ML20078A745
| ML20078A745 | |
| Person / Time | |
|---|---|
| Site: | Quad Cities  |
| Issue date: | 06/30/1994 |
| From: | Brandlund B, Ranganath S, Stevens G GENERAL ELECTRIC CO. |
| To: | |
| Shared Package | |
| ML20078A742 | List: |
| References | |
| GENE-523-A79-05, GENE-523-A79-0594, GENE-523-A79-5, GENE-523-A79-594, NUDOCS 9407010120 | |
| Download: ML20078A745 (29) | |
Text
{{#Wiki_filter:- - - _. O GE Nuclear Energy GENE-523-A79-0594 TECHNICAL SERVICES BUSINESS GE Nuclear Energy DRF 137-0010-7 175 Cunner Avenue, San Jose, CA,95125 Class H June 1994 Evaluation of the Indications Found atthe H5 Weld Location - in the Quad Cities Unit i Shroud i Prepared by: b/2bA r/7/W G. L95tevens, Senior Enginder # StructuralMM hantesProjects , l i M% a VeriSed by:@hl!! fund',Tenior Engineer
- (1/1/44 X. J. B StructuralMechanics Projects Approved By:
N' ^f ' ' k r Dr. S. RanganstK Manager StructuralMechanics Projects P
f 1MPORTANTNOTICEREGARDDVG CONTENTS OF THIS REPORT Pisase Read Carfully J This report was prepared by General Electric Company solely for the use of the j Communwenkh Edison Company (CECO). Theinformation contained In this reportis boHeved by t General Electric to be sa securate and true representation of the facts known, obtained or provided to GeneralElectric at the time this report was prepared. . The only undertakings of the Genersi Electric Company respecting Informstlan In this ; document are contained in the contract governing this work, and nothing contained in this ) document shallbe construed as changing sold contract. Tha use of this Information except as deRned by sold contract, or for any purpose other than that for which it is intended, is not authorized; nad with respect to anysuch unauthorized use, neither the GeneralElectric Compa not any of the contributors to this document makes anyrepresentation or warranty foxpress o impled) as to the completeness, accuracy or usefulness of the informstlen contained i i document or that such use of such information may not infringe privately owned rights; nor do i theyassume anyresponsiblityfor EsbiEtyor damage of anykind which may resuk from su; of suchinformation. : I
?
I l i l
l 1 1 Table of Contents 1 EAGE
- 1. 0 INTRODUCTION...... ......... . . ... . . .. . . . . .. .. . .. . . . . . .. . . . . . . . .. . . . . .. . . . . . . . . . . . . . . . . . . . . . . ...
- 2. 0 TEC HNIC AL APPRO ACH..... . . .. . . . .. .. .. .. . . . . . . . . .. . . . . . . . . .. . . . . .. . . . .. . . . .. . . . . . . .. . . . . .
2 .1 Allowable Flaw Size ..... .. . . . .. . . .. . . . . . .. . . . . . . . . . . . . . . .. . . . . .. . . . ... . . . . . . . . .. . . . . . . . 2.2 C rack Growth Assessment ... . . . .. .. . ... . .. .. . . . . .. . . .. . . . ... . .. ... . . . ... . . .. . . . .... . . .... .. l 2.3 Structural Margin Determination . ...... . ..... ..... ...... .............. . .. ... .. .......... . .. . . . 9 l
3.0 CONCLUSION
S .. .. ..... ........ .. . . . .. ... .. . .. . . .. . . . . . . .. . . . .. . .. . . . . ... . ... . . . . . . . . . . . ... . . . .. . . 4 . 0 REFEREN CE S . .. .. . . .... . . .... . .. . . . . .. . . . . . . . . . . . . . . . . . . . .. . . .. . . . .. . . . . . . . . . . . . . . . . List of Tables PAGE . TABLE 1: LIMIT LOAD EVALUATION FOR WELD H5. ............... ............. ..... 6 i TABLE 2: STRUCTURAL MARGIN RESULTS FOR WELD H5...... ... .... ............ I3 i TABLE 3: CONSERVATIVE ASSUMPTIONS INCLUDED IN LIMIT LOAD l EVALUATION . . . . .. . . . . .. . . .. . . . . . . . . . .. . . . . . . .. . . . . . . . . .. . . . . . . . . . . . . . . .. . . . . . . .i P List of Figures ; i PAGE FIGURE 1: QUAD CITIES UNIT 1 SHROUD HORIZONTAL WELD 1 CONHGURATI ON . . . . .... ..f.. . . .. . . ... . . . .. . . . .... . ... . . . . . .... .. . . .. . . . . . MGURE 2: LIMIT LOAD METHODOLOGY................... ..... ... ........... ................ 5 ' HGURE 3: H5 WELD DETAILS AND CRACKING LOCATION ........ ......... . ..... 7 HGURE 4: PLEDGE MODEL PREDICTION FOR QUAD CITIES 1/2................ . 10 HGURE 5: CRACKGROWTHRATE AS A FUNCTION OF SULFATE...............11 l MGURE 6: CRACK GROWTH RATE AS A FUNCTION OF CHLORIDE.............12 , i 4 i
. - . - . ,m.,__ ~ - _
7 l i
1.0 INTRODUCTION
During tha current refueling outage at Quad Cities Unit 1, core shroud inspections were , performed in accordance with recommendations given in GE Services Infonnation Letter No. 572, Rev.1 (SIL 572) [1]. During the initial portion of these inspections, crack indications were l visually detected in the vicinity of the H5 weld (see Figure 1). The cracking was located imrnediately below the H5 weld in the core plate support ring. The indications were predominamly circumferential, and were visible at all accessible locations (approximately 15 the circumference). SeMqua* examination by automated ultrasonic testing (UT) of the indications in all areas of the H5 weld accessible by the UT system confirmed the visual indications to be cracks. Based on the results of automated and manual UT exams a results of boat samples taken from the.H5 weld location provided by Commonwealth Edison Company (CECO), the probability of detection of flaws deeper than 1.24" on the core plate support ring side of the H5 we!d is very high, and no flaws deeper than 1.24" were detecte this reason, the bounding maximum flaw depth used for the purpose of this structural margin I assessmentis 1.24" [2]. i The purpose cf this report is to evaluate the indications found near the H5 weld from a structural standpoint. Limit load calculations are performed consistent with the previous : Screening Criteria generated for the Quad Cities Uit I shroud [3], and structural margins are detennined taking into account appropriate crack growth values and ASME Code, Section XI [4 safety factors. 9 I 1 j n
p 360* Shroud Head ' i
" H1 Flange -
1 H2 Top Guide - E ' H3 Support Ring l H4 Core Plats H5'
- I " H6 support Ring .
- H7 shroud Support Ring i
RPV SIDE VIEW "BOLLED-OUT VIEW
~
NOTE: NOT TO SCALE FIGURE lt QUAD CITIES UNIT I SHROUD HORIZONTAL WELD CONFIGURATION 2 '
2.0 TECHNICAL APPROACH i Tha Reference 3 report documents screening criteria developed for the Quad Cities Unit I shroud based on limit load and linear c!astic fracture mechanics (LEFM) techniques. The purpose i of that report was to develop criteria that a!! owed indications dim.d during visual inspection to be screened for ihrther evaluation. Since the criteria were based on visual examinations, all flaws were conservatively assumed to be through wall and allowable flaw lengths were calculated i using limit load and LEFM techn! ques. f This evaluation determines allowable flaw depth, since UT examination has confirmed that, the cracking is not through wall. The cracking was assumed to be 360' around the circumference l of the shroud for the pugoses of this evaluation, since the indications discovered were seen at all accessible locations. Similar calculations to those included in the Reference 3 report were performed for a fully circumfbmntial, part through-wall crack. Crack growth estimates were combined with the resulting allowable flaw size to determine structural margin. The results are l described in detail in the sections that follow. i 2.1 Allowable Flaw Size , The Reference 3 analysis conservatively included LEFM effects for welds H4 and HS due , to potential fluence effects. The fluence estimated for the H5 weld is low (3x10" a/cm') [l Since the irradiation level is low, the &acture tougimess is comparable to that of unirradiated material where ductile behavior govems. This is supported by studies performed by EPRI [6] l where the impact of fluents in the amount accumulated by the H5 weld is negligible. Therefore,- limit load calculations which use ASME Code, Section XI safety factors are the appropriate technique for evaluating structural margins for this location. j The limit load approach used here is depicted in Figure 2, as obtained from a net section collapse for 1=4n [4,7]. The neutral axis shown in Figure 2 is determined by equilib: sting the l l force resulting Rom the applied membrane stress, P., in the uncracked cross section with the i force resulting kom a stress equal to the flow stress in the remaining ligament (uncracked region) at the crack cross ser.tlon. 3 ,
- . -. - -~ - . --- - .- . -. .. -. - .-
-.r -
i Fct the case where et = 180' (i.e., 360* flaw), the following equatiens apply: g , tr(1-d/ t-P, l a,) 2-d/ t Pb '= (2-d/ t) sin , where: t = shroud thickness, inches d = crack depth, Inches
. = acta.h an,,. ;
E = angle that defines location of neutral axis l P. - applied membrano stress, psi P6' = failure bending stress, pd l 4 or = Sow stress of the material = 35
- i From Referenos 3, the faulted load condition wu determined to be limiting. The fkul load condition conservatively includes loading from both a design baals earthquake l main steam line break. For this load case, the membrana stress, P., was previously d l
be 0.278 kai and the bending arress, Pw was determined to be 2.337 kal. These stre result of deadweight, seismic and pressure loads. Per Section XI of the ASME Co i fheter of 1,4 fbr the thulted condition was applied to these stresses in the allowab calculations. The value of S. at 550'F for the 304 stainless steel shroud materiall Trial and enor solution of the equations given above using thess values is shown in Tab f l The results of Table I show that a crack depth of 96% (l.a., alt = 0.96) of the shr
' thickness can be solenned while still mali *=tning all ASME Code structural margin l
Cities Unit 1 shroud has a 2 inch wall thickness, and the H5 weld is backed by a The ! shown in Figure 3. The location of the observed cracking is also shown in Figum L minimum thickness through which the crack must traveres before reaching throl therefore 3 inches. Therefore, the allowable flaw depth in dds region, bued onf is 2.88" (i.e.,3" x 0.96). :
. .i 4 ,
i k l
~-- - - - ----_.--- -_______
,s.
NondnalStruss in the Uncracked Secton of P!pe p ,+p, Crack Langth = 2Rs 4 > N Flow Strees,sq l - t ! I -.
' d
+Y --...._---..........f a
+- !
4-~ ' y y'; 4"- >
/\ \!/ i j
\ 4-f \, '
,/
4- >I r' t-+ } +- *-- I 14 R7 4--- 4.,,,
\ q l
.............. y ,!
.....y . . . . . .
4 N-
-+ /.. _ .._____.........
-+ $
..1:
\ p,
.f 4-ou a '
Ms P,,, = #p8ed Membrane Stress in Uncracted sectlan Swees Detributionin P = Mptled Bending 6tsssin Unoracked section the Crocked section at he Pointof Co!!apee FIGURE 2: LIMIT LOAD METHODOLOGY 5
TABLE 1: LIMIT LOAD EVALUATION FOR WELD H5 l Cass 02: The neutral ages isleested suchinet a + p > s (tNo is enecked Wow) (1.d4 P,/8FAses
$s- (from References 4 and 7)
- 2. d4 Pi = (2'om)* (2- e) ak$
Sivent P. s 278 pel(Reference 3) P= 2,337 pel(Reference 3) amfety Factor, SF = 1A (for Faulted eenditions = bruiting ser Referenos 3) P.*8P = 200 pel P/8F = 3,272 pal
- b= 18,900 pel(at 5807 for 304 ss) 3b= $0.700 pels or as 180 ** 3.1415 rathens ,
3.1416 dA) / (2. dA) (Eqn. t) Tha* p=( 3.1175 . (Egn. 2) Pi = 3227s.s *(2.dt0eins P,' SoMng bystat andanon S tom!Egn.1) trom [Egn. 21 Oflerance (radans) (pel) = Ps' . Pe'6F (') m+$ezf - dA 61.047 87,775 84.5 YES 0.1000 1A754 57D75 53.003 79.2 YES 0.2000 1.3820 62.567 49,208 73.3 YES 0.3000 12794 47,408 44.138 06.5 YE8 0.4000 1.1830 41,534 38.252 08.1 YE8 05300 1.3311 34A3s 31,588 50.4 YE8 0 5300 0.8004 27,237 23Ae5 40.5 YE8 c.7000 0.70s4 18,see 15.416 28.8 YES 0A000 0.5036 0,283 5,082 15.1 YE8 ; 0.0000 0.2837 4,008 l 13.8 YES I 0.9100 0.2373 8.3W ' 4.00s 12.1 YE8 0A200 0.2104 7.280 8.284 3.013 10.5 YE8 03300 0.1830 YE8 5.284 2.013 8.9 CA400 0.1981 4,200 1.00s 7.3 YE8 0A051 0.1286 8A078 3.f72 1 84 YES OA600 8.6 YES 0.0073 3.282 9 03831 6
-l VESSELWALL TOP OF SHROUD A
2 22* 2 j 1.0" FILLET WELD I i WELD H5 LOCAT10N OF k CRACK INDICATIONS - l ! 1 4.0* CORE PLATE SUPPORT RING 4 l
\
1.0" FILLET WELDS l
? L8" C ;
l,, l . l J l FIGURE 3: H5 WELD DETAILS AND CRACKING LOCATION l l 7 l 1
._ . 2.2 Crack Growth Assessm:nt 4
Prior analyses [3] have conservatively used a crack growth rate of 5x10 inch / hour. Thl was intended to be a b +% value that covered both lmergranular stress corrosion cracking (IGSCC) and irr=*=daa assisted stress corrosion cracking (IASCC). More recent predictions made with the GE PLEDGE pridictive model where plant-specific water chemistry and other l 4 cffects were included suggest a crack growth rate of 1.32x10 inch / hour, as shown in Figure 4. Thus, the 5x104 inch / hour value based on the IOSCC/IASCC combination is defmitely conservative forthe H5 weld. A significant point to be made is that the observed cr. airs thus far is mainly due to prior operation at relatively high conductivities, including aggressive anions such as sulfates and chlorides. With the increased attention to IOSCC, most boiling water reactor (BWR) plants have l dramatically reduced their sggrenive anion input, thus assuring that future crack growth rates are much lower than that in the past. Quad Cities Unit 1 currently operates below 0.15 S/cm conductivity and 5 ppb chloride and sulfate combined. Figure 4 shows the dependence of the predicted growth rate on the conductivity based on th GE predictive model for IGSCC. Figures - i 5 and 6 show the deper4;;;c on sulfate and chloride species [9]. In all cases, the lower sulfate and chloride levels lead to dram =d fly lower crack growth rates. Additionally, Quad Cities Unit I has been operated on hydrogen water chemistry (HWC) since the third quarter of 1990 [8] which will help contribute to reduced crack propagation rates. Thus, any margin assessments 4 based on the two growth rates (5x10 4inch / hour for bounding values and 1.32x10 inch / hour based on the GE PLEDGE predictive model) are conservative. Pre-operational testing of BWR internals has demonstrated that high cycle fhtigue resulting from flow induced vibration is not a concem for the core shroud. Additionally, low cycle fatigue caused by therinal and/or pressure changes in the core region are negligible since all
. anticipated changes in these parameters resuh in relatively low stresses in the core shroud. This is fbrther supported by the lhet that no fhtigue cracking was observed from boat samples removed from the cracked areas of the Quad Cities 1 shroud, as well as the core shrouds of other BWRs.
Therefore, the impact of fhtigue on the core shroud is mncluded to be negligible, and is not considered to be a fbrther c eator to the crack growth values discussed here. The use of the automated UT system combined with enhanced manual UT provided crack depths at a number oflocations along the circuirA.uce. As expected, the crack depths varied along the circuwA ce of the inspected regions. .For conservatism, the bounding maximum flaw 8
depth cf 1.24" was used in the limit 1:ad evalunion described in this rep:n. Since the evaluati:n for limit load is based en the t tal structural atsa available, the more appropriate value to use is the average depth, not the bounding maximum depth. Therefore, the stmetural margins shown in the next section are likely to reflect even more conservatism. 2.3 Structural Margin Determination Since crack indications with a bounding maximum depth of 1.24" were estimated based on UT and boat sample evaluation, a maximum crack depth of 1.24" was conservatively used fo.r evaluating structural margin. Crack growth values corresponding to each of the two crack growth values identified above were added to this maximum flaw depth. Structural margin was assessed by comparing the remaining ligament to the required ligament obtained from the limit , load evaluation. The results are shown in Table 2. I
\
i l l 9 : l i
-_ , _ _ , - - - , ___I
Crack Growth Rate, in/h Leet 5_.- F[gg t .5. .-- -.200 MV. Cycle Ave. Cycle Ave. - 1.000E-04 -. .- - - . -
' ~~ # -~ =~'--'
5 000E -
" " m'--
FOI; - ,7 - - - -
~~
4.8 ' - # .
.g..mv--- l 1.000E -
- s
- .m !
.x
--{ [ n. .-
3 [--30tCmE l 1.000E-06 . . . - - -
"00.RE -
.--- - . f. -
..-- --_.=_--
.. ,, -=- ,
~ ~
l 0_00-1 9 662 l 1.000E-07 l 0 0.1 0.2 0.3 0.4 0.5 ; Conductivity, pS/cm I
)
PLEDGE: 15 C/cm2, 20ksi/in .. . . u . ) l
. 1 i
FIGURE 4: PLEDGE MODEL PREDICTION FOR QUAD CITIES 1/2 10 1
GENEv523 A19 05N, Rev.0
~
3 l J
# 1 1
l i 10000 - Sensluzed T304 Steintess Steel ! 288'C, 27.5 MPe/Th,15 c/cma E ;;;;;;;;:smm=====1.aivseri , 1000 ---- - gd M MC
===:
**'8 V 8CP
= = = = = == = = = = = = =igiiig __ -
;;;;;:ma __
- meme . ,c ,
l
---, M
- - m m ,
1 7 s - g F p 42'Y ECP ! I f f i 4 [ / i E
=d!EEE5 : ,-
W' [===:EE d = ======'"!= MF.o v scrl E . r s s s t O .- i n 6 1 s s s n .* l _.- J -
> +
. i g . .-
0 -
=
, L 1
to to se se so se 7e so to too tio 120 SULFATE (FFB) i
- I I
i i FIGURE 5: CRACK GROWTH RATE AS A FUNCTION OF SULFATE i i t 11 ! i l l t
GENE 523 A79 0594, Rev. 0 b 20000 St.sitised T304 StaWees Steel 2M'C,27J MPad,15 c/cnia
~
m - p Wi.e.1VBCP_
-" .os v ter 1000 g
g 1
,[ j F g !_8.3 V ECP g /
$ ,on ./ / / s # '
_- : 4.3 V ECP f ,- ,- s r ., O > r , ac .
/ s s
(< g, - l f f - f
/ I
^
6 ,: ,.. - _ , . . - -
.u....- .--- -
1 ro so ,o so ao so so to as CHLORIDE (FF5) 1 I FIGURE 6: CRACK GROWTH RATE AS A FUNCTION OF CHLORIDE 12
TABLE 2: STRUCTURAL MARGIN RESULTS FOR WELD H5 (Based on a 360*,1,N" Depth Flaw) Time Until Crack Crack Allowable Allowable Growth Growth Crack Final Crack Crack Depth is Case Rate Period Growth' Depth Depth Margin Reached' [lach/hourl Imonthsl finches 1 finchest finches 1 Factor2 [ hours (yrs)] 4 6 0.20 1.44 2.88 13.0 32,800 (i.1) 1 5x10 2 5x10d 18 0.60 1.84 2.88 9.7 32,800(4.1) 4 0.05 1.29 2.88 14.3 124,200 (15.5) 3 1.32x10 6 4 18 0.16 1.40 2.88 13.3 124,200 (15.5) 4 1.32x10 NOTE: (1) Crack growth is determined for each crack growth period assuming 8,000 - hours per year (=91% availability). (2) The margin factor is calculated by dividing the remaining ligament by the required ligament, as follows (for case #1, thickness = 3"); Margin Factor = Remaining Ligament / Required Ligament
= (3.0-1.44)/(3.0 2.88)
= 13.0 (3) The time until the a!!owable crack depth is reached is determined by dividing the minimum existing ligament by the crack growth rate, as follows(for case #1):
Tn* ne = Minimum Existing Ligament / Crack Growth Rate
= (Allow. Depth-Current Maximum DepthyCrack Growth Rate
= (2.881.24y5x104
- 32,800 hours or 32,800/8,000 = 4.1 yeara .
13
3.0 CONCLUSION
S This evaluation provides a structural margin assenment of the indications found near the H5 weld in the Quad Cities Unit I shroud. Limit load techniques and ASME Code, Section XI safety factors were used to demonstrate adequate structural margin for the next 18 month fuel cycle of operadon assuming a 360',1,.24 inch deep flaw at the H5 weld location. The structural margin results are summarized in Table 2. A list of a!! of the conservative assumptions used in the evaluationla provided in Table 3. The results of Table 2 demonstrate, based on limit load techniques, that approximately_a factor of ten is available in terms of required area for a 18 month fuel cycle of operation with a bounding maximum flaw depth of 1.24" In the H5 weld of the Quad Cities Unit I shroud. O e 9 4 14 l
TABLE 33 CONSERVATIVE ASSUMPTIONS INCLUDED IN LIMIT LOAD EVALUATION
- 1. A 360' crack wu assumed, even though only approximately 150' of the circumference was examined and bund to have cracking.
- 2. Crack depth was based on the bounding maximum crack depth which can be detected with bl;,h probability by UT rather than the average crack depth.
- 3. The bounding crack growth estimated for the next fuel cycle was included in the structural margin assessment. .
- 4. ASME Code pressure boundary safety margins were applied even though the shroud is not a primary pressure boundary.
e 15
. 1
4.0 REFERENCES
[1] GE Nuclear Energy, " Core Shroud Cracks," GE Services Information Letter No. 572, l Revision 1, October 4,1993. [2] Letter from M. D. Lyster (CECO) to W. T. Russell (NRC) dated 6/6/94, " Response to l NRC Request for AdditionalInformation Concerning Core Shroud Cracking at Dresden Units 2 & 3 and Quad Cities Units 1 & 2." GENE-523-02 0194, Revision 0, " Evaluation and Screening Criteria for the Quad Cities 1- [3] 2 Shrouds," W.F. Weitse, GE Nuclear Energy, San Jose, CA, March 1994. i [4] Ah Boller & Pressure Vessel Code, Section %f, " Rules for Inservice inspection for Nudear Power Plant Components," 1989 Edition, American Society of Mechanical , Eggineers, New York. * -
.~
[5] Letter from Sylvia Wang (GB) to Gary Stevens (GE), " Estimated Fluences at Shroud Nds H1 to H7 for Dresden 2 and Quad Cities 1 & 2," May 20,1994. f i ~2 :- ; [6] ,EMtl Raport NF4767, Project 2680-2, Evaluation of BWR Top-Guide Integrity,l ElR:ric Power Rr. search Institute, Component Reliability Program, Nuclear Power l
'D 5ision, Final Report, November 1986. l l := '
- a
[7] 5sepath R=ar-d and Hardayal S. Mehta, " Engineering Methods for the Assessment of ; M Fracture Margin in Nuclear Power Plant Piping," Elastic-Plastic Fracture: Second l
, Symposium, VoL>u H - Fricture Resistance Curves aaf Mnaarinn Anolientions.
ASTM STP 803, American Society for Testing Materials,1983. l 1' i GENE-NE-A00 05652-04, " Preliminary Safety Assessmen of Core Shroud Indications [8] for Cycle 13 Operation of Quad Cities Unit 2," D. K. Rao, OE Nuclear Energy, May 1994. l l A [9] EPRI Report TR 103515 Project 2493, "BWK Water Chemistry Guidelines - 1993 i Ravision, Normal and Hydrogen Water Chemistry," Elecuic Power Research Institute, ; BWRWaterc'- ' ' y Guidelines Revision Committee, February 1994. i , e . 4> 4 I 16 :; l
. ~,
E ATTACliMENT 4 L Structural Integrity Report RAM-94-159, Revision 0, dated June 11,1994, Evaluation of Haws in Cirrumferential Core Shroud Welds at Quad Cities Unit ! 1. i-V P f i e 1 1, r, - I 1 I
. >STRU - -
i > - 4 ASSOCIATES,INC. 3150 Almaden Expressway Fossil Plant Operations Suite 145 66 South Miller Road San Jose, CA 95118 Suite 206 Akron, Ohio 44333 (408) 978 8200 rAx (G)F8 BW (216) 864-8886 mm> rsa38 June 11,1994 rn ms> n as RAM-94-159 SIR-94-052 Revision 0 Mr. Thomas J. Wojcik Commonwealth Edison Company Ouad Cities Nuclear Station 22710 206th Avenue North Cordova, Illinois 61242
Subject:
Evaluation of Flaws in Circumferential Core Shroud Welds At Quad cities, Unit 1
Dear Tom:
StructuralIntegrity Associates (SI) has performed an evaluation of the flaw indications found in circumferential welds H1, H2, H3, H4, H6, and H7 at Quad Cities, Unit 1, in order to determine the ASME Code structural margin in each of these welds. The evaluation of weld H5 was performed elsewhere and will be submitted under separate cover. The evaluation performed here was designed to evaluate operation without repair of these welds for an 18-month operating cycle. The inspection and evaluation were performed following the approach used in the Inspection Criteria [1] and Screening Criteria [2] developed for Quad Cities, Unit 1, based on limit loud and imear elastic fracture mechanics (LEFM) techniques. The purpose of the screening criteria was to develop criteria that allowed indications discovered during visual inspection to be screened for further evaluation. The Inspection Criteria refined the screening criteria providing for minimum distributed sound material which would allow for operation for the specified operating period. Since the criteria were based on visual examinations, all flaws were conservatively assumed to be through-wall, and allowable flaw lengths were calculated using the appropriate limit load or LEFM techniques. The following sections of this letter report describe the methodology used in the initial inspection and the evaluation results. z= _:-. u= u. a = - = .-.- : ~ - - - -
~,_
Mr. T. J. Wojcik June 11,1994 Page 2 RAM-94-159/ SIR-94-052 Initial Inspection and Evaluation Methodology The inspection and evaluation approach employed at Quad Cities, Unit 1 provides the necessaryinformation for determination of the allowable flawlengths, including crack growth for the next operating period for all flaws observed, while assuming that the cracking is through-wall wherever it is observed. An initial sample of four to eight locations, spaced approximately evenly around the circumference of each horizontal weld in the shroud, represented the examination area for the initial in-vessel visual inspection (IVVI) for the core shroud. The sample was structured such that if sufficient sound metal was found visually to satisfy the screening criteria, the weld was accepted for continued operation for the next operating period. If sufficient sound metal was not observed, the IVVI was to be expanded as additional accessible locations were identified, given the physical constraints associated with the inspection, to other areas around the shroud, and was continued until sufficient sound metal was found, or until all accessible areas were inspected. This IVVI was performed on the outside surfaces of the core shroud for the H1, H2, H3, H4, H5, H6, and H7 welds and on the inside surface of welds H3 and H4. Additionally, ultrasonic examination (UT), using sophisticated state-of-the-art equipment, was performed on welds H2, H6 and H7 in order to corroborate the visual qualification. Acceptance Criteria The core shroud is a core support structure which provides lateral support for the fuel. The applicable codes, standards and classifications for the core shroud are as follows:
- The core shroud is classified as a safety-related component.
- The core shroud is not an ASME Code component. However, the original
. design is in accordance with the intent of Section III of the ASME Code.
- The evaluation of the core shroud was performed in accordance with the requirements of Section XI of the ASME Code,1989 Edition, Paragraph IWB-3142.4. [3]
Flaw Evaluation Results Following completion of the inspection of the H1, H2, H3, H4, H6, and H7 welds, flaw analyses were performed to demonstrate that the structural margins identified in the screening criteria were maintained for the actual flaw configurations which were identified. The flaw analyses were performed using limit load as the failure criterion for each of the STRUCTURAL INTEGRITY ASSOCIATESINQ
June 11,1994 Mr. T. J. Wojcik Page 3 RAM-94-159/ SIR-94-052 welds. The evaluation performed here takes into account the distribution of uncracked material around the circumference of the shroud, an approach less restrictive than assuming in the limit load analysis that the cracks are continuous. In addition, the H4 weld, which is the core beltline shroud weld, was also evaluated using LEFM fracture methodology to be consistent with the screening criteria [2]. The visual inspections of weld H4 were performed on the ID and the OD as separate examinations. The examinations did not overlap precisely at all azimuths. As a result, approximately 260-inches of circumferential length was examined combining the OD and ID results. The inspection length examined during the OD inspections was 162-inches, while 116-inches were inspected during the ID examination. A total of 15-inches of overlap occurred in the examinations. The flaw analyses for the evaluated welds were performed using limit load as the failure criterion for each of the welds. Substantial conservatisms were built into the flaw evaluation to account for the weld area examined and that area which was not examined, through-the-thickness crack growth and circumferential crack growth, and the screening criteria flaw prcximity criteria as applied to adjacent flaws. The specific conservatisms utilized in this evaluation are as follows:
- 1. A bounding crack growth rate (5x104 in/hr) through-wall and around the circumference vas applied to the cracks detected for the next operating cycle (18-months) for the structural margin assessment.
- 2. All inspected regions which are identified as cracked, whether by IVVI, or as reduced by UT in previously uncracked IVVI locations, are treated as through-wall cracks and grow by 0.625 inches at each end during the next 18-month operating cycle.
- 3. All uninspected regions associated with the IVVI examination are assumed to be cracked through-wall and are grown by a length of 0.625 inches on each side during the next 18-mcath operating cycle.
- 4. ASME Code pressure boundary safety margins were applied to these evaluations even though the core shroud is not a primary pressure boundary.
- 5. ASME Code, Section XI proximity rules for adjacent flaws were applied.
The conservative assumptions, described above were applied to each of the horizontal welds examined in this report. According to the screening criteria [2], the loading condition which governs the limit load analysis is the faulted condition. Table 1 presents tne membrane and bending stresses for the faulted condition which were used for the limit load analyses for each of the welds identified in the table. One notes from Table 1 that the highest loads are observed at the H6 and H7 welds, and th: lowest loads occur at the H1 and H2 weld
~
STRUCTURAL INTEGRFFY ASSOCIATESINC
i l l Mr. T. J. Wojcik June 11,1994
)
Page 4 RAM-94-159/ SIR-94-052 i locations. The limit load analysis was performed for all welds evaluated, the H1, H2, H3, ! H4, H6, and H7 welds. Table 2 presents the results of the IVVI inspection taken from Reference 4 for each of the horizontal welds evaluated in this report. The Table 2 IVVI results for the H5 weld are reduced by the UT results on that weld, as described in the conservatisms discussion above. The IVVI for H6 was performed on the OD. The UT examination of the H6 weld detected a single 2-inch flaw on the ID. Therefore, the visual inspection results were reduced by the UT results. These combined IVVI and UT results were used for the limit load analysis performed on weld H6. For weld H4, the IVVI results presented in Table 2 are ID results, representing an inspection extending a total of 116-inches in length. The ID inspection identified two 1/2-inch long flaws separated by 17a in circumference. The OD inspection of weld H4 extended a total of 162-inches and identified no flaws. Therefore, the ID inspection is limiting and is used in the limit load and LEFM analyses for weld H4. Table 2 also presents the IVVI sound metal locations (no flaws) which were used in the limit load analyses for all of the remaining welds (H1, H3, and H7). The results of the limit load analysis for each of the horizontal welds evaluated is presented in Table 3. One observes from this table that the factors of safety for the faulted condition (Table 1) range from 1.5 for weld H7 to 56.0 for weld H3. This compares to an ASME Code factor of safety of 1.4 specified for pressure boundary components under faulted loading conditions. One should note that the conservatisms utilized ere as described previously in this section. The H4 weld was evaluated using the methodology incorporated in the initial Inspection Criteria [1] for the ID surface IVVI inspection. The combined lengths inspected on the ID and on the OD far exceeded the minimum inspection lengths. A total of 162-inches of weld length were inspected on the OD by IVVI with no indications observed. A total of 116 inches of weld length were inspected on the ID containing a total or two indications, each approximately 1/2-inch in length at locations remote from one another. Consequently, a total of approximately 25 percent of the OD circumference and approximately 16 percent of the ID were inspected and found to be almost entirely defect free. The results of the limit load analysis for the H4 ID is presented in Table 3. The LEFM results for this location are presented in Table 4. Summary Based upon a review of the IVVI data for circumferential welds H1, H2, H3, H4, H6, and H7, as supplemented by the UT results for welds H2, H6, and H7, there is ASME Code margin for each of these welds under very conservative conditions to allow for continued operation for a minimum of one 18 month operating cycle. The analyses performed included limit load analyses under bounding design basis accident conditions, and LEFM for STRUCTURAL INTEGRrFY _' ASSOCIATESINC
Mr. T. J. Wojcik June 11,1994 Page 5 RAM-94-159/ SIR-94-052 the postulated highest fluence weld, and the evaluation was performed with the assumptions that all regions uninspected by IVVI or UT were cracked through-wall and that any cracking observed by UT was cracked through wall. Additionally, all uninspected regions were treated as through-wall cracks growing around the circurnference. ASME Code safety margins were used and were met or exceeded in all cases for the next 18-month operating cycle. Very truly yours,
' ~
hr R. Mattson, P.E. Associate
/mm attachments INTEGRITY ASSOCIATESINC
Mr. T. J. Wojcik . June 11,1994 Page 6 RAM-94-159/ SIR-94-052 References
- 1. GE Nuclear Energy," Recommended Inspection Criteria for the Quad Cities Unit I and 2 Shrouds", GENE-523-30-0294, March,1994.
- 2. GE Nuclear Energy," Evaluation and Screening Criteria for the Quad Cities Unit I and 2 Shrouds", GENE-523-02-0194, March,1994.
- 3. American Society of Mechanical Engineers Boiler and Pressure Vessel Code, Section XI,1989 Edition.
- 4. Commonwealth Edison Company, " Response to NRC Request for Additional Information concerning Core Shroud Cracking At Dresden, Units 2 and 3, and Quad Cities, Units 1 and 2", June 6,1994.
- 5. GE Nuclear Energy, " Quad Cities, Unit 1 UT Inspection Results, Weld H2, H6, and H7 Examination Summary Sheet", Report No. R-SO4, Data Sheet Nos. D-S05 through D-SO9, June 1,1994.
STRUCTUIUU. INTEGRITY ASSOCIATESINC
Table 1 Membrane / Bending Stresses Shroud Stresses Weld Location Membrane Bending H1 0.425 ksi 0.139 ksi H2 0.410 ksi 0.309 ksi H3 0.369 ksi 0.374 ksi H4 0.330 ksi 1.303 ksi H6 0.610 ksi 2.560 ksi H7 0.599 ksi 3.631 ksi NOTE: 1. All values are for the Faulted Condition. Per the Screening Criteria, the Faulted Condition governs for limit load and LEFM analyses for these welds. 4 l 1 Attachment to 1 RAM-94-159/ SIR-94-052 l INTEGRITY ASSOCIATESINC
Table 2
" Sound Metal" Locations Weld Location IVVI Locations H1 42 -49 , 132*-139o, 222*-229*,
322 -329 H2 42 -49 , 132 -139 , 222 -234 , 322 -329 H3 0 -85 *, 92 -1.t8 ,152 -160 , 165"-185 , 215 -225 , 260 -360* H4 83*-97 , 173*-187 , 260.14 -276.86 , 277.14 -280 , 355 -15 H6 15 -25 , 75 -85 , 105 -115 , 140 -145.4 , 146.38 -175 , 195 -205 , 255*-257 , 261*-265 , 285 -295 , 320 -355 H7 41 -49 , 135 -170 , 296 -304 , 320 -355 NOTES: 1. Values are from the " Shroud Visual Inspection Status", except as noted.
- 2. Values exclude identified indications.
1 1 j l 4 Attachment to 2 i RAM-94-159/ SIR-94-052
, M INTEGRITY ASSOCIATESINC
Table 3 Limit Load Factors-of-Safety Based Upor:IVVI Results Weld Location Factors-of-Safety H1 7.1 H2 5.8 H3 56.3 H4 5.6 H6 4.5 H7 1.5 Table 4 LEFM Factor-of-Safety Based Upon IVVI Results Weld Location Factor-of-Safety i H4 1.7 Attachment to 3 RAM-94-159/ SIR-94-052 STRUCTWUU. DITEGRITY ASSOCWESINC}}