NL-08-0228, License Renewal - Responses to 01/22/2008 RAIs

License Renewal - Responses to 01/22/2008 RAIs

ML080560267
Person / Time
Site:	Vogtle
Issue date:	02/21/2008
From:	Tynan T Southern Nuclear Operating Co
To:	Document Control Desk, Office of Nuclear Reactor Regulation
References
NL-08-0228
Download: ML080560267 (58)

Text

Tom Tynan Southern Nuclear

'Vice President - Vogtle Operating Company, Inc.

7821 River Road Waynesboro, Georgia 30830 Tel 706.826.3151 Fax 706.826.3321 SOUTHERN February 21, 2008 COMPANY Energy to Serve Your World*

Docket Nos.: 50-424 NL-08-0228 50-425 U. S. Nuclear Regulatory Commission ATTN: Document Control Desk Washington, D. C. 20555-0001 Vogtle Electric Generating Plant License Renewal - Responses to 01/22/2008 RAIs Ladies and Gentlemen:

By letter dated June 27, 2007, Southern Nuclear Operating Company (SNC) submitted a License Renewal Application (LRA) for Vogtle Electric Generating Plant (VEGP) Units 1 and 2, seeking to extend the terms of the operating licenses an additional 20 years beyond the current expiration dates.

By letter dated January 22, 2008 the Nuclear regulatory Commission (NRC) submitted eight Requests for Additional Information (RAIs) to SNC resulting from the NRC staff review of the LRA. The SNC responses to these RAIs are provided in the enclosure to this letter.

(Affirmation and signature are provided on the following page.)

I &-12 q,

U. S. Nuclear Regulatory Commission NL-08-0228 Page 2 Mr. T. E. Tynan states he is a Vice President of Southern Nuclear Operating Company, is authorized to execute this oath on behalf of Southern Nuclear Operating Company and to the best of his knowledge and belief, the facts set forth in this letter are true.

The NRC commitments contained in this letter are listed in the updated License Renewal Commitment List, to be provided concurrently with the first LRA amendment. If you have any questions, please advise.

Respectfully submitted, SOUTHERN NUCLEAR OPERATING COMPANY

~N \ T. E. Tynan

,k. <~

Vice President - Vogtle Sworn to and subscribedbefore me this ____ _ day of ihV #[+/-~2008.

Notary Public, Burke County, Georgia Expires January 13, 2012 My commission expires: My Commission TET/JAM/daj

Enclosure:

VEGP License Renewal Audit Question Responses cc: Southern Nuclear Operating Company Mr. J. T. Gasser, Executive Vice President w/o Enclosure Mr. T. E. Tynan, Vice President - Vogtle w/o Enclosure Mr. D. H. Jones, Vice President - Engineering w/o Enclosure Mr. B. J. George, Manager, Nuclear Licensing w/ Enclosure Mr. N. J. Stringfellow, Licensing Supervisor, Vogtle w/ Enclosure RType: CVC7000 U. S. Nuclear Regulatory Commission Mr. V. M. McCree, Acting Regional Administrator w/ Enclosure Mr. S. P. Lingam, NRR Project Manager - Vogtle w/ Enclosure Mr. G. J. McCoy, Senior Resident Inspector - Vogtle w/ Enclosure Mr. D. J. Ashley, License Renewal Project Manager, Vogtle w/ Enclosure State of Georgia Mr. N. Holcomb, Commissioner - Department of Natural Resources w/o Enclosure

Vogtle Electric Generating Plant Enclosure VEGP License Renewal RAI Responses - January 22, 2008

NL-08-0228 RESPONSE TO VEGP RAI 4.3-4 ATTACHMENT A FatiguePro Benchmarking Study for Vogtle-1/2

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 1.0 EXECUTIVE

SUMMARY

FatiguePro monitors fatigue usage for the Vogtle Units 1 and 2 Charging Nozzle and the RCS Hot Leg Surge Line using a uniaxial fatigue approach. This RAI response documents the performance of a benchmarking study demonstrating conservatism of the FatiguePro-calculated results relative to a more refined NB-3200 fatigue analysis for the two sample plant components.

The transients used in the analyses were developed to produce the range of stress response expected during plant operations. Two transients (one fast, one slow) were developed for the Charging Nozzle analyses. Two transients (one with global loadings during a heatup and cooldown cycle and one with asymmetric local thermal stratification loadings) were developed for the RCS Surge Line Hot Leg Nozzle.

The results of the comparison between the FatiguePro and the NB-3200 analyses are shown below. In each case, the FatiguePro analysis produced larger Salt and CUF than did the ANSYS NB-3200 analysis for the evaluated plant transients.

Charging Nozzle Transient Fast Transient Slow Transient Loss of Letdown with Delayed Return to Service Letdown Flow 50% Decrease and Return to Normal Analysis Tool FatiguePro NB-3200 Percentage FatiguePro NB-3200 Percentage Max Sait 296.8 261.2 88% 9.37 8.96 96%

Max S, 111.3 107.8 97% 11.71 13.3 114%

Max K, 3.333 3.333 100% 1.0 1.0 100%

CUF 0.0138 0.0102 74% 0.00E+00 0.00E+00 N/A RCS Surge Line Hot Leg Nozzle Transient Cooldown and Heatup MOP Heatup with Local Stratification Analysis Tool FatiguePro NB-3200 Percentage FatiguePro NB-3200 Percentage Max SaIt 14.465 8.8 61% 66.9 29.2 44%

Max Sn 23.5 10.5 45% 61.6 47.1 76%

Max K, 1.0 1.0 100% 1.767 1.0 57%

CUF 2.38E-08 0.OOE+00 0% 0.00012 1.50E-06 1%

1

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2

2.0 INTRODUCTION

/OBJECTIVE This paper documents the performance of a benchmarking study demonstrating conservatism of the FatiguePro-calculated results relative to a more refined ASME Subarticle NB-3200 [1]

fatigue analysis for two sample plant components; the Charging Nozzle and the RCS Hot Leg Surge Line Nozzle.

Transients were defined for each component and analyzed in FatiguePro [2, 3]. Using the same transient loading histories, each of the components was then analyzed for fatigue per the analysis rules in NB-3200.

For each of the two locations analyzed subject to the defined transients, FatiguePro produced higher fatigue usage than that from the NB-3200 analysis.

3.0 TECHNICAL APPROACH To benchmark the FatiguePro results for each component, a fatigue usage analysis was performed for the same two plant transients analyzed by FatiguePro, using the methodology of Subarticle NB-3200 of Section III of the 1986 ASME Code [1]. Applied loadings, including thermal, pressure, and piping moments in all three orthogonal directions were taken from the FatiguePro output, so that the two analyses would be comparable.

3.1 FatiguePro Analyses The transfer functions utilized by the FatiguePro software [6] conservatively account for the combination of time-dependent thermal stresses and instantaneous piping and pressure stresses using a uniaxial fatigue methodology. The stress time-history, accounting for thermal, pressure, and piping stresses, ultimately defines the peaks and valleys that are filtered into alternating stress pairs for input to the fatigue curve. Each stress pair is adjusted, as appropriate, by the simplified elastic-plastic factor, KI, and the ratio of modulus of elasticity (E) from the fatigue curve divided by E from the analysis, per the rules of ASME Section III.

The FatiguePro simulations defined a time-history of the local temperatures, pressures, film coefficients, and bending moments at the nozzle for two different load cases for each component.

These loads represent the loads assumed or considered by the FatiguePro analysis.

2

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 3.2 NB-3200 Analyses For the NB-3200 analysis, finite element analysis was performed using a 3-D ANSYS [4] model of each component. The VESLFAT program [5] was used to perform the fatigue analysis.

VESLFAT is an analysis program verified under Structural Integrity Associates nuclear QA program, which is used to perform the NB-3200 fatigue usage analysis. Attachment 1 includes a more complete description.

A 3-D ANSYS finite element model was used to perform the stress analysis for thermal and pressure loading. Manual calculations were performed to produce piping stresses. Thermal transient analysis was performed for each defined transient, and the thermal stresses were added to stresses due to pressure and piping loads, which were scaled based on the magnitudes of unit pressure and piping loads. Stress concentration factors (SCFs) were applied, as appropriate. All six components of the stress tensor were used for stress calculations.

3.3 Items of Comparison For each of the two plant component locations, comparisons were made between the FatiguePro and NB-3200 analysis results for alternating stress intensity, secondary stress range, Ke and fatigue usage.

Descriptions of the specific evaluations for the Charging Nozzle and RCS Surge Line Hot Leg Nozzle are provided in the following sections.

4.0 CHARGING NOZZLES The FatiguePro and NB-3200 fatigue analyses were performed at the FatiguePro-monitored location (safe end region), which was identified to be the critical fatigue location in the Charging Nozzle. A discussion of both the FatiguePro evaluation [7] and the NB-3200 evaluation [8] is presented in this section.

4.1 Transient Definitions The two transients simulated for this study were the 'Loss of Letdown with Delayed Return to Service' (fast transient) and 'Letdown Flow 50% Decrease and Return to Normal' (slow transient) [7]. These temperature transients are depicted in Figures 1 and 2.

3

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 Loss of Letdown with Delayed Return Charging Nozzle Temperature 600 -; r nCharging and letdown o.

500 400 300 200 200 400 600 800 1000 1200 Tiwe (scc)

Figure 1: Design Transient for Fast Transient Letdown Flow 50% Decrease and Return to Normal Charging Nozzle Temperature S

6 I-200 400 600 800 1000 1200 1400 1600 1800 Time (secs)

Figure 2: Design Transient for Slow Transient 4

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 4.2 FatiguePro Evaluation of Charging Nozzle 4.2.1 Fast TransientResults The local fluid temperatures and stress history for the Charging Nozzle for the fast transient developed by the FatiguePro analysis are shown in Figure 3.

The FatiguePro fatigue calculation develops a stress loading spectrum from the local peaks and valleys of the monitored stress history. For the simple simulation performed here, the stress history consisted of two stress pairs, derived from the down- and up-shocks of the Loss-of-Letdown Delayed (fast) transient. The details are reported in Table 1.

The Fatigue Details Report for the fast transient shows that the fatigue usage is determined by a maximum stress pair, with a (total) stress intensity range of 168.7 ksi. A second stress pair was found in the transient, but it did not contribute significantly to fatigue usage. A corresponding K, factor of 3.33 is applied by FatiguePro along with adjustment of the stress range for Elastic Modulus ratio, giving a cumulative usage factor of 0.01377.

4.2.2 Slow TransientResults The local fluid temperatures and stress history for the Charging Nozzle for the fast transient are shown in Figure 4. As shown in this figure, the stress cycle resulting from the Letdown Flow 50% Decrease and Return to Normal (slow) transient is below the fatigue endurance limit. The details are reported in Table 2.

5

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 Loss of Letdown wiith Dehyved Return (Fast)

-U-ýIF.

CýRO +-tv:CHG NOZ

-120012 60 1LR E- -200 ---- .- --------------------

00:05:00 0010:00 00t:i 5:00 ý0:20:(K0 00::25:00O DatetTimne 112f1 7/2007]

Figure 3: FatiguePro Local Temperatures and Stress for Fast Transient 6

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 Table 1. Fatigue Usage Results for Fast Transient Loss of Letdown with Delayed Return to Service Detailed Fatigue Report (CHRGNOZ)

Ordered Overall Range Table Level (%) Cycles Usage 99.50 1 0.01377 98.00 0 0.00000 96.00 0 0.00000 93.50 0 0.00000 90.00 0 0.00000 86.00 0 0.00000 81.00 0 0.00000 75.50 0 0.00000 69.00 0 0.00000 61.50 0 0.00000 53.00 1 0.00000 44.00 0 0.00000 34.00 0 0.00000 24.00 0 0.00000 14.00 0 0.00000 6.00 0 0.00000 Usage from Spectrum 0.01377 Removed Fatigue Usage 0.00000 Initial Fatigue Usage 0.00000 Total Fatigue Usage 0.01377 Peak Stress Range 168.664 Maximum Ke 3.33333 Rapid Cycling Usage 0.00000 7

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 Letdown Flow 50% Decrease (Slow)

- - CHRG NOZ - - 0t :CMRGNOZ Oq 28

-40. 24

- 0 ------- ------------------- 2 Oatefrime [12118/2007]

Figure 4: FatiguePro Local Temperatures and Stress for Slow Transient 8

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 Table 2. Fatigue Usage Results for Slow Transient Letdown Flow 50% Decrease with Return to Normal Detailed Fatigue Report (CHRGNOZ)

Ordered Overall Range Table Level (%) Cycles Usage 99.50 1 0.00000 98.00 0 0.00000 96.00 0 0.00000 93.50 0 0.00000 90.00 0 0.00000 86.00 0 0.00000 81.00 0 0.00000 75.50 0 0.00000 69.00 0 0.00000 61.50 0 0.00000 53.00 1 0.00000 44.00 0 0.00000 34.00 0 0.00000

.24.00 0 0.00000 14.00 0 0.00000 6.00 0 0.00000 Usage from Spectrum 0.00000 Removed Fatigue Usage 0.00000 Initial Fatigue Usage 0.00000 Total Fatigue Usage 0.00000 Peak Stress Range 17.7451 Maximum Ke 1.0 Rapid Cycling Usage 0.00000 9

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 4.3 Assumptions / Design Inputs for NB-3200 Analysis of Charging Nozzle 4.3.1 Assumptions

" Material properties in the FEM were taken at 350°F for consistency with the original Green's Function configured in the Vogtle FatiguePro software.

" Consistent with the original Green's Function analysis, the thermal sleeve and water gap were included in the FEM for the thermal analysis, but removed in the stress analysis.

" Stress components due to piping loads were scaled assuming no stress occurs at an ambient temperature of 70'F. As such, the FEM was created such that the structure is at a zero stress state at 70'F.

4.3.2 ASME Code Edition The analysis was performed in a manner consistent with the NB-3200 [1] fatigue analysis. The 1986 Edition was used, unless otherwise noted.

4.3.3 Heat Transfer Coefficientsfor Transients FatiguePro switches between two Green's Functions based on the range of the computed flow rate at the nozzle, as shown on Table 3 [6]. The heat transfer coefficient in the cold leg piping is constant, based on maximum flow rate. In the FEM for the NB-3200 analysis, the same film coefficients are applied based on the flow rate. The intent is to apply the same loads as that assumed in the FatiguePro analysis to isolate the effect of the stress combinations and fatigue analysis.

10

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 Table 3. Heat Transfer Coefficients for Charging Nozzle Region Flow Rate h (BTU/hr.ft 2"°F) h (BTU/sec-in 2.°F)

Charging Nozzle Flow Path, hl > 60 gpm 2332.8 0.0045 Charging Nozzle Flow Path, hl <= 60 gpm 1332.0 0.0026 RCS Cold Leg, h2 All 7361.3 0.0142 These heat transfer coefficients were computed such that they bound any value for the range of temperatures that the component could see, based on the given flow range.

The maximum heat transfer coefficient used in the design basis analysis for the Letdown Trip with Delayed Return to Service event was 1775.4 BTU/hr-ft 2.°F and the minimum value was 81.6 BTU/hr-ft 2-°F.

4.3.4 Geometry and MaterialProperties The Vogtle Charging Nozzles are composed of a stainless steel nozzle forging welded onto stainless steel charging piping. The nozzle and piping are ID counter-bored at the welded connection [8].

Material designations for the Charging Nozzle were provided by Westinghouse. The same material designations were used in FatiguePro and the NB-3200 analysis.

For the narrow annulus of fluid between the thermal sleeve and the nozzle, an effective conductivity was calculated [9]. This portion of the fluid was included in the FEM to accurately account for the heat transfer behavior between the bulk fluid and nozzle. Following the thermal analysis and prior to the stress analysis, the fluid elements in the annulus and the thermal sleeve were removed.

4.3.5 Finite Element Model The ANSYS program was used to perform the finite element analysis. An axisymmetric FEM of the Charging Nozzle was developed in the FatiguePro model (Figure 5).

A 3-D model using brick elements was then created for the NB-3200 analysis. The 3-D FEM differs from the axisymmetric FEM in that the nozzle and cold leg were modeled as the intersection of two cylinders. A quarter model was generated to be capable of calculating stresses due to symmetric loading (thermal, pressure). The quarter model is shown on Figure 6.

11

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 0

04 N

N,-

0j z3 H-Figure 5. Axisymmetric FEM of Charging Nozzle 12

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2

,IF N

N 0

z 0

H

.0 U

ri)

N 0)

J-3 0

Figure 6. 3-D FE Quarter Model of Charging Nozzle 13

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 For a consistent comparison between the FatiguePro model and the ANSYS model, stresses were extracted and linearized at the same longitudinal location monitored by FatiguePro. The only difference was the circumferential position, which is not differentiated in an axisymmetric model. The critical location was determined to be at the inside surface corner of the counter-bore in the nozzle to pipe region. This location was identified based on highest stress intensity due to thermal loading under high flow conditions.

4.4 Loading Calculations for NB-3200 Evaluation of Charging Nozzle 4.4.1 PipingInterface Loads In FatiguePro the stresses due to piping are based on an NB-3600 [1] approach that adds maximum bending stresses from the outside surface of the pipe to the transient stresses on the inside of the nozzle.

In the ANSYS analysis, the individual components of stress were calculated taking into account through-wall distribution.

Referring to Figure 7, and using the cylindrical coordinate system shown in the upper right of the figure, from general structural mechanics, the membrane plus bending stresses at the inside surface of a thick-walled cylinder are:

(T, = Axial stress due to bending moments = [Mb'sin(0) - Mc-cos(0)].(ID/2)/I oe = Circumferential shear stress (in 0 direction) due to torsion = Ma(ID/2)/J where:

ID = Inside diameter = 2.625 in OD = Outside diameter = 3.50 in 4 4 I = Moment of inertia = (rt/64)(OD - ID ) =4 5.035 in 4

J = Polar moment of inertia = 21 = 10.071 in Piping axial and shear forces were not used in the design basis analysis [10], and so were not provided in the design input. They are judged to have negligible effects on the alternating stress intensity of a fatigue contributing transient.

14

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 Y

UP X

+Mb Figure 7. Coordinate System for Piping Loads Table 4 lists bending moments on the Charging Nozzle for various conditions of operation [II].

Table 4. Moments for Vogtle Charging Nozzle Case Description T in-kips Case DescriptionRCL S1 S2 Ma Mb M, 1 Cold charging/letdown isolation 558 70 70 14.3 -21.8 23.2 2 Charging off 558 558 70 12.9 -18.6 25.5 4 Hot charging 558 500 500 9.3 -0.8 3.3 Case 1 has the largest SRSS of the moments. It also coincides with the worst case thermal transient (Loss of Letdown). Figure 8 shows the membrane plus bending axial stress and the stress intensity due to Case 1 moments as a function of the circumferential position around the pipe. The maximum stress intensities occur at 430 from horizontal, and at 2230, 180' apart.

15

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 Additionally, the maximum alternating bending moment in going from Case 4 to Case 1 is located at 47'. In the ANSYS finite element quarter model, the closest node to 430 and 470 along the critical section is located circumferentially at 450 from horizontal, or Node 2137.

Adjacent toNode 2137 on the outside surface is Node 2097. The path from 2137 to Node 2097 therefore represented the critical path for which to extract linearized and peak stresses from the model. Regardless of the circumferential path chosen in the FEM, the stress results were virtually identical because of the symmetric loading.

Piping Stresses Around Inside Circumference Due to Orthogonal Bending Stresses Plus Torsion Max at 223' 10 -- ----- -

-I-- -- I- ý -- I

-

ý -- L L

-

-IL Ji L -LL

- - - - - - - - - - - - - - - - - - - - - -

L - L - - L - L - - - .... :

8

1. _ -I__

____ F ~ -1::  :

r' -_-___%_-I T____

-~T t -

6 4

L - L ~ L i 2

a, 0

-2

-4 7 V I -I -- IT -

22

-6

-8

-10 Circumferential Position (Theta, degrees from horizontal)

I- Axial Stress - - Stress Intensity I Figure 8. Membrane Plus Bending Piping Stresses for Case 1 16

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 4.5 Stress and Fatigue Calculations for NB-3200 Analysis of Charging Nozzle 4.5.1 Summary of FatigueAnalysis Performed T hermal transient stress components were calculated in ANSYS using the thermal transients developed from FatiguePro.

Linearized stress components at Node 2137 (safe end inside surface) were used for the fatigue usage analysis. At the safe end, nominal stress components due to piping loads are multiplied by 1.8 to yield P+Q+F stress components.

The fatigue usage analysis considered all six stress components and was performed using the NB-3200 rules of Section III of the ASME Code.

4.5.2 Finite Element Analysis Using the loading history, analysis of the Fast and Slow transients was performed using the 3-D ANSYS FEM of the Charging Nozzle. Heat transfer coefficients are applied as appropriate based on flow rate. On the first time step, the model was brought to. steady state conditions to establish baseline temperature distributions, and thus baseline steady state stresses for the fatigue analysis. Stress analysis is performed using as input the temperature distributions calculated in the thermal transient ANSYS analysis. As described earlier, the water elements in the annulus, and the thermal sleeve were removed.

4.5.3 Stress Calculations Linearized stress components at Node 2137 (safe end inside surface) were used for the fatigue usage analysis. They were extracted in a cylindrical coordinate system to be consistent with the piping load stress equations defined earlier. The stress components from the thermal analysis were combined with stress components due to piping loads. Piping load stress components were computed based on the time-history of the individual moment terms, and using the closed form solutions and multiplied by an SCF. Thermal transient stresses were then combined with the piping load stresses for both membrane plus bending (P+Q) and the total (primary plus secondary plus peak) stress components (P+Q+F). The stresses for each of the six stress tensors were added together on an individual basis, representing the stress state at each time step.

17

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 4.5.4 Fatigue Usage Analysis In the VESLFAT program [5], transients that consist of both upward and downward temperature and pressure ramps are split so that each successive ramp is treated separately. The FatiguePro stress output for each transient was used as guidance to split the transient up into sub-transients.

These splits defining the peaks and valleys are shown on Figure 9. Each transient was run separately, and another case was run with both transients combined (one cycle each).

18

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 Loss of Letdown silthDelayed Return to Service

- C00GNOZ S-S DatatrflO [12/1 7/2007]

Letdowu Flow 50% Decrease

--- SOýG0 Noz A

24 C

20 16 B

a! W0050 00-100

%0i*m:o 00.200 002500 003000 Date/fre- [12118120071 Figure 9. Approximate Points of Peaks and Valleys of Transients 19

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 4.6 Results of Charging Nozzle NB-3200 Analysis 4.6.1 Loss of Letdown with Delayed Return to Service (Fast Transient)

Table 1 shows the FatiguePro output for the Fast Transient 'Loss of Letdown with Delayed Return to Service.' Table 5 shows the VESLFAT output.

Table 5. VESLFAT Usage Results for Loss of Letdown with Delayed Return to Service Fatigue Analysis using VESLFAT Fatigue Module Version 1.42 - 12/29/2006

(&VeslFatFatlp42)

Page 19 02-19-2008 16:24:41 I Load Sets Cycles Sn Ke Salt Nallowed Usage 1 2 LOLD B 1 107839 3.333 261237 99.77 0.0100226 3 LOLD C 1 2 1 LOLDA 1 41165 1.000 28199 1.0001E+06 .0000010 4 LOLD D 1 Total Usage 0.0100236 In order to perform an accurate comparison between the two programs, the FatiguePro Peak Stress Range requires adjustment to convert it into the alternating stress intensity, Salt. Peak Stress Range listed in the FatiguePro results (Table 1) does not reflect the required correction for the simplified elastic-plastic correction factor, K,, or the required correction due to Elastic Modulus. (Both of these corrections are made internally by FatiguePro before looking up the number of allowable cycles on the fatigue curve.) The VESLFAT output Salt does reflect these corrections. Maximum alternating stress intensity, Salt, from FatiguePro, followed by a comparison from VESLFAT, is therefore:

Max Salt (FatiguePro) = (Peak Stress Range)/2"KEratio = 168.664/2"(3+1/3)'28.3/26.8 = 296.8 ksi Max Salt (VESLFAT) = 261.2 ksi The maximum range of linearized membrane plus bending stresses is:

Max SN (FatiguePro) = (Peak Stress Range).PPSMAX = 168.664.0.66 = 111.3 ksi Max SN (VESLFAT) = 107.8 ksi Max Ke (FatiguePro) = 3.333 Max K, (VESLFAT) = 3.333 20

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 Fatigue Usage (FatiguePro) = 0.0137672 Fatigue Usage (VESLFAT) = 0.0101688 FatigueProresults are conservative 4.6.2 Letdown Flow 50% Decrease (Slow Transient)

Table 2 shows the FatiguePro output for the slow transient 'Letdown Flow 50% Decrease Transient.' Table 6 shows the VESLFAT output.

Table 6. VESLFAT Usage Results for Letdown Flow 50% Decrease with Return to Normal Fatigue Analysis using VESLFAT Fatigue Module Version 1.42 - 12/29/2006

(&VeslFatFatlp42)

Page 20 02-19-2008 17:46:05 I Load Sets Cycles Sn Ke Salt Nallowed Usage 1 1 LFD A 1 13273 1.000 8958 1.4532E+10 .0000000 2 LFD B 1 2 3 LFD C 1 4452 1.000 3238 3.5574E+10 .0000000 4 ZERO 10 Total Usage = 0.0000000 Maximum alternating stress intensity, Sa1t, from FatiguePro, followed by a comparison from VESLFAT:

Max Sait (FatiguePro) = (Peak Stress Range)/2"Kg'Eratio = 17.7451/2"(1.0)'28.3/26.8 = 9.37 ksi Max Sat (VESLFAT) = 8.96 ksi The maximum range of linearized membrane plus bending stresses is:

Max SN (FatiguePro) = (Peak Stress Range).PPSMAX = 17.7451.0.66 = 11.71 ksi Max SN (VESLFAT) = 13.3 ksi For slow transients where K, is 1.0, the value of SN does not affect the fatigue usage results.

Max K, (FatiguePro) = 1.0 Max K, (VESLFAT) = 1.0 21

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 Fatigue Usage (FatiguePro) = ZERO Fatigue Usage (VESLFAT) = ZERO FatijgueProresults are conservative or the same 4.7 Charging Nozzle Conclusions A detailed fatigue analysis was performed for the Vogtle Charging Nozzle for two different transients using stress results from a 3-D FEM and using the specialized VESLFAT software, developed independently of FatiguePro and in strict conformance with ASME Section III, Subarticle NB-3200. These results were compared to those from running the same transients in FatiguePro, which utilizes uniaxial fatigue calculations, digitized Green's Functions based on an axisymmetric FEM, and Ordered Overall Range (OOR) cycle counting to compute fatigue [ 12].

The FatiguePro analysis results were conservative and in excellent agreement compared to the more refined NB-3200 VESLFAT analysis.

For the Loss of Letdown with Delayed Return to Service transient ('fast' transient with large, rapid changes in temperature, resulting in severe stresses and high fatigue usage), the maximum alternating stress intensity computed by FatiguePro was 13% higher than VESLFAT, and the fatigue usage was 35% higher.

For the Letdown Flow 50% Decrease transient ('slow' transient with smaller, slower changes in temperature, with a benign effect on fatigue), the maximum alternating stress intensity computed by FatiguePro was 4.2% higher than VESLFAT. In both the case of FatiguePro and VESLFAT, zero fatigue usage was computed for this transient, because the maximum stress range was below the fatigue threshold.

The detailed benchmarking confirmatory analysis described in this evaluation demonstrates that the fatigue usage computed for the Vogtle Charging Nozzle by the FatiguePro software is conservative relative to an ASME NB-3200 fatigue analysis.

22

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 5.0 RCS SURGE LINE HOT LEG NOZZLES The FatiguePro and NB-3200 fatigue analyses were performed at the FatiguePro-monitored location (safe end region), which was the location identified by Westinghouse to be the critical location in the RCS Surge Line Hot Leg Nozzle.

A discussion of the FatiguePro evaluation [13] and the NB-3200 evaluation [14] is presented in this section.

5.1 Transient Definitions The 'Cooldown and Heatup9 transient starts with the plant at NOP/NOT with a bubble in the pressurizer: RCS Cold Leg (CL) temperature at 557°F, pressurizer level constant at 40%,

pressurizer temperature at 653°F and flow in the reactor coolant loops. At 0400 on 5/16/04, the RCS loop and pressurizer are cooled down at 100°F per hour and 200'F per hour, respectively, down to Cold Leg temperatures of 120'F and a pressurizer temperature of 439°F. At 0400 on 5/17/04, the cold leg temperatures are heated up at 100°F per hour and at 06:13:48 on 5/17/04, the pressurizer is heated from 439°F at 100°F per hour. The temperatures of the RCS loop, pressurizer and surge line are shown in Figure 10.

The "MOP Heatup with Local Stratification" transient simulated for this benchmarking evaluation was developed using the guidance of the Modified Steam Bubble Method Heatup Methodology provided in WCAP 14950 [15]. The transient was developed to produce global and local loading conditions at the RCS Surge Line Hot Leg Nozzle location.

The 'MOP Heatup with Local Stratification' transient starts at 00:00 on 5/15/04 in Mode 5 with the plant at cold conditions with the RCS and pressurizer at 100 'F. A plant heatup is simulated at 01:00 with RCPs on, raising the RCS and pressurizer temperatures at the rate of 100°F/hour to 180°F at 01:48. At this time, the RCS temperature is maintained at 180'F while the pressurizer continues to heat up using pressurizer heaters at 100°F/hour to a temperature of 440'F at 04:24.

The RCS and pressurizer temperatures are held at these temperatures until 08:00. At 05:00, flow from the Loop 4 RCP is stopped and then at 05:40 is restarted. This cycling of the Loop 4 RCP causes a local, thermal stratification cycle at the RCS Surge Line Hot Leg Nozzle. At 08:00, the plant enters Mode 4 with the RCS temperature being raised from 180'F to 340'F and the pressurizer temperature increasing to 450'F at 09:36. The plant continues its heatup in Mode 3 to an RCS temperature of 547°F and a pressurizer temperature of 653°F at 15:22.

The temperatures of the RCS loop, pressurizer, surge line and the state of the Loop 4 RCP are shown in Figure 11. The pressures of the RCS loop and pressurizer are shown in Figure 12.

23

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 RCS Surge Line Hot Leg Nozzle U TE443A -+-- TE450 A TE453 700C 600 400-e161200 5,170:00 5$1712.00 Date/Time [05/16/2004 0511812004]

Figure 10: Cooldown and Heatup Transient RCS Surge Line Hot Leg Nozzle

- TE443A TE450 A TE453 - -RCP_4 700 600 500 i 400 X E- ,n -

Date/'ime [05/15/2004 -. 05116/2004]

24

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 Figure 11: MOP Heatup with Local Stratification Transient RCS and Pressurizer Pressure 2400 2000.

1000.

1200 ..... . .

Figure 12: FatiguePro Plant Pressures for MOP Heatup with Local Stratification Transient 5.2 FatiguePro Evaluation of RCS Surge Line Hot Leg Nozzle 5.2.1 Cooldown and Heatup TransientResults The FatiguePro-computed local fluid temperatures and computed stress history at the RCS Surge Line Hot Leg Nozzle for the Cooldown and Heatup transient developed by the FatiguePro analysis are shown in Figure 13.

The FatiguePro fatigue calculation developed a stress loading spectrum from the local peaks and valleys of the monitored stress history. For the simple simulation performed here, the stress history consisted of two stress pairs, derived from the cooldown and heatup moment ranges. The details are shown in Table 7. This report shows that the fatigue usage is determined by a maximum stress pair, with a (total) stress intensity range of 27.7 ksi. A corresponding K, factor of 1.0 is applied by FatiguePro along with adjustment of the stress range for Elastic Modulus ratio, giving a cumulative usage factor of 2.376 x 10-.

25

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 Hot Leg Nozzle Local Temperatures and Stress M-U-M-BOT +--"- I-MTOp

• Str,:

t _NOZZLE CZ to 1M:00:00 Date/Time [05116/2004 -- 05/17/2004]

Figure 13: FatiguePro Local Temperatures and Stresses for Cooldown and Heatup Transient 26

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 Table 7: FatiguePro Fatigue Details for Cooldown and Heatup Transient Detailed Fatigue Report (HLNOZ)

Ordered Overall Range Table Level (%) Cycles Usage 99.50 1 0.00000 98.00 0 0.00000 96.00 0 0.00000 93.50 0 0.00000 90.00 0 0.00000 86.00 0 0.00000 81.00 0 0.00000 75.50 0 0.00000 69.00 0 0.00000 61.50 0 0.00000 53.00 1 0.00000 44.00 0 0.00000 34.00 0 0.00000 24.00 0 0.00000 14.00 0 0.00000 6.00 0 0.00000 Usage from Spectrum 2.37619e-008 Removed Fatigue Usage 0.00000 Initial Fatigue Usage 0.00000 Total Fatigue Usage 2.37619e-008 Peak Stress Range 27.7035 Maximum Ke 1.0 Rapid Cycling Usage 0.00000 5.2.2 MOP Heatup with Local Stratification TransientResults The FatiguePro-computed local fluid temperatures at the RCS Surge Line Hot Leg Nozzle and the stress history for the MOP Heatup with Local Stratification transient are shown in Figure 14.

The FatiguePro fatigue calculation develops a stress loading spectrum from the local peaks and 27

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 valleys of the monitored stress history. For the simple simulation performed here, the stress history consists of two stress pairs, derived from the cooldown and heatup moment ranges.

The Fatigue Details (Table 8) for the MOP Heatup with Local Stratification transient shows that the fatigue usage is determined by two maximum stress pairs. The larger of the pairs has a (total) stress intensity range of 72.5 ksi, where a corresponding K, factor of 1.77 is applied by FatiguePro along with adjustment of the stress range for Elastic Modulus ratio, giving a cumulative usage factor of 0.00012. The second stress pair producing an incremental cumulative usage factor of essentially zero.

RCS Hot Leg Nozzle Temperatures and Stress u -i-_T LI i HL-BOT A -Ltr:* N07--L_

U-cf)

E Date;'Tirne [05I15120O4]

Figure 14: FatiguePro Stresses for MOP Heatup with Local Stratification Transient 28

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 Table 8: FatiguePro Fatigue Details for MOP Heatup with Local Stratification Transient Detailed Fatigue Report (HLNOZ)

Ordered Overall Range Table Level (%) Cycles Usage 99.50 1 0.00012 98.00 0 0.00000 96.00 0 0.00000 93.50 0 0.00000 90.00 0 0.00000 86.00 0 0.00000 81.00 0 0.00000 75.50 0 0.00000 69.00 0 0.00000 61.50 0 0.00000 53.00 1 0.00000 44.00 0 0.00000 34.00 0 0.00000 24.00 0 0.00000 14.00 0 0.00000 6.00 0 0.00000 Usage from Spectrum 0.00012 Removed Fatigue Usage 0.00000 Initial Fatigue Usage 0.00000 Total Fatigue Usage 0.00012 Peak Stress Range 72.4919 Maximum Ke 1.76634 Rapid Cycling Usage 0.00000 29

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 5.3 Assumptions and Design Inputs for NB-3200 Evaluation for RCS Surge Line Hot Leg Nozzle 5.3.1 Assumptions

" Per [ 16], material properties in the FEM were taken at 290'F for consistency with the original finite element analysis performed by Westinghouse and documented in WCAP 14173.

" The thermal sleeve and water gap were included in the FEM for the thermal analysis, but removed in the stress analysis. Neither was specifically modeled in the Westinghouse finite element analysis.

" The FEM was created such that the structure is at a zero stress state at an ambient temperature of 70'F.

5.3.2 ASME Code Edition The analysis was performed in a manner consistent with the NB-3200 [1] fatigue analysis. The 1986 Edition was used, unless otherwise noted.

5.3.3 Heat Transfer Coefficientsfor Transients The heat transfer coefficients were provided by Westinghouse [ 17] to cover all cases in the monitoring.

In the FEM for the NB-3200 analysis, the same film coefficients were applied to the same convection surfaces.

5.3.4 Geometry and MaterialProperties Material designations for the RCS Surge Line Hot Leg Nozzle were provided by Westinghouse.

The same material designations were used in FatiguePro and the NB-3200 analysis.

30

• N L-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 5.3.5 Finite Element Model The ANSYS program is used to perform the finite element analysis. A 3-D FEM of the nozzle was developed by Westinghouse using their special purpose "WECAN" software as described in the WCAP 14173 [16].

A 3-D model was developed in a more current version of ANSYS using brick elements. The Structural Integrity Associates FEM differs from the Westinghouse FEM in that the mesh density is higher, the individual materials were modeled (Westinghouse modeled the entire structure as Type 316), the counter-bore between the nozzle and piping was modeled, and the thermal sleeve and water gap were modeled and included in the thermal portion of the analysis. A half symmetric model was generated to be capable of calculating stresses due to both symmetric and asymmetric loading (thermal, stratification, pressure). The model is shown on Figure 15.

31

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 Figure 15. Structural Integrity Associates ANSYS FEM 32

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 For a consistent comparison, stresses were extracted and linearized at the safe-end location monitored by FatiguePro.

5.4 Loading Calculations for NB-3200 Evaluation of RCS Surge Line Hot Leg Nozzle 5.4.1 PipingInterface Loads In FatiguePro the stresses due to piping are based on an NB-3600 approach that adds maximum bending stresses from the outside surface of the pipe to the transient stresses on the inside of the nozzle.

In this NB-3200 analysis, the individual components of stress are calculated taking into account through-wall distribution.

Referring to Figure 7, and using the cylindrical coordinate system shown in the upper right of the figure, from general structural mechanics, the membrane plus bending stresses at the outside surface of a thick-walled cylinder are:

(T, = Axial stress due to bending moments = [Mb'sin(O) - Mc-cos(0)]'(OD/2)/I c50z = Circumferential shear stress (in 0 direction) due to torsion = Ma(OD/2)/J Where for a 16" schedule 160 nominal pipe/nozzle size:

ID = Inside diameter = 12.812 in (nominal ID)

OD = Outside diameter = 16 in (nominal OD) 4 4 4 I = Moment of inertia = (7T/64)(OD - ID )4 = 1894 in J = Polar moment of inertia = 21 = 3789 in Piping axial and shear forces were not used in the WCAP 14173 analysis, and so were not provided in the design input. They are judged to have negligible effects on the alternating stress intensity of a fatigue contributing transient, and the approach is consistent with NB-3600.

The maximum alternating piping stress location is likely influenced chiefly by Heatup and Cooldown transients for the following reasons.

• The Vogtle plant operates with Modified Operating Procedure, which ensures constant outflow in the surge line, thus little chance of stratification.

0 According to WCAP 14173, empirical data suggest that "local stratification in the nozzle is significant only when the hot leg flow is stopped."

33

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 Also according to WCAP 14173, turbulent penetration from the hot leg flow causes the hot leg surge nozzle to be at the hot leg temperature, unless the surge line outflow is greater than 1505 gpm. Surge line flow greater than 1505 gpm is an unlikely condition, as it exceeds the maximum spray flow rate of 557 gpm per valve when fully open.

5.5 Stress and Fatigue Calculations for RCS Surge Line Hot Leg Nozzle 5.5.1 Summary of FatigueAnalysis Performed Thermal transient stress components were calculated in ANSYS. The transients considered are shown on Figures 10 through 12. A unit internal pressure load was calculated by ANSYS.

The fatigue usage analysis considered all six stress components and was performed using the NB-3200 rules of Section III of the ASME Code.

5.5.2 Finite Element Analysis Using the loading history specified in the FatiguePro evaluation, analysis of the 'Cooldown and Heatup' and 'MOP Heatup with Local Stratification' transients were performed using the 3-D ANSYS FEM of the nozzle. On the first time step, the model was brought to steady state conditions to establish baseline temperature distributions, and thus baseline steady state stresses for the fatigue analysis.

Stress analysis was performed using the temperature distributions calculated in the thermal transient ANSYS analysis as input. The water elements in the annulus and the thermal sleeve were removed. In addition, the edges of the hot leg piping and the surge line piping were coupled in the longitudinal direction to prevent local distortion of the planes of the cross sections.

5.5.3 Stress Calculations Linearized stress components at the safe end outside surface were used for the fatigue usage analysis. This location was selected, because for the thermal analyses it had the highest stress intensity on the outside surface of the top of the safe end region, which was identified as the critical fatigue location in the WCAP 14173. The critical fatigue section is located on the nozzle side of the weld above the cou'nter-bore. Stresses were extracted in a cylindrical coordinate system to be consistent with the piping load stress equations. In the ANSYS output in this 34

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 coordinate system, SX is radial stress, SY is theta (or hoop) stress, and SZ is axial stress. The stress components from the thermal analysis were combined with stress components due to internal pressure and piping loads. Piping load .stress components were computed based on the time-history of the individual moment terms from the FatiguePro analysis and using closed form solutions multiplied by an SCF. Pressure stress components were computed based on multiplying the time-history of the internal pressure by the stresses at the critical section due to the unit internal pressure load case. Thermal transient stresses were then combined with the piping load and pressure stresses for both membrane plus bending (P+Q) and the total (primary plus secondary plus peak) stress components (P+Q+F). The stresses for each of the six stress tensors were added together on an individual basis, representing the stress state at each time step.

5.5.4 Fatigue Usage Analysis In the VESLFAT program, transients that consist of both upward and downward temperature and pressure ramps are split so that each successive ramp is treated separately. The FatiguePro stress output for each transient was used as guidance to split the transient up into sub-transients. These splits defining the peaks and valleys for the second transient are shown on Figure 16.

35

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 Heatup Transient with Local Stratification 50000 600 40000 .

- 500 30000" I-_400 20000 ,

0.D w CL 10000- E z-- 200 0

0.00 1+00 11.00E+04 2.00 04 3.OOE+04 4.OOE+04 5.OOE+04 6.OOE+04 7.00E+04 8.OOE+04 9.OOE+04 1.00 +05 100

-10000 B

-20000 0 Time (seconds)

[I-SZ - Ttn Figure 16. Approximate Points of Peaks and Valleys of Transients for RCS Surge Line Hot Leg Nozzle 36

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 5.6 Results of RCS Surge Line Hot Leg Nozzle NB-3200 Analysis 5.6.1 Cooldown and Heatup Table 7 shows the FatiguePro output for the 'Cooldown and Heatup' transient. Table 9 shows the VESLFAT output.

Table 9. VESLFAT Usage Results for Cooldown and Heatup Transient Fatigue Analysis using VESLFAT Fatigue Module Version 1.42 - 12/29/2006

(&VeslFatFatlp42)

Page 10 02-19-2008 15:35:11 I Load Sets Cycles Sn Ke Salt Nallowed Usage 1 1 CD 1 10511 1.000 8776 1.4797E+10 .0000000 2 HU 1 Total Usage = 0.0000000 In order to perform an accurate comparison between the two programs, the FatiguePro Peak Stress Range requires adjustment to convert it into Salt. Peak Stress Range listed above in the FatiguePro results does not reflect the required correction for the simplified elastic-plastic correction factor, Ke, or the required correction due to Elastic Modulus. (Both of these corrections are made internally by FatiguePro before looking up the number of allowable cycles on the fatigue curve.) The VESLFAT output Salt does reflect these corrections. Maximum alternating stress intensity, Salt, from FatiguePro, followed by a comparison from VESLFAT, is therefore:

Max Salt (FatiguePro) = (Peak Stress Range)/2"KEratio = 27.7035/2"1.0"28.3/27.,1 = 14.5 ksi Max Salt (VESLFAT) = 8.8 ksi The maximum range of linearized membrane plus bending stresses is:

37

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 Max SN (FatiguePro) = (Peak Stress Range).PPSMAX = 27.7035-0.85 = 23.5 ksi Max SN (VESLFAT) = 10.5 ksi Max Ke (FatiguePro) = 1.0 Max Ke (VESLFAT) = 1.0 Fatigue Usage (FatiguePro) = 2.37619e-008 Fatigue Usage (VESLFAT) = ZERO FatigueProresults are conservative 5.6.2 MOP Heatup with Local Stratification Table 8 shows the FatiguePro output for the 'MOP Heatup with Local Stratification transient.'

Table 10 shows the VESLFAT output.

Table 10. VESLFAT Usage Results for MOP Heatup with Local Stratification Transient Fatigue Analysis using VESLFAT Fatigue Module Version 1.42 - 12/29/2006

(&VeslFatFatlp42)

Page 13 02-19-2008 14:37:52 I Load Sets Cycles Sn Ke Salt Nallowed Usage 1 2 STRAT A 1 47081 1.000 29184 777974 0.0000013 3 STRAT B 1 2 1 ZERO 10 21165 1.000 18651 4.7187E+06 .0000002 4 STRAT C 1 Total Usage = 0.0000015 Maximum alternating stress intensity, Salt, from FatiguePro, followed by a comparison from VESLFAT:

Max Salt (FatiguePro) = (Peak Stress Range)/2"K&'Eratio = 72.4919/2"1.76634-28.3/27.1 = 66.9 ksi 38

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 Max Salt (VESLFAT) = 29.2 ksi The maximum range of linearized membrane plus bending stresses is:

Max SN (FatiguePro) = (Peak Stress Range).PPSMAX = 72.4919.0.85 = 61.6 ksi Max SN (VESLFAT) = 47.1 ksi Fatigue Usage (FatiguePro) = 0.000119668 Fatigue Usage (VESLFAT) = 0.0000015 FatigueProresults are conservative 5.7 RCS Surge Line Hot Leg Nozzle Conclusions A detailed fatigue analysis was performed for the Vogtle RCS Surge Line Hot Leg Nozzle using stress results from a 3-D FEM and using the specialized VESLFAT software, developed independently of FatiguePro and in strict conformance with ASME Section III, Subarticle NB-3200. These results were compared to those running the same transients in FatiguePro, which utilizes uniaxial fatigue calculations, digitized Green's Functions, and Ordered Overall Range (OOR) cycle counting to compute fatigue [12].

The FatiguePro analysis results were conservative compared to the more refined NB-3200 VESLFAT analysis.

For the Cooldown and Heatup transient, the maximum alternating stress intensity computed by FatiguePro was 49.5% higher than VESLFAT, and the fatigue usage was below the fatigue threshold (zero) for both FatiguePro and VESLFAT.

For the MOP Heatup with Local Stratification transient (includes asymmetric loading), the maximum alternating stress intensity computed by FatiguePro was 129% higher than VESLFAT.

The principal source of conservatism, in the FatiguePro analysis was the much higher bending stresses computed in FatiguePro, compared to the alternate calculation used in the NB-3200 VESLFAT analysis. Other factors included the uniaxial fatigue methodology, and a much higher assumed ratio of the membrane plus bending to the total stresses in FatiguePro. This higher ratio computed a conservative value of the simplified elastic-plastic factor Ke.

The detailed benchmarking confirmatory analysis described in this evaluation demonstrates that the fatigue usage computed for the Vogtle RCS Surge Line Hot Leg Nozzle by the FatiguePro software is conservative relative to an ASME NB-3200 fatigue analysis.

39

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2

6.0 REFERENCES

1. American Society of Mechanical Engineers, Boiler and Pressure Vessel Code,Section III, Subarticle NB, 1986 Edition.

2. FatiguePro Generic Software, Version 3.01.00-03232, SI File No. EPRI-136Q-403.

3. FatiguePro Plant-specific Software for Vogtle, Version (Vogtle) 3.01.03-05347, SI File No. FP-VOG-503.

4. ANSYS, Release 8.1 (w/Service Pack 1), ANSYS, Inc., April 2004.

5. VESLFAT, Version 1.42, 02/06/07, Structural Integrity Associates, Inc.

6. SI Report No. SIR-95-073, Revision 2, "Transfer Function and System Logic Report, Transient and Fatigue Monitoring System for Vogtle Electric Generating Station", June 2007, SI File No. FP-VOG-403.

7. Structural Integrity Calculation Package: "Transient Definitions for Benchmarking Study for Charging Nozzles Monitored by Vogtle-1/2 FatiguePro", Rev. 0, February 2008, SI File VOG-07Q-301.

8. Structural Integrity Calculation Package: "Benchmarking Study for Charging Nozzles Monitored by Vogtle-1/2 FatiguePro", Rev. 0, February 2008, SI File VOG-07Q-302.

9. SI Calculation, "Charging Nozzle 2-D Model", Revision 0, December 1994, SI File No.

GPCO-23Q-30 1.

10. SI Calculation Package: "Charging Nozzle Green's Function," Rev. 0, 1994, SI File GPCO-23Q-302.

11. "Moment Stress Input for Charging Nozzles," SI File No. GPCO-23Q-261 (Westinghouse PROPRIETARY).

12. A.Y. Kuo, S.S. Tang, P.C. Riccardella, "An On-Line Fatigue Monitoring System for Power Plants: Part I - Direct Calculation of Transient Peak Stress Through Transfer Function Matrices and Green's Functions"; "An On-Line Fatigue Monitoring System for Power Plants Part II - Development of a Personal Computer Based System for Fatigue Monitoring," Design and Analysis Methods for Plant Life Assessment - PVP Vol. 112,

13. Structural Integrity Calculation Package: "Transient Definitions for Benchmarking Study of RCS Surge Line Hot Leg Nozzle Monitored by Vogtle-1/2 FatiguePro", Rev. 0, February 2008, SI File VOG-07Q-303.

40

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2

14. Structural Integrity Calculation Package: "Benchmarking Study for Pressurizer Surge Line Hot Leg Nozzle Monitored by Vogtle-1/2 FatiguePro", Rev. 0, February 2008, SI File VOG-07Q-304.

15. Westinghouse WCAP-14950, "Mitigation and Evaluation of Pressurizer Insurge/Outsurge Transients," July 1997, SI File No. VOG-07Q-202P.

16. Westinghouse WCAP 14173 - GTL and Stress Transfer Functions, "Global to Local Transformations and Stress Transfer Functions for Pressurizer Surge Line, Pressurizer Lower Head and Pressurizer Spray Line," Revision 3, November 1996, SI File No.

GPCO-23Q-223P.

17. Letter dated 1/4/2008 to T.E. Tynan (SNOC) from E.C. Arnold (Westinghouse), "Vogtle Surge Line Hot Leg Nozzle Model Inputs", SI File VOG-07Q-201P.

41

NL-08-0228 Response to RAI 4.3-4 Attachment A FatiguePro Benchmarking Study for Vogtle-1/2 ATTACHMENT 1 DESCRIPTION OF VESLFAT The Structural Integrity VESLFAT software program was used to perform the fatigue usage analysis in accordance with the fatigue usage portion of NB-3200. VESLFAT performs the analysis required by NB-3222.4(e) for Service Levels A and B conditions defined by the user.

The VESLFAT program computes the primary plus secondary and total stress ranges for all events and performs a correction for elastic-plastic analysis, if appropriate.

The program computes the stress intensity range based on the stress component ranges for all event pairs. The program evaluates the stress ranges for primary plus secondary and primary plus secondary plus peak stress based upon six components of stress (3 direct and 3 shear stresses). If the primary plus secondary stress intensity range is greater than 3Si, then the total stress range is increased by the factor K,, as described in NB-3228.5. The value of Sm is specified as a function of temperature. The input maximum temperature for both states of a load set pair is used to determine the temperature upon which S,, is determined from the user-defined values.

When more than one load set is defined for either of the event pair loadings, the stress differences are determined for all of the potential loadings, saving the maximum for the event pair, based on the pair producing the largest alternating total stress intensity (Salt), including the effects of K, The principal stresses for the stress ranges are determined by solving for the roots of the cubic equation:

s3- ((x +(Yy+(z)S2 +((Y (Y+ Gy Y z7x- T'y2 - *xz2 _yz2)s

- ((Yx Gy (z +r 2 T'xy *'xz "Cyz - (Yz "Txy2 _ GYy Tlxz2 _ (Yx Tlyz 2 ) "- 0 The stress intensities for the event pairs are reordered in decreasing order of Salt, including a correction for the ratio of modulus of elasticity (E) from the fatigue curve divided by E from the analysis. This allows a fatigue table to be created to eliminate the number of cycles available for each of the events of an event pair, allowing determination of fatigue usage per NB-3222.4(e).

For each load set pair in the fatigue table, the allowable number of cycles is determined based on Salt.

42

Vogtle Electric Generating Plant Enclosure VEGP License Renewal RAI Responses - January 22, 2008

Enclosure NL-08-0228 Vogtle License Renewal RAI Responses - 01/22/2008 RAI - 4.2.3-1 We have reviewed the information on page 4.2-10 of the Application for License Renewal and have noted a number of errors, the most significant of which is the numerical temperature value (-226 F) for the RTPTS for the Inlet Nozzle to Nozzle Shell Course Weld 105-121 D shown in Table 4.2.3-1. This value appears to be inconsistent with the other data provided for the material. In addition, notes B and E to this Table 4.2.3-1 appear to be incorrect and, in particular inconsistent with Note F to the same table. Please correct the information on page 4.2-10 and resubmit this page.

VEGP Response:

The -226 value on page 4.2-10 of the VEGP LRA should be -22. Notes B and E on page 4.2-10 should be the same as Notes B and E on page 4.2-12. The LRA will be amended to correct these errors.

A License Renewal Application amendment is required.

Page 1 of 11

Enclosure NL-08-0228 Vogtle License Renewal RAI Responses - 01/22/2008 RAI - 4.3-1 LRA section 4.3.1 states, "In addition to the original design transients, fatigue loading transients and issues subsequently identified are not parts of the original fatigue analyses. For the lower pressurizer head and surge line, thermal stratification and insurge/outsurge transients are evaluated (IEB 88-11)."

a.) The evaluation result for the pressurizer lower head was not discussed in LRA section 4.3.1.4, "Thermal Stratification of the Surge Line and Lower Pressurizer Head (IEB 88-11)." Please provide the limiting 60-year projected CUF value for the pressurizer lower head.

b.) An NRC safety evaluation entitled "Vogtle Unit 1 Safety Evaluation on Pressurizer Surge Line Thermal Stratification," dated April 12, 1990, states, "Applicant committed to revise applicable operating procedures to limit the system delta T (between the pressurizer head and the reactor coolant loop) for reactor coolant system (RCS) heatup (HU) to 320'F and RCS cool down (CD) to 300 0 F. The revised heatup and cooldown procedures ensure consistency between actual plant operation and the surge line analysis assumption."

Please discuss the procedures that have been used by VEGP and demonstrate the consistency between the recorded plant operational transient data and the assumptions (delta T limits of 320°F &

300°F during HU/CD) that were made and used in the surge line and pressurizer lower head thermal stratification analyses.

VEGP Response:

a.) The limiting 60-year projected CUF value for the pressurizer lower head is 0.00017 at the Unit 2 heater penetration.

b.) Vogtle procedure 12006-C, "Unit Cooldown to Cold Shutdown," Rev. 16, approved April 12, 1990 added a Caution before the step where the cooldown from normal operating temperature to 375 0 F is started. The caution is repeated before resuming the cooldown to 225 0 F and again before resuming the cooldown to less than 130 0 F.

The Caution states, "To reduce thermal stratification in the Pressurizer Surge Line maintain the Delta-T between the RCS and the Pressurizer Liquid Space as low as practical. The Delta-T of 300°F should not be exceeded."

Vogtle procedure 12001-C, "Unit Heatup to Hot Shutdown," Rev. 18, dated April 26, 1990 included a Caution prior to commencing the heatup. The Caution states, "To reduce thermal stratification in the pressurizer surge line maintain the Delta-T between the RCS and the pressurizer steam space as low as practical. The Delta-T of 320°F should not be exceeded."

Heatups and cooldowns with FatiguePro data readily available were reviewed. FatiguePro data was available from 1/1/97 through 02/18/08 for Unit 1 and from 6/30/95 through 02/18/08 for Unit 2.

Between the two units, 30 cooldowns and 32 heatups were reviewed. In general, the maximum Delta-T for each heatup and cooldown was in the 240°F to 260'F range. No heatups reached a AT of 320'F and no cooldowns reached a AT of 3000F.

Page 2 of 11

Enclosure NL-08-0228 Vogtle License Renewal RAI Responses - 01/22/2008 RAI - 4.3-2 Section 4.3.1.5.3 of the LRA includes the following assessment for the environmentally-assisted fatigue analysis for the surge line hot leg nozzle:

"The maximum design CUF determined from the ASME Code fatigue analysis for the VEGP surge line hot leg nozzle is 0.95. Applying the maximum Fen for stainless steel from NUREG/CR-5704 (Ref. 10) of 15.35 increases the maximum CUF value to an environmental fatigue adjusted value of 14.58, which is greater than 1.0. Therefore, a different demonstration method is used.

Using fatigue monitoring software, cooldown/heatup cycles for Unit 1 and Unit 2 from 6/30/95 through 10/9/05 were analyzed to determine the average CUF per HU/CD cycle. Using this data, SNC has shown that at the surge line hot leg nozzle the projected CUF for 200 HU/CD cycles is 0.00534 for Unit 1 and 0.00628 for Unit 2. "

Please discuss the changes in the heatup and cooldown procedures, specifically the implementation of the modified operating procedure (MOP), to mitigate pressurizer insurge/outsurge transients. Please explain how the impacts of MOP were factored into the calculation of the average CUF per HU/CD in the EAF analysis.

VEGP Response:

SNC did not specifically account for pre-MOP operation in the LRA because continuous outflow was in place from the beginning of plant operation and the other aspects of MOP were established at VEGP early in plant life. However, since the LRA was submitted, SNC has reconsidered this position.

Per WCAP-14950 (Page 2-6), the overall strategies to mitigate insurge/outsurge transients are:

• Maintain continuous pressurizer outsurge flowrduring heatup and cooldown operations.

" Minimize the system AT.

Farley and Vogtle were lead plants for proving the value of MOP as described in WCAP-14950. One of the reasons that Farley and Vogtle were lead plants for demonstrating the value of the modified operating procedures was that they were already using those procedures. Farley actually began operating during the first Unit 1 cycle with a pressurizer backup heater on in manual. This results in continuous pressurizer spray flow and continuous outflow from the pressurizer. Farley did this to equalize boron concentration between the pressurizer and the RCS. Minimizing surge line stratification was an unanticipated benefit. Early Vogtle-plant operators trained at Farley and discussions with them indicate that Vogtle has operated with a continuous outflow during heatups and cooldowns since initial operation.

VEGP procedure, 12001-1, "Unit Heatup to Hot Shutdown", Rev. 0, included the following precaution:

• The boron concentration in the pressurizer should not be different from the RCS by more than 50 ppm. Pressurizer backup'heaters may be energized as necessary to equalize the boron concentration.

VEGP procedure, 12002-1, "Unit Heatup to Normal Operating Temperature and Pressure", Rev. 0, included the following precaution prior to commencing the heatup:

Page 3 of 11

Enclosure NL-08-0228 Vogtle License Renewal RAI Responses - 01/22/2008

• It is recommended that the RCS temperature be maintained between 75°F and 125°F less than Pressurizer temperature.

VEGP procedure, 12006-1, "Unit Cooldown to Cold Shutdown", Rev. 0, included the following precautions:

• The boron concentration in the pressurizer should not be different from the RCS by more than 50 ppm. Pressurizer backup heaters may be energized as necessary to equalize the boron concentration.

" It is recommended that the RCS temperature be maintained between 75 0 F and 125 0 F less than Pressurizer temperature.

The above procedure steps demonstrate that VEGP has always operated with consideration of the overall strategies to mitigate insurge/outsurge transients.

The next significant change to VEGP heatup and cooldown procedures that was made to mitigate thermal stratification of the surge line was in 1990, a precaution was added to minimize the AT between the RCS hot leg and the pressurizer and to not let it exceed 320°F for heatups (300°F for cooldowns).

However, other than adding the steps to minimize the AT, no other change was made to the heatup and cooldown procedures at that time.

WCAP-1 4950 also describes the difference in the various heatup and cooldown methods. The primary difference between the standard steam bubble method and the modified steam bubble method is the modified steam bubble procedure starts at least one RCP before drawing a bubble in the pressurizer and leaves at least one RCP running until after the bubble is collapsed. The standard steam bubble method does not start any RCPs until after the pressurizer bubble is drawn and shuts down all RCPs before the bubble is collapsed. The modified steam bubble method results in the maximum system AT being 30°F to 40°F less than when using the standard steam bubble method.

WCAP-14950 Tables 5-15 and 5-16 compared the CUF for various locations during 200 heatups and cooldowns assuming that all the heatups and cooldowns were performed using either the standard steam bubble method or the modified steam bubble method. For fabricated 14s1 60 components, such as VEGP has, the bounding location on the pressurizer lower head is the heater penetration and the bounding location for the surge nozzle is the surge nozzle knuckle. For the heater penetration, the fatigue using the standard steam bubble method was 1.84 times the fatigue for the same location using the modified steam bubble method. For the surge nozzle knuckle, the fatigue using the standard steam bubble method was 1.88 times the fatigue for the same location using the modified steam bubble method.

VEGP now uses the modified steam bubble method. The original heatup and cooldown procedures allowed either the standard steam bubble method or the modified steam bubble method. When a first cycle shift supervisor was interviewed, he said that he always used the modified steam bubble method.

However, SNC has decided that although the operating procedures at VEGP included steps to mitigate thermal stratification from the beginning of operations, those steps may not have been 100% effective.

SNC has found instances both before and after FatiguePro was implemented where an insurge took place. We recognize that the percentage of such events before FatiguePro was implemented may have been higher than it has been since FatiguePro was implemented.

Therefore, SNC will adjust the initial fatigue (as of 1/1/1998 for Unit 1 and 6/30/1995 for Unit 2) to Page 4 of 11

Enclosure NL-08-0228 Vogtle License Renewal RAI Responses - 01/22/2008 increase it by a factor of 2 to account for potentially higher thermal stratification during that period.

Data collection for WCAP-1 4950 began on October 11, 1994, so by then the VEGP heatup and cooldown procedures would have been changed to require the modified steam bubble method. In any case, the increase in initial CUF includes heatups and cooldowns prior to FatiguePro data and heatups or cooldowns with FatiguePro data account for thermal stratification, regardless of what method is used for the heatup or cooldown. Therefore, establishing the actual date of the change is not necessary.

Prior to 1/1/98, Unit 1 had 23 RCS cooldowns and 24 RCS heatups. Prior to 6/30/1995, Unit 2 had 13 RCS cooldowns andl4 RCS heatups. The table below shows the current initial CUF and the CUF Increment since FatiguePro was implemented from FP-VOG-315, Rev. 1, "Baseline Evaluation and 60-Year Projection of Vogtle Plant Cycles, Fatigue Usage, and Environmental Usage. The table below then gives the revised initial CUF, the revised ending CUF, and the revised 60-year CUF Projection.

The CUF acceptance criteria for all 3 locations on each unit is a CUF less than 0.06515 (1/15.35 =

0.06515), which accounts for environmental effects.

Location Current Revised CUF Revised Revised 60-Initial Initial CUF Increment 10/9/2005 Yr CUF CUF CUF Projection Unit 1 Heater Penetration 0.00004 0.00008 0.00001 0.00009 0.00015 Unit 1 Surge Nozzle 0.00001 0.00002 0.00000 0.00002 0.00004 Unit 1 Hot Leg Surge Nozzle 0.00187 0.00374 0.00051 0.00425 0.00721 Unit 2 Heater Penetration 0.00003 0.00006 0.00003 0.00009 0.00020 Unit 2 Surge Nozzle 0.00001 0.00001 0.00002 0.00003 0.00005 Unit 2 Hot Leg Surge Nozzle 0.00111 0.00222 0.00094 0.00316 0.00739 This response results in a new commitment to revise the FatiguePro initial CUF values for the Unit 1 and Unit 2 hot leg surge nozzles, pressurizer surge nozzles, and pressurizer heater penetrations to double the current values and recalculate the current and projected CUFs. This change to FatiguePro will be implemented no later than two years prior to entering the period of extended operation.

This response requires an LRA amendment to correct the hot leg surge nozzle CUF values and EAF values in Sections 4.3.1.4, 4.3.1.5.3, and Table 4.3.1-3.

Page 5 of 11

Enclosure NL-08-0228 Vogtle License Renewal RAI Responses - 01/22/2008 RAI - 4.3-3 LRA sections 4.3.1.5.4 and 4.3.1.5.5 states the average Fen for charging nozzle and safety injection nozzle are 7.6 and 5.535, respectively. In response to the audit questions 4.3-5, VEGP stated that Fen values for normal charging, alternative charging and safety injection nozzles were computed from the actual plant events using an integrated strain rate (ISR) method as defined in the EPRI Report TR-1003083, Guidelines for Assessing Fatigue Environmental Effects in a License Renewal Application (MRP-47 Revision 1). ISR method calculates one Fen for one transient pair. Both charging nozzle and safety injection nozzle were designed to several thermal transients. Please justify how one average Fen value per nozzle could be used for more than one transient pairs having significant contribution to the CUF.

VEGP Response:

In the following discussion, PERIOD 1 refers to the period before FatiguePro data was available and PERIOD 2 refers to the period with available FatiguePro data. SNC's response to Audit Question 4.3-02, part c explained how an average CUF per event was established using PERIOD 2 data and how the CUF of the charging nozzles was determined for PERIOD 1.

Charging Nozzles Seven events from PERIOD 2 that were considered to be representative were selected. For each representative event, an integrated strain rate (ISR) Fen was calculated using the methodology described in SNC's response to Audit Question 4.3-05. The ISR Fen from three Charging and Letdown Shutoff events was averaged to determine the Fen value to be applied to each Charging and Letdown event. There were no Charging Shutdown Delayed Return events during PERIOD 2; however, these events were judged to be essentially the same as Charging and Letdown Shutdown events, so the same CUF and the same Fen are used for both types of event. There were only three Letdown Shutdown Delayed Return events, so the ISR Fen from all three were averaged to determine the Fen value to be applied to each Letdown Shutoff Delayed Return event. The Letdown Shutoff Prompt Return event with the highest incremental CUF was evaluated to determine the Fen to be applied to all such events. Table 1 below shows the seven events, the incremental CUF calculated for the event, the Fen calculated for that event, the incremental EAF calculated for each event, and the average Fen calculated for the type of event.

Table 1: Vogtle Environmental Fatigue Calculation for Charging Nozzle Event Incremental Fen Incremental Average CUF EAF Fen Charging and ousLetdown (Fr 1 Shutoff Unit 2 rnsen)1.10E-05 7.18 7.87E-05 (For < 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> transient)

Charging and Letdown Shutoff Unit 1 4.05E-04 6.70 2.37E-06 (For < 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> transient).

Charging and Letdown Shutoff Unit 1 (For > 27 hour3.125e-4 days <br />0.0075 hours <br />4.464286e-5 weeks <br />1.02735e-5 months <br /> transient - worst case)

Letdown Shutoff with Delayed Return to 5.31 E-03 6.71 3.56E-02 Service (Unit 2), a Letdown Shutoff with Delayed Return to 4.49E-03 7.14 3.20E-02 7.32 Service (Unit 2), b Letdown Shutoff with Delayed Return to 4.28E-06 8.13 Service (Unit 2), c 3.48E_05 Page 6 of 11

Enclosure NL-08-0228 Vogtle License Renewal RAI Responses - 01/22/2008 Event Incremental Fen Incremental Average CUF EAF Fen Letdown Flow Shutoff with Prompt 5.77E-07 8.44 4.87E-06 8.44 Return to Service (Unit 2 example)

Where Incremental EAF = Incremental CUF x Fen Table 2 below shows the event types used, the average CUF to be applied to each PERIOD 1 event of that type (UEVENT), and the Fen to be applied to all PERIOD 1 and PERIOD 2 events of that type.

Table 2. Average Charging Nozzle Incremental Fatigue Usage and Fen Values for Charging and Letdown Events Event Uevent Fen Charging and Letdown Shutoff 2.04E-04 6.57 Letdown Trip Delayed (average of two worst transients) 2.25E-03 7.63 Letdown Trip Delayed (Reduced Temperature Modes 3 and 4 only) 4.98E-06 7.63 Letdown Trip Prompt 9.27E-08 8.44 Charging Trip Delayed (assumed same as Charging and Letdown Shutoff) 2.04E-04 6.57 Charging Trip Prompt 1.59E-08 15.35 For each of the events in PERIOD 1, average incremental fatigue usages and Fen values from Table 2 were used to calculate the CUF. For PERIOD 2, the actual incremental fatigue computed in FatiguePro was used. The effective Fen was computed for each nozzle for the entire operating period. The lowest value calculated was 7.56 and the highest was 7.59. Therefore, 7.6 is considered to conservatively represent an effective value. In FatiguePro the allowable fatigue usage for the charging nozzles is then 1/7.6 = 0.1316.

Therefore, an ISR Fen value is calculated for each transient pair having a significant contribution to the CUF of the charging nozzles. From startup through 10/9/05, Unit 1 had a total of 26 events on the Normal nozzle and 48 on the Alternate nozzle. For the same period, Unit 2 had 13 events on the Normal nozzle and 63 events on the Alternate nozzle. The fact that the average Fen calculated for each nozzle was nearly identical, despite a significant difference in the number of events for the various nozzles confirms the validity of using the highest of those averages for future events.

Safety Injection Nozzles As stated in the response to Audit Questions 4.3-06 and 4.3-11, the Vogtle FatiguePro software monitors the safety injection nozzle as a cycle-based fatigue (CBF) location and the fatigue accumulation for the safety injection nozzle was shown to be based on the counted safety injection events. Also, the Fen to be used for VEGP safety injection events was calculated using a simulated event at a sister plant with a similar design and a stress-based fatigue (SBF) module for their safety injection nozzle. The Fen was calculated using the integrated strain rate method described in the response to Audit Question 4.3-05, yielding an overall Fen of 5.535.

Since additional thermal and pressure transients such as heatup and cooldown do not cause measurable additional fatigue usage to the safety injection location, the overall Fen for safety injection events is the only one needed for safety injection nozzles. Therefore, in FatiguePro the allowable fatigue usage for the safety injection nozzles is then 1/5.535 = 0.1807.

Page 7 of 11

Enclosure NL-08-0228 Vogtle License Renewal RAI Responses - 01/22/2008 RAI - 4.3-4 LRA section 4.3.1.5 stated that the environmentally-assisted fatigue on surge line hot leg nozzle, and charging nozzle was evaluated using fatigue monitoring software.

Please provide the benchmarking of the software using relevant transient data, proper 3-D model (cylinder to cylinder), and NRC endorsed computer code ANSYS. Please justify the use of the fatigue monitoring software to update the CUF calculation by using the monitored or projected transient data (cycles) and discuss the conservatisms in the calculation on a plant specific basis.

VEGP Response:

This RAI response documents the performance of a benchmarking study demonstrating conservatism of the FatiguePro-calculated results relative to a more refined NB-3200 fatigue analysis for the two sample plant components.

The transients used in the analyses were developed to produce the range of stress response expected during plant operations. Two transients (one fast, one slow) were developed for the Charging Nozzle analyses. Two transients (one with global loadings during a heatup and cooldown cycle and one with asymmetric local thermal stratification loadings) were developed for the RCS Surge Line Hot Leg Nozzle.

The results of the comparison between the FatiguePro and the NB-3200 analyses are shown below. In each case, the FatiguePro analysis produced larger Salt and CUF than did the ANSYS NB-3200 analysis for the evaluated plant transients.

Charaina Nozzle Transient Fast Transient Slow Transient Loss of Letdown with Delayed Return to Letdown Flow 50% Decrease and Return to Service Normal Analysis Tool FatiguePro NB-3200 Percentage FatiguePro NB-3200 Percentage Max Sat 296.8 262.8 89% 9.37 8.99 96%

Max Sn 111.3 108 97% 11.71 13.3 114%

Max Ke 3.333 3.333 100% 1.0 1.0 100%

CUF 0.0138 0.0102 74% 0.OOE+00 0.OOE+00 N/A RCS Surge Line Hot Leg Nozzle Transient Cooldown and Heatup MOP Heatup with Local Stratification Analysis Tool FatiguePro NB-3200 Percentag FatiguePro NB-3200 Percentage e

Max SaIt 14.465 9.7 67% 66.9 29.2 44%

Max Sn 23.5 11.1 7% 61.6 47.1 76%

Max Ke 1.0 1.0 100% 1.767 1.0 57%

CUF 2.38E-08 O.OOE+00 0% 0.00012 1.50E-06 1%

See Attachment A for the complete report on the benchmarking study.

Page 8 of 11

Enclosure NL-08-0228 Vogtle License Renewal RAI Responses - 01/22/2008 RAI - 4.7.3-1 The staff review of License Renewal Application Section 4.7.3 identified an area in which additional information was necessary to complete the review of the applicant=s Time Limited Aging Analysis evaluation.

Paragraphs two and three of Section 4.7.3 are shown here for aiding the questions that follow.

The nominal tube wall thickness is 0.040". The Steam Generator Program - Tube Integrity requires steam generator tubes be plugged if they have a 40 percent degradation from the nominal wall thickness (0.040*.4 = 0.016 or a wall thickness less than 0.024"). Based on the results of specific analysis for allowable tube wall thinning for the VEGP Model F steam generator tubes under normal operating and accident loadings, a minimum wall thickness of 0.014 inch is necessary to satisfy the stress limits of Regulatory Guide 1.121.

The minimum inspection-acceptable wall thickness for new tubes is 0.039".

The assumed general wall loss due to corrosion and erosion over 40 years is 3 mils, which reduces the tube wall thickness to 0.036". The corrosion rate of 3 mils is based on a conservative weight-loss rate for Inconel tubing in flowing 650 OF primary side reactor coolant fluid. The weight loss, when equated to a thinning rate and projected over a 40-year design objective with appropriate reduction after initial hours, is equivalent to 0.083-mils thinning. The assumed corrosion rate of 3 mils allows a conservative 2.917 mils for general corrosion thinning on the secondary side. Increasing the assumed corrosion rate by 50 percent from 3 mils to 4.5 mils has no effect on tube plugging criteria.

Sentences three, five, and six of paragraph three imply that 3 mils is the corrosion rate, not the general wall loss over 40 years. Please confirm that the 3 mils being referenced is the total wall loss over the 40 year design life.

VEGP Response:

SNC confirms that the 3 mils being referenced is the total wall loss over the 40 year design life. The term "corrosion rate" in sentences 3, 5, and 6 of the third paragraph of LRA Section 4.7.3 is the corrosion over a 40-year period, not an annual period.

Page 9 of 11

Enclosure NL-08-0228 Vogtle License Renewal RAI Responses - 01/22/2008 RAI - 4.7.3-2 The nominal tube wall thickness is 0.040". The Steam Generator Program - Tube Integrity requires steam generator tubes be plugged if they have a 40 percent degradation from the nominal wall thickness (0.040*.4 = 0.016 or a wall thickness less than 0.024"). Based on the results of specific analysis for allowable tube wall thinning for the VEGP Model F steam generator tubes under normal operating and accident loadings, a minimum wall thickness of 0.014 inch is necessary to satisfy the stress limits of Regulatory Guide 1.121.

The minimum inspection-acceptable wall thickness for new tubes is 0.039".

The assumed general wall loss due to corrosion and erosion over 40 years is 3 mils, which reduces the tube wall thickness to 0.036". The corrosion rate of 3 mils is based on a conservative weight-loss rate for Inconel tubing in flowing 650 OF primary side reactor coolant fluid. The weight loss, when equated to a thinning rate and projected over a 40-year design objective with appropriate reduction after initial hours, is equivalent to 0.083-mils thinning. The assumed corrosion rate of 3 mils allows a conservative 2.917 mils for general corrosion thinning on the secondary side. Increasing the assumed corrosion rate by 50 percent from 3 mils to 4.5 mils has no effect on tube plugging criteria.

Sentence three of paragraph three also implies that the entire 3 mils of wall loss will be consumed from the primary side of the tube only, which appears to conflict with the statement that there are 2.917 mils of allowance for general corrosion thinning on the secondary side.

Please confirm that the 3 mils of wall loss accounts for wall loss on both the primary and secondary sides of the tubes. If this is not the case, please indicate what the corrosion allowances are for the primary and secondary sides of the tubes. Also, please provide a basis for stating that a 2.917 mils allowance for general corrosion thinning on the secondary side is conservative (e.g., discuss the operating experience of the Vogtle Electric Generating Plant regarding secondary side general thinning of steam generator tubes).

VEGP Response:

SNC confirms that the 3 mils of wall loss accounts for wall loss on both the primary and secondary sides of the tubes.

The statement that a 2.917 mils allowance for general corrosion thinning on the secondary side is conservative came from the VEGP UFSAR, Section 9.:5.4.2.1.1. From Westinghouse report WNEP-8661, Rev. 1, the actual calculated tube ID plus OD erosion/corrosion calculated wastage for Alloy 600/690 tubes is less than 2 mils (1.77) over a 40-year design life.

Industry operating experience is that tube wall loss is usually due to local effects, typically wear of the tube against some structure (TSP, AVB, foreign object) or rarely, thinning of the cold leg tubes resulting from local off normal coolant chemistry conditions. The effects of degradation are detected and sized during inspections, and the Operational Assessment (OA) attempts to justify leaving them in service based on material loss rates added to the as-found (with NDE uncertainty) condition. The growth rates are usually derived from data from consecutive inspections for specific types of degradation at specific locations.

Operating experience at VEGP is that general corrosion of the secondary side of steam generator tubes has not been identified as an active degradation mechanism.

Page 10 of 11

Enclosure NL-08-0228 Vogtle License Renewal RAI Responses - 01/22/2008 RAI - B.3.14-1 In the License Amendment Request, Appendix B, Section B.3.14, Nickel Alloy Management Program for Non-Reactor Vessel Closure Head Penetration Locations, it is stated that currently, management of primary water stress-corrosion cracking in nickel alloys is a rapidly evolving area and as a result, program attributes have not yet been finalized. Further, where industry guidance has been developed, there are ongoing efforts to reach acceptable resolution of NRC staff concerns which may alter program requirements. Therefore, assessments for each of the ten aging management program elements are not included for this program.

Please identify when the assessments of each of the ten aging management program elements will be provided.

VEGP Response:

Assessments for each of the ten aging management program elements will be provided not less than 24 months prior to entering the period of extended operation for VEGP Units 1 and 2.

Future Action Commitment List item 12, sub-item (3) will be revised to add the following sentence:

"The inspection plan will include assessments of each of the ten aging management program elements defined in Section A.1.2.3 of NUREG-1800, Revision 1."

Similar revisions will be made to LRA Appendix A, Section A.2.14 and to LRA Appendix B, Section B.3.14.

A License Renewal Application amendment is required.

Page 11 of 11