ML23354A246

Cracking Assessment for Framatome RSG Channel Head Assembly

ML23354A246
Person / Time
Site:	Callaway
Issue date:	12/20/2023
From:	Framatome
To:	Office of Nuclear Reactor Regulation
Shared Package
ML23354A244	List: ML23354A246 ULNRC-06849, License Renewal Resolution for Commitments 34 and 35 Perform Evaluation of Crack Initiation and Propagation in Steam Generator Divider Plate and Tube-To-Tubesheet Welds
References
ULNRC-06849 86-9366726-000
Download: ML23354A246 (1)

Text

Enclosure 1 Cracking Assessment for Framatome R$G Channel Head Assembly

0402-01 -FOl (Rev. 021 03/12/2018) framatGme CALCULATION

SUMMARY

SHEET (CSS)

Document No. 86 - 9366726 - 000 Safety Related: EYes D No Title Cracking Assessment for Framatome RSG Channel Head Assembly PURPOSE AND

SUMMARY

OF RESULTS:

PURPOSE:

The EPRI SGMP (Steam Generator Management Program) investigated the crack initiation and propagation in the SG (Steam Generator) channel head assembly for Westinghouse SGs, therefore similar evaluation as presented in Reference [1] is needed for Framatome supplied RSGs (Replacement Steam Generators) in the United States. The purpose of this analysis is to demonstrate by flaw tolerance assessment that a postulated initial flaw in the RSG channel heads meets the ASME Section XI, IWB-3612 acceptance criteria (Reference [61) over the design life of the RSGs. The analysis is applicable to the RSGs at Callaway , Prairie Island Units 1&2, Salem Unit 2.

RESULTS:

The flaw tolerance evaluation of the RSG channel head assembly is performed with the stress profiles provided by a detailed finite element analysis. The bounding transients (along with the design cycles) and support loads out of these units are used in the analysis. It is concluded that the RSG channel heads are not compromised by the postulated circumferential or axial flaw initiating from the divider plate over the design life of the RSGs. In addition, it cannot be determined that the RSGs meet the requirement of at least 22 wt% Cr in the tube-to-tubesheet welds.

If the computer software used herein is not the latest version per the EASI list, THE DOCUMENT CONTAINS AP 0402-01 requires that justification be provided.

ASSUMPTIONS THAT SHALL BE THE FOLLOWING COMPUTER CODES HAVE BEEN USED IN THIS DOCUMENT: VERIFIED PRIOR TO USE CODENERSION/REV CODE/VERSION/REV EWes N/A No Page 1 of 31

0402-Ol-FOl (Rev. 021, 03/12/207 8) framatGme Document No. 86-9366726-000 Cracking Assessment for Framatome RSG Channel Head Assembly Review Method: Design Review (Detailed Check)

D Alternate Calculation Does this document establish design or technical requirements? YES NO Does this document contain Customer Required Format? YES NO Signature Block PIRIAIM Name and Title and Pages/Sections (printed or typed) Signature LP/LR Date Prepared/Reviewed/Approved Kaihong Wang K WANG All except Section 6.3.

Advisory Engineer 9/1 1 /2023 Jennifer Nelson IA NELSON LR All except Section 6.3.

Principal Engineer 9/12/2023 Stacy Yoder SL YODER 9/1 1 /2023 Section 6.3.

Engineer IV Sarah Davidsaver SB DAVIDSAVER R Section 6.3.

Advisory Engineer 9/12/2023 TA SCHMITT Tim Schmitt 9/12/2023 A All.

Manager Notes: P/RIA designates Preparer (P), Reviewer (R), Approver (A);

LP/LR designates Lead Preparer (LP), Lead Reviewer (LR);

M designates Mentor (M)

In preparing, reviewing and approving revisions, the lead preparer/reviewer/approver shall use All or All except in the pages/sections reviewed/approved. All or All except means that the changes and the effect of the changes on the entire document have been prepared/reviewed/approved. It does not mean that the lead preparer/reviewer/approver has prepared/reviewed/approved all the pages of the document.

With Approver permission, calculations may be revised without using the latest CSS form. This deviation is permitted when expediency and/or cost are a factor. Approver shall add a comment in the right-most column that acknowledges andjustifies this deviation. -

Project Manager Approval of Customer References andlor Customer Formatting (N/A if not applicable)

Name Title (printed or typed) (printed or typed) Signature Date Comments BJ WATSON Beverly Watson Project Manager 9/12/2023 Page 2

franiatGnie 040201F01 (Rev. 021, 03/12/2018)

Document No. 86-9366726-000 Cracking Assessment for Framatome RSG Channel Head Assembly Record of Revision Revislo n PagesiSectionsiParag raphs No. Changed Brief Description I Change Authorization 000 All Initial release.

Page 3

f raniatGnie Document No. 86-9366726-000 Cracking Assessment for Framatome RSG Channel Head Assembly Table of Contents Page SIGNATUREBLOCK 2 RECORD OF REVISION 3 LISTOFTABLES 5 LIST OF FIGURES 6

1.0 INTRODUCTION

7 2.0 PURPOSEANDSCOPE 7 3.0 ANALYTICALMETHODOLOGY 8 3.1 Finite Element Model and Stress Analysis 8 3.2 Flaw Tolerance Evaluation 8 3.2.1 Postulated Flaws 8 3.2.2 Stress Intensity Factor Solutions 9 3.2.3 Fatigue Crack Growth in Low Alloy Steel 9 3.2.4 Methodology for Flaw Growth Analysis I0 3.2.5 Acceptance Criteria II 4.0 ASSUMPTIONS 12 4.1 Unverified Assumptions 12 4.2 Justified Assumptions and Modeling Simplifications 12 5.0 INPUTS 13 5.1 Geometry 13 5.2 Material 14 5.3 Loads 17 5.4 Finite Element Model 18 6.0 CALCULATIONS 21 6.1 Finite Element Stress Analysis 21 6.1.1 Design Conditions 21 6.1.2 Thermal Analysis 23 6.1 .3 Stress Analysis 23 6.1.4 Steady State Stress on Tubesheet Primary Side Perforated Area 25 6.2 Flaw Evaluation 25 6.2.1 Applied Stresses 25 6.2.2 Fatigue Crack Growth Analysis 27 6.3 Tube-to-Tubesheet Weld Material 28 7.0 RESULTS AND CONCLUSION 29

8.0 REFERENCES

31 Page 4

f rarnatorne Document No. 86-9366726-000 Cracking Assessment for Framatome RSG Channel Head Assembly List of Tables Page Table 3-1 : Relevant Sources of Stress for Fatigue Flaw Growth Analysis II Table 5-I : RSG Lower Assembly Major Dimensions I3 Table 5-2: RSG Lower Assembly Materials I5 Table 5-3: SA-508 Grade 3 Class 2 [3/4Ni-I/2Mo-Cr-V} 15 Table 5-4: SB-166, Alloy 600 UNS N06600 [72Ni-l5Cr-8Fe] 16 Table 5-5: SB-163, Alloy 690 UNS N06690 [58Ni-29Cr-9Fe] 16 Table 5-6: SA-403 Type 304L [l8Cr-8Ni] 16 Table 5-7: Material Strength 17 Table 5-8: Enveloped Transients and Bounding Cycles 17 Table 6-1 : Equivalent Axial Force and Bending Moment at P1/P3/P5 25 Table 6-2: Equivalent Axial Force and Bending Moment at P2/P4/P6 26 Table 6-3: Axial Stresses Due to Seismic Loads 26 Table 6-4: Initial and Final Circumferential Flaw Sizes 27 Table 6-5: Initial and Final Axial Flaw Sizes 28 Table 6-6: Framatome RSG Tube-to-Tubesheet Welds Materials and Chromium Content 29 Table 7-1 : Final Circumferential Flaw Evaluation 30 Table 7-2: Final Axial Flaw Evaluation 30 Page 5

franatGrne Document No. 86-9366726-000 Cracking Assessment for Framatome RSG Channel Head Assembly List of Figures Page Figure 3-1 : Postulated Inside Surface Partial Through-wall Semi-elliptical Flaws 9 Figure 5-1 : RSG Lower Assembly FEM Dimensions 14 Figure 5-2: 3-D Solid Model 19 Figure 5-3: Meshed Finite Element Model 20 Figure 6-1: Displacementin Design Conditions Case I 21 Figure 6-2: Stress Intensity in Design Conditions Case I 22 Figure 6-3: Displacement in Design Conditions Case 2 22 Figure 6-4: Stress Intensity in Design Conditions Case 2 23 Figure 6-5: Crack Growth Path Line Locations 24 Page 6

franiatonie Document No. 86-9366726-000 Cracking Assessment for Framatome RSG Channel Head Assembly

1.0 INTRODUCTION

There are several cases of crack indications in Alloy 82/182 and Alloy 600 material in the divider plate assembly in Westinghouse SGs (Steam Generators) in operation outside the US (United States). US utilities want to avoid having to define and execute an Aging Management Program at units where the RSG (Replacement Steam Generator) channel heads contain Alloy 82/1 82 or Alloy 600 material. The EPRI SGMP (Steam Generator Management Program) funded work beginning in 2006 and ending in 2014 to investigate crack initiation and propagation in the SG channel head assembly, which concluded that visual examinations of SG divider plate and tubesheet welds are adequate for inspecting the structural integrity of the SG channel head. The US NRC agreed with this conclusion, but required plants to ensure SGs are bounded by the analysis performed by the SGMP and documented in EPRI Technical Report 3002002850 (Reference [1]).

However, the SGMP work is performed using a bounding case Westinghouse 5G. Because of geometrical and material differences (notably in the use ofPWSCC (Primary Water Stress Corrosion Cracking) susceptible materials) and potential differences in loading, results and conclusions presented in Reference [1] are not necessarily applicable to Framatome supplied replacement steam generators. It was determined by Framatome that additional analysis is needed to cover the RSGs supplied to the US fleet by Framatome SAS. In particular, the RSGs in the following plant/unit are identified for this investigation:

. Callaway Unit 1 (Model 73/19T, 4 SGs), or Callaway-1 thereafter

. Prairie Island Units 1 & 2 (Model FRA-56/19, 2 SGs each unit), or P1-1/2 thereafter

. Salem Unit 2 (Model 61/19T, 4 SGs), or Salem-2 thereafter 2.0 PURPOSE AND SCOPE The engineering analysis conducted herein is consistent with EPRIs SGMP (Reference [1]), which is performed to address similar concerns in SGs supplied by other vendors. The objective is to assess crack growth in the tubesheet to divider plate weld of the Framatome RSGs to determine necessary inspections following license renewal. The Alloy 600 and Alloy 690 materials for both base and weld metals that are used in the fabrication of the channel heads ofthese Framatome RSGs are documented in Reference [2].

Consistent with Reference [1], two regions are considered in the stress and flaw evaluation analyses:

1) The junction of the channel head and the tubesheet is assessed assuming that the divider plate and tubesheet have separated through-thickness over a length of 25% of the radius of the channel head from the triple point (approximately 1 5 inches).

2) The Alloy 82/1 $2 tubesheet cladding.

Stresses at the channel head/tubesheet juncture under various loading conditions are evaluated in the original stress analyses. Transient loads are selected based on potential contribution to fatigue crack growth from the postulated separation of the tubesheet from the divider plate. The tubesheet cladding is addressed by looking at the stresses in the cladding during 100% steady state operation. The analysis in Reference [1] concluded that the cladding is in compression during steady state operation, making PWSCC impossible. As long as the analysis of the Framatome supplied SGs demonstrates that the primary surface of the Alloy $2/1$2 cladding is in compression throughout the drilled region ofthe tubesheet at 100% steady state conditions, no further analysis is necessary, and it can be concluded that PWSCC is not a concern. The detailed 3-D stress FEA (Finite Element Analysis) is documented in Reference [3].

For the triple point region, an ASME Section XI fatigue crack growth analysis is performed using the stresses

( and metal temperatures) documented in Reference [3]. The objective ofthe Section XI analysis is to demonstrate that fatigue crack growth of the initial flaw over the design life of the RSGs meets the ASME Code,Section XI, Page 7

frarnatonie Document No. 86-9366726-000 Cracking Assessment for Framatome RSG Channel Head Assembly IWB-36 12 acceptance criteria. Note that the potential for a crack through the Alloy 82/1 82 cladding existing anywhere along the circumference of the channel head-to-tubesheet juncture is covered by this analysis, i.e., it is not limited to considering a crack only at the triple point location. The flaw evaluation is documented in Reference [41 3.0 ANALYTICAL METHODOLOGY 3.1 Finite Element Model and Stress Analysis A representative FEM (Finite Element Model) is developed based on the detailed review of the four units, and a set of bounding transient data are determined. The following is the general methodology of FEM development and stress analysis:

1) Building a 3-D model of half of the RSG channel head assembly with ANSYS (Workbench) (Reference

[5]). The model incorporates the SG bottom channel head, the divider plate, tubesheet, and associated welds. The model also includes the stainless steel channel head cladding and Alloy $2/i $2 tubesheet cladding. The 3-D model is converted into a 3-D FEM, with appropriate materials properties and boundary conditions. There are two FEMs consisting of thermal and structural elements, respectively so as to enable the thermal and structural analysis using AN$YS (Reference [5]).

2) Applying the design conditions ofpressure and temperature (as temperature affects the material properties only) to the structural finite element model and obtaining the deformation and stresses in the model. The deformation field is used to verify the expected behavior of the model and correct modeling of boundary and load conditions.

3) Applying the thermal loads resulting from the plant operating transients (in the form of transient temperatures and corresponding heat transfer coefficients versus time). Evaluating the results of the thermal analysis by examining the magnitude of temperature differences between key locations of the model. The time points of the maximum temperature gradient are those at which the maximum thermal stresses develop.

4) Applying the corresponding pressure and thermal loads (nodal temperature) at each time point identified in step 3 and other time points of analytical interest in the structural finite element model and obtaining the stress results.

5) Defining path lines near the triple point and extracting axial and hoop stresses, metal temperatures along these path lines for the subsequent fatigue crack growth evaluation 3.2 Flaw Tolerance Evaluation 3.2.1 Postulated Flaws As shown in Figure 3 1 an inside surface-connected, partial through-wall, semi-elliptical circumferential flaw and an inside surface-connected, partial through-wall, semi-elliptical axial flaw are postulated to exist before it penetrates to the channel head base metal. For the circumferential flaw, the initial flaw depths (a) are taken to be the thickness of the cladding (0.2 minimal or the actual cladding thickness of the flaw path line) from its inside surface and a flaw length equal to the divider plate thickness (1 = 2c = 2.0$). This assumption yields a longer initial flaw than ifusing the aspect ratio ofa/l = 1/6 recommended by ASME Code Section XI, Table L-3210-l (Reference [6]). For the axial flaw, the aspect ratio of a/i = 1/6 is assumed.

Page 8

framatGme Document No. 86-9366726-000 Cracking Assessment for Framatome RSG Channel Head Assembly (a) semi-elliptical inside surface axial flaw Both (a) and (b):

a0 initial flaw depth 10 = 2c = initial flaw length t = wall thickness R1 = radius ofthe channel head inside surface x = flaw propagation direction (b) simi-elliptical inside aoh x surface circumferential flaw Figure 3-1 : Postulated Inside Surface Partial Through-wall Semi-elliptical Flaws 3.2.2 Stress Intensity Factor Solutions With the stresses near the triple point provided by Reference [3J, the SIF (Stress Intensity Factor) (K) is calculated with the Framatome Excel-based macro code (AREVACGC 6.0) using the weight function method. The technical basis for this implementation is documented in Reference [7]. The weight function methodology is used to compute K for the postulated flaw geometries mentioned above in this analysis.

3.2.3 Fatigue Crack Growth in Low Alloy Steel Per Article A-4300 of Reference [6], the crack growth due to fatigue in LAS (Low Alloy Steel) is characterized by:

da C0(L\K1)1

=

where C0 and ii are constants that depend on the material and environmental conditions, AKj is the range of applied SIF in terms ofksWin, and da/dN is the incremental flaw growth in terms of inches/cycle.

From Article A-4300(b)(2) of Reference [6], the fatigue crack growth constants for flaws exposed to the primary water environment are:

A11 = K,nax Kmin. IfKmjn 0, use R = 0.

0R0.25, AK1<17.74 Page 9

franiatorne Document No. 86-9366726-000 Assessment for Framatome RSG Channel Head Assembly Cracking n=5.95 s=1.o Co = 1.02 x 1O2S AK1 17.74 n = 1.95 S=LO Co = 1.01 x 0.25 < R < 0.65, AK < 1 7. 74 [(3. 75R + 0. 06)/(26. 9R 5. 725)]025 11=5.95 S=26.9R 5.725 Co = 1.02 x JW2S AK1 1 7. 74[(3. 75R + 0. 06)/(26. 9R 5. 725)1025 12 1.95

$=3.75R+O.06 Co = 1.01 x O.65R1.OO, AK1<12.04 fl =5.95 S=11.76 CO = 1.02 x 1W12S z1K1 12.04 11 1.95 S =2.5 Co=1.01 Additionally, per A-4300(b)(2) of Reference [6], if the fatigue crack growth rate from light-water reactor environments is lower than that from air environments, the rate in air should be used. Per A-4300(b)(1) of Reference [6], the fatigue crack growth constants for flaws in an air environment are:

n3.07 Co = 1.99 x S is a scaling parameter to account for the R ratio and is given by S 25. 72 (2. 88 R) 307, where 0 R 1 and zlK1 = Km. Kmjn.

For R < 0, zlK1 depends on the crack depth, a, and the flow stress, o The flow stress is defined by o = 1/2(o +

ui), where 03;s 15 the yield strength and u1 is the ultimate tensile strength.

For 2 R 0 and Km Kmin (0. 8) x 12 oka, S = 1 and AKj = Kmax.

For R < 2 and K,nax Kmin (0.8) xl 1257Ca, S = 1 and AKj= (1 R) Kmax/3.

For R < 0 and Kmax Kmin >(0.8) x1.12&a, S = 1 and AKi = Km Kmin.

Where the (0.8) reduction factor is established by NRC in title 10 CFR 50.55a, item (xxviii),Section XI condition: Analysis ofFlaws, Reference [8].

3.2.4 Methodology for Flaw Growth Analysis For the crack growth analysis, the applied SIFs of the postulated axial and circumferential flaws are driven by hoop and axial stresses, respectively. The relevant sources of stresses for fatigue crack growth are summarized in Table 3-1.

Page 10

franiatGrne Document No. 86-9366726-000 Cracking Assessment for Framatome RSG Channel Head Assembly Table 3-1 : Relevant Sources of Stress for Fatigue Flaw Growth Analysis Inside Surface-Connected, Partial Through-Wall, Semi-Elliptical Circumferential Flaw Residual Axial Stress at Shutdown Condition Axial Stress from Transients Axial Stress due to Pipe Loads (Deadweight, Thermal Expansion)

Axial Stress due to Normal/Upset Pressure (Acting on Crack face)

Inside Surface-Connected, Partial Through-Wall, Semi-Elliptical Axial Flaw Residual Hoop Stress at Shutdown Condition Hoop Stress from Transients Hoop Stress due to Normal/Upset Pressure (Acting on Crack face)

For each transient, the cycles are assumed to be uniformly distributed through the design lifetime. The cycles from all the transients are sorted based on the time that they are assumed to occur. Fatigue flaw growth is calculated by considering the assumed sequence of total transient stresses which may consist of a collection of sub-cycles (peaks and valleys) within any transient. Through-wall metal temperatures are provided in Reference

[3].

The fatigue crack growth of the inside surface-connected, partial through-wall, semi-elliptical axial flaw is controlled by the values ofK and 4K at the flaw depth location. The axial flaw growth is conservatively taken to be self-similar so that the initial flaw aspect ratio of the axial flaw is retained during flaw growth.

The fatigue crack growth calculations in the tubesheet base metal are performed by the Framatome Excel-based macro Code AREVACGC 6.0 (Reference [7]), which is verified for use by running verification test cases. The results generated by the test runs are identical to those documented in Reference [7].

3.2.5 Acceptance Criteria The allowable flaw depth is determined using ASME Code Section XI, Appendix A methodology and IWB-3 600 acceptance criteria (Reference [6]).

Per ASME Code Section XI, IWB-36 1 2 (Reference [6]), any flaw exceeding the limits of IWB-3500 is acceptable if the applied SIf and the flaw size satisfy the following criteria:

1) for normal conditions K1 < Kii\Jl0

2) for emergency/faulted conditions and conditions where pressurization does not exceed 20% of the Design Pressure K1 < K1 / J2 where, K1 = the maximum applied total SIf (including stress due to weld residual stress, pressure, and thermal loadings) for normal and upset (N/U) conditions, including test conditions, or emergency and faulted (E/f) conditions for the end-of-evaluation period flaw size.

Kic fracture toughness based on crack initiation for the corresponding crack-tip temperature.

Per ASME Code Section XI, Article A-4200 (Reference [6]), figure A-4200-l, a lower bound curve for Kic is prescribed for low alloy/ferritic steels as:

K1 = 33.2 + 20.734 exp[0.02(T RTNDT)]

Page 11

frarnatGnie Document No. 86-9366726-000 Cracking Assessment for Framatome RSG Channel Head Assembly where T is the evaluated metal temperature, and RTNDT is the reference nil-ductility temperature of the material.

The maximum K1 is limited to an upper-shelf value of 200 ksWin. The RTNDT of 10°F is taken as the bounding value based on the non-ductile failure risk reports of all units (References [9], [ 1 0], [ 1 1 ] and [ 12] for Callaway- 1, P1-i, P1-2 and $alem-2, respectively).

4.0 ASSUMPTIONS 4.1 Unverified Assumptions There is no unverified assumption used in this analysis.

4.2 Justified Assumptions and Modeling Simplifications

1) The impact of the primary nozzles and manways is assumed to be negligible, consistent with the modeling in Reference [1]. The channel head is modeled as a continuous semi-spherical shell without any openings.

2) Cycle counts used in the original Design Reports will be used in the crack growth analysis and are assumed to cover any proposed period of extended operation.

3) The analysis considers a separation between the tubesheet and the divider plate from the triple point (i.e.,

the juncture of the tubesheet, divider plate, and channel head) extending toward the center of the tubesheet over 25% ofthe length ofthe divider plate. The separation is assumed to occur through the Alloy 82/182 cladding and modeled explicitly. Note that based on the symmetrical nature of the model, this actually simulates a separation at both triple points.

4) Weld material ofEQ3O8L/EQ3O9L or E308L/E309L is assumed to have the same properties as the stainless steel type 304 material such as $A-403 Type 304.

5) As discussed in Reference [1], prior analyses have shown that the tubesheet remains essentially isothermal during the thermal transients of the steam generator channel head due to the thermal interaction of the tubes and the tubesheet. Consequently, vertical through-thickness thermal gradients are seen only in the top 2-inch layer of the tubesheet. A coupling of thermal DOF is applied to simulate the condition in the thermal analyses.

6) The initial flaw is assumed to be 100% through the cladding, growing into the LAS of the channel head, i.e., the initial flaw depth (a0 in Figure 3-1) is the cladding thickness at the location where the postulated flaw is assumed. Both axial and circumferential flaw orientations are considered.

7) Residual stresses are not explicitly calculated as part of this project. Since the cladding welds (both Alloy 82/1 82 as well as the stainless-steel welds) are post weld heat treated, the residual stresses are expected to be minimal as discussed in Table 3-1 ofReference [13]and subsequent discussion ofcladding residual stresses in Reference [1 3] indicates that the residual stresses at the clad-base metal interface are only in the order of+ 2 ksi. Residual stress of +2 ksi is then conservatively assumed to exist through the entire thickness of the cladding, and no residual stress is assigned to the LAS material.

8) This analysis does not include PWSCC (Primary Water Stress Corrosion Cracking) crack growth analysis in any material. Initial cracks through the susceptible material are postulated as described above.

Page 12

If rarnatonie Document No. 86-9366726-000 Cracking Assessment for Framatome RSG Channel Head Assembly

9) Stresses for LOCA and seismic loading used in the crack growth analysis are determined by applying an equivalent axial force load to the model that leads to the same axial membrane force from both axial and bending.

10) The initial flaw (axial or circumferential) is assumed to be semi-elliptical and the crack shape (i/a ratio) remains constant during crack growth. The magnitude of crack growth increment is driven by the K1 values at the deepest point (AREVACGC 6.0, Reference [7]).

5.0 INPUTS 5.1 Geometry The major dimensions used to build the FEM are shown in Figure 5-1, with values listed in Table 5-1 For the tubesheet perforated region, the effective Youngs modulus and the effective Poissons ratio are calculated from Appendix A of the ASME Code (Reference [14]), consistent with the values determined in the original stress reports.

Table 5-1 : RSG Lower Assembly Major Dimensions Component Dimension Symbol Value Tubesheet OD(in.) A 135.65 ID, primary side, base metal (in.) B 126.33 ID, secondary side (in.) C 130.2 Height, overall (in.) D 39.01 Height, primary side, base metal (in.) F 13.82 Thickness(in.) F 21.34 Diameter,perforated(in.) G 121.350 Divider plate lane, solid (in.) H 4.3588 Cladding thickness, tubesheet surface (in.) I 0.315 Cladding thickness, ID surface (in.) J 0.20 Cladding height, Alloy 600/690 (in.) K 2.165 Fillet radius, primary side (in.) L 1.58 Channel Spherical head inside radius, base metal (in.) M 65.12 Head Spherical head outside radius (in.) N 70.32 OD, cylinder portion (in.) 0 135.65 ID, cylinder portion (in.) P 126.33 Height, base metal (in.) Q 50.1 Claddingthickness(in.) R 0.2 Divider Thickness(in.) S 2.08 Plate Weld toe height, on tubesheet cladding (in.) T 1.29 Angle(°) U 5.5 Weld toe height, on channel head cladding (in.) V 1.29 Angle(°) W 15

$S weld layer thickness (in.) X 0.2 Page 13

framatGme Document No. 86-9366726-000 Cracking Assessment for Framatome RSG Channel Head Assembly Figure 5-1: RSG LowerAssembly FEM Dimensions 5.2 Material Materials of the modeled components are listed in Table 5-2. The cladding and weldment materials are taken from Reference [2]. Note that $A-508 Cl. 3a is the prior designation of $A-508 Gr. 3 Cl. 2, i.e., they are the same material.

Page 14

framatGme Document No. 86-9366726-000 Cracking Assessment for Framatome RSG Channel Head Assembly Table 5-2: RSG Lower Assembly Materials Region Component Typical Material Base Metal Tubesheet SA-508 Cl. 3a or SA-50$ Gr.3 Cl.2 Channel head SA-508 Cl. 3a or SA-508 Gr.3 C1.2 Lower shell SA-508 Cl. 3a or SA-50$ Gr.3 Cl.2 Divider plate (& closure) SB-168, UNS06690 Cladding Tubesheet primary side flat surface (2) FM 82/182 Tubesheet primary side cylindrical surface E308L/E309L, EQ3O8L/EQ3O9L Channel head inside surface EQ3O8L/EQ3O9L Weld Divider plate to tubesheet cladding (2) FM 152 Divider plate to channel head (2) FM 152 Tubesheet to channel head (3)

Note:

(1) Equivalent SB-163 Alloy 690 UNS N06690 properties are used.

(2) Equivalent $B-166 Alloy 600 UNS N06600 properties are used (3) Base metal material properties are used in the analysis.

The FEA uses the thermal properties mean coefficient of thermal expansion (a), specific heat (C), thermal conductivity (k) and the mechanical properties modulus of elasticity (E), Poissons ratio (ji), density (p). The detailed values (thermal & structural) for these materials are listed in Table 5-3 to Table 5-6.

Youngs Modulus (E) [106 psi]

Poissons Ratio (ii) [unitless]

Density (p) [lb/in3]

Coefficient of Thermal Expansion (a) [106 inlin-°f]

Thermal Conductivity (k) [10 Btu/sec-in-°F]

Thermal diffusivity (Td) [in2/sec]

Specific Heat (C k / (p*Id) [Btu/lb-°F]

The Youngs modulus (E) at 100°F or 650°F is an interpolated value.

Table 5-3: SA-508 Grade 3 Class 2 [3I4Ni-lI2Mo-CrV]

Component: tubesheet, lower shell, channel head Temp,°F E t p x k C 70 27.80 0.3 0.2841 6.4 5.046 0.1057 100 27.64 0.3 0.2839 6.5 5.093 0.1081 200 27.10 0.3 0.2831 6.7 5.185 0.1148 300 26.70 0.3 0.2823 6.9 5.185 0.1202 400 26.10 0.3 0.2817 7.1 5.162 0.1259 500 25.70 0.3 0.2809 7.3 5.093 0.1314 600 25.20 0.3 0.2802 7.4 4.977 0.1366 650 24.90 0.3 0.2797 7.5 4.931 0.1399 700 24.60 0.3 0.2794 7.6 4.861 0.1426 Reference [ 15] assumed [ 16] [15] [15] Calculated Page 15

framatGme Document No. 86-9366726-000 Cracking Assessment for Framatome RSG Channel Head Assembly Table 5-4: SB-166, Alloy 600 UNS N06600 [72N1-J5Cr-8Fe]

Comnonent: tubesheet c1addin Temp,°F F i p a k C 70 31.00 0.3 0.3063 6.8 1.991 0.1055 100 30.82 0.3 0.3060 6.9 2.014 0.1068 200 30.20 0.3 0.3053 7.2 2.106 0.1106 300 29.90 0.3 0.3045 7.4 2.222 0.1140 400 29.50 0.3 0.3038 7.6 2.338 0.1166 500 29.00 0.3 0.3030 7.7 2.454 0.1184 600 28.70 0.3 0.3023 7.8 2.569 0.1221 650 28.45 0.3 0.3019 7.9 2.616 0.1224 700 28.20 0.3 0.3016 7.9 2.685 0.1243 Reference [1 5] assumed [1 6] [ 1 5] [ 1 5] Calculated Table 5-5: SB-163, Alloy 690 UNS N06690 [58N1-29Cr-9Fe]

Component: divider niate & its weld Temp,°F E ji p a k C 70 30.30 0.3 0.3063 7.7 1.574 0.1028 100 30.12 0.3 0.3060 7.8 1.620 0.1034 200 29.50 0.3 0.3053 7.9 1.759 0.1075 300 29.10 0.3 0.3045 7.9 1.898 0.1113 400 28.80 0.3 0.3038 8.0 2.037 0.1140 500 28.30 0.3 0.3030 8.1 2.176 0.1173 600 28.10 0.3 0.3023 8.2 2.315 0.1189 650 27.85 0.3 0.3019 8.2 2.384 0.1204.

700 27.60 0.3 0.3016 8.3 2.454 0.1218 Reference [15] assumed [16] [15] [15] Calculated Table 5-6: SA-403 Type 304L [l8Cr-8Ni]

Component: stainless steel cladding Temp,°F E i p a k C 70 28.30 0.3 0.2864 8.5 1.991 0.1151 100 28.14 0.3 0.2862 8.6 2.014 0.1157 200 27.60 0.3 0.2853 8.8 2.153 0.1209 300 27.00 0.3 0.2844 9.0 2.269 0.1246 400 26.50 0.3 0.2836 9.2 2.407 0.1286 500 25.80 0.3 0.2827 9.4 2.523 0.1313 600 25.30 0.3 0.2818 9.5 2.616 0.1334 650 25.05 0.3 0.2814 9.6 2.685 0.1348 700 24.80 0.3 0.2810 9.7 2.731 0.1358 Reference [1 5] assumed [ 1 6] [1 5] [ 1 5] Calculated Page 16

Document No. 86-9366726-000 f rarnatGnie Cracking Assessment for Framatome RSG Channel Head Assembly Table 5-7 lists the value ofyield strength (cry) and ultimate strength (i) at various temperatures per Reference

[15] for the tubesheet base metal.

Table 5-7: Material Strength

. Temp. Yield Strength u Ultimate Strength uu Material Component

[°f] [ksi] [ksi]

70 65.0 90.0 SA-508 Gr. 3 Cl.2 Tubesheet 600 54.7 90.0 (SA-508 Cl. 3a) base metal 650 53.9 90.0 700 52.9 90.0 As aforementioned for the reference nil-ductility temperature of the tubesheet base metal, the bounding value RTNDT 10°F is used in this analysis.

5.3 Loads The transient data (Normal, Upset and Test conditions) from the four units are reviewed. Enveloped transients along with the bounding numbers of cycles are listed in Table 5-8.

The RSG design life is designated in References [17], [18], [19] and [20] for Callaway-l, Salem-2, P1-i and P1-2, respectively.

Table 5-8: Enveloped Transients and Bounding Cycles Condition Abbr. Transient name Typical Occurrences Normal HU Heatup 1 OOF/hr 200 CD Cooldown @200F/hr 200 UL Unit loading 1 5- 1 00% of full power (5%/mm) 18,300 UU Unit unloading 1 00- 1 5% of full power (5%/mm) 18,300 SLI Step load increase of 10% of full power (from 90%) 2,000 SLD Step load decrease of 10% offull power (from 100%) 2,000 LID Large step load decrease with steam dump 200 FC Feedwater cycling at hot shutdown 2,000 RTR Reduced temperature return to power 2,000 Upset TRT Turbine roll test 20 LOL Loss of load 80 LOP Loss ofpower 40 LOF Lossofflow 80 RT Reactor trip 400 CRD Control rod drop 80 151 Inadvertent safety injection 60 EFF Excessive feedwater flow 30 OBE Operating Basis Earthquake 400 Page 17

frarnatonie Document No. 86-9366726-000 Cracking Assessment for Framatome RSG Channel Head Assembly Condition Abbr. Transient name Typical Occurrences Test PSHT Primary side hydrostatic test, primary side 15 SSHT Secondary side hydrostatic test, secondary side 10 PSLT Primary side leak test, primary side 200 SSLT Secondary side leak test, primary side 120 There are two types of supports in the R$Gs: the support ring at the tubesheet location; four support lugs attached to the lower outside surface of the channel head. The reaction loads at the support ring or lugs are conservatively used in the stress analysis of the lower assembly (including the channel head but not inlet/outlet nozzles) performed by Framatome $AS. Out of the four units, the support lug loads are identified to generate the highest axial force and bending moment at the triple point cross-section and is therefore used as the bounding loads for the subsequent fracture mechanics analysis. The comparison is documented in Reference [3], with the data taken from References [ 1 7], [ 1 8], [ 1 9] and [20] for Callaway- 1 Salem-2, P1- 1 and P1-2, respectively.

5.4 Finite Element Model The finite element model is developed with ANSYS (Reference [5]). The solid model is shown in Figure 5-2. The model is meshed with 3-D 20-Node SOLID9O elements in the thermal analysis, and with 3-D 20-Node SOLID1 86 elements in the stress analysis. The meshed model is shown in Figure 5-3.

In addition, contact elements (CONTA174 with TARGE17O) are inserted at the separation between the tubesheet Alloy 600 cladding and the divider plate Alloy 690 weld from the triple point extending toward the center of the tubesheet over 25% of the length of the divider plate. In the thermal analysis, nodes on either side of the crack are coupled and no thermal resistance is considered for the small crack.

Page 18

Irarnatonie Document No. 86-9366726-000 Cracking Assessment for Framatome RSG Channel Head Assembly Figure 5-2: 3-D Solid Model Page 19

f raniatonie Document No. 86-9366726-000 Cracking Assessment for Framatome RSG Channel Head Assembly Figure 5-3: Meshed Finite Element Model Page 20

frarriatGnie Document No. 86-9366726-000 Cracking Assessment for Framatome RSG Channel Head Assembly 6.0 CALCULATIONS 6.1 Finite Element Stress Analysis 6.1.1 Design Conditions The following Design conditions are simulated in the model by applying a uniform and reference temperature of 650°f throughout the model:

1) Maximum primary to secondary pressure

2) Maximum secondary to primary pressure The purpose is to provide a basis for verification of the correct behavior of the model, the structural boundary conditions, and to verify stress attenuation at regions away from the weld connections. For pressure loads only, the pressure on the hot leg and cold leg remains the same. The global coordinate system is shown in figure 5-2.

For the first case, the displacement and stress intensity are shown in Figure 6- 1 and Figure 6-2, respectively. For the second case, the displacement and stress intensity are shown in Figure 6-3 and Figure 6-4, respectively.

The detailed view of the local region in Figure 6-1 indicates there is 25% separation of the divider plate starting from the triple point. Accordingly the very high peak stresses occur near the crack front and the numerical value cannot be treated as the actual stress values due to the stress singularity at this region.

Unit: inch Scale: 80

.936E04 .014005 .027917 .041829 .055741

.00705 .020361 .034873 .048785 .062697 Figure 6-1: Displacement in Design Conditions Case I Page 21

frarnatGnie Document No. 86-9366726-000 Cracking Assessment for Framatome RSG Channel Head Assembly Unit: psi Scale: 80 81.7911 41409.9 82738 124066 165394 20745.8 62073.9 103402 144730 186058 Figure 6-2: Stress Intensity in Design Conditions Case I

.007373 .021759 .036145 .050531 .064917

.014566 .028952 .043338 .057724 .072108 Figure 6-3: Displacement in Design Conditions Case 2 Page 22

frarnatGme Document No. 86-9366726-000 Cracking Assessment for Framatome RSG Channel Head Assembly Unit: psi Scale: $0 90.5324 25260.8 50431.1 75601.4

• 100772 12675.7 37846 63016.3 88186.6 113357 Figure 6-4: Stress Intensity in Design Conditions Case 2 6.1.2 Thermal Analysis The results of the thermal analyses are evaluated to identify the maximum and minimum temperature gradients between critical locations in the model and the corresponding time points. These temperature gradients generate maximum and minimum thermal stresses, which in turn contribute to the maximum range of stress intensities in the model.

In the thermal analysis on the primary side (both the cold leg and hot leg channel), a forced convection HTC (Heat Transfer Coefficient) is calculated to have a range between 0.01 19 to 0.0136 Btu/hr-s2-°F (details in Reference [31). The HTC of 0.0136 Btu/hr-s2-°F is then used for the primary side surfaces. On the secondary side near the RSG tubesheet, an HTC ranging from 0.000802 to 0.001027 Btu/hr-s2-°F is calculated at 15% normal power without recirculation. The HTC of 0.001027 Btu/hr-s2-°F is applied for the secondary side surfaces. In addition, for the perforated regions the HTC on either side is reduced by a factor of 0.52 (Reference [30J in [3]) to account for the reduced surface area due to the perforation.

6.1.3 Stress Analysis The nodal solution from the thermal analysis is loaded into the structural analysis with ANSYS. Time points selected from the thermal analysis include those with max/mm temperature gradients as well as those where the internal pressure changes its rate. Internal pressure at each time point is added as the mechanical load. In the stress analysis, the symmetric boundary conditions are applied to both the vertical and horizontal cut-off sections.

Page 23

franiatorne Document No. 86-9366726-000 Cracking Assessment for Framatome RSG Channel Head Assembly As shown in Figure 6-5, a total of six path lines are defined for the crack growth analysis. Path lines (P1 to P4) are located on each side ofthe divider plate near the triple point. Path P5 is between P1 and P3, and P6 is between P2 and P4; both are through the center of the divider plate but are not shown in the figure. The axial and hoop stresses, and the metal temperatures along these path lines are extracted for each transient.

Front view A-A Tubesheet Cladding

.1-,--t,-,t (SA-508 Cl. 3a; SA-508 Gr. 3 Cl.2)

(fM82/1182)

Divider Plate Weld (FM 152)

Divider Plate (SB-168)

V Channel Head Cladding (stainless steel, extended to tubesheet cylindrical side)

Center piece right view corner detail Centerpiece iso view from the left Center piece left view corner detail Figure 6-5: Crack Growth Path Line Locations Page 24

frarnatonie Document No. 86-9366726-000 Cracking Assessment for Framatome RSG Channel Head Assembly 6.1.4 Steady State Stress on Tubesheet Primary Side Perforated Area In the tube-to-tubesheet region, the possibility that PWSCC may lead to a primary to secondary leak is investigated by determining the stress condition of the cladding surface at the steady state operation.

The first load step of the unit unloading transient (UU) is steady state operation at 100% power. Post-processing of stress results on the cladding surface over the tube-to-tubesheet region is conducted. The three principal stresses (cr1, a2, on each node ofthe surface are obtained and arranged such that c 2 .

For all nodes on the surface, it is demonstrated that J > Gi, indicating that the compressive stress conditions are justified.

6.2 Flaw Evaluation 6.2.1 Applied Stresses 6.2.1.1 Transient and Residual Stresses The cyclic operating stresses needed to calculate fatigue crack growth are obtained from a linear, thermo-elastic finite element analysis described in Section 6. 1 These cyclic stresses are developed for all applicable transients at a number of time points to capture the maximum and minimum stresses due to fluctuations in pressure and temperature. Enveloped transients along with the bounding numbers of cycles are listed in Table 5-8 for the RSGs design life listed in Section 5.3.

The transient stresses, metal temperatures and pressures on path lines (P 1 to P6) are collected from the output files generated in Reference [3] into spreadsheets. These spreadsheets are used by AREVACGC, which reads the appropriate transient stresses, metal temperatures and pressures for the applicable path line.

In addition, a residual stress of+2 ksi (Reference [13]) is added to the thickness ofthe cladding, and no residual stress is assigned to the mbesheet base metal material (see Assumption #6 in Section 4.2).

6.2.1.2 Sustained Stresses due to Support Loads Axial forces and bending moments from the RSG support may introduce sustained axial stresses at the flaw location. There are two types of supports in the RSGs: the support ring at the tubesheet location; four support lugs attached to the lower outside surface of the channel head. The reaction loads at the support ring or lugs are conservatively used in the stress analysis of the lower assembly (including the channel head but not inlet/outlet nozzles) performed by framatome SAS. As aforementioned, the support lug loads are found to generate the bounding axial force and bending moment at the triple point cross-section and are therefore used as the bounding loads for the subsequent fracture mechanics analysis. The total axial forces and bending moments are listed in Table 6-1 and Table 6-2.

Table 6-1: Equivalent Axial Force and Bending Moment at PIIP3IP5 Faxiai Mbendmg Mb force L oa d ing lbs in-lbs in-lbs Deadweight (DW) 1,100,000 1,767,767 8,192,709 Pressure 200,000 2,828,427 2,730,903 Thermal Exp. (TH) 1,200,000 9,333,810 5,461,806 OBE 2,800,000 16,122,035 218,472,237 DBE 4,000,000 1 8,3 84,776 327,708,355 Pipe Rupture 6,000,000 36,062,446 491,562,532 Page 25

f rarnatorne Document No. 86-9366726-000 Cracking Assessment for Framatome RSG Channel Head Assembly Table 6-2: Equivalent Axial Force and Bending Moment at P21P41P6 Loading faxiai Mbending Mb force lbs in-lbs in-lbs Deadweight(DW) 1,100,000 1,767,767 7,957,285 Pressure 200,000 2,828,427 2,652,428 ThermalExp.(TH) 1,200,000 9,333,810 5,304,856 OBE 2,800,000 16,122,035 212,194,260 DBE 4,000,000 1 8,3 84,776 318,291,390 Pipe Rupture 6,000,000 36,062,446 477,437,084 6.2.1.3 Seismic Event The effect of the seismic loads on fatigue crack growth is addressed by modeling the OBE (Operating Basis Earthquake) seismic event as a transient event. The axial stresses due to the axial force and bending moment from the seismic loads (OBE in Table 6-1 and Table 6-2) are calculated. The results are given in Table 6-3 where R, and R are the inside and outside radius, respectively.

Table 6-3: Axial Stresses Due to Seismic Loads Path Line OBE (+)

P1/P3/P5 P2/P4/P6 By axial force, [psi] 1,256 1,402 By bending moment, at 1?, [psi] 3,093 3,362 By bending moment, at R0, [psi] 3,363 3,622 The baseline through-wall axial stress distribution for each path line is obtained from the stress state at steady state conditions. The OBE seismic transient is only applicable to the evaluation of circumferential flaws, which are driven by axial stresses. Semi-elliptical axial flaws are driven by hoop stresses, and there are no significant hoop stresses due to seismic loads.

6.2.1.4 Crack Face Pressure Loads For the fatigue crack growth analyses, it is conservatively assumed that primary water gets into the flaw and the crack faces are subjected to pressure loads. Time-dependent pressure loads for each transient are included in the crack growth calculation.

6.2.1 .5 Combination of Stresses Axial and hoop residual stresses at the shutdown condition (+2 ksi over the cladding thickness) are combined with the transient stress results, and sustained stresses such as loads due to deadweight and thermal expansion, and pressure acting on the crack faces of the postulated flaws to obtain the combined stresses over each path line.

These results are used to perform the fatigue crack growth calculation.

Page 26

Document No. 86-9366726-000 f ratnatGriie Cracking Assessment for Framatome RSG Channel Head Assembly 6.2.2 Fatigue Crack Growth Analysis For every postulated flaw type, a crack growth analysis is performed with the life time cycles listed in Table 5-8 and the final crack depth (and length) is calculated. The results are documented in Reference [41. The maximum applied total SIF (K1) is calculated for both the initial and final flaw configurations and compared with the flaw acceptance criteria outlined in Section 3.2.5.

The following subsections provide the fatigue crack growth analyses and the flaw size evaluations for the postulated circumferential and axial flaws. Based on the Design Specifications of these units, the following load combinations are used in the calculation of the maximum K1.

Normal Conditions (including Normal, Upset and Test Conditions):

DW + Pressure + OBE + Thermal Emergency/faulted Conditions:

DW + Pressure + SRSS(DBE, Pipe Rupture) + Thermal 6.2.2.1 Circumferential Flaw for the circumferential flaw, the initial and final flaw sizes at each location are summarized in Table 6-4. The maximum SIf is calculated with the initial and final flaw sizes and the results are documented in Reference [4].

for the Emergency/faulted Conditions, transient stresses at the beginning of HU and at the end of HU during steady state are taken as the instantaneous stress conditions. The stresses are then combined with the external stresses due to support pad loads to calculate the maximum SIf to be compared with the acceptance criteria.

Table 6-4: Initial and Final Circumferential Flaw Sizes Path Condition Depth (a), inch Length (1), inch Ratio (i/a) Ratio (a/t) 1 Initial 0.330 2.080 6.303 0.061 final 1.338 8.435 6.303 0.246 2 Initial 0.200 2.080 10.400 0.041 final 0.659 6.851 10.400 0.136 3 Initial 0.330 2.080 6.303 0.061 final 1.148 7.238 6.303 0.212 4 Initial 0.200 2.080 10.400 0.041 Final 0.235 2.444 10.400 0.048 5 Initial 0.330 2.080 6.303 0.061 final 0.375 2.365 6.303 0.069 6 Initial 0.200 2.080 10.400 0.041 Final 0.203 2.114 10.400 0.042 6.2.2.2 Axial Flaw Similarly for the axial flaw, the initial and final flaw sizes at each location are summarized in Table 6-5.

Page 27

franriatGnie Document No. 86-9366726-000 Cracking Assessment for Framatome RSG Channel Head Assembly Table 6-5: Initial and Final Axial Flaw Sizes Path Condition Depth (a), inch Length (1), inch Ratio (i/a) Ratio (cilt) 1 Initial 0.330 1.980 6.000 0.061 Final 0.443 2.660 6.000 0.081 2 Initial 0.200 1.200 6.000 0.041 final 0.214 1.282 6.000 0.044 3 Initial 0.330 1.980 6.000 0.061 Final 0.357 2.140 6.000 0.065 4 Initial 0.200 1.200 6.000 0.041 Final 0.211 1.266 6.000 0.043 5 Initial 0.330 1.980 6.000 0.061 Final 0.331 1.983 6.000 0.061 6 Initial 0.200 1.200 6.000 0.041 Final 0.200 1.200 6.000 0.041 6.2.2.3 Resolution of Fatigue Crack Growth in LAS with the 2019 Edition of the ASME Code, Section Xl Note that the fatigue crack growth for the tubesheet base material is only implemented for ASME Code years of 1992 and 1995 with 1996 Addenda in AREVACGC (Reference [7]). For this analysis, the Code year of 1995 is utilized in AREVACGC. The difference between the applicable Code year of ASME B&PV Section XI, Article A-4300, year 2019, Reference [6] and the utilized Code year of 1995 is in the determination of AK when R < 0.

The following method is applied for computing AK when R < 0 per the ASME B&PV Code Section XI, Article A-4300, Code year 1995:

For2R0 AKiKmax.

For R < 2 AK= (1 R) Km/3.

The parameters R, Km, Kmin, and (0.8) x] 12u&a, are evaluated conservatively for the initial flaw size of the postulated circumferential and axial flaws for all transients evaluated. The computed Km Kmin for these transients are all less than (0.8) x]* 12u/&a. Therefore, the computed AKj performed by AREVACGC (Reference

[7]) with 1995 version ofASME B&PV Code Section XI yields the same results as the applicable ASME Code Section XI, year of 2019 (Reference [61). Therefore, there is no impact on the results of this analysis.

6.3 Tube-to-Tubesheet Weld Material A materials assessment is performed in order to determine whether or not the tube-to-tubesheet welds have a chromium content of 22%, consistent with the requirements of EPRI Technical Report 3002002850 (Reference

[1 ]) and with NRC License Renewal Interim Staff Guidance 201 6-0 1 (LR-ISG-20 1 6-0 1 ) (Reference [2 1 J) which identifies for units with thermally treated Alloy 690 steam generator tubes with mbesheet cladding using Alloy 600 weld material, a plant-specific aging management program (AMP) is necessary unless the applicant confirms that the industrys analyses for tube-to-tubesheet weld cracking (e.g., chromium content for the tube-to-tubesheet is approximately 22 wt % and the tubesheet cladding is in compression) are applicable and bounding for its unit, and the applicant will perform general visual inspections of the tubesheet region looking for evidence of cracking

( e.g., rust stains on the tubesheet cladding) as part ofthe steam generator program.

Framatome replacement steam generators (RSGs) listed in Section 1 .0 are evaluated for the chromium content of the tube-to-tubesheet welds. The methodology used to determine the chromium content in Reference [2] for the tube-to-tubesheet welds and the results of this evaluation are shown in text and Table 6-6, respectively.

Page 28

frarnatGnie Document No. 86-9366726-000 Cracking Assessment for Framatome RSG Channel Head Assembly

. Chromium content for each material is based on the ASME B&PV Code Section II minimum design requirements

. A weld dilution of 50% is considered for the Tube material and Tubesheet cladding material Table 6-6: Framatome RSG Tube-to-Tubesheet Welds Materials and Chromium Content Operating Unit Tubesheet Cladding Material Tube Material Cr wt%

Callaway-1 UNS W86182/ENiCrFe-3 (FM 182) SB-163-690 <22 P1-2 & P1-2 UNS W86182/ENiCrFe-3 (FM 182) SB-163-690 <22 Salem-2 UNS W86182/ENiCrFe-3 (FM 182) SB-163-690 <22 It is noted in Reference [1] Section 2.6.2, a more appropriate equation for weld dilution is:

%CTwet %Crtube(0.52) + %Crclad(O.4$) Eq. 1 Calculating the weld dilution for Alloy 690 and FM 1 82 using Eq. 1 yields a chromium content of the weld of below the 22 wt% threshold. It is noted Reference [1] Section 2.7 calculates a mean chromium content value using Eq. 1 for Alloy 690 and FM 182 of 22. 15 wt%. Therefore, it is possible a record search of the certified material test reports (CMTRs) for the RSGs (which is outside the scope of this task) with Alloy 690/FM 1 82 tube-to-tubesheet welds may identify chromium content values, when input into Eq. 1 will yield a chromium content which exceeds 22 wt%.

However, based on this evaluation, it cannot be determined that the Callaway- 1, P1-i & P1-2, and Salem-2 RSGs meet the requirement of at least 22 wt% Cr in the tube-to-tubesheet welds.

7.0 RESULTS AND CONCLUSION The flaw tolerance evaluation of the Framatome supplied RSG channel head and tubesheet (Callaway-l, P1-1/2, Salem-2) is performed with the stress profiles provided by a detailed finite element analysis. The bounding transients (along with the design cycles) and support loads out of these units are used. It is concluded that the RSG channel heads are not compromised by the postulated circumferential or axial flaw initiated from the divider plate for the design life ofthe plant, by the criteria per ASME Code Section XI IWB-3600 (Reference [6]).

The final flaw evaluation results are summarized in Table 7- 1 for the circumferential flaws and in Table 7-2 for the axial flaws (bold fonts for limiting values).

Page 29

f rarnatonie Document No. 86-9366726-000 Cracking Assessment for Framatome RSG Channel Head Assembly Table 7-1: Final Circumferential Flaw Evaluation Path P1 P2 P3 P4 P5 P6 Final Depth (a), inch 1.33$ 0.659 1.148 0.235 0.375 0.203 Flaw Length(1),inch 8.435 6.851 7.238 2.444 2.365 2.114 Size Ratio(lIa) 6.303 10.400 6.303 10.400 6.303 10.400 Ratio (alt) 0.246 0.136 0.211 0.048 0.069 0.042 Normal Kmax, ksWin 49. 1 37.2 52.5 30. 1 37.8 16.2 Kmax Event PSHT EFF RT EFF PSHT PSHT Allowable, ksWin 63.2 63.2 63.2 63.2 63.2 63.2 Emergency! Kinax, ksWin 56.1 40.9 52.2 19.5 32.2 20.6 Faulted Kmax Event HU end HU end HU end HU end HU end HU end Allowable, ksWin 141.4 141.4 141.4 141.4 141.4 141.4 Table 7-2: Final Axial Flaw Evaluation Path P1 P2 P3 P4 P5 P6 Final Depth(a),inch 0.443 0.214 0.357 0.211 0.331 0.200 Flaw Length(l),inch 2.660 1.282 2.140 1.266 1.983 1.200 Size Ratio (l!a) 6.000 6.000 6.000 6.000 6.000 6.000 Ratio(a!t) 0.081 0.044 0.065 0.043 0.061 0.041 Normal Kmax,ksWin 19.687 15.682 20.353 15.521 10.186 4.971 Krnax Event SSHT SSHT SSHT SSHT RI SSHT Allowable, ksWin 63.2 63.2 63.2 63.2 63.2 63.2 Emergency! Kmax, ksWin 3. 1 1 .5 2.7 1 .5 5.9 2.8 Faulted Kmax Event HU begin HU begin HU begin HU begin HU end HU end Allowable, ksWin 72.2 72.2 72.2 72.2 141.4 141.4 Based on the evaluation presented in Section 6.3, it cannot be determined that the RSGs at Callaway-1, PI-1!2, and Salem-2 meet the requirement of at least 22 wt% Cr in the tube-to-tubesheet welds.

Page 30

franlatGnle Document No. 86-9366726-000 Cracking Assessment for Framatome RSG Channel Head Assembly

8.0 REFERENCES

References identified with an (*) are maintained within Framatome SAS or the clients Records System and are not retrievable from Framatome Records Management. These are acceptable references per Framatome Administrative Procedure 0402-01 Attachment 7. See page 2 for Project Manager Approval of customer references.

1 . EPRI Steam Generator Management Program Report 3002002850, Investigation of Crack Initiation and Propagation in the Steam Generator Channel Head Assembly, EPRI, Palo Alto, CA. 2014.

2. Framatome Document 51-9268036-000, EPRI SGMP for Channel Head Aging ofthe framatome RSG Designs.

3 . Framatome Document 32-93601 1 1-000, framatome RSG Channel Head and Tubesheet Stress Analysis for Flaw Evaluation.

4. Framatome Document 32-9364633-000, framatome RSG Channel Head and Tubesheet Flaw Tolerance Evaluation.

5. ANSYS Finite Element Computer Code, Version 19.2, ANSYS, Inc., Canonsburg, Pa.

6. ASME Boiler and Pressure Vessel Code Section XI, Rules for Inservice Inspection ofNuclear Power Plant Components, 2019 Edition.

7. Framatome Document 32-9055891-007, Fatigue and PWSCC Crack Growth Evaluation Tool AREVACGC.

8. Code ofFederal Regulations, Title 10, Part 50.55a, Domestic Licensing ofProduction and Utilization Facilities, Codes and Standards, Federal Register Vol. 80.

9. *BUCRCANGV 1885 Rev. B, Callaway Unit 1 Replacement Steam Generators Section 13 : Non Ductile Failure Risk.

10. 1 746 Rev. E, Prairie Island Unit 1 Replacement Steam Generators Section 16: Non Ductile Failure Risk.

11 . 2596 Rev. A, Prairie Island Unit 2 Replacement Steam Generators Section 16: Non Ductile Failure Risk.

12. *BUCR5ANGV 1998 Rev. E, Salem Unit 2 Replacement Steam Generators Section 14: Non Ductile Failure Risk.

13 . EPRI-Framatome Project Agreement 10012545, Divider Plate Cracking Assessment for Framatome Replacement Steam Generators, EPRI Task ID 1-072913-04-01, May 29, 2020.

14. ASME Boiler and Pressure Vessel Code,Section III, Division 1, 1995 Edition with Addenda through 1997.

15 . ASME Boiler and Pressure Vessel Code,Section II, Part D, 1995 Edition with Addenda through 1997.

16. Framatome Document NPGD-TM-500 Rev D, NPGMAT, NPGD Material Properties Program, Users Manual (03/1985).

17. *M1 1 83 (Q) Rev. 3 Technical Specification for Replacement Steam Generators Callaway Plant Unit 1

18. *52RCMD50397 Rev. 3, Design Specification for Replacement Stream Generators Salem Unit 2.

19. *M530.0001005.40, Certified Design Specification for Replacement Steam Generators Northern States Power Company Prairie Island Nuclear Generating Plant Unit 1 .

20. *M530.000100705, Certified Design Specification for Replacement Steam Generators Northern States Power Company Prairie Island Nuclear Generating Plant Unit 2.

21 . NRC License Renewal Interim Staff Guidance LR-ISG-20l6-0l, Changes to Aging Management Guidance for Various Steam Generator Components, 2016, ML16237A383.

Page 31