ML17264B101

Rev 0 to Leak-Before-Break Evaluation of Portions of RHR Sys at Re Ginna Nuclear Power Station

ML17264B101
Person / Time
Site:	Ginna
Issue date:	08/29/1997
From:	Cofie N, Deardorf A, Markovits C STRUCTURAL INTEGRITY ASSOCIATES, INC.
To:
Shared Package
ML17264B100	List: ML17264B099 ML17264B101
References
CON-RGE-07Q, CON-RGE-7Q SIR-97-077, SIR-97-077-R00, SIR-97-77, SIR-97-77-R, NUDOCS 9711190074
Download: ML17264B101 (61)

Text

VENDOR'S DOCUMENT REVIEW 1

APPROVEO ~

DOCUMENT'Q'g gQ 5C3 PPROVEO

~ SUSMIT FINAL MANUFACTUPINOMAVI RCCEEO APFSOVKD AS NOTED MAI'F: CIiAN'KS ANDSUSMIT FNIALDDOC44SN ~ MANU FACIUINNOMATFHOCEFD lS A$'>ADVED NOT APPROVEO-COIIIII.'CTANO RESUBMIT REVIEWNOT REOUIREO VAIIUFACTURING MAYPROCEEO APPROVALOF 'THIS OOCUMEN I UOFS NOT REUEVE SUPPUER PRON FULL CI>LIPUAIICEVe'i 1 H CONTRACT OR PUll 0

R R~

UIRELIENTS TPP.

onTE:I ROC TER Ci XR Fi O.ECTRIC CORP.

POC 9:STITB, I4'(

Report No.: SIR-97-077 Revision No.: 0 Project No.; RGE-07Q File No.: RGE-07Q-401 August 1997 Leak-Before-Break Evaluation ofPortions ofthe Residual Heat Removal (RHR) System at R.E. Ginna Nuclear Power Station Prepared for:

Rochester Gas 4 Electric Corporation Prepared by:

Structural Integrity Associates, Inc.

San Jose, California Prepared by:

C..

arkovits Reviewed by::

Reviewed and Approved by:

. Deardorff N. G. Cofie Date:

Date: >>< ~7

'Tl7iii90074 9'7iiii PDR ADOCK 05000244 8

PDR Structural IntegrifyAssociates, lnc.

EXECUTIVE

SUMMARY

The residual heat removal system (RHR) at R. E. Ginna Nuclear Power Station was evaluated for leak-before-break (LBB)behavior in accordance with the NRC GDC-4 and NUREG-1061, Vol. 3.

The RHR lines considered in this evaluation are adjacent to the hot and cold legs of the reactor coolant system (RCS). They are 10-inch Schedule 160 piping, fabricated from Typ'e 316 stainless steel. The operating pressure for the RHR lines is 2235 psig and the operating temperature was conservatively chosen as'612.2'F forthe evaluation.

'BB was demonstrated forthe above piping in accordance with NRC margins. In this evaluation, circumferential flaws were considered since these are more limitingthan axial flaws. The evaluation consisted ofdetermining critical flaw sizes at selected locations on the piping in the vicinityofthe Component Cooling Water (CCW) piping to the reactor support coolers. The critical flaw sizes were calculated using the elastic-plastic fracture mechanics (EPFM) J-IntegraVI'earing Modulus (JfZ) approach, Leakages were then calculated through half the critical flaw sizes per the requirements ofNUREG-1061. The leakage evaluation was done for the affected nodal locations in the piping mathematical models provided by Rochester Gas &Electric Corporation (RG&E).

The predicted leakage for all the locations on the RHR lines considered in evaluation was at least 4.7 gpm considering the required NRC safety factor of2 between the critical flaw size and the leakage flaw size. This leakage should be easily detected by the present leak detection system at Ginna.

A fatigue crack growth analysis was performed to study the predicted behavior ofpostulated semi-elliptical, inside surface flaws. Postulated circumferential flaws of 15% ofthe pipe wall in depth, and with an aspect ratio (length to depth) equal to or larger than 10, were shown to grow an insignificant amount in depth and length during 40 years. The above postulated flaw sizes are slightly in excess ofthe maximum size permitted by ASME Code,Section XI,IWB-3514, and are conservative since such flaws would have been repaired during the preservice inspection.

Postulated flaws deeper than 15% ofthe wall were also studied, and shown to grow preferentially I

Structural IntegrityAssociates, inc.

through the pipe wall and result in leakage, rather than to extend an unacceptable amount'in length.

This result further validates the application ofLBB methodology to the prevention ofpipe rupture for this system.

The effect of degradation mechanisms which could impact the LBB evaluations were considered in

\\

the evaluation. It was determined that the probability ofwater hammer occurrence in the affected portions ofRHR piping is very low. In addition, RG&E has utilized EPRI guidelines and research

~ I results to prevent or mitigate water hammer in Ginna systems.

Corrosion is not an expected failure mechanism forthe 'system evaluated based on plant experience and RG&E's continuing erosion-corrosion monitoring program.

Structural IntegrityAssociates, inc.

Table of Contents

~Pa e

1.0 INTRODUCTION

l.i Background.

1.2 Leak-Before-Break Methodology.............

~ ~ ~

0 ~ ~ 00 ~ ~ ~ ~ ~ ~ 0 ~

~ 0 ~ ~ ~

0 ~ ~ 0 ~ 0 ~

~

~ ~

0 2.0 CRITERIAFOR APPLICATIONOF LEAK-BEFORE-BREAKAPPROACH...................... 2-1 2.1 Criteria forThr'ough-Wall Flaws 2-1 2.2 Criteria for Part-Through-Wall Flaws 2-2 2.3 Other Mechanisms.

2-2 3.1 Water Hammer.........................

3.2 Corrosion.....

3.3 Fatigue..................................

0 0

~

0 ~

~ ~ ~ 0 ~ ~ ~ 0

~ ~ ~ ~ ~ ~ ~ 0 ~ 0 ~ ~ 0 ~ 0 ~ ~ ~ ~ ~ ~ ~ ~ ~

~ ~ ~ ~ ~ ~ ~

~

0 ~ 0 ~ ~ ~ ~ 0 ~ ~ ~ ~ ~ ~ ~ ~ 0 ~ ~ ~

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~

~

~ 0 ~ ~ ~.. 3-2 3.0 CONSIDERATION OF. WATERHAMMER,CORROSION ANDFATIGUE............~......... 3-1 4.0 PIPING MATERIALSANDSTRESSES..................--. -.

4.1 Piping System Description.

4.2 Material Properties..............................------

4.3 Piping Stresses.

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ 0 4 1

~

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ 0 ~ ~ ~ ~ ~ ~

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ 0 ~ ~ ~ F 0 4 1

~

~ ~ ~ ~

~ 0 ~ 0 ~ 0 ~ 0 ~ ~ 0 ~ 0 ~ 0 ~

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ 0 ~ ~ 4 2 5.0 LEAK-BEFORE-BREAKEVALUATION~

5.1 Evaluation ofCritical Flaw Sizes..

5.2 Leak Rate Determination 5,3 LBB Evaluation Results and Discussions.........

~\\

~

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~

~

0 ~ ~ 0 ~ ~ ~ ~ ~ ~ ~ ~ 0 5-1

~ ~ 0 ~ 0 ~

0 ~\\ ~ ~ 0 ~ ~ ~ ~

~ ~ ~ ~ ~ ~

0 5-8 6.0 EVALUATIONOF FATIGUECRACKGROWTH OF SURFACE FLAWS......................... 6-1 7.0

SUMMARY

ANDCONCLUSIONS........~..........,......

~..-.

~

0 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ 00 ~ 0 ~

~ ~

~ ~

~

8.0 REFERENCES

. 8-1 SIR-97-077, Rev. 0 iv Structural IntegrityAssociates, inc.

ListofTables

~Pa e 4

I Table 4-1 Material Constants Used for Type 316 Stainless Steel in LBB Evaluation......~..........4-3 Table 4-g Calculation of Stresses for RHR Pipe Run from Valve 700 to Hot Leg.....................4-4 Table 4-3 Calculation of Stresses for RHR Pipe Run from Valve 721 to Cold Leg..............~.....4-5 Table 5-1 Summary ofCritical Flaw Sizes

~

~

~ ~

o5 9 Table 5-2 LBBEvaluation Results for Detectable Leakage....

~ ~ ~ ~

~ ~ ~ ~

5 10 Table 6-1, Plant Design Transients for RHR Piping..........~...........................................6-5 Table 6-2. Combined Transients For Fatigue Crack Growth Evaluation For RHR Line djacent to Hot Leg.........

A '

~ ~

~ 0 ~ 0 ~ 1 ~ ~ ~ ~ ~ 0 ~ 0 ~ 6 6 Table 6-3 Combined Transients For Fatigue Crack Growth Evaluation For RHR Line A 'jacent to Cold Leg. ~., ~

~ ~

~

~ ~ 0 ~ ~ ~ ~ ~ ~ ~

~ ~ ~ ~

~

~

~ ~ 0 ~ 0 ~ ~ ~ ~ 0 ~ ~ 0 ~ ~ ~ ~ ~

~ 6 7 Table 6-4 Combined Maximum and Minimum Stresses forFatigue Growth Analysis for RHR Line Adjacent to HotLeg....,

........,..............,.....................6-8 Table 6-5 Combined Maximum and MinimumStresses for Fatigue Crack Growth Analysis for RHR Line Adjacent to Cold Leg..............,.....................,.......................................

6 9 Table 6-6 Results ofFatigue Crack Growth...

....6-10 SE-97-077, Rev. 0 Structural IntegrityAssociates, Inc.

List ofFigures

~Fi ere Figure 1-1. Location ofRHR Line Adjacent to the Hot Leg..

~Pa e

~ ~ ~ ~ ~ ~ e 1A Figure 1-3. Location of the RHR Line Adjacent to the Cold Leg (Part A)

. 1-5 Figure 1-3. Location of the RHR Line Adjacent to the Cold Leg (Part B)..~.......................1-6 Figure 1-4. Representation of-Postulated Cracks in Pipes for FractureMechanics Leak-Before-Break Analysis.......

. 1-7 Figure 1-5. Illustration ofISI (UT)/Leak Detection Approach to Protection Against P'pe Rupture.

~

~

~

~ ~ e ~ e ~ ~ ~ ~ ee ~ e ~ ~ ~ ~ ~

~

~

dies

~ ~

~ ~

~ ~ e ~ ~ e ~ e ~ ee 1 8 Figure 1-6. Leak-Before-Break Approach Based on Fracture Mechanics Analysis with In-service Inspection and Leak Detection 1-9 Figure 5-1. Flow of Subcooled Water Through a Crack,.....

5-11 Figure 5-2. Leak Rate Versus Crack Size for RHR Pipe Run from Valve 700 to Hot Leg......5-12 Figure 5-3. Leak Rate Versus Crack Size for RHR Pipe Run from Valve 721 to Cold Leg....5-13 SIR-97-077, Rev. 0 Structural IntegrityAssociates, Inc.

1.0 INTRODUCTION

1.1 Background

This report documents evaluations performed by Structural Integrity Associates (Sl) to determine the leak-before-break (LBB)capabilities ofseveral locations on the residual heat removal (RHR)

System at R. E. Ginna Nuclear Power Station (Ginna). These evaluations are necessary because a pipe break at these locations could potentially affect the structural integrity of Component Cooling Water (CCW) piping to the reactor support coolers per Reference 29.

Two portions of the RHR line are considered in the evaluation and are shown in Figures 1-1 through 1-3 [1]. The first portion includes the piping from the hot leg ofthe reactor coolant system (RCS) to motor operated valve (MOV)700 (Node points 680 through 70 in Figure 1-1). The second portion extends from the RCS cold leg to MOV721 (Nodes points 960 through 8400 in Figures 1-2 and 1-3).

k 1.2 Leak-Before-Break Methodology NRC SECY-87-213 [2] covers a final broad scope rule to modify General Design Criterion 4 (GDC-4) ofAppendix A, 10 CFR Part 50. This amendment to GDC-4 allows exclusion from the design basis ofdynamic'effects associated with high energy pipe rupture by application ofLBB technology.

Definition of the LBB approach and criteria for its use are provided in NUREG-1061 [3]. Volume 3 ofNUREG-1061 defines LBB as "...the application of fracture mechanics technology to demonstrate that high energy fluidpiping is veiy unlikely to experience double-ended ruptures or their equivalent as longitudinal or diagonal splits." The particular crack types ofinterest include circumferential through-wall cracks (TWC) and part-throu )~1-wall cracks (PTWC), as well as axial or longitudinal through-wall cracks (TWC), as shown in Figure 1-4.

SIR-97-077, Rev.0 Structural tntettrlttrAssociatestnc,

LBB is based on a combination of in-service inspection (ISI) and leak detection to detect cracks, coupled with fracture mechanics analysis to show that pipe rupture willnot occur for cracks smaller than those detectable by these methods. A discussion ofthe criteria for application ofLBB is presented in Section 2 ofthis report, which summarizes the NUREG-1061 requirements.

The approach to LBB which has gained acceptance for demonstrating protection against high energy line break (HELB) in safety-related nuclear piping systems is schematically illustrated in Figure 1-5. Essential elements of this technique include critical flaw size evaluation, crack I

propagation analysis, volumetric nondestructive examination (NDE) for flaw detection/sizing, leak detection, and service experience.-In Figure 1-5, a limitingcircumferential crack is modeled as having both a short through-wall component, and an axisymmetric part-through-wall crack component.

Leak detection establishes an upper bound for the through-wall crack component while volumetric ISI limits the size ofundetected part-through-wall defects. These detection methods complement each other, since volumetric ISI techniques are well suited to the detection oflong cracks while leakage monitoring is effective in detecting short through-wall cracks. The level ofISI required to support LBB involves volumetric inspection at intervals determined by fracture mechanics crack growth analysis, which would preclude the growth ofdetectable part-through-wall cracks to a critical size during an'inspection interval. The objective of this fatigue evaluation is to limitpotentially undetected defect sizes to those which would be allowed under ASME Section XI rules. For through-wall defects, crack opening areas and resultant leak rates are compared with leak detection limits.

The net effect ofcomplementary leak detection and ISI is shown by the shaded region ofFigure 1-5 as the largest undetected defect that can exist in the piping at any given time. Critical flaw size evaluation, based on elastic-plastic fracture mechanics techniques, is used to determine the length and depth of defects that would be predicted to cause pipe rupture under specific design basis loading conditions, including abnormal conditions such as a seismic event and including appropriate safety margins for each loading condition. Crack propagation analysis is used to determine the time interval in which the largest undetected crack could grow to a size which would impact plant safety margins. Asummary of the elements for a leak-before-break analysis is shown SIR-97-077, Rev.0 1-2 Structural IntegrityAssociates, Inc.

in Figure 1-6. Service experience, where available, is useful to confirm analytical predictions as well as to verify that such cracking tends to develop into "leak" as opposed to "break" geometries.

In accordance with NUREG-1061, Volume 3 [3] and other NRC guidance on the topic, the leak-before-break technique for high energy piping systems in a nuclear power plant should include the followingconsiderations.

r

~

Elastic-plastic fracture mechanics analysis ofload carrying capacity ofcracked pipes under worst case normal loading, with safe-shutdown earthquake (SSE) loads included. Such analysis should include recent elastic-plastic fracture data applicable to pipe weldments and weld heat affected zones where appropriate.

~

Consideration ofpipes under limitload conditions for the piping system, as applicable.

~

Linear elastic fracture mechanics analysis of subcritical crack propagation to determine ISI (in-service inspection) intervals for long, part-through-wall cracks.

Piping stresses have a dual role in LBBevaluations.

On one hand, higher maximum (design basis) stresses tend to yield lower critical flaw sizes, which result in smaller flaws for leakage and a lower leakage rate. On the other hand, higher operating stresses tend to open cracks more for a given crack size and create a higher leakage rate. Because ofthis duality, the use of a single maximum stress location for a piping system may result in a non-conservative LBB evaluation. This LBB evaluation will,therefore, be performed in such a manner that the affected nodal locations for the piping models of the RHR lines willbe specifically addressed.

SIR-97-077, Rev.0 1-3 Structural Integrity Associates, lnc.

0 ISO 822I 0

2IO 0

yO Sg, K

l.4l7 l2ll7 y5

~> 2215/

3, tll llO Pgl 3$

< CONhl.

kl c351 3 S4ggy. ll

~t'nw 1-1, Zoca~ ~ Line Adjacent to g ot~g SfrugfIJIPf/jyfp~~gy~

SSOCI8feS /gg

I yl le y5 yO llu+O e)g.

R)t CO@7. SEE Pcnu A8Qoe'g,9

)bi QA(g,1d )

57 GIU-52

~,C~~w~ cP

S~e~~i 0 ~~~

co l030 72l 22i5 '

0 0Iq U

560 I'%NcnNN. ~

Sm,bake "V ON C.'M,l-~a%

SH'K7 Q.

/yQ

~o +os

~ 2'pp 8pI

@gal C~

Og 4~

<rr,

'>rp

<@br ASaa C

8 gO i,o lO L

go r

720 2215(

lt 480 A-t,o+p,4 ~

'-..(

Figure 1-2.

~

Location ofthe'HR Line Adjacent to the Cold Leg (Part A)

SE-97-077, Rev.0 1-5 Structural IntegrityAssociates, Inc.

e~r fA

$.)gp pQ

+0 b>

~g pfTA yj,l

~ gyre<~

gao bqq

.P a

867A I I O'S<~

8

~g Z 51

/2843 W.ICo7R si ibm 770 1ao 8p I

gr5'p

)pic 4)~

lm,oo

+yr

~@4 tG pf

~P p

gpss

>st) g' gQ

~ll ~355' gS<~

C'~r>>

Figure 1-3.

Location of the RHR Line Adjacent to the Cold Leg (Part B)

SIR-97-077, Rev.0 1-6 Structural IntegrityAssociates, Inc.

e 2a 28 R

2a E

~rrQP~k~yw

a. Grcumferential and Longitudinal Through-Wall Cracks ofLength 2a.

b. Circumferential 360'Part-Through-Wall Crack ofDepth a.

Figure 1-4.

Representation of Postulated Cracks in Pipes. for Fracture Mechanics Leak-Before-Break Analysis S1R-97-077, Rev.0 1-7 Structural IntegrtttrAssociatesinc,

0 1.0 0.9 (0

CO o.s 0

0.7 a.

0.6 UJ Cl o.5 K

0.4 tX:

0.3 0-0.2 X

IS[

Axisym.

Depth Thru-Wall Length Critical Flaw Size Locus 0.1 0.0 0

Leak Detection I

0.2 0.4 0.6 O.s THRV-WALLCRACK LENGTH CIRCUMFERENCE e

Figure 1-5.

Illustration ofISI (UT)/Leak Detection Approach to Protection Against Pipe Rupture SIR-97-077, Rev.0 1-8 Structural tntegrity Associates, inc.

Piping Stress Analysis Leak Detection System FRACTURE MECHANICS ANALYSIS

~ Critical crack size

~ Leak rates

~ ISI intervals Crack Detection Before Pipe Ruptures In-Service inspection 93371%

Figure 1-6.

Leak-Before-Break Approach Based on Fracture Mechanics Analysis with In-service Inspection and Leak Detection SIR-97-077, Rev.0 1-9 Structural IntegrityAssociates, Inc.

2.0 CRITERIAFOR APPLICATIONOF LEAK-BEFORE-BREAK'APPROACH NUREG-1061, Volume 3 [3] and GDC-4 (SRP 3.6.3) [2] identify several criteria to be considered in determining applicability of the leak-before-break approach to piping systems.

Section 5.2 ofReference 3 provides an extensive discussion of the criteria forperforming leak-before-break analyses.

The details of that discussion willnot be repeated here, but a summaiy of various requirements as applied to evaluation of the RHR line at Ginna is provided below.

2.1 Criteria for Through-Wall Flaws Acceptance criteria for critical stresses and critical flaws are:

1.

The flaw which is required to produce an "acceptable leakage rate" is smaller than the critical flaw length associated with the maximum stress (with SSE) by a factor of2.

2.

The stress required to make the "acceptable leakage rate" flaw critical is greater than the maximum stress (with SSE) by a factor of at least E2.

3.

The net section collapse criterion (NSCC) approach may be used to compute the critical flaw size provided a safety factor of 3 is placed on normal service stresses.

It has been found in previous LBB evaluations conducted by Structural Integrity Associates (SI),

that the second and third criteria stated above are not bounding. The method described in the first criteria provides a smaller leakage rate than the second criteria as was demonstrated in the LBB evaluation previously performed for high energy lines outside containment at Ginna [4]. Therefore, only the first criteria willbe considered in this report. Furthermore, the elastic-plastic fracture mechanics (EPFM) approach is generally conservative relative to the NSCC approach when applied to ferritic piping. Therefore, only EPFM principles willbe applied in this evaluation.

SIR-97-077, Rev. 0 Structural /ntegrity Associates, inc.

2.2 Criteria for Part-Through-Wall Flaws

~

~

NUREG-1061, Volume 3 [3] requires demonstration that a long part-through-wall flaw which is

, detectable by ultrasonic means willnot grow due to fatigue to a depth which would produce instability. over the lifeof the plant. This is demonstrated in a later section of this report, where the analysis ofsubcritical crack growth is discussed.

~,

29 Other Mechanisms NUREG-1061, Volume 3 [3) and GDC-4 [2] limitapplicability of the leak-before-break approach to those locations where degradation or failure by mechanisms such as water hammer, erosion/corrosion, fatigue, and intergranular stress corrosion cracking gGSCC) is not a significant possibility. These mechanisms were considered forthe RHR line at Ginna, as reported in Section 3 ofthis report.

SIR-97-077, Rev. 0 2-2 Structural IntegrityAssociates, inc.

3.0 CONSIDERATION OF WATER HAMMER,CORROSION ANDFATIGUE NUREG-1061, Volume 3 [3] and GDC-4 [2] state that LBB should not be applied to high energy lines susceptible to failure from the effects ofwater hammer, corrosion or fatigue. These potential failure mechanisms are thus discussed below with regard to the RHR line at Ginna,. and it is concluded that the above failure mechanisms do not invalidate the use ofLBB for this piping system.

3.1 Water Hammer Acomprehensive study performed in NUREG-0927 [5] indicated that the probability ofwater hammer occurrence in the affected portions ofthe RHR system ofa PWR is very low. In addition, RG&E has utilized EPRI guidelines and research results (References 30 and 31) to prevent, mitigate or accommodate water hammer events in Ginna systems.

3.2 Corrosion Two corrosion damage mechanisms which can lead to rapid piping failure are intergranular stress corrosion cracking (IGSCC) in austenitic stainless steel pipes and flow-assisted corrosion (erosion-corrosion) in carbon steel pipes. IGSCC has principally been an issue in austenitic stainless steel piping in BWRs [6] resulting from a combination of tensile stresses, susceptible material and oxygenated environment. IGSCC is not typically a problem forthe primary loop of a PWR such as the RHR system under consideration since the environment has relatively low concentrations of oxygen.

Erosion-corrosion is not anticipated to be a problem for this system since it is fabricated from stainless steel piping which is not susceptible to erosion-corrosion.

SIR-97-077, Rev. 0 3-1 Structural IntegrityAssociates, inc.

3.3 Fatigue

~

~

Known fatigue loadings and the resultant possible crack growth have been considered by the analyses reported in Section 6.0 ofthis report. Based on these results, it is concluded that fatigue willnot be a significant issue for the RHR piping at Ginna.

SIR-97-077, Rev. 0 3-2 Structural IntegrityAssociates, inc.

e

4.0 PIPING MATERIALSANDSTRESSES 4.1 Piping System Description The mathematical models for the RHR piping system at Ginna are shown in Figures 1-1 through 1-3 [7,8]. The piping is fabricated from SA-376 Type 316 stainless steel. The welds are fabricated using either the submerged arc welding (SAW) or shielded metal arc welding (SMAW)processes.

- The lines are fabricated from 10-inch schedule 160 piping. The operating pressure for the RHR lines is 2235 psig and the operating temperature was conservatively chosen as 612.2'F to correspond to the temperature at which the RHR piping analysis was performed [32]. Use of the actual operating temperature of550'F would yield larger critical flaw sizes and hence higher leakage rates.

4.2 Material Properties The material properties used forthe LBB evaluation are shown in Table 4-1. The elastic modulus (E), lower bound yield strength (a,) and ultimate strength (a) are taken from the Appendices of Section IIIofthe ASME Boiler and Pressure Vessel Code [10] at the operating temperature.

The flow stress is computed as an average of the yield stress and ultimate stress, although this does not influence the crack stability results. The true stress-strain curve is represented by the Ramberg-Osgood power law equation in the form:

a/e,, =a/a, +u(a/a,)"

The Ramberg-Osgood true stress, true strain parameters, a and n, were obtained using the relationship from Reference 27 as:

1 n=

ln(1+e)

SIR-97-077, Rev. 0 Structural IntegrityAssociates, lnc.

ln(1+e) a(1+e) 1n(1+a /E '(1+a./E) a(1+e) a(1+ a /E)

The J-integral versus crack extension (J-R) curves for flaw instability computations used in this evaluation represent the lower bound generic toughness values provided in the EPRI Ductile Fracture Handbook [9],for stainless steel weldments..For the critical fiaw evaluations, the J-R curve is input in the form ofa power law, as shown below:

J

=

C(ha)"

ha

=

crack extension The values for C and N obtained from Reference 9 are shown in Table 4-1.

4.3 Piping Stresses The piping stresses which are normally considered, in a LBB evaluation are due to pressure, dead weight, thermal expansion and Safe Shutdown Earthquake (SSE). Summaries ofthe pipe stresses for the RHR line are shown in Tables 4-2 and 4-3. These stresses are used to calculate the critical flaw size and the leakage rate through one-half the critical flaw size. For calculation ofcritical flaw size, the stress combination ofpressure, deadweight, thermal and SSE loads is used. For leakage calculations, the stress combination ofpressure, deadweight and thermal loads is used.

These stress combinations are shown in Tables 4-2 and 4-3 for the various nodal locations. These piping stresses are listed by their piping model node numbers, which are shown in Figures 1-1 through 1-3.

These node numbers, in general, correspond to the weld locations along the piping system.

Stress intensification factors based upon B31.1 piping Code for the RHR piping [28] were calculated and extracted from the stresses obtained from the piping stress reports [7,8]. This is justified because for the fracture mechanics evaluation, it is the stress in the weld which is of interest, and not that in the adjacent component.

The modified stresses excluding stress intensification factors are also SIR-97-077, Rev. 0 Structural IntegrityAssociates, Inc.

Table 4-1 Material Constants Used for Type 316 Stainless Steel in LBB Evaluation Property E (ksi)

~, (ksi)

(=ay) au (ksi) aflo (ksi)

Jt, (in-kipfin)

J~(in-kipfin )

N Value 25,240 (1) 18.8 (1) 71.8 (1) 45.282 (2) 0.776 (3) 3.81 (3) 0.99 (4) 5.0 (5) 6.033 (4) 0.391 (4)

Notes:

(1) Taken from Reference 10 at the operating temperature ofthe RHR system.

(2) Average ofa, and 0' (3) Determined using the procedure in Ref. 27.

(4) Taken from Reference 9.

(5) Maximum value used in the analysis.

SIR-97-077, Rev. 0 Structural IntegrityAssociates, Inc.

Input Stresses, psi Table 4-2 Calculation of Stresses for RHR Pi e Run from Valve 700 to Hot Le Pipe Run from Valve 700 to Hot Leg Calculated Stress, ksi IntensifiedStresses Unintensijied Stresses Load Combination Node Type Deadweight (DW) +

Pressure (P)

Therm al (TH)

DW+P+

SSE P

DW TH SSE i*0.75 P'W TH SSE P+DW+ P+DW+

TH+ SSE TH 680 30'aper Transit-ionn 6047 3121 18011 4.145 3.728 1.902 3.121 11.964 1.534 1.151 4.145 3.728 1.653 2.034 10.396 17.&11 7.415 50 Elbow 7030 60 Elbow 6263 1165 1719 15997 4.145 3.728 15644 4.145 3.728 2.885 2.118 1.165 8.967 1.719 9.381 1.000 4.145 3.728 1.000 4.145 3.728 2.885 1.048 2.118 1.547 8.967 9.381 16.628 7.661 16.774 7.393 70 304 Taper Transt tion 6197 1397 15402 4.145 3.728 2.052 1.397 9.205 1.534 1.151 4.145 3.728 1.783 0.910 7.9019 14.420 6.422 Notes:

D 2 I) Pressure Stress for Design Conditions (based upon pressure of24&5 psig); calculated with the followingequation: P

'here D, = outer diameter = 10.75 and D;=

D2 D2 r diameter = &.5.

rmal stress for Normal Operating Conditions.

D 2

3) 59cssurc Stress for Normal Operating Conditions (based upon pressure of2235 psig); calculated with the followingequation P

Do -D

4) stress intensity factor i, calculated for30'aper transition with the followingequation:

i = l.9 max or i = 1.3+ 0.0036DJt+ 0.225/t

~Where D, = 10.75 in. and t =.thickness = I.I25 in.

5)@ress intensity factor i, calculated for welding elbow with the followingequation: i =0.9/h

= I.Illwhere h = TR/(r) = 0.72&6; I'= nominal wall thickness = 1.125 in., R = hend radius = 15 in., and r = mean radius =4&125 in.

6)&0.75 cannot be <I.

7)Pll input stresses taken from References 7 and 8.

&)gr 93 I.I, thermal stresses do not include 0.75 multiplier.

9) Q:r 83 I.I, prcssure strcsscs do not include stress intensity factor, i.

Cb SIR-97-077, Rev. 0 4Q

Input Stresses, psi Table 4-3 Calculation ofStresses for RHR Pi e Run from Valve 721 to Cold Le Pipe Run from Valve 721 to Cold Leg Calculated Stress, ksi Node Type Deadweight (DW) +

Pressure (P)

Thermal (TH)

DW+ P+

SSE Intensified Stresses DW SSE i

i*0.75 p

s Unintensified Stresses DW Load Com SSE P+ DW+

TH+ SSE bination P+ DW+

TH 8400 910 920 930 950 960 Weld-o-let Elbow Elbo Elbo Elbow Weld-o-lct 7794 6868 5416 5735 5417 5209 9927 11560 8887 7066 5931 7038 10154 4.145 3.728 9170 4.145 3.728 9050 4.145 3.728 8715 4.145 3.728 7740 4.145 3.728 7451 4.145 3.728 3.649 2.723 1.271 1.590 1.272 1.064 9.927 11.560 8.887 7.066 5.931 7.038 2.360 1.534 2.302 1.111 3.634 1.111 2.980 1.111 2.323 1.111 2.242 1.534 1.151 1.000 1.000 1.151 4.145 3.728 3.171 4.145 3.728 2.723 4.145 3.728 1.271 4.145 3.728 1.590 4.145 3.728 1.272 4.145 3.728 0.924 6.470 2.051 15.419.

10.400 2.302 19.153 7.996 3.634 16.629 6.357 2.980 14.655 5.336 2.323 12.659 4.587 1.948 11.188 13.368 16.851 12.995 11.675 10.336 9.239 Notes D 2 1),Pressure Stress for Design Conditions (based upon pressure of2485 psig); calculated with the followingequation:

P

'here D, = outer diameter = 10.75 and Di =

Do D

/acr diatncter = 8.5.

2)~rmal stress for Normal Operating Conditions.

D 2

3) gessure Stress for Normal Operating Conditions (based upon pressure of2235 psig); calculated with the followingequation:

. P Do Di

4) press intensity factor i, calculated for30'aper transition with the followingequation: i = 1.9 max or i = 1.3+ 0.0036DJt+ 0.225/t where D, = outer diameter = 10.75 in. and t

~ thickness = 1.125 in.

5) press intensity factor i, calculated for welding elbow with the followingequation: i = 0.9/h

I.II I where h = TR/(r) = 0.7286; T = nominal wall thickness = 1.125 in., R

Wend radius = 15 in., and r = mean radius =4.8125 in.

6)4 0.75 cannot be <l.

7) II input stresses taken from References 7 and 8.

8) @r B31.1, thermal stresses do not include 0.75 multiplier.

9) Qr B31.1, pressure stresses do not include stress intensity factor, i.

SIR-97-077, Rev. 0 4-5

5.0 LEAK-BEFORE-BREAKEVALUATION The LBB approach involves the determination ofcritical flaw sizes, critical stresses and leakage through flaws. The critical flaw length for a through-wall flaw is that length for which, under a given set of applied stresses, the flaw would become marginally unstable.

Similarly, the critical stress is that stress at which a given flaw size becomes marginally unstable.

NUREG-1061, Volume 3 [3] defines required margins ofsafety on both flaw length and applied stress.

However, as explained in Section 2, safety margins based on flaw length have been found in previous evaluations to be the more conservative of the two and therefore, only the criterion based on flaw length willbe used in this evaluation.

Furthermore, previous evaluations [4] have demonstrated that circumferential flaws are more restrictive than postulated axial flaws. For this reason, the evaluation presented herein willbe based on assumed circumferential flaws.

5.1 Evaluation of Critical Flaw Sizes Critical flaw sizes may be determined using net section collapse criterion (NSCC) approach or J-IntegraUTearing Modulus (JfP) methodology. NSCC is particularly suited formaterials with a considerable amount ofductility and toughness such as stainless steel materials, since it assumes that the cross-section of the pipe becomes fullyplastified at the onset of failure. As such, for circumferential flaws, NSCC is less conservative compared to the JfZ methodology which is based on elastic-plastic fracture mechanics (EPFM) principles. The conservatism in the use ofEPFM was demonstrated on previous LBB evaluations for Ginna [4] and other similar evaluations performed by SL In this evaluation, the critical flaw sizes willtherefore be determined based on the J/I'pproach.

A procedure for using this approach for the assessment ofthe stability ofthrough-wall circumferential flaws in cylindrical geometries such as pipes is presented in References 11 and 12.

This procedure was used for the determination ofcritical stresses and flaw sizes in the RHR piping at Ginna, using SI's computer program, pc-CRACK

[13].

SIR-97-077, Rev. 0 5-1 Structural IntegrjtyAssociates, inc.

The expression for the'J-integral for a through-wall circumferential crack under tension loading

[18] which is applied in this analysis is:

R p2 a

a R

P J=fi a; +txa<e,c hi ri-t E

b b

t p,

where a

R a F R

b fi acc-t 4KR t ac P

a, 2a 2b hi Po effective crack length including small scale yielding correction nominal pipe radius pipe wall thickness elasticity factor [18,19]

applied load = tr~2z Rt; where g is the remote tension stress in the uncracked section Ramberg-Osgood material coefficient elastic modulus yield stress yield strain b-a crack length 2mR plasticity factor [11, 12]

limitload coriesponding to a perfectly plastic material Ramberg-Osgood strain hardening exponent.

Similar equations [11, 12] are used to compute critical flaw sizes for circumferential TWCs under g

bending stresses.

Crack extensions during stable ductile tearing in the EPFM analyses are SE-97-077, Rev. 0 5-2 Structural IntegrityAssociates, Inc.

I cons'ervatively not included in the critical flaw'length computations.

The piping stresses consists of both tension and bending stresses.'he tension stress is due to'internal pressure while the bending stress is caused by deadweight, thermal and seismic loadings. The critical flaw sizes (lengths) obtained with the tension model (a, ) and the bending model (ab) are combined to determine the actual critical flaw size (a,) due to a combined tension and bending stress using linear interpolation, as described by the followingequation:

<r 0'b c

I

+

b b++t rsb+

The results of the critical flaw size determination are presented in Table 5-1.

5.2 Leak Rate Determination The determination ofleak rate is performed using the Structural Integrity Associates program, pc-LEAK[14]. The methodology employed in pc-LEAKinvolves the determination ofcrack opening area (COA), assuming plasticity at the crack tip. Then, the flow rate is determined based on classical thermal-hydraulic expressions for single and two-phase flow.

Crack opening area under the influence ofsteady-state operating stress (combined tension and bending) is computed from References 15 and 16 as:

A =(<R )I<(8) 1+

a<

2 crab 3+COS8 E

a(

4 where crack opening area (in ) including plastic zone correction, assuming plane stress cz,

=

steady-state tension stress (psi) steady-state axial bending stress (psi)

E

=

elastic modulus (psi)

SIR-97-077, Rev. 0 5-3 Structural IntegrityAssociates, Inc.

R

=

nominal pipe radius (in.), and 8

=

the angle describing half the through-wall crack length (radians).

The term I,(8) is computed for varying R/t (pipe radius/thickness) in accordance with the equations ofReference 16.

The plastic zone correction for the effect ofyielding near the crack tip is incorporated by the

~

followingequation [15]:

2 8 +

KtOtel 2'y where 8,

=

effective half-length ofangle through-wall crack, assuming plane stress K(p<

stress intensity factor due to combined tension and bending cz=

reference stress In this evaluation, the flowstress which is the average ofyield and ultimate strength was appropriately used as the reference stress.

The flowrate through the crack is based on classical thermal-hydraulic methodology. The development of the approach is detailed in the followingsection. The methodology includes considerations ofboth liquid and vapor fiowofwater, including the consideration oftwo phase flow within the crack.

The crack is considered to have a total length of2a, either around the circumference or axially along the pipe wall. The crack has an average opening width w, and the flowpath length through the wall is taken as L.

SIR-97-077, Rev. 0 Structural IntegrityAssociates, inc.

4A DH =.

P where:

DH

=

hydraulic diameter A

=

cross sectional area P

=

perimeter.

For a narrow crack of length 2a, 4xA A

(2) (2a) a If w is the average crack opening width, then A=2aw DH =2W The frictional loss in the constant area channel willbe assumed to be that between parallel plates with a surface roughness.

The parameter ofinterest to characterize the flow resistance per unit of area is:

fL Keir + Kcxit= ~ Ki+

+ Kcxit DH where:

ff

=

effective total pressure loss coefficient K;

=

individual discontinuity total pressure loss coefficient f

=

friction factor L

=

flowpath length, (pipe wall thickness)

DH

=

hydraulic diameter K<<.=

exit loss coefficient = 1.0.

SE-97-077, Rev. 0 5-5 Structural IntegrItyAssociates, Inc.

The pressure loss coefficients for the entrance and flowdirection changes must be computed separately from the friction loss parameters.

For example, Reference 18 recommends a

discontinuity loss coefficient of0.5 for a sharp entrance crack with gaseous flow. Reference 19 recommends a value of2.7 to properly account for the vena contracta (reduction in cross section) when dealing with near saturated water entering a narrow crack.

t The friction factor for turbulent flow (Reynolds number > 4000) is determined from Reference 20:

1 8

2.52

= - 2 log,o (

+

~f

'.7DH Re/f.

where:

f

=

frictionfactor-8

=

surface roughness DH

=

hydraulic diameter Re

=

Reynolds number.

For laminar flowbetween parallel plates, Reference 21 recommends:

I'6 f=-

Re which occurs below about R, = 2000. In the transition range between 2000 < Re < 4000, a best estimate friction factor is used.

In the turbulent equation, an iterative approach must be taken to solve forthe friction factor.

Iteration is also required to determine the frictionfactor in the transition regime ofReynolds number.

Reference 19 recommends a value of5 pm (0.000197 inches) for the surface roughness offatigue cracks. For more tortuous paths, and extremely small crack opening displacements, additional losses might be input with increased values for K. However, this effect willbe quite small for crack opening widths which willproduce detectable leakage in a power piping system.

SIR-97-077; Rev. 0 5-6 Structural IntegrityAssociates, inc.

For the pipe region filledwith subcooled water, the flow can be determined by standard l

incompressible fiow methodology. For saturated stealn flow, the mass flow rate versus inlet total pressure may be determined directly from the charts offL/Dfrom Reference 22. Similarly, Reference 22 provides charts for the blowdown of water and steam-water mixtures. These are incorporated as tables in pc-LEAKI'14].

)

I In evaluating the flowof subcooled water, which flashes as the static pressure reaches saturation, a two-step approach is used. For the subcooled portion ofthe flow, the incompressible flowequation is used; fL PT,inlet Psnt = (Kinlet+I 0+

) Ii2P V DH where PT,i~et Psst Kinet 1.0 pressure inside pipe saturation pressure associated with water temperature in pipe liquid density velocity inlet plus discontinuity loss coefficient total to static pressure loss coefficient at the downstream end of the flow.

From this equation, the length (fLl/D)ofchannel to bring fluidfrom its subcooled condition to a flashing saturated mixture may be determined as a function of mass flux. This is illustrated in Figure 5-1.

In length L2, a two-phase homogeneous mixture flows and this length may be determined for saturated water from the Reference 22 charts. For smail values of~/D, the saturation flashing point may occur just at the exit of the crack, such that the flow can be approximately determined solely based upon flowofliquid water. When the inlet pressure is near saturation pressure, the flow

" may be approximately determined from the Reference 22 charts. In between, a combined flow situation exists.

SIR-97-077, Rev. 0 7 Structural IntegrityAssociates, Inc.

0

The leakage was calculated for an operating pressure of2235 psig and a temperature of612.2'F.

C Parametric evaluations showed that use oflower temperatures would produce higher leak rates.

The leakage results are presented graphically in Figures 5-2 and 5-3 as a function ofcrack size (2a) for the various locations on the hot leg side, as well as the cold leg side. Table 5-2 shows the predicted leakage as a function of the critical flaw size for each location.

5.3 LBB Evaluation Results and.Discussions As can be seen from Table 5-2, the calculated leakage through half the critical flaw size for locations adjacent to the hot leg is at least 4.7 gpm considered in this evaluation. The leakage increases to at least 16.5 gpm at these locations ifthree-quarters of the critical flaw size is considered.

Due to relatively high thermal stresses at the locations near the cold leg, the leakage through half the critical flaw size is relatively large (at least 13.4 gpm). This increases to at least 44.7 gpm when three-quarter the critical flaw size is considered. Itis believed that the leakage through half the critical flaw size can be determined by the leak detection system at Ginna which is capable of measuring 1 gpm leakage [24].

e SIR-97-077, Rev. 0 5-8 Structural IntegrityAssociates, inc.

Table 5-1 Summary of Critical Flaw Sizes CriticalFlaw Len th (2a), in.

Node

. No.

Hot Le 680 50 60 70 Cold Le 8400 910 920 930 950 960 Total Stress, ksi 17.811 16.628 16.774 14.420 15.419 19.153 16.629 14.655 12.659 11.188 Tension Stress, ksi 3.728 3.728 3.728 3.728 3.728 3.728 3.728 3.728 3.728 3.728 Tension 8.814 9.349 9.282 10A49 9.934 8.249 9.349 10.325 11.437 12.672 Bending 11.537 12.120 12.046 13.285 12.745 10.908 12.120 13.156 14.299 15.200 Combination 10.967 11.499 11A32 12.552 12.065 10.390 11.498 12.436 13.456 14.358 SIR-97-077, Rev. 0 5-9 Structural IntegrityAssociates, Inc.

Table 5-2 LBB Evaluation Results for Detectable Leakage Node No.

Hot Le Critical Flaw Length (2a) (in.)

One-quarter One-half Three-quarter Leakage at Fraction of Critical Flaw Length (gpm) 680 50 60 70 Cold Le 8400 910 920 930 950 960 10.967 11.499 11.432 12.552 12.065 10.390 11.498 12.436 13.456 14.358 0.60 0.73 0.67 0.69 2.36 2A4 1.96 2.00 2.01 2.03 4.71 5.74 5.31 5.64 16.2 15.82 13.46 14.14 14.69 15.00 16.5 20.2 18.8 21.2 54.8 50.1 48.93 53.5'7.5 SIR-97-077, Rev. 0 5-10 Structural IntegrityAssociates, inc.

Pwater Twater r

fL, Inlet +

D ~

Pwater eat flow of sub cooled water I

I I

I

~f D

I critical two-phase flow I

flashing point Figure 5-1. Flow of Subcooled Water Through a Crack SIR-97-077, Rev. 0 5-1 1

Structural tntagrltyAsscclatcslnc,

70.00 Leakage Evaluation Hot Leg 60.00 50.00

~Node 680

~Node 50

~Node 60

~Node 70 30.00 20.00 10.00 0.00 0.00 2.00 4.00 6.00 8.00 Flaw Size (in) 10.00 12.00 14.00 Figure 5-2. Leak Rate Versus Crack Length (2a) for RHR Pipe Run from Valve 700 to Hot Leg SlR-97-077, Rev. 0 5-12 Structural IntegrityAssociates, Inc.

Leakage Evaluation Cold Leg I

~ Node 8400

~Node 910

~Node 920

~Node 930

~Node 950

~Node 960 1500 1000 0.

Hair Slee (in) 10 12 14 16 Figure 5-3. Leak Rate Versus Crack Length (2a) for RHR Pipe Run from Valve 721 to Cold Leg SIR-97-077, Rev. 0 5-13 Structural InfegrlfyAssociates, Inc.

6.0 EVALUATIONOF FATIGUECRACK GROWTH OF SURFACE FLAWS In accordance with the NRC criteria [2,3] set forth in Section 2 ofthis report,'he growth of postulated surface cracks by fatigue is evaluated to demonstrate that such growth is insignificant for the plant life, when initialflaw sizes in excess ofthose meeting ASME Code Section XIIWB-3514 t're postulated.

Furthermore, the growth oflarger postulated initial flaws, in both depth and length directions, is studied to demonstrate that such flaws would tend to grow through the pipe wall (in the depth direction) to result in detectable leakage prior to significantly impacting safety margins by extending in length.

i I

s

~ r The stress intensity factors, K, corresponding to the point ofthe maximum depth ofa semi-elliptical crack are calculated using pc-CRACK'[13]. The Kvalues are calculated for each pipe size. for a reference 10 ksi uniform tension and pure bending stress, In each case, the stress intensity factors are determined for a conservative aspect ratio (a/Z) of0.1. The stress intensity magnification factors derived in Reference 23 were used to compute the Kvalue corresponding to the point of maximum length at the inside surface ofthe pipe.

Plant design transients forthe RHR piping [24,25] are shown in Table 6-1. The normal operational mode ofthe RHR system is that itis used for decay heat removal during the latter portions of cooldown. At about 350'F, the RHR system is initiated. However, the plant procedures [33]

require that the RHR system be pressurized and slowly warmed prior to use. This is accomplished by circulating flowthrough the RHR system into the letdown line. As a result, when return flowto the reactor coolant loop is initiated, there is no significant thermal transient. Therefore this transient is not considered in Table 6-1.

For the purpose ofcrack growth analysis, the transients shown in Table 6-1 are conservatively combined into fourteen (14) different load combinations based on their pressure and temperature ranges.

The combined transients and associated number ofcycles are shown in Tables 6-2 and 6-3.

The pressure and temperature values corresponding to these combined transients were used to t

linearly scale the pressure and thermal stresses corresponding to operating conditions. The axial SIR-97-077, Rev. 0 6-1 g Structurat tntegrtttrAssociates, tnc

stresses due to the pressure and thermal differentials for each of the transient categories are calculated as follows:

For an applied pressure ofP, the axial stress is calculated as:

D 2 c =P 0

I where D, is the outside diameter and D; is the inside diameter.

For thermal loads, hT, the stress is the maximum operating thermal stress, shown in Tables 4-2 and 4-3, factored by the ratio ofthe transient temperature to the operating temperature gradients:

hT Gt = Gmax,oper Topcf The calculated axial pressure and thermal stresses are presented in Tables 6-4 and 6-5. Tables 6-4 and 6-5 also show the total stresses, including the deadweight stress for Nodes 680 and 910 shown in Tables 4-2 and 4-3.

Using the Kresults calculated above with pc-CRACK

[13] and the transients in Table 6-2, the fatigue crack growth law recommended in Ref. 25 forstainless steel in a PWR'environment was employed to compute crack growth for various postulated initial flaw sizes. This crack growth is given by:

where C,n da/dN = C E S (dX)'hange in crack depth, a, per fatigue cycle, in./cycle material constants, n = 3.3, C = 2 x 10'in./cycle)/(psi ~in )"

SIR-97-077, Rev. 0 6-2 Structural IntegrityAssociates, Inc.

S Km<K~

R ratio correction factor = [1.0 - 0.5R ]

Km'~

environmental factor (equal to 1.0, 2.0, and 10.0 for air, PWR, and BWR environments, respectively)

K

-K.~psi~in and minimum and maximum values, respectively, of applied stress intensity factor Avalue of2.0 was used for the parameter E in the above equation. Two bounding R ratios of0.0 and 1.0 were used to calculate the crack growth. The R ratio of0.0 corresponds to a case where the effect ofresidual stresses is minimal while an R ratio of 1.0 conservatively represents the case where residual stresses contribute significantly to the total stresses.

In equivalent ksi units, the crack growth laws for these two R ratios can be written as:

da/dN = 3.177 x 10 (dX) forR = 0.0 da/dn = 5.083 x 10 (M) forR = 1,0 The analysis is performed forNode point 910 ofthe cold leg RHR loop since this location has the maximum thermal stress range as can be seen from Table 4-3. The stresses are cycled between maximum and the minimum stresses shown in Table 6-5. The weld residual stress is conservatively represented by a pure through-wall bending stress equal to the pipe material (SA 376, Type 316 stainless steel) yield stress at the operating temperature of612.2'F (S= 18.8 ksi).

For each pipe size and enveloping transient category, the appropriate scaling factors, based upon a reference stress of 10 ksi and actual stress values given in Table 6-3, are input to obtain the actual K values for the fatigue crack growth.

For the crack growth in the depth direction, the analysis is performed for three initial crack depths (a/tW.15, 0.5 and 0.7). In the length direction, the calculations are performed for depth-to-wall thickness ratios (a/t) =.0.15, 0.6 and 0.8. These ratios correspond approximately to the final a/t SIR-97-077, Rev. 0 6-3 Structural IntegrityAssociates, Inc.

ratios forcrack growth in the depth direction after 40 years or when the crack reaches 80% ofpipe thickness.

The fatigue crack growth analysis results are summarized in Table 6-6. It can be seen that postulated circumferential flaws 15% ofpipe wall by about 0.84 inches long (ala = 10) do not grow significantly in 40 years ofplant operation. Evaluation ofdeeper postulated flaws (50% and greater) for both R ratios, shows that such cracks'would grow through the pipe wall before extending significantly in length. In all cases, the crack would grow through-wall before extending in length more than 0.3 inches. Thus, detectable leakage would result before LBB safety margins are violated.

SIR-97-077, Rev. 0 Structural IntegrityAssociates, Inc.

Table 6-1. Plant Design Transients for RHR Piping Design Condition Design Transients Number of

=

Cycles h,P (psi) b,T(1)

('F)

Plant Heatup/Cooldown 1935 447 Level A PlantLoading/Unloading 14,500 0

58/5 Level B Test 10% Step Increase/Decrease Steady State Fluctuations Reactor Trip at Full Power Step Reduction 50% to 0%

Loss ofPower Loss ofLoad Loss ofFlow Primary Pressure Test Primary Leakage Test 2,000 Infinite 40 80 80 40 180 25/29 2485 300 2250 200 320 58/23 100 13/16 250 103/58 1250 113/53 340 92/37 (1)

First number represents hot leg and second represents cold leg.

SIR-97-077, Rev. 0 6-5 Structural IntegrityAssociates, Inc.

'able 6-2 Combined Transients For Fatigue Crack Growth Evaluation for RHR Line Adjacent to Hot Leg Hot Leg Cycles Load Case Load Combination Description Block Cycles Pslg P

~

Pslg oF TRM oF hP, psi hT, 'F Notes l0 l2 13 14 Pressure Test Leak Test Heatu /Cooldown+ Loss ofLoad (U )

Heatu /Cooldown+LossofPower(U

)

Hcatup/Cooldown + 50% Reduction (Up) 50% Reduction (U )+ Loss ofPower (Dn) 50% Reduction (U

+ Loss ofLoad (Dn 50% Reduction (U )+Loss ofFlow(Dn) 50%Reduction(U )+Reactor Tri (Dn) 10% Ste Increase (U ) + Reactor Tri (Dn) 10%Ste Incr(U )+10%Ste Deer(Dn) 10%Ste Deer(U )+10%Ste Deer(Dn) 10%Ste Deer(U )+10%Ste Incr(Dn)

Remainin (U )+Remainin (Dn) 2.5 (1) 45 45 377.5(2) 2485 2250 2350 2350 2350 2350 2350 2330 2330 2290 2290 2250 1550 1550 1910 1930 1930 2150 2150 2150 2150 547 547 660 650 6Q5 588 588 588 588 615 615 615 615 605 70

-70 70 70 70 547 547 520 547 547 592 592 590 547 2485 2250 235Q 440 420

]80 140 140 477 590 580 535 41 41 68 41.

68 23 23 58 Max hT assumed ForT, use Plant Loadin (I) For analysis purposes, 3 cycles are used.

(2) For analysis purposes, 378 cycles are used.

SIR-97-077, Rev. 0 6-6

Table 6-3 Combined Transients For Fatigue Crack Growth Evaluation For RHR Line Adjacent to Cold Leg Cold Leg Cycles Load Case Load Combination Description

" Pressure Test Leak Test Heatu /Cooldown+ Loss ofLoad (U )

Heatu /Cooldown+ Loss ofPower (U )

Heatu /Cooldown + 50% Reduction (U )

50% Reduction U )+ Loss ofPower (Dn) 50% Reduction U ) + Loss ofLoad n

50% Reduction

) + Loss ofFlow (Dn) 50% Reduction

) + Reactor Tri (Dn)

Block Cycles 2.5(1)

P~

Pslg 2485 2250 2800 2350 2350 2350 2350 2350 P~

Pstg 1550 1550 1910 1930 TlDtl oF 547 547 582 582 582 582 582 TALb oF 70 70 70 70 70 547 547 520 547 LIP, psi 2485 2250 2800 2350 440 420 hT, 'F 477 477 530 530 512 35 35 62 35 Notes Max hT assurited Max bT assumed 10 10% Ste Increase U )+Reactor Tri (Dn) 10%Ste Inc

)+ 10% Ste Deer(Dn 12 10%Ste Deer(U )+10%Ste Deer(Dn 45 2330 2330 2290 1930 2150 2150 568 568555'47 539 539 180 140 21 29 16 13 10%Ste Deer(U )+10%Ste Incr(Dn)

Remainin (U )+Remainin (Dn) 45 377.5(2) 2290 2250 2150 2150 555 557 551 547 140 100 10 (I)For analysis purposes, 3 cycles are used.

(2) For analysis purposes, 378 cycles are used.

SIR-97-077, Rev. 0 6-7

Table 6-4 Combined Maximum and Minimum Stresses for Fatigue Growth Analysis for RHR Line Adjacent to Hot Leg Hot Leg Stress Ranges Maximum Stress, ksi Minimum Stress, ksi Load Combination Th DW Total P

Th DW Total 4.145 1.789 1.653 7.588 0.000 0.000 1.653 1.653 3.753 1.789 1.653 7.196 0.000 0.000 1.653 1.653 4.671 1.988 1.653 8.312 0.000 0.000 1.653 1.653 4.170 1.988 1.653 7.812 0.000 0.000 1.653 1.653 3.920 1.921 1.653 7.494 0.000 0.000 1.653 1.653 3.920 1.921 1.653 7.494 2.586 1.789 1.653 6.028 cn CO 10 12 13 14 3.920 3.920 3.920 3.887 3.887 3.820 3.820 3.753 1.921 1.921 1.921 1.868 1.868 1.819 1.819 1.827 1.653 1.653 1.653 1.653 1.653 1.653 1.653 1.653 7.494 7.494 7.494 7.408 7.408 7.292 7.292 7.233 2.586 3.186 3.219 3.219 3.586 3.586 3.586 3.586 1.789 1.688 1.789 1.789 1.759 1.759 1.804 1.789 1.653 1.653 1.653 1.653 1.653 1.653 1.653 1.653 6.028 6.527 6.662 6.662 6.999 6.999 7.044 7.029 (1) A through-wall bending weld residual stress'equal to the yield stress was also applied.

4 Cb Cn SIR-97-077, Rev. 0 6-8

Table 6-5 Combined Maximum and Minimum Stresses for Fatigue Crack

~

Growth Analysis for RHR Line Adjacent to Cold Leg Cold Leg Stress Ranges (1)

Maximum Stress, ksi Minimum Stress, ksi Load Combination Th 4.145 9.149 3.753 9.149 DW Total 2.723 16.018 0.000 2.723 15.626 =

0.000 Th 0.000 0.000 DW 2.723 2;723 Total 2.723 2.723 4.671 10.166 4.170 10.166 2.723 17.560 0.000 2.723 17.059 0.000 0.000 0.000 2.723 2.723 2.723 2.723 3.920 9.821 2.723 16.464 0.000 0.000 2.723 2.723 3.920 9.821 2.723 16.464 2.586 9.149 2.723 14.458 3.920 9.821 3.920 9.821 3.920 9.821 2.723 16.464 3.219 2.723 16.464 2.586 2.723 16.464 3.186 9.149 8.632 9.149 2.723 14.458 2.723 14.541 15.092 2.723 10 3.887 9.552 2.723 16.162 3.219 9.149 2.723 15.092 cn

'~E 12 13 14 3.887 9.552 3.820 3.753 9.303 9.341 3.820 9.303 2.723 16.162 3.586 2.723 15.846 3.586 2.723 15.846 3.586 2.723 15.817 3.586 8.996 8.996 9.226 9.149 2.723 15.305 2.723 15.305 2.723 15.536 2.723 15.459 Cb e

Cb Cb Crl g

SIR-97-077, Rev. 0 (1) A through-wall bending weld residual stress equal to the yield stress was also applied.

6-9

Table 6-6. Results ofFatigue Crack Growth R

ratio Assumed Initial a/t Assumed InitialDepth (in.)

Final Depth (in.)

Final a/t Assumed InitialLength (in.)

Final Length (in.)

Change in Length (in.)

0.0 0.15 0.50 0.16875 0.5625 0.6275 0.5563 3.3750

. 0.1719 0.1528 0.84375 0.8441 3.3822 0.00035 0.00720 0.70 0.7875 0.9005 0.8000 4.500 4.5146 0.01460 1.0 0.15 0.50 0.16875 0.5625 0.9061 0.800 3.3750 0.2358 0.02096 0.84375 0.8490 3.4893 0.00525 0.1143 0.70 0.7875 0.9177 0.800 4.500 4.7479 0.2479 SIR-97-077, Rev. 0 6-10

7.0

SUMMARY

AND CONCLUSIONS Leak-before-break (LBB)evaluations are performed for the RHR system at R. E. Ginna in accordance with the requirements ofNUREG-1061. In the evaluations, circumferential flaws are considered since they are more limitingthan axial flaws. Critical flaw sizes and leakage rates through half the critical flaw sizes are calculated on a location specific basis for the RHR line at Ginna. Fatigue crack growth analysis was also performed to determine the extent ofgrowth of any pre-existing flaws.

Based on these evaluations, the followingconclusions can be made.

~

Predicted leakage through half the critical flaw size for the RHR line adjacent to the hot leg is at least 4.7 gpm.

~

Predicted leakage through half the critical flaw size for the RHR line adjacent to the cold leg is at least 16.2 gpm.

~

Fatigue crack growth ofsubsurface flaws is insignificantly small and therefore does not invalidate the leak-before evaluation ofthe RHR lines.

~

Based on the fact that the leak detection system at Ginna is capable ofdetecting 1 gpm leakage, leak-before-break has been demonstrated for the RHR line locations considered in this evaluation.

SIR-97-077, Rev. 0 7-1 Structural IntegrityAssociates, Inc.

1 S.O REFERENCES 1.

Rochester Gas &Electric Corporation Drawings:

a)

C-381-354 Sht.

1 b)

C-381-354 Sht. 3 c)'-381-355 Sht. 8 i

2.

S tello, Jr., V., "Final Broad Scope Rule to ModifyGeneral Design Criterion 4 ofAppendix A, 10 CFR Part 50", NRC SECY-87-213, Rulemaking Issue (Affirmation),August 21, 1987.

3.

l NUREG-1061, Volumes 1-5, "Report ofthe U. S. Nuclear Regulatory Commission Piping Review Committee", prepared by the Piping Review Committee, NRC, April 1985.

4.

Structural Integrity Associates Report No. SIR-85-034, Revision No. 1, "Fracture Mechanics Leak-Before-Break Evaluation ofR.E. Ginna Nuclear Power Station High-Energy Piping Welds Outside Containment".

5.

NUREG-0927, "Evaluation ofWater Hammer Occurrence in Nuclear Power Plants" Revision 1.

6.'. S. Hazelton, W. H. Koo, "Technical Report on Material Selection and Processing

~

~

~

~

~

~

Guidelines forBWR Coolant pressure Boundary Piping", NUREG-0313, Rev. 2, USNRC, January 1988.

7.

Westinghouse Stress Report, SDTAR-80-05-05, Rev. 1, RHR 2500, dated 3/4/81.

8.

Westinghouse Stress Report, SDTAR-80-05-26, SI-200 dated 3/20/81.

9.

EPRI Report No. NP-6301-D "Ductile Fracture Handbook", June 1989.

10.

ASMEBoiler and Pressure Vessel Code,Section III,Division 1, 1989 Edition.

11.

Kumar, V., et al., "Advances in Elastic-Plastic Fracture Analysis," EPRI NP-3607, August 1984.

12.

Kumar, V., et al., "An Engineering Approach for Elastic-Plastic Fracture Analysis," EPRI NP-1931, July 1981.

13.

Structural Integrity Associates, Inc., "pc-CRACK Fracture Mechanics Software", Version 3.0 - 3/27/97.

SIR-97-077, Rev. 0 8-1 Structural IntegrityAssociates, inc.

14.'pc-LEAK Calculation ofLeakage Rates From Through-Wall Cracks", Version 1'.0, Structural Integrity Associates, September 1996.

15.'

P. C. Paris, and H. Tada, "The Application ofFracture Proof Design Methods Using Tearing Instability Theory to Nuclear Piping Postulating Circumferential Through-Wall Cracks", NUREG/CR-3464, September 1983.

16.

R; Klecker, F. Brust, and G. Wilkowski,"NRC Leak-Before-Break (LBB.NRC) Analysis Method for Circumferentially Through-Wall Cracked Pipes Under Axial Plus Bending Loads", NUREG/CR-4572, BMI-2134, May 1986.

17.

Rohsenow and Choi, "Heat, Mass, and Momentum Transfer", Prentiss-Hall, New Jersey, 1961.

18.

SAE Applied Aerospace Manual,Section I, "Engineering Fundamentals - Part A, Incompressible Fluid Flow".

19.

EPRI Report NP-3395, "Calculation ofLeak Rates Through Cracks in Pipes and Tubes",

Electric Power Research Institute, December 1983.

20.

"Marks Standard Handbook forMechanical Engineers," Eighth Edition, McGraw Hill,New York, 1978.

21.

R.D. Blevins, "AppliedFluids Dynamics Handbook", Van Nostrand Reinhold Co., New

~

~

York, 1984.

22.

R.J. Lahey, and F.J. Moody, "Thermal Hydraulics ofBoilingWater Reactors", American Nuclear Society, 1977.

23.

Huget, W. and Esser, K., "Sos Intensity Factors for Slender Surface Cracks", SMiRT Conference, Paper G/F 3/4, Chicago, 1983.

24.

Ginna UFSAR, Table 5/1-4, Rev. 13 dated 12/96.

25.

BWIReport No. 222-7705-SR-2, "Transient Analysis ofTubesheet/Primary Head and Secondary Shell Assembly", Section 2.1, October 2, 1995.

26.

ASME Section XITask Group forPiping Flaw Evaluation, "Evaluation ofFlaws in Austenitic Steel Piping", Journal ofPressure Vessel Technology, Vol. 108, August 1986, pp.352-366.

27.

EPRI Report NP-5531, "Evaluation ofHigh-Energy Pipe Rupture Experiments", January 1988.

SIR-97-077, Rev. 0 8-2 Structural IntegrityAssociates, Inc.

28.

=

American National Standard, Power Piping, ANSIB31.1, 1973 Edition with Summer Addenda.

29.

RG&E Action Report No. 97-1235, 8/14/97.

30.

EPRI Report, TR-106438, 2856-02, "Water Hammer Handbook For Nuclear Plant Engineers And Operators", May 1996.

31.

RG&E Action Reports 97-0404 &0405.

32.

Westinghouse Electric Corp. Letter No. PT-PQ-1584 from P.S. Van Teslaar to J. C. Hutton (RG&E), "R.E. Ginna Seismic Upgrading Progam - Operating Transients Document Revision", June 11, 1982.

33.

RG&E Procedure 0-2.2, "Plant Shutdown From Hot Shutdown to Cold Conditions", Rev.

111, 10/21/96.

SIR-97-077, Rev. 0 8-3 Structural IntegrityAssociates, inc.

ROCHESTER GAS.AND ELECTRIC CORPORATION k

INTER-OFFICE CORRESPONDENCE October 15, 1997 To:

Subject:

~

File I

Evaluation of Structural Integrity Associates, Inc. Report No. SIR-97-077

References:

1. RG&E Procedure, EP-3-P-154, "Review &Approval Of Vendor Drawings; Design And Manufacturing Technical Documents", Rev. 0.

2.,Structural Integrity Associates (SIA), Inc. Report No. SIR-97-077, "Leak-Before-Break Evaluation of Portions of the Residual Heat Removal (RHR)

System at R. E. Ginna Nuclear Power Station", Rev. 0.

Per RG&E Procedure in Reference 1, I reviewed the subject report (Reference 2) for technical correctness, relevance, and applicability to Ginna Nuclear Power Station. Method of review consisted of independent verification of fundamental concepts and criteria of the

~

~

~

~

~

~

~

~

~

~

~

leak-before-break approach, input data, material properties, applicable loadings, effects of fluid/structure interaction on leakage quantification, and interpretation of results. In addition, I also discussed technical issues that were brought up by SIA independent reviewer and made sure these are resolved considering the existing RHR design basis, system requirements, transient operating conditions, and operating procedures.

Results of the review and evaluation are summarized below.

1. SIA has incorporated all RG&E comments in the final report.

2. Ginna plant specific input data, effects of procedural evolutions, technical programs, guidelines and regulatory commitments that were utilized in the leak-befog-break study have been reviewed and confirmed per SIA QA program. No findings were discovered.

3. The leak-before-break methodology in the evaluation of the RHR piping is based on sound fundamental engineering concepts utilizing EPFM (elastic-plastic fracture mechanics) approach for calculating critical flaw sizes and classical thermal-hydraulic equations for evaluating single and two-phase fiow. The approach is in accordance with Generic Design Criterion (GDC) 4 and NUREG-1061.

4. Results of leak-before-break evaluation (Reference 2) is applicable to Ginna due to the

~

~

~

following factors:

a.

Predicted minimum leakage of 4.7 gpm of the RHR line exceeds the minimum

SIA Report SIR-97-077 0

leak detection capabilities of Ginna inside containment.

b. Other degradation mechanisms affecting the RHR piping such as water hammer, fatigue, erosion/corrosion, etc.. have negligible impact on the structural integrity of the RHR pipe.

5. Subject report is therefore acceptable'to RG&E. It provides a technical basis that pipe rupture of the RHR pipe sections from the hot and cold legs of the reactor coolant system to MOV's 700 and 721 willnot occur.

The leak-before-break report (Reference 2), in accordance with requirements of GDC 4 can be submitted to the NRC as a basis for showing that dynamic effects of pipe rupture of the evaluated RHR pipe sections has a negligible probability of occurrence.

Prepared by:

A. P.

ochino Primary Systems Engineer Approved by:

Brian Flynn Manager, Primary Systems cc: George Wrobel

Structural IntegrityAssociates, Inc.

September 23, 1997 NGC-97-040 3315 Atmatten Expressway Suite 24 San Jose, CA 95118-1557 Phone:

408-978-8200 Fac 408-976-8984 ncoficstructint.corn Dr. Lee Rochino Rochester Gas &Electric Company R. E. Ginna Nuclear Power Station 1503 Lake Road Ontario, NY 14519

Subject:

Structural Integrity Associates Report No. SIR-97-077, Rev. 0, "Leak-Before-Break Evaluations ofPortions ofthe Residual Heat Removal,(RHR) System at R.

E. Ginna Nuclear Power Station".

Dear Lee,

Enclosed are two bound copies and one unbound copy ofthe subject report for your use.

We appreciate the opportunity to be ofservice to RG&E and you on this project. Please do not hesitate to call ifyou have any questions or comments.

Very truly yours, Nathaniel G. Cofie, Ph.D.

Associate sjl enclosure cc:

G. Wrobel (RG&E)

RGE-07Q-102 San Joan, CA Phone: 408478-8200 Akron, DH Phone: 3304t844888 Silver Spring, MD Phone: 301489-2323 Pompano Beach, Fl Phone: 954447.278t Talpel ~ Tahran Phone: 02488-5508 Charlotte, NC Phone: 704473-1 389