ML20245A103
| ML20245A103 | |
| Person / Time | |
|---|---|
| Site: | Limerick  |
| Issue date: | 04/30/1989 |
| From: | STONE & WEBSTER ENGINEERING CORP. |
| To: | |
| Shared Package | |
| ML20245A071 | List: |
| References | |
| NUDOCS 8904250050 | |
| Download: ML20245A103 (900) | |
Text
{{#Wiki_filter:O LIMERICK GENERATING STATION - UNIT 2 INDEPENDENT DESIGN AND ODNSTRUCTION ASSESSVIENT O DESIGN OBSERVATION REPORTS DOR-001 TO DOR-039 O d STONE & WEBSTER ENGINEERIN
- ,,= =33,
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK UNIT 2 - J.O.NO. 18138.00 INDEPENDENT DESIGN AND CONSTRUCTION ASSESSNENT Ot 'IDCA VOLUNE II BOOK 2 DESCRIPTION This Volume II Book 2 contains the Design Observation Report (DOR) packages for the DOR-001 to DOR-039 issued by SWEC during the Limerick 7 Independent Design Assessment (IDA). Each DOR package contains the followings o DOR PART I: SWEC Initiation o DOR PART II: Bechtel Response o DOR PART III: SWEC Response Evaluation O Where necessary, selected reference documents have been included with the DDR packages to insure a clear understanding of the DDR and its resolution. i l O o
ST0lt A H WEBStil IBG!IttillG C0tP0tlfl0i LI!!!ICI HIf 2 - J.D.10.18138.00 I H IPII H RT M SIGN A H CONSTIUCf!0f ASSISSE H f IDCA101,0E! IIB 0012
- MSIGI OBSiti17108 ftP0tf TH IBMI L l l0BSitilIl l
! TAB lltP0lfltl tifLE/MSCt!Pfl0iOf03SEtufl0iitP0tf l-l10.lIDEntlT! l l l l l l l-
!.....l.......:..l..........................................................................................l l l l l l l l 1 lD01-00l!0 lCiflL/ Sfl00fDi&L MSIGI Cilfit!A(SPEC 8031-C-!!5) MSCRIPHCitS & OEIS$!0ES l L l 2 lD08-002l0 lttACf0t BLDG BAf CALC 23.4 IIC0lSIStilf/HC0EPLift ; D0tsf*f HB0ISftAtt MSIGI AMG. l l 3lD01003!0lLACLOfCLIAIDIIRCf!0fFORfitDESIGIOf!!PAIS!0IAICHORS !
l 4lD08-004l0lLACI0FAPPR0ftlift180E11155f0IIS0LitC0lltifSBffitCitttttOfACALCRATI0I l l 5 lD02-005l0 lfIllft ILtutif 10HL 058011 CALC 22.2L66 APPIAIS TO Gift D10015t171f1ft ttSRf5 l l 6 lD08-006l0 lIIST LOCATI0I H GS FOR P H 51-2I0588 Ilf! H SCitPAIC!lS, IICORSISftICitS & 0H1551015 l l 7 lD08-007j0 lDIf!CIllCIIS Il flAct&BILiff of I D10 LOGICAL M SIGI PARAltftiS l l 8 lD01-008l0 lDIFICIIICitS Il f?It ftACIABILIff filt C0DLD CAUSE fitt PR0ftCf!05 DESIGI Pt0BLIES l l 9 lD08-009l0 l$tlSHIC ACCittlif!015 U$tD Il fit HS!GI Of BLOCl HLLS HAT BI HDittSf!HftD l' l 10 lD01-010l0 lIIS0fflCitif C0fft0LS TO IISitt CALCElfl015 Att C0lS!Sftff WIfi AS-HILf DtHIIGS l l 11 lD08-011l0 l CALC 23.1 DESIGI Of D-Lift IALL IIDHOAft FOR ftBSILI LOADS fl0E St!SHIC UtifS l I ! 12 lD01012!0 lHC05fitEID 8tfl0AS a ASSDEPf!0IS Il IfAL Of HIS$lt IIPACf Of SPtif PORD POEPHODSt 800f l l 131901013l0 lB!,AST fitSSURIS 10f CORSIMID II ITAL Of IISSLI D0021S3, CORD BI A SIGI!f!CHf LOG l l 14 lD01-014l0 lIISHilCIHf IIGIIttt!IG JOStillCAfl0i !$ P90f!HD II fit DISPOSit!05 Of ICR'S ! l '.15 lD01615!0 lMf!CIIHf fit!!IC&fl0i of IDIlf!0f $!!!LDIIG C05PHit CODI CORD C10$t DESIGI PROBLtBS ;
; 16 lD0t-016l0 l CALC 22.3 tit! LOADIIG: flACIABILiff DillCIIICits CORD CHSt HS!GI PROBLtBS l l 17 lD0t-017l0 l SPIC C-115 ' AC15 MtICfl01 TOR ALL0HBLt NILD SfttSSES FOR lACf0ID LDAD l
_l 18lD0s-018l0l CALC 6900t.14lifLICTS 111000t178 Alficiff RAfilGS 10t 21750 ICE PtilfRAf!0f COfHCTORS i l 19 lD0t 019l0 lP0f tlf!A!, DillCIIICT li fHI MSIGI Of HLDS Of TUBE Sf!!L f0 TL&f SHIACES l l 20 lD08 020l0 l CALC GBB 218-026 HC0titCfLT 50DELS 8.C. f02 CDISIHtiflM Of HtPHG StitSS l l 21 lD01 021l0 !!Sf0 LIS Htif Of B0P Q-f HCf!0IAL SifPOIITS & 70LIEARCES IS 10f CONfl0LLED l
. ! 22 lD08-022l0 l$PIC E-170 HC00::CfLY CLASS!$ ACCOIRATOR 74115 AS PIPHG, DSI iPT OtifS ItQ'D DSI LODS l l 23 lD02 023:0 l CALC St-69(131) COPPDfti COM AttillCIALLY CHIGHG IRifIAL C0fDif!085 50i-COR!ttuf!ft l l 24lD02-CH;0lCALCS17-C&!!3.4.8(ITALBAf!05OfP0OL$#tLL)C0ffA!! tit 0t3filtCORDAfftCfDtSIGI l l 25 lD0t-025!0 l CALC H !! 15 (til!N POEP TH) HS DillC!tiCIIS THAT CORD CHSt C0lfDS!05 & DESIGE IR1025 l
- l. 26 lD0t-826l0 l CALC 102.1 ItGLICTS Staffit!D GAHEA C0ift!!Gt!0I 70 TOTAL D0st titt if CHICLEl D00tui
! 27 lD01-02fl0 lI!SUff!CItif HEAIS TO ID APPLICAMI CALCS, HICH COULD CHSt fifHt DIS!GI Mf!CIHCitS l
'; 28 l DOR 028 0 l CALC 210-83 DOES 10f MB0lsfliff HIQUACf Of ALL IBftGRAL AffACistifS II ITS B00 Hit!!S l l 29 lD02-029l0 l SPIC H-12 (IHRSW PHP) 00lfills It!!CitfCI!S filf CORD Lia TO HSIGN PROBLIESl l
l 30 l DOR-030l0 lfSAR C08tAIIS ft!f HD TLEIS tit! IHCCDAICl!S ffAf BAf itSnf If HSIGN !!!025 l 31 !D08-031:0 ltEl POEP M SIGN DOES E T COEPLY WITE ISAI C0EElfEtif 70 RG 1.1 IPSI RE Witt8ttf5 FOR ICCS l l 32 lD02-032 0 lCLIC E-12-30 (EH5N POEP IPSB) C0171155 MilCIIICitS THAf CORD LLO 10 DESIGE PRDELES l l 33lD01-033l0l CALC 200612802 HHitSfiEAftS BLOHDH fl.0N & gifts IOR-CONS!!UfRt COEPltfEHf TEEP l l 34 lD08-034l0 l CALC 26.3 EffE0D Of AHLf!!BG Glitt!C IE!DHD MAft DOES IDT APPEAR 70 H APRICint l l 35 lD0:-035l0 l CALC !!3.08 LAC 15 IHO TO PR0ft AMHACf Of PIDEsfAL LIHB f02 AffACED Pl Aff0tB LODS l l 36 lME 036l0 lHACTOR EH $!!SEIC 50HL ClifAIIS StittAL ASSCEffl0f3 fHf BAf I0t Et APPROPRIATE l JD-LIH ELL IHLUli!05 Of POST-ACCIDiff n00HNG MD 107 CORSIHR !!!SEIC LODS l O jl 37 lD03-037:038 lD02-038l0 lLOCA SHEllGH SfMCfHE 100 EtiH0D0 LOGY l 39 lD08-039l0 l CALC 6300t.13 Oftt!,00 HAf tt SELICfl0i P200tHit IS Mf!CIHf, C00LD CAU$t SPRIOUS f t!PSl l tif f0f H C0ES
- - - - - - -___a._-- . - - _ _ - - ._ . - _ _ . _ _ _ _ _ _ _ _ _ _ _ , _ . _ _ _ _ _ _ _ . , _
l.- l STONE & WEBSTFR ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) i 4 . OBSERVATICE REPORT l .L Observation Report No. D OR- 001 Rev. O_ Review Plan LK- D - 1902 -S Rev. 1_ Reference AI No.: DAI-001 DAI-004, DAI-005 DAI-010. DAI-018 PART I - INITIATION
- 1. . Description of Concern There are discrepancies and omissions in the Civil / Structural Design Criteria (Spec. 8031-C-115) that could lead to confusion and possible design problems.
l
- 2. Supporting Information A. The design loads on roofs due to the probable maximum precipitation (PMP) are not clearly defined. The design criteria refers to Section 3.4.2.3 of the FSAR. This section states that PMP loads can be up to 296 psf. but are less than the roof desi Fn loads. This could lead the engineer to ignore these loads, but it is not clear that roof design loads do exceed 296 psf. In addition, response to 4.I-001 indicated that this PMP load in the FSAR was in error.
B. The criteria do not clearly enforce the requirements in the Design Assessment Report (DAR) for fatigue evaluation of connections in electrical raceway supports. The criteria refer to specification 8031-C-98 for desi.gn of electrical raceway supports. Specification 8031-C-98 states that loading combination (b) of Section 6.3 (Appendix A) need not be evaluated. The omitted load combination is one included in the DAR. Section 6.1 of Appendix A of C-98 states that raceways satisfy fatigue requirements, but does not give adequate criteria. It is not clear . that there is sufficient direction to - ensure that the commitments in the DAR are satisfied. C. Allowable loads for Unistrut clamps are not clearly given in Specification 8031-C-98, which is referred to by the design criteria. FSAR Section 3.7.3.1.5.1 states that manufacturer's allowables can be increased by 50%, but this requirement is not clearly given in the specification. Section 3.0(e) gives allowable values for some fittings, . but it is not clear how to use them, or what to use for fittings not mentioned.
................................=....................=..===.===========================
Additional Documents Attached: None s ................................... ..................... fW { Yom W1/ Date W - YY$ Lead Engineer Signature APM Sigr/ture / Date l) Page 1 of g Y
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT Observation Report D OR- 001 Rev. O CONTIMIATION: D. There is no clear direction in the design criteria for allowable rock bearing - pressures. Re ference to reports are given in Section 4.1.6 of the criteria, but this section is under the category of
' loads.' It would be preferable ' if the allowable bearing pressures were given in Section 5.1. The fact that this can be confusing is evidenced by the response to Action Item DAI-010, which states that bearing pressures in FSAR Table 2.5-3A arc appropriate. These values are different from those in the report referenced in the design criteria.
E. Section 4.5 of the design criteria gives accident pressure loads, but gives temperatures only for the containment. There are obviously elevated temperatures associated with the other pressure loads. These temperatures are generally short in duration and have little effect on concrete structures, but could be significant for restrained steel structures. F. The design criteria states that seismic category I struccures are designed in accordance with FSAR Section 3.8.4. Some structures should be designed in accordance with Sections 3.8.1, 3.8.2, 3.8.3 or 3.8.5 instead of Section 3.8.4. This could lead to use of the l vrong criteria ior some structures. G. Section 4.1 of the design criteria ctates that yielding is allowed for structures subjected to blast losdings. This is not consistent with FSAk Tables 3.8-9 and 3.8-10, which limits steel stresses to
.9 Fy in bending and .85 Fy in tension. Response to Action Item
'DAI-005 indicates that the approach taken in the design criteria was accepted by the ASLB. To avoid confusion, the FSAR should be revised to be consistent with the actual design approach.
O o Page el of JP Y f- - - - - _ - - _ _ _ _ _ _ _ -
]
'a- STONE & WEBSTER ENGINEERING CORPORATION I
LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT J AN 3 01989 Observation Report No. DDR-001 Rev 0 PART'II -
RESPONSE
- 1. Observation Concurrence Concur with observation X Do not concur with observation. Discrepancies and omissions have not lead to confusion and design j problems-(See explanation in " remarks" below). J
- 2. Response to Observation:
A. Causal Factor (s): N/A B. Extent of Condition: N/A C. Significance: N/A D. Corrective Action: N/A E. Action To Prevent Recurrence: N/A
==========6:=====================================================
Remarks We do not concur.that the discrepancies or omissions in the Civil and Structural Design Criteria (Specification 8031-C-115) have led to any confusion or design problem. Any such discrepancies or omissions are.not significant with respect to design adequacy and commitment compliance and are not of concern. Detailed responses to the items cited by SWEC as supporting information are as follows: 1 2A. The Observation Report states the design loads on roofs due to probable maximum precipitation are not clearly defined. There is r.o discrepancy or omission in the design criteria (C-115) with regard to PMP roof loads. The design criteria (Section 2.3) states that " All' structures shall be desigr. 7d for Probable Maximum Precipitation
.(PMP) so that the safety of Category I equipment, systems, or components are not endangered, as 2.4.2.3 of the FSAR." The O described in Section design criteria (Section 4.7) also states that Page jL of h
_= --- _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ______ _
l.. 9 STONE & WEBSTER ENGINEERING CORPORATION bc#l70 LIMERICK 2 IDCA ( J. O. NO. 18138) 1 OBSERVATION REPORT Observation Report DOR-001'Rev 0 CONTINUATION: Seismic Category I structures and G-listed equipment shall be designed to withstand loading resulting from retained rainfall- water from a storm of Probable taximum Precipitation." , FSAR Section 2.4.2.3 describes the retained rainfall. condition and the resulting roof load magnitude. The FSAR clearly requires that design provide for the condition in which all roof drains and scuppers. are blocked resulting in water ponding up to.the top of the roof parapets plus a - distance. equal to the_ head required to provide PMP flow over the parapet. However, the example given in the FSAR is not applicable for existing roof 1(~
'q- parapet heights and is not intended to specify the roof load nor report calculation' results. The.
example in the FSAR is based on a parapet height of 4'-3", passing of the entire PMP _ flow over only a one foot length of parapet, and a total water depth of 4'-9". This example results in a load that exceeds considerably the load possible for existing parapet dimensions. The maximum Category I structure parapet height is 2'-3", and available parapet length over which rainfall may flow is much longer than one foot. It is evident that the PMP can be discherged over the parapets, which have a length very much more than one foot, with a he-d of no more than one inch. The F5AR i roof load value is clearly inapplicable and has been deleted by FSAR change notice, LDCN No. FS-1522. Roofs of all category I structures were evaluated in Calculation 25.4 Rev 0 (Transmittal P-0046, Aug 24 1988), The calculation shows that the roofs are capable to support rainfall ponding to a depth more than 2 inches above the top of the parapets. This conforms ~ to the F S /.R. Since the existing design was consistent with the- design drawings and the roof system was designed for the appropriate loids, it is evident there has been no confusion or design problem. Page 4 of nf
}K L
l qDoW7!3 i' STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA ( J. O. NO. 18138) e , OBSERVATION REPORT Observation Report No. DDR-001 Rev 0 CONTINUATION: 2B. The Observation Report states the criteria do not clearly enforce the requirements in the Derign Assessment Report (DAR) for fatigue evaluation of connections in electrier.1 raceway supports. .The report refers to the fact that the criteria does not Tequire evaluation of a loading combination included in the DAR, and does n-t give adequate criteria as to how the requirement that raceways satisfy fatigue requirements should be satisfied. There is no discrepancy or omission in the design criteria. The design criteria clearly enforce the DAR requirement (Table 5.9-1) that electrical raceway support connections be assessed for fatigue for the load combinations D+E and D+E'+SRV+LOCA. This is shown by the following
.(%
Q discussion. The design criteria require that seismic category IE electrical raceway supports be designed as required by Specification 8031-C-98. Section 6. 3 of Appendix A'of Specification C-98 requires that supports be designed for the loading combination D+E'+SRV+LOCA. Specification C-98 Section 3.0 b. states that the loading combination D+E (load combination (b) of Appendix A Section 6.0) need not be considered. The loading combination D+E is satisfied however as it is enveloped by the loading combination D+E'+5RV+LOCA. This can be seen by the fact that damping criteria and acceptance criteria for both conditions is identical (DAR Section 7.1.8.1 and Table 5.9-1, FSAR Section 3.7.3.1.5.3). The design criteria enforces the DAR load combination requirements, and there is no discrepancy.
/
f I.3 Page j[ of ;t7
/k
,9
\.
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.D. NO. 18138) OBSERVATION REPORT Observation Report No. DDR-001 Rev 0 CONTINUATION:
, Specification C-98 gives adequate criteria for fatigue requirements. .iection 6.1 of Appendix A requires that connections.be evaluated for fatigue by calculating cumulative usage factors and comparing to appropriate fatigue curves as discussed in ' Cable Tray & Conduit Raceway Seismic Test Program' Addendum to Test Report No. 1053-21.1-4, Vol. 3, May 1980, by ANCO Engineers, Inc.
The criteria clearly requires calculation of cumulative usage factors for the load combination D+E'+5RV+LOCA, and comparison to fatigue test curves. Detailed methodology for the fatigue evaluation is outside the intent and scope of the criteria. j L I,/'N) Raceway support connections are designed to meet the DAR load combination including fatigue considerations. There has been no confusion or design problems. 2C. The Observation Report states allowable loads for Unistrut clamps are not clearly given in Specification 8031-C-98, which is referred to by the design criteria. It is acknowledged that Specification C-98 does not explicitly give allowable loads for raceway support Unistrut connections. The lack of explicit connection allowables in the Specification is not a discrepancy that has lead to confusion or a design problem. The FSAR, wh.4ch is cited by the Civil and Structural Design Criteria (Section 1.1) as the governing document, requires that manufacturers' allowables increased 50% for SSE conditions be used for the connections. The FSAR criteria has been used for design. There is no confusion or design problem. Section 3.0 paragraph e of Specification C-98 lists specific Unistrut fittings identified by manufacturer's numbers. Although it is not explicitly stated, these specific fittings are
~}
~/ obtuse and acute angle fittings used to connect members intersecting at angles less than 90 0
Page & of V5
}<
{ I L' dbM-r70 9 STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA ( J. O. NO. 18138) OBSERVATION REPORT Observation Report DDR-001 Rev 0 l: l CONTINUATION: degrees. This is evident from examination of' data applicable to the listed fitting numbers in the Unistrut catalogue. The . specification states the allowable load on the connection for. supports using these specific fittings. The allowable loads apply to connection assemblies consisting of a pair of complementary fittings identified -by the catalogue numbers listed. These allowable assembly loads have been used in the design. There in no confusion or design problem. 2D. The Observation Report states there is no clear direction in the design criteria for allowable rock bearing pressures. .The design criteria does not. reference allowable. rock bearing pressures in (, Section 5.1, Materials. A foundation evaluation report by M. J. Adair, dated April 1972, that includes recommended rock bearing pressures, is referenced in Section 4.1. 6, Earth, Rock, and Hydrostatic Pressure. The title of this Section implies it applies to rock pressure allowables. The April 1972 report is also referenced in Section 1.1 which requires that design conform to the more stringent requirements of the SAR and listed references. The report recommends lower allowable rock bearing pressures'than the FSAR and has been used in the design calculations. The report does not reflect the extent of ' final excavation and treatment of the rock resulting in ] the larger allowable pressures described and ; Justified in the FSAR. A detailed discussion of rock bearing pressures and their relationship to the Adair report and the FSAR is discussed in the original and supplemental responses to DDR-2. Although the allowables used in the design are different from the FSAR, they are lower than the FSAR allowables and their use could lead only to conservatism and not to any design problem. 2E. The Observation Report states accident not given outside the O temperatures are containment. There is no discrepancy or omission Page jL of dds 0 l
O STONE & WEBSTER ENGINEERING CORPORATION dbM/W LIMERICK 2 IDCA ( J. D. NO. 18138) OBSERVATION REPORT Observation Report No. DDR-Oel Rev 0 CONTINUATION: in the design criteria. The criteria require that structures be designed for load combinations that include accident temperatures. Althoelh temperature values are not given in the criter a for accidents outside the containment, v,he temperature effects are required to be considered in satisfying the- load combination criteria. The design criteris need not define the detailed temperature v a,'.u e s , the effects of which are considered in the design. In general, accident temperature eff.ects outside the containment have not been evaluated as Judgements have been made that these effects would not . prevent . the structures from fu11 filling their required functions. These Judgements were not documented. Os It is acknowledged the criteria would have been enhanced by adressing the extent to which temperature effects should have been evaluated. To respond to SWEC's concern we have performed extensive analysis, as discussed below, to clearly demonstrate the design adequacy. Accident temperatures are generally short in duration and have little effect on concrete structures. In response to the Observation Report we have assessed typical cases of steel members subsected to accident temperatures, confirming previous Judgements that the designs are adequate for accident temperature load combinations. The assessment has confirmed our previous Judgements that the design complies with SRP requirements for accident temperature effects. The assessment considers, in accordante with NRC Standard Review Plans 3.8.3 and 3.8.4, Part II Section 3.c.(ii), that temperature effects on steel structures can be neglected when it can be shown that they are secondary and self-limiting in nature and where the material is ductile. Corresponding to the analysis, a FSAR chknge notice, LDCN FS-1592, and an addition to the Civil and Structural Design Criteria, Specification 8031-C-115 (Addition 1 to ( Rev 5) have been issued to incorporate the SRP provisions. There is no design problem with regard to accident temperature effects. Page JL of abk 83 l' )K
4 b Jcenn!3 STOME & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA ( J. O. NO. 18138) OBSERVATION REPORT Observation Report DDR-Oel Rev 0 CONTINUATION: 2F. . The 3bservation Report states the design criteria requires- seismic category I structures to be designed in accordance with FSAR Section 3. 8. 4, whereas some structures should be designed in acccrdance with FSAR Sections 3.8.1, 3.8.2, 3. 8. 3, or 3. 8. 5. The design criteria, in Section 1.1, requires that all structures be designed in accordance with the FSAR. FSAR requirements for the n design of structures are given in Section 3.8, including the containment and its internal structures in Sections 3.8.1 through 3.8.3, other seismic category I structures in Section 3. 8. 4, and foundations of seismic category I structures () in Section 3.8.5. Section 3.1.1 of the design criteria specifically identifies FSAR Section 3.8.4 instead of 3.8 for the design of seismic category I structures. This appears to be an inadvertent typographical error. The Observation Report suggests this discrepancy could lead to the use of the wrong criteria for some structures. If this were the case, the containment and its internal structures and foundations would have been designed to criteria applicable only to cther category I structures. This is not the case. Section 5.0 of the design criteria C-115 explicitly defines design requirements for the containment and internals. Similarily, upon refering to Section 3.8.4 of the FSAR it is readily apparent that this section does not apply to the containment. Plant structures are designed 2.n accordance with FSAR Section 3.8. The discrepancy has not caused any confusion and there is no design problem. O I) Page _1 of abf JK
4 TEL NO:215-841-4578 #299 P02
- FEB-07 '89 10:31 ID PECO PROJECT MGT DIV STONE G WEB 5TER ENGINEERING CORPORATION LINERICK 2 IDCA I J. C. NO. 18138)
DBEERVATION REPORT Observation Report DDR-801 Rev e CONTINUATION:
- 20. The Observation Report states that yielding allowed by the criteria for structures sub;ected to blast imedings is inconsistent with the FSAR.
There is no discrepancy in the design criteria. The allowanoe of yielding for structures subjected to railead accident blast leads was accepted by the ASLt. The design criteria and the design of structures for blast loads is consistent with this licensing commitment. However, the FRAR which limits stresses to non-yielding values, was not changed. A FSAR shange notice, LDCN No. FS-1542, p has been completed to permit yielding of structures subjected to the blast loads. The design criteria discrepancies have not led to confusion er design problems. It is soknowledged that there are differtages between the design criteria and the FSAR, and there fre' areas where the criteria could be enhanced. For examp17, Sechtel Engineering has revised Specifications se31 C-98 (Rev 11) and 8831-C-115 (Rev 56 P-0385, Addition 1) (Transmittals No. P a 275, Dec 14,1988 and Jan 23,1?89) , and has issued FRAR change notices LDCN Nos. FS-1522, F5-1542, and FS 1592. 338399 359883853S$W55559883358385938885555599855358c.88833333333 9 ...
,Eco R e.\UE=?Mn6.e......A;d.A,,g
,,gneu,e,.,.ate ...h 1 A,p,o a ., osa,.
e.,e a of a n e ict w sw um 411nH GE 50:rt sa, M wt
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT Observation Report D OR- 001 Rev. _Q, Review Plan: LK- D - 1902 -S Rev. 1
'PART III - RESPONSE ETALUATION The response to this observation report has been evaluated as follows:
A. Causal Factor (s): Acceptable Not Acceptable (Explain): NOT APPLICABLE B. Extent of Condition: Acceptable Not Acceptable (Explain): NOT APPLICABLE C. Significance Acceptable Not Acceptable (Explain): NOT APPLICABLE
%/
D. Corrective Action: Acceptable Not Acceptable (Explain): NOT APPLICABLE E. Action to Prevent Recurrence: Acceptable Not Acceptable (Explain): NOT APPLICABLE
=======================================================================================
Additional Action Required: y No Yes (Explain) SEE INSERT O ---- {.Ym, 1 b' b,Willt 2 flIi 5' f Lead Engineer Signature '/ Date AP$ Signature / Date i Page 1 of _" 0
fa b-Although the Bechtel response is marked 'Do not concur', this is to point out that no design problems have resulted from the omissions and discrepancies identified in the Observation Report. The response does acknowledge that these discrepancies and omissions exist, and , eropriate corrective action has been initiated to correct and/or clarify the design criteria, the FSAR, and applicable design guides and specifications. It is not necessary to perform an ' extent of condition' since there is only one structural design criteria and it has been reviewed in total. Other Observation Reports (e.g. DOR-017 and DOR-064) address other deficiencies in the design criteria, and may result in additional corrective action. l 0 DOR - 00 i REV O PAGE 1A OF w30 m _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ . . . . . _ _ . _ . _ . _ _ _ _ _ _ _
. . ~. fA4EIofI, spjef O un oe mevarious rem ueo r os.
- 0 0R - 0 01._. LTE M '1 s ' . . . . .
2- - . -. _ 9Acf. 22 b R-2.._.E.*Y 0 .
- 3. . cAtc.#___11 s.t_S Rev0 _. ..
)
. 'l- .cA lc 4. 2 2: 5 A- 5 R*v0 . .
3- _ cd4c f 2234-' A8V8 __ .. _. _. 6- . c
.d t c f. 22 a A- 7_ _$ E V 0 .
{ 7 ..c atc g __2 1.s R -9 A s v o_ _ __ _ _
\
o ..%.qc g___A 1 s R-1 Re v o c,4 .c 4 11.s R-to Reve i 9 .1 10 CALC 4. . . . . 22 3 R-fl RE Y o. - .. _ ..
//. .c44.c4 2.2.32/1 AEro
. 11- ... . . .. CAL (4 21S.$-9 _ _
$GV0 _
am e e e . e. e-g .em-.a _=en * . ma - e__ a.n-
** 'd.m .
W w e's.6 DE e.es e e . ew RD WD + _ . - - s es.. mwp- .-. p.@
. . &*w-_
l hpe.g.g_oggyg _ q me e_. meNh e em yeeD.h +-e.e gy me_e_ ws ---.gm.m.mg..
. " . ' .*G e af Emm...
{ O l 3 . n -_.- _ _ . . _ . . _ _ _ _ . . _ _ _ _ .pg- l3 g . 3__ __ . _ . . . . _ . _ _ . _ .
(.' V l STONE & WEBSTFR .:NGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) 7m V)' OBSERVATION REPORT . I Observation Report No. DOR- 002 Rev. O Review Plan: LK- D - 1903 -S Rev. [ l Reference AI No.: DAI-010, DAI-Oll, DAI-013 DAI-016, DAI-017 PART I - INITIATION
- 1. Description of Concern The calculation for the reactor building mat (calc. 23.4) is inconsistent and incomplete. It does not completely demonstrate the adequacy of the design.
- 2. Supporting Information i A. Not all loads and loading combinations are evaluated. The mat portion ]
was not evaluated for hydrostatic pressures due to ground water, and footings were not evaluated for load combinations containing SSE loads. It is not obvious that these loads do not govern the design. B. When calculating bearing pressures on the rock, an approximate method of analysis is used. Since the calculated pressures slightly exceed
^
/T allowable values, and another method used in the calculation gives h higher bearing pressure, it is not conclusively demonstrated that the design is adequate. Furthermore, the bearing pressure is calculated at the interface between the ' Class A' fill concrete and
, the rock. Searing stress is not checked at the interface between the mat and ' Class A' concrete.
C. There are discrepancies between the calculation and the design drawings. One of these differences (relatir.g to an increase in wall thickness and increase in reinforcing steel) obviously lias no negative impact on the calculation. The other one (12" of fill concrete under the mat instead of 18" assumed in the calculation) is not conservative and could affect results. There is a concern that drawings can be issued or revised without assurance that they reflect conditions assumed in the design calculations.
.......................................c..................................=.===..==.===
Additional Documents Attached: None f 'o n f5 ' Y APM S(gnature 7 / Date Lead Engineer Signature / Da'te Page1of,([ I
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138)
, OssanVATION asPORT Observation Report DOR- 002 Rev. L CONTINUATION:
D. The calculation designs column footings using Unit 1 parameters, and assumes . that Unit 2 is similar. In the case of column E-29, the allowable bearing pressure is less than that for the corresponding Unit I column. If the loads carried by the Unit 2 column are the same as those for Unit 1, the footing size is too small. There is a concern that the design adequacy of Unit 2 footings have not-been demonstrated in the calculation, which is based solely on similarity to Unit 1 O O < i Page 3. o f If
r SIONE & IEBSTER ENGINEERING CORPORATION N J .1 ) h /3 O b M SSUED DATE
%' Mg M' O SERVATION REPORT NOV 211988
.bOV 21 198 Observation Report No. DOR-002 Rev. 0 PART II - RESPONSE
- l AMENDED COMPLETE RESPONSE l
- 1. Observation Concurrence:
X Concur with observation. The calculation, by today's standards, does not in itself demonstrate adequacy. The degree of documentation judged necessary when the calculation was originally prepared during the period frca 1971 to 1972 was sufficient to show design adequacy, and as discussed below, does provide the basis for demonstrating the design adequacy of the reactor building foundation mat-Ib not concur with observation (Note: if not in concurrence, explain in
" remarks" below)
- 2. Response to Observation:
A. Casual Factor (s) Engineering judgements were made that the impact of changes in the design ground water levels, allowable foundation bearing pressures, and drawing details did not warrant revisions to the calculations. It was not deemed necessary at the time to document the judgements. Preently there has been more emphasis for formally documenting many judgements and evaluations. B. Extent of Condition i In general, durirg . m design developmen and the construction process I and during the licensing process including the preparation of the FSAR, judgement has been exercised whereby calculation revisions to reflect minor changes, and other charges that have no inpact on the j design, may have been deemed to be unwarranted. These judgements may not be documented. With specific regard to the change in design ground water level, the condition applies also to base mat calculations for the control structure, a portion of the spray pond pumphouse, the radwaste building, and the turhine building. With regard to referencing the final, higher, allcMable rock bearing pressures, the condition applies also to fcundation calculations for all other safety related structures. With regard to considering drawing details of the concrete fill thickness, the condition does not apply to calculations applicable to any other location in the reactor building or to any other building. C. Significance The reactor building base mat and foundation design meets allowable stress limits when considering all applicable loads and loading combina-p tions, as required by the FSAR. In particular, the mat is adequate to g withstand hydrostatic pressure resulting frca groundwater levels given l in the FSAR, foundation bearing pressures are well witnin allowable l limits given in the PSAR, and footings were evaluated and are adequate l for SSE load combinations. The concern therefore has no significance. Page 3 of _ll _
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA [ (J.O. NO. 18138) ( . OBSERVATION REPORT Observation Report D OR- 002 Rev. O_ CONTINUATION: The following is provided as a detailed response to the supportirg information cited in the Observation Report, showing that the concern regarding adequacy of the reactor building mat has no significance. 2A. It is acknowledged that the calculation does not include evaluation of the mat for hydrostatic pressure due to ground water levels. The water levels were established after the calculation was originally prepared. For completeness the calculation has been revised (Calculation No. 23.4 Rev. 6) showing that this loading condition does not govern the design. Footirgs were evaluated for SSE loads and loading combinations. Footing bearing pressures resultirg separately from dead, live, OBE, and SSE loads are calculated on sheets 16 to 29 These pressures include applicable load factors for the OBE and SSE load combinations. Although not explicitly stated in the calculation, combined pressures for the OBE load combination govern over those for the SSE load combination. Accordingly only the OBE load combination combined pressures are calculated and used in the footing design calculations.
- 28. The method of analysis used to calculate bearing pressures on the rock beneath walls is acceptable and sufficiently accurate. It is acknowledged the calculated pressures slightly exceed the allowable values used in the calculation, but the exceedance are less than 2.5% and therefore are not significant. Two sets of
- computations are included in the calculation. The first set does not take credit for presence of c. crete fill beneath the mat and results in higher bearing pressures. The second set, which is more representative of the design shown on the drawings, takes credit for the concrete fill and is the basis of the pressures that exceed the allowables used by up to 2.5%. The calculation uses considerably lower allcwable bearing pressures (42 to 60 ksf) than those provided in the PSAR (90ksf). The lower allowables used in the calculation do not reflect the final excavation and treatment of the rock foundation that is described in the PSAR.
It is clear that if the allowable pressures given in the PSAR are used there can be no question of the adequacy of the wall foundation design. Furthermore, even if the lower allowables identified in the wall foundation calculation are used, the calculation conclusively demonstrates the design is adequate. Bearing stress on the Class A fill concrete at the fill - mat l interface is not checked and need not be checked as the fill concrete has considerable greater compressive strength than the l
/O rock. Bearing stress (pressure) is checked (evaluated) at the
' interface between the mat and the Class A concrete for use in the i design of footings as' discussed in "2A" above.
Page<11 of g q L _ _ ____ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _ _ _ _ __
]
STONE 6 VEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) \,.J OBSERVATION REPORT 73
)
Nd Observation Report D OR- 002 Rev. 0_ CONTINUATION: 2C. It is acknowledged the calculation uses an effective footing width, along part of the wall alorg column line D, that corresponds to a concrete fill thickness that is 6 inches greater than the thickness shown on the drawing. He calculation uses an effective footing width of 11 feet as compared to a more correct width of 10 feet. If the 10 feet width had been used, calculated bearing pressures would have increased an insignificant amount. Even for the lower allowable bearing pressures used in the calculation this minor discrepancy is of no significance. Drawings are revised and issued only after specific approval of Project Engineering to assure that design control procedures are met. R ese procedures include the requirement of Engineering Design Procedure 4.46, Section 'i.2, which stated, "When changes are made to approved drawings the responsible engineer shall review the possible impact of the changes on associated calculations, and by excercisirg appropriate judgement, determine whether it is necessary to revise the calculations". There is no requirement that calculations correspond exactly to the drawings
.3 when the resulting difference is judged to have no significant
() impact on the calculation. De increase in wall thickness and reinforcing steel was a field change that was reviewed by engineering and was determined not to warrant a calculation change. he charge was documented as FCR C-93 which was dispositioned and approved in accordance with project design control procedures. 2D. We calculation designs Unit 2 column footings based solely on similarity to Unit 1. It is acknowledged that the calculation does not include evaluation of column footings in a local region of Unit 2 where the allowable bearing pressures used in the calculation are lower than in Unit 1. This omission is of no significance with regard to the footings design adequacy as allowable pressures, as given in the FSAR, are identical for both Units. We have reviewed drawings, showing that the hydrostatic loading condition is not applicable to the Containment, the diesel generator building, , nor to the subgrade pits, manholes, and tunnels which contain safety l l related conponents. Base mat calculation revisions for the safety related reactor building (23.4 Rev. 6), control structure (23.4 Rev. ! 6), and spray pond punphouse (70-M Rev. 4), and for the non-safety related radwaste building (41-H Pev.1) have been ccrnpleted to include l n groundwater pressure evaluation. Review of the turbine building shows there is no safety impact as a result of the groundwater condition. l l Page 1 of ff i
NOV-22 '88 16:32 ID:PECO PROJECT MGT DIV TEL N0:215-841-4578 #777 P02 l-STONE & WE587RR INCINEERING C0AF0 RATION O U*'0"" '& . JL)cAIrro I 03s5NVAT105 REPOE! l 1 observation Report No. LORE nay, ,,g l6[ ! P&tt II - R5870555 00ftIEMD 1 :.'aQ. f- ., . ; A revier of foundation bearing pressure calculations, concreta
. drawings, and rock excavat. ion limits has been performed. The i
- review shows that no credit has been taken for concrete fill i~ thickness greater than that present at any safety related. structure other than the reactor building. The reactor building base met calculation has been revised (23.4 Rev. 6) to conform to fill ar;
- thickness shown.
t;;
. Foundation calculations for seismic load combinations for all safety related structures have been revised (15-L nov. 2, 23.4 Rev. 6, 42-3.4 Rev. 3, 42-A Rev. 6, 70-M Rev 4) to identify
~
411cWeble rock bearing pressures as those given in the FSAR. The
' limits are the same at both Units 1 and 2.
D. Corrective Action In recognition of the calculational improvement areas identifled by O SWEC, Bechtel has revised the calculations to document the previously
, acceptable engineering judgements as discussed in section C above. No .
I
' corrective action is required.
E. Action To Prevent Recurrences No action as required because present project design control procedures contain sufficient requirements for the development and control of design calculations, including the requirements that the calculations be conplete, s,upport the adequacy of the design, and be reviewed for ' the inpact of any subsequent drawing changes. Engineering management has also enphasized the need to fornally document engineering reviews and change authorizing documents and their potential inpact on existing calculations and related documents.
=..........n......n...............u..=.===.....--.=.=....=.....-.o Remarks None
- -& -)- - - %
O l-l0 n ib ket-
'FEcoasflswdign(ture
- o, M n-
/ Date
. E d L l M tile d n sashtel Apprevl1 sisaatire / Data ]
l Ta ge .Q,, a t .l,5 ,
.. . I
s . . . J.O. Ns. 18138.00 Limerick - Unit 2 1DCA
.( ~
N) LIMERICK IDCA (J.0.No. 18138.00) TELEPHONE RECORD SER: D TR- 001 Review Plan: LK-p-1902-S Rev. 1 Call Date: / R __jj_gjL Time: J:_gn jug /PM Incoming: Outgoing: A Participants Names St t .- : L.. Wille, B. Ebbeson Bechtel: S. Loo, E. Hughes, S. Sobkowski, C. Haines, H. Minkowksi PECo : Bob Brown Other : R. Parkhill, Ha.i-boh Wang
Subject:
DOR-002 g b Summary: The call was made to discuss Bechtel's rerponse to Observation Report DOR-002, in order to better understand the issues. The specific topics discussed are summarized in Attachment 1. Discussion will be sutunarized by item number.
- 1. Bechtel stated that higher allowable bearing stress values were iustified because post-excavation inspection and repairs to the rock imnroved bearine conditions. Reference was made to FSAR section 2.5.4.12 and a Bechtel report of Julv 1974.
Bechtel vill provide supplemental information to iustify the resposne. _ _ . . _ _ (. h , NI L Originator AsWistant Program Manage DOR - o o t REV o PAGE 7 OF 15
- 2. SWEC and Bechtel agreed that the method is approximate, and is appropriate if_the higher bearing stresses are allowed.
'([]) Resolution of Item I will resolve this item.
l 3. After discussion, it was agreed that the load combinations with'SSE and OBE give similar results. However, it was noted that the analysis on pages 29-1 to 29-15 appear to give results higher than those on pages 16 to 29 for the 'D' line walls. Bechtel will evaluate this issue further and advise SWEC.
- 4. SWEC pointed out that the allowable bearing stress on the fill concrete proposed by Bechtel is higher than the ACI-318 allowable value. Bechtel agreed to provide supplemental information.
O v ( . DOR - oo l REV_ o PAGE % OF I6
Attachmrnt 1 Telephone Conference Agenda Thursday, 12/8/88 1 PM EDT Related to Bechtel Response to DOR-002,
]
G'
- 1. The response to DOR-002 indicates that the all'ovable bearing pressure used in Rev. 4 of calculation 23.4, obtained from
' Report on Foundation Evaluation for LGS' by M. J. Adair, is not appropriate, and that allowables given in FSAR table 2.5-3A could be used. Note 2 of the FSAR table refers to information from Dames and Moore, 10/5/70 and 8/19/71. Revision 6 of calculation 23.4 also refers to there sources of information.
The report by Adair was issued in February of 1972, and it also references the Dames & Moore report. It would appear that Adair's recommended values were made based on knowledge of the Dames and Moore information and is more applicable. If it is now felt that the Adair report is not applicable, justification should be provided. Page 7 of the Adair report indicates that bearing capacities could be increased after excavation and examination of the rock, but the response provides no evidence that this was done.
- 2. The response indicates that the method used to compute bearing gg pressures under the rock is appropriate. The concern raised is that the loads or the footing are calculated using the
( V) area properties of the footings, not the walls. On pages 29-1 through 29-15 the area properties of the footings plus an area of concrete fill are used. With this method, the total load imposed on a given footing can be changedjby changing the size of that footing, or even other footings. This has a direct effect on the calculated bearing pressure at a given location.
- 3. The response states that SSE loads have been considered in the design of the footings, and that OBE stresses govern, although this is not stated in calculation. If one were to take the bearing stress for the D+L+SSE case on Sheet 29-13 at pt. 2 ('D' line wall) of 57.46 KSF, based on an 11' footing width, and factor it by 11/7 to get stresses at the footing to fill-concrete interface, the stress is 90.3 KSF. This l is greater than 71 KSF at this interface that is based on OBE loads and is used to design the footing on Sheet 61.
l l 4. Revision 6 of calculation 23.4 was issued to resolve the concerns of DOR-0,02 and gives an allowable bearing stress on the fill concrete of 288 KSF. What is the basis for this allowable bearing stress? [VD DOR - o o n REV o PAGE 9 OF 'S
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA ( J. O. NO. 18138) OBSERVATION REPORT Observation Report No. DDR-902 Rev'O 1 PART II - RESPONSE EVEELEMENTAL RESPONSE. Response to items listed in Attachment 1 to Telephone Record DTR-Oel are as follows:
- 1. The excavation'and treatment of the rock described in the FSAR and documented in a July 1974 Bechtel report substantiate the allowable rock bearing capacities (equivalent to 60 ksf for normal loads, 90 ksf for normal plus dynamic loads) given in Table 2.5-3A of the FSAR.
The ' Report On Foundation. Evaluation For Limerick Generating Station by M. J. Adair, dated Apr11'1972, contained recommended. allowable bearing capacities O which were acknowledged to be quite conservative, and-also indicated that these recommended' capacities might be safely increased to higher values bearing depending upon foundation conditions as revealed during
-construction. The report also gave recommendations for treatment of fracture zones and clay seems known to be present in the foundation rock.
Foundation treatment was carried out in accordance with the April 1972 report. During construction, excavation was carried out to sound unweathered rock as described in Section 2.5.4.5 and as shown in Figure 2.5-37 of the FSAR. Fracture zones and clay seams were treated as described in Section 2.5.4.12 of the FSAR. A July 22, 1974 Bechtel Report On Treatment Of Fracture Zones At Limerick Generating Station (Transmittal No. P-0273, dated Dec 13,1988) documents the treatment of the fracture zones and the clay seams.
- 2. As discussed in item 2 of Telephone Record DTR-001, the method used to calculate bearing pressures on the rock is approximate and appropriate as the higher bearing pressures, given in the FSAR, are allowed. Since there is significant margin between the maximum calculated bearing pressures and the high allowable rock bearing the method is appropriate O pressures given in the FSAR, and the existing design is satisfactory.
Page ik of j$
l I l .
&ck K7Y i
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) I OBSERVATION REPORT i Observation Report No. DDR-002 Rev 8 i PART II'- RESPONSE SUPPLEMENTAL RESPAMSE l
- 3. SSE loads were considered in the footing design. On sheets 16 to 29 factored bearing pressures acting on the bottom of the footings are calculated for the
. individual load components, D+L, OBE, and SSE. The.OBE load combination is slightly more than the SSE- load )
combination.. The calculation ststes on sheet 29 that the OBE loading co'4binatics acverns. Accordingly, onlyL l the OBE loading . combination combined pressures are 1 calculated, as shown on sheets 27 to 29.'The pressures ; shown on sheet 29 are enveloping pressures. Thus, for I example, the maximum pressure along wall'line D is 71 l kaf which is the pressure used, appropriately, to j design the footing on sheet 61. !
.)
The method used to calculate rock bearing pressures on O sheets 29-1 to 29-15 (P/A + Mc/I) results in different forces, per lineal foot of wall, at the footing-concrete fill interface and the concrete fill-rock interface. The resulting pressures at the two interfaces are not related linearly. By contrast, the discussion in item 3 of Attachment. 1 to Telephone Record DTR-801 considers the forces at the two interfaces to be the same, and results in a linear relationship between the two interface pressures. Thus, footing pressures derived by linearly factoring the calculated rock pressures, which are not linearly related to the calculated footing pressures, will always be different from the calculated footing pressures, and will not yield correct ~ values of footing pressures. This is amplified in the following paragraph. I For a structure such as the Reactor Building, in which the arrangement of walls and footings is not uniform, well forces are redistributed to the footings resulting in a difference between the lineal force at the bottom of the walls and that along the bottom of the footings.Similarily, the calculation on sheets 29-1 to 29-15 considers that footing forces are redistributed through the concrete fill to the rock, resulting in a O difference between the lineal force at the bottom of the footings and that at the rock surface. For wall Page JL of 15
1
..JAN-25 '89 14: 19 ID:PECO PROJECT MGT DIV TEL NO:215-641-4578 #170 P02 l
O GDCA N77 STONE & WERSTER ENGINEERING CORPORATION LINERICK 2 IDCA ( J. D. No. 18138) 055ERVATION REPORT Observation Report No. DDR-902 Rev e PART II - RERPOWRP ppPPLEMENTAL_RESPDNEE line D, a footing pressure'of 71 kaf is calculated en sheet 29. This pressure is asleulated by en appropriate method (P/A
- Mc/I) and is correct. Using the same
*ethod, a rock bearing pressure of 57.46 ksi is c.1-ulated on sheet 29-13. '
It may be true that formes at the festing and at the reek should be considered the esmo. However, to back calculste the festing pressure by LLnearly"'fsetoring the 57.46 ksf reek bearing pressure by the bearing vidth retic, 11/7 , will obviously give a different result (90.3 kst) and is not correst. Rather, in such a O case, the footing pressure would be 71 calculated on sheet 29. and the maximum rock pressure kaf as at line D veuid not be 57.46 kaf as calculated but would be 71(7/11) = 59. 5 hat. At other locations, where no concrete fill extets, rock pressures would be the same as the footing pressures calculated on sheets is to 29. Rock pressures se calculated would be within the FSAR limits.
- 4. 7he 248 kaf value referred to in the calculation represents the compressive strength (2 kaal of the fill concrete rather than the allowable bearing stress as stated. The calculatten has been revised to correct the statement (Calculatten 23.4, Rev. 7, transmitted Jan 23, 1989, No. P-23es). No comparison of bearing stress en the fill sencrete with the ACI 318 allevable is made nor required as bearing en the fill concrete is controlled by the much icwor maximum allevable rock bearing pressure (90 kat).
s s eas s e s sss s s s s ss s s s s s ss ss s s ss s sss e m eas s a ss e s s e s e s s e m a n e s.e s es se s
- e. A A = a. = = = a. e .-
PECe view Bignatur /Da e 8echte Appr al Signature /Date ra e 11 et as
1 STONE & WEBSTER ENGINEERING CORPORATION l LIMERICK 2 IDCA ( J. D. NO. 18138) MAR 0 21989 I OBSERVATION REPORT 1 Observation Report No. DDR-002 Kev 0 l PART II - RESPONSE SUPPLEMENTAL RESPONSE Response to SWEC supplemental item regarding method used on sheets 16 to 29 of calculation 23.4 to calculate pressure at the bottom of footings is as follows: The reactor building is a rigid box structure. The footings are not wide in proportion to their depth, i. e. the footings do not project beyond 45 degree lines emanating from the bottom of the wall faces. The footings are integral with and form a part of the rigid box structure. The footings are in effect an extension of the walls. For an assymetric and non-uniform arrangement such as the reactor building, vertical forces along the wall will not be constant at different elevations. The forces will redistribute by fx the deep beam action of the walls and footings acting together as ( ) a unit.
\,_)
The method used in Calculation 23.4 is similar to that used for a rigid mat footing. (Reference Foundation Design by Teng, pg 77). In the case of the reactor building, the design does not consider presence of a rigid mat, but the rigid wall / footing system exhibits similar behavior. It is noted that pressures at the bottom of the walls are calculated in Calculation 22.2c(k). If wall forces the footings, based on these the pressures are used, as suggested, to design pressures will still remain below the capacity of both the concrete footings and the rock foundation material, as shown by l the following example for the D line wall. Vertical pressure at bottom of 4 ft wall (el 177), from Calc 22.2c(k) Rev 7 Sht 18 for 1.25 (D + L + OBE) = 182.4 ksf Vertical force along bottom of the wall = 182.4 X 4 = 730 k/ft Vertical pressure at bottom
= 730/7 = 104 ksf of 7 ft wide footing
}
i fx l
<,) 1 Pa a l} of LS l
215327567e P.10/16
. t%R CE '69 12: 07 BECH'CL 3RD FLOOR D
o J3cdlo STQNE & WEOSTER ENGINEERING CORPORATION LINERICK 2 IDCA (J.D. No. 18138) OBSERVATION REPORT Oteervation Report No. DCR-ee2 Rev e PART II - RESPDMRt ELPfLMEKTAL RE3PM Moment in 3 ft thick footing projecting 2 f t beyond face of wall . 1/2 X 194 X 2e a geg gg.k Momsnt capacity of the femting with # 19 9 4 reinforcing bars (Ref Best F/Dwg C-174)
- 257 ft k > 298 Unf actored vertical f orce along bottom of vall (D + L
- QBE) . 73e / 1.25 a 544 k/ft Bearing width en rock, conside-ing a 12' depth of concrete fill = 7*2
- 9 ft Rock bearing pressure = 586 / 9 a 64.9 kaf 4 99 kaf allowable
.......s............. n...... ..s.............sa........ssenes...
Y VEco R 1ev Signature /Date
.,W -.Y ---- - -$-
techtel Approval Y. - --~' ignature Date
~ E* b Page ,d, of .!5.
STONE & WEBSTER ENGINEERING CORPORATION l LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT Observation Report D OR- 002 Rev. O Review Plan: LK- D - 1903 - S Rev. [ PART III - RESPONSE EVALUATION The-response to this observation report has been evaluated as follows: A. Causal Factor (s): X Acceptable Not Acceptable (Explain): The response states that judgements were made which were not documented in the calculation. B. Extent of Conditions X Acceptable Not Acceptable (Explain): The response indicates that all affected foundation design calculations have been reviewed and revised. C. Significance: X Acceptable Not Acceptable (Explain): [N Information provided in supplemental response (1/23/89) indicates that
\ higher rock bearing pressures are acceptable, and the design is adequate.
Supplemental response of 3/2/89 demonstrates that the mat design is adequate. D. . Corrective Actions X Acceptable Not Acceptable (Explain): All affected calculations have been revised. E. Action to Prevent Recurrence: X Acceptable Not Acceptable (Explain): No more calculations of this type are expected. Engineering management has emphasized the need to document engineering judgements.
.....................................................==..==.....=====.=================
Additional Action Required: X No Yes (Explain) 1
}L f. CL,n 3///M VAlitt -L D.WiIle alult1 Lead Engineer Signature
'/ Date AWM Signature / Date Page 6 of ,lg
l STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA l (J.O. NO. 18138) f/' OBSERVATION REPORT l. Observation Report No. D OR- 003 Rev. O_ Review Plan: LK- D - 1903 - S Rev. ,J_ Reference AI No.: DAI-021. DAI-023 PART I - INITIATION
. 1. Description of Concern There is a lack of clear direction for the design of expansion v.nchors.
- 2. Supporting Information -
There seems to be two conflicting documents for the evaluation of expansion. anchors, design guide C-2.40 (which is the one stipulated by the Civil / Structural design criteria) and specification C-64. Two instances have been observed where differences between the two documents have resulted in confusion:
'[\ '
- Design Guide C-2.40 states that shear and tension should be combined using a. linear interaction equation, while specification C-64 allows them to be combined vectorially (i.e. SRSS).
Design Guide C-2.40 gives reduction factors for anchor spacing
-* 1ess than the minimum, while C-64 does not.
Review of calculations indicates that the project is following C-64. In the case of reduced spacing, new allowables were calculated, since C-64 does not give adequate direction. The allowables calculated conflict with those in C-2.40. It is recognized that C-2.40 states that allowables therein can be changed if' test results demonstrate otherwise, but there is no evidence of any testing which would alter the design allowables. i i ' It is -not clear that the design basis has been adequately defined and implemented. None Additional Documents Attached: D O ,
& cCu nMa to W!r
/ Date l
Lead Engineer Signature / Dat'e APM Sighature Page 1 of 10 )
)
S'IONE & WEBSTER ENGINEERING CORPORATICH LIMERICK 2 IDCA WV29 M (J.O. NO. 18138)- OBSERVATION REPORT i g Observation Report No. DOR-003 Rev. 0 PART II - RESPONSE
- 1. Observation Concurrence:
X Concur with observation. Do not concur with observation (Note: if not in concurrence, explain in
" remarks" below)
- 2. Response to Observation:
A. Casual Factor (s) In 1979, the Civil / Structural Design Criteria had adopted Design Guide C-2.40, Rev 0 as the expansion anchor design criteria. 'Ihe installation and testirg of expansion anchors were required to be in accordance with Specification C-64. Early in 1982 Specification C-64 was revised to include allowable design loads for the anchors. The revision was made to provide a project specification that could be used by all disciplines to define allowable anchor loads. The allowable loads, including shear-tension O interaction, were identical to the allowable loads given in Design
-() Guide C-2.40, Rev. O. Reduction factors for anchor spacings less than the minimum were not included in the specification because these factors were not needed in the original anchor design process. The revision of the specification was reviewed and approved by the Chief Civil / Structural Engineer, ard thereafter Specification C64 was consistent with Design Guide C-2.40, Rev. O as there was no significant difference between the two documents.
As anchor installation proceeded, it becam apparent that to acconr:odate existing clearances it was expedient in same cases to permit anchor spacings less than the minimum, provided the resultirg reduced allowable loads would not be less than the design loads. Reduction factors for the reduced spacings were calculated usirg the overlapping shear cone method which is a nore refined method than the simpler and nere conservative linear reduction method given in Design Guide C-2.40. Specifications C-115 and C-64 were not revised to specify use of the overlapping shear cone method in lieu of the linear reduction method. Expansion anchor design criteria evolved in the industry during the period frcxn 1977 to 1984. tunsiderable testing took place during this time. As a result, additional information on the behavior of anchors became available, and Bechtel Design Guide C-2.40 was revised in Feb. 1984 to adopt an increase in the allowable shear loads, and use of a conservative, linear, shear-tension interaction method. There was no requirement or need to reflect the Design Guide revision in I Specifications C-115 and C-64 as these specifications provided a conservative and acceptable basis for design. However, in June 1984 Specification C-115 was incorrectly revised to require anchor design Page 1 of /6
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138)
'p OBSERVATION REPORT j u Observation Report 10R- 003 Re v . 0_._
CONTINUATION: to follow Revision 1 of Design Guide C-2.40. His revision introduced a difference between Specifications C-115 and C-64 as to the allowable shear load and the shear-tension interaction relationship. B. Extent of Condition The only design guide referenced in the C-115 design criteria is Design Guide C-2.40. All other design document references listed in C-115 have been reviewed by Bechtel Engineering and it has been confirmed that no other civil project specifications are inconsistent with the design document references listed in C-115. C. Significance. h e concern has no significance as both documents are supported by test data, and expansion anchor designs conforming to either document provide adequate factors of safety. SWEC cites two instances where Specification C-64 differs frcxn Design i Guide C-2.40. He followirg discussion shows these differences are ( refinements that are justified and are supported by test data.
- 1. Shear / tension Interaction Design Guide C-2.40 specifies that shear and tension be combined linearly, unless shown otherwise by test, whereas Specification C-64 requires the loads to be combined vectorially. The use of the vectorial method is supported by the followirg test reports which show that this method is more representative of tested behavior than linear interaction.
- a. Bechtel Power Corporation report "8856-C-101, Technical Criteria, Hilti HDI Shell Anchors, October 8, 1980".
- b. Teledyne Engineering Services report "3501-1 Revision 1, Summary Report, Generic Response To USNRC I&E Bulletin 's9-02 Base Plate / Concrete Expansion Anchor Bolts, August 30, 1979".
Since the vectorial method has a technical basis and the linear method is nore conservative, it is concluded that both methods are acceptable. Shear-tension interaction curves based on tests envelope interaction curves based on allowable shear and tension loads specified in C-64 O by a substantial margin. Le resulting high factor of scfety
\ makes the choice of a shear-tension interaction nethod to be of little engineering significance. The change in factor of safety resulting frcm the difference between linear and vectorial combination of shear and tension is small. j Page 3 of to
TEL NO:215-841-4578 #800 P02 NOV-29 '88 17:06 ID:PECO PROJECT MGT DIV I F.5/5
- tw !? '6313 EE BCXE WCETFS POWER 157 FLyp k
$70h*t & kTB5TER ENCIhTERING CORPOM:20N LIMERICK 2 3DCA U.0. No.18138) I 08tERVAT!05 EEP0tf dibc4tash Observat ten Report No. D_0ke{ Rev. O PART I! = R18P0s55 CONT _15DED
- 2. Reduction factors for Anenor spacing Design Guid6 C-2.40 requires use of a lirear method to cetermino reduced allowable loads for anchor spac! 4s less tnan the specified minirum. Specification C-64 does not address this subject.
Project calculations have used the method of overlapping shear cones, which gives larger allowable loads than C-2.40.
. The overlapping shear cone method is a mDre rigorous approach than the linear reduction method, and is supported, for exa@ le by the recorrnendations contained in a 1984 report by the ASCE Nuclear structures and Materials concittee titled " state of The Art Report on Steel Dtbedments". This report was prepared by an industry group that examined over 75 expansion anchor test reports.
ACI code 349-76 Appendix B also specifies the overlapping cone iO, method. As the overlapping cono method nas a technical basis and the linear method is note conservative, it is concluded that both nethods are acceptable. D. Corrective Action The Civil / Structural design criteria, Specification C-115, has oeen revised (C-115, Rev. S) to require that expansion anchor connections be designed in accordance with Specification C-64. Specification C-64 has been revised (C-64, Addition 3 to Rev. 27) to speelfy that reduced anchor all:wable loads to account for spacing less than the miniman apecified b6 determined by the method of over'apping shear cones. E. Action to Prevent Recurrence 16 action is required, as the revised Civil / Structural design criteria contained in specification C-115 refers only to specification C-64 for expansion ancnor design and does not reference Design GJice C-2.4C, and as the revised Specification C-64 contains req.11rements for detemining reduced allowable loads for roauced anchor spacings. me.es o n s.. . u s e....sm...... . ..a s a.m.m .no ..as .a m m.as es. . ....= e nse...= =s s.. .s a ma .. ... Pamarks None
...7.................................................................................
N. ' \ 2 *I SS . Y Uf2Yf$
/ Date Bechtel Approval $1gnatute / Da3e PECo Review $1gnature Page 1 of jo ._ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _- -
[ l AGENDA FOR DISCUSSION OF OBSERVATION REPOKT DOR-003 D)\ v SWEC's understanding of' anchor bolt usage at LGS is as fo: lows: l 1) The ' wedge anchors used are the ones no t ed ' in Section 2.2.1 of Spec. l C-64.
- 2) Sleeve anchors are used in block walls only. Also, Phillips sleeve anchors only were used.
- 3) Shell anchors were used only when specifically requested by Project.
Phillips shell anchors were primarily used, The following questions remain:
- What wedge anchors were actually used? Was there equal use .for all the manufacturers listed?
- Test data provided to SWEC does not cover all cases listed in Spec.
C-64.
- Test data for Phillips sleeve anchors in block walls (Sec. 5.1)
- Data for Ramset wed E e anchors, sizes 1/4", 3/8", 7/8".
Data for Phillips wedge anchors, sizes 1/4", 3/8". Typical shell anchor data.
- Wedge anchors need a safety factor of 4. Data for 3/4" Ramset wedge shows a safety factor of about 3.7 (both Teledyne tests and W. Virginia Univ. tests).
- The shear tension interaction test results (Teledyne report) do not demonstrate a vectorial combination. For the results available, the allowable values are low enough to still maintain the required safety factor when compared with the test data. In the absence of interaction tests, it is not clear that a vectorial combination of allov4bles will be adequate unless a safety factor greater than 5 for sf.mple shear or tension is available.
DDR - 003 REV O PAGE E OF IO Telecopied 2/24/89
i i 1 O Qt6%inb STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA MAR 0 21989 ( J. D. NO. 18138) OBSERVATION REPORT Observation Report No. DOR-003 Rev 0 PART II - RESPONSE SUPPLEMENTAL RESPONSE Response to items listed in telephone conference agenda received Feb 21, 1989 is as follows: Expansion anchor usage at LGS is as follows: o Wedge anchors used are as noted in Section 2.2.1 of Specification C-64 (Ramset, Parabolt, Phillips). Some Hilti wedge anchors may have been used before 1984. o Sleeve anchors are used in concrete block walls only. Phillips sleeve anchors only were used. o Shell anchors were used only when specifically requested, and approved by the Project. Only Phillips -g shell anchors were used. N-) The following responds to SWEC's questions:
- 1. All wedge anchors specified in Section 2.2.1 of C-64 have been used (Ramset, Parabolt, Phillips). Some Hilti anchors were used before 1984. On the order of perhaps 80% of wedge anchors used are Phillips.
- 2. The following test data has been provided (Transmittal P-0342 to SWEC March 1,1989):
o Phillips sleeve anchors in block walls - IOM from S.C. Desai to G.Katanics, Dec 19, 1977 (Chron No. 026368) with attached IOM FME-2338, July 7, 1977, and FME-2548, December 7, 1977, from J. R. Reiney, Jr to R. Elias. o Ramset wedge anchors (1/4, 3/8, 7/8) - Calculation C-1982-2, Rev 0, Mar 12, 1984, Shts 1 to 3, 10,12, 13, 15,25, 27, 62 to 65. o Phillips vedge anchors (1/4, 3/8) - Pittsburgh Testing Laboratory Report, Lab No. 105228, Mar 31, 1971. o Phillips self-drilling shell anchors - Pittsburgh Testing Laboratory Report, Lab Nos. 93110, Oct 31, 1956, and 94277, June 30,1958. o Phillips non-drilling shell anchors - Pittsburgh Testing lmboratory Report, Lab Nos. 96844, 96902, 97562, March 23, 1962. Page ja of lb
L f b o V STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA ( J. D. NO. 18138) OBSERVATION REPORT l Observation Report No. DCR-003 e.av 0 PART II - RESPONSE SUPPLEMENTAL RESPONSE
- 3. For 3/4" Ramset wedge anchors, .the West Virginia tests (Transmittal P-0300) demonstrate a minimum factor of safety of 5.75 ( 13.82 k minimum everage failure load / 2.4 k allowable load ). The Teledyne tests do show a factor of safety of about 3.7 , for tension load , but as noted in the test report minimum specified concrete strength was 3 ksi.
For LGS, factor of safety w131 be 4 or more as minimum specified concrete strength is e kai, and the anchor test failure load will be greater than shown in the Teledyne tests. Anchor capacities increase as concrete strength increases, as shown by the following references:- a, State-Of-The Art Report On Steel Embedments by ASCE Nuclear Structures and Materials Committee, Materials O., / and Structural Design Committee, June 1984. (Transmittal P-0344 to SWEC March 1,1989). Refer to Sections 4.1.3.3 and 4.1.2.1.
- b. Phillips Red Head Engineering Bulletin No. 102, Sheet 1, May 1,1973 (Transmittal No. P-0348 to SWEC March 1,1989),
- c. Bechtel Calculation C-1982-2, Rev 0, Sheets 10,12,13,15,25,27 (Transmittal No. P-0342 to SWEC March 1,1989).
I 1 O P.ge 2 cf a i i
)
2153275878 P.8/16 MAR 02 '89 12:05 BECHTEL 3RD FLOOR O dhCMW STONE & WEBSTER ENGINEERING CORPORATIDH LIMERICM 2 ZDCA ( J. D. NC. 16138) OBSERVATION REPORT Observation Report No. DDR-ee3 Rev 9 PART II - REEPOMBE SUPPLEMENTAL REEPOMBE.
- 4. If anchor allowables as given in C-64 are compared with the shear-tension interaction turvem presented in the Telodyne Report, it aan be seen that f acters of safety demonstrated by the tests (FS) are 4 or more for wedge anchors and 5 or more for shell anchors.
For wedge anchers at L58 fasters of safety will be 4 or more, es minimum specified senerete strength is 4 kui. as compared to the 3 kan strergth used in the Teledyne tests. A1 discussed in item 3, above, ancher capacities vill insrease with increased eencrete strength. For shell anthers at LOS, f acters of safety will be 5 or mere, considering the greater eencrete strength as discussed above, and sensidering the sneher embednents are equal to or greater than those usec in the Teledyne tests. I O
"zer.ikacirni;;tc;;
" " C N e, oat; '
em t; n EE;;;; $ yeh he nure o; e i wn r.,e .1. ea a
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT v Observation Report D OR- 003 Rev. 0 Review Plan: LK- D - 1903 - S Rev. 1_ PART III - RESPONSE EVALUATION The response to this observation report has been evaluated as follows: A. Causal Factor (s): X Acceptable Not Acceptable (Explain): INSERT A B. Extent of Conditions X Acceptable Not Acceptable (Explain): INSERT B C. Significance: X Acceptable Not Acceptable (Explain): im) INSERT C
%J D. Corrective Actions X Acceptable Not Acceptable (Explain):
INSERT D E. Action to Prevent Recurrence: X Acceptable Not Acceptable (Explain): INSERT E
=======================================================================================
Additional Action Required: X No Yes (Explain) l l n .=============== .======...======================= .========= ===..===== U L C C/Ln Lead Engineer Signature s/n/n
' / Date M* 4 b.Mlle AYM Signature sMs1
/ Date Page 1 of f
(
- t. _ _ _ _ _ - _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ - _ . ________ _ -_______ ___ _ _ -
DOR-003, Rev. 0 INSERT A
) The response indicates that both Design Cuide C-2.40 and Specification C-64 were used in the design of anchors. Initially these documents were in essential agreement but as time went by revision to one would be made without a corresponding change in the other. Revision 1 of Structural Design Criteria C-115 required C-2.40 be used for anchor design which introduced differences between C-115 and C-64.
In addition the overlapping shear cone method for reduction factors for reduced anchor spacing began to be used with no revision being made to the design guide or the anchor specification. INSERT B Bechtel stated they have reviewed all design documents referenced in Design Criteria C-115 and found no other specifications inconsistent with design documents listed in C-115. INSERT C Bechtel has submitted test data with their response to the Observation Report and the supplemental response of March 2, 1989 to support their position that design allowables are low enough to provide the required safety factor when using a vectorial combination of shear and tension. Reduction factors for anchor spacing by the overlapping cone method is generally accepted method and is enveloped by the linear method which has also been used by the project. (n)
\s INSERT D Structural Design criteria C-115 was revised to require expansion anchor design in accordance with Specification C-64. Specification C-64 was revised to include the method of overlapping shear cones to account for the effect of reduced spacing. Since all designs done previously conform to or are more conservative than C-64 the corrective action is acceptable. SWEC has reviewed the test data provided and has found the allowable loads in Specification C-64 to be acceptable.
I INSERT E With the above change in the Structural Design Criteria no recurrence of conflicting design documents for anchor should arise. 1 O 1 V DDR - oo 3 REV o PAGE lo 0F Io
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) l ;< 3 l (f OB3ERVATION REPORT - Observation Report No. D OR- 004 Rev. O Review Plan LK 1903 - S- Rev. 1 l Reference AI No.: DAI-050 PART I - INITIATION
-1. Description of Concern There may not be an appropriate mechanism in place to resolve comments by the checker of a calculation.
- 2. -Supporting Information l Calculation 100-B.06 contains a note indicating that input derived from another calculation appears to be inconsistent and the basis ' questionable'.
Subsequent review confirmed thet the input is very conservative. However, there is a concern that there may be inadequate review of calculations. Response to Action Item DAI-050 indicates that the supervisor's signature on the cover sheet indicates that the problem had been resolved. It would n seem that if this had been the case, the checker would have revised the i comments, or the supervisor would have added a note of explanation. Neither (O was the case. It is not clear that the concern was addressed at all. 4
................................. ..===..==== =====.=================================="
Additional Documents Attached: None.
..................,n...................................... ..........n............. .p .
ib l. Y n /o//1fM NlY S Lead Engineer Signature /' Ddte AFM Signfture '/ Date Page 1 of 1
dBck /#W STONE & WEBSTER ENGINEERING CORPORATION LINERICK 2 IDCA DEC 13 508 (J.O. NO. 18138) OBSERVATION REPORT Observation Report No. DOR-804 Rev e I PART II -
RESPONSE
- 1. Observation Concurrence:
Concur with observation X Do not concur with observation (Note if not in concurrence, explain in " remarks" below)
- 2. Response to Observation:
A. Causal Factor (s) N/A O B. Extent of Condition N/A , C. Significance N/A I
=================================================================
Remarks ; I In January 1982 when Calculation 100-B.96 was checked and l approved, Bechtel Engineering Department Procedure (EDP) 4.37 Rev 2 defined methods used te prepare calculations, l Ancluding checking of the calculatt a, resolution of any I comments, and approving of the stimulation. Specific requirements included the followings [See Continuation) Page(1) of ( 7) ; STONE & WEBSTER ENGINEERING CORPORATION
- - - - _ _ _ _ _ _ _ _ _ _ _ _ l
O ud O LIMERICK 2 IDCA ( J. O. NO. 18138) OBSERVATION REPORT Observation Report DDR-804 Rev 9 CONTINUATION: o Section 4.1.c required that the checker's responsibilities include " Initialing and dating each page of the original calculations after they are completely checked and all necessary corrections and additions have been model or attach initialed alternate calculations if used.' o Section 5.0, in the first paragraph required that
' Calculations that are the basis for establishing design criteria, basic dimensions, or other parameters shall be checked and submitted to the Group Leader for review and approval, and to the Group Supervisor for approval in accordance with this procedure. The Group Leader shall review and approve.all other design calculations prepared by f-his group for technical adequacy and conformance with design requirements.'
o Section 7.3.c required that the Originator
' Obtains concurrence of checker that originals are correct by obtaining the checker's initials on cover sheet and on each calculation sheet, unless alternate calculations are provided to validate the design.'
At present, calculation procedures are defined by Bechtel EDP 4.37 Rev 4. Procedures for checking and approval of calculations are similar to those described above and include an additional requirement regarding interfacing documents. Spacific procedures include the followings o Section 3.2.1. requires that the checker ' Rechecks all calculations made for corrections and revisions and resolves any comments with the originator. Initirls cover sheet and submits checked calculation package to the supervisor / group leader.' o Section 3. 3. a. is identical to EDP 4.37 Rev 2 Section 5.0, first paragraph. O Page( 3) of ( 7) STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA l
JoeA-/M O STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA ( J. D. NO. IDCA) OBSERVATION REPORT Observation Report DDR-804 Rev 9 CQRTIKUiTIQRs o Section 3.3.d. requires that the Group Supervisor (EGS) " Determine if interfacing or reference documents are applicable and sufficient. Also if those documents are affected by the calculations, the EGS shall take appropriate action to notify all others on a need-to-know-basis.' The referenced EDP's clearly require that comments by a checker regarding the originator's calculations, or regarding other referenced input calculations, be resolved prior to the checker's initia11ng of the calculation cover sheet to indicate resolution of the comment and verification of the calculation. During the course of calculation completion, comments made by the checker are resolved before calculation sign-off. Procedurally, this is sufficient. We have no procedural requirement that checker's 's comments, once made, be removed or revised after resolution 3 nor do we have any requirements for explanatory notation by the supervisor regarding checker's comments. I The foregoing statements show that there is an appropriate l mechanism in place to resolve comments by the checker of a l I calculation. In the supporting information section of this observation report, calculation 100-B.06 is identified as a calculation that contains a checker's comment questioning the basis of the calculation. The checker's comment after referring to apparent inconsistencies in input derived from another calculation stated, "Several other values show similar trends and although the discrepancy (?) is in a conservative direction, the basis is questionable." The apparent inconsistency referred to was that the input calculation showed larger anchor bolts, having larger embedmont in the concrete, to have smaller allowable tensile loads than smaller bolts having smaller embedments. O Page (4) of (7)
f o'~ Joc4 /W STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA ( J. O. NO. 18138) OBSERVATION REPORT Observation Report DOR-804 Rev e CONTINUATION: Bechtel in the response to Action Ites DAI-950 confirmed that the basis of the calculation is correct. The review , showed that the allowable loads derived by the input j calculation are consistent as they are controlled by the ) anchor pro 3ection (effective embedsent) below the anchor ! sleeve, and not by the anchor embedsent below the concrete { surface. j i SWEC cites a concern that there may be inde quate review of j calculations based on the following observations regarding ! calculation 100-B.86: o If the checker's comment had been resolved the checker would have revised the comments or the supervisor would have added a note of explanation. 1 o Although the supervisor's signature is on the cover sheet, it is not clear the checker's concern was addressed at all. Since the calculation is conservative there is no conclusive evidence that there may be inadequate review. Instead it must be concluded that the checker,in the performance of his duties, satisfied his concern as reflected by his signature on the calculation. Similarily, the supervisor who approved the calculation was satisfied that the calculation was conservative and technically adequate, and that the referenced input calculation was applicable and sufficient, I as reflected by the supervisor's signature on the { calculation. A representative sample from civil engineering calculations for seismic category I structures have been reviewed to determine if other similar checker's comments are present in the calculations. The review shows that no such comments are present. As the basis of the calculation has been reviewed and found to be conservative, and as no similar instances relative to checker's comments have been identified, it is concluded that the procedure requiring any checker's comments to be O resolved is satisfied, and calculations have been adequately reviewed. Page ( f) of 0)
TEL NO:215-841-4578 4916 P02
.. DEC-14 '88 13:10 ID:PECO PROJECT MGT DIV P. s 6 IU :; 'Bi 16:05 ECCHTEL WCL% PGER 151 FLOy O
Jhc4/# STONE & wtB8TER ENGINEERING CORPORATION LIMERICK 2 IDCA I J. C. NO. 15138) OBSERVATION REPORT Observation Report No. DDR 884 Rev e PART (1 2;RESPDMEE CONTJEMED
- 5. Gerrective Action ,
M/A O E. Action to Prevent Recurrence M/A senese==mssese...... sees.se...es.. .. es. mons neemmesassesse.=me O u....:. a&.& sb w.s s'.. % L A - n . a w P260 eview 51 nature / ate Rechted ApprovaI Signat re/Date Page(6) of (7)
J STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA gg (J.O. NO.'18138) OBSERVATION REPORT I Observation Report D OR- 004 Rev. O ' Review Plan: LK- D - 1903 -S Rev. 1
- PART III - RESPONSE EVALUATION i
The response to this observation report has been evaluated as follows: A. Causal Factor (s): Acceptable Not Acceptable (Explain): Not Applicable B. Extent of Condition: Acceptable Not Acceptable (Explain): Not Applicable C. Significance: Acceptable Not Acceptable (Explain): Not Applicable
/ i V - D. Corrective Action: Acceptable Not Acceptable (Explain):
Not Applicable E. Action to Prevent Recurrence: Acceptable Not Acceptable (Explain): Not Applicable Additional Action Required: x No Yes (Explain) It is agreed that there 19 no conclusive evidence that the calculation ! in question was reviewed inadequately. Since the calculation is conservati.ve, and Bechtel has performed a sample review of other calculations and found no similar checker's comments, this appears to be an isolated case. No other instances of this nature were identified during the IDA, which further substantiates the adequacy j of Bechtel's calculation review procedures. En l l [ l Lead Engineer Signature '
/' Date APM $4n Ve - - / / D/te Page 1 of 1 I
[
> STONE & WEBSTFR ENGINEERING CORPORATION.
LIMERICK 2 IDCA (J.O. NO. 18138)' p/(_ OBSEETATION REPORT .
. Observation Report No. DOR. 005 Rev. 0,, . , ,
Review Plan: LK 1903 -S Rev. 1_ Reference AI No.: DAI-129
'PART I - INITIATIOR 1.- Description of Concern The finite- element model used in Calculation 22.2L66 appears to give unconservative results.
2 .' Supporting Information The calculation states that the maximum vertical moment due to the 2 psi pressure. load is 3.53 ft-k/ft (Sheet 68). Simple hand calculations.
. assuming that the wall is' a one-way slab (10.5 ft span), '.ndicate that -
moments should be at least 3.97 ft-k/ft. Response to Action Item DAI-129 explained that the moments are reduced due to redistribution caused by. cracking. This explanation does not appear sufficient based on the following:
-The wall is very long in comparison . to it's height. 'Even if the' wall is cracked, the only way , for load to. redistribute is from the vertical direction to the horizontal direction. Since-the span is long, the wall still behaves as a one-way simple span between the top and bottom. Variations in stif fness will not affect the _ moment in a simple beam. Since' only one-half of the wall 'is modelled,- it is possibic that the boundary conditions at the right-hand edge (center of wall) of the model are not correct (Z direction motion should not be restrained at the edge). Note that the response to DAI-129 states that elements 10 through 36 are cracked, while the calculation (Sheet
- 68) does not state that elements 18, 27, and 36 are cracked.
Response to DAI-129 indicates that the moment in the 1.5 ft. wide strip between openings . (Elements 1, 10, 19, 28) acts as a beam. The moment found by averaging the two cases of simply supported at the top and fixed at tiie top (i.e. at joints 47 and 48) is 3.73 ft-k/ft. The simply supported case, which gives a moment of 4.82 ft-k/ft, would appear to be a more realistic estimate of the actual moments. This is considerably higher than the 3.53 ft-k/ft computed by the BSAP Program. l' ....................................................................................... Additional Documents Attached: None l Lead Engineer Signature (. &n fr/Mff/
'/ D4te APM Sig/fature
/ :-
/ Date Page 1 of 1 ,
l
.n
j l
.O occ 1 s i>88 l l
STONE & WEBSTER ENGINEERING CORPORATION l 2 LINERICK 2 IDCA ( J. O. NO. 18138) i OBSERVATION REPORT Observation Report No. DOR-905 Rev 2 PART II - RESPONSE
- 1. Observation Concurrence:
Concur with observation _E_ Do not concur with observation (Notes if not in concurrence, explain in " remarks' below) l l
- 2. Response to Observation:
A. Causal Factor (s) (' N/A I B. Extent of Condition f N/A I C. Significance N/A Remarks The finite element model used in calculation 22.2L66, and shown on sheet 67, gives reasonable and conservative results that are adequate for the purpose of evaluating this wall which has a much larger capacity than the demand. The model was used to calculate local moments near openings at one end of a concrete block wall, resulting from differential pressure loading. The global moment was I O calculated on sheet 49 of the calculation, j Page ( 2) of (9)
()DCHIN @ O STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA ( J. D. NO. 18138) OBSERVATION REPORT Observation Report DOR-905 Rev e CONTINUATION: The well evaluated in Calculation 22.2L66 is 10.5 ft high, and 58 ft long (aspect ratio = 5.5). The wall is supported at its top and bottom by building floors, and at each end by exterior concrete walls of the building. Two door openings and two smaller openings above the doors are located at the north'end of the wall. To evaluate the wall behavior for out-of plane loading in the region of the openings at the north end of the wall, the northernmost 38 ft (essentially half) of the wall was modeled, as shown on sheet 67 of the calculation. The model has an aspect ratio of approximately 3. Simply supported boundary conditions, which restrain out-of plane (Z direction). motion, were used at all four edges of the model. O .The modeling. of only the north half of the wall, using simple support at all four sides similar to the actual wall, was a reasonable approach for . calculation of the behavior near the openings. Even though only half of the well was modeled, and the boundary condition used at the south end of the model (22 ft away from the nearest door opening) does not exactly account for the out of plane displacement at that' location, the margin that exists between the calculated demand and the capacity of the wall makes the model adequate for the intended purpose. This is evident by the fact that the maximum bending moment in a uniformly loaded simply supported plate having the same aspect ratio as the model is very nearly the same as in a plate having the same aspect ratio as the wall (Ref. Roark and Young, Formulas For Stress And Strain, Table 26, Case la.). It is true that the wall is very long in comparison to its height. It is also true that the wall behaves, generally, as a one way simple span between the top and bottom, and that variations in stiffness will not affect the moment in the simple span. However in the presence of openings, and even l more pronounced near the ends of the wall where out-of-plane support is provided, two-way behavior is significant. Because there is two-way action, the extent of cracking affects the distribution of loads between the vertical and O the horizontal directions. This behavior is accounted for in the BSAP model analysis which determined elements le to 17, 19 to 26, and 28 to 35 to be cracked (The response to DAI-129 Page i of (j ) (,J ' 1 J
l d)Dc d /% 9 STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA ( J. O. NO. 18138) 1 OBSERVATION REPORT Observation Report DOR-905 Rev 0 ] CONTINUATION: j inadvertently stated that elements le through 36 are cracked, whereas elements 18,27,and 36 are determined by the analysis to be uncracked). The maximum moment calculated by the BSAP model analysis at the region of the door openings, l and ocurring in the wall strip between the doors, is 3.53 ( ft-k/ft. The one way slab moment applicable away from the ( influence of openings and support at the wall end is 3.97 ft-k/ft, as stated in the Observation Report. The moment near the openings is smaller than the moment away from the i openings because of the effect of the out-of-plane support at the end of the wall. As stated in the response to DAI-129, the 1.5 ft wide wall strip between the door openings acts as a beam spanning from the floor to the top of the openings. The beam is simply
\ supported at the bo'tos.
c At the top the beam end condition i is between a simply supported and a fixed condition. The actual condition is a function of the rotation and displacement at the top of the openings. The rotation is restrained by the extension of the vertical reinforcing bars
.into the short (3.5 ft) extent of wall above the opening.
The displacement is small because of the proximity of the top of the beam to the wall supports at the end and tcp of the wall. Considering the actual restraint, the simply supported end condition assumption yields too conservative an estimate. A more reasonable estimate is the average of the two conditions. The maximum moment in the beam occurs i between supports and should be between the simply supported at the top (4.83 ft-k/ft) and fixed at the top (2.62 ft-k/ft) conditions. The average value is 3.73 ft-k/ft which is close to the 3.53 ft-k/ft moment derived by the BSAP model analysis and well below the 7.70 ft-k/ft allowable capacity of the wall. To further support the adequacy of the model, a study (Calculation 101.105 Rev 0 transmittal No. P-0275 dated 12-14-88) has been performed by Bechtel Engineering in response to this Observation Report. The study uses a model with different boundary conditions, that permit displacements to e take place freely at the south edge. Results show the ( maximum moment near the openings increases only slightly, to a value of 3.80 ft-k/ft. The study confirms the adequacy of the model used in calculation 22.2L66. Page (g of (g ) s
_ DEC-19 '99 09:53 ID:PECO PROJECT MGT D!U TEL NO:215-641-4578 #945 PO4
' ;C, IT 'EB 0?:?? ECHTCL iEE'sn F %E ;!T % :6 SIJ O JDM1%9 STONE & WEB 8TER ENGINEERING CORPORATION LIMERICX 2 IDCA (J.C. No. 18138)
OBSERVATION REPORT Observation Report No. DDR-905 Rev 8 PART II - MEEPONer.fDMTINUED D. Corrective Action N/A O E. Action to' Prevent Recurrence N/A
-e..............................................................
k.. .,:. *., . - }. ] f.Y - .. .$ *. , .. $N.LNLhI PEC Review signature /D e ,Bechtel Approva Signat re/Date b Page (/) of (9) St- -
p AM / 7L/7 S'IONE & FEBSTER ENGINEERING 00 ICN U LIMERICK 2 IDCA s._n.e~
"~
(J .O. 10. 18138 ) REISSUED W08 m OBSERVATION REPORT DATE . Observation Report No. DDR-005 Rev. O PART II - AMENDED SUPPLEMEtGAL RESPONSE
- 1. Observation Concurrence:
Concur with Observation X Ib not concur with observation (Note: if not in concurrence, explain in " remarks" below)
- 2. Response to Observation:
A. Caushi Factor (s): N/A B. E. tent of Condition: N/A C. Significance: N/A O Corrective Action: N/A Q D. E. Action to Prevent Recurrence: N/A
========================================_=========================================
Remarks: The following is a supplement to the DOR-005 response to provide the additional information requested by SWEC.
'Ibe out-of-plane reaction at the wall upper left corner (see Reference "1") and the twisting moments adjacent to the corner and/or the large openings is typical of wall design analyses. These effects are provided for in the design of these walls by the addition of rabar, specifically ' trim' rebar. The trim rebars are provided in addition to the nominal wall reinforcement as called for on drawing C-608, " TYPICAL CONCRETE MASONRY WALL DETAILS" and uniquely for wall 103 on drawing C-763 sheet 1.
Two number seven horizontal rebars, continuous into the adjoining perpendicular wall, are provided at the extreme top of the wall to resist the corner reaction and its immediately adjacent twisting moments. Four number seven horizontal rebars, continuous into the adjoining perpendicular wall, are placed as close as practical across the top of the openings. These rebars, in conjunction with the vertical trim rebar and the nominal vertical and horizontal reinforcement, resist the twisting moments through shear friction in this region of relatively low bending roments. Since the trim rebars are typically provided, the design calculations do not normally include an explicit rcress evalt.ation for each and every wall opening or corner. p) Calculation file 101.105 Rev. O has been revised to include a stress evaluation for this Pagelof 9 ___-_. _ _ i
F.5 E me 0: 'e9 trias BttHTEL wr5TTM %CS 15- rLUS SMS & WCBSTER ENGIliCERING CORPORATION LIMERICK 2 IDCA p(M/ 7h'7 fs (J.0. IC.18138) OBSERVATION REPORT Observation Report No. DDR-005, Rev. O PART II - RESPONSE COWTINUED specifie location. (Calculation 101.105 nov. 2 hae been transmitted to SWEC via transmittal P-0355, Dated 3/7/89. The results of the evaluation confirm the design has provided adequate capacity for these effects. There are no blockwalle in the Unit 2 Reactor Building, the Spray Pond Paphouse, . or the Diesel Generator Building. There are several clockwalle in the Unit 1 and 2 common area. 'Ita blockwall reviewed and concented on in thie DDR is the only isolated instance of a blockwall modeled by the finite element snethod that did not include the entire wall in the analytical model. (1) Asferences S. Tinnehenko and S. Woirowsky-Krieger, '"Iheory of Plates and Shells", McGraw-Hill, 1959, pagee 120-123.
. O 1
casamens am m e s seu m s mamassa ma ss a m ma ss n e sse n esamm es s am a s s me m assenmassepasmasmanmemme d li n n f d *x r H w sl8] t9 uBoentel sia Approval 4 P kSignature' k M s hits
/Date /Date Onco 8evijwg51gnature Pagelef1 l
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) [ OBSERVATION REPORT
-Q)
Observation Report D OR- 005 Rev. _0_, Review Plan: LK- D- 1903 - S Rev. 1 PART III - RESPONSE EVAL ? TION The response to this observation report has been evaluated as follows: A. Causal Factor (s): Acceptable Not Acceptable (Explain): NOT APPLICABLE f l B. Extent of Condition: Acceptable Not Acceptable (Explain): NOT APPLICABLE C. Significance: Acceptable Not Acceptable (Explain): O .. NOT APPLICABLE D. Corrective Action: Acceptable Not Acceptable (Explain): NOT APPLICABLE E. Action to Prevent Recurrence: Acceptable Not Acceptable (Explain): NOT APPLICABLE Additional Action Required: X No Yes (Explain) l SEE ATTACHED PAGE 1
....................................................................................... {
Au f {lLn 3)</'t Lead Engineer Signature / Date
% CL 2 t 6 0. M lle AIY1 Signature sin /n
/ Date !
l l l Page 3 of 9 l w____._____.__
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT \_) Observation Report DOR- 005 Rev. O PART III - RESPONSE ETALUATION CONTINUATION: The finite element model of Calculation 22.2L66, which analyzed the block wal' with 2 psi pressure, appeared to give unreasonable results. Further review by SWEC showed that high twisting moments were produced near an upper corner boundary, which were not evaluated by Bechtel. 'Bechtel has subsequently submitted Calculation 101.105 Rev. 2 to demonstrate the adequacy of . the wall under the corner twisting moments. This calculation was done using the actual design pressure of 0.8 psi, instead of 2 psi. Since Bechtel has stated that no other block walls were analyzed with the finite element method a recurrence of this type (incomplete use of finite element results)'will not occur with block walls. O) 'u. 4 O Pap 3 of 1
5 V STONE & WEBSTER ENGINEERING CORPORATION l LIMERICK 2 IDCA
. /'~ (J.O. NO. 18138) 1' o. > ; OBSERVATION REPORT .
l Observation Report No. DOR- 006 Rev. O Review Plan LK- D - 1904 - C Rev. J Reference AI No.: DAI-032 PART I - INITIATION
- 1. Description of Concern During the review of Instrument Location Drawings for Instrument No.
PDT-51-2N058B, various discrepancies, inconsistencies, and omissions were noted which have been subsequently corrected by issuance of FCR's or DCN's.
- 2. Supporting Information The following problems were observed during the review of Instrument' Location . Drawing Nos. 8031-M-1539 and 8031-M-1540 specifically for that part of the drawings showing Instrument No. PDT-51-2N058B only:
a) Division I and Division II Instrument Lines were shown passing s through the same floor penetration, indicating an apparent separation problem. This problem was corrected by Bechtel A issuance of FCR MI-748-J dated September 23, 1988. 4 b) Instrument lines were shown passing through the wrong floor penetration. This problem was also corrected on FCR MI-748-J dated September 23, 1988. c) An instrument listed in the Instrument Penetration Schedule on M-1540 did not exist for Unit 2. This problem was corrected by Bechtel issuance of DCN No. 002 dated September 27, 1988, d) Several instruments listed in the Instrument Penetration schedule were inadvertently retained during incorporation of a change document. This problem was corrected on FCR MI-748-J dated September 23, 1988. CONTINUED ON ONE PAGE
............................................-...=...===...===..=================.======
Additional Documents Attached: None m
,J ........................................................(O~ */ BE
_0 CR yhdr ' Lead Engineer Signature / Date APM Signature /'Date Page 1 of 1 4
STONE & WEBSTER ENGINEERING CORPORATION i' LIMERICK 2 IDCA L (J.O. NO. 18138) > ( OBSERVATION REPORT Action Item No.: D OR- 006 Rev. O _ CONTINUATION: B e) Bechtel indicated in a telcon dated October 4, 1988, that routing of instrument lines classified "Q-P" are not required to be shown on Instrument Location Drawings. On Drawing 8031-M-1540, the low pressure sensing line for QP Instrument No. PDT-51-2N060B is not shown. The high pressure sensing line for the instrument is shown, however, and a correction associated with the routing of this line was . included. in FCR MI-748-J issued September 23, 1988. Based on a review of associated as-built isometrics, no hardware problems were identified as a result of these discrepancies. However, Bechtel has indicated that the location drawings, together with instrument detail drawings, support details, and 8031-M-830, form the basis for construction to install plant. instruments. Such discrepancies can create confusion concerning what should be (or has been) installed, and might have resulted in hardware problems for other cases not evaluated here. Also, if these drawings are intended to be retained to support and plan future modifications, they should be properly maintained. (
~ The full. extent of the existence of such discrepancies on Instrument Location Drawings is not known at this time, since the review was limited to sel'ected instruments within the scope o f ' t'h e IDCA. Bechtel should address.the extent of this condition in the response to this DOR.
i O . pa3e a of 2 -
STONE & WEBSTER ENGINEERING CORPORATION Jg /g ' LIMERICK 2 IDCA (J.O. No. 18138) JAN 2 51989 m OBSERTATIOR REPORT [8 - Offj Observation Report No. DOR- 006 Rev. [ PART II - RESPOW8R
- 1. Observation Concurrence:
X Concur with observation (I TEMS b, c & d) X Do not concur with observation (Notes: if not in concurrence, explain in " remarks" below) (ITEMS a & e)
- 2. Response to Observation:
A. Causal Factor (s) (b) Be cause penetration # 11 i s actually a floor grate pene-tration superimposed above'the subject instrument tubes shown on location Dr awing M-1540, the instrument tubes appear to pass through it. This confused the design and construction engineers. The penetration schedule however, incorrectly lists the instruments as passing through penetration # 11. O l FCR MI-32 incorrectly added these instruments to pene-V) tration #11 in the penetration schedule. The penetration schedule lists the instruments as a result of incorporating FCR MI-3 2. merefore, a location drawing misinterpretation by' the construction design engineers is the cause for the discrepancy.
'(c) Al though PCN 20409 d eleted instrument PDI-42-2R0 05 f rom the design, a DCN was not issued to delete the instrument f rom the M-1540 location drawing penetration schedule, and was not detected during the PCN approval process.
Instrument retention in the penetration schedule was an oversight by the PCN originating engineer. (d) FCR MI-604-J diagrammatically deleted instruments LT 2N08 0A , -2N09 5A ,' -2N00 4C , -2N017 and -2N02 7 from pene-tration # 11 a nd re-routed the instrument lines through penetration #18. Se FCR also deleted the instruments f rom penetration # 11 in the penetration schedule. Because the draf ter neglected to delete the instruments from the location drawing penetration schedule, and the ! checker failed to detect the omission, the discrepancy can be attributed to location drawing draf ting and ( checking errors. Remarks ' SEL SHEET 3 3 1 T- 7/11 Page/ of /f l JK /k
i l- STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) JMC//d[f ( (),/ l OBSERVATION REPORT Observation Report No. D OR. 006 Rev. O PART II - RESPOBSE l B. Extent of Condition Six out of 18 location drawings which list penetration schedules have been reviewed for penetration schedule accuracy. Two additional errors were identified during this review. C. Significance The location drawing penetration schedules are not intended to convey design requirements. Although the schedules are incorporated into controlled design documents, their contents are secondary to the design requirements reflected on the location drawing diagrams, and P& ids. The diagrammatic portion of the location drawing provides design information regarding tube routing and penetration location, size and elevation. P& ids provide primary point take-offs, while
/~N OADs provide the instrument's 0-status. The penetration d schedules were provided as a convenience to aid - the designer and construction engineer in identifying penetration tube quantities and tag numbers associated with the various tubes. They also facilitate with tracing instrument tube routes shown on the location drawings. Based on the above discussion, the identified discrepancies are considered to be insignificant. .
D. Corrective Action For causal f actor 'b', FCR MI-748-J was issued against location drawing M-1540 to delete instruments LT-42-215B,
-2N0 91B , -2N0 91F , - 2N0 81D , -2N4 02B , PDT- 51-2N0 60B and
-2N058B from penetration #11 in the penetration schedule.
For causal f actor 'c', DCN # 002 was issued to delete instrument PDI-42-2R005 f rom the penetration schedule on location drawing M-1540. For causal factor 'd', FCR MI-748-J was issued to delete instruments LT-422N080A, -2N09 5A , -2N00 4C , -2N017 and
-2N027 from penetration #11 in the penetration schedule.
Renwrke
'l 7 T- 7/11 Page g of Z
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ . _ _ _ _ _ . /)K /)G
"r '
~
~
gs.js gra ,,,,3 s'ob7 6 WER$TER ENG WIERING CORP 0KATION ,
- LIMERICK 2 !DCA U.o. NO.18138) .fOCA /d 99 9 ossitVATIos R D otT I
observatten Report Nc. A ot- h llev. g_ FART !! - 11370BFI
- DCN's 003 and 002 against location drawings M-15'34 and M-1543 respectively, heve been icsued to correct the two additional discrepancies identified in tne Extent of Condition section.
E. Action ~'o Prevent Recurrenew DCN 001 against Location Drawing M-1571 has been issued to add a note to reflect the design intent of the penetration schedules to preclude their use as design requirements. Remarks (a) Instrument locatior. Drewing M-1540, Rev. 18 depicts the blue l tube f or instruments LT-42-2N080A, -2N09 5A, -2N00 4C and -2N027 es originating from containment penetr6 tion X-4SA. The tube is then routed counter-clockwise to floor penetration G 8 without passing through floor penetration til. Location Drawing M-1540, Rev.18 correctly depicts the 'olue " instrument line as passing through fl0 and not #11 as stated in the DCR. Therefore there is no identified separation problem. FCR-MI-748-J corrected only the listing in the pene-tration schedule and not the actual tube routing. (e) The low pressure sensing line for instrument PDi-51-2N060B is not shown on location Drawing M-1540, as it originates from containment penetration X-66A on, location Drawing M-1554. Since instrument PDT-51-2N0605 wEriginally a Of statused item, the instrument lines were shown as required, However, this instrement was subsequently downgradod to a Op status. Although the applicable tubing isometric, electrica' scfematic . and QAD have been revised accordingly, the location drawing has not been revised since there is no regstrement to delete nor prohibit depicting Op lines on the locat,i.on drawings. Because the location drawings are not the lead design document in determining instrument 'O' status as are the QAD's, the drawing is acceptable in its current configuration, e p ' O(/hi
.... .......s .....................................................................
L L /x n Powsar am h dchtal Appr6 val Signaturg ~ / rate PECOTdfsW$ignatu / / Date S Psp 2. of E fd -+ 7 779 2_ _ _ _}_Y
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) O U-OBSERVATION REPORT Observation Report D OR- 006 Rev. O Review Plan: LK- D - 1904 - C . Rev. 1 PART III - RESPONSE EVALUATION The response to this observation report has been evaluated as follows: A. Causal Factor (s): X Acceptable Not Acceptable (Explain); SEE ATTACHED PAGE B. Extent of Condition X Acceptable Not Acceptable (Explain): SEE ATTACHED PAGE C. Significance: 3 ,;2ptable 4 Not Acceptable (Explain): r SEE ATTACHED PAGE
- D. Corrective Actions X Acceptable Not Acceptable (Explain):
SEE ATTACHED PAGE i i E. Action to Prevent Recurrences X Acceptable Not Aeceptable (Explain): l l SEE ATTACHED PACE l Additional Action Required: 2o N Yes (Explain) dM' L4ad Engineer Signature 2/u/81
/ Date Q h 4 . D.l W IIe.
MM Signature 2/2s/t1
/ Date Page i of 2
DDR-006 j A. The cause is acknowledged to be incorrect incorporation of change paper, oversight, and drafting and checking errors. B. The response states that the condition is limited to penetration schedule accuracy. Six (6).out of eighteen (18) drawings involving penetration schedule were reviewed by Bechtel and two additional errors in penetration schedules were found. C. The response states that the penetration schedules shown on the location drawings are not intended to convey design requirements. Only the diagrammatic portion of the location drawir.g provides design information regard'ng tube routing and penetration location, size and elevation. Although such errors can be confusing, no cases were found where the design adequacy of the installed instrumentation. was affected. Therefore, the problems noted have no design significance. D. The discrepancies identified by IDA were corrected by Bechtel by issuing appropriate change documents which were reviewed by SWEC. The change documents (DCN's) issued to correct the two additional errors found by Bechtel were not reviewed by SWEC. E. A DCN has been issued to add a note to the lead sheet of the location drawings to reflect the intent of the penetration schedules and preclude their use as design requirements. Bechtel did not concur with Items a and e of DOR-006. Regarding Item a, no separation problem was found. The error on the drawing was corrected by FCR-MI-748-J. Therefore, the response was considered acceptable. Item e deals with showing Q-Passive instrument lines on location drawings. Bechtel explained that there is no requirement to show Q-Passive instrument lines on those drawings. However, the line in question was originally classified as Q-functional and was therefore required to be shown. The instrument was subsequently reclassified to Q-Passive, and Bechtel opted not to delete it since there is also no requirement to delete nor prohibit depicting of Q-Passive lines on location drawings. This response was acceptable and the concern was closed. I i l 4 1 4 l DOR - 0 0 6 REV O PAGE 7 OF 7
l
'i SIONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138)
OBSERVATION REPORT c
'l Observation Report No. D OR- 007 Rev. _0 Review Plens LK- D - 1903 - MS Dev. L Reference AI No.: DAI-058, Rev. O l
PART I - INITIATION
- 1. Description of. Concern 3 I
There are deficiencies in . traceability of published radiological design ) parameters to design basis calculations that could u;d te possible design j problems. In accordance with commitments contained in the Limerick Quality Assurance Plan for Design and Construction, Section 3.2.1,'and FSAR Appendix 17.2B.II(h), traceability is required in accordance with 10CFR50. Appendix B, III-Design Control, and ANSI N45.2.11-1974. 1
- 2. Supporting Information A. RHR Pump cubicle. doses in the environmental design criteria specification, Ji.e., Specification M-171 are not traceable to design calculations via a controlled document, in accordance-with 10CFR50 Appendix B, III-Design Control and ANSI N45.2.11-1974.
(Section 5.1.4). O B. The RHR_ Pump compartment normal operating gamma dose rate calculated in Calculation S102.2 and published in Specification M-171 is not identified as being a result of this calculation and was not derived in response to its stated objective. Without recourse to the originator of Calculation S102.2, a person technically qualified in the subject could not find this dose rate and understand its applicability to the environmental design criteria specification. This condition is not in accordance with the requirements of ANSI N45.2.11-1974, Section 4.2. C. The final post-accident gamma pipe contact dose (based on diluted post-accident source) and the total post-accident gamma dose (i.e., pipe contact dose plus secondary containment cloud dose) do not appear to be supported and documented in a referenceable design calculation. Instead, they were calculated in the response CONTIN 0ED ON ONE PAGE som ma ma mm ase,e ma m a m men em m a ss a mm a samm mm m mmm mm u s u m mm mmm m m ma s a m ma m m m a ss u m s m u se s s a m m es se s s e s Additional Documents Attached: None eqs.. .,.....-..,.......................m........$............................m.
.(IGr iohlkt h W A H M, nf A,, ,1 aat.r. , ..t.
Gm L.ad . s r ,1..t.r. i wat. Page 1 of j
STONE & WEBSTER ENGINEERING CORPORATION l LIMERICR 2 IDCA (J. O. NO. 18138) p OBSERVATION REPORT i Observation Report No. DOR- 007 Rev. 0 4 CONTINUATION: to Action Item DAI-058. Also, the suppression pool volume as quantified in FSAR Table 6.2-1 Rev. 27 daes not appear to be 4 supported and documented in a referenceable calculation. This i volume. is inconsistent with the value quoted in Calculation j A00.4 and is used in the Action Item response to determine the ] dilution factor which, in turn, is used in the Action Item response to determine the final post-accident gamma contact dose. These conditions are not in accordance with 10CFR50 Appendix B, III-Pesign Control, and ANSI N45.2.11-1974 Section 5.1.4. O O Po8e a of I
zoc4/293 STON1 & WEBSTER ENGINEERING CORPORATION LINERICK 2 IDCA (J.O. NO. 18138)
,}. OBSERVATION REPORT OV 2 d nee {s-o W Observation Report No. D OR- 007 Rev. _0 PART II - RESPONSE
- 1. Observation Concurrence:
Concur with observation Do not concur with observation (Note: if not in concurrence, explain in _XX_
" remarks" below)
- 2. Response to Observation:
A. Causal Factor (r) B. Extent of Condition C. Significance D. Corrective Action O t E. Action to Prevent Recurrence ESEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEE Remarks
- 1. General. There are no deficiencies in traceability of radiological design parameters to design basis calculations that would lead to possible design problems.
It is not the Limerick Project's policy to footnote or otherwise cross-referenca every design value in our drawings and specifications to identify the source of the values. If there is a need to determine the source of a design value, we are able to do so (as we have done in response to several Als for this IDCA), but we do not maintain a source list for all these values on the design document itself. Further-more, thia degree of traceability is not required by 1DCFR50, Appendix B, Section III. That section discusses design control in general terms, but does not discuss traceability in specific terms to the extent that the design control and traceability methods used at Bechtel could be considered deficient. It should be noted that ANSI N45.2.11-1974 is not applicable to the design and construction of the Limerick plant and was not committed to during the design and construction phase as noted in FSAR Section 1.8, Regulatory Guide 1.64. The commitment to ANSI
.N45.2.11 in F3AR Appendix 17.28.11 (h) is limited to the PECo Engineering and Research Department for the operations phase of the plant. Although not applicable to Limerick, it is also noted that Section 5.1.4 of ANSI N45.2.11-1974 referenced . ,- s in the OR applies to documentation interface control with external organizations, and would hence not be applicable to the concern cited in the OR. Section 5.2.4, (V) . which applies to documentation interface control within an organization, does not contain such tra cability requirements.
Page 2_ of JE
870NE & WI43fER IN0!NttRING CORPORATION LINERICK 2 !DCA l (J.0, NO.18188) l ObsttVATION REPORT Observation Report No. ,,jL, OR .,Agl Aev. ,,2 ! 1 Remarke (continued) l
- 3. (a) The general discueston given above applies as well to thle specific item in the OR.
(b) The response above ales applies to this item. In addition, while the stated purpose of calculatten 8103.3 vse not directly to produce input for 1 specification M 171, since it generated data that could be used se M-171 l input, it wae' acceptable to use those resulta for that purpose also. It te not uncommen to use the results of a calculation for more than one purpose after making a review of the bases and assumptions for appropriatenose. (c) The values in M tT1 discussed in the response to DA!-Oes were derived so indicated in that reopense. These values were obtained in a direct and elmple manner (l.a., total integrated does rate divided by the volume reductica factor to obtain adjusted total integrated dese). Theref ore, when these values were entered into M-171 a emarate calculation was not prepared since it was not necessary. The suppreselon pool minimum water volume used in calculation A00.4 was the value given in PsAR Table 6.t-1. Subsequent to performance of calculation O A00.4. the value for suppresalt.n pool minimum water volume was revised by LDCN F8 408. The value presently given in Tabis 6.t-1 was calculated in Wechanical calculation Misc-6, Revleien C. Since the value used in A00.4 yielde the more conservative results, f t was decided to not revies A00.4 to reflect the revised input parameter.
.............. .......es...................... ..................................
O~hn S$b pede AetTp'w 8'ignaVure wlwic
/ Date (W /4, htt Tochtiri Approval' 81gnature ll/wos
/ Date Poge 4 d 5
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) N /. () OBSERVATION REPORT Observation Report D OR- 007 Rev. O Review Plant LK- D - 1903 _MS Rev. l PART III - RESPONSE EVALUATION The response to this observation report has been evaluated as follows: A. -Causal Factor (s): Acceptable Not Acceptable (Explain): Not applicable B. Extent of Condition: Acceptable Not Acceptable (Explain): Not applicable C. Significance: Acceptable Not Acceptable (Explain): Not applicable v) D. Corrective Action: Acceptable Not Acceptable (Explain): Not applicable E. Action to Prevent Recurrence: Acceptable Not Acceptable (Explain): Not applicable
......................................................................................=
Additional Action Required: y No Yes (Explain) The concern raised in DOR-007 concerning traceability of radiological destign paraueter to design basis calcular'ons is addressed in Bechtel response dated Nov. 23, 1938. Althoug,h Cales S102.2, A004, A00.7, a,d A00.9 are not referenced in Environmental Design Criteria Specification 8031-M-171, this is not contrary to Bechtel procedures. O uknhs}sa Lead Engine /!r Sigliature' ' / Date
~
APM Si ature N ' a1W
/ Da3
~
Page 5 of f
STONE Ea WEBSTFR ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138)
,- OBSERVATION REPORT .
L ) Observation Report No. D OR- 008 Rev. O Review Plan: LK- D - 1912 - MS Rev. [
~ .
Reference AI No.: DAI-195, Rev. O PART I - IltITIATION
- 1. Description of Concern There are deficiencies in traceability in the Fire Protection Evaluation Report (FPER) that could lead to possible design problems. The FPER does not identify or provides (1) A reference that identifies safe shutdown cables for each shutdown method in the fire areas where these cables are located.
(2) A reference that identifies the safe shutdown cables that are required to protected by fire barriers.
- 2. Supporting Information
,m, :
The condition identified in description of concerns is not in accordance ; U'}
~
with commitments contained in the Limerick Quality Assurance Plan for Design and Construction, Section 3.2.1, and FSAR Appendix 17.2BII(h), which requires traceability in accordance with 10CFR50 Appendix B, III-Design Control, and ANSI N45.2.11-1974. The FPER, Chapter 5 " Analysis of Capability to Achieve Safe Shutdown" includes the overall methodology (Section 5.1) and the descriptions for reactor shutdown methods (Section 5.2). Drawing 8031-E-1560 " Safe Shutdown Cable Data Base" lists the cables that are essential for safe shutdown and identifies the raceways through which each cable is routed, as well as the fire areas in which the raceways are located. The raceways that are required to be protected by fire barriers are listed in Drawing 8031-E-1406, Section 8. However, FPER Chapter 5 does not provide reference to these drawings that would provide traceability in accordance with 10CFR50, Appendix L, III-Design Contre 1 and ANSI N45.2.11-1974, Section 4.7. Design Analysis. Additional Documents Attached: None.
.... ... . ..... .............c....-.....,............... . ... .................. ;
o 1 . IIlAl6 4 Wr? APM Sigg6ture ' / Date
~
Q) Lead EngTnee'r Sighature / Date ' Page1ofd j l l l l
J-Dc a /z 7.2 STONE & WEBSTER EN0!NttR!NQ CORPORA?!0N LIMERICK 2 !DCA C.5 0 W &y- - (J.0. No.18188) M 23 O OBSERVAT!DN ttPORT Observatia Report No. ,,R., OR ,Jll2L Rev. .JlL. pAnr !! . nrapavar
- f. Observation Concurrence:
Concur with observation
.1L. Do not concur with ob6ervation (Note: if not in concurrence, explain in
" remarks
- below)
- 2. Response to Observation:
A. Causal Factor (s)
- 8. Extent of Condition C. Significance D. Corrective Action Z. Action to Preve'.4 Recurrence
.se ............sessasseessesssssess...s es....assenesessnessenessassessemos,.
Remarks The lack of reference to Drawings 8031-E-1560 and 8081-E=i406 in the Fire Protection Evaluation Report does not constitute a deficiency in traceability. The FPER merely documents the results et the safe shutdown evaluations. The safe shutdown evaluations were performed in accordance with specification 8031-0-35.. 'oeneral Project Requirements for Fire Barrier Review Program,' (transmitted to SWEC via 80890) which states the requirements for identification of saf e shutdown cables and identification , of the raceways that are to be protected by fire barriers. The Electrical Drawings 1 8031-E-1560 and 4081-E-i&O6 are referenced and described in this specification as to I their purpose in the Fire narrier Review Program. It is not the Limerick Project's policy nor is it a requirement of the Liesrick Quality -
.Assuraus Plan to footnoto or athervise cross reference svary source drawing in the TFER. Furthermore, this degree of traceability is not required by 10crR50, Appendix B, faction !!!. That section discueces design centroi in general terms. but does not discuen traceability in specific terme to the extent that the design control anc j traceability methods used at nachtel could be considered deficient. It should be noted that AN8! N46.2.11-1974 is not applicable to the design and construction of the Limerick plant and was not committed to during the design and construction phase as 1
-noted in PSAR Section i>8, regarding Result. tory cul6e i 64. The ecmmitment to AN51 N45.2.tiaif741r. TSAR AppeMir 17.11,!! (h) (changee to Appendit 17.2.!! 1. In Revision 55 cf the 78AR) is limited to the PECO Engineering and Research Department for the operations phase of the plant.
O
!)/l II .hR C& $ / N /Nff lll/8/Fk' Pec'o Rev e4 sig turs / Date .g W8echtel Approval $1gnature' / Date Pageal.of.L
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA-(J.O. NO. 18138) {n OBSERVATION REPORT Observation Report D OR- 008 Rev. O Review Plan: LK- D - 1912 - MS Rev. 1 PART III - RESPONSE EVALUATION The. response to this observation report has been evaluated as follows: A. Causal Factor (s): Acceptable Not Acceptable (Explain): Not applicable B. Extent of Condition: Acceptable Not Acceptable (Explain): Not. applicable C. Significance: Acceptable Not Acceptable (Explain): Not applicable ( D. Corrective Action: Acceptable Not Acceptable (Explain): Not applicable E. Action to Prevent Recurrence: Acceptable Not Acceptable (Explain): Not applicable Additional Action Required: x No Yes (Explain) The concerns raised in DOR-008 concerning traceability of documentation used in performing the analysis of safe shutdown cables is addressed in Bechtel response dated Nov. 22, 1988. Although drawings 8031-E-1560 and 8031-E-1406 are not referenced in the Fire Protection Evaluation Report (FPER), this is not against Bechtel procedures.
................ . ..... ...........................c.. .... . ......................
M li ~l 1 I $ Lead Enfin [er41'gn"aturel / Date APM signjltilre / Date Pcge } of }
STONE & WEBSTFR ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138)
-Ch OBSERVATION REPORT .
U Observation Report No. D OR- 009 Rev. O Review Plan: LK- D - 1903 - S Rev. 1 Reference AI No. : DAI-138 Rev 0 PART I - INITIATION
- 1. Description of Concern Seismic accelerations used in the design block walls may be underestimated.
- 2. Supporting Information In calculation 22.2L66, it was observed that seismic accelerations used to analyze the wall were incorrect for the following reasons:
In calculating vertical accelerations, the calculated horizontal frequency was used. Although this is proba'aly conservative for masonry walls, it could be unconservative if done elsewhere. O
- In factoring acceleration for .57. damping % 27., scaling factors based on the peak of the response spectrum curve were used.
These factors are not applicable at frequencies away from the peak. Resulting accelerations are too low. Additional Documents Attached: None (, n ll O . Lead. Engineer Signature / Date APM Sipature / Date Page 1 of 1 > - - _ _ _ - - - _ _ _ - _ . - - - _ _ - _ _ . - _ ~ . _ , _ _ _ _ _ _ _ _ _ _ _ - - _ _ _ . _ _ . _ . . _ . _
dtoal'#7DEC 19 1988 STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA ( J. O. NO. 19138) OBSERVATION REPORT Observation Report No. DOR-009 Rev 0 PART II - RESPOMEE
- 1. Observation Concurrence:
X Concur with observation (s). Item 1 Horizontal frequencies were used in some cases to cal-culate vertical accelerations that were used in the evaluation of block walls. This resulted in an over-estimation of accelerations for these cases. Item 2 Seismic accelerations used in the evaluation of con-crete block wal1s were,' in some cases, too low as a result of applying inapplicable scaling factors to parts of existing spectra in order +.o calculate higher damping spectral accelerations. Do not concur with observation (Notes if not in concurrence, explain in " remarks" below)
- 2. Response to Observations j A. Causal factor (s)
Item 1 Calculations sometimes included a consideration for the effect of vertical acceleration. In these cases, i horizontal vall frequencies were used to calculate vertical accelerations instead of the vertical wall (- ) frequencies. The causal factor is engineering Page ( 2) of ( 7)
fc4Nf1 O
' STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138)
OBSERVATION REPORT Observation Report No. DDR-209 Rev 0 PART'II -RESPONSE CONTINUED Judgements were made that it was conservative to use the horizontal frequencies, and use of the frequencies would result in overestimating the vertical accelerations. Item 2 The root cause is an incorrect application of a spectral interpolation methodology at non-applicable regions. Adjustments to response spectra for~ damping were made- because the original spectra were for low damping values intended for equipment qualification. O Later, block wall criteria were issued which allowed higher dampings than those for which spectra were available. To, find the appropiate responses for these higher dampings from the low damping spectra an interpolation method was used. This method was intended for application at spectral peaks but was, in some cases, also applied to regions of the spectra other than the peaks. For these cases, spectral responses could be undercalculated. B. Extent of Condition Items 1 & 2 A comprehensive review of all safety related block wall calculations was done to identify any block walls that used inappropriate scaling factors for damping or hori-zontal frequencies to determine vertical accelerations. A sample of other design calculat1ons was also reviewed for variances similar to those noted above. Both these reviews showed the variances to be limited to some safety related blockwalls. O Page (3) of ( ?)
DCA/47T O STONE & WEBSTER ENGINEERING CORPORATION , LINERICK IDCA (J.O. NO. 18138) OBSERVATION REPORT Observation Report DOR-909 Rev. O PART II-RESPONSE CONTINUED C. Significance Safety related concrete block wall evaluations, including those revised as described under Corrective Action, show that all walls meet the acceptance criteria specified' for safety related - block walls. The significance of each DDR item is discussed below. l Item 1 The revised calculations confirm the engineering Judgements originally sade that effects of vertical O' acceleration of the wall are minimal. Use of horizontal wall frequencies instead of vertical wall frequencies to calculate vertical responses conservatively' overestimated the vertical accelerations. Item 2 The revised calculations consider the effects of increased damping directly by using spectra calculated from time-histories. This did not change the responses of the most highly loaded I walls since the adausted spectral peak accelerations did not change. Changed wall loadings were confined to those walls that were seismically rigid and for these block walls the design did not change. 1 l It is concluded that both concerns noted in the DDR have no impact on the existing wall design. ) I O Page (y) of (7)
i DEC-28 '88 09:52 ID:PECO PROJECT MGT DIV TEL NO:21'5-841-4578 #991 P02 , i P. M etc E7 '96 14232 BECM WCSTRN PNCR iST ROOR StenMt? I BTONE &. WEBSTER ENGINEERING CORPORATION LIMERICK IDCA ' (J.D. 30. 181388 ,
)
QSSERVATION REPORT !
~
Observatten' Report No. DDR-teg bev. e
)
PART f f-RESPONAE CONTIMLIED D. Corrective Astion- ' The sigested safety reisted.. block well calculations.have been revised to address.(Item 1) , vertical sooeleratione used in design and (Item 2) 1 refloot use of higher desping asseleratione from l revised response spectra salsulated using floor o i
. .e ....i...
E.. Action to Prevent Resurrence . No further action is required. The condition sannot resur as all bleek well salsulations have been. revised (Etes 1) to document the senservatism in the esimulated vertical responses and (Item 2) the use si ~ proper response spectra omieulated using floor time histories. 4
........................................................,...c....
/
Remarks: None
- \\.W. L nutn _, ~S am O -
et Or)M1 UMIGS PECO Aevier Signature / Data _ Bechtel Rertow Signature /pate O Page (5) of (7)
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) p OBSERVATION REPORT j U/ Observation Report D OR- 009 Rev. 0_ Review Plan: LK- D - 1903 - S Rev. 1 PART III - RESPONSE EVALDATION The response to this observation. report has been evaluated as follows: A. Causal Factor (s):- X Acceptable Not Acceptable (Explain'J The cause is acknowledged to be engineering judgment and an incorrect use of interpolation methods. B. Extent of Conditions X Acceptable Not Acceptable (Explain): The statement that the condition is limited to block walls is reasonable, based on Bechtel's sample review and other calculations reviewed during the IDA. C. Significance: , X Acceptable Not Acceptable (Explain): It is stated that the walls are adequately designed, even when using the d higher accelerations. Revised calculations have not been reviewed. D. Corrective Action: X Acceptable Not Accentable (Explain): All affected calculations have been revised (see Attachment 1). These revised calculations have not been reviewed, but the corrective action is straightforward. E. Action to Prevent Recurrence X Acceptable Not Acceptable-(Explain): The likelihood of recurrence is low since all existing block walls have been evaluated already, and it is unlikely that more will be added.
................. _ ................................................................... l Additional Action Required: X No Yes (Explain) I I
i I l l ! l
............................................... ...... .. .. ...................... \
p hm f. W e :- Lead Engineer Signature
/
'h?
/ Date APM Sig;plature
/ ) l l? f
/ / / Date ]
)
Page ],_ of 1 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - . - - _ _ _ _ _ _ _ _ ___ . - - - - -- 1
DOR-009 Attachment 1 D
' List of blockwall evaluation calculation revised per DOR-009 File No. Calculation No. Rev. (New) 1 22.2 L1 IE8011/ACI702.1 1 2 22.2 L3 IE8011/ACI702.1 3 ;
3 22.2 LS IE8011/AC1702.3 11 ! 4 22.2 L7 IE8011/ACI702.5 1 5 22.2 L9 IE8011/ACI702.10 3 6 22.2 L11 IE8011/ACI702.12 1 7 22.2 L13 IE8011/ACI702.15 2 8 22.2 L14 IE8011/ACI702.17 3 9 22.2 L16 IE8011/ACI702.18 1-10 22.2 L18- IE8011/ACI703.1 2 11 22.2 L20 IE8011/ACI703.3 3 12 22.2 L22 IE8011/ACI703.4 1 13 22.2 L24 IE8011/ACI703.9 2. 14 22.2 L26 IE8011/ACI703.14 4 15 22.2 L28 IE8011/ACI703.20 3 16 22.2 L30 IE8011/ACI703.21 3-17 22.2 L32 IE8011/ACI703.22 1 18 22.2 L34 IE8011/ACI703.26 3 19 22.2 L36 IE8011/ACI703.27 1 20 22.2 L38 IE801:.CI703.30 4 O 21 22 23 22.2 L40 22.2 L42 22.2 L44 IE8011/ACI703.32 IE8011/ACI703.35 IE8011/ACI703.36 1 2 3 24 22.2 L46 , YE8011/ACI703.39 2 25 22.2 L47 IE8011/ACI704.4 2 26 22.2 L49 IE8011/ACI704.5 1 27 22.2 L51 IE8011/ACI756.1 2 28 22.2 L53 IE8011/ACI756.17 3 29 22.2 L55 IE8011/ACI756.21 3 30 22.2 L57 IE8011/ACI756.25 1 31 22.2 L59 IE8011/ACI756.30 7 32 22.2 L61 IE8011/ACI756.35 3 33 22.2 L63 IE8011/ACI756.42 5 34 22.2 L65 IE8011/ACI756.56 3 35 22.2 L66 IE8011/ACI756.57 5 36 22.2 L68 IE8011/ACI756.58 4 37 22.2 L70 IE8011/ACI756.59 4 38 22.2 L72 IE8011/ACI756.61 1 39 22.2 L74 IE8011/ACI756.62 3 40 22.2 L76 IE8011/ACI756.66 2 41 22.2 L79 22.2L79 1 42 22.2 L80 22.2L80 3 l l DOR - co 9 REV o , 1 PAGE 7 CW 7 o______----------_------------ - - - - - _ - - - - - - - - - - _ - _ _ l
STONE & WEBSTFK ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) h OBSERVATION REPORT , Observation Report No. D OR- 010 Rev. O Review Plan: LK- D - 1903 - S Rev. 1 Reference AI No. : DAI-177. DAI-155. DAI-053 { DAI-085 PART I - INITIATION j 4
- 1. Description of Concern )
It is possible that .. insufficient controls are in place to ensere that calculations are ccasistent with as-built drawings.
- 2. Supporting Information During the review of calculations and drawings, a number of discrepancies have been noted:
'(a). In calculation 23.3, the openings in the floor slab at El 201'-0" are smaller than those shown on drawing C-120.
(b). In calculation 22.1 A, the loads on column E-29 are calculated using information which is not consistent with the design drawings. I \ (,/ (c). Calculation 22.8 B designs a fillet veld to attach a bearing plate to an anchor bolt, while drawing C-196 shows a tack weld. j (d). Calculation 97.17.1 states that a note should be added to drawing .j E-1406-2, but the note does not exist on Sheets 2.4.1 or 2.4.1.1 i of E-1406-2. Although some of these discrepancies (particularly (b) & (c) above) will l nave negligible effect on the structural adequacy, it is not clear that t adequate control exists. Response to Action Item DAI-085 indicates that )' EDP 4.46 requires the responsible engineer to evaluate possible impact of drawing changes on calculations. However, Revision 0 of EDP 4.46 was issued on December 23, 1986, long after most civil drawings were issued. Even if a similar requirement was in effect at the time of the issuance of the civil drawings, it is not clear that the responsible engineer can
')
always make the assessment accurately. For example, it is not clear that the concern identified in (a) above will not affect the design calculations on sheets 54 and 55 of calculation 23.3. Additional Documents Attached: ) None
.......................................................... ......... ...o..............
l - W (. Lead Engineer Signature m //
'/ Date APM Sig/ature l
/ Date Page 1 of 1
. _ - - _ - _ - _ - _ - _ _ = _ - _ - _ -
/ !
JAN 161989 STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT Observation Report No. DDR-010 Rev 0 PART II - RESPONSE
- 1. Observation Concurrence:
Concur with observation X Do not concur with observation (Note if not in concurrence, explain in "remar'cs' below)
- 2. Response to Observation:
A. Causal Factor (s) N/A O B. Extent of Condition N/A C. Significance N/A
=====================r.===========================================
Remarks See Continuation Page 1 of [
~ ~
IdY![h $2!$NNLd!$ES$iNEbIskYLW - Q ((,f}& S ETQNE E WERETER ENGINEERING CORPORATION LIMERICN 2 IDCA ( J. O. NO. 16138) OBSERVATION REPORT Observation Report No. DQR-810 Rev 2 PART II - SEEPQMBE IONTTNUED D. Cberect19e Action N/A m E. Action to Prevent Recurrence N/A
.nsa sse ss ss ss s s s s s ss s s s s ss e s.e s e s s.e s ee s s s s a m a n... s s e ss e ne ss s s s
..LM., =l jh I k/49 P5Co', Review "55hnatU[l}[8'j,
~
re/Date Bec5te1 Ihh~o_ va l Sidnhtdre7Date ~ Page 3 of 1
dC11/63 ) O STONE & NEBSTER ENGINEERING CORPORATION LIMERICK'2 IDCA ( J. O. NO. 18138) OBSERVATION REPORT Observation Report No. DDR-Ole Rev 0 CONTINUATION: Remarks Controls are in place to ensure that. calculations are compatible with as-built drawings. Current controls are described in the following summary of the typical i design-construction-as-built process, o Calculations are prepared, checked, and q approved to support the design in accordance j with Bechtel Engineering Department Procedure (EDP) EDP 4.37 Rev 4. o Drawings are prepared and issued for construction in accordance with Bechtel EDP 4.46'Rev 0. As required by Section 5.1 of the EDP an. engineering N. check of the drawings is made for conformance to the design calculations. o As required by EDP 4.46 Rev 0 Section 7. 2, when changes are made to approved drawings, the responsible engineer reviews the possible
-impact of the changes on associated calculations and, by exercising appropriate Judgement, determines whether it is necessary to revise the calculations.
Calculations completed in accordance with the above described process are sufficient and consistent with the drawings. The process does not require, and it is not necessary, that the calculations precisely match the drawings. O Page 1 of _.3_ ;
IQ b6 4 M o~2 V STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA ( J. O. NO. 18138) OBSERVATION REPORT Observation Report DOR-010 Rev 0 CONTINU@TIQE In accordance with the above described procedures, engineering Judgement is exercised to determine whether a calculation change is necessary as a result of a drawing change. The decision regarding need for a calculation change largely depends on whether the responsible engineer needs to perform a computation to determine the significance of the drawing change. When no such' computation is necessary, in the responsible engineers' audgement, no calculation change is required.The calculations will generally be consistent with the drawings, but may not necessarily duplicate every exact detail of the drawings. Such duplication is not required.
,]
[ Calculation procedures have been defined by EDP'.4.'37 since v April 1974. Drawing procedures have been defined by EDP.4.46 since January 1982. Before the EDP's were -issued, Limerick Project Engineering Instructions (EII) were applicable. Throughout the Project duration procedures have not-required that calculations precisely duplicate the drawings. The procedures have required that engineering Judgement be exercised so that calculations and drawings are consistent. With this approach, the procedures cause the calculations to be maintained sufficient to demonstrate from an engineering perspective that the design requirements are met. It is noted that EDP 4.46 was initially issued as Revision 0 in January 1982, prior to the December 23, 1986 date referenced in the Observation Report. The initial issue of EDP 4.46 Rev 0 is identified as a Bechtel Thermal Power Organization document. The subsequent issue of EDP 4.46 Rev 0 in December 1986 is identified as a Bechtel Western Power document. O ~ Page )_ of jl
i v O JbC4/ W STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA-(J.O. NO. 18138) OBSERVATION REPORT Observation Report DDR-010 Rev 0 In the Supporting _ Information Section of the Observation-Report a number of differences between calculations and. drawings are identified. As discussed below, the. differences were reviewed and were determined, in accordance with Project procedures, not to warrant calculation changes, with the exception of one case where a calculation result was inadvertently not reflected in a subsequent revision of the { 4= drawing. (a) In Calculation 23.3, sheets 54 and 55, openings in the floor slab at' elevation 201'-8" are smaller then.those shown on Drawing C-120. By inspection of the drawing and the calculation, and without any computations, it is apparent that the design margin exceeds the force increases resulting from the larger openings. This is evident, with regards to shear, from the fact that the shear force increase is small and the typical slab O i reinforcing (#998" top and intermediate #6912 bottom =- 3.44 sq in) is much greater than the calculated shear reinforcing requirement (1 sq in). It is also evident, with regards to moment, from the fact that the calculation considers moments to be resisted solely by added reinforcing bars around the openings, and the fact that the moments can be. Judged to increase less , than the moment resistance provided by the typical I floor and wall reinforcing that. is also available to resist the bending. Therefore no calculation change is warranted. (b) Calculation 22.1A, sheets 5, and 10b to 12, calculates i loads on column E-29 usitg weight allowances which in j some cases differ from th e design drawings. As stated j in the Observation Report, these variances are small. I and have a negligible effect on structural adequacy. ] The changes that did occur in this area between the ' time cf the calculation and the drawing issue did not warrant a calculation revision. (c) Calculation 22.8B, sheet 27, designs a fillet weld to
- attach an embedded anchor plate to an anchor bolt at the RHR. Heat Exchanger support. The drawing shows a tack weld at this location. As stated in the Observation Report, this small variance has negligible )
effect on structural adequacy. No calculation change 1 O was made nor required because the calculation was conservative and the change had negligible effect on the design. Page jc of 3_ l
J&6A llsfeR rx k ,) STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA ( J. O. NO. 18138) , 1 OBSERVATION REPORT Observation Report DOR-010 Rev 0 CONTINUATION: (d) Calculation 99.17.1 Revision 4, completed on September i 18, 1987 states on sheet 12 that a note should be added i to drawing E-1406-2, but the. note does not exist on the drawing. The note was intended to require that installation of seismic category I raceway supports at the Diesel Generator ceiling be sobJect to Project Engineering approval. The note was required because standard raceway support designs detailed on the l drawing could potentially, under worst case conditions considered in the calculation, not satisfy allowable ; stress criteria for accelerations that could exist at this location. A drawing change was initiated but was inadvertently not completed. In response to DAI 053, a , drawing change adding the note was issued on October 4, l 1988. A field survey showed that no seismic category I [,,l t raceway supports were installed at the Diesel Generator Building ceiling prior to issuance of the drawing l change. A representative sample of civil engineering calculations for seiemic category I structures have been reviewed to dete.rmine if calculations containing statements regarding a need to revise a drawing are present. The review identified 4 such calculations. In all of the 4 calculations the required drawing revisions were implemented. l l (]) Page jLof j_
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. No. 18138) p.. ( OBSERVATION REPORT Observation Report D OR- 010 Rev. L Review Plan LK- D - 1903 -S Rev. -1 t PART III - RESPONSE EVALUATION l The response to this observation report has been evaluated as follows:(Item 'd' only) A. Causal Factor (s): X Acceptable Not Acceptable (Explain): The response indicates that a drawing change was initiated, but inadvertently was not completed. This is a reasonable explanation. B. Extent of Conditions X Acceptable Not Acceptable (Explain): The response states that a representative sample of calculations was reviewed j and no similar cases were found. Furthermore, no other cases were identified ! during the 'JDA review. C. Significance: X Acceptable Not Acceptable (Explain):, The response states that a field survey found no supports installed in the Diesel Generator Building ceiling. D. Corrective Action X Acceptable Not Acceptable (Explain): The applicable note has been added to the drawing. E. Action to Prevent Recurrence X Acceptable , Not Acceptable (Explain): None required.
.m.mmm.ame.mmmmmmmmmm..mm..m..mmmmm.mmmmm....m mmmm.... m....mmmm.....m.ms.m.m.m..m..ma l Additional Action Required: X No Yes (Explain)
)
SEE CONTINUATION SHEET mm. mass.amm....mme.mmmm..mmm.m ..mme.m.mmmmm.am....mmm.. . .m. mm.m..m.m.m.m .m.....a { nt- l l ) {
/ /' Date
~
~
1lead Engineer Signature ~// Date APM Signatufe Page _1 of d
l STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) fi d OBSERVATION REPORT Observation Report DOR- 3 Rev. O PART III - RESMMISE ETALUATION CONTINUATION: l The response indicates that engineering judgment is applied at the time of drawing change to determine if a change to the calculation is required. In three of the cases cited, it is agreed that the adequacy of the structure was not impaired. TFe fourth case (Item d) is somewhat ' dif ferent from the I other three in that it is a failure to incorporate calculation results onto a drawing, rather than a failure to incorporate drawing changes in calculations. For this case, corrective action was taken and an extent of condition was performed, and the condition appears to be an isolated case. The resulting condition of having differences between calculations and drawings is not a desirable one. There is the potential that the calculations could be used in the future to accept other changes (such as increased loadings), and if care is not taken to refer to the latest drawings, the cumulative effects of changes could be overlooked. For example, on Drawing C-269 (Detail 7) a 3/16" fillet weld is shown, which is smaller than that identified in the corresponding calculation, 17-0. Although the smaller weld was adequate at the time of drawing issuance, the design actually has Icss margin than indicated in the calculation. However, because no such O cumulative effects have been identified, there is no specific evidence to suggest that Bechtel procedures and practices do not adequately control the design process for civil calculations. I O s -, rw _3_
STONE & WEBSTFR ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138)
~ /T OBSERVATION REPORT k) Observation Report No. D OR- 011 Rev. O Review Plan: LK- D - 1903 -
S Rev. 1 l Reference AI No.: DAI-207. Rev 0 DAI-214 Rev 0 PART I - INITIATION 1
- 1. Description of Concern The calculation for the design of the 'D' line wall (EL 177' to El 217')
is inadequate with respect to tensile loads from seismic events. l
- 2. Supporting Information
- 1. Sheet 33(b) of calculation 23.1 (b) calculates tensile stresses at Elevation 217' due to seismic (OBE) loads. These tensile stresses are later reduced at EL. 177' by subtracting the weight of the wall between El. 177' and 217' (Sh 38(b)). This method is not correct, and underestimates the tension at El. 177. Response to Action Item DAI-207 indicates that the tensile stress at EL 177 is 14.6 KSF, which can be reduced by 3.6 KSF to 11 KSF. Even if the 14.6 KSF is correct (this cannot be confirmed), this reduction ir not correct rm since the seismic uplift force of 47396 K can only be reduced by the weight of the wall times the acceleration at El. 177', and could l
{) be tension as well ao compression.
- 2. Furthermore, the allowable shear stress at El. 177 was not reduced (using Equation 11-8 of ACI 318-71) to account for this tension.
Response to Action Item DAI-214 indicates that it is not a concern because of the low tensile stress. However, the stress used in the calculation was significant, and it indicates an oversight by the designer. Since both of these above items indicate a misuse of design principles, there is a concern that the design of this, and other, walls is not adequate.
....................................................................................===
Additional D'ocuments Attached: None. p (% c. (g,n n/r/u - HTW Lead Engineer Signature / D' ate APM Sigg%ture
/ Date Page 1 of
>R
SIONE & WSBSTER ENGINEERING CORPORATION ' (% l LIMERICK 2 IDCA ,f (J.O. NO. 18138) ,V I U OBSERVATION REPORT J AN 3 01989 Observation Report tb. DOR-Oll, Rev. O PART II - RESPOtGE
- 1. Observation Concurrence:
X Concur with observation Do not concur with observation. (Note: if not in concurrence, explain in " remarks" below).
- 2. Response to Observation:
A. Causal Factor (s) (i) The method used in calculating seismic tensile load in the "D" line wall underestimated the tensile stress at El.177'-0" because additional tensile sress, resulting from vertical seismic acceleration of the wall mass between El. 177'-0" and El. 217'-0", was neglected. (ii) The code required reduction of allowable shear stress due to axial tensile stress (using Equation 11-8 of ACI 318-71) was not incorporated because of an incorrect interpretation by the originator and checker that reduction is required only when there is a net axial tensile stress in the wall (load factor equal to 1.0 for all applied loads in the load combination). O 'Iherefore, the required reduction was not applied to factored loads. B. Extent of Condition * (i) Similar methodology is used to determine tensile stressas at walls along column line "D", "Mh", "14.1" and "31.9" of the Reactor and Control Enclosure. Review of all other Seismic Category I concrete structure calculations concluded that the correct methodology has been used in these calculations. (ii) Allowable shear stress reduction due to axial tensile stress is required at walls along column lines "D", "Mh", "14.1" and "31.9" of the Reactor and Control Enclosure only. Peview of all other Seismic Category I concrete structure calculations concluded that the concrete walls for these structures are in compression; therefore, this code requirement is not applicable for these walls. C. Significance The discrepancies within the calculations are insignificant because of conservatism (e.g. applied seismic load factor of 1.43 in lieu of 1.25) in the original design of these walls. The design calculations for walls at column lines "D", "Mh", "14.1" and "31.9" of the Reactor Enclosure have
=====.3===============_===========================================rr-----============
Remarks: O C Page l of 8 A
FEB-07 '89 10:36 1D:PECO PROJECT MGT DIV TEL
,._,~g30 "EVIE rcratcFiicC' wtsTW PMA '15iTCCCR ,NO:215 ,841-4578
...#29g,P0g . . _ _ .
.,+d" FICliC & WE6 STER ENGINEERING CORPORATION.
M LIMERICK 2 IDCA f' I __.T ' _-_ ...__. . . W.0r IC.- 18138 F - H 4 -- , --
/ b-OBSERVATION REPORT ,
7 s. observation toport ns. DOE-011, Rev. O j
'PART II - _ RESPONSE CORTINUED
~ been revised demonstrating the structural adequacy of these walla using
.gg.vy.i.i n J udvluvi .c4 G- s q. ism mMis e.tr ;;z x s--" :c M. -
D. Corrective Action The desion calculations (includi calculation 23.1.0) for alls at column lines "D", "Mh", "14.1' and "31.9 of the moector enclosure have been I revised to demonstrate the structural adequacy of these walls and are available in pottstown and san Francisco. These cel M *
- revisions
. are performed.using an appropricts method to calculate the well axial tensile stress and using a reduced allowabla sheer streen.pse ACI, 381-71
.. :... equation 11 4 .. , . ,
"; y, ,.; . . . .. ..
E. Action to Provent securrenos . .
", ne results of this observation report and the calculattert serisions
. . ..have been discussed.with the involved engineers.
~
.t. : . ro further action is required because: - -
The observed discrepancies wre limited to the design of four exterior concrete walls of the Reactor Enclosure. De discrepancies are shown to be insignificant by calculation revisions. The calculation revisions have demonstrated the structural adequacy of those walls using the appropriate methodology. - 1 e 4 g.g, r00 .hd L d@' }$/#r uA 2 ' L/m4 ths/4 sechtel ApptwaI signaturfs fonte PEco pov: et 'signates 7 /Date
. s-
. Page 3 of br 74 .
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT OD Observation Report D OR- 011 Rev. O J Review Plan: LK- D - 1903 - S Rev. _J_
'PART III - RESPONSE EVALUATION The response to this observation report has been evaluated as follows:
A. Causal Factor (s): X Acceptable Not Acceptable (Explain): The cause is an oversight by the engineers, and by a misinterpretation of the ACI code. B. Extent of Condition: J Acceptable Not Acceptable (Explain): A review was performed to determine that the concern is limited to the reactor building. C. Significance: X Acceptable Not Acceptable (Explain): The response indicates that the revised calculations demonstrate
~'
that the structure is adequate, even after correcting the errors. SWEC has not reviewed these calculations in detail. D. Corrective Actions X AcceptaMe Not Acceptable (Explain): The response states that all affected calculations have been revised. A cursory review of Calculation 23.1(b) and 22.2c(k) indicate correct methods have been used. E. Action to Prevent Recurrence X Acceptable Not Acceptable (Explain): All affected calculations have been revised, and no further calculations of this nature are anticipated. Additional Action Required X No Yes (Explain)
.................................................a.......=' ====..==============..=====
< IL e a L 3Ah/ Datc ctsksp.nnite alales Lead Engineer Signature ' ' AP4 Signature / Date S
Page 1 of g Jk
LIST OF CALCULATIONS REVISED FOR DDR #11 22,2C (b) Rev. 9 22.2C (d) Rev. 8 22.2C (h) Rev. 9 22.2C (3) Rev. 11 23.1 (b) Rev. 4 23.1 (b-1) Rev. 4 23.1 (g) Rev. 4 23.1 (h) Rev. 5 42-3.2.2 Rev. O I O DOR - o / l REV O PAGE 6 0F f
l 1 1 STONE & WEBSTTE ENGINEERING CORPORATION ) LIMERICK 2 IDCA j l (J.O. NO. 18138) hV ossxRVATION REPORT . Observation Report No. D OR- 012 Rev. _j Peview Plan LK- D- 1903 - S Rev. j D - 1908 - S Rev. 1 Reference AI No.: DAI-126 Rev. 0 PART I - INITIATION
- 1. Description of Concern There are unconfirmed assumptions and methods used in evaluation of structures for missile impact on the Spray Pond Pumphouse Roof.
- 2. Supporting Information In the review of two ' areas subject to tornado missile loadings, the following concerns were identified:
a) In calculation 70-H, the roof of the Spray Pond Pumphouse (SPPH) is evaluated for aircraft impact, but not for tornado missiles. It is not clear that the aircraft impact (at 15") is more severe than the tornado missiles given in FSAR Table 3.5-4. Re m nse to Action Item DAI-126 suggests that the velocities given b (c k Table .3.5-4 can be reduced by 20% for vertical impact, as suggested by the Standard Review Plan (SRP). This may not be appropriate, since the FSAR has taken exception to the SRP in other areas, including the spectrum of missiles considered and allowable ductility ratios. b) In evaluating the SPPH roof for the aircraft impact,.the analysis considered the target mass as the entire mass of the beam and supported slab. Actually, an ' effective' mass, not total mass is appropriate. Response to DAI-126 indicated that the results i would not be affected if the missile mass were reduced by 38'/.. l There does not seem to be a rational basis for reducing the l missile mass. The response also indicates that the calculation of target mass is in accordance with BC-TOP-9A. Review of this document did not identify any guidance for the case in question.
.....................................................=.........==..=.====.=============
Additional Documents Attached: None b [. bw //- /'/ -// Mk Lead Engineer Signature / Date APM Sig6ature / / Date Page 1 of 1 (
o STONE & WEBSTER ENGINEERING CORPORAaAuN LIMERICK 2 IDCA G 20/76 ( J. D. No. 18138) FEB 0 81989 OBSERVATION REPOP.T Observation Report No. DDR-12 Rev 0 PART II - RESPONSE
- 1. Observation Concurrence:
X Concur with observation (Item A) The basis for determining the aircraft as governing over the tornado missiles in the evaluation of the spray pond pumphouse roof for missile impact is-not documented in the calculation. _K_ Do not concur with observation (Item B) The methodology used in the analysis of the spray pond pumphouse roof for aircraft impact is acceptable. (See " remarks" below) k 2. Response to Observation: A. Caussi Factor (s): (Item A) Missile effects for the spray pond pumphouse (SPPH) were analyzed in two calculations. Calculation 70-H addresses aircraft impact and calculation 25.1-T addresses tornado missile impact. In Calculation 25.1-T the tornado missiles were determined to be less critical for the SPPH than the aircraft missile . This was determined by comparing the maximum tornado missile energy with the aircraft energy for the case of a normal impact. As done, the comparison is applicable to walls, but does not specifically address the roof for which the design criteria prescribes an aircraft impact angle of 15 degrees. For the case of a roof impact, the aircraft was also Judged more critical than the tornado missiles based on the comparison for the case of a normal impact. This Judgement considered the aircraft impact angle of fifteen degrees, and the tornado missiles vertical l l , velocities of 80% of their horizontal velocity as l \ Page ik of $_ t
STONE & WEBSTER ENGINEERING CORPORA Y N LIMERICK 2 IDCA (J.O. No. 18138) OBSERVATION REPORT Observation Report No. DDR-12 Rev. O CONTINUATION: specified in SRP Section 3.5.1.4. However, the originator and checker of the calculation did not confirm this assumption with the FSAR. B. Extent of Condition: (Item A) The determination of tornado missiles as less critical thw 'he aircraft missile, based on the
, comparison of tb missile energies only for the case of a normal inipa c t, was made in Calculation 25.1-T. This condition applies to the SPPH, the reactor building and the j control structure. No other calculation has used this comparison of missile energies.
N C. Significance: (Item A) A calculation revision has been performed to document the comparison of the tornado missiles with the aircraft- missile (See Corrective Action). The calculation confirms and documents the original Judgement, applicable to the SPPH, the reactor building, and the control structure, that the aircraft missile at a 15 degree impact angle is more severe than the tornado' missiles wi+h a maximum vertical velocity of 80% of their horizontal velocity. An FSAR change has been issued to state this vertical velocity. Therefore, the concern has no significance. The Observation Report suggests that reducing tornado missiles velocities 20% for the case of vertical impact on a roof, as provided in the SRP, may be inappropriate as the FSAR has taken exception to the SRP in other areas, including the spectrum of missiles considered and allowable ductility ratios. The missile spectrum description in the FSAR (Table 3.5-4) is identical with that for Missile Spectrum A of the SRP except that the wood plank missile velocity in the FSAR is slightly higher and the automobile missile frontal area is not defined. Allowable ductility ratios are described in BC-TOP-9A which is referred to in the FSAR (Section
-( 3.5.3).
Page 3 of 9 {
r STONE & WEBSTER ENGINEERING CORPORATION
-- LIMEhiCK 2 IDCA (J.O. No. 18138)
OBSERVATION REPORT Observation Report No. DDR-12 Rev. O CONTINUATION: As allowable ductility ratios given in the SRP (Section 3.5.3) are the same for tornado missile vertical velocities ranging from 70% (SRP 3. 5.1. 4 missile spectrum II) to 100% (SRP 3.5.1.4 missile spectrum I) of the horizontal velocities, there is no apparent relationship in the SRP between the vertical velocity reduction and ductility ratio. The vertical velocity reduction in the SRP is an independent characteristic. The use of a missile vertical velocity of 80% for Limerick is consistent with the SRP and is appropriate. D. Corrective Action: (Item A) Calculation 25.1-T Rev. I has been revised to address the comparison of the tornado missiles with the () rx aircraft missile for the case of a roof impact at the reactor building, the control building and the SPPH. To document, the tornado missile vertical velocity criteria used, an FSAR change has been issued (LDCN FS-1623) to add to Table 3.5-7 the provision that tornade missiles vertical velocities shall be taken as 80% of their horizontal velocities. E. Action to Prevent Recurrences (Item A) Missile analyses of structures are complete. Calculation 25.1-T Rev. 2 (transmitted to SWEC via Transmittal No. P-0330 dated Feb 13, 1989) has been completed and LDCN FS-1623 has been issued. Therefore, no further actions are required.
=================================================================
1 I l O Page 4 of 9 ) l l l
\ Je STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA
( J. O. No. 18138) OBSERVATION REPORT Observation Report No. DDR-12 Rev. O CONTINUATION Remarks: (Item B) We do not concur that there are inappropriate methods used in evaluation of the SPPH roof beam for. aircraft impact. The analytical model used, including its target mass, was a reasonable representation of the complex problem, and the application of BC-TOP-9A was appropriate. The -SPPH roof evaluation in Calculation 70-H consisted of two separate analyses. In the first analysis, the steel roof beam was analyzed as absorbing all of the aircraft impact. In the second analysis,-the roof slab was analyzed as absorbing all of the impact, as a two-f~s. way slab supported by.the building walls. I The roof beam analysis, which is questioned in the Observation Report, calculated effective mass by the method given in Bechtel Topical Report BC-TOP-9A Revision 2 which is referenced as the applicable procedure for design of missile barriers in Section 3.5.3 of the FSAR. Although the topical report focuses on tornado missiles,the underlying principles provided in the report using the energy balance method are equally applicable to the phenomenon of an aircraft impact. The topical report addresses the case of an automobile impact, and this case is similar to an aircraft impact. Methods to calculate target effective mass are given in Section 3.4 of the topical report. As stated in this section, The effective target mass during impact varies from a low value at initial contact and generally increases to an upper limit during or at the end of the impact duration. Due to the complex phenomenology associated Page SI of 9
r o STONE & WEBSTER ENGINEERING CORPORATION G&en57 LIMERICK 2 IDCA
" (J.O. 18138)
OBSERVATION REPORT Observation Report No. DDR-12 Rev. O CONTINUATION: with missile impact, no general analytical solution is available to evaluate the effective coupled mass on a continous time basis. The average effective mass can, however, be estimated, utilizing the results- of-impact tests on reinforced concrete beams wherein the measured . maximum structural response was used to back-calculate the average mass during impact. Based on these data, the following formulae provide a lower limit estimate of_Me- (which results in an upper limit O estimate impact)." of kinetic energy after For steel beams Me is given in Section 3.4 by equation (3-17) as equal to (Dx + 2d) Mx , where Me = Average effective mass of target during impact Dx = Maximum missile contact dimension in the x direction (longitudinal axis for beams or slabs) d = Depth of steel beam Mx = Mass per unit length of beam i The beam analysis calculated the target effective mass according to the applicable equation (3-17) given in BC-TOP-9A, resulting in an effective mass equal to the entire mass of the beam and supported slab. SWEC's suggestion to use a 0.33 factor (Reference DAI-126) in computing Me would be correct only if the missile's contact length were to be so small as to qualify as a concentrated load. As the contact length I of the aircraft missile is longer than the the beam span, the entire length of the beam is correctly used l l in computing Me in accordance with BC-TOP-9A. Page _k_ of T 1 L 1
FEB-13 '89 13:58 ID:PECO PROJECT MGT DIV TEL N0:215-041-4578 #382 P03 t STONE & WERSTER ENGINEERING CORPORATION LIMRICK 2 IDCA (J.0. 18138) RESERVATION REPORT i observation Report No. DOR-12 Rty, e CONTINUATION: l Response to DAI-126 indicated that the strain energy ] requirements would not be affected if the missile mese j were reduced by 38X . As stated earlier, because the aircraft mass is not concentrated'on the beam span, the 38% reduction faster wculd be a emneervative factor { crampati. bis wLth EWEC's suggested f actor of 2. 33, 1 O t f i l DOR - o / A REV O PAGE ~7 OF 9
===...r.....e......s====as===...............s======se..........
p~PECoo . % .tM A u.nn [@iew Signature /Dat'e 6 nhaLkla$w Bechtel Approval signatur(/Date l
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT j V Observation Report D OR- 012 Rev. O Review Plan: LK- D - 1903 - S Rev. ,1_ FART III - RESPONSE EVALUATION The response to this observation report has been evaluated as follows: (ITEM A) A. Causal Factor (s): , 1 Acceptable Not Acceptable (Explain): The cause is assumed to be that the originator and checker did not confirm an assumption by checking the FSAR. B. Extent of Conditions y Acceptable Not Acceptable (Explain): The calculation in question applied to the SPPH, reactor building, and control structure. Other calculations did not use this method. C. Significance X Acceptable Not Acceptable (Explain): It is agreed that when the tornado missile velocities are reduced by 207. the aircraft impact will govern the design. ( D. Corrective Action X Acceptable Not Acceptable (Explain): It is not agreed that there is not a relationr, hip between allowable ductility ratios and missile velocities. However, the FSAR change initiated will address the relaxation of commitments and will be reviewed by the NRC. E. Action to Prevent Recurrence: X Acceptable Not Acceptable (Explain): All affected calculations have been revised and are considered final. No recurrence would be expected. , 1 Additional Action Required: X No Yes (Explain) i SEE ATTACHED PAGE 0 /L Et L Lead Engineer Signature tb3lP1 '
/ Date 0(Ak L. 0. Mile.
AW1 Signature 2h4le9
/ Date Page 1 of j
i DOR-012, Rev. O ITEM B The response indicates that the effective target mass is consistent with Topical Report BC-TOP-9A. Although not addressed in the calculation, it can be shown that the contact area is quite long for a 15* impact angle. Tornado missiles are smaller and would engage a smaller target area, but this was checked (using 807. of horizontal velocity) and it is not more critical than the aircraft impact. l O i O DOR - O / J REV o PAGE 9 OF 9
STONE & WEBSTTR ENGINEERING CORPORATION LIMERICK 2 IDCA oc (J.O. NO.'18138)
.(
i V OBSERVATION REPORT . l Observation Report No. D OR- 013 Rev. 0 Review Plan: LK- D- 1908 - S Rev. 1 Reference AI No.: DAI-068 Rev. O PART I - INITIATION
- 1. Description of Concern Blast pressures have not been considered in the evaluation of missile door 293, and it could be a significant load.
-2. Supporting Information l
Review of Wooley Document 8031-A-11-90-8 indicates that the railroad accident blast pressure of 16.4 psi is not considered in the evaluation of the missile door. Response to Action Item DAI-068 states that the force produced by this pressure (282 KIPS) is small compared to the missile l load. However, the effect of the load is determined by its time-history of application, as well as the magnitude. The missile load is for a very
/f,\ short duration. Since the calculated capacity of the door to a concentrated k- load is only 209 KIPS (Sheet 18), this blast pressure load could be significant.
L 1 I Additional Documents Attached: None l l ............................................................. ......................... Y 5 {. n //- /4 N ' APM Sighature / Date j Lead Engineer Signature / Date Page 1 of I
t t)D64 /604
'3AN 0 5193g !
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK.2 IDCA ( J. O. NO. 18138) OBSERVATION REPORT Observation Report No. DDR-013 Rev e PART II - RESPONEE i l
- 1. Observation Concurrence:
JL_ Concur with observation Do not concur with' observation (Note if not in concurrence, explain in ' remarks" below)
- 2. Response to Observation:
A. Causal Factor (s) Engineering Judgement was made that the railroad accident blast pressure loading is less than the missile load and that these effects are not concurrent. O The Judgement was based on data available in the ic11owing calculations: o Bechtel calculation 101.65 Rev 1, which had shown the response to missile loads to be j elastic. o Bechtel calculation 42.9 Rev 0, which had computed the blast pressure loading. o Bechtel Calculation 101.64 Rev 0, which had computed the natural period and pressure resistance of the door in the process of evaluating tornado effects. It was not deemed necessary at the time to document the Judgement. q l $ j E3335353335533333333333333333333333333333s352385353EE332R333333EE Rom.rks None l () Page 3 of je_ L
- OI)6#' ! )
O l STONE & WEBSTER ENGINEERING CORPORATION LIMERICK IDCA ( J. D. NO. 18138) OBSERVATION REPORT Observation Report No. DOR-013 Rev 2 PART II - RESPONSE CONTINUED B. Extent of Condition The condition applies to other missile doors. The condition does not apply to the associated building structures as they are evaluated by calculations for pressures equal to or exceeding the railroad accident blast pressures. C. Significance Calculation 101.65 Rev 2 performed by Bechtel in response to DAI-068, confirms and documents the original Judgement that the pressure loads are less O than the missile loade and that the loads are not concurrent. The calculations show the door response to pressure is elastic. The calculations also show the door response to missile loads is elastic. A comparison of these calculations show that the door missile response exceeds the door pressure response. Calculated responses are based on the following parameters: Dynamic Load Missile Pressure Load Distribution Concentrated Uniform Peak Force (k) 2940 11.7 psi x(Door Area) = 201 Duration (sec) .000598 .135 Door Procerties Natural Period (sec) .072 .072 Static Resistance Force (k) 209 34.53 psi x(Door Page 3 of 6 zu
O l V STONE & WEBSTER ENGINEERING CORPORATION LIMERICK IDCA ( J. O. NO. 18138) OBSERVATION REPORT Observation Report No. DOR-013 Rev 0 Responses were determined -using curves relating structural response to dynamic loads. For t..* concentrated missile loading, figure 3-1 of BechtsA Design Guide C-2.45 Rev 0 (Ref 1 on sht 2 of Calculation 101.65) was used. For the pressure loading, figure 6-7 of " Structures To Resist The Effects Of Accidental Explosions" , Dept Of Army TM 5-1300 (Ref 4 on sheet 21 of Calculation 101.65) was used. The above discussion demonstrates that the Judgement exercised in the evaluation of the missile doors was adaquate in that the railroad accident blast missile was the governing design load. It is evident from observation of the load-response curves that the blast pressure load, which has a peak force of 1/3 the door resistance and a duration of O nearly twice the door period, is less critical than the critical blast missile load, which has a peak force of 14 times the door resistance and a duration of 1/120 of the door period.' The blast pressure on door 293 is 11.7 psi computed in 1 Calculation 42.9, is less than the 16.4 psi pressure l used for the building design. The 16.4 psi pressure is the pressura computed in Calculation 42.9 for the Unit 1 & 2 Diesel Generator Building location nearest to the railroad. The door, being further away from the railroad, is subjected to less pressure than the nearest point on the buildings. The reference in the response to Action Item DAI-068 to a 282 k pressure force conservatively applied a door pressure of 16.4 pai. If the calculated pressure of 11.7 psi is applied, the door pressure force is reduced to 201 k. l
..g. 9 o, e )
sh
aw-w m u m u m w e ae ~1 - -1, .m O dbent&o l BTONE & WE85TER EN3!NEERING CORPORATION LINERICK IDCA (J.O. NO. 14134) OBSERVATION REPORT Observation Report No. DDR-813 Rev e PART II - RESP.DMSE.CQRTIRUED Calculation 181.55 Rev 2 aise evaluates door 294 for pressure the railroad accident and shows that the pressure leading on the door is less critical than the missile leading. By similarity, the other doors, which are farther away from the railroad and therefore sub;ested to less pressure, are also adequete. D. Corrective Action No additional corrective action is required. O Calculation 101 65 Rev 3, prepared in response to this Observation Report, demonstrates that the railroad assident blast pressures were sorrectly Judged to be less eritissi than the railroad aseident missiles. E. Action to Prevent Resurrence No action is required. 35533333sssass3333333333333333333333333333333333333333333333333
= al am . ==== d --
Mm.C A= ke. 1,..ture .r
.e.Ae1 A,,,o.., 1,nat re
! o. ...e c oi i L _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA l (J.O. NO. 18138)
!n) . OBSERVATION REPORT . (/ -
Observation Report D OR- 013- Rev. O Review Plan: LK- D - 1908 - S Rev. 1 PART III - RESPONSE EVAIRATION The response to this observation report has been evaluated as follows: i Causal Factor (s): X Acceptable Not Acceptable (Explain): A. The response states that an engineering judgment was made that blast pressures do not govern the design. The subsequent calculations confirm,this judgment. B. Extent of Conditions X Acceptable Not Acceptable (Explain): The response indicates that the condition applies to other missile doors, but not to building structures. This has been verified.
- C. Significance: X Acceptable Not Acceptable (Explain):
The calculations performed (101.65, Rev. 2) ' demonstrate the ability of the Q doors.to resist the blast pressure loads. D. Corrective Action X Acceptable Not Acceptable (Explain): Calculation 101.65, Rev. 2 and applicable portions of Calculation 101.64, Rev. I have been reviewed, and were found to be adequate. The blast pressure loads were taken from Calculation 42.9, Rev. 2. This calculation was not reviewed. E. Action to Prevent Recurrence: X Acceptable Not Acceptable (Explain): All missile doors have been evaluated already. Additional Action Required: X No Yes (Explain) I-
- f. ps A f .
Lead Engineer Signature / Date APMSign'at[e /'Datd f Page y,_ of I 1
s STONE & WEBSTTR ENGINEERING CORPORATION LINERICK 2 IDCA (J.O. N0' 18138) 4 (. OBSERVATION REPORT . Observation Report No. D OR-__014 Rev. O ! Review Plan: LK- D - 19TF - S Rev. T Reference AI No.: DAI-256 Rev. O DAI-283 Rev. O PART I - INITIATIOE
- 1. Description of Concern Insufficient justification is provided in the resolution of NCR's.
- 2. Supporting Information In the review of NCR's, it was observed that in responding, the rationale given does not' provide adequate . technical justification. The response.
to NCR 5448 (Sheet 8) gives ' design requirements are met' and the response to NCR 10724 gives ' cutting bars will not impact structural integrity of slab' as . the technical basis of acceptance. Although it may be true that calculations are not necessary in these cases, a more detailed justification should be provided. i V It is noted that NCR 10682 does provide a calculation for resolution of the NCR. Response to Action Iter DAI-283 states that this was done because the evaluation is more complex than for NCR 5448 and NCR 10724. Reviewing NCR 10682, it appears as though calculation was provided only because the client requested additional information -for evaluation of the nonconforming condition. It does n% appear that there is adequate documentation of engineering judgments made in resolving nonconformances.
...........................................................................=.........==
Additional Documents Attached: . None. [. bw // - /'/ M APM Sigtiature / Date Lead Engineer Signature / Date Page 1 of _1 l 1
)
g\/ d'24/7/ STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA ( J. O. NO. 18138) OBSERVATION REPORT J AN 3 0 N69 Observation Report No. DOR-014 Rev 0 PART II - RESPONSE
- 1. Observation Concurrence:
Concur with observation X Do not concur with observation (Note: if not in concurrence, explain in " remarks" below)
- 2. Response to Observation:
A. Causal Factor (s): N/A B. Extent of Condition: N/A C. Significance: N/A O i D. Corrective Action: N/A (v) E. Action to Prevent Recurrence: N/A
=================================================================
Remarks Justification provided in the resolution of nonconformance reports (NCR's) is sufficient and is in accordance with Bechtel Engineering Department Procedure (EDP) 4.61. which includes the following requirements: o The responsible engineer decides on appropriate disposition of the NCR and completes the NCR Form (Exhibit A). Close out Action is reviewed by the Group Supervisor and approved by the Project Engineer (Section 3.1 e), o The Group Supervisor confirms the correct technical solution before forwarding the NCR to the Pro]ect Engineer for approval (Section 3.1h). Detailed justification of the engineering disposition is not required by the EDP. When in the Judgement of the responsible engineer a calculation is not
/~N warranted, it is sufficient and in accordance with the
(_,) procedure that the responsible engineer's disposition Page al_ of 7
n Y-' Joc4/7/V STONE & WEBSTER ENGINEEING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT Observation Report No. DOR-014 Rev 0 CONTINUATION: be in the form of a statement such as " design require-ements are met", or " cutting bars will not impact structural integrity of slab". The responsible Engineer initials the disposition attesting to the fact that it is appropriate. The Group Supervisor reviews the dis-position and also initials it attesting to the fact that it is technically correct. When no calculation is performed nor referenced in the disposition, the Group Supervisor's review includes examination of supporting documents. The Pro 3ect Engineer reviews the disposition and signs the NCR form attesting to his concurrence with the disposition, and attesting to the fact that the disposition has been prepared in accordance with EDP requirements. (G) , The Observation Report cites three examples of NCR dispositions to support the concern that insufficient Justification is provided in the resolution of NCR's. NCR's 5448 and 10724 are cited as examples where insuf-ficient Justification of the engineering disposition is provided. NCR 10582 is cited as an example where detailed Justification (a calculation) was provided, not because of the complexity of the problem, but because the client requested additional evaluation. The following discussion shows that justification provided in these NCR's was adequate. NCR 5448 identified 4 pump support anchor plates having anchors bent more than permitted by the drawing. The engineering disposition required that the bolts be nondestructively tested to determine the effects of bending. On the condition that no bolt cracking was found, the disposition required that each anchor plate be extended by adding a welded plate and three addi- ! tional bolts at the top and bottom. The revised anchor plate assembly has 10 bolts as compared to 4 bolts in the original design. The bent bolts were considered to remain effective in resisting applied loads, although l at a small reduction in capacity. No calculation was ]'
- deemed to be warranted because it was apparent by s'
[ inspection that the revised assembly had a greater cepacity than considered in the original design.
- l. Page 1 of 7 w_-____-__-________ _ _ _ l
I L o
\ '-
g4 nj Ll STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J. O. NO. 18138) OBSERVATION REPORT Observation Report DDR-014 Rev 0 CONTINUATION: The Justification given in the engineering disposition is that " repaired base plates meet design require-ments." The disposition Justification is sufficient and meets the EDP requirements. NCR 10724 identified cutting of 4 E-W reinforcing bars rather than 3 bars allowed to be cut by FCR C-11968F. The FCR had also permitted cutting of one N-S bar. Review of Calculation 22.4g showed that cutting of the l additional bar would not prevent the slab from perform-ing its design function. No calculations were needed to i reach this Judgement. Justification given in the engin- I eering disposition is " Concur with field's recommenda-tion. Since cutting top 4 #9 (E-W direction) and 1 #9 (N-S. direction) bars will not impact strength of slab." (~') ( j The disposition Justification is sufficient and meets the EDP requirements. NCR 10682 identified local deviation of the contain-ment wetwell liner plate surface from specified tolerances. The deviation was located near a platform attachment. The engineering disposition, based on review of the deviations- and the design requirements was to use as is, and to provide a revised pistform attachment detail that would bridge over the area of liner plate deviation. No calculation was needed or prepared to make this disposition. Subsequently, the NCR was reopened by a client request to provide a more ' detailed description of the liner deviation, A detailed measurement of the plate surface was made and plotted by field engineering. During the field measurements a nonconformance in the installation of the revised platform attachment was identified and reported in the reopened NCR. The new nonconformance was that the revised attachment connections were not centered on the liner plate embedded stiffeners as required by the original engineering disposition. This additional nonconformance introduced complexity that could be resolved only by means of engineering calculations. It is this complexity that is referred to in the response to DAI-283. Engineering disposition of the reopened NCR (_- reiterated the original disposition of the plate Page 4 of 7
I 1 1 FEB-06 '89 15:18 ID:PECO PROJECT MpT DIV .TEL NO:215-841-4578 #293 PO4 ! I a O ; l lk"&fTlW 5tCNE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA j (J.Q. NO. 18134) DOSERVATION REPORT Observation Report DOK-514 Rorv e CORTIMUATION: surface nonconfermance deviation se *use es Le'. The revised platform attachhment nonconformance was also dispositioned a use se .ie', with a referense to Caicu-istion No. 113-1-53 as documenting .the.- estytmeering i evaluation. It le noted that,the calculatica'lsedresses" { only the platform attachment deviation, ',1 mad isot the ] liner plate surface deviation. The' disposition gusti-ficatten is sufficient and moete EDP requirements. Detailed Justification (a calculation) Le. inskuded in O the new disposition new nonconformance. because of the semplexity of the j l I I
...========... I REMARKS l
mo l l -
. t,ida & HP>lvA %_) . ),*JL]W1 flvol64 l PEco M WLew Signature /Date techtel Approval Signature /Date 1
I O . DOR - o / 4 REV o Pans 5 op ,7 w-
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA y, (J.O. NO. 18138)
" -(x OBSERVATION REPORT' l
Observation Report D OR- 014 Rev. _0, Review Plan: LK- D - 1911 - S Rev.,1 PART III - RESPONSE EVALUATION The response.to this observation report has been evaluated as follows: A. Causal Factor (s): Acceptable Not Acceptable (Explain): Not applicable. P. Extent of Condition: Acceptable Not Acc ptable (Explain): Not applicable. C. Significance: Acceptable Not Acceptable (Explain): Not applicable.
~f D. Corrective Action: Acceptable Not Acceptable (Explain):
Not applicable. E. Action to Prevent Recurrence: Acceptable Not Acceptable (Explain): Not applicable.
...............o.......................................................................
. Additional Action Required: X No Yes (Explain)
See attached page. O & E. (L~ zMer Lead Engineer Signature / Date
% C4 dte & a w ife APf1 Signature able?
/ Date Page i of 1
l l
- l. STONE & WEBSTER ENGINEERING CORPORATION l LIMERICK 2 IDCA (J.O. No. 18138)
^ f%
i A OBSERVATION REPORT V Observation Report DOR- m Rev. O PART III - RESPONSE ETALUATION CONTINUATION: It is agreed that there are no requirements in Bechtel's QA system that would require a detailed justification on the NCR. This does not cause tehnical problems since the procedures do require the group supervisor to review the disposition and concur with it. Assumedly, he would be able to discuss the detailed justification with the responsible engineer if the justification were not clear. Although this system makes it difficult to recreate the logic used to accept the nonconformance, no incidences were found during this review where the resolution was not appropriate. 1 p - d DOR - o 14 REV o PAGE 7 OF 7
i 1 STONE & WEBSTFR ENGINEERING CORPORATION LIMERICK 2 IDCA l \ (J.O. NO. 18138) OBSERVATION IRPORT . Observation Report No. D OR- 015 Rev. ,p, ] Review Plan: LK- D- 1903 .__}is_ Rev. J j l
- i Reference Al No.: DAI-265 Rev. O j i
FART I - INITIATIGE
- 1. Description of Concern There is a deficiency in the verification of a non-standard radiation 1
-shielding computer code that could lead to possible design problems. [
EDP-4.37 requires that "non-SCP" computer codes be verified.
- 2. Supporting Information San Francisco Radiation Protection Group Report SFRP-76-05 is the reference document for QAC, a non-standard radiation shielding computer code, which has been used in Calculation S102.1. Report SFRP-76-05 contains the mathematical basis, an input description, six sample problems, and a program listings but this report does not appear to include verification of results obtained for these six sample problems via the QAC code. Without O a comparison of QAC results with results obtained by other valid means, V it cannot be established that QAC is performing correctly.
............................................................................= ==.....==
Additional Documents Attached: l None.
. == . ...............
O .. ........ .. ........................................ NY5 All,w lAA ffh & APMSigng(ure
/ Date Lead Epgipe'r figt,a~tuf6 /~Date Page 1 of 1
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA fpgg /pg& ( , (J,0, No.18138) MAR - 31989 l OBSERVATION REPORT (S.agg f) l, Observation Report No. _.Q ,OR- M Rev. O PART II - RESPONSE
- 1. Observation Concurrence: ,
Concur with observation H Do not concur with observation (Note: if not in concurrence, explain in
" remarks" b61ow)
- 2. Response to Observation:
A. Causal Factor (s)
- 8. Extent of Condition C. Significance D. Corrective Action E .' Action to Prevent Recurrence
................n...................................................................
- Remarks AMENDED RESPONSE The following is an amended response prepared in accordance with agreements made during a telecon on December 15, 1988. (Reference DTR-002, dated December 19, 1988, DC# IDCA 1499 ) The response was supplemented in accordance with a telecopied request from G.
Hirst (SWEC) to S. Yim (Bechtel) on January 5,1989. The response was f urther amended per the telecopied request of G. Hirst to S. Yim on February 3, 1989. There is not a deficiency in the verification of the non-standard computer code QAC which was used in Calculation $102.1. To comply with requirements in EDP 4.37 for using non-standard codes, the QAC code is referenced via i;alculation S000.1 to a San Francisco Radiation Protection Group Report, SFRP-76-05, which provides necessary background information such as mathematical model and capabilities. Further, to comply with EDP 4.37, a program verification for the QAC application to $102.1 is performed
~
within Calculation $102.1 (page 9) where results for a 30" diameter pipe source are shown to be in good agreement with the standard computer code GRACE II. (Note that DAI-?65 response erroneously indicated that the QAC code verification is contained in Report SFRP 76-05. ) As per the EDP, the checker of the calculation has the option of i'urther verifying results by alternate meanc if deemed necessary. Therefore, in calculation $102.1, the originator and checker have confirmed the overall correctness ( .of the QAC code and the adaquacy of its documentation and verification for that calculation. Page 1 of .7
7_-_________ 1-1 STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA 1h N /3h O V (J.O. No.18138) OBSERVATION REPORT l Observation Report No. D OR- 015 Rev. O PART II - RESPONSE CONTIN,,3 Q Supplement to DDR-015 Response: The non-standard computer code QAC was verified for its application in Calculation S102.1 by c0 aparison to standard computer program NE348 GRACE II results on page 9 of that calculation. The comparison was made for the case of direct shine from a 30" HBB pipe with a dose point location at the midplane at a distance of 16 feet. Direct shine is also the dominant contribution for radiation in the labyrinth entry coder evaluation. Since the QAC algorithm for direct shine in the same regardless of dose point location, comparing midplane results against GRAa II should be sufficient QAC verification for this application. However, a Revision 1 to Calculation S102.1 has i been performed to further verify the appropriateness of the assumption that QAC is also : valid for non-midplane dose point locations. Revision 1 to Calculation $102.1 as transmitted to SWEC via S0514 compares the results of the QAC runs in Revision O to results from the standard computer program code NE650 CYLSEC for all dose point locations and labyrinth lengths. The results are all in reasonable agreement (i.e., within about 20%) as shown in the table in Revision 1. m This confirms the applicability of the QAC code to the situation in Calculation $102.1 and gives further confidence in the verification of QAC for any other non-midplane
\ -
application.
- The standard computer program code CYLSEC, which is used for comparison, has been extensively validated by comparison to other industry and Bechtel standard computer program codes: GRACE I, GRACE LI, QAD-CG, QAD-P5A, PIPEND, and ANISN. A description of the validation calculations performed has been ca" mitted to SWEC via S0514.
Responses to Supplementary Questions:
- 1. The Validation Report is Section III of the NE650 CYLSEC manual. A copy of the manual title page signoff sheet has been transmitted to SWEC via S0569.
- 2. Section 2.3 of the Validation Report summarizes validation calculation N.650.3.
TLat calculation hrul two different objectives: (1) to compare CYLSEC and QAD-CG results in a r elatively complex geometry, and (2) to demonstrate proper functioning of several specific CYLSEC features and computational methods by edparison to hand calculations. The first objective used the first two geometries described in Section 2.3. The second two geometries were used for validating features such a ray traces, symmetry determination and integration limits calculations. Box sources were used for the second set of geometries to simplify hand calculations for boundaries, ray lengths and angles.
- 3. As noted in Section 2.6, no published results of solutions to the ANS 6.2.1 I radwaste shielding benchmark problems were available at the time of the rm validation calculation N.650.6. Therefore, no specific comparisons were made.
In fact, the benchmarking efforts for that particular sample problem were discontinued by the ANS 6 subcommittee. Page 1 of 2. l
)
J
STONE s WEBSTER ENGINEERING CCAPORATION LINERICK 2 IDCA MM Q (J.O. No. 18138) OBSERVATION REPORT Observation Report No. _A_ OR _ pig _,Rev. O PART II - RESPONSE CONTINUED 4~. a. Calculation N.650.9 was performed at the time NE650 Version A1 was the
" current" production version of CYLSEC. 'the calculation was not revised for any subsequent releases of Ch.SEC since its only purpose was to compare results and computation time from the early test Version 1 to the first production version A1. Therefore, Section 2.9 properly refers to Version A1.
The current (1989) versions of CYLSEC are 83 on the UNISYS mainframe computer and C1 on the microcomputer. l
- b. The programs NE650 SCAN and NE650!N are not used with the current (1989) versions of CYLSEC. NE650 SCAN was an optional pre processor program to scan a CYLSEC input file and check for inconsistencies and completeness ;
of input data. NE650!N was also an optional pre-processor program which presented a question and answer sequence for simple geometries and prepared an input file for CYLSEC.
- c. The program errors reported for Version A1 were:
#1) In order to change samma source energies or source s+rengths for individual samma groups, data for all groups needed to be re-entered.
#2) For shield slabs normal to the z axis with a box source, no credit was given for. attenuation through the slab for certain dose point locations outside the source perimeter.
#3) For dose points located inside a cylindrical penetration normal to a source cylinder or normal to the z axis of a box source, samma rays which pass through only a portion of the shield wall were.
improperly attenuated. The dominant gamma rays which pass through the end of the penetration toward the source were properly attenuated. Reported error 85-3 for Version A3 was that for cylindrical penetrations parallel to a source cylinder, dose points in certain areas which were both behind the shield wall and to either side of the penetration would be improperly attenuated. The most important dose points directly behind or above/below the penetration were properly attenuated,
- d. Version B1 of CYLSEc was an enhancement of the prior Version A4, which had corrected all previous error notices. CYLSEC wan enhanced to include multiple source and internal shield capabilities. Also, a " simple" input option for typical shielu ng geometries, and an option to save input responses on a file were added. The NE650 SCAN and NE650AL'D ancillary O programs were deleted. Their features were incorporated dinctly into CYLSEC. Options such as suppressing screen prompt display and printing dose summaries on the screen were added. Fortran coding was also aonverted from Fieldsta to ASCII.
Page 1 of _2.
STONE & W188TER ENGINEERIMO CORPORATION
- LINERICK 1 IDCA (J.D. N3.18188) -
1tX/? /$2 A O OBSERVAT10M SEPORT Observation Report No. L. OR _,g,1L, Rev. ,,2., 1 I PART II - RESPONRI (EHfTlM1XED
- 4. d. (continued)
Version St includes a modification to the
- simple
- Input option to allow convenient dose point specification.
The current (1999) Version 88 on the mainfrant also imeludes some minor . i modifications to the output format. The microcomputer Version C1 is the same computationally as the mainframe version 88. In susmary, th's Version 81 of CYLetc has evolved into the current Version SS (mainframs) and Version C1 (microcomputer) through the above sequence. l l
- e. Calculation 5103.1. Revision 1 uses the microcomputer Version C1.
Response to Additional supplementary Question: In response to the previous request for information on calculations performed to validate the CYLSEC (Nt650) code, a copy of the test portion of the validation Report i for that code was provided. It describes, in general terms, the calculations and heir relation to the developmental seguence of the code. The table from the Valida-tion Report entitled " Validation Manual for cyLsEC NE650* (transmitted to SWBC via 30688) provides a summary of the appliashility of the calculattens performed to the . various code versions. Except for Calculations N.650.9. N.650.10. and N.660.11 (which are not relevant to the recent code versions), all calculations have been updated to be applicable to sede versions up to sad including Version C1. That would also include previous versions such as the current mainframe version ns. . I l /Wth Tr9 J n }.f .fr k lr MJZ/ 3 hM s ate
. Rep .a ene ur. / oate phiei appeovai signature
. 1 o,1
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) 't
' OBSERVATION REPORT Observation Report D OR- 015 __ Rev. 0_
Review Plan LK A- l on't - MS Rev. j_ PART III - RESPONSE HIaLUATION The response to this observation report has been evaluated as follows: A. Causal Factor (s): Acceptable Not Acceptable (Explain): NOT APPLICABLE B. Extent of Condition: Acceptable Not Acceptable (Explain): NOT APPLICABLE C. Significance: Acceptable Not Acceptable (Explain): NOT APPLICABLE D. Corrective Action: Acceptable Not Acceptable (Explain): NOT APPLICABLE E. Action to Prevent Recurrence: Acceptable Not Acceptable (Explain): q NOT APPLICABLE 1 l l
========== s==-=============================---=-s===-=======s-=================-=====-
Additional Action Required: X No Yes (Explain) SEE INSERT A ) O --- -- ---------------- ---------- ---------------------- --- - ,
---- ---- E--- r------/ Date dm oc& w AIK Signature a we & ~/'Date
~
Lead Engineor' Signature Page 1 of 1 a
\ STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA i (J.O. NO. 18138)
. ( ~) .
U OBSERVATION REPORT l Observation Report DOR- E Rev.,_g, PART III - EESPGsSE ETALUATION CONTINUATION: INSERT A DOR-015 raised the concern that non-standard computer program QAC, which was used in Calculation S102.1 Rev. O, was not verified in accordance with EDP 4.37. To address this concern, Bechtel agreed to benchmark the QAC results in this calculation by performing Calculation S102.1 Rev. 1 using the CYLSEC (NE650) computer code and to provide evidence for review that the version of CYLSEC used in Calculation S102.1 Rev. I has ur.Jergone a formal verification process. Accordingly, Bechtel provided a copy cf the CYLSEC Validation Report, the programmer's' manual title page sign-off sheet with the appropriate signatures and dates, and a full discussion in the third amended response to DOR-015 explaining the developmental sequence of revisions to the code. Based upon the fact that acceptabic agreement between CYLSEC and QAC results was achieved in Calculation S102.1 Rev. 1, the statement in the third amended response that Calculation S102.1 Rev. 1 uses CYLSEC Version C1, and'the indication in Table 1 of the CYLSEC
/'T, Validation Manual that the CYLSEC Validation Report is applicable to Version
() C1, it ir concluded that the QAC program produced acceptable results in Calcular su S102.1 Rev. O based on the simplified source / shield model employed in this calculation. It is noted that the verification of QAC results in Calculation S102.1 Rev. O by means of a comparison with CYLSEC results in Calculation s102.1 Rev. 1 is based upon a particular geometry application for which QAC was intended. However, according to Report Number SFRP-76-05, there are other distinctly different source / shield geometry applications of QAC, and these have not been tested in this comparison. Therefore, it must be concluded that the benchmarking of QAC results in calculation S102.1 Rev. O does not constitute a generic verification of QAC, and the validity of QAC results in other calculations in which this code may have been used should be established. v Pyc. - 7 of 7
STONE & WEBSTFR ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) n OBSEITATION REPORT .
/'N -) Observation Report No. D OR- 016 Rev. jl,.
Review Plan LK 1912 - MS Rev. _1_ Reference AI No.: DAI-187. Rev. O _ PART I - INITIATION
- 1. Description of Concern There are deficiencies in traceability in the Fire Loading Calculation No. 22-3, Revision 1, dated December 4,1987 that could lead to possible design problems.
- 2. Supporting Information The condition identified in description of concern is not in accordance with commitments contained in the Limerick Quality Assurance Plan for Design and Construction, Section 3.2.1, and FCAR Appendix 17.2BII(h),
which requires traceability in accordance with 10CFR50 Appendix B, III-Design Control, and ANSI N45.2.11-1974. Review of Fire Loading Calculation No. 22-3, Revision 1, dated Dec. 4, 1987, indicates the following concerns regarding traceability:
/N A. The following inputs to the calculation for the RHR pump B cubicle (s are not clearly traceable:
(1) Lube oil quantity, reference calculation page 74 (2) Cable insulation quantity, reference calculation page 74 B. No area evaluated in the calculation includes traceability for the quantities of combustibles indicated. . For example cable insulation is not cross-referenced to an issued drawing, computer-listing, or to actual field collected data, reference calculation pages 27, 31, 59, etc. C. The letters noted on page 1 of the calculation are not traceable, They are neither identified nor attached to the calculation. These letters form the basis for revision 1 of calculation No. 22-3. The action item response explained that the referenced letters are not now retrievable, but that the values used in the calculation were determined to be correct. Additional Documents Attached: None l a
)LeaaEng:.nlerSignature
& , m ll?2}9R
/ Date-90LW%nk k?M Signature nInkt
/ Date i Page1off(
L __ __ _______________ - - __ _ _ - _ _ - _ _ _ _ -
1 l l l STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J. O. NO. 18138) im OBSERVATION REPORT Observation Report No. DOR- 016 Rev. O CONTINUATION: l l l l The response to the action item provided references for the inputs ncted in A and B above. However, Calculation No. 22-3 does not provide the definitive traceability indicated by the response, and personnel l unfamiliar with these documents might not be able to reconstruct l the traceability to the vendor drawing or to the indicated series I of electrical drawings. Also, reference should be made more l specifically to the exact drawing including revision, not a series l of drawings in order to provide the required traceability. O O O P"oe a of a , 1 - - _ _ _ _ - _ _ _ _ _ _ _ - _ - _ _ _ - _ - _ - _ _ - _ _ _ - _ l
4 STONE & N1887ER ENGi>ERRINC 00RPORM20N 'L CA / 77 LIMERIC4 3 IDCA (J.O. NC .18188 i {pg hB1RVAT!014 REPO)tf Observation Report No. 1,OR .,9jj,,Rev. _g,, PAAT !!..._EEAPONEE
- 1. Observation concurrence:
Concur with observation
.J;L, Do not scncur with observation (Note: if not in sencurrence explain in
*romarks' betow)
- 2. Response to observation: )
l A. Causal Factor (s). j 1
- 4. Estent of Condition C. Significance D. Corrective Action E. Action to Prevent Recurrence s e s e s .s ss s ss s s s s s s ss e m e r e r e s s on s s e n s s s s s e s s e s e s s eis s e n e s s e mies s e s siis es s e e ,s e s s n es es s e s Remar ke There are no deficiencies in the traceability of combustibles inf ormatica used in the fire leading calculation 12-8. Revleien i s cated December 4.1987, that could lead to possible design problems. -
All combustible listed in the subjeat calculation can be traced to project controlled doenents/drawinge such as equipment vendor drawing and :a.ble tray layout drawings. regineare can find these documents through the Project Decument Centrol Center. Appli:t.ble revlalon numbers of documents a*e> not required eines the calculation le in accordance with the latest information available at t>e time cf the calculation revielen. Changes in combustible loads mt;st be coordinated with the project fire prote: tion engineer for the purpose of upicting the subject calculation. Also note that the requirement to identify sources of cata used in calculations was not in effect when the original calculation was performed. ($ sit EDP 4.3'? in 1979. ) rer t h erroore , this degree of tracasbility is not specifically required by ic;rR60 Appendix 8.!!! or the Limerick Quality AlsJrance I'lan for Design and Construction. Section 3.2.1. App!!cability of F8AR Appoidir 17.2511(b) (changed to 17.2.11.1 in Revision 65) and the commitment to AN5! H45.2.11 therein is !!mited to the P100 Engineering and Research bepartment for the operation of the plant, essness...... ss ssss s s s sss s s s ss ss sss e s.ms nises sa sis m u s....miis s e m miss s s s a s s a.s s...m as s e I 2b 'f ' g hlf /bl/M Faco Review Signature // Phts Rech Approval Signature / bate Page ),of g.,
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA i (J.O. NO. 18138) p OBSERVATION REPORT Observation Report D OR- 016 Rev. O_ Review Plan: LK- D - 1912 - - MS Rev. l_ PART III - RESPONSE EVALUATION The response to this observation report has been evaluated as follows: A. Causal Factor (s): Acceptable Not Acceptable (Explain): Not Applicable B. Extent of Condition: Acceptable _ Not Acceptable (Explain): Not Applicable C. Significance: Acceptable Not Acceptable (Explain): Not Applicable n D. Corrective Action , Acceptable Not Acceptable (Explain): Not Applicable . E. Action to Prevent Recurrence: Acceptable Not Acceptable (Explain): Not Applicable Additional Action Required: X No Yes (Explain) The concerns raised in DOR-016 concerning traceability of documentation used in performing fire loading calculations is addressed in Bechtel response, dated January 20, 1989. Although the specific drawings and vendo' information is not referenced in Fire Loading Calculation Number 22-3, this is not against Bechtel procedures. 4.........................................................................
\ 11rAA Lead Engine'er" !Iignkture 2 2W
/ Date APMSignpure
$ bh/
'/ Da'te Page d of 3
STONE & WEBSTF3 ENGINEERING CORPORATION LIMERICK 2 IDCA 1 (J.O. NO. 18138)' l p ossxxvarios anroar . Obserystion Report No. D OR- 017 Rev. 1
; itW.aw Plan: LK 1- 1903 -
S e Rev. L Reference Al No.: DAI-125 Rev. O L PART I'- INITIATION .
- 1. Description of Concern- I 1
There does not appear to be adequate direction for allowable weld stresses for' factored load in the design criteria-(Specification C-115).
- 2. Supporting Information In reviewing Calculation 22.2L66, the allowable stress for a fillet weld on Sheet . 63 -is given 'as .52Fyw, ' where Fyw is the yield stress (60 ksi) '
for : thel E70 electrode. This gives ~ an allowable stress of 31.2 kei. On sheet 71 the allowable stress- is given as .52 (70 kei), or 36.4 kei. Since there is no direction given in the design criteria (Specification C-115), there appears to be no consistent criteria. In , the design of cable tray supports, the allowable weld stress is given asi .1.25 times the working stress allowable for factored loads (see Sheet 5 of Calculation 99.13.21 for example) This gives an allowable stress of 26.25 ksi.
. Response to Action Item DAI-125 indicates that the. allowable stress for factored loads can be as high as 1.7 times the working stress allowable of 21 kai or 35.7 ksi, in . accordance with Section 2.8 of the AISC code.
This value is in fact used in the design of missile resistant doors (see Sheet 24 of Wooley Report Number 37672, for example). However, this' is not specified in the design criteria, and ' it seems that if-this approach is used, factors of safety for weld design will be less than those for design of the members. For example, the allowable tensile stress of .85 Fy'in the design criteria is only 1.42 times the working stress allowable. The design criteria is unclear with regard to allowable weld stresses for accident and extreme environmental conditions, and there appears to be a variety of different allowable values used in the' calculations.
...........................................................................==....==...=
Additional Documents Attached: None. . {. Y zn 88 APM SignRure / Date Lead Engineer Signature / Date Page 1 of f,, l
c S'IONE & WEBSTER ENGINEERING CORPORATION I LIMERICK 2 IDCA I U (J.O. NO. 18138) V OBSERVATION REPORT FEB021980 Observation Report No. DOR-017, Rev. O PART II - RESPONSE
- 1. Observation Concurrence X Concur with observation Do not concur with observation. (Notes if not in concurrence, explain in " remarks" below)
'2. Response to Observation A. Causal Factor (s)
For the design accident and extreme environmental conditions, the design criteria (C-115) specifies allowable stresses for steel structures (.90Fy for bending, 0.85Fy for tension, and 0.50Fy for shear). The specified allowables are not appropriate for welds. C-115 paragraph 5.2.3 does require that the design shall be "in accordance with referenced codes and standards", and the AISC Specification and AWS (Dl.1) code are listed ( i as references. As different allowable weld stresses can be inferred from the design criteria, different and inconsistent allowable weld stresses have been used. 'Ibe cause of the inconsistency, cited in the Observation Report, is that the criteria do not explicitly address weld stress allowables. B. Extent of Cordition A survey of 71 calculations, representative of all (on the' order of 4000) civil calculations for steel structures, has been performed to identify the allowable stresses used for design accident and extreme environmental conditions. 'Ihe survey showed that different and inconsistent weld stress allowables were used, but only one calculation (101.73) used an allowable (37.5ksi) inconsistent with NRC Standard Review Plan (SRP) 3.8.4. Similar infrequent uses of weld allowables inconsistent with the SRP are expected in other calculations. I C. Significance In accordance with section 3.8.4 of the NRC's Standard Review Plan (SRP), a stress increase factor of 1.6 over normal code allowables is acceptable for design accident and extreme environmental conditions, and for the same condi- 1 tions a factor of 1.7 is acceptable for load cases which include pipe rupture j missile, jet and pipe whip (local) effects. 'Ihis results in E70 electrode allowable stresses of: 4 . .........................................................=........ .............. l ,h Remarks: Page 3 of f
- - - . - - - - - - _ _ _ . . . u
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) f (,/ OBSERVATION REPORT Observation Report No. DDR-017, Rev. O PART II - RESPONSE OWIINUED Fall = 1.6 (.0(Fu)) = 1.6 (.3(70)) = 33.6 ksi For load combinations without pipe rupture local effects Fall = 1.7 (.3(Fu)) = 1.7 (.3(70)) = 35.7 ksi For 16ad combinations including pipe rupture local effects he normal code allowable stress is 21 ksi (AISC Specification). Test have shown a minimum fillet weld factor of safety of 2.2 (1983 AWS code commen-tary, section 8.2). B us shear yield occurs at a minimum of 46.2 psi (2.2 times 21 ksi). With the stress increase factor of 1.7, under Design Accident and Extreme Environmental Conditions, the ratio of the allowable weld to the yield stress (35.7/46.2 = .77) is less than C-115 permits for members subjected to bending (0.90), tension (0.85), and shear ( 0.5Fy = .87). F/g y fT Consequently, it can be seen that although the maximum allowable stress d increase factor based on the SRP is greater for weld design (1.7) than for tensile stress in members (1.42 as stated in the Observation Report), the factor of safety is greater for welds than for members. For the missile-resistant doors referred to in the Observation Report (Ref. Wooley Report Number 37672, sht. 24), the use of an allowable weld stress of 35.7 ksi, corresponding to a stress increase factor of 1.7, as discussed above, provides a factor of safety greater than for the members. Bis allowable is acceptable for missile door design and meets t.he door acceptance criteria which permits yielding. In calculation 22.2L66, since the allowable of 36.4 ksi was used for a "no good" determination, and the re-evaluation on the following pages used an acceptable allowable stress (ksi) of 31.2 ksi, the calculation portion using the higher allowable is no longer applicable. As referenced in the Observation Report, an allowable stress (26.25 ksi) equal to 1.25 times working stress is used in Calculation 99.13.21 for the design of cable tray supports. He allowable used is satisfactory as it is less than the allowable stress discussed above. O I
- V Page 3 of 5
ID:PECO PROJECT MGT D1V TEL NO:215-041-4578 #293 P02 FEB
- M'
._ . .-06 '09 15:16: .uw. LCML .Li?JW d ' ' 5 - -
l O 1 l STONE & WEBSTER ENGINEERING CORPORATION j j L1MERICK 2 IDCA (J.C. NO. 18138) !*[f [ i OsSElWATION REPORT Observation Report No. D3-017, Hev. O PART II - RESPONSE CXHTINUED The calculatiori survey (Ref. E, Extent of Condition) of 71 calculations showed that all of the calculations except one used allowable stresses consistent with the 8RP limitations. The exception is calculation 101.73 which evaluated RNR fisat Exchanger Supports for SSE loading combination, ustng allowable stress of 37.5 kei. This allowable is 124 above the 33.6 kai allowable that is appropriate, as discussed above. It is acceptable from a safety standpoint as the resulting factor of safety would remain greater than for menbers (ratio of allowable weld stress to the yield stresa a 17.5/46.2 = 0.41), The use of the allowable stress of 37.5 kei in calculation 101.73 has no impact as the weld stress derived Esma the load shown in the calculation is only 13.3 kai. . O Although allowable weld stresses used in the reviewed calculations are different and not consistent, they provide acceptable factors of safety that are no less than those prescribed for members by the design criteria. Based on the review, it is expected that all other weld calculations are accepcable, and the concern has no significance. D. Strective Action ' The design criteria will be revised by February 37, 1989 to include weld allowables for design accident and extress environmental conditions. An allowable stress increase factor of 1.6 has seen adopted for loads that do not include pipe rupture local effects and 1.7 for load cases which in accordance include pipe rupture 3.8.4 with Section missile, jet or!s of the h1t Standard Review Plan.ipe whip (local) e No welding modif toations or calculation revisions are needed because of 1 the ins 19nificanoe of the inconsistency. E. Action to Prevent Recurrenot The design criteria will be revised to include weld allowables. No further action is required. O . %. ........ ..............................................e..-............. m.o nrvtow signature o li.M.ML - t14l88 '
/pate
& nirk e4.14sMs1 stehtel Approval signature / /Date PU M
STONE & WEBSTER ENGINEERING CORPORATION ] LIMERICK 2 IDCA ! (J.O. NO. 18138) ) OBSERVATION REPORT Observation Report 1 0R- 017 Rev. 1 Review Plan: LK- L - 1903 1 Rev. L PART III - RESPONSE EVALUATION The response to this observation report has been evaluated as follows: A. Causal Factor (s): y Acceptable Not Acceptable (Explain): The cause of the use of a variety of different weld stress allowables is acknowledged to be inadequate direction in the design criteria. B. Extent of Condition: 1 Acceptable Not Acceptable (Explain): A total of 71 calculations (out of approximately 4000) were reviewed, and only one case was identified where allowable stresses higher than those in the SRP were used. C. Significance: Y Acceptable Not Acceptable (Explain): C\ The one case where a higher allowable stress had been used was found to V be insignificant. Even in this case the allowable was only 12% greater than the SRP allowable. D. Corrective Action: Acceptable Not Acceptable (Explain): The design criteria will be revised to provide the appropriate allowable stresses. No calculation changes are necessary. E. Action to Prevent Recurrence: Y Acceptable Not Acceptable (Explain): None required, cince the design criteria will be revised. Additional Action Required: 2o N Yes (Explain) Q .......................................................................................
$ D.NINt m {. Yn Lead Engineer Signature
'/ Date APM Signature 2.
/ Date Page di of I l
- _ , _ _ - _ - _ _ _ _ - _ _ _ - - _ - _ _ _ _ _ - - _ - _ - - . _ _ _ _ - - _ _ = _ _ _ _ - _ _ _ ._- _ - . - _ __ ___ __-_-___ _ - _ _ _ _ _ - . - _
STONE & WESSTTE ENGINEERING CORPORATION ,. H n ' LIMERICK 2 IDCAL 1 (J.0. NO. 18138). OBSIRTATION REPORT .
-Observation Report No. D OR- 018 - Rev. l'
. Review Plan: LK- D - 1903 -
E Rev. L . Reference AI No.: DAI-077,- Rev. O PART I - INITIATION
- 1. Description of Concern Calculation 6900E.14 ~does not reflect accurate ampacity ratings. l for the' 2-1750 MCM penetration conductors- per phase and should incorporate actual vendor data. .
l
- 2. Supporting ~Information
' Upon review ' of the _tesponse concerning calculation 6900E.14, the
. following was determined:
A. The penetration vendor, as documented to IDCA via transmittal S-0413, has " revised the ~ original 1200 ' ampere rating (Sheet
- 4. of Enclosure .1 and Figure 7A to the calculation) to ^ allow the 2-1750 MCM penetration conductors to carry 1680 amperes
. continuously, and B. The increased 1680 ampere rating can.now be properly protected by the existing protection scheme ' as depicted on time-current.
coordination sheet 7A to the calculation which shows use of redundant 51 relays.. However, the calculat{on still indicates the previous 1200 ampere' rating and improper design inputs could be extracted from the exieting calculation. l 4 Additional Documents Attached: None
==.......... .......................................... . ... .....................
Lead Eng'ineer SignacMe A /Ef Fate 7fff { h l 7 APM Sip 6ature
/ / Date Page 1 of i 1 .
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. No. 18138) 010291988 ( 5- 0558 ) l [] OBSERVATION REPOEf V i Observation Report No. D OR-018 Rev. O ; PART II - RESPOESE
- 1. Observation Concurrence:
x Concur with observation Do not concur with observation (Notes if not in concur-rence, explain in " remarks" below)
- 2. Re sponse to Observation:
A. Causa1 Factor Design schematic review determined the need f or increased penetration ampacity capability. ' B. Extent of Condition The original value of 1200 amps as stated in Calculation 690 0E .14, Re v . 6, applies to the penetration conductors A of both Re actor Re circulation Pump Motors of Unit 2. Id Full load current for each pump motor is 960 amps. This upgrade applies only to those limited number of medium , voltage penetrations protected by relays (as opposed to f uses or breakers). C. Significance Since the overcurrent capability of the penetration conductors of 1680 amps has been confirmed by the vendor and the conductors are properly protected by the existing protection scheme, there is no impact on plant operability or plant safety. D. Corrective Action Ca lculation 6 900E .14, Re v. 6, will be updated by March 15, 1989, to reflect the penetration conductor current rating of 1680 amps. ( Con t . ) r ( ( ....................................................................................... Remarks T- 6/3 5
5 TONE & E llttR INGIPttRING C0ktonAfgog LIMERICK 2 IDCA (J.O. NO. 18138) [.$- 0837) OBSEITATION RDORT Observatten Report No. J R-0!g gey, g PART IT = ISSPOSSE l E. Action To Prevent Recurrence The change committed in the corrective action section covers all medium voltage penetrations, ao no additional action is required. O l
. =...................................................
.. .....................c.
AT Eco Review lignature / W lla(n / Data k I Rahtd
. a l ApTrWEF A d $ hlighatur'ay Al N A/ Dated f H + gg T- 6/3 5
#83' 1 *I 1
STONE & '1EBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) l OBSERVATION REPORT C Observation Report DOR- 018 Rev. O Review Plan: LK- D- 1903 - E Rev. 1 PART III - RESPONSE EVALUATION ] The response to this observation report has been evaluated as follows: A. Causal Factor (s): X Acceptable Not Acceptable (Explain): The plant design required increased penetration ampacity capability necessitating vendor concurrence for this requirement. B. Extent of Condition: X Acceptable Not Acceptable (Explain): The extent of condition only pertains to the medium voltage penetrations protected by relays. C. Significance: X Acceptable Not Acceptable (Explain): ('j% Vendor confirmation has concurred with the penetration's overcurrent capability and the plant's safety or operability is not impacted. D. Correcitive Action X Acceptable Not Acceptable (Explain): A commitment has been made to update Calculation 6900E.14, Rev. 6 by March 15, 1989 to reflect the penetration conductor current rating of 1680 amperes. E. Action to Prevent fecurrence: X Acceptable Not Acceptable (Explain): No actioni to prevent recurrence is required.
.........................................................................-=..--=...-..=
Additional Action Required: X No Yes (Explain)
...................................................... --- - ....... . <.-...-==..
- f
( Lead Engineer Signature / Date APM Si ture
'/ Date Page 1,of f i
l (
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. No. 18138) . onssavATIon REPORT s Observation Report No.- D OR- 019 Rev. 0; Review Plan: LK- D- 1903 - s Rev. 1 D - 1904' - s' n
-Reference AI No.: DAI-029, Rev. O PART I - INITIATICE
- 1. Description of. Concern There is a potential deficiency -in the design of welds .of tube steel to flat surfaces.
- 2. Supporting Information Review.of calculation 99.13.21 and Sheet 3.4.8.6. of drawing E-1406-2 indicates that the. length of : fillet weld made along the edge of a - tube s teeli member is not as long as assumed in the calculations.
Response to Action Item DAI-029 (Revision 0) confirms'this situation. The response indicates that the design in question is adequate based on conservatism but this has not been' demonstrated numerically. / \ . s a m m e s s e ss e s s a m m m m a m e s mus em a s s a m a s s o m s m e m a s s e s s m a a m m e s s a m m m m a s a s s a s s a s s a s s a s s a m a s s e m e Additional Documents Attached: None masasemasosummessnessmuseumsmessnespassanssammasummassumes seasses samassammesessemos
'{ $n /Z.f/'l P Lead Engineer Signature '/ Date FM Sig)(6ture /
/ / Date Page 1 of 1
f[ l
' FEB 0 '31989 STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA
( J. O. NO. 18138) 1 OBSERVATION REPORT Observation Repor. No. DOR-019, Rev.0
~
G PART II -
RESPONSE
REISSUED R
- 1. Observation Concurrence: DATE MAR 1 6 1989 X Concur with observation u'me Do not concur with observation (Note:
m3 % w .
- 1 not in concurrence, explain in " remarks" below)
- 2. Response to Observation: _
A. Causal Factor (s)
' Calculation 99.13.21, qualifies standard cable tray
- support (Type S-8.4). which consists of welded tube steel members. Weld symbol and size are shown on Design Drawing E-1406-2, for the tube to. tube connections, h6 wever the drawing does not show a required weld length.
In accordance with the applicable code velding symbol (AW5 A-2.4), " Fillet welding extending requirements beyond abrupt changes in direction of the velding shall be indicated by means of additional arrows pointing to each section of the Joint to be welded, except when the ! weld-all-around symbol is used." The code requirement is that, when velding across the end of tube steel members, weld need only be provided along the straight length of the tube steel face, rather than the full nominal width of the tube steel, ; unless the design specifies otherwise. Engineers misinterpreted the term " abrupt changes in
. direction" as requiring veld along the full nominal width of the subc steel. Misinterpretation of the code resulted in a calculation which assumed longer velds than actually provided by the design drawing.
O Page _1 of jL
. STONE AND WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.D. NO. 18138)
OBSERVATION REPORT I Observation Report No. DDR-019, Rev. O PART II - RESPONSE CONTINUED B. Extent of Condition ' Tube steel members are used predominantly for commodity supports. These include Piping supports, which are designed by the Plant Design Group, and Electrical, HVAC. Instrumentatier., and Multi-Commodity supports, which are designed by the Civil Group. Tube steel is also occasionally used in miscellaneous structural applications, designed by the Civil Group, such as CRD
~
support steel, equipment bracing, and bottle storage racks.
) The condition does not apply to Pipe Supports. The Pipe Support Installation Specification, P-319-2, specifically requires in Section 4.14.2 that tube steel welding wrap around the corners, except when Joining tubes of the s4me size, where only the flat surfaces require welding.
With respect to HVAC supports, the -concern applies potentially only to calculations completed before May of 1987. At that time a note ( Note 21. Drawing C-1362-
- 2. Rev. 1) was added to the design drawings. The note specifically stated "Except for all-around welds, welding shown for tube steel shall, as a minimum, be on the flat surfaces only ( i. e. no radius welding is required) unless noted otherwise."
All Unit 2 Civil calculations for commodity supports which use tube steel (approximately 300 calculations) were reviewed. The review identified 43 calculations that had considered a weld length including the radius of the tube steel when the drawing weld symbol would require velding along the " flats" only. l l O Page _3_ of ]_
- l. . . _ _ _ _ - _ _ _ . . _ _ _ - _ _ _ _ _ - _ _ _ - _ _ - _ _ _ _ - _ _ -
m W-l' . O
~ .
STONE AND WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA ( J. D. NO. 18138) OBSERVATION REPORT
- Observation Report No. DOR-019, Rev.0 PART II -
RESPONSE CONTINUED A review of all drawings for miscellaneous Safety Relat'ed structural steel applications (approximately 400) identified 22 drawings that include welds for which no weld length.nor all-around symbol is shown. A review of 19 calculations applicable to 16 of these drawings identified it calculations wh1ch considered welds longer than required by the design drawings. C. Sign'ficance i As discussed under Section B, above, the condition does not apply to piping supports, and review of all O calculations tube steel for all identified other 43 inappropriate weld lengths were assumed. commodity calculations supports using in which Review of 25 fgh of these calculations shows the weld designs are / acceptable by inspection when considering the correct weld length, associated reduction in connection capacity, and the stress levels indicated in the calculation. The remaining 18 calculations have been revised to consider correct weld lengths,and the revised calculations demonstrate ' hat
. the design criteria is met. The concern thus has no significance with respect to commodity supports.
With respect to miscellaneous structural applications, review of 19 calculations identified 10 calculations in which inappropriate weld lengths were used. Review.of these le calculations shows that the weld designs are acceptable by inspection, using a basis similar to that discussed above for commodity supports. Thus the assumption of inappropriate weld lengths in these 10 calculations'is not significant. Similarly the concern is expected to be of no significance with respect to miscellaneous structural application calculations not reviewed. () Page _T of _7_
FES-09 '89 14:31 !DtPECO PROJECT MGT D:V TEL Not215-E41-4575 :!41 PC2 3 '89 16:41 BECHTEL ESTRN % CR 157 FLCCR #. WiC O gpj74 BTONE AND WEBSTER ENGINEERING. CORPORA 710N ! LINERICK 2 IDCA (J.D. NO. 18138) OBSERVATION REPORT Observation Report No. DDR-819, Revie - PART - REBPONBE CONTINUED D. Corrective Action , As demeribed in Sectson C above, la commodity support calculations have been revised to consider correct veld inngthe. 4 E. Action to Prevent Recurrence O In response to this Observation Report, general notes have been added to Drawings E-1496-2, C-1425, and C-1426 to slarify veld lengths required far tube steel end connections, for Electrical. Instrumentation, and Multi-Commodity Supports, when no veld length nor all-around symbol is shown. No action is required for HVAC supporte, since a note was previously added to the HVAC standard support detail drawings. In addition to the Corrective Action described above, a training session was conducted with the Civil Group to discuss this Observation Report and action to prevent recurrence. No further action is required. Ne 5...............................S.S......s..............e5.WS..ma Rosarks: None
^
p _ _ j j / i PEco pp" v&ow Signature 1/Ddte Bechtel Review Signature /Date i P a g e [., o f ,,?,
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT Observation Report D OR- 019 Rev. _0_ Review Plan: LK- D - 1903 -S Rev. _1, PART III - RESPONSE EVALUATION The response to this observation report has been evaluated as follows: A. Causal Factor (s): X Acceptable Not Acceptable (Explain): The cause is a misinterpretation by design engineers of the code requirements for the welding of tube steel. B. Extent of Condition: 1 Acceptable Not Acceptable (Explain): Bechtel has reviewed all commodity support calculations which could be affected, and the majority of miscellaneous structural steel calculations using tube steel. C. Significance: y Acceptable Not Acceptable (Explain): ' Ph Bechtel has found that although a number of calculations required revision,
) -- no increases . in veld size were required. These revised calculations were not reviewed by SWEC.
D. Corrective Actions X Acceptable Not Acceptable (Explain): A total of 53 calculations were found to use the incorrect weld length. All but eighteen were found to be adequate by inspection for the reduced weld length. The other eighteen were revised to evaluate further the effects of the reduced weld length. None of these calculations were reviewed by SWEC. E. Action to Prevent Recurrence: 1 cceptable A Not Acceptable (2xplain): Bechtel has added notes to applicable drawings and has conducted training sessions for the engineers. Additional Action Required: 1 No Yes (Explain) C.......................................................................................
& C. %n 2bc//9 Lead Engineer Signature '
/ Date 0 & % 0.klilie 2/17/t1 APH Signature / Date Page 6 of 7
k7YACMm6NY / t 1
\j FEBRUARY 15, 1989 BECHTEL POWER CORPORATION DDR-019 CORRECTIVE ACTION LIST OF REVISED CALCULATIONS CALCULATION NUMBER REVISION NUMBER 094.008.060 2 099.013.021 1 099.013.025 4 099.014.010 6 099.014.080 3 099.014.108 4 099.017.040 3
- 099.019.128 1 099.022.020 4 099.022.026 1 099.022.085 2 099.022.071 1 099.022.120 1 099.022.'123 1 099.023.032 1 112.001.011 3 099.022.015 5 099.017.033 1 l
l O DOfL._pi9 - REV o PAGE 1 _08. 7 I
-_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ a
STONE & WEBSTER ENGINEERING CORPORATION O LIMERICK 2 IDCA
- h. (J.O. NO. 18138) onsnavATION REPORT Observation Report No. DDR- 020 Rev. O_
Review Plan LK- D - 1903B - MC Rev. 1 Reference AI No.: DAI-173. Rev. O PART I - INITIATION
- 1. Description of Concern The analysis for the ' wide flange member in calculation CBB-216-C26 Rev. 3 incorrectly models the boundary conditions for consideration of warping stresses resulting in unconservative results.
- 2. Supporting Information The FAPPS computer coding error for the cantilevered wide flange member of support GBB-218-H26 results in the consideration of warping stresses at the free end of the beam. In reality the warping stresses occur at the fixed . end, and should be considered in conjunction with the other stresses at that location. As a result of this error,
/~T the total stress at the fixed end is miscalculated on the Q unconservative side in the existing calenlation.
I i l
===============================u================.======.===============================
Additional Documents Attached: None k l
==== == =================
C)=....=====..m===.==============.==.=.==================== 0&W Lead Engineer Signature
/2/a/18
/ Date APM Sig1;p6ture
/ /^ /ffdV
' / Date l Page 1 of [ l
S'IONE & WEBSTER ENGINEERING CORPORATION p LIMERICK 2 IDCA (J.O. 10. 18138 ) {J OBSERVATION RETORT MAR 011989 Observation Report DOR-020 Rev. 0 ) CDMPLETE AMENDED RESPONSE
- 1. Observation Concurrence:
%.W REISS DATEJ-/ st X Concur with observation *MC I Do not concur with observation (Note:
h if not in concurrence, explain in " remarks" below) 3
- 2. Response to Observation:
A. Causal Factor (s) A review of pipe support calculation GBB-218-C26, Rev. 3 and an interview of the designer and checker of the referenced calculation reveals that an incorrect input selection was made in modelling the boundary conditions ; for the consideration of warpirg stresses. B. Extent of Condition Q To determine the extent of condition, a random sampling review of seismic d category- 1 pipe support calculations that used the FAPPS computer program was conducted. We random sample survey size was selected to demonstrate with a 95% confidence level that 95% of all FAPPS calculations do not have warping stress input coding errors that would result in a pipe support design that did not meet existing design criteria. Out of total population of 2833 Iirge Bore Seismic Category 1 pipe support calculations, 64 pipe support calculations using the FAPPS ccrnputer program were randomly celected. A Lirerick Unit 2 database printout of all the Seismic Category 1 Large Bore Pipe Supports was used to assign a sequence number of 1 to 2833 to each of the pipe support calculations. A random number generator was used to select the 64 calculations to be reviewed for the FAPPS input of the warping stress boundary conditions. h e random number list was used to select calculations to be included in the survey. If a selected calculation did not have an FAPPS computer analysis, the next random number was used to select the next calculation. 'Ihis process was used to determine the scope of the 64 calculations surveyed. The results of scientific rarxiom sample demonstrate that calculations are free of warping stress boundary condition input errors that would result in a pipe support design that did not meet existing design criteria. C. Significance The wide flange member in calculation GBB-218-C26 has been reanalyzed using the FAPPS computer program with correct coding for the boundary condition for warping stress consideration. Based on the revised FAPPS output, the v ...=...............=.. .............==...==..........==..===.=. _ ..=.............=.... Remarks: 1 Page J of 5 l l t i
SIONE E, WEBSTER ENGINEERING CORPORATION [^b LIMERICK 2 1DCA (J.O. to. 18138) (g/ 1 hfhh OBSERVATION REIORT Observation Report DOR-020 Rev. O PART II - RESPONSE CXNTINUED weld calculation and wide flange check has been performed and the resulting total stresses at the fixed joint remain within acceptable limits. Our random sample review results also demonstrate the lack of a generic concern. Of the 64 calculations reviewed, two calculations contained an input error for the warping stress boundary condition. Both of these calculations, GBB-211-CH5 and JBD-289-C6 have been reevaluated with the corrected input boundary conditions and the results demonstrate that the existing designs remain within acceptable limits. The maximum warping normal stress after correction of the warping stress boundary condition for the affected members in Calculations GBB-211-CHS, Rev. 4 and JBD-289-C6, Rev. 3 is 99 psi. The maximum total shear stress (pure shear, torsional and warping) in each of the four members is 195 psi for member 4 of Calculation GBB-211-CHS, 532 psi for member 10, 663 psi for member 11, 479 psi for member 13 of Calculation JBD-289-C6 respectively, with a design limit of 12,760 psi. We, therefore, consider the concern identified in this Observation Report to be not significant. /n\ U D. Corrective Action Pipe Support Calculations GBB-218-C26, GBB-211-CHS and JBD-289-C6 have been revised with the correct FAPPS input for the boundary condition. E. Action to Prevent Recurrence A group meeting of the Pipe Support Design Group was conducted to review this Observation Report and to reemphasize the need for attention to detail for FAPPS input data. Warping stress input coding was reviewed in detail. Based on our review results and the concern being insignificant, no further action is required.
==================================================================================
h, F p R4v'iew Signature 3[s[89
/Date
%T $
Bechtel ApTroval Sigpature h( 4 3 l .fr$
/Date C
Page 3 of S
)
l. l l ST6dE & WEBSTER ENGINEERING CORPORATION l j- LIMERICK 2 IDCA ; ( (J.O. NO. 18138) OBSERVATION REPORT Observation Report D OR- 020 Rev. O_ Review Plan: LK- D -1903B - HC Rev. J_ FART III - EESPONSE EVALUATION The respor.se to this observation report has been evaluated as follows: A. Causal Factor (s): 1 Acceptable Not Acceptable (Explain): SEE ATTACHED PAGE B. Extent of Conditions y Acceptable Not Acceptable (Explain): SEE ATTACHED PAGE C. Sa p> ficance: X Acceptable ,_,,yst Acceptable (Explain): SEE ATTACHED PAGE D. Corrective Actions X Acceptable Not Acceptable (Explain): , SEE ATTACHED PAGE E. Action to Prevent Recurrence X Acceptable Not Acceptable (Explain): SEE ATTACHP P 7E Additional Action Required: X No _ _Yes (Explain) m U.......................................................................................
- RhJAJ Lead Engineer Signature sHe9
/ Date GGLL s p.wite sWes AIS Signature / Date Page d of f
_ _ . - - = _ _ _ - . - i STONE & WEBSTER ENGINEERING CORPORATION l- LIMERICK 2 IDCA - (J.O. NO. 18138) OBSERVATION REPORT Observation Report DOR- 020 Rev. O j PART III - EESPONSE ETALUATION 00NTINUATION:- A. Casual Factor (s): Acceptable The response acknowledges that the preparer and checker made an incorrect input selection in modeling the boundary conditions for warping stress considerations. B. Extent of Condition: Acceptable Bechtel's random sample review demonstrated a 95/5 confidence level that support calculations do not have warping stress errors that result in designs exceeding design criteria requirement. In addition the results for the two calculations identified by Bechtel's ~ review that had input coding errors show that the impact on the support design margins is negligible i.e. the maxinram normal warping stress valt after correcting the input errors is only 99 psi. Therefore the Bechtel survey also demonstrates a 95/5 confidence level that the pipe support calculations arc free of input coding errors for warping stress that would result in a significant erosion of design margin. C. Significance: Acceptable Reanalysis of the three calculations shows that all three supports meet the design criteria requirements. The revised calculations have' not been reviewed by SWEC. D. Corrective Action: Acceptable Bechtel has revised the three calculations (GBB-218-C26, GBB-211-CH5, JBD-289-C6) to correct the input error for warping stress. E. Action To Prevent Recurrence: Acceptable Bechtel conducted a group meeting to review DOR-020 and warping stress input coding for the FAPPS computer program. O Pap g of 1
3 L STONE & WEBSTTR ENGINEERING CORPORATION
-rw- LIMERICK 2 IDCA-1 I (J.O. No. 18138) %) .
OBSERVATION REPORT . Observation Report No. D OR- 021 Rev. O_ Review Plan LK- D - 1903 - C Rev. L Reference AI No.: DAI-054. Rev. O DAI-345. Rev. O PART I - INITIATION
- 1. Description of Concern
' Concerns exist that an effective, consistent, proceduralized and well-documented program to establish BOP Q-functional instrument setpoints . and tolerances has not been established by Bechtel for this project. This could lead to the establishment of inappropriate setpoints for these BOP instruments. Also, some discrepancies and/or inconsistencies were identified during the review of two specific BOP setpoint calculations and associated methodology,
- 2. Supporting Information In response to Action Item DAI-054, Bechtel provided information which describes the program used to develop Limerick Unit 2 BOP Q-functional setpoints and setpoint tolerances. This information was provided in OV transmittal S-0292 to IDCA. After review of this document, certain clarifications were requested and were supplied by Bechtel as Notes of Telecon dated 11/8/88, and were fo rwarded to IDCA . via transmittal S-0373. Review of these transmittals resulted in' concerns that the documentation of the basis for Q-functional BOP setpoints and tolerances may be inadequate and/or inconsistent.
Setpoint calculations and related documents were then reviewed for two specific BOP Q-functional instruments. This review identified additional discrepancies and/or inconsistencies which further support overall concerns about the adequacy of the Bechtel Q-functional instrument setpoint program. These discrepancies are described as follows: A. BOP Instrument PSH-12-004D (Calculation No. 1266-1):
- 1. A radiation error of *1% is included in the final error computation of PSH-12-004D for the seismic case, but no radiation effect is pre.sent in the Spray Pond Pump House, tw mwn fg(,gggyA ,gAgi,,,,,, ,,,,,, ,,,,,,,,,,,, ,,,,, ,,,,, ,,, ,, ,, ,,,,,, , ,,, ,,,,,,,,, , ,,
Additional Documents Attached: None O.......................................................................................uln QCLic n/ules QGMck o.ws LeVd Engineer Signature / Date M M Signature / Date i Page 1 of 10
l
- STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J. O. NO. 18138)
OBSERVATION REPORT Observation Report No. , DOR. 021 Rev.O CONTINUATION:
- 2. An error value was computed for PSH-12-004D for the LOCA case s equipment in the spray pond pump house , however, experiences no LOCA environment.
- 3. Mechanical Calculation M-12-29 and setpoint Data Sheet 16A both shov the setpoint for instrument PSH-12-004D as 138.25 psig. Calculation 1266-1 shows the setpoint as 138.5 psig.
- 4. Setpoint Data Sheet 16A and Calculation I266-1 both show the setpoint tolerance for instrument PSH-12-004D as
- 21.9 psi.
Instrument setpoint Index M-600 shows the tolerance as *15 psi.
- 5. No documentation was found which described how the setpoint tolerance of 121.9 psi for instrument PSH-12-004D, as shown on setpoint Data Sheet 16A, was developed.
O 6. The setpoint tolerance in this calculation is calculated based I on a drif t value of i2% per year. Therefore, this tolerance is valid for one year only. PECo's Preventive Maintenance (PM) program indicates a recalibration frequency of 7 years for PSH-12-004D. It is very unlikely that the pressure switch will hold its calibration within the calculated setpoint tolerance for 7 years. B. BOP Instru*aent TSH-HH-76-223B (Calculation No. 660): This calculation, which Bechtel has stated is typical for BOP HVAC safety-related instrument loops, is little more than a data sheet which contains far less detail than Calculation No. 1266-1. I The following concerns were noted:
- 1. Sources of input data are not identified.
- 2. It is not indicated whether any assumption or unconfirmed input has been used in the calculation.
- 3. Instrument uncertainties are inconsistently applied to the different loop components. For example, line voltage stability and repeatability are applied to the temperature transmitter but not applied to the temperature switch even though the manufacturer specifies them for both of the instruments.
L 4. There is no evidence that environmental and seismic effects were considered in the calculation of setpoint tolerance. DDR -oal REV o CONTINUED ON '1W0 PAGES PAGE a OF do
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA l (J. O. NO. 18138) A OBSERVATION REPORT Observation Report No. DOR- 021 Rev. 0 CONTINUATION: l
- 5. Although the manufacturer did not specify any drift value for these instruments, the subject was not adequately addressed in the calculation.
- 6. The calculation shows a nominal trip setpoint for instrument TSH-HH-76-223B as 114 F vith a tolerance of 12 F. This suggests an acceptable upper setpoint limit of 114 + 2 = 116*F, which is in conflict with 115' committed in FSAR section 9.4.2.2.1.
This discrepancy was previously identified in Action Item Number DAI-345. Bechtel's response stated that the instrument provides only an alarms since the alarm could actuate at a temperature above the FSAR limit of 115', Bechtel should explain the purpose of this alarm and justify that actuation above the FSAR limit is meaningful or should resolve the setpoint/F3AR discrepancy. The following overall observations are therefore provided:
- 1. Bechtel has stated that no formally documented projdet procedure g exists to establish the methodology or documentation requirements to be used in the development of BOP Q-functional setpoints and setpoint tolerances. Also, the method used to verify the developed values against the instruments supplied, and to perform the Unit 2/ Unit 1 comparison, was not documented. Lack of such programmatic definition can result in:
a) inadequate and/or inconsistently calculated values for setpoints/setpoint tolerances and difficulties in traceability of inputs. b) inconsistencies / discrepancies such as those identified in sections A and B for the two specific instruments reviewed. I c) differences in approach depending on which group is responsible for particular instruments or for phases of design development. j provided In the samples reviewed, calculation 1266-1 a reasonably detailed and methodical approach, notwithstanding the discrepancies identified. HVAC loop tolerance calculations I such as 66C, however, are much less detailed and barely more than a simple data sheet. For non-loop HVAC instruments, J Bechtel stated that "HVAC utilized vendor data to develop instrument tolerances which were compatible with the process tolerances specified in the ISDS." There is no evidence that this determination of non-loop HVAC instrument tolerances was documented in a traceable manner. O CONTINUED ON ONE PAGE PAGE 3 OF 20
. STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J. O. NO. 18138)
OBSERVATION REPORT
'$v)_ Observation Report No. DOR- 021 Rev.O CONTINUATION:
d) inadequate direction to individuals not normally involved with instruments, such as mechanical HVAC engineers, for those inputs to the program for which they were responsible.
- 2. Bechtel stated that, for non-HVAC BOP setpoints, Calculation 1266-1 should (but did not) document the basis for all BOP instrun'ents classified Q-functional. Bechtel ' indicated, however, that two additional instruments (PT/PISL-59-252A and B) are missing from the calculation, and that the calculation would also have to be revised to reflect a change in the type of pressure switch used in four places on Unit 2 versus Unit 1.
In view of these discrepancies, it is difficult to establish that all BOP Q-functional instruments are enveloped. The A/E is generally required to develop plant-specific setpoints for instruments provided in such systems as leak detection and process radiation leakage monitoring. _ None of these is evident in calculation 1266-1. In summary, it does not appear that a consistent, well-documented and traceable calculation program exists to support BOP Q-functional O~ setpoints. Considering the variances in content noted in the calculations reviewed, and the various discrepancies found, it is impossible to establish confidence that all such setpoints for which Bechtel has responsibility are appropriately established and acceptable, i 1 il 1 1 I 1 1 t~ i DOR -oal REV o PAGE 4 OF 20 ( w__ - -
l STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA - (J.O. NO. 18138) !
,~N OBSERVATION REPORT 1 v l Observation Report No. DOR-021 Rev._Q_
PART II -RESPONSE
- 1. Observation Concurrence:
AMENDED RESPONSE The following is an amended response to clarify the interface between GE and Bechtel in determining the setpoints for the Leak Detection System as requested during a meeting between SWEC and Bechtel held in San Francisco on March 21, 1989. x Concur with observation Do not concur with observation (Note: if in concurrence, explain in " remarks" below)
- 2. Response to Observation:
(]* Causal Factor (s) G A formally documented project procedure for the development of Balance-of-Plant (BOP) setpoints did not exist. A unique formal documented procedure for setpoint determination was not developed because the engineering and design practices and procedures in effect for the Limerick Project were considered adequate to meet the design objectives. BOP setpoints were determined based on methods using sound engineering practices and judgement, accounting for items as, instrument performance, process and environmental conditions, and , unique concerns, in a generally conservative manner. The bases for i I determining BOP setpoints are contained in the design records, vendor documents, equipment qualification documents, and files for the individual instruments and systems. The methods used were presented in the white paper transmitted to SWEC via transmittals ; S-292 and S-0373. The results of these methods, i.e., instrument j setpoints, tolerances, and in some cases calculation numbers, were j documented on Instrument Setpoint Data Sheets (ISDS) and Instrument ! Loop Tolerance Sheets (ILTS) in accordance with the engineering ] practices and procedures used on the Limerick Project.
======================================u=================================
( ! V Page 5 of 30 CS8932217/1
STONE &-WEBSTER ENGINEERING CORPORATION
. LIMERICK 2 IDCA
, (J.O. NO. 18138) [Md4 / f/2 E OBSERVATION REPORT Observation Report No. DOR-221 Rev._Q,, HPART II -RESPONSE
.B. Extent of-Condition Instrument setpoints and tolerances for the BOP instrumentation were determined using the methods discussed _in the white paper transmitted to SWEC via transmittals S-0292 and S-0373. The lack of a unique, formal, documented procedure for setpoint determination is common to all BOP instruments.
.C. Significance Section 2, Overall' Observations, Paragraph 1-The lack of unique, formal, documented procedure for determining instrument setpoints can result in:
- The lack of a single, formal, and traceable document demonstrating the-inputs, assumptions, considerations and O calculations used in determining an instrument setpoint. This situation ~in and of itself does not mean that setpoints were determined. inappropriately. However, future questions concerning the-bases for setpoints or any changes to the setpoints would require reconstituting the original' inputs, assumptions, considerations and calculations.
- Some variation in the inputs, assumptions, considerations, and how calculations are handled in determining instrument setpoints. As discussed in the white paper transmitted-to SWEC via S-0292 and S-0373, consistent and acceptable methods were used in determining instrument setpoints. Some variation inLthe way setpoints are determined must be expected even when a formal procedure is used. There is no cookbook method for determining instrument setpoints.
Calculation I-266-I was performed to show that during design basis conditions bistable pressure switch (Barton) tolerances were adequately enveloped by the process tolerances. This calculation demonstrated that the overall switch tolerances were adequately enveloped by the process tolerances. This~ calculation does not indicate a difference in approach between HVAC and Control Systems in the design development of process setpoints. The Barton 0 ----------------------- Page b of 20 CS8932217/2
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA MM j p r
,_ (J . O . NO. 18138)
I
\ OBSERVATION REPORT Observation Repurt No. DOR-211 Rev. O PART II -RESPONSE CONTINUED pressure switche s were the only non-loop instruments used for safety-related functions. All other instruments, both Control Systems and HVAC, were analog instruments used in loops.
Since HVAC utilized analog instruments in loops, it was necessary to document the accumulated accuracies for each instrument and then calculate the total loop tolerance to show that margin existed j between the process tolerance and the loop tolerance. It was not 4 considered necessary to provide additional details in the calculations to identify existing project records since a similar concern as with the Barton pressure switch did not exist. Therefore, the methodology used by HVAC was not, an indicated in the DOR, a difference in approach from Control Systems but rather the necessity to address analog loop instruments differently than non-loop instruments. The methodologies used by Control Systems and HVAC engineers were sufficient to establish instrument
,-sg setpoints which were appropriate for their intended application.
O The instrument engineers who compiled the data for the ILTSs and ISDSs were intimately familiar with the types of instruments being used and relied on their experience with HVAC instrumentation for those inputs to the setpoint development process for which they were responsible. The checker and reviewer had the same working knowledge of the system and the performance of the instrumentation. The results of this effort were documented in various calculation files and a copy of the ISDSs and ILTSs were transmitted to PECo. Additional calculations wete performed in response to this DOR. Calculation for two typical HVAC analog instrument loops were performed. These analog loops contain 7 of the 20 instrument types used in the HVAC instrument loops. A third calculation for a typical non-HVAC analog instrument loop (Rosemount transmitter / trip unit) was performed. In all three cases the process tolerances adequately enveloped the instrument loop tolerances. These calculation were provided to SWEC in transmittal S-0696. l The time intervals for " drift" (or " accuracy") from the l manufacturers of the equipment represented in the sample HVAC loops l is for a six month period. The sensing element (i.e., TEs and FEs) i accuracies are related to manufacturing tolerances and are not time-^^ pendent. The sample HVAC loop calculations have identified pO
...=........................s...........................................
Page 7 of 30 CS8932217/3 L__ _- ____.
1 STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) ZO A' / /c2
/~N
( ,) OBSERVATION REPORT Observation Report No. DOR-211 Rev. O PART II -RESPONSE CONTINUED the drift interval where appropriate and the resulting loop tolerances are within the published ISDS process tolerances for a six month interval. Evaluating the drift intervals at 18 months to be consistent with the actual calibration frequency could result in the loop tolerance exceeding the ISDS process tolerance. A review of the extent and significance of this condition will be performed and is expected to demonstrate that there is no adverse impact to plant safety, although some setpoints may be revised. Calculation I-266-1 and the n>n-HVAC instrument loop calculation cover all the safety-related BOP trip instruments for which Bechtel supplied and was responsib)2 for determining setpoints. The two calculations for the HVAC instrument loops cover a large percentage of the instrument types supplied for HVAC loops. Consequently, we are confident that the safety-related BOP setpoints are adequate. k_- ! Section 2, Overall Observations, Paragraph 2 Calculation I-266-1 verif?.ed that at least a positive 1% margin existed between the procers tolerance and the instrument tolerance for all non-loop BOP instruments classified Q-functional (i.e., only the Barton pressure switches were required to be reviewed). ! The remaining two additional BOP analog loop instruments were not missing from I-266-1, which only addressed non-loop instruments. ! These two instruments wera purchased by Control Systems and are 3 utilized in the Primary Containment Instrument Gas System. A calculation was issued, as a result of this DOR concern, to confirm the results that the original methodology provided a margin between the process tolerance and loop tolerance. This calculation verified that such a margin is provided. j i The type of pressure switch (Barksdale) used in four places on Unit i 2 versus Unit 1 were purchased from GE as a qualified replacement switch for one of the Barton pressure switch applications. The process setpoints were part of GE's Reactor Protection System scope and were calculated by GE. The switches are covered under the NSSS scope of the response to DAI-054. fO,=======================r========================================== Page W of 20 CS8932217/4 i - - - _ - _ _ _ _ _ - _ _ _ _ - _ _ _ _ _ _ _ - 1
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA
'(J.O. NO. 18138) 2 3 1989 OBSERVATION REPORT <$ ~ o 74 7)
Observation Report No. DOR-221 Rev. O PART II~-RESPONSE CONTINUED The Leak Detection System (LDS) and the Radiation Monitoring System (RMS) were supplied by GE and General Atomics, respectively, as well as was other safety-related equipment supplied by other equipment manufacturers. The RMS equipment was purchased by Bechtel and the necessary instrument setpoints and tolerances were identified in the purchase specification. It was the responsibility of the vendors, who were the most knowledgeable in the operation and design of their equipment, to verify that their instrumentation provided an adequate margin between the process tolerance and instrument tolerance. This information was submitted by each vendor and was reviewed by Bechtel for compliance with the specification. For the LDS, Bechtel calculated the various process setpoints and process tolerances based on GE design specifications. GE was responsible for establishing instrument tolerances in their design specifications and purchasing instrumentation that met these tolerances. () Therefore, the scope of the BOP Q-functional instrument review was limited to only the Barton pressure switches in calculation I-266-1 and various analog-loops. { l Based on this review, with the exception of the HVAC instrument drift intervals coordination with calibration intervals, the identified discrepancies are considered to be insignificant. I Additionally, Bechtel considers that the program was consistent and ! that the instrument setpoints and tolerances were sufficiently , documented and do support the Q-functional setpoints for which 1 Bechtel had responsibility. Therefore, the original setpoints are 4 considered-to be appropriately established and acceptable for these BOP instruments. The significance of the discrepancies identified by SWEC in Item 2 - Supporting Information of the DOR are hereby discussed: Section 2, Items A.1 and A.2. Calculation I-266-1 was performed to show that during design basis conditions, the Barton pressure switch tolerances were adequately enveloped by the process tolerances. c)..=_=______________________________________==___==== Page 9 of 30 CS8932217/5
p IDCA/9/e? STONE-& WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT j. Observation Report No. DOR .Q21 Rev. O PART II -RESPONSE CONTINUED I l Bartonprov[dedtheradiationtolerancesfortherangesof' O to 3 X 10 Rad. Totgl Integratpd Dose (TID) (normal operating conditions) and 3 X 10 to 4 x 10 Rad. TID (accident conditions). The tolerances were 11% and 12%, respectively. These tolerances were used in the calculation and showed that during these conditions, the overall switch tolerances were adequately enveloped by the process tolerances. The inclusion of radiation tolerances for areas.where no radiation is anticipated'provided additional margin to prevent the exceeding of the design limits. Section 2, Item A.'3. The process setpoint value was not changed in I-266-1 during incorporation of the effects of the revised process setpoint value shown in Mechanical' calculation M-12-29, Revision 1. This is O attributable to a lack of attentiveness to detail by both the originator and checker. The small difference in values'(0.25 psig) is considered to be insignificant because the revised process setpoint value still meets'Barton's recommendation that the setpoint fall within the 10% to 90% range of the switch. Section 2, Item A.4. ISDS 16A showed the process setpoint for'PSH-12-004A through D as 138.25 i21.9 psig (Rev. 4) and the current revision of the Setpoint Index (M-600) shows the process setpoint to be 138.5 psig with a ; process tolerance of 115 psig. PECo has verified that the actual - instrument setpoint was corrected to 138.25 121.9 psig per Startup ; Work Authorization (SWA) 16A-37, dated 26-Sep-84. A subsequent l check, as a result of this DOR, has identified that M-600 and the i ISDSs still disagree with some setpoints. A review of these discrepancies shows that the actual instrument setpoints are . correct and that the discrepancies are limited to the lag in updating M-600 and/or backfitting the ISDSs with changes that have been made by.PECo on Unit 1. PEco engineering revises M-600 per the revised ISDSs, which had not occurred for thes- instruments. c;,,=...................................................................... Page IO of 30 CS8932217/6
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) [gf / . is ,) OBSERVATION REPORT Observation Report No. DOR-221 Rev. O PART II -RESPONSE CONTINUED PEco's current procedure (T&L 11-000406) emphasizes that M-600 be kept current. The differences that exist between the ISDSs and M-600 are considered to be insignificant. Section 2, Item A.S. The process tolerance of i 21.9 psig for PSH-12-004A through D is the difference between the maximum spray pond level at the minimum pump flow and the minimum spray pond level at the maximum pump flow. Minimum design flow = 900 gpm per vendor print 8031-M-12-61-3 (pump head = 390 ft.) Maximum design flow = 8,500 gpm per vendor print 8031-M-12-85 (pump head = 270 ft.) ps
's s Difference in pond level = 18'-9" Process tolerance = 390 - 270 -18.75 = 101.25 ft. or i 21.9 psig Therefore, the adequacy of the process tolerance in ISDS 16A is clearly demonstrable.
Section 2, Item A.6. PEco's Preventative Maintenance (PM) program specified a seven (7) year calibration frequency for PSH-12-004A through D. As noted in Section D, Corrective Action, this ha's been corrected to 18 months as a result of this DOR. Had the situation gone uncorrected, monthly surveillance testing of the RHR Service Water (RHRSW) Pumps would have discovered any potential drift problems with these switches that could have prevented safe operation of the RHRSW System. Drift in the positive direction (138.25 psig +) would not 1 have been identified during the monthly testing, however this only affects the low end of the low flow trip. This trip is provided as a precaution against unanticipated low flows. However, there is no design mode of the RHRSW system which requires flow rate below the switch setpoint. Drift in the negative direction (138.25 -) would c;),._____________________________________________..__________ Page !! of do CS8932217/7
STONE & WEBSTER' ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) MM gg OBSERVATION REPORT Observation Report No. DOR-921 Rev. O PART II -RESPONSE CONTINUED affect the high and of the low flow trip and would have been identified during the monthly testing as it could have prevented normal operation of the associated RHRSW pump. A Bypass switch is-
.provided in the control room to enable the operator to bypass the high pressure (i.e., low flow) trip of the affected pump (s) and restart the pump (s). Sufficient status information is provided in the control room to enable the operator to ascertain that a false low flow' trip has occurred and to effect a manual bypass and-restart.
All of the switches covered by calculation I-266-1 have a calibration frequency of 18 months. The current revision of this calculation is based on a 12 month calibration frequency. 'The effects of an 18 month calibration frequency were reviewed and it was concluded that a positive margin is still provided~between the process tolerance of the ISDS and the switch tolerance for each switch for the worst-case event (seismic). This calculation will () be formally revised to reflect the 18 month calibration frequency. Based on the above discussions, there is no significance regarding the 12 month drift in calculation I-266-1 and the 18 month calibration frequency for each switch. Section 2, Item B Paragraph, B.1 The ILTS was included in the ISDS to provide a recoverable. basis for tolerance data in case of future modifications of'the loop. This degree of documentation is considered to be sufficient to establish the: instrument tolerances. The information contained on the1ILTSs (e.g. Inst. Function, Tag No., Manufacturer, Model No., Range and Span) are sufficient to identify the. instrument and its use and location in the plant. Accuracy data was taken from project catalogs, specification data sheets, or vendor print documentation existing at the time. Section 2, Item B.2 No assumptions or unconfirmed inputs were used in calculation 66C. c;),.....................-................................................ Page 11 of 20 CS8932217/8
Uf STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) [$d [f/d OBSERVATION REPORT Observation Report No. DOR-D21 Rev._Q_ 9 PART II -RESPONSE CONTINUED l Section 2, Item B.3. t A~ review of the Reactor Enclosure'KVAC system ILTSs verifies that. I consistency exists in applying values. The HVAC instrument engineer considered the manufacturer's specification as well as the material specification in calculating loop tolerances and applied j these values consistently throughout the ILTSs. The instruction j manuals for tho' subject transmitter and the switch-are dated # May 1977 and February 1977, respectively, in the project files. A repeatability tolerance and line voltage stability tolerance is specified for the transmitter. For the switch, the specificat:on lists no independent repeatability value and an acceptable power supply variation range for which no accuracy deviation is
. specified. Therefore, it would be reasonable to assume that the switch would operate properly, with no deviation, as long as the poser supply was maintained within these limits. Since the_ data lq available from the-manufacturer's instruction manuals did not i Os specify these inaccuracies for all the instruments as independent variables at the time of the calculation, the engineer did not apply this value.
The current (1987) bulletin'of the manufacturer shows a line voltage stability tolerance of i 0.2%.of span and a repeatability tolerance of 0.1% of span. Using the additional tolerance in the original ISDS calculation 66C results in an increase of i 0.11*F, well within the margin available. Section 2, Item B.4. The same data sources and information were used to determine the acceptability of environmental and seismic qualification. If there were any deviations from specification data sheet accuracies, the variances would have been addressed as required and documented in the Qualification Report and/or the Qualification Report Summary Sheet. Since any variances from the original data were not identified'during qualification, the specified accuracies were considered acctpth51e without any further investigation or documentation. The.e is no significance for this item.
,=======================================================================
O Page 13 of Ao CS8932217/9
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA
-t (J.O. NO. 18138) [ [ M / f/.2' y OBSERVATION REPORT Observation Report No. DOR-221 Rev._D_
PART II -RESPONSE CONTINUED Section 2,zItem B.S. Drift, when not specified separately by the manufacturer, is assumed to be included in the general data given'for " accuracy". Section 2, Item B.6. The high-high temperature setpoint of 114*F i 2*F for instrument' TSH-HH-76-223B is used only'to provide annunciation of abnormal pump compartment temperature conditions. After a< design basis accident, this alarm provides no design basis control' functions or-information which must be used to maintain the compartment within., the design temperature limits. The lead and standby unit coolers are 100%' capacity each and start automatically at-100*F and 110'F, respectively, to maintain the 115'F maximum compartment' design temperature. Because there is no control function associated with this alarm for providing additional cooling, the high temperature O alarm is useful only to alert the operator that an abnormal condition exists in the compartment. During normal operation, such a condition could be caused by the loss of the normal ventilation with no unit coolers operating. During an RHR. Pump test, this
~
condition could be caused by failure of the cooling water or multiple failures of the unit coolers. The resultant compartment temperature caused by these abnormal conditions is not expected to elude detection by stabilizing at a temperature above 115'F. The
~
alarm will still actuate, albeit at a slightly higher temperature. During a design basis accident, it is desirable to keep the temperature setpoint as high as possible to preclude' spurious
-alarms since operator action is not available to determine the cause of the alarm.
D. Corrective Action
- 1. As of January 1st, 1989, for all future changes to existing setpoints for safety-related instruments, existing setpoints that are listed in the Technical Specifications and FSAR, and selected existing non-Q instruments considered important to the safe and reliable operation of the plant, as well as determining setpoints for newly added instruments in the same categories, calculations will be performed to support the change and the 1:
()_.......___________________________________..................__ Page IH of 20
.CS8932217/10
_ - - _ _ - _ _ - _ - - = _ _ _ _ __ - 1
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) [gA / fQ OBSERVATION REPORT Observation Report No. DOR-D21 Rev. O PART II'-RESPONSE CONTINUED revised setpoints. Similarly, calculations will be performed to support the setpoints determined for any newly added instruments or newly added instruments.whose setpoints are listed in the Technical Specifications or FSAR. These calculations will be done in accordance with existing Bechtel and PEco procedures for preparing and documenting design calculations. The objective of this corrective action is start documenting in a formal and traceable manner the inputs, assumptions, considerations and calculations used in determining instrument setpoints.
- 2. Calculations for two typical HVAC analog instrument loops were performed. These analog loops contained 7 of the 20 instrument types used in the HVAC instrument loops. A calculation for a typical non-HVAC analog loop was performed. The objective of this corrective action is to demonstrate that the methods used to determine instrument setpoints discussed in the white paper transmitted to SWEC via S-02'73 and S-0373 were adequate and O conservative, i.e., the process tolerance adequately enveloped the loop' instrument loop tolerance. The calculations were performed and documented in accordance with Bechtel procedures for design calculations.
l 3. Drawing M-600 has been revised to show an instrument setpoint and tolerance of 138.25 +/- 21.9 psig PSH-12-004A-D, consistent with the ISDS and' actual instrument calibration. L 4. The calibration frequency for PSH-12-004A-D has been revised to I- an 18 month cycle, consistent with the bases for the process tolerance.
- 5. The non-HVAC ISDS's were checked and revised to correct minor discrepancies, where setpoint values were revised during operation of Unit 1 but were not yet reflected in the IDSDs.
The updating of M-600 is a continuing process and lags the actual engineering document / calculation that is used to justify the change.
- 6. Calculation I-266-1 will be revised to show a process setpoint value for PSH-12-004A-D consistent with mechanical calculation M-12-29 and the 18 month calibration frequency by March 31, 1989.
c;),______________________________________________________ Page IJF of 30 CS8932217/11
V ' STONE-& WEBSTER ENGINEERING CORPORATION LINEltICK 2 IDCA-OdA/fM (J.O. NO. 18138) e , OBSERVATION REPORT Observation Report No. DOR-221 Rev._Q_ PART II -RESPONSE CONTINUED
- 7. The Q HVAC instrument drift intervals versus the instrument calibration frequency will be evaluated to confirm acceptability. The instrument calibration will be readjusted if required and/or establish new calibration frequencies to meet
.the current instrument' drift intervals. This work will be completed by Unit 2 Fuel Load.
E. . Action to Prevent Recurrence
- 1. PEco will develope a unique setpoint determination procedure to used for determining instrument setpoints for at least the four classes of instruments and setpoints discusstd in Corrective-Action No. 1. This procedure will be based on current industry standards, NEDC-31336 -- General Electric company Instrument Setpoint Methodology,-and current NRC guidance. This procedure will be implemented by June 30, 1989 and will supercede
() Corrective Action No. 1 at that time.
- 2. In early.1988, PEco identified as a long tern' configuration management issue, the need to reconstitute the design basis calculations for instrument setpoints for its nuclear stations.
PEco is: developing a program to reconstitute the design basis calculations, using a generic setpoint determination procedure applicable to both nuclear stations, for instrument setpoints prioritized on each particular instruments importance to plant safety. This is a long term program, extending over the next 3-5 years. PECo is targeting to complete program definition by the and of December 1989. Objectives of the program are:
- reconstitute the design bases calculations for instrument setpoints in a traceable and retrievable manner.
- provide a consistent means for determining and documenting instrument setpoints for the entire PEco Nuclear Group.
- better integrate engineering and station procedures associated with setpoint determination, change, and control.
c;)....................................................................... Page /b of do CS8932217/12
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA [OM /fQ (J.O. NO. 18138) OBSERVATION REPORT Observation Report No. DOR-pil Rev. O PART II -RESPONSE CONTINUED improve the consistency, throughout the nuclear group, of setpoint documentation issued to the stations. O l
== ---
=========_
j wd 3 '3 8Y &
- Beclifel Approval Signature /Date
, ) )/b]f 18f/h)
'Eco Review Signature /Date {
Page i7 of M CS8932217/13
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) im OBSERVATION REPORT , Observation Report DOR- 021 Rev. _0 Review Plan: LK A - 1901 - c Rev. j PART III - RESPOESE EVALUATION The response to this observation report has been evaluated as follows: A. Causal Factor (s): X Acceptable Not Acceptable (Explain): See attached. B. Extent of Condition: X Acceptable Not Acceptable (Explain): See attached. C. Significance: X Acceptable Not Acceptable (Explain): D) See attached. D. Corrective Actions X Acceptable Not Acceptable (Explain): See attached. E. Action to Prevent Recurrences y Acceptable Not Acceptable (Explain): See attached. Additional Action Required: X No Yes (Explain) OO' Lhad Engineer Signature 4l3fC9
/ Date 0 M ' S.l).WillL 4l3l81 AI4 Signature / Date .
l t Page,lfof)_0 i l. l .l
t STONE fu WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT v PART III - RESPONSE EVALUATION CONTINUATION: l A. The causal factor is acknowledged to be the absence of a formally documented project procedure for the development of BOP setpoints. B. The condition is applicable to all BOP instruments. C. The lack of a formal, documented project procedure might have resulted in inappropriate establishment of BOP Q-functional setpoints and setpoint tolerances. However, based on the sample evaluated during this review, no evidence was found that the BOP, Q-functional setpoints were established inappropriately. Additional calculations were performed by Bechtel in response to this DOR, which included both HVAC and non-HVAC analog instrument loops. These calculations were reviewed, and in all cases, the process tolerances adequately enveloped the instrument loop tolerances. For the Q-HVAC instruments, a disconnect was identified between the vendor-supplied drif t interval and the Limerick calibration interval. As a result, the project agreed to evaluate the effect of extending drift to 18 months and, for any instance where loop tolerance is O found to exceed the ISDS process tolerance, conduct a review of extent b of condition and significance prior to fuel load. The discussion regarding other points raised in the DOR was reviewed and the discussion provided with regard to significance of each item was considered reasonable. D. As of January 1, 1989, for all future changes to existing setpoints for safety-related instruments, existing setpoints that are listed in .the technical specification and FSAR,,and selected existing non-Q instruments considered important to the safe and reliable operation of the plant, as well as determination of setpoints for newly added instruments in the same categories, calculations will be performed to document the changes and the setpoints. All other discrepancies identified in this DOR will be corrected by revising the appropriate documents as explained in the response. E. The project will develop a unique setpoint determination procedure by June 30, 1989, to accomplish the action in item D, above. Also, the project has committed to development of a long-term program to reconstitute design basis calculations for instrument setpoints in a traceable and retrievable manner over a 3-5 year period. Priority will be placed on completing calculations for instruments important to plant safety. i (g' Based on the information reviewed including that presented in this response, it is clear that a formal procedure was not utilized for establishment of BOP setpoints. Also, design basis calculations for those setpoints did not exist in a consistent manner which CONTINUED ON ONE PAGE. ) Page L4 of M I
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) O 1 OBSERVATION REPORT U Observation Report No. DOR- 021 Rev. 0 ) PART III - RESPONSE EVALUATION CONTINUATION: 1 establishes a clear basis for those setpoints. The information , available, including the details in this response, do appear to indicate - that current setpoints were appropriately and conservatively established based on engineering judgement and known instrument performance in a generally conservative manner, in accordance with the described engineering practices in effeet for Limerick at the time. For the sample reviewed, it appears that the current setpoints are acceptable. However, . the action which has been committed to is essential to positively demonstrate this for all instruments and provide a basis for future changes. b uj l l 1 Page d of do
l l i ! STONE & WEBSTFR ENGINEERING CORPORATION l LIMERICK _2 IDCA f~~N (J.O. NO. 18138)
< i OBSERVATION REPORT Observation Report No. D OR- 022 Rev. O Review Plan: LK ,JL_- 1908 - MC Rev . ,_1_
Reference AI No.: DAI-188 DAI-190. DAI-192 PART I - INITTATION
- 1. Description of Concern Design Specification 8031-M-170 incorrectly classifies the ASKE III accumulator tanks' pressure boundary as " piping".
The accumulator tanks for the Mainsteam Isolation Valves (MSIV'S) and Mainsteam Safety Relief Valves (MSRV'S) are permitted to be designed to ASME III, Class 3, Service Levels C and D by the Design Specification 8031-M-170, and these service levels were implemented in the Design Report. j The ASME III Code requires that the use of these se'rvice levels be reviewed l by the owner. It has not been documented that the owner's review was obtained. This concern addresses only compliance with the ASME III Code and not loss of the pressure retention function of the tanks. The design report does not include all the loads that are required by
,c3 ASME for design purposes.
! )
~.J'
- 2. Supporting Information Classification of the Accumulator Tanks as " Piping" The design specification, paragraph 41f, invokes ASME III, NCA-1273, which allows certain miscellaneous items that fall into the categories of fluid conditioners and flow control devices to have their pressure boundaries considered as piping. The accumulator tanks are neither fluid conditioners or flow control devices. NCA-1273 provides examples of such devices and the accumulator tanks are not typical of the examples that are listed.
Also, ASME has deliberately refrained form providing a definition of a tank. Instead, if the item is referred to in the applicable documents as a " tank", ASME contends that it should then be considered a tank. This is the case for the accumulators, as they are referred to as " tanks" throughout the design specification and design report. CONTINUED ON 2 PAGES
.............................................................................==========
Additional Documents Attached: None fey ......................................................... . ......................== Lead Engineer Si'gnature 12}p)lgg.
/ Date APM Signature l U
(
/ Date Page 1 of 3
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. No. 18138) OBSERVATION REPORT V Observation Report No. __y_ OR- g Rev. _jl_, PART 1 - INITIATION CONTINUATION: l The MSIV accumulator tanks outside the containment are normally supplied with compressed air and do not have inspection ports as required by ASME III, ND-3363. No evidence exists that they are excluded from this requirement by having the compressed air supplied with a dew point of
-50*F or less and having the Certificate Holder's Data Report contain the statement "non-corrosive service".
The Design Report's use of piping equations from article ND-3600 of the code is misapplied as only tank equations, tank load combinations, and tank allowable stresses are appropriate since the accumulators are classified as " tanks". , Use of ASME III Code Service Levels C and D In response to DAI-192, Bechtel established that the design of the tanks would assure pressure retention by use of the ASME III, Subsection ND-3600 rules. Although the rules of ND-3600 are for piping design and are not the same as the vessel design rules of ND-3300, the conclusion that pressure retention will be maintained is acceptable, although the use of Service Levels C and D' permits deformation of the tanks. Because of this permitted deformation, which the code cautions may V necessitate the removal of the component for repair, the code requires that the owner review the use of these service levels for compatibility with established safety criteria (NCA-2141). This requirement is found in paragraphs NCA-2142.2 (b.) (3) and (4) of the code. In response to a request for documentation to show that the owner's review was conducted, Bechtel provided a copy of Engineering Department Project Instruction, EDPI-4.55.1, Rev. 4. This document requires that " Initial Issues of the material requisition shall receive client approval prior to issue for use." Bechtel orally stated that the client has approved Revision 0 of the design specification. Pcges 4 and 5 of the design specification contain the conditions that allow the use of Service Levels C and D. Page 4 in its entirety is Revision 1, whiae page 5 in its entirety is Revision 3. No evidence has been provided to show that Revision 0, which the client approved, contained Service Levels C and D, or that Revision 3 of the design specification, which does contain Service Levels C and D, was approved by the client. CONTINUED ON 1 PAGE O Page 2 of T>
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBERRVATION REPORT {} l V l Observation Report No. D OR- 022 Rev. 0 PART 1 - INITIATION CONTINUATION: _Ex,clusion of Loads in Design Report Article ND-3111 (f.) of the ASME III Code requires that reactions of supporting lugs, rings, saddles or other type of supports be taken into account in the design of a component. The design report for the accumulator tanks does not include these loadings. Bechtel has provided a supplemental calculation (No.18240-5504, dated 11/11/88) in response to DAI-190 for consideration of the ring (U-bolt clamps) and saddle loadings on the accumulator shell. Two errors exist in the supplemental calculation as follows:
- 1. The radial expansion of the tank under pressure was erroneously derived, whereas reference 2 in the calculation (page 448, case Ic.) provides the correct equation for radial expansion.
However, the derivation in the calculation produced conservative results.
- 2. The operating pressure of 100 psig was used to determine the radial expansion. The design pressure of 200 psig should have been used. The result is a small part of the total stress and O would not significantly affect the results.
Also, the supplemental calculation did not combine the resulting stresses with other stresses for comparison to allowables as depicted in Section ND, Tables ND-3321-1 and ND-3321-2. In its present state, the design report contains a mix of vessel and piping equations and allowables and is remiss by the exclusion of the saddle and ring loads. It therefore is not a complete ASME III design report. 1 O l l l Pege 3 of S
STONE & WEBSTER ENGINEERItKi CORPORATION QDCAl7/-,, m .pg 4
.1 ) REISSUED I, DATE FEB 2 71989
\ OBSERVATION REEORT Observation Report No. IDR-022, Rev. O PART II - RESPONSE CDMPLETE AMENDED RESPONSE
- 1. Observation Concurrence:
X Concur with observation - The design report does not include all the loads that are required by ASME for design purposes. X Do not concur with observation - The tanks are " incorrectly" classified as piping.
- 2. Response to Observation:
A. Causal Factor (s): The vendor design report did not address all loads on the tanks because of conservatism included in both the design and the report that compensated for the loads not considered. We surmise that the design report preparer exercised undocumented judgement that the support and pressure reactions of the U bolts and clamping beams were acceptable based on their design q (i.e.: contact shoes for the U bolt and sizirg of the clamp beam). B. Extent of Condition: The design report is used to qualify all of the accumulators purchased under the M-170 specification, and these accumulators are unique in that they are classified as flow control devices (see Remarks). Therefore, this concern is an isolated case. C. Significance: Based on the results of Calculation 18240-S504 Revision 1, which demon-strates ASME III code compliance for the concerns cited in this Observation Report, the concerns are not considered significant. D. Corrective Action: In response to the questions addressed in the Observation Report, Bechtel has revised calculation 18240-SS04 to include the concerns below:
- 1. The derivation of the radial expansion of the tank under pressure is correct and conservative. The calculated results are exactly the same as those from case 1B on page 448 in calculation reference 2.
However, these results are about 18% more conservative than those from case 1C on page 448 in calculation reference 2. A positive statement has been added to the calculation comparirg the two formulae p) ( for radial expansion under pressure, and documents the use of the conservative case. . Page 'lof % 1
) gjDd 4/ 7 /[ l SIONE f WEBSTER ENGINEERING CORPORATION l LIMERICK 2 IDCA I ex (J.O. NO. 18138) (U ) OBSERVATION REPORT Observation Report tb. DDR-022, Rev. O PART II - RESPONSE CONI'INUED
- 2. Se design pressure of 200 psig rather than the maximum operating pressure of 100 psig has been used to calculate the pressure induced stress. He calculated stress is also included in the tank vessel stress evaluations. Since the pressure is an insignificant load in i this case, there is no impact on the completed analysis. l
- 3. The resulting stresses due to these local effects are combined with the other stresses compared with the Code allowables. These stresses meet piping Code allowables, and the vessel allowables as depicted in Section ND, Tables ND-3321-1 and ND-3321-1 as well.
- 4. The Accumulator Design Report WPE 241056-10 (8031-M-170-13-4) will be revised to apperd Bechtel calculation 18240-SS-04 Revision 1 thereby providing a canplete document. The calculation cover sheet also notes the calculation's applicability to the vendor report.
Calculation 18240-SS04 Revision 1 is available for review in both the Ibttstown and San Francisco offices. E. Action to Prevent Recurrence: O)
\
v Based on the corrective action taken and the lack of significance, no further action is required.
===================================================================================
Remarks: We do rot concur that specification 8031-M-170 is incorrect in classifying the MSRV and MSIV accumulators as " piping". Although Specification 8031-M-170 refers generically to these items as tanks, it is clear in sections 4 and 8 that the specification specifically classifies these " tanks" as parts of a piping system. In particular, Section 4.1.f states that "These parts are classified as flow control devices, as defined in (ASME Section) NCA-1273." The legitimacy of this interpretation was verified before the specification was issued for purchase with the Bechtel Codes and Standards Group l . This decision was also concurred with by the Limerick Site Resident ASME Code Inspector 2,3 Because these accumulators are designed as piping parts, the requirement for inspection openings (ND 3363.2) is not applicable. The requirement that the owner review the specification as required by paragraphs NCA-2142.2(b)(3), and (4) was fulfilled before the specification was issued for I purchase. Ibcumentary evidence of the Owner's review of the initial revision as well as later revisions is shown in the references 4,5,6,7, and 8. n v Page s of S
l I FEB E7 'M 15: 1C BECHTEL WESTRN F0WER iST FLOOR P.7<8 anem1/l 570NE 5 WEBETER ENGINEERING CORPORATION t.IMrRICK 2 IDCA (J.O. m . 14L36) OBSERVATION REPORT l Qbeervation Report. No. !lDR-022,' Rev. 0 1 PART TT - RESPONSE 00tff!WED
- 1) IOM from W R. Brith, Sr., (Bechtel R&E/ Codes 4 Stendartis), to R. E. Elias,
" Code Starrpi.ng Requirements for MSIV/MSRV Accumulator Tanks, Limerick Project Job 403L", tbc. No. 111945, Detober 20,1981 (transmitted to SWEC via transmittal S0577).
- 2) IQM from T. M. Gdtn (Sechtel STPD - Tield Construction), to R. Elian/B. Posta,
" Limerick Generation StatLon Unita 1 and 2, Job 8031 MSIV and MERV Accumula-tora", toe. No. FL71116, November 10,1981 (tranarLtted, ta 34EC. via transmittal 80577).
- 3) ION from R. E. Enrich, (M S.B.I. & I. Co.), to A. Aren, (Boehtel 5tmer Corp.
Aasist. Project Engineer), " Classification of MSIV & MSR7 Accumulators", Doc. No, r170801, October 30,198L (tranar.itted to SfEC via tasaar.ittal S0577j.
- 4) Latter frort R. A. Mulford (PE00) to R. H. Elias (Sechtel), "I,imerick O' Generating Station, Units 1 & 2, MSIV/MSRV Accumulator Tanks, Specification 3031 +l-170, Rev. O, PI2-12747, Doc. No.114700, November 30, 1981, (transmitted to SWEC via transmittal S0577). ,
- 5) Istter from R. A. Mulford (PEco), to R. H. Elias (Bechtel), " Limerick Generating Station, Uni'ts 1 & 2 MSIV/MSRV Accumulator Tanks, Specification 8031=N.170", Pla-12740, Doc #114247, November 18, 1981 (Transmitted to '
SWEC via transmittal 30$77). ,
- 6) Latter from W. C. McDaniel (Bechtel) to R. A. Malford (PEco), " Job. No. 6031, Philadelphia Electric Company Limerick Generating Station, Units 1 and 2, Accumulator Tanks P.O. 8031-M-170", BLP-25305, Doc. #118217, Tebruary 3,1982 (transmitted to SWEC via transmittal 80$71).
- 7) lettdr from R. A. Mulford (PECo) to N. C. McDaniel (Bechtel), " Limerick Genereting Station, Units 1 and 2, MSIVA1SRV Aceumulater tanks, Spectf tea-tion 8031-M-170, Rev. 2", PIA-13013, Doc. #119894, February 16, 1982 (transmitted to SWCC via transmittal 19171).
- 8. Letter from R. A. Mulford (PECo) to W. C McDaniel, " Limerick Generating Station, Units 1 and 2, M$IV and MSRV Accumulator Tanks, Specification 8031-M-170, Rev. 3", PLB-13417, tcc. 4128651, June 8,1982 (transmitted to SWEC via transmittal S0577). l
_ Y PEco Review Signature /Date Big ~natDre /Date Bechtef4pf61 Pagelof1 L__ _ _ _ _ _ _ _ _ _ _ _
STONE & UEBSTER ENGINEERIh0 CORPORATION
~. LIMERICK 2 IDCA
[Ns,/') (J.O. NO. 18138) OBSERVATION REPORT j Observation Report D OR- 022 Rev. 0 l Review Plan: LK- D - 1908 - MC Rev. 1 ) PART III - RESPONSE EVALUATION The response to this observation report has been evaluated as follows: l
!. Causal Factor (s): Acceptable Not Acceptable (Explain): j l
SEE ATTACHED PAGE B. Extent of Condition: Acceptable Not Acceptable (Explain): SEE ATTACHED PAGE C. Significance: Acceptable Not Acceptable (Erplein)? SEE ATTACHED PAGE e I \_ D. Corrective Action: Acceptable Not Acceptable (Explain): SEEATTAClkEDPACE E. Action to Prevent Recurrence: Acceptable Not Acceptable (Explain): SEE ATTACHED PAGE
=============================================================s=========================
Additional Action Required X No Yes (Explain) SWEC has agreed to close-out the concerns; but with the record showing that our position on the classification of tanke as piping sub-assemblies is unchanged, that is, they should have been qualified as " tanks". However, the close-out of the concern is based on PECO having the authority to classify them otherwise as allowed by ASME. O ==============================================================================sas====== 21~d-fL4~ Lead Engineer Signature 4/0/ Date Of/Utt% mlIe shh7 APH Signature / Date Pa g e ,], o f _{,
DOR-022 Rev. O A. Casual Factor (s): Acceptable Bechtel has agreed that the Design Report did not analyze all loads on the accumulator shell. B. Extent of Condition: Acceptable With the tanks being considered as piping sub-assemblies, the concern is an isolated case and no further action is required. Also, based on SWEC's reviews of Bechtel designs of other ASME III components, such as RHR heat exchanger supports, the loading . that was omitted in the design report is considered irolated. C. Significance: Acceptable Bechtel supplemental calculation results (Calculation Number 18240-SSO4 Rev. 1) indicated that.the accumulators are still qualified. A i D. Corrective Action: Acceptable
- 1. Bechtel revised supplemental Calculation 18240-SS04 to include a clarification as to the conservatism of the radial expansion of the accumulator shell.
- 2. Bechtel appended the supplemental calculation to the original report in order to establish linkage'between them.
- 3. PECO agreed to the classification of the accumulators as piping sub-assemblies.
i l E. ActfontoPreventRecurrence: Acceptable Based on the corrective action taken, no generic preventatives are j nececsary because of the isolated nature of the concerns. 1 l I l 1 l l O tj DOR - o 3 A. REV o PAGE T OF 2
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) D OBSERVATION REPORT Observation Report No. D OR- 023 Rev. O_ Review Plan: LK- D - 1903H - MC Rev. _1_ Reference AI No.: DAI-141, Rev. O PART I - INITIATION
- 1. Description of Concern In the analysis for the design forcing function (Calculation SR-69(131)
Rev. 1) using computer .: ode RVCL(EE805), the approach of artificially changing the initial condition to avoid supersonic flow may not be conservative for piping in the suppression chamber.
- 2. Supporting Information Calculation SR-69 (131) Rev. I was performed because anomalous results, i.e. supersonic flow condition, were observed in Rev. O. The fix employed in Rev. I was to change the initial condition from a physically present condition to an unreal initial condition. This approach is questionable because the initial condition ' used in the design calculation is unreal and arbitrary.
Bechte?'s response to DAI-141, Rev. O is based on load comparisons. Loads Q() generated from the computer code RVCL (NE805) (Calculation SR-69(131) , Rev. 1) are compared with the loads generated from computer codes with the capability to analyze shock waves. Comparisons were made with the GE computer codes RVFOR04 and RVFOR05, which have been fully documented in the MARK I and II containment programs, and the industry accepted and well -tested computer code RELAP5. This
- approach is acceptable and would provide confidence on the adequacy of the design loads, provided that adequate comparisons were made. However, review of result comparisons showed that comparisons with RVFOR codes j were made only for SRV piping in the drywell. Loads were not generated for piping runs in the suppression chamber using GE computer codes RVFOR04 I
and RVFOR05 (Ref. I and 2.). Comparisons of piping loads in the suppression l chamber were found in Reference 3 using the RELAPS computer code. However, the comparison was not very favorable, and was attributed to the quencher modelling in RELAP5. CONTINUED ON ONE PACE
...........................--................................ -=====..==....===========
Additional Documents Attached: None l O W JAJ
~~-~~---
~~~~'---------~~~~~-~~~~--'----------~~~~~~~~--~~~~~-- ~~ ------~~~--~~~~-~~
amer-
/ 7/67 Lead Engineer Signature / Date APM Signar/are
/
/ Date
\ 10 Page1of)f_ Y wn g
STONE & WEBSTER ENGINEERING CORPORATION i LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT LJ Observation Report No. 0 l D OR- 121 Rev. CONTINUATION: Without attributing undue significance to the details of quencher modelling, there still exists one unsettling fact, i.e., RELAP5 results also showed { that piping loads in the suppression chamber are substantially higher j' for the case with realistic air density compared to that found with artificial initial air . density. Since both the RELAP5 and the RVCL codes showed that piping loads in the suppression chamber are higher for the realistic initial condition (i.e., initial air dencity) than those corresponding to the artificial initial condition, conservatism of piping loads in the suppression chamber has not yet been demonstrated and should be resolved. REFERENCES O[N Sechtel Calculation SR-69(147) Rev. O " Comparative Study of SRVDL 1.
'A' Forcing Functions Using RVFOR04"
- 2. Bechtel Calculation SR-69(147) Rev. 1 " Comparative Study of SRVDL
'A' Design Forcing Functions Using RVFOR05"
- 3. Bechtel IOM File Number SR-69(230), DC#142484, "LCS Job No. 8031, LCS MSRV Line A Sensitivity Study, Calculation SR-69 (145), (146),
(147) and (147) Addendum" from H. H. Safwat to W. C. McDaniel dated December 9, 1982. C \. U JO Page d of Y Y
l STONE & WEBSTER ENGINEERING CORPORATION LINERICK 2 IDCA O /374 (J.O. NO. 18138) v OBSERVATION REPORT ($'O 7/ 3) Observation Report No. D OR- 023 Rev. O ! AMENDED RESPONSE The following is an amended response prepared in accordance with agreement reached between SWEC and Bechtel in meetings and telephone conferences on March 7, and March 10, 1989. PART II - RESPONSE
- 1. Observation concurrence:
Concur with observation _ZL Do not concur with observation (Note: if not in concurrence, explain in
" remarks" below)
- 2. Response to Observation:
~
A. Causal Factor (s)
~
B. Extent of Condition C. Significance D. Corrective Action E. Action to Prevent Recurrence
====================================================================================
Remarks The concern expressed in DOR-023 is that the approach of artificially increasing the initial fluid density in the Safety Relief Valve discharge line (SRVDL) to avoid the occurrence of supersonic flow may not be conservative for piping in the suppression , chamber. The discussion below addresses the conservatism in question: a) Be,chtel's calculation SR-69(146) Rev 0 documents a sensitivity study f or SRVDL A in which the effect of changes in initial fluid density, SRV mass flow rate, SRVDL inlet pressure, heat transfer, etc. on forcing functions were evaluated. Pipe run segments 6, 7, 8, 9 and 10 are in the suppression chamber area. Bechtel calculation SR-69(131) Rev 0 documented the original analysis in which the initial fluid density was low and the calculated flow exceeded sonic velocity during the first 0.041 seconds of the transient in the first 39 feet of piping, e out of a total of about 150 feet from the SRV to the quencher. In this ( calculation, the SRV opening time was 0.02 seconds and heat transfer to the pipe wall was not allowed 1.e. the pipe wall was treated as adiabatic. Case 7 and 8 of SR-69(146) were identical except that limited heat transfer (only half the Page 3 of V
.S'IONE & NEBSTER ENGINEERING CORPORATION LINERICK 2 IDCA (J.O. No. 18138) IBCA[W74 OBSERVATION REPORT Observation Report No. ._p_. OR- 023 Rev. O PART II - RESPONSI CONTINUED value of the calculated heat transfer coefficients used for conservatism) was allowed in case 7 whereas it was not allowed in case 8. Also the SRV valve opening time for these cases was 0.025 seconds as opposed to 0.020 seconds in calculation SR-69(131) . Case 7 is the final design case.
Therefore, case 8 of SR-69(146) and the original case in SR-69(131) Rev 0 are
. identical except that the SRV opens faster in SR-69(131 ) and the initial densities were different. Comparing the calculated peak forces between the two cases, one finds that except for pipe segment number 10 (last segment) and the blowdown force, all other forces are significantly higher (about 40 %) when the initial density is artificially increased. Segment 10 and the blowdown forces are 126 and -255.10 kips respectively, for case 8 of SR-69(146); and 125 and
-254.99 kips respectively, for the low density case. It should be noted that the.value for segment 10 shown in figure 14 of SR-69(131) Rev 0 is 146 kips but this is a spurious spike in the time history.
From the discussion in the above paragraph, increasing the density artificially resulted in'the predicte) forcing functions being much higher for pipe segments O 6 through 9, which did not contain the water leg. Also, the predicted wave and blowdevn forcing functions were reduced by a very sme.Il amount in the last pipe run segment that contained the water leg initially. Since the fluid forces in the piping segments containing the water leg initially are primarily governed by the water leg momentum changes during its expulsion from the pipe, the effect of initial density change is expected to be small for these segments. Peak water rate flow (and hence the peak force on segment 10) occurred at 0.32 seconds into the transient. During the period the calculated flow exceeded sonic flow (o to 0.041 seconds), the water leg did not even start to move. Comparing the results of cases 7 and 8 of SR-69(146), the effect of changes in pipe wall heat transfer is much greater than that from changes in the initial density for the last pipe segment. b) The following steps were taken to ensure that the calculated pipe run forces were conservative:
- 1) Forcing function predictions from RVCL were compared with those from the RVFOR04 and RVFOR05 computer codes from GE and the RELAP5 code.
- 2) Forcing function predictions from RVCL were compared with available test data for code validation purposes and to establish conservatism.
- 3) Conservative data (modeling and boundary conditions) were used to generate design forcine; functions.
- 4) Structural evaluations based on the results of the various codes above w.,. .ad..
I6 Page 4 of W
?
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA 3b [34 (J.O. No. 1siss) p ( OBSERVATION REPORT Observation Report No. _A_ OR- 023 Rev._0.L PART II - RESPONSE CotrrINUED The higher peak force calculated by RELAP5 for segment 10 was attributed' to quencher modeling details. For pipe segments where there is no water leg, the results from RVCL for the artificially increased density case were much higher than those from RELAP5 with lower density. It is in these segments that the effect of supersonic flow would be expected most. In the laat segment, as discussed in the paragraph above, water slug dynamics dictate the forcing functions to a large extent. These dynamics ar's very sensitive to quencher modeling. For the design analysis we believe that the conservatism used in modeling and selection of boundary condition data (item 3 above) do assure a large conservatism. Comparing the design forcing function with those for f atigue evaluations where more realistic. conditions are used, one gets an' estimate of the added. conservatism due to modeling and boundary condition data. .! c) The occurrence of supersonic flow in real SRVDL flow situations is very unlikely and as such there has been no real impetus to experimentally investigate this phenomenon. There is no test data available to show the effect of supersonic flow on forcing functions. The approach taken by Bechtel for the design analysis is believed to be very reasonable. Supplemental DOR-023 Response DOR 23 is related to the design load adequacy of SRV discharge line in the wetvell area. Specifically, the forcing function on pipe segment number 10 has been raised as the main concern. The agenda for telephone conversation item listed four points. Bechtel's responses to the four points during the telephone conversation on March 10, 1989, are briefly summarized below: l Point 1: Bechtel agrees that the water slug dynamics is basically an inertia dominated motion. Therefore it depends on the pressure upstream of the water slug, pipe wall friction and the discharge flow area. Also the last segment peak force is governed by water slug dynamics. Bechtel believes that the difference in the last pipe segment force between the j two RELAP5 runs is mainly due to the difference in the discharge flow i areas. Bechtel's calculation SR-69 (145) Rev. O documented this. Bechtel f added that based on their previous experiences, both results from l experimental and numerical (computer simulation runs) have shown that the stooper the change in cross section is, the larger the effects on the dynamics of the water slug as the gas / water interf ace passes through the change in cross section. Stone & Webster requested Bechtel to make a O couple of confirmatory runs on any computer code to quantify the effect of initial density on last segment forcing functions. Page 8 of 8
STONE & WEBSTER ENGINEERING CORPORATION LINERICK 2 IDCA / (J.O. NO. 18138) OBSERVATION REPORT G Observation Report No. _.p__ OR- 023 Rev. O PART II - RESPONSE CONTINUED i Calculation SR-69 (150) Rev. O was accordingly prepared and transmitted to SWEC via transmittal S0707. In this calculation, the base case for the evaluation of the effects of the initial density on the pipe segment peak force was chosen to be the design case for SRVDL "A", case # 7 in calculation SR-69 (146) Rev. O except for changing the valve opening time from 25 milliseconds to 45 milliseconds. The valve opening time was increased to avoid supersonic flow occurrence with low initial density. Thus, in the calculation SR-69 (150), with this new base case, the initial density was varied. To keep the total' mass of the initial air in the line the same for all of the cases in these sensitivity runs, the initial line pressure was adjusted as well. All other input to the RVCL code remained the same as the original design case [ case 7 in calculation SR-69 (146)]. The table below gives a summary of the results of calculation SR-69 (150). Initial Initial Spike Pressure Peak Force Magnitude case Densit{ Ibm /ft psia kips kips i O.055 16.5 113.8 171 11 0.06 18. 113.8 179 111 0.072 21.66 113.7 163 (base case) iv O.08 24.00 113.8 117 From the above table it can be seen that the effect of the initial density on the peak force is insignificant. Comparing Case 7 of Calculation SR-69 (146) to lii (base case) above, we find that the longer valve opening time, as expected resulted in a slight decrease in the peak force (114.03 kips versus 113.7, respectively) . The emergence of spikes in the force time history for cases i through iv and their random magnitudes, while the original case (#7 in Calculation SR-69 (446) did not exhibit a spike, is an indication of spurious nature of the spike. Point 2: Bechtel showed that in calculation SR-69 (146) cases 1 and 8 are similar except for initial air pressure in the discharge line. Therefore, if the spike is a real phenomena, it should be of the same order in the two cases. The peak forces for cases 1 (including spike) and 8 are about 200 kips and 125 kips, respectively. The results of the table shown above . also confirms the random / unreal nature of the spike, i The base case of SR-69 (150) and t.se Case 7 of SR-69 (146) are identical except for the valve opening times. The peak force of SR-69 (150) (base Is Page (a of & Y
STONE & WEBSTER ENO!NEER!WO CORPORATION 1.runa:On a inca 2DCA/f7p (J.0. NO. is100) OBSERVATION REPORT Observation Report No. ]L OR=,,,921., Rev..,g, [ ART II - RERPellAE.AbMTIMtitD stee) is tilghtly smaller than the peak force of 3R=69 (146) which is to be esposted because of the larger valve opening time of SR-69 (150). This supports the senclusion that the spike in SR-69 (150) is spurious as there is no reason for the forces to increase by about 50 kips due to a longer valve opening time. The spike in the segment 10 foros is sensidered to be spurious. However, if the spike were present in the forcing function. It was taken into secount in performing the structural analysis of the lines, calculation ; 8/8041/D-Ott is an example of this analysis, transmitted to swsc via j transmittal 8070s. ( I r.inte a and 4: In the March to conference call stone and wenster stated that reopendias
- t. ,. int i ui ,.seive ,eint. . .n. 4 ai.e.
O i g# lO e h s /, @5 { zumr w,# L ( l neo n.vi., sign.tur. i t. sechtoi i,prov i sign ture / o.t. IO
,aget.e,9 o
TELEPHONE CONFERENCE ITEM RELATED TO BECHTEL RESPONSE TO DOR-023 The concern raised in the Observation Report DOR-023 Rev. O is regarding the design load adequacy of the SRV discharge line in the wetwell area. Specifically, the concern is regarding the blow down load for piping segment l number 10 and water clearing forces. Maintaining structural integrity of the SRV discharge piping inside the wetwell is required to avoid steam bypass and condensation instability. Bechtel's response addresses primarily the design load for piping segments above the suppression pool water level. Questions still remain on the contentions made by Bechtel regarding the water clearing forces and blowdown loads on Segment Number 10. Specifically, the following items need to be clarified.
- 1) The water slug dynamics is basically an ibertia dominated motion s as such the controlling factor is the. pressure at the steam-water interface, the friction factors, and the discharge flow area.
It is not clear why a quencher modelling difference would contribute to the large load difference observed in the comparison. In particular, why the two RELAPS runs, which presumably used the same quencher model, produce 'results that exhibit different trend compared . to that shown by the RVCL codes? Rerunning the RELAPS code with an equivalent quencher model, to that used in RVCL that Q Q comprises , equivalent friction factors and discharge area may be an alternative to resolve the issue.
- 2) It is stated that the higher load found in the calculation SR-69(131) Rev. O is a spurious spike. What is the basis of this contention?
- 3) The blowdown and water clearing forces for Segment Number 10 were not compared with GE computer codes RVFOR04 and RVFOR05. It is known that the water clearing forces can be calculated in the GE computer codes. Since the GE code REFOR is much closer to RVCL code both in analytical model as well as in numerical schemes, a complete comparison with RVFOR05 would be more meaningful and conclusive.
- 4) It was stated that the water clearing forces calculated for Segment Number 10 were not significantly af fected by the change of initial air density in the RVCL runs. However, the comparison was based on Case Number 8 which was not used in the design load. Since the conclusion derived from this comparison is not consistent with the RELAPS results which indicated that the water clearing ,
force is very sensitive to the initial air density, it would be ( advisable and meaningful to address the sensitivity of the water l clearing force using Case Number 7, and not Case Number 8. O DOR - o J 3 REV o PAGE E OF 10 i Telecopied 2/23/89
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT Observation Report DOR- 023 Rev. 0 Review Plan: LK- D - 1903H - MC Rev. 1 PART III - EESPONSE ETALUATION 4 The respont.e to this observation report has been evaluated as follows: A. Causal Factor (s): Acceptable Not Acceptable (Explain):
-NOT APPLICABLE B. Extent of Conditions- -Acceptable Not Acceptable (Explain):
NOT APPLICABLE C. Significance: Acceptable Not Acceptable (Explain): b v NOT APPLICABLE D. Corrective Action: Acceptable Not Acceptable (Explain): NOT APPLICABLE E. Action to Prevent Recurrence: Acceptable Not Acceptable (Explain): NOT APPLICABLE Additions 1 Action Required: X No Yes (Explain) SEE ATTACHED PAGE q G 90hknL Lead F.ngineer Signature
"/$r / Date ocLic& D.wals 4/cln AMI Signature / Date 9 /o Page d of d s< n
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. No. 18138)
-O-O OBSERVATION REPORT Observation Report DOR- 023 Rev. O PART III - RESPOWSE ETALUATION CONTINUATION: i
\
DOR-023 Rev. O was written because of the concern on the design load adequacy for piping in the suppression chamber, specifically regarding the piping load on segment 10. As was explained in D0R-023 Rev. O, Bechtel's justification was based ' on load comparison. Loads generated from the computer code RVCL (NE805) (Calculation SR-69(131) Rev. 1) are , compared with the loads generated from the industry accepted computer code RELAP5. However, comparisons of piping loads in the suppression chamber, namely pipe segment 10, was not favorable. As a result DOR-023 Rev. O was written. The initial response did not adequately resolve this concern. Therefore a telephone conference agenda was prepared for a focussed discussion. As a result of the telephone conversations, Bechtel agreed to perform additional confirmatory computer runs to demonstrate that changes in the initial air density would not adversely modify the driving pressure at the steam-water interface, thereby af fecting the blow down loads on pipe segment 10. f'N The supplementary response which is supported by Calculations SR-69(150) Rev. O and S/8031/D-025 Rev. 1, Bechtel shows that the main effect of () the water clearing loads was not significantly altered by the initial air density change. This conclusion was derived from the Calculation SR-69(150) Rev. O using RVCL (NE805) computer code. Since all confirmatory runs were made within the computer code applicability zone, i.e., without complications of supersonic flow conditions, and adequate density differences, the results are believed to be credible. In addition, Bechtel stated that the forcing function was not manually filtered to remove any spikes that may be present, and the forcing function in its entirety was used in the structure analysis, as is demonstrated. in the Calculation S/8031/D-025. It can therefore be concluded that the concern on the piping design load adequacy on the SRV discharge piping in the suppression chamber has been adequately demonstrated and the issue raised in DOR-023 Rev. O is adequately resolved. i O /0 g /c 3 o.g e. f 0 [ p n ! I
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) (gj OBSERVATION REPORT Observation Report No. D OR- 024 Rev. O Review Plan LK- D- 1903- - S Rev. J R'ference e AI No.: DAI-218. Rev. O DAI-233. Rev. O PART I - INITIATION
- 1. Description of Concern Calculations for the evaluation of pool swell effects contain errors which could affect the design.
- 2. Supporting Information In calculation 17-C, the output from the computer program SS*PSAM.
IMPACT was used as psf. when it is actually psi. This leads to an underestimation of laads on the structure. Response to Action Item DAI-218 indicates that effects of this load will be minor because of the short duration. (Note that the response states that the duration 'is .000031 seconds when it is actually
.000081 seconds).
s
) In calculation 113.4.8, pool swell drag loads on the grating are not utilized because they were erroneously considered to act only on the solid area of grating rather than the total area.
Application of the loads to the total area will lead to larger drag loads.
............................................................................===...===..
Additional Documents Attached: None W (. Clk:,n /t./Nlff f D4te
' ' / /f/W
'/ Date Lead Engineer Signature APM'Sigpture /
Page 1 of d I
I: l SIONE & WEBSTER ENGINEERING CORPORATION (' LIMERICK 2 IDCA (J.O. No. 18138) V OBSERVATION REPORT Observation Report DOR-024, Rev. 0 PART II - RESPONSE 4M 2 51989
- 1. Observation Concurrence:
l X Concur with observation Do not concur with observation (Note: if not in concurrence, explain in " remarks" below)
- 2. Response to Observation:
A. Causal Factor (s) S e misinterpretation of the pressure output for the computer program SS*PSAM was caused by assuming that the impact pressures were in the same units as the calculated drag pressures. The pressure units were correctly presented in the computer documentation included as an appendix to this calculation. The misinterpretation of the grating area (solid area in lieu of gross or total area) to be used for calculating poolswell grating drag loads (p) resulted from a misinterpretation of the criteria presented in NUREG 0808, Appendix A. The location of the footnote indication was literally mis-leading in that it indicated that the NRC had apparently adopted the solid area criteria, where as the intent of the footnote was a simple reference to the previous document providiro the general criteria for poolswell grating drag load. B. Extent of Condition The misinterpretation of the impact pressure units in program SS*PSAM was limited only to the computation of poolswell loads for the design of the wetwell platforms in Calculations 17-C. Other sub-systems within the wetwell subjected to the poolswell effect were evaluated using different methodology and therefore are not affected. The misinterpretation of the grating area was limited to the computation of poolswell loads used in the evaluation of the pedestal liner plate in Calculation 113.4.8, Revision 0. All platforms within the wetwell were originally designed in 1978 (Calculation 17-C) using the total area of the grating.
===================================================_===============================
Remarks: f) i O
' rage Q of (a
l l STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. No. 18138) f 3 OBSERVATION REPORT
;C}
Observation Report DOR-024, Rev. O PART II - RESPONSE CONTINUED C. Significance l The significance of the inadvertent use of PSF in lieu of PSI pressure units l is inconsequential since the poolswell impact loads do not govern the design of the wetwell platforms as confirmed by calculation using the correct engineering units. The poolswell drag loads, which are applied independently of the impact loads, govern the design. The significance of the misinterpretation of solid vs (gross) total area for the computation of poolswell grating loads in the evaluation of the pedestal liner plate disposition of 11CR 11733 (Calc 113.4.8, Revision 0) is inconse-quential. Calculation 113.4.8 has been revised demonstrating that the poolswell loads acting on the grating is smaller than those used in the original design,-and the pedestal liner plate meets Category IIA criteria for applied poolswell drag loads (for additional discussion see DOR-035). The determination of grating poolswell drag loads for the design of the platforms used the (gross) total grating area and thus did not need further revision or evaluation. p- D. Corrective Action O Calculation 17-C Rev. 7 has been revised to demonstrate that poolswell drag loads govern in lieu of the impact loads with the correct interpretation of pressure units (PSF vs. PSI). (Calculation 17-C, Revision 8, has been transmitted to SWEC via transmittal No. S-0596, dated 1/27/89). The evaluation of the pedestal liner in Calculation 113.4.8, Revision 0 has been revised to include a detailed evaluation documenting that the liner plate conforms to Category IIA criteria for poolswell drag loads based on (gross) total gratirg area. E. Action to Prevent Recurrence No further action is required because of the following:
- The observed discrepancies were isolated instances which are shown inconsequential by detailed calculation revisions. The evaluations conclude that the wetwe'l platfonns meet their design criteria.
(Calculation 113.4.8, Revision 1, has been transmitted to SWEC via transmittal No. S-0596, dated 1/27/89). O Page ] of i
. JAN-30 '89 14: 17. ID.:PECQ,PROJE T_MGT,DIV TEL NO:215-B41-4578 #213 PO4 o STONE & WEBSTER ENGINEERING CORPCEATIQj LIMEAICK 2 IDCA amy
- (J.O. NO 18138)
OBSERVATION EPORT Observation Mport Na. ICR-024, Rev. O I PART II - ESPONSC CORrINUED
- The resulta of the observation report and the calculation revisions have been discussed with the involved ergineers.
ii O i O --- - stuin n a,h- a -a- n-- 3 s, pe- d.w- M ,.. ,,, - . , ,... - n,-.,,,,a-. . , .. Pagelofi
-Z .
STONE & WEBSTER' ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT J Observation Report. DOR- 024 Rev. O Review Plan: LK- D - 1903 - S Rev. 1 PART III - RESPONSE EVALUATION The response to this observation report has been evaluated as follows: A. Causal Factor (s): X Acceptable. Not Acceptable (Explain): See Attached Page. B. Extent of Conditions X Acceptable Not Acceptable (Explain): See Attached Page. C. ' Significance: X Acceptable 'Not Acceptable (Explain): See Attached Page. D. Corrective Action: Acceptable Not Acceptable (Explain): S'e e Attached Page. i E. Action to Prevent Recurrences X Acceptable Not Acceptable (Explain): See Attached Page.
...............................m................. m====================================
Additional Action Requirec.: X No Yes (Explain) 3
.]
i i m a s s am me s as s am a m a amm e s e ss a.m m a s a s s m a n e s s m a n e s s e m a s s ma m m as sa ma m a ss s m a s a s sm a a m m us sss s m u s a btra. {. Yn .1. 6 9 $ ' h D.N$$t- .2fl $$
' Lead Engineer Signature
'/ Date APh Signature / Date 1 l
Page [ of 1
I V A. The cause was determined to be due to a misinterpretation of units (PSI rather than PSF). Pool swell loads consist of drag and impact loads. Drag loads are given in PSF and it was apparently assumed that impact was in the same units. A misinterpretation in application of area of the grating bars (solid area) rather than total area of the grating bars was due to misinterpretation of NUREG 0808. This is considering likely as the NUREG is not entirely straight forward in its requirement to use solid area. B. The misinterpretation of units was limited to the wetwell platforms that used the program SS*PSAM. The response stated that this method was not used for other wetwell subsystems. The misinterpretation of units is considered likely to be unique to pool swell due to magnitude of pressure loads encountered. The misinterpretation of solid grating area was limited to one calculations other calculations involving pool swell were reviewed by Bechtel and did contain the correct area. It was noted during review of this calculation that a memo was attached that stated the use of pool swell loads on the total area rather than the solid area was too conservative and that the calculation in question could utilize the bar area. This memo is considered the direct cause and appeared to be specific to the one calculation. This type of misinterpretation (p) v is considered isolated due to the unique nature of pool swell loading. C. It was determined that the misuse of units vae inconsequential as the loading condition with the incorrect units did not govern the design even when the units were corrected. This was done by a more detailed check on the dynamic load factor. The dynamic load factor found was low enough to offset the mistaken units, l The use of the area of grating bars versus the total area occurred in a single calculation (113.4.8) that checked an attachment to the Category IIA pedestal liner. That calculation was revised and the revised calculation (also the subject of DOR-035) showed that the liner was capable of resisting the applied loads with a reasonable factor of safety. , D. Calculation 17-C has been revised using the correct units. It is noted that revised calculation utilized a half sine wave impulse instead of a triangular impulse. However, the result is conservative. The pedestal liner calculation (also the subject of DOR-035) has been revised to include the proper area of grating. E. It is accepted that the occurrences are isolated to the calculations l that were corrected. Due to the unique nature of the type of loading, no further action is required. 1 O DDR - 0 A Y REV o PAGE G OF 6
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138)
,A OBSERVATION REPORT Observation Report No. D OR- 025 Rev. O Review Plan LK- D - 1903 - MS Rev. J Reference AI No.: DAI-059, Rev. O PART I - INITIATION
- 1. Description of Concern There are deficiencies in the RHRSW Pump TDH Calculation (No. M-11-15, Rev. 1)'that could lead to confusion and possible design problems.
- 2. Supporting Information
- a. Additional pressure drop due to throttling of valve F068, determined in calculation No. M-12-30, is not considered in the RHRSW pump TDH calculation No. M-11-15.
- b. The basis for the selection of RHRSW system modes of operation used in- the calculation are not provided. Bechtel letter to PECo, BLP-22627, attached to calculation No. M-11-15, gives i
[3 a list of operating modes for the ESW and RHRSW systems and V .does not provide the basis of selection of these modes. is also noted that the evaluation of the RHRSW mode during the It winter operation that may require throttling of the cold RHRSW system to maintain an acceptable reactor coolant temperature leaving the RHR heat exchanger is not considered. This would be a reduced flow mode for the system and it should be ensured that ' the flow in this mode meets the minimum flow requirement of the RHRSW pump. Furthermore, the modes where both RHRSW and ESW pumps of a loop are operating, for example mode 8 (sheet 207 of calculation No. M-ll-15), the ' pump curves should be j
~
reviewed for parallel operation of RHRSW and ESW pumps to ensure that one of the pumps is not providing a high pressure in the system such that the other pump can not overcome both the system resistance and the system pressure provided by the other pump. CONTINUED ON 1 PAGE m aammmmmmmmmm mmmmmmmmmmmmmm mmmmmmmmmmmmmmmmmmmm mmmmmmmmmmmmmmmmmmmmmmmmm mm m mmmmmm==== mm Additional Documents Attached e i maammmmmmmmmmagemmmmmmmmmmmmmmmapsmmmmmmmmmmmmmmmmmmmmmmmQ ========= ========= ' f == m AJAf $$ \ / Ledd Enfin _r%j; nature ~/ Date APM Sig6ature / Date Page1of,$,
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT l O Observation Report No. 3 OR- g Rev. O PAB.T 1 - INITIATION CONTINUATION:
- c. Determination of the pressure drop for the orifice on sheet 173A of the calculation is not traceable. It is not clear how a pressure drop of 240 inches is obtained in the second phase i using a beta ratio of 0.595, flow of 540,000 gph, and pipe inside !
diameter of 19.25 inches. Using these values on the Daniel Slide Rule yield 400 inches of pressure drop and appears to be an incorrect method to determine the pressure drop. However, from Figure II-III-4 of the ASME Fluid Meters book (sixth edition, 1971, pages 201), the overall pressure loss across the orifice should be about 64% of the maximum differential pressure (0.64X400
= 256 inches). ;
- d. The RHRSW system pressure drop for loops A and B flow paths, including the governing case for the RHRSW pump sizing, are not addressed in the calculation purpose and conclusions. This is, however, done for the ESW system, i
O i l l I O Page A of 3
s STONE & WEBSTER ENGINEERING CORPORATION LINERICK 2 IDCA y l (J.0. No. 18138) 4 jggg l OBSERVATION REPORT pff Observation Report No. D._ OR- 025 Rev. O PART II - RESPONSE
- 1. Observation Concurrence:
Concur with observation _.11._ Do not concur with observation (Note: if not in concurrence, explain in
" remarks" below)
- 2. Response to observation:
A. Causal Factor (s) B. Extent of Condition C. Significance D. Corrective Action E. Action to Prevent Recurrence O .................................................................................... Remarks AMENDED RESPONSE The following is an amended response in accordance with the telephone conference call held on February 10, 1989. Refer to Telephone Record No. DTR-010 (DCIl IDCA 1792) .
- a. Additional pressure drop due to throttling of valve F068 is not required to be considered in Calculation M-11-15. The purpose of this calculation is to determine the ESW and RHRSW pump operation conditions under various modes of oporation and to assure that the required flow is provided by the ESW and RHRSW pumps. Calculation M-12-30 was performed at a later date to determine the available NPSH for RHRSW pumps and to introduce additional pressure drop in the RHRSW flow path to assure adequate NPSH. Calculation M-12-30 showed that even with the RHRSW system aligned and operated in the most restrictive mode with the additional pressure drop in the system, the RHRSW pumps still provided the required flow of 9000 gpm. Although the additional pressure drop causes the RHRSW pump to operate near its design condition, it will not deviate from the Calculation M-11-15 conclusion that the RHRSW pump can provide the required flow under all modes of operation.
- b. The listing of modes in the BLP is the result of a joint effort between Bechtel and PECo to identify all modes of operation. The selection of various modes of p operation reflects the design basis of system alignments for plant operation Q .during emergency and normal cooldown conditions and the possible system alignments that may result due to a single active failure. This is stated in Page _,3_ of 1 ,
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA MM /[$~h (J.O. No. 18138) OBSERVATION REPORT Observation Report No. _D_ OR- 025 Rev. O PART II - RESPONSE CONTINUED flow in the common spray header. The purpose of this plot is to be able to establish various ESW pump operating points considering the added RHRSW flow from the common header point to the spray pond header. This family of curves ha: accounted for the resistance (Cg to cm) from the common heaaer back to the ESW pump. The resistances were converted to resistance coefficients. See Sheet 244' lines 16 to 31 and Sheet 246.
- b. A family of RHRSW system resistance curves was plotted on the RHRSW pump curve considering different ESW flow. The purpose of this plot is to be able to establish various RHRSW pump operating points considering the added
'ESW flow from the common header point to the spray pond header. This family of curves has accounted for the resistances (c to cm) from the common header back to the RHRSW pump. See Sheet 245 lines 2 to 14 and Sheet 247.
- c. To converge the ESW pump and RHRSW pump operation, a third plot was generated using RHRSW flow as ordinate and ESW flow as abscissa.
The intersection points between the ESW system resistance curves and the ESW pump curve were plotted and a curve was drawn on Sheet 248. A second curve was generated from the intersection points between the RHRSW system resistance curves and the RHRSW pump curve. The intersection point of these two curves defines the flow of RHRSW pump and the ESW pumps. In Mode 15, the ESW pump flow is found to be 6550 gpm which satisfies the required flow of 5999 gpm (see Sheet 1B) and the RHRSW flow is 10,500 gpm - which satisfies the required flow of 9000 gpm. The pump flows do not place the pumps in a dead-headed or run-out condition. To see the effect on the ESW pump due to combined ESW and RHRSW flows in l the common spray header, refer to Sheet 246. With no RHRSW flow in the I common spray header, the ESW pump operating point is 7300 gpm at 213 feet TDH. With combined ESW and RHRSW flows in the common spray header,'the ESW pump operating point is 6550 gpm at 237 feet TDH. A similar conclusion can be drawn for the RHRSW pump from Sheet 247. This change in pump operating point is due to the combination of increased friction loss fron. the common point to the spray headers and decreased friction loss between the pumps and the common point. l 1 O Page 9 of _$_ { 1
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA g g g/ , (J.O. No. 18138) {N OBSERVATION REPORT Observation Report No. D OR- 025 Rev. 1 PART II - RESPONSE CONTINUED t BLP-22627. It is not considered necessary to document in further detail the process or selection in the BLP or calculation. Acceptable reactor coolant temperature leaving the RHR heat exchanger is controlled by valve HV-C 2F048 (A or B) which is used to bypass' flow around the heat exchanger. Throttling on the RHRSW side is not required. { The ESW and RHRSW pumps do not operate in parallel except to share a common spray network at the spray pond. The effect of combined ESW and RHRSW flows in the common spray header on the system resistance curve was evaluated for both ESW and RHRSW pumps and shows an acceptable result. c, Note that there are two scales for the beta ratio on the Daniel Orifice Calculation Slide Rule. The top scale applies to vena contracta taps to determine the maximum pressure drop and the bottom scale applies to pipe taps to determine the overall pressure drop with pressure recovery. The bottom scale ! was used to obtain the overall pressure drop across the orifice for the system O(,/ resistance curve. The pressure drop of 240 inches was obtained using the beta ratio (bottom scale) of 0.695, specific gravity of 1.0, flow of 540,000 gph, and pipe diameter of 19.25 inches. Using the top scale will yield a pressure drop of 400 inches.
- d. Unlike the ESW system which consists of numerous parallel flow branches in series with common supply and return headers, the RHRSW system has one flow path.
Therefore, evaluation to determine the governing case is not needed for the RHRSW system as was done for the ESW system in Calculation M-11-15. Both the RHRSW A and B loops are analyzed in the M-11-15 calculation. As mentioned in b above, the RHRSW system does share a common spray network with the ESW system. The effect of the RHRSW pump operating condition with or without the ESW system operating was evaluated. The flow rate for each of the analyzed modes is indicated on sheets i A and 1B of Calculation M-11-15 and shows that acceptable flow is provided by the RHRSW pump. SUPPLEMENTAL RESPONSE
- 1. Effect of Pressure Drop Between Pumps and Common Point The changes in system resistance for the portion of piping system upstream of the common ESW/RHRSW return piping was accounted for in the calculation. This can be explained by going through the steps or logic of the calculation. Mode p 15 is used as a typical example.
O a. A f amily of ESW system resistance curves was plotted on the ESW pump curve. Each resistance curve represents a different RHRSW flow added to the ESW Page I of 7
BTOM & WESSTRR EWO!NEERING CORPORAT10N 129(/7 j'gfg/_ ( I.!MRICK t IDCA (4.0. WO. 18188) Q- CB8ERVATION Rtp0RT Observation Report No. ,,jL, OR _Q2)_ Rev._Q., NAT .II.= BR4POMER CONTIMII54
- 2. Effect of MR Heat Exchanger outlet Valve throttling The result of RNR heat exchanger outlet valve (RNR8W) throttlant, does not have to be incorporated in calculatlen M-11-15 because, as Calculation N-12-30.
Revision 0 has concluded on sheet it, the actual field test has demonstrated acceptable system performance using simultaneous operation of RNR8W with ESW with the most restrictive flow path including the valve throttling. The effect of RHR heat exchanger outlet valve (RNR8W) thrott!!ng is a slight increase in flow I of the R$W pumpe on the loop. This is beneficial since more cooling flow is avatiania t. r..ev. ts. h.ai io.d. O 1 esoseeeeeeeeeeeeeeeeesessnes.....ssenessssssssssssssssssssssssss . ................. O ^A n hl%M .l3/det acu WL M SAM -_ ' PEco Review Signature / Date
- i i fff / Bechtel Approval signature / Data Page 6 of $,,,
N.] STONE & WEBSTER' ENGINEERING CORPORATION LIMERICK 2 IDCA
.,x
-(J.0c No. 18138) .
OBSERVATION REPORT Observation Report D OR- 025 Rev. 0 1 Review Plans LK J__- 1903 - MS Rev. ~1_ ! PART II.I - RESPONSE EVALUATION The response to this observation report has been evaluated as follows: A. Causal Factor (s): Acceptable Not Acceptable (Explain): NOT APPLICABLE B. Extent of Condition: Acceptable Not Acceptable (Explain): NOT APPLICABLE C. Significance: Acceptable Not Acceptable (Explain): 1 NOT APPLICABLE
.D . Corrective Action: Acceptable Not Acceptable (Explain):
NOT APPLICABLE E. Action to Prevent Recurrence: Acceptable Not Acceptable (Explain): NOT APPLICABLE Additional Action Required: X No Yes (Explain) SEE INSERT A o v .. .... - - ... L.................................................................... LeiaVEngindeN Signature t15 %]93M
/ Date oCAsk% & wie /s/dn APM/ Signature Date g Page 7 of T i .. _ _ _ _ _ _ _ _ . _ _ _ _
7 l I L STONE & WEBSTFR ENGINEERING CORPORATION 1.IMERICK 2 IDCA (J.O. No. 18139) OBSERVATION REPORT Observation Report DOR- 025 Rev. 0
.PART III - EESPONSE ETALUATION CONTINUATION:
INSERT A
- a. As explained in the Observation Report responsa, the lack 'of consideration of additional pressure drop due to the throttling of RHRSW heat exchanger valve F068 in Calculation Number M-11-15 is not a concern since it is demonstrated in Calculation Number M-12-30 that the RHRSW pumps provide the required design flow with the RHRSW system aligned in the most restrictive mode with the additional pressure drop in the system. However,.it is noted that Calculation Number M-11-15 does not reference Calculation Number M-12-30 for the throttled valve conditions,
- b. The lack of basis for the selection of RHRSW system modes of operation used in the calculation is not a significant concern as these modes are jointly selected between Bechtel and PECO.
Throttling of Valve F048 in the RHR system to maintain an acceptable reactor coolant temperature during the winter operation is acceptable. O As explained in Bechtel's supplemental response, it appears that the calculation addressed the situation where the varying flows in either' loop affect both the other (either ESW or RHRSW) as well as the combined flows. The effect of throttling RHRSW heat exchanger valve F068 is also considered acceptable since the response explains that an acceptable system performance was demonstrated in an actual field test with simultaneous operation of ESW and RHRSM pumps in the most restrictive flow path. However, a minor adjustment of resistances within calculation Number M-11-15 to show the impact of throttling would have made this calculation complete.
- c. Determination of the pressure drop for the orifice using the Daniel Slide Rule is still not clear. However, the difference between the calculation pressure drop (240") and the pressure drop calculated by SWEC (256"), as mentioned in DOR-025, is 16". This difference, with respect to the overall pressure drop is not considered significant and the concern is resolved.
- d. The lack of clear statement in the calculation purpose and conclusion with respect to the RHRSW pressure drop for loops A and B flow paths, including governing case for the RHRSW pump sizing, is not a concern since both the RHRSW loops are analysed in the calculation. However, it is noted that the RHRSW system does not have one flow path but has three distinct paths (flow to the spray network, the spray pond winter bypass, or the cooling towers).
O Pope .1 Of 1
c f - L f STONE & WEBSTFR ENGINEERING CORPORATION I LIMERICK 2 IDCA
,, (J.O. NO.' 18138) l
[V' ) - OBSERVATION R2 PORT o Observation Report No. DDR- 026 Rev. O Review Plan: LK- D - lE - MS Rev. T_ Reference AI No.: DAI-353. Rev. O PART I - INITIATION
- 1. Description of Concern Radiation Protection Calculation S102.1, which examines the shield design adequacy of the labyrinth entranceway to the RHR Loop B pump cubicle from the Radiation Zone II stairway, neglects to address the scattered gamma contribution to the total dose rate in the stairway zone at the cubicle doorway (the direct shine contribution of which has already been determined to exceed the Zone II limit by 1.5 mrem /hr ovpr a small region at the cubicle doorway).
- 2. Supporting Information Action Item DAI-353 raised the concern that Calc. S102.1 neglects (n) the contribution of scattered gamma radiation in the evaluation of the RHR Loop B cubicle entranceway shield design adequacy. This action item points out that Calc. S102.1 calculates a maximum dose rate of 4 mrem /hr at the doorway between the Zone II stairway and the cubicle entranceway 'this dose rate being 1.5 mrem /hr greater than the Zone II limit. The action item response acknowledges that the calculated dose rate exceeds the Zone II limit, but asserts that only a small portion of the doorway is af fected. This response provides a qualitative argument to support the assumption that scattered gamma radiation would contribute insignificant 1y to the total doorway dose rate, thus obviating the need for a detailed scatter analysis. In a situation in which the direct shine dose rate contribution exceeds the established zone limit (even if this zone limit is exceeded over a small region), scattered radiation should not be ignored and should be addressed. In the absence of definitive in formation in the calculation, it cannot be verified that the contribution from scatter radiation is not significant in this case.
maam m mmm m me m,mm a mm m mm m mm m m m m m m m m m m mm m m m m m m mm e = = = mm mm m mm m m m m m m m m m m m a m a m u s m a m m m m m m mm mm m m m a Additional Documents Attached: None
/ / A \ )
APM Egtpfture / Date Lea'd Engindet ' Signature / Date Page 1 of k
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA NM /[f[ (J.O. No. 18138) f (.5-0572 ) k- OBSERVATION REPORT Observation Report No. D OR- 026 Rev. .2_ PART II - RESPONSE
- 1. Observation Concurrence:
Concur with observation _XI_ Do not concur with observation (Note: if not in concurrence, explain in
" remarks" below)
- 2. Response to Observation:
A. Causal Factor (s) B. Extent of Condition C. Significance D. Corrective Action E. Action to Prevent Recurrence O, 3555555555333533355533333555555EEEEEEEEEEEEEEBBBBBBESSEEEEEEEEEEEEEEEEEEEEEEEEEEEEEE Remarks The previous response to DAT-353 provides a qualitative discussion on the reason that scattering should be secondary to direct shine and that a specific scatter calculation was unnecessary and, therefore, not performed during the labyrinth design analysis. To further confirm the validity of that reasoning, the following simple, conservative scatter estimate is provided for a dose point at the outside center of the RHR room l doorway. Since all RHR piping radiation sources are located east of the inside face of the labyrinth wall, the only potential scatter surf aces, as shown on the following page, would be:
- 1) labyrinth ceiling below roof slab,
- 2) f loor inside labyrinth up to west RHR room wall,
- 3) back wall behind labyrinth at column D, l 4) triangular area at top of doorway,
! 5) triangular area at bottom of doorway, and l 6) portion of the back side of the doorway. Note that all of these areas are not entirely visible from any single point at the outside of the doorway, but it is conservatively assumed so for this estimate. It is also assumed that the incip*nt dose rate is constant for the entire surface. The l [ calculated dose rate from a't. unshielded 30" HBB RHR pipe of 24.2 mR/hr is used for
\~ -) all surfaces.
Page il of bL
STONE & WEBSTER ENGINEERING CORP! RATION LIMERICK 2 IDCA dddA /[f[ (J.O. NO. 18138) F OBSERVATION REPORT Observation Report No. _p_ OR _Q25_,Rev. O PART II - RESPONSE CONTINUED e 3 s e u a .,. s .p
\
i
/, \'
/' \ d'I s j%
\
\ g
\
s \ o , L /
\
\ s
.n ,
Iph Ii y \ [/ v
\ \
M
\
g _ _ _ - ./ /
- /
/
\ //sn \/
\ g
/'
,w---.---- *
\
l '.
~
\
\ , / x ,
- - \s ->
7; g ,-;,/ , ,
, : / - - a -
+- ~ =
.-- - a _-
~
/
\
. z-
& ~
=- ..
_- .- 'l j , ' ye k. /
\
/ ,
- t ..i@ ,
"/
/ _.
\/ -
1 O . i Page d of I f l k L L'
l. STONE & WEntTtR ENGINEERING 00RPORAT!DN OCA /I f[ 1.!NERICK 2 IDCA
, (J.0. NO. 18188) 8 OBSERVATION SgP01T
.3 $ =-
- gy.$"
Olis2rYatieti Repert Ne '.jL"OR e,24&e, Aov. A, Um Z w b. FART .II - RIspostAt 00MTIERD- . .
.. c.scattee. doses are estimated using the simple feraala
. . - . . . -. .. ~ . . - -
- Ur;
i
;@A,Q D = D, r dA / RI r a,(4,,0,$)
where dA is the scatter surf ace ares.,
' R is the distanee from the ares centreld o, is the aanma albedo for 8,. O and 4 angles at i NeV p, is the acident dose rate, surf ace areas, distances and angles are es:timatad from the figure en the previews Page. Albedes for concrete at i WeV are ottimated using the mothetelegy'in Table C.T !
of ANs standard ANs!/ANS-4.4-1988: . for horison'tal surf aces, a,(60,48,ac) a 4.4 x 10*3, for vertical surf aces, e,(10,80 5) e n.1 x 10'3 Applying those albedos to their respective scatt.or areast and swaning, vill yield a ecatter dose at the doorvay center of uo.7 sA/hr basec on the above conservative methodelegy. since the peak direct shine in the upper right stirner of the dw>rvsy, as calculated using the QAC code in Revision 0 of 5102.1. is etrudy 3.86 nA/hr (s!!ahtly higher than the sene objective), the relatively naall lasroment for scattering would not change any conclusions. For the majority of the doce area, the diroet shine la low (less than t . 29 mPe/hr) . Thus any small increment from saatteritig w!!! niso ncit change the results or the concluslen that the proposed labyrinth design is acceptable with some areas of higher dose rate in not-normally-accessed locations, s!nce it was properly judged during the original calculation that scattering voald not be a significant cencern, and since the results of this conservative satinate agres, there Asi no need te modify the calcul? tion to include a rigorous scatter analysis. S W e m m e e e s s e o s t e S 5 5 9 3 5 3 3 3 E S B S E S S S 5 F r E N R W F 3 0 9 3 8 3 3 3 513 e e ei g e n e e in e a s e s e, e s p o s e g e s W e e t w e g e 3 4
-' 3
_1 - v Q{ i sco neview signature / pate htel Approval signature / cate p Page1(,of,(g
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT
%/
Observation Report D OR- 026 _ Rev. J ' Review Plan: LK- D - 1903 - MS Rev. 1 PART III - EESPONSE EVALUATION The response to this observation report has been evaluated as follows: A. Causal Factor (s): Acceptable Not Acceptable (Explain): NOT APPLICABLE B. Extent of Condition: Acceptable Not Acceptable (Explain): NOT APPLICABLE C. Significance: Acceptable Not Acceptable (Explain): NOT APPLICABLE ! D. Corrective Action: Acceptable Not Acceptable (Explain): NOT APPLICABLE 1 E. Action to Prevent Recurrence: Acceptable Not Acceptable (Explain): NOT APPLICABLE Additional Action Required: X No Yes (Explain) SEE INSERT A
.. t....m ............................................ . ... . ....................
O . teae EerE 14ee r Si Y mature O
, eate APs S1E nat e , eate j Page i of j f
)
l i STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA I (J.O. NO. 18138) OBSERVATION REPORT Observation Report DOR- 026 Rev. O PART III - RESPONSE EVALUATION CONTINUATION: INSERT A Bechtel's response to Observation Report DOR-026 Rev. O adequately demonstrates that the scattered gamma dose rate contribution to the RHR pump cubicle B entranceway dose rate is not a significant concern. However, this response does indicate that Calculation S102.1 is based, in part, on a judgement (unstated in the calculation) that the scattered gamma dose rate contribution is insignificant. Such a judgement is a design assumption. According to EDP-4.37 Rev. 2, design assumptions shall be clearly stated so that they may be understood by the checker. This issue is being addressed by Observation Report DOR-073. O l O Pa y & oE }_ .
STONE (a WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) { q e a ( ,/ OBSERVATION REPORT s 1 l Observation Report No. D OR- 027 Rev. ,0, i Review Plan: LK A- 1901 g . Rev. 1 l 1 Reference Al No.: DAI-185, DAI-319 DAI-354 PART I - INITIATION
- 1. Description of Concern There is a concern that there are insufficient means to identify applicable structural calculations, which could lead to future design deficiencies.
- 2. Supporting Information During review of calculations, it was observed that there is no I conclusive method to identify the most current applicable calculations for a particular structural feature. Specific examples are:
Calculation 10.N for platforms in the Spray Pond Pumphouse is identified on the cov'er sheet as ' final'. However, there are a number of ' supplemental' calculations identified in the response to Action Item 7-~3 DAI-185 that also apply to these platforms. There (' )
' does not appear to be any ef fective , system of cross referencing these calculations.
- Calculations .11-B, 11-B-1 and 11-B-2 all concern the design of penetration X-45B. Action Item DAI-319 '
identified design deficiencies in calculation 11.B. The response indicated that the pages in question were no longer part of the calculation, as reflected on the cover sheet. However, these pages were not removed from the calculation or marked ' void'. The response stated that sheets 9-12 were the applicable pages. After review of pages 9-12 of calculation 11-B, Action Item DAI-354 was initiated. Response to this AI indicated that these pages also were not applicable, and provided pages from calculation 11-B-2. There appear to be the final calculations for Unit 1. Unit j 2 calculations have not been completed. CONTINUED ON 1 PAGE
..................................................................................=....
Additional Documents Attached: None
.......es.........c................p...................... ....... .................. /~'\
l [. h r.-- Lead Engineer Signature
/7. M
/ Date APMSign/ture
/
/ / Date Page 1 of J_ j
t STONE & WEBSTER ENGINEERING CORPORATION f LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT
/
Observation Report No. D OR _027 Rev. O pg 3 , CONTINUATION: Although not previously identified by an action item, the following was also observed. Review of the embedment design for the RER heat exchanger supports is performed in calculation 22.8B (Rev. 5), approved 5/25/84. This calculation is designated as final. However, calculation 101.73 appears to reevaluate these embedments. 9 O Page e2 of 7
(~ L STONE & WEBSTER ENGINEERING CORPORATION l LIMERICK 2 IDCA 1 3 1 ( J. O. NO. 18138) M3 01989 OBSERVATION REPORT Observation Report No. DOR-27 Rev 0 l PART II - RESPONSE 1
- 1. Observation Concurrence:
Concur with observation X Do not concur with observation (Note: if not in j concurrence, explain in " remarks" below) l
- 2. Response to Observation:
A. Causal Factor (s): N/A l B. Extent of Condition: N/A C. Significance: N/A () D. E. Corrective Action: N/A Action to Prevent Recurrence: N/A assesss========sss====semass===ssan=====ssassamasus===ssmammassas Remarks There are sufficient means for knowledgeable engineers assigned on the project to identify and effectively use applicable structural calculations. Use of the appropriate calculations will not lead to future design deficiencies. The future use of these calculations by others and implementation of cross-referencing enhancements will be determined at the time of document turnover. The following specific observations are made for each item in the DDR. , o Item 1 of the DDR addresses multiple final calculations for the Spray Pond Pumphouse. These include " final" calculations 70.N, 102.4.1, 102.4.2, and 102.4.3 which were identified in DAI-185. Some of these calculations are for platform sizing, some for field changes, and some a check of the f platform for changed input parameters. The term " final" for calculations does not mean Page _3_ of 9 i
< M STONE & WEBSTER ENGINEERING CORPORATION 1
LIMERICK 2 IDCA ( J. O. NO. 18138) OBSERVATION REPORT Observation Report No. DDR-27 Rev. O PART II RESPONSE CONTINUED: the last or end calculation. It's procedural meaning is a calculation that forms a basis for construction. A final calculation may be revised due to changes in design criteria input parameters, or construction changes. The term " final" is procedurally synomonous with the term " confirmed" as defined in paragraph 8.2 of EDP 4.37, Revision 4 and paragraph 2.1 of EDPI 4.37.1. These calculations form a complete basis for construction. Although there may be some overlap between the calculations, they are not contradictory and will be consolidated at time of document turnover to PECo. O o Item 2 of the DDR addresses calculations 11-B, 11-B-1 and 11-B-2 which include the design of penetration X-45B. Originially Calculation 11-B Revision 9 was transmitted to SWEC. This calculation is a generic design calculation for penetrations based upon conservatively assumed loads and initial system design data (committed calculation). Inadvertently, the transmittal included copies of sheets 60 through 185 from Revision 8 of the calculation. These sheets 'are identified as removed on the cover sheet of Revision 9, and are in fact removed from the original calculation. o The response to DAI-319 acknowledged that sheets 60 to 185 were superseded and were inadvertently transmitted. The response identified sheets 9 to 12 of Calculation 11-B as applicable to penetration X-45B. After review of these sheets, SWEC initiated DAI-354. The response to this DAI did not identify sheets 9-12 as inapplicable. Rather, these pages were described as part of the generic Calculation 11-B which was O identified as based on assumed loads. Page y of 7 i
) FEB-07 '89 10:33 ID PECO PROJECT MGT DIV TEL NO:215-941-4578, n299 PO4
. . :p -
i p' '
= _=- _ . _ . _ . . =
l ,t j O'
-- _ p '5 m ruaZurtRrno ConioR1 Trex L.IntRICK.2 IDCA . .
(2.5. NO. 18138) OBSERVATION REPDAT Observatten Report No. 00R-27 Re's e e:. .. . u _ , __ _ ,,_,1 ueni -
.um. ye n ,
** 1 .
^
The ensponse specifically identified-M ' ' en&oulettene based on final loads ior Unit t { se Cm1 = 1 =* i a=a 11-5-1, for pentration 45-5,
^
med smLonLasten 11-5-2. der penetrations X-leMS. The Unit i saleulations vers, peewided stase, ma stated in the response,
, . . , , , vertS&ention af Unit 2 penetrations, based on
.,;. . dLeak Leeds, was set exposted to be complete ,.'
.s. 2. setta, the one of 1988. In iact, the Unit 2
_. t .: up..Eani&am&ausse including thome for penetratten ,. ., N=49-t see sercomplete.' .....,.y..,,,
. ::: .,-m; .
- .Es emosesy Calculation 11-5 1s a sommittee
.satsuletten for Unit 1 and 2 based en assumed loads. Osloulations 11-5-1 and 11-5-2 are O'- -
final based on salsulations for Unit i penetrations final loads, similar calculations
.. have been sempleted for'. Unit 2.
e Ites 3 of the DDR has not been addressed
?
* -. earlier.in a DAI.* . It relates to multiple 7 ,omiculations for RNR heat enchanger supports.
In this case calculation 22.43 was the original design document, marked final based upon confirmed input. Calculation 181.73 was subsequent'y made to recono11e revised vender los.de on the same RNR heat exchanger supports destyned in Caloulatten 22.45 and was aise marked final. As discussed under item 1, multiple final calculation may be used for d&fferent parts of a design. This osos confirms the point.
....... . . . . ..........................sq=...................
& ' rt.N.
'PEco R M ew Signaturs/Date O tl!W K$lYh ] }M Iho{$9 Bechtel Approv'al sigasture/D'ate Page ,J.,,,of 7
' :. { , . ..'.J-
- :.:m.,: .-. .,f
' m;p. .' 3;;
. .e 4 . , ,,
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) m OBSERVATION REPORT Observation Report LOR- 027 - Rev. L Review Plan LK- D - 1903 -S Rev. 1 PART III - RESPONSE EVALUATION The response to this observation report has been evaluated as follows: A. Causal Factor (s,: Acceptable Not Acceptable (Explain): Not Applicable B. Extent of Condition: Acceptable Not Acceptable (Explain): Not Applicable C. Significance: . Acceptable Not Acceptable (Explain): ( Not Applicable D. Corgtective Action: Acceptable Not Acceptable (Explain): Not Applicable E. Action to Prevent Recurrences Acceptable Not Acceptable (Explain): l Not Applicable l Additional Action Required: y No Yes (Explain) ) See Attached Page OMfL--------------- ----------------------- ---------------------------------------------- Lead Engineer Signature 2/4//1
/~Date Q M ' L 0. w iI k Alt Signature
- t/n/t1
/ Date l Page k of 1 l
[ _ _ - _ - - - _
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) \ t Q OBSERVATION REPORT Observation Report DOR- 027 Rev. O PART III - RESPON8E ETAIAATION CONTINUATION: The response states that there are sufficient means for ' knowledgeable' engineers assigned to the project to identify and effectively use applicable structural calculations. By this, it is assumed that Bechtel means engineers who are familiar with the history of the project, not engineers who are
' knowledgeable' in structural engineering. It is not clear that a
' knowledgeable' engineer could find the applicable calculations by himself without help from someone involved with the project. This much is implied in the response by the statement ' implementation of cross-referencing enhancement will be determined at the time of document turnover. '
Although there do not appear to be clear methods to identify applicable calculations, Bechtel engineers are able to identify the applicable calculations if given enough time, and consequently there is no concern related to the design adequacy of the plant because of this situation. During the IDA review, no instances were noted where there were not sufficient calculations to justify the design. O 1 V Page 7 of 7
l l STONE (m WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) (,) ossurvATIoK axPonT Observation Report No. DOR- 028 Rev. ,0, Review Plan: LK- J _ _1gj L un Rev. ,1_ Reference AI No.: DAI-306 FART I - INITIATIoK
- 1. Description of Concern Stress calculation 2-10-83 does not demonstrate the adequacy of all the integral attachments within the stress problem boundaries.
- 2. Supporting Information The local stress evaluation for pipe support CBB-218-H58 was performed using the ME-210 Computer Program. The moment term ML was input in the computer program as f t-lbs, when the proper input was in-lbs.
This resulted in an unconservative calculation for the strees at the attachment point. No local stress evaluation was performed for the stanchion to elbow gO connection at support CBB-203-H12. As a result of these errors, the technical adequacy of the integral attachments is not sufficiently demonstrated in this calculation.
....................................................==.. ..==..========================
Additional Documents Attached: None p ................................... .................... N 90kdM n/ts/tr<. /2//C)bl? Lead Engineer Signature /~Date APM S % nature
'/ Date Page 1 of 1
STONE & WEBSTER ENGINEERING CORPORATION (,_) LIMERICK 2 IDCA (J.O. No. 18136) hhh V[ f,
.C/ OBSERVATION REPORT M I 9 ISO 9 Observation Report No. DDR-028 Rev. O PART II - RESPONSE
- 1. Observation Concurrence X Concur with observation Do not concur with obs3rvation (Note: if not in concurrence, explain in " remarks" below)
- 2. Response to Observation:
A. Causal Factorf1) (a) The local pipe stress evaluation for an integral attachment of pipe j support GBB-218-H58 was performed using the ME-210 Ccunputer Program. h e units for moment ML was input in donputer program as f t-lbs. instead of in-lbs. The analyst inadvertently selected the wrorg units for M L. g- This was an oversight by the analyst and the checker. i (b) A local stress evaluation was not documented for the stanchion attach-ment of pipe support GBB-203-H12 to an elbow. Pipe Support GBB-203-H12 was identified as being relocated during the as-built reconciliation program. Undocumented judgement was exercised about the acceptance of local piping stresses at the integral attachment, based on the fact that the pipe support is only a gravity support with a lubrite plate between the stanchion and a sliding surface which would generate minimal moments at the welded attachment, thus producing relatively negilgible local stress. B. Extent of Cordition To confirm that this is not a recurring error, a representative survey of 29 large pipe Seismic Category 1 strees calculations were reviewed, out of a total population of 139, to check for:
- 1) Correct units for input values to ME-210 local stress calculations.
- 2) Documented evidence that the technical adequacy of integral attachments was sufficiently demonstrated in the calculation.
======================================================================================
Remarks: O V Page a of 4
. AN-25 '89 10 ?"' ID PECO PROJECT MGT DIV TEL NO 215-841-4579 s165 P02 P,F3
'evi 19 '9912:55 KC4EL LESTRN POL.ER iST PLOM O S70NC & IEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.C. No. 18138) hd f4- [ [
. OBSERVATION REPOE Observation Report.' No. DDR-028 Rev. O PAR II_ - RESPONSE CORR 1NUED The survey consisted of a review of every fifth stress calculation based on a database printout of stress calculations sorted in an ascending order.
It was determined that all integral welded attachments have been correctly evaluated for the local stress effects and are within acceptable stress limits. It was also determined that ME-210 input- values used the proper units (in-lbs.). In conclusion, the referenced stress calculation and 29 other arbitrarily selected stress calculations have been revicwed for technical adequacy of the integral attachments, and it has Deen determined that all reviewed stress calculations have correct usage of the ME-210 computer program and adequate documentation for integral pipe attachments. C. Significance
'Ihe integral attachments identified in this Observation Report were re-evaluated and detemined acceptable. A representative survey did not identify O any other discrepancy. "Ihe concern identified in the Observation Report is j en isolated occurrence. The concern is not considered significant as l discussed in correction action Delow. ]
D. Corrective Action
- I
'Ihe localized effects due to pipe support GBB-218-M58 has been re-evaluated I by using correct units for the moment M t,, and it has been determined I that the piping stresses at this location are still ws. thin all allowable {
limits and are acceptable. The localised offacts due to pipe support GBB-203-H12 have been evaluated, and it has been determined that the piping stresses at this location are well within allowable limits. '!he remaining pipe supports in Calculation 2-10-83 with integral welded attachments had been properly documented in the calculation for their localized stress effects and are acceptable. Stress Calculation 2-10-83 has been revised to include tt.e evaluation of local stress effects due to pipe integral attachments of pipe tapports GBB-218-H58 and GBB-203-H12. I l'. Action to Prevent Recurrence Based on the fact that the concern is not significant and is not generic, it has been detemined that no further action is required. lO -- N/ Asc=~ 4 /19 4 -- b b& lMi destes PEco Review Sign 4ture /Datle Becntal Approval Signature /Date Page l of 1
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) l f~h
. OBSERVATION REPORT Observation Report D OR- 028 Rev. 1 Review Plan: LK- D -_1903 A 1 Rev.1
_PART III - RESPONSE IVALUATION The response to this observation report; has been evaluated as follows: A. Causal Factor (s): ,,,_1_Ac ce p table Not Acceptable (Explain): Engineering oversight and undocumented engineering judgment are listed as i the causal factors. B. Extent of Conditions X Acceptable Not Acceptable (Explain): Based on the conclusive' findings of Bechtel's sample review (20% of SEISMIC CAT 1 Stress Calculations selected arbitrarily), it is reasonable to conclude that these concerns were limited to the stress calculation in question. n C. Significance X Acceptable Not Acceptable (Explain): The response states that the revised calculation for 2-10-83 now demonstrates ! the adequacy of the integral welded attachments. Revised calculations have not been reviewed. D. Corrective Action X' Acceptable Not Acceptable (Explain): The affected calculation has been revised to incorporate the corrections / additions for local stress evaluation. Although the calculation has not been reviewed, the existing Bechtel procedures should provide sufficient guidance to e'nsure acceptable design. E. Action to Prevent Recurrence X Acceptable Not Acceptable (Explain): Since the concern is not generic, it is reasonable to conclude that the problem will not recur.
...................................................................n.....==============
Additional Action Required: X No Yes (Explain) Lead Engineer Signature / Date APM Signature V / Date Page ,5{, of 1
STONE'&, WEBSTER ENGINEERING CORPORATION
-LIMERICK 2 IDCA (J.O. NO. 18138)
L,O OBSERVATION RFN RT V l
- Observation Report No. DOR- 029 Rev. 0_
Review Plan LK- D - 1906 - MS Rev. L Reference AI No.: DAI-220. Rev. O PART I - INITIATION l '. Description of Concern There are deficiencies in the RHRSW Pump Specification'(No. 8031-M-12, Rev. 4) that could lead to possible design problems.
- 2. Supporting Information
- a. The pump data sheet does not require the NPSH test for the RHRSW pump. However, the pump data sheet shows that the manufacturer.
has indicated a required submergence of three feet. The NPSH/ Submergence . curves used in calculation No. M-12-30, Rev. 0 (Sheet ~ 5); shows that the submergence requirement is only applicable up to a flow of 10,600 gpm. The required pump NPSH must be _ met beyond this flow. The pump system curves on sheet 17 of the calculation indicate that ' the system flow . is more than 10,600 gpm in certain modes. Also, the NPSH test is required by the Hydraulic Institute Standards. (Ref. 2.0.b.(3) of the specification).
'b. The maximum particle ' size of suspended solids is not specified.
The throttle ' bushing with spital grooves to prevent suspended solids from reaching the shaft seal area and the pump's espability
.to handle 1 inch size particles mentioned - in the action item response are not indicated in the pump vendor drawings / Operation and Maintenance Instructions.
samsmemmassssammassummmmmmmmmmmmmmmmmmmuasammmmmassammassammmmmmama ssassam a ssmusssamass Additional Documents Attsched: None mass s'ammmasamosaammmmmmu msmespessam mam e asummusemasspysmsmana gns . muasssssssssssssssa O su l I AltA h Le(d Eiigin'eer Signature'
/ Date APH Sig6ature
/ Date Page 1 of 1
STONE & WEBSTER ENGINEERING CORPORATION O LIMERICK 2 IDCA (J.O. No.18138) g , g gg9 ~ [g ,' p p OBSERVATION REPORT-Observation Report No. _JL' OR _Q29_ Rev. O PART II - RESPONSE
- 1. Observation Concurrence:
Concur.with observation. _JJL. Do not concur with observation (Note: if not in. concurrence, explain in.
- remarks" below)
- 2. Response to Observation:
A. Causal Factor (s) B. Extent of Condition C. Significance
\ D. Corrective Action E. Action to Prevent Recurrence
=====,. =============================================================================
Remarks We do not concur with the concern stated in DOR-032. The bases for this statement are given below.
-a. As indicated on sheet 6 of Calculation H-12-30, the minimum NPSH available for the worst case (i.e., maximum RHRSW flow) is 41.2 feet. This corresponds to a flow rate of 11,200 gpm at which the RHRSW pump can be operated and still meet j the required NPSH (reference sheet 5 of the calculation). Sheet 17 of the calculation shows that the mani.i.um possible RHRSW flow is 11,150 gpm. This flow is slightly less than 11,200 gpm and, therefore, the NPSH criterion will be satisfied.
We have reviewed the Hydraulic Institute. Standards (HIS) and find that the Standards only provide the basis for the determination of NPSH and do not mandate an NPSH test. Our experience is that NPSH tests are usually mandatory only when evaluating a new pump model. I Iage _d of 1, 1
STONE & WE88TER EN0!NEERING CORPORATION
. - . . . i,g ggRieg. a...g pex ._ .
> . -- 10CA /7 h. -
(J.0.'NO. 18138) ; ossERVAT!0N REPott observatten Report No. A on ,,ggg.,Rev ,,p,_ l FAk? !! - REEPOMat 00MimitD
- b. .The spiral grooved throttle bushing la ites 88 en tyren Jackean drawing 1-F-satt, Revielen A, and the Materlate of Construction list, page 3, both a part of the installation, operatten and maintenance manual (vender print 8081-N 12-61-8BR, transmitted to sWEC vis 80064). The pump's sapability of handling one inch sine particles was discussed and documented in a telecen between nechtel and Byron Jackson en November ti, 1988. This document was transmitted to sWEC via 30554. No revielen to either the pump vender drawings or operation and l instruction manual is required to further document this capability, saced on the condittens cited above, we believe the specification to be adeguate and its existence in the present form would not lead to possible design problems.
I I 1 O 4 1 l l
. skim e,(P2 1,L /m tr i M A hCd&d R.vi. si natur. / o. .
/Qtviappr.v.isiinatur.<
... 2 of .
l i STONE & WEBSTER ENGINEERING CORPORATION O LIMERICK 2 IDCA
/j
\ (J.O. NO. 18138)
OBSERVATION REPORT Observation Report _3_OR- 029 Rev. L Review Plan: LK- D - 1906 -MS Rev. L 1 PART III - RESPONSE EVALUATION The response to this observation report has been evaluated as follows: A. Causal Factor (s): Acceptable Not Acceptable (Explain): , NOT APPLICABLE 1 B. Extent of Condition: _ Acceptable Not Acceptable (Explain): NOT APPLICABLE 1 C. Significance: Acceptable Not Acceptable (Explain): NOT APPLICABLE D. Corrective Action: Acceptable Not Acceptable (Explain): NOT APPLICABLE E. Action to Prevent Recurrence: Acceptable Not Acceptable (Explain): j NOT APPLICABLE
.............................................................=====.......=..=.======..=
Additional Action Required: y No Yes (Explain) SEE INSERT A 1 i O . . ..
.....L...................................................................
- nh 2lls]P9 C % d te 6 0. W ili e Alki Signature zlu.lc1
/ Date L(ad Engin'ee} 'Sig'nhture / Da'te i
Page _L{_ of g_
.,m INSERT A
- a. Bechtel's response to Observation Report DOR-029, . Rev. O demonstrates that the lack of an NPSH test for the RHRSW pump is not a significant concern as the NPSH criteria is satisfied for the worst case. It is also noted from Item d of Bechtel's response to Observation Report DOR-032, Rev. O, dated February 8, 1989 that the NPSH curve applicable to the Limerick two-stage pump has been obtained from the vendor and the curve is the same as the catalog data used in Calculation Number M-12-30, Rev. O.
- b. The RHRSW pump's capability to handle 'one inch size particles is not a significant concern as the same is confirmed by Byron Jackson (pump manufacturers) in a telecon with Bechtel, dated November 11, 1988.
1 1 O DOR - 0.19 R EV_.o PAGE F OF S
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) m
) OBSERVATION REPORT Observation Report No. DOR- 030 Rev. 0,_
Review Plan: LK- D- 1901 - MS Rev . _0_ Reference AI No.: DAI-014. Rev. O DAI-208. Rev. O PART I - INITIATION
- 1. Description of Concern The FSAR contains text and tables which have been found inaccurate. These inaccuracies may result in design errors.
- 2. Supporting Information '
\
Following are examples of noted inaccuracies found during the reviews
- 1. Table 6.2-6 consistently uses 10,000 CPM for the RHR pun:p flow rate. RHR flow rates depends on the operating mode of the individual pumps and may be up to 11,000 GPM. (FSAR Figure 5.4-14)
- 2. The typical NPSH calculations in section 6.3.2.2.4.1 uses rounded
( ) off values for atmospheric pressure whicH are inconsistent with V the values used for atmospheric pressure , for NPSH calculations in sections 6.3.2.2.1.1 and 6.3.2.2.1.2.
- 3. FSAR Sections 6.2.4.3.1.4, compliance with General Design Criteria 57, includes a discussion on the Control Rod Drive Lines.
However, Table 6.2-17, page 5 for penetrations X-37A-D and X-38A-D (CRD Lines insert and withdrawal) indicate the applicable general design criteria is GDC-55. The body of the FSAR and the data in Table 6.2-17 appear inconsistent for these penetrations. This item will be corrected by an LDCN in accordance with the response to DAI-208.
.........................................................................== ..=...=====
Additional Documents Attached: None l
...s.. ...................
o .. ,.......... ................... .................... .. my h) jf, l l S Le'ad Erfgife'ETISignatbre '/ Date APM 'Sig;{ature / / Date Page 1 of I
l STONE & WEBSTER ENGINEERING COR OR T O 1 . LINERICK 2 IDCA , gg7 (J.O. No. 18138) I p) r OBSERVATION REPORT Observation Report No. D OR- 030 Rev. O PART II - RESPONSE
- 1. Observation Concurrence:
Concur with observation
.JUL Do not concur with observation (Note: if not in concurrence, explain in
" remarks" below)
- 2. Response to Observation:
A. Causal Factor (s) B. Extent of Oondition C. Significance D. Corrective Action E. Action to Prevent Recurrence Remarks AMENDED RESPONSE The following is an amended response prepared in accordance with agreements reached between SWEC and Bechtel during a telephone conference on February 28, 1989.
- 1. The RHR flow rates depend on the operating mode of the individual pumps and may be up to 11,000 gpm; however, the data shown in Table 6.2-6 for Cases A, B, and C are all for the same RHR long term containment cooling mode. The data in Table 6.2-6 are presented for three variations of Mode B (RHR pumps aligned to flow through the heat exchanger) which are described in FSAR Section 6. 2.1.1. 3 . 3 .1. 6.
For Mode B the flow rate is 10,000 gpm which is the value depicted in Table 6.2-
- 6. The discussion in FSAR Section 6. 2.1.1. 3. 3.1. 6 is discussing the containment heat removal capability of the RHR system. Consequently, in this section the dedicated LPCI pumps and the other RHR pumps are referred to separately. The l pumps that are aligned to flow through the heat exchanger are referred to as the RHR pumps (10,000 gpm) and those injecting directly to the RPV as the LPCI pumps (11,000 gpm) . It is only the pumps that are aligned to flow through the heat exchangers that are referred to as RHR pumps and included in Table 6.2-6.
- 2. The RHR system has many operating modes and it would be excessive to show NPSH p calculations for all of these. Consequently, the FSAR shows only two typical 4 NPSH calculations for the RHR system in Section 6.3.2.2.4.1. Because these calculations are only typical, as stated at the beginning of this section in the FSAR, the numbers for the head equivalent of the atmospheric pressure above the suppression pool are shown with a precision of only two significant figures.
Page _a_ of I
2pcd /8W I sTONI & WsSBTER EN0!NEtt!NG CORP 0 RATION LIMERIM 2 IDCA I - (J.O. NO. 18130) OSSERVATIM REPORT Observation Report No. ,,L, OR ,,,gELRev. 9 PART !! - RESPONRE 00rrIMI5D The values are securate, however, to the two significant figures shown when the standard atmospheric presaura is converted to feet of head at the temperature of the water in the suppression pool. The values for atmospheric head shown in sections 8.3.3.3.1.1 and 6. 3. 2. 3. i . 3 are shown to three significant figures becaves these are for the NPCI system which has only the two operating modes shown, and these are the actual values which were obtained from the design calculations. Note, that in the introduction to this section the PSAR does not describe these calculations as typical,
- s. The inconsistency cited in the DOR could not lead to design errors since the PSAR reflects other design basis documents and evaluations. Nevertheless, the PSAR has been revised to clarify the editorial inconsistency between paragraph 6.g.4.3.1.4 and Table 6.p717. Although the design of the CAD insert and withdraw line penetrations represents a justified departure from the esplicit requirements of the 800, this discussion is more appropriately included under the subheaaing.
O Evaluation Against criterion S5 (consistent with Table 4.g-17). gince the time the FSAR was initiated. evolving interpretation of the GDC applicable to the subject penetrations were not translated to the FAAR. To ensure that the potential for F8AR inconsistencies is minlaised, measures were taken in 1988 to place greater emphasis on the design as described in the 78AR. These measures included procedural enhancements, additional training and designatten of an individual in each engineering discipline to be responsible for the accuracy of the F3AR. A complete comparison of the 000 reflected in the test and table has identified no sir,ilar occurrences of this type, conclusion With regard to the esamples cited in the DOR one can see the FsAR has provided values and information to generally describe the subject design. Use of this information could not result in design errors since in these cases the values and critoria presented in the FSAR reflect aspects of other design basis . documents or evaluations which preceded the F8AR and appropriately describe the design in grenier detan. . IiL h a isq hlsMMArt 1649 PEcokeviewsignatbre / Data secht i Approv 8 signature / Date
- v. ,e 2. ef s
I STONE & WEBSTER ENGINEERING CORPORATION 1 LIMERICK 2 IDCA (J.O. NO. 18138)
- A.
OBSERVATION REPORT Observation Report D OR- 030 Rev.O_ Review Plan: LK- D - 1901 - MS Rev. 0_ PART III - RESPONSE EVALUATION The response to this observation report has been evaluated as follows: A. Causal Factor (s): Acceptable Not Acceptable (Explain): NOT APPLICABLE B. Extent of Condition: Acceptable Not Acceptable (Explain): NOT APPLICABLE C. Significance: Acceptable Not Acceptable (Explain): I NOT APPLICABLE ( D. Corrective Action: Acceptable Not Acceptable (Explain): NOT APPLICABLE E. Action to Prevent Recurrence: Acceptable Not Acceptable (Explain): NOT APPLICABLE
.............v.........................................................................
Additional Action Required: X No Yes (Explain) SEE ATTACHED PAGE l . .... . ............................................................................. Lead Eng'iti'eer Signature
, z# M3 6~IDate
~
QCAM & D.Wille_. 3/zv/r7
/ Date
/ MM lignature Page 1 of f L____ _ _-_ _ - - - - - - - - - - -
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REP 027 Observation Report DOR- 030 Rev. O PART III - RESPONSE ETAIEATION CONTLWUATION: The response addresses the details of the concerns and provides sufficient information for clarifications, corrective actions, and whether they were significant cases. The response is acceptable. ITEM 1: The response clarified that it is only pumps aligned to flow through the heat exchangers that are considered as RHP. pumps in FSAR Table 6.2-6. ITEM 2: The response justifies the degree of accuracy based upon the intended usage as " typical". ITEM 3: An FSAR LDCN has been initiated to correct the concern. Comparison of text and table on CDC has identified no similar occurrence of this type. O Pa3e s et c
1 STONE & WESSTER ENGINEERING CORPORATION LIMERICK 2 IDCA p (J.O. NO. 18138) OBSERVATION REPORT Observation Report No. DOR- 031 Rev. O Review Plan LK- D- 1901 - MS Rev. 1 Reference AI No.: DAI-014 Rev. O PART I - IIITIATION
- 1. Description of Concern The design of the Residual Heat Removal System (RHR) pumps has not been demonstrated _'to comply with the requirements of Regulatory Guide 1.1, Net Positive Suction Head for Emergency Core Cooling and Containment Heat Removal Pumps, in accordance with the FSAR licensing commitment of FSAR Section 1.8.
- 2. Supporting Information p Regulatory Guide 1.1 requires that operability of the ECCS and containment heat removal pumps be demonstrated under conditions of maximum expected suppression pool temperatures without the benefit of considering pressurization of the containment. The plant unique analysis indicates a suppression pool temperature of 212.5*F may be achieved under accident conditions (FSAR Table 6.2-6). ' 'Ihe design calculations reviewed did not contain an analysis of the pumps NPSH available at this temperature for all RHR modes of operation required for mitigating an accident. All accident modes should be evaluated at the highest expected temperature because operator action determines the operating mode durations during which time the suppression pool may reach the maximum expected temperature. Also, the typical calculations in the FSAR for RHR NPSH during accident modes do not use this maximum expected temperature, reference FSAR Section 6.3.2.2.4.1.
Additional Documents Attached: None
. ....... ... ......................................... ... ...,n.. .................
77 Al4m ld $ ~ Lehd E[gifieeYSignatdfe / Date APM Sigr(ature ' '
/ Date Page 1 of [
L-_-_-_________-_______
STONE & WEBSTER ENGINEERING CORPORATION
.'7 " b C A N W LIMERICK 2 IDCA $- 8 Nb (J.O. No. 18138) l G
) OBSERVATION REPORT MAR 1 0 1989 I
l 1 Observation Report No. D OR _931_Rev. O PART II - RESPONSE !
- 1. Observation Concurrence:
Concur with observation
,JK._ Do not concur with observation (Note: if not in concurrence, explain in
" remarks" below)
- 2. Response to observation:
A. Causal Factor (s) B. Extent of Condition C. Significance D. Corrective Action E. Action tciPrevent Recurrence U .................................................................................... Remarks
. AMENDED RESPONSE The following is an amended response which has been prepared in accordance with agreement reached between SWEC and Bechtel during a telephone conference on February 28,- and March 1,1989.
Regulatory Guide 1.1 requires that " Emergency core cooling and containment heat removal systems should be desianed [ emphasis added) so that adequate net positive suction head (NPSH) is provided to system pumps assuming maximum expected temperatures of pumped l fluids and no increase in containment pressure from that present prior to postulated loss of coolant accidents." The Limerick design meets these requirements for all operating modes of the RHR system as explained below. The maximum suppression pool temperature of 212.5 F shown for Case C in FSAR Table 6.2-6 cccurs at a time when the space above the suppression pool is pressurized to 16.7 psig as is shown in this same table. However, no credit is taken for this pressure in the NPSH calculation and it is the engineer's judgement the maximum suppression pool temperature of 212 F is the appropriate temperature to use '; for NPSH calculations considering the requirements of Regulatory Guide 1.1. l Consequently, this is the temperature that is used for the NPSH calculation shown for Case 2 in FSAR Section 6.3.2.2.4.1. In any event, the change in available NPSH due to this 0.5 F difference in temperature is insignificant compared to the 22.1 feet of available NPSH shown in the calculation. Page 2_ of f_
1
.zoca/8 VT sTONI & W ASTER EWo!NEERINO 00RPORATICH 1.!NtatcK t spCA l (J.0. No. 18188) -
658tkVATION REPonT Observation Report No. IL, OR g3L.Rev. sL, PART !! - RERP0 Mat omff!MURD _. Because NPSN is a function only of the pump inlet conditions it is not necessary to 4 calculate the available NP8M for all the operating modes of the RNR system that differ only on the discharge side of the pump. It is only necessary to calculate the NP8H for inlet flow conditions that envelope all operating modes of the system. Consequently, only two calculations are shown in FSAR Section 6.8.t.2.4.1 to demonstrate that the Limerick design meets the requirements of Regulatory Guide 1.1 for all operating modes of the RNA system. For cases 8 and C in Table 6.2-4 where the temperatures are shown as 311.8' F and tit 8' F, the NPSN Case t in FSAR Section 6.4.2.t.4.1 shows the NPSN available. In cases B and C the flew from one RNR pump is through the heat enchanger and one LPCI pump injoets flow directly into the RPV. The design flow rate through the heat exchanger for the containment heat removal mode la 10,000 sps. Calculation 8081-H-51-8 (transmitted to SWEC by transmittal No. 80032) shows, and the PSAR states in Sostion 6.8.8.2.&, that there is sufficient system resistance te preclude RNR pump runout in the I.PCI mode. Consequently. Case 2 in Section 6.8.2.2.4.1 shows the NP8N available for this RNR mode of operation at a j suppression pool temperature of tit' F. The maximum flow rSte for any mode of the RNR system is 11.000 3pm as shown for Mode A-2 in FSAR Figure 5.4-14. For this modo the maximum temperature is shown as 180' F. Hence the NP8N for this mode is covered by case i in F8AR Section 6.8.t.3.4.i. Case t in FSAR 8ection 4.4.2.t.4.1,can also be used to show the NPSM available at the maximum flev rate of 11.000 spa and tit' F. Because the head less due to flow is I proportional to the square of the flow, a 10% increase in flow rate from 10.000 spm to 11.000 3pm will only increase the flow h ud loss by 1.58 feet from 7.54 feet to 9.12 feet which contributes an insignificant decrease in the total NPSN available as shown in calculation 4031-Nisc=65, Rev. 0.(transmitted to Swtc via transmittal 50666). Documentary evidence that the ECC8 pumps meet the requirements of Regulatory Guide 1.1 is provided in Calculation 8081 MIBC-66. This calculatten summartres the available NPSK margin for all the 20C8 pumps at the maximum operating condition of each system. In addition. an i.DCN which was prepared for other reasons, but happened to involve the P8AR sections that presented the Np4N calculations, is in process and incorporates the NP8N values as shown in Calculation Sosi-N!80-65. .
..............e s s ema s s e s s e n e s s m e n e.....s e ...... m e s e s s e s s e s s e s s m e n e s e s s e n.
...s m s h .3J, h h ffn I /Mfl I . H .
PEco Review Bignature / Date techtel Approval Signature / Date Page d,of [
STONE & k'EBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA i (J.O. NO. 18138) j { OBSERVATION REPORT Observation Report D OR- 031 Rev. O_ Review Plan: LK- D - 1901 - MS Rev. O PART III - RESPONSE EVALUf.? ION J- _The response to this observation report has been evaluated as follows: A. Causal Factor (s): Acceptable Not Acceptable (Explain): NOT APPLICABLE B. Extent of Condition: Acceptaole Not Acceptable (Explain): NOT APPLICABLE C. Significance: Acceptable Not Acceptable (Explain): NOT APPLICABLE D. Corrective Action: Acceptable Not Acceptable (Explain): NOT APPLICABLE E. Action to Prevent Recurrence: Acceptable Not Acceptable (Explain): NOT APPLICABLE
.....................................................................................=.
Additional Action Required: X No Yes (Explain) INSERT A A .
....E................................u....................................
vah-- Lea'd Egineer Bignature ElnM '
/ Dade (Y1 x & o. d e sh4n AYM signature / Date Page 4 of [
_ _ = _ - _ _ _ - _
r l STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT Observation Report D,0R- 031 Rev. O PART III - RESPONSE EVALUATION CONTINUATION: INSERT A The concern that the RHR system design . did not demonstrate compliance with Regulatory Guide 1.1 has been resoli-ed. Bechtel has issued Calculation 8031-MISC-65 which documents the engineering judgements that the RHR systems meet the conditions 'mposed by Regulatory Guide 1.1. Additionally, an FSAR Change LDCN has been initiated to resolve the concern. O O Prie E of f
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA l (J.O. NO. 18138) l OBSERVATION REPORT Observation Report No. DOR- 032 Rev. 0_ Review Plan: LK- D - 1903 - MS Rev. L Reference AI No.: DAI-024 Rev. O DAI-025 Rev. O PART I - INITIATION
- 1. Description of Concern There are deficiencies in the RHRSW Pump NPSH Calculation (No.
M-12-30, Rev. 0) that could lead to possible design problems.
- 2. Supporting Information
- a. The basis for a three feet drop in pond level for the spray pond winter bypass operation is provided by engineering judgement. The maximum level drop is critical, using a three feet drop, the difference between the available NPSH and the required NPSH is only about one foot. Therefore, the available NPSH may not meet the pump required NPSH if the level in the pond decreases more than three feet. Also, if the operator
-_. switches from the winter bypass mode to the spray mode based
[%/ upon a pond temperature, the pond level may decrease before the pond temperature increases to the limit that would require the operator to switch to the spray mode. The predicted pond level at the time of switchover frc,a the winter bypass mode to the spray mode is not provided. Therefore, more than engineering judgment is required to establish a pond level decrease of three feet.
- b. Reference for the " minimum initial pond water level" is not provided in calculation No. M-12-30 Rev. O. In accordance with FSAR Table 9.2-9, the storage volume of the spray pond is 29.60 X 106 gallons and not 28.92 X 106 gallons as stated in the action item response. Also, the minimum pond level should be calculated using the normal minimum storage volume and not the storage (design) volume. Conversion of the minimum storage volume to pond level is also not addressed in calculation No. M-12-26.
CONTINUED ON 1 PAGE
..............................................................=========.===============
Additional Documents Attached: None tw .. ......... . ...............,..................... . ...... .............. ... llpg l j) l
/'Date~ APM Sig/ature l
l/$
' / Date Lea'd En[infr'Sigitdtured Page 1 of i
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT Observation Report No. DOR- 032 Rev. O PART 1 - INITIATION CONTINUATION:
- c. The calculation conclusion is not supported by a comparison of NPSH required versus NPSH available. Tr.e NPSHA of 37.18' mentioned in the action item response is for the post-LOCA l' mode. The calculated system flow for this mode is 10,700 gpm and the corresponding NPSHR from the vendor curve is 35'.
Therefore, a comparison between 37.18' and 26' (at design flow of 9,000 gpm) is not correct. A ccmparison of NPSH available versus NPSH required should be made in the calculation for various modes of pump operation at the calculated system flows and not the design flow.
- d. A catalog data sheet for a single stage pump is used in the calculation, whereas the RHRSW pump is a two stage pump.
According to the action item response, pump NPSH requirements for single - stage or two stages are the same, provided the impellers are the same. However, the single stage pump catalog data used in the calculation is not certified by the vendor to be applicable to the Limerick two stage pumps. 8 O l Page 3. o f h
STONE & WEBSTER ENGINEERING CORPORATION ff)g 7 g p LIMERICK 2 IDCA (J.O. No. 18138) @ . g lg39 l f OBSERVATION REPORT
$~Oh5f)
Q) observation Report No. D OR- 032 Rev. O PART II - RESPONSE
- 1. Observation Concurrence:
Concur with observation _IL. Do not concur with observation (Note: if not in concurrence, explain in
" remarks" below)
- 2. Response to Observation:
A. Causal Factor (s) B. Extent of Condition C. Significance D. Corrective Action E. Action to Prevent Recurrence 3 j ssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssmussssssssssas Remarks
- a. A three feet drop in pond level for the spray pond winter bypass operation is a conservative assumptio6. The conditions which cause a drop in pond level are the design basis freezing which produces a maximum of 15 inches of-ice on the pond surface (FSAR 9.2.6.5.3) and the water volume needed to fill the RHRSW system piping. We have estimated that it takes about 225,000 gallons of water to fill- the RHRSW system piping, including the RHRSW side of the RHR heat exchanger. This amount of water will cause the water level in the spray pond to drop by about an inch. Note that the RHRSW system piping is normally filled.
with the exception of those portions of the spray network which are located above ground. If the operator switches from the winter bypass mode to the spray mode, the pond level will decrease slightly in order to fill the drained spray network. Since it takes approximately one inch of pond level to fill the entire RHRSW system, the pond level decrease during the switchover is negligible, but certainly less than one inch. Note that with the switchover to the spray mode the RHRSW pump will operate at a lower flow due to higher system resistance through the spray piping network. The NPSH required will therefore be less. There are no other identifiable conditions that could reduce the spray pond water level more than about one inch when operating in the bypass mode. Fond level decrease due to evaporative losses should be negligible. Hence, the leve! l in the pond will not decrease any further and the three foot assumption is l O. conservative. l ( Page 3 of 0
I STONE & WESSTER ENGINEERING CORPORATION J d ( // / 7 5 0 LINER!ct 2 IDCA . (~J.C. N0 J 18184) OBSERVATION REPORT i J ! Observatten Report Nr. .} OR Atl. Re v ..JL PART If - RERPDMR5 COWTINUED
- b. Sheet & of Calculation H-12-80 provides a reference; 1.e.. Technical Specifi- l sation 3.'r.1.3.a. which shows a minimum pond water level at elevatten 250' for single-unit operation. This value is currently being revised to 180' 10' for ,
two-unit operation. l
. In ew- BA!-Ote action item response, we refer to FSAR Table 9.2-t3 (note the reference to P8AR Table 9.t=8 was a typographical error) which shows the minimum storage volume of to.92 s 10' gallone for two-unit operation. This volume sorroepends to a pond level that closely matches the two-unit minimum initial level used in calculation H-it-80.
calculation N-12-25 calculated the pond volume for five different pond levels. The minimum storage volume een be converted to pond level by interpolation between the rive calculated levels. j
- c. A comparison of NpSW required versus NP8W available was performed in calculat19n N-12-80. However, it was done by other means, as discussed below. Calculation-H-12-30 concluded that the maximum RHR$W flows are below the allowable flows for O various modes of pump operation. The allowable flow was determined by first calculating the NP8H available and using this WP5H available to obtain the q
J correspond!n flow by using the vender NP8M required curve. Hence, a comparison ] of maximum low versus allowable flow is the same as a comparhon of NPSH ! required versus NpSN available, f i
- d. As ineicated in our response to CA! 220 the NPSH test was waived. A certified l NPSW required curve vos het requested. However, an HP8H curve that applies to the Limerick two-stage pump was recently obtained from the vendor. The curve is the same as the catalog data used in the calculation.
l l
)
........e...........................................................................
\ en
- A, $]
PEco Review signature / Date Rechtel Approval signature / Date Page V,. of Ie ,
[' L I' STONE & WEBSTER ENGINEERING CORPORATION E LIMERICK 2 IDCA (J.O. No. 18138) OBSERVATION REPORT Observation Report D OR- 032 Rev. O Review Plan LK- D - 1903 - MS Rev. L PART III - RESPONSE EVALUATION The response to.this observation report has been evaluated as follows: A. Causal Factor (s): Acceptable Not Acceptable (Explain): NOT APPLICABLE B. Extent of Condition: Acceptable Not Acceptable (Explain): NOT APPLICABLE C. Significance: Acceptable Not Acceptable (Explain): NOT APPLICABLE ( D. Corrective Action: Acceptable Not Acceptable (Explain): NOT APPLICABLE E. Action to Prevent Recurrence: Acceptable Not Acceptable (Explain): NOT APPLICABLE
.....................................................=........==..=====...=========.==.
Additional Action Required: X No Yes (Explain) SEE INSERT A
,Ai .. ...................................................................................
0iar~~ Lefd Engineer Signature a]2\. 99
/'Dat'e ocLLs ontle.
AWM Signature ddn
/ Date Page [ of I
INSERT A ,
- a. Bechtel's response demonstrates that a three feet drop in pond level j
O for the spray pond winter bypass operation is a conservative assumption and is, therefore, not a concern.
- b. The lack of reference for the " minimum initial pond water level' (for 2 units) is not a significant concern as its value is being added in the technical specification 3.7.1.3a. This technical specification is already referenced in Calculation Number M-12-30, Rev. O.
- c. As explained in Bechtel's response, a lack of comparison between the NPSH available and 'the NPSH required in the calculation conclusion is not a concern since a comparison cf maximum flow versus allowable flow with regard to NPSH requirements is already made in the calculation.
- d. The use of- catalog data for a single stage pump in the calculation is not a significant concern since the NPSH curve appl / cable to the Limerick two-stage pump has been obtained from the vendor and Bechtel confirmed that the . curve is the same as the catalog data used in Calcu,1ation Number M-12-30, Rev. O.
( ( h
' DDR - 03 A REV o PAGE 6 OF 6
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT U Observation Report No. DOR- 033 Rev. O Review Plan LK- D- 1903 - MS Rev. L Reference AI No.: DAI-199 Rev. O PART I - INITIATION
- 1. Description of Concern Calculation No. 2006 contains an error which underestimates the j blowdown flow and produces non-conservative compartment temperature transient result. The peak temperature for Case 12 may exceed the environmental design criteria.
- 2. Supporting Information As stated in Action Item No. DAI-199-0, the reverse blowdown flow for the main steam line break is underestimated because of error in determining the piping frictional loss. The frictien i loss in the intact line must be adjusted with respect to the flow rate in order to obtain the correct equivalent loss at the break location.
The response to this action item has confirmed and agreed that a 1/9 factor should be applied to the friction factor of the intact line in the reverse blowdown calculation. (~N The response aleo indicates that this error will have no impact on the calculated compartment peak pressure and temperature because the peaks have already occurred during the inventory blowdown period before the period when ,the reverse blowdown flow is considered. This statement is confirmed to be correct for the Case 11 pressure and temperature transients and the Case 12 pressure transient during which time the peak pressures and temperature do occur during the inventory period and are substantially higher than the calculated pressures and temperature for the period when the reverse blowdown is considered. However, this is not true for the Case 12 temperature transient. The Case 12 result clearly snows that the peak temperature occurs during the period when the reverse blowdown is considered. The incorporation of the correct friction factor will increase the peak temperature for the Case 12 Study and it may exceed the environmental design criteria. Additional Documents Attached: None
......................................(........................
'L WWerr O Lea'd hfgideir Sighature
. > winds
'/ Date 6 ~
APM Sigj(ature ' / Date Page 1 of 1
370NE & WE5875R ENGINEERING CORPORAt!CN 1.1NERICK t IDCA Jpgjyg (J.0. No. 18188) ossamiroN arreMAN 311989 [S- off 9)
. vtO Observation Repor t No.1 OR _g)),, Rev. ,_CL, PART II - RERPDNat <
- 1. Observe,t.Len concurrence:
.II.,. Concur with observation
, Do not concur with obserystion (Note: if net in concurrence, explain in
'reamrks" below)
- 2. Response to Observation:
A. causa! Factor (s) An error was made by originator and checker.in est'imating the frictional Affect of para 11st branches.
- 3. Estent of Condition Error affects the peak temperature of calculation 2006 case 12 compartment pressure and temperature study only.
The case of blowdown flows from parallel main steam lines through a cor. mon cross header is a unique case'of blowdown flow path, since this is the O only case of this physical configuration, this error is a unique and laelated case. C. significance None. Due to the location of the postulated high energy line break being in the math condenser area, ard the fact that there is no safe shutdown equipment located in the break . compartment or in the adjacent steam venting plenum, the case it results are non-consequential for the safe shutdown operation of the plant.
- D. Corrective Act105 Calculation 2006 will be updated to address the error and its effect on the results. No correction to environthental design criteria is required.
E. Action to Prevent Recurrence since this error was a unique and isolated case and hac no significance, action to pr.v.at r.currence i. 3et r. quired. Remarks a .....,;.......y.<;,............................................................... G* , f M b d67 reco Reviev si natur.- i cato
/ AfR./mr cP# .eht<i A,,r var signature-i i/xm i cate Page of l M Pop 2 of }
L - _ - _ _ - _ _ _ _
1 STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) [U OBSERVATION REPORT Observation Report D OR- 033 Rev. L Review Plan: LK- D - 1903 1 Rev. L l PART III - RESPONSE EVALUATION 1 The response to this observation report has been evaluated as follows: l A. Causal Factor (s): 1 Acceptable Not Acceptable (Explain): 1 The response has concurred that an error was made by the originator and checker in estimating the frictional effect of parallel branches. l B. Extent'of Condition: J Acceptable Not Acceptable (Explain): The response has concurred that this error will ef fect the calculated peak temperature of the case 12 study in Calculation 2006. The response also indicates that the case for blowdown flows from parallel lines through a common cross header is a unique case of the blowdown flow path. It was concluded that this error is, therefore, a unique and isolated case' . C. Significance: X Acceptable Not Acceptable (Explain): The response has indicated that this error has no significance because the ( 4 break location is in the main condenser area and there is no safe shutdown equipment located in the main condenser area. D. Corrective Action X Accept able. Not Acceptable (Explain): The response has agreed to update the calculation to address the error and its effect on the result. E. Action to Prevent Recurrence: X Acceptable Not Acceptable (Explain): The response has indicated that action to prevent reciarrence is not required because this error is a unique and isolated case.
.............y.%.......................................................................
Additional t.,. ion Req dred: X No Yes (Explain) (IN h Lead Enginee'r Signature 9 2ll99
/ Date odat fx 0,uhlI<-
A#M Signature 1lult1
/ Date Page .)_ of 1
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138)
</~X j OBSERVATION REPORT Observation Report No. DOR- 034 Rev. O Review Plan: LK- D - 1903 -
S Rev. L Reference AI No.: DAI-298 PART I - INITIATION
- 1. Description of Concern The method of analyzing a generic embedded plate in Calculation 26.3 does not appear to be applicable.
- 2. Supporting Information )
Page 23 of the calculation evaluates the plate for the worst position of applied tension. The analysis is done assuming that the plate and studs behave as a rigid frame. This results in moments at the
. junction of the plate and studs which reduce the bending rtress in the plate. Since the studs cannot resist this moment, this type of behavior is not possible.
Response to Action ~ Item DAI-298 states that since the studs are t embedded in concrete 'they will not see any bending'. This does V not explain how the studs can be used to apply moments to the plate.
.....m.mm..m...m.ame.a.................mmm...m..mmmm...m....m.....m..m.....mmmmm.sm..m==
Additional Documents Attached: l None. N mem..mmmmmmm..m.. mas......mmmm..mm.p.m.mm.m..m..m......m..m. . .m.. mmm..mmmmmm..ma (. Yn / Lf**h _ apt. Signatgre '
/'Date Lead Engineer Signature ' / 'Da t e Page 1 of g
S'IONE & WEBSTER ENGINEERING CORPORA s (dpALC ,j m _,, n LIMERICK 2 IDCA ,_ M'd W w- - - - (J.O. NO. 18138) REISSUED W 0 9 1999 OBSERVATION REPORT DATE j Observation Report No. DDR-034, Rev. O PART II - RESPONSE AMENDED COMPLETED RESPONSE l
- 1. Observation Concurrence:
X Concur with observation Ib not concur with observation (Note: if not in concurrence, explain in " remarks" below)
- 2. Response to Observation:
A. Causal Factor (s)
'Ihe rigid frame method was used in Calculation 26.3 to determine the tension load capacity of the generic embed based on the allowable bending stress of the embedded plate. We concur that this methodology was incorrectly used for P.is application as the predicted rooment acting on the studs was not conside, M in the evaluation because of an inadvertent error by the originator of the calculation.
p' B. Extent of Condition V We have reviewed all other embed calculations within the project and I have concluded that the application of the rigid frame method is unique to this embed. - C. Significance l 'Ihis observation has no significance as the generic embed's c..pacity is governed by the studh pull-out and shear capacities rather than the bending stress in the plate. 'Ihis is evident from evaluation of the embed in Calculation 26.3 Revision 7 for combined tension, shear, and noment loads equal to maximum allowable embed loads given in the calculation. The evaluation uses the resulta of finite element analyses of the embed usirg Bechtel standard computer program " BASEPLATE" (ME-035). (Refer to sheets 35.1 to 35.24 of Calculation 26.3 Revision 7). Althotqh the finite element nodel analysis considers only the case of the wide flange attachment member centered on the plate, the nodel represents close to the nost critical position of the attachment with regard to the plate stress since the attachment flarge is nearly midway between anchor lines. It is apparent, from the plate stress computed in Calculation 26.3 Revision 7, sheet 35.24, that the plate design will accommodate any position of the attachment, and plate stresses will not control the design.
========================================================================r.===========
Nrbo Page l of [
l l tt701 P02
, , , _ . g .gg. 39 11:50 !D:PECO PROJECT MGT Dlu TEL NO:215-641 4579 p.3/3 A 89 '8916:3e WCCHTcL ESTRN PO4R iST floor O
"" "5 mci"fis (J.O. NO. 18138)
" """ "VDC 4 /G 95
. OBSERVATION REPORT Observation Report 00R-034, Asv. O PART !! - ACEPONSE CONTINUED 1
Allowable embed loads given in calculation 26.3 account for interaction. The allowable loads, given for normal conditions, are summarized on sheets 34-35, and are defined by the interaction curves on sheets 40-41. The curves on sheet 40 are derived to limit anchor shear, resulting from b m ial shear loads and torsional moment, to acceptabJe limits when + considering the presence of anchor tension. The cu:ves on sheet 41 are ; derived to limit anchor tension, resulting from tension load and bending sement, to acceptanie limits when considering the presence of anchor shear. l As stated in the calculation, the allowable loads must satisfy concurrently the interaction curves on sheeto 40 and 41 and the biaxial moment interaction equation on sheet 34 (Ms/M + My/M < 1.0). '!he curves are derived considering the attac,hment member as concentric. Thus, as stated in the calculation, any eccentricity must be accounted for.- To demonstrate more explicitly that interaction is accounted for, the calculation has been enhanced by calculating anchor shear-tension interaction O
- ratios for traxitrun embed loads given by the interaction curves on sheets 40-41, and for the loads of the exatrple on sneet 42 (Refer to shw ts 35.22 t.o 35.23, 41.1, 42.1, and 42.2 of Calculation 26.3 Revision 7). These added efeputations show the interaction ratio to be within acceptante limits and demonstrate that interaction is properly accounted for. .
Dnbeds installed in'the plant are evaluated for specific applicable attaen-rnant loads using the allowable loads as given in Calculation 26.3. An exa.tple of such an evaluation for a pipe support attachttent is shown in Calculation 22.5-3-12 Revision 0, sheet 6. . D. Corrective Action Calculation 26.3 Revisions 6 and 7 have been completed to delete the rigid frame evaluation, to include an evaluation using computer program "BASEPIATE", ,
'1
.cd to enhance calculation of the effects of combined loads.
No other corrective action is required, because this is a unique occurrence. E. Action to prevent Recurrence . No additional action is required, because (1) this was a unique application of the rigid frame method in the design of one generie embed type, and the con-dition cannot recur because calculation 26.3 Asv. 5 for this generic err. bed has been revised as discussed in item 'D' above. (2) The results of the observation Report and the calculation revision have been discussed with involved engineers. O ..........the M&w#1 PDCo Aevistwjgnatute /Date au a e 4%
...................................................................L....
Beental Approval 6(ghg)(fre 7 /Da t.e' Pap 1 dl
l l l STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT Observation Report D OR- 034 Rev. O Review Plan LK- D- 1903 - S Rev. 1~ PART III - RESPONSE EVALUATION l 1 The response to this observation report has been evaluated as follows: A. Causal Factor (s): X Acceptable Not Acceptable (Explain): The cause was determined to an incorrect methodology on the part of the i originator in utilizing a rigid frame model for an embeded plate that resulted in the studs resisting bending moment. B. Extent of Condition: _ 1, Acceptable Not Acceptable (Explain): All other embed calculations were reviewed and no other use of this model was identified. C. Significance: X Acceptable Not Acceptable (Explain): ( SEE ATTACHED PAGE D. Corrective Action: 1 cceptable A Not Acceptable (Explain): The incorrect methodology used was unique in nature and as such correction of the calculation is acceptable. E. Action to Prevent Recurrences y Acceptable Not Acceptable (Explain): As this model was unique, discussion with the responsible engineer is considered acceptable to prevent recurrence.
....................................m...........-=........m============================
Additional Action Required: y No Yes (Explain) O .smosesses==memana.ammeness.nemessammmmmmmmmmmm.sammespeammmmmmessnessamesassans. mss === U L f.C L sA/ Date J1 o%x s0.uk sMn Le'ad Engineer Signature ' APV Signature / Date Page f of [
-STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) s OBSERVATION REPORT s Observation Report DOR- 034 Rev. O PART III - RESPONSE EVALUATION C0lrFINUATION:
C.- Significance: Acceptable l The product of the calculation was a set of limiting loads that could ] be applied to the embed plate. As the revised calculation, using an acceptable finite element model, showed that the li.niting loads were still within design allowables the effect of the incorrect model is not significant. It is noted that an interaction of shear and tensile loads was l performed using the finite element model and, when using the actual concrete strength of 4000 psi rather than the.3000 psi in the original calculation, the studs were shown to be adequate for interaction. Although the response to DAI-478 indicates that it was necessary to rely on the 4000 psi concrete to justify a moment higher than that in the calculation, the conclusion of that evaluation is still valid since the shear was low in that case.
\
1 O Pag e r o f s_ . l
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) Q OBSERVATION REPORT , Li Observation Report No. DOR- 035 Rev. 0_ Review Plan LK- D - 1903 - S Rev. 1 Reference AI No.: DAI-243 PART I - INITIATION
- 1. Description of Concern Calculation 113.4.8 does not provide sufficient information to demonstrate the adequacy of the pedestal liner for loads applied by an attached platform.
- 2. Supporting Information In the calcul ; ion, the following items were not checked: ,
l The /C vertical bars, which the calculation assumes anchors the ..ner, is welded to the liner, but the welds were not checked. The liner and 3/8" vertical bars are not checked for
' pullout'.
There is no check on the strain in the liner at the ' hinge' i formed at the point of load application. Response to Action Item DAI-243 states that these items do not have
~
to be checked, based on the overall behavior of the liner. However, this behavior is not consistent with the assumed behavior in the calculation, and the mechanisms of load transfer have not been checked for this situation either (e.g. weld stress o- liner stress). Furthermore, the response states that the liner behaves elastica 11y, in membrane action. It is true that the liner has not yielded in tension, but the liner must first yield in bending before membrane tension can develop. The displacement which results (approximately is") is much greater than the elastic deflection of the panel.
.............................................................................-.......==
Additional Documents Attached: None.
............m.........m............p..m.....m....m...a....... ...m-u. ...mm.m.s......m.
( (Nn /tfleh/ ead Engineer Signature '
/ Date APM Signature l
/ Date Page 1 of 3
i
)
STONE & WEBSTER ENGINEERING CORPO LIMERICK 2 IDCA (J.O. tb.18138) ' l REISSUED yAs o a 1989 l")' C OBSERUA} ION RT IDATE
\)j h f h$ Observation Spor DOR- , Rev. O PART II - RESPONSE g AMENDED COMPLETE RESPONSE
- l. Observation Concurrence: l Concur with observation X Do not concur with observation (Note: if not in concurrence, explain in " remarks" below)
- 2. Response to Observation:
A. Causal Factor (s): N/A B. Extent of Condition: N/A C. Significance: N/A D. Corrective Action: Not Required E. Action to Prevent Recurrence: tbt Required O ====================--------=========-------=======--------========--------=== d Remarks:
- W do not concur with the observation. Calculation 113.4.8, Rev. O, generated for the disposition of NCR 11733, does demonstrate the adequacy of the pedestal liner plate. The conservative assumptions relating to the load carrying node of the liner plate and the engineering judgement applied to the items of concern in this observation report were sufficient to demonstrate the adequacy of the liner plate for the 2.91 kip applied load. This has been verified by the additional detailed calculations in response to SWEC's concerns and included as Revision 2 to Calculation 113.4.8 (Transmittal tb. P-0355 Dated 3/7/89 ). This revised calculation demonstrates the capacity of the pedestal liner at the platform attachment to be 11 kips on the basis of Seismic Category IIA design criteria.
The pedestal liner plate was designed and constructed as a continuous cylindrical plate with 3/8" vertical and horizontal stiffeners as shown on Drawing C-296, Revision 14. Individual plates approximately eight feat high encompassing 90 degrees of curvature utilized edge stiffeners as construction aids for initially bolting and subsequently welding them together. The 3/8" attachment lug for the access platform at elevation 217'-0 3/8" is attached at a location not immediately adjacent to local stiffeners of this continuous liner plate as shown on Drawing C-296. The membrane analysis method used in Calculation 113.4.8 is consistent with the local liner plate behavior for transferring the lug load to the global circumferential (hoop) membrane load resistance of this continuous cylindrical plate. Calculation 113.4.8 has been revised to confirm the initial method and ( t judgements. ( Page l of 4
, rW: M u 12:11 20:sTCt. WES Ft4 50 Is af r_:& F. N
. ==.- .
< STONE r. WEBSTER ENGINEERING CORPORATION \)DCAllo90 LIMERICK 210CA (J.O. No. 18136)
OBSERVATION PZK)RT Observation Report COR-035, Rev. O PA)C II - REMARXS COh"rINUED 2e followirg coments address the concerne stated in the Observation Report: Item 1 The boundary conditions of the local models were defined at the adjacent stiffenors. Se weld of the stiffener to t.he liner plate needs to be checked to assure the assumod toundary conditions are met and the local analysis is valid. This eneck was made by judgement in the initial calculations. An explicit check was made in the revised yieldline theory analysis. Item 2 Since the bounday conditions cf t.no local analysis need to De checked to assure the assumed conditfor.s are met, the " pullout" of the liner and 3/8" vertical bara must be checked. In the initial analysis the bearing in the circumferential direction of the stiffener on the pedestal concrete was calculated and "ptilout." was assessed on an enginooring judgement basia. An explicit check was made in the revised yieldline theory analysis. O C'- Item 3 A check on strain control for these types of Seismic Categen IIA analy-ses is typically made by calcu:atirq a " ductility" ratio. Por this case, it would be a ratio of tre total displacement of the lu;; under the maximum applied load relat:ve to the maximum elastic displacement of the lug. ne revised calculation indicatsa a ductility ratio of 5.4 for an ultimato load capacity cf 11 kips. ne stif fener to liner weld check (Item !) and the stiffener pullout check (Item 2) are required to verify assumptions used :n the local plate analysea and are not tne global failure trode. If eit.ncr of these " local failures" occurred, the liner would continue to safely carry the load through membrane action. These platfom attach-ment lu;s are isolated cases of attactvnerets to the pedestal liner which required non-linear analysis, i in summary, the revised calculation cont:ms the adequacy of the platform attach-ment lug and tne local liner plate to sustain the conservative platfo= poolswell drag load. M e Sco Review Signature /Date gentalApprovalSW6ature/ /Date' Page 2 ot j E_ -- -- -- )
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138)
-O G OBSERVATION REPORT Observation Report D OR- 035 Rev. O Review Plan: LK- D - 1903 -S Rev. 1 PART III - RESPONSE EVALUATION The response to this observation report has been evaluated as follows:
A. Causal Factor (s): 1 Acceptable Not Acceptable (Explain): The response states that engineering judgements (undocumented) were used, but do not appear in the calculation. B. Extent of Condition X Acceptable Not' k eeptable (Explain): The configuration for the platform attachments to the liner plate are the only use of this type of non-linear analysis. C. Significance: X Acceptable Not Acceptable (Explain): The revision to Calculation 113.8.4 (Rev. 2) was performed
^
to demonstrate that weld stresses, pullout and ductility ratios are'within allowable levels. D. Corrective Action: 1 cceptable A Not Acceptable (Explain): The calculation has been revised to check the pull-out welds , (liner to lugs) and ductility ratio. A detailed review of this calculation was not performed. 1 E. Action to Prevent Recurrence: X Acceptable Not Acceptable (Explain): No action is required, since no attachment of this nature to the liner are anticipated. Additional Action Required: X No Yes (Explain) 1 O....................................................................................... L e c m _ .s A / u aace a o. tv<//c s/,s+r Lead Engineer Signature '
'/ Date APM Signature / Date Page 1 of j
- _ _ _ = - _ _ - _ _ . - _ _ _ _ - - _ _ _ _ _ _ _ - _
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA s (J.O. NO. 18138) ( ) E OBSERVATION REPORT Observation Report No. DOR- 036 Rev. 0_ Review Plan LK- D - 1903 - S Rev. J Reference AI No.: DAI-051 DAI-055 PART I - INITIATION
- 1. Description of Concern There are several assumptions made in the model used to perform the seismic analysis of the reactor building which may not be appropriate.
- 2. Supporting Information
- Calculations 24.1 A through D use soil springs based on a rectangular foundation. Results of these analyses are used to design structures. Calculation 24.1E, which develops amplified response spectra (ARS) used for subsystem design and qualification, uses soil springs based on a circular foundation. Response to Action Item DAI-052 states that an equivalent circular formulation is appropriate, but this does not seem to be true since the rocking
, , , stiffness is obviously greater in the E-W direction than in the I k N-S direction. \_)
- The response also states that the question ' of soil spring formulation is really irrelevant, since the structure is founded on rock and there , pre negligible soil-structure interaction effects. This would seem to be true, but resulting ARS peaks from the circular base model can be significantly lower than those from a model having a fixed base, especially in the vertical direction. This can be seen from the plots included in calculation 24.1E. It is agreed that the LCS foundation rock is in the ' rigid range', and the results from the fixed base model are appropriate, but it has not been demonstrated that the results from the model using springs based on the equivalent circle are appropriate.
CONTINUED ON 1 PACE
.....m..............m......m...........m....................m..mm..m...m..mm.mm.....m.m j Additional Documents Attached:
None.
.mm..........m...m.....m...mam..m.mpm.s....maeam......m.. m... . m..mam.....mam.. / }
j
! 9% l
/ / Date Lead Engineer Signature '/ Date APM Sigrp(ture Page 1 of 1
STONE & WEBSTER ENGINEERING CORPORATION l LIMERICK 2 IDCA (J.O. NO. 18138) . OBSERVATION REPORT f"'N Observation Report No. DOR- 031 Rev. O p , CONTINUATION: In the development of the seismic model of the reactor building, interior walls are not included in the calculation of the building stiffness. Response to Action Item DAI-055 states that modeling is based on industry practice, judgement and experience. This ! does not adequately explain the omission of these walls, which clearly contribute to the stiffness. Although it is agreed that neglecting these walls is conservative for design purposes, the results of the seismic analysis could be affected. O Page 1 of 9
t l-VDeg /4C S'IONE & WEBSTER ENGINEERING CORPORATION ) LIMERICK 2 IDCA REISSUED I DATg yAR 0 21989 fN (J.O. NO. 18138) OBSERVATION REPORT ^' M6' m u. Observation Report DDR-036, Rev. O PART II - RESPONSE l AMENDED COMPIfrE RESPONSE l ; i
- 1. Observation Concurrence: i Concur with observation X Do not concur with observation (Note: if not in concurrence, explain in " remarks" below)
- 2. Response to Observation:
A. Causal Factor (s): N/A B. Extent of Condition: N/A C. Significance: N/A D. Correction Action: Not Required
/O g
E. Action to Prevent Recurrence: Not Required Remarks: # We do not concur with this observation. A detailed response to the items cited by SWEC as supportirg information is provided as follows: )
"Ihe assumptions used in the seismic models and analyses of the reactor building are appropriate for their intended use.
Durity the duration of the project, three complete seismic analyses were performed. j In 1971-1972, seismic stick models were developed for the east-west horizontal, ) north-south horizontal, and vertical analyses of the reactor building. Soil spring ' stiffness was developed based on the rectangular base of the structure. Soil dampirg was not included. Instructure acceleration response spectra was generated and included in Spec G-14 as Appendix C. The seismic structural responses (forces, I displacements, and' accelerations) were used for the seismic . design of the structures j and structural subsystems. 1 In 1976, soil damping was incitxied in the analyses. A fixed base analysis of the . building dynamic properties was performed to compare with the properties obtained l from the model with soil springs and dampirg. 'Ihe complete seismic analysis was l then performed usirg the soil structure interact 1on tredel. Instructure acceleration l
/' response spectra was generated and incorporated in Specification G-14 as Appendix F. {
s 5 l i Page 3 of 9 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - - - - - . - _ _ _- I
SIONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. to. 18138) [ OBSERVATION REPORT b g Observation Report DDR-036, Rev. O PART II - RESPONSE 00tEINUED R e reactor building was reanalyzed for seismic responses in 1981 during the hydrodynarrnic load assessment. Se purpose of this reanalysis was to incorporate the concrete nodulus of elasticity determined during the inplant hydrodynamic load test, verify and possibly reduce the impact of the previous analyses (minor changes were made in building stiffness, equivalent circular base was adopted) ard develop instructure seismic acceleration response spectra campatible with the hydrodynamic load effort (included in Spec. G-19). A fixed base seismic analysis was performed for ccrnparison purposes. In addition, local decoupled structural models were generated for local vertical responses. Item 1: he effect of using either rectangular or equivalent circular based soil springs can be seen by comparing the building dynamic properties, in particular building frequencies, with their respective fixed base analyses. For the more critical N-S direction and rectangular based springs (see Table 1,1976 analyses), frequencies reduced by 6.3% for the first node, 1.4% for the second mode, and 2.3% for the third mode. Similar analyses in the N-S direction using equivalent circular based soil springs (see Table 1, 1981 analyses) show frequency reductions of 6.1% for the first node, 1.6% for the second mode, and 2.5% for the third mode. A maximum difference of + 0.2% for the reduction effect on individual frequencies, is well within-the numerical accuracy of these analyses. These results show essentially identical resulting frequencies for either the rectangular l
/7 or equivalent circular based soil springs with the soil damping included.
V Acceleration response spectra were generated for the 1981 fixed based and flexible base acalyses. R ese spectra show some differences in the values of peak accelerations. Feak acceleration variations of 30 to 50 percent could be expected in a comparison of these spectra, depending on the relative impedance between the structure and foundation. Item 2: During the development of seismic models there was a point at which the addition of increasingly smaller member stiffness will not significantly affect the global lead path and thus have negligible effect on the global l structural response. This occurs because the relative stiffness of the added members is very small and does not significantly change the building properties and response. The practice of broadening instructure acceleration response spectra is an adequate procedure to cover such uncertainties. As an example of the building stiffness increases in the N-S direction model, a comparison of the 1976 and 1981 fixed base nodel frequencies can be used. Between elevation 177 and 201 of the 1981 model, the shear area was increased by 7.8%, and the moment of inertia was increased by 14.2% above the 1976 model. In addition, the concrete nodulus was substantially Page Q of 9
tw u m. w w w.m is.- m r - 810NE & WES8TER ENGINEfMNG CORPORATION LIMERICK 2 IDCA (J.O. NO. .l4183) h( fh' OBSERVAT30tt REPORT Observation Report No. DDR-036 Rev. 0 PART II - RESPONSE C0tiTI!Cg increased (see note 4 of Table 1). These etenges resulted in a fixed base frequency increase of 8.5% for the first node, 7.6% for the second node, and 7.3% for the third nede. The eemparison discussed above between the previous 1976 and 1981 codel illustrates the degree of sensitivity of the buildino model to stiffness increases. Inclusion of the partial walla between elevation 177 and 201 would not significantly change the seismic analysis results, since the stiffness change associated witn these walls is reistively small cenpared to the changes includs! in the 1981 model. Mditionally, the omission of these walls in the suilding modele is representative of the types of " approximations in acchling techniques used in solemic analysis" described in Regulatory Guide 1.122 (Development of rioor Reponse spectra). In sumnary, the assumptions made in the nsactor Builditg seismic analyses models were appropriate for the intended use of the seismic responses. Ade% ate evaluation and control was exercised to oesure the required criteria conserve ic > were set. O
................................................... .... g....................
/&fudQHw PECcVRevier 11 gnat'ure
$/e7
/Date 4A/m *FCnLi w-n
/Bechtel Approval Signaty /Date Pagelofd
S7ONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA f A / h/ (J.O. NO. 18138) - f
) { D[ /
(O) OBSERVATION REPORT Observation Report No. DOR-036, Rev. O PART II - RD4 ARKS 00tfr1NUED Table 1. Summary of Reactor Building N/S Horizontal (DBE) Properties and Fesponses Analysis Freq.(hz)/ Peak (5) Base (5) Base (5) Danping (%) Accel. g's Shear Overturn k x 106 k-ft 106 1971 (1) 2.49/ - 1.166 0.128 16.06 Soil springs, 7.75/ - (node 12) no soil damp 11.53/ - 1976 3.02/ - NA NA NA Fixed Base 8.30/ - 12.75/ - 1976 (1) 2.83/5.0 1.037 0.120 15.14 Os Soil springs, 8.18/5.0 (node 12) soil damp 12.45/5.0 1981 (3,4) 3.30/ -- NA NA NA Fixed Base 8.98/ - 13.78/ - 1981 (2,3,4) 3.10/4.7 1.082 0.130 16.80 Soil springs, 8.84/5.6 (rode 12) soil damp 13.43/6.2 l (1) Soil springs based on Rectangular base. (2) Soil springs based on Equivalent Circular base. (3) Building stiffness increased due to refined calculations. (4) Concrete modulus of elasticity increased by 40% to account for age hardening of concrete and to correlate with in-plant test results. l (5) Based on ABSS of seismic modes. NA Not Available. O Page6of1 1
STONE & WEBSTER ENGINEERING CORPORATION
' LIMERICK 2 IDCA i (J.O. NO. 18138) ,
/O OBSERVATION REPORT l LJ Observation Report D OR- 036 Rev. O Review Plan: LK- D - 1903 - S Rev. T
~
PART III - RESPONSE EVALUATION Th2 response to this observation report has been evaluated as follows: A. Causal Factor (s): Acceptable Not Acceptable (Explain): NOT APPLICABLE B. Extent of Condition: Acceptable Not Acceptable (Explain): NOT APPLICABLE C. Significance: Acceptable Not Acceptable (Explain): NOT APPLICABLE b D. Corrective Action Acceptable Not Acceptable (Explain): NOT APPLICABLE E. Action to Prevent Recurrence: Acceptable Not Acceptable (Explain): NOT APPLICABLE Additional Action Required No Yes (Explain) SEE INSERT ! l l CONTINUED ON WO PAGES [. ni 3 f O N/ 6 3!/ N Lsad Engineer Signature
'/ Date A b Signature / Date Page 1 of 3_ [ ~ . _ _ _ _ _ _ - _ - _ _ _ - _ _ _ _ . _ _ _
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) g OBSENATION REPORT ( ) Observation Report DOR- 036 Rev. O i l PART III - RESPONSE ETALGATION COMINUATION: l ITEM 1 The issue, the use of soil s;eings based on a circular base, was , addressed by comparing natural frequencies for the five horizontal l (N-S) models developed during the course of the project. The comparison of the 1981 fixed base model and the 1981 model with soil springs indicates that the fundamental frequency is 6% lower , for the model with soil springs. Higher modes show even smaller differences. This confirms that there should be little difference in results between the soil spring model and the fixed base model. The response addresses differences in resulting ARS from the two models by stating that variations in results could be expected
' depending on the relative impedance between the structure and foundation.' This is interpreted to imply that 'radiational damping' ;
is the major cause of the differences. However, the information ! presented for the N-S responses does not confirm this statement. Comparison of the composite modal damping (Table 1 of the response) for the first three v. odes of the circular spring model (4.7%, 5.6% and 6.2%) do not indicate damping levels appreciably higher than n the DBE concrete damping of 5%, which is assumedly used for all ( ) modes of the fixed base model. v To further evaluate the response, SWEC selected a case where the largest differences in'ARS between the two models were noted. This occurred at node 25 (El. 313') of the vertical model. Comparing the OBE 0.5% spectra, the peak for the fixed base model is approxiniately 9.3g (Sheet 583 of Calculation 24.1E), while the peak for the soil spring model is only about 4.8g (Sheet 300 of Calculation 24.lE). Review of results provided the following information: The ARS peak occurs at a frequency of about 8 Hz. Review of the modal frequencies for the soil-spring model (Sheet 128 of Calculation 24.1E) indicates that this peak is caused primarily by response in mode 13 (f= 7.72 Hz). A review of the modal participation factors indicates that this is a dominant mode. Frequencies for the fixed base model were not available, but the corresponding mode would be expected to have a slightly higher frequency. For the OBE case, the composite modal damping ratio is given as 4.11%. This is significantly higher than the constant 2% damping used in the fixed base model. The difference muct be due to damping in the soil springs. This difference in modal damping will cause higher resulting response for the (~3 fixed base model. t
'V CONTINUED ON ONE PAGE haft 3 of 3
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA I (J.O. NO. 18138) OBSERVATION REPORT O. Observation Repc,rt DOR- E Rev. J
/
PART III - IISPONSE ETALUATION CONTIEU& TION: h Review of the response' spectrum (0.57. damping) of the input time history (FSAR Figure 3.7-4) shows a large sharp peak at a frequency slightly less than 8 Hz. This could obviously cause significant differences in response for models with slightly different frequencies. ) Review of computer input / output fo- the soil-spring model indicates that the spe,ctra were alculated at frequency intervals consistent ,ith BC-TOP-4A, which is the licensing commitment. Near the peak, the spectral accelerations were calculated at frequencies of 7.5, 7.72, and 8.0 Hz. It is important to note that the spectra was generated right at the modal frequency. Althcugh the spectra could be somewhat higher (or lower) at other frequencies near the undamped modal frequency, the frequencies used are reasonable and meet the licensing commitments. Based on the above review, it is con-luded that the differences between the results are explainable, and both the fixed base and soil-spring results are acceptable (one being more conservative O than the other). The differences are not entirely due to soil V structure interaction effects, but these effects are a significant contributing factor. ,. ITEM 2: It is agreed that the omission of the shear walls in question does not significantly affect the results of the analysis. 1 i l J Pap 9of 7
i
- l. STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 1DCA m (J.O. NO. 18138)
- (N OBSERVATION REPORT
Observation Report No. D OR- 037 Rev. 1 Review Plan LK- D - 1912 -
S Rev. A Refe rence AI No. : DAI-355 FART I - IIITIATION
-1. Description of Concern ,
When evaluating the effects of post-accident flooding for the 'D' line wall, seismic loads were not considered.
- 2. Supporting Information Design criteria specification C-115 requires that post-accident flooding loads (B oi be considered in combination with SSE seismic loads (E'). This was not done in the design of the 'D' line wall (Calculation 23.1 (b ) ., sheet 1(b)), where SSE loads were omitted.
Response to Action It.tm DAI-355 indicates that this load combination does not govern the (esign, but this has not been demonstrated for this structure. ,
.........m......m.......m.m.m.............m...............m====en======================
Additional Documents Attached: None.
=as====s=======s===============smmmymmenemmassomssesseses a mands ==samanesessammans b
{. $n Lead Engineer Signature
/ t*h!
'/ Da t e APM Sign 6ture l
/ Date Page 1 of [
SIME & WEBSTER ENGINEERING CORPORATION - LIMERICK:2 IDCA ~ f) (J.O. No. 18138) y/Q OBSERVATION REPORT JAN 3'O 1989.- Observation Report No. DDR-037, Rev. 0 4 PAIC II'- RESPONSE
- 1. Observation concurrence:
Concur with observation W'. X Do not concur with observation. (Notes if not in concurrence explain. in'" remarks" below)
- 2. Response to Observation:
y A. Causal Factor (s): N/A B. Extent of Corx11 tion: N/A C. Significance: N/A D. Corrective Action: N/A E. Action-to Prevent Recurrence: N/A O -
- We'do not concurs with the observation. There is no discrepancy in the load.
combinations.used in the design of the "D" line wall of the Reactor Enclosure .
-(calculation 23.l(b)) as discussed below.
The initial design of the "D" line wall was performed in 1971. Calculation 23.1(b), Revision 0 was performed considering all applicable load combinations as required at that time -(ISS-PSAR, Appendix C, Section C.2.5.2.2) . From a_ review of PSAR
. Appendix C, it is apparent that the post-accident flooding load (load combination:
U = l'.0 (D+L+Bo+E' )) was not a required design load. When the post-accident flooding event was defined and became a required design
< condition for IES, it was evaluated based on engineering judgement and was not documented. It was concluded that this design condition has negligible effect on the wall design because of the following:
- The post accident flooding loads are applicable only to the ECCS and RCIC corppartments within the Reactor Enclosure and between El.177'-0" and 201'-0". The hydrostatic load (A) used in load combination U = 1.0 (D+L+A)
. was-judged te be equivalent to (or more conservative than) the post-accident flooding load in the design of wall "D".
O' Page a of f
FEB-07 '89 10:33 ID PECO PROJECT MGT Dl~U T
"EL NQ:.215-B41-4578 #293 P05'
.~ .g.T'"*'"2"""'"'*"""
'[ 8'IONE & WEBSTER ENGINEERING CORPORAT!0N
. _ _ _ _ _ . _ _ _ _ _ .__ .. .__.g,1MgRICK-2 10CA--. .-- --- -- n- -
.- (J.O. NO. 18138)
-{r=*f
/7 7-castRVATION. rep 0RT ..
eservation Report MD. EX51-437, Rev. 0
. PART !! - RESPONSI CONT 1WCD
* 'me well design was governed by load combination U .1.0 (D4Ma%S+F+E')
*ra *** --- LM m!5t :: -Fr:: pm::= 1-:4f eg a te 3.. y.1 ei,y -
. J . .between ti.177'-O' through 21}'-W. It was judged that the 'F': load applied
,e . - . to a longer wall span is more critical than the post-sooident Standing load.
.w.:..x; 1 . .
i '
, me above evaluation conclusion is further confinned/ supported by the enlaulation l' .erovision perfomed as the corrective wtion for DOIt-011. Reactor Inc3anure well 1 0design. calculations (including calowlation 23.1(b)), which are affected by post-t2iaccident flooding loed, have been revised to demonstrate the walls amo structurally adequate using the current required load combinations. 'these =1mim*iaa revisions
..-' included the assesament of walls for tho' load combinetton U = L.4 (D* Lee 4'). --
c.; . 2. u..,... . . . . . .
.m.t. .:.- .
" ' ' ~ ~
p
'y'Wi-?"I- ,' , ' . ,
. . ... y. .....$.. . .......... .. ........._
M E).O Bhlr# ca4 -)-dL )u theldt 1/6(19 (v~'7 iso ... mo wi., sign.ture
<= t. =cnt.1 wrom ugn. cum o.=
Page l ot S
. 4'.*..'.g . . . , .
=
.v.,....
- a. =s.
[i.i
l STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT a Observation Report D OR- 037 Rev. O Review Plan LK- D - 1912 - S Rev. O_ PART III - EESPONSE ETALUATION Thz response to this observation report has been evaluated as follows: A. Causal Factor (s): X Acceptable Not Acceptable (Explain): The response states that the load combination was not in the PSAR, when the original calculations were done. When these loads were added to the criteria, a judgement was made that they would not impact the design, but this was not documented in the calculation. B. Extent of Conditions X Acceptable Not Acceptaole (Expla M): The response correctly states that post-accident flood loads are limited to the ECCS and RCIC cubicles. C. Significance: Acceptable X Not Acceptable (Explain): The response statos that this load combination has negligible effect on 9 the design. This is not consistent with revised Calculation 23.1(b), which (V states on Sheet 2(b) combinations. that this is one of the three governing load However, the revised calculation does demonstrate that the original design was adequate. D. Corrective Action: [ Acceptable Not Acceptable (Explain): Bechtel has revised the reactor enclosure wall calculations which are affected by post-accident flood loads (see Attachment 1). These calculations were not reviewed by SWEC. E. Action to Prevent Recurrence: X Acceptable Not Acceptable (Explain): There is no required action to prevent recurrence, since all concrete structures have been designed. A Additional Action Required: X No Yes (Explain) 0 0bwa C. fjl ,m. z)y'h? M$k (w- D.W 'lle- 3]nl87 Lasd Engineer Signature
/ Date A9M Signature / Date Page f of [
/f 7YAcM'rtsE&7~ l l
C i FEBRUARY 15, 1989 BECHTEL POWER CORPORATION DDR-837 CORRECTIVE ACTION 1 1 LIST OF REVISED CALCULATIONS CALCULATION NUMBER REVISION NUMBER l 923.901(b) 4 823.991(c) 4 823.801(d) 4 e23. eel (f) 5 023.901(g) 4 DOR o37 REV o PAGE F OF f
l. L l-
. STONE & WESSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.0, NO. 18138)
OBSERVATION REPORT ( Observation Report No. DOR- 038 Rev. O Review Plant LK- D - 1903 A - MC Rev. 1 Reference Al No.: DAI-134. Rev. O __ PART I - INITIATION
- 1. Description of Concern .
The LOCA submerged structure load methodology used by Bechtel on LGS may not be conservative when the load is calculated using the pressure field determined from the wave e'quation.without considering the presence of the submerged structure and the associated wake j effect. This concern applies to all calculations wherein, the l submerged structure loads are generated using the equation:
.F,,=[FdA 2, Supporting Information Bechtel used the equation II)
,F,=.fPd_A_
jG to calculate the submerged structure load. But, without using
\ physically . measured pressure field around the submerged- structure ,
the pressure field was calculated using the improved chugging methodology which did not ~ consider the presence of submerged structures in the suppression pool, nor.did it include the viscosity, the boundary layer separation, and the wake effects. The wake effects have been proven to be the dominant contributors of drag load (submerged setucture load) in all practical applications. In other words, the improved chugging methodology does not have the right governing equation (it neglects the presence of boundary layer and wake), nor does it have the right boundary condition (the submerged structure itself is absent in the formulation), to calculate the l pressure field adequately. CONTINUED ON 1 PAGE l
....................................................................................... j Additional Documents Attached: List of references.
....................M..............P............s........ .. -
.....n... ......a
, 2L]A.& apo/n
( u Lead Engineer Sign'acure // Date
~
\
/ Date APM Sig/'ature /
Page 1 of ~)'/
~
1 l 1
i STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT l Observation Report No. DOR- 038 Rev. O PART 1 - INITIATION CONTINUATION: , Because solving the pressure field considering the wake effect and the presence of submer,ed e structures is a formidable task, an empirically correlated formulation is used instead for engineering applications. The industry accepted submerged structure load calculation method is using the following equation: F, = 1/2 Cp d l U l U A + Cg d U' V + CL d lUl U A (2) where Co, Cg, and CL are standard drag, acceleration drag and lift coefficients respectively; d is the flanid density U and U' are flow velocity and acceleration respectively; A is the projected area and V is the volume of the submerged structure. This is the method NRC reviewed (Reference 1) and subsequently issued acceptance criteria (References 2 & 3). It should be noted that Equation (2) is used in all industries in calculating flow-induced drag and lift loads (Reference 4 - 6) for the reason explained above. Theoretically, the load can be calculated using Equation (1), using a analytically calculated or empirically determined pressure field, O but the pressure field has to be solved with the presence of the
) submerged structure and, more importantly, include the wake and the vortex shedding effects. Without considering these effects, the loads could be underestimated by an order of magnitude.
Therefore, conservatism of the submerged structure load during a postulated LOCA has not been demonstrated and should be resolved. I f V Page 2 of M
Attcchmtnt to DOR-038 , Rev. O ib
.(_,/ References
- 1. F. J. Moody, Analytical Model for Estimating Drag Forces on Rigid Submerged Structures Caused by LOCA and Safety / Relief Valve Ramshead Air. Discharges, NEDE-21471, revised by L. C. Chow and L. E. Lasher, September, 1977
- 2. USNRC, " Mark II Containment Lead Plant Program Load Evaluation Report and Acceptance C rit e ria", NUREG 0487 and Supplements
- 3. USNRC, " Safety Evaluation Report - Mark I Conatinment Long-Tern, Program - Resolution of Generic Technical Activity A-7", NURFG 0661, l- July, 1980
- 4. T. Sarpkaya and M. Isaacson, Mechanics of Wave Forces on Offshore Structures, Van Nostrand Reinhold Co., NY, 1961 l
l S. R. D. Blevins, Flow-Induced Vibration, Van Nostrand Reinhold Co., NY, 1977
- 6. H. Schlichting, Boundary-Layer Theory, M c C r a w - H ill, New York DOR -o 3E REV o PAGE 3 OF M
{
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA 1h A [ 7/ 3 (J.O. NO. 16138) / 7 g s OBSERVATION REPORT MAR 2 31989 LS-#74 9/ Observation Report No. ,,JL OR __QlB_ Rev. O AMENDED RESPONSE The following is an ameaded response prepared in accordance with agreement reached between Stone & Webster Et. gin-eri - Corporation and Bechtel in meetings held on M'rch 22, 1989. l PART II - RESPONSE
- 1. Observation Concurrence:
Concur with observation _21_ Do not concur with observation (Note: if not in concurrence, explain in
" remarks" below)
- 2. Response to Observation:
A. Causal Factor (s) B. Extent of Condition C. Significance D. Corrective Action E. Action to Prevent Recurrence l Remarks We do not concur with the concern stated in DOR-038. Based upon theoretical analysis, , experimental test, independent consultant study, and NRC review and acceptance; Bechtel 1 feels that the acoustic chugging methodology does provide conservative submerged l structure loads. Our technical pr u ition is presented in the following. 1.0 Introduction and Background The force exerted on a submerged structure by a fluid in motion is given by l F-fpdS [1] A U Page 2L of .2.4 l l
STONE & WEESTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. No. 18138) ((M [h} x OBSERVATION REPORT ( N, - Observation Report No. ._D_ OR- 038 Rev. O PART II c M PONSE CONTINUED where p is the fluid pressure acting on an area increment dS - ndS, n being the inward-pointing normal unit vector at :.be location of the area increment dS. If static pressures are excluded from p, chen F is the dynamic load (force) due to fluid motion. This treatment of the forco on a submerged structure neglects contributions due tc tangential shear forces since they are significant only for Reynolds numbers less than unity.2 For large Reynolds numbers (10 8 - 105 ) , the flow field past a structure resembles potential flow at the upstream side, followed by the development of a wake on the downstream side.2,3 Therefore, the surface integral given by [1] cae be divided into the sum of two integrals. The first is an integral over the structure surface outside the wake region where the pressure p = p, is given by potential flow theory; the second is an integral over the structure surface inside the wake where the pressure p - p,,.' Thus, [1] becomes F-fp,dS+fpdS [2] f- S, S,y ( ( where S, is that portion of the structure surface where the pressure is 61ven by perential flow theory while S,, is the remainder of the structure surface. Hence, the total structure surface area is the sum of a " potential flow region" plus a " wake region", i.e., S = S, + S,,, By adding and subtracting the integral of p, over the wake region of the surface, [2] can be modified such that the submerged structure force is given by F-fp,dS+f(p - p,)dS . [3] S S, The second integral in [3] is all but impossible to evaluate. Therefore, we set f(p, - p,)ds = f pC AAlU= lU, [4] 8 w
- Vector quantities will be indicated by bold type.
Page 5 of )_H 4
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA , /f(( (J.O. NO. 18138) OBSERVATION REPORT V- Observation Report No. p_ OR- 038 Rev. O PART II - RESPONSE CONTINUED where U. = U.(t) is the uniform flow velocity, and A is the projected cross-section area of the submerged structure. Note that [4] is a defining equation for the unsteady drag coefficient C3 which reduces to the steady drag coefficient Ce for U. constant in time. Values for CA are obtained from experiment. A somewhat limited conclusion 5 drawn from the comparison of the unsteady drag force Fp -fpCAlU,lU, A [5) with experimental data is that for cylinders in unidirectional flow acceleration CA< 2Cp for 108 < Re < 105 . In this Reynolds number range, Co - 1.2.8 The expression for the force on the submerged structure given by [3] thus becomes F-fp,d5+fpC,AlU,lU, [6) S where the pressure field p, is determined from the equation of potential flow in conjunction with a set of suitable boundary conditions. For an inviscid fluid described by potential flow, Euler's equation can be transformed into an expression relating the pressure field p., the time derivative of the velocity 7 potential d4/dt, and the gradient of the velocity potential V4 p, - p +fp(V4) - f(t) [7] Here f(t) is an arbitrary function of time which can be set equal to zero without loss of generality. Taking into account d' Alemberts paradox,s the integral term in [6] can be written as [P,dS-pf dS - spV 3 [8] S l 1 O Page 1 of D
STON! & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA
-[hd[ / f /3 (J .O. NO 18138)
OBSERVATION REPORT x) Observation Report No. ,,JL OR- 038 Rev. O PART II - RESPONSE CONTINUED The acceleration volume VA = xVg is equal to the sum of the structure volume Vg, pluS the classical hydrodynamic mass Na divided by the fluid density p. Hence M [9] x=1+[s The total force on a submerged structure can thus be written in the form F - spV 3
+fpCAlU,lU, A
[10] [ ] Since dU/dt is parallel to U., it is convenient to express the submerged structure force given by [6] in the following form CAD A F- 1+ g fp,dS [11] where D is a characteristic dimension of the structure and Sc = DU.-2dU./dc is the Strouhal number.10 Under certain conditions the force acting on the submerged structure given by the integral term in (6) (which is also known as the acceleration drag force) can be significantly larger than the unsteady drag force given by [5) . It is obvious, however, that when St >> C AD/2xV g 3 , the submerged structure force is ; essentially given by the integral of the pressure field p, (which is determined by the ( equations for potential flow) over the structure surface S. The submerged structure force given by [10) is the general app. roach which has been reviewed by the Nuclear Regulatory Commission (NRC) staff.1 12,13 The submerged I s tructure load acceptance criteria are based primarily upon the methodology presented in General Elmetric Report NEDE-21730.1' However, implementing thir methodology for submerged structure loads due to chugging and condensation oscillation is difficult because of the necessity to find the velocity and acceleration at the location of the i structure due to a condensation source at every vent. Since a similar problem has ! already been solved for the evaluation of the suppression pool boundary loads due to I f chugging with the acoustic chugging methodology,15 28 there has been interest in
\ applying this same methodology for submerged structure loads. However, the acoustic ;
chugging methodoleg,y solves for the pressure field in the suppression pool rather than ! Page _1 of M i
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA [pj /f] , (J .O. NO. 18138) i OBSERVATION REPORT i l V- Observation Report No. _D.,OR- 038 Rev. 0 l 4 PART II - RESPONSE. CONTINUED
)
the velocity field. Therefore, to use this methodology, the relationship between ! submerged structure loads based on pressures and thoso based on velocities and accelerations must be understood. l
't . 0 Acoustic Submerged Structure Ioad Methodology For a linear isentropic fluid where the velocity is everywhere small compared to the sonic speed c, the equations governing fluid motion reduce to the scoustic wave equation."
[]2p - -4xq [12) where q is a source density term and []2 ,92 , c-2 23f g g2 is the d'Alembertian operator. In the case of unsteady steam condensation (chugging and/or condensation oscillation) O-V q is determined by the dynamics of the steam-water interface. To a good approximation the unsteady condensation source density can be represented by a point source. For a chugging or condensation oscillation source at a location given by to - [r,0,z) in cylindrical coordinates q - p6(x - x o)S(t) [13) where here S(c) is the time-dependent source strength and 6(x - ox ) is the Dirac delta function. The nature of S(c) for chugging is impulsive with damped oscillatory behavior, while for condensation oscillation the source strength is mainly oscillatory. For chugging and condensation oscillation, there is a point source given by [13] at each downcomer exit. The solution to [12), together with suitable boundary conditions, uniquely specifies the pressure field p. The boundary conditions for a Mark II-type suppression pool are: (1) p vanishes at the free surface, O Page 1 of M
7 L i STONE & WEBSTER ENGINEERING CORPORATION l l- LIMERICK 2 IDCA JDCA /9/3 (J.O. NO. 18138) OBSERVATION REPORT' q l Observation keport No. D, OR ._Q)l Rev. 0 l PART II - RESPONSE CONTINUED 1 (2) the normal component of Up (or equivalently the nornal component of the fluid velocity u) vanishes at the suppression-pool bou 4dary,* and (3) the normal component of Vp (or normal component of the fluid velocity u) vanishes at the submerged structure surfaces. Note that for the case when the pressure is time-independent, [12] reduces to the equation for potential flow and p -+ p,. Therefore, for a time-dependent pressure field, p, is the acoustic analog of the potential-flow pressure field and thus is a solution to [12] satisfying the above boundary conditions. In general, it is practically impossible to obtain an analytical expression for the pressure field which satisfies the boundary condition at both the su'. merged structure and the suppression-pool boundary surfaces. Thus, we relax the bountcry condition at the submergef structure surface and obtain an approximate pressurn field px which , satisfies the first two boundary conditions listed above. As long as the submerged structure is smaller in size than the principle wavelengths of the pressure response,
. pa is a good approximat!.on to the true pressure field.
V(O When the acoustic wave equation [12) is a good approximation to the fluid equations of motion and the iluid acceleration is essentially constant over the submerged structure volume, we can write using [8] fpedS-afpadS . [14) The above mpression is the required relation between the pressure field p, which satisfies ; three boundary conditions listed above and the scoustic pressure field px which satisfies the only first two boundary conditions. Thus, we can evaluate the submerged structure force via CAD A F- x+ 3 fpdS A [1H 3 .
- This is rigorously true only for tha. case of a rigid boundary and/or structure surface. For simplicity we shall assume all submerged structures nre rigid. The containment can be treated as flexible and thus the normal velocity cem;,onent of the s
x boundary equals that of the fluid. Page 3 of D
STONE & WEBSTER ENGINEERING CORPORATION U."l** :. 'd&^> IDCA /9/3 OBSERVATION REPORT Observation Report No. _,[L OR- 038 Rev. O ; PART II - RESPONSE CONTINUED which follows from [11) and [14) . For the case of incompressible potential flow normal to the axis of a cylinder in the flow field with a uniform time-dependent velocity field U.(t) at infinity" dU [Pg,dS-2pV3[. [15] Ther:efore , comparison of [15) with [8] shows that for a long cylindrical submerged structure x - 2. Since the submerged structures generally have a cylindrical geometry, we infer that x - 2 is both an adequate and conservative connection between the pressure field p, which satisfies the boundary conditions at the submerged structure and the approximate pressure field px which does not. O 3.0 A Specific Example We now consider a specific example of a cylindrical, rather than an annular, suppression pool which has radius a, water depth L, and contains a single cylindrical submerged structure of radius b located at the center of the suppression pool and parallel to its axis. Let the depth of water be less than the length of the submerged structure. We choose this specific arrangement because an analytical expression for the acoustic pressure field can be obtained from [12) which satisfies the proper boundary conditions at the suppression-pool free surface and boundary, as well as at the boundary of the submerged structure. Hence, the pressure field p, is given by the solution to [12). The force on the submerged structure, given by [11), is 1 C A F- 1+ 2St P ,dS . [16) where for cylinders x - 2. For 108 s Re s 10 ,5 C <3 2Cp - 2.4; and the submerged structure force is essentially given by the integral of p, over the structure surface to within 5% for Je h 48/w. We consider a single peint-source located at x o = [r o,do,zo ) . The pressure field p, is O given by the solution of [12) subject to the boundary conditions at the free surface I
l STONE & WEBSTER ENGINEERING CORPORATION l LIMERICK 2 IDCA J p rsj/ (J.O. NO. 18138) O OBSERVATION REPORT L). Observation Report No. _]L, OR- 038 Rev. O PART II - RESPONSE CONTINUED and boundary of the suppression pool, and also at the surface of the submerged18 structure. Hence, the pressure at the surface of the submergcd structure (r - b) is Pe(b.f ,z) = 4wpc 8 E 9,(r o, do,zo )t,(b, # ,z)F,(t) . [17] The eigenfunctions 9,(r,# ,z) are functions of the suppression-pool geometry including boundary conditions and are 9,(r,$,z) - A, f (r) cos[m(f - fo)) cos[(1+b)(wz/L)] [18) t where A, is a normalization cou. tant and N = 1, m, n is shorthand way of writing the three eidenfunction indices. The radial eigenfunctions f,(r) are given by J'(wy b/a) fan (#) ~ m("7mn#I*) ~ N (wy m mnb
/a) m("Tan# !"}
where the prime denotes differentiation of the Bessel iunctions with respect to their arguments, and the eigenvalues wy, are defined by the roots of J'(wy ) N'(wy l nn I") ~ m("7 mn I"} "m(#7mn) ~
- l The function F,(c) determines the time dependence of the pressure field and is ,
dependent upon the suppression-pool or submerged structure boundary conditions only through the pressure-field eigenfrequencies. The time-dependent part of the total i pressure field F,(c) is given by
-A (t-t')
FNID) ~ 8(* ) *i"I"N( ~ }) w.l N Page d of M
1 STONE & WEBSTER LIMERICKENGINEERING CORPORATION' f 2 IDCA /f/3 (J.O. No. 18138) OBSERVATION REPORT p Observation Report No. D_ OR- 038 Rev. O PART II - RESPONSE CONTINUED where S(c) is the time-dependent 'sourc. strength due to unsteady steam condensation (chugging and/or condensation oscillation), and A, is the suppression-pool damping constant which is a direct result of structural damping. The eigenfrequencies w, for the pressure field satisfy
'w 2 'ry 2 ,,2 N mn
,c ,
,a ,
+ (1 + h) A -
[22) Consider now the approximate solution pa te the true acoustic pressure field which is obtained from' [12) and subject to the bound Try conditions at the free surface and boundary of the suppression pool, but not the surface of the submerged structure:
-( PA(b,#,z) - 4xpc2 g ,A (go,g0s2 y 0)I^N(bs0*E)I^N(t) . [23]
As before, the source is located at zo= [r o,Po,z o). The eigenfunctions for the approximate acoustic pressure field t^,(r,#,z) are functions of the suppression-pool geometry and boundary conditions and are tan(r,,#,z) - A*,J,(wpo r/a)cos[m(# - So))cos[(1+b)(wz/L)) [24) where A A
, is a normalization conatant and as before N = 1, m, n. The eigenvalues wp.
are defined by the roots of J'(wpan) - 0 . [25] The function F",(c) determines the time dependence of the approximate pressure field and has the same form as befcre: ( Page M of D = _ _ _ _ _ _ - _ _
_ STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA
'(J.O. No. 18138)
..[ M M / fG
. . OBSERVATION REPORT-Observation Report No. .JL OR- 038 Pev. O PART II - RESPONSE CONTINUED
" A
-A (t-t') A (t) - Si(t' ) e- sin [w (t-t')) de' [26]
W N where w", is the eigenfrequency for the approximate pressura field. Examination of.[20] reveals that in the limit as the ratio b/a goes to zero, wy. -* wp., and hence p4 . + p,. Thus, if b/a is small, .one might expecc p3. to be ' a good approximation to p,. As an estimate of how well pg approximates p, in a Mark II-type BWR suppression pool, let a - 13.4 m which is the radius of a typical suppression pool. The submerged structure with the largest radius in a suppression pool is a support column which'has a radius b - 0.53 m. Hence, b/a - 0.04. The condensation source function S(c) will generally.be significantly different from zero for frequencies less-than 100 Hz while acoustic speed for the chugging. sources is greater than 420 m/s.20
'Thus, from'[22) the' maximum value for wy , s 20. Examination of tables of cross product Bessel functions 21 (20] shows that. the largest difference between wy, and up.
O occurs for m - O and n - 6 where wy. - 20.069 and wp - 19.616. Thus, to within 3%, V w", = w,, and therefore A^, = A, and I',(c) = T,(c) . To estimate x in [14},, we shall compare ~the pressure at the surface of the submerged
' structure at a location r - b for any # and z, i.e., [b,#,z) obtained from [17] and
[23). It is convenient to make this comparison mode by mode. Therefore, define the parameter c such that f (a)f (b) - (1 + e.)J,(sp.)J,(p.b/a) [27) For this comparison, the source has been placed at a radius ro - a in order to maximize c.. Evaluation 2 of [27) yields con - 0.0, e,3 - - 0.024, c 2o - - 0. 0 5 2, and c o3 - - 0.070. In general: i i
' h - 0.071 ... m - 0
[28)
'an"(=1.0-1.12 ... m > 0 It is observed that for a given m, c increases slowly with increasing n. However, I. f- the modes for large n make lesser contribution to the pressure field due to the 1/w,
~1 term contained in F,(c), since as n increases, the eigenfrequency w, also increases but at a much faster rath. At large values of m, c = 1.0 for any n.
. Page d of M
k% ' STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA 1obA / /3 (J.O. NO. 18138) i OBSERVATION REPORT Observation Report No. _D OR Q.16 Rev. O PART II - RESPONSE CONTINUED If g were equal to unity for all m and n, then it can be seen from [14) and [27] that a would be equal to 2. However, since the m - O modes yield a value for % that is less than unity,'s must be less than 2 for the complete pressure field. 'The amount by which n is less than 2 cannot be reliably estimated from this example because of the special geometrical relationship between the suppression pool boundary and the submerged structure. Mora enlightenment comes from experimental evaluation of s.
-However, reflecting upon the above example and the case of potential flow, we infer-that x - 2. Hence,_ the submerged structure force given by the acoustic methodology is F-2fpds [29]
A where p4 is the acoustic pressure field which is a solution to [12), and satisfies boundary conditions at the suppression pool free surface and boundary. To evaluate the submerged structure force, [29) was approximated by F - 2Ep AS . [30) A 1 This approximation is justified provided that the size of the surface area increment AS is chosen such that the acoustic pressure p3 is essentially constant over AS. Nodalizat. ion of the submerged structure surface was performed in accordance with the procedure used in conjunction with the generic submerged structure methodology.11'18'1 4.0 In-Plant Submerged Structure Test At the request of Pennsylvania Power and Light (PP&L), SRI International conducted experiments 23 in the Susquehanna Steam Electric Station (SSES) suppression pool to f- determine loads on cubmerged structures. The experiment consisted of a " balloon" device, teferred to as the bubble source, inflated with a combustible mixture of hydrogen and oxygen. An ignition source would detonate the mixture producing r.n oscillating gas bubble which would rise to the surface of the suppression pool. The Page J.i of ).3
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA Z bA[ /h (J.O. NO. 18138) OBSERVATION REPORT v Observation Report No. _R, OR- 038 Rev. 0 , 1 PART II - RESPONSE CONTINUED oscillating bubble produced (acoustic) pressure waves in the suppression pool similar to those produced during SRV air clearing from an SSES T-quencher. Various locations in the suppression-pool and on the support columns were instrumented with pressure transducers to measure the resulting pressure field. To simulate the SRV air-clearing bubble as closely as possible, the " balloon" device was calibrated at Karlstein in the Federal Republic of Germany in the same test tank in which the /P&L SRV discharge quencher tests were performed. Although, these tests were expressly for the air-bubble oscillation loads, they can be applied to chugging and condensation oscillation since only the nature of the source is different, not the physics of the acoustic transfer function.* To make pressure calculations with the acoustic chugging methodology, a source function S(t) first needed to be generated. This was done in the following manner. The Fourier transform of the solution to [12) can be written in the following form p(xlw) - H(xlx [w)S(z o lw) o [30) where H(xlx o lw) is the acoustic transfer function given by 2 W N (*0
! ") 'N(*!")
H(xlx0 !") ~ 2 [wN ' (" + iAN }2} l l where, as before, t,(x]w) is the flexible-boundary eigenfunction, and t*,(xlw) is the i complex conjugate of the flexible-boundary eigenfunction.** Note that the acoustic transfer function H(xlx lw) o is completely known for a given suppression-pool geometry and fluid-structure interaction parameters. The bubble source was placed between two support columns in the suppression pool at [r o ,fo,z o) - [9.14m,15*,1.09m). The elevation z - 1.09 m was chosen to be equal to the T quencher arms. The measured
- This is only approximately true since the SRV air-bubble rises thus generating a movirg source while the chugging and condensation oscillation sources are fixed in space.
o ** For a complete description of the acoustic transfer function, see Chapter 5 of Reference 12 or 13. Page .lf of ]_i 1
p , q g STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2'IDCA I: (J .O. NO. 18138) M h /f/h i O ~ OBSERVATION REPORT U Observation Report No. .lL OR _Q)1_ Rev. Q_, PART II - RESPONSE CONTINUED pressure 1 response from Sensor P12, located at (r,#,z] - [9.14m,15',0.45m], was
. transformed to the frequency domain to obtain p(xo lw). The source function S(x lw) o was l empirically obtained via [29] by division of both sides of the equation by the known
' acoustic transfer . function H(xlx olw) . This source function in conjunction with the transfer function could then be used to determine the pressure field at any desired
. location.
The comparison
- of the measured pressure time-histories and power spectral densities
'with the predictions by the acoustic chugging methodology at all sensor locations was
. excellent. These test results indicate that x = 1.3 for this specific arrangement of source and' submerged structures, and for a source which essentially contains only a.
9-Hz component. However, we believe it would be imprudent to accept a value for x less than 2 based on this single test with a source which has a completely different frequency content from the steam condensatic,n sources. Nonetheless, these experiments
'do demonstrate that . for x -2 the acoustic chugging methodology does provide a conservative evaluation for submerged structure loads.
O' 5.0 Conclusions' We have~ demonstrated that the application of tne acoustic chugging methodology to the generation of submerged structure loads is adequate and conservative provided that x.
- 2. In addition, a comparison between the theoretical and experimental submerged structure pressures shows that for x - 2 the submerged structure loads generated by the acoustic methodology are conservative. A detailed examination of the application to submerged structure load evaluation by an of the acoustic independent chugging consultant methodolog7' was performed with the same conclusions.
This additional work to develop and license a methodology which would result in significantly lower predicted submerged structure loads was performed outside of the Mark II Owners Group under sponsorship of the Pennsylvania Power and Light Company (PP&L) and the Philadelphia Electric Company (PEco). The methodology received extensive review and scrutiny by the NRC and its consultants. As SSES was the lead plant in licensing, NRC concurrence is documented in t' e SSES Safety Evaluation 2s Report.as The summary of the methodology included in the M., Design Assessment Report is the same as that provided for SSES and its application was identical. NRC Staff i approval of this approach for MS is indicated by their approval of the LCS Safety Evaluation Report without additional comment or question. O respo* T n se.his comparison is proprietary to PP&L and thus cannot be included in this Page S. of 23
STONE & WEBSTER ENGINEERING CORPORATION l LIMERICK 2 IDCA
/" ..
(J.O. NO. 18138) [h(A/h3 OBSERVATION REPORT (- j Observation Report No. _D_. OR- 038 Rev. 0 ' PART II - RESPONSE CONTINUED l
>l' 6.0 NOMENCLATURE A -
Projected cross-sect!cn area of the submerged structure. C3 - Unsteady drag coefficient. l Co Steady drag coefficient. F - Dynamic load (force) due to fluid motion on a submerged structure neglecting contributions due to tangential shear. I Ma - Mach number. Na Hydrodynamic mass, p - True dynamic pressure field which satisfies the proper boundary conditions at the pool free surface and boundary, as well as the submerged structure (s) ((~}/ boundary. p, - Pressure field inside the wake region which forms at the downstream side of a submerged structure. p, - Pressure field given by potential-flow theory which satisfies the proper boundary conditions at the suppression pool free surface and boundary, as well 6. at the submerged structure boundary. Re - Reynolds number. S - Total surface area of the submerged structure. S, - Area of the structure surface indde the wake, j S, - Area of the structure surface outside the wake region where the pressure is given by potential flow theory. St - Strouhal number, u - Acoustic fluid velocity. _ U.(t) - Uniform flow velocity. A p V3 - Acceleration volume. V Vs Structure volume. Page .Q of 2.4
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA l (J.O. NO. 18138) ([ / f[3 L OBSERVATION REPORT
'g- -,
1 i V Observation Report No. .JL OR 9.16 Rev. ,_,Q. l FART II EMPONSE CONTIETED p - Fluid density, s - Ratio of the acceleration volume to the structural volume. 6 - Velocity potential, 9,(xlw) - Eigenfunctions of the Helmholtz equation V2 ,, + g,2 - O for a suppression pool with flexible boundaries. ()2 - d'Alembertian operator ( []2 , 9 2 - c -22jg 3 g2 ),
7.0 REFERENCES
- 1. H. Schlichting, Boundary-Layer Theory, sixth edition, McGraw Hill Book Company, New York, 1968, p 104 ff.
- 2. Ibid, p 22,
- 3. L.' D. Landau and E. M. Lifshitz, Fluid Mechanics, Pergamon Press, London, 1959, p 169.
- 4. 'F. J. Hoody, Forces on Submerged Structures in Unsteady Flow, Proc. ANS Topical Meeting on Thermal Reactor Safety, July 31 - August 4,1977, Sun Valley, Idaho, p 3-516 ff.
- 5. Ibid, p 3-592.
- 6. H. Schlichting, op cit, p 17.
- 7. L. D. Landau and E. M. Lifshitz, op cit, p 19.
- 8. Ibid, p 34.
V Page d of 2.3
I STONE & WEBSTER ENGINEERING CORPORATION l LIMERICK 2 IDCA (J.O. NO. 18138) / } ( OBSERVATION REPORT
'v Observation Report No. ._D_ OR- 038 Rev. O FART II - RESPONSE CONTINUKp
- 9. . L. M. Milne-Thomson, 2*heoretical Hydrodynamics, 5th ed. , MacMillan Press 1.td. , London, 1968, pp 246-247,
- 10. H. Schlichting, op cit, p 32.
- 11. , Mark II Containment Lead Plant Program Evaluation and Acceptance
. Criteria, NUREG 0487, Division of Systems Safety, Office of Fuclear Reactor Regulation, U.S. Nuclear Regulatory Commission, October 1978,
- 12. , Mark II Containment Lead Plant Program Evaluation and Acceptance Criteria, NUREG 0487, Supplement 1, Division of Systems Safety, Office of Nuclear Reactor Regulation, U.S. Nuclear Regulatory Commission, September 1980.
&r5- 13. C. Anderson, Mark II Containment Lead Plant Program Evaluation and
'~
Acceptance Criteria, NUREG-0808, Division of Safety Technology, Office of Nuclear Reactor Regulation, U.S. Nuclear Regulatory Commission, August 1981.
- 14. R. J. Ernst, T. C . Peterson, and G. H. Salas, Mark II Pressure Suppression Containment Systens -- Loads on Submerged Structures, NEDE-21730 Class III, General Electric Company, December, 1977, (Company proprietary).
- 15. C., K. Ashley II, N. M. Howard, E. Rabin, An Approach to Chugging, Mark II Task A.16 Improved Chugging Methodology, Rev. 1, San Francisco Power Division, Nuclear Engineering Staff, August 1980.
- 16. , Mark II Improved Chugging Methodology, NEDE 24822-P, General Electric Co., May 1980, (Company proprietary), Prepared for the Mark II Utilities Owners' Group by Bechtel Power Corporation under contract with the General Electric Company.
- 17. Ibid, p 3-7 ff.
S. Eskinazi, Principles of Fluid Mechanics, 2nd ed., Allyn and Bacon, O' 18. Inc., Boston, 1968, pp 302-306. Page 8 of 23
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J .O. NO.18138)
~
/3 OBSERVATION REPORT Observation Report No. 1 OR 038 Rev. O PART II - RESPONSE CONTINITED
- 19. G. K. Ashley II, et al, op cit.
- 20. , Generic Load Definiclon Report, NEDE 24302-P, General Electric Company, April 1981, (Company Proprietary).
p
- 21. H. F. Bauer, Tables and Graphs of Cross Product Bessel Functions, MTP-AERO-63-60, George C. Marshall Space Flight Center, June 24, 1963.
- 22. E. Elawar, Mark II Chugging Loads, Force on Submerged Structure, Nuclear Staff Calculation AP-81-162, Bechtel Power Corporation, July, 7,1981.
- 23. C. R. Abrahamson and A. Hashemi, SSES In-Plant Tests to Measure Submerged Structure Loads and Pool Frequencies, SRI International, Project FYC 5881-p 50,. April 1980 (Pennsylvania Power & Light Proprietary).
U
- 24. J. M. Healzer, Submerged Structures Load Study, SLI - 8116, S. Levy, Inc. ,
September 1981.
- 25. , Safety Evaluation Report Related ?.o the Operation of Susquehanna Steam Electric Station, Units 1 & 2, NUREG-0776, Supplement No. 2, Office of Nuclear Reactor Regulation, U.S. Nuclear Regulatory Commission, September 1981.
-26. Limerick Generating Station Units 1 & 2, Design Assessment Report, Philadelphia Electric Company, Section 4.2.2.5
- 27. Susquehanna Steam Electric Station Units 1 & 2, Design Assessment Report, Pennsylvania Power & Light Company, Section 4.2.2.5 O
v Page U of @_i
A STONE & WEBSTER ENGINEERING CORPORATION LINERICK 2 IDCA . [pC A / p/3 (J.O. No. 18138) z OBSERVATION REPORT Observation Report No. D OR- 038 Rev. O PART II - RESPONSE CONTINUED l i I i l l [ SIGNATURE PAGE ONLY] O l
'l i
I l
===========sss=ss== sass ========================================sss===================
If PEco Review Signature / / Da'te G Bechtel/ Approval# Signature / Date Page.)_[_of d4 { L 1
' STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) q OBSERVATION REPORT l Q) . .
Observation Report D CR- 038 Rev. O Review Plan: LK- D- 1903A - MC Rev. I - PART III - RESPOE3" M MiATION The response to this observation report has been evaluated as follows: A. Causal Factor (s): Acceptable Not Acceptable (E, plain): l N/A B.- Extent of Condicion: Acceptable Not Acceptable (Explain): N/A
-C. Significance: Acceptable Not Acceptable (Explain):
N/A []J L . D. Corrective Action: Acceptable Not Acceptable (Explain): N/A E. Action to Prevent Recurrence: Acceptable Not Acceptable (Explain): N/A
. . .C.O.N.T.I.N.U.E.D. .. .O.N. .'1W0. .P.A.C.E.S. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Additional Action Required: X No Yes (Explain)
, ~g.......................................................................................
U Rh),JaL " 4Ney 9 (L K & h wille- 4/s/c9 Liad Engineer Signature / Date apt, Signature / Date Page M of 3
STONE fu WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. No. 18138) A OBSERVATION REPORT Observation Report DOR- 038 Rev. O PART III - RESPONSE ETALUATION CONTINUATION: DOR-038 was written because of the concern on the design load adequacy for piping submerged in the suppression pool _during a LOCA event. Bechtel's justification is that the methodology was previously accepted by the NRC. Specifically, Bechtel stated in the conclusion section of the response, "the summary of the methodology included in the LGS - DAR is the same as that provided for SSES and its application was identical", and, "NRC concurrence is documented in the SSES Safety Evaluation Report". Since Bechtel's LOCA submerged structure load methodology is equivalent to the NUREG 0487 accepted method, except that the standard drag, the lift force, and the interference effect on acceleration drag a e omitted, this implies that the NRC has accepted for SESS; therefore, for other Mark II plant applications, that the aforementioned effects, namely, standard drag, lift and interference effects can be ignored. The equivalence of the methods can be established using the following simple derivations. Bechtel's LOCA submerged structure load methodolgy is to use the equation (1) F_=2fP A d S_, where PA is the acoustic pressure field without the presence of the submerged structure. It is noted that, using Green's theorem, the surface integral given by Eq. (1) can be converted to a volume integral F=2[tfp dV (2) Since in the acoustic theory, V p =fU_', Eq. (2) becomes F,=2hU' dV (3)
~ 2Eg U' v Comparing Eq. (3) with the industry and the NUREG 0487 accepted method.,
F_ = Cg g U' V + 1/2 Cp q (Ul U_ A + 1/2 C L glul 2 A$ (4) shows that Bechtel's method is the same as the NUREG 0487 accepted method but omitting the flow-induced vibration /lif t effect, ignoring the standard drag term, and setting Cg = 2, which is the same as ignoring the interference effect on acceleration drag. However, since the method has been reviewed and approved by the NRC [1], no further action is considered necessary. To implement Bechtel's methodology, it is advisable to use Eq. (3) or ]
/O t
Eq. (4) for the following reason: The surface integral contains the unit surface vector, therefore using7_p A_ will cause a cancellation effect, q and many more eigenfunction terms are required to obtain convergence. CONTINUED ON ONE PA3E pgEg p - 3 j L _ _ _ _ _ _ _ _ - l
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA i (J.O. No. 18138) OBSERVATION REPORT Observation Report DOR- 038 Rev. 0 PART III - RESPONSE EVALUATION CONTINUATION: On the other hand, as is shown above, the force given by Eq. (4) reduces l to the acceleration drag and, since acceleration drag constitutes the major submerged structure force contribution during CO and chugging, one would expect comparable load magnitudes to be obtained whether the load is calculated using the acoustic method Eq. (2) or using the industry accepted method. In fact, one would expect to 'see the same calculated loads when _ the standard drag and the interference effect are removed from l the industry accepted method. The load magnitudes obtained by the Bechtel l methoo appear to be about 1/10th the magnitude obtained by using the NUREC 0487 accepted method. This difference in loads, which results from implementation on Limerick 2 of the Fechtel methodology, has not been completely explained. Omission of the standard drag, lift, and interference effects can account for only a fraction of the discrepancy. Specifically, convergence of the eigenfunction series in the algebraic sum approximation to the vector integral given by Eq. (1) remains unresolved. However, the project's position is that the methodology is identical to that approved by NRC for SSES. Therefore, the issue was not pursued further.
REFERENCES:
[1] Safety Evaluation Report related to the operation of Susquehanna Steam Electric Station, Units 1 and 2, NUREG-0776, Supplement Number 2, Office of Nuclear Reactor Regulation, U. S. Nuclear Regulatory Commission, September 1981. O Pap B O
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) r i OBSERVATION REPORT i Observation Report No. DOR- 039 Rev. L l Review Plan LK- D- 1903 - E Rev. 1 Reference AI No.: DAI-099 _ l PART I = INITIATION
- 1. Description of Concern The' existing thermal overload (0.L.) heater selection procedure and implementation as outlined in Calculation 6300E.13, is deficient it could lead to spurious tripping of the Class 1E motors (non-MOV's).
- 2. Supporting Information The ' deficiency exists since the procedure does not properly address the effects on the 0.L. heater sizing for the following conditions:
- a. Operation at Minimum Operating Voltage The protection for operation at minimum sustai.2d operating voltages is provided by the degraded undervoltage protection scheme. The undervoltage scheme, outlined in Calculation ;
[' 6300E.23, page 3, states that the minimum undetected 480V
! voltage may be less than 90 percent of the motor rated voltage.
To maintain - 90 percent at the '480V level re' quires a minimum of 92 percent at the 4KV bus. However, undervoltage relay DEV 127Z is set to drop out at 90 percent, i.e. 2 percent below the required minimum. Any tolerances due to calibration, D.C. variation, temperature changes, and repeatability could lower the detection point further. Therefore, an undervoltage condition of 90 percent at the terminals of the motor control center supplied motors (460V base) is possible and, indeed, is acceptable since 460V motors are designed to run continuously at 90 percent of the rated voltage, and can perform their safety related function at 90 percent or above. However, the 0.L. heaters must be sized for this undervoltage condition to preclude spurious tripping. CONTINUED ON 1 PAGE Additional Documents Attached: None n ................................... ..................... ....... . .............. .. xn O PF/ Lead Engineer Signatifuire
+ /'Date
/n APM Sigftature
(
'/ Date Page 1 of 1
STONE & WEBSTER ENGINEERING CORPORATION i LIMERICK 2 IDCA' l _ (J.O. NO. 18138) i
- i. OBSERVATION REPORT i i Observation Report No. DOR- 039 Rev. 0 l PART 1 - INITIATION l CONTINUATION:
Based on the above, implementation of the existing procedure could result in spurious tripping as illustrated by the following example: Upon use of the overload heater (0.L.) selection q tables in the calculation, it is possible to select a coil number 1017 for a motor with a full load current I (FLC) of 12A. Consequently, this motor will draw 13.33A (12.0A/0.9) at 90 percent voltage. Therefore, this 0.L. selection could result in a motor trip in approximately 420 r*conds (7 minutes).
- b. O.L. Heater Tolerances Thermal 0.L. heaters have a minus 10 percent tolerance, established by the manufacturer, which should be addressed
- when sizing the heaters.
Referring to the example in part a , for the motor with a FLC of 12A, using a number 1037 heater coil, and incorporating the minimum tolerance, the minimum trip value is 106.3 percent times the tolerance 10.2A X 1.25 X 100 12 = 106.3A 106.3A X 0.9 = 95.63 percent This indicates that the heaters could trip below the full load rating of the motor, again resulting.in spurious tripping. The effect of the tolerance will cause the heater to trip in a shorter time than the time due only to the undervoltage condition, i.e. tripping time will decrease below 7 minutes. The example for the coil number 1037 in the above is for illustration only and may not define the worst case. O Page d of T
STONE & WEBSTER ENGINEff.ING CORPORATION LIMERICK 2 IDCA gg y (J.O. No. 18132) (] ossam m los axreer Observation Report No. (5 - 07M) D OR- 0 3 9 Rev. 1 FART II - RESPOBBB
- 1. Observation Concurrence:
AMENDED RESPONSE: Following is the amended response per SWEC 3/18/89
" Telephone Conference Items related to Bechtel Response to DOR-039" and per clarifications at a meeting in San Francisco between SWEC, Bechtel and PECo on 3/21/89.
Concur with observation X Do not concur with observation (Note: if not in concurrence, explain in " remarks" below)
- 2. Response to Observation A. Causal Factor (s)
,_ B. Extent of Condition ws C. Significance D. Corrective Action E. Action to Prevent Rect rence
==============================================================
Remarks (Calculation 6300E.23, Page 3, does not contain the statement that the minimum undetected 480 V voltage may be less than 90 percent of the motor rated voltage). The calculation used for the selection of thermal overload heaters is Calc. 6900E.13 (EE-84 LGS), not 6300.E13 as stated in Part I.l. The calculation is not deficient for the following reasons: 14 Operations at Minimum Operatino Voltages Per the requirements of BTP PSB-1, it was necessary to identify the limiting ratings of loads connected to the Class lE buses and to provide an undervoltage scheme that will prevent sustained operation outside of that rating. fg That analysis (Calculation 6300E.23 Rev. 2) identified a ( ) level of 90% on the 480V MCC as the limiting voltage for component operability. The undervoltage relays on the 4kV buses are set accordingly. (Cont.) pag,,3,og { T-20/3-1
STONE & WEBSTER ENGINEERING CORPORATION ! LIMERICK 2 IDCA (J.O. NO. 18138) osssavaTIou REroer Observation Report D OR _0 3 9 Rev. 1 CONTINUATION: 1 The voltage regulation studies have shown that for all l anticipated grid conditions, and for all plant design basis events, that the voltage levels within the plant will be restored to 100%. Therefore, to impose an artificial value of 90%, based on the equipment manufacturers rating, is not consistent with the anticipated voltage levels. The example cited for compensating for a 90% voltage condition is not necessarily correct for all loads. For some motors, a decrease in voltage will result in a decrease in current, whereas, if you assume a constant output, then a decrease in voltage would result in a non-linear inverse proportional increase in current (factoring in motor inefficiencies). Each motor and its connected load would have to be analyzed individually, however, the point is moot since there are no sustained voltage degradations.
- 2. Overload Heater Tolerances The following further substantiates the reasoning for not O including tolerance in the sizing of the TOL.
The selection criteria and procedure for thermal overload (TOL) as outlined in calculation 6900E.13 are based on the selection table for motors with 1.15 service factor (SF) in Cutler-Hammer publication No.15412, Figure 27. For motors with 1.0 SF with the given motor nameplate full-load Amps (FLA), identify the heater coil number where motor FLA f alls between the minimum (Imin) and Then, the maximum (I based on t$ax) amoere range e Cutler-Hammer selection in the selection table. the guide, select heater coil one size smaller with the coil mounted in the same position as the larger coil for motors with SF 1.15. The ultimate tripping current is determined by the product of Imin x 1.25 (but not to exceed 115% of motor FLA). Note that the Cutler-Hammer publication makes no mention of heater coil tolerance. The tolerance of the TOL is not a constant value. It is normally distributed and is meaningful in a statistical sense only (see Transmittal S-0582). O (Cont.) T-20/3 Page 3 of T_ i I
' STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA g /%y (J.0, No. 18138)
OBSERVATIGE REPORT Observation Report No. D 0R-03 9 Rev. O CONTINUATION: Additionally, there are other compensating factors present in the design that should neutralize the effects of negative i heater coil tolerance. Some of the factors are: a) Positive margin in motor horsepower rating by selecting 1 standard motor size (larger than brake horsepower) due to normal design practice b) Actual ambient temperature (where motor control centers are located) usually less than the design ambient at 40*C The superimposition of an additional 10% of " margin" on the selection of the thermal overloads would essentially render the motors unprotected.
- 3. Summary Operation of 460 V continuous' duty motors at the sustained 90% voltage is not considered since
~
Both offsite power sources (Station Auxiliary Transfor-mer and Safeguard Transformer) are each equipped with automatic load tap changers which closely regulate the 4.16 kV bus voltage at 1.03 p.u. of 4.16 kV and thereby maintain the MCC bus voltage at 1.065 p.u. (1 p.u. = 460V). The diesel generators are equipped with a fast acting regulator to maintain the same 1.03 p.u. of 4.16 kV bus voltage The effect of the negative tolerance of the TOL heaters is not spe'cifically considered as a general requirement since it is neutralized by Design positive margin usually available in electric motors (motor horsepower greater than brake horsepower of driven equipment)
- Operating current of the motor usually below its nameplate rating due to higher (than nameplate) operating voltage applied
- Actual ambien' temperature of TOL heaters usually below
-- desi n ambient temperature
; //f[lhWh
'b Signature Sl22ldi
/ Date YAkklUL Bechtel Approval Signature / / Date PECo 1-20/3 Pagej[ofj[
L-_-_ _
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) onsxITATION REPORT Observation Report D OR- 039 Rev. L Review Plan: LK- D - 1903 -E Rev. L PART III - ERSPONSE EVALUATION The response tc this observation report has been evaluated as follows: A. Causal Factor (s): Acceptable Not Acceptable (Explain): SEE ATTACHED SHEETS B. Extent of Condition: Acceptable Not Acceptable (Explain): SEE ATTACHED SHEETS C. Significance: Acceptable Not Acceptable (Explain): SEE ATTACHED SHEETS D. Corrective Action: Acceptable Not Acceptable (Explain): SEE ATTACHED SHEETS E. Action to Prevent Recurrence Acceptable Not Acceptable (Explain): SEE ATTACHED SHEETS
. 9"H!P!P. P!. . J."D. !8. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . - . . . . . . . . . .
c Additional Action Required: No X Yes (Explain) See Page 3 of this evaluation. O -/WLhas Lead' Engineer Signdture e im
/'Date 4%A & 0. wale.
APH Signature 4in/n
/ Date Page k of 1
L STONE & WEBSTER ENGINEERING CORPORATION ( LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT Observation Report DOR-- y Rev. A PART III - EESPONSE EVALUATION CONTINUATION: A. Causal Factor (s): Acceptable The causal factor is the present Class 1E 480V Motor Control Center (MCC) Thermal Overload (TOL) design criteria which does not incorporate minimum trip level considerations due to a 90% MCC minimum acceptable voltage- and a TOL tolerance of +0%, -10%. This selection criteria could cause spurious tripping of Class IE 460V continuous duty motors during specific operating scenarios. B. Extent of Condition: Acceptable The extent of condition concerns the present selection of all Class IE MCC supplied continuous motor TOL sizes. C. Significance: Acceptable The significance of the present selection criteria. could cause potentially undersized TOL's leading to the possibility of spurious tripping due to the degraded voltage scheme which ensures a minimum O 90% voltage and the lack of consideration of the -10% TOL negative h tolerance, as confirmed by the MCC vendor. Selection of-each TOL should ensure the availability of its continuous Class 1E 460V motor under all conditions of motor operation, including the existing degraded voltage scheme. D. Corrective Action: Not Acceptable { The project has stated that the present TOL selection ' criteria does not take into account the tolerance or sustained 90% voltage for the following reasons:
- 1. As outlined in Calculation 6900E.13, TOL sizes were based upon a Cutler Hammer selection table for motors with a 1.15 Service Factor (SF).
- 2. For motors with a 1.0 SF, with a FLA between the maximum and minimum ampere range, the heater coils were selected one size smaller than . the larger coil for 1.15 SF motors. Using this method, the ultimate tripping current was determined by the product of the minimum current x 1.25 which would not exceed /
115% of the motor FLA.
- 3. The TOL tolerance is nc' a constant value as it is normally '
l distributed and meaningful in a statistical sense only.
,O CONTINUED ON ONE PAGE Pa$m -
7 dT
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA i (J.O. NO. 18138) OBSERVATION REPORT l [\ Observation Report DDR- OJ9Rev. O l PART III - RESPONSE EVALUATION CONTINUATION:
- 4. Motor horsepower rating was selected with a positive design margin by using motor sizes larger than brake horsepower.
- 5. The actual ambient temperature for MCC locations and its TOL
. heaters was less than design ambient of 40*C.
- 6. The regulation of both offsite power sources, usinE the automatic load tap changing capability, will maintain the MCC bus voltage at 1.065 p.u. (1 p.u. = 460V) and the diesel' generators will maintaiu 1.03 p.u. of the 4.16 KV bus voltage.
- 7. The motor operating current will usually be below its nameplate rating due to the higher (than nameplate) operating voltage applied.
We agree that there are additional margins not incorporated in the sizing criteria and that minimum operating' voltage should occur only rarely, both of which provide some basis for not considering the
) effects of heater tolerances and low voltages in the design U calculation. However, we consider that to provide additional -{
confidence that sufficient margin exists, the calculation should address this condition ' or, as an alternative, actual field testing of these heaters should provide objective evidence that the design accommodates the low voltage operating conditions and heater tolerances.* E. Action to Prevent Recurrence: Not Acceptable
)
The calculation should be revised to incorporate criteria which includes both the 90% voltage and the TOL negative tolerance or the field testing of these heaters should provide objective evidence that the design accommodates the low voltage operating conditions and tolerances.
~
*In discussions between the project and SWEC, the project has not committed to any corrective action since it considers its selection criteria adequately addresses the effects of heater tolerance and low voltage.
O P"J e
%dl
\
l _ _ _ _ _ _ _ _ _ . _ _ _ _ _ . I
O LIMERICK GENERATING STATION - UNIT 2 INDEPENDENT DESIGN AND ODNSTRUCTION ASSESSMENT O DESIGN OBSERVATION REPORTS DOR-040 TO DOR-079 STONE & WEBSTER ENGINEERING CORPORATION
1 !. STONE & WEBSTER ENGINEERING CORPORATION 1 LIMERICK UNIT 2 - J.O.NO. 181T8.00 ) INDEPENDENT DESIGN AND CONSTRUCTION ASSESSMENT pc IDCA VOLUME II BOOK 3 .
.( f DESCRIPTION J
1 This Volume II Book 3 contains the Design Observation Report (DOR) ' packages for the DOR-040 to DOR-079 issued by SWEC during the Limerick 2 Independent Design Assessment (IDA), Each DOR package contains ] ' i the following: { I o DOR PART I: SWEC Initiation o DOR PART II: ! ehtel Response t l o DDR PART III: SWEC Response Evaluation (T g,,). Where necessary, selected reference documents have been included with the DOR packages to insure a clear understanding of the DOR and its resolution. l l d,--
-l Sf0HI AfD WEBSitt !!GINIttING C0tP0tifl0N p
Q LlHIRICI DNIT 2 - J.0.10.18138.00 INDEPINDElfDESIGNANDC0HSf200f106ASSESSEINT J 1 1001VOLDEIIIB0013 DISIGNOBS!tilfl0HHIP0Rt i TABINDt! l l l l0BSlliltl l 1 l TAB ll!P0Hfltl f!!LI/DISCilffl0BOfOBSiliATIONREP0tf l ) lN0.lNUEBitlfl l I l l l l l !
!.....l.......l..l...........................................................................................
' ' 40 l DOR-040 0 CALCS 23.1 & 17-1-3 OHIfTHIREALLOADS(OPillf!IG&ACCIDtif)FR0HDIS!GIOfSTRUCfERES l '41 lD02-041l0 l0flLI!ATION Of f!EPitif0lf LISS THAN fit POSTULif!D ftHPitifDit FOR fit Atta l l 42lD08-042l0l CALC 63801.08(DGTOLT!!GSfDT)00EflIIDISCitPARCIISTHATHattSIfINCONCLUSIft l i l 43lD0H-043l0lCALCS/8031-2302 IIP 0lIITIALf!EPDICafOfPIPIWALLAtRPi10ZILEIS50fCOISitfATIft l l 44 lD02 044!0 l CALC H-51-9 F0HHULA IS IIAPPROPHIAft & INPUTS Att INCONSISftEt WIfl AS-BUILT PIPIEG l l 45 lD02-045l0 l CALC 2006 5tfiOD f02 HELB BL0fDOWI 18001515f117 & LESS CONSilfATIft THAI INDUSftf STD !
l' 46 lD01-046l0 l CALC 2006 IIPUf5 UfDtt!Sf!BAft itLB BLOWDOWIILON, & COEPitfHilf PitSSitt & ftEPitATURE l l 47 lD01-047l0 lftsf IIP 0lf LACIS tilDilCI filf 10S!E00tf IEIffERS WERE iiiOSID TO AGING T-H 08 S-f LOCA l l 48lD02-048l0l$PECC-115DESIGHPitSSUltSGIft!ASSf!DCTURALDESIGICRItttIADiffitFR0HSPICH171 l l'49lD02-049l0lDISIGilf!DPOStACCIDiffINDICATORSLf-61-215&-235 TRIPfl0BCLASS!!BUSUP05LOCA ! l 50 lD02-050l0 lf0 CALCS QUALIff SUPP08f5 Of PIPIIG IIf TIDING SEISHIC ttQUIttutifs 70101-SEISHIC !!GION l O
\_/
l 51 !D02-051l0 lDISCllPallCf BEff t!I iSAk 00HElfElit 70 20-1.47 (QUESfl0B 421.33) AND DWGS t-360 & t-622 l l 52 lDOH-052!0 l CALC DLA-212-Cl2 0 HITS TCt CS 2208-0 fall AffACHE!Hf LOADS & fCR DID 50f RiftttICE CALC ! l 53lD02053l0lQ-FBICTUBIIGSUPP0tfGUIDifCI10fAPPLIEDCONSISfilfN/ftSTINGflafF0HHSSUPP0tfBASISl l 54 lD02-054l0 l CALC GBB-203 C12 DOES 107 QUAL!!! hcl CBB-203-112 WILD,AP0filflALDESIGEDilICIENCY l l 55lD01-055;0lCALCGBB-219-078 SICTIONB0DULOSTALUISSWITCHID,FIELDTOL50fAPPLIEDTOStifGANGLE l l 56 lD0H-056l0 lIAGLI SIGIAL flHit DillHIC CAPABILIff SIGNIFICAlfLY L0ftt flAl STAftD Bf GI & LGS SQRt l l 57 lD02 057l0 lPIPI SUPP0lf LOADIIGS Of IHBIDDED PLAf t$ Hlf E0f Ilit St!I SUfflCItifLY ttillWID l l 58 l DOR-058l0 lflLLif WILD 01 DWG C-942 SHOWI SHALLII THAR CALC 17-A-3 itQUltID, COULD Bt OftBSittSSED l l 59 lD08-059l0 lH0DULUS Of ILA$f!CIff II SIISHIC ANAL Of h BLDG & SPIAT POND POEPE00S! Att INCOESISftlf l l 60 lD01-060l0 l CALC 17-I-17 OHITS DEAD & Lift LOADS, OSIS 50E-PEAK tiffitt LOA)$ & OffALIDAftD IIPUTS ! l 61lD01-061l0l CALC 25.91TORITALOlf!0IOfPOSSIBLEDifMILLHEADDROPHafOSE80EAPPLICABLEMtfiODS l
.l 62 lD0H-062l0 lIQ PACKAGI D-102 LACIS tilDIXCE filf TALf! 2l017B CAN f5f0fl0i ADER SEISEIC/EID201,0 ADS l l 63lD01063l0l$0HIIllGERCALCSLACISPICC-97SIISEICitSPONSISOHHAfl0N&LOADELIEllATIONBASIS !
! 64 lD02 064!0 lDISIGI Of 50102 AILS II CALC 22.3J-l DOES 10f APPIAR TO Hitt DESIGH CRIERIA REQUlttitifS l l 65 lD01-065l0 l CALC H-59-7 USIS 10 THAL DW PitSSUl! TO SI!! ADS HIfR0Gli SUPPLY STSftH; HAT 507 Ettf ISAR l l 66 lD01-066l0 l12' 90011071 Hi-51-25017B, CONDOIf Elf Bt ACTIHG AS !!GID SUPP0tf IHTALIDif!IG QUAL '!
l 67lD0H&7l0lSPICSP-401&402 DISft!B0ftLOADS5050BtfWEEEPAlllDSNUBBERSC0Hftatt70STDit-IT-97i l 68 lD0H-068l0 l CALC 2-10-84 008fII0!S TO CONTAIK StitRAL (22025 Affit AS-BUILT llc 0fCILIAfl0i l l 69 lD01069l0 lCALCS 101.73 & 22.8.B OElf PIPI SUPP0lf AffACHHilf LOADS & fl0 Lift HIf EDGE DISt& ICE l l 70lD0H-070!blCALCS 2-10-83&2-1084!!!015 HAT Glit 50EC0XStilflit ACCELitafl0F PIPI StitSSES l l 71 lD0H471l0 lGI HHA !!Lif SEISEIC CAPABILIff Dif!IID Af LOWit LITILS THAf Il DihGH t!P0tf l
'l 72 lD01-072l0 lfDI SIGNID 00EPLift, BUT E0D!flCAfl0HS TO CORRICf Dif!CIIIT 00fDIfl0E NOT COMPLtitD l l f3 lD02-073!0 lCALCS S102.1 & S102.2 0 Elf SIGN!f! Calf Iff0f5 & FAIL TO CLEARLY Statt DESIGN ASSUHPfl0f5 l l 74 lD01074l0 ! CALC 5102.1 OHITS SIGK!flCANT INPUTS & FAILS TO DESCRMI HOW C0KilG0taf!0f IS 50 DELED l l 75l DOR-075l01811LIMIf01SillIGHTTUBI20580fSPICIFIED15f0BtANCE02INSTALLATIONCt!TitIl l l 76 lD01-076!0 lINSOfflCIllf CUf-Off it!QUINCf EAT HAi! Bit # OSID TOR f!Et HISi0tf FLOW TIANSIllt AHALYSISl l 77 lD04 077l0 li!ND02 HANUAL I0f tifl!WID & APPR0ftD AS !! QUI!!D BT SPIC H-123 l
(" l 78lD02-078l0l CALC 63001.18 t!SULTSIICORHICfLfIKCORP01(f!DINTOftANSf0REttLOADINGSUEtattTABLIS l l 79 lD02-079l0 l CALC 6300!.18: ADIQUACf Of SAftGUAID M d f0RE!H RIQUlt!S F0ffitt SUSTAKf!ATION l
i l-l
- j. STONE & WEBSTER ENGINEERING CORPORATION L
LIMERICK 2 IDCA (J.O. NO. 18138) k OBSERVATION REPORT Observation Report No. D OR- 040 Rev. O Review Plan LK- D- 1903 - S Rev. 1 Reference AI No.: DAI-213 DAI-434 PART I - INITIATION 1.- . Description of Concern No ' evidence was found indicating that thermal loads (operating and accident) have been considered in the design of structures.
- 2. Supporting Information
- In none of the calculations reviewed were thermal loads considered, although they are included in load combinations given in the FSAR and design criteria. Although these loads often are insignificant and need not be considered (e.g. interior concrete walls and slabs or- bolted steel members subjected only to operating temperature loads), instances where it is not obvious that these loads are negligible were found. .For example:
O- o In the design of the 'D' line wall below subgrade V (calculation 23.1 (b)), operating thermal loads were not considered. Response to Action Item DAI-213 states that these loads are not significant, because the diesel generator building is adjacent to the 'D' line wall, and that for other walls belok grade the soil ' moderates' the temperature gradient. It is also stated that above grade exterior walls do include these operating temperature gradients. However, it has not been demonstrated that temperature gradients are. always negligible for below grade walls. For example, a four foot thick wall with a 35*F gradient (90*F to 55"F) experiences a thermal moment of about 120 ft-k/ft if it is assumed to be uncracked. Even if this moment is reduced by a factor of three to account for cracking, the moment of 40 ft-k/ft is not insignificant. 1 (Note that the vertical iesign moment for the 'D' line wall is only 282 ft-k/ft). Futhermore, ic was not observed r in the calculations reviewed that above grade exterior CONTINUED ON 1 PAGE me.. .mm.......... ...........mm.......................m...p.........m........m.==..... Additional Documents Attached: None, p .........mmm.................... .. .. ...................... .... ................. C. C L /z b t/P/ /Hb/Ahl Lead Engineer Signature /'Date APM Signpfure / / Date i Page 1 of E
STONE & WEESTER ENGINEERING CORPORATION LIMERICK.2 IDCA (J.O. NO. 18138) (r OBSERVATION REPORT Observation Report No. DOR- 040 Rev. O PART 1 - INITIATION CONTINUATION: concrete walls and roof slabs con' sider thermal gradients. For example, calculation 22.2C(h), which designs the 'D' line wall above grade, does not appear to consider these loads. In calculation 17-A-3 (drywell interior steel framing at El. 252'), thermal loads are never considered, even though the temperature can reach 340*F. Response to Action Item DAI-434, states that thermal stresses in this framing are not significant due to the design of the connections which allow for thermal growth. However, it is not agreed that the intermediate cross members do not require consideration of thermal stresses. There are connections which allow little local distortion (e.g. Section C and H drawing C-927). For connections using clip angles, distortion will be low in the axial direction. b (f i i O Page 1 of 1 i
{' - l l STONE & WEBSTER ENGINEERING CO AI159 ~ ' rN LIMERICK 2 IECA gm/ OC NN
-]
is (J.O. to. 18138) ggg l OBSERVATION REPORT DATE M 3 'm Ob rvation Peport No. COR-040, Bev. O' PART II - AMENDED CCMPILTE RESPONSE A w r
- 1. Observation Concurrence: REISSUED '
i DATE MAR 16198 Concur with observation y --
. , _ fg ._
/ JJ /
X Do not concur with observation (Note: if not in concurrence, explaM in " remarks" below)
- 2. Reponse to Observation: _-
A. Causal factor (s): N/A B. Extent of Concition: N/A C. Significance: N/A D. Corrective Action: N/A
.O E. Action to Prevent Recurrence: N/A V
........................................=.........................................
Remarks: We do not concur with tne Observation Peport that themal loads (operating and and accident) were not considered in the design of structures. ~hermal loads are of ten insignificant and in such instances need not be specifically documented in the design calculations. However, for structures where tnese loacs are significant, tne ef fect of tne thermal load are incluced in tne design (i.e. Calculations 17-R-5 t.e.rougn 17-R-7 was performed for the box beam themal study of containment drywell platfoms at El. 272' , 286' , and 296' , respectively. Calculations 13A and 19-L-1.2 were performed for the themal analysis of the containment structure, etc). , A detailed response to the items cited by SWEC as supporting infomation is provided i as follows:
- 1. Peactor Enclosure 'D" line wall.
In the design of the "D" line wall, themal loads need not be considered t (calculation-th! (b)) because of the following: Accident Temperature: In the design -f concrete strJCtures, rapid temperature fluctuation has an insignificant effect due to the low themal conductivity of concrete. O Page 3 of T
o &#/76 STONE & WEBSTER ENGINEERING CORPORATION ) LIMERICK 2 IDCA J (J.O. NO. 1813) OBSERVATION itEPORT 1
- \
Observation Report No. DDR-040, Rev. 0
]
PART II - RESPONSE COtTTINUED l I' thermal expansion presented above. Hence, the intermediate cross members are structurally adequate for the plant accident conditions. It is apparent that for a conservative containment drywell operating temperature change of 85 F (65 F to 150*F), the-thermal effect on the
- intermediate cross members is insignificant.
Two typical and a specific types of framing connections for the floor framing system at El. 252' have been reviewed for thermal effects. All three types have been found adequate to accommodate the loads due to thermal effects under accident temperature of 340 F. The three types of connections evaluated and their behaviors are discussed below:
~
(1) 03nnections as shewn on Sections G and H of Drawing C-927. Qp - Sections G and H are used for only one cross beam (W14X43) which is rigidly connected to Box B3am #11 and Box Beam #12 near plant azimuth 315 . At Box Beam #11, all other cross beams are supported on the top of the box beam via a C8 channel section, and are connected using bolted connections '(Section C of Drawing C-927). The cross beams, therefore, offer negligible lateral restraint to Box Beam #11 and the bolted connections provide inherent relief for themal effects. It is apparent that portion of the cross beam thermal expansion will be accommodated i by lateral displacement at Box Beam #11. At Box Beam #12, all other cross beams are framed into the box beam near the top flange. The cross beams and the sliding connection at the containment wall, therefore, offer minimal rotational restraint to Box Beam #12. It is concluded that the remaining thermal expansion will be accommodated by rotation deflection of Box Beam #12. (2) Connection as shown on Detail 3 of Drawing C-942, Sheet 1, and (3) Connection as shown on Detail 4 of Drawing C-942, Sheet 1 Connections (2) & (3) are the typical welded connection used for attaching cross beams to the box beams where the top of steel elevation is the same. At these typical connections, the cross beams are attached to the webs of the box beams. It is concluded that the thermal growth of the cross beams will be accommodated by bendirg deficction of the rels.cively O flexible web of the box beams. b l l Page 4 of f
.a . -
STONE & WEBSTER DIGINEERING CORPORATION O ( bC'Mi159' LIMERICK 2 IDCA
> l (J.O. No. 18138)
OBSERVATION REPORT Observation Report No. DDR-040, Rev. O PART II - RESPONSE CCNTIMJED
' Accident temperatures are generally short in duration and have little inffect on concrete structures. Hence, accident thermal loads were not considered for concrete structure design, except the contairinent structure. -
Operating Temperature: Operating temperature difference between wall face ( were expected to be small and thermal effects were considered to be of minor significance. We have calculated and concur that a 35'F differential temperature in a four feet thick wall resialts in a 40 kip-ft/ft moment. However, this moment is small in camparison with the maximum design moment of 568.9. kip.ft/ft (page 137 of calculation 23.l(B) revision 3) and the walls ultimate moment capacity of 742 kip.ft/ft. It was inadvertently stated in response to CAI-213 that above grade exterior walls include operating temperature gradient. For D-line wall, . calculation 23.1(b) and 23.l(b-1) for 1 elow 2 grade and calculation 22.2c(h) < for above grade have been revised to show the adequacy of the wall for O- operating temperature gradient. In addition other reactor enclosure exterior V walls and the roof slab and all other Seismic Class I structures calculations (see Table 1) except Spray Pond Pump House have been revised to demonstrate that walls below grade and above grade including roof slabs are structurally adequate considering the operating-temperature gradient on them. Spray. Pond Plmtp House calculations have been reviewed, and it is judged that walls and roof slab are adequate'for operating temperature loads.
- 2. Drywell interior Steel Framing at El. 252': ]
I Thermal expansion of the intermediate cross members (Section "G" & "H" on J Drawing C-927) identified by SWEC in the observation are not a concern 1 because of the following: I
- The framing is designed to allow for thermal expansion. The intermediate ,
cross members connected to the framing are not fully restrained, thus the j thermal effect is ne For a conservative accident temperature ) change of 275*F (65*gligible. F to 340 F), a typical 13.5' cross member will exper-ience a thermal growth equal to approximately 1/4". Approximately 1/8" growth is acccamodated by the corresponding growth of its supporting box beams. . Balance of thermal growth (1/8") is accannodated in accordance with USNRC Standard Review Plan 3.8.3 and 3.8.4, Part II Section 3.c.(11), "In the above factored load combinations, thermal loads can be neglected when it can be shown that they are secondary and self-limiting in nature and where the material is ductile". With the design details provided for tnis i framing, thermal expansion of the steel is considered secondary and self-limiting. Minor local member deformations, movement or distortion of member section or connection components and anall lateral displacement of the supporting beams, will be sufficient to acconnodate the accident Page_[ofj
. <..~-~a ~- ~ ~ ~ ~ ~
P.2 tva 13 sogg4:49 310038 & WEBSTRA DEINt3R3410 6"' ', . ,%wg LZetBRICK 2 ZDCIL (J.C. 2 . 18138) / [ CASSRYATI0tl REPORT observation Report No. 0016-040, Rev. 0 PAltr II - REsponl85 00EIWsp 1 y The structural steel outside containment have been evalueted and found adequate for thermal effects as discussed in response to j DDR 001, item 22. i O ___. ___ ___ _. . .= _. w% &SW &llS{$9 &V d 3./kg -
' Faco mM say ssgnature /Date nochtel Agsww s:engture /Date O
Pageloid
STONE & WEBSTER ENGINEERING CORPORATIJi []
-(/
LIMERICK 2 IDCA (J.O. NO. 18138) [ L f OBSERVATION REPOP.T Observation Report !b. DOR-040, Rev. O TABLE __1_ List of calculations revised for operating temperature gradient - 22.2C(b) Rev. 9 22.2C(d) Rev. 8 22.2C(h) Rev. 9 22.2C(j) Rev. 11 23.l(b) Rev. 4 23.1(b-1) Rev. 4 23.l(g) Rev. 4 23.l(h) Pev. 5 22.4N Rev. 3 22.4S Rev. 5 42-3.1 Rev. 2
; 42-3.2.2 Rev. O l g#
- i I
i Page 7 ofi
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) 7-( 3' )
%./ OBSERVATION REPORT j l
Observation Report D OR- 040 Rev. O_ Review Plan: LK- D - 1903 -S Rev. L PART III - RESPONSE EVALUATION j The response to this observation report has been evaluated as follows: A. Causal Factor (s): Acceptable X Not Acceptable (Explain): 1 The response indicates that thermal loade were considered only when they were considered significant. However, as an example, thermal moments in exterior concrete walls are not insignificant, as demonstrated by the revised calculations. I B. Extent of Condition X Acceptable Not Acceptable (Expl31n): For concrete structures, all exterior walls and roof slabs were reviewed, and calculations revised as necessary. For steel structures, thermal effects were typically considered for. drywell framing, except for the case cited. Steel outside the containment has been addressed in the response to DOR-001. C. Significance: X Acceptable Not Acceptable (Explain): The reviews and revised calculations demonstrate that concrete structures d are still adequate. The response adequately explains why thermal stresses in the drywell platform cited are not significant. D. Corrective Action: Acceptable Not Acceptable (Explain): Concrete calculations and evaluations of steel framing have been performed. SWEC did not review these calculations in detail. j E. Action to Prevent Recurrences y Acceptable Not Acceptable (Explain): This is not necessary since no new concrete structures are expected, and typical steel structures have been evaluated.
............................................................................====..=====
Additional Action Required: y No Yes (Explain) O 4 J
/%u C. OJ n 3/okt Lead Engineer Signature '/ Date oCCK tw p.Wdk shdn A#1 Signature / Date Page 1 of [
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT v' Observation Report No. DOR- 041 Rev . 0_ Review Plan: LK- D - 1907 - C Re v . ,1__ Reference Al No.: DAI-221 Rev. O _ PART I - INITIATION
- 1. Description of Concern The utilization of a temperature less than the postulated temperature for the area is not consistent with the DOR Guidelines. The use of less than the postulated a rea temperature in Arrehenius equation in other applications could lead to a non-conservative design.
- 2. Supporting Information Per NUREG 0588 for both Category I and II plants paragraph 1.5 (2) states " Equipment located in general plant areas outside containment where equipment is not subjected to a design basis accident environment should be qualified to the normal and abnormal range of environmental conditions postulated to occur at the equipment location."
Paragraph 5.2 of enclosure 4 to IE Bulletin No. 79-01B (DOR (~3 Guidelines) states "The environment in the test chamber should be
; ') established and maintained so that it envelopes the service conditions defined in accordance with section 4.0 above." ~
10CFR50.49 paragraph (d) (8) states: " Margins must be applied to account for unqualified uncertainty such as the effects of production variations and inaccuracies in test instruments. These margins are in addition to any conservatism applied during the derivation of local environmental conditions of the equipment unless these conservatism can be quantified and shown to contain appropriate margins." Paragraph 3 (2) o ~ NUREG 0588 for category I plants states "In lieu of other proposed margins that may be found acceptable, the suggested values indicated in IEEE Std 323-1974, Section 6.3.1.5 should be j used as a guide." (See Note 1) IEEE Std 323-1974 suggests the use of 15 F as a margin for i temperature.
)
CONTINUED ON 1 PAGE 1 Additional Documents Attached: None l l l
/2.!7 I s
[
/
Lead tngineer Signature / Date APM Sigydure '/ Date Page j o f _7_ }
I STON'E & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT Observation Report No. DOR- 041 Rev. O COIrfINUATION: Addition of 15"F margin to 112.6*F (from 30 days of postulated profile through 180 days) is equivalent, to 127.6*F which is not enveloped by 122'F testing for 30 days. Therefore it is questionable if this testing can be' considered in LGS Calculation E-856-2. Additionally, use of 115'F instead of 120*F as a LOCA temperature effectively creates a negative margin condition which does not meet the intent of 10CFR50.49. Note 1 - -Alternative to 15'F margin suggested in IEEE 323-1974 can be found in NUREG 0588 for Category II plants, and can be applied to LGS. l
)
O Page A of I
A STONE 6 WE3 STER ENGINEER!NC CoEPoKAT!o3 LDERIcK 2 IDCA. (J.o. No. 18138) _
/ f-
, ossanoman israer U" # 75T) O , observation Report No. D 01 41 Rev. O
)
j not tr - ass.reget
- 1. observation concurrences l
'X concur with observation Do'not concur with observation (Notes if not in conegtrrence. explain in " remarks" below) 2.- Response to Observation:
The following is an amended response prepared in accordance with agreement reached between SWEC and Bechtel in a telecon on March 21, 1988. A. Causal Factor (s) Calculation E-856-2 included in EQ Pkg. E-182) used 115'F instead of 120*F as LOCA temperature for the temperature devices (RTD's) listed in Bechtel.EQRR M066/47.00 revision 3. The environmental conditions listed in Bechtel EQRR M066/47.00 revision 3 are properly' based on Specification M-171. Calculation E-856-2 was prepared to_ demonstrate that the high temperature testing by the vendor envelopes the Limerick LOCA profiles shown in Bechtel EQRR M066/47.00 (120'F for 180 days) and Bechtel EQRR M066/45.00 (115*F for 180 days). Use-of the incorrect LOCA temperature is attributed to hur.an eircr on the part of the originator and checker of the calculation. B. Extent of Condition Of approximately 70 total similar calculations a representative sample of 20. calculations has been reviewed. For each calculation reviewed the proper LOCA temperature has been used and adequate margin exists to comply with the Category II requirements of NUREG 0585. In each case the required post-DBE duration is adequately enveloped by the duration calculated after considering 15'F margin. In our opinion the selected sample of 20 calculations is representative of 70 calculations to demonstrate that adequate margins exist for qualification of Limerick equipment. Therefore the cited condition is concluded to be an isolated case. C. Significance
- The temperature devices listed in Bechtel EQRR M066/47.00 and M066/45.00 are required to conform to Category 11 requirements of NUREG 0588. As discussed below adequate margin exists to meet the intent of NUREG 0588 Category Il requirements, Paragraph 3.(2)
(and Limerick Specification 8031-G-18, Paragraph 6.3.6): Page 3 of 7 Misc 35-1
STONE 6 '=T5 STER ENCDrEERING CORPORATION 1.DtIRICK 2 IDCA (J.O. NO. 18138) Jpg /f/f
/"; 085EITATION REFotT b observation Report No. L oR- 41 Rev. O PART II - RESPOSSE_
o NUREG 0588, Category II, Paragraph 3.(2)(a) and (b) The RTD test specimen was exposed to three temperature profiles as follows: 9
- First Phase, 122*F for 30 days
- Second Phase, 150*F for 100 days
- Third Phase, 255'F for 6 hours The test peak temperature is 255'F whereas the Limerick required peak temperature is 120*F. Therefore adequate peak temperature margin was applied.
The revised calculation E856-2 (Transmittal S-0615) considered longer duration in lieu of a 15'F margin. Actual Limerick long term requirement is 120*F for 30 days followed by 112.6*F for 150 days. The calculation shows that the second phase testing of 150*F for 100 days is equivalent to approximately 798 days at
- 120'F. It is our judgement that testing st an elevated
( temperature to reflect an appropriate margin is equivalent C in terms of stress to testing at a lower temperature for a longer period. The judgement is further supported by applying a 15'F margin to the second phase testing of 150*F for 100 days. The calculation E856-3 (Transmittal S-0690) shows that the second phase testing is equivalent to approximately 275 days at 135'F. o NUREG 0588, Category II, Paragraph 3.(2)(c) Since adequate margin exists the factors of " number of items tested and the number of tests performed in the hostile i environment" do not require further consideration. Number of items tested is significant only when a lesser margin is required to be justified. o NREG 0588, Category II, Paragraph 3.(2)(d) and (e) The performance characteristics of the RTD was demonstrated following each stage of test (i.e. radiation, aging, seismic andLOCA). However, the performance characteristics were not monitored during the LOCA test. This omission is acceptable because in the Limerick application the RTD's are not required to function during and after (except 2 RTD's) a Design Basis Accident. The RTD's (except 2 RTD's) control the air cooler fans of the unit coolers in the pump The compartment unit cooler maintains the (V~] compartments. Misc 35 2 "9' A Of .2. I
sfoWE 6 WEssTsa EWo!Wssa!Nc CoaNaA71ok unmarcs :scA (J.o. No. 18134) JDCM / 9/ f
~ ' ' ~
' d$$ET&TIW a3M ,,,
0 88 . cai asU d'ge a4::f n-%.5.N .h.
- p.g y
.y 5 (+. .,
. . ;... 4,.
. ,.,'I 0
5 j ,,
- - ff raat
- tir.rassenMR copetsWEB a
,~
. <s
.uw s
- . :.ws : , an ;w a tua - . _
j ,'
~
%;o 4 NUREG 0548 Category !!, Paragraph 3 (t)(d) and (a)'(Cont'd)'
,, , e uu design temperature if the normal HVAC fails during normal '
% ..,u plant operation. Pump start initiates start of compartment -
air cooler fan bypassine the RTD circuitry. The otter two RTD's rovide a signal to control room indicatore for the
- e. inlet ir temperature to the filters of the Reactor
" Enclosure Air Recirculation System (RERS . The only y' m' .. , n r filtration component which requins cent ol room temperature es'. - s, igfg indication for system monitoring and operation is the charcoal adsorber. T erature indication for this c;; ' component is provided i the control room. The inlet' air
~ w-~ to erature indication to the RER$ filters is a convenience c- but is not used by the control room operators to-
. ,,disla{eanyaction.
, ini fa. -
, .a It is concluded that adegaats margins were incorporated during the qualification test and the lack of monitoring performance characteristics durin LOCA test is accostable O for the Limerick application. Th RTO's listed in tie Bechtel EQRR N066/47.00 meet the intent of NUREG 0508 Rule 10CFR60.4g and Limerick Catacory Speelfication !! requirements, 8081 G-18. Therefore the RTO'a are qualified t.o perform their intended safety function for the Limerick application.
D. Corrective Action Calculation E 856 2 has been revised and a sample review of similar calculations has bun performed.- MtiontoPravantRecurrgnee E.
~ '
Since the review of calculations discussed in iten B above demonstratedthatthesubjectconcernisanisolatedcase,which is not sigfttficant, there is no need.for any additional action.
.... nu. n u..
h.7 n 3/uh9 n.nu..u..we....e===a.g.aaaaaa""*"""f W
<m O ,=ewie. p . v .ese sesua *,,s.~t a.-t -
Fase 1, ef 3 Misc 35 3
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK a IDCA (J.O. NO. 18138) OBSERVATION REPORT Observation Report DOR-041 Rev. O Review Plan: LK- D - 1907 - C Rev. L PART III - RESPONSE EVALUATION The response to this observation report has been evaluated as follows: A. Causal Factor (s): X Acceptable Not Acceptable (Explain): Reason for calculation preparation is clearly stated and consistent with the purpose stated in calculation E-856-3. Reason for use of an incorrect LOCA temperature is attributed to human error on the part of the originator and checker. B. Extent of Condition X Acceptable Not Acceptable (Explain): An I.C. sample of 20 of 70 calculations was reviewed to assure that in each case the required post-DBE duration adequately enveloped by the duration af ter considering 15 F margin; therefore, it appears that correct temperature plus adequate margin exists in the calculations reviewed. Project concluded that this was an isolated case. C. Significancor X Acceptable Not Acceptable (Explain): Project response states that the number of items tested and the number of tests performed in the hostile environment (Paragraph 3.(2)(c) of NUREG-0588, G Category II) does not require consideration, since the calculation has shown CONTINUED ON NEIT PAGE. D. Corrective Action X Acceptable Not Acceptable (Explain): Calculation E-856-3 revision to E-856-2 has been reviewed and accepted. Sample review supports conclusion that error was an isolated case. E. Action to Prevent Recurrence: X Acceptable Not Acceptable (Explain): No further action required. Additional Action Required: X No Yes (Explain) 0 CfL A & D. kl/Il e 4/4/r1 G[.............................................................. CGCC e d EngVne~er Signature 4/9/s1
/ Date AIM Signature / Date Page i of _],
STONE & WEBSTER ENGINEERING CORPORATION ~ LIMERICK 2 IDCA f-4. . (J.O. NO. 18138) i \ OBSERVATION REPORT Observation Report DOR _041 Rev. 0' PART III - RESPONSE EVALUATION CONTINUATION: C. Significance that adequate margin exists. Except for 2 RTDs that provide a signal to control room indicators f,or the inlet temperature to the filters of the Reactor Enclosure Air Recirculation System, which are convenience displays only, none of the RTDs have to operate after a DBA. Therefore, it is not significant that the performance of the tested RTD(s) was not monitored during testing. Note: . Project response includes a statement that "The only filtration component which requires control room temperature indication for system monitoring and operation is the charcoal absorber." The qualification of the devices used to measure this parameter was not pursued. !
~ ,} i P ap , a t ,
___l i
]
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA
/N -(J.O. NO. 18138)
OBSERVATION REPORT Observation Report No. DOR 042 Rev. O Review Plan: LK- D - 1903 - E Rev. L 1 Reference AI No.: 113. 114. 115. 149 & 150 PART I - INITIATION
- 1. Description of Concern Due to the discrepancies identified below, the ruults contained in Calculation 6380E.08, Rev. 2 " Diesel Generator Voltage Regulation Study" are inconclusive.
- 2. Supporting Information
- 1. The subject calculation analyzes the following diesel generator DBA ,
loading possibilities: e Starting of all 480V loads with the diesel generator output voltage held at 1.0 Per Unit (PU). e Starting of all 480V loads when the initial diesel generator A voltage is allowed to dip in accordance with the following initial
- Q, voltage dip formula supplied by the diesel generator manufacturer:
E" d KVA x 100 1 + SKVA x CX"d e Starting the RHR pump motor after all other 480V loads are applied with the diesel generator voltage held at 1.0PU.
- Starting the RHR pump motor after all other 480V loads are applied with the diesel generator voltage allowed to dip in accordance with the above manufacturer supplied formula.
The above loading possibilities do not reflect the actual designed loading sequence. The designed loading sequence dictates t.ha t the 1250 HP . RHR pump motor be energized by the diesel generator first; 3 seconds later, the 480V loads are energized; next the remaining loads are sequenced. The subject calculation does not analyze this
. . . . . . . .. . . c.o".r.I."H".n. .o.a. .?.o..uc.m . . . .. . . .. . .. . .. . . . . .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .
Additional Documents Attached: None, j p ............ ...................... ..................... O f)-/ 4 g/z//n ....h..................../2/2 '
// Date Lead Ing[ineer Signatfre '/ Date APM Sigrfditure Page 1 of J.Q l
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) ()l OBSERVATION REPORT Observation Report DOR- 042 Rev. O PART 1 - INITIATION CONTINUATION: sequence. Of particular concern in this sequence are the starting voltages for all motors supplied by the 480V buses. The subject calculation does not address this concern. This calculation does not demonstrate that the motor terminal voltage of these lou voltage motors does not drop below their minimum starting voltage ratings for the duration of the motor starting period. Also, the appropriateness of the above voltage dip formula for time periods beyond the initial 0.1 second loading should be investigated. This formula was applied by the diesel generator manufacturer to study initial generator voltage dips after a load block was applied. It should only be applied to the initini addition of a load block to the diesel generator. It determines generator reactance for a specific instant in time af ter a lead block is applied. TL ratio of generator transient and subtransient reactances will con cir,ue to change with time after a load block application. This change sill tend to prolor.g the voltage dip durations.
- 2. The subject calculation indicates that motor terminal voltage will fall below motor minimum starting voltage ratings during the beginning O
V of the motor starting time period. For example, Case CS5D14102A of the calculation determines that the voltage at MCC 10B218 will fall to 77.17 percent for a short period. Allowing for the 6 percent maximum MCC voltage drop determined by calculation 6470E.11, Revision 5, the 460V motor , terminal voltage will fall to 75.7 percent ; motor base voltage). These 460V motors have a minimum starting voltage rating of 807. cf rated voltage. However, the duration and impact of this voltage dip is not addressed in the subject calculation. Depending on the duration of this dip, the motors may or may not successfully start prior to the tripping of protective overload relays.
- 3. As discussed in the above items, motor terminal voltage should be calculated for all worst-case design scenarios. These terminal voltages should be calculated for 4KV, 460V and lower voltage motors to insure that motor voltage ratings are maintained. Any voltage transients below motor voltage ratings should be evaluated for technical acceptability. The subject calculation only analyzes voltages for buses connected to diesel generator D14. The calculation does not confirm that diesel generator D14 .has tu worst-case system loaos i
and voltage profile when compared to the Limerick 2 diesel generators. l l 4. The subject calculation requires correction to a 0.112% vc h sge drop l which was calculated between the diesel generator and 4160-V switchgear. l As indicated by Bechtel in response to Action Item DAI-115, this voitage drop actually should be calculated at 0.237.. This voltage drop should , be considered in the above item 3 evaluations. l
\
") CONTINUED ON ONE PAGE Page 1 of jp l
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) f g OBSERVATION REPORT Observation Report DOR- 042 Rev. L ! PART 1 - INITIATION CONTINUATION:
- 5. Appendix M of the subject calculation depicts only the total quantities of MOV loads in equivalent KVA which are operational during the initial energization of the 480V buses by the diesel generator under DBA conditions. The calculation does not document how the Appendix M MOV load totals were derived. The Appendix M MOV load totals for each diesel generator 5.ppear to constitute an average of approximately 20 percent of the tctr1 connected MOV load. If the Appendix M MOV load quantities are inaccurate, the voltage drop created by loading the diesel generators with the 480V loads could be adversely affected.
- 6. The RHR pump motor and RHR Service Water pump motor should be checked to confirm that the motor starting terminal voltages are above 807.
of the nameplate ratings and that the 80% motor starting time-current curves can be applied for setting protective relays. This confirmation is not contais.ad in the subject calculation. Due to the above discrepancies, we consider that resolution of these concerns and discrepancies should include consideration of the generator synchronous, f- transient, and subtransient reactances and their associated time constants ! to correctly model generator terminal voltages during all motor starting \' time periods.. 4 O Page 3 of f 0
1 STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA [p(g /g)[ (J.0. NO. 18138) g 018BETATIM M
@ O H [)
Observation Report No. D 01- 042 Rev. O FMnt II - REBMESE I l i AMENDED RESPONSE Following is the amended response prepared per " DOR-42 Bechtel Response Review" dated 2/13/89 f rom SWEC to Bechtel, and , subsequent telecon from SWEC (E. Heneberry/R. Sibulkum/G. Nolan) l to Bechtel (E. Kang /K. K. Chan/A. Kar/I. Don-Doncow) dated ' 2/14/89. I 1
- 1. Observation Concurrence:
Concur with observation X Do not concur with observation (Note: if not in concur-rence, explain in " remarks" below)
=======.....=================.....................................
Remarks: [)
\'
- 1. Calculation 6380E.08 analyzes the voltage regulation of the Class lE onsite ac power system, when supplied from diesel generators, based on the expected loading condi-tions during a FORCED SHUTDOWN (resulting from a LOOP without a LOCA) and a DBE (LOOP concurrent with a LOCA). Under*a LOOP without a LOCA, the 480V loads will be energized first. The RHR pumps are not automa-tically sequenced during this event. Under a LOOP with a LOCA the RHR pumps are started first followed by the 480V loads. The purpose of this analysis is to verify that the minimum voltage requirements at the equipment terminals are satisfied during load starting and steady state operation to ensure proper equipment performance.
The simplified model used for the purpose of this cal-culation takes into consideration the results obtained in the diesel generator manufacturer's computer study 8031-M-71-491-3 (Transmittal No. S-0133). Therefore, this calculation should be used in conjunction with the manufacturer's' study for reviewing the approach in the calculation and interpreting its results and conclusions. Keeping this in mind, the results contained in this cal-culation are considered to be conclusive for the reasons described below: O (See continuation sheet) T-10/6-1 "'seILofjf I
srows & wastrR zucIwxrRInc CORPORATION LIMrRICK 2 IDCA [ggg. /gp[ . (J.O. NO. 18138) g _ ) h OB8ERTATICE REPORT h-Od[ Observation Report No. D OR. 042 Rev. O I AMENDED RESPONSE Following is the amended response prepared per " DOR-42 Bechtel Response Review" dated 2/13/89 from SWEC to Bechtel, and subsequent telecon from SWEC (E. Heneberry/R. Sibulkum/G. Nolan) to Bechtel (E. Kang /K. K. Chan/A. Kar/I. Don-Doncow) dated
?/14/89.
- 1. Observation Concurrence:
Concur with observation X Do not concur with observation (Note: if not in concur-rence, explain in " remarks" below) n================================================================= i Remarks: (O) 1. Calculation 6380E.08 analyzes the voltage regulation of the Class lE onsite ac power system, when supplied from diesel generators, based on the expected loading condi-tions during a FORCED SHUTDOWN (resulting from a LOOP without a LOCA) and a DBE (LOOP concurrent with a LOCA). Under* a LOOP without a LOCA, the 480V loads will be energized first. The RHR pumps are not automa-tically sequenced during this event. Under a LOOP with a LOCA the RHR pumps are started first followed by the 480V loads. The purpose of this analysis is to verify that the minimum voltage requirements at the equipment terminals are satisfied during load starting and steady state operation to ensure proper equipment performance. The simplified model used for the purpose of this cal-culation takes into consideration the results obtained in the diesel generator manuf acturer's computer study 8031-M-71-491-3 (Transmittal No. S-0133). Therefore, this calculation should be used in conjunction with the manufacturer's study for reviewing the approach in the calculation and interpreting its results and conclusions. Keeping this in mind, the results contained in this cal-culation are considered to be conclusive for the reasons described below: O (See continuation sheet) Page,1l_of,j,p T-10/6-1
STONE & WEBSTER ENGINEERING CORPORATION LDERICK 2 IDCA JggA / (J.O. NO. 18138)
) OBSERVATION REPORT l
Observation Report D OR- 042 ra v . ,0_, CONTINUATION: , Rsmarks: 2.1 For the purpose of this calculation, the in-house TE503 program is used to calculate the voltage level on the Class lE distribution buses for the two scenarios in item 1 above. The program does not calculate the voltage dip duration, nor is the intent of this calcu-lation to estimate its duration. To account for this, the manufacturer's computer study calculates voltage and f requency deviations and recovery times for each block of loads not only for the initial 0.1 second after each block is applied, but for the entire period until a 100% recovery takes place. To achieve this, the model in the study considers the different parameters y required to represent the diesel engine, the governor, the generator (including the transient and sub-transient reactances), the excitation system, applicable time constants, and the characteristics of the applied loads. , /} s_/ By using the load data and the data extracted from the inertia and speed / torque curves for motors and driven equipment, the program is able to calculate the demand on the diesel generator; thereby producing a more accurate estimate of the deviation and recovery times for the volta'ge and the frequency. Colt Industries confirmed that the acceleration time for the modeled motors (RHR, Core Spray, and Emergency Service Water l Pumps) is determined by the computer program used in their study from the data provided for these motors i and their loads. (See pages 3,5, and 11). The ! diesel generator performance is further verified during the Limerick preoperational test program. As demonstrated in the manufacturer's computer study, even for the worst voltage dip, the generator will recover its voltage to 100% in less than one second. Taking credit for this fast recovery, the assumption is made that the " swing bus" (used in the calculation for diesel /4kV bus) is an infinite bus. Therefore, although Calculation 6380E.08 did not simulate the actual loading sequence for the RHR pumps versus the 480V loads during a DBE, because of the model used, the calculated values for voltage drops at the 480V buses will not be affected even if the sequence is q reversed. %-) (Cont.) T-10/6-2 page f of ,Jp
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA M (J.O. No. 18138) O osssavArrou asroar Observatiet Report D OR- 04 2 Rev. O_ CONTINUATION: Remarks: 2.2 For each case studied, two computer runs were made. One with the generator voltage at the maximum dip and one with the generator voltage restored to normal (1 p.u.). The cases that consider the maximum voltage dip are intended to demonstrate that adequate voltage will exist at the motor control center to eliminate the possibility of dropping out the contactors. Also, the voltage recovery is fast enough to prevent the tripping of the protective overload relays during a voltage dip. In the cited example, Case CSSD14102A of the calculation, per vendor analysis the recovery to 100% after starting the 480V loads occurs in approximately 1/3 second. Re covery to 90% occurs even faster and, therefore, tripping of protective overload' relays will not occur. f-~g 2.3 Only the cases that consider the diesel generator voltage (j at 1 p.u. are used to demonstrate that adequete voltage will exist at the motor terminals during starting and steady state operation of these loads. Note that the results obtained in all applicable cases (the two 100 HP, 480V motors do not start simultaneously with the rest of the 4'80V loads but are delayed by 20 sec, and 25 sec.) satisfy'the minimum voltage requirements for starting and operating the 4kV and 460V motors. This is true even after-the voltage drop between the diesel generator and the 4.16 kV switchgear is corrected from 0.112% to 0.23% and a maximum voltage drop of 6% between the motor control centers and the motor terminals is considered. Moreover, additional conservatism is built into the model because the system model assumes that: l
- The longest feeder cable and the largest anticipated voltage drop to a motor is used for all the motors.
- All motors accelerate their loads and come up to I speed simultaneously; thereby maintaining loaded l rotor current on all motors for the same length l of time.
l
- All motors are exactly matched to the acceleration
.,\-} requirements of the driven equipment.
l 1 l (Cont.) T- 10/ 6 -3 Page 6 of 10
STONE & WEBSTER ENGINEERING CORPORATION
- 0. 1 )
OBSERVATION REPORT Observation Report D OR- 042 Rev.0_ CONTINUATION: , Remarks: However, none of the above asrumption are universally true because:
- Feeder cables to motors are generally shorter than the used maximum length, so that the motor will generally experience voltage drops smaller than postulated.
- As some motors come up to speed, the current require-ments change from locked rotor to full load. This '
in turn reduces the total current and voltage drop I through the system impedances so that motors which have not yet accelerated will likely experience higher operating voltage than calculated. Motors are manufactured in standard size with the result that some motor are conservatively sized for the driven equipment. In order to reduce the number of cases to be studied,
- O the loads and impedances on each ci the four safeguard
\> channels were examined. The channel D and diesel generator D14 have the largest diesel generator loads, 480V load center load, motor control center load, and the longest cable run to most heavily loaded motor control centers. Therefore, if diesel generator D14 and channel D have satisfactory voltage regulation, the voltage regulation for other channels will also be satisfactory. Since the loads for both units are equivalent except for the common loads, which are pre-dominantly fed from Unit 1, no conditions are anti-cipated where a Unit 2 worst case loading on the diesel generator will exceed the Unit I worst case loading. Therefore, the results obtained from Unit 1 are con-sidered to be conservative for Unit 2. 2.4 As indicated in item 2.3 above, the minimum voltage requirements for starting 4kV and 460V motors are eatisfied even after the voltage drop between the diesel generator and the 4.16 kV switchgear is corrected from 0.112% to 0.23%. 2.5 Appendix M load figures are accurate. Although records for the MOV data collection were not kept, the MOVs which are operational during the initial energization of the 480V buses under DBE conditions O were identified and their total load recorded in (Cont.) I T-10/6-4 Page 7 of 10
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) [hh .2[ OBSERTATICE REPORT
. Observation Report JOR-_042 Rev. L CONTINUATION:
Rema rks: Appendix M at the time this calculation was prepared. To support this, division A was selected for a sample review. All needed MOVs were identified and their total resulting load was calculated. This calculated load matches the MOV load used in Appendix M (Transmittal No. S-0656). 2.6 From the data for voltage dips in the manufacturer's analysis it can be concluded that the starting terminal voltages for the motor loads on the 4.16 kV buses (including the RHR and RHR service water pumps) are above 80%. Moreover, the analysis indicates that "the amount of voltage dip is also not significantly influenced by the load on the unit"; therefore, the sequence in which these loads are applied is not significant. Relay setting calculation 6900 E.05B (Transmittal Nos. S0096 and S0111) confirms that 80% starting time current curves are applied. Clarifications for Sections 2.2 and 2.2 f-ss b To determine voltages expected at motor and load terminals, the table in the " Summary of Results" of Calculation 6380E.08, sheet No. 79, must be used along with Table 1 in Attachment
- 10. 2, shee t 1 of Calculation 6300E.20 (Transmittal Nos.
S0037 and S0096). The feeder cable voltage drops in Table 1 for the different voltage levels and the 0.23% (0.24% on a 4kV base) voltage drop between the diesel generator and the 4.16kV switchgear must be subtracted, as applicable, from the voltages in the " Summary of Results" to obtain the expected voltages at the load terminals. Keeping this in mind, a review of the " Summary of Results" indicates that the expected voltages at the load terminals for all voltage levels are adequate for all cases where the diesel generator voltage level is 1 p.u. (Note that Case Nos. CS5D14103 and CS5D14104 are not applicable because the starting of compres-sor K101 and pump P210 is delayed by design). When the diesel generator is experiencing the maximum voltage dip during load starting, the worst voltage condi-tion on the 4.16kV bus occurs for Case Nos. CS5D14104B and CS5D14102F. For these cases, the expected voltage at the terminals of the 4kV motors during starting of these motors is 82.06% - 0.24% - 0.5% = 81.32% (4kV base). This voltage
- is above the 75% minimum thac is required for starting 4kV l motors. The remaining buses are 480V with no starting l
((~' motor loads for these two cases. The voltage on bus 15 is (Cont'd) l T-10/6-5 Page.1,of_LO i
i i giop W 5 Vsita ENGINICRENO CORPORAT10N LIMERICE 81DCA (J.o. No. 18138) yjgg O = = = = = Observaties Report No. n.,,01-g Rev. A l CONT 150ATION Remarks: acce) table as it will' remain above the 75% minimum voltage requared at the motor terminti (voltage at motor terminals for worst case, C85D14102F, is 78.104 - 0 244 - 1.56 = 76.36%). For buses 16 and 18 the voltage at the motor l terminale may drop below 756 horst case is bue 18, Case No. CS5D14102F, waere the voltage at the motor terminals may drop to 74.004 - 0.244 - 1.54 = 72.264). Although this , , vohtage is momentarily below the minimum requirement of 75% ] for running motors during the starting of other motors, no . I impact on the motor operation is ex fast voltage recovery (less than 1/pected 2 secondbecause to 90%ofatthe the diesel generator terminals). Furthermore, the condition in these two cases malt only occur wnen the RNR pumps are manually started s;.nce under an automatic load sequencing , the RHR ptimps are always started first, before the 480V loads a.re co.nnected. Two other cases, CS5D14104A and CS5D14102A, may result in voltages at the 460V inoter terminala that are below the - minimum 806 requirement during the starting of these motors. O - The worst case may occur for motors starting on bus 16, Case No. C55D14104A, where a minimum voltage of 78.204 - Again, because of (i.244 - 4.06 = 74.04% could be expected.
? yhs very fast recovery to 984 voltage at the diesel generator terminals, no adverse impact on the motor operation is expected.
Similarly, these momentary voltage dips are not expected to have any adverse effect on the operation of non-motor loads (e.g., battery chargers) and 120V ac equipment and components. I i
...y.y . y . y ..p........
O YNt %\e9 - A + R r. 6 M 4 f 1b/rf / Data Signature / Date Sethiel Appteval 513M ture PECe Ravi T-10 Fase ,0,,,of,LO ._~_-- _ - - - _ _ _ _ _ _
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) A OBSERVATION REPORT Observation Report D OR- 042 Rev. 0_ Review Plan: LK- D- 1903 -E Rev. L l PART III - RESPONSE EVALUATION The response to this observation report has been evaluated as follows: A. Causal Factor (s): X Acceptabic Not Acceptable (Explain): The results contained in Calculation 6380E.08, Rev. 2, " Diesel Generator Voltage Regulation Study" in conjunction with the amended response to this DOR, provide for satisfactory documentation to ensure that the required voltage is provided to all Class IE equipment when supplied by the emergency diesel generators. B. Extent of Condition X Acceptable Not Acceptable (Explain): The extent of the condition pertains to motor starting voltages and motor and load operating voltages when the loads are supplied by the emergency generators. l C. Significance: 1 cceptable A Not Acceptable (Explain): CN When safety-related loads are supplied by the emergen.:y diesel generators, the voltages applied at the terminals of these Ivads are within their nameplate ratings. D. Corrective Action. y Acceptable Not Acceptable (Explain): No corrective action is required E. Action to Prevent Recurrence: Acceptable Not Acceptable (Explain): Not applicable Additional Action Required: X No Yes (Explair.) l No additional action is required 1 o ............... .-...................................................................... WAlTD Garunm3h>l83 o c A z A D. m fle- slmin Lead Engineer Signature / Date APM Signature / Date l Page to of 10 j _ - _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ l
i STONE & WEBSTER ENGINEERING CORPORATION
' LIMERICK 2 IDCA g (J.O. NO. 18138)
OsssxTATION REPORT Obs'ervation Report No. DOR- 043 Rev. O Review Plan: LK- D - 1903A - MC Rev. L . 1 Reference Al No.I DAI-210. Rev. 0 ) PART I ,, INITIATION 1
- 1. Des 3ription of Concern bse of exponential' temperature decay of the pipe wall near RPV nozzle is not a conservative practice and may lead to unconservative transient analysis.
- 2. Supporting Information In Calculation S/8031-2302 Rev. O temperature of pipe wall near RPV nozzle !
has been datermined. by exponential temperature deca:/ using Standard
- Exponential Temperature ' Decay equation. This will lead to unconservative analysis since water inside the pipe near the RPV nozzle will be at or close to RPV -temperature and pipe wall temperature can not be predicted
-using Standard Exponential Temperature D':ay equation. -For this reason GE -has recommended in the Design Specification Document Number 22A2919 in comment- section sheet. number .18 that " Pipe lines which normally have no flov such as RHR supply and return lines will remain at ambient drywell
' temperature away from the vessel during most vessel transients. It is recommended that analysis be based on these most severe transients.(meaning transients given in Design Specification Document) since it would be difficult to predict the actual transient in all portions of the piping."
In this specific instance 300*F has been considered to' be' the temperature at the LPCI RPV nozzle / pipe connection at 2. 279 ' . sway from RPV wall when RPV is at temperature of 406*F. Based on the above recommendation from GE it can be stated that reducing temperature applying exponential decay in region I (region, where pipe temperature is influenced by RPV temperature) is not conservative' practice and may lead to unconservative results.- Additional Documents Attached GE Design Specification Document Number 22A2919 Sheet 18. q ............................ ........................... ...... . ................... Lead Engineer Sighature
/Zl22ltff-
/ Date APM Siptature
(
/ Date
)
-Page 1 of d
L GEN ERAL h ELECTRIC
.. c... 22A2919 .c.. ., 6 O . . . . 18 j i
l
- 1. Description The thermal transients provided in this APPENDIX are derived from the thermal'tran-sients established for nozzles on the reactor vessel which are used for ASME Code Section III design calculations on the reactor vessel. The piping systems for which.
themal transients are provided are shown on the piping diagram (Figure 1). Tabulated values of thermal transients are provided in Table II for each pipe line shown in bold lines on Figure 1. A thermal transient table is pre' ,.3,.;J. for each mede of normal operation expected and for upset, emergency and fault ccaditions postulated over the 40 year life of the plant. Piping lines shown in narrow lines on Figure 1 have thermal transients identical to the lines to which they are connected. Table 1 lists each pipe line by system and shows which thermal transiants in Table II apply to the particular line.
.2. Commenes -
The tabulated values of thermal transients show fluid temperatures in the pipe line as they change with time. It should be noted that the thermal transiencs are based on O condition at the vessel nozzle and depending on final piping arrangement may be lees severe for portions of piping which are amote from the vessel. Pipe lines which normally have no flow, such as RHR supply and return lines, will remain at ambient dr>vell temperature for portions of the pipe away from the vessel during mest vessel transients. It is recommended that analysis be based on these most severe transients since it would be difficult to predict the actual transient in all portions of the piping. i A V DOR - o V 3 REV O PAGE 2 OF fl _ _ _ _ - _ - _ - _ _ _ _ . _ _ . _ . _ _ l
n S'10NE & WEBST!:R ENGINEERING CORPORATION LIMERICK 2 IDCA [h I i (J.O. No. 18138) A/ - J AN 2 41989 OBSERVATION REPCRT Observation Report No. DOR-043 Rev. 0
.P_ ART II - RESPONSE
.i
- 1. Observation concurrence:
Concur with observation X Do.not concur with observation (Note if not in concurrence, explain in " remarks" below)
- 2. Response to Observation:
A. Causal Factor (s) N/A B. Extent of Condition N/A C. Significance N/A D. Corrective Action N/A E. Action to Prevent Recurrence N/A Remarks GE's statement in their design specification is only a recommendation rather than mandatory requirement. Bechtel designs Nuclear Cla'ss 1 piping in accordance with the Bechtel design specification, not the GE design specification. In nest cases, Bechtel uses the most severe transients given in the Bechtel design specification to perform nuclear class 1 stress analysis. wm i Pagelof ll _ _ _ _ - _ _ _ _ _ _ l
h V STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA V) (J.O. to. 18138) OBSERVATION REPORT o Observation Report No. DOR-043 Rev. 7 PARI II - RESPONSE COffrINUED For a few isolated cases, when appropriate, Bechtel calculates a store realistic temperature decay alorg the dead leg branch line with stagnant water using exponential temperature decay equation. For the LPCI system, Bechtel has considered the effect of insulation and stagnant water, as well as the conductivity of pipe wall as described below. For bare pipe, the temperature decay is normally presented ass T(x) = Too+ (Ts - Too)e-mx where m = g, kA h = metal surface film coefficient, BTU /hr FI 2 ep p = perimeter of pipe outside circumference, FT
.s k = metal thermal conductivity, BTU /hr FT 'F A = metal cross sectional area, FI2 The kA term represents heat flux in pipe's axial direction. This term tends to decrease the rate of decay. A 10% increase in the kA value considers additional axial heat flux due to insulation ard water. The hp term represents heat loss through the pipe's outside surface. With the presence of insulation, the hp term can be represented by a new relation in terms of the thermal conductivity of the insulation and the insulation thickness, namely:
hp = 2 41' ki n. inn in CD W where kins = thermal conductivity of insulation Dins = outside diameter of insulation layer Do = outside diameter of pipe Hence, the coefficient m becomes: m2 = 2 Tf kins
/
1 in'Dg hA T'T i Do
\
.................. ........................................... {cormaoco} .
9N. 1 Page 4 of ll
I JGH-26 '89 17:10 ID:PECO PROJECT MGT DIV TEL N0:215-641-4579 8193 P02 __ STONE & WEBSTCR ENGINEERING CCSPORATION G0ck/VI LIMERICK 2 IDCA (J.O. ND. 18138) OB8ERVATION REPCRT Observation Report No. D3R-043 Rev. O PART II - RESPCN82 00tft!W8D Therefore, the above equation incorporates the effect of water and insulation. W is classic heat transfer calculation for steady state condition is justified for LPCI system as follows:
- 1. '!he !JCI flow conduit ente.1de through the RPV nossle to the reactor core shroud via a thermal sleeve and coupling (see Drawing 8031-N-1-B11-2010<:-003.5 transe.itted to Sec on i
1/19/88 via 5-0578).
- 2. The discharge of !JCI flow is baffled such that secondary flow penetration from within the reac6cr abroud into the LPCI conduit is partically obstructed. See Drawing listed
'( above, view "YE".
- 3. The LPCI conduit exit is recessed frw. the shroud below such that it is partially isolated from the aulk upward flow within the ehrcud.
The calculated temperature at that location is 223'r, and 300'F was used for the parpose of conservatism. Therefore, nechtel's approach is adequate. In addition, the analysis of the LPCI piping was performed in accordance with AEME Section III 1974 code where Ti is included in Sn lEQ.10) calculation. ASME section III 1979 Summer Addenda is allowed for the analysis of LPCI piping. If the 1979 Code is used, T1 wil1, be excluded from the ns calculation such that the fatigue damage would be significantly reduced. The reduction wuld be more than outficient to caigensata for the stress increase due to the therma'6 shock of 406*F to 50'r. W na negew signatursg
\b a we /uate mMt. w Nu sechtel Approval ;iignatWe /Date Pagelof,lj,
l 4 l l- ) i l STONE & WEBSTER ENGINEERING CORPORATION l i
.( q4 LIMERICK 2 IDCA j j (J.O. NO. 18138)
OBSERVATION REPORT FEB 131989 Observation Report No. DOR-043 Rev. O PART II - RESPONSE SUPPLEMENT In the LPCI analysis (ME 913) the reduced transient (300*F - 50*F) is only applied to the TTJ's @ data points 100, 140B and 150. Among these 3 data points, the maximum usage factor occurs at data point 150, where the cumula-tive usage factor (CUF) = 0.6493. The major portion of this CUF value is contributed from load pair 2-3 where the UF = 0.5580. This analysis was performed in accordance with ASME Section III, 1974 Edition. The following demonstration calculation recalculates the UF for load pair f' 2-3 at node 150 by: V] 1) Using ASME Section III, Summer 1979 Addenda
- 2) Using transient data of the full shock (406*F - 50'F) 6T1 = 7 291*F AT2=I 74 F respectively for loadcases #2 & #3, Ta -Tb = T 132*F from run #85068 (10/13/82)
- 3) Using S,at operating temperatures 406'F, not at design temperature.
Previous transient data used for loadcases #2 and #3 is: AT1 = T 206*F AT2=T 50'F T-Tb = T 92'F C3 = 1.35 C3 = 0.6 K3 = 1.87 Previously calculated Sn and Sp for load pair 2-3 is: Sn = 140,381 psi K, - 3.333 Sp = 299,408 psi DOR - o 4 3 REV O PAGE le OF II
FEB-21 '89'11:04 ID:PECO PROJECT MGT DIV TEL NO:215-041-4578 a461 P22 FEE 13 ' O ;1 06 iECH C'. 4 tTRH W R 137 orp : -, Q,) . Recalculate Sn by excluding ATt tarms En = 140,381 - 1 E oC 471 range 2(1 V )
= 140,381 - 1
- 28.3
- 9.11 * (2
- 206) 2(0.7)
- = 140,381 - 75,871 = 64.510 psi Sm = 15,434 pst G 406'F K, a 2.311 Recalculate 5p by usin8 higher transient data i sp = 299.408 +.1/1.4 * (1.87) * (28.3) * (9.11) * (2) * (291-206) .
+ 1/0.7
- 28.3
- 9.11 * (2)(74 50)
+ 1.87
- 1.35
- 28.3
- 9.11
- 2 * (132-92)
= 299,408 + 58,562 + 17.679 + 52,068 a 427,696 psi salt
- 2.311/2
- 427.696 = 494,203 psi < 499,013 psi A --
, Previous Eq (14), Salg using ASME III '74 As shown above, the recalculation of the alternating stress intensity, using-higher charmal transients, yields results bounded by the existing calculation. The conservation of the ASME III '74 approach was recognised by the calculation originator / checker and supervisor and, as such, the existing calculation is acceptable and conservative.
=======.=====...============ ...=========... =========================================
0at6 ?tCo lleview Signature x z_/,3 's 9
/qate '
. l .,A zuo >/nlee SechtW1 Appr6 val signerure ' /Date' Page 1 of ll
l l gg STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (U 1 (J.O. NO. 18138) f OBSERVATION REPORT MAR 0 91980 Observation Report No. DOR-043, Rev. 0 l PART II - ADDITIONAL RESPONSE SUPPLEMENT i' As previously verified that by using the ASME Code Section III 1977 Edition with the Suamer 1979 Addenda, the alternating stress calculated for LICI, TIJ's @ dp 100, , 140B, and 150 is lower than previously calculated using the 1974 Code. Therefore, I the previously calculated CUF of 0.6493 remains valid. By using the 1979 Code, the thermal stress ratcheting is also to be checked in accordance with NB-3653. ( A Tl ) range ,5 oC C4 The transients of concern are the 406'F to 50*F step charge and the 50'F to 406*F step change.
~
.. Tavg = 406 + 50 = 228'F 2
O Q Sy = 20,484 psi @ 228'F for SA358 TP316L E = 28.3
- 106 psi oC = 9.11
- 10-6 in/in/'F
= 1.3 for austenitic material C4 To find y', x should be calculated first x = ( PD ) (1)
E & P = max. pressure for the condition = 250 psig D = 12.75" t = 0.687" x = ( 250
- 12.75 ) ( 1 ) = 0.1133 2
- 0.687 20,484 y' = 1 = 8.83 (The relation of y' and x with the given pairs of y' x and x values matches with the case 1 in NB3222.5)
O Page T of if ; 1
TE' 0:215-S"I-"575 #I y,:#_;; 33 c3::: l'sien.3 5 SCIE 7 T Ll'c Conservation Report te. DOR.043, Rev. PART !! - RESPONSC CohT!*lC
.. ( A T1 ) FAnge 4 (-- 8.83
- 20,484 )
- 1.3 = 1303*r 0,7
- H. 3 p. u Max ( A T 1) carge a 2 x 291 = 582* F < 1303* F, .', acceptable
/
I l I
. I asemessessassesassammesessenessamesasassessemesassaaneeeemanesanesammemmaammas lo O W N $$$ 3'QEb PECU Rey tow Signature / pat 4 Becntal AppstpM1 St(nature /Date Page l of 1
\ e2esacsta axn; c:c ,nm tsict ca, so ww tta
STONE & WEBSTER ENGINEERING CORPORATION
' LIMERICK 2-IDCA (J.O. No. 18138)
OBSERVATION REPORT []
%J Observation Report D OR- 043 Rev. O Review Plan LK- D -
1903A - MC Rev. 1 PART III - RESPONSE EVALUATION The response to this observation report has been evaluated'as follows:
-A. Causal Factor (s): Acceptable Not Acceptable (Explain):
N/A B. Extent of Condition: Acceptable Not Acceptable (Explain): N/A C. Significance: Acceptable 'Not Acceptable (Explain): N/A D. Corrective Action: Acceptable Not Acceptable (Explain): N/A E. Action to Prevent Recurrences _ Acceptable Not Acceptable (Explain): l l Not Required
. CO.
.. .NT.
.. .I.NU.
...ED.O. N. ONE.
. P. A.G. E. .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . = . .
Additional Action Required: No J_Yes (Explain) Bechtel to complete supplemental calculations for all similar_ cases and incorporate supplemental calculations as part of the calculations of record for Unit 2 Class 1 stress calculations for each affected case. 9Ll~~lJa:A 4HM . Ch k % b.w/Ile q/n/e1 Lead Engineer Signature / Date APM Signature / Date Page [0,, of /_/
l STONE'& WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA~ (J.O. NO. 18138) OBSERVATION red' ORT Observation Report DOR- 043 Rev. O PART III - RESPONSE ETALUATION CONTINUATION: The response to DOR-043 lacked adequate technical justification for using exponential temperature decay to reduce temperature 'in the dead leg connecting to the RPV nozzle. The response did not adequately address l the technical basis for reduction of temperature in the step transient from 406'/50*F to 300 /50 F. Bechtel subsequently transmitted a response supplement which recalculated Salt stress intensity based on 1977 ASME Code, Summer '79 Addenda (which is the applicable FSAR Code requirement for LGS Unit 2). Using the 1977 coue '79 summer Addenda, the response supplement recalculated reduced Sn value in Eq 10 of code subsection NB-3653. A reduced Sn value yielded a lower Salt stress intensity, which in turn reduced _ cumulative usage 1'ac to r (CUF). However the response supplement failed to provide evaluation of NB-3653.7 (Thermal Stress
~
Ratchet). Bechtel transmitted an additional response supplement evaluating the Thermal Stress Ratchet effect. The total response consists of the original response and the two supplements. Provided that . the supplemental calculations will be made part of the Unit 2 Class 1 stress calculation and the stress report (since FSAR commitment is ASME III 1977 code up to '79 summer addenda), the response for this I case is considered acceptable. These documents still have _not been completed by Bechtel, and therefore not reviewed by SWEC. Bechtel's response to DOR-043 indicates that Bechtel has in a few additional isolated cased reduced temperature in dead legs by applying an exponential temperature decay equation. Therefore, the action taken for this case must also be completed for the additional cases where the exponential aecay equation was utilized to reduce temperature in dead legs.
~
O Py e .U of .U
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2'IDCA' (J.O. NO. 18138) l ,[] OBSERVATION REPORT _$ ) l Observation Report No. DOR- 044 R e v . _0__ Review Plan: LK- D- 1903 - MS Rev. l_ Reference AI No.: DAI-066 Rev. O PART I - INITIATION
- 1. Description of Concern The calculation for the RHR system pump Net Positive Suction Head (NPSH) for Mode A has incorrect results because the formula used is not appropriate and uses inputs that are inconsistent with the as built piping isometrics. Also the RHR System pump NPSH calculations do not provide documentation that supports the FSAR licensing basis for available NPSH. In addition, the calculations do not address all operating modes.
-2. Supporting Information The RHR system pump NPSH calculation; M-51-9, for Mode A applied a method to calculate NPSH available that adds the velocity head term (V2 /2g) which is inconsistent.with the Hydraulics Institute Standards and is non conservative. The standard shows that the velocity head /" - term is not included when determining the NPSH available for proposed I installations. The formula used in the /SAR is consistent with the standkrd.
The calculation for Mode A is not consistent with the as built piping isometrics with regard to piping run lengths. This iticludes omission of the piping length between the strainers and the inboard penetration isolation valve. Also the calculation acknowledges the 30"X24" reducer but " neglects" its pressure drop without documented justification. It is agreed that the formula used in the calculation and the current design appears to provide adequate e.vailable NPSH however this approach of calculating NPSH could lead to design errors. CONTINUED ON 1 PAGE Additional Documents Attacned: None o) Le(d ITFg~inhkr S:.'gnature (v 8
/ Date Bk
...yi.........................................................................
Lm whzirr APM Sigplature aHW ' / Date Page 1 of g
l-(- STONE & WEPSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) r~ OBSERVATION REPORT ( Observation Report No. DOR- 044 Rev. O CONTINUATION: l Additional concerns from the review of the RHR system pump NPSH calculations M-51-9, 10..and 11 are the following:
- 1. The calculations do not provide the documentation that supports .
the licensing basis for the FSAR demonstration of conformance I with Regulatory Guide 1.1, reference FSAR Section 6.3.2.2.4.1.
- 2. NPSH calculations do not address all operating modes of the RHR system. Only modes A, B, and D are addressed. All modes of RHA should be addressed to confirm that adequate NPSH fa ,
available under all operating conditions including runout , I flow. m s Page 1 of T
ST;NE & WEPSTER ENGINEERING CORPORATION i LIMERICK 2 IDCA l (2.0. NO. 18138)
; OBSERVATION REPORT w/
Observation Report No. DDR- 044 nev. O I PART 1 - INITIATION 00NTIEUATION: Additional concerns from tha review of the RHR system pump NPSH calculations M-51-9, 10, and 11 are the followings j
- 1. The calculations do not provide the documentation that supports the licensing basis for the FSAR demonstration of conformance with Regulatory Guide 1.1, reference FSAR Section 6.3.2.2.4.1.
- 2. NPSH calculations do not address all operating modes of the RHR system. Only modes A, B, and D are addressed. All modes
' of RHR should be addressed to confirm that adequate NPSH is available under all operating conditions including runout flow.
O
\d O
,... a o. ,
u
STONE & WEBSTER ENGINEERING CORPORATION fDCoA? /OY4 LIMERICK 2 IDCA 3eM 3 (J.O. No. 18138) MAR 101989 OBSERVATION REPORT Observation Report No. D OR- 044 Rev. O PART II - RESPONSE AMENDED RESPONSE The following is an amended response which has been prepared in accordance with agreement reached between SWEC and Bechtel during a telephone conference on February 28, and March 1,1989.
- 1. Observation Concurrence:
_1g_ Concur with observation _ZL, Do not concur with observation (Note: if not in concurrence, explain in
" remarks" below)
- 2. Response to Observation:
A. Causal Factor (s) Velocity Head Term The Hydraulic Institute Standard requires that the NPSH be determined by subtracting the vapor pressure from the total pressure, and defines the
.tita] pressure as the sum of the velocity head and the static head. The causal factor f or ' the inadvertent addition of the velocity head in simulations 8031-M-51-9,10, and 11 is the originator's and checker's misinterpretation of the equation for NPSH as presented in the referenced textbook . The specific inclusion of the velocity head term in the example cited in the Hydraulic Institute Standard when combined with the expression f or NPSH in the text referenced in the calculations resulted in the calculations' adding the velocity term in a way that effectively added the velocity term twice.
B. Extent of Condition Velocity Head Term This is an isolated occurrence. The textbook referenced in these calculations for RHR NPSH was only utilized by this particular Bechtel engineer for this purpose. This conclusion is, based on a review of all the NPSH calculations (nine calculations) for all the ECC systems. This velocity head term is included only in the three calculations identified above. kJ
*R. M. Olson, Essentials of Ennineerinn Fluid Mechanics, third Edition. Intext Educational Publishers, 1973.
t Page 3 of ,1
fh W Y STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. No. 18138) OBSERVATION REPORT Observation Report No. D OR- 044 Rev. O PART II - RESPONSE CONTINUED C. Significance Velocity Head Term The significance of this error is small because the velocity head for liquid flow in piping at the low velocities found in these systems is usually less than one foot. In particular, in Calculation 8031-M-51-9 the velocity head term contributes only 0.36 feet of head loss. When considering less than one foot of head relative to the over 30 feet of excess NPSH available in this case, the effect is negligible. D. Corrective Action Velocity Head Term O Calculations 8031-M-51-9 and -10 have been superseded by Calculation 8031-Q M-51-62. This new calculation does not include the velocity head term. E. Action to prevent Recurrence Velocity Head Term There is no need for an action to prevent recurrence because this l calculation method was sa isolated case, and the Bechtel Mechanical Standards already specify that the methods of the Hydraulic Institute Standards shall be employed. Remarks As-built consistency We do not concur with this item of the DOR. When the original calculation 8031 -M-51 -9 was performed in May 1972, the only piping information ; available were the piping sketches. Based on discussion with project personnel, when the as-built piping isometrics became available the calculation was reviewed for consistency with these new drawings and the differences were considered not to be sufficient to justify revision of the calculation (refer to item 3 of DAI-066 response for details). As for
.s the calculation neglecting the 30"x24" reducer, the author of the calculation purposely did this because this specific fitting was tabulated along with others and the word " neglect" was written next to it in the tabulation of losses. Hence, the author of the calculation, as well as the checker, each had to make a conscious decision as to the significance Page 3 of $
STONE & WEBSTER ENGINEERING CORPORATION l LIMERICK 2 IDCA (J.O. No. 18138) O Q OBSERVATION REPORT Observation Report No. _A OR- 044 Rev. O PART II - RESPONSE CONTINUED of the fitting. We can only conclude the author and the checker used their judgement and technical familiarity with the subject to appropriately identify and treat the fitting. Without assistance from the author or checker, we were able to quickly come to the same conclusion. To further underscore the insignificance of the fitting and the appropriateness of the author's and checker's judgement in response to this DOR, we note the reducer involved in the subject calculation would only add 4.35 feet of equivalent length to the approximately 475 feet of equivalent length of suction piping. Additional Concerns we do not concur with the additional concerns presented in the DOR. Regulatory Guide 1.1 requires that
- Emergency core cooling and containment heat removal systems should be desinned (emphasis added] so that adequate not positive suction head (NPSH) is provided to system pumps assuming maximum expected temperatures of pumped fluids and no increase in O containment pressure from that present prior to postulated loss of coolant accidents."
Because NPSH is a function only of the pump inlet conditions it is not necessary to celculate the available NPSH for all the operating modes of the RHR systel. At is only necessary to calculate the NPSH for inlet flow conditions that envelope the discharge side operating modes of the system. Therefore, the two calculations shown in FSAR Section 6.3.2.2.4.1, which are based on calculation 8031-M-51-62, show that the Limerick design meets l the requirements of Regulatory Guide 1.1 for all operr. ting modes of the RHR system where suction is from the suppression pool. For the shutdown cooling mode where suction is from the recirculation loop, calculation 8031-M-51-11 shows the NPSH available. The maximum flow rate for any containment cooling mode of the RHR system is 11,000 gpm as shown for Mode A-2 in FSAR Figure 5.4-14. For this mode the maximum temperature is shown as 180' F. Hence, the NPSH for this mode is presented as case 1 in FSAR Section 6.3.2.2.4.1. Case 2 in FSAR Section 6.3.2.2.4.1 can be used to show the NPSH available at the maximum flow rate of 11,000 gpm and 212' F. Because the head loss due to flow is proportional to the square of the flow, a 10% increase in flow rate from 10,000 gpm to 11,000 gpm will only increase the flow head loss by 1.58 feet, from 7.54 feet to 9.12 feet which contributes an insignificant decrease in the total NPSH available as shown in calculation 8031-Misc-65, Rev. 0 (transmitted to SWEC via transmittal S0668). It is also not O necessary to evaluate as a separate case the condition of pump runout flow. It is shown in the calculations, and it is stated in the FSAR, that Page .I of .3.
2pcA/WYG 519NE & NESSTER ENO:NEERING 00aPORATION LIMBRICE 8 ItcA (J.O. NO. 18188) onstaVAT!0N REPORT Observatten Report No. jL OR ,,,A M _Rev. jk PART II - RERP0MBE gDWTINUED __ the oystem resistance is sufficient to prevent runeut. Theref ore,11,000 spa is the maximum flew that needs to be considered. Documentary evidence that the ECCS pumps meet the requirements of Regulatory Guide 1.1 is provided in calculatten 8081-M!sc-65. Revision 0 , (tranesitted to SEC via transmittal soses). This calculation summarises the available NPSH margin for all the Ecc8 pumps at the maximum operat'ng condition of each system. In addition, an LDcN which was prepared for i ether roteens, but happened to involve the rsAR sections that presented the NPSM calculattens, is in process and incorporates the NPSH values as shown in calculation 8081-N!8C-65. i The Limerick plant meets the requirements for design presented in Regulatory Guide 1.1 for all operating medes of the RHR system and the calculattens adequately demonstrate that the requirements of Regulatory cuide 1.1 have been sost. l l seeses sene s sessn e sesse s s e n e st e e s s e s s s s se m a n e s s e s s a s e n s a s s e s s e s s n e s s n e s e s s e s s e s s e s s e n e s k}
\O c- ,
% -b b , / h 1 7 W / // /
, f PEco paview signatere / Date J Sechtel Approval Signature / Date Page .h of 1
, STONE & WEBSTER ENGINEERING CORPORATION.
LIMERICK 2 IDCA (J.O. NO. 18138)
~
OBSERVATION REPORT Observation Report D CR- 044 Rev. 0__ Review Plan: LK- D - 1903 __ES_ Rev. 1__ PART III - RESPONSE EVALUATION The response to this observation report has been evaluated as follows: VELOCITY HEAD TERM: A. Causal Factor (s): X Acceptable Not Acceptable (Explain): The cause is explained to be an inadvertent addition of the velocity head term based upon human error. B. Extent of Condition X Acceptable Not Acceptable (Explain): The extent of condition was identified to be an isolated case based upon a review of "all the NPSH Calculations (nine calculations) for all the ECC Systems." C. Significance: X Acceptable Not Acceptable (Explain): The significance of this error was shown to be small and the effect was determined to be negligible. D. Corrective Action X Acceptable Not Acceptable (Explain): Calculation 8031-M-51-9 and -10 have been superseded by calculation 8031-M-51-62 which does not include the velocity head term. l E. Action to Prevent Recurrence: X Acceptable Not Acceptable (Explain): Because this was an isolated case and standards are in place that specify that methods of the Hydraulic Institute Standards shall be employed, action to prevent recurrence is not required.
................................................................====............=...===
Additional Action Required: X No Yes (Explain)
..p................................................................... t Mito 3/2 3/99- % c 4 A & D. W ille. s/de Lead Engilieer $igna'ture / Date' AFM fignature / Date j Page 1 of _$
STONE & k'EBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT Observation Report DOR- 044 Rev. O PART III - RESPONSE ETALUATION CONTINUATION: j AS-BUILT CONSISTE? ICY: The response provides the information that the reviewed calculation has been superseded and therefore the cor.cerns about the i inconsistencies with the as-built condition have been resolved. ADDITIONAL CONCERNS: The response indicates that a calculation has been performed (8031-MISC-65) which documents the enveloping conditions of Regulatory Guide 1.1 compliance. This resolvec the concern that Regulatory Guide 1.1 was not supported by a design calculation. The response indicates that an LDCN is being processed which af fects the FSAR sections about NPSH and Regulatory Guide 1.1 and will incorporate the NPSH values shown in the calculation. It is accepted that the RHR pumps design does have adequate NPSH under the maximum condition of Regulatory Guide 1.1. O 8 - - l 1
STONE & WEBSTER ENGINEET.JG CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) I J 1 OBSERVATION REPORT j Observation Report No. ' DOR- 045 Rev. Q_ Review Plan: LK A - 1903 1 R ".1 Reference AI No.: DAI-198 Rev. O l PART I - INITIATION
- 1. Description of Concern l The method used to compute blowdown flow in calculation No. 2006 for high energy line breaks (HELB) during the inventory period is no~ consistent with and less conservative than the industry accepted computation method. This may result in nonconservative compartment pressure and temperature transients for use in the environmental design criteria.
x
- 2. Supporting Information The response to action item no. DAI-198 Rev. O states that the inventory blowdown period was determined based on the ANSI standard, ANS-58.2
" Design Basis for Protection of Light Water Nuclear Power Plant against Effects of Postulated Pipe Rupture". For the inventory-blowdcwn, however, instead of a step function as recommended in the ANSI standard, it was assumed to be linearly re.mp'ed down to
( the next po,ssible maximum blowdown flow rate. The response states that this is a more realistic yet still conservative representation of the physical phenomenon, but provides no technical justification.
. It is noted that methods used to calculate break flow require NRC approval in accordance with the Staff Position contained in BN-TOP-4, revision 1, October 1977, page E12.
l
.............................................................................m.........
Additional Documents Attached: None 'gc)
/ - a k 12l2/ Date Lead Engineet Sigdature 081 t
2Mfr'
'/ Date APMSignpture Page1offt
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.0, No. 18138) Z[ [A OBSERVATION REPORT "" O 7[ / Observation Report No. ,,12._ OR- 045 Rev.'O AMENDED RESPONSE The following is an amended response provided in cecordance with agreement reached between SWEC and Bechtel in meetings on March 20, and 21,1989. PART II - RESPONSE
- 1. Observation Concurre'.tce:
Concur with observation _XL, Do not concur with observation (Note: if not in concurrence, explain in
" remarks" below)
- 2. Response to Observation:
A. Causal Factor (s) B. Extent of Condition C. Significance D. Corrective Action E. Action to Prevent Recurrence ss==ssss===sssssssssssss===ss====================================================ss= Remarks The method used to calculate the inventory blowdown is based on ANSI Standard 58.2, which has been modified to " ramp" the blowdown rate down from its maximum to the steady-state rate. This is less conservative than the step-function method used by I the ANSI standard. However, Bechtel believes that this " ramped" approach is still not only a conservative, but also a more realistic method to estimate the blowdown flow rate. In order to demonstrate that a " ramped" blowdown is indeed a conservative representatfs of the true blowdown, a calculation was performed to simulate the mass flow ratt from a vessel through 275' of a 11.374" diameter pipe to the atmosphere. The vessel was assumed to have a constant pressure and temperature of 1024 psia and
- l. 547.5 F respectively. The " minor" or local loss parameter K was taken to be equal to
- 16. For simplicity, the working fluid was taken to be air rather than steam. The calculation was performed using the method of characteristics to simulate the flow through the pipe. The mass flow rate at the pipe discharge as a function of time was taken to be the vessel blowdown.
Page 1 of fg_
STONE & WEBSTER ENGINEERING CORP 02ATICN LIMERICK 2 IDCA N [ [ (J.D. NO. 18138) OBSERVATION REPORT Observation Report No. D OR- 045 Rev. O PART II - RESPONSE CONTINUED It was observed that the blowdown rose steeply from zero, reaching its maximum value at time t p This steep rise from zero is an artifact of the initial conditions of zero velocity imposed throughout the pipe. Aftec the blowdown had attained its maximum value, it steadily decreased, reaching a steady-state value at time t .2 The first derivative of the blowdown with respect to time (slope) was everywhere less than zero for times greater than t,. In addition, the second derivative was observed to be greater than zero for tgitit. Hence, 2 the blowdown curve in the range t it5t 2 3 is concave upward. Therefore, ramping the blowdown between t3 and t2 "III 'everywhere bound the true mass flow rate. From this simple calculation we conclude that our deviation from the ANSI standard is still conservative, but better represents the true blowdown. In addition, blowdown reruits obtained using this method and which clearly reflect the ramped down blowdown behavior, are presented in FSAR Table 3.6-6 and Figure 3.6-10, which was reviewed and accepted by the NRC. This satisfies the NRC statement made in their acceptance of BN-TOP-4. SuDDiemental Response tt DOR-045 The-following is provided in response to SWEC concerns presented in meetings between SWEC and Bechtel on March 20, and 21,1989. Bechtel's modified method, based on ANSI Standard 58.2, for determining the high-energy line blowdown mass flow rate is a conservative method. Bechtel's method linearly ramps the blowdown mass flow rate from a peak value for frictionless flow to a value for flow with friction. These mass flow rates are derived from Hoody's diagram " Pipe Maximum Steam / Water Discharge Rate". This method is less conservative than using step functions as given in the ANSI Standard, but bounds the realistic blowdown mass flow rate. In demonstration of this fact, a calculation was performed to determine the transient blowdown mass flow rate for a typical case of high-energy line break. This calculation demonstrates that the ramped mass flow rate used in the original calculation 2006 Rev. 2 Case 1 is conservative. These transient calculations were performed using Bechtel's l Standard Computer Program GAFT. GAFT is a Bechtel Standard Computer Program which utilizes the method of characteristics to calculate transient ideat-gas flow. l Validation of the GAFT program was performed by comparing calculational results with the industry standard computer program RELAP5, and with Hoody's sample problem on page 405 of " Thermal Hydraulics of a Boiling Water Reactor". Examination of the GAFT User's manual by Stone & Webster Engineering Corporation personnel to confirm that GAFT is suitable for these types of calculations was also performed. Weighting factors are l "O used for improving the convergence of the pressure calculation. Default values of the . weighting f actors are built into the program. These f actors are described in the GAFT l Program User's Manual, Rev. 2. Section 4.3 and 4.4. Page .1 of .{a.
l ZDM /924
. ,, .n .t -
., l' . . L . ~ .. .. ,. .- STONE & WEB 871R ENGINEERING CORPORATION
~ '
' "Nj[e$j$jikdN..'.
pa
'N y.t.1
.' ~.(J ;-
LINERICK t IDCA. 0..No. 14138). ..;;. . i .. . . N , ff'f.64,
.. .N% N..Ti h* 7 . ?.. . . ' . ' e C* !' ' j'
- 5[ Ypg;393fi,'$1lk25.'N'.,j,%.M{'0BSth.vh!0N"RIPORT f ,s. .. .;u!Fsyv . .--
. m n .. . . . , . . . . . .
I* h:: # 'h n:. b- :r
' * * P
#". e. * *l - *. * '; '
. ;**'l"h? 4 '4. fj'.y.*v:(+'~. .+ , : * ' r*:r, '.~ ; . :% .*
." *z.,
+
1.,..c...w
- s. s -. 9-@!'
- .:.p ..:
,+ ,
. Observation Report No.1 OR .ME, Rev.1 d,- &4.iGs:3.t: . , J .W .
~ :, . . ,
..a.i..}' yk.1T w.:.d' 3(FART 11 = RESPOMBE ChMTIE!1D
! .in.h.s -' r;' ' ~.. 4. . :1.:. .
r .~ -
- .. 3 , -
'y ' To evaluate the effect of a small flow reelstance coefficient on the comparison with p :.the linear ramped blowdown mass, additional calculatten of two cases of high-energy is c:line break was performed. These two cases were selected to be those with the smallest j.- flev resistance coefficients. The calculational results demonstrate that linearly iK '@, ramping the., blowdown mass flow rate is still conservative. These calculations are '
5 transmitted'to' stone'& Webster Engineering Corporation via transmittal 80743. M'g&UMi$ fi4 ~ $h?l0.' ~. ~ . $l -
' +
b%
- )
l
,r
?)b*
*~
v
'X t4 A W lb fhff J/h///9 i
. \' 'f ev. ...4f.
Sechtel Approval signature / Date E ' :te.vl ow'. 81gna re Date -
~~gj.y :,. .
.h..
- ~ ' ~
rsee'.!LorJe
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA j (J.O. NO. 18138) O.
- j'
\ OBSERVATION REPORT Observation Report D OR- 045 Rev. J ' Review Plan LK- D - 1903 - MS Rev. J PART III - RESPOESE EVALUATION The response to this observation report has been evaluated as follows: A. ::ausal Factor (s): Acceptable Not Acceptable (Explain): NOT APPLICABLE B. Extent of Condition: Acceptable Not Acceptable (Explain): NOT APPLICABLE C. Significance: Acceptable Not Acceptable (Explain): /3 NOT APPLICABLE D. Corrective Action: Acceptable Not Acceptable (Explain): NOT APPLICABLE E. Action to Prevent Recurrence: Acceptable Not Acceptable (Explain): NOT APPLICABLE
=======================================================================================
Additional Action Required X No Yes (Explain) SEE ATTACHMENT O == ====== = ============ ============================================================
% /A) h, h[ [
Lead'Enktfieer Signature ' ' APW Signature / Date
~
/ Date '
Page 1 of (, l __________________.__._.________._._______.___.______m_ _ _ _ _ _ _ _ _ _ . _ . _ _ _ . _ _ _ _ _ . _ _ . _ _ _ _ _ . . . _ _ _ _ _ _ _ _ _ _ _ _ . _ _ . _ __ _ _ ___
STONE & WEBSTER ENGINEEUNG CORPORATION LIMERICK 2 IDCA (J.O. No. 18138) OBSERVATION REPORT Observation Report DOR- 045 Rev. O 1 PART III - RESP 0BSE EVALUATION CONTINUATION: The response has provided calculations which were performed to demonstrate that the ' ramp' blowdown flow during the inventory period is a more realistic method to estimate the blowdown flow rate. These calculations were performed using Bechtel's computer program GAFT which has been qualified by comparing program results with RELAPS results. Input variables such as ' weighing factor pm1' and ' weighing factor pm2' were also clarified by the response as factors used to improve the convergence of pressure calculations. The selection of these variables will not impact the calculation results. Three representative cases with friction factor range from (1/d=4.7 (smallest f1/d used in Calculation Number 2006) for the main steam line break to f1/d=18 for the HPCI break were analyzed in these calculations. The results indeed demonstrated that the linear ramp down of the blowdown flow during inventory period is still conservative for all three cases. The response is acceptable. O 6 O P a 0 e ,4 0 4-G
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA' (J.O. NO. 18138) OBSERVATION REPORT Observation Report No. D OR- 046 Re v . 0__,,, Review Plan LK- D - 1903 - MS Rev. 1 Reference Al No.: DAI-197 Rev. 0 DAI-201 Rev. O l. i- PART I - INITIATION i 1
- 1. Description of Concern Calculation NO. 2006 for high energy line breaks (HELB) contains i
inputs which underestimate the blowdown flow and may result in non-conservative compartment pressure and temperature transients for use in the current environmental design criteria.
- 2. Supporting Information A. The response to DAI-197 Rev. O has concurred that the valve friction factor of K=1.95 used in the calculation is higher than the vendor provided value of K=1.20. The response has indicated that the calculation is still conservative since it has not included the effeet of the valve closing and flow restrictor, and that the overestimate of the valve k value will be offset if these effects were considered.
O The position regarding the flow restrictor is incorrect because calculating choke flow at the flow element a,nd eliminating all piping friction loss downstream of the flow element will result in higher blowdown flow. For the valve closing effect, the response indicates that "with a 507. valve open position, the friction k is about twice as large as that of a full open valve, and the increase in the friction factor is enough to offset the original overestimate of the full open valve frictional length." While this statement is judged to be correct, it still means that the blowdown flows are underestimated during the period that is equal to the valve closing from the fully open to the 50% open position plus the valve closing signal delay time. Also, with a postulated single active failure of one of the two isolation valves, the remaining valve has to close more than 50% to offset the overestimated k value of two fully open valves. CONTINUED ON 1 PACE
=======================================================================================
Additional Documents Attached: None t ar->dzW O - - - - - - - - - - - - - --
< 2 a Lead' En'gnekrVSignsture '
/ Date APM Signat4re /'Date Page 1 of d
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT Observation Report No. DDR- 046 Rev. O PART 1 - INITIATION CONTINUATION: B. The response to DAI-201 Rev. O has concurred that the valve closing history table used in the calcualtion to demonstrate the acceptance of an additional 2 second delay on the valve closing signal is in error.The response indicates that the addition of a 2 second delay on the valve closing signal will not affect the calculated peak temperature and pressure. This is claimed because the previous blowdown flow is calculated based on the valve being fully open for the entire valve closing period of 13 seconds and if the throttling effect of closing the valve is considered, then even with the additional 2 second delay, the blowdown will still be less than the blowdown previously calculated. The response further indicates that since the blowdown is significantly reduced at 13 seconds as the valve has nearly completed closing, the blowdown flow between 13 to 15 seconds will not be significant enough to impact the peak temperature which had been already reached. The statement regarding a significant reduction of blowdown flow at 13 seconds is incorrect because without a correct valve
~O) closing history table, the valve opening at the 13 seconds mark is indeterminate. Also if we assume a 20% valve opening at 13 seconds, then the calculated blowdown flow (based on K=50 which consists of friction k for one 20% open valve and k=20 from HPCI piping) is about 60*/. of the full flow (calculated based on the valve fully open condition). The blowdown flow between 13 to 15 seconds, therefore, can not be' deemed as insignificant.
O
.V Page 1 of IA
STONE $6 WEBSTER ENGINEERING CORs0RATICN LIMERICK 2 IDCA (J.O. No. 18138) g g 4 jggg OBSERVATION REPORT ~O Observation Report No. p._ OR- 046 Rev. O_ PART II - RESPONSE
- 1. Observation Concurrence:
Concur with observation IL, Do not concur with observation (Note: if not in concurrence, explain in ,
" remarks" below)
- 2. Response to Observation:
A. Causal Factor (s) B. Extent of Condition C. Significance D. Corrective Action p E. Action to Prevent Recurrence U .................................................................................... Remarks AMENDED RESPONSE The following is an amended response prepared in accordance with agreement reached between SWEC and Bechtel in a telephone conference on February 28, 1989, and in meetings between SWEC'and Bechtel on March 20, and 21, 1989. Calculation 2006 contains inputs which underestimate the initial blowdown flow rate.. However, the analysis approach does not result in non-conservative compartment pressure and temperature transients for use in the current environmental design criteria even with the additional two second delay time as discussed below. This is based on the fact that the slightly lower initial blowdown flow rate was conservatively applied j without considering the effect of reduced flow due to valve closing. In response to this DOR, we performed further analysis to confirm the conservatism and validity of our original assumptions. Our evaluation concluded that blowdown flow rate is significantly reduced at 13 seconds, the instant the blowdown was ended in the original calculation. Calculation 2006, Case 3, was reevaluated with the following changes:
- 1. The additional two seconds delay to account for a total three seconds delay time,
- 2. The correction of valve friction factor (full open k value (adjusted to 12" diameter pipe) from 3.0 in Calculation 2006 Revision 2 Case 3 to 2.4), I (Q) 3. The incorporation of the isolation valve closing characteristics (as given in l Reference i of the calculation Revision 4) on one of the two valves in series (the second valve is considered to fall open).
Page .,3_ of M J
STONE & WEBSTER ENGINEERING CORPORATION g LIMERICK 2 IDCA (J.O. NO. 18138)
.[M((/ h7 [
OBSERVATION REPORT Observation Report No. D OR- 046 Rev. O PART II - RESPONSE CONTINUED Case 3 of a HPCI break in the isolation valve compartment was selected for further study because there is more potential for the compartment temperature to rise in this case, whereas in cases 1 and 2, the peak temperatures calculated are already close to the theoretical maximum of 3009 following an isenthalpic expansion of blowdown steam to 14.7 psia.
It was found that the e ak compartment temperature was reduced slightly in the revised calculation as expectea. Comparison of peak pressures and temperatures and blowdown data for the revised and original calculations are presented in the following Tables 1 and 2. Table 1 Comparison of Blowdown Data HPCI Break in Isolation Valve Compartment Calculation 2006 Case 3 Calculation Revision 2 Reevaluated Case 3 l ( Total K Blowdown Enthalpy Valve Total K Blowdown Enthalpy Valve Time open (12 in) (Ib/sec) (BTU /lb) Open (12 in) (Ib/sec) (BTU /lb) 0.000 100% 0 2940 1192.4 100% 0 2940 1192.4 0.135 100% 0 1272 1192.4 100% 0 12"' 1192.4 0.230 100% 0 902 1192.4 100% 0 Je 1192.4 0.475 100% 16.2 328 1192.4 100% 0 351 1192.4 3.000 100% 13.2 351 1192.4 5.400 80% 13.62 347 1192.4 7.800 60% 14.54 337.3 1192.4 10.200 40% 17.92 302 1192.4 12.600 20% 37.40 21 5 1192,4 l 13.000 100% 16.2 328 1192.4 14.000 100% 16.2 0 1192.4 15.000 0% m 0 1192.4 0% m 0 1192.4 (Note: Blank spaces are conditions not evaluated for that case) The following should be noted: Calculation Revision 2 Reevaluated Case 3 Signal Delay Time: 1 second 3 seconds g valve closing: Both valves full open One isolation valve A during the closing full open, one valve duration of 12 seconds closing in 12 seconds Page 9 of Il
STCHE & WEESTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. No. 18138) /f OBSERVATION RFPORT Observation Report No. D OR- 046 Rev. O PART II - RESPONSE CONTINVED Table 2 Comparison of Peak Prassures and Temperatures HPCI Break in Isolation Valve Compartment Calculation 2006 Case 3 Revised Original Calculation Case 3 for Case 3 Isolation Valve Compartment 263.623 F 256.195*F Elev. 217' @ 13.62 seconds 6 13.0 seconds 16.0891 psia 16.0888 psia 6 0.185 seconds @ 0.18 seconds Steam Venting Tunnel 260.142 F 252. 944'F Elev. 241' 8 13.18 seconds 6 13.0 seconds 15.5354 psia 15.5353 psia 6 0.188 seconds @ 0.19 seconds Pressure is not a concern because it peaks before the valve begins to close. Bechtel standard computer code FLUD (NE017), current release 7.2, was used to perform the reanalysis. Supplemental Response to DOR-046 An additional study was performed to incorporate ottr response to the concerns discussed in the telephone conference call between Bechtel and SWEC on February 28, 1989. The l results of this study further confirmed our previous conclusion that the peak temperatures and pressures calculated in calculation LGS 2006 Rev. 2 for HPCI breaks are conservative. The Calculation 2006 Revision 2, results exceed those of both studies performed for the DOR response. The DDR response studies include a total of three seconds isolation delay time and a more realistic valve friction flow loss. The following describes the additional study performed for this supplemental response. The blowdown mass used and the results are presented in Tables 4 through 6. I An additional calculation of high energy line break in the HPCI/RCIC/RHR isolation valve compartnant was performed to incorporate the following changes in blowdown data O to satisfy concerns item 3B, 3C of " Telephone Conference Items Related To Bechtel Response to DOR-046" (IDCA DC #1794, dated February 22, 1989). Page 1 of Il
i l STONE & WEBSTER ENG15ERING CORPORATION LIMERICK 2 IMA (J.0, NO. 18108) ((gg /Q 7 bl OBSERVATION REPORT j Observation Report No. D OR- 046 Rev. O PART II - RESPONSE CONTINUED (1 ) Revise the frictional blowdown to include additional blowdown mass due to minor interpolation discrepancy of the Hoody's Pipe Maximum Steam /Nater Discharge Rate. See Item 3B. (2) Revise the frictional blewdown per Item 3C to include the " inventory steam" as simulated in original calculation 2006 Revision 2. The frictional blowdown changes were made by linear interpolation of: (A) Additional mass flow rate of 14 lbm/sec at full open valve position, (B) A flow rate of 8 lbm/sec at valve K=16.2, which corresponds to a valve stem position of 50%, and (C) A flow rate of O. Ibm /sec at valve full close position. The so called " inventory steam" of 164 lbs assumed in Revision 2 of Calculation 2006 was simulated by adding the " residual steam" to the compartment during a one second period af ter the valve is closed. This case is the same as Case 3 of Calculation 2006 g Revision 2. (\ ] The flow model and the method used in this revision are identical to that described in Calculation 2006 Revision 2 and the original DOR-046 response. See the following Table 4 and 5 for the blowdown mass changes, and Table 6 for the summary of results of various cases. The. results of this study further confirmed our conclusion that the peak temperatures and pressures calculated in staff calculation 2006 Rev. 2 are conservative. The calculation 2006 results exceed those from this revised calculation which includes a total of three seconds isolation valve delay time and a more realistic valve flow loss. It should be noted that the compartment peak temperature or pressure in a high energy line break event is not only a function of the " total" energy discharSed into the compartment. It is also a function of the " rate" of energy added to the compartment, if a flew path communicating to outside is available. l l NJ j l Page }r of 11
STONE & WEESTER ENGINEERING CORPORATION LIMERICK 2 IDCA Jhdg / h 7
.(J.0. No. 18138) f OBSERVATION RE MRT Observation Report No. D OR- 046 Rev. O PART II - RESPONSE CONTINUED Table 4. Additional Blowdown Steam Mass Additional Blowdown Hass Valve Moody Diag Residual Position System Interpolation Steam in Pipe Time Open K DOR-046 Item 33 DOR-046 Item 3C (sec) (Ibs/sec) (lbs/sec) 0.0 100% 0. O. O.
0.135 100% 0. O. O. O.23 100% 0. O. O. 0.475* 100% 0. 14.0 0. 3.0 100% 13.2 14.0 0. 3.001 100% 13.2 0. 5.4 80% 13.62 11.5 O. 7.8 60% 14.54 9.2 0. 10.2 40% 17.92 6.3 0, 12.6 20% 37.4 3.0 0. 15., closed 0.0 0. 15.001 closed 328. 16.0 closed 16.001 closed O.
- end of initial stage non-frictional blowdown I
l Page ~I of Il
I i STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA M(A [ (J.O. NO. 18138) 1 O OBSERVATION REPORT Observation Report No. D OR- 046 Rev. O PART II - RESPONSE CONTINUED Table 5. Comparison of Total Blowdown Mass Rate Original Supplements! Valve DOR-046 DOR-046 Item Position System Rev. 2 Revised SB and 3C Time open K Case 3 Case 3 Revised Case Enthalpy (sec) (Ibs/sec) (Ibs/sec) (lbs/sec) (Btu /lb) 0.0 100% 0. 2940. 2940. 2940. 1192.4 0.135 100% 0. 1272. 1272. 1272. 1192.4 0.23 100% 0. 902. 902. 902. 1192.4 0.475* 100% 0. 328. 328. 365.0 1192.4 3.0 100% 13.2 351 .0 365.0 1192.4 3.001 1192.4 ( 100% 13.2 5.4 80% 13.62 347.0 358.5 1192.4 7.8 60% 14.54 337.3 346.5 1192.4 10.2 40% 17.92 302.0 308,3 1192.4 12.6 20% 37.4 215.0 21 8.0 1192.4 13.0 328.0 14.0 0.0
- 15. closed 0.0 0.0 1192.4 15.001 closed 328.0 1192.4 !
16.0 closed 1192.4 16.001 closed 0.0 1192.4 4
- end of initial stage non-frictional blowdown Table 6.
Summary Results of Peak Compartment Pressures and Temperatures Original Original Supplemental Calc 2006 Response O DOR Rev.2 Case 3 Case 3 Case 3 HPCI max P 16.0891 psia 16.0888 psia 16.0888 psia Page T of _CL i
STONE & WEBSTER ENGINEERING CORPORATI!N LIMERICK 2 IDCA (J.O. No. 18138) [M h OBSERVATION REPORT Observation Report No. D OR- 04f_Rev. O PART II - RESPONSE CONTINUED RCIC 8 0.185 sec 0 0.18 sec e 0.18 see RHR
!so V max T 263.623 F 256.195 F 257.687 F Compt 6 13.62 sec 0 13. sec 8 13. see Steam mar P 15.5354 psia 15.5353 psia 15.5353 psia Venting 8 0.188 sec 0 0.19 sec 0 19 sec Tunnel umx T 260.142 F 252.944 F 254.650 F
@ 13.18 sec @ 13. sec @ 13. see Supplemental Response to DOR-046 The following is a supplemental response prepared in accordance with the agreement y reached between SWEC and Bechtel in a meeting held in San Francisco on March 20, and 21,1989.
(v) DOR-046 expressed concerns with the effects of additional HPCI/RCIC isolation delay time, the valve friction K value, and the residual steam in the pipe downstream of the isolation valve on compartment peak pressures and temperatures. These concerns were expressed for all HPCI and RCIC high energy line break cases, in addition to the case 3 HPCI break in the isolation valve compartment, which has been reanalyzed in response to SWEC concerns. To resolve further concern with all HPC' ACIC high energy line break cases, a calculation reflecting DOR comments was performed for other cases of HPCI and RCIC breaks. This calculation confirmed that the peak compartment pressures and temperatures calculated in the original calculation 2006 Rev. 2 are conservative, with the exception of case 1 HPCI break at the pump room. In case 1, the peak temperature increased about 1* F, which is considered to be insignificant. This increase is partially due to the modifications of the computer program since 1980. This reanalysis is transmitted to SWEC via transmittal S0743. For better comparison of the effect made in the revised calculation, the exact compartment flow mcdeis of the original calculation 2006 Rev. 2 were reproduced for the HPCI/RCIC line break cases using the current version of computer programs. In the reproduced case 3 which uses the current version of FLUD computer program (Version 7.2) the peak compartment temperature is about 3.6 deg lower than that of calculation 2006 Rev. 2. The lower temperature is the result of the change made in the heat transfer calculation to the heat sink. The latest version of the FLUD program uses the /~ " implicit method" in the transient heat sink heat transfer calculation compared with ' g] the " explicit method" used in the older version of the FLUD program. Page _3 of M
STOWE & WEldTER EN0!NEERING CORPORATION l I.!WERICK 2 !DCA
,(J.0, NO. 18134)
[p([ /h 055ERVAT!0N REPORT Observatten Report NJ. .]L OR ,,dORev. 9 . PART !! - 1 M PCNAE CONTINVID The output of the reproduced case, the original case, and the DOR-046 first and second responses. are air.n transmitted to SWEC via s0743. In the F'LUD program, the steam condensation heat transfer rate is determined by the method presented in the " Handbook of Heat Transfer'. edited by Rohtenov and Hartnett. Once the condensation has occurred, no re-evaporation is considered, which is conservative. a=====........ =====================essnessenesser=. esses =========================== k_ 7,9 63 /h Yd2d. thy PEco Reviev 81gnatuts b Date Bechtel Approval signature / Date Page /.f. of .L.,
[, 1 STONE & WEBSTER ENGINEERING CORPORATION I LIMERICK 2 IDCA (J.O. NO. 18138) O OESEld,ATION EEPORT
%Y Observation P.eport D OR- 046 Rev. 0 PART III o EESPOBSE EVALUATION l
The response to this observation report has been evaluated as follows: A. Causal Factor (s): Acceptable Not Acceptable (Explain): NOT APPLICABLE B. Extent of Condition: Acceptable Not Acceptable (Explain): NOT APPLICABLE C. Significance: Acceptable Not Acceptable (Explain): NOT APPLICABLE O-D. Corrective Action: Acceptable Not Acceptable (Explain): NOT APPLICABLE 1 I E. Action to Prevent Recurrence: Acceptable Not Acceptable (Explain): NOT APPLICABLE
...............................................................................=.......
Additional Action Required: X No Yes (Explain) SEE ATTACHMENT
- o. ...$ .L
. { .....g......h..............................................................
wf o!S9 CCLK L p wale sh>ln
/ Date Li(d Et.gitieVr Signature ' / Date AN Signature Page J{,of M
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT Observation Report DOR- 046 Rov. O PART III - EESPOESE ETALUATION CONTINUATION: The ObscTv3 tion Report concerns regarding the non-conservative calculation inputs were satisfactorily addressed by the response. The response provided a calculation which re-evaluated Case 1 to case 5 of Calculation Number 2006 with a revised blowdown flow. The results showed that the calculated peak pressure and temperature shown in Calculation Number 2006 are still conservative for all cases with the exception of Case 1 temperature which is increased by l'F. The response stated that this is insignificant and is partially due to modification of the COPDA program. This is deemed to be acceptable since Case 1 was performed without taking credit for heat sinks. If heat sinks were to be utilized, then the reduction of the peak temperature would be more than 1*F. For Case 2 to 5, the re-evaluated cases indeed showed a lower calculated peak temperature and an identical calculated peak pressure as compared to the results of the original calculation. But for Case 3, which is the only case utilizing heat sinks, the re-analysis was performed using e the latest version (7.2) of FLUD code. Since the old version used in b the original calculation is more conservative (which has been confirmed by the response), the re-analysis should be done with tne old version of the FLUD code in order to have a meaningful comparison. However, the response has indicated that the difference in the two versions is an improved convergence technique in the heat sink heat transfer calculation from an ' explicit method' in the old version to an ' implicit method' in the new version of the FLUD program. Methods used for modelling heat sink steam condensation heat transfer and re-vaporization fraction are not modified. Besed on our experience on numerical solution of heat transfer problem, the implicit method is always a preferred and a more accurate method than the explicit method. Therefore the re-evaluated Case 3 using the latest version of FLUD code is appropriate. O Pa0 c c Of u -
STOWE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA' (J.O. NO. 18138) OBSERVATION REPORT U j Observation Report No. - DOR- Ot.7 Rev. L i Review Plant LK-_D - 1907 - C Rev. 1 Reference Al No.: DAI-178'Rev. O PART I - IEX M Tig
- 1. Description of Concern Based upon test report 108026, the basis for requiring a conduit seal was not provided by the vendor. It is not evident tha* tearing of Rosemount 1153 Series B transmitters exposed the term 2nal block inside the test units to the aging Temperature-Humidity or Steam-Temperature (LOCA) environments.
- 2. Supporting Information Review of EQ package E-1 for Rosemount 1153 Series B transmitter and subsequent site walkdown identified that a totally sealed conduit installation is not utilized for the subject transmitters. The Action Item tesponse confirmed that sealed conduit installations were not utilized, and stated that nene of the , transmitters are
.O required to function under a HELB condition in the area where they are located.
The response did not address the possibility of moisture buildup and entry into the terminal block cavity of the transmitter through flexible conduit under normal fluctuating temperatures and high relative humidities, on the basis that the subject locations are not subjected to high humidity and temperature fluctuations during normal or abnormal conditions. It is noted, however, that at 90% R.H. , only a 3*-4*F ambient temperata-e reduction is needed to reach saturated (100% R.H.) conditions, so wide fluctuations are not necessary. CONTINUED ON 1 PAGE
....................w.e.. *v.eweyet42gn.L...............................................
Additional Documents Attached: None l l
...................................,.................. 4.... ... ................ 4 t 0, t ? r / 2 d l M adgngineer Signature / Date APM Sigfature '
/ 'Date I)
Page 1 of f Y(pttt 4 L . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - - _ _ _ - - _ - _ - _ - _ - _ - _ - - _ - - _ . - - _ _ _ _ _ _ _ _ _ _ _ _ _ _ - - - _ _ . - _ - - _ . -
STONE & WEBSTER ENGINEERING CORPORATION l LIMERICK 2 IDCA. (J.O. NO. 18138) OBSERVATION REPORT Observation Report No. DOR- 047 Rev. O ! PART 1 - INITIATION CONTINUATION: Additionally, the response addressed only the HELB condition. Other plant conditions such as Moderate Energy Line Crack, which could lead to higher relative humidities and water spray, must be addressed. Test report 108026 did not establish threshold environmental conditions beyond which a totally a,ealed conduit installation is required. Testing of Rosemount 1153 Series B Group B transmitters did not expose the terminal block inside the units to the aging temperature-humidity steam / temperature (LOCA) environments. The following information is from RMT Procedure 108026 and attachment II to report (RMT Procedure 1802):
- a. Paragraph 5.3 of Report 108026, page 9 states: "Unless specified otherwise in the individual test paragraphs, each test unit was configured for test as described below."
- b. Paragraph 5.3.1 Conduit Connections states: "One of the two conduit openings was sealed with a stainless steel pipe plug h-14 NPT. The Threads were coated with a thread sealant, NEOLUBE 100 (T.M. of Huron Industries, Inc.).
O The _ plug was tightened to 150 in-lbs, and sealant was cured per manufacturer's instructions. removed during the test program. The plug was not An adapter, NPT-to-compression fitting, was installed in the other conduit opening, and was sealed and torqued as described above. Its purpose was to allow connection of tubing to simulate conduit in the portions of test, such as steam / temperature, where this interface was . important. Pigtail leads, 22 AWG copper with KAPTON (T.M. of Dupont, Inc.) insulation, were connected to the terminal block inside the transmitter, and brought out through this conduit opening."
- c. Paragraph 6.3.1 of RMT Procedure 1802 states: "For the l Temperature-Humidity Test (Para. 13.3) and steam temperature !
test (Para. 19), the leads will be enclosed in metal tubing, connected to the unit with a \" tube to h" NPT SWAGELOK adapter. Since a totally sealed conduit installation was utilized during testing, and since only the HELB condition was addressed in the Action Item response, the use of an unsealed conduit installation for the subject transmitter has not been demonstrated to be acceptabic. 17 Page 1 of Mr
l STONE & WEBSTER ENGINEERING CORPORATION gg/ j l LIMERICK 2 IDCA T
, (J.O. NO. 18138) M Q 4 jggg OBSERVATION REPORT ($" O 7 7)
Observation Report No. DOR-047 Rev. O PART II -RESPONSE
- 1. Observation Concurrence:
AMENDED RESPONSE: The following is an amended response prepared in accordance with agreement reached between SWEC and Bechtel on March 8, 1989. The following is an amended response prepared in accordance with agreement reached between SWEC and Bechtel on March 22, 1989. Concur with observation X Do not concur with observation (Note: if in concurrence, explain in " remarks" below)
- 2. Response to Observation:
q i N / j, A. Causal Factor (s) B. Extent of Condition C. Significance D. Corrective Action E. Action to prevent Recurrence
=========================u==============================================
Remarks: Review of the Rosemount installation requirements, and confirming discussions with Rosemount, has established that Rosemount is concerned with the accumulation of moisture on the terminal side of the electronics housing causing an electrical leakage path. The Rosemoant housing meets the NEMA 4X requirements (Rosemount Report 2823A). Operation of the electronics portion is protected by a hermetic seal which separates the electronics side from the terminal side. As long as moisture accumulation cannot occur, humidity is not a problem. A recent Equipment Qualification walkdown verified that Rosemount units qualified by EQ package E-1 have flexible conduit entry into the bottom of the m
-)=============================================== ,,
Page 3 of oe
)K CS8932410/1
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) gg /pg q;
, OBSERVATION REPORT Observation Report No. DOR-047 Rev. O PART II -RESPONSE CONTINUED terminal side of the housing. This installation configuration will prevent any moisture building in the terminal side of the unit.
The Rosemount Model 1153 Series B transmitters installed without conduit seals are only located outside the primary containment. The units may see a meximum ambient temperature of 1159E and a maximum of 90% RH during normal plant operation. By LGS definition (FSAR Section 3.6) a Moderate Energy Line Break (MELB) cvent is the breaking of a line with fluid at Itss than 2009F and less than 275 psig concurrently -if either the temperature or pressure condition noted is exceeded, the event is defined as a HELB. Thus a MELB event would produce only very warm water, not steam. The potential of moisture buildup during normal operating conditions as a result of temperature fluctuations at up to 90% RH is very".ow. Even if it should occur, the exiting bottom entry flexible conduit configuration will preclude moisture buildup in the terminal side of the transmitters. It ('g.is thus concluded that the installed condition is acceptable. N) Sunclemental Resoonse to DOR-047 (Item numbers below correspond to those used by SWEC in their Telephone Conference Items Related to Bechtel Response to DOR-047, telecopied February 24, 1989.)
- 5. a. No credit was thken for the rigid or flexible conduit providing a qualified seal for the Rosemount units addressed in this observation Report. Therefore, no conduit qualification was necessary.
- 5. b. The Rosemount transmitter conduits are not sealed to provide protection from the effects of moisture condensation or the effects of direct spray. A thin film of condensation is not expected to occur during normal plant operation for the reasons discussed in item 5.e. Rosemount confirms that their published errors are applicable to the 1153B units installed at Limerick without qualified conduit seals as long as the conduit installation configuration will prevent accumulation of moisture in the terminal compartment. (Refer to transmittal S0692.)
l
========---======---=====-==-===m=---------=---==--==--=----===--=======
l , G G Page 4 of f4 CS8932410/2
STONE & WEBSTER ENGINEERING CORPORATION fpg7gg LIMERICK 2 IDCA (J.O. NO. 18138) t 'h OBSERVATION REPORT l ' observation Report No. DOR-047 Rev._0_ PART II -RESPONSE CONTINUED As regards SWEC's concern expressed in the March 22, 1989 meeting with Rosemount's telecon (Transmittal S-0692), Rosemount's position was that if the transmitters were exposed to a steam environment associated with a HELB or LOCA event, then qualification was predicated on the use of conduit seals. If the transmitters were not required to function during or after exposure to a steam environment assnciated with a HELB or LOCA event, then operability at humidity conditions up to 95% RH could be assured if the user ensured that moisture accumulation in the terminal side of the housing was not possible. The effect of direct water spray due to a MELC (i.e., MELB) causing water intrusion into the conduit is not considered because the likelihood of this event causing an instrument
#ailure is very small. The beses for this conclusion are as follows: ,m lV ) 1. A break must occur.
- 2. There must be a clear path from the pipe crack location to either the open conduit end, the intermediate junction box (if any), or the flexible conduit.
- 3. The spray must have sufficient energy and orientation to reach the cable tray /J box / flexible conduit.
- 4. The conduit tray must be open to the spray (i.e., no fire barrier wrapping).
- 5. The conduit leaving the cable tray is horizontal with !
internal cables which reduce the free opening of the conduit entrance for water to enter. .
- 6. The open conduit (at the cable tray /J box / flexible conduit) must be associated with a component required for safe shutdown for the specific MELC event. j
- 7. The water must reach the susceptible portion of the instrument and cause damage such that it will not function when called upon to do so.
l
===============================================r================r=======
n L ,) l Page di of } hE 1 CS8932410/3 l l
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA ZOM [ h$ (J.O. NO. 18138) [~ i OBSERVATION REPORT
\/~
Observation Report No. DOR-047 Rev. O PART II -RESPONSE CONTINUED
- 8. For Rosemount transmitters, the conduit is a bottom entry, thereby allowing a reservoir beneath the transmitter to collect water which may have intraded into the open end of the conduit.
- 9. A MELB is limited to piping at operating conditions of 2009F (or less) and a maximum pressure of 275 psig.
Therefore, a MELB could only generate a stream of very warm water, not steam, and would have only a localized effect near or under the pipe break. Normal systems are usually evailable after a MELB to provide shutdown capability in addition to the safety-related systems. MELBs are typically slow events; therefore, time would be available for operator actions to be taken to repair the failure. Even if a Rosemount transmitter were adversely affected, if the transmitter is such that it is required for
- s. safe shutdown, a redundant instrument is available. Single
\')
I failure of the counterpart instrument would have to occur concurrently with the postulated failure due to MELB.
- 5. c. A HELB will not cause Rosemount instrument errors, or degradation, which could mislead the operator and prevent him from safely shutting down the plant. The function and location of the Rosemount transmitters have been evaluated for their participation in the mitigation of a HELB accident. The results of this evaluation are sumaarized below.
- 1. Most of the Rosemount Type 1153B transmitters are not located in a HELB area and are, therefora, not exposed to HELB conditions requiring the use of a qualified seal.
The response to Item 5.b confirms than these transmitters will meet the published performance specifications without the use of qualified seals.
- 2. Three of the Rosemount transmitters are located in potential HELB areas but do not mitigate the consequences of a postulated HELB.
o PT-55-2N050 is located in Room 180 and monitors HPCI Pump 20P204 discharge pressure. This pressure moni-
========================================================================, /*h
{ ) 17 Page b of M
)K i CS8932410/4 1
{ STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) M /f ( ) OBSERVATION REPORT Observation Report No. DOR-047 Rev. O PART II -RESPONSE CONTINUED toring instrument loop provides the following:
- a. Analog Control Room indication o'.' HPCI Pump discharge pressure via PI-55-2R631 on panel 20C647.
- b. Analog input to the Unit 2 Safety Parameter Display System (SPDS) via Plant Monitoring System (PMS) point number E2428.
- c. Provides HPCI pump minimum flow protection by opening Minimum Flow Bypass to Suppression Chamber valve HV-55-2F012 at an increasing pump discharge pressure of 125 psig coincident with a pump discharge flow of less than 300 gpm.
- d. Initiates HPCI Pump Compartment cooling by starting the lead HPCI Pump Compartment Unit
[')N (, Cooler (2A or 2BV209) at an increasing pump discharge pressure of 125 psig.
- e. Energizes status light E41-DS3 HPIC High Pressure on Auxiliary Equipment Room HPCI Relay Panel 20C620 at an increasing pump discharge pressure of 125 psig. <
None of the above provide any mitigation against a postulated HELB in this room. If a HELB is l postulated in this room, the HPCI system items in this room are also assumed to be lost. The control room operators' primary sources of indication of this event are from alarm signals initiated by level cwitch LSH-55-240 detecting compartment flooding, temperature sensors TE-55-2N024, 28B/D, 29B/D and 30B/D detecting steam flooding and HPCI turbine isolation valve closure signals which have sufficient distance from each other so as not to be affected from one HELB break jet or pipe whip.
========================================================================
O o Page 3 of CS8932410/5
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA ,[h(/f /h (J.O. NO. 18138) p ( ) OBSERVATION REPORT v' Observation Report No. DOR-047 Rev. O PART II -RESPONSE CONTINUED o PT-50-2N055G is located in Room 179 and monitors the pressure on the downstream side of rupture disk PSE-50-2D001. This rupture disk is located on the exhaust side of RCIC Turbine 20S212 and'provides turbine casing protection. The rupture pressure for PSE-50-2D001 is 150 psig. This pressure monitoring instrument loop provides the following:
- a. Initiates Control Room alarm PAH-50-212 (RCIC TURB EXH DIAPHRAGM RUPTURED) at an increasing pressure of 10 psig.
- b. Digital input representing RCIC Division 3 Isolation to the SPDS via PMS point number E2455 at an increasing pressure of 10 psig.
- c. Initiates RCIC system isolation by closing O Inboard Steam Supply Line Isolation valve HV-49-F007 at an increasing pressure of 10 psig.
- d. Trips the RCIC turbine by closing Trip and Throttle valve HV-50-212 at an increasing pressure of 10 psig,
- e. Initiates Control Room alarm (RCIC OUT OF SERVICE) at an increasing pressure of 10 psig.
- r. Energizes status light E51-DS1 (RCIC DIVISION 3 ISOLATION INITIATED) on Control Room panel 20C648 at an increasing pressure of 10 psig.
l None of the above provide any mitigation against a postulated HELB in this room. If a HELB is L postulated in this room, the RCIC system is assumed to be lost. The control room operators' primary sources of indication of this event are from alarm signals initiated by level switch LSH-49-240 detecting compartment flooding, temperature sensors TE49-2N011, 21A/C, 22A/C and 23A/C detecting steam
============u ----======================================================
17 Page 3 of 447 CS8932410/6
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA g g j (J.O. No. 18138) i C\ ( OBSERVATION REPORT Observation Report No. DOR-047 Rev. O PART II -RESPONSE CONTINUED flooding and RCIC turbine steam line isolation valve closure signals. o PT-50-2N053 is located in Room 179 and monitors the suction pressure for RCIC Pump 20P203. This i pressure monitoring instrument loop provides the ! following:
- a. Initiates Control F.oom alarm PAL-50-230 (RCIC PUMP SUCT LO PRESS l at a decreasing pump suction pressure of 20" Hg vac.
- b. Provides RCIC pump protection by closing the RCIC turbine Trip and Throttle valve HV-50-212 at a decreasing pump suction pressure of 20" Hg vac.
("% None of the above provide any mitigation against a postulated HELB in this room. .If a HELB is postulated in this room, the RCIC system is assumed to be lost. The control room operators' primary sources of indication of this event are from alarm signals initiated by level switch LSH-49-240 detecting compartment flooding, temperature sensors TE-49-2N011, 21A/C, 22A/C and 23A/C detecting steam flooding and RCIC turbine steam line isolation valve closure signals. In summary, none of the above three transmitters are required to operate to ensure the operator has received information. Transmitter error will have no affect on operator response to the hazard event.
========================================================================
/O Page 3 of CS8932410/7
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA {gg 7pj l (J.O. NO. 18138)
/m D) OBSERVATION REPORT Observation Report No. DOR-047 Rev. O PART II -RESPONSE CONTINUED In response to SWEC's concern 3 expressed in the March 22, 1989 meeting, the following list of Unit 1 transmitters were reviewed for equivalent Unit 2 transmitters with the following results:
Unit 1 Unit 2 Eauivalent Transmitters PDT-51-1N058A/ B/ C/D PDT-51-2N058A/B/C/D located in Rm 475 PT-59-152B PT-59-252B located in Rm 475 PDT-41-1N088C/D PDT-41-2N088C/D located in Rm 370 PDT-41-1N089C/D PDT-41-2N089C/D located in Rm 370 r~S ( ,) FT-51-1N015D FT-51-2N015D located in Rm 370 LT-49-1NO35A/E LT-49-2NO35A/E located in Rm 27'9 l FT-11-011A/B No equivalent Unit z device FT-11-013A/B No equivalent Unit 2 device FT-11-015A/B No equivalent Unit 2 device The locations identified for the equivalent Unit 2 transmitters are all non-HELB locations, even though the equivalent Unit 1 transmitters were located in HELB areas. In response to SWEC's concern 4 expressed in the March 22, 1989 meeting, the temperature element sealing has been addressed as follows. The procedure " General Project Requirements for Environmental Qualification Walkdown of Safety Related Equipment (G-47)" (SWEC Transmittal S-0760) was issued for PECo approval May 1988. The associated equipment walkdowns, presently nearing completion, have established the need to install conduit seals in the temperature elements (TEs) identified above in the discussion of
========================================================================
fr%
? )
C/ i7 Page 10 of Hi JK CS8932410/8 _ _ _ _ _ _ _ _ l
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA I (J.O. No. 18138) yj /f)g i OBSERVATION REPORT Observation Report Fo. DOR-047 Rev. O PART II -RESPONSE CONTINUED transmitters PT-55-2N050, PT-50-2N055G AND PT-50-2N053. This sealing is in progress and scheduled to be completed before fuel load.
- 5. d. The HELB analysis did not take credit for the operability of the Rosemount transmitters. A single active failure was assumed in addition to the loss of the Rosemount transmitters disabled by the HELB event. The original response to DOR-047 was based on these conditions.
- 5. e. !!oisture will not accumulate on or within electrical cc.'ponents that are located within the power block during normal plant operation. The bases for this conclusion are as follows:
- 1. The relative humidity of the plant air will be less than 95% due to the heatup of the supply air by the supply air r~ fan motors and the internal room heat loads due to (T) lighting, motors, piping and electrical cables. These heat loads will reduce: the relative humidity of the air within the plant ever; if outdoor air conditions are assumed to have 100% relative humidity.
The reactor enclosure supply air system fan motors (v202) are in the air stream. The heat load from these motors will raise the air temperature about 59F. With outside air conditions of 959F, this would reduce the relative humidity from 100% to 86%. Heat loads from lighting and other operating equipment will further reduce the relative humidity within the plant. Because the power block is always a source of heat generation and for the reasons described above, a reduction of three to four degrees Fahrenheit will not result in 100% RH.
- 2. In order for condensation to occur on electrical components, it is necessary to have a large and rapid increase in the supply air temperature. This increase in supply air temperature would have to be accompanied by a significant increase in absolute humidity contained of the supply air. This rapid and large temperature and
========= ============================ ==========================
Page ll of bh ( CS8932410/9
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) [OON [ q OBSERVATION REPORT (u) Observation Report No. DOR-047 Rev. O PART II -RESPONSE CONTINUED humidity increase would be necessary for components within the plant to be at a temperature below the dew point of the supply air. In addition, the component temperature would have to remain below the dew point temperature and not heat up along with surrounding environment. These conditions will not occur as explained in the following paragraphs. The LGS normal HVAC system cools the plant using an all-outside air supply system. The reactor enclosure has a average air change rate of approximately six per hour which will provide indirect communication of indoor component with the changing outdoor conditions. The turnover rate for ECCS compartments below grade is approximately one per hour. Due to the relatively slow rise in outdoor air temperature and the resultant slower rise in indoor temperature due to the low turnover rates, /^T there will be time for components within the plant to become warmer as the supply air temperature changes. ( ,/ During the winter months or at night time, the outdoor air temperature will be lower than 659F for significant period of time. This means that the supply air heating coil will be in operation which will further reduce the relative humidity c ' the air inside the plant. Automatic temperature control of the heating coil will minimize temperature increases of the supply air during these time periods. Furthermore, the supply air is mixed within the room by the HVAC registers and diffusers which will also hinder rapid temperature increases within the plant.
- 3. Electrical components such as conduits have a large external surface area relative to their internal volume.
Therefore, if any condensation were to occur, most would occur on the exterior of the component. Any condensation on the interior surface of these components would be a thin film resulting in an insignificant quantity of moisture. It would also quickly evaporate due to the large surface area available to allow convective heat transfer from the room air to warm these components up to the room temperature.
===================================----------==========================
V $7 Page lA of Eb
)ll CS8932410/10 .
l
l STONE & WEBSTER ENGINE 2 RING CORPORATION LIMERICK 3 IDCA g (J.0. No. 18138) OBSERVATION REPORT observation Report No. DOR-gd,2, Rev. JL PART II -RISPCHEE CONTiMUED It is concluded that the conditions necessary for an . accumulation of noisture to develop over a period of time will not occur for the reasons discussed above. l
- 4. Most electrical equipment is energised and therefore itself a souro' of heat generation which will hinder the formation of condensation.
- 8. For Rosemount transmitters, the conduit is bottom entry and would provide a reservoir for condensation accumulation from any moisture running down the inside of the conduit. Such postulated condensate would evaporate as explained in Item 3 above. l O
1
)
JAA} e
& b gscopidsignatur /eate $6chter/gprovar scenature /Date q W fl81 Page d of 1 G l
)Id.
C88932410/11
c
/^ -t TELEPHONE CONFERENCE ITEMS \,
FOR BECHTEL AMENDED RESPONSE TO DOR-047 l
- 1. Telecon with Rosemount requires clarification or correction since transmitters should operate after a LOCA. ,
i
- 2. Available EQRR's for Unit 1 transmitters show that PT-50-LN055G !
functions include initiation of RCIC isolation. PT-50-1N053 EQRR ; initiates a RCIC turbine trip. These functions should be addressed in amended response unless the functions are unique to Unit 1. If unique, please provide statement to clarify. Unit 1 equivalent to ' PT-55-N050 is not included in EQ report.
- 3. Available EQRR's for Unit I transmitters show that level transmitters, dif ferential . pressure transmitters, and flow transmitters in. addition to the pressure transmitters listed in the amended respense are also located in HELB areas, the equivalent Unit 2 transmitters should be addressed in the response.
- 4. Response should confirm that conduit seals are utilized on the temperature elements referenced or that the error associated with their exposure is acceptable, and cannet mislead the operator.
.O V
i 1 DOR - 0 'I7 REV o PAGE / '/ OF /7 Telecopied 3/24/89 L _ . . _ . _ . _ _ _ _ _ . _ _ _ _ _ _ _ _ . _ . _ __ __ . _ _ _ . . . _
STONE & WEBSTER ENGINEERING (~RPORATION LIMERICK 2 IDCA (J.O. No. 18138) .s C) OBSERVATION REPORT Observation Report D OR- 047 Rev. O Review Plan: LK- D - 1907 - C Rev. _1_ PART III - EESPOESE EVALUATION The response to this observation report has been evaluated as follows: A. Causal Factor (s): Acceptable Not Acceptable (Explain): N/A B. Extent of Condition: Acceptable Not Acceptable (Explain): N/A C. Significance: Acceptable Not Acceptable (Explain): N/A D. Corrective Action: Acceptable Not Acceptable (Explain): N/A E. Actiot to Prevent Recurrence: Acceptable Not Acceptable (Explain): N/A
. W M M .'DE. M t................................................................
Additional Action Required: X No Yes (Explain) SEE CONTINUATION PAGES O acts gef-LFad Enginedr 8'ignature y/,/n
/ Date acces ,. w.w<
APWSignature y/,/,,
/ Date IS 17 Page g ot 8 JC }t
STONE fu WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138)
~
OBSERVATION REPORT Observation Report DOR- 047 Rev. O PART III - RESPONSE ETALUATION COFFINUATION: Review of Rosemount reports contained in LCS EQ Package El for 1153 Series B and walkdown of Unit 2 resulted in a concern about use of conduit seals. Rosemount tests apparently utilized conduit seals or equivalent during qualification testing. Based upon how the transmitters were tested, and that . under normal conditions, the transmitter would be subjected to 90% RH, an Observation Report was generated. This Observation Report addressed the above concerns. In addition, since the basis for requiring a conduit seal was not established by the vendor, this was also pursued. Project response established that Rosemount was cc.nerned with accumulation of moisture in the te rminal side of the transmitter. The response stated that during a recent EQ walkdown the Rosemount units qualified by EQ Package E-1 have flexible conduit entry into the bottom of the terrr.inal side of the housing and that "this installation configuration 'will prevent any moisture building in the terminal side of the unit." Discussions addressed the following:
- 1. The qualification of flexible and rigid conduit at LGS. This was resolved with the project based upon their statement that no credit was taken for the rigid or flexible conduit providing.
a qualified seal for the Rosemount units addressed in the Observation Report.
- 2. Conce: n that setpoint errors could occur if a possibility of thin condensation buildup .could exist during normal MELC, or direct spray conditions, and whether Rosemount would stand behind their published errors if a qualified conduit seal is not used.
This was resolved with the project in the following manner.
- a. Rosemount stated that they would stand behind their published set point error of 1/4% for the 1153 transmitters at the plant maximum normal humidity condition of 95% RH. Therefore this concern was adequately addressed.
- b. Project provided a supplemental response which indicated i
that there was little possibility of reaching 95% RH during l plant operation due to room heat loads from lighting, motors, ! piping, and cable. This response adequately addressed the concern about setpoint ertors. O
- h. O Pv dof xle Je
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA [O (J.O. NO. 18138) V OBSERVATION REPORT Observation Report DOR- 047 Rev. O PART iII - EESPONSE EVALUATION CONTINUATION:
- 3. The transmitter errors created by HELB conditions and the potential for such errors to mislead the operator. This was resolved with the project in the following manners
- a. Project clarified that only 3 transmitters were located in a potential HELB zone and they did not have to operate to mitigate the consequences of a postulated HELB.
- b. Project provided information about the functions of each of the three transmitters,
- c. Project stated that none of the three transmitters are required to operate to ensure the operator has received information, and that transmitter error will have no af fect on operator response to the hazard event.
- d. Further discussion with project resolved that TE's identif3ed p as being the operators' primary sources of indication of Q steam flooding were identified as requiring conduit seals during equipment walkdown; that the sealing is in progress, and it is scheduled to be completed before fuel load.
- 4. The accepted loss of transmitters under HELB, MELC, or spray conditions and the effect of single failure on the ability of the remaining systems to being the plant to a safe shutdown.
This was resolved based on the project statements that MELC or spray would not hinder the transmitter (s) performance, and that no credit was taken for the operability of the transmitters during HELB. They also stated that a single active failure was assumed in addition to the loss of the Rosemount transmitters i disabled by the HELB event. ) 1
- 5. Insufficient evidence that 100% RH could not occur was resolved with the project in the same manner as 2 above.
k 1 O . l I P y e. ,n o f a?
, n
/'N STONE '. WEBSTER ENGINEERING CORPORATION
( LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT Observation Report No. DOR- 048 Rev. L Review Plan LK- D - 1909 - S Rev. L Reference AI No.: DAI-476 PART I - INITIATION
- 1. Description of Concern Design pressures given in the structural design criteria (soteification C-115) do not agree with those in specification M-171.
- 2. Supporting Information Design pressures on pages 20 and 21 of specification C-115 do not agree with those in Table 1 of Specification M-171, ' Environmental Qualification Service Conditions'. Some of the values in C-115 appear to be too low, which could lead to underdesigned structures.
Response to Action Item DAI-476 indicates that the review of calculations indicates that the correct pressures are usually used. However, pressures given in Specification A-11 appear to give the incorrect pressure for steamtight doors. The response did not address the design of the doors. l l l 1 l Additional Documents Attached: None
\
...................................p..................... ............................
- f. sw Lead Engineer Signature
/E!2.3h
' / Ifa t e APM Sig6ature / Date Page 1 of I
n STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA pgg 0 3 M89 ( J. O. NO. 18138) OBSERVATION REPORT Observation Report No. DDR-48, Rev. O PART II - RESPONSE
- 1. Observation Concurrence:
X Concur with observation Do not concur with observation (Note if not in concurrence, explain in "resarks" below)
- 2. . Response to Observation:
A. Causal' Factor (s) Reactor Building accident condition steam flooding design O. pressures given in Section 4.5.2 of the Civil / Structural Design Criteria,-Specification C-115 Revision 4, correspond to pressures originally. calculated by the Mechanical Group-and- given to the Civil Group. Subsequently.the pressures were reevaluated and civil calculations were completed to evaluate the structures for.the revised pressures. Later some pressures were reevaluated, and Mechanical l Specification M-171 was issued for- use in environmental l; qualification- of systems. Since the pressure differences from the structural calculations were small, and since any pressure increases were Judged insignificant compared with
~
available structural margins, it was Judged' that the calculations did not need to be revised. Specification C-115 was not updated to reflect the final pressures. Since no additional structural evaluation was required, it was Judged that C-115 did not require revision. B. Extent of Condition There are no other conflicts between the Civil / Structural design criteria and Specification M-171. Review of reactor building structure evaluation calculations shows the calculations are performed for pressures equaling or exceeding the pressures given in Specification M-171, and also exceed the current pressures which are given in the FSAR (Table 3.6-7), O except as follows: Page JL of JL
w kJ NCGC4 /-74& STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA ( J. O. NO. 18138) OBSERVATION REPORT Observation Report No. DDR-48, Rev. O CONTINUATION: Pressure Pressure Used In Specified Compartment Calculation In M-171 RWCU valve / pump co'partments m 2. 7 2. 8 Regenerative Heat Exchanger Compartments 2. 7 2. 8 Non-regenerative heat exchanger compartments 2. 7 2. 9 [_\ v C. Significance As discussed above, under " Extent of Condition", pressures used in the st'ructura' calculations equal or exceed current pressures as given in the FSAR, and also exceed pressures
/
given in M-171 except at the identified compartments. It is apparent by inspection of the structural calculations relating to these compartments that the structural acceptance criteria will be met when considering the slightly higher current pressures, and the available structural margins. The Judgement that calculation revision for the small pressure increases was unwarranted was reasonable. As the calculations show the structure has adequate capability to withstand the current pressures, the difference between design pressures given in Specification C-115 Rev 4 and those given in Specification M-171 have no significance. Specification A-11 covers furnishing of steamtight doors. Pressures given in the Specification are pressures that were applicable at the time the Specification was prepared. These pressures are greater than the current steam pressures and (~g thus meet or exceed current requirements. i Page 3 of G L_--_-_-______-__________-_-___________-_-__-_
FEE-09 '89 14:31 ID:PECO PROJECT MGT LIJ TEL No:215-64; :F.c :;3.;; 402 __
, ca e r e ta .e et.t M i t.L we.S i mi NLe_k ; e i
- Le F ;0.;0 0 O Dc 4r7 w STONE & WEBSTER ENGIa.EERIHO CORPORATION LIMERICK 2 IDCA (J.O. NO. 181341 I
OB8ERVATION REPORT I Observation Report No. DDR-48 Rev 8 CONT 1Ry4 TION: D. Correct / e Action
]
5pecifiestien C-li5 has been revised (Rev. 5, Transmittal P-0275 to SWEC 12-14-88) to specify the current steam flooding pressures as given in the FSAR. He calculation revisions are needed since the calculations use the current ) pressures, or show that adequate steuetural margin is 1 available when using pressures only slightly less than I current pressures. I E. Action to Prevent Recurrence . No further action is required as specif1:stien C 115 has been corrected to reflect the current pressures.
........................... .. ....essass==== van ====.as=========
REMARK 5: None O e-An.o:p@se.L. C. L .n.,.,e,sk.
.. e.,.\nnppr...
t., . s .. . m iank,1,n._.a,a1 Page 1 of I /
l STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) O U)' OBSERVATION REPORT Observation Report DOR- 048 Rev. O Review Plan: LK- D - 1909 -S Rev. 1
~
PART III - 2ESPONSE EVALUATION _ The response to this observation report has been evaluated as follows: A. Causal Factor (s): X Acceptable Not Acceptable (Explain): (See next page) B. Extent of Conditions X Acceptable Not Acceptable (Explain): (See next page) C. Significance: X Acceptable Not Acceptable (Explain): 1 (See next pa'j) D. Correctf.vc Action: X Acceptable Not Acceptable (Explain): (See next page) E. Action to Prevent Recurrence: X Acceptable Not Acceptable (Explain): (See next page)
.............re............................===......===.......=.=..=.=.=..======.======
Additional Action Required: X No Yes (Explain) o g .......................................................................................
% C. fLL Lead Engineer Signature z/z//n
/'Date O(E k Rr D.tMdle AYM signature 2}2.4lt1
/ Date age b of I - - - - _ - - - _ - - - _ - _ - - - - - , - - - - - - _ - _ - - - _ - -
L. DOR-048
,m i )
v A. Causal Factor (s) - Acceptable The cause was determined to be a judgment by the responsible engineer that the structural design criteria (C-115) did not need to match the mechanical criteria (M-171) as the dif ferences were not considered significant when considering available structural calculation margins. B. Extent of Condition - Acceptable The response stated that there were no other conflicts between the Structural and Mechanical criteria. The two items that effect the design criteria are pressure and temperature. The absence of temperature requirements were addressed subject of DOR-001. C. Significance - Acceptable The response stated that the structural calculations that utilized pressures less than the current mechanical criteria requirement were reviewed and the margin was adequate to offset the small difference in pressures. As this check is straight forward, these calculations were not reviewed. The revised critieria was checked against the FSAR and found to be in agreement. (O) The effect on steam tight doors of Specification A-11 and Watertight doors of Specification A-15 was reviewed. In each case in secondary containment the door specification listed pressures equal to or above the required pressure. D. Corrective Action - Acceptable The Structural Design criteria has been revised to be in agreement with the required pressures listed in the FSAR. E. Action to Prevent Recurrence - Acceptable The immediate concern over inconsistencies between the mechanical and structural requirements has been resolved. Although Bechtel did not commit to update the design criteria in the future if the pressures in M-171 change again, but this is considered unlikely to happen at this point in the project. O DOR - o 45 REV d PAGE 6 OF 6
4 STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA ( (J.O. NO. 18138) OBSERVATION REPORT Observation Report No. DOR- 049 Rev. O Review Plan: LK- D - 1907 - C Rev. T Reference Al No.: DAI-157, Rev. 1 PART I - INITIATION
- 1. Description of Concern Tripping of transmitters LT-61-215 and 235 from the class IE bus upon LOCA is not consistent with their designated use as Post-Accident indicators.
- 2. Supporting Information
- 1. Electronic Transmitters LT-61-215 and 235 do not perform an active safety function and, per the response to the referenced Action Item, are tripped from their Class 1E power supply upon a LOCA. However, drywell sump level is shown on FSAR Table 7.5-3 as a post-accident parameter. Therefore, these instrument loops will be unavailable post-LOCA to perform their post-accident monitoring function.
- 2. Also, the subject transmitters are not tripped from the Class 1E bus in the event of a HELB in the area where the transmitters are located.
The HELB failure mode for these transmitters tterefore 3 has the potentiel to affect the Class IE power supply, which also supplies power to other Q-functional loads. Assuming that these loads are redundant,
. postulation of a single failure which affects the redundant Class 1E power supply can result in the loss of certain Q-functional loads in both divisions during the HELB. The transmitters EQRR's were deleted from the E/Q program es a result of DAI-157 Rev. li an analysis is required to assure that all equipment required to mitigate the HELB remains functional without the neeJ for tripping. these non-safety transmitters from the Class 1E bus. Bechtel must confirm that such analysis has been completed.
l
=======================================================================================
Additional Documents Attached: None O --- Dwaa sh------=---.=----=R=
)Gaf Engineer Signature
/ Date
= = = ' =-=n4k APMS/gnature
~
' / Date Page 1 of $_
STONE &' WEBSTER ENGINEERING' CORPORATION LIMERICK 2 IDCA (J.O..NO. 18138) 1pM / fg WMR 2 4 $89 OBSERVATION REPORT - p 7f.Q Observation Report No. DOR-049 Rev. O PART II -RESPONSE
- 1. Observation Concurrence:
AMENDED RESPONSE: The following is an amended response to clarify the statement (3.3.3) in EG&Gs Report whi'h c was included in SSER 4 per the meeting between SWEC and Bechtel held in San Francisco on March 8, 1989. The following is-an amended response to provide additional clarification regarding conformance to R.G 1.97. Concur with observation x Do not concur with observation (Note: if not in concurrence, explain in " remarks" below)
========================================================================
(j Ramarks:
- 1. FSAR Section 7.5.2.5.1.1.2.4 provides the issues used to substantiate deviation of the Limerick system from the R.G.
1.97 criteria. Section 7.5.2.5.1.1.2.4.3 provides the-discussion for variables B8 and C6, Drywell Sump Level and Drywell Drain Sumps Level, respectively. The reason for this FSAR section is to document the classification of these two R.G. 1.97 variables as Category 3 devices. The basis for reclassifying the drywell sump level (Floor Drain Sump and Equipment Drain Tank) is that each sump drain is isolated at the primary containment penetration on receipt of the initial containment isolation signal which establishes containment integrity. This initial containment isolation signal is initiated on high drywell pressure and/or Jow reactor water level (Level 2). Subsequent to the initial containment isolation, the sump level indication continues to operate until the reactor level and pressure have decreased and the ECCS initiation signal occurs. The ECCS signal automatically starts the Core Spray and RHR systems to cool and replace the inventory of water in the reactor. This signal is initiated on high drywell
=========================================----===========================
0 Page jl_ of T CS8932315
l STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA IOCN / h g (J.O. NO. 18138)
- OBSERVATION REPORT Observation Report No. DOR-042 Rev. O PART II -RESPONSE pressure coincident with low reactor pressure and/or low reactor water level (Level 1) and indicates that a gross release of reactor coolant has occurred (large break LOCA).
The signal would not be generated for a small break LOCA. This scenario also assumes that the reactor coolant has condensed, collected on the drywell floor and drained to the suppression pool via the downcomers. Both sumps will be flooded via floor drains to the Floor Drain Sump and via the overflow line to the Equipment Drain Tank. Because the capacity of each sump is small in comparison to the volume of water released and is assumed flooded at the start of the ! ECCS, the drywell sump level indication is irrelevant in any post-ECCS condition. In addition, upon receipt of the ECCS signal, the diesels are started and all non-essential loads. are shed from the class-1E busses. Because the drywell level indication is considered a non-essential load and serves no useful purpose for the post-ECCS event, it is shed from the class-1E bus. FSAR Section 5.2.5.2.1.3 states that this (g'~) system is energized by Class-1E power and that it will automatically be isolated in the event of a LOCA. As defined in R.G. 1.97, there are other accidents in addition to the post-ECCS event. The Drywell Sump Level Monitoring System (DSLMS) will remain operable during these accidents and provide the control room operator with meaningful level indications (until the sump tanks are full if the accident has resulted in containment isolation). This information will assist the operator to identify the leakage source. Therefore, the drywell sump level instrumentation, as designed, serves as a category 3 variable but does not provide a useful monitoring function after the post-ECCS event or after the sumps are full if the containment is isolated. This is consistent with the BWROG position and has been accepted by the NRC as described below.
//~\========================================================================
\_J .
Page 3 of T CS8932315
STONE & WESSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) 2O[A/bQ
- OBSERVATION REPORT Observation Report No. DOR-012 Rev. 0 PART II -RESPONSE A detailed review and technical evaluation of the Limerick licensing submittal for conformity to R.G. 1.97 was performed by EG&G under contract to the NRC, with general supervision by the NRC staff. The EG&G report contained in Supplement No. 4 to the Safety Evaluation Report concluded that "...the Sump Tank Level Monitoring System that is specified in R.G. 1.45 a method of determine leakage from the Reactor Coolant System.
Once the drywell sump systems (i.e. drains) are isolated for accident conditions, no useful post-accident information would be available and the operator is able to tell that the tank is full by use of existing instrumentation." This refers to the existing tank level indication available in the control room on the ERFDS CRT (points E2192 & E2193) and other parameters which would indicate leakage into the drywell(i.e. -Drywell pressure, Drywell temperature, Primary Containment Radiation). These indications will allow the operator to monitor tank (~mg level until full and/or to infer that the tanks are full. (j Therefore, the EG&Gs evaluation is consistent with the DSLMS design and meets R.G. 1.97 as a Category 3 indication for the Drywell Sump Level Indication. This approach was similarly docketed and approved on many other BWRs including Peach Bottom Atomic Power Station (PBAPS). The following conclusion on this subject is stated in the NRC Staff's SER (Transmittal letter J.F.Stolz (NRC) to E.G. Bauer (PECo), dated June 3, 1985) on PBAPS compliance with R.G. 1.97: "We conclude that the alternate instrumentation supplied by the licensee will provide appropriate monitoring for the parameters of concern based on the following: a) for small leaks, the alternate instrumentation is not expected to experience harsh environments during operation; b) for larger leaks, the sumps fill promptly and the sump drain lines isolate due to the increase in drywell pressure, thus negating the drywell sump flow and drywell drain aumps flow instrumentation; and .c) this instrumentation neither automatically initiates nor alerts the operator to initiate operation of a safety-related system in a post-accident situation. We, therefore, find the alt'Jnate Category 3 instrumentation provided acceptable." [Q v j --- _ ================================================================ Page 4 of T CS892315
MAR-24 '89 10:32 ID:PECO PROJECT MGT DIV TEL NO:215-641-4578 #968 P02 c ZotA / %Q STONE & WEBSTER IMGINEERING CORPORATION LIMERICK 2 IDCA (J.0. No. 18136) OBSERVATION REPORT Observation Report No. DOR-D.11 Rev. D PART IT -RESPO M
- 2. The subject tJnit 2 LTs are not located in a HELB zone as indicated in the, DOR. They are located at a floor elevation of 217 feet in Room 370 which is outside of the HELB sone.
The transmitt9rs are isolated from their Class-1E power vources by means of approved isolatien aavices (fusa and circuit' breaker in series per FSAR Section 8.1.6.1.14(a).4). The transmitter cables, cable routing and terminations have been analyzed with regard to safety system impact. There is no adverse safety system impact. O . O ,. ... ... __.................._....... AAdin ........_sAr/n
............... N. (MALuiwr wwn Pedo Rev.iGv Signature /Date [Bechtel appr6 val signature' / bate Page I ofd
l' O STONE 6 WELSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT . Observation Report DDR- 049 Rev. 0 l Review Plan LK- D - 1907 -C Rev. ~1 PART III - EESPONSE ETALUATION l The response to this observation report has been evaluated as follows: A. Causal Factor (s): Acceptable Not Acceptable (Explain): l l N/A B. Extent of Condition Acceptable Not Acceptable (Explain): N/A C. Significance: Acceptable Not Acceptable (Explain): N/A D. Corrective Action: Acceptable Not Acceptable (Explain): N/A E. Action to Prevent Recurrence Acceptable Not Acceptable (Explain): N/A Additional Action Required: X No Yes (Explain) See continuation pages. F O WLt LeadlEngWneer Signature uMn
/ Date
&& onne AMi Signature sluin
/ Date Page 1 of [
l l r IL STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) V OBSERVATION REPORT Observation Report DOR 049 Rev.O PART III - RESPONSE ETALUATION CONTINUATION: , . Review of Unit 1 Equipment Qualification Review Records (EQRRs) and l responses to questions resulted in a concern about drywell equipment and
- drain tank level indication. These questions arose since the instrument-l loop functioned to meet -both R.G. 1.97, Post-Accident Monitoring, and R.G. 1.45, Leakage Detection, Requirements. Two transmitters, one for each tank, performs both (i.e. , R.C. 1.97 and R.C. 1.45) functions. These transmitters are powered from a Class 1E power source from which they are separated on receipt of a LOCA signal.
Two concerns were addressed: One concern addressed was that the two instrument loops would be unavailable post-LOCA to perform their post-accident monitoring function. The response stated that the drywell sump systems are isolated at the primary containment penetration on receipt of an accident signal- to establish containment integrity and that this fact renders the drywell sump level indicotion irrelevant in a post-LOCA condition. Since R.G. 1.97 and R.G. 1.45 do not address if these instrument loops should be available during and after a LOCA, additional clarification was pursued. The NRC approval for reclassification of the post-accident functions of the drywell drain tank and sump to Category 3 appears to be based upon acceptance of EG&G Idaho's Technical Evaluation Report about LGS R.C. 1.97 Compliance. EG&G's report finds that the tank level monitoring system that is specified in R.G. 1.45 a method of determining leakage from the reactor coolant system. It continues to state that "Once the drywell sump systems are isolated for accident conditions, no useful post-accident-information would be available and the operator is able to tell that the tank is full by using the existing instrumentation." Since the subject instrument loops perform both R.G. 1.97 and R.G. 1.45 functions, the EG&G statement is unclear. This remains unclarified. Through review of additional drawings, it was concluded that the signal used to trip the instrument loop from the Class 1E bus was a LOCA signal developed from ECCS initiation. This satisfied a concern that the separation of the instrument loop was from the signal used to close the drain tank and sump isolation valves. Since the ECCS LOCA signal is used to trip the instrument loop from the 1E bus, it is assured that the instruments would be available for monitoring [
\
1eakage except when a LOCA signal is present. This supports the project position that "since the capacity of each sump is small in comparison to the volume of water released, and is assumed flooded at the start of CONTINUED ON ONE PAGE. Page. 7 of !
~ ,
1
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT Observation Report DOR- 049 Rev. O PART III - RESPONSE EVALBATION CONTINUATION ECCS, the drywell sump level indication is irrelevant in any post-ECCS condition." A second concern, based upon the review of Unit I documentation, addressed the effect of a HELB and the failure mode for the drywell drain tank and sump transmitters. Additionally. 'it was requested that assurance be provided that a single failure for the HELB condition would not affect the Class IE power supply. The project response indicated that the Unit 2 transmitters were not located in a HELB zone and that they were' isolated from the IE power ' supply by means of an approved isolation device. This second concern was resolved satisfactorily. Page T of T L_______________________________________ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ __ _ _ _ _ _ ___ _ _ _ _ _
STONE & WEBSTER ENGINEERING CORPORA"' ION
'O LIMERICK 2 IDCA k
(J.O. NO. 18138) OBSERVATION REPORT Observation Report No. D OR- 050 Rev. 0_ Review Plani LK- D - 1903B - NC Rev. L Reference AI No.: DAI-130 Rev. O PART 1 - INITIATION
- 1. Description of Concern Calculations were not performed for the qualification of supports on the piping involving the extension of seismic requirements into the non-seismic region (marriage isometric) as bounded by stress calculation 1-21-99. .
- 2. Supporting Information The response to DAI-130 states no calculations are required for the supports on the " marriage" isometric cf stress calculation 1-21-99 based on the conclusion that the reaction loads are negligible.
This position does not comply with the requirements of FSAR section 3.2.1-(d) which states that " design and design control for such items are carried out in the same manner as that for items directly important to safety." The conclusion that the reaccion loads are negligible does not preclude the necessity for a design calculation. In this regard, the design approach is deficient. l I Additional Documents Attached: None
\
9t t ) R Lead Engineer Signature
- +sla.
/ Date APM Sigfiature M mLMsr / Date Page 1 of 3
E & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA V h"
- (J.O. NO.'18138)
.IU 1 <
OBSERVATION REPORT J AN 2 41999 Observation Report No. DOR-050 Rev. O PART II - RESPONSE
- 1. Observation Concurrence:
X Concur with observation , _.Ib not concur with observation (note: if not in concurrence, explain in " remarks" below)
- 2. Response to Observation A. Causal. Factor (s)
Design calculations were not performed for pipe support structures included in calculation 1-21 -99, because the reaction loads are judged to be negligible with respect to the . capacity of the members used to support the pipe. This judgement was exercised and documented in the Civil Department's response to the Plant Design coordination memo that supplied the pipe reaction loads to the Civil Department. p . B. Extent of Cbndition Bechtel Engineering has reviewed all the Seismic Category I valve pit piping in the yard area for both Limerick Unit I and Unit 2 pipe supports designed by Civil for piping systems analyzed by Plant Design. No other instances of Civil steel used as a primary supporting structure for the pipe were identi-fied. Other areas of the plant (eg: reactor building, spray pond pump house, diesel generator buildings) do not require a review, because both the piping and pipe support calculations are generally the responsibility of Plant Design. C. Significance (Cor4Tir4UED) Remarks: 1 O Pagelof 4 , __ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ .]
JAN-24 '8913:59 ID:PECO PROJECT MGT DIV TEL NO:215-841-4578 #154 P02
. g STONE 6 WES! TIER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.C. No. 18138)
OBSERVATION REPORT d%+ M7A observation Report No. DQR=050 Rev. O s FMll - AERPQHSt (continued) Se lack of a design calculation for the pipe support structures identified in this DQR is not significant. When review 1 thc capacity of the mercers
. A shown on drowing c-43 Detail.1 against the des gn loads supplied by Plar,._
Design, the civil Engineering Group Supervisor exercised reasonable
---engineering-judgement that the steel members provided-were spre than"-" - - -
adequate, by inspection, for the loads. Purthemore, this judgement was documented in a memorandum that was processed via existino communication control procedures such that traceability and retrievabillty were provided.
-Drcorrective-Action - --
- :.... Calculation 2PSSCs104 Rev. O dated 1/19/89, available. in Pottstown, has .
been prepared to document the acceptability of the structures identified on drawing C-43 Detail 1 for the loads provided by Plant Design in calculation 1-21-99. Se piping loads on the structure represent only 206 Q(s ef the structure's design capacity. Se calculation demonstrates the coundness of the engineering judgement exercised in the design of the supports. . E. Actions to Prevent Recurrence No further action is required based on the oorqplete review conducted and the oorrective action performed. l l l l
.I mens amassessens========ewuss sesssmessmamm e nsemaname=====ese ssammesssssas em as sam a m.. .
. a h j LM M P r.4 Daview 51gnature /Date Bechtel Approval Signature / / /Datte l
Page 3 of 'l L _
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA 1 (J.O. NO. 18138)
, OBSERVATION REPORT Observation Report D OR- 050 Rev. L i
Review Plan: LK- D- 1903B - MC Rev. L 1 PART III - RESPOBSE EVAIRATION The response to this observation report has been evaluated as follows:
. A. Causal Factor (s): X Acceptable Not Acceptable (Explain):
The cause is acknowledged to be the use of engineering judgment (as documented in the civil response to the plant design memo that supplied the support loads) instead of preparation of calculations. B. Extent of Condition X Acceptable Not Acceptable (Explain): Based on Bechtel's review (limiting the sample to valve p!t piping in the yard area since no other instances of civil steel used as a primary supporting structure for the pipe existed), the cited case was the only instance where civil steel was the primary support. C. Significance: X Acceptable Not Acceptable (Explain): The engineering judgment exercised by the civil group was valid, which now O has been documented in the Calculation 2PSSC-104 (not reviewed by SWEC). The concern was limited to this particular instance only. D. Corrective Action: X Acceptable Not Icceptable (Explain): The required calculation has been prepared to document the acceptability of the supports on the marriage isometric. E. Action to Prevent Recurrence X Acceptable Not Acceptable (Explain): Since the concern is not generic, and all existing valve pit supports have been reviewed, it is reasonable to conclude that the problem will not recur.
........m...m........................-- ...................... .....--=================
Additional Action Required: X No Yes (Explain) 1 s==============.ss==================================== ===============================
^1'" 1 < "" -
.O '
- 1 - ' " - -
Page_6/_ of I I L_______-_______..___________ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _
N f a STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA , (J.O. NO. 18138) D OBSERVATION REPORT \ v) t
' Observation Report No. D OR-' 051 Rev. _q Review Plan LK- D - 1905 -
C Rev. 1 Reference AI No.: DAI-236
' PART I - INITIATION Description of Concern 1.
A discrepancy exists between the. FSAR compliance regarding Regulat;ory Guide 1.47 and schematic diagrams 8031-E-360, Sh 3, Rev 14 and 8031-E-622, Sh 3. Rev 18, 4
- 2. Supporting Information The FSAR commitment for Regulatory Guide 1.47 (FSAR Qaestion & Response 421.33) states that the condition of pump breaker not %nnected, pump breaker locked out, and transfer switch out of position initiates a rystem level out-of-service annunciation.
The Unit 2 design for the RHR pumps is presently different from the Unit 1 design and not in accordance with the FSAR. Also, the Unit 2 design does not-meet the intent of Regulatory Guide 1.47 in that bypass indicacion
', is provided at the component level only. The dif ference between the Unit V 1 and Unit 2 design which is a result of an intended plant enhancement, also represents a human factors discrepancy.
..m.......m===..m.m......m....m..m...m.m.......=mm..........m======....m...============
Additional Documents Attached: None mm......mmmm....mm.m..mmm..mm.msmany.aus.mmm.mmmmm.m..m. = mm.m . m....mus..m.am..mma YlJ s h ~
, e <e O.
te e E>*1neer e1Snat re eate .ee S1 4:a ,.= r c - Page 1 of k
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA 7/) M /7f (J.O. NO. 18138) OBSERTATICE IRPORT FEB 161989 @ "/#8) - V Observation Report No. D OR-051 Rev. O PART II - RESPONSE
- 1. - Observation Concurrence X Concur with observation Do not concur with observation (Notes: if not in concurrence, explain in " remarks" below) 4
- 2. Re sponse to Observation:
A. Causal Factor (s) The discrepancy between the implemented Unit 2 design change and the FSAR commitment (FSAR Ouestion & Respons e 4 21.3 3 ) resulted from a failure to recognize that detailed information for the des'.gn that was being modified was included in the FSAR. Because an existing equivalent Unit 1 temporary modification was intended to be made permanent, the originator and checker in the electrical discipline incorrectly assumed that Unit I was in compliance with the FSAR and its review f or the Unit 2 modification was not needed. Co nsequently , (~ s the design change document (DCP 2006) that was issued for (_ ) implementing the modification in Unit 2, incorrectly indicated that an FSAR change was not required. B. Ex tent of Condition DCP 2006 is one of about 57 design changes that were issued f o r Uni t 2 under a review program of Uni t 1 temporary circuit alterations (TCAs) f or their applicability to Unit
- 2. This program was a one time ef fort intended for the review of TCAs that were to become permanent modifications in Unit 1 and for the implementation of equivalent modifica-tions into Unit 2, as applicable, to make both units the same.
Because the possibility for occurrences similar to DCP 2006 can not be ruled out, all 3 2 of the othe r Unit 2 modification initiated- by the electrical discipline based on TCAs were reviewed for compliance with FSAR commitments. It was felt that only modifications originated by the electrical discipline needed a review since the error occurred in a modification (DCP 2006) i ssued by this discipline. The review revealed that all 3 2 modifications were either in conpliance with FSAR commitments or properly identified applicable FSAR changes. Therefore, based on the results of this review it T- 14/9
'8' b ' '
I l L STONE & WEBSTER ENGINEERING CORPORATION l LIMERICK 2 IDCA ss.v. NO. 18138) [pdjy i OBSERVATION REPORT Observation Report D OR. 051 Rev.0_ CONTINUATION: 1 is concluded that the identified discrepancy can be considered an isolated occurrence and no further investigation of the extent of - this condition is warranted. Moreover, provisions l exist in project procedures for identifying and evaluating similar conditions (Unit 1/ Unit 2 d if ferences, FSAR compliance, and human f actor concerns). C. Significance The significance of the identified condition is that, had it gone undetected, it would have resulted in a noncompliance with the FSAR. However, the identified differences between Units 1 and 2 for the "RHR OUT OF SE RVIC E" and the "RHR PUMP MOTOR OVERCURRENT" alarms are not considered to have a safety significance. Although the Uni t 2 design does not f ully conform with the PSAR commitment and may present a human f actor discrepancy, this design does provide the operator the means for recognizing the effects of an inoperable condition of the RHR system. The Uni t 2 design also satisfies
,Q BTP ICSB 21 in providing individual indicator lights on a
(_ ,/ control room panel to help identify what function of the system is out of service, by pa s sed , or otherwise inoperable. D. Corrective Action Provisions exist in the ongoing programs for identifying and evaluating conditions such as the one addressed in this DOR. Th e Un it 1/Un it 2 Di f f erences program (Transmittal No. S0619 ) along with other programs, which are governec by project or engineering procedures, have provisions such that differences between the two units are reviewed for their safety implications, ef fect on the PSAR and Technical Specifications, and for their impact on operator transparency (i ncluding human f actors) . The procedures governing those programs are EDPI 4.25.3, " Tracking Control Form" f or Units 1 a nd 2 (Transmittal No. S0619); EDPI 4. 7 3.1, " Modification Design Chang e Package s" f or Unit s 1 and 2" (Transmittal No. S0619); a nd E DPI 4.7 3.2, " Project Design Change Package Procedure f or Uni t 2" (Transmittal No. S0619). In addition, discrepancies between the two units are also identified by PECo Operations , Field Engineering, and during the Preoperational Te st Pr ogram, by Startup Engineering. As a result, the discrepancy addressed in this OOR was fs identified (i n a startup field report) a nd P C R- 20768 ( ,) (Transmitta l No. S0619 ) was issued to correct this condition. Page 3 of b T- 14/9
STONE & Wtl8TER INGINttRIM CORPORATION LIMERICR 2 IDCA (J.0. 50. 18138) IMCN / 78[ 058EttAT108 rep 002 [V Observation Esport No. L OR- E Rev.O_ COP 33NUATION: i The PCR eliminates the identified dif ference between the two I units for the subject RHR system alarm. his modification l will ensure compliance with the FSAR commitment and eliminate ! ' the human f actor concern. E. Action to Prevent Recurrence Existing procedures are considered adequate for identifying, . reviewing and evaluating discrepancies between the two units. Furthermore, EDPI 4.21.2, " Discipline Licensing Coordinators (DCL)" for Unita 1 and 2 (Transmittal No. 80619), was revised and personnel trained accordingly to help i eliminate errors that result in discrepancies between design and FSAR commitments. Revision of this EDPI and , training of personnel was the result of recommendations , f rom a study that was conducted (independent f rom the issue in this DOR) to determine underlying root causes leading to e rrors in the FEAR. O l I e $~N $f % hNuL] *b N iechtet APPr6 vat lignatura YINW
/ Date '
l FtcoRevisfignature[ / Date T- 14/9 Fage 1 of fg,
i STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) A OBSERhTION REPOPT I Observation Report D OR- 051 Rev. A Review Plan: LK- D - 1905 -C Rev. L PART III - RESPONSE EVALUATION The response to this observation report has been evaluated as follows: A. Causal Factor (s): X Acceptable Not Acceptable (Explain): SEE ATTACHED PAGE B. : Extent of Condition: y Acceptable Not Acceptable (Explain): ! SEE ATTACHED PAGE C. Significance: 1 Acceptable Not Acceptable (Explain): f SEE ATTACHED PAGE i ( D. Corrective Action y Acceptable Not Acceptable (Explain): SEE ATTACHED PAGE i E. Action to Prevent Recurrence X Acceptable Not Acceptable (Explain): SEE ATTACHED PAGE
........................................................==...=====...==.==..=========== l Additional Action Required: X No Yes (Explain)
O C LedJ Engineer Signature CL,W zheln / Date APA Signature CC&st fw DMlle / Date 1/z4ln Page [ of jg_ n _ . _ _ . _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ -
DOR-051, Rev. O A. The response stated that the non-compliance with the FSAR commitment for Unit 2 resulted from the assumption that an existing equivalent Unit-1 temporary modification, which was intended to be made permanent, was- in compliance with the FSAR and the review for the Unic 2 modification was not needed. Consequently, the Design Change Document _(DCD 2006) issued for implementing the modification in Unit 2 incorrectly indicated that an FSAR change was not required. B. 32 of the approximately 57 design changes for Unit 2 as a result of a review program for Unit I temporary circuit alterations which were to be made permanent were prepared by the electrical discipline. The response stated that all 32 modifications were reviewed since the error occurred in a modification issued by this discipline. The review determined that all 32 modifications were either in compliance with commitments addressed in the FSAR or properly identified as requiring FSAR changes. This discrepancy appears to be an isolated occurrence. Furthermore, project procedures for identifying and evaluating Unit 1/ Unit 2 differences, FSAR compliance and human faccors concerns are in place. C. The significance of the discrepancy identified is that, if gone undetected, it would have resulted in a non compliance with the FSAR;- however, the difference between Unit 1 and 2 for the "RHR Out Of Service" and "RHR Pump Motor Overcurrent" Alarms do not represent a safety significance.
'D . The discrepancy addressed in this DOR identified (in a startup field report) and Project Change Request (PCR) Number 20768 was issued to correct this condition. The PCR eliminates the difference between the two units, removes the human factors concern and ensures full compliance with the commitment stated in the FSAR.
E. Review of current Engineering Project Instructions by SWEC (EDPI 4.21.2 and 4.25.3) indicates that existing procedures are adequate and if followed should assist personnel in eliminating errors that result in discrepancies between design and licensing commitments. Additionally, the training of personnel regarding these instructions should reinforce the prevention of recurrence of this type of condition. O 'ooa - PAGE lo s' OF aev o fo
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA
. (J.O. NO. 18138)
OBSERVATION REPORT O) t
'V Observation Report No. D OR '052 Rev. O Review Plan LK 1- 1903B - g Rev. J Reference AI No.: DAT-263 R.v. O PART I - INITIATION
- 1. Description of Concern
- 1. Pipe support calculation DLA-212-C12, Rev. 1 is deficients it does not consider all applicable loads.
- 2. Engineering approval of FCR CS-2208-C does not list the affected calculation (DLA-212-C12, Rev. 1) which must be revised to include tank attachment loads.
l l
- 2. Supporting Information The response to DAI-263 indicates that Utiit 2 hanger calculation DLA-212-C12, Revision 0 referred to the Unit I hanger calculation DCA-112-2C, Revision 1, which had incorporated the tank attachment loads.
Ho+: eve r , Revision 1 of the Unit 2 calculation DLA-212-C12 omitted the tank attachment loads. Therefore, the calculation is deficient in not
\ addressing all applicable loads.
The response also indicates that the oversight (not including the tank loads) was recognized by engineering (Ref. FCR CS-2208-C). However, the engineering disposition of the FCR does not list the affected calculation DLA-212-C12 Revision 1 as requiring revision in accordance with Bechtel procedure EDP 4.62. Therefore, there is no assurance that the calculation 1 would have been revised to reflect the tank attachment loads based on this FCR. l Additional Documents Attached: None l l
...................................p............................. . ............. ....
f e Y * ' Lead gineer Signature / D4te APM Sigo6ture / Date Page1of[g_, I
[ d /7/O STONE & WEBSTER ENGINEERING CORPO TION a j LIMERICK 2 IDCA (J.O. 10. 18138 ) k
'( ) OBSERVATION REPORT DATE __
mag 0 81989 t _ Observation Re . . DDR- 52, Rev. O PART II - RESPONSE CDMPLETE AMENDED RESPONSE l
- 1. Observation Concurrence:
l X Concur with observation - Calculation DLA-212-C12 Rev.1 does not ' address the accumulator tank loads. X__ Ib not concur with observation - Engineering approval of FCR's does not require that affected calculation numbers be listed on the FCR (See Remarks). l
- 2. Reponse to Observation A. Causal Factor (s):
Pipe support calculation DLA-212-Cl2 Rev.1, dated 12/22/87, was based on field drawirg DLA-212-H12 Field Rev. OF4, dated 11/16/87, which did not clearly identify the accumulator attachment points. Therefore, the originator and enecker of the Rev. 1* calculation did not include the accumulator loads. (m L)
' B. Extent of Condition:
A field walkdown was performed to verify that accumulator 2A003E is the only accumulator attached to a pipe support - the remaining 13 accumulators all are attached to civil structural steel. In order to assess other Unit 2 designs for multi-commodity supports originally included in the base design, 7 stress calculations wherein the Unit 2 pipe support design drawings were based ca the identical Unit 1 designs were reviewed for the followirg attributes: The revision "O" of the Unit 1 design drawings was issued based on a Unit 1 design. The Unit 2 revision "O" drawing indicated an attachment (of any type) utilizing the pipe support as the " parent" structure. The same attachment is now shown on the current revision of the Unit 2 design drawing. The scope of the review included 83 pipe supports and identified six that possessed the above attributes. All six calculations currently document the consideration of the attached commodities. No discrepancy similar to the concern stated in this Observation Report exists. (See Table 1 for survey scope). O Based on the survey results, the concern is considered to be an isolated (j case. Other comrrodities may attach to pipe support structures, but are programmatically controlled as discussed in the response to IAI-263. Page d. ofd
P.5/B
/lAR90'8912814 BECHTEL WESTRN POWER iST FLOOR STONE & WE88TER ENGINEERING CORPORATION
! [h O LIMERICK 2 IDCA (J.O. H0. 18138) D h OBSERVATION REPORT Observation Report No. DDR-052, Rev. O PART II - RESPONSE ColfrINUED C. Significance: '
As discussed in the DAI-263 response, this pipe support is based on the Unit i design such that the accumulator attachment loads were included in Revision 0 of Unit 2 calculation DM-212-C12. Inclusion l of the attachment loads into the Rev. 2 calculation did not affect the design and all results remain within acceptable limita. Attached small pipe hangers 87-DCA-418-E3-H4, -E7-H2, and H10 were considered in the Revision 1 calculation thereby denenstrating that other attach-sent loads were included in the design. We do not consider the concern identified in this Observation nsport as significant. D. Corrective Action: Calculation DM.212-C12 Revision 2 has been issued as a result of incor-poration of FCR Cs-2200-C. Drawing DM-212-M12 Revision 2 has been issued to specify the accumulator attachment to the pipe support structure. Both the drawing and calculation are available for review in the Pottstown office. E. Action to Prevent Aseurrences since this concern is not significant and an isolated case, no further corrective action is required. Remarks: EDP 4.62 is modified for project use by EDPI 4.62.1 Rev.19, 8 specifically, Paragraph 3.4.3 of EDP 4.62 (which discusses the identification of calculation numbers on FCR's) is revised by EDPI 4.62.1 Rev.19 as follows:
" 2.1 Paracraoh 3.4.3, PCR Processina, revised to read,#
Reviews possible impact of change on associated calculations and by exercising appropriate judgement, determines whether it is necesary to revise the calculations. " As noted on FCR C8-2208-C Block 10.2, the civil Structrual (CS) FCR was coordinated with Ray Scott, a Plant Design discipline pipe support engineer. During the review process, it was detsrmined that the tank was attaching to the support and the tank loads had been accounted for in the original revision 0 calculation in accordance j with existing poject procedures. Revision 2 of the calculation was issued to ' incorprete these loads as a result of the FCR incorporation into the design drawing. l
....... ...e ....................................a.....si......... ..................
W Y g
}; _
O " PECO ReVim, Signature/Date 2 e
' sechtel Appf0 Val 51gnaggre ' /Date Pegslof,[g,
I SIONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA.
$^'
/ - (J.O. NO. 18138)
?%
OBSERVATION REPORT Observation Report No. DOR-052, Rev.'0 TABLE 1 SURVEY SCOPE Stress NumLc of supports Support Drawings That Calculation Have Attachments Based (System) on Unit 1- Designs 2-01-03 '49 DBA-206-H14 Rev. 1 , (Main Steam) GBC-201-H15 Rev.'l l i 2-20-01 6 None !
~(Core Spray) 2-10-10 10 DCA-204-H11, Rev. 3-(RHR) 2-10-05 7 DCA-418-H1 Rev. 2 O- (LECI) 2-10-06 7 DCA-418-H8, Iwv. 2 (LPCI)- -
DLA-212-H12, Rev. 3 2-10-07 7 None (LPCI) 2-10-08 7 None (LPCI) i TUTAL 83 6 J I 1 1
-1 l
l LO Pap 1 d .1 l
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138)
. OBSERVATION REPORT Observation Report D OR- 052 Rev. O Review Plan: LK- D - 1903 B - Mc Rev. 1 PART III - RESPONSE EVALUATION The response to this observation report has been evaluated as follows: (ITEM 1)
) A. Causal Factor (s): X Acceptable Not Acceptable (Explain): The response acknowledges that the originator and checker did not include the accumulator tank loads because the attachment was not clearly identified on the drawing. B. Extent of Condition: 3 Acceptable Not Acceptable (Explain): Bechtel reviewed all other accumulator tank supports and determined that they were attached directly to structural steel. In addition, Bechtel reviewed a sample of Unit 2 multi-commodity supports where the Unit 2 drawing was based on the identical Unit 1 design. The reviews showed that this concern was an isolated case. C. Significance: X Acceptable Not Acceptable (Explain): 1 l Inclusion of the loads had no impact on the qualification of the
/']
V supports, since the Unit 1 design considered the loads, and the Unit 2 design (Rev. 0) was identical to the Unit 1 design. D. Corrective Action: 1 Acceptable Not Acceptable (Explain): Bechtel has revised Calculation DLA-212-C12 and Drawi'.g DLA-212-H12 to account for the tank loads. This calculation and drawing have not been reviewed by SWEC. E. Action to Prevent Recurrence: 1 Acceptable Not Acceptable (Explain): Since the review indicates that this concern was an isolated case and it was determined to be not (Ignificant, no further action is required. Additional Action Required: X No Yes (Explain) SEE ATTACHED SHEET O ....................................................................................... W ~Cd Lead Engineer Signature 3l's/M.
/ Date
%& Ar o. mile. sla/r1 A191 Signature / Date Page 8 of 6 l
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) ( OBSERVATION HPORT 052 Rev. O Observation Report DOR- l PART III - RESPONSE ETALUATION CONTINUATION: l l I ITEM (2) - In accordance with Bechtel procedure EDPI-4.62.1, the responsible engineer determined that it was not necessary i i to revise the calculation. Since there is no procedural requirement to revise the calculation, no further action is required. N' l O e.,,1 s i
I STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA m (J.O. No. 18138) ( 'I C' OBSERVATION REPORT Observation Report No. D OR- 053 Rev. O Review Plant LK- D - 1904 - c Rev. 1 l Reference AI No.: DAI-365 Rev. O a PART I - INITIATION
- 1. Description of Concern Guidelines used to locate Q-functional tubing supports have been applied in a fashion not consistent with the testing used as a basis for establishing maximum support intervals.
- 2. Supporting Information Several cases were identified where installed tubing support intervals along a particular axis were greater than the intervals specified on drawing 8031-M-830-G038. The Action Item response acknowledges that these arrangements are not addressed by Anco testing results (8031-M-255) and associated calculation A-198, which are given as a basis for the maximum support intervals specified. This inconsistency has apparently occurred because support spans were applied in a linear fashion without regard to changes in direction, rather than independently along each axis.
. 4 s ) Since 8031-M-255 and A-198 cannot be used as a basis to justify configurations such as the examples cited, a deficiency exists in that, for certain installed tubing configurations, no basis exists for the as-built design. Bechtel, therefore, modeled a theoretically possible configuration given in the original Action Item, utilizing worst-case conditions. The response concluded that all Limerick tubing support configurations could be demonstrated to be conservative and acceptable based on that model. However, the following concerns apply:
a) Bechtel should confirm that the case modeled represents a true worst case for all installed Q-functional tubing at Limerick, since that model is now being offered as the basis for acceptability of all Q-functional tubing support spans. b) The model should be formally documented and made a part of the design basis for Limerick tubing installations, in lieu of 8031-M-255 and calculation A-198.
...r...................... ............................................................
Additional Documents Attached: None A N{u Sr dcB 4?/f/s/ APM Signture I ** 4 W
~ / Date Le/d' EnJfineer Signature / Date Page 1 of _4.
'll L <
1 i L SIONE & WEBSTER ENGINEERING CORPORATION py LIMERICK 2 IDCA ff
%)Vf i
- A l- (J.O. NO. 18138) l l
U- OBSERVATION REPORT J AN 31 1989 Observation Report No. DOR-053, Rev. 0 l OBSERVATION REPORT l
- 1. Observation Concurrence:
X Concur with observation Do not concur with observation (Note: if not in concurrence, explain in " remarks" below)
- 2. Reponse to Observation:
A. Causal Factor (s): A lack of specificity on drawing 8031-M-830-G038 allowed support spans to be installed in a linear fashion, rather than independently along each axis as intended by the testing results and calculations. B. Extent of Condition: To confirm the extent of condition and to justify the selection of the worst A case configurations, a 100% survey of "O" functional tubing iscmetrics was Id performed to identify both horizontal and vertical "Z" configurations as depicted in IRI-365 Rev. O as well as other support configurations which deviated from the intention'of M-830-G-038. 'Ihis survey was used to develop worst case configurations for evaluation using ME-101 dynamic analyses to qualify the tubing layout and to calculate reaction loads. Spectral envelopes of the containment structure, reactor building and control structure were used to provida conservative dynamic input accelerations and included both seismic and hydrodynamic spectra. C. Significance: Our analyses completed to date demonstrate that tubing stresses remain within ASME III NC3650 stress limits for all loading conditions and that reaction loads at the tubing restraints remain within the design values used for the support qualification. Based c1 our 100% survey of the tubing installations, and subsequent analyses that will demonstrate acceptability of the installed configurations, we have concluded that this concern is not significant. D. Corrective Action: Project Calculation A-198 will be revised and issued by 2/27/89 to formally document the analyses described above. Calculation A-198 was selected as the appropriate document to include these additional dynamic analyses since it is currently referenced in drawing M-830. h =====================n==================================2======================= Remarks: Page ,2 of 4
i
)
FEB-07 '89 10 35 ID:P
. .._. - .. .--. ..ECO PR.OJECT .MGT D 1 V TEL N.O:215-841-978.... #299 PO8 . . . - . .
ge910:44 BCCHTCL WESTRN POWER iST FLOOR P. 34
. a. ... - -
.. . . .s .
, g-
, ._T l,(. ' =
.st E-WE88TER ENGINEERINCTIORPOMTIOK-~
~ ~ ~ ~ ~ - - ' ~ ~ ~
s l l C.7 LIMERICK 2 IDCA i t j (J.C. NO. 18138) l ygy
. . Observation Report No. EIM-053, Rev. 0 l W W ff = RESPONSE 00tffINUED l, ~:
_ .: 3 . . .,. . . . . . . . - -
- .: c: , . .. . .
..d.
. L. . Action to Prevent Recurrenos -
- - l
~. .
l Drawing 9031-N-830 4034 , will be r'evised by 2/27/89 to specifionlly limit 1
. - .:.. - - the tubing support intervals.to spans that have been W"y documented j
.as monoptable Desed on the revised calculation A-198. Sinos both the '
installation drawing (M-630) and the technical basis (calculation A-198) wilk be revised, no further corrective action is required. 5 v,
. .ns' .
s: 2 '.. ........................................ .............................. l 1h L {n - ,%hbhh4 ' M kH l]Mlff PECO RtWMMilf uture ;. 7Date BedKtel Approval Signature / Late ui.u.yh. .: ; -j:,; -
; Pagelofl
' @.,.P.
;;j,' . ,..
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT b Observation Report D OR- 053 Rev. l _0_ l Review Plan LK- D - 1904 -C Re v . ,1_, l PART III - RESPONSE EVALUATION The response to this observation report has been evaluated as follows: A. Causal Factor (s): X-Acceptable Not Acceptable (Explain): A lack of specificity on Drawing 8031-M-830-G038 allowed support spans to be installed in a manner inconsistent with the testing results and calculations used as a basis for the spans. B. Extent of Conditions X Acceptable Not Acceptable (Explain): 100% survey of Q-fuctional tubing isometrics was performed to identify th. worst case configurations for evaluation. Evaluation results affirming acceptability of these worst cases will envelope all Limerick Unit 2 Q-functional instrument support locations. C. Significance: X Acceptable Not Acceptable (Explain): Since subsequent analyses demonstrated that tubing stresses are within
.p allowable limits and support reaction loads are within design values for 9 worst cases, the installed configuration is acceptable. Therefore, the concern is not significant for Unit 2.
D. Corrective Action: X Acceptable Not Acceptable (Explain): Project Calculation A-198 will be revised and issued by February 27, 1989 to document the additional analyses which support acceptability of the existing design. E. Action to Prevent Recurrences .X Acceptable Not Acceptable (Explain): Drawing 8031-M-830-G038 will be revised by February 27, 1989 to clarify the requirements for support spacing consistent with revised Calculation A-198.
............................................. .......===.. ..=============.============
Additional Action Required: _g_No Yes (Explain) C....................................................................................... octCt Lhad Engineer Signature 1Mn / Date ocLts wak zlu/n A h Signature / Date Page j of j I
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (d ~J
) (J.O. NO. 18138)
OBSERVATION REPORT 03servation Report No. DOR- 054 Rev. L Review Plan LK- D - 1903B - MC Rev. L Reference AI No.: DAI-316 Rev. O PART I - INITIATION
- 1. Description of Concern The integral weld attaching the stanchion to the run pipe (Hanger GBB-203-H12) is not qualified in calculation CBB-203eC12, and the potential for a design deficiency exists.
- 2. Supporting Information The amended response states that the qualification of the stanchion and the stanchion to base plate weld was performed by comparison to Pipe support calculation Standard (PSC3)-10, which indicated that the subject stanchion dimensions and loads a e enveloped.
The amended response states that the integral attachment weld was qualified r% by comparison to the PSCS-10 baseplate weld. This comparison, which is-not documented in the calculation is not appropriate for this situation. The loads on the integral weld at the pipe are greater than the loads at the baseplate. Typically in veld designs, the bending moment terms have a significant impact on the required weld size, but there is no calculation for these thrms. In addition, the calculation does not determine section properties of the integral weld, nor does it calculate the design margin for the weld at the baseplate. Based on the above reasoning, it is inappropriate to qualify the integral weld based on the comparison with the baseplate weld.
================================================================================-======
Additional Documents Attached: None O
'rka k&su wn/a---
Lead Eng er Signature / D4te ----------------- ----------- APM Sig6ature #
/ Date ---
Page 1 of {
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA / p (J.O. NO. 18138) Q OBSERVATION REPORT J AN 2 4 1989 Observation Report No. EOR-054 Rev. 0
]
PART II - RESPONSE l i
- 1. Observation Concurrence: l l
X Concur with observation Do not concur with observation (Notes if not in concurrence, explain in " remarks" below)
- 2. Response to Observation:
A. Causal Factor (s) A review of the pipe support calculation GBB-203-Cl2 Revision 3, reveals i that the designer and checker made an undocurnented judgment in qualifying the weld betweeen the stanchion and the pipe elbow by comparing the size of the weld with the Pipe Support Calculation Standard (PSCS)-10 for welds between the stanchion and the base plate, based on the following facts: Pipe support GBB-203-H12 is a gravity support and is designed with a lubrite plate between'the stanchion and the 'F' type spring. Due to the relatively small gravity load and the presence of a lubrite piste, the weld between (A) the stanchion and the pipe will experience a small bending moment due to friction. B. Extent of Condition ; Based on the Limerick Material Iabor Control System (MILS) database, all Seismic Category 1 integrally welded stanchions that are similar in con-figuration to the PSCS-10 standard detail were reviewed for adequacy of the integral weld.to the pipe. This survey identified 20 additional seismic category 1 supports, using integrally welded stanchions, all of which were reviewed and determined to be adequately qualified and within design limits. We, therefore, conclude that this Observation Report concern is an isolated Case. C. Significance The integral weld on Pipe Support GBB-203-H12 has been re-evaluated for adequacy and has been been determined to be within acceptable limits. A review of the remaining seismic category 1 pipe supports with stanchions welded to the run pipe did not identify any other discrepancy. The concern is, therefore, not significant.
========================================================================================
Remarks: Page 2 of 4
FEB-03 '89 10:16 ID PECO PROJECT MGT DIV TEL NO 215-841-4578 #268 P09 O STONE & WEBSTER ENGINEERING CORPolWTION LIMERICK 2 IDCA gh f[ tw h/ (J.O. tJ. 18138) v QBSD V". TION REPORT Observation Report No. DDR-054 Rev. 0 , PART II - RESPONSE QQNTINUED D. Corrective Action Pipe support calculation 088-203<12 has been revised with adequate documentation for evaluation of the integral weld attaching the stanchion to the run pipe. E. Action to Prevent Recurrence asaed on the fact that the concern is not significant and is not generic, it has been determined that no further action is required. O , 0 -
~/N<A % 2ltlA1 PECo Review fignature /Date (f& mbhu Bocycl Approva;. Signature f
/' 24/t)
/Date Pagalof 'l
l 1 b i STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) l l i OBSERVATION REPORT Observation Report A R-O Ost, Rev. L Review Plan: LK- n - 1onin _gn_, Rev. L j PART III - RESPONSE EVALUATION l The response to this observation report has been evaluated as follows: , l A. Causal Factor (s): ,_ L Acceptable Not Acceptable (Explain): The response confirms that .; qualification of the weld was based on undocumented engineering judgement based on the use of the PSCS-10 standard detail' . 1 B. Extent of Conditions y Acceptable Not Acceptable (Explain): The Bechtel review of all similar calculations (Seismic Cat, 1 ' integrally welded stanchions similar in configuration to the PSCS-10 standard detail) determined that similar welds for all other supportr, of this type had been properly qualified. C. Significance: y Acceptable Not Acceptable (Explain): The evaluation of the weld indicated that the weld met existing design j] Q requirements. D. Corrective Action: Acceptable Not Acceptable (Explai.): Calculation GBB-203-H12 has been revised to document the qualification of the weld (calculation not reviewed by SWEC). E. Action to Prevent Recurrences y Acceptable Not Acceptable (Explain): Since the concern is isolated and assessed to be not significant, no action to prevent recurrence is required. Additional Action Required: y No Yes (Explain) I 1 Af09- Q ' h D. Nikt 2fl$$$ l Lead Engineer Sigriature / Date APY Signature / Date l I l Page i of 3
l STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA l (J.O. NO. 18138) OBSERVATION REPORT l Observation Report No. D OR- 055 Rev. O Review Plan LK- D - 1903B - HC Rev. 1 Reference AI No.: DAI-227 Rev. O i PART I - INITIATION
- 1. Description of Concern In calcualtion GBB-218-C78, revision 2, the section modulus values Sw 2 and Sw 3 for the weld of the rear bracket to the base plate were switched, resulting in an erroneous weld design. In addition, the field tolerance was not applied to the worst case swing angle. This could result in violating the maximum tolerance permitted by the component manufacturer.
- 2. Supporting Information Although the inadvertent switching of Sw 2 and Sw 3 in the analysis of the rear bracket to base plate weld in calculation CBB-218-C78 Revision 2 produced a conservative result, a concern exists that this may not be true for other weld designs with the same error.
Calculation CBB-218-C78, Revision 2 omits the inclusion of the field installation tolerance of 3 in the computation of the worst case swing angle. In order to include this tolerance, reanalysis was necessary to
'~1 replace the conservative values of pipe displacement in the calculation with actual values. This resulted in a decrease of 0.1 , and the swing angle then became 4.990 with the field t ole ran'ce included. Although this is marginally less than the allowable 5, a concern exists since this may not be true for other cases where the field tolerance has not been applied to the worst case swing angle.
Additional Documents Attached: None i
........ 3.........................p....................... p.... 9 ..=.........=..
i.e a
)%CJw) Vh ngineer Signature
/ 27
/ Wati
)
APM Signa 6re f
/' Da t e Page 1 of k
1 dM I3/h STONE & WEBSTER ENGINEERING CORPORATI h ag1E LIMERICK 2 IDCA REISSUED M=ud DATE , (J.O. No. 18138) _7 gd \d OBSERVATION REPORT MAR 01 1989 ) s Observation Report No. DDR-055, Rev.'O PART II - RESPONSE CDMPLETE AMENDED RESPONSE i
~
- 1. Observation Concurrence:
]
l X Concur with observation l l Do not concur with observation (Note: if not in concurrence, explain ! in " remarks" below) 1
- 2. Response to Observation:
A. Causal Factor (s) (a) The designer inadvertently switched the section nodulus values of referenced weld in pipe support calculation GBB-218-C78 Rev. 2. The checker of the calculation accepted the switched values, based on the fact that the final results of the switched section moduli were conservative. (b) Pipe Support Calculation GBB-218-C78, Revision 2 considers the n installation tolerance of 3' in the computation of the swing angle V in the local Y direction only. However, the installation tolerance in the computation of the swing angle in the local z direction was not considered. This was an oversight by both the designer and the checker. B. Extent of Condition To determine the extent of condition, a random sampling survey of calcula-tions for Seismic Category 1 pipe Eupports with struts and snubbers was conducted to check for:
- 1. Possible s" itching of weld section modulus values in weld calculations.
- 2. Consideration of the 3' field tolerance in the computation of swing angles.
The survey consisted of a randan sampling of pipe support calculations selected to demonstrate with a 95% confidence level that 95% of the pipe support weld calculations do not have either the weld section moduli switched, or omission of the 3' field swing angle tolerance that would result in a pipe support design that did not meet the design allowables. 64 pipe support calculations with snubbers or struts were randomly selected out of total population of 2833 Seismic Category 1 Large Bore pipe support calculations. A Limerick Unit 2 database printout of all
=========================-----=====================================================
O Remarks: b Page 2 of i
t STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) [ [ h OBSERVATION REPORT Observation Report No, DOR-055, Rev. O PART II - RESPONSE CONI'INUED the Seismic Category 1 Large Bore pipe support.s was used to assign a a sequence number of 1 to 2833 to each of the pipe support calculations. A random number generator was used to select the 64 calculations to be - reviewed for above mentioned errors. The random number list was used to select calculations to be included in the survey. If a selected calcu-lation did not have a strut or snubber, the next random number was used to select the subsequent calculation. E is process was used to determine the scope of the 64 calculations surveyed. The results of our scientific random survey demonstrate that the pipe support calculations are free of instances where the weld section moduli were switched when performing the weld calculation; however, one calculation, EBD-212-C18, Rev. 3, did not include a doc nted check for the swing angle in-the non-binding direction of the clamp end. We Revision 3 calculation did include a check for the binding direction swing angle and included the + : 3' fieldtolerance).Actualvaluescalculatedafterthesurveyindicatea 1.6' swing angle plus a 3' tolerance yields a 4.6' total versus the 6' manufacturer allowable, therefore, the design remains adequate. Since only one case of an undocumented swing angle check was identified and no cases of a switched weld section moduli were found, we conclude that the concerns identified in this Observation Report are isolated cases. ; C. Significance: In pipe support calculation GBB-218-C78, the referenced calculated weld is conservative, and the worst case swing angle with field tolerance of 3' is within the tolerance permitted by compenent manufacturer. In pipe support calculation EBD-212-C18, the swing angle at the clamp end in the non-binding direction is within design limits permitted by the manufacturer. he results of a review of 64 randomly selected pipe support calculations also demonstrate the lack of generic concern. We do not consider the concerns identified in this Observation Report to be significant. , l D. Corre %ive Action Cr.iculation EBD-212-C18 has been revised to address the swing angle plus the field tolerance in the non-binding direction. (Calculation EBD-212-C18 l Rev. 4 transmitted via P-0340 dated 3/1/89). Based on our review conducted, and the lack of significance of the concern, no further corrective action is required. l Page_)_,oE(o
l SIONE & WEBSTER ENGINEERING CORPORATION !- LIMERICK 2 IDCA (J.O. No. 18138)
>Q Q ,
Observation Report No. DDR-055, Rev. O PART II - RESPONSE COffrINUED E. -Action to Prevent Recurrence Based on D above, no further action is required. O a
/oate Beentei w rwal s g ature /oate g w Review signacure Page L/ ofi I q
l
. _ - _ - _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ f
l STONE & WEBSTER ENGINEERING CORPORATION I LIMERICK 2 IDCA g (J.O. NO. 18138) l OBSERVATION REPORT l Observation Report D OR- 055 Rev. O Review Plan LK- D -1903B - MC Rev. 1 PART III - RESPONSE EVALUATION The response to this observation report has been evaluated as follows: , l A. Causal Factor (s): X Acceptable Not Acceptable (Explain): i SEE ATTACHED PAGE B. Extent of Conditions , X Acceptable Not Acceptabla (Explain): SEE ATTACHED PAGE C. Significance: X Acceptable Not Acceptable (Explain): SEE ATTACHED PAGE O D. Corrective Action: X Acceptable Not Acceptable (Explain): SEE ATTACHED PAGE E. Action to Prevent Recurrence X Acceptable Not Acceptable (Explain): SEE ATTACHED PAGE Additional Action Required: X No Yes (Explain) i
.............a.........................................................................
O 9hLtd Lead Engineer Signature 3Mn
/ Date DCLkA 0.wk akt'e1 APF Signature / Date Page [ of (
A STONE & WEBSTER ENGINEERING CORPORATION ' LIMERICK 2 IDCA (J.O. NO. 18138) O OBSERVATION REPORT Observation Report DOR,- 055 Rev. 0 l PART III - RESPONSE EVALUATION CONTINUATION: 1 A. Casual Factor (s): Acceptable The response acknowledges:
- a. The designer inadvertently switched the weld section modulus values for the referenced weld in Calculation CBB-218-C78 Rev. 2.
- b. Due to an engineering oversight, the installation tolerance was not considered for the local s direction swing angle calculation.
B. Extent of Condition: Acceptable Bechtel's random sample review indicates (with a 95/5 confidence level) that the support calculations do not have either the weld section modulus values switched or an omission of the field tolerance that would result in designs that exceed design allowables. The review indicates that the concerns were
]
j isolated and not generic. No occurrences of switched section modulus values were identified. One calculation (EBD-212-C18) did not document the check for the non-binding direction swing angle. However, the review showed that the swing angle (including the ' tolerance) was less than the manufacturer's limit and had no effect on the skewed loads used in the calculation. Therefore, there were no occurrences of either concern that could have resulted in a deficient design. C. Significance: Acceptable The response indicates that the concerns had no impact on the technical qualification of the designs; therefore, the concern is determined to be not significant. D. Corrective Action: Acceptable The response indicates that calculation EBD-212-C18 has been revised to address the swing angle and field tolerance in the non-binding direction. E. Action to Prevent Recurrence: Acceptable Since the concerns were not generic and were determined to i be not significant, no further action is required. Po.y e c o,'n 6
I l i STONi: & WEBSTER ENGINEERING CORPORATION
'fm., LIMERICK 2 IDCA
) (J.O. NO. 18138)
(J OBSERVATION REPORT Observation Report No. DOR- 056 Rev. j Review Plan: LK- D - 1907 - C Rev. 1 Reference AI No.: DAI-337 Rev. O PART I - INITIATION
- 1. Description of Concern The correct dynamic capibility levels of the HP series Eagle Signal Timer are reported to be significantly lower than those stated. in GE documents DA 304A3630 and Limerick SQRT analysis for H12 and H23 and other panels (CE DRF Number A00-01560).
- 2. Supporting Information Review of original documentation (DRF Number A00-02095) resulted in response that DRF Number A00-02245 was the correct qualification documentation for the HP Series Eagle Signal Timers supplied to Limerick.
DRF Number A00-02245 shows two drawing numbers for the HP Series Eagle Signal Timer; dedication drawing (referenced in DA 304A3630) DB 188C8233, and 145C3043. DRF Number A00-02245 contains DRF Number A00-1084 which [,,) U includes QTR-HPS-Model 8 Qualification Test dated May 18, 1978. 1 of the report states that the fragility testing was conducted at 2.5g's. Page Page 3 of QTR indicates that contact bounce occurred for a 1.5 millisecond duration. Page 4 shows vibration along front to back axis of the timer as it would normally be mounted, and states: "at a fixed 33 hz, the timer: Failed. N.C. contact bounced continuously with clutch de-energized. When energized the clutch contacts did not bounce". GE .TI3547 details the inspection, testing, marking, and documentation to dedicate the Eagle Signal Timers DB 188C8233 to safety related application per DA 304A3630 that were sold to Limerick 1 and 2 via memo of shipment 320-73190-419. DA#304A3630 does not state that chatter, noted above, existed during fragility testing. DA 304A3630 states that for seismic (paragraph 2.4), timer is capable of 3.0 g's horizontal, and 1.75 g's vertical. This contradicts the fragility levels of less than 2.5 j g's noted above. I 1 CONTINUED ON 1 PAGE ) I Additional Docements Attached: None O ---------- ----------------------- -------------------------- -- -- ----------------- 1
/ ~ na ass ~
4e~adAngineerSignature /'Date 2 APM Sigfature i nkr
'/ Date i
Page 1 of LI
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) OBSERVATION REPORT Observation Report No. D OR- E Rev. O CONTINUATION: DRF Number A00-02245 references DRF-01560; Limerick SQRT. This analysis of Limerick H12 and H23 and other panels shows that in Table A page 8 the timer. (145C3043) has a limir. of 10 g's in all axes. The Limerick l SQRT report transmissibility plots and SQRT forms show timer (145C3043) mounted in panels where horizontal "g" loads exceed 2.5 g's, which contradicts the fragility levels noted above. O \ O Page 1 of II
l. l IDCN /f3) /n (s -o ?/f) ov l STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA
-(J.O.NO.18138)
OBSERVATION REPORT PART II . RESPONSE - AMENDED
- 1. Observation Concurrence:
Concur with observation X Do not concur with observation Mee " Remarks", below) Remarks
- 1) Per attached PIR 84-367, the referenceable DRF for DA304A3630 qualification is DRF A00-02245. DRF A00-02245 cross references DRF A00-1084-61 and DRF H13-37. Per PIR 84-367, DRF N13-37 contains the documentation to support seismic qualification for the Eagle Signal Timer O Model HP-5 for GE dedf cation drawing DA304A3630. The Eagle timer referenced by SWEC in DAI-337 and 00R-056, per MOS 218-9812E(GE Requisition No. 320-73190-419), was purchased as a replacement timer for MH application E11A-X028A in Limerick 2 panel H12-P617, per DA304A3630P008. Rev.1, Catalog Model #HP55A6.
'~
- 2) DRF H13-37 provides trie seismic test results for Cofrentes panel H13-P628, which monitored installed Eagle S 4nal Co. Motor Driven Timer Model HPS2A534(GEPPD145C3043P016). This was a dual axis seismic test:
lateral (front-to-back)/verticalandlongitudinal(side-to-side)/ vertical. The required response spectra (RRS) ZPA's were 2.5g in the front.to-back, side-to-side and vertical directions. The test response spectra (TRS) ZPA's were approximately 3.8g front-to-back, 5.2g side-to-side and 5.2g vertical. See attached copies of RRS/TR$ plots per DRF H13-37.
~.
- 3) Per DRF H13-37, Section B EWA: EAD-05, Rev. O, the required test chatter maximum allowable limit was 10 msec. Chatter was monitored during the test and no failures were reported.
- 4) The GE dedication drawing DA304A3630 states the seismic capability of the timer as 3.0g horizontal and 1.75g vertical - ZPA (55 damping.
multifrequency broad-band standard response spectra shape per ANS!/IEEE C37.98-1978). This difference from the DRF H13-37 test requirements /~tsults will be reconciled and any required document revisions will be implemented by 3/31/89. DOR - oSG REV o PAGE 3 OF If _ _ _ _ . _ _ _ _ _ _ _ _-_x________m_______m
ZDM 1933 U STONE & WEBSTER ENGINEERING CORPORATION. ) LIMERICK 2 IDCA (J.0. NO 18138) OBSERVATION REPORT Observation Report No. 00R-056, Rev. O Remarks (Continued) 5)' The SQRT report for Limerick 2 panel H12-P617 DRF A00-01560-3, Vol.1 Rev. 0, shows the E11A-K028A relay seismic capability as 8.00g-(F-B), 4.00g (S-5) and 2.50g (V), alth: ugh DRF.H13-47 is referenced in the Remarks column. The SQRT teport will be revised as required to reflect the correct seismic capability limits per DRF H13-37, by 3/31/89.
- 6) The worst case location panel H12-P617 maximum expected accelerations are 3.09g(F-8),1.479($-5)and1.00g(V),perDRFA00-01560-3.Vol.1.
Rev. O. Per Item 2) above, these acceleration levels are bounded by the DRF H13-37 test results.
- 7) Added response The following responds to the SWEC request per 3/15/89 telephone conference memo for revised response, per item number, a) " Confirm the actual seismic capability of the relay".
GE concurs with the SWEC assessment that the Eagle Signal Timer ' qualification front-to-back, level 3.6gper DRF H13-37 side-to-side andis 3.83 approximately(3.8gZPA. vertical input at the bottom.of the tested panel). Although tie DRF W13-37 qualification test report states that the chatter which occurred above the 10 msee acceptance level at the 5.2g IPA side-to side test condition was causedbythetestfacilitydrivebottomingout(andthereforethe test results were accepted), GE concurs that the basis for this conclusion is not adequately documented and the 5.2g test results should vot be used. b) "Corfirm the expected accelerations for all GE timer locations". The attachment 3 table lists all (24) of the Eagle Signal timer applications at Limerick 2 and compares the tested timer capabilities (ZPA input at the bottom of the tested panel) per a) above with the expected accelerations for the Limerick timers at their actual panel locations (usingthe Volume 1, Revision 0)panelaccelerationgridsinORFA00-01560-3
. These data are not design verified. As noted, only 2 of the timer applications are acceptable in all 3 directions.
DOR -050 REV cr PAGE 4 OF 16
IDcA /933 O STONE & WEBSTER ENGINEE4ING CORPORATION LIMERICK 2 10CA (J.0.NO.18138) OBSERVATION REPORT Observation Report No. DOR-056, Rev. 0
- 7) b) Added response, Continued However, the tested Eagle signal . timer (GE PPD 145C3043P016,'Model No. HP52A534) was located at 75 inches height and 32 inches from the j left side in the test panel (Cofrentes panel H13-P628). A comparison L was made between the tetted timer panel location and the location of all of the Limerick timer applications. In all cases the Limerick )
timers are at lower paAel locations than the tested timer. A comparison (unverified) has been made of the tested panel TRS at the bottom with the floor RRS for the Limerick control. room panels, and the TRS 'enevelopes the RR$ at all frequencies. Thus, the Limerick timer applications can be qualified based on their lower panel locations than the tested timer, as'= < ing the tested panel and Limerick panels can be shown to be structurally similar. This panel similarity evaluee tion is in process, in accordance with the methodology described in DRF A00-01560-3, Volume 1, Rev. O, and is scheduled to be completed by 4/14/89. If the panel similarity cannot be confirmed, then it would be necessary te analyze the panels to determine the psuedo in-structure TRS at the tested timer panel location and the in-structure RRS at tne Limerick timer panel locations to show that the TRS enevelopes the RRS and thereby qualify all of the Limerick timer applications. GE will perform these additional analyses if necessary to confirm the qualification of the Limerick timers, c) All appropriate SQRT forms and other affected GE documentation will be revised to correct the identified errors by 4/28/89.
~
d) 5echtef is in the process of reviewing their records to identify any Eagle Signal timers procured from GE NE, San Jose per PPD 14503043 or DA304A3630, either by Bechtel or a Bschtel subsupplier, if any are identified which used the erroneous GE seismic capability limits for the timer, the expected acceleration levels will be checked against the corrected capability limits and corrective action taken if necessary. The scheduled completion of this effort is 4/7/89. e) The 10 msec chatter criterion is considered a standard acceptance level .for typical control instrument applications, and no deleterious 4 effect is expected on the safety related function of devices controlled by the timer. However, GE is in the process of confirming O
- this expectation for all of the Eagle Signal timer applications on Limerick. The scheduled completion date is 4/14/89.
DOR E56 REV o PAGE __fi OP- 86
._-._____a_ ____--_______._______.____m_
IDcg 79y STONE & WEBSTER ENGINEERING CORPORATION /O !.!MERICK 2 IDCA O (J.0.NO.18138) OBSERVATION REPORT Observation Report No. 00R.056, Rev. 0
- 7) Added response, Continued f) The differences between the tested timer, 145C3043P016, and the Limerick timer reviewed, DA304A3630P008 are judged to be insignificant for dynamic qualification purposes. However, GE is in the process of performing a similarity analysis to confirm and document this judgement. The scheduled completion date is 4/28/89.
g) GE will perform a review of the correctness of the seismic capabilities reported for all of the cther relays identified in the Limerick 2 SQRT report for control room panels. DRF A00 01560 3, Volume 1. This review is scheduled to be completed by 5/12/89.
- 8) Added response to further address significance.
Weareconfidentthat,oncompletionofthecommittedactionsper7) above, all Limerick 2 Eagle Signal timer applications will be shown to be G qualified to perform their intended safety functions. Therefore, the V significance of this observation is of no concern. Additional documents attached:
- 1. PIR 84 367, Qualification of the Eagle Roset Timer (DA304A3630),
10/24/84.
- 2. Figures 83,84,8-7and88fromDRFH13-37(RRS/TRSplots).
- 3. Capability vs. Expected Acceleration for Limerick 2 Eagle $ignal Timer Applications. ,
NOTE: Only Attachment 3 is submitted With this amended response. The other attachments were submf tted with the initial rMponse.
.W^'~*
PEco Rov ture y Da e R 0 hot.rt.
- ec te Ypprovy1 signature / U6te Pljf 3/LV/R \
O F* OE M~51'gnature AFptova 3/444r
/ Datd DOR - os G REV o PAGE G OF if
TEL No. w 40,8_,.925 d952 Feb 09.89 16:31 P.04 Compo_ne~hr.
- lPD/PWA E sevaence nve- '-
QUALIFICATION ANALYSIS AND ,,, j,' 710 g4. 3g 7
<- 'OESIGN ENGINEERING - 710
( . 3 GRAM IN FC Rh*.ATION R EQU EST / RE LE ASE hj[g ~~ V'" m:
,,,9 gfp --
io' TrMcy* l
% (A, h PRO.60 N.G.Luria L. Pohl M/C 429
- . s o,; :A;: mro.At: s to n:;tet attutact
, 3ggg777g I 10-24-84 various titCT l
.inlification of the Earle Reset Timer ('DA304A3630)
LFoRMAT10H REQUE STED/itELE ASED he Eagle Reset Timer Model HP-5 DWG. DA304A3630 is qualified to the requirements ontained therein. The attached Quality Assurance Audit Report has established similarity f the now purchased commercial parts to those previuosly available as Class 1E. The edication procedure will complete the qualification process for all purchased parts against WG. DA304A3630 . Supporting doc' u mentation for the environmental levels given on the above ientioned drawing is contained in DRF A00-1084 Index 61. The documentation required to
- upport seismic qualification is contab f in DRF H13-37.
NOTE:'THE REFERENCEABLF DRF FOR DA304A3630 QUALIFICATION IS DRF A00-02245 NJ DOR -- 054 REV o PAGE 7 OF /C DISTRIBUTIb: n o r * *. .Lastsave a: E= cwt es co ics,a w i . . . . . . .i _ q N.G.Luria n iw. n -a. Q B.Bohn (w/o attachments) R. Garrison o ,.1__ n a wes. M 8 -n-n e -... O '* a-D.Dunbar (w/o attachments) _1 R.Pcnse (w/o attachments) n -u- n -a. file n n nav u"
L~ - xm 3E V A V\ N
- T C
~ 6 D
\
t B/ - D1 O T ( _ 7 _ 0 \
'V _
8T . ^ 7A4 7 :
/
2S V
/E .f l
3S- / L EUG SPN l/ / .( O A HFP POh
. I
. /
/
N t A f_ D I E . S g; S ._ L _ _. A ' R _ E T A L
~
B E ~ S ' P T ~ - SS TE NT ~ E RT FS CO CPG O N D 2 - g I OO2 i f e 2 o N>Q A o* -
l!\Iill 4' a ' b3 3' !!. u o * $ '* " 7r o o
- m e ~
- g l mom iji!
- ]
qDGb N N ML" t% N A1 b N' @. N ow 4 2
,li j;>l
- l Il. ! E 4
- - 5U
,( w '
9R T i
'q. ' o C l E N P S
3E V
\ oCI I
A N y R T \
\ E C ,
N O / E V
\ G
=6 ii iA l, o t
B/ D1
;gi
. i' N' !i T
S E e 0 i g. L g ! T s a
^
h P
% , S l
/ I f O
t l}.6 o 8 \ X - 7T . A t
/A4 ,
ff ' 4 O u 2 L
/S A A l a
3E , 3U
'/v r
S- D e
//
t EL a. SUG " E 1 APN I s 1 . LS 8
,J H -
- S 2 PFP 6 P
OM ~ _ F A s e O#D - t n e T ~ /
!!q r
r U - r. ..
;~ -
l
- o O .
[l! . C L ijii
~
i ,
?'
A 8 1 R ' l . 8 E
,i e
T - ,
~
r _ A u g L i F 8 2 6 !q P _G SS T O L _ TE NT
, l i
7 _ H E .
~
RT . - FS 2 OO - CPG 00
,,1 i
i ' 0 - [g 2 1 4 3 g o1
~ ' '
N O 0 } _ 1 1 no" 1 e $< O m>@ a _9 o.n _ ll- 1!l' !l!l L
j j l l .l!
! ! i! I I am am z.
' ii I a 8 ,d gv" 3 ;o* Ie. pm.'&w!
i 1E n
'.oa-MDnAxNN 9Rt6 s
.hi3s s g$ % gT + 2 Gf 4 O
I l i' 5U l 9R
\
\ .
T C N E x
\
/
N P S 3E V - U ' \ C I A N R T C
% E N
O s\
/ \ E G
6 3'f I
\ ,
B/ - L T D1 - 1i ysS S
- \g' E 0 T d'
r
/
ti
/
S I O e I i ' X w T A A4 iII L\ /
/
w
" L a S A A n E '
U n aS-7L N/ D a n
/UG M/ E A 3PN / S -
/ I / S c O 3 FOPP.
E*A S*D n a n e A H P n_ N ' i
/.
I , m m T G e N a r O L - B 2 w G il!l.' I, pi - P G T O SS L TE ,' T Z NT H E RT FS 2 OD - CPC - 00 _ ' 4 . 1
~
0
~
- 2 1 D1 N -
D _.. O . 1 co" I $e mm< 0 T$m 3 O % I. -
l :l mm - ~
- f r i.
ug M ,,em 2cr oe m* ; $! i "5 Eo6~ HQhh g~
- V k, Dk 3k$ N s h* A ,% M 2 N
; 6M
- - 5U 4R
]t11_' T
- 1C C E
N P _ S 3E V A
\ \ C I
A N R y T E _ C N O h E G
=6 B/ gl 5
b , T DI 3 S e I s 0 ._ s E a T h
~
% r ' '
S P f O
/ I o h X t u
T . m A O 8A4 ,
/
7 - L l a
/S h/ A n 3E U i
[N/ d
/S- / D u 3L t UG " E i
g EPN I . S n o S S 1 L AF= 4 3 HO 2 P*N *D 6 P F s O f f e t
.lI n
T ' e r U r O o C T - G 8 N 8 O - L e
" r u
8 g i 2 F 6 g G l P O T L SS TE ' Z NT H E ' RT - FS . 2 OO . CPC . 00 1 J - 0 2 1 . O1 N _ O s I g2 3 ogc 2m< O jC -: o* \ LJ Il 1 { lt.l j .Nr ;l r'lI! . (!i i!
1
.ZDM /933 I
1 00R-056 1 Attach. 3 CAPABILITY VS. EXPECTED ACCELERATION FOR LIMERICK 2 EAGLE SIGNAL TIMER APPLICATIONS LOCAL i ACCELERATION EXPECTED CAPABILITY (1) ACCELERATION-(2) PANEL NO. DEVICE NO. DWG./PART NO. f-b s-s- v f-b s-s v ACLEPTABLE H12P609 G31-R616A 145C3043P004 WWW W77 No H12P611 G31-R6160 145C3043P004 3.8 3.6 3.8 5.04 .91 .67 No H12P617 E11A-K028A OA304A3630P008 3.8 3.6 3.8 1.50 .72 .89 Yes H12P618 E11A-K028B 145C3043P008- 3.8 3.6 3.8 1.50 .72 .89 Yes H12P628 821C-K08A 14SC3043P004 3.8 3.6 3.8 2.51 4.99" 1.00 No H12P631- B21C-K08C- 145C3043P004 3.8 3.6 3.8 2.51 4.99 1.00- No H12P644 E32-N600B 145C3043P004 3.8 3.6 3.8 2.C1' 4.99 1.00 No H12P644 E32-N600F 145C3043P004 3.8 3.6 3.8 2.41 4.70 1.06 No H12P644L E32-N6018- 145C3043P005 3.8 3.6 3.8 3.24' 6.00 1.06 No H12P644 E32-N601F. 145C3043P005 '3.8 3.6 3.8 2.51 4.99 1.00 No: H12P644 E32A-K498 145C3043P003 3.8 3.6 3.8 3.24 6.00 1.06 No H120644' E32A-K49F '- 145C3043P003 3.8 3.6 3.8 2.51 4.99 1.00 No
.?v544 E32-N6028 145C3043P013 3.8 3.6 3.8 2.51 4.99 1.00 No
.12P644- E32-N602F 145C3043P013 3.8 3.6 3.8 2.41 4.70 1.06 No H12P645 E32A-K46 145C3043P003 3.8 3.6 3.8 2.51 4.99 1.00 No H12P545 E32-N604 OA304A3630P009 3.8 3.6 3.8 3.24 6.00 1.06 No H12P646 E32A-K49K 145C3043P003 3.8 3.6 3.8 3.24 6.00 1.06 No H12P646 E32A-K49P 145C3043P003 3.8 3.6 3.8 2.51 4.99 1.00 No H12P646 E32-N600K 145C3043P004 3.8 3.6 3.8 2.51 4.99 1.00 No H12P646 E32-N600P 145C3043P004 3.8 3.6 3.8 2.41 4.70 1.06 No
-H12P646 E32-N601K 145C3043P005 3.8 -3.6 3.8 3.24 6.00 1.06 No H12P646 E32-N601P 145C3043P005 3.8 3.6 3.8 2.51 4.99 1.00 No H12P646 E32-N602K 145C3043P013 3.8 3.6 3.8 2.51 4.99 1.00 No H12P646 E32-N602P 145C3043P013 3.8 3.6 3.8 2.51 4.99 1.00 No j NOTES:
.(1) AccelerationcapabililyistheZPAinputatthebottomofthetested ranel, in g ur.its.
(2) Local expected acceleration (ZPA) is at the location of the device in the Limerick panel, in g units. (3) Alterr. ate part DA304A3630P004 (4)' Alternate part DA304A3630r005 (5) Alternate part DA304A3630P003
- 6) Alternate part 0A304A3630P013 1
DOR -or6 REV o PAGE 11 _OF 15
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. No. 18138) OBSERVATION REPORT Ob.;ervation Report D OR- 056 Rev. O Review Plan: LK- D - 1907 - C Rev. 1 PART III - EESPONSE EVALUATION The response to this observation' report has been evaluated as follows: A. Causal Factor (s): Acceptable Not Acceptable (Explain): N/A B. Extent of Condition: Acceptable Not Acceptable (Explain): N/A C. Significance: Acceptable Not Acceptable (Explain): p N/A b D. Corrective Action: Acceptable Not Acceptable (Explain): N/A E Action to Prevent Recurrence: Acceptable Not Acceptable (Explain): N/A
.9 NE M .91.T N . M ................................................................
Additional Action Required: x No Yes (Explain) O .... Q ............ g 4sp, p ................................................... uw n wmc sgn 1x.L Lead Engineer Signature / Date APM Signature / Date Page d of _lI
I STONE & WEBSTER' ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) . j A OBSERVATION REPORT Q Observation Report DOR- 056 Rev. O h PART III - EESPONSE NTALUATION CONTINUATION: 1 The response is acceptable for the following reasons. GE has committed to further evaluation for several topics, as noted: j
- 1) . Review ofilocumentation' originally referenced by GE for dedication and quali'ication of Eagle Signal Timers, Type DB188C8233P008, ,
indicate; that DRF A00-02095 was the wrong design file, thus generating an Action Item (AI).
- 2) Response to AI indicated that DRF A00-02245 was the correct reference; GE would revise these references in qualification documents.
- 3) Review of DRF . A00-02245 and its cross reference DRF A00-1084, revealed further seismic capability discrepancies, leading to.
issuance of the OR.
- 4) Original response to the OR gave the correct file as DRF H13-37, also cross-referenced by DRF A00-02245; quoted the seismic capability and chatter limits; noted that Panel H12-P617 was A the worst. case location and that the required acceleration was d bounded by the timer capability; and that several documents would be revised (See DOR Response Items 1-6).
- 5) Review of these documents resulted in the following:
a) The timer capability quoted in the OR response was incorrect, since values for two axes were as the result of a test facility malfunction, resulting in contact chatter exceeding the test criterion of 10 msec. (Response 7a). b) Review of actual timer location in LGS panels found that expected seismic accelerations exceeded both the corrected timer capability and in some instances, the original, higher (incorrect) levels. (Response 7b). It should be pointed out, however, that GE has, so far, conservatively compared response curve ZPA for the base of the tested panel (which contains the applicable relay), with the expected ZPA at the actual device locations for the Limerick panels. The response indicates that the tested timer was located higher in the test panel, than the timer for any LGS panel; and that the test panel baars ' EPA is higher than any LGS panel base ZPA. Thus, the LCS t.ime r may be qualified, based on their lower location than the tested timer, provided that test and candidate panels can O be shown to be structurally similar. (DOR Response 7b). CONTINUED.ON ONE PAGE Pop e ij of -is
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) [D OBSERVATION REPORT j Observation Report DOR- 056 Rev. O PART III - RESPONSE EVALUATION CONTINUATION: ACTION: GE will perform a structural similarity evaluation between the test and appropriate LGS panils. If this cannot be confirmed, GE will determine tested timer TRS and show that it envelopes the worst LGS timer RRS. I c) ACTION: GE will revise all applicable documentation to show correct Eagle Timer (Model 145C3043) data. (DOR Response 7c). d) It 'is possible that these Eagle Timers may have been purchased by, or through, Bechtel, to the original incorrect qualification levels, ACTION:, Bechtel will review their records for purchase' by them, or their 'subvendors, of the Eagle Timer. to incorrect seismic capability levels and verify the qualification status, as necessary. (DOR Response 7d). e) The OR questioned whether the 10 msec. chatter level would affect devices controlled by the timer, ACTION: GE will confirm that the ' safety-related function of equipment controlled by the Eagle Timers are not jeopardized by the 10 maec. chatter level (DOR Response 7e). f) The OR noted differences between the LCS timer model and the tested model, ACTION: GE will perform a similarity analysis by comparing the LGS timar model to the tested version. (DOR Response 7f). g) Since, for the two relay applications reviewed, significant discrepancies were acknowledged, the OR requested an Extent of Condition review for all Class 1E timers used in GE control room panels at LGS, ACTION: GE will review the seismic capability of all relays listed in DRF A00-01560-3, Volume 1, the general reccrd file for control room panels. (DOR Response 7g). O Pag e s & is
STONE & WEBSTER ENGINEERING CORPORATION h LV LIMERICK 2 IDCA (J.O. NO. 18138)
- f. OBSERVATION REPORT Observation Report No. D OR- _057 Rev. O Review Plan: LK- D - 1909 - S Rev. 1 Reference AI No.: DAI-478 PART I - INITIATION
, 1. Description of Concern There may not be sufficient review of loadings _ from pipe supports on embedded plates.
1
- 2. Supporting Information Spring . Hanger GBB-218-H26 is attached to an embedded plate, the detail.
of which is shown on Detail 19 of drawing C-603. The calculation which provides _ limiting loads for this plate is calculation 26.3. Sheet 41 of this calculation-lists a moment capacity of 16 ft-k. The mechanism for evaluation of support loads is the use of coordination
.% prints. The coordination print for GBB-218-H - gives a moment of 209
, in-k (17.5 ' f t-k), which exceeds the capacity of the plate as calculated in calculation 26.3, but it was accepted by the Civil group.
Response.to Action Item DAI-478 explains that because the concrete strength has increased since the time of the calculation, the capacity is actually greater than the applied moment of 17.5 f t-k. Although this may be true, there is no evidence to suggest that this evaluation was done at the time of coordination print review.
... ................................................................................=.=.
Additional Documents Attached: None (3 ........................................................... U Scr'cu soa sias essescu iz.z14e
....b.................. if1 ed Lead Engineer Signature / Date APMSigpture / / Date '
Page1ofI _-- _ - - _ - - - - - - - - - - - - -- - - - - - . - - - -- - - - - - - - - - - - a
.m STONE & WEBSTER ENGINEERING CORPORATION I I LIMERICK 2 IDCA V (J.O. NO. 18138) OBSERVATION REPORT M30 W Observation Report No. DDR-057, Rev. O PART II - PISPONSE
- 1. Observation Concurrence:
Concur with observation X Do not concur with observation (Note: if not in concurrence, explain in "ranarks" below) !
- 2. Response to Observation:
A. Causal Factor (s): N/A B. Extent of Condition: N/A C. Significance: N/A D. Corrective Actfon: N/A (
\ E. Action to Prevent Recurrence: N/A
==-----=================,=========-----===========-------==========-----============
Remarks: , Structural members, including embedded plates, which support pipe harger attach-m<nts are evaluated as part of the Structural Irad Verification Program. This program assesses pipe support loads on Civil designed structures, both steel and concrete, to assure <xrnpliance with the Civil / Structural design criteria. The program is as follows:
- Pipe support drawings are prepared by PJant Design Group and coordinated with the Civil Design Group.
- The Civil Group reviews the affected structural members for adequacy based on the loads supplied with the coordination print. Tne review utilizes existing calculations. When deemed necessary by the reviewer, additional calculations are prepared. In some cases, an undocumented assessment is performed by the reviewer to record the coordination loads and the reviewer's justification of his approval of the coordination print.
- Pipe support coordination drawings are annotated to reflect any comments and are signed by the responsible Civil Engineer to document that the pipe support loads can be accommodated by the affected structural members.
O .O Pagelof I
, FEE-96 '89 15:18 ID:PECO PROJECT MGT DIV TEL N0:215-841-4578 #293 P05
~~
, -9? 28A0.: BCCHTry#_9Pt FOt4Es :1~ r;.0CF P. M SMNS & WEBSTER ENGINEERING CORPORATION 1.1MERICK 2 !DCA (J.C. NO. 18136) l OBSERVATION REPORT Observation Report No. DcR-057, Rev. O PART II - RESPONSE CONTINUED l
The Observation Report states that there is no evidence to suggest that evaluation of pipe support GB8-216-H26 was done at the time of coordination print review. The response to CAI-478 described a otsis for acceptance of the pipe support loads shown on the coordination drawing but did not identify any supporting calculation done at the time of the coordination print review. An assessment was generated to confirm the engineering judgement exercised during the review and approval cycle of this support. Se assese:nont does not compute the embed plate ce;ecity but evaluates it as adequate fo- the applied loads. The evaluation takes credit for the fact that the actual design concrete strength is greater than that considered in Calc. 26.3. which developed generic embed plate designs. he above discussion shows that there is sufficient review of loadinge from pipe supports on embedded plates. O I i e 599999 E B S S S t e merW5G W W WWWa pS B S S SS man ett e ss ag gBBB BBE NESW ymasem p S tama SSS Es e n gyg g sy g yper us s , f S. ' e ! h reco e vr.w stenature aste =cntei wrova:. signatw mate Page 3 of I
l STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2.IDCA (J.O. NO. 18138) l ' .. s - OBSERVATION REPORT l ) (O Observation Report D OR- 057 Rev. J Review Plan LK- D - 1909 - S Rev. l - !~ PART III - EESPONSE EVALUATION The response to this observation report has been evaluated as follows: A. Causal Factor (s): Acceptable Not Acceptable (Explain): NOT APPLICABLE B. Extent of Condition: Accepta.le Not Acceptable (Explain): NOT APPLICABLE C. Significance: Acceptable Not Acceptable (Explain): NOT APPLICABLE D h D. Corrective Action: Acceptable Not Acceptable (Explain): NOT APPLICABLE E. Action to Prevent Recurrence: Acceptable Not Acceptable (Explain): NOT APPLICABLE
=======================================================================================
Additional Action Required: _X,No Yes (Explain) SEE ATTACHED PAGE
=======================================================================================
. % J a
/ Date Lead Engineer Signature / Date AlWI Signature Page 1 of I
STONE & WEBSTER I:NGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138) ' [ OBSER'JATION REPORT Observation Report DOR- 057 Rev. O PART III - RESPONSE ETALUATION CONTINUATION: The Observation Report was written because loads applied to an embed plate exceeded the design capacity of the plate. This condition wen accepted during the review of the coordination print without writtea technical justification for the acceptance. Additional information has been provided to indicate that a higher concrete strength existed for this plates there fo;;e , it is clear that for this case the plate is acceptable. This adequately resolves the technical concern for the loading on this plate, and the-Observation Report is closed. I i, i v Papsof(
H t. STONE & WEBSTER ENGINEERING CORPORATION. LIMERICK 2 IDCA (J.O. NO. 18138) 7
'\ OBSERVATICE REPORT Observation Report No. D OR- 058 Rev. O Review Plan LK- D - 1903 - S Rev. 1 l
( Reference AI No.: DAI-430 PART I - INITIATION
-1. Description'of Concern )
The size of ' a ' fillet welo shown . on drawing C-942 is smaller than that specified in the design calculation. i
- 2. . Supporting Information- l Sheet' . 29 . of ' calculation .17-A-3 computes the required fillet weld' size for the connection ' of a W24 member to' the box beams to 'be 5/16", but Detail 4 on drawing C-942 shows a k" weld. Using the methods in calculation 17-A-3, the k" weld'would be overstressed.
~ \
4
-J
)
1 1 I p namemass mos es semanne s senes sem ommsessnesam mamam ens asemosom as sam m messem a ss e ssa mm e ns sa mmem
-Additional Documents Attached: None semesessessassmemenessmannssumssomspumassesammensammamassenessammme smassnessasssmamens
%() htY for Beere Etttsso i2-inr6 Lead Engineer Signature / Date APMSipature
/
/
h
/ / Date Page1ofJ{
1 h STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA ( J. O. NO. 18138) J AN 3 01989 OBSERVATION REPORT-Observation Report No. DOR-058 Rev-0 PART I - RESPONSE
- 1. Observation Concurrence X Concur with observation
____ Do'not concur with observation (Nete: if not in concurrence, explain in " remarks" below)
- 2. Response to Observation:
A. Casual Factor (s): The
Description:
of- Concern in~ the Observation. Report indicates'the weld shown on-detail 4 of Drawing C-942 is smaller than the weld specified in calculation 17-A . 3, z aheet 29. The fact is that this is not the case O since both the drawing and the calculation conclusion specify the same size weld (1/4").. Notwithstanding the above, the calculation made to size the weld computes s'5/16" weld, using AISC Manual Table-XIV. The computation is correct. The difference between the computation result and the conclusion appears to be a mistake by the originator and checker of the calcula-tion. B. Extent of Condition The. calculation applies to' skewed welded plate connee-tions between rolled sections and supporting bex beams at platforms inside the containment. This type of con-nection is common only inside.the containment. Connec-tions in other areas generally use bolted connections with clip angles which are not susceptible to the des-cribed condition. A11-(11) . platform-beam -design calculations (file 17A), and a representative sample (52 out of 315) of other steel design calculations (files 17 and 113), inside the containment, were. examined to identify and review similar calculations which used AISC Manual Table XIV. Some calculations examined were prepared or checked by the same engineers as Calculation 17-A-3. No recurrence of the condition was identified. Page _2_ of 4
a N299 Pp6
--.FEB ,..: :.sID:
._ -. . ,-.'89 10:34.,, ,v. ..PEC.,O
.s m.P,RO,J,ECT ri .
f1GT 4 D,i.v.TEL
;- NO:--215-641-4578 STONE & WESSTER. ENGINEERING CORPORATfOlt
~
.M LINER!cx 2 !DCA
[,'" tJ.D. RO. 18138)
. OBSERVATION < REPORT .
Oteervatten Report No. 00R=00D h O' ,
,m _ .o.,
--1
.- Therefore, the sited sensern to constuded to be as "isoisted ones.
~h~
C. 51'gnif1Eenset
'Thie' esteulation dieerspeney has . . .been determined to by
"'~~~~~~
en isolated instenee.The connection has been resenkysed 4 in more detail as described under D, corrective Ast&as.-
. . " . '. The re' analysis shows the 1/4' weld estisites the ese6ge
';; . + - . eriteria.; The soneern therefore has ne signifiennes.
*D.CerrectiveAhtene O YaToulaEten reanalyse the eennection.
17-A 3 Nov 8 has been completed to
# El ' Action to Prevent Resurrenost N
'..n.e o " action'is or ine ett., required beesuise sen.orn ene .'it. er t.he lack' os signifi-
.ter.in.tton i. ..
an tentated onee.
- v:. , . :. .
RENARMS nons d.hi.1 % 4.M dbe/%. - A # @/s1
~
1 . O.
. et
.- ,s.e a_ e, .3_ .
l .. c . .-< ._
STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA (J.O. NO. 18138)
\
OBSERVATION REPORT Observation Report D OR- 058 Rev. O_ Review Plan: LK- D - 1903 - S Rev. L l- PART III - RESPONSE ETALUATION, The response to this observation report has been evaluated as follows: A. Causal Factor (s): X Acceptable Not Acceptable (Explain): The cause is acknowledged to be a mistake in concluding a weld size smaller than the computation shows. l l l B. Extent-of Conditions X Acceptable Not Acceptable (Explain): The statement that the condition is an isolated case is reasonable based on Bechtel's review of all the box beam design calculations and a representative sample of other steel design calculations. C. Significance X Acceptable .Not Acceptable (Explain): The discrepancy is determined to be an isolated instance and the connection in question has been reanalyzed and shown to be adequate (with the acceptance of a 6% overstress). D. Corrective Action X Acceptable Not Acceptable (Explain): The affected calculation has been revised (17-A-3, Rev. 8) to reanalyze the connection. The revised calculation has been reviewed by SWEC. E. Action to Prevent Recurrences __g_ Acceptable Not Acceptable (Explain): i No further action is required based on the determination that this is an isolated case. ses.............s..s.G.ss....sss....... ass..ss......m..s...s.as..........s....s........ Additional Action Required: 2o N Yes (Explain) l I i
..................s......s..s..s.................s...............as..............s.s...
)L rCL. 2/u/n C(alt
- 4 b. WiIIe 2lwlt1 1;ead Engineer Signature / Date 'APO Signature / Date Page 1 of 1
i l STONE & WEBSTER ENGINEERING CORPORATION LIMERICK 2 IDCA 7 (J.O. NO. 18138) OBSERVATION REPORT l l Observation Report No. D OR- 059 Rev. O Review Plan LK- D - 1903 - S Rev. _1_ Reference AI No.: DAI-035 Rev 1 PART I - INITIATION
- 1. Description of Concern The modulus of elasticity used in the seismic analysis of structures is not consistent between the Reactor Building and the Spray Pond Pumphouse (SPPH),
- 2. Supporting Information During the review of the seismic analysis of the reactor building (24.1 series calculations), it was noted that the modulus of elasticity value used was considerably higher than that given in the ACI 318-71 Code.
Response to Action Item DAI-035 indicated that the higher value was based on tests, and was appropriate for the 4000 psi concrete used at Limerick, and was used for all seismic analysis of major structures. Contrary to this, it appears as though the seismic analysis of the SPPH uses the value specified in ACI 318-71. (Sheet 5 of calculation 70-A) ( ) (_) Since the SPPH is also constructed of 4000 psi concrete, there appears to be an inconsistency in the two analyses.
.......................................................=.......===...=.=.==============
Additional Documents Attached: None . l I h _ $4Mt foR Beas Eggswa Lead Engineer Signature 12- 29-tt
/ Date APM Sigj6ture
^
Ob7ff I 4 Date Page 1 of [ t
S'IONE & WEBSTER ENGINEERING CORPORATION /Q LIMERICK 2 IDCA (J.O. NO. 18138) ( /t U 7fG f m OBSERVATION REPORr W 2 5 409 Observation Report No. DDR-059 Rev. O PART II - RESPONSE
- 1. Observation Concurrence:
Concur with observation X Do not concur with observation (Note: if not in concurrence, explain in " remarks" below)
- 2. Response to Observation:
A. Causal Factor (s): N/A B. . Extent of Condition: N/A
.C. Significance: N/A D. Corrective Action: Not Required E. Action to Prevent Recurrence: Not Required A ==================================================================================
Remarks: We do not concur with observation. The use of different concrete modulus of elasticity values was recognited during the seismic analyses of Limerick structures. Detailed response to the items cited by SWEC as supporting information is provided as follows: In 1971 when the seismic analyses of the containment and the reactor building were initiated, it was recognized that although ACI 318-71 applied to these special structures, additional code requirements were being developed in the ASME B & PV Code, Section III Division 2, and in ACI 349. Rese structures are special in that they are heavily reinforced and of massive concrete construction with unique design configura-tion. Rey support the bulk of structural sub-systems, piping systems, and equipment. l 2erefore, special attention to the seismic analyses was not only warranted but was prudent. In that light, key parameters such as the concrete modulus of elasticity were thoroughly evaluated by appropriate testing and the tested values were incorporated in the seismic analyses as permitted by ACI 318-71. De seismic analyses of the remaining Seismic Category 1 structures, such as the Spray Pond Pumphause (SPPH), used the ncminal concrete modulus value provided by ACI 318-71. Rese structures are of conventional reinforced concrete construction for which the provisions of ACI 318-71 directly applies and for which no special O Page l of 5 i
3 30,-lep;fy i,5 p u, peg , g (E3J,,m,T,p Q y,,TEL NO:215-841-4578 #213 P07 SZEE t, WEBSTER ENGINEERING CORERA7:ON I LIMERICK 2 IDCA j/t f V (J.O. NO.18138 ) Y be l}}