ML20086B548

Fracture Mechanics Evaluation,Braidwood,Unit 2 Residual Heat Exchanger Tube Side Inlet & Outlet Nozzles

ML20086B548
Person / Time
Site:	Braidwood
Issue date:	11/30/1991
From:	Bamford W, Jambusaria H, Lee Y WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP.
To:
Shared Package
ML20086B542	List: ML20086B541 ML20086B548
References
MMDT-SMT-193, NUDOCS 9111200238
Download: ML20086B548 (25)

Text

.

f.

MMDT-SMT-193 FRACTURE MECHANICS EVALUATION BRAIDWOOD UNIT 2 RESIDUAL HEAT EXCHANGER TUBE SIDE INLET AND OUTLET N0ZZLES November 1991 W. H. Bamford H. Jambusaria Y. S. Lee WESTINGHOUSE ELFCTRIC CORPORATION Nuclear and Advanced Technology Division P.O. Box 2728 Pittsburgh, Pennsylvania 15230-2728 e 1991 Westinghouse Electtic Corp.

WPF0863/111191:10 g >J .

'!8? b Q

TABLE OF CONTENTS

1.0 INTRODUCTION

1.1 Code Acceptance Criteria 1.2 Geometry l

2.0 LQADING CONDITIONS. FRACTURE ANALYSIS METHODS. AND MATERIAL PROPERTIES 2.1 Transients 2.2 Stress Intensity Factor Calculations 2.3 Fracture Toughness 2.4 Thermal Aging 2.5 Allowable Flaw Size Calculation 3.0 SUBCRITICAL CRACK GROWTH 3.1 Analysis Methodology 3.2 Crack Growth Rate P.eference Curves 3.3 Residual Stresses 3.4 Stress Corrosion Cracking Susceptibility 4.0

SUMMARY

AND P.ESULTS 4.1 Flaw Evaluation Charts Construction 4.2 Conservatisms in the Flaw Evaluation

5.0 REFERENCES

WPF0863/111191:10 i

y ..

SECTION

1.0 INTRODUCTION

This fracture mechanics evaluation has been carried out to c9termine the largest size of indications which can be accepted according to the rules of Section XI, paragraph IWB 3600 for the residual heat exchanger 111et and outlet nozzles. The results of this evaluation are presented in flaw evaluation charis in Section 4, and the technical basis for construction 6f the charts is contained in the remaining sections.

1.1 Code Acceptance Crittria The evaluation procedures and acceptance criteria for indications in austenitic stainless piping are contained in paragraph NB 3640 of ASME Section XI.[1] The evi sation procedure is applicable to all the materials within a specified distance from the weld centerline, vrt, where r - the pipe nominal outside radius and t is the nominal wall thickness. For example, at the RHX nozzle, this distance is calculated to be 1.62 inches, which encompasses regions of the heat exchanger, as well as part of the RHR line.

All the materials in this region are Type 304 stainless steel.

The evaluation process begins with a flaw growth analysis, with the requirement to cons; der growth due to both fatigue and stress corrosion cracking. For pressurized water reactors only fatigue crack growth need be considered, as discussed in section 3. The methodology for the fatigue crack growth analysis is described in detail in section 3.

The calculated maximum flaw dimensions at the end of the evaluation period are then compared with the maximum allowable flaw dimensions for both normal operating conditions and emergency and faulted conditions, to determine acceptability for continued service. Provisions are made for considering flaws projected both circumferentially and axially.

In IWB 3640 the allowable flaw sizes have been defined in the tables based on maintaining specified safety margins on the loads at failure. These margins are 2.77 fcr normal and upset conditions and 1.39 for emergency and faulted i&F0863/111291:10 1

l L _ __ _

1 conditions. The calculated failure loads are different for the base metal and the flux welds, which have different fracture toughness values, as discussed in section 2. The failure loads, and consequently the allowable flaw sizes, l are larger for the base metal than for the welds. Allowable flaw sizes for welds are contained i separate tables, in lWB 3640, i

1.2 grometry L e geometry of the residual heat exchanger is shown in Figure 1-1, with the details of the inlet and outlet nozzies of the tube side shown in Figure 1-2.

The notation used for surface flaws in this work is illustrated in figure 1-3.

The fracture and fatigue crack growth evaluations carried out to develop the l handbook charts have employed the recommended procedures and material properties for stainless steel as prescribed in paragraph IWB 3640 and l Appendix C of Section XI.

Vr70L$3/111291:10 2

,4 .

• e '

s O'

4 5* - 45* >- .

b ,' MM **N h ,d ag [g a a c, fvg i O j I '

• 4 " b M G 56ESr0E 41 "O'

> .M * '*

n ' j, - -~~

, '*e[V'h 2%v ,

~$ E "** **~*

f z -w c c.t

~i 6

u& -

l tea.o. '

%ev ,g' .x . -

7 ._w _ 7 ' .". . . ~ . t w yig  ;

y .y m oces aie.x so - w _1 " ~ J see e s+aaoea ( Nc175CEO m 49 j ec ar,sg g -

a nr. ANewca sees C h ( C2 < t

• i* T. 2 %1 6 ' e ' cs:I ?

St Sw'C .wfe SEE t'hg - A' owc $$db akf[ 2 gre a set t f C wo ~~

49 44*CC e : N

-= wega gMM TL. - fWN ti ~~

j

~, os ,%v, ,s - t rCs ,l - -

a,a w..

w -

,v,'/',

w/

I

, ._ asM o o _ _I ,

1 ma s .. .!

d,

,:,.osc.! a

n. /. L, p  ;

m N I

~ ,{ 3]

_t -

$ i ( :@Tha I-i Q

. w m

• di I -- l$w,$t SIDE VI W i <

i

, v, r n< -

~ n v,

,- s*~ , 17. I

,i pV $

p M t / i -! . riiN 'h

"~'l' ( [y I Qy

'.. . ' gl .' jw' l

\y <g;

~

i .- s* r

- - n g' . , -

i

,u

p. , ~ .1,,5 so i i

- .r , 2 _ - l

\, r %p~ -- ,  ! g ,

-pJ  : .

. ~, ... , s ,8

.. . :. .. _ m oe' < o * *.a y y _e ,. -ms ,, .u a cac 5'a' ,

Figure 1-1. Geometry of the Residual Heat Exchanger wpF0863/111191:10 3

4 VESSEL WALL

-e -

PEINFORC& B T II

~

('

"' ""ll' O.40"k"" thickness ,

]k lHRU WALL = .1" WTdf i

figure 1-2. Geometry of the Tube Side Nozzles (Inlet and Outlet Nozzles are Identical)

WPF DM3/111291:10 4

t' . ,

l' '

Wall Thickness t

= r r%

l

=

h

.1-- Y

- e a l

Figure 1-3. Typical Notation for Surface Flaw Indications WPF0863/111191:10 5

.e

t SECTION 2.0 LOAD CONDITIONS, FRACTURE AMALYSIS METHODS AND MATERIAL PROPERTIES The loading conditions used in the analyses described herein were taken directly ham the equipment specification. The fracture analysis methods are the most advanced which are now available, and the material properties are the latest available properties contained in the ASME Code.

2.1 Transients and load Conditions The design transients for the residual heat exchanger are very miniinal, because this component operates only during plant shutdown conditions.

Therefore the only transient conditions which it experiences are the startup and shutdown of the system, which coincides with the shutdown and startup of the plant, respectively. The appropriate limiting load conditions for the location of interest are discussed next.

The loading conditions which were evaluated include thermal expansion (normal and upset), pressure, deadweight and seismic (OBE and SSE) loadings. The RHR piping forces and moments for each condition were obtained from the ASME Code Section 111 calculations previously performed by Sargent and Lundy and Westinghouse [2]. These loads were compared with the Equipment Specification design loadings for the heat exchanger nozzles (G-679150 Rev.1) and found to be bounded by them. Therefore the design loadings were used in this analysis.

Residual stresses were not used in this portion of the evaluation, in cou.pliance with the Code guidelines. A further discussion of residual stresses is contained in Section 3.2. The stress intensity values were calculated using the following equations:

S I - P, + P, SI = +h(Hj+H . y )ca) 2 y

2 l

WM 0863/111191:10 0

where F, = axial force component (membrane)

M,, My , H, = moment components (bending)

A = cross-section area Z = section modulus The section properties A and Z at the weld location were determined based on the minimum pipe dimensions. This is conservative since the measured wall thickness at the weld is generally larger.

The following load combinations were used.

A. Normal / Upset - Primary Stress Pressure + Deadweight + OBE B. Emergency / Faulted - Primary Stress Pressure + Deadweight + SSE C. Normal / Upset - Total Stress Pressure + Deadweight 4 OBE + Normal Thermal D. Emergency / Faulted - Total Stress Pressure + Deadweight + SSE + Normal Thermal 2.2 Stress Intensity Factor Calculatinns One of the key elements of the fatigue crack growth calculations is the determination of the driving force or stress intensity factor (Kg ). This was done using expressions avsilable from the literature. In all cases the stress intensity factor calculations utilized a representation of the actual stress profile rather than a linearization. This was necessary to provide the most WF4M3/111191:10 1 7

l

i accurate determination possible. The stress profile was represented by a cubic polynomial:

• +A +A o(x) =Ao+A 1 2 3 (2-1) where x is the coordinate distance into the wall t = wall thickness o - stress perpendicular to the plane of the crack A, - coefficients of the cubic fit for the surface flaw with length six times its depth, the stress intensity factor expression of [McGowan and Raymund [3))a,c e was used. The stress intensity factor Kg (c) can be calculated anywhere along the crack front. The point of maximum crack depth is represented by c = 0. The following expression is used for calculating Kg (c), where d is the angular location around the crack.

Kr ($) =

(cos2 $ + 2 sin $) 24 ( A Roo+ AH 1 3 (2-2)

• AH 2 2

' AH)3 3 The magnification factors Ho (c), H,(o), H 2(o) and H3 (c) are obtained by the procedure outlined in reference [3].

The stress intensity factor calculation for a semi-circular surface flaw, (aspect ratio 2:1) was carried out using the expressions developed by [Raju and Newman [4)). Their expression utilizes the same cubic representation of the stress profile and gives precisely the same result as the expression of

[McGewan and Raymund]***** for the 6:1 aspect ratio flaw, and the form of the equation is similar to that of [McGowan and Raymund]* above.

WPF0863/111191:10 8

. - - - - - - . - - - - - - _ - - . - - - - - - - - - . - --- ---._.-~.-----.-----._-----___._--.-----__----a

e .

The stress intensity factor expression used for a continuous surface flaw was that developed by (Buchalet and Bamford [5)). Again the stress profile is represented as a cubic polynomial, as shown above, and these coefficients as well as the magnification factors are combined in the expression for K, a . c. .

I, = {ia Af1 + AF 1 2

+ -

AF 2 3

+ a ' A> F, (2-3) where F , 2F ' Is ' I 4are magnification factors, available in {5].

3 2.3 Fracture Touchness The residual heat exchanger is stainless steel type 304, The weld at the nozzle was made by the shielded metal arc process, as verified by the shop traveller, and the weld procedure referenced therein.

The fracture toughness of the base metal has been found to be very high, even at operating temperatures [6], where the J i, values have been found to be well over 2000 in-lb/in . Fracture toughness values for weld materials have been 2

found to display much more scatter, with the lowest reported values significantly lower than the base metal toughness. Although the J i , values reported have been. lower, the slope of the J-R-curve is still large for these Representative values for J,, were obtained from the results of J

Ic cases.

Landes, et. al. [7], where the following values were obtained, and used in the development of the fracture evaluation methods:

[ for shielded metal arc welds: J,, = 990 in Ib/in 2 ,)..c..

2.4 THERMAL AGING Thermal aging at operating temperatures of reactor primary piping can reduce the fracture toughness of cast stainless steels and, to a lesser degree, stainless steel weldments. Because of the lower operating temperature (400*F) j of the residual heat exchanger, and the fact that the materials are type 304 stainless (not cast), thcrmal aging in this component will be negligible.

( WPFOM3/111191 10 l

l

o. ..

J 2,5 Allowable Flaw 111g_ Determination The critical flaw size is not directly calculated as part of the flaw evaluation process for stainless steels. Instead, the failure mode and critical flaw size are incorporated directly into the flaw evaluation technical basis, and therefore into the tables of " Allowable End-of-Evaluation Period Flaw Depth to Thickness Ratio," which are contained in paragraph IWB 3640.

Rapid, nonductile failure is possible for ferritic materials at low temperatures, but is not applicable to stainless steels. In stainless steel materials, the higher ductility leads to two possible modes of failure, plastic collapse or unstable ductile tearing. The second mechanism can occur when the applied J integral exceeds the J !c tractuie toughness, and some stable tearing occurs prior to failure. If this mooe of failure is dominant, the load carrying capacity is less than that predicted by the plastic collapse

~

mechanism.

The allowable flaw sizes of paragraph IWB 3640 for the high toughness base materials were determined based on the assumption that plastic collapse would be achieved and would be the dominant mode of failure. [However, due to the reduced toughness of the shielded metal arc welds, it is possible that crack extension and unstable ductile tearing could occur and be the dominant mode of failure. This consideration in effect reduces the allowable end of interval flaw sizes for flux welds relative to the iustenitic wrought type 304 vessel and piping materials, and has been incorporated directly into the evaluation tables.]*d'*

wpF0863/111191:10 10

j? ..

4 SECTION 3.0 FATIGUE CRACK GROWTH In applying Code acceptance criteria as introduced in section 1, the final flaw size a f is defined as the flaw s'.ze to which the detected flaw is calculated to grow at the end of a specified period, or until the next inspection time. This section will examine each of the calculations, and provide the methodology used as well as the assumptions.

3.1 Analysis Methodoloav The methods used in the crack growth analysis reported here are the same as those suggested by Section XI of-the ASME Code. The analysis procedure involves postulating an initial flaw at specific regions and predicting the

- growth of-that flaw due to an imposed series of loading transients. The input required for a fatigue crack growth analysis is basically the information necessary to calculate the parameter AKg which depends on crack and structure geometry and the range of applied stresses in the area where the crack exists.

Once AK y is calculated, the growth due to that particular stress cycle can be calculated by equations given in section 2.2 and figure 3-1. This increment of growth is then added to the original crack size, and the analysis proceeds to the next transient. The procedure is continued in this manner until all the transients known to occur in the period of evaluation have been analyzed.

The only transients considered in the analysis we're the startup and shutdown

of the RHR system. These transients are spread equally over the design lifetime of the vesse' .

! Crack growth calculations were carried out. for a range of flaw depths, and L three basic types. The first two were surface flaws, one with length equal to six times the depth and another with length equal twice the depth. Third was a continuous surface flaw, which represents a worst case for surface flaws.

l 1

WPF0363/111291:10 11

i 3.2 Crack Growth Rate Reference Curves i

l The reference crack growth law used for the stainless steel was taken from  !

that developed by the Metal Properties Council - Pressure Vessel Research Committee Task Force in Crack Propagation Technology. The reference curve has )

the equation:

l l

= CFS A K" (3-7) where Yd = crack growth rate, inches per cycle C -

material coefficient (C - 2.0 x 10~I9)

F = frequency coefficient for loadings (F = 2.0) 2 S -

R ratio correction coefficient (S = 1.0 - 0.502 R )-4.0 n - material property slope (=3.0321)

AK - stress intensity factor range, psi /in This equation appears in Section XI, Appendix C (1989 Addendum) for air environments and its basis is provided in reference [8), and shown in ,

figure 3-1. For water environments, an environmental factor of 2 was used, bastd on the crack growth tests in PWR environments reported by Bamford [9).

3.3 Residual Stresses Since the residual heat exchanger vessel-to-piping welds were not stress-relieved, residual stresses are clearly present. For fatigue crtck growth analyses, these stresses were included directly.

In general the distribution of residual stresses is strongly dependent on the degree of constraint of the structure. The stiffer the structure the higher the residual stresses. For a thin walled large diameter pipe the residual stresses will be lower than a small diameter thick-walled pipe. This-has been found by a number of investigators and there is general agreement that- the distribution of residual stresses is tansile near the surface, and then WPF0863/111191:10 12

t compressive near the center of the wall after which it reverses to become tensile at the outer surface.

The residual stresses were taken from work reported by General Electric-[10) and E. Rybicki [11], which included both measurements of residual stress and finite element calculations. Both approaches were found to be in agreement,

- and included a_ range of pipe sizes from 4 inches to 28 inches in diameter.

The stresses were found to peak at the weld, as shown in Figure 3-2 for a 10 inch diameter pipe, the through wall distribution of residual stresses used in this analysis was taken from the work of Rybicki, and is shown in Figure 3-

. 3. This distribution is for a 10 inch schedule 160 pipe with a thickness of 1.125 inches, which is a much stiffer configuration than the 14'ird diameter, 0.375 inch thick junction at the heat exchanger nozzle.

3.4 Stress Corrosion Crackina Susceptibility In evaluating flaws, ali mechanisms of subcritical crack growth must be evaluated to ensure that proper safety margins are maintained during service.

Stress corrosion cracking has been observed to occur in stainless steel in

- operating BWR piping systems. The discussion presented here is the technical basis for not considering this mechanism in the present analysis. The residual heat exchanger tube side nozzles are exposed to only primary coolant water.

For all Westinghouse plants, there is no history of cracking failure in the reactor coolant system loop piping. For stress corrosion cracking (SCC) to 4

occur in piping,_ the following three conditions must exist simultaneously:

high-tensile stresses, a susceptible material, and a corrosive environment.

Since some residual stresses and some degree of material susceptibility' exist

! in' any stainless steel piping, the potential for stress corrosior, is minimized by proper selection of a material immune to SCC as well .as preventing the

~

occurrence of a corrosive environment. The material specifications consider l- compatibility with the system's operating environment (both internal and f external) as well as other materials in the system, applicable ASME Code l rules, fracture toughness, welding, fabrication, and processing.

L f

WPFOS$3/111291:10 13 l -. - , .

. , , _, _.m . . - -_ _ . . . . . _ . . _ , . . . _ . . , _ . _ , - . _ - . _ _ . , _ . _ _ _ - _ _ . _ . -

. . l

_ l l

.1 l

j

e. )

.000 /

~

• Lineer interpoietion is recom.

!f 700 -

~

rnanded to account for ratio clopenoence of meer environinent

[

/

• • '* '*' O.2 5 < A < 0.65 tor 4 /

500 -

insison ioce: [k p /

$ = (1.01 X 10'110 A Ki '8

~

dN 2

& D I

~

0 2= 3.75 R + 0.06 ,

[ / )

M=Kmn )Krn, , (f g l 200 - l

/

/ \

Sutzuffe<s flows /

g 100 - fair ennronmenti i U

[ 8. . (0.0267 x 10 ) A Ki 4 3.726

) p 70 -

e .,  ;

{ - *4 j

l 50 -

Determine t%e AK si which ttw tow changes try calculation of

/

C4 43 -t 3 *&

g ~

the intersection of the two curws. I/ g l s i j Surfecu flows _/

(w.we reactor 20 - '"vironment) k

'3 appliceb6e for j O

'0.25 < <0 5 f R 3 0 65 10 -

g . K,,, /K,n,,

-l L 2 -

!u .I *

=

• 'I""'i"*'i'"'8****"*d 5 - m -

to account for A retio decenance a f $-

- of water enwronment curws,for -

- g

/ 0.25 < n < 0 e5 for swoo .> ope:

R R o

1. 8 = (1.02 X 104) Og Ag5 95

~

U-f '

,' an Ifk IIk Oj = 26.94 5.725 I R

• Km ,n IK ms, I

I 1 I I I I I I Ill l' I l- 1 IIII 1- 2 5 7 10 20 50 70 - 100 Strese intensity Factor Range (AK3 kai [.)

Figure 3-1. Reference Crack Growth Rate Curves for Stainless Steel in Air Environments (8].

WPF0863/111191:10 14

_. . . -- w . + -

- -a s a -- -, .

- # ',4 4. c a i e t~

450*

- 25 16 9

EO \

13 01 -

21 09 1 06 PtECE NO. t eitCE NC 3 (10) .

7 03 l

1 l 2

~ 1; I

[

2 'l) j \

h .

I 1

! 33 f I g

.I \\

a% -

Q l l i l ~

'21's l

\

\\

4 01 - l

-2813 , k l

(* M - pitLtt -

3% 16 i

(- 6 01 l

42,19 I f l- 7 08 1 t 1 l (C 6 ia l to 7 n t- 10 2 m l - 10 4 ia.:

~

10 9 n) to a sa l 10 d in t 1.524 cm 1016 cm 0501cm 0 600 cm - t 018 cm 1524 em 2.037 l

' Figure 3-2. Maximum Principal Surface Residual Stress for a

-10 inch Schedule 160 Pipe [10]

WPF0863/111191:10 15

-a q s_

L h

9 to' SCMEDULE 540 PIPE '

Location. O se' rnout wtLD CENTERLINE

/

t0-

/

j - AIL AL $7488 g - CIRCuuFEM NTi&L statst a

'9 /

- 4 /

I 4

/ ,

g / s-

• /

'g

/

/ LOC Af TON A

f .- g,.ers- t or stacsus l Tg T w rv .g- =Q l i

. . . s m- -*

a \

\

i \. . i i

-to -4C 0 . to 40 60 usiouat. sincss. i.4 Figure 3-3. Through Wall Distribution of Residual Stress in a 10 Inch, Schedule 160 Pipe, in a Cross Section Adjacent to the Weld Center Line [11]

WPF0663/111191:10 16

.l. ;

i SECTION 4.0 4.1 Flaw Evalgation Charts Construni j The acceptance criteria for surface flaws have been presented in Section 1.

For flaw evaluation in stainless steels, only the fatigue crack growth results must be calculated. The allowable flaw depths were determined directly from the tables in IWB 3640.

The first set'of data required for surface flaw chart construction is the final flaw size a,. As defined in IWB-3611 of ASME Code Section XI, a, is the flaw depth resulting from growth during a specific time period, which can be the next scheduled inspection of the component, or until the end of design li fetime. Therefore, the final depth, a, af ter a specific service period of time must be used as the basis for evaluation.

The final flaw size a, can be calculated by fatigue crack growth analysis, which has been performed covering a range of postulated flaw sizes, and flaw shapes. The crack growth calculational methods have been discussed in Section

3. The results of the crack growth calculation showed that growth for a complete range of crack sizes, up to 60 percent of the wall thickness was inconsequential for the entire service life of 40 years. This was expected, since the region sees so few cycles, i The allowable flaw size for stainless steel is obtained directly from tables

! in paragraph IWB 3640, so the evaluation process is very straight forward.

The allowable flaw size is calculated based on the most limiting transient for all normal operating conditions. Similarly, the allowable flaw size for emergency and faulted conditions is determined. The theory and methodology for the calculation of the allowable flaw sizes have been provided in Section 2 and Reference 12 Allowable flaw sizes were calculated for a range of flaw '

shapes.

The two basic dimensionless parameters, which can fully address the characteristics of surface flaw, have been used for the evaluation chart construction. Namely, WP*0863/11129 h10 17 j

- -r  ;

..* <c ~ e ._-

t o - Flaw Length divided by the circumference, t/c  !

o Flaw Depth parameter a/t where, t - wall thickness, in, a -- flaw depth, in, t - flaw length, in, c = pipe circumference, in.

The flaw evaluation chart for the residual heat exchanger inlet and outlet nozzles is shown in Figure 4-1. The chart has the following characteristics:

5 o The flaw length / circumference e/c was plotted as the ab: cissa from 0 l

to .S. For values of t/c which exteed 0.5, use the results for e/c L - 0.5.

e The flaw depth parameter a/t was plotted as the ordinate.

o The upper boundary curve shows the maximum acceptable flaw depth based-on flaw evaluation, beyond which no surface flaw is acceptable for continued service without repair. This upper bound curve has been determined by tne fracture and fatigue evaluations described herein, using Tables IWB 3641-5 and IWB 3641-6, for shielded metal arc-welds, o Any; surface indication which-falls-below the boundary curve will be-acceptable by the code rules, based on the analytical justification nrovided herein. However, IWB-2420 ci ASME Section XI requires future monitoring of such indications.

i A detailed example on the use of the charts for a surface flaw is presented below:

i I

l WPF0863/111291110

!. 18

i .[ .

14rf_qce Flaw Example Now suppose an indication is to be evaluated using the charts. For the circumferential orientation:

a = 0.10" ( = 6.1" t - 0.40" c 44.0" The flaw characterization parameters then become:

a/t - 0.250 t/c = 0.139 Piatting these parameters on the surface flaw evaluation chart of Figure 4-1, it is quickly seen that- the indication is acceptable.

4.2 Conservatisms in the Flaw Evaluation ,

The stress and fracture analysis results presented herein have been structured to be conservative at each step, to ensure that the final result will be conservative.

The stresses applied t, the heat exchanger nozzles were taken from the vessel equipment specification loads, which represent bounding loads for the structure. The actual loads for the Braidwood Unit 2 heat exchangers (2) are about 60 percent of the design loads. n The residual stresses used in the analysis were taken from a combination of measurements and analysis for a 10 inch schedule 160 pipe. The smaller pipe diaweter and larger thickrass (1.125 inches) for this pipe mean that the residual stress distribution used here will be very conservative relative to the heat exchanger nozzle.

Since the publication of the flaw evaluition criteria and methodology for stainlets steel fl2] a number of experiments have been carried out on large fracture t ughness specimens and full size pipes with both submerged arc welds a

WPF0P.63/111191:10 19

. . .. .. , _ - ~

and shielded metal arc welds (13). These experiments have shown t' t the fracture toughness from these larger specimens is higher than the toughness values used in the development of the flaw evaluation methods. Therefore the flaw e aluation results presented here are conservative.

i A fu' 1er conservatism is added to this fracture evaluation by using the fracture criteria for a class 1 piping system for a class 2 component. There are presently no flaw evaluation criteria for class 2 components, but presumably if they were to be developed, smaller margins could be justified, with resulting larger allowable flaw sizes.

1he indication depths from toe inspections have been compared with the thickness of the pipe, with no benefit taken of the additional thickness resulting fro.n the large fillet weld on the outside surf ace of the nozzle. As shown in Figure 1 2, this fillet weld is immediately above the indicetions, and to the actual percentage flaw penetration is smailer than that reported.

Wf 0563/111291:10 20

. . 1

.c . . l 1

, 1 l

.60 _

v

' !:! E :jijfti t ilil l:ig!!!!i li! iiii

! :r i :. i t..l .!!-

.. .d. gf .9, ,

ti iti:m!!i:?Iiip:

-. j. f: ;p1L.3..--jt 2....- di!!! [li 1

i j j,:, ,{t: t

.. uij!".. k.;t. "[. .i- .: fill zij.

o n!!i..il.1;. 4;.:f 1.:c:i,jll ..

. ., a'

u:lll

}n ,.2 i in :n

. tt s. e a

.;;.tttttr rts

.,.t: . r i

-* p j 4 q -.j :;f..-

j yjjtjj };jji

+

.g -p r 4:

4  : .

il yh. i ...

~

i.

E l-t .J. ..

t..

. u l

i.*tii.l.l[. 11.f.

t

• 0 '

t. -. .

. . .i .

F.,

g ' i i . t

l:ll.. # j[g., ji.3; h.;f{

U 2

H 2i; E[-

1..

1 r

3 L2 . . .,%

4 3i: ty..g .g 1-4 Q

.Jil !!t Ld

-t Eij l m EllE llati l}flj,Wjjli::

. ,1.

pM.tjiil

g.; i

31

its i. ,ii. ?ji:

!:: a:

3.2 l *:.

l  : b  ::1.. t  ::. t .  :.

a1 40 l

~.:

..t.

p:

...pu: -

tt ip 2, .

it tv u

--:n

.m p+llilih l':!i l.jj ;;p1!!!ill',.

t::

- - - - . .i,..if'

.g

.,v.

,Ia

!Lt MT lj']. +

{+;

4: :I lL! 4 Ii

p. .

.:h::fl t i.x y,.t.b :t a !:.

y 1

I

.30 i.t

-1:E pn.c1 3-

~

=-di: u a itfn

> 1.pt dit:11 :m :  : . .

a .

m

,f W i i o 3

• 12'

'  ::i-.nt dii y

• d~i t01 ~

< .20 . . .".U. .

\ a -2 .j..:tLin" h i

w  : f.:

7:y

{

j,1:

fiji 1 %l!!jstd .

q<t

~ . 4 fi!!

i!% tij =f:  :

Wi hi -

.10 ,

.g.g ";t+1 q;;p).7

77 .

g 3.

+ d;j

,- -U; i

A 4

N}{NjE iig!O!$  : m

.]

I I 4'

0 0.4 0.5 0.1 0.2 0.3 0

F1.AW LEFGTH / CIRCLHTERENCE 1

1 figure 4-1 Flaw Evaluation Chart for Braidwood Unit 2 Residual Heat Exchanger Tube side Nozzles WPF0863/111191:10 21

.'- o

.. p SECTION 5.0 ,

REFERENCES

1. ASME Code Section XI, " Rules for Inservice inspection of Nuclear Power Plant Components," 1983 edition (used for updated code allowable limits); 1983 edition Winter 1985 Addendum (used for flaw evaluation of austenitic stainless steel piping); 1989 edition (used for reference crack growth curve, stainless steel).

2. Jambusaria, H.,

• Residual Heat Exchangers: Braidwood Unit 2,"

Westinghouse Report No. 031804 Rev. O, Jan. 24, 1986.

3. McGowan, J. J. and Raymund, M., " Stress Intensity factor Solutions for internal Longitudinal Semi elliptic Surface F' - in a Cylinder Under Arbitrary Loading", ASTM STP 677, 1979, pp. 365 380.

4. Newman, J. C. Jr. and Raju, I. S., " Stress Intensity Factors for Internal Surface Cracks in Cylindrical Pressuro Vessels", ASME Trann ,

Journal of Pressure Vessel Technology, Vol. 102, 1980, pp. 342-346.

5. Buchalet, C. B. and Bamford, W. H., " Stress Intensity factor Solutions for Continuous Surface Flaws in Reactor Pressure Vessels", in Mechanics of Crack Growth, ASTM, STP 590, 1976, pp. 385 402.

j 6. Bamford, W. H. and Bush, A. J., " Fracture of Stainless Steel," in l Elastic Plastic Fracture, ASTM STP 668, 1979.

l 7. Landes, J. D., and Norris, D. M., " Fracture Toughness of Stainless Steel Piping Weldments," presented at ASME Pressure Vessel Conference, 1984.

l l

8. James, L'. A., and Jones, D. P., " Fatigue Crack growth Correlations for l Austenitic Stainless Steel in Air," in Predictive Capabilities in Environmentally Assisted Crtcking," ASME publication PVP 99, Dec. 1985.

WPfD863/111191 10 l 22

. ., 1 0 em +* O j v i 4

9. Bamford, W. H., " fatigue Crack Growth of Stainless Steel Piping in a Pressurized Water Reactor Environment," Trans ASME, Journal of Pressure Vessel technology, Feb. 1979.

10. " Studies on AISI Types 304, 304L, and 347 Stainless Steels for BWR l Application, April-June 1975," General Electric Report NED0 20985 1 September 1975.

11. Rybicki, E. F., McGuire, P. A., Herrick, E., and West, J., "The Effect of Pipe Wall Thickness on Residual Stresses Oue to Girth Welds," ltant

@ 1, Journal of Pressure Vessel Technology, Vol 104, August 1982.

12. " Evaluation of Flaws in Austenitic Steel Piping," Trans ASME, Journal of Pressure Vessel Technology, Voi. 108, Aug. 1986, pp. 352 366.

13. Wilkowski', G, et. al., " Analysis of Experiments on Stainless Steel Flux Welds," Battel;e Columbus labs report for USNRC, number NUREG/CR 4878.

April 1987.

l l

Wf 3S63/111191:10

,