Presentation to NRC by Vermont Yankee Regarding ... · uwl~tergyAgna d- -a Overview-Technical...

iteinteEntergy, Nuclear

Presentation to NRC Staff Regarding ReactorPressure Vessel Nozzle Environmental Fatigue

Analyses for License Renewal

Vermont Yankee-Nuclear Power Station- (VY-NPS)

~Enteigy

NRC Public Meeting, 1/8/20081 1

1 tergy

Instroduct'ilon,

Jay Thayer

Vice: Pres-idenat Entergy


uwl~tergyAgnad- -a

*Overview- *Technical Presentation

*Open Discussion3* Summary/Follow-up Actions

*Closing Remarks

~Enteigy


nierg T echnical ,, Pres,,nr t'a ti,-o)njil

Grand Gulf

--Umýý

" Open Technical Questions

* Nomenclature

" Industry Experience

" VYNPS Fatigue Analysis Methodollogy

" Conservatisms in VYNPS Ana lysis Approach

" Basis for Acceptabil1ity of VYNPS Approach

" Confirmatory Analyses


ditegj

I -."'I ..If

I

I

I Opeinii T chinicalF" Que-,-st-fijo~ns

I

I TEntergy


iteigy Op e n Tie-,.c--hnicalI Qu e stilns

!d~ RAI 4. 3.3-2 Question'sm Use of axisymmetric. model for vessel

I nozzles.I m Use of component stresses in lieu of

principal stresses.I m Use of Green's Functions for thermaltransients.

IEnterý


1 ergy

I

N~onieneia tu1-1r-e

EnteW


iteigyjNI om,, ,/encIIa-

Terminology requiring clarification0 "21D" vs. 3D

mNozzle corner contour effectsE "1 D virtual stress"

m Use of a single stress difference vs. using 6stress components

m Nozzle corner, blend radius & innerradius are interchangeable terms.


teigy Nv,"o-)fmiie rnon cr;I turej

I iI

Z

L)

W0n

Z

I

I

I

Z

cc

I Entfergy

VYNPS Feedwater NozzleNRC Public Meeting, 1/8/20089 9

teigy

VYNPS Fatigue AnalysisU

Mettho dorlogy

NRC Public Meeting, 1/8/2008

Enter,

10

Ite

iI

rgyY P Fatigue AnalysilsrMetVoYoNNgy) SJ

M, 1` h! odag

1. An axisym metric nozzle finite element model(FEM) is develop-ed from plant-specific drawingsand material specifications.

2. Heat transfer coefficients and boundary conditionsare established for the FEM based on the RPVCertified Design Specification and Stress Report.

3. The thermal stress response (i.e., Green'sFunction) is developed for a step change intemperature. A scalable pressure stress isseparately obtained by applying a unit pressureload to the FEM.

I

I

I

I Entery


ntegyV/Y NIPS, Fatiue- A nralysis1

4. Thermal transients are based on appropriateDesign Specification and BWR-4 thermal cycledefinitions.

)<ANO,

1~ 5. Thermal stress histories are obtained (usingGreen's Function integration) for each thermaltransient. The Green's Function results are

7 reviewed for appropriateness.IRiver Bend

-73 6. The associated pressure for each thermal stresspoint is determined based on linearly scaling the

-- 7 unit pressure stress response based on the__ transient, pressure time history.

LfEnteWg


te-igy V'ryN týij gr/ýN 7, ý !') Pow S -Frr:: ure AnalysiW / -is

N

..a t~~.ir 7. Stresses due to attached piping loads areconservatively calculated and are scaled based-on RPV temperature for each transient.

I 8. NB-3200 fatigue analysis is performed for thecollected thermal transients stress histories,conservatively combining the thermal, pressure(with a scaling factor to 'account for'nozzle,contour effects), and attached piping stresses,and using conservative cycle projections todetermine the 60-year CUF.

I :2

The VYNPS methodology is the same as the approach usedfor most CLB BWR RPV nozzle fatigue analyses..


,t~VYPSFaigjue Analsis

Gra?~d Gulf' 9.I

Maximum fatigue life correction factors (F enmultipliers) are calculated for water chemistryconditions (including power uprate effects)expected to occur over the 60-year operatingperiod.I

10. Environmental fatigue effects are calculated bymultiplying the CUF for 60 years by the Fen,I

I

I ~Entiegy


1 tergy

FF V/YNPSAnalysi

A p_-p r oachI

F:ý:EnteriW


*Constant (bound ing) material properties are used.

*The FEM mesh is much finer than that used in theCLB stress analyses for VYN PS, which results inh Iigher peak stresses. [1,2,4,11,14]

*Thermal stresses are calculated using-'41 '61conservative heat transfer coefficients based on

bounding flow rates. [4,11,14]

- Attached piping stresses are always combinedwith thermal stresses with like signs., such that

C.'yc"ýthey always maximize alternating stress ranges.


ditegy .Conservlatlis,.,m,,s Mn VYNPSAnal~i k E~ FO% htI

II . - - I , - I,,

11

*Bounding values for pressure' and temperature (atpower uprate conditions) are assumed for theentire 6,O-year period of plant operation. [9,- 13, 161

I

*The entire stress time history for each transient isgenerated, compared to selected transient pointsgenerated in the CLB stress analysis. Thisensures that maximum peak stresses are used.

I *Ke is calculated consistent with current ASMVECode methodology, which ensures alternatingstresses are maximizend. Ke was not required forthe VYNPS CLB vessel fatigue analys-is.[1, 2, 5,9,10,13,16]

[ygI


tergy Co-)n js e ýrva-,tji s-,ms min VYN,,P- jS)A na is AW noaI

I

ran Glf

IThe number of transient cycles for 60 years usedin the analysis is conservative relative to thenumber of transients experienced to-date andexpected through 60 years of operation (to bemonitored). 1191

I *Design basis transient, severity definitions wereused.

I 0 Bounding F en multipliers werevalues for temperature, straincontent that were selected tomultiplier. [10]

calculated usingrate and sulfurmaximize the F en

I

11 Entergy


i tergy

io

1-11.1-1 i , 91

tr e rtdu jrv El' i, e-Fj -11-'--,, --,; r- in -m

I ýEntergy


itergy Ind'st ry ExerienceK

m VYNPS considered methodologyI previous license renewal applican

used byits(used by:m VYNPS selected the methodolog)I

m Dresden/Quad Citiesm Nine Mile Point Unit 1I

m Oyster CreekI m Ginna

m Point Beachm FarleyI Palisades

I ~EnteqVy m Millstone 2/3NRC Public Meeting, 1/8/2008 220

Enteigy


tergy Basirs.frAcptblt ofIt

ulf

*The nozzle FEM techniques and analysis methodsused to establish stress histories are consistent withthe current design and licensing bases (CLB) forV/YNPS.

I *The use of Green's Functions for calculatingthermal transient stresses is well establishedthroughout the industry (since 1986). [6]

0 The multiple conservatisms in the analysismethods.I

*The comparisons performed between the VYNPSapproach and the- classical ASMVE NB-3200approach show that the VYNPS approach providesequal or higher alternating stresses and fatigueusage. [18, 20]

ýEnteW~I


teigy

IGadGulf

UJs-e of AxsymnrcMoe

I

~EnteWgI


I~Generalul

*The CLB RPV nozzle fatigue analyses for VYNPSare based on the use of axisymmetric modelS. [1, 2]

- This approach has been. an industry standard formany years and is the basi~s for most existing

I- nozzle fatigue analyses in U.S. operating reactors.

*The approach VYNPS used for environmentalIfatigue calculations is consistent with the VYNPS

-Ltegy


itergy U~se of Axis0 -,ymmetri

- -~ .,2-, -' 1 4

Technical Considerations

*Nozzle contour effects have been accounted for in the VYNPSaxisymmetric models.

"Thermal transient stresses were shown in BWRVIP-108 tobe azimuthally uniform in the nozzle/RPV contour regionfor a variety of RPV nozzles using full 3D (non-axisymmetric) finite element models

*For example, refer to the next slide. [3]

*Therefore, no adjustment for nozzle contour effects isrequired for thermal stresses.

" Pressure stresses were accounted for by increasing theradius of the reactor vessel in the axisym'Metric model (tobe consistent with the CLB analysis) and/or application ofan additional multiplier to fully account for nozzle contour

pressure stress effects in the nozzle/RPV contour region..


UJse o xsymtiiteWgMo iderl

Technical Considerations(typical example of results from BWRVIP-1 08)

AN

'c0pz EA OPK 0JZ

Core Spray Nozzle. Blend Radius. Inside SurfaceZLE. THERMAL

35000O

30C-00

25000

220000

10000

5D200

NRC Public Meeting, 1/8/2008 02Auirnutti (degreei

26

Use of A xis ymm erriciteWgMdel,1

GrndLffITechnical Considerations - Pressure Stress

*Multipliers on vessel radius have traditionally been used forCLB analyses to account for nozzle contour effects.

I*The multipliers applied to environmental fatigue

calculations are consistent with the maximum values usedhistorically in the VYNPS CLB. [2, 4, 51

.1"a

*The VYNPS nozzle analyses appropriately considerednozzle contour effects on stress..

*The finite element model accounts for material and geometricdiscontinuities of the nozzle corner.

I

I En~itelgy

*A multiplier of 2.0 was used to account for the sphericalmodeling effect.


d erg3(

R

Use of Component St,,r-e--sse.s.-


~tegyUse of Component Stres-es

General Case (3D stress) [21]

(--ad 6411f

L_ * .'r- It can be shown that the complete state ofstress can be determined by knowledge of

Y stress vectors on any three perpendicularCTY planes.

*It is conventional to consider the three mutuallydI::ý perpendicular planes as faces of a cube of

PRE_ infinitesimal size, a stress element.T ZY The state of stress can be conveniently written

CFX as a matrix or a tensor:

dtz TCzy T * nysxcmoetsaeidpnet~T~60p C/X'X TXZ Z,~

I~SYEntY yZ


Uý

ulterg Ue of C, -mp-.onen Stre-,sses

*There exists an orthogonal set of axes 1, 2, 3 called principal axes withGul respect to which the stress state elements are all zero except for those

in the principal diagonal:

[Gl 01

-0 0 73

L~i In other words, there always exists a set of mutually perpendicularplanes with zero shear stress.

*The stress intensity, SI is defined as:

SI =MAXIMUM (cyl - 2, G2 3 1 ~31)

__ *Thus, if a stress difference based on component stresses (i.e., cy -CTX)equals SI, then shear stresses are negligible.

EnteTgý


;iieigy Us-e of ComponetSrs s

*The NSSS vendor practice for BWR CLB nozzleanalyses has traditional~ly been to use component(Sx, Sy, Sz) stresses, for two reasons:

*ASMVE Code Section 111, NB-3215(d) states, "nmn-7 1pressure component calculations, the t, , and r directions

may be so chosen that the shear stress components arezero and a,, G2, and Cy are identical to at, cy1,'and oyr-"

*Experience indicated that shear stresses were negligible.

I The VYNPS CLB analyses used componentstresses, since shear stresses are negligible. [2,4,5]

I :Entergy


-Technical Considerations -Feedwater Nozzle

*The impact of using component stresses was evaluated forthe nozzle corner and safe end locations for all three nozzlesanalyzed for VYNPS.[7, 17]

*The feedwater nozzle has the highest environmentally77_ adjusted CUE (0.639 at the nozzle corner and 0.256 at the

safe end).*The thermal stress response to a 400OF step- change in

temperature for the to nozzle locations evaluated is shownon the next 2 slides. [7, 17]

*For both feedwater nozzle locations, excellent agreementexists between SI and the maximum component stressdifference for the Green's Function.

*Based on this close agreement, using SI would not change-~ the calculated CUF.

EnteWg


I t er1Y. UJ/ s-e aof C o, mn p) on len t' S-ttr e s,- s-,es-

IFeedwater Nozzle, Corner Stress DifferenceComparison

Total Stress Intensity

30000

... ~w +

25000

20000 - - -

15000 --- Sz-s

0 100 200 300 400 500'EnteriW Time (sec)


tergy U s e o)f C om,,-tp )o, n-,lent S tr e.ss e- s

m

~AJ

Si

Feedwater Nozzle Safe End Stress DifferenceCornparison


7a0000

60000

50000

40000

30000

20000

'A

U)

I) - - - sz-sx

I; - -Sz-SY

I -S1

A1

0

-10000

TEnteWg

NRC Public Mee

0

ting, 1/8/2008

100 200 300 400 500

Time (sec)

34

U tergy Us,ýe i -,of Comrrponre.,nt Strfes se s-7

Technical Considerations - Core Spray Nozzle

" The core spray nozzle has the next highest environmentally adjusted CUE(0. 167 for the nozzle corner and 0.059 for the safe end). [16]

" The thermal stress response to a 400OF step change in temperature for the twonozzle locations evaluated is shown on the next 2 slides. [7, 17]

" There is very close agreement between SI and the maximum component stressdifference at the nozzle corner.

*Based on this close agreement for the nozzle corner, using SI would not change thecalculated CUF.

" The safe end also has very good agreement at the peak value for the Green'sFunction. However, there was a difference (up to 50%) at decay stress values..0 Due to the difference in stresses for the safe end, a confirmatory evaluation was

performed using SI. This resulted in a calculated increase in environmentally adjustedCUE of 0.003

" The total environmentally adjusted CUE increased from 0.059 to 0.062, or 5%. [-9" This increase is small, is within the accuracy of the analysis, and is enveloped by

conservatisms in the analysis.a This small difference is a result of the Green's Function process, where the key

comparison is the peak stress value." A significant margin of 0.833 exists with respect to the ASMVE Code limit of 1.0.


i teigy Use of Comonent, Stresses

Core Spray Nozzle Corner Stress DifferenceCornparison Total Stress Intensity

30000

25000

20000

15000

10000

5000

U) 0U)

-5000

-10000

-15000

-20000

-25000

-30000

SNRC Public Meeting, 1/8/2008

100 200 300 400

Time (see)500

36

iteigy Uiis-e o 0f ,Ca'0 i"m-f( p'- ojne l..,nt, S tr-e s-ý sje--,s

rCore Spray Nozzle Safe End Stress Difference.Cornparison


I

II.

i-

80000

70000

60000

50000

40000

30000

20000

10000

0

-10000

-20000

-30000

-40000

-50000

-60000

-70000

-80000

I

I

I Entergy Time (sec)


riergY Use of Compo_,nrent Strsses

Technical Considerations-Recirculation Outlet Nozzle

" The recirculation outlet nozzle has a environmentallyadjusted CUE of 0.084 for the nozzle corner and 0.018 forthe safe end. [13]

" The thermal stress response to a 400OF step change intemperature for the two nozzle locations evaluated isshown on the 2 follow-on slides. [7, 17]

" For the safe end location, there was excellent agreementbetween SI and the maximum component stressdifference for the Green's Function.

*Based on this close agreement for the safe end, using SIwould not change the calculated CUF.

I

EnteW~


-Recirculation Outlet Nozzle Safe End StressDifference Comparison

120000

1000080000

6000 f i- - - SZ-Sx

20000___________________ __

-4000 -

-100000-

-1 20000

~-~Eterg -14000Time (sec)


itergY. U~se of Co-mfp-oinient Stresses

Technical Considerations- Recirculation Outlet Nozzle

*For the nozzle corner loc 'ation, there was a difference of 10% betweenthe peak values of SI and the maximum component stress differencefor the Green's Function.

The difference between SI and the maximum component stress differencehas a negligible effect because:

*The most significant thermal transient (Improper Start causing reverse flow) wasmodeled directly in the FEM due to its unique characteristics.

*In. the nozzle corner, the thermal stresses are small compared to the pressurestresses. [13]

*Due to the difference in stresses for the nozzle corner, a confirmatoryevaluation was performed using SI. This resulted in a calculated increase inenvironmentally adjusted CUF of 0.003.

*The total environmentally adjusted CUF increased from.0.084 to 0.087, or3.5%. [7]

*This increase is small, is within the accuracy of the analysis, and isenveloped by conservatisms in the analysis.

*A significant margin of 0.913 exists to the ASME Code limit of 1.0.~EnteiTy


tergY. Use- of Cmoent StresesI -I

7-

RecirculDiffereni

50000 ------ -

30000 I

20000

30000

-10000

-20000-

-30000

-50000

-60000

-70000

NRC Public Meeting, 1/8/2008

ation Outlet Nozzle Corner Stressr~e Comparison


'Entei

41

:h ter~y

II A -. If

I

I Use, of Green $1ýsF~u-n~ct'io)n-, A-pproach

I

I ~Eter~y I


itergy Ue of Green's, Function,Ap pwr o-a-,,c;h

Hi

I

*The use of integration functions, such as Green'sFunctions, are a well-established mathematicaltechnique. [6]

m This approach is used to establish the correlation, or stressresponse, to a unit step change thermal transient.

m From the stress response to the unit step transient, a stresshistory can be easily integrated for any thermal transient.

* The method is accurate and reliable, and has provenmathematics behind it.

0 "Duhamel's Formulas" in most college engineering text books.m Similar to integrating' the area under a curve, the only limitation

is the size of the integration time step.m The VYNPS Green's Functions utilize time steps as small as

0.01 second.~EnterWg


UsA --e of Green's FunctioniteWgApproach

I, Green's FunctionsGrand- Gu If

OL

450

400

350

300

250

200

Q,

-4--C,)

Time (sec) 92825r0

iNote: A typical set of two Green's Functions is shown, each for a different set of heat transfer coefficients (representing different

flow rate conditions).Enterg


Use. of Green's FunctiontergA /,A1ppro.ac-hF

Green's Function Integration Process

I*1

0

aIII

I 0o0 zoo Wo Mo I= Im 0 1400 1500 I=0 2000 M0 4W NO0 800 1000 1200 1IM IM010 I 2000ft =

* To compute the thermnal stress response for an arbitrary transient, the local fluid temperature is deconstructed into a series of step-loadings.* By using the Green's Function, the response to each step can be quickly determined.* By the principle of superposition, these can be added (algebraically) to determine the response to the original load history.

" The result is demonstrated in the figure on the right." The input transient temperature history contains five step-changes of varying size, as shown in the figure on the left." These five step changes produce the five successive stress responses in the figure on the right. By adding all five response curves,

the real-time stress response for the input thermal transient is computed.* The Green's Function methodology produces identical results compared to running the input transient through the finite element model.* The advantage of using Green's Functions is that many individual transients can be run with a significant reduction of effort compared to running

all transients through the finite element model.I _`EnteWg


i tergy

rwul Gulf

I,ii

Addtioal-onfr toryRive /o r kd-

7Entclgy


nte_)n Cofr atoiry A nalyi

Le;W'GadifiA benchmarking calculation will be performed ofthe limiting component (feedwater nozzle) asfurther confirmation of the VYNPS fatigue analysis

i approach.

*The feedwater nozzle was chosen because:*It has the largest number and the most severe

Ruerlkndtransients.

*It has the largest fatigue usage.*Results will be bounding for the core spray and

71- recirculation outlet nozzles.*The analysis will be performed using the existing

axisymmetric model.


W ntergy. Con fim oyA ayi

Grad ul *All defined- transients will be eval'uated using thefinite element model.[ *-d All six stress components will be used to computefatigue usage via ASMVE Section'1II NB-3200 methods.

*The CUF results will be compared to CUF resultsfrom the previous environmental fatiguecalculations.

I EnleWg


d ergy

I

Op en DisC-ussion

EnfeWg


otergy

I

I

01 9;

olait ,.5, ck 1/0 vs,-,' to)i nuII

I

I "EnterVy


E

in tegtovlson

*VYNPS nozzle fatigue analyses were performed usingmodeling techniques (axisymmetric) and methodologiesthat are consistent with the CLB.

*The methods used in the. VYNPS nozzl e fatigue analyses''INO are consistent with classical ASMVE Code Section III NB-

3200 methodology.*Conservatisms exist in the analysis approach that bound

any uncertainties, which are small when compared to theanalysis results.

Bounding design basis transient definitions*Bounding 60-year cycle counts

-*Bounding heat transfer coefficients*Significant margin (0.36) remains to the ASME Code

allowable value of 1.0 (maximum CUF is 0.64).

I ~EnteW


d ergy

G' "-I 60f,

)References


ztergy Reoe 0e-:A ' e - r ,l e-.,

I1. Chicago Bridge & Iron Company, Reactor Vessel Stress Report,-1GW6ý-YnkeReactor, cnrt9-6201, October 1969.

2. Teledyne Engineering Services Technical Report TR-6400-1,Fatigue Analysis Summary of Vermont Yankee Reactor Vessel,

;1NO VYC 378, October 1985.S3. EPRI BWRVIP-1.08: BWR Vessel and Internals Project, "Technical

Basis for the Reduction of Inspection Requirements for the BoilingWater Reactor Nozzle-to-Vessel Shell Welds and Nozzle BlendRadii, Report 1003557, October 2002.

14. Calculation VY-1OQ-301, Rev'. 0, "Feedwater Nozzle FiniteElement Model and Heat Transfer Coefficients", March 2004.

5. Calculation VY-1OQ-303, Rev. 0, "Feedwater Nozzle Stress andFatigue Analysis", March 2.004.

6. Kuo, Tang & Riccardella ASME PVP Paper, "An On-Line FatigueMonitoring System for Power Plants", presented at the 1986Pressure Vessels and Piping Conference and Exhibition, Chicago,

I ~ IL.


itergy R-eferences-.772

7. Letter, Entergy to USNRC, "Vermont Yankee Nuclear PowerStation, License No. DPR-28, License Renewal Application,Amendment 33, BVY 07-082, dated December 11, 2007.

8. Calculation VY-1 6Q-301, Rev. 0, "Feedwater Nozzle StressHistory Development for Green's Functions",3 July 2007.

9. Calculation VY-16Q-302, Rev. 0, "Fatigue Analysis of FeedwaterNozzle" July 2007.

10. Calculation VY-16Q-303, Rev. 0, "Environmental FatigueEvaluation of Reactor Recirculation Inlet Nozzle and VesselShell/Bottom Head"19, July 2007.

11. Calculation VY-1 6Q-304, Rev. 0, "Recirculation Outlet NozzleFinite Element Model"11, July 2007.

12. Calculation VY-16Q-305, Rev. 0, "Recirculation Outlet StressHistory Development for Nozzle Green Functions"11, July 2007.

13. Calculation VY-16Q-306, Rev. 0, "Fatigue Analysis ofRecirculation Outlet Nozzle" , July 2007.

1.Calculation VY-16Q-308, Rev. 0, "Core Spray Nozzle FiniteElement Model"3), July 2007.~Enteigy


>itergy R-,eferene

15. Calculation VY-1 6Q-309, Rev. 1, "Core Spray Nozzle Green's Functions"G~dG~dfDecember 2007.

16. Calculation VY-1 6Q-31 0, Rev. 1, "Fatigue Analysis of Core Spray Nozzle",December 2007.

17. Report SIR-07-438-NPS, Rev. 0, "Comparison of Component StressDifference to Total Stress Intensity for Recirculation Outlet and Core SprayNozzles", December 2007.

18. Calculation QA-2000-1 02, rev. 0, "Evaluation of Turbine Roll Event:ANSYS, vs. FatiguePro", May 1995.

19. "Reactor Thermal Cycles for 60 Years of Operation," Attachment 1 ofEntergy Design Input Record (DIR) EC No. 1773, Revision 0,"Environmental Fatigue Analysis for Vermont Yankee Nuclear PowerStation," July 2007.

i:;'zý20. EPRI Materials Reliability Program: Second International Conference on_____Fatigue of Reactor Components (MRP-84), "Fatigue Evaluation of a BWR

Feedwater Nozzle using an On-Line Fatigue Monitoring System", March2003.

21. Shigley, Joseph E., "Mechanical Engineering Design," Third Edition,McGraw-Hill Book Company, New York, 1977.

7='-Entertgy


Presentation to NRC by Vermont Yankee Regarding ... · uwl~tergyAgna d- -a Overview-Technical...

Documents

Transcript of Presentation to NRC by Vermont Yankee Regarding ... · uwl~tergyAgna d- -a Overview-Technical...

Presentation to NRC by Vermont Yankee Regarding ... · uwl~tergyAgna d- -a *Overview-*Technical...

Documents

Transcript of Presentation to NRC by Vermont Yankee Regarding ... · uwl~tergyAgna d- -a *Overview-*Technical...

Presentation to NRC by Vermont Yankee Regarding ... · uwl~tergyAgna d- -a Overview-Technical...

Transcript of Presentation to NRC by Vermont Yankee Regarding ... · uwl~tergyAgna d- -a Overview-Technical...