ML20199H233
| ML20199H233 | |
| Person / Time | |
|---|---|
| Site: | San Onofre  |
| Issue date: | 05/31/1986 |
| From: | Russell M EG&G IDAHO, INC. |
| To: | NRC |
| Shared Package | |
| ML13324A929 | List: |
| References | |
| CON-FIN-A-6808 NUDOCS 8607030233 | |
| Download: ML20199H233 (66) | |
Text
~~
.:: L~ , _ u. __
, - _ . _ . ~ . - .=- __- - n.. .
og 3
. t...
T~
Attachment 1 i
J TECHNICAL EVALUATION REPORT.F0il THE SAN ONOFRE NUCLEAR GENERATING STATION UNIT 1 SEISMIC CRITERIA FOR LONG TERM SERVICE FOR PIPING, MECHANICAL EQUIPMENT, ELECTRICAL EQUIPMENT AND RACEWAYS AND ASSOCIATED SUPPORTS 4
1 l
! . M. J. Russell i
May 1936 EGLG Idaho, Inc.
Idano .: alls, Idahe 83415 i
I Prepared for the U.S. Nuclear Regulatory Commission Washington, D.C. 20555 Under CCE Contract No. DE-AC07-76IO01570 i FIN No. A6808 0
. q,6Y~,4/f
~
00 s
mu ..i. a =._we L m w u re .
- r. .v ~_- - r--. - - . ~ m - L e = = - ~.m u .
.n V
SUMMARY
The initial USNRC Safety Evaluation Report (SER) c:vering the Leng Term Service (LTS) Criteria and Methodology for the San Oncfre Nuclear Generating Station Unit 1 (SONGS 1) was published with a number of
{ unresolved issues pending. These involved some of the criteria and methods I for piping and piping supcorts, electrical raceway supports and electrical
! e:uipment su:: Orts. All the SER issues within the sc::e of this c:cument nave ceen resolve:, with :ne results presentec here. Three c:her issues are discussec. The first, concerning electrical equipment, was included in the LTS review by the Return to Service review of SONGS 1. The second,
- n:e-n' g ice su::crt :en: rete an:hcr b:10 criter'a, was generated dari ;
} tre LTS calculati:n audits. The thire, con:erning :i;e su: port trunr.icns 4
atta:nec to elbows, was generated curing the Systematic Evaluation Program (SEP) review concu::ed price to the beginning of the Return :: Service i
(RTS) program. All of these issues were resolved.
i l
4 l
I i
4 i
FIN No. A6808 11 .
g p. ,.@ x m ,gg.W y .;.%c-M c M 6 % %. . ca .;; ..w nq sk.%dwAMun A%Bm2.e : -_a .
e -
g f
CONTENTS -
SUMMARY
............................................................... 11
- 1. INTRODUCTION ..................................................... I
- 2. RESOLUTION OF ISSUES ............................................. 3 2.1 Large-Bore Pipi ng Strain Is sues ( 3.1.1) . . . . . . . . . . . . . . . . . . . . 3 2.2 Small-Eore Piping Confirmatory Analysis Review (3.2) . . . . . . . 6 2.3 Pi pe Support Weld Cri teri a ( 3. 3.4) . . . . . . . . . . . . . . . . . . . . . . . . . 6 2.4 Electrical Raceway Analysis Methods and Criteria (5.) ...... 7 2.5 Elect.-ical Ecui: ment Analysi s Matheds and Cri eria . . . . . . . . . E 2.6 Pipe Support Concrete Anchor Bolt Criteria ................. 5 2.7 Pipe Sup: ort Trunnions Attached to Elbows .... ............. 1C
- 3. CCNCLUSIONS ...................................................... 12 1
.*. AOr"M".lw..
.., ;d ....................................................... .0 l
ApDEN0!X ----
A--TECHNICAL EVALUATION OF, THE. STRAIN EASED METHODO.0GY exurCScu- av, w . .Rn,, cm. L.
.e 0.J.. n: .rv- R.,iIn, tu,i d ..u.,I c.m.. r---u n ,., Sc. ,2.y.
f SEISMICALLY QUALIFYING THE LARGE-50RE PIPING OF THE 2 ..n O N. s.-- n:N.J C.:3s , . . . G : .gi:n- n.i .t.
. ,4u, S.initw
. . . c N U N ti
. 1 r.. .um LCNb" TERM SERVICE ................................................ k-I
- a.m a t, p n.O g .4w ;u n t. X O,.. I a Rc.c. = e a tu s. P it P t.i4c .e d i sn.J R. .T
. ge t35 A e n.RR..q w g .o.n v en is .e.NiC e. . . . . . . . . . . c . t.
n..r *. u tiwAA == pa ; W m . ., e."**Sw.w.q
' ^ It 4w Farc J'wffWas m.urgmeiw.-at-M a k. t. J1L .516;Mio ....
1
.i I
tii
mee.mmee nw-n-ann-: m-~ce : emn n 2 _e : ~m ~ n- ~ aut- ~ v-t ' I 1
I I
TECHNICAL EVALUATION REPORT FOR THE SAN ONOFRE NUCLEAR .
GENERATING STATION UNIT 1 SEISMIC CRITERIA FOR LONG TERM SERVICE FOR PIPING, MECHANICAL EQUIPMENT. ELECTRICAL ECUIPMENT AND RACEWAYS. AND ASSOCIATED SUPPORTS
- 1. INTRODUCTION In Novemoer of 1934, San Onofre Nuclear Generating Station, Unit 1 (SONGS 1) resumed power operation in accordance with specified terms (Reference 1), the terms being implementation of the Return to Service (RTS) Flan. Diant operatten was autnorized proviced trat a seismic evaluation program ce completed and resulting plan cocifications be imolemented prior to start-up from the the cycle IX refueling outage. The Long Term Service (LTS) Pian is intenceo to demonstrate capability to reacn and maintain a cold shutdcwn condition, including capacility for accicent mitigation, in the event of the reevaluation basis earthquake, the 0.67 g mocifisc housner eartacuake.
Scu:nern California Ecison Co. (SCE), :ne licensee, has fermula ec tne LTS ;ian and is new seeking a:croval of the plan frem ne Nuclear Regulatory Commission (NRC). Two meetings were held (in Rosemeac, California on February 12, 1955 anc Ee:hesda, Maryland on Fecruary 27, 1955) at which SCE : resented a su.T ary description of the LTS clan t the NRC staff anc their consultants. Inis was followec cy a formal sucmitta'.
of :ne LTS Flan (Reference 2). As a result of the meetings and review of the sucmittal, several issues were icentified where the criteria and methodolcgy differ from current incustry practice as excressed in :ne Stancard Review Plan (NUREG-CSCO, Reference 5), the NRC Regula cry Guices, the criteria developed as a result of the NRC Systematic Evaluation Program (SEP, References 4 and 5), and LTS requirements develeced during the RTS review process (Reference 6). Issues associated with the proposed criteria and methodology were identified and discussed in Reference 7.
1
___ ~..+ :e w: :=. ww.c v ,;+. x: w :- w o x xx =- a ~ -
,- 6
~
This was followed by a series of submittals.and meetings dedicatec to the resolution of the issues (References 8 through 15). The results of this process were reported in a Safety Evaluation Report (the Septemoer, 1985 SER, Reference 16). This SER was published before all of the issues could be fully resolved in order to allow the analyses subject to tne criteria and methodology to begin. The resolution of some of the issues identified in the September,1985 SER as incomplete is the topic of the sections which fellow in this re; ort. These se:tions include a cross-reference to ne pertinent sections of this SER as an aic in tracing issues. Some of the unresolved issues identified in the Septemoer, 1985 SER are not discussed here because they are beyond the scope of this work.
Their resclu icn is discussec in 0:ner Te:hnical Evaluation Re:Or:s 'ssued by c:her NRC ::nsultants. In aedition, all audit and case-by-case review work done to satisfy :ne Septem:er, 1985 SER recuirements has been includec in an auci report puolished in the same time frame as this report. These topics are not ciscussed here. Three accisional issues ciscussec here are not related to tne September, 1985 SER. The first, concerning electrical
- ecut; men; anc an:ncrages, oas raised in :ne SCN35 1 RTS .o :rgram SER (Re'eren:e 1). The secenc, : n=erning pice support ::n rete an:nce bel; criteria, was genera:ec curing tne aucit of LTS calculations. The :nied,
- rcerada; p4:e support trunnions attacr.ed to elbcws, was generatec curirg
2
MirN.' #3&KL4*AR%Xup!MMMMWh@%G%WD?%tt ??@&NMMT3;WGW :V Q-e 2 .- RESOLUTION OF ISSUES 2.1 Larce-Bore pioine Strain Issues (3.1.1)
Section 3.1.1.2 of the September 1985 SER made the acceptability of the equation proposed for the calculation of strain used in evaluating large-bore pipe contingent upon its being consistent with Section NE of the ASME Coce. A comparison of the equation, as defined in Reference 17, with the Section NE requirements was made. The results of the comparison, wnich are presented in Appendix A to this report, indicate that the equation is consistent with Section NS of the ASME Code, and is acceptacle on that
- asis.
Section 3.1.1.1 of the September, 1955 SER identifies five failure mechanisms which must be addressed if the criteria allows strains between 1% and 2%. These mechanisms are identical to those identified in Volume 2 of NUREG-1061 (Reference 18) as the failure mechanisms which nonlinear 010'9; analysis should ensure against. Iney are:
- 1. The onset of clastic tensile instability
- 2. Lcw-cycle fatigue or plastic rateneting
- 3. The onset Of local or system buckling
- c. Excessive ceformation (resulting in more tnan a 15% recu:ti:n, er less as recuired by system performance, in cross-sectional fl:w area)
- 5. Functional failure of pipe-mounted equipment.
In response to these requirements, the licensee added two additional checks (Reference 17) to the equation evaluated in Appendix A. The local buckling concern was addressed by an additional limitation on the calculated strain. The check limited the strain to less than one-fifth the pipewall thickness to mean pipe radius ratio. This limitation is a l
l 3 l
t l
l i consultant recommendation found in NUREG-1061 (Reference 18). The j low-cycle fatigue concern was addressed by a fatigue check based on Markl's
} correlations on moment loading fatigue tests, which are also the basis for I. the ASME Class 2/3 stress intensification factors. This check involves a l calculation of the seismic usage factor using Markl's correlation with the l
! number of cycles for the earthquake set at five. The calculated usage
! factor would be limited to 0.25. The usage factor was found to be ;
reasonable, but the number of earthquake cycles was questioned. l l
} Section 3.9.2 of NUREG-0800 (Reference 3) requires ten maximum stress cycles in the absence of an analysis of a synthetic time history used for system analysis to provide a basis for fewer cycles. The licensee provided i
i such an analysis in Reference 17. This was reviewed from the viewpoint thac it should be a production analysis and four differences were noted:
l
] The analysis was one-dimensional, had a four second duration, was done j using a time history corresponding to the 0.67 modified Housner ground
} spectrum with input accelerations increased by a factor of 7.85. Each of 1
j these differences was conservatively evaluated in terms of the change in
- input acceleration required to compensate for it. Assuming two equal
{ l
] horizontal and a vertical input at 2/3 the horizontal, the one-dimensional
! input needed to be 1.56 times the three-dimensional horizontal input to obtain equivalent response. Considering the fifth power relationship i between stress and the number of cycles to failure in the Mark 1 correlation, the input acceleration needed to be increased by a factor of l l
! 1.20 to obtain in 4 seconds what would be obtained in 10 seconds without j the factor (10 seconds is the minimum NUREG-0800 duration). Using peak f
j spectral acceleration as a basis, the input ground spectrum needed to be increased by a factor of 1.73 to obtain a response equivalent to that for l
j the most severe in-structure spectrum calculated for SONGS 1. Since the l product of these three factors, 3.24, is considerably less than the 7.85
! increase in input acceleration actually used, the analysis done is !
I conservative in comparison to a production analysis. Therefore this I analysis can be accepted as a basis for using 5 peak stress cycles in the i fatigue check, per the requirements of Section 3.9.2 of NUREG-0800. The concern about functional failure of pipe mounted equipment is alleviated !
because the methodology is not proposed for use with the nozzles of j mechanical equipment, which includes all line mounted equipment. Stress I
l 1 4 1 ;
SY4fMMEM5fWif*IMEM&M5MW2M!5.VdT*T4EM@jMC'4Jr Et ' ' #K9M/CM-_%?MM%.
- levels for the noz:le's are limited to the SEP allowable stresses. This limitation on stress would also implicitly limit accelerations imposed on the equipment, so that the loadings on line mounted equipment would not be increased as a result of the strain criteria. The remaining concerns are addressed by'the limitation on strain represented by the original strain criteria. The test result's indicated in Reference 17 demonstrate that piping systems will not. buckle, suffer plastic tensile instability, or I
suffer excessive deformation under dynamic loads at the strain criteria limit. The strain criteria, supported by the two acditional checks, meet the September, 1935 SER requirement that the five failure mechanisms be prevented.
Ine final aspect of the strain criteria discussed here is not explicitly discussed in the September, 1985 SER. These criteria have been evaluated as nonlinear from the regulatory viewpoint. Nonlinear criteria are not controlled to as fine a level of detail by regulation as are linear criteria. Therefore, the evaluatien of nenlinear criteria decends more heavily en a concensus process. To succort this prc:ess an acditional review was performed by ancther NRC consultant, the results of wnien are cresented in Acpendix 3. These results indicated a need to provice a:ditional censervatism in the calculation of support, flange and valve Ioacs anc system displacements using a linear analysis methodology applied
'- the range of plasti: "esponse between 1% and 2's strain. The licensee ,
i agreec te provide acditionai conservatism for system discla:emen s in Reference 17. This was cene, and the results were audited and fcune a::eptable. The require. ent for additional ::nservatism for su: pert leacs was incluced in the cr'teria used for case-by-case reviews of ne strain a:plicati:ns b: ween 1% and 2% performed by the au ner of this re:c t.
This recuirement has been mot. Results of the aucit and case- y-:ase reviews are ::ntained in an audit report published in conjunction with tnis report.
The licensee has modified the strain criteria and methodology to meet the requirements of the September, 1985 SER, and these criteria and methode*ogy are accertable..
l 5 l
.5#%%;,b#%MiffRRTM:@D!M%WSMdde#3in-?;gpMW3tG3MX ~\:.h i' ?$Rhhh $l t
2.2 Small-Bore Picine Confirmatory Analysis Review (3.2)
Section 3.2 of the Septem:er, 1985 SER m:kes the final acceptance of the small-bore piping criteria and methodology contingent upon acceptability of a confirmatory analysis provided in Reference 11. Review of the analysis shewed that it contained analyses of all of the permutations of piping geometry included in the criteria, and a range of pipe si:es representative of the population of small-bore pipe si:es in SONGS 1. A finite element analysis, per the large-bore criteria and methodology was performed. Stress results from the finite element analysis were ccm ared to stresses calculated using the small-bore criteria and methecciogy. The comcarisons cemonstratec :ne acceptacility of the small-bore piping methodology.
One change was made in the criteria. The original allowable stresses were written in terms of the RTS large-bore piping allowable stresses.
These were revised to correspond to LTS allowable stresses. Based on the
'favoracle compar4sen between small-bore and large-bore (finite element) meineds Of calculating stress, this is acceptable. The Se: tem:er, 1955 SER requirements have been me:.
2.3 Pi:e Sue ort Weic Criteria (3.3.1)
Section 3.3.4 ef :ne Se::em:er, 1955 SER makes acceptance of :ne cipe succort weld criteria contingent ucon the licensee croviding reasonable assurance :na: :ne weld material strength is greater than or ecual to :nat of the base metal.
An associated issue was raisec during discussien of the pice succor:
weld criteria. This issue related to the possibility that brittle weids may have been created in stainless steel piping because of the use of either submerged arc or flux core welding on the piping. Both of these issues were resolved in References 19 and 20. The Procedure Qualific:-tion Record for A-36 steel pipe support welds presented in Reference 20 a demonstrates that,the weld metal is stronger than the base metal. In Reference 19, a list of all piping welds proceduras used at SONGS 1 is 6
MsXitMC*r 22iG*; MM ;WW2 LE%W%2hW2M"MMMMMP~dWM2WtW5 FMG.1[
e presented, along with a designation description that identifies the type of 4 , weld.used in each procedure. This demonstrated that neither submerged are nor flux core welding was used on any SONGS 1 piping. The September, 1985 SER requirement and associated requirements have ceen met.
2.4 Electrical Raceway Analysis Methods and Criteria (5.)
As noted in Section 5 of tne Sectember, 1985 SER, the review of electrical raceway and support analysis methods and criteria (proviced in Reference 12) was incomplete at the time of its publication. This review has been completed, with the following results.
Review of the raceway support evaluation criteria and metnods yielced I
the fcilowing. Allowable stresses are consistent with the provisions of Section 3.9.3 of NUREG-0800 (Reference 3), which are invoked by Section 3.10 of NUREG-0800. Although Section 3.10 pertains to mechanical ,
and electrical equipment only, it is applicable to the review of raceway sup orts because of the similarity between suen succor:s and some of the electrical ecuipment supports coverec by Section 3.10. Concrete anchor bel:_ allowable stresses are basec on a facter of safety of 4.0 for wedge
- yce anc 5.0 f: snell type anchors. This is consisten: with current industry practice, as expressed in IE-Sulietin 79-02 (Reference 21), and is acce: table. The static and dynamic analysis methods are acce: able based
- n correspondence :: Regulatory 3; ice (R.G.) 1.100, ar ancillary cocument to Section 3.10 of NUREG-0800. Cam:ing values of 7% for conduit and 15%
for cable trays were proposed. R.G. 1.61 cefines an acceptacle damping value for boitec structures of 7%. Since concuit su:perts are bolted, the 7% damping value is accepta:le. R.G. 1.61 also allows :ne use of damping data higher tnan specifiec, creviced the data used is supportec by documented test cata. A substantial test program was undertaken, as documented in Reference 12, unich demonstrated that a 15% damping value for cable trays is acceptable.
The adequacy of the raceways is ensured by the application of a 4 in.
. lateral deflection limit. This criterion is based on the test results presented in Reference 12. The test cor figuration of most interest here I.
4 7
-r --
consisted of a four tier cantilevered support raceway system typical of the raceway systems found at SONGS 1. The first three layers were cable trays, with the bottom layer (furthest from the support base) consisting of multiple conduits. Realistic and severe dynamic tests were performed, with resulting relative deflections above 4 in. occurring. Although there was localized plastic response, buckling of the cable tray side rails, the trays did not collapse. Static relative-deflections of up to 5 in were also applied, with the same results. Continuity testing was performed before, during and after the raceway system testing, with no loss of function (continuity loss or shorting) observed. These test results provide an adequate basis for the raceway deflection criteria.
The review results indicate that the electrical raceway criteria and methods are reasonably consistent with current industry practice, and are acceptable on that basis.
2.5 Electrical Equipment Methods and Criteria As noted in the Return to Service (RTS) SER (Reference 1),
modifications to electrical equipment supports had been completed prior to the initiation of the RTS program. It further stated that detailed review of methods and criteria would for analyzing equipment and supports be completed under the Systematic Evaluation Program (SEP) seismic review.
The review of criteria and methods for equipment anchorage has been performed using the same regulatory guidance as used for electrical raceway supports. Allowable stresses and analysis methods are consistent with Section 3.10 of NUREG-0800. Concrete anchor bolt criteria (4.0/5.0 Factor of Safety) are acceptable per IE Bulletin 79-02. Damping values, 4% for welded and 7% for bolted anchorages, are consistent with Regulatory a Guide 1.61. The review of criteria and methods for the equipment will be incorporated in the Unresolved Safety Issue (USI) A-46 Program., as discussed in Reference 20. The equipment anchorage criteria and methods are reasonably consistent with current industry practice, and are acceptable on that basis.
8
~-
.u - &nspd:.c@hn _
2&Bmc esw-a = .d:Q&a <~&wknStum&c
- o. *.
2.6 pipe Succort Concrete Anchor Bolt Criteria Aucits of calculations involving application of pipe support concrete anchor bolt criteria resulted in a difference of opinion betwen the auditor and the licensee in interpreting the criteria. To resolve this problem, the licensee proposed a revision to the criteria. This was presented at a meeting held October 8-11, 1985. The handouts pertaining to this proposal are attached in Accendix C. A Factor of Safety (FCS) of 2 woulc be acplied, provided that:
- 1. The base plate in question has at least four anchor bolts, and i
not more nan half are simultaneously subjectec to tensile leads l
l l 2. Loads greater than :nat associated with FOS = 4 are redistributed J
- o adjacent supports
- 3. Adjacent supports remain elastic and, if appropriate, meet the FOS = 4 under the new loac 4
The ancnor bolts meet visual and ultrasonic (for ecceccent) recuirements.
This crocosal was suoported by the following evidence. Recent analysis of anchor bolt test data ey the Seismic 0 alification Utility Grouc (SCUG) precictec a failure rate of 1% for FOS = 2. Tests of anchor bolts in SCNGS 1 concrete all showed slip (noncatastrophic) type failures. The SONGS 1 test results, cocumented in Reference 22, are significant because
- ney show that concrete anchor bolts are capable of carrying limitec load following the onset of slippage. Following the proposal, a review of the basis for the FOS = 4 criteria was uncertaken. It was found (Reference 23) that approximately half of the FOS of ? was specified to accommodate two effects associated with seismic loads: degradation (cracking) of concrete due to the earthquake, and degradation of anchor bolt response due to dynamic loading relative to that of static loading. In addition, no consideration was given for the potential of the associated piping system to accommodate degraded performance of isolated supports. Based on this, o
9
QNof tc.. u .;;i .- c 4%.; ,Hgfgds!.w -
,.9 + @ q v 0;ty & M R ; g _ p r'. g e g P 2 % e s v i # b the proposed criteria were accepted. A case-by-case review was performed to establish that either (a) the associated piping and supports have sufficient margin to compensate for degradation of performance of the subject support, or (b) the support is located in an area where significant cracking of the concrete is not likely. This case-by-case review has been performed, with the results documented in the audit report discussed in Section 1.0 of this report. The review established that at least one of tne criteria (a) and (b) above was met in every case, and that the applica:icn of the F05 = 2 criteria was acceptable in every case.
2.7 Pice Su: cert Trunnions Attached to Elbows One of the issues raised during the Systematic Review Program (SEP) review concerned the use of pipe support trunnions welded to elbows (References 24, 25, and 26). The proolem witn this practice is that neitner the ANSI nor the ASME piping codes contain stress intensification factors (SIF) and flexibility facters for such components. The licensee
'crocesec me:neds for calculating the SIF and accounting for lack of a cefinec fiexibility f actor in a letter (Reference 25). These methces were found acceptacle in an NRC meeting summary (Reference 26), excect for one ce:adi. The licensee hac commit:ec :: removing all rectangular trunniens, anc all circular trunniens with a diameter less than one-half that of the elbcw. The NRC meeting summary indicated that all circular trunniens with ciameter creater than ene-naif tne elbcw ciameter should be emovec. Tne issue has since been reviewec in terms of this difference of coinien, with the conclusion that the preccsal to remove all circular trunniens of diameter less than one-half that of the elbcw, as stated in the licensee's letter, is correct. One assumoticn of the piping ccdes is that cending mcments predominate over axial leadings in piping ccmoonents. The codes make note of this assumption, and place responsibility on the piping analyst to ensure that it is correct. In the particular case of a pipe support trunnion attached to an elbow, this assumption may be questionable where the elbow diameter is significantly larger than the trunnion
! diameter. This is because the axial load in the trunnion is controlled by the code limitation placed on bending moment developed in adjacent piping l 10
m_
y- -
v*-, - ~--s
- xx.s: .. , "
-: '= v 2r =wr : =u vn.: -
c :. :w mww= m-m. cm .~= w. w. u components as a result of the axial icad. If those associated components -
are of a significantly larger diameter, a disproportionate axial load may be developec in the trunnion. This could result in a " punch through" type of failure in the elbow. Maintaining comparable trunnion and elbow diameters would prevent this. type of failure. Therefore, the proposal made in the licensee's letter (Reference 25), which was confirmed to be completed in a later letter (Reference 27), is acceptable.
e B
11
iN-W,!hyn$$;;.' . 76.4_@fg:K*dlpi..~;
.-' . - t.,. . . gsi@@i%tydith).};, _ " ' . 04.3NNKNWild?OE
.. s e *
- 3. CONCLUSION All issues in the scope of this report have been resolved. The criteria and methods for piping and piping supports, electrical raceway anc equipment, and supports have been modified as necessary and reviewed sufficiently to meet the requirements of the current LTS and previous SERs.
4 l
1 l
1 i
12 l
g ... n a u ,.w. w e e u w ,- m .:: ,._ . g . x._ ,. a , _, .. . . m w. ;w .
0 0
- 4. REFERENCES
- 1. Letter, D. G. Eisenhut (NRC) to K. P. Baskin (SCE), dated Novem er 21, 1984, transmitting Contingent Rescission of Suspension anc Su: porting Safety Evaluation Report.
- 2. Letter, M. O. Medford (SCE) to J. A. Zwolinski (NRC), dated March 12, 1985,
Subject:
Long Term Seismic Criteria and Methodology.
- 3. NUREG-0800, "Stancard Review Plan for the Review of Safety Analysis Re: orts for Nuclear Power Plants, LWR Editien," July 1951.
?. Letter, W. Paulsen (NRC) to R. Die:ch (SCE), dated Septemoer 20, 1982, Suoject: Staff Guicelines for Seismic Evaluation Criteria for SEP Grouc II Plants.
- 5. NUREG/CR-0093, "Ceve'c; men: Of Criteria f:r Seismic Rev'ew of Seie:ted Nuclear Pcwer lants," May 1978.
- 6. Letter, H. R. Denten (NRC) to K. Baskin (SCE), cated Fecruary 8,1984, Suoject: Procesec Restar: Plan for San Oncfre Nuclear Generating Station, Uni: No. 1.
- 7. Letter, D. Crutchfield (NRC) to K. Baskin (SCE), dated March 27, 1985, Su: ject: Seism'c Criteria anc Me: hec 01 gy - San Onofre Nuclea-Generating Station, Unit 1.
S. Summary of A:ril 2-5, 1935 Meeting, E. McKenna/T. Cheng u C. Grimes, ca ec Acril 15, 1985.
- 9. Le::er, M. C. Mecferd (SCE) to J. A. Zwolinski (NRC) dated April 15, 1955, transmitting " Responses to 0:en Items frem the April 2-3 Mee: ng and the Methcdology Test Proclems."
- 10. Summary of A:ril 29-3C,1955 Mee:ir.g, C. Grimes to 0. Crut nfielc, catec June 12, 1935.
- 11. Le:ter, M. O. Mecfore (SCE) to J. A. Zwolinski (NRC) datec Acril 30, 1955, transmitting " Review and Development of Small Scre Pi:'ng anc Tucing Criteria" and " Evaluation of the Refuel'ng Water S:crace Tank f:r Long-Ter: Service."
- 12. Letter, M. O. Mecford (SCE) to J. A. Zwoiinski (NRC) cated May 14, 1985, transmitting reference documents.
13 Letter, M. O. Medford (SCE) to J. A. Zwolinski (NRC) dated June 4, 1985, transmitting 1) Long Term Service Status Report, 2) Technical Basis for Piping Strain Limits and Develo ment of Linear Elastic Analysis Methodology, 3) Revised Test Problems and 4) Recert on Studies of Soil Modulus and Damping.
mm .. p.x .c _
o e .
14 Summary of June 10-11, 1985 Meeting, C. Grimes to D. Crutchfield, dated July 11, 1985.
- 15. Summary of July 1-2, 1955 Meeting, T. Cheng to C. Grimes, dated July 22, 1985.
- 16. Letter, H. Thempson (NRC) to K. P. Easkin (SCE), dated September 19, 1985,
Subject:
Long Term Service (LTS) Seismic Criteria and Methodology - San Onofre Nuclear Generating Station, Unit 1.
- 17. Letter, M. O. Medford (SCE) to G. E. Lear (NRC), dated March 25, 1986, Sucject: Occke: No. 50-206, SEP Tooic III-6, Seismic Design Considerations, Technical Basis for Stress-Strain Correlation, San Onofre Nuclear Generating Station, Unit 1.
- 18. NUREG-1061, Volumes 2 and 4, " Report of the U.S. Nuclear Regulatory Commission Piping Review Committee," Acril 1985 and Decemoer 1954 (respectively).
- 19. Letter, M. O. Medford (SCE) to G. E. Lear (NRC), dated April 16, 1986,
Subject:
Occket No. 50-206, SEP Topic III-6, Seismic Design Consicerations, Welding Procedures, San Onofre Nuclear Generating Station, Unit 1.
- 20. Letter, M. O. Medford (SCE) to G. E. Lear (NRC), dated May 19, 1986,
Subject:
Occket No. 50-206, SEP Tootc III-6, Seismic Cesign Considerations, San Cnofre Nuclear Generating Station, Uni: 1.
- 21. USNRC IE Eu11etin 79-02, "Cice Succor: Ease Plate Designs Using Concrete Expansion Anchor Bolts," Revision 2, Novemcer 3,1979.
- 22. Letter, D. F. Kirsch (NRC) to K. P. Easkin (SCE) dated Decemcer 7, 1934,
Subject:
NRC Inspection of San Onofre Units 1, 2, and 3.
- 23. Memorancum, J. P. Knight (NRC) to E. L. Jordan (NRC), ca ed Fecruary 15, 1950, Suoject: IE Eulietin 79-02, "?i:e Suo:or-Easemiate Designs Using Concrete Excansion Ancnor Bolts"--?ac crs of Safety.
24 Memorancum, W. T. Russell (NRC) F. J. Miraglia (NRC), dated Decemoer 29, 1982,
Subject:
Summary of Decemoer 1-3, 1982 Meeting, SE? Taoic III-6, " Seismic Design Consicerations."
- 25. Letter, M. O. Medford (SCE) to D. M. Crutchfielc (NRC), dated I Novemoer 21, 1983,
Subject:
Docket No. 50-206, SEP Topic III-6, Seismic Design Considerations, San Onofre Nuclear Genera-ing Station, Unit 1.
- 26. Memorandum, W. T. Russell (NRC) to F. J. Miraglia (NRC), dated July 7, 1983,
Subject:
Summary of May 2-4, 1983 Meeting - SEP Topic III-6,
" Seismic Design Considerations," San Onofre Nuclear Generating Station, Unit 1.
14 ,
J
SMPMdiSML:;;;"' :7ET4II.MSIf3Ei@7d?!35!:li@!6fJ:R6fMIAIf.T3];I~ ) TREGRWLi@$$Mb 1
- 27. Letter, M. O. Medford (SCE) to D. M. Crutchfield, dated June 14, 1984, I
Subject:
Docket No. 50-206, Criteria for Seismic Reevaluation of Pipe Supports and Mechanical Equipment, San Onofre Nuclear Generating Station, Unit 1.
l
. O w
e 15
?MM$$%%4, Gr. i z. :.d*MpkM.%2R :-MSMdM9es:?%:9?? Zh MA hSPY APPENDIX A TECHNICAL EVALUATION OF THE STRAIN BASED METHODOLOGY PROPOSED BY SCUTHERN CALIFORNIA EDISCN CO.
FOR USE IN SEISMICALLY QUALIFYING THE LARGE-BORE PIPING 0F THE SAN ONOFRE NUCLEAR GENERATING STATION UNIT 1 FOR LONG-TERM SERVICE A-1 r
m , v x w.c:. w w n u m m r. v- - ..~ m .n w ,- c -~ w.m.-u e =m- u. _. ,= w a ,vu-w <
- e APPENDIX A
, TECHNICAL EVALUATION OF THE STRAIN BASED METHODOLOGY PROPOSED SY SCUTHERN CALIFORNIA EDISON CO.
FOR USE IN SEISMICALLY QUALIFYING THE LARGE-BORE PIPING OF THE SAN ONOFRE NUCLEAR GENERATING STATION UNIT 1 FOR LONG-TERM SERVICE Southern California Edison Company (SCE) has paoposed the followin; me:nocology for calculating strains o be used in conjun -ion wi-h the large-bore strain criteria of the San Onofre Nuclear Generating Station Unit 1 (Reference 1 on page A-9):
Calculate a total strair according to:
e total = Ks K c /E m t (1) where p0 e- =
q v.e: i (Ma - Mo)
(2) e is a Class 2/3 s ress basec on :ne 1950 Edit. ion, Winter 1930 Accenca of the ASME Eciler anc Pressure Vessei Ccde (ASME Code). All terms are cefined there.
K = 1.0 for 3.4 c s Sy s (2) 1.0 I1 ~ ' ' #
- 1.0 for Sy < 3.a og s mSy
,q , _ l
= 1/n for mSy < 3.4 c g K3 is similar to K, as defined in Section NS-3653.6 of the 1983 Edition, Winter 1984 Addenda of the ASME Code.
The difference is that Sn/35m has been replaced by A-2
% . _ n. .~ .t~ -m . i .- = .. w -w a n , - x -. x:- - - ~ -+ > - - ~ - ~
= !
3.4 oE/3Y' where Sy is the ASME yield strength at temperature. Tacle NS-3228.5(b)-1 provides values of m and n.
K, = 1.0 Carbon Steel 2.0 Stainless Steel E = Young's Modulus at temperature.
The total strain is then ccmpared to a maximum allowable strain to determine acceptability. These strain limits are defined in Referen:e 2:
"*o 'er car:cn anc stainless steel, with an extensien to Zia for stainless steel alicwec, previce: tha accitional recuirements are met. The recuiremen .s are defined in Reference 2, as are the criteria and me:necoicgy applicacie to the calculation of the elastic stress, e . g Reference 2 also requires nat this methodology be consistent with Section NS of the ASME Code for acceptability.
Evaluation cf the pr0 posed methoccicgy wil' censis: Of an examina-fon Of ea:n term in Ecuation (1) above. Wnere pessible, the evaluation.will in:iace direc : m:aris:n to curren: ASME methedclogy per the Referen:e 2 recuirements. The first term to be examined is a= . Its suitability is based on the form of the criteria. Since it is an ASME Class 2/3 term, it re: resents a calculated princical stress. :: is more suitable nan :ne equivalent Class 1 term, wnich expresses orincical stress intensi y,' anc
- nerefore coes not correspont as well with the criteria. As an excression of primary plus bending s .ress, c g is =cre ap recriate than a peak stress expression because of the ASME practice of allowing local ciste -1:n in se,ere icw probability events such as eartnquakes. Tne 1 and 2%
allowable strain values specified also indicate that the primary plus secondary stress term is more appropriate than a peak stress term.
The next term to be evaluated is Ks . Its uncti n in Equation (1) is to correct the linear-elastically calculated stress, og , for plastic response. In this, it is similar to the K, term defined in NS-3653.6 of the ASME Code. Since these two terms are similar not only in function, but l also in form, the evaluation c/ K, will be based en a c = paris:n to A-3
W $kN!?b&$$ibS$!$5 SON 50lSb$?$$$$uMN!kND$ $5NNS5?SWE5NT t
K7 e
In the area of function, k differs s
from K in two ways. First, e
K, is used in evaluating thermal leads, which are secondary, wnile K s is used in evaluating seismic loads, which are defined as primary loacs in current ASME methodology. This is significant because the treatment of primary loads is more conservative than that for secondary loads. However, there is evidence that the K, term provides conservative results at the comconent level regardless of whether the loading is displacement or load controlled. The results presented in the figure in Reference 1 entitled
" Strain Concentration Effect: (Experimental)" indicate that K, is conservative for both load and displacement controlled leading. The Accencum to Volume 2 cf NUREG 1051 (Reference 3) contains the results of a detailed and extensive investigation into the effects of str ng-meti n earthcuakes on industrial (531) piping. Reference a cescribes the tes fng of a prototycical nuclear piping system to seismic acceleration levels well above -hese to wnicn it was designed. Ecth of tnese references indicate
- nat piping has seismic margin far in excess of that credited oy curren; me necclegy. One source of this margin could be cam:ing. As the Oi:ing system rescense increases cas: the 00 int wnere plastic response initfa es in the limiting ccm:enen s, :ne danping associa:ec with such res:0nse coulc come into play, limiting further increase in system resconse, anc leading
- a rapic attenua.icn of res;cnse tack .0 the elastic range. 5;ch a mecnanism woulc be intrinsic :: the piping material and woulc ce resent in all ci ing systems. This is consis ent with :ne field experience presen .ec in Reference 5 anc wi-h the tes results cresen ec in Referen:e 4 Note tha the rela-ively icw damping levels reported in Reference t were li-i ec te maximum acceleration levels asscciated with the Design 53E. This is im:Ortant :ecause tne anomaly in the test results is associatec with accelera-icn levels far in excess cf Design SSE levels. In:u accelerati:n levels mus- be sufficien-ly above the SSE levels c account for the kncwn margin in the current ASME methodology before observable damage (plastic deformation) can be expected. That such acceleration levels were applied wi'thout observed damage is anomalous.
1 1
l l
A-4_ . _ _
l
NYMihMSd%%4HMFdn!/MIEMcW9 MkkS$$'M@di: < J._.; .
~ EJtMSDF The only other mechanism which could explain these results, load redistribution resulting from plastic response, is associated with piping system configuration (geccetry and support location), and is therefore net necessarily present in all. piping systems. If this mechanism were the source of excess margin, it should have been identified by Reference 3.
Among the large number of piping systems surveyed, there must have been systems which could not respond via load redistribution, and supports which would have failec as a result of the load redistribution. That no such failures were icentified is clear evicence that material camping is the source of excess margin. This being the case, treatment of the seismic inertia leading as secondary is reasonable because the material damping cacability is irtrinsic to all 0 cing systems.
4 The second difference between the function of sK and K, is that K, is used to correct an elastic peak stress intensity for plastic response, while K s is used to correct a principal, primary plus bending stress. This difference can be decomposed into twc parts: the difference
- *
- etween stress intensity and principal stress, and the difference between ceak and :rimary plus bending s: ess. Stress intensity previces a more a:: urate cescriptien of three-cimensicnal stress states tnan prin:ipal stress, an: would therefore provide a =cre accurate preciction of plastic response. The cifference between peak and , primary plus bending stress will be examinec in terms Of the corresponding strains. The peak strain occurs at a Ocint in tne section, as :::osec :: :ne ;rimary plus bencing strain, wnicn is obtainec by an averaging crocess across the section. As the leading on :ne section increases through the ocin; of initial yielcing, tne ceak strain will experience a larger excursion from the linear-elastic reference than the crimary clus tencing strair, because cf the avera;ing.
As the loacing is further increased, tne excursion of :ne crimary clus bending strain frem the elastic reference can be expected to approach that of the peak strain, but because of the averaging process, it would not exceed the peak strain excursion. Based on these two arguments, the application of 'a peak stress intensity correction factor to a principal, primary plus bending stress is acceptable.
l i
M no * *
@ tm:4?,; %'a%%yFAW, p .-
h m c~e*WM%)t1Mit-70.
.m?.: m%. - - g'i!M. .W. - MMWam. . _
- 7: W- llF E i B =' m_.n.
M ~ _ m~~ .
In form, Ks differs from K, in the independent variables used to calculate it. K is a function of Class 2/3 parameters, while K, s
depends on Class 1 parameters.
The difference in form between Ks ""d K, will be evaluated based on the idea that, given the same component with the same moment loading, the Ks term should be equal or conservative relative to the K, term. Conservatism requires that K 2 K, s =
(4)
Figure 1 shows the functional form of Ks (and K,). As shown in Figure 1, Equation (d) will be met for a specific component with a specific leading if 3.4 cE , Sn (5)
Sy 35m utilizing the fact that all the terms are positive, Equation (5) can be rewritten as
-q
"'!' s1 (5)
( 'e - n : J -)
(nm/cy This ferm, with a ratic of calculated stress diviced by a ratic of allewable stress, is c'onvenient for the evaluation. Table 1 shews that, basec en the ccculation Of ciping ma arials used in the plant and their cperating temperature ranges, the ter: 35m/Sy varies frem 1.71 :c 2.72.
(Ne:e na. Sm should be based on cesign and not opera-ing temperature.
This was net cene due to a lack of data, but is essentially neglect'ng a conservatism cf the ASME Cede, and should result in a small errer). Tne ra fc Sn/3.4 c requires a cit more work. Frem N2-3653.1 of the W54 ASME Cece, .
Sn = C PoDo 1 2t
+C 2 Demi 21
+C 3 Eab ( 6 aaT -6Tbb)
(7)
A-6
'&%skd$E?dE3\@$bNY?$N$if$$$$N5%fiMWS$$$$EN@C'2A:5S5?TN!5I$$2% -~2 To avoid an involved and time-consuming population calculation, the assumption that the earthquake lead is strongly predominant is made. This allows:
- 1. Neglecting the pressure term
- 2. Neglecting the thermal term
- 3. Setting Mg = 2M [Mg is the Class 1 full-range moment. M is the Class 2/3 half-range moment. One implication of the assumption is that inclusion of the ceadweight load, M, in Equation (2), is negligible. This al= lows :ne simplification M=M a+ M,].e Note that Item 2 above is reasonable only if there are no severe thermal transients occurring during the earthquake. This is the situation durine normal operation. Also note that neglecting the deadweight load is
- reasonable because the proposed methodology limits this loac to ASME Level A allowable stresses. An acplication of :ne assump-ion to Ecaations (2) anc (7), in comoination with the fact tha:
Z = 2I/Co (5) results in v
3.?c = 3.c (0.75ij) (i) and S
n
=C 222M (10)
Division of Equation (10) by Equation (9) gives A-7 , _ _ _ _ _ _ . . ,
- WEMMWWF8EIE!M41W5MMMafiE9fMK4W]fMW MM%ERWMEM&M#dM
~
A Ub/2//86 WCB Doc 1047k Disk' Job 25119 Proof _,
djd 0' SCC Sn "2
3.4 ce= 0.751
- 11)
Table 2 presents the range of Sn/3.4 Ee calculated for a range of components allowed by the ASME Code. This is a conservative representation of the population of components in the plant. Eased on Table 2, a range of 0.59 .o 1.71 has ceen chosen for Sn/3.4 c . Recucers have been neglected in the range because of the disparity in maximum values. The very large ratio fer reducers reflects the large allowable variatien in gec=e:ry, and coes not represent be :0:ulation of reducer gec=e: ries ir.
One p'. ant. It is cer:ainly a :co conservative representation of :ne range of other component types. Based on :ne chosen range of the stress ratio, and noting sna: all ccmcinations of componen type and material are likely to occur in the plant, the range of (Sn/3.4 cE)/(3Sm/Sy) is 0.2 :: 1.0.
This shows the calculation of Ks is conservative in comparison to that of Ke*
The next fac:cr in Ecuation (1) to be examinec is K . This fac Or came a:0ut as a result of a reques :na: SCE confirm :ne methec:1 gy Oy an ap:lica icn of it to existing test results. When this was done, the ccmcarison for carben steel showed geoc results, but the stainless steel esults showed the :redicted strains :: be ics by a factor of 2. This 1,ed
- c :ne K term. (hete that tne EG&G expression of Ecuatien (1) is not icentical to SCE's, but is algebraically equivalent). Althougn this difference may come from :ne unavoida:le inclusion of peak strain effects in the ex:erimental results (strain gages are surface mounted). it c0uic also result from tne fact that tne definition of the yield stress for stainless steel, as compared to that for carbon steel, is more of an artificial construction. Stainless steel does not have as well-defined a yi. eld point as carbon steel. In any case, the correction does achieve the desired result.
Since the term-by-term examination of the proposed methodology was favorable, the methodology is consistent with section NB of the ASME Code, and does meet the requirements of Reference 2. It is acceptable.
fTM 4id@NMSKE @Mi$$fXA F}M &46SS N UNisisf@l@ E W ?'E & $:35hWiSAF@&$n M b REFERENCES"'
- 1. Letter, M. O. Medford (SCE) to G. E. Lear (NRC), datec March 25, 1986,
Subject:
Decket No. 50-206, SEP Tooic III-6, Seismic Design Considerations, Technical Basis for Stress-Strain Correlation, San Onofre Nuclear Generating Statican, Unit 1.
- 2. Letter, H. Thompson (NRC) TO K. P. Baskin (SCE), dated September 19, 1985,
Subject:
Long Term Service (LTS) Seismic Criteria and Metnocology - San Onofre Nuclear Generating Station, Unit.1.
- 3. U.S. Nuclear Regulatory Ccemission, Recort of the U.S. Nuclear Reculatory Commission Picine Review Committee, NUR5G-1061, Volume 2, Accencum, April 1985.
. U.S. Nuclear Regula:Ory Oc--ission, Studies: :ynamic Restense cf Pro:::ycical Dicinc Syste s, NUREG/CR-3593, August 1934 e
25119 A-9
eacemven=.wenx.ww y
--m-r=.mm:c~ . . . uan w = . ~x.n. c w.a wn~w,-
y . _ . . --
t +
3 . %" _
4 if 3.40s/S y = Sn /3S...
/
. then K, = K."
4/n i _______;________________
,A/ i' i . ,/
%gg
, N j
~
q r
/ i i b
\
K, or Ks / !
4 64 y g. _.
N
% T.
Ds
%+ N' 4
/
>6
,/
- s ,
/
l l
/
/
- .! ./
1.0
/
i <
. , ! .- /
l /
i
- N
.- . 1
\
1.0 m S" 3.4 0c -
or 3S m Sy _
Figure 1. Functional relationship: K, (Sn /3Sm ) & K, (3.40s/S y)
A-10 m, ire 85-1
.2:s w w.v; m- w w m , w w. m . M s y, w a c.s.,e..:
. . ;. ;ga,wz s.1. ,,. gu.g. .., .__ g.y, g ..;_;, . x.g.,,,,
o >
TABLE 1. VARIATION IN MATERIAL PROPERTIES FOR SONG 5 1 PIPING MATERIALS AND OPERATING TEMPERATURES Material T*F Sy Sm (35m/Sy)
A312 TP304L 100 25.0 16.7 2.00 200 21.3 16.7 2.35 TP304 100 30.0 20.0 2.00 575 18.5 16.7 2.71 TP316 100 30.0 20.0 2.00 570 19.1 17.3 2.72 A106 Gr3 and A53 Gr8 100 35.0 20.0 1.71 (using SA-106) 340 30.6 20.0 1.96 545 27.2 18.2 2.01 Total Range (1.71 - 2.72) e A ~1
- ewerewrwcs. xw enn:~ww.wmv =- ~ n~ .w=w =6+weeen1+EO.uGETM e > l TABLE 2. VARIATION IN STRESS PARAMETERS FOR TYPICAL PIPING COMPONENTS Range of 0.59C / max 2 ,
Comconent (1.0, 0.751) Pertinent ASME Code Sections Straight pipe 0.59 Table NS-3681(a)-1 Figure NC-3673.2(b)-1 Girth butt weld
- 2 0.237 in. 0.59 Table NR-36S1(a)-1 t < 0.237 in. 0.41 to 0.59 Figure NC-3673.2(b)-1 Girth fillet weld 0.78 to 1.24 Table NB-3681(a)-1 (socket weld) Figure NC-3673.2(b)-1 Figure NC-a427-1 Weicec transition 0.37 NS-3683.6 NC-3673.2 Recucers 1.51 to 4.30 NS-3633.6 Figure NC-3673.2(b)-1 LR elbows 1.71 NS-3683.7 Figure NC-3673.2(b)-1 Branc.9 connections 0.90 to 1.57 NS3633.8
("stu " tyce) Figure NC-3673.2(o)-1 Fo ged tees 1.18 to 1.57 NS-3683.9 Figure NC-3673.2(b)-1 Total range (excluding reducers) 0.59 to 1.71 --
- a. The 1953 Edition, Winter '34 Adcenda of the ASME Coce was used
- nroughout. This is not strictly correct, since (0.75i) shoulc be based on
- ne 1980 Edition, Winter '50 version. However, the error is negligicie.
. l 1
l I
1 A-12
n
_B% at__ .a,L5.a: ,a.-. ms. -
l-_..:_.
. i_r. .; ..
- r.s..~,. , & .~ m_. z.u.a ,.+..gs.p- _e?u. rv:s s. .r.rfi_wt::a...ir.,.,c.':
. _ ~ a _=. v. >. ~__ a Jn. wm:W;gt;:st.r:s l
APPENDIX B n,
b..Mut Own: l E...N,.z......
r' 1 e innin nli:.R..
C.. in n n cSPnN,-
Con- v u: N-- v:
e E-;
N - d ? 6 d t $ @ S$ N Ei5 $, N N O b k 8 h E d $N N 8 A D N hil N M ib N h D I N $ A N " N E
$. 0. habnbaugb baaocialu, $nc.
4625 CEMETERY ROAD e HILLIARD, OHIO 43025
.m 614/876-5719 .
Nove=ber 29, 1985 Dr. T. M. Cheng p u m 4e s _e.
- - - 4 , d4
~-s US Nucles: Regulatcry Cc 'ssics Vashington, DC 20555
Subject:
San Onofre U=it 1, Stress-Strain Cc= elation for Large Scre Piping Seis=ic A alysis, Meeting with Applicant, Sethesda, 11/26/85 Daar Tc=:
Frier to the 11/26/85 =ee ing, I reviewed docu=ents sent te =e by Marh
?.ussell (IOaG Idah:) in la a Cet:ber, 1985. In particular, I reviewed te Southe= Califc=ia Iiiscs Cc=pa y Report No. 01-0310-1459, ceteber 7, 1985, nsas C =fre Nuclear Generating Statics Unit 1, Leng Te= Se: rice Seis:10 Re-evaluatic Progra=, Teeb=ical Basis for Stress-Strai: C =elati =", herei e af ter refe=ed to as the Repert.
Se Repert, page 4, shevs alta=ative pathvays to evaluatics of pipi:g s/s-te: acceptance. My :=ents cc ce= the pathway " g < str ' ' d '
s", but de equatics fer calculating c, sheva c p.1 ef de Repert, and its defi iti::s, l
is an e s - -' a ' par ,of that pathvay. 2' ce that equa:1:n is a 1980 ASMI C:de, Secti: III, Eq. (9), I v'" refer to it as Eq. (9).
?-ic; a de 11/26/85 =ee.ing, I prepared de en:1: sed seve: pages cf :::es.
S ese vere discussed vith the Appli " -t at de 11/26/85 =eeting. It is =y = der-s A. ding that the Applicant cc--' tted a rev se the Report, as he dee=ed appre-priate, to respend to =y ce=ents. This letter is written to expa:d a. bit c:
=y notes.
Page 1 of =y notes gives ext =ples of hev =uch the Report strain 1'-4 ts vill increase the allowable y, as ce= pared to 1980 Code 4 7' ' ts. Fer carbon steel t
I u
' Mning. ..s y . Is~ 4p'pL2.@Mlit;;gW?ps;g :Mh;m:g-&sg ynn:.qu, wg in-the increase over Level D 'd dts is by a factor of about 1.5; fer stainless' steel by a facter cf about 2. With respect to the SIP 'd-d t of 1.S5. for Class 1 pi;-
3 ing, the increases are by facters of about 2 and 2.6 for carbe: and stai less steel, respectively. Because the Applicant proposes to use PVRC da: ping, the in-
! creases, as co= pared to prior seis ic a:alysis :sthods and r, 7'-d ts, =ight be another 3C". =cre.
'@d's these increases are substa tiAl and sh:uld be accc pa 'ed by appro-priate c:=sideratics of piping systa= effects, va dee -=' e proposed strais
. a:d ec.uivalent CF,-l' ' ts are a ste:. in the directie: =f .:: vi?' g ::re reliable piping syste s.
Page 2 of y : tes s"- --izes :y views == the ec diticus under whid the pro-posed 'd-'ta are acceptable fer evaluatic: of piping pressure boundaries, except bolted-flanged jcists. Page 7 ef =y :::es picks up c= :: -press =e- bcundary as-pects and bolted fla=ged jcists. Felleving are cc :ents c= te seve: ite:s c=
Fage 2 ef =y =ctes. ,
(1) Equati = (C), p.1 cf Repcrt, contains ? a:d M, A icad' gs which are differ-est ths: typical es.rthquake leads. I view the str='- -' ts as questi::able f:: P, M.A 1:adings and vant to pr: vide assurs=cs that ::st cf ef is due te M .
=
(2) I have : t -' tad this to envelop respense spectru= =eceds. In =y viev,
?',7.0 ds.= ping is appr:priate vid de ":ultilevel resp::se spectru: =ed:d".
1 (3) Page 3 of =y notes is inte=ded to show dat ce proposedev -li=its c uld '
lead to buck'd g (cc pressive vrinkli:g) u= der static =c=e:: 1cading and
=ight do so under earthquake leadi=g. Ms is not to say that buckling vill occur but rather, 1: our present state of k=culedge, it vould be prudent to
-it applicatic: of the high (,-li=ita to D t 6 50. As indicated c the ms
l e: -sw me:=w:m x. e:. L.x.,y;~.rm m eni- _-EAwlG=i h m -A ~n~uc- A
\ ,
_3-top of p. 5 of =y notes, the 'd 't which the Applicant proposes to use fer strai > 0.01, in fact, 7'-d ts D/t to 40 at E = 0.01.
(4) 2ese are intended to explicitly state what, I presu=e, is the intent of (5) the Applicant. Bob 3es ack brought up the point that the Applicant should review velding procedures used at SONGS-I to obtain assurance that the veld-
=ents, as well as base =aterials, are reasonably ductile.
(6) I brought up this ita= because the Report is not clear ce just what veuld be d: e. Se verbal resp se at the 11/26/85 =eeting vas dat they vculd 00:p17 vid the liC-36CO (Class 2 piping) cf the C0de; i.e., ei-ler put S.W s in Code Eq. (9) er in Code Eqs. (10) and (11). 'Jith the stra'- -'ts c:
Eq. (9), it v'" he less restrictive than Code Eqs. (10 and (11). I view putting SAM 8s i: Eq. (9), with the strad' -' ts, as acceptable for SC:iGS-1 evaluatic=s.
(7) Se Repert (p. 13) indicates that a fatigue check night be =ade whe: =i:g stra ' - -' ts greater than 0.01 fer s * ' ess steel. Se 1cuer p:::ica cf P. 5 cf y =ctes represents :7 quick leck at the questi::: Should a fatigue chech be - d- ave: f: S 6 0.01? Sis questics involves the ass =ed :.::ber of "significa:t" cycles anticipated frc: the earthquake. Se Report, p. is, states:
"... the ==ber of significant cycles for M dified Ecusser Ear quake (=) vd " be ass =ed 2 be 5. 21s value was re=== ended as a realistic esti= ate is (32."
Refere=ce 8 is NURIG 1C61, Vol. 2. Zacving the se=ewhat "eff-the-top-cf-the-head" basis for = 5, I rsised the questics (2)(c) ce p. 6 cf =y =ctes.
Hopefully, the Applicant vd" provide a bit =cre quantitative defense of a See folleving c p. 6 cf =y notes
. - _ - - _ _ . .B- 4 . . . _
j =_, ~~ a _
. . . a.m ,, - . :. m ,
~ ,
o ~ ~..:_.- c. :n ~ w n=
l . a
' ~
-b ,
4
= = 5. This does='t mean that I think = 3 5 is a bad assu=ptic: but what verries =e a bit is that for the Modified Ecusner Earthquake = 20 ight be a =cre defensible assu=ptic=. Oat assu=ptic=, cf course, is directly involved in =y questic: of whether a fatigue check should be :ade eve for d 4 0.01.
4 F = =y quick icek at the questic , I a: not sure of the answer. . Pre-su= ably, the Applica t v either refute de need for :y ite= (7) or is-clude it is sc:e for= 1: his revised Report.
Pere 3 ef rr retes Ois data quan ifies =y ec ce= abcut using te strad- ' ta for M 1 cads g
(ite: 1 et page 2) and de D/t 't (ite: 3 of page 2). It also is relevant to p. 7 cf =y =ctes.
? ace i of _
r netes I
'is data, quantifies =y cc::e= about using te stra'- ' ' ' :s f:: M A
1**
(ite: 1 := page 2) and is relevant to p. 7 cf =y actes.
Face 5 ef -v netes Oese have been previcusly discussed in ce=ecti = viS ite:s 3 and 7 c:
page 2 of :y ::tes.
?cre 6 ef v netes Ite= (1) has been discussed in ce=ection with ite: 6, page 2 cf :y ::tes.
Ita= (2), (a)-(d), are perhaps se=astics or editerial. Hevever, I read this pertic cf the Report as saying that aither the Optic: 1 cr Optics 2 check v'"
bo =ade. I think both are needed. Ita= (2)(c) was previously discussed in cen-ju=ctic: vith ite= (7) c= p. 2 of =y =ctes.
B-5
...w .-
. ,_ u , ,
w =
.: .x --
a . w ,:-_ _u ,~.-
.a.=.. ~
.x
. s.
l?
. ' SP Ite: (3) vas raised because the Report did not specifically address the question. The verbc response at tke 11/26/a5 =eeti:g was "Yes". ':hs.t is sur-ficient as far as I a:'cencerned. .
Ite: ~(4) stems back to the definitics of i c p.1 of the Report:
" i' r Stress intensification factor as listed in Fig. NC-3673.2(b)-1 of ASME SE V Code,Section I;I, Subsection NC, 1c80 Editi:n, Winter 1980 Addenda,"
Ihe potential proble= here is cat this may tie the Applicant's ha d to the ex-te=t Sat he r* not use any data er a:alyses which =ight shev that the C:de i-facters are "- eccessarily restrictive. I dea- "'s : t desirable frc: N?.C' s standpoint. I would suggest an additics to the defi=1 tics of "i" such as:
- "Any deviation fres Ceds-epecified 1-fact.
- rs vd" he addressed := a case-by-case basis ~ vita E C review and c::currence."
t~' Pese 7 ef -v .etes .
F1 st, le =e note dat I have act raised a new p ble:. Esc 1:sure 2 d. .is sc:e pc -icas fr:= N"?S 1C61, Vol. 2, in which the sa:e pr:ble. ' is discussed.
Since vriti=g that perti:= cf NUPm 1C61, I have attended a nu ber of =eetings
- Code revisiens to ::re read stica'y evaluate earthquake lead ' rs,1:0'M g (so far, in v='-) f:: the ansvers to the questi::s := p. 7 cf :/ ::tes. Indeed, in :y viev, if 0 had the anesers to these questi::s, the C de V:uld have been 6- red bj : v.
C: the bottc= cf p. 7 ef :y notes, I included a si=ple cenceptua' - ey ple of the possible pr:bles. It obvicusly is act a re=stic exa:ple. *devever, in U S j l
F.eport No. 04-0310-CC63 for So. Calir. Edisc (12/23/S3), I ses c p. 30 an ex-a:ple of where, apparantly, a support lead has i= creased by a facter C.91/0.41
= 2.2 in going frc= a 7' ear (elast:,c?) a:aly:is to a non near m-m'ysis. I do r
B-6
cy m - r ___ ~ . . :. ax
_ e. i .m .
. m. _ n. _ -
2 : ,
3 :v - - -r ? , =
nst k cv how good er appropriate Sis co=parison is but it does not lead me to the ec=clusic: that there is no possible proble=. -
Hopefully the Applicant, working vid the specifics of SONGS-1 large bere piping, ca develop a basis for adequately answering the ques.ic s I have rais-Ed. I thi=k NRC should insist c it. The proble: is not unknov: a=c=gst pip-ing designers and NUREG 1061 is a public docu=ent, Oc=, even vi S this le=gthy letter, I suspect there ray be aspects of =y
=ctes not adequately exp'="ed; if so, please let me know. I was pleased viS de Appliesst's respenses at de 11/26/85 =eeting in dat he chose to ec sider
=y ec:=ents c thei: tech ic 1 :erits.
! curs yery' t .:17, CO W R..
EC?/:r E. C. Ecca'caug I cc: Lt.. 16:g Shieh (~dNL)
Enclosure:
EOR Netes (7 pages) ,
pertions frc= NUREG 1C61, Vol. 2 l
B-7 ,
i 3 p__zr - t f:_ H w.- - - _ . m _ .. ,u_ _._u _.a ..
O s de Nadt $r2u !!
/ahle 1: Co.wpwssars el P<o,,a:e.l Q l. ,iis us iJS
~
//j'2z./g.r J9i0 Cedc fr. (9) /.,'oert/r
/
5, k, I 910 '* d* 'E l ' ~ '*
, Ta ,.,p. 4 S,y f/to ' Te e, I 0 0 f f f >t psi f 'I [* s ! A. S
/.2 f3 g) /. 55,, . y 6.
Ioo Iroao 3:00 o 2 ?.S 30 ??: C. so 0.00123 '
. $~100 0 f. 3 0 0 018o Jt os o 17 soo 3souc 3 e (3
.a 500 /Sooa 2:30s 27,3 2f ?71 .s~ s o .:0vf7 s
?? 50 a f. 0 0 2.0/30 /ros o : 7200 3 s aco T
I l
h e ICO 170 a o 30:00 2. T. 3 /5342 ?.323 Q. 2s 2: 3
+2 5tro 1. S il G. clos h tt900 3.333 ,0.020e 22:so litro tri:o b I I
l E [*0 IS?so 19?co 25.2 9700 3. 3.r3 O.20:51 l
,, i l
- j ; If7 3 2. 23 ; J. sto - j -
n .
,l's?: 1 21 30 y , I t. . .774s0 , i.!!: r : 2::o n tist":
2) r:e cd:ir ,0.esaf s r ik parhealre y .r !.< e oi c, et.a i ~<u:
! A c :-ir d e n n ,v K, = !/n - ma a s n u- k;. F.1/owsn; whics are h e .,-k a is t, i,, ofI? .ke
- saban de /, / .~ .r e d 21. he :iainless :fedt 1
I B-8
un w_, w _ ,= - .n = .,a . ,m - 6 E t, a .c. .,u.= . - - . ,-~--:- _
2 cf 7 Souther: California Edise: Co=pany Report No. 01-0310-1459, October 7, 1985 .
1 (Hereinafter called Repert) l Ihe procedure described in 2.0 and 3.0 of the Report is dee=ed accept-able for evaluatic: of piping pressure boundaries (except bolted-fla=ged joints) provided:
by the equatic: shev: c= p. I cf the Repert, a, least (1) In 50%calculating of C," is due(,to earthquake leading. ;
(2) In calettlating =c=ents due to earthquakes, a respense spectru =ethed is used, with da: ping not exceeding that specified in Ccde Case N-411 (?VRC rece.=endaticus).
(3) Dia=eter/v= thickness ratic, D,/t, does =ct exceed 50.
(4) Materials are "'<e SAIC6 Grades A, 3, C cr 1: tke SA312 ?/pe 304. (No quenched and te=pered ferritic steels or celd verked aut.tenitic s' h less steels.)
(5) Jcists are tutt velded er girth f' " et velded.
(No threaded, or seal-uelded threaded.)
(6) Seis=ic ancher =cve:ests are adequately censidered. ,
(7).2e c"-"'**'ve usage factc; due to SSI dces =ct exceed 1/3.
W 5
9
! B-9
.ru wn.n,.-. c. .
.c w -e. n ., =. w n. ., a . n c ,~x ,.:,.,,.,=s::
,, i, , ~ .. n* '...r'-..
~
~
Ta b/t 2
- Shiro Mown,</ Lod res /r en M iyk+ F,ya hem NURfC/C it - 0Z a l
. gg.~ 0, , r,
~
o,./t S, re s t "' _
. r< y .,.J. g,
, 3 3
) ;,, .
- r. ksi la)
Mz ;
to, -ici o ; i .u,,
Mn n~ah n, Ma,a k.:
x, .
S
- 01) - l 10.7S 1RI Sir l 't oo '
/soo l 0.133 l (35) 17C .
990 I/70 f.J t I
ill) - Z l0.7 S , 0.350 29 7 +9 ,
1700 2l c o
- l. (35) ' 1210 treo 1700 1. Y o A
3 11 - 3 14.00 l c. 2 :, o 5 c.s E3 32oo 3Ys o s
- I (15) 12+0 ; 3 t'd 2*37 f. 31
- i ,
% Z s.c o I 0.255 77.1 57 S/00 El 00 h UI)-1 I
(!!) !I50 ?!3 0 9ffo I.?S C o) - 3 12.75 0. i s o j n 7, 99, ? j go " .; g e " ,
s I i
- (35) 275 at: [2] l 9/
b I U 2 )-I 3 2 0. c 0 ' O.25 71.0 't,2.) Syso (J:) ~
l 3070 ?950
(.9 ) -'3 i 50 0.337 12. f 42 2 t, o 3 :, o (35) 217 ?00 S2 1.16
. (
$s.31 $ 1.3 5 $S.JI
'a . 1 I N 3)-10 y,50 2 337 . . .
yo gya 30 g n
0 -
(3s) !?f 777 / Ti J L i, /. f i. i i
4 . ! l t J, 3)-/ 3 I i G, [* .0.Zh3 . e 01.1 I
(
ff DC 400 3 ,
i ', L ;. un ,. sto . .i e s ils l 6 7 5 ,
, l. ii .
- 8askluro I fewjers:;te wr;<sarm) /
aweed
'Ai fit:? orrrey o
, m d:<.2 /
1,, .ra d fe:t r;,nusi:, /V: iM, ,
- Ak 4.1 l' t) , >* t M* *1 int VM < ON Q lU Ch, T' J W;Aito!V!W f:.; m g re, onh,
- /
M ,,
w w hb / -
~ ~
f ( # 6 y v M*
. i /
d 42 at I
.1 f'
~ : ~ ,
/
/
/ w B-10
~~
-sa sa.: _u :. A~ a me _= s.w:_m erar+m n : . . ._.
e= xi : -=w =c-e u w x *y Tis b It 3: Shehc Moma.,/ A cod 7~es,Ls co's flhows, C/e,ts,sy , Zn - Non t, f
- O Pro m Nortf6 / Cit- 014 t, ex s ep /. waro ga y, 0,,,, f, pgt 0.75 i ,'
t S,,
, Teit ' f,,, ,i,J g, l
}. ;,1, tar.
g,; M, M, A'f* - *l' R g, t -
a .o .xr,, to, tuy M,, i..a.y ,
(21) - 2 to. 3 Yf f. 2to 22.7 I.10 50
- 2e Y 26/ +
gs1 in) 17: It! t 215 1.51 (2 2) - 7 (*.'?2 0. n z 11.3 1.23 27. C fti YTo &
% R.9 (15) 394 VI7 + (15 j,95 i:2 ,' -Il !. 5 + E p. uo 11.7 ". :! ??. 6 205 :o i +
RsG '
(2:) II: 17?+ 2C t. 2 +
2 (11) -l ? f . 3 'tf 0.;10 22.7 Z :3 f le. 0 I?+ to:
s Q
2s4 (37) 11 1 /C ? Zif /.52 (22)-:s lo. l ? 3 2. H: I f. 3 !.bl , 3 % 'o l2h? 24?
e:& l Of) 375 J73 447 /. " o I
~
! s. si is. a: e * *:. 1 f 1 L::) -If 6.?+i o.:20 22.7 1. 7 c 37.7 ;91 20s t ;
R=1 (ic) 152 /sys' 21: ' i:t 2.79 (22)-!A 4. J Y r 0. :10 21.7 /,7 0 37.7 17 1 2eZ+ g ns 1 t i
- l. (3o) l 737 ; ; tat e 1
- 1 [ -:s l J. o i k ( :) - 17 l l, . ! FS 0. .18 ' 2. 7. 7 i 0, 3 O f. f. f /7f ' I J o i- l i ,
w d.J '
l lJo) ! /47 ! /4? iI /:,: ' 1:t j :.
- o l' 1
< (::) - IS : 6.;93 3 . ':3. i+.3 ; 1. ; 5 ! i t, 4 141 l 3 t! !
s
[ ' I
% e it s 1 l (30) 21? 3:3 - l:1 , Sti 3
- 2. ? I ei
%s h' Al? O I' !!. lr :f 3. ?? ? ? I. I
- .+2 Y 4. 0 l t 12 2 0 , l 12 7 oi
- s : i '
(50 7/ / 173 '/:33 .:167 3. ~.
i i . l . .I k
' WAK -d f*- 19 45- 25 ? C*'* *d
$A M,,7 I!/ */IC s l ,
j' ~ ' - ~ ~*s'*'
h W, 8 .
ca a W
.B-11 _ _ _ _ _ __ ._
~
>.aw n, m , :.~, :s m .u u , . - c n-n -- - - -nw. .- . ..-~.
_ - . ~ . .y3.- f
~
t* ~
ju 6 f, O. 2 [ C b t'C}c f
f+ ~+
4.te1 0.00: C. o e ? 0. 0 / 0.02 E eg, ps;
- Jone0 6 e 3et /1e 0e 0 200004 600oso Qj f y .-o S o ta 200 /d6 YC 20 D/4,,, ' O. +/%
Ya St o es t
.a M o h r ie l-* $ d'!. T. ! Is k ,: S/sierter s ilos/
Gr . -* 0. s / 0. e 1 0.27 3
/fd* ?3 [* N b P.is:h*J Ear, : N
- s" :- ~&u d An. My 137 Ill o J!:
)
ave,.... .
N, - [7. e en pse, - taia.2 Y i.< :,,u.., sitei w; s ea x>:, i > a u* ';
= l?.4 vo ']{Eeg - t71:a) Y .4e si.i.,le:: :sre!, s = 12.3 no 'ys.'
Wi = Co !e a J/s we ble n a.n bu . I dpro a uio s ,
.: / ,
?/. n Cu s lc; -i e'- f.s o'/ve :
~ . . r. _-
. Pioe W terim! o Elbow Sending Ans.
25.
s /q y ...
0 -.av:
,5
'N '
- n. .
g r,
cc-.o i
0 . 'e'- ' ~
s .n ZO f%*00 ,
Q w rinCh~~ q t
'"** * 0
,'c .
e Pressure 4 %.
_ 't . s 7. 3 V F to . s 3 7 Q o genc..in9 (In )
a ~
1 W \ .* i (
- g 45endin9 (Out ) .
^o elN' -'
2
/ j x Thrust 1 N 5
H _j
~
N Na .l l l i l' 10 '
l -
'103 10 4
r-atigu e Life' a s n,,, , u, , ,,
F.- vinvchi
% FIG.3, FATICt,JE RELA 71CNs'CFIP@fsPfMfl[hENTS AND PIPg MATgggAL 2
- d ~a F 4o8
. ., reau r, v.,,.i 4 CIo #F.
87 [
B-12
. _.w w- :x;,u m 4-% v.1.6 a w;.imic,c.w x _
,..:,.. ., ,6 - ~~,w : ill-6 er 7 l
Questions by ICR, 11/26/85 (1) Hov v seis=ic ancher noticus be ha:dled? If included as part of NC-3600 Iqs. (10) and (11), vd " Pg represent range of =c=ents er as-plitude of =c=e=ts?
(2) Recort No. 01-0310-1459 Section 5.0 (i) Ecv doos optic: 1 address icv-cycle fatigue?
(b) Ecv d e s Option 2 address ce=pressive vr' '< g?
(c) Where do you anticipate " tensile plastic instab ty" =ight occtrr7 (e.;;., is an elbow?, straight pipe, what?)
Ecv d:es either opti:= 1 cr 2 address whacever this concer:
=1ght be? .
(d) What are the u=its of M7 Is M a =c=ent ra=ge er a:plitude?
(e) Vnere, in WREG 1C61 Vol. 2, is "5 sig=ificant cycles" 'd "-
fled as a realistic esti= ate for the Modified Hous=er M .h-quake? What is the basis for it? (e.g., does it ce=e frc= a cunulative ds= age evaluatic 7)
.(3) ')ces 0.75i have a lever beund cf 1.0?
(4) W ha ? uJill kt ul t$ Ae f 5 50 r SanCN 0*"MCCN'"A (c) Fa r- m o me n is an a n ac) ? .
(f) 7 e m. m n h a n ren ?
O B-13
us--m =.: w.: .2.. _ .mm.m. .r. erex amw c.a. n : w. r-ww -~ - ~. . s, . -- = -- -
7 cr 7 Diselacement and Icadines of Picine Sistems Stress idmdts c: y as calculated by Code Eq. (9) are intended to li=it plastic defor=atics of ec=ponents in piping syste=s se that results frc= an clastic piping systa= a alysis vd7' be reasonably valid for esti=ates of displace =ents and leads everywhere in the piping systa=. Having increased the stress 'd-dt c: y by facters of up to tve, it is not apparent that an elas-de piping a alysis vd" predict, er cess rvatively bound, the real re-spense of the piping systa=. Sis raises questicus such as:
(1)'id the pipe displace =uch far.her than es-d ated by the elastic a-a'-
ysis and =aybe hit sc=e safety-related equip =ent and da: age it?
(2) '4 hat vd be the sc::le. lcads c valves, pu:ps, heat e== hangers, ve:,-
sels, etc.7 (3) tihat vill be the leads c piping supports?
(1) '4d valves with =cter cperaters =-ad- operable?
(5) '.imat "" he the =c:ents c belted-ranged jcists:
O s pcssible prehle: is d " ustrated by the fcileving si:ple ec ceptual ev ple.
v/
}W I hua c, s
y A M, y 1 / - VM M C ::;:
%.- L 9
2 m y*
'.
- A (>ffL - - - - -
-/
. >m Ms WLl[ Ms WL/i S e end =c=ents double due to the hi=ge. real piping syste:s, the ends
-ight
- be va.lves or no: les er bolted-flanged joints.
f 8-14
&.w% un .m.6se:tm:M 1,.; . _wsk,wn .M an.k c5 wivW-A.2%.k:2& w . a ._
, , _ Knc/otset L i
. l I
& fracfr 50 ser NUdfG Jolo/, Y' l. 2, p apr o
F22 a / z-2g l i
I
- 2. Nozzle, Flanced Joint, and Succort Loads The elastic piping system analysis that is used to evaluate pressure boundary adequacy is also used to calculate nozzle, flanged joint, and support loads.
The limits to Code Equation (9), shown in Table 2-3 were selected so that gross plastic deformations would be restricted and an elastic piping system analysis could be defended as usable to calculate loads. The potential Ccce .
4 criteria discussed in Item 1 may, of course, permit higher limits than shown j in Table 2-3 and are intended to do so. The applicability of an elastic piping -
system analysis for calculating nozzle and support loads is no longer apparent.
What appears to be needed is a simple technique whereby the leads, for piping i systems where Code Equation (9) limits are exceeded, would be scaled up. Refe r-ences 2-21 and 2-22, as well as the ANCO data, should be reviewed to see if ,
i a basis for such a technique can be found.
- 6. Preblems ,
a
- There is no apparent pr
- blem with the pr pesed change in Item 3. Hewever, the type of change discussed in Item 1 needs to be acecapanied by a simple way to determine leads as discussed in Item 2.
Axial stress is not limited by Code Equatica (9).
2.5.4 Censultant Suggestiens f r Research Programs A program should be initiated with the specific cbjective of develeping a simple way to establish loads from an elastic analysis of piping systems where the stresses in the piping exceed the present Code Equation (9) limits. This program shcula include, but not be limited to, a review of high-level test data 1
and the Campbell [Ref. 2-21] and Breman [R,ef. 2-22] reports.
The end product of this research should be the development of pseudolinear-elastic seismic analysis methods that conservatively and yet reasonably approxi-mate the results of nonlinear time-history seismic analyses for piping systams.
Design procedures that fully incorporate inelastic energy absorption and strain capability of piping systems should be developed. Two candidate techniques are !
the cynamic-to-static margin ratio technique of Reference 2-21 or the use of increased allowable stresses such as discussed in Reference 2-22. However, soth techniques need further development and more comparison with nonlinear cime-history analyses of realistic piping systems.
4 B-15
$. O. kodabaugk 8.uociates, $nc.
4625 CEMETERY ROAD e HILUARD, OHIO 43026 614/876 5719 January 22, 1986 l Mr. T. M. Cheng, MS566 US Nuclear Regulatory Commission 7920 Norfolk Avenue Sethesda, MD 20814
Subject:
San Onofre Nuclear Generating Station Unit 1, Technical Basis for Stress-Strain Correlation", So. Cal. Edison Co.
Report No. 01-0310-1459, Revision 1, January 1986
Dear Tc=:
The revised Report does a good job of addressing all of my co:sents and i questions except those on page 7 of my notes; a copy of that page is enclos-e.d.
The revised Report addresses my page 7 questions in three places:
(1) On page 18: "Solted-flanged joints are qualified per the require =ents of NC-3658 (Analysis of Flanged Joints) of the Code C1].u i
(2) On page 21 under the heading " Concern on Pipe-Mounted Equipment Qalifi-cation".
(3) On pages 21 and 22 under the heading " Concern on Accuracy cf Elastic Pip-
. ing Response Analysis".
On page 2'1 the revised report states: -
" Elastic piping response analysis vill predict conservative piping bound-ary loads, even when the elastica 11y-calculated stresses exceed the Code limit. "
If " conservative" means the boundary leads vill not be under-esti=ated, and if the quoted sentence is accurate, then my page 7 questions (2), (3) and (5) are answered. However, my page 7 includes a simple conceptual example in which an elastic analysis would under-estimate boundary leads by a factor of two. Fur-ther, as I pointed out at the 11/26/85 meeting with the applicant and in =y 11/29/85 letter to you, in the EDS Report No. 04-0310-0063 for So. Calif.
B-16
.---__ - - _-. , ~ _ _ _ - - - - - - -
= ,
Edisc: (12/23/tG), there is an exa:ple (p.30) of where a support load appar-ently i= creased by a facter of 0.91/0.41 = 2.2 in going frc: a linear (elas-tic?) analysis to a non-linear analysis. O ese data indicate that the 'quet-ed se=tence is inaccurate.
My page 7 questics (1) cc ce=s calculated displace:ents and is addressed c: p. 22 of the revised Repert. Oe first sentence poses a =iner proble: t:
=e because, at least in =cs suclear pcuer plants, there are a ==ber cf es-sential piping syste s which operate at a:bient to:perature. It is =ct ap-i pe. rent to =e "=* d-such syste:s the ther 'y-induced calculated displace-
=ents uculd be larger tha: te ear c quake-induced calculated displace =ents.
Ecvever, y =ajcr proble= is that, granting that elastic-calculated ther:al displace =ents are larger ca elastio-cale aated earthquake displace =ents, are the latter accurate when the calculated elastic stress is up to 4.0 ti=es the
- aterial yield streng-2 and'per=issible :::ents c elbevs see up a 2.5 ti=es
-le experi=e:->"y-dete: ' ed =axi = :::en capacity?
Oe see::d sentence cf d e paragraph :: p. 22 of the revised Rep::: does
- t cc:vince e that large displace =ents ca=ct occur. It is ct apparent ,o e da: --
", s r s (e.g., 0.02) =ecessarily i= ply s==' displace:ents. Se dird seEtence is = ice but, if the :agnitude of seis=ic displace:ents casse: be es ' ated ( : bc=ded), I de 30t see hev adecuate clearances f:: seisni: 1:ad-ing = be es ablished.
My page 7 questics (4) appears to Se at least partially addressed by de l, parag aph c: p. 21 ef the revised Report headed "Conce= c: Pipe-Meunted Iquip-ent qualification". Mest valve operaters are qualified in ter:s of accelera-ticus so the questics here is: Will an elastic piping systa= analysis v th -
elastically calculated stresses.up to 4.0 ti=es the =aterial yield strength ap-prepriately esti=ste (or beund) the acceleratic:s the valve operater vill have 1
i B-17 J
ym _ -~ -
- - - m:
\
' 1 e l e -y -
1 to withstand? The only clue I have comes frc: the previously cited EDS re-port, p. 31, which indicates that acceleraticas calculated by a 14 ear (elastic?) analysis can be lever than those calculated by a non-linear a:al-ysis. This, cf course, suggests that an elastic a:alysis vi'7 not necessari-ly bound valve operator acceleraticas.
J i
In su==ary, it appears that c e respc=ses in the revised Report to =y page 7 questi:ns are inadequate. If, for exa=ple, se:e k=culedgeable i=ter- I vener vere to raise these questic=s, I dink the discussics centai=ed in the
~
revised Reper; veuld not convi=ce a c:=petent judge tha: ce applica.st (cr NRC) had adequately addressed the questi =s.
l 1
1 I vish I had a si=ple ansver to =y page 7 questions, but I do n:t. One '
=ight cc duct sen '4 ear analyses for all piping syste s where the high allev-able stresses are needed; but pres = ably that veuld be prchibitively expensive and ti=e-cens=1:g. :f I vere he applicant, and dee:ed dat -le ::n-linear analysis appr:ach vas i pra::i:al, I "' 'e I veuld frs=e =y response to =y pags 7 questi: s ale:g -le fell:ving li es.
(1) Act"*' Es.rdquake Experience Cite what actually happened to pipi=g syste=s during severe eart-quaker, pein.1.g cut dat equip =ent attached te d e piping syste=s (pu=pe, j valves, heat exchangers, e::.) vers ::t da= aged; supports did n:: fail; a:d adjacent equip en vas not da: aged.
(2) Si=ulated Earthquake Iests Atte=pt to quantify the relaticeship between what was =easured in these tests and what would be calculated by the Report procedure (e.g., response spectra analysis using FVRC da: ping). Shov ratios of:
B-18
g~-- - - - - . - - _ - _ - --.
l
. , _y a
(a) Displace =ents (b) Boundary Loads .
(c) Accelerations Hopefully, this would show that, at least for the tested piping syste:s, the Report analytical methods reasonably bound the sensured responses.
(3) ES Nc=-Linear A=alyses (Report No. 0/.-0310-0063)
"his reper; vculd see= te previde se:e da a,fer a technical respc=se to my page 7 questic=s. For exa=ple, on p. 30, nede 24, I-directics,the s
elastic and =en-linear leads are 0./.1k and 0.01k, respectively. But what is ce actual hardware represented by cede 2/., I-directics? Is 1: perhaps i
a ik-rated snubber, a chered for a 1k 1 cad? Mere gener=y, what design ec:servatis:s are ' volved between the calculated leads frc: the pipi=g syste: a:alysis and the actual hardware desig ? The ES Repert does =ct give any displace:ent data but presurably the data is av=' % ble and could be used te respend 2 =y page 7 questic: (1) for displace =ents.
I recog_ ice that the hind cf app;cach cu-ad ='cVe is heuristic and does
- lead te a:/ genery applicable criteria. McVever, I ^'-> - apprcach prevides =cre relevant respenses te y page 7 questic=s ca: ary 21:g 1: es revised Repert.
1
! curs very truly,
~
bd ECP/: -
E. C. Redabaug:
Enc 1: Page 7 of ICR Notes f 9 B-19. i
e e 1 7 er 7
~
fnm a ER M.:hs Diselaceme .t and Leadines of Piring Svstems Stress 7'-dts c: 0," as cal =ulated by Code Iq. (9) are intended +a 'd
't 4
plastic defer:atics of c =;c=ents 1: piping syste=s so dat results f::= a elastic piping system analysis v7 he rease ably valid for esti=ates of displace =ents and loads everywhere in the piping systa=. Having i= creased the stress -' t c c',' tf facters of up to tve, it is net appare:t that as elastic pipi:g =- 'ysis v'7' 7: edict, c: ce=servatively bcusd, the real r w
., sp :se of the piping systa=. Ois raises questi==s sus as:
(1) V'" the pipe dis;12:s ::s farther es-'--ted by the ' elastic - '- .
ysis and =aybe hit se=e safety-related equip ent and da: age it?
(2) Pst v'" he the ::::le leads c: valves, pu=;s, heat e ds gers, ves- "
sels, etc.7 (3) Pat v'" he the leads c piping supperts? -
'(!.) '."" valves vith : ter cperat::s re-*'a cperable?
(5) ?st "" he de :::ents c: bcited-fis .ged jcists?
"he pessible pr:ble: is illustrated t/ the fellevi g si=ple c:: cept m' em:;1s.
o I
W M -/}x h/,,, c, k #
\
I MC3 Pe
)M ==ep Cgr ^
,))
{
.u__ _ - .. . .. . J Lt j M . WL/ l Ms W/Y 7:e end :::ents d:uble due to the ge. I: real piping s/ste:s, the ends )
=1ght be valves :- ----'as c telted-flanged ' cists.
O e
B-20
-- ~
.- 2 - --
r
- O O E. d. d?odaBaugf1 c%ociataa, Gna.
4825 CEMETERY ROAD e HILLIARD, CHIO 43026 614/876-5719 February 13, 1986 Er. Long Shieh I4vrence Liver:cre National Laboratory F.C. Ecx 808 Liver: ore, CA 94550
Subject:
San Cncfre Unit 1, Review of Licensee's Stra'"
Based Methodology for Seisnic Ivaluation of Large 2 re Pipe
- car Dr. Shieh
This letter-repert is sub itted to c:=plete de se:pe of verk specified in LLNL ?.0. 82S18C5 dated 11/13/85.
The verk cen isted Of the f:ll:ving steps:
i1 Review cf d::u:ents provided to :e by Mark Russell (IO&G Idahc). :n parti-cular, : reviewed 23 Scudern Calif:rnia Idisen C::pany Rep: t :::. C1-C31C-1459, dated Cet:bar 7,1985, " San Cnefre :iuclear Generating Station 21: 1, long Ter: Service Seis:1c Re-evaluatic: ?r:gra=, Technical 3 asis for Stress-Strain Correlaticn". This review took place during 11/21/35 and 11/25/25 and lead t :y preparing a set of notes (7 pages, included herewith as par:
of S ciesure 2) for discussi:n vi h de Licensee at a :eeting in 2ethesda en
- 9. *i 24 2.,4 (2) Meeting with SC:lGS-1 Licensee (50. Calif. Idisen), Bethesda, 11/26/25. 2.e amtendance list is included herewith as helosure 1. A nu=ber of subjects vere discussed, including the cuestions raised by :y set of notes (Fer : ore
. details, see " Meeting Su==ary-Seis=ic Reevaluation Criteria", Eileen Mc enna, dated 12/9/85. )
e B-21
.t ___w--- -e._ _
4 2
(3) Folleving the =eeting, I wrote a letter, dated 11/29/85, to Dr. T. M. Cheng (NRC), which fer: ally trans=itted =y set of notes to him, expanded a bit on those notes and indicated =y understand
- g that the Licensee (Applicant) had cc 4 tted to provide a revised version of Report No. 01-0310-1459. That let-ter is i=cluded herein as h elosure 2.
(4) On January 18, 1986, I received a c py of Revision 1 of the So. Calif. Idisen Reper No. 01-031C-1459, " San Cnefre Nuclear Generating Statien Uni: 1, L:ng Ter: Service Seismic Reevaluation progra=, Technical Basis for Stress-Stra.in Cerrelatics", January 1936. I reviewed this revised reporn and cencluded -hat i, satisfacterily addressed all of =y questiens (see Inclosure 3) except f:
these en p. 7 cf =y n tes (See last page of E cl. 2). : then wrote a letter dated 1/22/56 to Dr. T. M. Cheng (helesure 4 herev12) indicating :y reserta-tiens concer-Jng dat portien of de revised repert.
1 (5) My le :er dated 1/22/56 uas discussed by telephcne en 1/31/26, vit represen-a ive s cf NRC, Sc. Cm " ". Idison, :pell and :yself.
- 3) On Tehruar/ 13, 1966, I received a copy of a Supple ent to te 30. Ca r . Idi-
- sen Repert N0. C1-C31C-1459, Revisien 1; ;.he Supple
- ent is dated February 7,
,. A. .
il As indicated in helo:ure 3, I dee: that ce Licensee had (by Revisien i :f s
Report 01-0310-1459) satisfacterily respended to all cf :y suggestiens and ques-
- tiens exespt these en p. 7 cf =y notes. The re: tining question, briefly, is:
1 Will an elastic analysis provide sufficient guidance for esti: sting leads en a-tached equip ent and for displace:ents of the piping syste:7 I viev the supple-
=ent to Report 01-0310-1459 as the Licensee's respense to this questien; the re-
=ainder of this letter gives =y cc:nents on the Supple:ent and :y recc :endati:ns.
B-22
. _ - ~ . ~ - - - _
o o. -
-3 <
A. Pire Dis lace ents The Supplement, in 5.0, addresses this portion of =y p. 7 questions and ar-rivas.at what I dee: to be an acceptable respense; i.e., =ultiply elastic-calcu-latsd displace =ents by 2.7 for pipe interference checks.
- 3. Loads en Eeuir:ent The Supple =ent, in 4.0 pIpI 3 CIT'!DA?Jf MADS, addresses the leads and accele.-
a:icns perti:n of =y p. 7 notes.
In general, I view the centents of 4.0 of the Supple:ent as a :uch better respense :: p. 7 cf :y =ctes the. con.ained in P.eper; ::c. Ci-0310-1450-?.1. 50:e spaci.fic c:nnents fellev.
Pr. 4.1:
I recognize that there may be se e controversy over the quantitative inpli-
.. 4_..,a . e.e . ._4 ,4_.. -
4
.. ... ,ys d s. ". -". .v e. , - _d. a. e - .'..a'. *. a'.. ~ a..' _' s a reas:=ing respense :: :y concerns.
Par. / 2 a
...,] s e ... e.
. . . . . . 4_3. nn. .e.e
_....._.f,
- 3. e.r . . e . ..,
. . . .. . . 3 2 4 .a. . . .... . _ . ,. .e
.s
.e s.. _4, .2 . . .........,.4.-. - . ..-..3
- . .. , + .. _4 _ . .....-
2, . 4 .
4 4 ,.
.. . . a.,.. .fs_4.,n h: pod da ce licensee : uld da-a-d -a se=e ratics as suggested in :y 1/22/26 le-1 Ocr (Incl:sure 4), p. vever, I ree:g.1:e de aceracy cf .he qu::ed sentence and, in viev of te tine in whid de Licensee had available :: .re.are : : his respense s
- an not surprised ,dat he did not evalua'.e de test resul s.
An unfortunate feliev-en is that the subsequent ;c agraphs under 4.1 are no-releve. to the ques:1cns raised by p. 7 cf =y notes. 2.ey es relevant to se pressure bounda:/ integrity of piping.
B-23
_ _ - s, e o Par. 4.3 .
1 e
I regard this as the =est convincing respense te p. 7 cf my notes. Neverthe-less, it is an " engineering judg=ent" type cf respense net a: enable to quantifi-entien in the fer= cf criteria. This leads =e back to Par. 2.0 cf the Supple =ent.
Par. 2.0 a '.-. u*. , .' .. s. a.-*.' ~ .# - -..s,__..-
- a 'ad a.xc-a.ssed. c .... .e
. . ' _' ar , ' . e ^. . "a 2 s '.. = _' .. _' _' _'
less steel as pessible invalidating loads and displace =ents obtained frem an elas-
._.,-4,._4,
.. . . . ..,.s_.e anal, s' .. "."..a. S" e .. _' a._.- a. .'., , _d .. 2. v* , a. . . .
. d a'.e _' ~s di s e" s s e s s .. ..' . .
.'a ,_'
_ _4_4 . =_ . . a s _' ' _ __ . a..d. 'v. _-'.v-c/c a .a. ." a_ _ .- a. c.' . a. .' . .
". . .. ". ev e , . '...' v a. a . . - ." .' a. -
.. .a . u. . . , n a_ _.a. g _.e _,, .. .
. . -. . u.g . ; ,a3 ...
. . . . ._n.
. .1. , __ u._, e _ a .< ,, , .r._.
e, g. rs .< _
tien of ("a), see p. 20 cf Ci-031C-145~ -R1. In y experience, the usage fraction U can be uch less than 1.0; e.g., 0.1 cr even 0.05. Te cover that pessibility, I
....u_e_ .. _.va. . . . a s s"- e "..a ., p'"a) ava' ' a x..i a. .#.-- . ". a. .. -
- c ". ... . . e _= . " .3" e.h.. . ..u_' .d ' a. a_ _' -
,a
. . . . . 1.m. .s . ,_,.; . . ~
. . . . . .. v..,=,
....,; s .. a ... .. .. _.u_ 3. a.c. . ._ , _ _ ,_ s. .. ...
,,.,,.c..,.,...4. .. ... ..._....,,....a..,,..e.u.. . . . .... . .. .u.. ...-4 __._.
.. .. .., n . _a.,._a.. .. . . . . a __ _a.a ..
. . .. . .. .e...r.._.. ._e .
, ne. ........a..
~....u...
. . _,..a. .- s ..- a . .e .. . .e.,,. . . . . - . . .
..u .:r i ~. a ._ a. .. _ . ...
,_s . ._._._.ar_s__ e ;.
... s 4
. _ . . . s ...
3 . . s. S- s .. , __4 . . . .
~_.,.4._
. . . . . _. a s .
. . . . . . . . . ..,... ,. ,..t -a 4 . . <-:
._ ,.. .........n
~.. . = . -"es._'-..~._'s~.~.~.~..~.".s_'..-*_.'..*.*.s'....~j.'..'_'s.
g . . . . . . '.i~ ..~j '_s .'.a. .- ' '. . a . .c a. a. s .. . .. 5-
__4._., .
- 4 u s.*.__. ' v a C .C 4 s'.-= ._ _'.n _' _'_ ~ _d. *. ".
w.e .. , a~.
.. a-a.
. .. .,_' v , a ' ' ".. .a '., '. s .. a. a. d a. d .'.-- .
50:*05-1 is 0.01257 Cne aspect of the questien is that, in :y view, NRC shculd
..e. a',, " -._ds "-.e '-a. vievd..- e* u.:r 1 c d. '.e *. d_ a a s -a. ...e-a._'
'.".a. w'.' _' '/ ya ,__ abla. . ' ' . . ",_ e, i it ay be .:essible te de sc. S t at least a careful appraisal is needed cf what has been de cnstrated by si:ulated ear hquake tests with respect to boundary leads, displace =ents and accelerations.
8 24
. -5_
C. Accectance of 0.01' strain Linit Pages 3 and I, cf =y notes show cc=parisons between test data and the propos-ed criteria at a strain of 0.01. *he right eclu=n, Ma/M2 , vould be ebviously ac-captable if the values did not exceed 1.0. (Divide the M,/M2 #*tiO8 f*# 8t'i*-
less steel by 2 for this ce=parisen.) The ratics are greater than 1.0; hence, to defend the adecuacy of an elastic analysis to predict beundary 1: ads, it is necessary .c inveke engineering judg=ents such as:
(a) piping systa= g ess distortion requires =cre than ene plastic hinge.
( w. ) :.. . ,._u_.~ s, .s.e . . ,. s s"~-a. .e. - d .. '" e-. c. _' ', e_ _' = " su_= _" /' .-~. _' " a s a . -
additicnal =argin.
(c) Ear-2 quake-type leadi.gs are less severe -dan s atic 1:adings used in the tests.
These types of judg=ents, in conjunction with the discussi:n in 4.0 cf the Su . _' . a. .. , ar .
. s"' a d . ' . ' - .-a_. *-
. . .. a-.....
. '.a. ... .. sa.d .' .e.' _= ".
.. _ .. e. s..e_' .
-.a_
. - _ .. ... . , - ~
v.e.i.
- 2. .*ce e--- #
==tveen 0.01 and 0.02 strain Reper Nc. 01-0310-1/.59 (beth the 10/7/85' and revised versi:ns) cen.ain under n ... g.:-.: a .a ., g_- - _
.a a _
-- . , a r_s .
,.g~. a. .,. .- , c. g . ~ . a ,. . s . ..
.a - . .
_ e _a _ a ,;...._ _ . ... , ...
.s..... ,
2.0 fer review cn a case-by-case basis where calculated strains f:: s.a'-'ess steel exceed 1 percent and/cr deviati: s frc: 9.e stated stress-s rai . ::.-- e-lati:n =e9.:d=1:s~.r are a:: lied." ..
- n he precedd g (C), we have indicated cur accepmes of strains up .c 0.01. We hsre give suggestions for accepta.ce of strains up to 0.02. If Supple =ent Table 3-1 is cc:plete, we are addressing four " piping problems" with a taxinun strain of 0.0128.
B-25
_._.s __ . u o _s :. s - u .y , ~. m.. . i ._ , _ .mems ., __ m. _ . _ _
_..,p..
e o (a) Establish a ' d u= required =argin, I, for leads and acceleratic=s obtained frc the elastic analyses: e.g.,
I w 1 t- ((- 0.01)x170 (1)
Equation (1) is a= engineering judg=ent equation fer=ulated se tha X = 1.0 at G = 0.1 and I - 2.7 for e = 0.02 for sc=e continuity with 5.2 cf the Sup-ple=ent. For e = 0.0128, I = 1./.76.
(b) Reviev leads en supports, no::les and belted-f'a ged .'cints. Review acceler-ations of valve operaters. If these 1cada are less than 1/I ti=es acceptable
_ cads, ".e ,d..d
_. ~.. /os *, e . ( c. r _d ,._4..., . . a. ) .<.."'d 'e ac-a
- . - . r '.ed. -
(c) If checks (a}/(b) are act satisfied, censideration =1ght be give te -"- 'ng a pseudo-inelastic analysis in which locations c' d-d*cated gross plasticity (e.g., an elbev) veuld be replaced with a constant =c:ent spri g. If these give cale"'ated 1cada and acceleraticus less than acceptable leads and accel-
. . . . , _ . , , w. -.. . _4..._ .e .. ,_,_ (e- y 4_. _4 3 .. . . w _ . ) .e . ~._, .a w. _..-.. . .. ed.
I, Rece endations (a) ".R0m shculd n accept the criteria in Reper, No. 01-0310-1450-R1 fer strains up
.c v.w1.
rs u.., .c .e,-- ' .. ' a .". e a. .. v'.a* and . 0. a'^, s.' w. u.' d eva_' ".a ,,e
. - '..T. . , . eaer. ,_' ..' .., .ey ne. ~.~
, _4. ..4.g,. ;. . . w _i . _ ..4 ,
_4 .,,0 4. ....w ,, .w
-.s,,, s e 4 4 sw ... .d 4-
. _.5 .. . ... (' 3. \ . .
j
(' <. '> ~..~.e di s r _% - a.. e.. . c.".e 4. a- das-
, . - .. d'ad '- #,,
, .1, c.# '.e .c"--'a-a-* ..-'.a .S. a. r .. . .
.., a w
.e. -- e e ~. 4. _a. .. .. a. .. ." .. _ .ar . ." ".e a c .- a *, ' . a.r, . .. .. _d . e . ' ' .
w .: a,, ".e ....sa.... ._'~a, . .m.a,..- ..
....,, <.-"'d *-
- t, ~ .. c ..s _' d a.- .".c . v'"i -v"' .', i u - i 4.c, c, -n. , a..d.
its Supple:ent as gene-ic*"y acceptable.
F. =c en-isi 21:a-d s cf "a--c r Ove- _ecnse va-iz:
.. - o. , _ e .. a_..,-
. a.. .
4 . .a .. .e et ... r .e d e_4_ .,.e __ 4 _. ,,. .. a . . - di -o. .,a_ .. .. .. u.
. . . .. .. .. . .e , . u. . .
.r w _- , ,. .-a_ ... .
3 . . . ,; _4__r,.
.. . c... s . _4 . , . .. . e e s a ., e ,4 a.. . a .
- a -.-a. s _ _' ..3
_. -a . .. .. _' ,. e _
. .'.a. ".e-a a. r_' ' a d .-
.,w .". ,4.:r- 1, s"- -". a s ".a . a. x -a. . .c. .e. e d.
, ,w. e . . . .,_4 .. w7 3 w.e w.e .m.e. r, .. .,,.a- _e .r +w ..
. .. o4-
. . . , . d. add ' .' - C_ - a. s ,. .a..' .. a-
.6 . .
(e*6*f -
- ""e
- - - s ') "-- tid a *"e
-- - --eul"* - -'
5 r-r-- o
'-- +
/****c*s a d "e' a a' ' ad a-"d- . - r - a-.
- =* -
- w. c . e ". ,= 4.<, w._. .s here is a sig .ificant bcdy of ec petent judg=ent, such as that expressed in Vol. 2 et m' REG 1C61, that indicates the answer to the questien is "Nc" er "Probably net".
Acccrdd gly, ve dee: it appropriate to apply engineering judg-
=;nt to establish less stringent criteria. Indeed, in view of our rapid.ly B.25
~ . . - _ .. u - _,_ . . _ .
_w_ _ _
e 7-e increasing knevledge of the effects of es.rthquake leadings on piping syste=s, if additional restraints were added to SONGSd piping, in a few years there =ight ba a :otivation to re=cve so=e of those restraints to obtain increased reliabili-ty.
l
?CCs Very truly, 2.C. B-la.kw A E. C. R:dacaugn
- ECR/::
Enclesures:
11/26/85 (1) Attendees, (2) ECR to T.M. Sethesda Meeting,/29/85 Cheng Letter, 11 (3) 50;!35-1 Resp:nses via Revised Repert (4) ECR : T.M. Cheng Letter,1/22/56 cc: ?. M. Cheng 9
B-27 - __ _ . . _ . _ . _ _
a way- - _ --g y--- -
2_. --
1 .
e , !
\
S.$.fadabaug$n81sodan, $ne. l 4425 CSIETRY ROAD e HILL!ARD. CHIO 43025 *
$14/876 5719 February 22, 1986 Dr. Lc=g Shieh I.avrence Liver:cre Natic-=' Laberatcry F.O. Box 808 Liver:cre, CA 94550
Subject:
SONOS-1, Meeti=g at I=pell c 2/20/86 tear Dr. Shieh: ,
My letter..to you dated. 2/13/86, p. 4 under the discussics of Par 2.0 of de Supple =ent to Sc. Calif. Edisen Repert No. 01-0310-1459-R1, ecstains the i= plied cuestics: In dcing the fatigue check under 5.2 of de Repert, ud ' ' (Ua) ever be taken as greater than 0.257 At the =eeting, I u= der-
, steed ^=' "e ansver was dat (Ua) veuld :ot be taken as greater than 0.25 a-d that the Supple ent vd" be revisec to so indicate.
The pertic: cf =y 2/13/86 letter c: " Acceptance of 3etween 0.01 and C.02 Str*d-" ec >d s sugges.icts to the NRC-reviewer a-4, thus, does =ct di-rectly involve the Reper cr its Supple =ent. At de :eeting, I:pell pre-sented t e enciesed a.1:er:a.ive to obtain the argin, I, fer leads asi acceleratic=s ebtained frc: an elastic analysis. My respcsse was dat y suggestics in =y 2/13/86 letter appeared to be =cre defensihle vith re-spec- to E::S Repcrt No. C4-031C-CC63 (12/23/83) and, " 'ess se e specific p chle: arcse with its applica:1c (e.g., vculd recy. ire added' res rain s er beefing-up supports), I would prefer to stich v-th =y p;cposed I. -
! curs very tru.ly, -
I S. C IklaF<>u) l E. C. Redabaug=
cc: T. M. Cheng /
Z:cl.
i f
e B-28 -- - _-. - . - _ _ . - --.
. . _ - _ n. m. . .n = -. .u ,
.)]be A.2 bib
&pcH; :2)2e/cc.
DISCt!SSION OF THE MINIMUM REQUIRED MARGIN, X, FOR LOADS AND ACCELERATIONS .
OBTAINED FROM THE ELASTIC ANALYSIS. .
- INCREASE the leads / accelerations by a facter of 2.0 by assuming a hinge in the middle of a fixed-fixed beam for mid-point static leading:
ML ML = 2.0 7 "T Reduce the loads / accelerations by a factor of 1.5, which is the ratio of the dynamic margin against failure to the static margin against failure ,
I Therefore, X at a = 2%
=
h = 1.33 ,
- Linearly interporate X for e in 1 - 2% .
X = 1 + (d -0.01) 33
- Check the load /acceleratien increase in the vicinity of all locations which satisfy the fo11cwing ecnditions:
o Pipe strain is greatar than 1% (Stainless Steel) o Neminal pipe si::e is greater than 2 inch (large bore) o X is greater than 1.1
' Example: 4= 1.23%,X = 1 + (0.0123 - 0.01) x 33 = 1.09 No check is required a= 1.5% , X = 1 + (0.015 - 0.01) x 33 = 1.17 ,
Check is required e
4
Pw=*w=" ,
. mm 0 o APPENDIX C PROPOSED REVISION TO PIPE SUPPORT ANCHOR BOLT CRITERIA e
e 9
o e
. -a. . .v . 1 ~< amw : a. .u. ,n.eu.a. .n . .=u ~: - , _ - ,. . . - _ - =a,_= _ . . ,rsaa
_ = . _ _ .
a s .
REVISED ANCHOR BOLT CRITERI A OcToSER 9, 1985
. 9 9
l l
i e
a ,: .- ...,c..-
. . . . . .. . , :: . . . _ a ... :~,-,--.---..._._.n.-,-... .- . .. .. -- - - - . . , . ~ . . ~
a .
LTS ANCHOR BO_T CRITERIA 0 FACTOR OF SAFETY OF 2.0 ACCEPT /BLE PROVIDED:
A0J ACEtJT SUPPORTS ARE OUAL 1F LED TO A FACTOR OF SAFETY CF 4.0 MlfilMUM OF 4 Af1CHOR BOLTS NO MORE THAtt 1/2 0F THE AtlCHOR BOLTS ARE S I MUL TAf;EOUSLY If; TENSI Oli 0 OTHERWISE USE FACTOR OF SAFETY OF 4.0 OR 5.0 AS AFR. l CABLE.
e C-R -- _- _ - - - - - -
M #k % C lt h tss M ui & w :: . m , 2 + a m W M ?~E E = @ w s -- e m=-
D t
SUfe%RY OF NRC REVIEW / CONCERNS 8 bust PROVIDE OVERALL FACTOR OF SAFETY OF 4 OR 5 FOR THE ENTIRE SUPPORT ASSUME ANCHOR EOLTS WITH FACTOR Cf SAFETY OF 2.0 FA!L REDI STRIBUTE LOADS TO OTHER ANCHOR BOLTS DEMONSTRATE OVERALL FACTOR CF SAFETY C: 4.0 OR 5.0 1
l l
C-4
c .y _ n - ..... - _ _ -. - -. -.~ - -.- ,. _ -- - , - - ,
<~
t RE\/ ISED ANCHOR BO_T CR ITERI A 1
0 FACTOR OF SAFETY OF 2.0 ACCEPTABLE PROVIDED 1
ALL CONCI TI ONS OF ORI GI NAL CR I TERI A ARE MET SL IP ' TYPE FAILURES -
9 . WEDGE ANCHORS S SLEEVE . ANCHORS LOADS GREATER THAN FACTOR OF SAFETY OF 4.0 ARE REDI STRIBUTED TO ADJACENT SUPPORTS ADJACENT SUPP0 TITS WITH NEW LOADS ARE QUAL 1FiED WlTH FACTOR CF SAFETY OF 4.0 ULTRASONIC-TESTING HAS BEEN PERFORMED AT THE SITE WITH 100% SATISFACTORY RESULTS ON EM3EDMENT VISUAL INIPECTION 9 CONCRETE OR GROUT CRACKlNG e THREADS FULLY ENGAGED 0 CORROSION 6 CHECK IF BOLTS ARE HAND TIGHT i
I l .
I C-5
\ myx%2 e;-1&;aretm-u :;xwsm nimw.-:- . . . - m a w me' w = ssis.v m ew.
- o
?
REVIEW OF TEST DATA e AN SIS OF ANCHOR BCLT TEST DATA PERFORhED BY
[ 2900 TENS 10N TESTS, 1600 SHEAR TESTS)
ULTIMATE STRENGTHS ARE GENERALLY LARGER THAN STRENGTH I N MANUF ACTURER' S CATI, LOG STATIST l CAL ANALYSIS INDICATES A FAILURE RATE OF 1% FOR FACTOR CF SAFETY OF 2.0 VISUAL INSPECTION CAN EE USED TO IDENTIFY CASES 0F POOR WCRKMANSHIP O NRC SONGS-1 TEST DATA (HILTI IMIK - BCLTS - 1984)
ALL TESTS GREATER TH AN M'ANUF ACTURER' S CATALOG (AVERAGE 1.33 TIMES CATALOG V ALUE)
ALL SL I P TY PE F AI LUR E S -
5 FAILURE BODES WEDGE AND SLEEVE TYPE ANCHORS HAVE SLIP TYPE FAILURES SHELL TYPE ANCHORS HAVE CONCRETE CONE FAILURES C-6
.,--x v_n man -,-- ---n- nn ..~ -m
. J -
.~
'MPi cal lead- hflecYion C.ueve. Se 5/9 " um , n / /
Ty PR. Co n v-* ht Awe.ho/
i
\
l Lo -
I
\ s g- I
' We dg e. c.d S l eeve. Ty pe raees Anc. Ley t U l
)
~
H 2
.oso ,o75 .joo Ias
.o25 .
Defle.r.,bn (isubRA)
C-7
-, -, = v a