ML13231A187

Steam Dryer Support Bracket Flaw Evaluation

ML13231A187
Person / Time
Site:	Nine Mile Point
Issue date:	08/09/2013
From:	Oberembt C, Sommerville D Constellation Energy Nuclear Group
To:	Office of Nuclear Reactor Regulation
References
Download: ML13231A187 (68)

Text

ATTACHMENT (2)

NINE MILE POINT UNIT 1 STEAM DRYER SUPPORT BRACKET FLAW EVALUATION Nine Mile Point Nuclear Station, LLC August 9, 2013

File No.: 1300596.301 Structural

) Integrito Associates, Inc."

Project No.: 1300596 CALCULATION PACKAGE Quality Program: 0 Nuclear E] Commercial PROJECT NAME:

NMP I Steam Dryer Support Bracket Flaw Evaluation CONTRACT NO.:

NMPNS Purchase Order 7722655, Rev. 3 CLIENT: PLANT:

CENG Nine Mile Point Nuclear Station, Unit I CALCULATION TITLE:

Nine Mile Point Unit 1 Steam Dryer Support Bracket Flaw Evaluation - N1 R22 Document Affected Project Manager Preparer(s) & Checker(s)

Revision Description Approval Signatures & Date Revision Pages _Signature & Date 0 1 - 46 Initial Issue A-i - A-1I Responsible Engineers:

B-i - B-i0 D. V. Sommerville D. V. Sommerville 6MAY2013 6MAY2013 C. Oberembt 6MAY2013 Responsible Verifiers:

H. L. Gustin 6MAY2013 M. Walter 6MAY2013 Page 1 of 46 F0306-O1 RI

Vjnow If OW Assf f Table of Contents 1.0 IN TRO D UCTIO N .................................................................................................... 5 2.0 OB JEC TIV E .................................................................................................................. 8 3.0 METHODOLOGY .................................................................................................. 8 4.0 A SSUMPTIO N S ..................................................................................................... 12 5.0 D ESIG N INPU TS .................................................................................................. 13 6.0 CA LCULA TION S .................................................................................................. 13 6.1 Flaw Characterization and Growth ............................................................. 13 6.2 LEFM Analysis for Flaw Stability ............................................................ 16 6.3 Limit Load Evaluation ................................................................................. 16 6.4 LEFM Analysis for VPF Fatigue Crack Growth ......................................... 17 6.5 Qualitative Assessment of Radial Indication in 1-587A ........................... 20 6.6 Functional Analysis for Faulted Load Case ................................................ 20 6.7 C omputer Files ............................................................................................ 20 7.0 CONSERVATISMS .............................................................................................. 43 8.0 CO N CLU SIO NS .................................................................................................... 44 9.0 REFEREN C ES ...................................................................................................... 45 Appendix A STEAM DRYER SUPPORT BRACKET LEFM EVALUATION ......................... A-1 Appendix B VPF FATIGUE CRACK GROWTH EVALUATION ........................................... B-1 File No.: 1300596.301 Page 2 of 46 Revision: 0 F0306-OIRI

VA=" MwWft~ ASSOO&A ft List of Tables Table 1: Comparison of RFO21 and RF022 Inspection Results ................................................. 21 Table 2: Comparison of EVT-1 Flaw Dimensions Reported in RFO21 and RF022 .................... 22 Table 3: Comparison of UT Flaw Dimensions Reported in RFO21 and RF022 ......................... 23 Table 4: EVT-1 Length Flaw Dimension Changes Reported RF022 versus RFO21 .................. 24 Table 5: UT Flaw Dimension Changes Reported RF022 versus RFO21 .................................... 24 Table 6: LEFM Results for Crack Cases Evaluated ..................................................................... 30 Table 7: Summary of Limit Load Benchmark Analysis Results .................................................. 31 Table 8: ASME B&PV Code Primary Local Membrane + Bending Stress Check, RPV Shell, Steam Dryer Support Bracket 1-587A, 1-587D .......................................................... 33 Table 9: ASME B&PV Code Primary Local Membrane + Bending Stress Check, RPV Shell, Steam Dryer Support Bracket 1-587B .......................................................................... 34 Table 10: ASME B&PV Code Primary Local Membrane + Bending Stress Check, RPV Shell, Steam Dryer Support Bracket 1-587C .......................................................................... 35 Table 11: Estimated AKI values applicable for High Cycle FCG ................................................. 39 Table 12: Summary of Project Computer Files ............................................................................ 42 File No.: 1300596.301 Page 3 of 46 Revision: 0 F0306-OIRI

Vanw NVbY,-Anodaf List of Figures Figure 1: Photographs of NMPI Steam Dryer Support Bracket Configuration .............................. 6 Figure 2: Schematic of Steam Dryer Support Bracket Configuration ............................................. 7 Figure 3: Schematic of Quarter Elliptical Comer Crack Configuration ......................................... 10 Figure 4: Boundary Correction Factors for Remote Tension (top) and Remote Bending (bottom) for a Com er C rack in a Plate ............................................................................................. 11 Figure 5: Overlay of RFO21 Inspection Data ................................................................................. 25 Figure 6a: Overlay of RF022 Inspection Data with NDE Uncertainty and 1 Cycle of SC C G row th ....................................................................................................................... 26 Figure 6b: Overlay of RF022 Inspection Data with NDE Uncertainty and 1 Cycle of SC C G row th ....................................................................................................................... 27 Figure 7a: End of Interval Flaw Configurations ............................................................................ 28 Figure 7b: End of Interval Flaw Configurations ............................................................................ 29 Figure 8: Crack Cases Considered for Limit Load Evaluation ...................................................... 30 Figure 9: Comparison of 1100539.401, Rev. 1 Crack Case EVT-1 #1 Results from RFO21 Evaluation (a), rerun in 2013 (b), and rerun using elastic material model for the saddle (c) ............................................................................................................... 31 Figure 10: Steam Dryer Support Bracket FEM Mesh (top) and Boundary Conditions (bottom) ...... 32 Figure 11: Limit Load Results for Steam Dryer Support Bracket Crack Case 1 -

Bounds 1-587A, 1-587D - I Cycle SCC Growth ......................................................... 33 Figure 12: Limit Load Results for Steam Dryer Support Bracket Crack Case 2 -

Bounds 1-587B - 1 Cycle SCC Growth ..................................................................... 34 Figure 13: Limit Load Results for Steam Dryer Support Bracket Crack Case 3 -

Bounds 1-587C - I Cycle SCC Growth ........................................................................ 35 Figure 14: Orientation of Paths to Extract Linearized Stresses in RPV Shell ............................... 36 Figure 15: High Cycle Fatigue Crack Growth (Calculated using Austenitic FCG Rates in Air) ........ 37 Figure 16: RF022 Inspection Photographs Showing Contact Locations on Top Surface of Support Brackets ......................................................................................... 38 Figure 17: 2-D Edge Cracked Finite Width Plate LEFM Solution for In-Plane and Out of Plane Shear ............................................................................................................ 40 Figure 18: 2-D LEFM Solution Perpendicular Plates Subjected to Axial Force, Bending Moment, and Uniform Membrane Stress in Semi-Infinite Wall ................................................. 41 File No.: 1300596.301 Page 4 of 46 Revision: 0 F0306-O1RI

Ca~buw" N*fOAsockat WO

1.0 INTRODUCTION

Reportable indications were identified in three of the four steam dryer support brackets at Nine Mile Point Unit 1 (NMP1) during the Spring 2011 Refueling Outage (RFO21) in-vessel visual inspections (IVVI) [1]. A supplementary ultrasonic (UT) examination of the dryer support brackets was also performed in RFO21 [2]; however, the UT procedure was not qualified at the time. Consequently the UT was a best effort performed in order to obtain volumetric data that could be used to estimate the depth of the indications identified visually. The support brackets were previously inspected in 2001 using a visual inspection technique without any reportable indications [2]. The three steam dryer support brackets with reportable indications in RF021 were 1-587A, 1-587B, and 1-587C.

Cracking in steam dryer support brackets has been previously observed in other operating Boiling Water Reactors (BWRs) as documented in the Boiling Water Reactor Vessel and Internals Project (BWRVIP)

Inspection and Evaluation Guideline written for vessel brackets and attachments (BWRVIP-48-A) [3].

BWRVIP-48-A [3] provides general guidance regarding performance of a flaw evaluation for the steam dryer support brackets. During RFO21 Constellation Energy Nuclear Group (CENG) contracted Structural Integrity Associates, Inc. (SI) to perform a flaw evaluation of the indications identified in the NMP I steam dryer support brackets [4]. The flaw evaluation demonstrated that all flawed support brackets would retain the required structural margin for at least one cycle of additional operation considering both stress corrosion cracking (SCC) and fatigue crack growth (FCG) from system cycling.

FCG caused by vibratory loading was qualitatively assessed and considered not to be applicable to this location.

CENG re-inspected the four steam dryer support brackets during the Spring 2013 refueling outage (RF022) using EVT-1 and a qualified UT tool and procedure [5, 6]. All four brackets contained reportable indications using UT. Three of four brackets contained reportable indications using EVT-1.

Further, CENG performed laser profilometry of the top surfaces of the support brackets and the mating surface of the steam dryer in order to quantify contours on each surface to determine if there was any evidence of unusual fretting or wear [24].

The UT and EVT-1 inspections performed during RF022 [5, 6] reported flaw dimensions larger than reported in RFO21 and identified new indications that were not reported in RFO21 [1, 2].

Consequently, CENG has contracted with Structural Integrity Associates, Inc. (SI) to perform a flaw evaluation of the indications identified in the NMP l steam dryer support brackets during RF022.

Figure 1 is a compilation of photographs which show the general configuration of the support brackets.

Figure 2 is a schematic showing the nominal dimensions of the steam dryer support bracket configuration. The materials and component names used in this report are contained in Figure 2 for reference.

File No.: 1300596.301 Page 5 of 46 Revision: 0 F0306-O1RI

Veombwftw hbOtAsochfWO Right Side View Lett Side View Note: The photographs used here are from 1-587C.

Figure 1: Photographs of NMP1 Steam Dryer Support Bracket Configuration.

File No.: 1300596.301 Page 6 of 46 Revision: 0 F0306-OIRI

!V8Dw"ftve bgUvAuoc~ftb

_ _ _ RPV Attachment Weld (Alloy 182) 21/2 11/4 I RPV Shell (SA-302, Gr. B) 7 Steam Dryer Support Bracket 7" (SA-240, Tp. 304) 4Saddle (SA-240, Tp. 304)

NMP1 4.6201 1 Figure 2: Schematic of Steam Dryer Support Bracket Configuration.

File No.: 1300596.301 Page 7 of 46 Revision: 0 F0306-OIRI

2.0 OBJECTIVE The objective of the analyses documented in this calculation package is to determine whether the flawed NMP 1 steam dryer support brackets can be left in-service, without repair, for one additional operating cycle.

3.0 METHODOLOGY The methodology used for the present flaw evaluation follows the same methodology used for the RFO21 flaw evaluation [4] with the following exceptions:

1. A single load case from the RFO21 evaluation is rerun to confirm that the archived supporting files produce results consistent with those published in Reference [4]. This step is performed since the 2011 evaluation was performed using ANSYS Version 11.0 [22]; whereas, the present evaluation is performed using ANSYS Version 12.1 [23] and APDL input files created for one version of ANSYS can sometimes produce errors and unexpected behavior when the commands are used in a later version of ANSYS. Once it is shown that similar results are obtained between the two ANSYS versions using the 2011 input file then the input file is considered to be acceptable for use on the current project.

2. Flaw sizes and overall support bracket condition reported from the RF022 inspections are used

[5, 6, 24].

3. Length/depth evaluation factors appropriate for the RF022 inspection techniques are used.

These are addressed in Section 6.1 of this report.

4. Stress Corrosion Crack (SCC) growth in the depth direction of bracket 1-587B is determined using the BWRVIP- 14-A, K independent crack growth rate of 2.2x10 5 in/hr [7] rather than the value of 5x 10- 5 for both length and depth directions as was done in Reference [4]. The bounding crack growth rate between Alloy 600 [21 ] and stainless steel [7] is used for brackets 1-5 87A, 1-587C, 1-587D.

a. Considering the fabrication details for the lug and the location of the indications, crack growth is expected to remain in the stainless steel material. This item is addressed further in Section 6.1.

5. The load application location has been moved from the locations considered in the American Society of Mechanical Engineers (ASME) Boiler and Pressure Vessel (B&PV) Code,Section III, design stress analysis [13], to a location consistent with the wear patterns apparent on the top surface of the support brackets. The change in load application location is conservative with respect to the previous analysis [4] since the location maximizes the torsional moment applied to the cracked section.

a. For the flaw stability calculations (Linear Elastic Fracture Mechanics (LEFM) and Limit Load) the forces applied to the top of the bracket are applied at the outer most comer of the bracket and the forces applied to the side of the bracket, in the design stress report

[13], remain applied in the same location as considered in the design stress report.

File No.: 1300596.301 Page 8 of 46 Revision: 0 F0306-OIRI

b. For the vibration fatigue evaluation, discussed below, only loads on the top of the bracket (radial and tangential to vessel) are considered. Omission of side loads is supported by the inspection data showing no evidence of normal operating contact and wear on the sides of the brackets.

c. To implement the change in load application point, the load is applied in the current analysis using point forces applied on a single node rather than the distributed nodal loads used in the previous evaluation. This method of applying load to the structure results in very large local stresses that do not affect the results at the location of interest but will result in large plastic strains local to the point of application. To eliminate solution convergence issues at this location, the previous model was changed, for the present analysis, to use an elastic material model for the entire saddle. The region of the bracket containing the crack plane continues to have an elastic perfectly plastic material model defined. A single case from Reference [4] is rerun to validate that using an elastic material model for the saddle will not introduce significant changes to the results reported in the evaluation performed in 2011 [4].

6. An additional LEFM crack model is used for this evaluation. The quarter-elliptical edge crack model given by Raju and Newman [8] is used in this evaluation in order to more accurately evaluate the comer cracks. The work documented in Reference [4] conservatively treated all cracks as edge cracks. In the present analysis this approach continues to be used in some cases and is augmented by the comer crack solution in some cases. Although the expected failure mechanism for both stainless steel and Alloy 600 materials is net section collapse, because of the potential for significant constraint at the crack plane, it is possible that large hydrostatic stresses will develop (large stress triaxiality) which could retard plastic flow. A hydrostatic stress is the average normal stress acting in the three orthogonal directions at a point in a material [25, pg.

258]. The stress state at a point in a material can be defined as the sum of the hydrostatic and deviatoric stresses. It can be seen that presence of hydrostatic stresses can apparently raise the yield stress since the hydrostatic component prevents shear strain. The deviatoric component of the stress state is responsible for shear deformation; thus, in the presence of large hydrostatic stresses the yield point can appear to be increased. This condition could result in failure by fracture prior to net section collapse [25, pg. 275]. LEFM was selected in Reference [4] since it is conservative and quicker to implement than an elastic-plastic fracture mechanics (EPFM) solution.

a. The quarter elliptical comer crack solution for a finite body subjected to remote tension and bending loads [8] takes the form of:

K, =(S, + HCSb)F )r (1)

Where: St is the remote membrane stress, psi Sb is the remote bending stress, psi File No.: 1300596.301 Page 9 of 46 Revision: 0 F0306-OIRI

AU Ms Ha they I MW H, is the bending boundary correction factor, dimensionless Fc is the tension boundary correction factor, dimensionless a is the crack depth, in Q is the shape factor for an ellipse, dimensionless Note that Q = 1+ 1.464(a1 , for a/c < 1 (2)

Figure 3 illustrates the crack configuration and body dimensions considered for this solution.

The boundary correction factors can be determined through evaluation of curve fit expressions developed from the finite element analysis results obtained by Raju and Newman or through the use of tables and figures provided in Reference [8]. For this analysis the figures provided in Reference [8] are used. Figure 4 is an excerpt from Reference [8] containing the appropriate boundary correction factor figures for the quarter elliptical crack case and flaw aspect ratio for which this solution is used.

Note that the limits of applicability for this solution are based on the available finite element solutions from which the curve fit expressions for the boundary correction factors are obtained. The published limits of applicability are:

0.2 < a/c < 2, a/t < 1.0, c/b < 0.5 Raju and Newman state [8]:

For all configurationsfor which ratios ofcrack depth to plate thickness do not exceed 0.8, the equations are generally within 5% of the finite element results, except where the crackfront intersects afree surface. Here the proposed equationsgive higher stress intensityfactors than the finite element results, but these higher values probably represent the limiting behavior as the mesh is refinednear the free surface. For ratiosgreaterthan 0.8, no solutions are availablefor directioncomparison; however, the equationsappear reasonableon the basis of engineeringestimates.

h - b -1

.t. T a . . t T I-. c -

(c)Corner crock Figure 3: Schematic of Quarter Elliptical Corner Crack Configuration 181.

File No.: 1300596.301 Page 10 of 46 Revision: 0 F0306-OIRI

Vanacfw hbFUrAsOcWa ft 2

I 0

(a] a/c = 0.2. (b) a/c - 0.5.

FcHc (a)a/c = 0.2. (b) a/c = 0.5.

Figure 4: Boundary Correction Factors for Remote Tension (top) and Remote Bending (bottom) for a Corner Crack in a Plate 181.

File No.: 1300596.301 Page 11 of 46 Revision: 0 F0306-OI RI

V ftgrtjrAss oCb e .*

7. The possibility of high cycle FCG caused by the recirculation pump vane passing frequency (VPF) vibration is evaluated. This is added to the present evaluation because of current industry efforts to understand the possible effect that recirculation pump VPF vibration may have on the steam dryer caused by vibratory loading transferring from the reactor pressure vessel (RPV) to the steam dryer across the steam dryer support brackets.

a. An upper bound of the vibration loading acting on the dryer support bracket is taken to be the value of friction force that can be created between the steam dryer and the support bracket. Once the excitation at the RPV causes a force greater than the available friction force then relative motion would occur between the two components and the force transferred across this joint would be limited to the friction force.

b. A conservative value of friction coefficient is defined for this evaluation.

c. An evaluation is performed to show that if FCG at the VPF was occurring then the crack growth predicted to occur during a single operating cycle would be significantly larger than the height of the bracket leading to failure of the bracket. Since this has not been observed to occur during the operating interval between the 2011 and 2013 inspections then it can be inferred that high cycle fatigue crack growth is not a relevant crack growth mechanism for this location at NMP 1. This evaluation and the observation of no apparent evidence of fatigue crack growth in the 2013 inspection data supports the qualitative evaluation performed in 2011 that concluded vibration induced FCG was not applicable.

8. An evaluation interval of 1 operating cycle is used.

A detailed description of all other aspects of the analysis methodology is provided in Reference [4].

4.0 ASSUMPTIONS The same assumptions as used in the Reference [4] flaw evaluation are used for the present flaw evaluation with the following exceptions:

1. Geometry Assumption 5 of Reference [4] is no longer applicable since the loads on the top surface of the bracket are applied to the outermost comer node rather than eccentric by 0.25 inches. The change in saddle dimensions in order to achieve an eccentric condition is no longer necessary.

2. A dynamic coefficient of friction of 0.5 is used to predict a postulated upper bound vibratory loading on the steam dryer support bracket. The value of 0.5 is supported by the discussion included in Section 6.4.

3. NMP 1 will operate with an average reactor coolant conductivity < 0.15 PS/cm over the evaluation interval considered in this evaluation; therefore, the K independent normal water chemistry (NWC) SCC crack growth rate (CGR) of 2.2x10-5 in/hr [7] can be used for cracking in stainless steel materials.

File No.: 1300596.301 Page 12 of 46 Revision: 0 F0306-OIRI

Vftww Anodeft WO ahftffy 5.0 DESIGN INPUTS The same design inputs as used in the Reference [4] flaw evaluation are used in the present flaw evaluation except:

1. RF022 inspection data are used [5, 6, 24],

2. NDE evaluation factors applicable to the inspection techniques used in RF022 are used [9, 10],

3. NMP 1 has 5 recirculation pumps, each with 5 vanes, operating at a rated speed of between 720-1141 revolutions per minutes (RPM) [ 11].

6.0 CALCULATIONS The calculations performed for the following aspects of the evaluation are documented in this section:

1. Flaw characterization and growth

2. LEFM evaluation for flaw stability

3. Limit Load analysis

4. LEFM evaluation for vibration FCG

5. Qualitative evaluation of radially oriented indication in 1-587A

6. Functional evaluation of support brackets for faulted service condition 6.1 Flaw Characterization and Growth Both UT and EVT- 1 inspections were performed for all four steam dryer support brackets. Table 1 provides a summary of the RFO21 and RF022 inspection results, for each method and bracket, in order to enable an understanding of observed differences between inspections. Figure 5 is an overlay of the UT and EVT-1 inspection data from RFO21 (2011) for brackets 1-587A, 1-587B, and 1-587C. No visual indications were reported in 1-587D during the RFO21 inspections and UT was not performed on 1-587D during RFO21. Figure 6 is an overlay of the RF022 inspection data for brackets 1-587A, 1-587B, 1-587C, and 1-587D. Although not shown on Figure 6 the UT data sheets report that the location of the flaw tips in all four brackets are well into the stainless steel side of the Inconel weld; the minimum distance between the weld fusion line and the crack tips varies between 0.58 inches and 0.88 inches [6].

There was a reported change in flaw dimensions or newly reported indications in several of the brackets; however, it should also be noted that some indications had shorter reported lengths in RF022 and many of the indications exhibited reported lengths comparable to those reported in RFO21. Considering the newly reported indications and the apparent changes in dimensions of some indications, additional growth during the next operating interval cannot be ruled out.

Tables 2 through 5 tabulate the reported indications from RFO21 and RF022 and calculate the change in reported dimension as well as the average reported change in crack dimension for all indications. These values are also compared with the I cycle predicted crack growth using the BWRVIP- 14-A [7] upper File No.: 1300596.301 Page 13 of 46 Revision: 0 F0306-01RI

Cj UUcW tf bftV* Assocbt. kas~

bound NWC CGR. The average change in flaw dimension is less than the crack growth that is predicted for each indication using the bounding crack growth rates in Reference [7].

The initial flaw sizes shown in Figure 6 are increased to account for NDE uncertainty as well as crack growth from SCC. It is shown below that FCG from system cycling is negligible over the next operating cycle. It is also shown, based on the limited information available, and the conservative evaluation performed in Section 6.4, that FCG from vibration is expected to be negligible over the next operating cycle. The flaws in brackets 1-587A, 1-587C, and 1-587D are grown for one fuel cycle using the bounding Alloy 600 and stainless steel CGR. The flaws in bracket 1-587B are grown for one fuel cycle using the bounding stainless steel CGR.

From BWRVIP-03 [9, Section 3.1], the appropriate EVT- 1 length evaluation factor, for measurement by ruler, is 0.2 inches. Each visually detected flaw length is increased by 0.2 inches per crack tip, prior to calculation of SCC growth. Further, the depth of the visual indications that were not detected by UT was assumed to be the minimum detection limit of the UT method [6] equal to 0.2 inches; therefore, all visual indications that were not detected by UT are assumed to have a depth of 0.2 inches prior to crack growth.

A UT depth sizing factor was provided by the NDE vendor for the tool and procedure qualification performed with the Electric Power Research Institute (EPRI) [10]. The specific numbers are not reported here since the qualification resulted in both length and depth uncertainties for both mechanical and electronic scan directions and since the NDE vendor reports the correct evaluation factor with each reported dimension on the NDE data sheets [6].

SCC growth is considered using the bounding, K-independent, NWC crack growth for the stainless steel base material reported in BWRVIP-14-A [7], and for the Alloy 600 is reported in BWRVIP-59-A [21].

The SCC CGRs are:

• Length direction: 5.0 x 10- 5 in/hr (stainless steel and Alloy 600)

" Depth direction: 2.2 x 10-5 in/hr (stainless steel)

" Depth direction: 5.0 x 10-5 in/hr (Alloy 600)

The total SCC growth added to each flaw tip, for the evaluation interval, is:

Length: 1 Cycle: Acscc = 365.25.2.24.5.0xl 0-5 = 0.877 in/tip Depth: 1 Cycle: Aascc = 365.25.2.24.2.2x10- 5 =0.386 in/tip (SS)

Aascc = 365.25.2.24.5.0x10- 5 = 0.877 in/tip (Alloy 600)

File No.: 1300596.301 Page 14 of 46 Revision: 0 F0306-OI RI

System cycling FCG is calculated assuming:

• 6 Startup/shutdown cycles (3 cycles per year)

• 5 Seismic cycles (1 event with 5 cycles assumed for the event)

Note that CENG communicated for the 2011 flaw evaluation [4] that care is taken during installation and removal of the dryer to prevent significant impact loading between the dryer and support bracket; therefore, no installation removal loads are considered in the evaluation. The entire load (both deadweight and Seismic) considered in the original stress report is conservatively assumed to act for both the startup/shutdown and seismic events; thus, 11 cycles are considered. The KIEQ calculated in the LEFM section below is treated as the range of stress intensity factor (AK1 ) used for the FCG calculation.

This assumption considers that the loads cycle from a no load condition (i.e. dryer removed) to the full load (dryer installed and seismic event) which is a reasonable assumption for the startup/shutdown load and conservative for the seismic load since for the seismic event the DW loads would act as mean loads.

For this calculation, the effect of residual stresses is to increase the R-ratio but has no effect on the AK; therefore, an R-ratio near 1.0 will be conservatively assumed in order to maximize the effect of mean stresses.

Considering a AKI = 27 ksi-in°5 [4] (This value is consistent with the value predicted in the present evaluation of 28.3, see Table 6), the FCG rate for Alloy 600 at 600 'F, with a R-ratio of 0.9, from Figure C-8410-2 of ASME XI, Appendix C, 2010 Edition [12] is da/dn = -1.5E-4 in/cycle. No FCG curves for austenitic stainless steel in water are currently available in the ASME Code; however, noting the similarity between the air FCG curves for Austenitic stainless steel and Alloy 600, and that the air and water curves for the Alloy 600 converge for AK > 20 ksi-in°5 , the air FCG rate from Figure C-8410-1 of ASME XI, Appendix C, 2004 Ed. [12] for 550 'F and a R-ratio of 0.9 is used. Consequently, the FCG rate for the austenitic stainless steel base material, da/dn = -1.5E-4 in/cycle. An R-ratio of 0.9 is considered acceptable and conservative since the applied stresses, given in the original stress report [ 13],

are on the order of 16 ksi and the yield stress of the material at operating temperature is on the order of 18 ksi (the weld residual stress is conservatively assumed to be at yield); thus, since the applied stress intensity factor is proportional to stress, the R-ratio can be estimated as 16/(18+16)=0.47. The crack growth rates described above are well supported by the data published in GEAP-24098 [14] prepared by General Electric under contract to the Nuclear Regulatory Commission (NRC).

Finally, the calculated FCG for a one cycle evaluation interval, conservatively calculated assuming the edge cracked LEFM solution described above and that all SCC crack growth has occurred first (results in a larger AK1 ), is given as:

Aa = 11.1.5x10-4 = 0.0017 inches This value would be added to each dimension of each flaw. Considering that this value is almost negligible and that the assumptions for crack size, LEFM solution, and loading are extremely conservative, this value is considered to be an upper bound. Consequently, system cycling FCG is considered to be insignificant for the evaluation interval. Note that the AKI value used in the argument above was from the 2011 evaluation; this value remains representative of the range of AKI predicted File No.: 1300596.301 Page 15 of 46 Revision: 0 F0306-OIRI

Coo"wa MWO AsoC&A WO using the 2013 inspection data and presented in Table 6; therefore, the above argument remains applicable for the present evaluation.

The end of evaluation interval flaw sizes considered for this flaw evaluation are shown in Figures 6 and

7. Figure 8 shows the three crack cases evaluated for the limit load analysis.

6.2 LEFM Analysis for Flaw Stability Appendix A contains the MathCAD file listing used to perform the LEFM evaluation. The analysis performed considering the 2013 inspection data is similar to the analysis performed considering the 2011 inspection data with the addition of the comer crack configuration discussed in Section 3.0 above. The following two cases are chosen to bound all cracks in all four support brackets:

1. A 2.734 inch deep edge crack located across the short axis of the lug.

2. A 1.637 x 5.574 inch (a x c) comer crack superposed with a 1.240 inch deep edge crack. The comer crack is on the long axis of the lug and the edge crack is on the short axis of the lug. The effect of the two superposed cracks is treated by adding the K, values calculated for each crack separately.

Table 6 summarizes the calculated KIEQ for each crack case and reports the allowable fracture toughness.

All brackets are shown to possess sufficient structural margin for one cycle of operation as-is.

6.3 Limit Load Evaluation As discussed in Section 3.0, the point of applied loading is changed in the present analysis to be at the furthest location away from the centerline of the bracket. This maximizes the possible bending moment and torsional moment applied to the crack plane.

Two benchmark cases are performed. The first case is performed since the version of ANSYS used in 2011 was ANSYS Release 11.0 [22]; whereas, the current version of ANSYS installed on the SI workstations is Release 12.1 [23]. Since APDL files prepared for one version of ANSYS can generate errors or unexpected modeling behavior when used in a different version of ANSYS, this benchmark case is performed to confirm that the input file is functioning properly in ANSYS Version 12.1 [23].

The second case is performed to confirm that the change in material model used for the saddle introduced no unexpected effects on the analysis results.

Figure 9 compares the Von Mises stress on the crack plane for the original results documented in Reference [4] and for the two benchmark cases performed for the present evaluation. The acceptance criterion used for the benchmark is that the general stress distribution and the maximum displacements and stresses differ by no more than 5%. Table 7 summarizes the Von Mises stress and the maximum displacement for each case evaluated. The Von Mises stress and the maximum displacements for all cases agree to within 1.3% and 0.7%, respectively; therefore, the input file from the 2011 evaluation can be used and the model change made to accommodate a point load applied in the subsequent limit load analysis does not affect the results of the analysis.

The three crack cases shown in Figure 8 are evaluated using the same limit load methodology used in Reference [4]. Figure 10 shows the finite element model (FEM) mesh used for this analysis. All crack cases are run using a 0.1 inch mesh size on the crack plane. Figure 10 also shows the boundary File No.: 1300596.301 Page 16 of 46 Revision: 0 F0306-OIRI

Can" tnMwh ANDO&As f conditions applied to the FEM for all crack cases. The displacement and pressure boundary conditions are identical to those used in the Reference [4] evaluation. The lateral loads applied to the side of the bracket are the same as applied in the Reference [4] analysis. The only change made for the present evaluation is to apply the loads defimed in the original stress report on the top of the bracket at the extreme edge of the top of the bracket. This placement is a conservative decision to apply the loads using the maximum eccentricity possible, in recognition of the contact patches evident in the inspection photographs showing wear patterns on the outer edges of the support brackets. These wear patterns can be observed in Figure 16. The lateral and vertical loads on the bracket are defined such that they are additive; in other words, care is taken to make sure that the moments created by the vertical and lateral forces act in the same direction, in order to evaluate the bounding condition.

The acceptance criteria used for this analysis are [4]:

1. The ANSYS solution converges (meaning that the applied load can be supported by the net section and collapse does not occur),

2. The total strain does not exceed the minimum specified elongation at rupture for the bracket material,

3. The primary stress limits in the adjacent RPV shell are satisfied.

Figures 11, 12, and 13 present contour plots of the Von Mises stress, Hydrostatic stress, and Von Mises Strain on the crack plane, for each crack case considered. Tables 8, 9, and 10 present the maximum Primary Membrane plus Bending stress intensity in the RPV shell. Three paths were reviewed for each crack case and the membrane plus bending stress intensity reported in this calculation package is the largest of the three. The paths evaluated were located at the top of the bracket, the bottom of the bracket, and at the location where the stress intensity on the inside surface of the vessel, at the toe of the attachment weld, was the largest. Figure 14 shows the path locations for each crack case.

Review of Figures 11 through 13 reveals that all three crack cases produced converged ANSYS solutions demonstrating that collapse did not occur. The elastic core remaining in the net section can be observed in each figure. Further, the maximum Von Mises strain for all three cases is on the order of 5% which is significantly less than that the reported 40% elongation at rupture for this material [4].

Finally, the results presented in Tables 8, 9, and 10 reveal that the primary local membrane plus bending stress intensity in the RPV shell is less than the allowable limit for all crack cases. Thus, for the limit load analysis, all steam dryer support brackets are acceptable, as-is, for one additional cycle of operation.

6.4 LEFM Analysis for VPF Fatigue Crack Growth The normal operating range of recirculation pump speeds for NMP I is 720-1141 RPM [11]. Each pump has 5 vanes [11 ]. Thus, the normal operating recirculation pump VPF is:

720 RPM 1min VPFL,,w = - .--. 5 vanes = 60 Hz 1 60sec 1141RPM lmin VPFHIgh =- 5 vanes = 95 Hz 1 60 sec File No.: 1300596.301 Page 17 of 46 Revision: 0 F0306-OIRI

VSM" MW AssooAfta f# WO Conservatively assuming an 80% capacity factor and the lower value for normal operating VPF, the approximate number of fatigue cycles that can accumulate during one operating cycle is:

Cycles= VPFL,,w . 3 6 5 .2 5 days . 2 4 hr . 3 6 0 0 sec . 2 yr .0.80= 3.03x10 9 cycles yr day hr Operatingcycle To illustrate the order of magnitude of FCG expected from a high cycle fatigue mechanism, such as VPF induced vibration, a sample calculation is performed conservatively using the FCG rates for Austenitic stainless steel in air from the 2004 Ed. of the ASME B&PV Code,Section XI, Appendix C [12]. A AK from 1 to 5 ksi-in° 5 is used. Existence of a threshold stress intensity factor for FCG is ignored for this illustrative calculation. Figure 15 shows the cumulative FCG calculated for a given AK. For a AK as low as 1 ksi-in0 .5 the predicted FCG would be on the order of 8 inches, which is equal to the height of the bracket. Even higher AK values would yield crack growth on the order of tens or hundreds of inches.

The critical observation from this example is that if FCG was occurring at the VPF then the bracket would be expected to fail within one operating cycle.

Inspection photographs were reviewed by CENG staff in order to determine if there was evidence of side loading on the brackets that could be indicative of vibration loading. No evidence of lateral wear or scouring was observed. Consequently, VPF vibration loading is assumed to exist from the deadweight loading of the steam dryer acting on the top of the brackets. The load will be introduced into the bracket by the friction force between the dryer and the support bracket. This force can act in the radial and tangential direction, depending on the direction of the relative acceleration between the steam dryer and the support bracket.

Coefficients of friction are not readily available in literature for stainless steel materials (SS) in water.

Reference [16, pg. 12] states that GE-Hitachi has experimentally determined the coefficient of friction for 304 SS sliding on 304 SS with deoxygenated water as a lubricant to be close to 0.5. Further, the dry (non-lubricated) coefficient of friction is reported to be about 0.42 for hard steel sliding on hard steel, and about 0.57 for mild steel sliding on mild steel [17, pg. 3-23]. It should be noted that vibration can lower the coefficient of friction, as shown in the results of testing of motor operated valves provided in Reference [ 18]. It would be difficult to quantify the amount of reduction in the dynamic coefficient of friction; however, this information provides further support that coefficients developed without vibration would likely bound the actual coefficients when vibration is considered. Thus, a dynamic coefficient of friction equal to 0.5 is selected for this evaluation; however, it is expected that the effect of vibration would possibly reduce the coefficient of friction further. For the purposes of this evaluation the higher the coefficient of friction the larger the postulated vibration loading; therefore, a coefficient of friction of 0.5 is considered to be conservative for this analysis.

Recognizing that a vibratory load magnitude is not known, the range of stress intensity factor contributed by VPF vibration is estimated by calculating edge crack and estimating comer crack AKI values for the range of observed crack sizes in 2011 and 2013. These values are reported in Table 11.

Further, the end of interval flaw sizes from the 2011 evaluation [4] and the present evaluation are also used to estimate AKI values, also reported in Table 11. Since the flaw depths reported in 2011 and 2013 File No.: 1300596.301 Page 18 of 46 Revision: 0 F0306-OIRI

V w"tAsMfat WO exhibit little difference, the conservatively calculated AKI values are very similar. When estimating the AK, for a comer crack it is seen that the AKI is less than the AKTH for FCG presented in Reference [9, Figure 9.3]. These results are in agreement with the observation of no apparent high cycle FCG during the previous operating cycle and the observation from Figure 15 that if the AK1 was larger than the threshold value then significant FCG would be expected.

A similar estimate of AKi is made for the end of interval sizes for the 2011 and 2013 inspection data.

Using the edge crack configuration FCG would have been expected for the 2011 indications but was not observed from the 2013 inspection data. Using the comer crack configuration no FCG would be expected for the 2011 indications. Using the comer crack configuration for the 2013 indications the AK1 values are generally within the range of the AKTH for FCG. Considering the conservatisms used in this analysis it is expected that the actual AKI is less than the value calculated here; however, it is acknowledged that accurate vibration amplitude information is not currently available.

Considering a reduction factor in the AK1 based on the fact that the vibration loads used in this analysis are expected to be approximately a factor of -2 high (based on a conservative friction coefficient) and the end of interval flaw sizes are expected to be a factor of -2 high based on the comparison of the average observed crack dimension changes and the predicted crack growth using the bounding CGR, shown in Tables 2 through 5, then the end of interval AK1 values for the 2011 and 2013 data are both less than the range of AKTH for FCG given in Reference [9] of 2.5-4.5 ksi-in° 5 (for the expected R ratio range of 0.5-1); these results are shown in Table 11. The reduction factor is estimated from the fact that K, is proportional to load and proportional to crack size raised the V2 power; thus, the reduction factor is approximately 2(2)0-5 = 23/2 = -3. The estimated AKI considering the expected end of interval SCC crack growth (based on comparison between 2011 and 2013 inspection data rather than bounding CGR) and a lower coefficient of friction considered to be more representative of two vibrating surfaces is seen to be less than the threshold value for FCG in stainless steel [9]; this is shown in Table 11 as the "best estimate" value.

Appendix B contains the MathCAD file listing for the VPF fatigue crack growth calculations.

Finally, the results of the laser profilometry performed on all four steam dryer support brackets showed no evidence of wear that would be indicative of sustained vibratory loading [24].

The results of this evaluation show that if high cycle FCG from VPF vibration were occurring in the NMP 1 steam dryer support brackets then very large crack growth would have been expected to occur subsequent to the 2011 inspection. This was not observed during the 2013 inspection. Further, using the information available, the expected AKI for the 2011 and 2013 inspection data were calculated and shown to be similar; thus, the FCG behavior in the subsequent operating interval is expected to be similar to that during the previous interval. It is important to note that detailed information regarding vibration loading is not currently available; therefore, this results contained in this section can be considered to be an estimate only.

File No.: 1300596.301 Page 19 of 46 Revision: 0 F0306-01RI

6.5 Qualitative Assessment of Radial Indication in 1-587A During the EVT-1 inspections of 1-587A performed in RF022 a portion of the radially oriented indication reported during RFO21 was reported to be on the face of the bracket which is oriented radially inward. The two radial cracks appear to be the same crack, simply wrapped around the edge of the bracket. The previous evaluation performed in Reference [4] remains applicable for the present evaluation; therefore, please refer to Section 7.6 of Reference [4].

6.6 Functional Analysis for Faulted Load Case The same evaluation performed in Reference [4] remains applicable for the present evaluation; therefore, please refer to Section 7.5 of Reference [4].

6.7 Computer Files Table 12 lists all computer files used for this analysis. All computer files are filed in the project files.

File No.: 1300596.301 Page 20 of 46 Revision: 0 F0306-OIRI

ICaw" M~p# Anssifft Table 1: Comparison of RFO21 and RF022 Inspection Results.

Steam Dryer RFO21 (2011) RF022 (2013)

Support EVT-1 UT EVT-1 UT Bracket ______________________ ___________ ___________

B 1 indication on 0 1 indication on

• I new indication on 1 indication along bottom face of lug bottom of lug bottom of lug right bottom face of lug face

• 1 indication along 0 1 indication on bottom of lug along lower left e 1 indication along

• 1 indication along face of lug bottom face of lug left lower left face of 1-587A lug a 1 new indication on center bottom of lug front face a 1 indication of bottom face of lug o 1 indication along bottom of lug

• 1 indication on . 1 indication on 0 1 new indication on o 1 indication on right side of lug top right side of lug top lower side of left right side of top face face face face

• 1 indication along

• 1 new indication on 9 1 new indication on upper portion of lug lower side of right lower side of right right face face face a 1 indication across 0 1 indication along 0 1 new indication 1-587B entire bottom lug bottom face along bottom face face 1 indication on right side of top face

• 1 indication on upper side of right face 0 1 indication on 1 indication across a 1 indication on

• 1 indication on right side of lug top lug top face right side of top right side of top face face face f 1 indication on 1-587C

• 1 indication along upper portion of lug 9 1 indication on

• 1 indication on upper portion of lug right face upper side of right upper side of right right face face face

• No reported

• Not inspected

• No reported 1 indication on left 1-587D indications indications side of bottom face File No.: 1300596.301 Page 21 of 46 Revision: 0 F0306-OIRI

Table 2: Comparison of EVT-1 Flaw Dimensions Reported in RFO21 and RF022.

1-587A 1 0.875 0.875 1-587A 2 NR 0.991 1-587A 3 2.081 2.044 RFO21 length limited because end 1-587A 4 1+ 1.657 of flaw could not be seen 1-587A 5 NR 0.362 1-587B 1 0.355 0.694 1-587B 2 3.229 3.229 1-587B 3 2.5 2.5 1-587B 4 NR 1.70 1-587B 5 NR 1.69 1-587C 1 1.371 1.364 1-587C 2 3.25 3.782 1-587D NR NR Notes:

1. Dimensions are in inches.

2. NR = Not reported.

File No.: 1300596.301 Page 22 of 46 Revision: 0 F0306-OIRI

V8nwhc &*I*brWAs eW Table 3: Comparison of UT Flaw Dimensions Reported in RFO21 and RF022.

Length 1.9 NR Visual indication was reported here in 1-587A I RF022 Depth 0.4 NR 1.4 inches of length in RFO21 was Length 2.5 2.04 from uninspectable regions on either 1-587A 2 side of bottom face.

Depth 0.4 0.6 0.75 inches of length in RFO21 was Length 1.35 2.01 from uninspectable region on side of 1-587B 1 top face Depth 0.28 0.26 This location was not inspectable by Length NR 1.74 UT in RFO2 1; however, no EVT- 1 1-587B 2 indication was reported here either.

Depth NR 0.24 Length NR 2.5 Visual lcto indication nRO was reported at this 1-587B 3location in RF21 Depth NR 0.46 1.5 inches of length was from Length 2.5 0.65 uninspectable regions on both sides of 1-587C 1 top face in RFO21 Depth 0.25 0.54 0.7 inches of length was from Length 3.5 4.5 uninspectable region on top side of 1-587C 2 face in RFO21 Depth 0.2 0.25 Length NI 1.13 1-587D 1 Depth NI 0.7 Notes:

1. Dimensions are in inches.

2. NR = Not reported.

3. NI = Not Inspected File No.: 1300596.301 Page 23 of 46 Revision: 0 F0306-OIRI

Table 4: EVT-1 Length Flaw Dimension Changes Reported RF022 versus RFO21.

1-587A 1 0 1-587A 2 0.991 1-587A 3 -0.037 1-587A 4 0.657 1-587A 5 0.362 1-587B 1 0.339 1-587B 2 0 1-587B 3 0 1-587B 4 1.70 1-587B 5 1.69 1-587C 1 -0.007 1-587C 2 0.532 Mean 1 Cycle bounding crack 0.877 in.

growth: L0.519 in.

Table 5: UT Flaw Dimension Changes Reported RF022 versus RFO21.

1-587A 1 -1.9 1-587A 1 -0.4 1-587A 2 -0.46 1-587A 2 0.2 1-587B 1 0.66 1-587B 1 -0.02 1-587B 2 1.74 1-587B 2 0.24 1-587B 3 2.5 1-587B 3 0.46 1-587C 1 -1.85 1-587C 1 0.29 1-587C 2 1 1-587C 2 0.05 Mean 0.241 in. Mean 0.117 in.

1 Cycle bounding crack 0.877 in. 1 Cycle bounding SS crack 0.386 in.

growth: growth:

1 Cycle bounding Alloy 600 0.877 in.

crack growth:

File No.: 1300596.301 Page 24 of 46 Revision: 0 F0306-OIRI

Cam"I~ MsoW M AONA W-0D.

0.70 TYP-t-0.80 TYP UT Uninspectable regions UT Indication EVT-1 Indication Notes: )Dameae to frontcorner ofsupport bracket show in INF-11-19R3 [1c). Thisis not on the same plane asthe indtcatonsnearthe mttachment weld-Flw l detected by EVT-1onbottom front surface of bracket This Is noton the sarneplane asthe Indications nearthe aftthrment weld

) Thisindication was not deteCted b*UT.

()This indication was not detected byLT.

O The end of this indication was not discernaible by EVT-1The indication appeared to run underneath the saddle.

O This indication was not detected bytUTbecause no UTscans were performed which could interrogate this region.

O The length of this indication is defined byn sumdeng the length ofthe both branches reported in INF-11-15R2 [Ia].

Figure 5: Overlay of RFO21 Inspection Data.

File No.: 1300596.301 Page 25 of 46 Revision: 0 F0306-OIRI

V3~dano MbWfyAssoC&A~ ftcQ

.200Th Z9 II

.87 .877 (2.814) .00 1.937 1

'1700 '

Ast-1-587A 1-587B Notes:

1. Bounding value of both stainless steel and Alloy 600, K-independent, Normal Water Chemistry SCC crack growth rate used for 1-587A.

2. Stainless steel, K-independent, Normal Water Chemistry SCC crack growth rate used for 1-587B.

Figure 6a: Overlay of RF022 Inspection Data with NDE Uncertainty and 1 Cycle of SCC Growth.

File No.: 1300596.301 Page 26 of 46 Revision: 0 F0306-OIRI

Vj~IsvwaehMW f#Assocaf W-0 r-B, 4.

4.

3.

I4 5.

460 TK-1-587C 1-587D Note:

1. Bounding value of both stainless steel and Alloy 600, K-independent, Normal Water Chemistry SCC crack growth rate used for 1-587C and 1-587D.

Figure 6b: Overlay of RF022 Inspection Data with NDE Uncertainty and 1 Cycle of SCC Growth.

File No.: 1300596.301 Page 27 of 46 Revision: 0 F0306-OIRI

Caw"MIfNASSOdefta W-'

1-587A 1-587B Figure 7a: End of Interval Flaw Configurations.

File No.: 1300596.301 Page 28 of 46 Revision: 0 F0306-OIRI

Cm,"d kfty Asochf hic?

1-587C 1-587D Figure 7b: End of Interval Flaw Configurations.

File No.: 1300596.301 Page 29 of 46 Revision: 0 F0306-OIRI

C3OMMtM No f# AssOchM *-'

Case I Case 2 Case 3 1-587A, 1-587D 1-587B 1-587C Note: The element size used for the ANSYS analysis is 0.10 inches which results in a nodal dimensional accuracy of +/- 0.05 inches when defining the crack sizes in the FEM.

Figure 8: Crack Cases Considered for Limit Load Evaluation.

Table 6: LEFM Results for Crack Cases Evaluated.

05 Acceptable Crack Case KIEQ, ksiin~ KI, Alowable, ksi-in0 .

Crack Case___________ n (Y/N) 2.734 in. edge crack 28.3 Y (located across short axis of lug) 1.637 x 5.574 in. (a x c) comer crack combined with 4.83 + 19.24 = 55.6 1.240 in. edge crack 24.07 Y (comer crack is on side, edge crack is across top) 24.07 File No.: 1300596.301 Page 30 of 46 Revision: 0 F0306-OIRI

Vn8cfw" NO#*AVwsocaft hn lIt*zH 1Iu.fl

-r 4w, 3

S .fl4.~II1

h. A 0?

Io 11964 2100 321"4 4116 124.417 4752 4144 1411.5 14446 21274 17107 12542 17112 42411 a)Von Mises stress from 1100539.401, Rev. 1 [4] b)Von Mises stress rerun in 2013 using input deck run on ANSYS VI 1.0 from 1100539.401 archives [4] on ANSYS V12.1.

I 4)Vo s M e 142i4 23I2 7aera m ef WI 714 744 2714 *1*C1~lw VATI.214 c) Von Mises stress using elastic material model for saddle.

Figure 9: Comparison of 1100539.401, Rev. 1 [41 Crack Case EVT-1 #1 Results from RF021 Evaluation (a), rerun in 2013 (b), and rerun using elastic material model for the saddle (c).

Table 7: Summary of Limit Load Benchmark Analysis Results.

Max. Von Mises CaeDisplacement,  % Max.% Cag Case Stress, psi Change in.spacmet 1100539.401 Rev 1, EVT-l#1 41261 - 0.0745 -

(ANSYS VI 1.0)

EVT-I#IRe-run2013 41803 1.3 0.0746 0.13 (ANSYS V12.1)

EVT-1 #1 Re-run 2013 with Elastic Material Model for 41799 1.3 0.0740 -0.67 Saddle I I II Note: Percent change is calculated relative to the baseline case which is taken as 1100539.401, Rev. 1 case EVT-1 #1.

File No.: 1300596.301 Page 31 of 46 Revision: 0 F0306-OIRI

Vew" OW f# ANDO&A *-*

ELEMENTS AN TYPE NUM U

F pRES-NORM

-15000 -13333 -11667 -83~3

-14167 -12500 -10 -9167 -7500 NMP1 STEAM DRYER SUPPORT BRACKET FLAN EVALUATION Figure 10: Steam Dryer Support Bracket FEM Mesh (top) and Boundary Conditions (bottom).

File No.: 1300596.301 Page 32 of 46 Revision: 0 F0306-OIRI

VON(" No fNAssocu W-0 T-z~

AN a)Right side view of the FEM showing Von Mises b)Von Mises stress on crack plane, looking toward stress. Displacement scalin2 = 15X. RPV surface.

c)Hydrostatic pressure on crack plane, looking d)Von Mises Strain on crack plane, looking toward toward RPV surface. RPV surface.

Figure 11: Limit Load Results for Steam Dryer Support Bracket Crack Case 1 - Bounds 1-587A, 1-587D - 1 Cycle SCC Growth.

Table 8: ASME B&PV Code Primary Local Membrane + Bending Stress Check, RPV Shell, Steam Dryer Support Bracket 1-587A, 1-587D.

File No.: 1300596.301 Page 33 of 46 Revision: 0 F0306-OIRI

COO W NY ANbgr&A W-dmt'*

z4:3 fif 32M 4221*

a)Right side view of the FEM showing Von Mises b)Von Mises stress on crack plane, looking toward stress. Displacement scaling = 15X. RPV surface.

c)Hydrostatic pressure on crack plane, looking d)Von Mises Strain on crack plane, looking toward toward RPV surface. RPV surface.

Figure 12: Limit Load Results for Steam Dryer Support Bracket Crack Case 2 - Bounds 1-587B -

1 Cycle SCC Growth.

Table 9: ASME B&PV Code Primary Local Membrane + Bending Stress Check, RPV Shell, Steam Dryer Support Bracket 1-587B.

File No.: 1300596.301 Page 34 of 46 Revision: 0 F0306-OIR1

C3oneiruI~ IbAwvAsocM W-0~

300-2 S0 ! Ie-41 3

-M -100C. *s121G

• e1

- 355 2

• 2146 RPVsu sufae A" It 2CI cl: 1 7:5 ISPI STSXMn -6,0 43*6 tVAWTIMS 50PI?4ACXGC a)Right side view of the FEM showing Von Mises stress. Displacement scaling = 15X. RPV surface.

II :15 30 API 3 201 175 -2 7T6-I SU.S 3z*001 £71052 (A2 Do5 -. 320902

• 450504 01564l U2355 22324

.0240 1_'

• 140 d M0514k ostatic pressure on crack plane, looking d)Von Mises Strain on crack nlane. lookine toward toward RPV surface. RPV surface.

Figure 13: Limit Load Results for Steam Dryer Support Bracket Crack Case 3 - Bounds 1-587C -

1 Cycle SCC Growth.

Table 10: ASME B&PV Code Primary Local Membrane + Bending Stress Check, RPV Shell, Steam Dryer Support Bracket 1-587C.

File No.: 1300596.301 Page 35 of 46 Revision: 0 F0306-OIRI

Vshbft"WWA Amocef W ANS~YS 12. 1 AN' AV* 32 2013 SW -!

0 1: 7: 0 3

-Ix 5~5St 195X:.3269792 8W .114.0 1 SM A4,525 P'ATH11.1 U5 3172i m 14

- 41525 I-1 40.108 10557 21293 31919 &2546 9.353 19680 26606 35233 47899 NW51 MM93 095!) 3UPPORT BRRLI!? !VAUMT10H4 MLAW WWI1 91141 MYER13 SUP"T 55ACKE1312 96.3 EVAS 10N Case 1 Case 2 (1-587A 1-587D) (1-58713)

AN Styr (AM5 alp - 274%1 1 67.053 12638 25208 31119 50349 635' 19922 31493 44064 56635 92W4 13TEAM99R52'S SRACTWI2.'T FLAWE*VALtAT1OS Case 3 (1-587C)

Figure 14: Orientation of Paths to Extract Linearized Stresses in RPV Shell.

File No.: 1300596.301 Page 36 of 46 Revision: 0 F0306-0IRI

~~Cam"W* hbOl AssocifW-1000.0 A

At 100.0 10.0 1.0 0 1 2 3 5 AM, "i* o

- 1 Cycle VPF FCG (500 F) - 1 Cycle VPF FCG (550 F)

Figure 15: High Cycle Fatigue Crack Growth (Calculated using Austenitic FCG Rates in Air).

File No.: 1300596.301 Page 37 of 46 Revision: 0 F0306-OIRI

Va~bnoyO VAssodaf br' I 1-587C I 1-587D I Figure 16: RF022 Inspection Photographs Showing Contact Locations on Top Surface of Support Brackets.

File No.: 1300596.301 Page 38 of 46 Revision: 0 F0306-OI RI

V&OW3I hbgI# Anadocf ft Table 11: Estimated AK1 values applicable for High Cycle FCG.

Dimension, Edge Crack, Corner Crack, Condition ear in ksi-in°- ksi-in°'0 2011 0.2-0.4 3-6 0.3-0.6 Reported 2013 0.2-0.7 4-9 0.4-0.9 2013 1.0-1.5 5-17 0.5-1.7 End of Interval 2015 1.0-3.0 5-50 0.5-5.0 Best Estimate - 2013 0.5-0.75 1.7-6.0 0.2-0.6 End of Interval 2015 0.5-1.5 1.7-17 0.2-1.7 Notes:

1. AKITH for austenitic stainless steel is in the range of 2.5 - 4.5 for a R ratio of 1 - 0.5.

2. "Best Estimate" AKK values are estimated by taking a factor of improvement defined by the assumed conservatism in the load of 2 and growth rate by 2. The K, is directly proportional to load and proportional to the square root of crack size; therefore, a AK1 reduction factor of 23/2 = 2.8 = -3. This is purely an approximation. Insufficient information exists currently to quantify expected vibration loading.

File No.: 1300596.301 Page 39 of 46 Revision: 0 F0306-OIRI

Vj~owbr Assohf W-0' fNYMp~

Figure 17: 2-D Edge Cracked Ln frlae LLR ane and Out of Plane Shear [151.

File No.: 1300596.301 Page 40 of 46 Revision: 0 F0306-O1RI

Comm& ~N ANgdlyAssce k Figure 18: 2-D LEFM Solution Perpendicular Plates Subjected to Axial Force, Bending Moment, and Uniform Membrane Stress in Semi-Infinite Wall [151.

File No.: 1300596.301 Page 41 of 46 Revision: 0 F0306-OIRI

Vsow ANk*grty A .

Table 12: Summary of Project Computer Files Filename Program Purpose 29APR2013 - LEFM of MthCAD Contains LEFM calculations for Bracket.xmcd abracket flaw stability 29APR2013 - LEFM of Bracket- Contains LEFM calculations used VPF.xmcd MathCAD for high cycle fatigue crack dgrowth calculations FCG - FIV.xls MS Excel Calculations for Figure 15 Flaw Tabulation.xls MS Excel Crack summary and flaw growth calculations Crack Sketch - 1 cycle (A600 and Autodesk Inventor Flaw overlay sketches SS).idw AutoeskInvntor Flaovrlayskeche Crack Sketch 587B (1 cycle Autodesk Inventor Flaw overlay sketches SS).idw AutoeskInvntor Flaovrlayskeche Crack Cases.idw Autodesk Inventor Crack Lodalyicase sketches used for Limit Load analysis Input file for benchmark cases NMP 1_SBBracket-i1 .inp ANSYS documented in Table 7 Input file for crack cases NMP1_SBBracket-2.inp ANSYS documented in Section 6.3 File No.: 1300596.301 Page 42 of 46 Revision: 0 F0306-OIRI

VOWWW~0 AnOoMtM&a*

7.0 CONSERVATISMS Although the methodology utilized for this flaw evaluation is considered to be consistent with the guidance of ASME XI, IWB-3600 [12], it is acknowledged that the component evaluated in this analysis does not clearly fall under the existing flaw evaluation rules given in ASME XI, Appendix C. Further, there is currently lack of detailed data regarding the amplitude of the possible recirculation pump VPF vibration at the steam dryer support brackets. Consequently, this section is included to clearly identify some of the conservatisms inherent in the methodology used for this evaluation.

1. The Level A/B RIPD is neglected, which is conservative since this load acts opposite to deadweight (DW). This assumption increases the load considered in the evaluation.

2. A friction coefficient of 1.0 is assumed for the differential thermal expansion loads, consistent with the original stress analysis, such that substantial axial loading is introduced into the bracket.

This assumption increases the load considered in the evaluation.

3. A friction coefficient of 0.5 is assumed for the vibration fatigue crack growth evaluation despite the fact that the actual coefficient of friction is expected to be lower than this value. This assumption increases the vibration loading considered in the evaluation.

4. The bounding, K-independent, NWC, IGSCC crack growth rate is applied for crack growth.

This CGR was shown to predict an average crack growth larger than observed for the steam dryer support bracket indications between 2011 and 2013 by approximately a factor of 2 to 3.

5. A linear elastic fracture mechanics analysis is performed to assess the likelihood of unstable crack propagation in the bracket as opposed to a more complex but more appropriate and less conservative elastic-plastic fracture mechanics analysis. The fracture mechanics calculation is performed in addition to limit load because of the high constraint and resulting stress triaxiality which might retard plastic flow sufficiently that failure would occur by fracture prior to collapse.

6. A bounding Level A structural factor for membrane loads is considered for all loading (membrane, bending, shear) and for the limiting load combination which considers Level B loads.

7. All loading on the top surface of the bracket is assumed to occur on the farthest comer of the bracket rather than distributed across the observed contact patches. This assumption increases the applied torsional moment on the cracks.

8. Plastic collapse is evaluated without considering contact between the crack faces which, if considered, would react additional load for cracks subjected to compression and would provide a frictional reaction load for cracks subjected to torsion or direct shear.

9. Fatigue crack growth, from system cycles, is assessed by considering all load cycles to consist of both a full DW and Seismic contribution despite the fact that for a seismic event the deadweight would act as a mean load rather than a direct contributor to the AK.

10. An R-ratio of 0.9 is used for the system cycles FCG assessment despite the fact that the R ratio is shown to be closer to 0.5.

File No.: 1300596.301 Page 43 of 46 Revision: 0 F0306-OIRI

VgAffAnoft. -

8.0 CONCLUSION

S The results of the flaw evaluation documented in this report support the following conclusions:

1. The indications reported in the NMP 1 steam dryer support brackets 1-587A, 1-587B, 1-587C, and 1-587D are acceptable, as-is, for one additional cycle of operation.

2. Using the limited available information, high cycle fatigue crack growth caused by vibratory loading at the recirculation pump VPF is not expected to occur during the next operating cycle,

3. SCC growth is expected to be the dominant crack growth mechanism over the next operating cycle.

Considering the lack of detailed information regarding the amplitude of the possible recirculation pump VPF vibration and the appearance of active SCC growth in the steam dryer support brackets, SI recommends that:

1. CENG re-examine all four steam dryer support brackets using both visual and UT methods during the next refueling outage to confirm that the behavior of the flaws remains bounded by the flaw growth evaluation documented in this report or to acquire the necessary inspection data to re-evaluate the flaws, as necessary.

2. CENG should include the contact locations on each dryer support bracket in the inspection scope to determine if SCC initiation has occurred at or near the contact location caused by the fretting or large surface residual stresses that could result from the contact or wear at this location.

2. CENG have a contingency repair option available during the next refueling outage.

3. CENG attempt to more accurately characterize the amplitude and direction of any VPF induced vibratory loading on the steam dryer support brackets in order to remove uncertainty in the high cycle fatigue crack growth evaluation performed in this calculation package.

4. CENG consider refining the present analysis using the results of an effort to more accurately quantify the high cycle fatigue loading and to simulate the effect of the crack shape, crack interaction, and expected residual stresses. This effort could include:

a. Presence of residual stresses and their effect on retarding SCC growth through the thickness,

b. Contact along the crack faces preventing crack faces in compression from passing through each other.

c. Use of FE LEFM to more accurately simulate interaction between the cracks and growth to SCC based on crack front K distributions rather than single point K estimates at the surface and deepest point along the crack front.

File No.: 1300596.301 Page 44 of 46 Revision: 0 F0306-OIRI

9.0 REFERENCES

1. Westinghouse Indication Notification Forms, SI File No. 1100539.204:

a. NMP1-RFO-21-INF-11-15R2, April 2,2011

b. NMP1-RFO-21-INF-11-16, March 30,2011

c. NMP1-RFO-21-INF-11-19R3, April 4, 2011

2. Design Input Request, Rev. 3, SI File No. 1100539.201.

3. BWRVIP-48-A: BWR Vessel and Internals Project, Vessel ID Attachment Weld Inspection and Flaw Evaluation Guidelines, EPRI, Palo Alto, CA: 2004. 1009948.

4. SI Report 1100539.401, Rev. 1, Nine Mile Point Unit 1 Steam Dryer Support Bracket Flaw Evaluation.

5. Westinghouse Indication Notification Forms, SI File No. 1300596.202:

a. NMP1-RFO-22-INF-13-05, Rev. 2, April 26, 2013

b. NMP1-RFO-22-INF-13-06, Rev. 0, April 26, 2013

c. NMP1-RFO-22-INF-13-07, Rev. 3, April 26, 2013

6. Design Input Request, Rev. 1, SI File No. 1300596.200, UT Data.

7. BWRVIP-14-A: BWR Vessel and Internals Project, Evaluation of Crack Growth in BWR Stainless Steel RPV Internals, EPRI Report 1016569, September 2008.

8. Newman, J. C., Raju, I.S., "Stress Intensity Factor Equations for Cracks in Three Dimensional Finite Bodies Subjected to Tension and Bending Loads," NASA Technical Memorandum 85793, April 1984.

9. TR-105696-R1 1 (BWRVIP-03) Revision 11: BWR Vessel and Internals Project, Reactor Pressure Vessel and Internals Examination Guidelines. EPRI, Palo A lot, CA: 2008. 1016584.

10. EPRI Letter 2013-051, "New NDE Demonstration for UT of Steam Dryer Support Bracket/Lug."

11. Design Input Request, Rev. 1, SI File No. 1300596.200, VPF input.

12. American Society of Mechanical Engineers Boiler and Pressure Vessel Code:

a.Section XI, 2004 Ed., No Addenda.

b.Section XI, 2010 Ed., No Addenda.

13. Combustion Engineering Report No. CENC-1 142, NMP Calc. No. SOVESSELM026, SI File No.

1100539.206.

14. Hale, D.A., Yuen, J., Gerber, T., "Fatigue Crack Growth in Piping and RPV Steels in Simulated BWR Water Environment," GEAP-24098, January 1978, General Electric.

File No.: 1300596.301 Page 45 of 46 Revision: 0 F0306-OIRI I

wft fbtgd'lW Asso90 tso.°

15. Tada, Hiroshi, Paris, Paul C., Irwin, George, R., The Stress Analysis of Cracks Handbook, 3 rd Ed.,

ASME Press, 2000.

16. United States Nuclear Regulatory Commission, "Safety Evaluation by the Office of Nuclear Reactor Regulation Core Plate Hold Down Bolt Inspection Plan and Analysis Entergy Nuclear Operations, Inc. Vermont Yankee Nuclear Power Station Docket No. 50-271," ADAMS Accession No. ML120760152, SI File No. 1101291.211.

17. Mark's Standard Handbook for Mechanical Engineers, 1 t1hEdition.

18. Letter from P. Swafford of TVA to V. McCree of NRC, "Request for Regulatory Conference or Public Management Meeting," June 2, 2011, ADAMS Accession No. ML111930423.

19. Barsom, J. M., Rolfe, S. T., Fracture and Fatigue Control in Structures, 3 rd Ed., ASTM, C. 1999.

20. Design Input Request, Rev. 1, SI File No. 1300596.200, UT minimum detection depth.

21. BWRVIP-59-A: BWR Vessel and Internals Project, Evaluation of Crack Growth in BWR Nickel Base Austenitic Alloys in RPV Internals. EPRI, Palo Alto, CA: 2007. 1014874.

22. ANSYS Mechanical and PrepPost, Release 11.0 (w/Service Pack 1), ANSYS, Inc., August 2007.

23. ANSYS Mechanical APDL and PrepPost, Release 12.1 x64, ANSYS, Inc., November 2009.

24. Design Input Request, Rev. 1, SI File No. 1300596.200, Laser Profilometry Input.

25. Dowling, N. E., Mechanical Behavior of Materials, 2 nd. Ed., Prentice Hall, 1999.

File No.: 1300596.301 Page 46 of 46 Revision: 0 F0306-OIRI

V8Uu"r~hb AmrsocbkstW Appendix A STEAM DRYER SUPPORT BRACKET LEFM EVALUATION File No.: 1300596.301 Page A-1 of A-1I Revision: 0 F0306-01 RI

c~~owdfsoc OWN O -

+/-

46 For the coiguration shown above the stress intensity factor at the crack plane is determined by superposition of the foMowing LEFM solutions found in The Stress Analysis of Cracks:

1. Edge cracked finite width plate subjected to in plane and out of plane shear stresses (pg 73)

2. Seri-ininite 90 degree plate intersection with edge crack at intersection subjected to axial force, moment, and membrane stress (pg. 307).

From pg. 73:

KUM T -,'-a-f FUI 2 3 1-122 -- 0-561.(~ +005 o

-~s..) +090 -s.(~

FI-=

a b

KMMl Tf1 -a/~-FMf FMI= 2b ýF 7a File No.: 1300596.301 Page A-2 of A-1I Revision: 0 F0306-O1RI

IC- ( - Fla ao A) + "-F*p(A) + - -FIR (A)J I)[ r-a .-vPA) + + '-1~p(A) a A(a,W) -

w Flu(A)= -A _-[.ois + 0.069-e - A)]

-A -[0156-0 .96-c (1A]

=p(A) 3 -[0379 + 0.624-A - 0.062-e Gl-A)]

4j-(- A)2 FUp(A) - ' [0.126 - 024-A - 0.023-(l - A),]

A) 2 9-

-005 - 0.72-e 9 A F1(A) =

4A0-_( A)2 FU(A) 1 3 [-O0 + (1 - A)4.(0 0.2A + o.8A 2)]

3 2

4o- A)

File No.: 1300596.301 Page A-3 of A-I l Revision: 0 F0306-01RI

• OCf kn,*

For Hoop orentation (i-e n:urierentia orientation in RPV) i* 0,1_3 (1.077'"

1.337 .

Range of crack sizes to evaluate.

1.637 W =2.5 in b W (0.431~

A..

a.I I0.535I iw 0.655 k..o.:)

T '_

16250 2-5-9 T = 812.5 (106250-,- 1z.75-2.5)-(1.

S(2525 3)T -1580 12 1-122 561{2 + 0.0J5- + o.180- . -

FII..

I, I

b b1.833)

{1803.6" 2133A4I KII-I - 2639A4

,373312) 1.152 I 7ra 2-b Fif - I 1.27

,1-565j File No.: 1300596.301 Page A-4 of A-11 Revision: 0 F0306-OIRI

COjNWfRWRYANeDhurA bmc'*

(1162" 113701 116711

,.2275.,

213.44 1000---

2 (Pressure stress) a-= 14978 psi 7.125 20000 lb P (Force per unit thickness) P = 2500 m m

M =5000-2 + 16250-U2 + 15M (Moment per unit thickness) M = 8789 in 0.021 Flai.- .001[ + 0.069-e 0.017 I

Fla= 0.013 A.i

,9x 10-A..o A.1-A Ao 1 -[0.156 -

9.( Aý.

0.067-er"0

[0.113 0-179')

0-145 1

+ 0.624-A. - 0.062-e (TAJI, [2.298

,0.078),

3.072

[0-075 Fpip I FIp= 4-799

,10-978)

-0.012 FU-i = 1 1- 1 - 0.24-iAi - 0023-(1 - AJ] ni~p= -0.19 FA-l- A,)2 ,,-0.25j File No.: 1300596.301 Page A-5 of A-1I Revision: 0 F0306-OIRI

7.111 8-641 Flmi D 1 [2.005 -0.72.e9( 1:I)

S12117 j ~2 Aj.( ,25-063)

(-0-571" 0.2A. + 0.8(A)21 Fm -0.42I M+(I- A.)4.-577 -

3 -01 321

-ýA1 2 28 23191 KI1 = Tý-j- (C1aHa + P li+ M -FluiJ

{,31714 I psi-,Fm 50286

,116198) 3600 S2016I p m KH 2. - T7ý-aj F11a-1 + -Fup, + -Fllmrý KI_2 = psi--

- 1, w W2

,-9120.)

An equivalt* may be gven by:

u := 03 0 23853 (2 (Ku + Kq 2)2 + I- I u -(KI2]

.]' 322 psi I~e%-+ [(Ký) Kleq =

50360

,116355, File No.: 1300596.301 Page A-6 of A-11 Revision: 0 F0306-OIRI

!3~ufVWW hb*IW AssOC&Ae bn*

L JV.RR 118M~

Kleq

/

Kil 60 00 7 Km X3 54000 220DO

• . a--r-. -- ~ -

1 1.1 1.2 13 1.4 15 1.6 1.7 1.8 1i9 2 a

Using a solution for a quarter eiptical crack in a finite body subjected to remote tension and bending loads (Raju and Newman):

a = 1.637-rn c = 5.574-in a C 0.697 a

- = 0.294 -

S0-655 c

8-rn 2.5-in Crudely iterpolatig between the alc=0.2 and aWc=0.5 graphs in Figure 7 of NASA Technical Memorandum 85793 gves a maximum Fc at the deepest point. This value is approximately 1.75. At the surface Fc wotd be approximately 1 From Figure 8. using the same crude interpolation gives a Fclc=-,-O.4 at the surface and -1 at the deepest point.

Qý= 1 + 1.464_(I i I-65 Q = 1-194 File No.: 1300596.301 Page A-7 of A-Il Revision: 0 F0306-OIRI

CSIM" OWfO AnOWAssoibr In 6-M 6-M b 3, Remote bending stress is given as: Sb--psi Sb_ Sb = 2109--

4-2-52 in .2 El b-t p b Remote tension stress is given as: St St= 125---

2-b-t 2-5-S . 2 2 M

K .sudfe=(St-125 + 0.4-Sb)- a 02 me a . m*. b,. 0-5 Kdeep :- (St-1.75 + 1-Sb)- F--- Kdeep - 492 20.5 SQ .2 El Note that the corner crack case gives a 10 that is -10 smaller than the edge crack (5 versus 50). This solution does not consider torsional loaing or shear and may be at the edges of the applicability of c/b; however, it gives an estimate of the benefit of more accurately treating the comer crack configuration.

For Axial orientation (i.e. longitudinal axis of RPV)

(1.077"*

I1.240 in Range of crack sizes evaluated 1-971

,2.734)

W- 8 in b :=W 0.135" a 0155 A := w A 02

,0-342, T==

12750 2-5-S Tr = 937.5 S(16250 1S.75-2*5).(1 + 42)

TI = 4640

+ 2-5-4) 12 File No.: 1300596.301 Page A-8 of A-11 Revision: 0 F0306-OIRI

CON~W9fw lakftyAssC&A bic

ý0-342) 18750 2.5-8 T 937.5 2

A (16250-1 75-2-5)-(125 .4 42)0 TI 4640 12 FU-1.122 - o0,-5s6 bj + 0 Jo-. +

1.127" 1.129 1.142 Fb (1944 YX1 26651

ý.3209.)

t,1053, FM- 2-b '

1.01I 1-026 Kila..Th if 1000-* 213-44 2

(Pressure stress) cr- 749 psi 2-7.125 20000 (Force per unit thickness) Ib P25 2.5 P =8000 in I - in M 15000-5.5 + 18750-225 + 5000-1 oment per unit thickness) M - 51875 2.5 in 0.071 0.058 Fla. l .A0018+0 .069-e5(~jJ Fla 0-034 o,0o25, "0-353" 0-334 Film 1A-1i[ 006-e(T--j01 01M6

,0116)

File No.: 1300596.301 Page A-9 of A-11 Revision: 0 F0306-O1RI

Vao M ft M ochft b

( A, 1.535" FIp1 0379 + 0.624-A. - 0 .06 2 -e I-KAi]i 1-.5331 FHp 1.637

-A)2 1-9971 S-0.18 01279" Flpip 1 - 0.24-A 0023-(1 - Aj.] 0.258I A.). 9-

{0.132)

"6 .1 97" Flbi - 1 2.005 - 0.72-e (Ti50] /6-o105 6.057

-A)2 64)

Filmn' 1 L-[.OM + (I - A.)4[0 -577 - 01A2K + 0-.S(AJI) F~m - 0.197

-c0.136

{

13022' KYi a-aFla'i + " -FIpi+ M TFmi 13654 psi-4FM 16915 21314)ý r5824"¶ K_ 2. 5{ FIO +P + MMFpip Fbmi -57561 151741 1.4243J An equWatent KI may be given by:

v .= 03 18317" t,283 l.

2 + (KClli + KEL2i)2

+

KMe [= 19240 I psi.4FM K~q=23411 File No.: 1300596.301 Page A-1O of A-11 Revision: 0 F0306-OIRI

Ca~bgr"WMIAssocif kx*

Kleq KMi CI1090W 4400

-2000-

1. 1.4 1.5 1.7 1.9 2.1 23 2.4 2.6 2-8 a

The static facture tougness for unirradiated stainless steel is taken from BWRVIP-76 as 150 ksi--"0.5. The required structural factor for Level A/B conditions is taken as 2.7. Therefore.

the allowable kacture tougness for stainless steel is given as:

150 KIallowable := ksimi -

KIUalowable = 55.6 2.7 Considering a faw oriented on either axis of the support bracket the applied stress intensi factor is less than the allowable fracture toughness.

File No.: 1300596.301 Page A- 1I of A-l Revision: 0 F0306-O1RI

Appendix B VPF FATIGUE CRACK GROWTH EVALUATION File No.: 1300596.301 Page B-1 of B-10 Revision: 0 F0306-OIRI

CjwntwIM" gdty Msoelafts hL 9

For the configuration shown above the stress intensity factor at the crack plane is detelrmned by superposition of the fowing LEFM solutions found in The Stress Analysis of Cracks:

1. Edge cracked finite width plate subjected to in plane and out of plane shear stresses (pg. 73) 2- Semi-infinite 90 degree plate intersection with edge crack at intersection subjected to axial force, moment, and membrane stress (pg 307).

From pg. 73:

KUM= T-.4/-FU 1.2- 0-561.(+/- + 0-095-(+/-) +010a F11 =

F--Ib KMMl T,-,fr--aFMI File No.: 1300596.301 Page B-2 of B-10 Revision: 0 F0306-O1R1

Ia =m4", - .Fk(A) + -FPTA) + Ž F lm(A)J K~In rwa(a-FIlc7(A) + P-FUp(A) + M-F*

W W

A(aW) - -

w (A) I- -A 10.019 + -0.069e FA

.-A Flip(A) L 1t0.156 - 0.067- e-A 9.9- ( A)

FIp(A) ./ 1= 10A 7[9 + 0.624-A - 0.062-e- ý --

F~p

)3 2 6 - 0.4 -A ".2,(l-A 3

X-(A_- A) 2 FlM(A) -[-0.O22 + (I - A)4-(0.77 - OlA + ORA)]

3 qXA - A) 2 File No.: 1300596.301 Page B-3 of B-10 Revision: 0 F0306-OIRI

Cjjo8 m efah &Of*l Asudsc*f, WOc For Axial stress oieitat(in .e.kow~jtuinW ais of RPV) i:= 0,1_3 0.2 W= 2 in b:- W (0.025'~

a 0.098 A I w 0.198 0.5 0-.25 T w0.0 2-5-S T =0 psi it (1500o0.-5.5)-(12 + 42)0-A"= 1476 psi 12.(9-2-53 +2-5-S )

12 1.122]

a -a.

I 1.1243 S1-133 b  !.1143) psi-,FN Kilo l I 7r a X-7c) FilM

{,1.027) 1.0037, 1.015 File No.: 1300596.301 Page B-4 of B-10 Revision: 0 F0306-OIR1

CA=" W**y Assof n.*

(1171l Km. -r-t-Fmi KM 2196 psi-4FM

ý.3S01) 0"*213-44 2

(Pressure stress) psi 2-7.125 p.. IL-150D0 Ib (Force per unit thickness) P a 3000 2-5 in M R-*15000-5-5 Ib-in (Moment per unit thickness) M a 16500 2-5 in (0.425" 1o0.1251 FIo. a1 -[.019 + 0.069-e 0.045 o0.033)

"0.641" S0.412I F Iq-

0-156[ - 0.067-.e k 0.307 01264, 1p-1 379 + 0-624-A. - 0.062-e (

1.606 FIpa 1552 (0.655"'

I0-351 Flip. -a0 1 2[6i- 0.24-A.i - 0.023-fl - A1)5]

Fp 0.23

-A 2  %,,016 File No.: 1300596.301 Page B-5 of B-10 Revision: 0 F0306-OIRI

-0 Camw"m thqlyg Assodftkh.

9A16 Fla. 1 3 .05- 0052-0.72e-9(1 6.039 2

.6.063) s3)11-901 "

F ;1 3-O.- + (I - A)410[-577 012K

- + 0-8 (A ] Film - 0.063 2 lý4J 14)

-A 1) r260-'

IF P

+Flmi +M K.I= I3415 psi.4iM 4642

ý,5464, e583'ý 14381 KU _2 -M

. i laa{F

+ P + FilmM KII 2 =i psi-4FM 1221 85 10 A equiviaent 19 may be given by:

33015 I -. )2]

K eq . [( 2 + (K uI + [A 2i)2 +

KIeq/4-3/ ,I 160591 1.106, File No.: 1300596.301 Page B-6 of B-10 Revision: 0 F0306-OIRI

Ibcfamw W llga AuSOdafte kx.0 2

a For Circunmerential oikNo (i.e. bencng abou thin axis of lug) i.' 0,1_ 3 0.71 a 1-5 in 11.,

W 2-=5 b:-W a

A:- A-I w 0-61 V- 0-5 1,0.8) 2-5-9 T-"0 psi File No.: 1300596.301 Page B-7 of B-10 Revision: 0 F0306-OIRI

2+ 4 2)0-TI-(15000-R-5-5)-15 T= 1476 psi 12 (1.124"'

a.2.-

I I 1-15 I 1-352 b S1.833)

KU_1=

KUI I T- TW a TH psi-4' r 1.003"i Fil 4rl2-b 1-034I 1-209

{1173 CM~ -= T1-,ýFajFffl 2265 I 3873 psi-s/in

,5792) 0-213.44 2

ar (Pressure stress) cr=0 psi 2-7-125 (Force per unt tckness) 3, P = 938 in M:= -5 9

(Moment per unt tthckness) Ib- in M = 2109 in 0.14 FIc :-jA - 1 0.0 9 + 0-069-e 1.{:J 0.03 1

Fia= 0.015 9x 1073j File No.: 1300596.301 Page B-8 of B-10 Revision: 0 F0306-01R I

14 4 CJ~fcauwe hMey2y Assodftes khL 0-4241

.10-156 MaA.

I A.

- 0.067-e (AJ

=I FFIo I0.247I d= 0.1271

,~0.078),

- 12. ,1-631 Fip. .[ 1 3 L F0-379 + 0.624-A. - 0.062-e FIp-1.711 3.s45 /

10978s, (0367 /

Flip1 3 .[0o 12 - 0.24-A. - 0.023-(1 - AJ)' I0.168I FIp - 0 .09 A-j2 F9"I- ,-0.925, FIm =

-A) 3 -[2.005 - 0 . 7 2-e 2

] [6.135 6.714 10.232

ý25063)

Fim 3[**0 m2+ (I - A.) 0-577 - 0.2A 0(A )I Film

[ 0.711

-+ -.06 4 A~~1~A~)-2.8311

[

2291 K L =;- {u Fl a.i+ P 4022I

- F~p j+ M -FIc%) KI- pSi-,Tm 10626 31521)

(299 *'

KIl2=I 7f-FUU +P Fpi+M + M -17

,,-3171),

File No.: 1300596.301 Page B-9 of B-10 Revision: 0 F0306-O1RI

!C MStXUaMM*kg*y AswdOE k*1~~

An eqisvaet K] mnay be 4mn by:

(r2694 Ele-

= (K-)+ (Klull + KU_2~) 2 + I - psa-4 ia -q 11622

,k32428)ý 320DXO

/

KIeq /

KII 2 40 0 0 /

/

7

/

KmU M000

-- -- L 0.4 0.6 0.7 0.9 1.A 13 I1 1.6 1.8 2 a

File No.: 1300596.301 Page B-10 of B-10 Revision: 0 F0306-OI RI