ML17053C002
| ML17053C002 | |
| Person / Time | |
|---|---|
| Site: | Nine Mile Point  |
| Issue date: | 08/01/1980 |
| From: | Dise D NIAGARA MOHAWK POWER CORP. |
| To: | Robert Carlson NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| Shared Package | |
| ML17053C003 | List: |
| References | |
| 10CFR-050.55E, 10CFR-50.55E, NUDOCS 8010150041 | |
| Download: ML17053C002 (166) | |
Text
LN 7 NlAOARA LN NlAQARAMOHAWICPOWER CORPORATlON/300 ERIE BOVLEVAROWEST, SYRAcuSE. N.Y. l3202/TELEPHONE (315) 474 l5ll August I, 1980 Office of Inspection and Enforcement Region I Attention:
Mr. R. T. Carlson, Chief Reactor Construction and Engineering
, Support 8ranch U. S. Nuclear Regulatory Comiission 631 Park Avenue, King of Prussia, PA 19406 Oear Mr. Garison:
Re: Nine Mile Point Unit 2
Oocket No. 50-410'.
Enclosed is the final report concerning the Nine Mile Point Unit 2 biological shield wall in accordance with 10 CFR 50, Paragraph 50.55( e) (3).
This ma ter was initially reported to your staff on May 30, 1979 as involving a defect in the base ring to outer shell plate weld.
Investigations subsequent to the initial notification revealed that there was a potential deficiency in other welds of similar geometrical configuration.
The increased scope of the biological shield wall review was reported to your staff on August 24, 1979.
After extensive engineering evaluation, it has been determined that the condition of the biological shield wall welds could not have adversely affected the safety of operations of the Nine Mile Point Unit 2 plant had it remained undiscovered.
The basis for this conclusion is discussed in detail
-in Section VI, Analysis of Safety Implications, of the enclosed report.
Also included in the report is a description of.the deficiencies and the corrective action taken.
Very truly yours, NIAGARA MOHAWK POWER CORPORATION Oonald P. Oise Vice President Engineering Oirector Office of Nuclear Reactor Regulation U. S. Nuclear Regulatory Commission
- ington, O.C.
20555 Rash 80>Ox 5 00 tl PEF:ja Enclosure xc: Oirector of Inspection and Enforcement U. S. Nuclear Regulatory Commission Washington, O. C.
20555
TABLE OF CONTENTS
~Pa a No.
I~
IHTRODUCTXON B
C Statement of Problem and Summary of Coaclusioas Overvie~ of Report Revisions to Interim Report DESCRIPTXON OF BIOLOGXCAL SHIELD WALL A
Bo Ce Di Eo F~
Go Ho Physical Descriptioa Fuactioaal Requirements Summary of'esign Criteri'a aad Liceasiag Commitmeats Summary of Analytical Techniques
'ummary of Stresses for Design Loadiags Summary of Fabrication and Erectioa Strategy Specification Requirements Qh. Program Requirements 2,
2 3
4 5
6 6
IXI+
. DESCRZPTIOH OF'ROBLZH A
Bi Statement of Problem Metallurgical Discussion 9
~
10 EHGXHEERZNG EVALUATIONAHD CORRECTIVE ACTION 12 ka Bi C+
Di Introduction Overall Approach 1
Stress Analysis 2~
Fracture Mechanics Xaaer Wall to Stiffener Evaluation 1
Approach 2~
UT Techniques. Employed 3~
Hap of Defect Sizes and Locations 4i Summary. of Evaluation Results aad Repairs Required 5
Example Calculation Cover Plate to Stiffener Evaluation 12 12 12 13 2Q 20 20 21 21 21 1
Approach
. 2 UT Techniques Employed 3i Nap of Defect Sixes aad Locations
'4.
Summary of Evaluation Results and Repairs Required 5
Example Calculation 23 24 25 25 26
I
- I
I
~
~
e
~
C
~Pe e Ne.
E Stiffener to Stiffener Evaluation I
Approach 2
Hap of Defect Locations 3
Summary of Evaluation Results and Repairs Required 4i Example Calculation F
Conservatisms In EngineeringIEvaluations QJALZTT ASSDRANCE/CORRECTXVE ACTION 27 28 28 28 30 A
Bo Co Di Eo Introduction Summary of Weld Problems Shy the ProbZems Vere Not Discovered in the Shop Corrective Action for the BSW Actions to Reduce the Possibility of Recurrence 33 33 34 34 34 ANALYSIS OF SAFETT IMPLICATIONS COHCLUSXOHS APPENDICES 40 4l A
Revisions to Xnterim Report:
History of Events 42 44 45
~
~
I 0
LIST OP TA3LZS Table Number TLele Inner VaII to Stiffener UT Indication Data Cover PLate to Stiffener UT Indication Data Stiffener to Stiffener UT Indication Data
J
h
~
I LIST OF FIGURES Fi ure Number Title Reactor Building Configuration 3
Base Detail Elevation Cover Plate to Base Plate Veld 8'ertical Stiffener to Inner Wall and Cover Plate Velds Horizontal Stiffener to Vertical Stiffener Velds I
Horizontal Stiffener to Innex'ali and Cover Plate Welds.
Cover Plate to Base Plate Weld Toe and Root Indications 10 Horizontal Stiffener to Innex Wall Veld Indic'ation Surface and Subsurface Defects 12 Defect Size 13 Stiffener to Innex'ali Straight Beam and 4~egree Angle Beam Scanning Patterns Calibration Block Used to Establish Primary Refex'ence Level 15 16 Calibration Block Used for the Examination of Fusion and Heat Affected Zone in Vertical Stiffenex Map of Innex'all to Stiffener UT Indications 17 18 Subsurface Defect Cover Plate to Stiffener Weld Zones 19 Cover Plate to Stiffener Weld Zone 2 Beam A'ngles
'J
Pi e Number 20 TLtla Reference Blocks Hap of Cover Plate to Stiffener QT Indications 22 Example of Cover Plate to Stiffener UT Indication-Types of UT for Stiffener to Stiffener Welds Hap of Stiffener to Stiffener UT Indications Pipe Restraint Wth Gusset Shear Stress Due to Temperature Differential
'l
~
\\
4 I
INTRODUCTION A
Statement of Problem and Summary of Conclusions
/
During nondestructive examination of the biological shield waLL (BSW) at the Jobsite, weld defects* were discovered in a number of shop welds.
Investigation has shown that the weld defects existed primarily in backing bar welds; only minor defects were discovered in double bevy; welds.,
ALL shop weld points were evaluated 1n accox'dance with AWS DL I and original PSAR commitments and either -were shown to be acceptable or were repaired.
Based on the engineering evalu-ation, it has been determined that the BSW weLd defects could not have adversely affected the safety of operations of the Nine Kt1e Po1nt 2 plant had the weld defects remained und1s-coverado
- Note:
Defect" is used throughout this report to describe a weld discontinuity and should not be construed to mean that a weL6 discont1nuity is not acceptable B
Overview of Report The purpose of this report is to provide the following:
1 ~
A description of the BSW, including its specification and Quality Assurance px'ogrsm requirements 2
A, detailed technical description of the weld problems.
30' presentation of the results of an engineering evaluation and a corrective act1on plan, including a discussion of the overall approach, summary of the evaluation results and repairs required, and example calculations The Quality Assurance corxective action.
5.
An analysis of safety implications.
C'evisions to Interim Report Portions of the Znterim Report of April 15, L980, require revisions to refLect the final closure plan action which was subsequently adopted and clarifications to maintain con-sistent terminology.
Ea the interim report, it was inadvex tently stated that the eng1neering evaluation was not in accordance with AWS Dl LE As stated 1n this report, all UT evaluations and acceptance of indications axe in accordance with AWS Dl,l, paxagraph 3.7.6 Details of interim report revisions are listed 1n Appendix A to this report.
ZI DESCRIPTIOH OP BIOLOGICAL SHIELD WALL A
Physical Description The BSW is supported by the reactor pedestal (Figure 1) and is attached to the pedestal by means of embedded anchor bolts (Pigux'e,2)
~
The BSW is an extremely stable structural system because of the lateral support at the drywell floor in the reactor building and the star truss system near the top of the dryweII.
The BSW consists of two concentric steel cylinders connected by horizontal and vertical stiffeners (Pigures 3 and 4).
The BSW is 48 ft 4 inches high and has an inner radius of 14 ft 3/4 inch and an outer radius of 15 ft 9 1/4 inches.
The BSW was fabricated in three rings, each approximately 16 feet highs Each ring was shop fabricated in three 120 degree sectionsi The inner and outer shells and the stiffeners are 1 I/2 inches thick, A537 Class I steel plates connected by full penetration welds The space between the shells will be filled with nonstructural high density concrete for neutron radiation shielding purposesi The BSW is penetrated by air duct openings, inspection openings, instrumentation line and pipe sleeve penetrations, and doox'penings for various piping-systems Attached to the wall axe pipe restraints, a
BSW extension to support the star truss and stabilizer, clip angle supports for floor beams, and insulation support brackets.
The full penetration welds used in the BSW ax'e both single bevel (with backing bars) and double bevel types The follow-ing table lists the various weld configurations, their abbre-viations, and the figure in which they are shown.
Weld Joint Abbrevia-tion Pigure No.
Cover plate to base plate Vertical stiffener to inner walI Cover plate to vertical stiffener Vertical stiffener to horizontal stiffenex Horizontal stiffener to vertical stiffener Cover plate to horizontal stiffener Horizontal stiffener to inner waIl CPBP VSIW CPVS VSHS HSVS CPHS HSXW 5
6 6
7a 7b 8
8 B
Punctional Requirements The functional requirements of the BSW are:
I Provide shielding against radiation from the reactor vessel o 2
Provide anchorage support for pipe restraints, pipe supports, flooring beams, and insulation.
3 ~
Provide support for the star truss/stabilizer system.
4.
Protect the reactor pressure vessel from pipe whip, get impingement, and missile loads
J
REACTOR BUILDING CONFIGURATION REACTOR REACTOR PRE5SURE VESSEL (R.P.V.)
STAR TRUSS TOP OF BSW EL. 314'-1~/g" EL. 316'-1 I
\\ 28'1/ l ~
67'-4" I
EL. 302'-0 BEND LINE PRIMARY CONTAINMENT REACTOR SUPPORT SKIRT BIOL 0 GlCAL SHIELDWALL EL. 266'-6~/q" EL 240'-0 15'-3" DRYWELL BEND LINE REACTOR P'EQESTAL 4l 4I MAT EL. 175'-0 7
FIGURE 1
J
BASE DETAIL 3'-5 >is"
-8 sic" lNNER WALL GROUT AREA Q'o:<
~:o'::
1>/a"
- PASE PLATN
'.:>:i'4?o..!
'o::-'>/~"
COYER PLATE NASHER PLATE EL. 265'-5 >lz" 3)'I SOLE PLATE '"":0 ANCHOR BOLT PEDESTAL 21'I 31$
P
3RD RING 17'-6 I/2" 2ND RING
)5'-) S/I6"
)ST RING
)5'-6 3/16"
)gt 0 3/gll R INNER WALL CONCRETE COVER PLATE HORIZONTAL STIFFENER
'ASE PLATE STAL FIGURC 3
VERTICAL STIFFENER
)g.0 1'-8 1/2".
3 1/2" (TYP.}
1 1/2" (l'YP.}
FIGURE $
O.
C C
COVER PLATE TO BASE PLATE WELD COVER PlATE 7/S" l/4"45'll BASE PLATE WASHER PLATE FIGURK 5
I
VERTICAL.STIFFENER TG INNER SALL AND COVER PLATE WELDS 1 >/2" lNNER WALL i/8" 4$
'ERTICAL STIFFENER I 1/2" COVER PLATE
'I/O" 324
~(244+84)
FIGURE 6
/
HORIZONTAL STIFFENER TO VERTICAL STIFFENER WELDS
)/Qll 4 $ 0 VERTICAI.
STIFFENER HORIZONTAl STIFFENER HORIZONTAL SjlFFENER VERTICAL STIFFENER -,
b
P
HORIZONTAL STIFFENER TO INNER WALL AND COVER PLATE WELDS
'I I/>"
80 I 1/2"
/)
30o
$4o INNER WALL COVER PLATE HORIZONTAL STIFFENER FIGURE I
C Summary of Design Criterfa and Lfcensing Commitments The BSW is desfgaed fa accordance with the AISC Manual of Steel Construction for normal operating load conditions.
For abnormal/extreme environmental load combinations, the allovable stresses are fncreased fa accordance with the factors specified in the NMP2 PSAR.
The following loads have beea considered in the analysis aad design of the BSW:
1 ~
Deadload. and seism1c loads 2+
Accident temperature cases consisting of the maximum temperature dffferentials between the inaer and outer walls occurring as the result of a-loss~coolant accideat (LOCA) 3 Accident pressure dffferent1al betveen the inner and outer walls occurring as the result of a LOCA Pipe restraint loads occurring as the result of restraining pipes folloving a postulated rupture 5
Jet fmpfngemeat loads resultiag vhen pressurized flu1d from a ruptured pipe strikes the BSW.
D.
Summary'f AnalytfcsL Techniques The structural analysis of the BSW vas performed by the finite element method using the computer program STRUDL The BSW was modeled using a 180 degree model with the appropriate boundary coadft1oas for the symmetric and antisymmetric loads.
Analyses for geaeral loading conditions vere conducted usiag priacfples of superpositioai The inner and outer walls as weLl as the horfzontal and vertical stiffeners were modeled using isoparametrfc elements After the computer runs of the fndfvfdual load cases were made, the stresses were combfned in accordance with the appropriate load combfnatioa equation from the HMP2 PSAR, Section 12e5e2e8~3a The effect of accident temperature conditions vas studied taking into account the concrete inside the BSW, It was concluded that the coacrete would crack uader these conditions and that the effect of including concrete on the stresses in the BSW fs insignificant For other loading coadftfons, the stresses in the BSW vould be reduced ff coacrete vere included; hence, ft vas considered conservative not to include the effect of concrete
k
E
'ummary of Stx'esses for Design Loadings The various loads aad load combinations considered in the BSV analysis aad design meet the requirements of PSAR Section 12.5.2.8 3.
The stresses which vere considered include those due to deadload (D),
accident temperature (T), accident pressure (P), seismic (E), pipe restraint loads (Rr), and jet imp1ngemeat loads (R]).
In all areas: of the BSQ,, the stresses due to the various Roads'were combined in accordance with the following governing equations from the PSAR:
I 6S < I OD + I~ OT + I~ OP le8S
< I~ OD + I OT + loOP + I ORr + I OR) + I OE where S fs the allovable stress based oa AISC Msaual of Steel Construct1on The two load combination equations above x'eflect abnormal/
eztreme environmental coaditfoas and govera the BSV desiga The stresses ia the BSV for load combination equations from PSAR'ection 12 5 2 8 3 besides those listed above, such as for normal operating conditions, are very lov The conditions which control the design of the BSV aad under which the stresses approach the allovables are the accident cond1tfons.
The accident temperature and p1pe restraint loads pxoduce the highest stresses governing the BSW design The accfdeat temperature loads produce longitudinal compressive stresses fa the faner and outer walls fn the 25 to 30 ksi range and longitudinal tensile stresses fn vertical stiffeners in the 15 to 20 ksi raage The pipe restraint loads produce compressive or tensile in~lane stresses ia horizontal and vertical stiffeners wh1ch are located directly behind the pipe restxaints.
Because of the large number of pipe restraints attached to the BSV, and because each xestraint can have a number of load1ag direc-tfoas and magnitudes, the stress x'ange in the stfffeners behind the restrafats varies from approximately 5 to 35 ksf in either tension or compression Only stfffeners fn the imme-
'diate vicinity of a pipe restra1nt are stressed near the allowable stx'ess during a pipe restraint loading, The stresses due to deadload, seismic, pressure, and get imp1ngemeat loadfngs axe, 1a general, less than 8 ksf when
- combined,
P.
Summary of Fabrication and Erection Strategy On the basis of a shipping and economic study, it was determined that the BSW should be fabricated and shipped in nine 120 degree sections, each approximately 16 feet high.
These sections would then be assembled to form three 360 degree rings which would be stacked and welded together to form the waLL~-
The fabrication sequence of each ring (consisting of three 120 degree sections) was as follows:
2%
k fabrication jig was erected to align and hold the shield wall plates during fabrication The inner shell plate was erected and welded to the fabrication jig 3a The horizontal and vertical siffeners were welded to the inner shell platei 4~
The cover plates were welded to the stiffeners 5
The three 12~agree sections were removed from the fabrication. jig for shipment.
6 The sections were transported to the site for field assembly..
The field assembly of the nine BSW sections occurred in temp-erature controlled enclosures, separate from the containment buildingi ln the assembly
- sequence, the three sections of each ring were erected and rounded up using jacking spiders.
The inner waLL and horizontal stiffener welds at each vertical seam were welded, and the cover plates at the vertical seams were weldedi Pollowing fitup and welding of the first ring, it was inverted in order to level the base plate which had distorted during fabrication, The base plate was leveled by attaching shim plates, depositing weld metaL, and machining the surface Two intermediate postweld heat treatments of the first ring were performed while it was in the inverted position.
Weld joint reezmdnation by ultrasonic and magnetic particle testing was performed on the shop welds, and a substantial amount of repair was accomplished While work was progress-ing on ring I, rings 2 and 3 were assembled, fit together, and girth welded Ring I was inverted and rings 2 and 3 were stacked on ring I~
HM and engineering evaluation of defects were accomplished and repairs were made as required.
Upon completion of all fitup and repair welding, the completed BSW was heat treated and prepared for transport and placement on the reactor pedestal in the containment building.
'J
G Specification Requirements The BSW fabricat1on syecificat1on has the following technical, workmanship, and inspection requirements.'
All fabrication work was performed uader the fabricator's QA program 20 Allwelding, weldiag yrocedures, and welder qualifica-tioas shall be in accordance with AWS Dl l.
30 Za addition to the AWS Dl I requirement for l00 percent visual insyectioa, all full penetration welds were required to receive additional NDE 1n accordance with the follow1ag oyt1ons:
a Radiographic or ultrasoaic inspection b~
Progressive magaetic particIe inspection at I/3, 2/3, and 3/3 weld )oint thickness.
(The fabricatox'hose to employ the UT option as well as MT of the root pass
)
4 The quality of workmanship shall conform to the require-ments of the AZSC Code of Standard Practice for Steel Buildings and Bridges, i976.
5 Nondestructive examination of welds shall be ia accordance with AWS Dl I, Section 6
Nondestructive test operators shall be qualified ia accordance with SNT-TC-LA, Recommended Practices, Nondestructive Testing, Personnel Qualificatioa and Certification, l975 6
The steel shall conform to the applicable AS'pecifi-cation as given, aad shall be traceable at all times to a specific heat number.
All plates shall be UT'ed to AS'IK A578, Level I Stiffener and base plate steel shall aot exceed 0 OL percent suLfur.
7 The weld f11ler metal shall.coaform to the requirements of AWS Dl l (1975 edition) ~
H QA Program Requirements The QA Program imposed by the specif1cation required certain actions by both S&W and the Seller, Cives Steel Company.
The fabrication specification was reviewed and approved by S&W's QA Department (Quality Systems Division) to ensure the inclusion of ayyropxiate QA/QC requirements.
'S
The specification and Test, Xnspection, and Documentation (TID) Report required the following:
Gives SGW 1 ~
Compliance with Appendix B 10CFR50 2.
. Submittal of QA program, 3.
Transmission of QA requirements to identified subcontractors 4
Conformance of NDT to AVS Dl I 1975 5
Submittal of welding and NDT procedux'es.
1 l.
Qualification by survey and audit of Cives as Seller 2.
Perfoxmance of inspections (over a 26~onth period)'overing specific TXD attributes 3
The following attributes address weld quality:
Velding Procedure Electrode Contx'ol Procedure Qualification of Velders Veld Preparation Veld Inspection Random Check of Pabrication Completeness Xnspection of Surface Defects HDT Test Operator Certificationa.
HDT Xnspection Procedures HDT Xnspection of Velds Reports of NDT Tests In addition, SSV's Procurement Quality Assurance Department performed an annual evaluation of the Seller's quality history'ncluded in this evaluation was a review of:
Prior Quality Progxsm Audits Seller Surveillance Activities The nature of frequency of haxdware unsatisfactory inspection reports and nonconfoxmances Results of audits by other sources (i,e., Client,
- CASE, ASME, HRC) when available Seller's responsiveness and cooperation in resolving questions or px'oblems related to Quality Assurance.
The above data was evaluated for tx'ends which would indicate a
need for an audit, survey, or other Stone 6 Vebster Management action The above summary constitutes the involvement of the ShV shop inspector, and comprised our normal QA/PQA effort on this type of structure,
i'
Below is a summary of the man-hours expended by PgA oa the Nine Mile Point Nuclear Station - Unit 2 BSW (Contract No NMP2-S204G):
Vendor Oives Tesk District Su rviso Out~f-Plant Related Time Hours 159 Z of Hours 9
1 Final Document Rev1ew-95 Inspection Prep/Report Writing-80 Updating and MaiataQd.ag Specification '4 189 10 8 Travel Time 420 24.1 In-PIaat Time Records Vex'if1catioa Systems Verification Status Developmeat Hardware Inspectioa 72 0*
0 613 Total In-Plant Time CIerical Su rt 685 265 39 ~ 4.
15,2 Typing, filing, reproducing inspection reports, updating office copy of specification, processing veador documentation, etc.
Other e
Totsl 25 1,743 1.4 100 0
The above figures do not iaclude headquarters support activities
'o quaIify supplier, aud1ts, certify inspectors, monitoring inspection reports for negative trends, review quality assur-ance program manual x'evisioas, etc.
~ree audits were performed but aot chaxged to purchase order.
J
III DESCRIPTION OF PROBIZH A.
Statcmeat.of Problem Numerous NDE fndfcatfoas which werc zejectable to Secefon 6.19 of AWS Dl~l, and hence to the fabrication specification, were discovered fn shop velds after the nine BSW sections were NDE'ed ia the shop aad shipped to the jobsite.
The initial indicatioas werc discovered with aa MZ iaspectioa, duriag field work, fa the toe azea of the cover plate to base plate veld joint.
UT results, from a sample UT examination of this veld, shoved that although no UT indicatioas vera present fn the veld toe, reflectors vere present in'he zoot area near the bickfng bar (Figure 9).
As a result of this, the weld joiae was 100 perceae UT examined, repairs were made, aad a sample was removed for metallurgical analysis.
Approx1mately 20 percent of the length of the CPBP veld vas found to be zejectable by the reexaminatfon.
Following the discovery of the cover plate to base plate weld iadications, visual indicatioas on the horizontal stiffener to inner waII (HSIW) velds where backing bars had been removed were identified during fnspection of the three thizdmfng sections (Figuze 10).
As a result, the quality of backfng bar wclds ia the entire BSW was questioned.
An engineering 1nvestfgation was performed on the HSIW wclds by examining the root of thc veld using progressive grinding and'agnetic particle testing,to determine the depth and leagth of the defects (Thc MT results are presented 1n Appendix B.)
The initial results showed an approximately 22 percent defect. rate.
Due to such a high defect rate in backing bar velds, ft vas concluded thee all weld configu-zat1oas should be investigated A sampling plan approach which employed UT inspection was developed for examination of the various BSW veld configu-rations~
In addition, four thfzdmfng HSIW specfmens vere removed for metalluzgical analysis The results of thc sampling plan showed that ll of the 18 weld configuratfons dfd aot meet expected confidence levels Subsequently, the decision was made eo perform UT inspection of all shop weld joints oa the BSW.
The results of the sampl1ng,plan UT did,
- however, show that HSIW weld joint indications occurred with less frequency. than the frequency encountered during.the iaitial MT investigatfon It was concluded that dur1ng investigation of the weld indfcatfons, grinding caused the cracks to propagate because of the jo1at restraint;
- hence, a
misiaterpretatioa resulted vhich overestimated ehe original crack size+
A detailed tfm~equenced h1story of events fs presented in Appeadix C.
l
COVER PLATE TQ SASE PLATE WELD TOE AND ROOT INDlCATlONS COVER PLATE ROOT INDICATIONS INITIALTOE INDICATION'ASE PLATE FIGURE 9
HORIZONTAL STIFFENER, TO INNER %ALL
, NELD INDICATON.
HORIZONTAI STIFFEN ER INNER WALL ROOT INQICATION BACKING BAR REMOVED IN OPENING FIGURE 10
l l'
B ~
Metallurgical Diecussioa Horizontal Stiffener Co Inner Wall and Base Plate to Cover Plate Welds A total of five metallurgical specimens which represented the worst case UT indicatioas found during the initial engineering investigatioa were removed from these type welds (single yee with backing bar) to determine the aature of Che iadicatioas Pour specimens were removed from the third ring horizontal stiffeaer to inner waII (HSXW) weld aad one from the similax'over plate to base. plate. (CPBP) weld.
-Metallurgical evaluation.
of three of the specimens (two HSIW, one CPBP) showed that all of the indications were welding x'elated (e.g,, slag inclusions or lack of fusion) and were insignificant ia size aad effect.
A crack was found in each of the Cwo remaining HSlW specimens, and the cracking condition was investigated by two independent consultants~
Xt was determined that the most probable cause of crack initiatioa was hydrogen and that the cxack propagation was a result of the constraint of the structure.
The hydrogen possibly originated from moisture which was aot driven off by the preheat, but which was retained on the backing bars (The fillermaterial was not coasidered to be a. possible hydrogen source
)
There was no evidence of czack initiatioa or crack growth subsequent'o Che completioa of welding in the shop.
Since. the last shop welds were completed ia December 1978, it has been determined that there was insignificant risk of crack initiaCion or furthex" crack propagation between the time of the investigation (late Hay 1980) and the time of postweld heat treatment (PWHT) (mid-July 1980)., All possibility of further cracking in the unloaded structure has been eliminated by the PWHT (The possibility of crack pxopagation in the loaded condition is addx'essed in Section IV.B,2 )
Pux'thermore, two intermediate PWHT's previously pezfoxmed on the first ring demonstxated that the PWHT process itself did aot induce crack initiation or propagatioa and Chat reinspection aftex'WHT was aot req'uired Therefore, the engineeriag evaluation and corrective action, presented in the next section, shows that this condition,was adequately detected, evaluated, aad repaired.
Cover Plate to Stiffener Welde Theze was a high refection rate (17 percent, 22 percent, and 12 percent on rings I, 2, and 3, respectively) during Che initial reinspection of the cover plate to stiffener welds to the standard AWS Dl,l criteria, Subsequent investigation by excavation determined that the indications were in both the weld metal and in the stiffener plate base metal.
The refection of the base metal indications was largely due Co difficultyin interpreting the AWS Dl.l UT results.
The laminar-type plate indications were acceptable to the ASTH A20 plate UT criteria, ASTM A578 Level I, aad to the AWS Dl.l edge preparation criteria.
10
I
However, plate indications acceptable to A578 Level 1 and later sub)ected to AWS Dl I standard weld UT acceptance criteria would be respectable.
Since Che initial AWS Dl.l UT does not precisely categorize by type or size, or locate these plate indications, the UT inspectors conservatively assumed that the indications were in the weld metaL and, therefore, respectable.
As a result, some rework of these welds was performed in accordance with the standard AWS Dl.l UT acceptance criteria prior to-establishing appropriate engineering. criteria as used, in the engineering evaluation addxessed in Section IV.
UT reexamination of some xewelded excavations resulted in unexpected stiffener plate lamellar separations which were attributed. Co repair weld shrinkage stresses.
In summary, it was found that the first appx'oach, using the standard AWS Dl I UT inspection acceptance criteria, was requiring a greater amount of repair than was necessary, and Chat the repairs in many cases were ineffectual in Chat the repair itself created a severe weld restraint condition and plate Lameliar. Cearing concern.
Therefore, a different approach was deveLoped to better define the indication size and location where-possible as it relates to its metallurgical environment, as discussed in Section IV.
An engineering evaluation of Che size and location data limited the repair Co that required to meet design requirements.
Then, to avoid the concern of lamellar tearing due to repair weld shrinkage
- stxasses, stxict welding controls were imple-mented Ganera1 Weld ualit The welds Chat have been discussed in this xeport were made in the shop, or during fieLd rewoxk, repair or-replacement, using either the'floored process with gas shielding, ox the shielded metal~re welding process.
ALI welds examined metallurgically exhibited overall high quality.
No cracking has been found in weld metal; all observed cracking has occurxed in the heat affected zone (HAZ), or else outside the HAZ in-the base metal.
The stabilization of the cracking condition as a result of time and the PWHT, the evaluation of this condition by stress and fracture mechanics analysis (addressed in Section IV), and stxict welding controls after PWHT ensures adequate weldment quality.
IV+
ENGIHEEEING EVALUATIONAND CORRECTIVE ACTION A
Introduction The resolution of the BSW weld defect px'oblem is divided into two main phases - UT of all accessible shop welds and engi-neering evaluation of UT indications.
I A 100 percent UT reexamination has been performed on accessible shop welds.
(Approximately 90 percent of the.
shop welds are accessible for'T )
Either a standard'UT in accordance with AWS Dl 1, Section 6.19 was performed where access permitted ox a special UT in accordance with AWS Dl 1 was performed 20 Based on the, UT data, an engineering evaluation was pezfoxmed to determine which indications were acceptable as is and which required repaixs The evaluation con-sisted of both a stress analysis and a fracture mechanics analysis to establish technical Justification for the acceptance of inconsequential defects All PSAK commit-ments xegazding loads, allowable stxesses, and other technical requirements were maintained The UT and engineering evan.uation ax'e in accordance with AWS Dl.l, paxagxaph 3.7.6.
In addition, the results from the metallurgical analyses of the five weld specimens which were removed from the BSQ have been factored into the overall plan for resolution of the problems B
OveralL Approach I
Stress Analysis The UT of the inner wall. to stiffener welds provided weld defect sizes, locations, and orientations.
A stx'ess analysis consistent with the following steps was performed in the vicinity of each weld defect:
a The maximum tensile stress due to the combination of individual load case stresses (in accordance with the PSAR load combination equations) was detezndned b~
The auuehnna tensile stress was factored by the ratio of the original weld azea to the zeduced weld area.
This reduced weld area resulted from the weld defect.
12
c The resulting stzess was compared to-the factored allowable stresses If the x'esultfng stress were less than the factored allowable stx'ess, the weld defect vas acceptable from a stress analysis view point; ff greater, the weld defect vas repaired-If a defect vere showa to be acceytable from a stzess analysis viewpoint, a fracture mechanics analysis was yerfohaed usfng the stress and defect information to determine the defect's acceptability.
If a defect vere determined to. be easily accessible and repairable vfthout undue hardship, it vss repaired even though determined to be acceptable based oa an engineering aaalysfs 2
Fracture Mechanics In the industry today, a conservative fzactuxe mechanics analysis fs aov a regular means of assuring the integx'ity of, welded structux'es which realistically contain some discontinuities fn,the material, such as slag, yozoefty, or Lack of fusfoa Such an analysis provides a sound basis for establishing acceptability criteria for the discontinuities aad thus can eliminate unnecessary repairs.
Por most structural steels under normal design conditions, linear elastfc fx'acture mechaaics (LEFM) might not be applicable.
Thex'efore, in this analysis we uee both LEAM aad a techaique by Dovtfng and Towaley (Reference 1)'his method,, known as the ~rftezia Approach, hae been used by others (References 2 and 7), and covers the spectrum of conditioas from brittle fracture (where IZFH fs applfcable) to completely plastic failure (where some form of limit analysis has to be used)
The structure fe built of ASSAM A537 Class 1 steel, which has exce1.1eat fractuze toughaess in the longitudinal direction In cases Were ft fs requixed, the directional properties fn the through~hie%ness dfxectfoa axe considexed.
Por the purpose of this analysis, the vaxiety of discon-tinuities whfch might be eacountered in a welded stzucture can be reduced to two ma)or types:
suxface and subsurface defecteo 13
l
~ '
~ ~
It is assumed throughout the analysis that the applied stress is a tensile stress perpendicular to the defect, the applied loads are dynamic and there is no cyclic loading which could initiate fatigue cracking In this
- analysis, surface and subsurface defects (Figure ll) are defined as in ASME XI, Division 1 The following relations describe LEFM approach as it applies to this structure:
I In the case of a surface defect at the root of a weld, "the stress intensity factor is given by l
r ' V'('/W"~-
In this equation, the defect size a ~
ha + a (Pigure 12), where ha is the actuaL defect size and a is the stressed portion of the backing bar The stress S
S lied + S id~
The fre~urface correc-
..factor PS ~ '1 + 0 12 (I-<) ; the shape factor PE 1/~Q (see ASME XI, Appends A)
Pactor P>
is the correction for finite thickness, t, of a plates 2t ca P
~ tan
'g.
i~
Zt (la)
In. the case of a weldment with a backing bar, t ~ (thickness + a Factor P+ accounts for the stress field gradient caused by changes in geometry of the stressed structure In this case the factor applies to defects emanating from the root'and the toe of T~Lds Analyses of such points were performed in References 8 and 9.
The most conservative of the values from these papers were used in the calculation.
Factor P (a/a ) describes the stress distribution in the backing bar:
P (a/a ) ~ -
(a/a ) -1 + </2 - sin a /a 2
B 2
-1 N
The fracture criterion can be presented as (Ib)
(2)
To be conservative, the dynamic fracture toughness, Kld, is used in this analysis.
K d is calculated from the Sai?orsWorten relation (Reference 10):
K 15 873 (C
)'here C is the Charpy V~otch impact energy.
v 14
Pl
SURFACE AND SUBSURFACE DEFECTS SURFACE DEFECT (B(a) 2a SUBSURFACE DEFECT (B>a)
FlGURE ll
~ '
~]1/
s~t HORlZONTAL STlFFEN ER BACKlNG BAR lNNER WALl.
FlQURE 12
P
~ '
Por subsurface
- cracks, the stress intensity factor is defined as:
K
~ M FES ~ca
- Here, M is the correction factor for membrane stress (see ASME XI, Appendix A).
Pactor P is the E
same as in Equation 1 ~
The defect size, a,. is defined in Figure 12~
The TwoWriteria relation for the critical applied.
- stress, Sp, is:
J 2
-1 S
~ S x cos P
u 2
2 exp x x (S
-S
)
k res S
2 The ultimate stress is:
S
+S 2 - 2<1 (1-t/t) -~2" (1-t/t),
where S
and St are the yield strength and tensile ys ts strength, respectively.
The critical stress, S, is calculated from the following LEFM relation:
X d 1 + 4.593(a/g)
Sk ~
Zt [1.12Ftp(a/tg)]
ttn 't 2t In the cases where bending and membrane stresses are acting on a defect, the stress intensity factor for a surface crack is given by the following LEPM equation:
I SW bb G
E~
where M is the correction factor for bending stress (see ASME XI, Appendix A),
S is the membrane stress and Sb is the bending stress.
The other parameters are defined as above.
The LEPM equation for a subsurface crack is:
K
~ (M S
+ M Sb)PE+Ca (The two~riteria method is not used in cases where bending and membrane stresses are acting simultaneously because of some uncertainties in applying this method to a nonuniform stress field )
15
J
Za conclusion, it should be emphasized that the fracture mechanics analysis is very conservative.
Conservative assumptions are used for the defect size and all defects are assumed to be sharp, which for the large ma)ority of cases is more severe than the actual case 'lso, it is assumed that the stresses due to various mechanical and thermaL loads act simultaneously and that the applied ten-sile stress-is perpendicular to the defect.
The lowest value of the fracture toughness is used throughout the structure Thus, the analysis provides additionaL assur ance for the safety of the BSW structure~
To illustrate the approach outLined in the report, a typical example is given below, which evaluates the effect of exclu-sion of I/8 inch from the root area of a backing bar weld.
The material, A537 CLass 1 steeL, has minimum tensile strength S
70 ksi and minimum yield strength S
50 ksi.
The lowest xracture toughness can be expected in tPe through-ta thickness direction The postulated defect is located in the EC (Figure 12), where toughness in the through-thickness direction, according to SM experimental data, is'etter than in the base metal; the base.
metal toughness is used here for the sake of conservatism.
Based on the experimental Charpy impact tests performed at 90 P and available published data (Reference ll), the Charpy energy at 100 P is:
C
> 20 fr. Ib IZPM A oach The stress intensity factor for this case is given by Equation 2 (see Section IV.B.2)
The parameters in the equation are given below:
a << LLa + a Por this particular example a
<< 0.090 inch.
Thus, in this example:
a << 0 125 + 0 090 << 0 215 inch.
The maximum average tensile stress is given as S li d 25 ksi.
0 applied The residual stress after PWHT at 1,100 P is assumed to be 10 percent of the yield stre11gthy that is+
S
<< O,l S
<< 0 1 x 50 << 5 ksi.
res 's 16
Therefore, PS ~ 1.12 S~S
+ S
~ 25 + 5 ~ 30 ksi applied res PE is given in Appendix A to ASHE XZ, but it can be also expressed by the following equation:
1 1
P 1 + 4e593 (-) '
Oe212(S/S Were E is defect length P> is given by'quation 1a (see Section IV.B 2)
P0 ~
1 328 (in this csee) 8(a/+ )
8(
')
0 ~86 (see 8ecnion ZV 82,.E,qcanion Ih)
- 0. 215 0 090 Pinally, note that the'efect is assumed to be infinitely long so that a/<
~ 0
- Thus, X
~ le12PP(a/a
)S ~ca tan 1 + 4 593(-)
0 212(S/S
)
2 1 ~ 12 x 1.328 x 0.86 x 30~0.213x 2 x 1.59 0.215'an 0
215am 2 x 1.59 1 + 4 593 x (0) '
212 (
)
le65 30 2 50 33.2 ksij/inch The fracture toughness of the material is calculated from Equation 3 (see Section IViB 2):
2 6 13 ~ 823 (20) 48.8 ka1~1ach 0.375 Kl ~ 33.2
< KZd 48.8 The critical defect sIze evaluated from Equation 1
(see Section ZV.B.2) is:
da
~ 0.47 inch.
cr The evaluation is performed by an iteration process until a calculated stress intensity factor equals the fracture toughness of the material.
17
~
~
TwoMriteria A oach The critical applied str'ess is given by Equation'5 (see Section IV.B 2) ~
The parameters in this equation are given by Equations 6 and 7 (see Section ZV.B.2), and S
~ 5 ksi.
The critical defect size is then evaluated fBm Equation 5 (see Section LV.B 2):
da
~ ~ 0.42 inch cr The actuaL defect size is shown to be much,l'ess than the criticaL defect size and is, therefore, acceptable.
Further more, the ratio of the critical crack size to the actual crack size is large.
- 3. 36.
cr 0.42 ha 0.125 Therefore, a I/Stanch defect at the root of a weld sub)ected to 25 ksi. average-tensile stress is acceptable.
18
I
REHRENCES 1 ~
Dowling y Ae Re and Townley, C~ H. A.
The Effeet on Defects on Structural Failure:
A TwoWziteria Approach The L'nteznational Journal of Pressuxe Vessels and Piping, 3, 2,
- 1975, p 77-107.
2
- Chell, G
C A Combined Linear Elastic and Post"Yield Fracture Mechanics Theory and Its Engineering Applications, Fracture Mechanics in Engineering Practice, Editor, P
StanLey.
Applied Science, London, England.
3 ~
Harrison, R.
P A Unified Approach to Failure Assessment of Engineering Structures Fzacture Mechanics in Engineering Practice, Editor, P
Stanley Applied Science,, London, England.
h Muscati, A. and Turner, C
E Poet-Yield Fx'acture Behavior of ShaIIMotched Alloy Steel Baxs in Three-Point Beading Fxactuxe Mechanics in Engineering Practice, Editor, P. Stanley.
Applied
- Science, London, England 5 ~
- Townley, C
H A
The Integrity of Cracked Structures Under ThermaL.
Loadingi Fracture Mechanics in Engineering Px'actice, Editor.,
P Stanley.
Applied Science, London, England 6
- Roche, R
L Analysis of Structures Containing Cracks - Some Comments on the J Integral Criterion The Znternational Journal of Pressure Vessels and Piping, 7, 1979, p 65 7'loom, J M
Prediction of Ductile Tearing of Compact Fxacture Specimens Using the R-6 Failux'e Assessment Diagrams The Inter national Journal of Pressure Vessels and Piping, 8, 1980, p 215-231.
8~
Usami, S.
et aL~
Transactions of the Japan Welding Society, April 1978.
9<<Guerney, T. R Finite Element Analysis of Some Joints with the Weld Transverse to the Direction of Stress Welding institute Research Report, E/62/75, Mazch 1975.
10e Sailors y Ro Ho and Corteny He To ASM STP513y p 164o LL~
Lents, J S~
Journal of Pressure Vessel Technology, Vol'00, February 1978, p 77.
19
P
C Inner VaLL to Stiffener Evaluation 1
Approach Based oa UT data, veld defect sizes,.Locations, aad orientatioas vere obtained for evaluation.
The special UT performed from the inaer wall was the sole basis for evaluation of indications aad subsequent accept or zevork dispositioas Stresses in the vicinity of the defects were evaluated as a result of reduced veld area due to the defects All PSAR commitments regarding allowable.
stressea were maintained.
After it'wae shown that the stresses vere at an acceptable level, the stress and defecC information vas used in a fracture mechan1cs evaluatioao 2.
UT Techniques Employed All horizontal and vertical sciffeners to inner walL velds vere 100 percent examined from the inner wall surface using special ultrasonic techniques in addition Co those described in ASS Dl.l These techniques were especially effective in detecting plana~ype discon-tiauities located ia fusioa and heat affected zones of these welds The examiaatioa employed 5 MHz 1/2&nch diameter, straight beam transducer and 4 MHz, 8 x 9 mm 45&egzee angle beam traasducers to scan, the welds as shown ia Figure 13 The tx'ansducer frequencies and sizes were selected to provide optimum sensitivity, resolut1on, aad minimal iaterfezeace fzom nea~ield effects Two calibration blocks as shown ia Figures 14 and 15 were prepared for establishing primary reference sensitivity levels The calibration block in Figure 14 contains a flat bottom slot 1/8 inch wide,. vh1ch was used to establish the. reference level for the straight beam examination The primary reference level fox'he 45&egree angle beam examinat1on vas obtained by using the slot in the calibration block oriented at a 4~agree angle.
In addit1oa, a part of Che fusion zone of vertical stiffener along a 45&egree bevel Chat does not get adequate coverage by angle beam Cxaneducer vas examined by straight beam transducer The reference seasitivity level for this examination was established by using the notchWamecond calibration block shown in Figure 15.
A distance amplitude cuzve (DAC) was established for each examination aad alL the indicatioas above 20 percent DAC vere recorded.
All recordable indicatioae were further evaluated by additioaal ultrasonic examination.
The responses from veld discontinuities vere compared to the responses from various size reference reflectors which simulated the orientation and location of the discontinuities.
Duriag this comparison, corrections vere made for any diffex'ences in second attenuation 20
4'
STIEEENER TO INNER WALL STRAIGHT BEAM AND 45'NGLE BEAM SCANNING PATTERNS INNER WALL VERTICAL STIFFENER
>>/*"
45 454 r
r
/r/rr 04 04 INNER WALL 304
~ ///rrr HORIZONTAL STIFFENER
/
454
/
/r r
PO QO 45'5 45'IGURE l3
4
CALIBRATION BLOCK USED TO ESTABLISH PRIMARY REFERENCE LEVEL 4tl iS j8 11 8
611 411 tt 8
FIGURE 14
A I
P
CAI.IBRATION BLOCK USED FOR THE EXAMINATIONOF FUSION AND HEAT AFFECTED ZONE IN VERTICAL STIFFENER VERTICAL STIFFEN ER O
~ey.
O O
O'NNER
- WALL, 311 3)l FIGURE 15
characteristics between the reference block and stiffener material+
3.
Map.of Defect Sixes and Locations A map of defect sixes and their respective locations is included as Figure 16.
(The defects are.shown on s,.developed view of the BSW fz'om the perspective of'being outside the BSW, Looking toward the inside,) ~ A total of 23 indications (foux in the first ring, two in the. second ring,, and 17 in the third ring) were evaluated and are shown on the map Details of the inner wall to stiffener UT indication'ata are presented in Table 1
4.
Summary of Evaluation Results and Repairs Required ALL 23 indications were evaluated by stzess and fracture mechanics analyses and were determined to be acceptable However, ten indications were reworked because they were accessible from the inside of open compartments and it was detezmined that rework would not be hazmful to the structure The remaining 13 indications were not zeworked because of one or more of the following conditions where rework of the acceptable welds would be more detrimental than the presence of the defect:
a RemovaL of a cover plate would be required to gain access to the indication b
Cutting of an access hole in a stiffener would be required in order to chase an indication into an
'd)scent compartment c~
Cutting through the inner wall would be required to gain accessibility to the indication d
Excavation of a significant amount of weld metal oz base metal would be required to remove the indication, thus imposing large through~hickness shrinkage stress on the inner wall.
The repair status of each indication is summarized in Table 1 ~
5 Example Calculation Subsurface Defect in the Inner Wall of the BSW The following wxample issustrates an actual case of a reported indication in the inner wall of Ring 1.
The indication was interpreted as being a subsurface defect parallel to the surface of the inner wall (Figure 17) ~
za
7 t
sef.lx set@ 0 sn to (sersr I ~
snreo
(
tn'nce Intr
(
I ffe ffe lr I
tt'er
(
et ~lo te er I
~I Sle'X of-ex tlr.tX Ieo. IX I.OX I I~
'let'
'0 Ser.ll'l lee 'I r
lte er lel'OX
'Iot IX Sle'X ter IX steer ne er I-ti W
Sea' 14 ler.r sff'0xr-nr.tx ter.t'lp slr.tx.
ter.0 tet StI'0~
04 ItrIX ttr4x.
styx.
nr.ex.
0%
IlrIX tar.ag s4r4+
tte'sr FIGURE 16 - MAP OF INNER WAI.t. TO STIFFENER UT INDICATIONS
~>>
set Ip llolleo Qg. eot. oooe, oe oocl offteeto fg foloeafftalllt OIIOIICOIICttffCOIIIIlaffs
- 55. Ioloofftalffo.Alfferoocelff cone naffl tltox
TABLE 1 INNER WALL TO STIFFENER UT INDICATION DATA Indica)f~n No, Length (Inches)
Thru-Thick Depth (Inches)
Accept/Rework Dis osition Basis for Dis osition l-l 1-2 1-3 1-4 2-1 2-2 10 (int)(2) 13/16 1
5/8 28 1/4 1 7/8 1/8 1/8 1/8 1/8 1/8 1/8 Accept Accept Accept Accept Accept Accept Inaccessible>
Inaccessible, Inaccessible, Inaccessible, CP removal CP removal CP removal CP removal Inaccessible, CP removal Inaccessible, CP removal required required required required I
required required 3-1 3-2 3-3 3-4 3-5 3-6 3-7 3-8 3-9 3-10 3-11 3-12 3-13 3-14 3-15 3-16 3-17 5/8 37 1/2 (int) 23 (int) 5 1/8 3/4 22 1 3/8 26 5
29 3/8 29 3/8
'3 (int) 5/8 18 18 24 45 1/2 1/8 1/8 1/4 1/8 1/8 1/8 1/8 1/8 1/8 1/8 1/8 1/8 1/8 1/8 1/8 1/8 1/8 Accept Accept Accept Rework Rework Accept Rework Accept Rework Accept Rework Rework Accept Rework Rework Rework Rework Inaccessible, CP removal required Excessive weld metal or base metal removal required Excessive weld metal or base metal removal required Accessible, only small amount of metal removal required Accessible, only small amount of metal removal required Excessive weld metal or base metal removal required Accessible, only smell amount of metal removal required Excessive weld metal or base metal removal required Possible burn-through Accessible, only small amount of metal removal required Excessive weld metal or base metal removal required Accessible, only small amount of metal removal required Accessible, only small amount of metal removal required Inaccessible, CP removal required Accessible, only small amount of metal removal required Accessible, only small amount of metal removal required Accessible, only small amount of metal removal required Accessible, only small amount of metal removal required Notes l.
Indication numbers correspond to numbers in Figure 16.
2.
(int) Indicates the defect is intermittent in the given length.
3.
All indications were acceptable baaed on engineering eyaluations.
r
SUBSURFACE DEFECT 1 1j2" REPORTED UT INDICATION 1.32"
~2a 1 I/2 HORIZONTAL STIFFENER INNER
%ALL FlGURE 17
1
LEPM ADDroach The stress intensity factor for this case is given by Equation 4 (see Section ZV.B.2).
Por subsuxface
- defects, the defect size, a, in Equation 4 is half of the actual reported defect size (denoted 2a).
Note that the actual defect size, 2a,
@as reported as less than 1/8 inch, but it ms conservatively assumed here that 2a~ 1/8 inch and therefore a ~ 1/16 inch.
The defect length, according to the UT report, wig~ 3/4 inch.
- Hence, a/<~ 0.083.
The stress, once again, includes both applied and residual stresses.
The applied stress in this example is 16 ksi, and the residual stress, as in Section is 5 ksi.
Thus S
S lied + S
~ 16 + 5 ~ 21 ksi.
applied res Factor M
~ 1 (defect parallel to the plate surface)
Pactor P is given by the same equation as in the example in Section ZV.B.2 above.
Thus p
~ 9.13 ksi ~in 21 cx 1/16
~ sn 1 + 4.593 (0 ~ 083) 0.212(50)
Since the fracture toughness of the material in the through-thickness direction at the minimum temperature when the stxess might develop is:
K d ~ 48.8 ksi ~n(see Section ZV.B.2)
Thus, Q is Q51ch less than The critical defect size evaluated fran LEFM Equation 4, (see Section ZV.B.2) assuming that the ratio a/~is constant, a
~
1 3/4 inch and 2a
~ 3 1/3 inch.
cr cr Two>>Criteria A roach The critical applied stress is given by Equation 5 (see Section ZV.B.2).
The parameters in this equation are given by Equations 6 and7 (see Section ZV.B.2).
S
~5 ksi res The critical defect size is evaluated by an iteration process fran Equation 5:
a
~ 1.7 inch, cr so that:
2a
~ 3.4 inch.
cx'herefore,the 1/8 inch defect sub)ected to 16 ksi stress is acceptable.
22
I
D Cover Plate to Stiffener Evaluation 1.
Approach The cover plate to stiffener welds were evaluated in a similar manner to the inner wall to stiffener welds; namely, with a stress analysis employing mapped UT data, effective weld area reduction, and stress comparisons, and with a fracture mechanics analysis employing stress and UT data The engineering evaluation limited the amount of cover plate to stiffener weld repair to only that which was necessary to meet the design and PSAR comad.tments.
This Limited repair was necessitated by the following events.
The relatively high percent refection rate (17 percent, 22 percent, and 12 percent for Rings L, 2, and 3, respec-tively) during the initiaL reinspection of the cover plate to stiffener welds was determined by excavation to be due to.1ndications both in the weld metal and in the stiffener plate base metal The refection of the base metal indications was Largely due to difficultyin interpreting the UT results These laminar'ype indi-cat1ons were acceptable to the plate UT criteria, ASKS A578 Level I, and to the AWS Dl 1 edge preparation criteria However, if the same indications were in'the weld metal, they were respectable to the AWS Dl.l standard weld. UT acceptance criteria The standard AWS Dl.l UT did not preciseLy locate these indications; therefore, it was conservatively assumed that the indications were 1n the the weld metal Some rework of these welds was required in accordance with the standard AWS Dl 1 UT acceptance criteria.
UT reexamination of the reworked cavities showed unexpected stiffener pLate lamellar separations which were due to the weld shrinkage stresses induced by repa1r This approach, using the standard AWS Dl.l acceptance
. criteria, required a greater amount of repair than was necessary; the repairs in many cases were ineffectual since the repa1r itself created new weld indications.
A UT program, described 1n Section IV.D.2, was developed to better define the indication size and location On repairs which were necessary, welding techniques were implemented to mitigate the weld shrinkage stress problem It should be noted that some cover plate welds can be UT examined only when the ad)scent afte~oncrete cover plates are attached 23
r
2~
UT Techniques Employed All shop cover plate welds were examined by ultrasonics ia accordance with the requirements of the American Welding Society Code, AWS Dl.le Ia addition, all indicatioas re)ected by the staadard AWS criteria were further examined by additional UT techniques to establish their.relevancy aad to better quantify their nature and size~
Most of the AWS respectable iadicatioas wex'e margina1.
aad data is aot available which relates AWS defect ratings to actual flaw'ize In order to develop suitable procedures which would provide'ufficient iafoxmation for stress aaalysis aad fracture mechanics evaluatioa of the discontiauities, the
,weld area was divided into three separate zones as shown in Pigux'e 18, aad the weld discontinuities tabulated for each zone.
Indications in Zone l were aot reexamined, since the ultx'asonic responses from weld discontinuit1es may'ave been influenced by'umerous small but acceptable plate laminations contained within the hor1zoatal aad/or vertical stiffeaers~
All ind1cations in Zone l were removed and the welds ware repaired.
Indications, in Zone 2 were examined using, in all cases, both 45Megree aad 7~egree beam angles as shown in Pigure l9 Refereace blocks containing artificial reflectors which simulate the orientation, size, and location of crit1cal flaws as defined by fracture mechanics analysis were used during the examinatioa.
They are shown ia Pigure 20 Other test parameters, such as transducer frequency, probe size, and instrument cal1bratioa, were addressed to provide the resolution necessary for proper interpretat1on during. the examination.
Duxing the exami-nation, all responses were recorded ia decibels relative to the responses from each reference block reflector.
Iad1catioas ia Zone 3 were exam1aed to establish that their physical sins d1d not exceed a 1/2-inch through-wall dimension~
This criterioa was determined by fracture mechanics analysis as a meaningful threshold for gathering data The use of a 4 8Hz, 8 x 9 mm, 7~agree angle beam probe within the halfskip distance provided a sound beam within Zone 3 which was less than the 1/2-inch through-wall size, therefore, other estimated limits of reflectoxs larger than 1/2 inch are meaningful by the 6 dB drop methods Also, the separation of probe positions at the 6 dB limits from point source reflectors is far less than from the 1/2Mnch reflectors Probe spacings were xecorded for the 6 dB limits from all reflectors in Zone 3
The results of all examinations were xeported for further engineering evaluation as stated below.
Ct
ZONE 1
ZONE 3 COVER PLATE ysi STtFFENER ZONE' FlGURE 18
t
COVER PLATE TO STlFFENER WELD ZONE 2 BEAM ANGLES COVER PLATE 7P'0 8
4S STIFFENER FIGURE 19
1, P
I/8" DEEP I/4" DEEP y
11
- a. 45'EFERENCE BLOCK 411
'I 3/
+ SIMULATED BACKING BAR
- b. 70'EFERENCE BLOCK FIGURE 20
l
~'
Inte retations The responses from all indications in Zone 2 were less than the I/8~h deep notch in the 45&egree reference block and also substantially less than the 7~agree reference block reflector.
The response from all reflectors in Zone 3 did not indicate discontinuities with through~ll dimensions exceeding.
I/2 inch Evaluation The through~11 size of all Zone 2 discontinuities may be considered to be less than I/O inch and the through~ll size of all Zone 3 discontinuities less than I/2 inch These values were used as part of the engineering disposition for acceptance or refection 3.
Map of defect sizes and locations A. map of defect sizes and their res'pective locations is included as Figure 21.
(The defects are shown on a developed view of the BSV from the perspective of being outside the BSW, looking toward the inside )
411 indications in Zones 2
and 3 which require an engineering evaluation are shown The first ring had 71 defects requiring evaluation; the second ring,. 51 defects; and the third ring, 121 defects.
Additional cover plate to stiffener. data. are summarized in Table 2~-
4 Summary of Evaluation Results and Repairs Required The total length of cover plate to stiffener welds examined was approximately 48,100 inches Of the total length
- examined, 243 UT indications totalling approximately 1,470 inches were discovered in Zones 2 and 3, which was approximately 3 percent of the total length examined.
Through stress and fracture mechanics
- analyses, 228 indications were acceptable as-is; 15 indications totalling approximately 142 inches were reworked h summary, including indication sizes and dispositions, of the indications is presented in Table 2.
In Zone I of the cover plate to stiffener welds, 48 indi-cations totalling approximately 450 inches were discovered and reworked.
25
Srvetl-srr-oa-srv te Ier'I0 IOV 1%
mr.0%
Ier.t)p
\\sac III nt Io ter.sr nl
'I I
I ns er
~).1
).2 Irr ).2
).0
-I
).I)I ae're tfr'll'llte tlr'ar
-I tet'll'10
~) J
~2
~)
~2 attr uat
.10
~I ns'Ir on sr er ar et Iv re.av s.aP
.IO' I
.10
-10 lie ev Irrle Irr OO
).Sl ll Iro.$ %
I~
s F110
~t
.~
.Il
.11
.Sl
~I
).1st
-0 I
~ ar Ierloa Iev 11a tel 4%
Ier.lg
~ v Ir'er tulet
~III ter.e
'I Ilr.IN 22 nersr ne el
- 2. ~
2.11
~l QI.l )
~I
~I
~2
~2
~2 Iel'll'.1 aer
.Cl ~S
/
-~
~2
~'
I' Ir.e-lit%a I~I' 1 tn.ly; nreett tl 're
-IS
'I
-2 I.)2
.C i.)
41
~2
~2
~I'rt'el.syr
-trr.sa-n 'lrÃrlff n
Sar.ota se tst'll' ee' O'
I
~rl 0 ffl ntffr n 'tr nr'Ir na tr ne
~
t sa'4'Io'S
~I so reffa'ln ss'u ta lr n ra I tr
~
sl.sv
~t re.te.
FlGURE 21-MAP OF COVER PLATE TO STlFFENER VT INDICATtONS Iloreo Qg ~ rel, oooe. oe oocr orlroeo gg relearlrAnle. WIOICCIWterrlCalle tlAffl grS. tete errrurro. Arrre.coecnff corn reeffs
~-
T48IH 2 ~ COVER PLATE TO STIFlEHER UT IHDICATIOH 04Th (kll fadfcatfoas sere acceptable froude 411 engineering evaluation unless othersfse noted )
ladfcatfon
~!la.
1 Length
~Zona 2
~?eh
" ladfcatfoa
~Ão. I Length
~Inch
)
I I I 2 I 3
. 14 IS I&
I 7 I&
IW 1-10 I II 1-12 I 13 1-14 I 15 I 16 I 17 I 18 I 19 1-20 1-21 1-22 I 23 I 24 I 25 I 26 1-27 1-28 I 29 I 30 I 31 I 32 I 33 I 34 1-35 I 36 I 37 (3) 1-38 I 39 I~
I&I 142 1<3
= 1~
IAS I~
1-47l~
I&9 1-50 I 51 I 52 I 53 I 54
.I 55 I 56 I 57 1-58 I 59 1%0 2
2 2
3 2
3 2
2 3
2 3
2 3
2 3
2 3
2 3
3 2
2 3
2 3
2.
3 2.
2 3
3 3
2 2
2 3
2 2
3 3
3" 2
3 3
2 3
3 3
2 2
2 3
3 2
3 2
33' 2
I 26 2I I/4 30 I/O I
12 3/4 11 I/O 4 3/8 2, 3/4 I I/2 7
4 5/8 18I 5 I/2 I I/4 I 3/8 I 3/8 I I/4 2 I/2 I I/8 10I I/8 4I, I/4 2 3/4 I 3/4 I I/2 3
8 I/O 4 I/2 8 I/4 I 3/4 3 1/4 4 I/O 15 7 I/O 2 5/8 7 I/8 27 I/2 2I I/O I I/O II I/2 8 3/8 3 I/2 2 I/2 5 I/2 17 3/8 2 I/4 I 3/8 4 I/4 2 I/4 7
8 I/8 I&I 1&2 1&3 I~
1&5 1%6 1&7 I&8 1-69 I 70 I 71 21 2~2 23 2&
2 S 2&
2~7 2M 2 9 2 10 2 11 2 12 2 13 2 14 2 15 2 16 2 17 2 18 2 19 2>>20 2 21 2~22 2 23 (3) 2 24 2 25 2 26 2 27 2 28 2 29 2 30 2 31 2~32 2~33 2 34 2-35 2-36 2-37 2 38 2-39 2~
2-41 2&2 2&3 2-44 2-45 2W6 2&7 2&8 2
3 2
2 2
2 2
2 2
2 2.
2 2
2 2
2' 3
2 2
3 3
2 3
2 2
2 2
2, 2
2 2
3 2
3 2
2 3
3 2
3 2
2 2
3 3
3 2
3 2
3 3
2 2
2 3 '
I/8 I I/O II 3/8 I 3/4 22 3/8
'7 3 I/2 2 I/?
3 I/O I 3/4 3/4 I I/4 I
2 I/4 25 3/8 I I/2 I I/4 3 I/2 4 I/2 I I/2 I 3/4 2 3/8 I 3/4 16 5 3/4 2 I/O 4
3 I/2 8
4 3/4 I 3/4 14 3/8 2 I/O I 3/4, 8 7/8 4
2 I/2 3I 3/8 I
2 3/4 I I/2 IS I/O 6 I/8 2I 5/8 2 I/2 I I/8 I 3/4 3
2I 3/8 I I/2 4 3/4 15 6 3/4 24 3/4 Hotes (I)
Indfcatfoa aunbers correspond to aunbers fn 2fgure 21.
(2)
Zona 2 indicatioas have a through thfckcess depth of I/4 fnchl zona 3
fadicatfoas have I/2 inch.
(3)
Indication uas unacceptable froa an engineering evaluation and revorked, (4)
Indication vas acceptable froa an eagfneerfag evaluatioa but reworked because ft was ed)scent to a zoae I iadicatfoa,
r J
~
~
Indication
~Ra.
1 2-49 2 50 2-51 3-1 3 5 37 (4) k4 (4)'
(4) 3 10 3 ll 3 12 3-13 3-14
. 3-15 3 16 3-17 3-LS 3 19 3 20 3 21 3 22 3 23 3 24 3-25 3 26 3 27 3 28 3 29 3 30' 31 3-32 3-34.
3 35 3 36 3-37 3 38 3-39
~I (3)
~2 (3)
~3 (4) 3-41
~8
~9 3 50.
3 SI 3 S2 3 53 3 S4 3-SS 3 56 (4) 3 57 3 SS 3 59 (3)
~2
~3 KocES 2
.3 2
2 2
3 3
3 2
2 3
3
.2 3
3 2
2 3
2 2
2 2.
2 2
2 2
2 2
2 2.
3 2
2 2
3 2
2 2
2 2
2 2
2 2
3 2,
2 2
2 2
2 3
2 2
3 2
2 2
3 2
I 3/8 I I/2 40 5 I/4 13 3/4 10 3/8 2 7/8 3 I/2 33 5/8 10L2'/2 I I/O 18 I/2 4 I/4.
28'8 3/4 2 1/8 14 I/2 4 1/8 3 I/8 2 3/4 I I/2 2
2 3/4 2 I/2 2 5/8 3 I/O II 3/8 3 3/4 I
2 3 3/8 I
2L 5/8 3 I/8 3'5/8 5 I/O 10 3/4 I I/8 3 I/2 3 3/8 I
5 I/2 10 I/8 I 5/8 8 I/8 III I/2 L
3 3/4 4 7/8 15 I/2 I 3/4 I I/2 5 I/4 32 I/2 20 7/8 8 3/8 I
1 3/4 5 I/4 I I/4 I I/2 I I/2 Indication
~Ha.
~5
~6
~7
~8
~9 3 70 3-71 3 12 3 73 3 74 3 75 3 76 3 71 3 78 (4) 3-79
~0
~L~ (3)
~4
.~ (4)
~&
~7 (4)
~8 (4)
~1 3 92 3 93 3 94 3 95
~7 3 98
~9 3
LOO 3 101 3 l02 3'03 3 104 3 IOS 3
L06 3-107 3 108 3-109 3 110 3 lll 3 112 3-'113 3 114 3 ILS 3 116 3 lll 3 LLS 3 119 3 120 3-121 2
2 2
2 2
3 3
3 2
3 3
2 2
3 3
2 3
2 2
2 2
3 3
3 3
2 2
22' 2
2 2
2 2
2 2
2 2,2' 3
2 2
3 2
3 3
2 2
2 2
3 2
2 3
Length
~Zaehes I I/2 4 I/4 37 10 I/2 10 I/2 3 3/4 4 I/2 2
6 I/2 4 I/4 6 3/4 2 I/O 3
3 3/4 23 L/4 I 3/4 15 I/2 3/4 7 I/4 3/4 2 3/4 7 I/2 18 I/2 3 I/4 19 21 3/4 4 I/O I I/2 7/8 3/4 1/8 4 I/2 I 3/4 I
L I/2 3
15/16 I 3/4 2 3/4 II 1/8 3/4 II 3
2 3/4 I
5/8 I
7/8 I 1/4 I
5/8 9
57 3/4 2 I/2 (I)
Indication ncncbers correepond to nccnbera in Figccre 21 ~
(2)
Zone 2 incU,caticma have a throccBh thicbneaa depth of I/O inch, tone 3
indicationa have I/2 inch.
(3)
Indication vae ccnacceptable'roe an engineering evalnaticm and reccorked.
(4)
Indication vaa acceptable froca an enBineering evalnation bot revorked becanae it we adfacent to a !one I indication.
r 4
5 Example Calculation Surface Defect in the Cover Plate This example illustrates an actual UT indication in the cover plate located in Zone 2 of the weldment (Pigure 22) ~
The assumed defect depth in Zone 2 is I/4 inch (see Section IV.D.2)~
The tip of the defect does not reach the
- surface, but since the distance between the tip and the surface is less than half of the assumed defect depth, the defect is categorixed as, a surface defect (see ASME XI)~
Therefore, the total depth is 0 325 inch.
This weldment is exposed to combined membrane and bending stresses due to a postulated accident when the outside wall is at a higher. temperature than the inner wall In such a case Equation 8 (Section IV.B 2) gives the stress intensity factor The parameters in this equation are determined as follows:
a ~4a + a 4a ~ 0 325 inch; a
~ 0 181 inch (in this case)
~ 0 325 + 0 181 < 0 506 inch Applied membrane stress, S
appl., is - 2.7 ksi The residual.stress, S
, is assumed to be 5 ksi; therefore, the total membrane stress is:
res'
~S appl+S
~
2 7+5 ~23 ksi m'.
m res Bending stress, Sb, is 33.3 ksi, so the total maximum stress is!
S'
+ Sb 2o3 + 33'3 35 6 ksi The length of, the indication according to the UT report is 4.25 inches, so a/g
~ 0 119 and PS 1 + 0 12 (I~/g )
1'93 2
P
~ 1'4 (Section IV.B 2, Equation la) 8
~ 0.84 P
~ 1'04 (in this case)
G P(- ) ~ 0 885 (Section IV.B.2, Equation Ib)
E 1 + 4 593(a/i ) '
0.212(S/Sys) 1 65 2
I P
EXAMPLE OF COVER PLATE TQ STIFFENER UT INDICATION
. COVER PLATE 3f ~
1.5" a a=
'/i" 0.325" STl F FEN ER UT INDICATlON
='/+(1.5-1.3-1/8) =0.325" FlGURE 22
1 E
0,985 1.65 35.6 1 + 4+593(0.119)
- 0 212(555
}.
K ~(PPS
+MS)PP()P
~ (l 093 x 1 04 x 2.3 + 0.84 x 33.3) x 1.04 x 0 885 x 0.985 x~
~ 34 98 ksi ~in It is conservatively assumed that the critical fracture toughness is the same as in the through-thickness direction of the inner plate, that is, K
~ 48.8 ksi ~in Therefore, Q ~ 34.98 is Less than R
~ 48 8 Id Stiffener to Stiffener Evaluation Approach Investigation has demonstrated that the overall weld quality of the stiffener to stiffener welds is very good Shile all of the stiffener to stiffener welds previously received a 100 percent UT examination in the shop, all of the accessible welds were reexamined by UT in the field.
(Over 40 percent. of the stiffener to stiffener welds were accessible for a UT examination
)
Only minor indications were discovered during the reexamination, and the inci-dence of occurrence of the weld indications was extremely
~
Low Less then-3 percent of the length of welds reinspected were re)actable to the AWS Dl.l standard UT (Pigure 23),
which is within the range of repeatability expected from a UT reexamination No defects were discovered as a
result, of the special UT examination (Pigure 23).
St&xcturally, the stiffener to stiffener welds are less critical than the cover plate and inner wall to stiffener welds The load transfer mechanism is primarily one of shear along the stiff ner to stiffener weld. It has been determined that a ma)or fracture mechanics problem does not exist in this particular configuration, Even with extremely conservative assumptions, such as neglecting the inherent structural stability provided by the restrained, ceLLula~ype configuration of the stiffener to atiffener welds, a crack of approximately 3/4 inch throughMhickness depth can be tolerated for the entire length of the weld UT data bas shown that no weld defect even approaching 3/4 inch can be expected 27
TYPES OF UT FOR STIFFENER TO STIFFENER WELDS VERTICAl.
/
/'
//
HORIZONTAL STIFFENER STANDARD AWS Dl.l UT HORIZONTAL STIFFENER HORIZONTAI.
STIFFENER VERTICAL STIFFENER 00 HORIZONTAL STIFFENER
~s~r '
) i 4$i I
r 0 0404 SPECIAL UT (FOR WELDS INACCESSISLE TO STANDARD AWS Dl.l UT)
FIGURE 23
In order to absolutely demonstrate the acceptability of the BSW, the structural integrity of the iaaccessible stiffener to stiffener welds was addressed ia the follow ing manner:
Sheax'tresses due to temperature aad pipe rupture loads were used to determine the required veld area.
(The stresses due to deadload, seismic, aad pressure loads weze negligible.)
Considering the unlikely occurrence of a large veld defect, the weld area'vailable vas detezmined based oa the largest defect found by UT for this configuration, a 1/4 inch through-thickness indication discovered in the ianex vali to stiffener examination.
By coasezvatively assuming that the defect extends for the entire length of the stiffener, the veld ax'ea available is approximately 84'percent of the original veld area; 80 percent was used for the analysis.
For each inaccessible stiffeaer, it was showa that there was adequate weld area available to react to the postulated accident loads.
2.
Map of Defect Locations The locations of the indications discovered by UT are shown ia Figure 24.
(The defects are shown on a developed view of the BSV fran the perspective of being outside the
- BSW, lookiag toward the inside.)
Additional stiffener to stiffener indication data are summarized in Table 3.
3.
Summary of Evaluation Results and Repairs Required All of the indicatioas originally disc'ovezed by the staadard AQS Dl.l UT vere reworked.
The MT data which vas collected during the rework was iavalid because of the crack aggravatioa due to grinding as described ia Section III.B.
Therefore, the maximum UT indication size of 1/4 inch fzom the stiffener to stiffener innez'ali data (Table 1) vas conservatively used in the analysis.
All of the inaccessible stiffener to stiffener welds vere detexmined to be acceptable based on eagineeriag evaluations sind1ar to that presented in Section ZV.E.4.
4.
ExampXe Calculation For veld )oint 278/12B )read as "weld joint at elevatioa 278'-S 3/4", azimuth 12 29', below the horizontal stiffener"), the minimum required weld areas are determined for the followiag tvo loadings:
a.
Shear Stress due to Pi e Ru ture (Figure 25)
Consider the restraint with applied moment M of 24,000 in-k and pullout load P of 744 kips.
The equivalent force couple F at the extreme flanges is
~ 298 k M
s 28
t
trl'0>>
trna>>
3%
tro'1>>
ttriW l.~
$100I'l
~
tttitO') tr'll' i'"
~0
~or~Sr'l I
$00
~
L
~
lr~'.0>>
nrrl tortr>>
52
~ ~
515 S-ll
$0$'0>>
ttr0>>
$00 0>>
trr 5>>
5-5 lloyd
- t. ~
tlirt>>
SMrr lit'10't 5'4>>
1110 ~ >>
ttt stre 510.$ %e trl'I>>
I nr 5-1
.srr I>>.
Srt'o>>
tIt'Itsy S-t
-trl'I>>
tI ~ wrt
'IrrI>>
5-~
trrI>>
tII'.IrSS ln'll'tr'SV tt ~'ll'rtlt'lolno
~rrt, 0000, ol tort orrrroro nnoorrrawo.wrorllonclrrl cortt nlnl
~mt Mrlkrrro,Jrrlt<orntrn IOTIItllrll
7 r
4 TABLE 3 - STIPPENER TO STIPPENER UT INDICATION DATA Indict~a No.
l-l 1-2 (3) 1-3 (3) 1-4 1-5 (3) 1-6 (3) 1-7 (3) 1-8 2-1 2~2 2~3 2-4 2-5 2W 2-7 3-1 3~2 3~3 3-4 3-5 3-6 3-7 3M 3-9 3-10 3-11 3-12 3-13 3-14 3-15 3>>1.6 3-17 3>>18 3-19 3-20 3-21
~
3-22 (3) 3~23 Length (Inches) 3 3/8 10 (int) 9 7/8 (int) 1, 1/2 4 1/4 4 1/4 16 1/4 11 1'/4 7
5 1/4 9 3/8 3
3 1/4, 3 3/4 (3) 10 3/4.
7 3/4 3% 2 1/4 3
3
.3 3
4-1/2 4 7/8 6
5 1/2 2 1/2 6 1/2 7 3/4 8
4 1/2, 3 3/4 Thtoogly~
Thick.
Depth Inches)
O-I/8 1/8-1/4 0-1/8 1/8-1/4 1/4-1/2 0-1/8 1/4-1/2 1/4-1/2 1/2-3/4 1/4-1/2 1/2 3/4 1/4-1/2 1/2-3/4 O-l'/8 0
1/4-1/2'0-1/8 0-1/8 0-1/8
.1/8-1/4 0-1/8 1/4-1/2 0
1/2-3/4 0
0 1/4-1/2 0
1/8-1/4 1/8-1/4 0-1/8 Notes 1.
Indication numbers correspond to numbers in Pigure 24.
2.
The length and through>>thickness depth were determined by the progressive magnetic particle investigation of the UT indications.
. 3-Indi.cati.on was.reworked without data having been recorded.
(int) - Indicates that the UT indi.cation is intermi.ttent along the weld length.
PlPE RESTRAlNT V/lYH 6ussET
.I
~~A p, II I)
II II II Ii I'
~ 4et GUSSET (BOTH ENDS)
II PIPE RESTRAINT I
I I
I I
)
I I I
I s = SPACING BETWEEN RESTRAINT FLANGES H = CLEAR DISTANCE BETWEEN HORIZONTAL STIFFENERS by. = FLANGE'ENGTH a = LENGTH OF UNIFORMLY DISTRIBUTED LOAD w = UNIFORMLY DISTRIBUTED LOAD t = THICKNESS OF STIFFENER FIGURE 25
4
The stress Q in the flanges is F+P P 18.6 Rs1 t$
From simple beam theory, the reaction at pointAi RA> is R
~
wa A (2H -a) 205 k 2H The required net weld area.
A t is net'..
A
~ ~ 5' in A,
2 net a~
whereCS factored allovable shear stress The percent veld area required for R is A
x 100X ~ 221 net weld b
Shear Stress due to Accident Tem erature (Figure 26)
The. free increase in length of cover plate AB (4~) is equated.to thy sum of reduction in length of cover plate AE (4 ~) dug to developed force plus the vertical displacement
( 4~) of the stiffener.
4 where 4~
4~
w st
'AH + 4AE
~ Q(4T) H (H/ )
~THtAE,2 4~ ~.THt d
d (1.2)
+-
3EI A
G st This equation reduces to 1
Q(4T) E H
d 3 12d X
t 2 A
+ +
3 I A
st st where T <
Q E a d ~
A ~
w AstI 0
st 4T ~
shear stress coefficient of thermal expansion modulus of elasticity distance from inner wall to cover plate area of wall section area of stiffener moment of inertia of stiffener increase in temperature of outer wall 29
r 1
SHEAR STRESS DUE TO TENlPERA'JURE DlFFERENTlAL:
HORIZONTAL STIFFENER INNER WALL I
VERTICAL STIFFENER d
COVER 1'LATE INNER WALL o+
COVER ILATE (b)
FIGURE 26
J
- ~
t
In this case, t
9,2 ksi The total force, P, developed in the weld is F ~ ~
st
~ 24I 5 k" M
'xA d
8 E
The total fox'ce, P, is reduced by the allowable pressure force, Fall
, to give the net force, allow'
<<F-Fal
~615k net allow where Pall allowable foice due to stiffenex'eaxing on the inner or outer wall The required net weld area is A
~
net P
net ~
~
1 9 in2 S
The percent weld area required is A
x IQOX ~ 7.5X weld.
Therefore, the total percent we'd required is 22 + 7'
~ 29.5 percent Based on a net weld area of 80 percent (total weld area less 20 percent for indication area),
the factor of safety FS against shear for the weld is net area 2+7 requix'ed weld area F.
Consexvatisms in Engineering Evaluation Numerous conservative assumptions which were made in the stress and fricture mechanics analyses are listed as follows:
Stress Anal sis 1
In the thermal stress analysis, it was assumed that while the innex'nd outer walls are being heated during an accident condition to px'oduce the maximum temperature differential, the stiffeners remain at the same (opexating) temperatuxe; hence, higher thermal stresses result.
30
20 In the heat transfex'nalysis to produce the accident temperature differential across the shield wall, the vertical stiffeners were ignored; hence, a greater temp-erature differential occurred, ind thus the stresses were overestimatedi 30 Zn the LOCA annulus pressurization analysis, no credit was taken for effects of heat sinks; only flow diverter door venting out of the shield wall penetrations was assumed; the blowdown calculation included the effects of inventory and subcooling; and the break opening time was "assumed instantaneous.
- Hence, a higher annulus pressure resulted In the p1pe rupture analyses, if a dynamic analysis was not performed, a dynamic load factor of 2 ind a stitic 1mpact factox of 1.3 were used; minimum material yield property factox's of 1 3 for bolts and straps and I 1 for aluminum honeycombs were used.
Hence, higher stresses due Co pipe rupture loads resulted Zn Che stress
- analysis, the peak values of various loads, such as accident tempex'ature, accident
- pressure, and pipe
- rupture, were combined in accordance with PSAR load combination equations resulting in conservatively h1gh stresses ln the event of an accident,
- however, the peak load values occur at different times Practure Mechanics Anal s1s The ultrasonic responses from all weld d1scontinuities were far less then the responses fram known reflectors in calibration blocks.
20 Zf there was any doubt about the location and orientation of. a weld discontinuity, Che worst Location and orientation were assumed.
30 It was assumed that all the defects were sharp, which was
'usually not the casei Xn cases where Chrough~hickness propex'ties were applicable, the lowest known value of the fracture toughness in through-thickness direction was used.
Dynamic rather than static fractuxe toughness was used in the analysis although actual strain rates are expected to be at least an order of magnitude lower Chan those corres-ponding to dynamic fracture toughness.
t r
6 It was assumed that the. postulated failure conditions 0
develop at 100 F although the temperature for a postulated accident's expected to be greater, than 135 F.
The fracture toughness corresponding to the 100 F temperature 0
was used in the calculations 70 Zt was assumed that the applied tensile stress was perpendicular to the plane of the assumed defect.
P Mote that assumptions I, 2, 3, and 7 result in high values of stress intens1ty factors and assumptions 4, 5, and 6
give a very conservative estimate of fracture toughness.
Ia add1t1on to the above conservatisms another margin of safety is 1ntroduced by limiting the size of allowable defect'to a fraction of the critical defect size for every weld which has 'been evaluated.
32
t
(
~
~
~
~
Vo QUALITY ASSURANCE/CORRECTIVE ACTION A
Introduction Through the inspection/rework program described within this report the biological shield wall (BSV) will comply with all PSAR commitments and criteria.
However, in order to reduce the possibility of recurrence of the type of problems encoun-texed with the BSW, a plan of actioa has been developed The purpose of this section is to. summarize the nature aad.
possible causes of the weld problems, to discuss the condition of ehe BSV while it was ia the sellex's shop, to explain why the problems were noe discovered in the shop, to review the corrective action which was implemented, and to present a plan of action aad certain steps already implemented to reduce the possibility of recurrence B~
Summary of Veld Problems The two types of welds with evidence of problems were the single bevel backing bar welds, which included horizontal seiffeaex'o inner wall (HSIV) and cover plate eo base plate (CPBP) welds, aad the cover. plate to stiffener welds Horizontal Stiffener to Inner Wall (HSIV) and Cover Plate to Base Plate (CPBP) Welds These weldi exhibited crackiag which propagated from the weld root As discussed in Sectioa IIIC, it was coacluded that the most probable cause of cracking initiation was hydrogen The hydrogen possibly originated from moisture which was not driven off by preheat Ie was also concluded that the source of the hydrogen was not from the weld fillermaterial ox the handling of the weld fillermaterial.
The cracks appeared to have propagated as the result of )oint restraint In addition, it is now believed that duriag inves>>
tigation of the weld iadicatioas, grinding caused ehe cracks to propagate, resulting in misinterpretation which over-estimated the original crack size.
Other weld indications encountered ia these welds included general worhnanshi~elated indications, such as slag inclusions and lack of fusioa.
The presence of this type of indication was shown thx'ough engiaeering evaluation to be acceptable
without compromising the weld )oint integrity.
33
7 I
Cover Plate to Stiffener Feldh ('ac~g ~a The problems encountered in the cover plate to stiffener weeds were in both the weld metal and base metal:
The cxacking and workmanshi~elated indications which were px'esent, wex'e caused by the same reasons as those found in the HSIW and CPBP welds.
Howevex ~ as discussed in Section IIIC, the stiffeners had plate laminations which were acceptable to a plate UT acceptance criteria (hSTB A578, Level I) but re)actable ta a weld QT acceptance criteria~
Such laminations axe inherent.'in plate material and are, the result of plate manufacturing processes Some stiffeners also exhibited lamellar tearing which was the result of a combination of the weld point con-figuration, weld shrinkage stresses in the plate through-th1ckness direction, and the plate laminations 4
C Why the Problems Were Not Discovered in the Shop Although defects such as slag or lack of fusion occurred in the shop, it is possible that some indications did not mctst in the shop but occurx'ed at some later time due to hydrogen inducement, erection/handling
- stresses, or field welding.
C1ves inspection system did identify 1ndications which were repaix'ed in the course of manufactuxe; however, it is possible that. some indications were missed It should be noted that some indications may have been noticed by Cives but passed in the belief that they were acceptable under the Code.
Investi-gation of this matter is continuing, and any substantive new information w111 be submitted to the Commission D
Corrective Action fox'he BSW In order to resolve the shield w'all problems, an extensive program of ultxasonic ezamination. and engineering evaluation, described in Section IV, was conducted The BSW was analyzed or reworked to assuxe that all safety requirements are met.
E Actions to Reduce the Possibility of Recurrence Pro am Enhancements Previousl Im lemented There have been certain lou~ange S&W Standard QA program modif1cations that occurred within the program to meet continuing new industx'y demands, reguLations, codes and standards, and modifications suggested by S&W program audits, as well as ASME, licensees, and regulatory audits
k r
The following describes certain significant Lou~ange modi-fications put into effect by S&W since August, 1978.
S&W PQA Inspector training, qualification, and certification in NDE discipLines were intensified in 1978 The resultant program modifications introduce more extensive training to cover all NDT disciplines within the division (and PQA district offices) and most specifically certification in accordance with ASNT-TC-1A guidelines While PQA Inspectors are not normalLy performing hand~n.inspection, this enchanced training should assist in detecting supplier NDT problems Xmplemented - August 1978 S&W PQA has now implemented a fozmal PQA inspection planning. activity This function provides for PQA engineering revie~ of prospect specifications, codes and standards, and past vendor performance with an output consisting of an inspection plani This plan allows the inspector to concentrate his efforts on surveillance activities and documentation Implemented - ApriL 1979 (QAD-T.14, Revision A) 30 S&W inspection reporting system, which allows centralized data analysis and input into the Quality XnfozmatLon Center, has been modified to provide a variety of inspection report fozms ~
These newly developed forms allow for simplified computer entzy and general ease of completion since they are geared to specific inspection actions such as hardware inspection, process surveiLLance, and system surveillance.
S&W's PQA effort continues to emphasize seLLer surveillance and conformance evaluation toward effective implementation by the vendor of his contractual quality control zesponsibilities These new report formats are designed to facilitate this approach Implemented - February 1980 (QS-14.2, Revision 0) 4 The S&W QA procedural system has been restructured to provide flexibility in accommodating unique prospect requirements without sacrificing the inherent advantages of a standard program.
Implemented - June 1980 (Prospect Pzoceduze No 64)
Additional Pro ram Enchancements Not Yet Full Im lemented Ovex'all, the Nine Mile Point - Unit 2 (NMP2) Quality Assurance (QA) program has been effective in identifying deficiencies.
Nevertheless, the following enhancements are being added to this pxogram so as, to reduce the possibility of recurrence of material and fabrication-related deficiencies:
Xmproved communication and systematic evaluation of current QA problem events Additional engineering invoLvement AdditionaL direction to PQA inspection and project engineering personnel More extensive process surveillance in Seller's facility Procedure and Program improvements These additions are detailed below:
l Improved,Communcations and Systematic Evaluation of Current QA ProbLem Events The S&W Pxoject QA Manager will establish a program to review, on,a month~o~onth basis, all Category X
and other selected Nonconformance and Dispositions (N&D), Engineering and Design Coordination Reports (E&DCR), Problem Reports, Inspection Reports (IR),
and audit findings (HRC, Client and S&W) for identifi-cation of potential generic problem areas as weLL as "short and long term txends that affect quality.
This information will be processed into a monthly summary report that contains a tread analysis and a listing of N&Ds and E&DCRs to be reviewed at a meeting chaired by the S&W Project QA Manager, This meeting will take place monthly unless there axe no substantial problem ax'eas to review Mandatory attendance for the Potential Problem Review Meeting (PPRH) will include the S&W Project Engineer, PQA
- Manager, Superintendent FQC, or their designees.
HMPC QA and Px'oject personnel will also be in atten-dance This committee wiL review the listed N&Ds and E&DCRs for generic type problems and provide immediate evaluation for corrective action in the form of, but not limited to, specification change and/or additional inspection requirements for the fieL'd site or the specific Seller's shop.
The results of these meetings will be documented Initial Implementation - August 1980 FuLL Implementation - October 1980
/
2.
Additional Engineering Involvement The S&W PQA Division has reviewed its operations to determine where system improvements might avoid seller problems As a result of this review, the folLowing will be applied to all new major Category I orders and selected orders already awarded Seller Selection Process Expand the present S&W, QA survey program to include verification of the seller's manufacturing capability.
The survey team will consist of a qualified quality assurance auditor, the responsible Engineer on the
- project, and the engineering specialist as required The team willverify the presence of adequate engi-neering, design, manufacturing, and quality assurance capabilities Post-Award Activities b
Include S&W Engineers on project and/or S&W equipment specialists as members of the PQA Post-Award Conference.
Include S&W Engineers on project and S&W engi-neering assurance personneL as required in audits for recertification.
Initial Implementation - August 1980 Full Implementation - December 1980 3
Additional Direction to PQA Inspection and Project" Engineering Personnel S&W PQA and project engineering personnel will review existing and future Category I specifications for problems inherent in heavy section weldments, including the QA/QC requirements related to.
Welding joint design Welding controL and inspection Fabricator NDE procedure review and approval Fabricator conference to manufacturing plans Vendor personnel qualifications Exercise of stop work procedures under appropriate circumstances Initial Implementation - September 1980 FuIL Implementation - November 1980 37
4 More Extensive Process Surveillance in Seller' Pacility A review of md.sting aad future Category I specifi-cations will be performed and a determiaation made of those ordexs that will require, increased S&W PQA surveillance.
In order. to implement this increased surveillance, S&W will prepare a plan tailored to each pex'tineat order aad submit it to HMPC for concurrence Initial Implemeatatioa - August 1980 Initial Specificatioa Review aad Revised Iaspectioa Plans will be implemented by November 1980 5i Othex Px'ocedure and Program Improvemeats ae For future and selected existing Category I orders where ultrasonic testing (UT) is utilized, the specification will establish hold points at which, prior to performance of any UT, the S&W PQA inspectox',
aided by Nondestructive Testing, (HDT) Divisioa engineers, will evaluate the effectiveness of the seller's UT techniques, operatox's, and general implementation of the test pax'ameters All UT px'ocedure qualifications (techniques) will be appxoved by the S&W Nondes-tructive Testing (NDT) Divisioa eagiaeers at the preestablished hold point as required by specificatioa fox each diffexent configux'atioa.
b For future and selected adstiag Category I specificatioas S&W QA will e'stablish notification points for application of all NDE methods other than UT The NDT Division Engineer willverify NM applicatioas on a random basis after coordinatiag
'he activity with the S&W PQA iaspector.
Procedure qualificatioas (techniques) will be reviewed on a selected basis dependent on observations at the preestablished notification point.
c Require S&W PQA attendance at any pre~ward meeting between S&W Engineers oa prospect aad seller.
d.
Require sellers to notify S&W Pro)cot Management of aay changes in their quality assux'ance managements Initial Implementation - August 1980 Initial Review of Existing Specifications will be Implemented by November 1980 38
t
Increased Surveillance and Audit Additionally', the HMPC ~lit@ Assurance Department wU.1 increase its surveillance and audit program to verify'mplementation of the foregoing program enhancements.
Initial Implementation - September 1980 39
r
VI~
ANALYSIS OF SAFETY IMPLICATIONS Based on an extensive eagineering evaluation; it has been deter-mined that the BSW weld defects could not have adversely affected the safety of operations of the Nine Mile Poiat 2 plant had the weld defects remained undiscovered.
In the context of the safety evaluatioa and 10CFR50 55(e), the term "und1scovex'ed" raChex'han "uncorrected" is used since the evaluatioa has showa that some weld defects need aot be corrected~
The basis for this conclusion ie discussed in deta11 below for'ach weld coafiguxatioa As d1scussed'hroughout this report, the primary weld problems involved the following backing bar-welds:
(l) horizontal stiffener to inner waLL and cover plate Co base plaCe and (2) cover plate to stiffener.
Although minor weld indications were discovered in other veld )oint configurations, the iacidence of occurrence of the indicatioas is within Che range expected for repeatability of a reexamiaation by UT These insigaificant weld defects have a
negligible effect on the structural integrity of the BSW and would noC have been a safety hazard if the defects had gone uadiscovered.
Horizontal Stiffener to Inner Wall and Cover Plate to Base Plate Welds As discussed in Section IV.C of this report, the horizontal stiffener to inaer wall welds were evaluated based on QT examination results.
Fx'om Table 2; all weld indicatioas are acceptable based oa stress and fracture mechanics analyses Therefore, if the weld indications had gone undiscovered,
- a. safety hazard would not have existed The cover plate to base pLate weld, which was Che initial problem discovered, was repaired based oa 100 percent UT.
Evaluation has shown that, even if the defects had been undiscovered, a gross structural failure could not have resulted and, therefore, a safety hazard vould not have existed Cover Plate to Stiffener Weld As d1scussed in Section IV.D of this report, AWS Dl,l weld defects were grouped according to location of the defect 1n the weld joint.
(Refer'to the discussion of Sect1on IV.D.2 and Figure 18) ~
Those veld indications located in Zone l of the weld point were repa1red, aad,those indications located in Zoaee 2 and 3 were evaluated.
From Table 3 it may be seen that all but 17 indications in Zones 2
and 3 of the cover plate are acceptable based on engineering evaluatioa.
By.a realistic consideration of the margins which exist in the analysis, it has been determined that these 17 indications could not have propagated and would not have been a safety hazard.
Addressing indications found and repa1red ia Zone l of Che weld, it has been determined that these indicat1ons would not have been a
safety hazard had the 1ndicatione remained undiscovered Evaluation of the worst possible defect which could have occurred 1n Zone l, taking account for realistic margins which exist in the analytical procedures, shows Chat Che gross structural integrity of the BSW would be maintained and a safety hazard would, therefore, aot exist.
40
P
VII+ CONCLUSIONS The exhaustive investigation and corrective action outlined in this report demonstrates that the BSW will provide radiation shielding and maintain structural integrity for all conditions described in the PSAR All shop weld points were evaluated in accordance with both AWS Dl 1 and the original PSAR commitments and either were shown to be acceptable or were repaired.
It has been determined that the BSW weld defects could not have adversely affected the
'safety of operations of the Nine Mile Point 2 plant had the weld defects, remained. uncorrected The QA program'as been modified to reduce the possibility of recurrence of future weldmelated problems such as those discovered in the BSWi 41
P P
APPENDIX A REVISIONS TO-INTERIM REPORT ITEM NO.
LOCATION IN INTERIM REPORT INTERIM REPORT VERSION CORRECTED VERSION
- p. 1, sect. I.C.> 2nd paragraph Stiffener to stiffener evaluation using inner wall UT data Stiffener to stiffener evaluation using accessible stiffener.to stiffener weld UT as well as inner wall UT data Figure 2 "inner wall plate" and "outer wall plate" "inner wall" and "cover plate" Figure 2 2 inch grout thickness between sole platy and top of pedestal 3 inch grout thickness Figure 3
"inner wall plate" and "outer wall cover plate" "inner wall" and "cover plate" Pigure 5
washer plate to cover plate and washer plate to base plate weld details are shop welds washer plates were removed and reattached using weld details shown in final report Pigure 5
Figure 6 title "Vertical Stiffener to Inside and Outside Mall Plates" "Vertical Stiffener to Inner Mall and Cover Plates" Pigure 6
"inner wall plate" and "loose outside wall cover plate" "inner wall" and "cover plate" Pigure 8 title uNorizontal Stiffener to Inside
'nd Outside Wall Plates" uHorizontal Stiffener to Inner Wall and Cover Plates" 10 Figure 8
- p. 5, sect.
IV.A, paragraph 2
"inside wall plate" and "loose outside wall cover plate" Same as Item No.
1 "inner wall" and "cover plate" Same as Item No.
1
LOCATION IN INTERIM REPORT INTERIM REPORT VERSION CORRECTED VERSION
- p. 6, sect.
IV.A, 1st paragraph
- p. 6, sect. IV.h, 2nd paragraph
- p. 6, sect.
IY.h, 3rd paragraph 1.
defect sizes larger than 1/S inch.will be mapped 2.
the 1/8 inch criteria deviates from AWS Dl.l Same as I~em Ho.
1 Same as Item Ho, 11, Ho.
1
- l. all defects, regardless of size, which underwent engineering evaluation were. mapped 2.
defects are allowed in welds if the provisions of hWS Dl.l, paragraph 3.7.6 are met Same as Item Ho.
1 Same as Item Ho. 11, Ho.
1
APPENDIX B Bm mXS ~ ZpurZ ZC Dam( )
Total Total Length Defect Eeeeieed
~hee eh 1/4-3/8
'3/8-1/2 1/2-3/4 3/4-1 3(2) 3(3) 204 283 2172 950.3 475.1 460.4 273.1 391.6 31.5 35 3 2
0 0
0 0
0 0
0 0
0 60 28 104 10 1.
I/8 - 1/4 is read "1/8 inch ~ defect 5 1/4 inch" 2.
This data does not include the data showa on the next line.
3.
Ã2 data in the 30 ccnupartments made accessible by cover plate removal.
4.
A11 units are in inches.
44
J r
t
~'Histo
'of "Events The discovery of veld defects in the biological shield wall and sub-sequent action taken can be described in three separate phases:
dis-covery of a potential problem, engineering investigation, and sample plan approach.
A.
Phase I - Discove of a Potential ProbLem (Hay 1979)
Based on UT indications M the cover plate to base plate velds and visuaL indications discovered in the third ring horizontal stiffener to inner waLL velds, the quality of backing bar velds for the entire biological shield wall was investigated.
B.
Phase II-En ineerin Invest ation (June 1979 to August 1979)
The purpose of the investigation vas to detexmine if a veld quality pxoblem existed for the biological shield waLL backing bar welds.
The sequence of events for the second phase is as follows:
June l.
A specimen was removed from the base plate to perform a. metallurgical examination of cover plate to base plate indications.
June July Z.
UZ and HT inspections of the horizontaL stiff-ener to inner wall welds for all three rings were performed.
The inspections were performed on random accessible velds.
Based on unacceptable defects discovered by EV in the third ring horizontal stiffener to inner walE welds, additional HT inspction vas performed on.all accessible third ring horizontal stiffener to inner wall weld points (approximately 2,000 inches out of a total of 6,000 inches).
July 3 ~
The ccnrer plate to base plate veld points vere dispositioned to require l00 percent UT inspection and repaired as required.
The UT was performed in accordance with AWS Dl.l with a I/8 inch exclusion allowed at the root of the~veld (based on engineering evaluation).
This UT vas caapleted in September 1979.
By evaluation of data obtained to that tMe, it vas concluded that the horizontal stiffener to inner mall velds on the first and second rings were acceptable but the third ring horizontal stiffener to inner wall welds vere respectable and vould require caaplete reinspection and rework, as required.
P C
lk
~ <
.C.
Phase III-Sam le Plan A roach (August 1979 to December 1979)
Since the initial engineering investigation was primarily concerned with the root of the welds, inspections were performed using EZ methods.
Subsequent investigations to detenaine the quality of the entire volume of the welds were made using UT methods in accordance with the sample plan.
The purpose of using a sample plan approach was. to verify the quality of backing bar welds by using a more rigorous, systematic approach.
A nationally recognized sampling approach using confidence levels consistent with levels previously employed was chosen.
4 The sequence August-Sep tember September-October of events for the third phase was as follows:
l.
An effort was made to establish confidence levels based on the data available fzom the engineering investigation.
An additional 91 inches of HF inspection on the first ring horixontal stiffener to inner wall welds was performed to fulfiLLthe root sample size requirements.
Based on the EV sample plan, the horizontal stiffener to inner wall welds for the first. and second rings were acceptable and for the third ring, regectable.
h 2
Thirty cover plates were removed from the third zing to provide accessibility for reinspection and rework of the horizontal stiffener to inner wall welds.
October November 3.
- The sample plan approach was extended to all weld configurations, including weld foints without backing bars..
The weld points were caapiled into 18 weld groups based on the ring (i.e., first, second, oz third), the point configuration (i.e., single bevel weld with backing baz or double bevel weld without backing bar), and the thickness of the plates being connected (i.e.
1 1/2-inch to 1 1/2-inch plates or 1 1/2-inch to 2-inch plates).
November 4.
Two specimens weze removed fran the third zing horizontal stiffener to inner wall welds to perform a metallurgical examination.
~ (Two additional specimens of the horizontal stiffener to inner wall'welds were removed for metallurgical examination, one in Earch 1980 and one in April 1980.)
46
4 r
November-December 5.
UT inspection of the 18 veld groups vas performed by a certified Level EZ inspector. 'T data previously taken during the engineering investigation was also incorporated into a sample plan approach.
Based on the sample plan UT data of record, ll of the 18 veld groups vere re)ected, 4 vere accepted, and 3 vere not applicable to the sample plan.
Due to the high number of re)ected weld groups, it was determined that all velds, vhich vere accessible for inspection, would be inspected by UT.
47
P Jl