT GE INDEPENDENT REVI FBY DATE r
UTILITYREVI Y
DATE ANIIREVIEW BY DATE PAGE:~OF:~
FOES VS@4 REV l E3-13
...:,;:::;:;:::,:;,:::,:::,:.::,;:,,:::::::;:,.",,;::::::GF:.Niiclea'r..:Eri erLgy-::
EXAMINATION
SUMMARY
SHEET REPORT NO.:
PROJECT:
0 S
PROCEDURE:
REV~ FRR:~
N/A SYSTEM:
WELD NO.:
CONFIGURATION:
EXAMINER: J, BRIGGS EXAMINER:
/
EXAMINER:
0 G
LEVEL:~
LEVEL:m~
LEVEL:MS WELDMPE:
REV~ FRR:~
REV:~ FRR:~
~l OMT OPT OUT IIVT IICIRCUMFERENTIAL O LONGITUDINAL O OTHER DATASHEET NO.(S):
CALSHEET NO.(S):
During the visual examination of the referenced weld, no recordable indications were detected.
SUMM Y B GE REVIEWE Y
LEVEL DATE LEVEL DATE GE INDEPENDENT REVIEW BY UTILITYREVI BY E3-14 DATE DATE ANIIREVIEW BY DATE PAGE:~OF:
3 FERMNCOREV E
XXX SAFETY ANALYSXS
SUMMARY
TVA and GE performed an analysis of the core shroud cracks identified in Unit 3 to show that restart and resumption of operation for at least one cycle would be acceptable.
The analysis was based on the fracture mechanics limit load based screening criteria and evaluation techniques applicable to BFN.
The analysis reports used for these assessments are available on-site for review.
The screening criteria establishes the allowable flaw lengths for the various girth and axial welds on the core shroud.
The evaluation techniques provide guidance for evaluating inspection results.
The screening criteria and evaluation techniques are conservative and bound the BFN inspection results since they are based on the presumption that only visual inspections will be conducted.
As such, allowable flaw lengths were established assuming that all flaws would be through-wall.
Only the H5 weld had any significant flaw indications so it is the only weld that was evaluated.
The remaining welds were determined to be acceptable since no indications were found or the analysis for the H5 weld bounds their condition.
The H5 weld is in a low fluence area (i.e.,
below 3. 0 x 10 n/cm )
so Linear Elastic Fracture Mechanics (LEFM) evaluation techniques are not needed.
Therefore, the total flaw length used for determining the acceptability of postulated flaw growth is 416 inches, not to exceed 104 inches in any quadrant.
TVA examined approximately 414 of the H5 weld.
In the weld length examined, 264 was found to be cracked.
TVA considers that any cracks in the unexamined portions of the weld would be similar to those found (e.g.,
26% of the unexamined weld was assumed to be cracked).
The deepest crack had a depth of 0.68 inches and a length of 1.63 inches.
The longest continuous crack was 32.2 inches with a maximum depth of 0.62 inches.
An evaluation of the H5 weld was performed to conservatively estimate the extent that the cracking may propagate during operation.
ASME Section IX proximity rules were used in evaluating the data since flaw characteristics were determined by UT.
Two methods were used to distribute postulated flaws in the unexamined portion of the H5 weld.
First, the entire postulated flaw length was assumed to be adjacent to the nearest observed flaw (single indication method).
- Second, the postulated length was divided into smaller postulated flaws that were evenly distributed within the unexamined area (distributed indication method).
The length of the E3-15
postulated flaws in the unexamined weld area was based on the average length of the observed flaws.
To account for the uncertainty in depth sizing by UT, TVA added 0.3 inches to the flaw depths.
The flaw growth rate was estimated using conservative values that have been accepted by NRC (5.0 x 10~ inches/hour of hot operation).
Calculations were performed using both evaluation methods to determine if the postulated flaw sizes would meet the acceptance criteria through an operational cycle (assumed to be 12,500 hot operating hours).
The results of the evaluation are shown below:
Flaw T e
S inci le Indication Method Results Distributed Indication Method Results Total postulated flaw
- length, inches 211 234 Allowable total flaw
- length, inches Safety margin, percent Maximum postulated flaw length per quadrant, inches 416 49 84 416 82 Allowable flaw length per
- quadrant, inches Safety margin, percent 104 19 104 21 E3-16
A
~ l
~4
~'
ENCLOSURE 4
TENNESSEE VALLEY AUTHORITY BROWNS FERRY NUCLEAR PLANT (BFN),
UNITS 1g 2q AND 3 LIST OF COMMITMENTS TVA will inspect 100% of the accessible areas on core shroud welds H1 through H7.
The inspections will be conducted using the best available technology (i.e.,
Smart-2000
- system, suction cup scanners, Westinghouse 1250
- camera, etc.) per the following schedule:
Unit 1 the inspection will be completed prior to restart.
Unit, 2 the inspection will be conducted during the Unit 2 Cycle 7
(U2C7) refueling outage.
2 ~
A structural margin analysis of the core shroud inspection results discussed in item 1 above will be performed to determine if the plants can resume operation without repair 3 ~
4 ~
Subsequent core shroud examinations will be performed as described in GE Service Information Letter (SIL)-572 or until repairs are made.
TVA will evaluate the need for inspecting the vertical core shroud welds, and the shroud support plate and leg welds (i.e., welds H8 through H11) for possible inclusion in future refueling outages.
The evaluation will be based on BWROG and GE recommendations.
5.
TVA will develop plant-specific core shroud repair procedures that are based on BWROG repair plans determined to be acceptable by NRC.
6.
The Units 1 and 2 core shroud inspection results discussed in Item 1 above will be provided within 30 days of the completion of the core shroud inspections.