ML20203E120
| ML20203E120 | |
| Person / Time | |
|---|---|
| Site: | San Onofre  |
| Issue date: | 09/11/1997 |
| From: | Haslinger K, Nadgor B, Wrenn J ABB COMBUSTION ENGINEERING NUCLEAR FUEL (FORMERLY |
| To: | |
| Shared Package | |
| ML19317C790 | List:
|
| References | |
| S-PENG-DR-003, S-PENG-DR-003-R01, S-PENG-DR-3, S-PENG-DR-3-R1, NUDOCS 9712160390 | |
| Download: ML20203E120 (76) | |
Text
. .
ssg-W-51-/8 -l
- l It is hereby certified that the analyses described Report 25 Pages in this design report have been property and Appendix A 6 Pages completely reconciled with the requirements of Appendix B 2 Pages Section ill of the ASME Soller and Pressure Appendix C 9 Pages Vessel Code,1989 Edition (no Aldenda) l DESIGN REPORT NO, S-PENG-DR-003. REV. 01 ADDENDUM TO THE PIPING ANALYTICAL STRESS REPORT FOR RECEIVEDCDM SOUTHERN CALIFORNIA EDISON NOV 211997 SAN ONOFRE UNITS 2 AND 3 gg ABB COMBUSTION ENGINEERING NUCLEAR OPERATIONS COMBUSTION ENGINEERING, INC.
WINDSOR, CONNECTICUT This document is the property of Combustion Engineering, Inc. (C-E) Windsor, Connecticut, and is to be used only for the purposes of the agreement with C E pu suant to which it is furnished.
The Design Analysis is complete and verified. Management authorizes the use of its results.
Prepared By: B. Nadoor SC9O[ Date: 09!///87 Cognizant Engineer Checked By: J. T. Wrenn >- I' LA ">s Date: .9!//!f7
/ /
Independent Reviehr f j)
Approved By: K.H. Hamlinnar / Il1 - d il 66 Date: 9 ////f 7 '
/
Supervisor, Structural Integrity and Testing (
This Design Report is certified to be in compliance with the requirements of the ASME Boiler and Pressure Vessel Code, Section til, Division 1, Nuclear Power Plant Component, _1989 Edition, up to and including the (NONE) Addenda, vru : w g yn m .fqt.y e en.,, L MN1
- h%v Certified By R T. Lubin P.E.
g*1. muovun-w Registration No. 15253 0: E
""*"'".o**.a".s*.**8'"*"8'***"***
. . , un i w
- Stste of connecticut -
- 04. vesmer w n m Date 9/t/ /f 7 assecos .
LWRY
- tes oromen as sepen ny CMas 3"
cm. .
i _
m L'Lf O '"*""'"*""~
sca spam asw e ses poetsencs sousowatsam 9712160390 771212 31 OR ADOCK 050
b-PENG-DR-OO3, Rev. 01 Page 2 of 25
_ . . _ _ . . . . . . _ . . . _ . . . . - . . . . - - - - - . < .- - "" ~
I RECORD OF REVISIONS 1
Rev Date ! Pages Changed Prepared By Reviewed By Approved By 00 6/26/97 Original B. Nadgor C.L. Mendrata K.H. Haslinger 01 1,2,4,5,14,16 B. Nadgor J. T. Wrenn H. H. Hastinger 9////87 20,23-25, Appendix B r
ABB Combustion En0 i neering Nuclear Operations
i S-PENG-DR-00:4, Rev.01 Page 3 of 25 ABSTRACT The Southern California Edi*on, San Onofre Unita 2 and 3, Hot Leg RTD Mechanical 'lozzle Seal Assembly (MNSh), which is a replacement of the "J" weld between an inconel 600 instrument nozzle and the hot leg pipe, is designed and fabricated under the requirements of Reference 5.1, Project Plan No. S3 NOME IPOP-0156, to satisfy the requirements of the ASME Code, Section Ill. The structural integrity ano acceptability of th'e design is established by the results of the detailed structural and thermal analysis contained in this report.
All stresses and cumulative fatigue usage factors within the scope of this report are satisfactory and meet the appropriate requirements from the ASME Boiler and Pressure Vessel Code,Section III,1989 Edition (No Addenda) (Reference 5.91.
ABB Combustion Engineering Nuclear Operations 1
)
S-PENG DR 003. Rev. 01 Page 4 of 25 i
TABLE OF CORIENIS Stcitan Pace No.
R e co rd o f Re visi o n s . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Abstract...........................................................................................................................3 TableofContent'a..............................................................................................................4 1.0 Introduction.......................................................................................................................6 2.0 Dealgninputs....................................................................................................................6 3.0 Analysis.......................................~....................................................................................8 3.1 iV N S A D e scri p t io n . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 3.2 Loading Co ndi t io n s . . . . . . . . . . . . . .. . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 3.3 S t r e s s C alcula tio n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 3 3.4 FelsmicLoada......................................................................................................22 4.0 S u m m a ry o f Result s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 3 0.0 References.....................................................................................................................24 LIST OF FIGURES 9 GUM. DESCRIPTILN EAGE 1 MNSASketch......................................................................................................8 2 C o m pr e s s no n Co lI a r . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1 3 UpperRs.nge.......................................................................................................22 LIST OF APPENC!CES A ENDIX At CO D E D ATE RECO N CILIATIO N . . . . . . . . .. . .. . . . . . . . . . . . . . . .. . . . . . . . . .. . ... . . . . . . . . . .. . . . . .. . . . . .. . .. . . . . . .. . A.1 APPENDIX B: AS S E M BLY D RAWIN G . . . .. . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B- 1 APPENDIX C: QU AUTY A S S U RA N C E FOR M S . . . .. . . . .. . . . .. . . .. . . . . . . . . . . . .. . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . .C.1 ABB CombustlOn Engineering Nuclear Operatior.s
S.PENG DR 003, Rev. 01
- Page 5 o* 25
- - - ~ - -- --
1.0 INTRODUCTION
1.1 OBJECTIVE The objective of this design report is to present the results of the evaluation of the Mechanical Nozzle Seal Assembly (MNSA) to be installcd on the hot leg RTD nozzle at the Southern California Idison (SCE), San Onofre Unita 2 and 3.
The MNSA is a mechanical device that acts as a complete replacement of the "J' weld between an inconel 600 inc,trument nozzie and the hot leg pipe, its function is to prevent leakage and restrein the nozzle from ejecting in the event of a through wall crack or weld failure of a nozzle.
The potential for these events exists due to Prirnary Water Stress Corrosion Cracking.
The MNSA for the hot leg RTD nonle has sirnitar designs and operate under the same conditions as the MNSA for the sido pressurizer RTD nozzle. Therefore, the analysis for the hot leg RTD MNSA is sirnitar to the analysis for the side pressurizer RTD nozzle MNSA.
This Revision is performed to modify the methodology used in calculating the load on the bolts connecting the MNSA to the Hot Leg.
1.2 ASSESSMEb'T OF SIGNIFICANT_DESICN CHANGFE This report presents the detailed structural end thermal analyses required to substantiate the adequacy of the design of the SCE. San Onofre Units 2 and 3 Mechanical Nozzle Seal Assembly as a replacement of the nozzle 'J' weld. This analytical work encompasses the requirements set forth in Reference 5.1 and is performed in accordance with the requirements of the ABB CENO Quality Procedures Manual QPM 101 (Reference 5.2).
1.3 ANALYTICAL METHOD Sta. Ward methods of elastic analvsis were used in this evaluation. This analysis follows the iequirements of the ASME Code Section III for Class 1 components and is anatyred for a 10 year life.
ABB Combustion Engineering Nuclear Operations
S PENG-DR 003, Rev. 01
- Page 6 of 25 2.0 DESIGN INPUTS .
2.1 grt FCTION OF DESIGN INPUTS 2.1.1 The Mechanical Nozzle Seal Assembly is considered a pressure retaining component. The design pressure is 2500 psi and design temperature is 650'F. Operating pressure and temperature are 2250 psi and 611'F, respectively (F;eferences 5.3, and 5.4, pg. 4).
Ambient design temperature in 120*F, Reference 5.7.
2.1.2 MNSA materials, msterial proporties ano stress allowable limits are given below and are taken from References 5.8, 5.9, Table I 1.2, pgs. 24 25, 46-47, and Tables 15.0 and I-0.0.
ham Material Compression Collar SA 479, Type 304 Lower Flange SA 479, Type 304 Upper Flange SA 479, Type 304 Trp Plate SA-479, Type 304 Hex Bolts SA-453, Grade 660 Hex Nuts SA 453, Grade 660
'De Rods SA-453, Grade 660 Material Allowable Stress Thermal Expansion Coeff. Modulus of Elasticity Sm (kall (a) IX 10inlin/'F) (E) IX 10' psil
~
650*? 120'F 200* 400*F 011' 70*F 650*F F F SA 453. Grade 660 26.8 8.27 8.39 . 8.95 28.3 25.0 i
SA 479, Type 304 16.2 8.60 8.79 9.19 9.54 28.3 25.0 2.1.3 Hot leg RTD nozzle and hot leg pipe materials are taken frun References 5.5, 5.6,5.17 and 5.19. Material properties and stress allowable limits art, taken from Reference 5.9, Table 1.l.1, pgs. 8 9, and Tables 15.0 and I-6.0.
Component / Thermal Expansion Coeff. Modulus of Elasticity Matenal la) (X 10
- In/in/*Fj (E) (X 10' pst) 400*F 611'F 70'F 650'F Nozzle SB 166 7.57 7.83 31.0 28.4
~
Thermowell Inconel 7.57 7.83 31.0 28.4 Hot tog pipe SA 616 Gr. 70 Sm @ 650'F = 18.4 ksa ABB Combustion Engineering Nuclear Operations
, AB __ -
S PENG DA 003, Rev. 01 Page 7 of 25 I
2.1.4 The bolts and tie rods have the following dimensions (References 5.8 t.nd 5.16): !
TIE ROOS l BOLTS '
0.500-20 UNF.2A O.37516 UNC.2A Major diameter 0.b000 in 0.3760 in Minor diameter 0.4405in 0.3005 in Basic pitch diameter 0.4075 in 0.3344 in Minorarsa 0.1480 in' O.0670 in' Stress ares 0.1599 in' O.0775 in' Rod length 6.25 in 2.1.5 Norrie dimensions are taken from Reference Drawings 5.5 and 5.6,6.17 er f 5.19:
Pressure Oiameter 0.993 in Length of nozzle toverafil 8.375 an Shell thickness 3.75 in min 4.125 in max Length of Thermowell O.b in Pressure Area = la r'l O.7 74 in' 2.1.6 The Mechanical Norrlo Seal Assembly design provides for 0.005 e sches of compression of the Grs' oil seal (cap between Upper and Lower Flanges, Reference 5.8). Such compression of the Grafoil creates pressure of about 3,500 psi, according to Reference 5.13. It is deemed to be sufficient to seal the possible leak area on the Nozzle, because achieved pressure exceeds the design pressure of 2500 psi at design temperature of 611'F. Sealing capabilities of Grafoil were verified during the hydrostatic test at 3,125 psi and three thermal cycles from near ambient temperature to 650*F and 2,500 psi vith borated water.
2.1.7 The hydrostatic test pressure conditions are not analyzed in this calculation because the seal components are not exposed to these conditions in acrvice. Hydrostatic testing was performed as part of seal qualification program (Reference 5.20).
2.2 ASSUMPTIONS 2.2.1 If no crack is present, it is assumed that the MNSA is not loaded during normal operating conditions. The only load would be exponenced if the nottle is subjected to a 360' through-wa!! crack at or above the J weld. This load would be equal to the internal pressure of the system against the area of the nozzle. This load would act against the top plate and distribute through the rest of the essembly back into the Hot Leg pipe. After this event occurs the load would become cyclical from essentially zero at Cold Shutdown to its maximum at Normal operating conditions.
2.2.2 A coefficient of friction of 0.30 for Grafoil on inconel 600 is assumed. "aferenu,5.10, Table IV shows tests results of Grafoil on stainless steel of 0.20 for 6 pressure of 12 psi.
For this apphcation, the load generated to compress the Grafoil seal is significantly larger than 12 psi. Extrapolating the information in Reference 5.10, a factor oi 0.30 is considered a reasonable assumption.
ABB Combustion Engine 0 ring Nucl0ar Op0 rations
' ~
t,. .. 3 .. v:. , . . ., 5 W,u-
~ ~~ ~cTW . .
SfENG.DRIQ3, Rev.01 3.0 ANALYgIS 3.1 MNRA DERL . 'nON The MNSA is a mechanical deviet that ogw. ps a complete ._ ': .= el the 'J' weld between er* Innonel 800 huvument romanie and the i et les pipe, h repieces the sessne function of the ,5 ' -
weld using a Orefet sosi arv:: M et th e noosio evwide desneser to the outer pipe surtees. .,
The MNSA ales repieces the weld seuensth ty rasens c( treeeded faseeners engaged Irl tapped holes in the muser pipe owfoes, end e eestronas piste deed in panee by teveedW 16e roda .Ns feature prevents tne nonte from electing from the pipe, should the *J' weld fell or the nesse develop a circumlerentesi ereek. A poneral etwtch of 6 MNGA la depleted in Figure 1 below.
l i
FIGURE 1 MNSA Concept Sketch for Hot Leg RTD Nozzle j
n .e R e ond Top Plote [ , [ ,,
, l Compression Coller N
("" C 3 y [{j .
h a rien neod Bolls l '
upper rionge l ,'
l 4 i..
ll -
! Lower Range rn1 s , ,
~
sffA6% % $
f 5
l s
l%
/
% ,s
/NN AN N l N ,, .
.i ,
N A k f. .)
a 1 - .
~
Seal Retointtr Geo fc' - Sec' (sp:it) (spilt) l i ABB Combustion Engineering Nuclear Operations
i AB S-PENO DR-003, Rev. 01 Page 0 of 25 I
I l
3.2 t OADING CONDITIONS 3.2.1 t_oading Due to preaturn The applicable loading is due to the pressure pushing against its entire crosJ section. From Section 2.1, the pressure area of the hot leg RTD nozzle assembly is 0.774 in'. Therefore, the load is:
Load = (2500 psi) (0.774 in') = 1,935 lbs 3.2.2 Loadmg Due to Thermal Fungg Under operating condrtions, it is assumed that the tie rod temperature increases from a reference temperature of 70'F to 120'F, and that the nozzle /thermowell is a perfect heat source, as well as lower flenge, and reaches the operating temperature of 611 *F (Section 2.1). These conditions produce the maximum gap closure between the nozzle /thermowell and the MNSA top plate. On the other hand, a more reasonable temperatures distribution is selected based on engineering judgment it is a~ Jmed that the tie rod temperature increases from a reference temperature of 70'F to 200*F, and that the nozzle /thermowell, as well as the lower flange, reach the temperature of 'Just' 400'F under operating conditions. These conditions produce the minimum gap closure between the nozzle /thermowell and the MNSA top plate.
The thermal expansion (maximum and minimum gaps) due to the displacement of the nozzle is calculated as follows (Reference 5.12, pg. 53):
8 due to thermal expansion = L a AT where: L is the length of the component (References 5.5, 5.0, 5.8, 5.17, and 5.19) n is the thermal expansion coefficient (Section 2.1)
AT is the differential temperature (as applicable)
Maximum / Minimum Closure = Nozzle + Thermowell- Lower Flange Tie Rod section The length of the tie rod section 'nvolved in thermal expansion may be calculated as follows:
L tie rod = L nonle + L thermowell . Shell thickness L lower plate = 8.375 + 0.5 . (3.75 or 4.125) - 0.365 = 4.76 in max and 4.385 in min.
So:
Maximum Closure:
4 (8.375 in. 3.75 in.)4 (7.83 X 10 in.An. 'F) (611-70)*F +
(0.5 in.) (7.83 X 10 in./in.
4
- F) (61170)'F .
(0.305 in.) 19.54 X410 in./in. 'F) (6117VF .
(4.76 in.) (8.27 X 10 in./in. 'F) 8120-70)*F = 0.018 in, Minimum Closure:
4 (8.375 in. 4.125 in.)4 (7.57 X 10 in./in. 'F) (400 70)'F +
(0.5 in.) '7.57 X 10 in./in.
4
'F) (400-70)'F -
(0.365 in.) (9.19 X 10 in.An. 'F) (400-701'F -
4 (4.385 in.) (8.39 X 10 in./in. 'F) (200-70)'F = 0.006 in.
I II p ABB Combustion Engineering Nucl0ar Operations
.*a..a -3 , e7 . e, sp
- y* y *.7 i .:aye .- -
SfWNHN6403 Flev.Oi i Page 10 of 25 i
A said gap steuld Le set to , allow for free thermol esqb6M=on et the neeste, but not to sesorsed the gap oaeraed for 6n the fosowing sections.
3.2.:P t'm -- === ear the km 1mn ftM M !
A cold pop between the de rods and the top plate should be est to account for the thermal esponelen of the nosale. If the nosele is ereceed, the impset loed would produce onesses en the tie rods and top plots whleh need to be conaldered A setting of 0.0203,,0.005.8s soonassionded for the hot see MNSA. It le resegnhed that the low end of INe tones le lees then the enamenum eleewa eineined in the prowlous seenon shoe the Idest oonedons used to <% ttu madmum cassure ers not aneciposed dwing operadon. the 0.015 minimum gap is concluded to be eseeptetto, The mandmum esed gap setting of 0.025' inmostos that a gap J 0.025-0.00s = 0.019' can esist dwing norw4J operation. Therefore, the sonenes due to the impost lead ese deseemned ensurming a gap of 0.019*. The stiffnesses of the tie rods and top pleen ese teken into conalderation in the calcadonon of the stresses.
~
,,J. .L . -
\/
f}%
The total defleetion due to the 'wnpoet lead is dete# mined below. It to eseumed that the one gy e 1 Impact le corwerted Into poesntial spring energy, and that the total ding.' J eement le aquel to the amours of semiar= ment showed by the gap plus the deflection of the nozzle and 9dNSA after impact. Equet6 ens are taken from floderence 6.18, pgs. 317 and 3 20.
Kincue Energy = Spring Energy 1 1
-m2 V* = 2K Ax' where: V=]2a:
1 m(n s) = 2 K bx' .
wheres F=ma 1
K Ax' Fs=~2 Assuming: a w Gap + Ax, and a friction coefficient of 0.30; then P a 0.7 l'.,, , where l',,,,is 2
load octing on the nozzle (1,935 lbs), and ABB Combustion Engineering Nuclear Operations
- - .-,,..m ,.,, .. .w,..,, - , - --,_,,,.....,....,-,.,,a- .., , , , , . , . ,
ABB S-PENG-OR-003, Rev. 01 Page 11 of 25 1
0.7F (Gap + b) = 2 K k' 1
7K Ax' - 0.7F G p- 0.7F & = 0 13.11 in order to deterrnine the Ax, the stiffness of tha tie rods and the top plate are calculated.
Stiffness of 4 Tie Rods (Section 2.01:
3 AE (0.0775in )(25.0X 10' , )# lbf K =4 =4 =1718400 1 4.5 lin in (the offective length of the rod is calculated as follows:
Ef. Length - L tie rod min (from Section 3.2.2) Thickness of upper flange +
+ Thickness of top plate = 4.385 0.75 + 0.875 = 4.51 ini Stiffnets of the Too Plate (Section 2.0)!
The equations for calculating the deflection of the top plate are found in Referenco 5.11, Table
- 24. Case 1a:
w a' y= D ( C, C, "
L, '}
where:
g,a 25.0X 10' '" O (0.875)'in'
-=
D 12(1-y ) 2 2 = 1533700 lbf 12(1- 03 )
and C,, C,, L,, and La are constants, and are calculated uning the equations of Reference 5.11, pgs. 398 399 where:
a = 1.906 in b = 0.56 in l r, = 0.665C in t = 0.875 in l y = 0.3 E = 24.8 X 10' psi l Ci = 0.7781 l
C, = 1.4149 La = 0.0264 L, = 0.2924 l
Solving for the stiffness of the top plate:
K ,, = 3
= 6886500 5 2.(C,L' - l' )
D C, ABB Combustion Engineering Nuclear Operations
A S PENG-DR@3, Rev. 01 Page 12 of 25 Q,plarmination of souivalant atiffnennt I
K,, = ; 3
= 13752405 Ku K+
The ecuation (3.11 previously developed is used to solve for ax:
1
-K Ar'- 0.7F Gap-0.7F Ar = 0 2
Hence, using a F = 1,935 lbs, and Gap = 0.019 in :
Ax = 0.0072 in Solving for the impact force:
Force, = K,,,, Ar = 1375240$(0.0072 in) = 9,902 lbf = 9.9 Alps in ABB Combustion Engineering Nuclear Operations
S PENO DR-003, Rev. 01 Page 13 of 25 3.3 $ TRESS CALCULATION 3.3.1 STRESSES DUE TO THF IMPACT LOAD Strest in the tie rodg From Section 2.1.4, A = 0.0775 in' impact Force = 0.9 kips / 4 tie rods = 2.48 kips Stress = 2,48 kips / 0.0775 in' = 32.0 ksi Stress = 32.0 kol < 2Sm = 53.6 kal Shear strema In the threads (0.37518 UNC-2A/2R)
The tie rods pass through the top plate and are held in place with nuts at the top and at the bottom. The top nut is the only one being loaded during impact. The nuts are of the same material as the rods. Therefore, the extemal thread area is considered.
From Reference 5.18:
As = n n Le Kn max (1/2n + 0.57735(Es min Kn max)) = 0.288 in' where:
n is number of threads per inch = 1G Le is the length of engagement (nut thickness) = 0.5 in (Ref. 5.8)
Kn ma.t is maximum minor diameter of internal thread = 0.321 in (Ref. 5.18)
Es min is minimum pitch diameter of external thread = 0.3287 in (Ref. 6.18)
Impact Force = 9.9 kips / 4 nuts = 2.48 kips Shear Stress = 2.48 kips / 0.288 in8 = 8.611 ksi Shear Stress = 8.61 kal < 0.6Sm = 16.08 ksi On the other side, the tie rods thread into the Upper Flange and are lot,ked down with lock washers. Since the Upper Flange is manufactured from a lower strength material than the tie rods, the strength of the Upper Flange thread is the critical element of this connection. Therefore:
From Reference 5.18:
A: = x n Le Ds min [1/2n + 0.57735(Ds min . En max)) = 0.414 in' where:
n is number of threads per inch = 10 Le is the length of engagement. Assume,it er,uals 0.5 in (Ref. 5.8)
En max is maximum puch diameter of internal thread = 0.3401 in (Ref. 5.18)
Da min is minimum maior diameter of external thread = 0.3643 in (Ref. 5.18)
Impact Force = 9.9 kips / 4 tie rods = 2.48 kips Shear Stress = 2.48 kips / 0.414 in' = 5.990 ksi Shear Stress = 5.99 ksi < 0.6Sm = 9.72 kai fAinimum allowable length of engagement of tie rod into the Upper Fiango may be calculated as s simple proportion:
ABB Combustion Engineering Nuclear Operations
S-PENG DR-003, Rev. 01 Page 14 of 25 )
Le m = (Shear Ctress / Allowable Stress) x Assumed Length of Engagement =
= (5.99/9.72) x 0.5 = 0.308 in.
l Shaar streat in the her botta (0.500 20 UNF-2A1 See Section 3.0.2.3, page 19. l Siteam in the too o!att The shear area is = n (D) t = x (1.33 in) (0.875 in) = 3.600 in' where: D is the diameter of the thermowell - 1.33 in (Reference 5.19) t is the thickness of the top plate = 0.875 in (Heterence 5.8)
Shear Stress = 9.9 kips / 3.656 in' = 2.708 ksi Sheer Stress = 2.71 kol < 0.6 Sm = 9.72 kol Bendi..a strema-The impact load it distributed over the eres of the top plate in contact wifh the thermowell. Conservatively, the impact load is app!!ad at the location of the outer rodius of the thermowell. From Reference 5.11, Table 24, Case la:
r, = 0.665 in (Reference 5.19) a = 1.906 in (Reference 5.8) b = 0.50 in (Reference 5.8)
W=F / (2 n) r = 9.9 kips / (2 m) 0.665 in = 2.37 kips / in bla = 0.2938 K. (b/a) = 0.6122 M=%Wa = 0.6122 '2.37 kips /in) (1.906 in) = 2.765 kips-in o = 6 (2.765 kips-in) / (0.875)8 in' = 21.669 ksi Bending Stress: o = 21.67 kal < 1.5 Sm = 24.3 kal 3.3.2 STRESSES DUE TO NORMAL OPERATING CONDITIONS f atter weld f ailure occural 3.3.2,1 TH Rods
- 1. At the threadeu area (Reference 5.8h Threaded area = 0.0775 in.
P = 1.935 kips / 4 = 0.484 kips a = 0.484 kips / 0.0775 in' = 6.25 kal < Sm = 26.8 ksi ABB Combustion Engineering Nuclear Operations
l S-PENG-DR-003, Rev. 01 Page 15 of 25 Fatious Analvgig For fat!gue, assuming that the nozzle is cracked through and that the load cycles are from 0 to 2500 psi, the cycle on the tie rods would be from 0 to 6.25 ksi. Reference 5.9, Table I 9.1 gives a f atigue life of infinite cycles (> 10").
- 2. Imp Plate Connection to T1e Rods The tie rods pass through the top plate and are held in place with r.sts at the top and at the bottom. The top nut is the only one being loaded. Because the tie rods and the nuts are the same r.interial, the external thread area is conside red.
From Reference 5.18:
As = x n Le Kn max (1/2n + 0.57735(Es min Kn max)) = 0.288 in' where:
n is number of threads per inch = 16 Le is the length of engagenient (nut thickness) = 0.5 in (Ref. 5.8)
Kn max is maximum minor dianieter of internal thread = 0.321 in (Ref. 5.18)
Es min is minimum pitch diameter of extemal thread = 0.3287 in (Ref. 5.18)
Pressure Load = 1.935 kips / 4 tie rods = 0.484 kips Shear Stress = 0.484 kips / 0.288 in' = 1.681 ksi Shear Strees = 1.68 ksi < 0.6Sm = 18.08 kal
- 3. Uoner Flance Connection to Tie Roda The tio Rods thread into the Upper Flange and are locked down with lock washers. Since the Upper Flange is manufactured from a lower strength material than the tie rods, the strength of the Upper Fiange thread is the critical element of this connection.
Therefore:
From Reference i,.18:
As = x n Le Ds min [1/2n + 0.57735(Ds min En max)] = 0.414 in' where:
n is number of threads per inch = 16 Le is the lengtil of engagement. Assume, it equa!s 0.5 in (Ref. 5.8)
En max is maximum pitch diameter of internal thread = 0.3401 in (Ref. 5.18)
Ds min is minimum major diameter of external thread = 0.3643 in (Ref. 5.18)
Impact Force = 1.935 kips / 4 tie rods = C.484 kips Shear Stress = 0.484 kips / 0.414 in' = 1.169 kai Sheer Stress = 1.17 ksi < 0.6Sm = 0.72 ksi I I ll Il I I I I ABB Combustion Engineering Nuclear Operations
S PENG DR 003, Rev. 01 .
Page 16 of 25 i 3.3.2.2I00 Plate Shear streen' The top plate will become loaded and the shear force will be equal to 1.935 kips. Shear i stress is proportional to one determined in Section 3.3.1 t- (1.935 kips / 9.9 kips) x 2.708 ksi = 0.529 ksi t = 0.53 ksi < 0.6 Sm = 9.72 kal
_Bendina in the Too Plate:
Dending stress is also proportional to corresponding stress determined in Section 3.3.1 o = (1.935 kips / 9.9 kips) x 21.669 ksi = 4.235 ksi Bt.iding Stress: a = 4.23 ks' < 1.5 Sm = 24.3 ksi 3.3.2.3 Bnit stressen Damien sizino Under normal operating conditions, frrur 0.500 20 UNF 2A be,us are loaded by 1,935 kips. The load acting on each bolt equa!: 1,935 / 4 = 0.484 kips. Stress area = 0.1599 in' (Sectiran 2.11.
Therefore:
Stress = 0.484 / 0.1599 = 3.027 kai < Sm = 26.8 ksl.
s Bolt Pre load The bolts are being pre-loaded to 30 ft-lbs fReference 5.8). To determine the load in each bolt, the following equation is used (Reference 5.15, pg. 302):
T = 0.2 F d ; hence F =T / 0.2 d where: T !s the applied torque = 360 in-Ibs d is the nominalinajor bolt diameter = 0.50 la.(Section 2.1)
Therefore, F = (360 ir. pounds) / (0.20) (0.50 in) = 3.600 kips.
himum Bott Lead Due to the flexibility in the design of flanged connection between the MNSA and the Hot Leg, the impact load during ejection of the nozzle will increase the load of the bolts. The stiffness of the flange relative to th6 stiffness of the bolts will determine what percentage of the impact load will be transmitted to the bolts. The total load on the bolt can be expresses by (Reference 5.21):
ABB Combustion Engineering Nuclear Operations
S-PENG-DR 003, Rev. 01 Page 17 of 25 j l
r a y"
Fmax = F,,,w +
.K + K, ,
Q l
I The stiffness of the components is calculated below.
i Stiffnnan of 4 Her Head Botta (Section 2.0h !
l The stiffness of the bolts la calculated using the same method described for the tio rods in Section 3.2.3. Dimensions are taken from Reference 6.8.
2 AE (0J599/n )(25.0X '"10. ) fy Kw = 4 =4 =9546300 1 1.675fn in 1 the effective length of the bolt, assuming 0.5 in of thread engagement:
I = thread engagement + lower flange + upper flange + washer =
= 0.5 +0.365 + 0.75 + 0.06 - 1.075 in Stiffneta of Overalt Flance The Hot Leg RTD MNSA has two cor sponents which represent the flanged connection to the Hot Leg, the Upper Flange end the Compression Collar. The stiffness of each of theses components is calculated using the same method described for the Top Plate in Section 3.2.3.
Ithe following equations are found in Reference 5.11 Table 24, Case Ia: all dimensions are taken from Reference 5.8).
w a' Csk Et*
y= -
D ( C, -L):
3 D= 2 12(1-y )
where C , C,, Ls, and L are3 constants, which are calculated using the equations of Reference 5.11, pgs. 398 399.
Upper Flange:
I Since the flange does not have a rectangular cross section, the dimensions are selected to
' produce the lowest flange stiffness.
e = 1.906 in b = 1.048 in r, = 1.048 in t = 0.75 in I y = 0.3 E = 25.0 X 10' psi C = 0.4358 3
C, = 0.57/3 L3 = 0.0112 L, = 0.2809 l
ABB Combustion Engineering Nucl0ar Operations
RF S PENG-DR-003, Rev. 01 Page 18 of 25 l Et 3 25.0X 10' # (0.75)'In' D= = '"
= 965800 lbf 12(1 -y ') 8 12 (1 - 03 )
Solving for the stiffness of the top plate:
2n r' = 4572900 K., y, = ".,(C,L' - LQ D C, Compression Collar:
Since the collar does not have a rectangular cross section, the dimensions are 1, elected to produce the lowest collar sitif.tess, a = 1.046 in b = 0.846 in r, = 0.621 in t = 0.75 in y = 0.3 E = 25f, X 10' psi Ci = 0.1864 C, = 0.1946 L, = 0.0086 L, = 0.2685 D=
Et
= 25.0X 10' " (0.75)'in' = 965800 lbf 12(1-y ) 2 12 (1 - 03')
Solving for the stiffness of the compression collar:
K _ ,,,, =
= 13246000 3
-D ( C,' -L ) 3 Equivalent flange stiffness:
The Upper Range and Compression Collar act in series against the bolts. The effective stiffness of the overall flange is calculated below.
Kj,,,,, = ,
=3399000 4
in K, K, Therefore, the maximum boit load is
, i ABB Combustion Engineering Nuclear Operations
S PENG.DR-OO3, Rev. 01 Page 19 of 25 l
l l
' 5 l 9546300 99 Fmax = 3.6 + - = 5.425 kips (9546300 + 3399000) 4 Tantite strina Stress due to the maximum bolt load is based on stress area = 0.1599 in' (Sectiori 2.1)
Stress = 5.425 / 0.1599 = 33.927 ksi < 2Sm = 53.6 ksi.
Shear strean in the threads 10.500-20 UNF 2Al2B1 The tie bolts thread into the Hot Leg Pipe and are locked down with lock washers. Since the Hot Leg Pipe is manufactured from a lower strength mtterial than the bolts, both external and internal threads must be checked. Therefore:
For butt thread, from Reference 5.10' As = n n Le Kn max (1/2n + 0.57735(Es min Kn max)) = 0.4 in' where:
n is number of threads per inch = 20 Le is the length of engagement. Assume,it equals 0.5 in Kn mar. is maximum minor diameter of internal thread = 0.457 in (Ref. 5.18)
Es rnin is minimum pitch diameter of external thread = 0.4619 in (Ref. 5.18)
Bolt load = 5.425 kips Shear Stress = 5.42b kips / 0.4 in' = 13.56 ksi Shear Stress = 13.56 kal < 0.6Sm = 16.08 ksl For pipe thread, from Reference 5.18:
As = n n Le Os min 11/2n + 0.57735(Ds min En max)) = 0.541 in' where:
n is number of threads per inch = 20 Le is the length of engagement. Assume, Le = 0.5 in En max is maximum pitch diameter of internal thread = 0.4731 in (Ref. 5.18)
Ds min is mintmum major diameter of external thread = 0.4908 in (Ref 5.18)
Bolt load = 5.425 kips Shear Stress = 5.425 kips / 0.541 in' = 10.028 ksi Shear Stress = 10.028 kai < 0.6Sm = 11.04 kal The minimum allowable length of engagement of hex head bolt into the Hot Leg Pipe may be calculated as a simple proportion:
Le,,,, = (Shear Stress / Allowable Stress) x Assumed Length of Engagement =
= (10.028/11.04) x 0.5 = 0.454 in.
ABB Combustion Engineering Nuwaar Operations
S.PENO-OR-003, Rev,01 Page 20 of 25 Str===== Ai= to thermal ernanalon!
The thermal expansion of the upper fienge and the lower flange could produce stresses in the botts. Since both the upper and lower flanges are of the same material, SA 479 Type 304, the thermal expansion is:
Dimensions for the upper and lower flanges are taken from Reference 5.8:
'Jppet flange thickness = 0.75 in.
Lower flange thickness = 0.365 in.
Expansion = (0.75 +0.365) in. (9.54 X 10 in/in/*F)(61170)*F = 0.00575 in, it is assumed that the bolt growth occurs in the area of contact v.ith the clamp assembly, then the thermal expansion of the bolts is:
8 Expansion = (1.115 in.) (8.95X 10 in/in/'F)(61170l'F = 0.00540 in.
Therefore, the stress in the bolt:
Stress = 68 E / L = [(0.00575 0.00540) 25.0 X 10' psi) /1.115 in.
Stress = 7.848 kai Primary + Secondary = Pre load + Thermal Primary + Secondary = 33.93 + 7.85 = 41.78 ksl < 3 Sm = 80.4 kol l Fatloue Umane Factor For a primary plus secondary stress of 41.78 ksi, a stress f atigue usage f actor is calculated us;ng l a stress concentration factor of 4 (Ref. 5.9, N8 3232.3) and a modulus of elasticity ratio, E /
j E n., = 30.0 /25.0 = 1.2. The attemating stress intensrty, S, is ca!culated to be:
S, = 4 i S,,,, / 2 (E /E )) = 4 (20.89 ksi (1.2)) = 100.27 ksi l
[
The number of allowable cycles Nm , was determined using Figure I-9.4 (MNS< 2.7S,,) of l
Reference 5.9 and equals 900. This transient is evaluated for 500 cycles of heatups and l cooldows J (References 5.3. 5.4, pg. 4), therefore:
U = 500 / 990 = 0.505 U = 0,305 < 1.0 i
l l
ABB Cornbustion Engine 0 ring Nuclear Operations
ABB S-PENC DR-003, Rev. 01 Page 21 of 25 i
l 3.3.2.4Comormazion cotiar MOURE 2. COMPRESS.Otd COLtAR
- soar 1
1
\ __
'- JMA;)
c air \ l Il l \ i l._, _ . , , ,
. .L q"i e t ser -*
S ett Sheatsitass The bolt preload,14.4 kips, acts as a shear force on the surf ace 0.180 inch inborsrd of the 2.453 diameter. See Figure 2.
The shear area is equal to (x)(D)(t) = (s)(2.093 in) (0.312 in) = 2.052 in 8 Therefore, the shear stress through the section Shear Stress t = 14.4 kips / 2.052 in' = 7.02 ksi Sheer Stress t = 7.02 kol < 0.6 Sm = 9.72 kol Baarina stress The bolt preload,14.4 kips, acts also as a bearing force oli the surf ace between outside diameter of the Compression Collar (2.453 in, see Fig. 21) and inside diameter of the Upper Flange (2.006 l in).
The bearing area is equal to (x/4)(D*,,w d*,n,,) = (x/4)(2.453 -82.090 )8 = 1.27b in' Therefore, the bearing stress Bearing Stress = 14.4 kips /1.275 in' = 11.294 ksi Beadng Strous = 11.29 ksl < Sy = 17.9 kal l
(For SA-479, Type 304 SST, Sy @ 650'F = 17.9 kai fro Hef. 5.0, Tabte 12.2) 3.3.2.5 Uncetflangc' The load is the same as defined above for the Compression Collar,i.e.,14.4 kips. The shear area equals to (n)(D)(t) = (x)(2.096 + 2x0.2085)(0.438) = 3.458 in' (Reference 5.8)
Therefore, the shear stress through the section:
Shear Stress t = 14.4 kips / 3.458 in' = 4.164 ksi Shear Stress t = 4.16 kai < 0.CSm = 0.72 kal ABB Combustion Engineering Nuclear Operations
l l
l S-PENG DR-OO3, Rev. 01 Page 22 of 25 FIGURE 3. UPPER FLANGE H 20e -
W E'; d'# 0.750' t ;
- l{
o n ss- ~~
W m
= any =-
Due to the proximity of the bolts and support surface, bending stresses are small and are neglected.
3.4 EtisMIC LM Seismic loads are not considered within the scope of this analysis. A scismic Qualification test program will be performed which will address the impact of seismic loads in the MNSA.
(Reference 6.14).
ABB Combustion Engineering Nuclear Operations Im ..-
_.._m.__ _m- . _ _ . __ . . ____m.__ _ _ - _ . - - . . _ _ _ _ _ _ _ - - - - _ - - - - - - - - - - - - _ - - - - - - - - - - - - - - - - - - - - - - - -
l S.PENG-ORC 03, Rev. 01 Frpe 23 of 25 ,)
- u --
4.0 fil.[MM&M "c RESULTS All stresses are satisfactory and meet the appropriate allowable almits sai for?h M Section 1:1 of the ASME Boiler and Pressure Vessel Code (Referenes 5.9).
The results presented below were determined using the assumptions datined and justified in Section 2.0. There are r'o additional contingencies or asssumptions that are applicable to these results.
Results of this analysis due to the impact load are sumanarizcd below Component Stress Calculated Stressiksil 1 Allowable Stress (ksil Tee Rods Tensse 32.0 53.6 Thread Shear 8.61 10.08 Upper Flange Thread Shear 6.99 9.72 Top Plate Shear 2.71 9.72 Bending 21.67 ~ 24.3 Results of this analysis under normal operating conditions are summarized below:
Condition Stress Calculated Allowable Usege Stress (ksi) Stress (ksi) Factor Tie Rods ,T ensilo 6.25 26.8 0.0 inreed Shear 1.08 16.08 0.0
^
Upper Fiange Thread Shear 1.17 9.72 N/A Shear 4.10 9.72 N/A Oesign Sizing 3.03 26.8 N/A 0.50-20 UNF Botts Tensile 33.93 53.0 N/A Thread Shear 13.66 10.08 N/A Prtmary + Secondary 41.70 80.4 0.505 Hot Leg Pipe Thread Shear 10.03 11.04 N/A lop Plate Shear 0.63 9.72 N/A Bending 4.23 24.3 N/A Compression Collar Shear 7.02 9.72 N/A Beanng 11.29 17.9 N/A ABB Combustion Engineering Nuclear Operations
S PENG DR 003, Rev. 01 Page 24 of 25 5.0 f7FERENCES 5.1 ABB Project Pen No. S3 NOME IPOP-0150, *MNSA for PZR and RTD Nozzles",
Revision 00.
5.2 ABB Combu6 tion Engineering Nuclear Operations Quality Procedures Manual QPM-101, Latast Revision.
5.3 ' Analytical ReWrt for Southern California Edison San Onofre Unit No. 2 Piping," Report No. CENC 1065, March 1970.
5.4 *Analytler' Report for Southern California Edison San Onofre Unit No. 3 Piping,' Report No. CENC 1507, May 1982.
5.5 CE Drawing E235178, Revision 4, " Piping Assembly, San Onofre ll, Piping".
5.0 CE Drawmc E235 763, Revision 4, " Piping Assembly, San Onofre 111, Piping".
5.7
- Design Specification for the Mechanical Nozzle Seal Assembly (MNSA) San Onofre Units 2 & 3*, Specification No. S3-NOME SP-0049, Revision 01.
5.8 ABR Drawing No.
- 1. E-MNSA 228-003, Revision 04, " Hot Leg RTD Mechanical Nozzle Seal Assembly *
- 2. E-MNSA 228-004, Revision 04,
- Mechanical Nozzle Seal Assembly
- 5.9 American Society of Mtchanical Engineers Boiler and Pressure Vessel Code, Section lit, 1989 Edition (No Addo 0 5.10 Union Carbide Grafoil, " Engineering Design Manual," Volume One, Sheet and Laminate Products, by it.A. Ho svard.
5.11
- formulas for Stress and Stra'n," Raymond J. Roark and Warren C. Young, Fifth Edition, 1975, McGraw-Hill.
5.12
- Mechanics of Matesals," Beer and Johnson, McGraw Hillloc.,1981.
5.13
- Test Report for Verification Testing of 60D Nozzle Seal Assembly *, Report No. TR.
PENG-012, 8tev. 00, February 95.
5.14 S.'esmic Qualification of the San Onofre MNSA Clamps for Pressurizer instrumentation Nozzle i.nd RTD Hot Leg Nczzles, Report No. TR-PENG-033, Rev. 00.
5.15 ' Fundamental of Machine Component Design," R.C. Juvinell, John Wiley & Sons, Inc.,
1983.
5.1 C
- Marks' Standard Handbook for Mechanical Engineers." E.A. Avallone and T. Baumeister lit, N6 nth Edition, McGraw H;:1.
5.17 CE Drawing E235 717, Revision 1,
- Piping Assembly Modification. San Onofre 11 & !II, Piping ",
ABB CombuStlon Engineering Nuclear Operations
S-PENG-DR-003, Rev. 01 l Page 25 of 25 5.18 ASA B1.1 1982, Unified Screw Threads,1982.
5.19 Thermowell CE Primary Loop (09 287), Drawing No. H33385-9002, Rev. C.
5.20 " Test Report for MNSA Hydrostatic and Thermal Cycle Tests.* Test Report No. TR-PENG-042, Rev.00.
5.21 "Daltimore Asymmetric LOCA Analysis for Upper Flange, Gird Beams, and CEA Shrouds",
Analysis Number 8067 640 73, July 1980.
ABB Combustion Engineering Nuclear Operations 1
S PENG-DR-003, Rev. 01 Page A1 of A6
- i- ,,,
APPENDIX A CODE DATE RECONCILIATION
S PENG DR-003, Rev. 01 I ,
Page A2 of A6 construcelan code omt. nacenciEation for RCE Machanical _NNbhA330NSA)
The purpose of this reconciliation is to demonstrate fulfillrnent of the requirements for use of a later edition of the Construction Code for SCE's Mechanical Nozzle Seal Assembly.
This is intended to a!!ow the use of ABB Combustion Engineenng's Mechanical Nozzle Seal Assembly, which was built to a later Code edition, at SCE.
In accordance with Southern California Edison Company (SCE) Purchase Order No.
OC267001, and San Onofre Units ll and lli Reactor Coolant Pipe and Fitting P.oject Specification No.1370-PE-140, Rev. 09 and No. 01470 PE 140, Rev. 04, correspondingly, the Original Construction Code associated with Design and Procurement for the Mechanical Hozzio Seal Assembly is the 1972 Edition through Summer Addenda (hereinafter referred to as the Original Code). The Original Construction Code associa ed with the installation is assumed to be the same as for Design and Procurement. The ASME Section XI program at SCE is governed by the 1989 Edition, No Addenda (hereinafter referred to as the Section XI Code). The Construction Code used for the Mechanical Nozzle Seal Assernbly project is the 1989 Edition, No Addenda, of the ASME Code, Section 111 (hereinafter referred to as the Replacement Code).
The SCE Mechanical Nozzle Seal Assembly Project involves both Repair and Replacement activities in accordance with the Section XI Code. Article IWA-4120 states that Repairs may be performed in accordance with later editions of the Construction Code, or Section lil, alther in its entirety or portions thereof". The Reptw-,nent Code is therefore acceptable for the Repair activities, which includes Installation.
The Original Construction Code for Design and Procuroment is the 1972 Edition of the ASME Code as noted above. The Section XI Code (Article IWA 7210) specifies that Replacements shall meet the requirements of the edition of the Construction Code to which the original component or part was constructed, unless the ic!!owing alternative is adopted (Article IWA 7210 (c)):
"(c) Alternatively, replacements may meet all or portions of the requirements of later editions of the Construction Code, provided that the following requirements are met.
(1) The requirements affecting the design, fabrication, and examination of the replacement are reconciled with the Owner's Specification.
(2) Mechanical interfaces, fits and tolerances that provide satisfactory performance are not changed by the later edition of the Cnnstruction Code.
(3) Modified or altered designs are reconciled with the Owner's Specification (Purchase Order) through the Stress Analysis Report, Design Report, or other ABB Combustion Enginecilng Nuclear Operations
l l
S-PENG DR-003, Rev. 01 Page A3 of A6 suitable method which demonstrates the satisfactory use for the specified design and operating conditions, whichever is applicable.
(4) Materials are compatible with the installation and system requirements."
These four requirements are addressed individually in Paragraphs (1) through (4), below:
BCAuirement
"(1) The requirements affecting the design, f abrication, and examination of the replacement are reconciled with the Owner's Specification (Purchase Order)."
Dirtuasion The Owner has specified the Original Construction Code ao the 1972 Edition, through Summer Addenda, of the ASME Boiler and Pressure Vessel Codo, ABB Combustion Engineering Nucteer Operations, acting as the Owner's Agent, prepared a design specification for the Mechanical Nozzie Seal Assembly in accordance with the Owner's Purchase Order, namely Design Specification fcr the Mechan! cal Nozzle Seal Assembly, Spec!fication No. S3 NOME SP-0049, Revision 00.
The Design requirements, as specified in the Owner's purchase order land Reactor Coolant Pipe and Fitting Specifications) for ASME Code Class 1 components, are per Article NB-3000 of the Original Code. The f abrication and Installation requirements are per Article NB-4000 of the Original Code. And, the Examination requirements are per Article NB-5000 of the Original Code. Similar Articles specify the Design, Fabrication and installation, and Examination requirements of the Replacement Code. The corresponding Articles for Design, Fabrication and installation, and Examination requirements of the Replacement Code are Articles NB-3000, ND-4000, and NB-5000, respectively.
An itemized comparison of each of the requirements of Design, Fabrication and Examination (called inspection in the original Code) for the Original Code and the Replacement Code is provided below:
Design The basic design requirements defined in Article NB-3000 of the Original Code are incorporated in Article NB-3000 in general, and in particular Article ND 3200 of the Replacement Code. Between 1972 and 1989 many more design critoria, categories and definitions were added to the Replacement Code, resulting in a more comprehensive Design Code. Thus, the significant differences between the two Design Code editions are the volume of written material and the editorial / acronym changes.
ABB Combustion Engineering Nuclear Operations
S-PENG OR-003, Rev. 01 Page A4 of A6 I
Overall, it can be observed that the Replacement Code is more prescriptive concerning vessel design than is the Original Code. It is therefore concluded that, with respect to Design, the Replacement Code is reconciled to the Owner's Specification.
Embrication and inntattation The intent of the Fabrication and Installation requirements defined in Article NB-4000 of the Original Code are also evident in Article NB4000 of the Replacement Code. Sirri.lar to the Design requirement reconciliation described above, the Fabrication and Installation requiremer'ts defined in Article NB-4000 of the Original Code lack the depth associated with those of Article NB-4000 of the R. placement Code. Once again the original intent of the Original Code is maintained in the Replacement Code, but with a significant increase in breadth of material content.
Additionally, the nuclear industry (including the Nuclear Regulatory Commission) acceptance of the Replacement Code requirements is evidence that it provides the same level of safety, if not greater, than the Original Code.
Framination Similar to the Fabrication and Installation requirements, the intent of the Examination requirements defined in Article NB 5000 of the Original Code and Article NB-5000 of the Replacement Code are essentially the same. The Examination requirements defined in Article NB 5000 of the Original Code lack the depth associated with those of Article NB-5000 of the Replacement Code in terms of the examination procedures and techniques. Once again the original intent of the Ori@nal Code is maintcined in the Replacement Code, but with a significant increase in the technical area and most significant changes in the acceptance standards.
The acceptance criteria in the Replacement Code may seem less stringent at first glance, but further examination proves the Replacement Code is at least equivalent to the Original Code. Additionally the nuclear industry (including the Nuclear Regulatory Commission) acceptance of the Replacement Code requirements is evidence that is provides the same level of safety, if not greater, than the Original Code.
Overall, it can be observed that the ReplacemJnt Code is more prescriptive concerning vessel examination than is the Original Code. It is therefore concluded that, with respect to Examination. the Replacement Code is reconciled to the Owner's Specification.
ABB Combustion Engineering Nuclear Operations
S PENG DR403, Rev. 01 Page A5 of A6
- 2. Baguirement
"(2) Mechanical interfaces, fits, and tolerances that provide satisfactory performance are not changed by the later edition of the Construction Code."
Discunzion The relevant interfaces, fits, and tolerances are associated with the seal between the Split Packing (Grafoil) of the Assembly and the Mechanical Nozzle. The Mechanical Nozzle Seal Assembly acts as a replacement pressure boundary, instead of the nozzle to hot leg pipe weld. The Mechanical Nozzle Seal Assembly is installed over the interface between the Mechanical Nozzle and Hot Leg Pipe O.D., and requires no modification of the Mechanical Nozzle for installation, in summary, the interfaces, fits, and tolerances that provHe satisfactory performance are evaluated in the Mechanical Nozzle Seal Assembly design report and consequently are in accordance with the Replacement Code.
- 3. Ranuirement
"(3) Modified or aftered designs are reconciled with the Owner's Specification (Purchase Order) through the Stress Analysis Reoort, Design Report, or other suitable method which demonstrates the satisfactory use for the specified design and operating conditions, whichever is applicable."
D11Cutslan This Design Report has been prepared and demonstrates that the modified design is satisfactory for use for the design and operating conditions spreified in S3-NOME-SP-0040 and the Owner's Specification (Purchase Order).
- 4. Raoulrament
"(4) Materials are compatible with the installation and system requirements."
Discussion The SCE Mechanical Nozzle Seal Assembly is fabricated from SA-479 Type 304 austenitic stainless steel, and SA 453 Grade 660 high alloy, high temperature botting material, which are compatible with the Hot Leg RTD Nozzle material of SB- ,
166 (inconel). The Original Code does not have any material specification for SA.
479 Type 304 austenitic stainless steel or SA-453 Grade 660 high alloy, high temperature botting material, therefore no comparison can be made between the Original Code and the Replacement C. ode.
ABB Combustion Engineering Nuclear Operations
l S-PENG DR-003, Rev. 01 Page A6 of A6 Because the Replacement Code has been accepted by the nuclear industry (including the Nuclear Regulatory Commission) and the Assembly's materials are similar in composition to the Mechanical Nozzle material, it is evidence that the Mechanical Nozzle Seal Assembly material, SA-479 Type 304 and SA 453 Grade 660, is acceptable for use as designated by the Owner's Specificatlun. it is therefore concluded that, with respect to Material, the Replacement Code is reconciled to the Owner's Specification.
ConrJunion it has been shown in the preceding paragraphs that the Replacement Code requirements are at least as prescriptive or more prescriptive than those of the Original Code. Therefore, it can be concluded that the requirements concerning the Construction Code date change for the SCE Mechanical Nozzle Seal Assembly are
, satisfied.
a ABB Combustion Engineering Nuclear Operations
S-PENG-DR-003, Rev. 01 Page B1 of B2 w
APPENDIX B ASSEMBLY IRAWING y
s
E r i e _
I < f ,
3 c .. i i _
l
, s.1 fi !
a!i g l!
si !! lll i
hhi!ii,ipiul l(i il'i 6
i I
f' 2l 1 .
i ,c .
9 4 l{i ii # -
'9 gjg
! i i E u >
4 i - ,
l r-c 1y
._th
,g ef$ei
- ---' i p s g e' g _
~ ,47 y~ .'
.g ,
- d. g ~
e g/e
. O-O- ~g4 l l
l
S-PENG DR-003, Rev. 01 Page C1 of C9 i
E i
APPENDlX C OUALITY ASSURANCE FORMS f
i 1
4 i
- - - - - ~
S PENG DR-003, Rev. 01 Page C2 of C3 DESIGN REPORT REVIEW CHECKLIST laktractinaat The independent Reviewer is to complete this checklist for each Design Report. This checklist may be incorporated into the Daign luport or maintained separate.
Title:
Addendum to the Piping Analytical Stress Report for Southern California Edison San Onofre Units 2 and 3 Document Number S-PENG-DR-003 Revision Number: 01 Yes N/A
- 1. Have all drawings been pnepared and independently reviewed in accordance with QP 3.7? 121 0
- 2. Are Checklists for the Design Analysis (QP 3.4) and Drawing (QP 3.7) review attached to Ed O the Design Report or on file with the CEO as quality records?
- 3. Have the analyses been separately prepared and independendy reviewed in accordance with QP 3.47 ; or Ed O
- 4. Is the analyses to be independently reviewed in conjunction with the compilation and Ed O verification of the design report?
- 5. Have all applicable TCRs, DCRs, NCRs, etc. been listed and reconciled in the Design Rena, O id
- 6. Are all applicable drawings and analyset used for design and construction in agreement Gd O with. and identified and described ir the Design Report?
- 7. Are the coacct revision levels of all desigr. output documents listed? O id
- 8. 11 ave provisions been made for a copy of the Owner's Review of the Design Report to be O E2I attached to the Design Report?
- 9. Does the Design Report contain sufficient details and references to permit certification by a Registered Professional Engineer (RPE)?
dO
- 10. Is the Design Report in accordance with the format requirements of the procedure? Ed O Comments (if any)
Checklist completed by:
Independent Reviewer i knnted Nanx g, p
/ 6 5tgnsture Dare V
ABB Combustion Enginee."ng Nuclear Operations
ABB l
S-PENG DR-003, Rev. 01 Page C3 of C9 Verification Plan
- Addendum to the Piping Analytical Stress Report for Southern California Edison San l Onofre Units 2 and 3 Document Number: S-PENG-DR 003 Revision Number: 01 lastruenans Desenbe t!.c method (s) of verification to be ernployed, i.e., Design Review, Asternate Analysts, Qualification Testing. a combination of these er an ahernative. Tbc Design Analysis Verification Checklist is to be used for all Design Analyses. Other elements to consider in formulating the plan are: methods for checing calculations:
comparison of results with similar analyses, etc.
Descrintion of Verification Method:
plethod o f vev 1.fsceh. ton is dest 3n review; mcla.d 3 :
t/e<<ly Lha.L appvop <a.,A.e amIy4csa.I meLhocis ac<s a, sed co w ee.hIy,
- Vers (y +ka.L atI dechnscal paa.msLe<.s assocta.ted wi6h 64e aalydica.( me4haals u.se<e covvecLty
.selec4 cal .fvoon 6 nn.c ea. h (< .sou rce.s,
. - j?e.v, ew n a.m es,s c.a. I calcu isLoom .for a ecu ta.Cy -
Venticanon Plan prepared by: Approved by: ,
J.T.7.' m } , Q , ,, g.
K. H. Haslinger g{ f , (1) yf tr%t Refeer enmed name and signatura Managment approver pnnted name and s:p=uidl V
ABB Combustion Engineering Naclear Operations
f S PENG DR-OO3, Rev. 01 Page C4 of C9 Design Analy:Is Verification Checklist (Page i of4) lastructions: The Irdeper. dent Review.t is te complete this checklist for each analysis and it is to be incorporated into the completed analysis, if a major topic area (generauy unnumbered, bold face type such as Use of Computer Software) is not applicable, then N/A next to the topic may be checked and :he check boxes for all items under it may be left blank.
Where there is no check box under N/A (not applicable) for a numbered item, such a res;wnse is generally inappropriate.
if N/A is checked in cuch a situation, document the basis at the end of this checklist in the Comments section.
Title:
Addendum to the Piping Analytical Stress Report for Southern California Edison San Onofre Units 2 Antt 3 Document Number: S-PENG-DR-003 Revision Number: 01 Yes N/A Overall Assessment
- 1. Are the results/conclusioes correct and appropriate for their latended use? 7
- 2. Are alllimitations and coattagencies on the resalts/ conclusions documented? [
Assignment of Cognizant Engineers,ledependent Reviewers and Mento.s I. Have Cognarma Engmeen. '" Reviewm and Menen. lf applicable, tra asigned mi approved by n . sw.3 g
- 2. If there are udtiple cognizant Engineers, has their scope been documerued?
O Gif
- 3. If the are mulapie Independent Ravwwers, has their scope been docurmersed?
O Clif
- 4. If the witi be nadtiple Management Approvers, has erir scope been documeoned?
g g
- 5. If an Irakpendese Revnmer is the superviser has the e propnam level of approval been documented?
O I Use of Competer Software (if For softwe which has baci. validated under QP 313:
O
- 1. Is the sotware applicable ibe tids analysis?
O
- 2. If that me significant changes in the nuk af softwm use, ha the Program Manager (s) tm.en consulted and have they initialed Mie approvals Fxtion of the Design Analysis la-Process Approvals form? O O For softwm which has not been validased under QP 3.13:
O 8 is the compvoer type, program name and revisic.a Wh hed?
O
- 2. Is the docins.antauon sumcient for trs % dependent Reviewer to concur that the software is appropriate for the analysis?
Q
- 3. Is the documernam sufficient for the Independ:nt Reviewer e concur that the results are corh :r?
O
- 4. If the A= -- 1 is liw.,w 4 by reference,is there assmance that the software acanny used is identical to that in the reference? O
- 5. If spadsheeu hm been usedJe h documenunon susciera for the M-M Reviewer to concur thu k rush a y comee ABB Combustion Engineering Nuclear Operations i
l
S-PENG-DR-003 Rev. 01 Page C5 of C9 Design Analysis Verification Checidist (Page 2 of 4)
Design AnalysisContents Yes N/A Objective of the Design Analysis
- 1. Has infonnamn neassary w dcrine the tak been inckaw or referencee g
- 2. lias the mason wtry the analysis is bemg periwmed w sevhed been h*==a-M g
- 3. lies the applicabilty and lascaded use of she results been docurnenter
[
Assessment of S3galAcast Design Changes
- 1. Have signiacas designeamed charsges em trdght kupact Ihis analysm tres considerte g 2, if any such changes have been idemifwd, have dry been m%ussely addressed?
IO AnalyticalTechalques (Methods)
- 1. Are the analyucal technignes (mcewds) descrobed in sufficent detail to judge ' wir appropnaseness?
[
11 Have anatyucal techmques snewporated bw reference to gerstric analyses, lead plant analyses or pectious cycle analyses hee" previouslyven ned?
O I
!!!. For moddiceacas er departures fronn prevmusty approved analyucal techruques or Convenuonal Er:gmiring Analysis Procedures (QP 3.19):
O I A. Are eey6mneni.dandjusiir c O O B. Have try been approved by Management initiating the Design Analysis in-Promss Approvals form?
O O IV. If supemded approved analyncal endiaiques ce Erigmaering Analysm Procedures are imed, is their usejusufied and approve #
D I V. Des the due ofissue of refuenced approved procedures or Er:gmeering Analysis Procedurcs twdate their use in this analysis?
I O Selection of Design loputs
- 1. Are lhe designloputs documenace g
- 2. Are the design inputs conecdy eclected and treccable to their source" 3.
4.
As redemscus as ducts as possible to the crismal source w documents containing collectxw= tabulations ofinputs?
Is the reference notshon appropnmely specine m the informauon utilised?
[
- 5. Are it benes foe sele: tion cf all design inputs documensed?
6.
7.
Is the veri 6 cation sanus of 96:n inputs transmined Rom causomers appropnate and 4- --
Is the verihm staan of design inputs trussuned frorn ABB CENS appropnate and documented?
O I g
- 3. Is the use of customer controlled suurces such as Tect Specs, UF;"'s, etc. aAwized, and does the authorization specify snendment level. revision number, etc.? O I Assumptions
- 1. If tbcrc at no = _. , ~-- =. is th.s docurrrnsed' O d
- 1. Are an ---% identified andjustifur I O
- 3. Are asstaripions which must be cleared by C'iNO or the customer listed on a Contingencies and Assumptions form?
O I
- 4. Is a peocess in place which assures that assumptrms whteh must be cleared by the customer will be included in transmittals to d= cunon=?
g [
ABB Combustion Engineering Nuclear Operations
I S-PENG DR-003, Rev. 01 Page C6 of C9
. _ _ - _ . _ - - - _ _ ~
Design Analysis Verification Checklist (Page 3 of 4)
Rasmits/ conclusions Yes N/A
- l. Are as results contamed in or referenced ici the Resutur encison section?
l [
1 Are aB lenhetions no 0. residts/cocclusions and their applicabihty documented in this section?
[
- 3. An as coe:ingeneses on the resets the must be ckared listed in the results/concitsion secdon and on a Contrngencies and Anompeor. form?
O I
- 4. Is a process la place whkh assures sat those contingencies which are the conomer's re3cnsibEgy to clear will be included in
- m _m.r, O d
- 5. lies a comparison of the results with those of a pwrion cycle or similar analysis been made and sigruficant difTerences explemec O d Other Elesments I. lieve appkcah e Codes kg. ASME Code) and standards been.pproprissely referencel and apsdied?
I O 1 is the informasion from relevant therecue searchedeckground data adequately & ----- -f andreferenced O
- 3. Are band caicadanons cocecs and appsopriately docusnenser I O
- 4. Is di applicabic computer oasput and input included?
O 5 Is all computer soeware used iderened by assne and revision identi6 cation?
g tk Are all micro 6che envulares identifsod with the analysis number and nurnber of sheets
- O
- 7. Are an Sles on CD-ROM lesneiend by the pash name?
O I 7 Are all cornputer disks identified with the analysis number? g References I. Are mit nemences used to perform the andysis lisied?
g 2, As thshferences a dincs as possibic and appropnase to the source?
[
- 3. 11 tie alumnos notadon speci5c to the informataan milised, inci ding nysion level er date of' sue,u and where appropnate, identinomion of the locadon of the infonnation in the refennce, such as page, table or paragnph number?
Forum / Format I. Is the document legible, aproducible and in a ibern suirahie for Eling and retrieving as a Quality Recore g
- 2. An all pages identified wtth the document number, including revnen msnbcr9 [
- 3. LM all pages have a unique page msuber' [
- 4. llave att changes beee authenecased by the inicals and date of both the Cogmzant Engineer, L as A.a Reviewer said. if mElmired, by Meissement? O M*
For a avision to a completed analyss O
- 1. Where practical have changes and additions beett identifkd by mechanisms such as vcrucallaies etc.? g
- 2. Where practical have deletbas been identified by rnechanisms swh as strike outs etc.?
O@
- 3. Have ind'ecaucas of cha1ges in prevk is revisions been amoved O@
- 4. Has a Record of Revisions page been added or revised, and dces it contain the extent of the revision?
7
- 5. Docs the distnbution or die *cvnion include those on the distribution of the previous revision?
[
ABB C amaustion Engineering Nuclear Operatior:
+ i ABB S-PENG DR-OLN 'Rev.01 Page C7 of C9 l
1 l
l Design Analysis VerificatioriChecklist (Page 4 of 4)
Yes N/A For a"compicts nyhnon";
[
L Have the tiets and document number tren presemd wideut change?
O IL Ilas the revhion number beca k .~M by onc?
O For a
- pare change packass":
gI
- i. - a bered - t -na? a
- 2. Are instrucoces pnmded kg the huanion and deletion of revised pages?
O
- 3. Has a new Title Page been prepared with the Package Centene re$nting the change package?
O
- 4. Has the original TR!c Page been retaned to presave the approval record' i
O S. Has a new Desism Analysis In-l% cess Appewals !brm been prepared?
O
! 6. Han the onginal Daign Analph In-f% cess Approvah Ema been raamed to pnserve the approval record?
O e - Ofany) l l
l l
l l
l ABB Cornbustion Engineering Nuclear Operations
S-PENG-DR403, R,tv. 01 Page C8 of C9 Independent Reviewer's Comments Comment Reviewer's Commest Response Author's Response Response Number Required? Accepted?
Checklist completed by:
Independent Reviewer J. T. Wrenn I
/~'* Ie 191_w g[ff/$7 Prmted Name i i Sistature Date Y
Its8 Combustion Engineering Nuclear Operationc
4 i.
l S-PENG-DR-003, Rev. 01 Page C9 of C9 Design Report Review Certificate This Design Report has been reviewed by the undersigned in accordance with the requirements of the ASME Bot:ar and Pressure Vessel Code, Section 111, Division 1, Nuclear Power Plant Component,1989 Edition, no Addende, and to the best of the reviewer's knowledge and belief is based upon the Design, Service, and Testing Loadings stated in the design specification.
Stress Report Vendor: CE Nuclear Ooerations Report No. S-PENG-DR-OO3 Revision: 01 -
Date: $///[i __
01370-PE-140 Revision: 09 Design Specification:
Date: 12-14-93 Design Specification: 01470-PE-140 Revision: 04 ,
Date: 12-14-93 S 3-NC,ME-SP-OO*9 Revision: 01 Design Specification:
Date: 6-24-97 Plant Owner: Southern California Edrson San Onofre 11 & 111 Designee: Combustion Engineering, Inc.
ABB Combustion Engineering Nuclear Operations Windsor, Connecticut Certified by: 661 / . Pl( hl Pmfessional Ennineer ![!!
Name k Title Date CT I L 9'n
. .. . . . . . . , . . . . . . . _ . . . _ , amnee ABB Combus^ ton Eng. :ering Nuclear Operations l
3 6 3-ya; November 5,1986 -
- 1.
1 MR. J.' T. REILLY SUBJECTh Corrective Action for the Unit 3 Pressurizar Nozzl March 1986
> A failure analysis for the Unit 3 Pressurizer nczzle failure in March 19u has i
'been performed. The results of the analysis are documented in the of the failed nozzle, has been thoroughly reviewed by Dr
- the University of California, San Diege. international well-known e for your information. His review letter is also attached iThe failure analysis concludes that the heat which contains the failed nozzle '
has several metallurgical characteristics prone to pure water stress corrosion cracking.(SCC).
should be replaced. However, due to a higher service tem:All nozzles made rature, two out of the three: nozzles in Unit 3, which ara located in tiie vapt. space of tiie pressurizer, should be replaced in the upcoming outage. However, the remaining.two pressurizers could nozzles, locatad in the water space of.the Unit 2 and 3 years from nc,w. be replaced at a later date but no later than three
- are described in this repsrt.The tpecift.
- ations for a pure wa,ter SCC resistant In:ene ,
specifications for the future replacement nozzles.They should be included in the im The conclusion reached by this failure analysis rejects the recommendation from C-E stating that the Unit 3 failure is an . isolated case. The rejection established theories and data for Inconel-600 pure water s cracking :.nd C-E's field operation experience.
- extensive experimental work on this subject is not usad in this failureAlso, Mr. L.
analysis since applicable thewater to pure dataSCC.
he collected in strong acidic solutioW are not r C A1.
C. CK[U CCt3870I/shg-Attacoments cc: Harold B.-Ray.
O. E. Shull
.H. E. Morgan.
'W. C. Marsh-K. L. Johnson:
S.-R; Gosselin- J. - L. .Reeder R. L Krieger M. P. Short COM Files a
. - - - - .- _ -~ . - - - .- - -. - ,. .
INTERIM FAILURE ANALYSIS FOR SAN ONOFRE UNIT 3 PRESSURIZER INSTRUNENT N0ZlLE C. Chiu, S. Gosselin-Introduction On February 27, 1986, instrument nozzle as a result' of a leak investigation.a vapor space leak wa prompted by a long suspected reactor coolant leak of 0.15-0.2 gpm from theT vapor space of the pressurizer (Reference 1).
brought to cold shutdown and the broken nozzle was groundAout. Consequently, the replacement nozzle was then welded in place.
A small portion of the nozzle (1/4 x 1/8 x 1/8 inch) containing the fracture surface and the remainder of the removed .iozzle were sent to Comb
-Both of them submitted metallurgical examinations an repo.ts (Referenc.es 2, 3 and 4).
Subsequent to the review of C-E's report, Mr. L. McKnight proposed that certain metallurgical and laboratory tests be conducted on Inconel-600 material.
Reference 3. The results of the test are documented in Purpose The purpose of this report is to review the con-lusions and the supporting arguments contained in the C-E and McKnight's reports. Additionally, the experimental data performed by other scientists regarding Inconel-600 IGA or feasibility of various failure scenarios hypothesized in the C-E's Analvsis The summarized results of C-E's laboratory examination as documented in Reference 2 ara below:
1)
_ The cracking mode on the fracture surface of the small sliver was entirely intergranular, characterized by well defined grain facets. An intergranular crack mode is characteristic of Intergranular Stress Corrosion Cracking (IGSCC).
larger piece of the nozzle. There was no e;iaence of cracking in the
'possible to tell where this cracking started.Because of the size of the sliver, it wa 2)
There were no signs of cyclic / fatigue-induced failure.
3)
The same design, material form, and fabrication techniques have been used with partial penetration instrument nozzles on other C-E supplied components including the San Onofre 3 pressurizer.
indicate.any pressurizer. anomalies during the fabrication of the San Onofre 3The records do no
- 4) '
The failed nozzle was fabricated from an Inconel-600 forging and was stress relieved at IC75'F for a period of 11/2 hours.
stresses-typically associated with a. rolled tube do not exist.Therefore, the 5)
Themicrostructureoftheremovednozzle(notthesifNercontainingthe crack) is characterized by very large grains, sizes 0-2 (ASTM Stan with small grairi boundary carbides, a lot of very fine intragranular precipitatss, and a narrow denuded zone adjacent to the grain boundar .
6)
CE believes that Inconel-600 tubing with a high greater than 55 Kst, yield IGSCC. strength has been associated with poor res,istance to pure water of IGSCC. 60.9 Kst, well into the range that as tubing has p This heat yields the highest yield strength among those of the five heats of material used at seven CE plants.
7) also used to make three (3) other nozzles asin the Un well as one in San Onofre Unit 2 and five (5) at another C-E plant. No problems longer than have Unit 3.been encountered at the other two plants which ilave' op nozzle of the SONGS Unit 3 pressurizer was due to a f Since the nozzle IGSCC would is made be theof mostAlley 600mechanism.
probable material with a high yield strength, pure water Further, .C-E believes that since this same heat of material has been used without cracking in other plants with longer service, this Unit 3 nozzle had a unique set of conditions that resulted in the IGSCC, believed to be a generic problem. Hence, this crack is not McKnicht's Reoort & Test Data The preliminary report by Mr. McKnight agreed to C-E's conclusion that the fracture mode is intergranular and sugststed a series of tests be done to investigate the metallurgical property of Inconel-600 and the failed nozzle The tests were subsequently performed by C-E under Mr. McKnight's supervision.
The results of tne test are summarized below.
1)
The nozzle material was forged in accordance with SB166 and had be annealed carbon at is content 1675'F withtoo considered a high.
carbon centent of 0.065% The to 0.07%.
amount of intragranular carbide precipitation wa:; obrerved.As a result, a 2)
There is 4.
boundaries. chromium depletion zone adjacent to 'the grain r ' cessing 1, sequences and due to the welding technique til the nozzle.
3)
Contrary 'to C-E's belief, McKnight believes that it is impossible to identify any corroding media responsible for the observed failure on thesurface and fracture service deposits because has been reinoved. the material has been deconta
P
- 4) .The tsmaining position of the removed nozzle was used to prep'are
- four specimens, i.e., 2, 2,:3 and 4. These four specimens were
.L first- pre-treated t#yteld different grain boundary morphology and
- then tested for their susceptibility to IGSCC. The pre-treatment, l grain boundary carbide morphology, and the test results for these four specimens are tabulated in table 1. In .ddition, Your more .
specimens made of standard Inconel-600 with a' lower carbon content
~ (0.03%) were prepared; two solution annealed and the other two solution annealed and then sensitized. The test.results of these -
four additional specimens are also tabulated in Table 1, but in the parentheses.
- 5) Based on the results of Table ~ 1 it appears that if the carbides tend to be larger size and less continuous in the grain boundary, there
-is less tendency for SCC.
' In summary, Mr. McKnight believes that the material of the failure nozzle close to the weld may have been exposed to elevated temperature in the realm i
l of 2000*F during welding.
some of the material may have.been sensitized through theUpon cooling do
- Howev'er, he stited that for'IGSCC to happen, .t would have been necessary'the1200 -
existence of a corrosion environment, a very high residual stress or applied >
stress,1.-and an unfavored microstructure, as evident of the test results in Table -
The mechanical properties of typical Inconel-600 as well as the
- failed nozzle are tabulated in Table 2. *.
past Research on-Inconel-600 IGSCC Inconei-600 has been known to be susceptible to IGSCC in the caustic environment (References 5, 6 and 7), in the lead doped water (Reference 8),
in the resin intrusion environment of H 50. solution (Reference 9), and in the pure water environment these references-are summar(zed i below. References 9, Significant.
10 and 11).findings in 1)
After an extensive study on the microstructure effect on SCC resistance, Airey stated-in Reference 12 that "the maximum improvement in SCC resistance correlates with a semi-continuous grain boundary carbide- s
> precipitation and phosphorous segregation tc the grain boundaries." This
. maximum improvements is associated with annealing at 1200'F (10-100 hours)
[- and 1300'F (10-24 hours).
2). When annealed at the top of the sensitization range (1200'F to 1600*F) in -
which carbides tend to precipitate at the: grain boundaries, large discrete precipitates period of time. due to an agglomeration process were observed in n very short '
Low temperature annealing (1200-1300*F) generates fine, semi continuous carbide precipitation and it does not agglomerate even after 100 hours0.00116 days <br />0.0278 hours <br />1.653439e-4 weeks <br />3.805e-5 months <br /> (Reference 12).
1
- 4
- - . - - - - - ~ .-- .-- - . - - _ . - - - - - -
- 3) Data taken by Bandy and Vanrooyen (R f ~
C e erence 13
- strain performance rate- i.e. 1 .in pure water have e demonstrated regarding Inconet-600's th )
is ~ proportion (al to the system tempe ,
4 cold worked Inconel-600 rature. - Moreover, withit pure growth rate seems that wat the l'
several other combinations
'shown in Figure 1.seems to accelerate of the material environmental-condition.
IGSCC process and
! .. H 2
The data = by Randy and Van '
- 4) - Bandy and Vanrooyen a found th t test with 0.05%C Inconel-600-is into be indep material. seems good agreement with that .
2 5)- Bandy and Vanrooyen's data show a' l increase in pH of the primary water L of lithium hydroxide to the test owering mediumin crack growth r i 6) .
Page and.McMinn (Reference 9) did a
- Inconel-600 environment. and Inconel-690 in the sin experimental comp 24 ' hours, and a combination for 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, three e of the form 840'F for mild ~ degree of sensitization may have bee i
e annealed A11oy-600 specimens. oy-600. ~ This fact suggests that
- 7) n present in all the mill
. slow strain rat-The data by Page and McMinn show that t
-(200 ppb oxygen)e at 600*F. However test for Inconel-600 with a-high pu n the SCC occurrence iswhen 1150*F or 840'F.
. independentppm, oxygen level increases SCC to 16 sur tion
- 8) In.1978 of whether or nei it is sensitize _
(pipe-connection) at the recirculatirat Duane Arnold BW determined to be the high residual on-inlet ress stnozzle.ound on an Inconel-6
.(Reference the existing crevice, and, 10, NUREG-0531). stayed in the sleeve a probably, The rootthecause c,orrosi*t were n anced material insufficient evidence to Reference indicatr resin intrusion incident 14 also concluded tha 9) contributed significantly to crack initiati e that sensitization is a factor that s
on or propagation.
A detailed study on the microstructu
~ --Inconel-600.has beenreperformed
. Laboratories. effects on e primary bywater EPRI SCCand of Batt The studyJexplains.when a semi continThe results of the s precipitation will' improve : SCC ce.
resistanuous intergranular carbi examination of the fracture surface ,
Based on a detailed result :in a reduction inurces crack-tip to generate strecarbides, y
dislocations, be localized corrosion will be hindered ss state.
the- theory:of film rupture SCC model)strain -
e rate exists J
localized oxidation-film - based
.e a f -=, - -
4
~
~10) The role of?the' tensile stress in the crack inttiation: time:has been-
- correlated by (predictions of_the mechanistic film-rupture model developed by EPRI.LThe time against all- available data.-crack' initiation-related to primary basedwater on .the model are calibrated SCC. .;
Based on the model, it is believed' that the. time-to-failure, beyond which SCC cracks will occur, is' inversely proportional: to the r. train rate. -
- 1
- 11) Based on the experiments by Bandy' and' Rooyen, and Coriou (Reference 16),
EPAI Steam Generator Reference Book (Reference-17) stated that the !
threshold for cracking of mill arnealed material in high temperature water is estimated to be 0.6 to 0.8 times the room temperature yield strength.
12): Based'on the fact that IGSCC ha: been observed with no difficulty in non-sensitized (as well as sensitized) condition and in highly pure i
deoxygenated (as well as oxygenated) water, it appears-that the effect of sensitization on SCC is minor or insignificant (Reference 17). This view is shared by S. M. Bruemmer et al sfter a review of the data documented in References 17, .18 and 19. They concluded that "Significant Chromium i depletion SCC."
or impurity segregation at grain boundaries is not essentist for by S. M.This-observation Bruemmer et al. is consistent with the mechanistic uodel developed L
- 13) C-E believes that.there are two common denominators that are characteristics of pure water SCC prone tubing. A paragraph from the l paper (Reference 21) by Mr. Owen of C-E is quoted below.
"3.1 Hichly susceptible tubino "There
'Coriou'are pronetwotubing.
common denominators that are characteristic of 2
Cracked tubing removed from plants such as Obrigheim , Doel II, Ringhals IIr and Trojan have exhibited a characteristic microstructure and yield strength. The high yield strength (-60KSI, 413MPa) is set by the fine grain sizo ASTM - 9 to 11). The fine grained microstructure- is the result cf a
' low temperature ( 1700'F, 926*C) final anneal which also dictates a very specific carbon inventory. 'Coriou' prone microstructures typically exhibit a preponderance of intragranular carbides, few if any intergranular carbides and little or-no solid solution carbon. '
"The absence of grain boundary carbides has been -shown to be-
' undesirable for resistance to 'Coriou' type SCC. - This undesirable '
microstructure, combined with the high yield strength-where elastic stresses can build and persist, apparently above the threshold required for. crack initiation, constitutes the ultimate-metallergical condition for 'Coriou' susceptibility. This condition develops as a direct result of employing low temperature-at final anneal."
Discussion-
[ Based on'the current understanding of the phenomena and mechanism of pure-water _ stress corrosion cracking as briefly discussed in the previous section, several-key points raised by CE and Mr. McKnight are discussed here.
. _ - _- 2.~ _
.__ __ ___ _ _ _ _ ~ _ _ . _ _ . _ _ __ _ _ . -
4 .g..
?
cl., '
55 Ks1 Yield Strenoth Limit g
As the understanding of the author, based on several conservations"-
with IE's metallurgists, CE has a 55 Ks1 upper yield strength limit for CE's steam generator tubes for many years.:
~
small grein size and.high residual- stress pers and, therefore, results in a higher susceptibility to SCC. This - a correlation has not been studied by other researchers but it is-consistent with susceptibility tothe SCC. current understanding in Inconel-600 upper limit f: One evidence to support the validity of this n that both W and Framatone steam generators have extensive purawater SCC Tailure experience whereas CE steam -t
'd and Framatone do not have. this uppor limit where F 2.
! Isolated vs. Non-isolated Case
- CE believes that the Unit 3 leakage is an isolated case because i
- with a yield strength of 60.9 Ksi and none of the oth exhibited symptoms of failure.
St. Lucie-2.
The location of these nine nozzles are tubulated !ib ,
3 nozzles Unit 3 vapor space 1 nozzle- Unit 2 water space 4
4 nozzles .St. Lucie-2 vapor space .
1 nozzle. St. Lucie-2 water space
- -strain Because the time-to-crack-initiation is both highly temperature and rate dependent (References 22 i
the water at the bottom of the pressu,rizer tends to have a longe life because of axial temperature stratification.
that has experienced fewer startups will last loager.Also, Notea nozzle that '
the strain rate. is non-zero only during a startup and the (References 22, 24). time-to crack-inittuion is infinite when the str i
Since St. Lucie-2 has only experienced about _17 startups (based on the data provided by St. Lucia-2 technical staff) since its i
commercial operation, whereas San Onofre Unit 3 has already has not experier.ced a similar failure. experienced about 22 star lower than those for the nozzles in the vapor spac pressurizer, the life is probably 2 to 4 times longar Speidel's data,. life increases by a factor of 2 for eve (based on ry 10*C 1
reduction in solution temperature, Reference 23).
Based on the'above discussion,-it is expected that one of the three
. nozzles, from the heat of 60.9.Ksi yield strength, located in the vapor. space of. the Unit 3 pressurizer will fail first. - Because this
-Inconel 600-pure water SCC, treating the Unit 3 pressuri e failure-incident as an isolated case is technically questionable'.
__. __ - _ . _ _ _ . _ _ _ _ __. _ __ _ ~ _
- 3. Forcino vs. Mill nnealing j
The failed nozzle was machined out of a forging, whose processing temperature is unknown and could be a little lower than the temperature typically used for mill annealing -(1900'F-2000'F). The effect of.
a lower processing temperature would be that a smaller amount of intra-boundary. carbon could be going into solution and diffuse to the grain granular i is more susceptible to SCC.As a result, the grain boundary carbide con l
- 4. ;
Hioh Carb3 Content of the Failed Nozzle The carbon content of the failed nozzle is 0.07%, within the specification steam generatorrange tubes.of Inconel-600, but higher than what is typical for A high content of carbon greater than 0.03% may be associated with a high intragranular carbon content, thus reducing the resistance to IGSCC. The solubility of carbon content at about 1800*F is 0.03%.
Therefore, for a low temperature forging, about 0.04% of carbon will not dissolve into Inconel solution and eventually migrate the grain boundaries.
However, this effect is considered minor by Bandy and Vanrooyen for Inconel with carbon content between 0.01% and 0.01.L Moreover, three out of six heats of CE's Inconel material used at its pressurizers have carbon content greater than 0.07% without any failures. Table 3 tabulates the properties for the pressurizer nozzles. of all the heats of Inconel material used by CE This fact also suggests that the effect of carbon content negitgible. on SCC susceptibility in pure water is minor, if not 5.
McKnioht's Exoeriments at CE The results of the experiments perfomed at CE under the supervision of McKnight, as tabulated in Table 1, are judged to be not useful or related to the at Unitunderstanding
- 3. This is becauseof the pure water stress corrosion cracking problem that there t enough difference in the SCC solution such that the conclusion drawn from the data ma applicable to the case of pure water SCC.
data show that the sensitization reduces the SCC resistance in an acit solution after solution annealed. Meanwhile, the sensitization process does not reduce the SM resistance in pure water and caustic solution after mill-annealed at various temperatures (see data in Reference 17),
I
. . ~ '8 -
- i
~ Relevant-Data-for Failure Analysis.-
. considered relevant to the' failure analysis.The following data
-1) .0.218-inches.-
The= outside; diameter of the failed tube-is 1.05 inches with. thickne I
'l 2)_-and 'The' configuration of the failed tube' with respect to the pressurizer the 1/8" stainless liner is shown in Figure 2.
- 3) The unit has been in operation since August,1983. Total operation time at the system approximated 600 days.temperature of 650*F in pressurizer before failure is i
at a high temperature. It has gone through about 22 pressurization cycles i,
- 4): Typical met.hanical properties of Inconel-600 are listed in Table 3
, 5) According to Westinghouse, all the instrument nozzles in pressurizer are ;
made of stainless steel-316, together with a $1-316 presrt'rizer liner.
IGSCC has not been; observed en these nozzles for sorm years, k Failure Analysis As discussed in the previous sections,. all laboratory a grain size, and copious intragranular carbide precipitation is susceptible to pure water SCC in.a high strain rate and/or local stress environment.
Based on the da';a collected and analyses performed on the failed tube, we k
.that:
1)
The yield strength 60.9 Ksi is beyond the acceptable limit (55 Ksi).
2)- Copious intragranulat earbides- are precipitated inside the grain.
3)- The location of ths highest location of the tube end inside surface. stress point is at the six and twelve o' clock With the residual stress from!
welding (estimated ir. Appendix B) and the pressurized hoop stress combined' (estimated in Appendix C by elastic model), the_ actual stress level is I probably 50% above if plastic deformation is the yield strength-(based on_a first order estimation)-
considered. The nozzle hoop stress is considered very small as compared to other stress components (Appendix A).
The cra
'close :k area tc:the observed at'highest with the the nozzle, stress.and is at the'five= o' clock location, very
- 4)' Thefaverage~ strain rate during plant start-up is 4 8 x 10ec.*/s If .
. the tube-is' as susceptible to pure water SCC as' the'~ mill annealed Inconel-600, star. up (Appendix the_D).. crack will; initiate after about 17 to 18 times of-
'15) The4 i grain: size estimated by McKnight's SEM picture of the cracked sliver
- itself, will notireduce the susceptibility to pure water SCC.s?
i
?
,-..w , ? +
-9 With the above information and the discussion in the previous se:tions, it pseems one toreasonable to conclude that the failed nozzle has a material composition pure water SCC.
whetcas a tube with mill-ann 1aled Inconel-600 (which is also prune to pure wcter SCC failure) will have failed after about 17 startups. This cor..:1usian, plus the fact that there is no sound evidence suggest'ing that the failed n1zzle has been subjected to a different operation condition than the other three nozzles located in the Unit 3 pressurizer vapor space, implies that another nozzle /ailure is expected, probably, within one or two years.
The hypothesis of a pre existing crack on the failed nozzle which had propagated not likely. during operation until the March,1986 incident) (is possible but This is because that based on the observed crack configuration '
its origin stress. seems to be at the inside edge of the tube, an area of the highest It is believed that a pre existing crack at this location would have been detected during the NDT of the weld after its completion.
Recommended Actions To prevent recurrence, the following corrective actions should be taken:
1)
Replacement of all nozzles from the same heat as the failed nozzle (Heat No. 54318). All three Unit 3 nozzles located in the vapor space should be inspected and then replaced as soon as possible within one year. The two other nozzles (one for each unit replaced within about three years) from now. submerged in the water should be 2)
The replacement resistant nozzle to pure-water should have material characteristics proved to be SCC.
Two candidates should be considered. One is Inconel-600 with low yield strength and a high annealing temperature
(>1900*F)
CE's plantsand crn past.
in the be selected from one of the heats which were provided to The other is Inconel-800 which has been used in the KWU steam generators with no known pure water SCC problem for more than 20 years (Reference 17).
3S70I
- l. ,. ...
i.
13BLE_1 ficKnieht er CE incopel - 600 ICSCC Test Res* sits With the ?:
~
M Ilozzle fe.071 Cl, I 1:
i Sl'EclMEN I PflE-TREATMENT. I i *
- I '1 I "
l MORPMOLOGY PflBILITY I .1
'l I
'As-received 1
l MACllESilNt 1
45 1* ? 3_CMORIOE i E I ~ -*f G -li ACIO I -:
SULF h a ACID l'
l- l Cont inese s ca rbide t 24 'r-EEE I .I l-1 I at C8 + Cr i Mo I 72 ~^ T_ . 2
-I.
1 I depletion i I flo l
i l 8 I I I Ito ,
I 2 I l i I I I I As-received + 1200*r senaltization i Semi-continssous i i i I for four. hours I i l l _I I I Carbide at C8 10 0 I No i l I I i lio i
I- l I i 3 I l I 1 I
I i I i
l' Solution(2annealed quenched at 2000*r more samples, 0.03%C)+ water i Semi-continuous i No I
~l I l
1 I
I Carbide at C8 I i les i Ito
~l I 4
, 1 I I (Iso n I 3 l' I i _I (Ile) l I i i
I Solution annealed quenched at 2000*F
+ sensitization + water I Continuous at 1200*F i Me i
- t. l t l Carbide at C8 I I Yes I I I1. For.i hours (2 more samples, 0.03%C)l + larger grain I I Yes i -
(Yes) I i I size i i t. (Yes) i 1
I I i
O 38738 t
. . . . . . .=. .. - . _. .. . .
TABLE 2 Typical (and the failed nozzle) A11oy-600 Properties
_ Room Temperature
'600'F Yield Strength 50.4 Kst (60.9 Ksi)* 40.9 Ksi-Tensile Strength
'110.3 Ks1 (108.0 Kst)* 101.3 Ksi Elongation 35.9%(36%)* 40.4%
Reduction In Area 56,9%(70%)* 53.3%
Note:
The actual strength depends on the actual annealing temperature.
The higher the annealing temperature, the lower the yield strength.
Based on the. data specified on the certified material test report.
1 e
4 3870I
A*
'$, 9 O IMAGE EVALUATION b
%fg;4+#
.. 1Es11AneErcur.33 ///zg/vf g
g e4, g)%+*' *%;f$p l.0 Em E T .-
"lj Ha i,i [m En Ji
~; -
i.25 _g_ i.6 4 150mm >
Ti (:f,.&
4tf,(,,',f__==_ .#3
"'ifj!!s5"'* .
l' ,
" e
- TABLE 3 Inconel-600 Materials Used by C-E Heat Grouc g,79,, y
-Yield (psi) 1*
c . .07 60.9K
- .05 36K
.06
~
38.5K 4
.08 -
52.7K
.09 38.5K
.09
_ 51.6K
- Note that Group I contains the five suspect nozzles and the failed nozzle.
l
=
t:
=
l l
38701 !
h
1 UPPER HEAD HANWAY *
- f UPPER LEVEL AND PRt.SSURE
( j
" ~
gggg N _ __
%,p NOZZLE
_ po PT'b LEAKING
% N0ZILE h MANWAY PRESSURIZEP' TOP VIEW OF FOUR LEVEL AND PRESSURE CARBON STEEL Noms 3 7/8" THICK g
N0ZZLE INCONEL v//
I
/
/
5/8" CRACK
, y f FACTORY WELD
~./ PRESSURE BOUNDRY LOCA FLOW 0.005" CAP h p INCONEL PUDDY LIMITER ANNULUS T 1/2" DEPTH OF WELD AT CRACK INCONEL PUDDY-N0ZZLE DETAIL
' '; A N0ZZLE STAINLESS 1"1kEL WELD CLADDINC 1/8" MINIMUM THICKNESS
+ '
CRACK -
SECTION A-A FIGURE I 4
..~
s WhM TM
's s 1L g 2
\. }N\f N i g I NkG- Oh l
i
/ s N M{}b s s 1 N
- s. .
% y A,p a 4 Aac h 3 .4
- Aem wAk & Ap h Floure 2 Front View of the Crack Configuration !
}
4
l Apoendix A Stress Distribution cf Instrument Nozzle at Location Away From Welding The tensile stress of a thick pipe when internal pressure P can be g
expressed by the following formula (Reference 25).
a2P S i b2 t = b8 a3 (1 + F) (Aal) where Py = internal pressure = 2250 psi a = 1.sside radius = 0.307 inch b = outside radius = 0.525 r
= radial 'distiice from center The maximumt S . occurring at the inside diameter, is:
S t = (0.307)2 2250 -
0.5252 max 0.5253 - 0.3072 I3 0.3072)
= 4588 psi (A*2)
The minimum St . ecurring at the outside diameter, is:
S t x 2250 irin
- 20.5253x 0.3072 - 0.3072
= 2338 psi (A*3) 38701
Appendix B Residual Tensile Stress Due to Welding During weldin weld recess. g, many passes of weldment were progressively applied to the butt The average width-of each pass.is about 1/8 inch. The radial shrinkage hoop stressofinlast thepass at each vertical welding plane will generate a tensile nozzle.
Note that one side the weldsent was attached to Inconel-batter weldment whichto was' was connected welded the tube. to the thick wall and the other side of t Because the wall is much stiffer than the tube, the tube, not shrinkage the wall. in the weldment will result in deformation in only the linear approximation The residual tensile stress can be approximated by the SR = AT* *t =E R (Bal)
AT = temperature change of the weldment - (2300*F - 650'F)
- = 8.6 x 10 ' in/in.*F, thermal expansion cosfficient t = radial width of one pass weldment R = mean radius of the no::le E = Young's modulus of Inconel-600 1
S p, 1650 x 8.6 x 10 ' x 5 x 29.5 x 10' O4
= 130.8 Ksi > 60.9 yield strength (B'2)
Conclusion Since the ca'iculated residual stress with linear approximation is greater than the yield strength, the tube was plastica 11y deformed during welding.
However, after deformation the residual stress would reduce to a level approximately equal to the yield strength of 60.9 Ksi.
Appendtx C . '
Stress Of stribution-and Strain Rate Around the Nozzle Inlet
-The stress 1 field-around the nozzle inlet area can be detersfr.ed by_
superimposing the following three stress components. '
il) The tensila-stress caused by the internal force applied to the
-i i
2250 psi.- well, which was subjected to a pressurizer pressure of pressurizer The stress is in both the circumferential and : ,
longitudinal directions-of the pressurizer cylinder.
- 2) The ' residual _ stress induced by welding. -
- 3) ^
The tensile stress in circumferential and longitudinal directions of the tube due to the internal pressure of the nozzle. This stress _is considered negligible as-compared to the welding residual stress _ and. the pressurizer hoop stress as described in (1).
The pressurizer hoop stress end. will result in a local high tensile stress at the sharp edge of the-nozzle sharp edge location which -is_ perpendicular to the nominalInhoop stress.T other inlet. words, it occurs'at the 12 o' clock and 6 o' clock location of the nozzle tensile stress en the sharp edge of the nozzle end.Meanwhile, the lo The highest concentration will be at the 3 o' clock and 9 o' clock locations. The numerical valves of the calculated below.. tensile stresses at 3, 6, 9 and 12 o' clock of the sharp edge nozzl Nozzle End Tensile Stress at 6 and 12 o' clock Locations-Based on the analysis documented by R. Peterson (Reference 26), the stress concentration factor for a- small hole in a large plate, which is in an uniaxial, circumferential tensile condition, is 3.0. Knowing.that the inside radi Js of the pressurizer cylinder is 48,125" and the thickness of the wall is 3.873".
.The hoop stress at the inside surface of the pressurized is calculated-as follows:
S t max = -
(br + ar ) Pg (see Appendix A) sa na
. , J 48.125 +'3.875): + 48.1252-(48.125 + 3.875)z-- 48.1253 x 2250 = 29.1 Ksi (Cal)
_ Including the stress concentration factors of 3, the maximum local stress
= at the 6 and.12 o' clock . location of the sharp-edged, ' nozzle end is:
S t max at_r.ozzle end = 87.3-Ksf -
e Nozzle End Tens 11' Stress at 3 and 9 o' clock- Locations' the stress concentration factor for a small hole in a la is in an uniaxial longitudinal tension condition is also 3.0. The longitudinal tensile stress can be obtained by force balance.
- S g, (s st) Pi (s b3 3 a3)
= 13.4 Ksi (C 2) the 3 and 9 o clock locations of the nozzle end is:Accountinyfort S
t max at nozzle end = 40.3 Ksf (C=3)
Suoernosition of Elastic Stresses by Linear Elastic Model At the 12 and 6 o' clock locations of the nozzle edge, the km:1 pal stress' at the 2250 psia can be calculated by the following formula:
Sp = ((87.3 + 60.9)8 + 13.48)b
= 148.8 Kst (C+4)
The 8.0 an'd 16.0 Ksi are tin minimum and maximum compressive, residual the longitudinal force exerted to the pressurizer wall. stresses caus Its direction is perpendicular inside pressurizerto that wall.of 87.3 Ksi, but on the same plane containing the At the 3 and 9 o' clock locations of the nozzle edge, the principal stress is calculated by the same method as that for the 12 and 6 o' clock locations.
S p, min = ((40.3 + 60.9)' .+ 29.12)b = 105.3 Ksi (C+5)
Note the tensile stresses caused by the nozzle internal pressure in either its circumferential or longitudinal direction are neglected because of their small magnitude compared to the tensile stress generated by the pressurizer internal pressure at the tip of the nozzle edge.
Actual Stress The actual local stress at the 6 and 12 o' clock locations will be lowe than 148.8 Ksia, but greater than 60,9 Ksia, if plastic deformation is considered.
of the' plastic deformation allowed by the surrounding A elastic m first order estimate of the actual stress is the average of the yield strength 105 Ksi. and the stress calculated from the linear elastic mode. That is,
.. ~. . . . ._.
- - . . _ . _ _ _ _ - -_ _ _ . . _ .. . . ~ . _ _. _ -___ ____. .._
Local Strain Rate In the Hihh Stress Area the 12 and 6 o' clock location on the edBased on the above analysis o average strain rate is determined here.ge of the Inc,nel-600 tube 'end. -The-highest stress point,- the area of high stress will beha Moreover,- the maximum strain is limited by the elastic b surrounding material. .
change during pressurization can be calculated as follows:With this rel 3
S E
. 29.1 Xst 30 x 10' ps1 = 0.97 x 10-3 (C+6) i The average time, is: strain rate, assuming an average of 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> of pressurization S
sn 0.97 x 10-3 8 x 3600 -- = 3.3 x 10 '
1 4
4 1
3S70I1
)
,-t
. Appendix D- '
Time-To-Failure Prediction -
Based on the model proposed by Y. G.-Garud (Reference 22), the Ihcone pure water model. SCC failure model can be approximated by the following failure i It I a AgP dt g = )1 0 (01) z where a t = critical value, beyond which SCC is considered in existence A = constant
- k = strain rate tg_= time-to-crack-initiation From the above equation, t is:
f t
g,(_A__ *t ) /.P
/E (D*2).
Based is 8.547 on secondSpiedel's and P =data 0.638,for mill-annealed Inconel-600 pure water SCC at 644
, (a /A) g t
f = 8.547/(3.3 x 10 8) " = 139.0 hours0 days <br />0 hours <br />0 weeks <br />0 months <br /> No.- of Startups = 139/8.0 = 17.3 times 4
3S701
e
, References .
L2.
" Metallurgical: Examination' of a Pressurizer Instrument Nozzid from San Engineering OnofreInc.,, Nuclear April 1986 Generating Station - Unit 3," TR-MSD-131 Combus i
3.-
CE letter from R. P. Daigle to J. T. Reilly, S-CE-10199, April 4,1986 4
M. 5. Mostafa and L. E. McKnight, " Metallurgical Examination of '
Station - Unit 3," Report No. SCE-160419, July 7,.'1986 L. McKnight and Associates, Pressurizer Ins
' 5.
I. L. W. Wilson, et al., " Caustic Stress Corrosion Behavior of Fe-Ni-Cr pp. 70-84 (1976) Nuclear Steam Generator Tubing Alloys," Nucitar Tech
~6.
G. J. Thous', " Relationship Between Acid Intergranular Corrosion and Caustic pp. 20-26 Stress (1977) Corrosion Cracking of A11oy-600," Corrosion, Vol. 33,
- 7. A.~
R. Mc!1ree, H. J. Michels, "The Stress Corrosion Behavior of Fe-Cr-Ni Vol. 33, pp. 60-67 (1977)and Other. Alloys in High Temperature Caus 8.
G. P. Airey, "The Effect of Carbon Content and Thermal Treatment On the
- SCC pp. 129-135 BehaviorNACE of Inconel A11oy-600 Steam Generator Tubing," Vol. 35, No.
(1979) 9.
A11oy-690 and 600 In Simulated Boiling Water Rea Vol. 17A pp. 877-887, Metallurgical Transactions A (1986) i
- 10. H. A Domain et al., "Effect of Microstructure on Stress Corrosion Cra of A11oy-600 In High Purity Water," Co rosion, Vol. 33, pp. 26-37 (1977)
L11. J. Blanchet, et al., " Paper G-13,- Presented at the International
[ .Firminy, Conference FranceOn Stress Corrosion Cracking. And Hydrogen Embrittlement.
(1973)
- 12. G.-P. Airey, "The Stress Corrosion Cracking _(SCC) Performance of Inconel-Alloy 600 in Pure & Primary Water Environments," "In Proceedings of the International Symposium On Environmental Degradation Of Materials In
-Nuclear NACE(1984) Power Systems - Water Reactor, South Carolina,22, August 1983,
, 13. R. Bandy and D. VanRooyen, " Mechanisms of Stress Corrosion Crack
-Intergranular Attack In A11oy-600 In High Temperature Caustic and Pure Water," Vol. 7, No. 3, J. Materials for Energy Systems, ASM (1985) 14.Piping NUREG-0531, " Investigation and Evaluation of Stress-Corrosion Crac of LWR Plants," US NRC, February 1979
-.m--- - - - , -. , , - , - .,--
. ~
References (Continued) 15.Project S.M. Brueamer et al., " Microstructure Effects 2/63-4, Final Report August 1986
-16. H. 1566 October Coriou, et al., " Sensitivity To Stress Corrosi
- 17. A.R. McIlree et al., " Primary Side Stress Corrosion Cracking," Chapter EPRI' Steam Generator Reference Book (1985)
- 18. "Instru: tion Manual, Pressurizer, San Onofre Unit No. 3," CE Book No. 72370 and SC23-919-74-0, SCE,.0ctober 1977
- 19. (1981)
E. Serra, " Stress Corrosion of Alloy 600," EPRI Report No. NP-2114-SR
- 20. D. Vanrooyen, Corrosion Journal, Vol. (1) p. 329 (1975)
- 21. G. P. Airey, " Optimization of Metallurgical Variables to Improve the Stress Corrosion Resistance of Inconel-600," EPRI NP-1354 (1980)
- 22. C. rt. Owens, " Preventive And Corrective Actions The Primary Side Stress Designed Steam Generators," SessionCorrosion Cracking Perform 5-1, 1984 Generator Problems in Stockholm, Sweden (1984).Cnnference on Steam
- 23. Y. S. Garud, " Development of a Model for Predicting Intergranular Stress EPRI (1985) Corrosion Cracking of Alloy-600 Tubes In PWR Primary Water,"
- 24. M. O. Speidel, " Overview of Methods for Corrosion Testing As Related to PWR Steam Generator and BWR Piping Froblems," Predictive Methods for Assessing Corrosion Damage To PWR Piping and PWR Steam Generators, eds. H. Okada and R. Stachle, pp. 31-44, NSCE (1982)
- 25. :1 Coriou, et al. , " Influence of Carbon and Nickel Content On Stress Corrosion Cracking of Austenitic Stainless ' teel Alloys In Pure Or Chlorinated Water at 350"C," Proceeding of ,nference on Fundamental Aspects of Stress Corrosion Cracking, NACE-1, Houston (1969)
- 26. Company, J. F. Harvey, Theory and Design of Pressure Vessels, Van Nostrand Reinh New York (1985)
- 27. R. E. Peterson, Stress Concent.ation Fa; tors, John Wiley and Sons
, 1974 38701 l
~ ._ . . . _ . . - -- -- ~
4 ve.nsirrUr t.;A1.IFORNIA. SAN DIEGO
'e
.uva.sm==.umt=cau. - .un
- .unemmeme. h unn unuu . una cava DEPARTMENT OF APPLIED MECHANICS AND E
- ENGINEERINC 3CIENCES. MAIL CODE S-010 .
. LA JOLLA. CAttFORNIA 92093
. ?
30 October -1986
-Mr. John S. Steibel Vice President-Torrey Tech Inc.
- P.O. Box 786 -
.Solana Beach,-CA 92075 .
I Dear Johns Re Interim Failure Analysis for. San onofre Unit 3 Pressurizer Instrunent Nozzle. ,
Per -your request,
.S. I-have reviewed the nemo from C. Chiu Gosselin .on tho' exanination of t.nd the cracked pressurizer
.and instrument' nozzle in-San Onofre Unit 3. The supporting documents papers which assessnent, telated is to this topic attachen have also been included in my herewith.
I ~ agree with ' the conclusions reached by C.Chiu and their report. S. Gosselin in ,
backCround infornation Their analysis of the problem and is quite complete and accurate. the reviou of the Sincerely.
J4.e d 7~f, Q Mas s ourt T. S iunad
-Adjunct: Professor Materials Science & Engineerint and Nuclear Energy '
enc 18 .
.g,. , --
UNIVERSITY OF CALIFORNIA. SAN DIECO starttrv e paves
- tav xt e am awctLrJ
- nrvtAstat + SA:: )
/n -N A __
4 Disco . sAw rnexcisco (. [ sAwrA mannARA
- SANTA CRUE h y ,/
/
DEPARTMENT OF APPLIED MECHANICS AND ENGINEERING 3ClENCES MAIL CODE B 010 LA JOLLA. CAL 1?ORNIA 92093 C 0!DIENTS ON INTERI>l FAILURE ANALYSIS FOR SAN ONOFRE. UNIT PRESSUnIZER INSTRUMENT N0ZZLE by M. T. SI!!N AD University of California in San DieSo 30 October, 1986 The Onofre San follouinC Unitare 3 my connents on the interim failure analysis for pressurizer instrument nczzlet
- 1. The susetptibility of high nickel alloys, such as Inconel-600, to intercranular stress corrosion crachi.2n in pure water at elevated tenperatures has been uell established since the pioneerinn work failure of Coriou over twenty years aCo. The time to tensile stress is neverned by a number of variables, including the level and the strain rate, the of structure the alloy conposition operations, resultinn from of prior heat treatments and forninC the alloy, surface purity and oxyCen content treatment, and pil of the unter, temperature. Most of the experience and the PWR steam has been obtained f r o r..
nunber of generator Inconel-000 tubing. There are also a cases of IGSCC failures in Swedish .%'R instrumentation tubing and core screw ennponents.
Inconel No failures have been observed with Incoloy-800. such
- 2. The C-E conclusions failed noz;1e material,are based upon the examination of the other nozzics made and upon the fact that a number of not from the identical heat of material have that failed after longer exposures. The statement to the effect somewhat "thispersuasive.
crach is not However,believed to be a generic problem" is the postulated "umique set of it is unclear conditions that as to whatinwere resulted ICSCC in San Onofre Unit 3". The question is the exposures in the San Onofre units will whether lonner undamaged nozzles too, if the result in IGSCC in the susceptibilities of these stress levels or the relative nozzles are The accelerated somewhat different.
acid solutions arestressnot corrosion tests in strong relevant to this problen.
alkali and
- 3. It is important to note that the heat of material for the cracked nozzle has too hinh a yield strencth of 60.9 Ksi, well into data.
CE the range It where poor resistance to ICSCC is indicated by yield has been observed that Inconel-600 tubing with strenCths ICSCC in pure unter. greater than SS Ksi are nost s u-s c e p t ib l e to
.u.nu 4 There is little doubt that the material of the failed nozzle 4
close to the durins the weld weldincwasoperntion.
exposed to sensitizing temperature rance stresses in _ the The presence ,of high tensile l
susceptible structure of result in IGSCC the alley could ;
in the Inconel-600 nozzle material.
- 5. The dis'cussion
- heat of Inconel-600, of the nine intact nozzles made from the same is well supported by theoperating experimental in other units and locations, evidence.
time for IGSCC is both strain rate and The threshold Hence, nozzles that temperature dependent.
lower temperatures have experienced fewer stkrtups ane/or The cracked nozzle was will have correspondingly loncer lifetimes.
650'F exposud at the system temperature of before for approxinately 600 days and 22 pressurization cycles it failed.
to The crack is located at the location where the stress is highest (estimated to be 50% aboveclose the yield strennth), resultin:
and the pressurized hoop stress. from residual stresses fron weldinn
- 6. In CANDUGernany (KUU) and Canada reactors), (for the boiling version of the the alloy Incoloy-800 was following its highly specified Dotton HTGn successful application in the stean generator tubinC (no leaks in 7 Peach operation). years o f.
two decades tiso,on extensive studies in France durinn the past water of Inconel-600,caustic Incoloy-800, stress corrosion and IGSCC in pure steel and type 316 staininss stresseshave provided very useful information on tbc thresho)d tests werefor stress co rosion cra'ching of these alloys. The carried out at 350'c (662
- F) on specinens under constant tests showed strain and under constant load. The results of these that corrosion, in NaOH for Incoloy-300 resistance to caustic stress cracking occurs so'utions below 10% concentration, no (280 MPa, 40ksi).even at stresses exceedin5 the yicid strength corrosion Inconel-600 showed susceptibility to stress crachinn pure in all concentrations of caustic and in Heat water treatment improvedasthe at stresses low as one-hnif caustic stress the yield stren6th.
resistance of Inconel-600. The corrosion cracking Super (Ph.Berge, Phoenix fast breeder reactor issteam nonerator tubing for the all Incoloy-800.
17(4), 291 (1973). et al, Corrosion 33(12), 425 (1977); Nucl. Ener5y 7 In very thesuccessfully KWU steam generators Incoloy-800 tubing has been used since 1969 in about specification complies with ASTM 20 PWas. The narrower compositional specifications 163-66, with somewhat nicxel (32 to 35%) toward hicher mean and Cr (20 to 23%) contents and lower permissibic carbon contents of (0.03%). The stabilization ratios to titanium-to-carbon prevent and C+N are m12 and =8, respectively, sensitization at welds. All KWU steam tubes in order aretoshot peened with class balls on the outside ncnerator nenerate dianeter mitigate compressive surface residual stresses Incoloy-800stress corrosion. to in KWU The decision reactors is ca:ed upon to continue to use extensive tests in laboratory experiments in the results a model of steam 3
p.g. ;j D '
~' concrator and the many years of excellent perfornance in operatinn-PURs with Incoloy-800 tubinc. In contrast, 600 Inconel-tubine, in intergranular cracks. KWU reactors showed both in,t e rnal and external
- 8. Canadian studies have shown that tests in pure wat'er_and in diluteTNaOH tests solutions provide a better sinulation of SCC than -
maximun inconcentration concentrated caustic of free solutions hydroxide which that far exceed the can form in heated- crevices.
available The lat ter - concentration 'is governed by the temperature superheat (difference between- the saturation caustic and the primary coolant tenparature). In dilute-solutions be -practically immune and in pure water Incoloy-800 was found. to Inconel-600. and far more resistant to SCC than 369 (1978). (R.S.Pathania . and J. A.Chitty, Corrosion 34(11),
- 9. The results of an extensive study of SCC of steam generator tubinn materials in a cyclic steam environment were reposed in 1975 carried out by ORXL and Southern Nuclear Engineerinn. -The tests were nenerator in a loop in steam taken directly fron a steau superheator (Bartow Plant, Florida Power). These tests of yielded results that differed radically fron the results conventional chlorido.
laboratory tests conducted in boiling Mg-Cr (e.n., The alloys that resisted SCC contained at least 22%
Incoloy-800). Nickel alloys susceptibility that showed Inronel-630, withto16% SCC all contained low anount's of Cr (e.g.,
Cr).
(J.P.Cammond. et al, ORNL-5031).
10.The EPRI balanced assessment " Steam Cencratorof Reference Book" does not give a Incoloy-800 for PWR steam Concrator tubing.
the relative merits of Inconc1 and Paine, et al, (p.4-4) "
stato that Incoloy-300 is less desirabic than available to US Inconel-690 utilitics and due totoitsa lesser corrosion data bank-SCC," suscept ibility to caustic alloy in the while EMU acknowledninn PRRs since 1969! the excellent performance of this (p.4-3) that for Inconcl-690 (theirThe same authors also admit
- preferred alloy) senerator operating experience is~1acking." (!).This data"bank steam certainly is inavailable to US utilities, experience as well as the long contrast, the Peach Botton and Ft.St.Vrain HTGRs. In be. both Inconel-600 and Inconel-690 ausceptible to ICSCC at 238'C in a resin intrusion water have been found to environment'(Pace & Minn, Met.Trans.AINE, 17A (May 1986) 877).
11'. I concur with the actions recommended by C. Chiu and S.-Cosselin, as namely, that the nozzles the_ failed nozzle (heat No. 54313) be replaced. made from_the same heat The replacenent 600 selectednozzles should bc Incoloy-800 alloy, or Inconel-
'IGSCC. from CE specified alloy that is resistant to f.
Q o
_ . y ;I
. \
m-si 4 n.d an./, soA .
} ""
t Y.j/ s ns j a
, ' 1 8" '
m.a. my me
. \
. N lm . ,
5 l use , g s
t .
j ,,, W" E 7
.. NA7Awr .
I Swen: Mrs
(*)
. . s,.-
uson:aA th)
Fig. ?6. .
i Comparison of reelstance to stress cracking of Alloy 600 steel in deserated caustic sode solutione at 350 . Alleyct 800, and T-316 steintees .
(b) effect of caustic soda concentration (a s:(a) effect of strees (Neon = 100 g/1);
4 I
0.8 EO.2} II*f
- III [I*g*,*f af),
i
~
8 t
P 1
I
- i a vl l
400 -
INCONEL800 a s ss s - * '
NOT - 50 CRACKED CRACKED 300 g AS RECEIVED # 0 Ig \
\
s CRACKED g -
850' C,4 h A A
~ 40
. s g 200 -
30 E
\ "
$ 0 E
% E Y \
" \
NO CRACKING -
20 100 -
\
\
\
10 G bG .
I I l l l l l l 0 10 20 50 100 700 500 1000 5001UM HYDROXIDE CONCENTRATION (9 NaOH/kg H O) 2 Tig. ;t8.
Effect of applied stress and sodium hydroxide concentration on strass corrosion cracking of Inconal 600 C-rings and grassurized capsulas at 290 to 300*C (554* to 572*T) (Ref.115) 3t.ge,4 4.
142
t 1
i 400 -
Ch && &&
g '
INCOLOY$00 q t (SANICRO.30)
NOT' -- 50
' CRACKED CRACKED 300 -
AS RECEIVEc # 0 l '
8
' 800 C 5J h 5 0 4g 850' C,4 h A l .
A
@ l CRACKING
$ 200 '
~
30 $
@ \e e G w
Ng 20 100 %
NO CRACKING %
10 G bG I I I l I I l I 10 o 20 50 100 200 500 1000 3001UM HYDROXIDE CONCENTRATION (g NaOH/kg H O) 2 Fig. 27.
Effect of applied stress and sodium hydroxide concentration on strase corrosion cracking of Incoloy 800 (Sanicro 30) C-rings and pressurized capsulas at 290* to 300*C (554' to 572*T)
(Raf. 115) 3e.rp ,4 */
e 141
m.i .
- ~~. ~
avess w a . m s n a si
. LICENSEE EVENT REPORT (LER[ amn.m anim
-_=... ...._m
~ 1AN e., ONDPRE NUCLEAR GrpreATING STATION. UNIT 3 4
015 io 10 to i 31612 1 or Oi 4-
<- PRE 150Riin eve =,save m ESTRUidhi N0ZILE PRESSURE BOUNDARY IIAK -
6s= ===... m . . , ears m M w .v s= ,asic m asinv 6vs. m - -
_E___ . . . .un a
=m
== .
SAN ON0FRE UNIT 2 _ 0 la 1010101316 1 0'2 ' fl 7 a6 '8 6""' O 013
~~
Q1 11 1 11 3 86 0 15
. 1010_,10 , , , I, 1 i ,-
3 naam m.m.., mramm. ,
--. ,n ,,,, _ ra.rt m sm. . , . aaO _
m.o. ,n ,.., _
w n,
.. , m,.., _
,m ,
JgWdgi'gyggQB
_ NEA,. ,., = ,
E.et ,U I,W, _
" ' St.784.,m.il St.73 H8H.wtHA, 4
h[ E.88,.lH He ,
..,n,,,
T St.73 H2,,U, SS.73 H3H.448H.l
...,,,H,.., ...... . ..,
....................e.,
- v. . ... ... ,
H. E. MORGAN. STATION MANAGER
_ 71 11 4
. . . . . .; = g gg.
.=.o.,
38 6 8 -1 6 21.
i
- .=. ggg n on W K Ll d490 Y l lII i 11 b l i iii iii
.. . . u .
WB
. , o .uvo n . i i i ii i i i MIN
, , , _ , c .. i .. i pas g'g;':'gJ. .g,P -
p e.o . . . . . .
A=i..a m . . ao v a ..
- " " " - i ;
. iv- .. . u u i On February 27, 1986, with !! nit 3 in Hot Standby, a small' Reactor Coolant System (R pressure nozzle. boundary leak was observed in-a 3/4 inch diameter pressurizer level instrument- i An Unusual Event was declared and cooldown to Cold Shutdown was accomplished I within the limitations of the Technical Specification using normal operating procedure It was determined by dye penetrant testing that a crack extended from the end of the
.Subsequant independent investigation by both Combustion En California Edison has determined that'the nozzle material, its installation ar.d subsequent opera' ion, contributed to pure water intergranular stress corrosion (IGSCC) attack and rew%ted in crack origination and-propagation through the wall of the nozzle.
The affected nozzle was completely cut out, including the weld, and replaced with a ne nozzle having metallurgical properties less susceptible to IGSCC. [I Therespace water are two additional nozzle on Unit vapor 2 similarly sosce no nles and one water space nozzle on Unit 3 a affected.
i The nozzles in the vapor space will be '
replaced; the norzles in the water space will be replaced pending further evaluation.
Th2 leakane i subsecuent was identified by normally monitored RCS leak rate detection paramete investigation.
All plant systems performed as designed to bring the Unit tc Cold Shutdown. -Therefore, neither the health and safety of plant personnel nor the heali:h and safety of the public was affected by this event. ;
t
.s.. *~ ,
_..,c,
-m , , . - - ,-m -+ w--*,
--_- --~
~
- E fA *"" W l
LICENSEE EVENT REPORT (LER) ==ov[oE l
TEXT CONTINUATION **'n****YEmSe
- s. _ . . . .. , . _ ,,, -
SAN ONOFRE UNIT 3 NUCLEAR GENERATING STATION, .... : c
~::..
m::;,, -
uxr ne mer. i c1s1010101311 6 2 _ 8l6 -
003 11 -
Ol On February 27, e==.1986,vaa === _uc sm uwii on i 02_
1 ar 041, with Unit 3 in Hot Standby, a Reactor Coolant System (RCS) pre ssure boundaryCode Component leak PIR)was level observed (EIIS SYSTEM CODE AB) in a 3/4 instrument nozzle. inch diame temperature of 545'F and an Rr.5 pressure of 2250 The Unit psi. was in Mo<ie 3 at an RCS The discovery was made during an inspection bein suspected vapor space leak from the Pressurizer.g performed to locate the source of a The inspection was initiated after activity in the containment atmosphereobserving a higher than expec both of which are normally monitored parameters.
This combination Pressurizer. alerted operations per,sonnel to a potential vapor e space le A contair. ment entry was made and subsequent inspection identified a pr leak appeared to be located between the nozzleTheand the P .
, in the
'i minute.
annulus area of the nozzle assembly, and was estimated at approximately ;i noted. Engineers checked for possible vibration of the instrument piping, but E none Based on these findings an Action statement was entered pursuant to Technical Specilication
,'be .in Cold Shutdown Limitingwithin Condition the next thirtyfor Operation hours. (LCO) 3.4.5.2, which o required Event was terminated.At 1250, an Unusual Event was declared and cooldown was init.
The Unit entered Cold Shutdown at 1655 on FebruaryAt 28, 1986. 1445, the Unusu OntheMarch on 6, 1986, instrument nozzle. with Unit 3 in Cold Shutdown, dye penetrant ormed tests (PT) w into the pressure boundary weld.As a result, a crack was identified in the nozzle, extending The crack was axial to the nozzle starting from the end ef the nozzle attached sketch). inside the pressurizer, extending outward approximately 5/8 inch depth.
of the nozzle Theassembly.
crack extended beyond*the weld approximately ma 1/8 inch in To ensure that the other instrument nozzles on the Pressurizer were not affected, a PT was performed on the nozzle closest to the leaking nozzle.
the PT showed no defects in the other nozzle. The results of
)ther two instrument nozzles on the head of the pin addition, this adjacent nozzle and the
)utside of the Pressurizer. ressurizer were visually inspected from tgain following depressurization, and no evidence was observed.
of leakageThis inspection w
'he affected nozzle was completely cut out including the entire weld ias installed by the NSSS vendor (Combus' tion Engineering)
The new nozzle in accordance with the endor's fabrication specifications.
arch 10, 1985. The installation of the new nozzle was completed on i
~'
.. . . . . - . . . . . . . . . - . . - - , . . . . - - . ... ~ . . . . . ~ . . - . . . _ . . . . . - . . ~ - . . . .
. . ~ -
~
9e . C P.esa 364 A
" m~ms TnEs Y" LICENSCE EVENT REPORT (LER)
TEXT CONTINUATION s m ass. a a ns
. .m .. #
- m. . i.e u,,...,,,, ,.. n,
. p *.=h . > ==n.
SAN ONOFRE NUCLEAR GENEkATING STATION, ,
UNIT 3 0 151010101 31612, 816 - -
venm m . ..- .ninace.mia w cm 01013 _011 013_ oF 014, I 1
Subsequent metallurgical examination of the affected nozzle and weld joint material in '
the area of the crack by Combustion Engineering (CE) detensined that intergranular stress corrosion cracking (135CC) of the nozzle material had occurred. Since the cracking has occurred only in the subject nozzle and not in any of the nine other nozzles fabricated CE concluded that the San Onofre Unit 3 cracking was the result involving a unique set of conditions.
Subsequent to the CI analysis and findings, SCE confirmed oy independent analysis, that the weld joint cracking was due to Pure Water IGSCC. SCE analysis, however, concluded that two other nozzles from the same heat of material, which are located in the vapor space of the Unit 3 pressurizer, could be susceptible to similar cracking. Therefore, these nozzles will be replaced during the upcoming refueling outage, presently scheduled to commence 13 January 1987.
Tt:o locations. pressurizer nozzles from the same heat of material are also located in water space In these locations, they are not subjected to the same environment as trie nozzles in the vapor space. The potential for these water space nozzles for Pure Water IGSCC is less than for the vapor space nozzles and the need for their replacement continuesiinder evaluation. If indicated by the results of this evaluation, SCE will i, replace the water space nozzles at a later date, yet to be determined.
' Any potential nozzle cracking would result in the development of very small leaks, similar to the one seen initially. Such leakage would be detected at vary low leak rates by containment airborne monitors and other reactor coolant 1cakage monitoring. Until replaced, or otherwise resolved, all suspect nozzles will be verified to not be leaking at normal operating inspected at refueling conditions outages. following each cold shutdown and they will be carefully n; ~ ~ I x.
. . _ . . ._. _m . - - - -- - ^ - ~ ' ^
ec !" pes. 480FA St /* *' .
44& sougMAA HESUL AT90tY SDhpreggpq' LICENSEE EVENT REPORT (LER) macum one == suome TEXT CONTINUATION **'* "a*
%....e 4.s
- - . i.e un ...es i e.. :::
.... o ,,,*.*.% %.,",.;. .
SAN ON0FRL NUCLEAR GENERATING STATION, UNIT 3 Turv ger mee, enees as veevees, was seesteemet NRC Perm M4A's) (17) 0_L510loIon.16R . 83 g..Jo 0 o 11 0.3_ m _Q.8 ..
UPPER HEAD I
- MAWAY l UPPER LEVEL AND PRESSURE N0ZZLE ,
SECOND
\ l f %#
N0ZILE y N0ZZLE . PT'D DETAIL LEAKING N0ZILE
, MANWAY PRESSURIZER TOP VIEW OF FOUR PRESSURIZER g LEVEL AND PRESSURE CARBON STEEL N0ZZLES j , 3 7/8" 1 HICK l
m N0ZZLE j INCONEL .
I j
( .
l 5/8" CRAM j
g g rTACTORY WEI.D
.., PRESSURE SOUNDRY .
IDCA TLOW t 0.005" CAP y INCONTL PUTTY LIMITER ANNULb5
- 1/2" DEPTH OF WELD
- AT CRACK N0ZZI.E DETAIL A
N0ZILE STAINLESS hTEEL
- - WELD CIADDINC 1/8" MINIMUM '
i TNICKNESS 4
+
l l CRACK SECTION A-A - -
j _ .
~
IC pam asaa *
!