ML19332D539

Core Thermal-Hydraulic Methodology Using VIPRE-01.

ML19332D539
Person / Time
Site:	Oconee
Issue date:	10/31/1989
From:	Kenneth Jones, Koontz D DUKE POWER CO.
To:
Shared Package
ML16152A876	List: ML16152A875 ML19332D539
References
DPC-NE-2003-A, NUDOCS 8912040110
Download: ML19332D539 (105)

Text

pu;;gu .y ;,mam_ _ _ m mm.__,,_.m mm ,,g._, ., m . , .. . . . _ , - . . , _ , _ _ _ _ _ , , , _ _ _ _ _ . _ _ _ _ _

p

]l!j i. -"#** -l e

LL -+ . t a + . [L d A , ,

a th  !

7 g 7 .

.m

]

o Li 1

i

(

l *

.J 4 3

a

] _ _ . . - =._ _ -

_. s

.i 1

a4 1

3

1 LM w.

,a l

]:

LM 4

q g  !

.t J-

.g u

3

?

i F

S .

- r

.e- g912040110 891121

. pea ADOCK 0500gg9 P .

>'

s _;

.,' 4 t j,\) ' -

I, l l1 ,

DPC-NE-20034A _

• l lfg s i

DUKE f0WER COMPAW .

^

OCONEE NUCLEAR STATION CORE THERMAL-HYDRAULIC METHODOLOGY USING VIPRE-01 l3 Original Report: August 1988 Approved Report: October 1989 I K. B. Jones D. R. Koontz ..;

I g

i n

Duke Power Company l , Design Engineering Department Nuclear Engineering Section l:

ET Charlotte, North Carolina

(;E.

I .

I:

"I l

w -

g-y

-;g x9 i$

ABSTRACT i

j This report presents Duke Power Company's methodology for using VIPRE-01 for.

. performing thermal-hydraulic analyses in support of Oconee Nuclear Station The VIPRE-01 thermal-hydraulic: methodology and models

~

licensing activities.

are presented along with the results of sensitivity studies used in determining

~

the acceptability of the various input criteria. This report meets the  ;

licensing requirements addressed in the. Safety Evaluation Report for EPRI NP-2511-CCM, VIPRE-01, ref. 3.

{

i i

i i i 11 l

?

1 Table of Contents 1.0 15TRODUCTION- 1 12.0 GODE DESCRIPTION 2.

l 3.0- STATION DESCRIPTION 3 4,0 CORE MODELING 3 4.1 - STEADY-STATE -SINGLE PASS MODEL COMPARISONS - 6 4.2 TRANSIENT MODEL COMPARISONS 6 4.3- TRANSITION CORE MODEL COMPARISOMS 7 4.4- RESULTS

SUMMARY

8

5. 0 ~. VIPRE-01 DATA 8 5.1 AXIAL NODING 8 5.2 ACTIVE FUEL LENGTH. 9 5.3' CENTROID DISTANCE 10 5.4 . EFFECTIVE CRO'SSFLOW GAPS 10 5.5- SPACER GRIO FORM LOSS COEFFICIENTS 11 5.6 CORE BYPASS FLOW 11

5. 7- INLET FLOW DISTRIBUTION 12

'5.8 VIPRE-01 CORRELATIONS 12 5.8.1 PRICTION PRESSURE LOSS 12 5.8.2 TURBULENT MIXING 14 5.8.3 TWO-PHASE FLOW CORRELATIONS 15 5.9 REFERENCE DESIGN POWER DISTRIBUTION 17 5.10 AX!AL POWER OISTRIBUTION 17 5.11 HOT CHANNEL FACTORS 18 5.12 FLOW AREA REDUCTION FACTOR 19 5.13 BWC CRITICAL HEAT FLUX CORRELATION 19 I

111

hh$h v^[

b + Table of Contents-(Continued)

, Section Pace

[L 6,0 OCONEE THERMAL-HYORAULIC ANALYSES 22 i

22 l -

6.1

SUMMARY

22 -l 6.2. THERMAL-H'!DRAULIC DESIGN CRITERION

, -6.3- CORE SAFETY LIMITS 23 23 6.4 PRESSURE-TEMPERATURE ENVELOPE 6.4.1 REFERENCE POWER DISTRIBUTION 24 6,4.2 CORE POWER 25  ;

6. 4,3' RCS FLOW 25 3 ,

6.4,4 CORE INLET TEMPERATURE 25 {

6,5 GENERIC MAXIMg ALLOWABLE PEAKING LIMIT CURVES 26 27 6.6 PUMP COASTDOWN TRANSIENT ANALYSES

7.0 REFERENCES

31 APPENDIX A Safety Evaluation Report A-l' .l k Appendix B Responses to Requests for Additional B-1 -!

Infonnation i i

IV i J

ib

"{

, W

~

List of Tables

~

j j.

Table Title Pace -j

. 3-1 Mark BZ Fuel Assembly Data....................... 32 4% Ope rati ng . Condi ti on s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

. -- a O.- '4-2 Comparison of 64 Channel and.8 Channel Model-Steady-State Results........................ 34 s 4-3 '- Comparison of 64 Channel and 8 Channel Model Transient-Results.......................... 35

- . - 1 4-4 Comparis'on of 64' Channel and 9 Channel Transition g

Core Model Steady-State Results...... ........... 36 i 4-5 Comparison of 64 Channel .and 9 Channel Transition Core- Model Transient Results . . . . . . . . . . . . . . . . . , . . . 37

-l' l 5-1 8 Channel Model Axial Node Length' F S e n s i ti v i ty S tudy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 5-23 8 Channel Model Inlet Flow Distribution 1 Sensitivity Study................................ 39 I t 5-3 8 Channel Model. Two-Phase Flow Correlation And Friction Multiplier Sensitivity Study. . . . . . . 40 - q 5-4 8 Channel Model Turbulent Momentum Factor S e n s i t i v i ty . S t udy '. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

'6-1 RP S T ri p Functi on s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42  ;

1 I

3

x , x e

a

?

.-)'-

g A% -

( List of Ficures, i Fioure' T_1 t le pace 1

. 4-1L 64 Channel Model Eighth Core Representation. . . . . . 43 i

4-2L 64 Channel Model Hot Assembly Detail............. 44-4-3 8 and 9 Channel Model Hot Assembly Detail........ 45 s 4 8 Channel Model. Eighth Core Representaiton....... 46 1

4-5 9 Channel Model Eighth Core Representation. . . . . . . 47 .j i

5-1 VIPRE-01 v s . LYNX 2 M/P CHF. . . . . . . . . . . . . . . . . . . . . . . 48  !

5-2. VIPRE-01 vs. LYNX 2 Mass Velocity ................ 49 J 5-3 VI P RE-01 v s . LYNX 2 Qua l i ty . . . . . . . . . . . . . . . . . . . . . . 50 -

s 5-4 Measured vs. Predicted CHF.... .................. 51  :

5-5 Measured-to-Predicted CHF vs.-Quality............ 52 li 5-6 Measured-to-Predicted CHF vs. Mass Velocity. . . . . . 53 Mea sured-to-Predicted CHF vs. Pres sure. . . . . . . . . . . 1 5-7 54 6-1 RPS Core Protection Safety Limits. . . . . . . . . . . . . . . . 55 6-2 RPS P-T Core Protection Envelope................. 56' i 6-3 High Temperatures MAPS.......................... 57 6-4 ' Low Pressure MAPS.............. ................. 58 6-5 Fl u x- to- Fl ow MAP S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 l

6-6 Typical Two Pump Coastdown Transient Results..... 60 h

vi

-- , w a .:n ,

4 k p O I(; u ,

1.0 INTRODUCTION

Duke Power Company's Oconee Nuclear Station reactor core thermal-hydraulic design and licensing analyses have traditionally u nd very conservative methods

. to-establishthemaximumpermissiblecorepowerandpowerdistrib0kionfor-various combinations of core outlet pressure and reactor outlet temperature to ensure that DNBR criteria are met. Conservative " closed-channel" computer f codes have been used for Oconee Nuclear Station thermal-hydraulic analyses I using the methodology described -in reference 1. Crossflow computer codes which can predict flow redistribution ef fects within an open lattice reactor core, can more realistically predict the local fluid properties and thus, the departure from nucleate boiling ratio (DNBR) in the hot channels of the core.

.d.

R This report presents the procedure used to apply the VIPRE-01 computer code for thermal-hydraulic analyses of Oconee reactor cores and fulfills the i requirements addressed in the SER for using .VIPRE-01 for licensing analyses, ref. 3. The geometric representation of the core is-illustrated and discussed along with the models and empirical correlations used to determine friction pressure losses, coolant mixing and subcooled voids. Descriptions of the methodology used to determine the thermal-hydraulic iimits which define the I

t regions of safe operation in terms of power level, reactor coolant temperature and pressure, and power distribution are included in this report. The Oconee l thermal-hydraulic analyses will continue to treat un:ertainties, tolerances, and measurement errors conservatively. The methodology used to perform generic Oconee thermal-hydraulic analyses is discussed in the report. The need to perform the thermal-hydraulic analyses in conjunccion with a reload arises

^

b. .

L; >

h . when.there is a change inlthe fuel assembly design, a change in input assump-L: w tions of the generic analysis. .or- a change in the regulatory criteria, t -

i 2.0 CODE DESCRIPTION I

l i

VIPRE-01,'ref. 2, is an open cnannel, homogeneous equilibrium, thermal-hydraulic code-which features diversion crossflow and turbulent mixing to calculate the-

~

-l

' departure from~ nucleate boiling ratios (DNBRs). The code accepts input data  ;

which defines the geometric, hydraulic and thermal characteristics of. the core, and permits the user to select correlations and solution methodologies.

s Generally, core representation is made by inputting parameters defining and describing the number of channels and subchannels within the model and their -l individual channel and subchannel characteristics,.such as flow area, wetted -

1 and heated perimeters, adjacent channel data, and centroidal distances between j adjacent channels.' Hydraulics of the code are defined by crossflow resistances f determined from gap dimensions through which;the channels communicate, spacer grid locations and form loss coefficients, mixing coefficients, two phase flow j correlations, friction pressure losses, and inlet flow distributions. Thermal I

<v modeling of the reactor core is a function of the core radial and axial power distribution, core power, operating conditions, hot channel factors, heat  ;

transfer correlations and correlation limits. VIPRE-01 was designed to perform steady-state and transient thermal-hydraulic analyses of nuclear reactor cores for normal operating conditions and several accident conditions. The VIPRE-01

(: code has been reviewed by the NRC and was found to be acceptable for c '- ': referencing in licensing applications with the limitations addressed in ref. 3.

l 1

w

p I --

! 3.0. STATION DESCRIPTION Oconee Nuclear Station consists of three, Babcock & Wilcox (B&W) designed, pressurized water reactors with each reactor rated at 2568 Mwt. Each reactor.

core consists of 177 fuel assemblies with each assembly having 208 fuel rods, 16 control rod guide tubes, and an instrument tube arranged into a 15 x 15 l array. Eight non-mixing vane spacer grids provide lateral stiffness and fuel {

rod positioning. Typical dimensions and characteristics of the current Oconee  !

i in-reactor fuel assembly designs are given in Table 3-1.

i 4.0- CORE MODELING 'l L

Traditionally, core thermal-hydraulic analyses have been performed using  ;

multi-pass analyses. In a multi pass analysis, fuel assemblies and the subchannels of the hot assembly are modeled in separate simulations and

-sometimes in different computer codes. A more direct approach involves only a single pass. In a single pass analysis, the hot subchannel and adjacent j subchannels are modeled individually with larger and larger channels modeled toward the periphery of the core; the result is that all thermal-hydraulic DNB calculations can be performed using one code. VIPRE-01 has the capability to perform single pass analyses.

An Oconee Nuclear Station reactor core is geometrically modeled using eighth-core symmetry with the center assembly modeled as the " hot" assembly, Figure 4-1. The hot assembly is the assembly in which the minimum DNBR (MDNBR) can be expected to occur. The hot assembly is divided into subchannels with boundaries formed by fuel rods and guide tubes within the assembly, Figure 4-2.

1. m v

.I k ,

The hot assembly contains the " hot" subchannel (i the subchannel which h e.,

yields the MONBR oflthd. core). To conservatively determine the MDNBR for the

%^

core, the models use a high, relatively flat radial pin power distribution- 4 I along with.-the= application of hot subchannel factors and-reduced hot-subchannel flow area. The derivation and application of these factors will be discussed

,. in more detail in' Sections 5.11 and 5.12< I 1' h Selection of single pass models for performing' thermal-hydraulic analyses I requires the development of different size models and compt.risons of,the different models at various operating conditions.-Three different size models l were developed and compared for modeling Oconee Nuclear Station Fuel: 1 64 Channel Model 9 Chanael Model .

I 8 Channel Model i

~All'three.models were developed assuming eighth-ccre symmetry. The 64 channel  ;

model consists of 36 subchannels making up the hot a

,sembly with the remaining 28 channels individually modeling the rest of the assemblies in the eighth-core segment. The 64 channel model is depicted in Figures 4-1 and 4-2. The 8 and 9 j channel models were formed by including two rows of subchannels around the hot

^

subchannel (Channel 1) and lumping the rest of the hot assembly into one '

channel (Channel 7), Figure 4-3. The remaining 28 assemblies were either lumped into one large channel, Channel 8, in the case of the 8 channel model, i

l

_ _ _ _ __ _ __j

u r .

k-o Figure 4-A., or into two large channels, Channels 8 and 9, in the case of the 9 ~

channel model, Figure 4-5. The 64 and 8 channel models were compared to confirm the accuracy of the 8 channel model which will be used for steacy-state Land two-pump coastdown analyses. The 9 channel model will be used to evaluate j potential transition core effects of differing fuel assembly types. As Table l

'l 3-1 shows, the different fuel assembly designs only incorporate minor changes in the basic Mark-BZ fuel assembly-. design. The use of the 64, 9 and 8 channel models is not limited to the fuel assembly designs listed in Table 3-1; moreover, the 64, 9 and 8 channel models will be used to predict and evaluate i f the thermal-hydraulic effects for future fuel assembly designs.

For-illustrative purposes, the number of assemblies lumped together to form Channels 8 and 9 of the 9 channel model, Figure 4-5, was based on the Oconee 1 Unit 1, Cycle 11 core. The number of assemblies lumped together may vary with the cycle specific core configurations being evaluated. The assemblies are j arranged iri a' manr.er which will give conservative DNBR results.

l To determine the modeling detail required to accurately evaluate the hot channel local coolant conditions and the minimum departure from nucleate boiling ratio-(MDNBR), the 64, 9 and 8 channel models were run using the operating conditions stated in Table 4-1. The RECIRC numerical solution option was chosen to calculate the results. The VIPRE-01 SER, ref. 3, pg. 17 c- m

(

states that the RECIRC numerical solution is acceptable for licensing calcula-tions. IFe first two operating conditions, Cases 1 and 2, correspond to the high temperature and the low pressure safety limits associated with the Reactor Protection System, ref. 1. The case 4 operating conditions correspond A .

l^ ,

< to'the initial conditions for the two pump coastdown transient. The case 3 operating conditions correspond to the operating conditions occuring at the limiting MDNBR during the two pump'coastdown transient (i.e. , the limiting l statepoint in Figure 6-6). The Case 3 operating conditions are used to R develop the maximum allowable pin peaks discussed in Section 6.5. Additional details of the Reactor Protection System will be discussed in Section 6.

I i

- 4.1 STEADY-STATE SINGLE PASS MODEL COMPARISONS i

The 64 and 8 channel model results are compared in Table 4-2. Results for the l Case 1 operating conditions show that the 8 channel model conservatively 4

predicted the MDNBR by 1.2% when compared to the 64 channel model MDNBR.

Results for the Case 2 operating conditions showed the 8 channel model exhibited a 0.44% conservative difference in MONBR when compared to the 64 channel model. Likewise, the 8 channel model exhibited a 2.2% conservative change in MDNBR for Case 3.

i 4.2 TRANSIENT MODEL COMPARISONS ' l l l The two pump coastdown transient is the most limiting DNB transient; therefore, E the two pump coastdown transient was chosen to make a comparative study between the 64 and 8 channel models. The development of the transient modeling details are presented in Section 6.6. The transients were performed using the initia' operating conditions from Table 4-1, Case 4. The 64 and 8 channel model transient results are presented in Table 4-3. Throughout the transient, the 8 channel model produced conservative MONBRS in comparison to the 64 channel l

e c-

'model. The limiting MDNBR cbserved for the 8 channel model occurred at 4.1 seconds where the MONBR = 1.216 (i.e., which is conservative in comparison to the 64 channel ~model MONBR of 1.234 also occurring at 4.1 seconds). 1 I

4.' 3 TRANSITION CORE MODEL COMPARISONS I  :

As mentioned earlier in Section 1.0, a thermal-hydraulic analysis must be

~

performed whenever there is a change in the fuel assembly design, a change in ')

' Input assumptions of the generic analysis, or a change in the regulatory criteria. Combinations of different fuel assembly designs in a reactor core constitutes a mixed (transition)- core which must be evaluated to determine its effect on thermal-hye ;ulic performance; Transition core effects are determined  !

by comparing results of a thermal-hydraulic (T-H) analysis explicitly modeling  ;

the mixed core with that of a T-H analysis for a non-mixed core.

If the comparison shows the FONBR is adversely affected, then a penalty must be assigned to that particular operating cycle.

'The 64 channel and 9 channel transition core models were. compared on a steady- .

state and transient basis to ascertain the accuracy of the 9 channel model.

Table 4-4 presents a comparison of-the 64 and 9 channel models steady-state results. In all cases, the 9 channel model produced conservative results.

The steady-state runs using the Case 3 operating conditions produced the lowest MDNBRs, with the 9 channel model predicting MDNBRs 1.9% more conserva-tive than the 64 channel model. A comparison of the 64 chanrel and 9 channel ,

model two pump coastdrwn transient results using the Case 4 initial operating

(

l i

's ' >

y JT:

conditions revealed'that the 9 channel model again produced conservative i s

MDdBRs (see-Table'4-5); therefore, the 9 channel- model will be used to assess any-future Oconee Nuclear Station reloads involving transition cores.

i 4.4 RESULTS

SUMMARY

1 In all of the studies and comparisons performed between the 64,'9, and 8 channel models, the 9 and 8 channel models consistently produced conservative results. Duke Power Company will use the smaller channel models to perform thermal-hydraulic; analyses since the 64 channel model requires an extensive amount of. computer processing time. The 64 channel model will be used if a i l

situation arises which req 61res the 1-2% conservatism currently available with l g the 8 and 9 channel models. The larger model would only be used for cycle 1

specific evaluations requiring the additional margin. j 5 '. 0 .VIpRE-01 DATA l'

'{

The fuel assembly data used to develop each of the input parameters, such-as- '

' flow area, wetted and heated perimeters, centroid distances, and gap widths s

are given in Table 3-1. Other important VIPRE-01 input is discussed in-detail in the subsectio,ns which follow.

5.1 AXIAL NODING Given the axial power shape and a specified heated rod length, VIPRE-01 determines the axial power factor fo'r each axial nede, ref. 2. The node length determines how well the code approximates the axial power shape, the shorter 1

L __ -

p w 7

.n .

L the node lenCth,- the better the approximation of.the curve. Volume 4 of.the  !

J VIPRE-01. manual states as a general ~ rule that nodes on the order of 2 or 3-l inches long are recommended in the region where MDNBR is likely to occur, ref. ,

l

2. Calculations involving node sizes smaller than 2 or 3 inches require more l

y computer processing time without gaining significant increases in the

! accuracy of results.  !

Results of an axial node length sensitivity study performed with the 8 channel-I steady-state model are presented in Table 5-1. A comparison was made between a i

three-inch node length, uniformly applied to the axial length of the rod from j 4.125 to 142.125 inches, and two ranges of two-inch axial node lengths applied B to the rod at elevations ranging from 32.125 to 94.125 inches and 81.125 to I

143.125 inches. As Table 5-1 shows, the three inch node lengths produced slightly conservative MDNBRS; therefore, the three-inch node -length will be used for all Oconee Nuclear Station thermal-hydraulic analyses. ,

5.2 ACTIVE FUEL LENGTH Uranium fuel both densifies and swells when irradiated. Densification effects are predominant at low burnup and swelling effects are predominant at higher burnup. Fuel densification decreases the active fuel length while fuel swelling tends to increase the active length. ,

P-e 6

]

gl -

5. 3 CENTROID DISTANCE The location of each subchannel or channel'is defined by numbering all the l

. channels, inputting connecting channel numbers, and defining'the distance between centroids of adjacent channels. The centroidal distance in a normal I

i l

square array, is the-subchannel pitch. The centroidal distance determines the $

[- length over which the crossflow exists and defines the lateral pressure  ;

gradient in the crossflow momentum equation. The centroidal distance for a l channel cut by a line of symmetry is the same as the centroidal distance for the complete channel, ref. 2. For the lumped subchannels, the centroidal distance is increased from its individual subchannel value in proportion to the

,! number of rod rows between channel centroids. Likewise the centroidal distances between lumped assemblies is increased in proportion to the rows of assemblies between the lumped channel centroids.

g 5.4 EFFECTIVE CROSSFLOW GAPS Crossflow resistances are calculated by inputting connecting channel information and crossflow gap widths. The product of the gap width and the axial node length defines the lateral flow area between channels. The gap widths are easily calculated given the rod pitches and diameters. The gap width for a fuel assembly or any lumped channel is the sum of the subchannel gaps through which the two assemblies communicate.

u Le 3

h 5.5 SPACER GRIO FORM LOSS COEFFICIENTS ,

l i

ss ,

Form loss coefficients are used to account for the unrecoverable pressure losses caused by the abrupt variation in flow area and turbulence at a spacer grid. The. Mark-BZ fuel assemblies have six intermediate zircaloy spacer grids

.and two inconel end grids. Form loss coefficients determined for the different types of subchannels (i.e. unit, thimble tube, peripheral, instrument guide tube, and corner channels) and for the overall grid are used in the thermal-hydraulic analyses. Spacer grid form loss coefficients are developed from full size fuel assembly flow tests performed by the vendor. Individual subchannel l I form loss coefficients are determined analytically by the vendor from the overall grid form loss coefficients. i 5.6 CORE BYPASS FL3W Core flow is equal to the total reactor coolant system flow less the bypass flow,.which is defined as that part of the flow which does not contact the effect!ve heat transfer surface area. The bypass flow paths are the 1) core shroud, 2) core barrel annulus, 3) control rod guide tubes and instrument tubes, and 4) all interfaces separating the inlet and outlet regions of the reactor vessel. A typical value of the design bypass flow is 9.0%; however, the bypass flow rate is dependent on the number of control rod and burnable poison rod assemblies in the core since they act as guide tube plugging devices. The actual core bypass flow must be verified each cycle to assure that it is less than that used in the generic analysis.

m '

1 Li j

-i g.

L O 5.7 INLET FLOW DISTRIBUTION L

C L ,' -VIPRE-01 allows the user to specify the core inlet flow maldistribution. The Oconee core thermal-hydraulic analyses include a 5% reduction in inlet flow 4

{

to the hot assembly.to conservatively represent the results obtained in B&W's L a

1/6-scale Vessel Model Flow Test, ref 1. More restrictive flow maldistri- j bution factors are used for operation with less than four reactor cool".nt -!

l pumps. Table 5-2 shows that the use of a 5% inlet flow maldistribution  !

produces slightly conservative results compared with a uniforr. inlet flow distribution.

5.8 VIpRE-01 CORRELATIONS Empirical correlations are used in the VIPRE-01 code to model turbulent mixing _ >

E and the effects of two phase flow on friction pressure losses, non-equilibrium ,

i subcooled boiling, and the relationship between the quality and void fraction.

The correlations which have been selected for use in the Oconee ,

thermal-hydraulic analyses are discussed in the subsections which follow, b

s 5.8.1 FRICTION PRESSURE LOSS

( ,

Pressure losses due to frictional drag are calculated for flow in both the '

( axial and lateral directions. In the axial direction the friction pressure loss-is calculated by L

y.

fi ,

-i r

]

G8 v'

. -dP

' IZ

• f i: 2g D c h m; ,

W where f= friction f actor determined from an empirical correlation

-defined by user input 1 G= Mass flux, ibm /sec-ft*

{

v'= specific volume for momentum, ft.8/lbm g= force-to-mass units conversion factor,~32.2 lbm-ft/lbf sec2 l c

rn - -

4 Dh = hydraulic diameter based on wetted perimeter, ft.

l o

, j Based.on the recommendation in ref.'2,-vol. 4 of the VIPRE-01 manual, the -

default Blasius smooth tube friction factor expression j f = 0.32 Re-0.25 + 0.0 1

1 will be'used to calculate the friction: pressure loss for turbulent flow.

l Based on sensitivity study results given in Table 5-3, the friction pressure loss for two phase- flow will be calculated using the EPRI two phase friction 'l multiplier.

In the lateral direction the pressure loss is treated as a form drag loss that

- is calculated by AP = Kg lwlw v' 2S2g' 1 where Kg = loss coefficient in the gap between adjacent channels w = crossflow through a gap, Ibm /sec-ft v' = specific volume for momentum, ft 2/lbm S = gap width, ft g

c = f rce-to-mass units conversion factor, 32.2 fb / ec a l

4

m ..  ;;j, r

( ,

W iWhen rod arrays are modeled as lumped channels the effective crossflow resis-tance is the sum o'f the resistance of theirod rows between the lumped channel-centroids. .The lateral loss coefficient becomes .

Eij = N Kg where N is the number of rod rows between lumped channels and gK is the nominal 1 drag coefficient for a single gap. Crossflow resistance coefficients,are not

' precisely known, but sensitivity study results discussed in Volume 4 of ref. 2

-  ! show that for applications where the axial flow'is predoniinant relative to crossflow, crossflow' resistance has an-insignificant effect on mass flux and DNBR. A-subchannel drag coefficient, gK , of 0.5 will be used with the coeffi- l cient_for-lumped channels calculated internally by the code based on the input l 1

centroid distances between' lumped channels and the standard subchannel fuel rod pitch, i

1

=

5.8.2 TURBULENT MIXING The VIPRE-01 transverse momentum equation includes terms to calculate.the i exchange of momentum between adjacent channels due to turbulent mixing. Two a turbulent momentum-  !

parameters must be input to include turbulent mixing:

factor (FTM) and a turbulent mixing coefficient (B).

I- l l

The turbulent momentum f actor (FTM) defines how efficiently the turbulent cross-

- I. flow mixes momentum. FTM can be input on a scale from 0.0 to 1.0, where 0.0 l indicates that the crossflow mixes enthalpy only and 1.0 indicates that it l I l

P d - -

4 Ev .r--%. x. 4 w 3

[

I 6 . mixes enthalpy and momentum with the same strength. In actuality, some propor-1 l

~ tion of enthalpy an@ momentum mixing does take place; therefore, turbulent momentum f actors of 0.8 and 1.0-are probably. more representative'of actual-I ..# . crossflow effects. Sensitivity studies discussed >in Vol. 4 of ref. 2 show t

that' changes in the. fraction of momentum mixing have negligible impact on.the  ;

flow field; therefore, FTM = 0.8 is recommended, ref. 2. Sensitivity studies

'I

- y.

o using the B channel model were performed for the Case 1 and 2 operating o

-conditions given 1niTable 4-1.

~

The runs.were made using FTM = 0.0, 0.8 and ,

1.0. The results of the. analyses are presented in Table 5-4. Since the L

[ results show that MDNBRs. for an FTM = 0.8 ' lie between the MONBRs for FTM = 0.0 t

F and 1.0, and since F1H = 0.8 more realistically assumes some momentum mixing, an FiM = 0.8 will'be used in all future Oconee thermal-hydraulic analyses.

l I- '

Turbulent crotsflow between adjacent channels is calculated by l

L w' = SSG I

a where w' f s the turbulent flow per axial length, S is the turbulent mixing lg i coefficient, S is the' gap width, and G is 'the average mass. flux of the adjacent channels. Based upon vendor predictions of mixing test results, a mixing coefficient of , will be used for all Oconee Nuclear Station core thermal-l hydraulic analyses.

)'

5.8.2 TWO-PHASE FLOW CORRELATIONS Two correlations are used in VIPRE-01 to make two phase flow predictions. The first correlation is referred to as the subcooled void correlation. It uses a i .

I '

I ou:J o _mortel to calculate the flowing vapor mass f raction including the effects o.' .abcooled boiling. Once the flowing vapor mass fraction is calcu-lated, the bulk void correlation is applied to calculate the void fraction

~

itcluding aity ef fects due to slip, ref. 2, Vol.1.

I sensitivity studies were performed using three different cenbir,ations of ,

subcooled void and bulk void correlations to evaluate their effects on the hot channel local coolant conditi6ns and MDNBR.

Subcoo nd Void Bulk Void Levy Zuber-Findlay Levy Smith EPRI EPRI I

The hot channel local coolant conditions and MONBRs are given in Table S-3 for the Case 1 and 2 operating conditions. As Table 5-3 shows, the combination of the Levy subcooled void correlation and the Zuber-Findlay bulk void correla-tion yields slightly conservative resu'.u. Section 3.3 of Vol. 4 of the VIPRE-n1 raanual, ref. 2, presents the results of VIPRE-01 predictions of the Martir, void fraction tests at high pressure (1565 and 1991 psia). Of the two phase flow correlations evaluated, the Levy /Zuber-Findlay combination compared most fevorably with the test results. The Levy subcooled void correlation and the Zuber-Findlay correlation will be used for Oconee thermal-hydraulic analyses.

I I '

I 1

I 5.9 REFERENCE DESIGN POWER DISTRIBUTION l

The reference design power distributions are shown in Figurt 5 4-1 thrcugn 4-5.

I 1

The power distributions were designed to be conservatively high and relatively flat in the vicinity of the :.>t subchannel. The pin power peaking gradient within the area of the hot subchannel is approximately 1%. The pin power distribution was verified to be conservative by comparison with predicted physics power. distributions. The reference design power distribution was N

developed using a radial-local hot pin peak, F,H, of 1.714 and an assembly power of 1.6147. Tne F H = 1.714 is the same reference pin peak used in the methodology discussed in reference 1. The two pump coastdown transient is analyzed as discussed in Section 6.6 using the reference design power 1 distribution. A different design power distribution may be used to add or delete margin in the two pump coastdown transient. As discussed in Section ,

6-5 and 6-6 maximum allowable peaking (MAP) limits are calculated to define combinations of radial and axial peding that provide equivalent DNB protec-tion.

5.10 MLIAL POWER DISTRIBUTION The axial power, shape used to develop the results presented in this report was a chopped cosine axial power shape. Predicted and actual axial power shapes vary for cycle specific reloads and transients since they are functions of control rod .:ositions, xenon transients, et:. 'he effect on DNB of different l axial flux shapes is taken into account as discussed in Section 6.5.

-I I

I' . - _-

i I

I A routine has bean added to the VIPRE-01 code to generate axial power shapes with inlet, symmetrie, or outlet peaks. The routine is based on the following constraints on an axial power shape I

F(x) is continuous from (B,E)

F'(x) is continuous from (B,E)

I _L f F(x) dx = 1.0 E-B' B where F(x) = axial power shape as a function of the I axial location. x I B,E = beginning and ending normalized location of the active fuel length I The reference 1.65 axial flux shape is generated using the new axial shape routine.

5.11 HOT CHANNEL FACTOR The local heat flux factor, F q

, and the power factor, Fq , are conservatively applied to the' hot subchannel (i.e. , the instrument guide tube subchannel) of the hot assembly to compensate for possible deviations of several parameters from their design values. The local heat flux factor F " = , ref. 1.,

incorporates variations in pellet density, pellet cross-sectional area, weight per unit length, local enrichment, and local outer clad diameter. It is only used in the computation of the surface heat flux of the hot pin when calculat-i I ._ . _. _ . . . .

s n

L-L- ing the DNBR for the hot subchannel, ref. 1. The power factor, F

• Q - .

ref.1., accounts for variations in average pir power caused by differences in

{- the absolute number of grams of U-235 per rod. The loading tolerance on U-235 per fuel stack and variation on the powder lot mean enrichment are considered

{

in determining the factor, ref.1. Fq is applied to the heat generation rate h- of the hot pin of the hot subchannel. Both factors are applied to the hot subchannel during steady-state and transient thermal hydraulic analyses.

E- However, in the determination of maximum allowable peaking limits (as will be discussed in Section 6.5), the local heat flux factor, F ,"q is increased by

{

applying two additional penalties. First, a 1 Secondly, a penalty is applied to account for axial nuclear uncertainty, ref.1. The F " for calculating maximum allowable 9 .

peaking limits for Mark-BZ fuel is ref. 1. j

[ , _

5.12 FLOW AREA REDUCTION FACTOR The hot subchannel flow area is reduced by to account for variations in as-built subchannel coolant flow areas.

[ 5.13 BWC CRITICAL HEAT FLUX CORRELATION The BWC critical heat flux (CHF) correlation, ref. 4, will be used for Oconee thermal-hydraulic analyses. The BWC correlation was originally developed for B&W 17 x 17 Mark-C fuel. Subsequently, as discussed in ref. 4, B&W showed that the BWC correlation can be used for 15 x 15 Zircaloy grid Mark-BZ fuel.

J

(

i F

1he BWC correlation was developed by B&W using the LYNX 2 crossflow computer code, ref. 6. To justify use of the BWC correlation with the VIPRE-01 code the Zircaloy grid CHF test results given in ref. 4 were predicted using VIPRE-01 l

I and compared with B&Wi s LYNX 2 results. The VIPRE-01/BWC results for all 211 i i

data points were used to determine a DNBR limit which provides a 95%

probability of precluding CNB at a 95% confidence level.

I Figures 5-1, 2, and'3 show the B&W LYNX 2 versus VIPRG-01 calculations for the BWC Measured-to-Predicted (M/P) CHF, mass velocity, and quality at the CHF location, respectively. These figures show that the VIPRE-01 coolant condi-tions and BWC CHF predictions are essentially the same as B&W's LYNX 2 predic-tions. Figure 5-4 shows the measured CHF versus the VIPRE-01 predicted CHF for all 211 data points demonstrating that the overall prediction of the correlation is correct. The ratio of measured-to-predicted CHF is plotted versus local quality, mass velocity, and pressure in Figures 5-5 through 5-7, respectively. These figures show that there is no bias in the correlation relative to the important fluid parameters. Calculation of the design DNBR limit is based on the assumption that the M/P CHF values are normally I distributed. This was verified statistically using the 0-prime test. '

I A DNBR limit is calculated so that cores can be dasigned to operate below the CHF. The DNBR limit is the lowest DNBR that can be calculated (for any core condition) for the limiting pins in the core and ensure with 95% confidence that 95% of the limiting pins are not in film boiling. The design DNBR limit was calculated using the following expression developed in ref. 4:

i @

s :r L

I 1.0 DNBR Limit = gfp _ g

,y,p, where R/P = mean measured-to-predicted CHF ratio (f s' K

N,i,P = ne-sided tolerance factor based on

[' degrees of freedom (N), confidence level (T), and portion of population protected (P),

o = standard deviation of measured-to predicted CHF kSci values For the VIPRE-01/BWC combination the design DNBR limit is 1.161. The ONBR ,

limit is calculated as shown in the following.  !

(

N = 211 R/P = 1.0076

[~

E 211, 0.95, 0.95 = 1.832 o = 0.0797 DNBR Limit = 1.0F 6 - 18 2 (0.0797) = 1.161

{_

For all Oconee thermal-hydraulic analyses using VIPRE-01 and the BWC

[

correlation, a design DNBR limit of 1.161 + margin will be used.

The applicable range of variables for the BWC correlation are:

[L 1

E F

, Pressure 1600 < P < 2600 psia

' Mass Velocity 0.43 < G < 3.8 Mlbm/hr-ft8

-Quality -0.20 < X < + 0.20 6.0 QfffjLI}4ERMAL-HYDRAULIC ANALYSES 6.1

SUMMARY

l A thermal-hydraulic analysis of the Oconee reactor cores is necessary to defiu the core thermal margin and acceptable operating limits.

The crossflow code thermal-hydraulic analysis methods used to derive the core safety and operating limits are the same as the previously approved methods in ref.1. The safety and operating limits are used to ensure core protection against anticipated transients and steady-state operation. Some of the Reactor Protection System (RPS) trip functions incorporate these safety limits as setpoints which would trip the reactor prior to exceeding the thermal design limits. A list of RPS trip functions is given in Table 6-1. The safety limits are derived from thermal-hydraulic analyses based upon various combinations of power, pressure, temperature and flux-to-flow limit.. A new analysis is performed for a reload g core whenever there is a significant change in the fuel design, a change in the ,

) input assumptions of the generic analysis, or a change in the regulatory I

criteria.

6.2 THERMAL-HYORAVLIC DESIGN CRITERION ,

I The thermal-hy'draulic design criterion is that no core damage due to DNB occur during steady-state operation or anticipated transients. DNB is defined as

m V

the point where bubble generation on the clad heat transfer surface forms an insulating blanket over the surface heating area, thus, causing a large clad

(, surface temperature rise. The departure from nucleate boiling ratio (DNBR) is defined as the ratio M the critical heat flux at a point on the rod to the k actual heat flux at the same point. DNBR is calculated using Babcock and

{ Wilcox's BdC Correlation. The minimum DNBR (MDNBR) is limited to 1.161 +

margin as previously explained in Section 5.13.

5.3 CORE SAFETY LIMITS Core safety limits are determined to protect the core during steady-state operation and anticipated transients. The coro safety limits prevent overheating and possible rupture of the cladding which would release fission products to the coolant. Fuel clad overheating is prevented by restricting operation to within the nucleate boiling regime where clad temperature is only slightly above the coolant temperature. Two core safety limits directly provide DNB protection:

1. Pressure - Temperature Envelope Figure 6-2

2. Power - Power Imbalance Limits Figure 6-1 6.4

) PRESSURE-TEMPERATURE ENVELOPE The Pressure-Temperature (PT) envelope defines a region of allowable operation in terms of reactor coolant system (RCS) pressure and vessel outlet temperature. j l

< [

r- _.;.

p The PT envelope provides ONB protection as well as protection for the RCS.

The three reactor trips that define the region of allowable operation as shown 1

in Fig. 6-2 are:

1. . High temperature trip

[ 2. Low pressure trip

3. Variable low pressure trip To ensure that the PT envelope provides DNB protection, PT curves are determined for reactor coolant (RC) pump operation. The PT curves-are the combinations of RCS pressure and vessel outlet temperature that yield the design DNBR. limit (BWC correlation limit plus margin) or the BWC correlation quality limit. The PT envelope must be more restrictive than the most. limiting PT curve as shown in Fig. 6-2. 1 The PT curves are calculated using che 8 channel model discussed 1.1 Sectioi.  !

4.0. The VIPRE-01 input that is used to calculate the generic PT curves is discussed in subsections 6.4.1 through 6.4.4 which follow.

4 i

6.4.1 REFERENCE POWER DISTRIBUTION

(

t The reference power distribution discussed in Section 5.8 and shown in Fig. 4-3 and 4-4 is used to calculate the PT curves. The reference axial power profile used to calculate the PT curves is a symmetric chooped cosine with a peak to average value of A different reference axial power shape may be used as 1

necessary to ensure that the MDCit during a two pump coastdown transient is greater than the design DNBR limit. The axial power shape can change as a 1

l m m. . .

n

'Y s-result of rod motion, power change, or due to a xenon transient.,. Power -

power imbalance limits, ref.1, provide protection for the core from the effects of skewed axial power distributions. To determine the power power imbalance limits maximum allowable peaking (MAP) limits are calculated as discussed in Section 6.5. i

'6.4.2 CORE POWER The maximum power level for 4 pump operation,112% FP, is set by the high flux trip setooint with adjustment made for uncertainties 6nd margin. The maximum power level for ,

The PT curves are caiculated for the maximum power levels for pump operation.

6.4.3- RCS Flow The generic Oconee thermal-hydraulic analyses will be based on an RC' flow of 366,080 gem (104% of the design flow of 88,000 gpm/ pump) which is lower than the measured flow for any of the three Oconee units. This value could be j increased for a cycle specific analysis to take credit for the flow margin at a particular unit.

6.4.4 CORE INLET TEMPERATURE For a given core power, flow (number of operating RC oumps), and pressure the vessel outlet temperature at which the MDktsR equals the design DNBR limit 4 u

7 .

• . m a1

)

defines a point along a I: a fe. VIPRE-01 is run at several pressures to

(

$- determine the' core inlet temperatures that yield the design DNBR limit.

I 6.5 GENERIC MAXIMUM ALLOWABLE PEAKING _ LIMIT CURVES II In order to provide DNB protection for axially assymetric and symmetric power distributions, a series of maximum allowable. pin peaks ?.re calculated such that the MONBR limit is obtained. Maximum allowable peaking (MAP) limits are calculated in the form of lines of constant MDNBR for a range of axial peaks with the location of the peak varied from the bottom to the top of the core.

Thisisperformedforaxialpeaksofl ,

l I ,.

The axial peaks were generated using the new axial shape '

a routine discussed in Section 5.10. The maximum allowable peaks arc nultiplied i I by their respective axial peaks to obtain Total Maximum Allowable Peaking ,

Limitt (i .e. , MAP limits). The MAP Limits are plotted for each axial peak t.nd l

X/L to form a set of MAP limit curves. The MAP limits provida a basis for ,

equating the symmetric and asymmetric power distributions. MAP limits are q compared in a maneuvering analysis with peaks resulting from design power

[ transients as discussed in ref. 1, Two sets t* generic MAP lim t curves are

' determined. One set is used to determine the DNB operational offset limits, l

and the other . set is used to determine the Reactor Protection System (RPS) DNB offset limits. <

m.

1 t .

E .

Operational MAP limit curves are developed in the same manner as the RPS MAP limits based on the two pump coastdown transient as explained in the following

~

section. A typical set of Operational MAP limit curves generated with the 8 channel model is shown in Figure 6-5.

If any negative peaking margins (predicted peaking greater than the appropriate 1'

MAP limit) are determined during a maneuvering analysis, ref. 1, the MONBR will be calculated for the limiting predicted power distribution. The predicted radial power distribution and axial flux shape is input directly into the VIORE-01 code, i

6.6 PUMP COAST 00WN TRANSIENT ANALYSES The flux / flow trip prevents the core from violating the DNBR criterion during a loss of one or mere reactor coolant (RC) pumps. DNB protection is also provided by pump monitors which provide an immediate trip signal on loss of electrical power to the pump motors. The pump monitors at Oconee will trip '

the reactor for a loss of two or more reactor coolant pumps from above 55%

power. Thus, it is conservative to determine the flux / flow trip setpoint assuming the -

The two pump coastdown transient is analyzed using VIPRE-01 to assure that the 1.'.61 + margin design DNBR limit is not violated af ter the loss of one or more

f-RC pumps. The VIPRE-018 channel model was used with rods 1-4 (see Fig. 4-3) modeled using the conduction model available in the VIPRE-01 code. For f

steady-state analyses " dummy" rods are used with the power (heat flux) applied  %

directly to the coolant. During a transient; however, once the reactor is tripped, the neutron power generated in the fuel decreases rapidly, but the

' thermal power reaches the coolant with some * .ne delay through condsction and

(

convection from each fuel rod. To model the conduction and stored energy effects the VIPRE-01 conduction model is used.

Conduction through the gap between the fuel pellet and the clad is determined i

using the gap conductance model in VIPRE-01. The NRC concluded in the VIPRE-01 SER, ref. 3, that the fuel rod heat conduction model is acceptable for licers'ng analyses.

To select the input for the conduction model sensitivity studies were performed varying the following input parameters:

Pellet / clad gap Gas composition Pellet radial power profile l Sensitivity studies discussed in the VIPRE-01 manzal, ref. 2, shew that the gap conductance model is most .. sitive to the specified gap width. Maximum and  !

uinimum gaps, based on preditted pellet densification and clad creepdown, were j evaluated. Stt. dies were also performed to determine tne effect of the fill gas composition and fission gases released into the gap on the transient DNBR results. Cases were also run to determine the effect that the pellet radial power profile has on the transient DNBR results.

O w -,

( e i

Based on the sensitivity study results, the generic VIPRE-01 two pump coastdown analyses.will be performed using the following conservative conduction model input:

Nominal Rod OD Nominal Clad Thickness Maximum Pellet / Clad Gap Minimum Pellet Diameter 1

Minimum Prepressure Nominal Plenum Volume Helium and Nitrogen Fill Gas Uniform Pellet Radial Power Prth ie Heat transfer correlations are used by the VIPRE-01 code only when the conduc-tion model is specified. Convection and nucleate boiling correlations are selected since only conditions up to the point of DNB are normally analyzed. i The default single-phase forced convection correlation, the Oittus-Boelter correlation with the leading coefficient compatible with the EPRI void model, l

will be used for Oconee pump constdown analyses. The Tho9 subcoaled and ,

satursted nucleate boiling correlation will be used. A sensitivity study showed that the choice of nucleate boiling correlations made very little d$fference in the pump coastdown DNBR results.

I ,

Generic pump coastdown transient analys,es are performed for each unit to verify that the flux / flow trip setpoint provides ONB protection for the loss of one or b more RC pumps. The flux / flow trip setpoint also provides overpower protection  !

y ij:,

c

[ for three and two pump steady-state operation. The generic analyses are

' N performed using the reference power distribution (F H = 1.714) shown in Fig. 4 3 and 4-4 along with the cosine reference axial power shape. The 1

reference radial and axial power distribution may be changed to add or delete

, margin during tr.a two pump coastdown transient. To ensure that the DNBR criterion is met during a pump coastdown transient with any possible axist flux shape, Operational MAP limits are ca'.culated as previcusly discussed in stction 6.5. -

e 1

i l ~

B

7.0 REFERENCES

l I

1. Oconee Nuclear Station Reload Design Methodology II, DPC-NE-1002, Duke i Power Company, Charlotte, NC, March 1985, 1

I 2. J. M. Cuta, et, al., "VIPRE-01: A Thermal-Hydraulic Code for Reac+.or i l

5 Cores," EPRI-NP-7311-CCM, Vol. 1-5, Battelle Pacific Northwest l Laboratories, July 1945.

I i

3. Letter from C. c.. Rossi (NRC) to J. A. Blaisdell (VGRA), " Acceptance for l

I- Referencing of Licensing Topical Report, VIPRE-01: A Thermal HydraMic Analysis Code fer Reactor Cores," EPRI-NP-2511-CCM, Vol. 1-5, Mav 1, 1986.

4. BWC Correlation of Critical Heat Flux, BAW-10143-A, Babcock and Vi!cox,

. Lynchturg, VA, Aoril 1985.

l 1

5. LYNX 2: Subchannel Thermal-Hydraulic Analysis Program, BAW-10130-A, L Babcor.k and Wilcox, Lynchburg, VA, July 1985.

I r

I

'I I- l l

I l

a TABLE 3-1. MARK-BZ FUEL ASSEMBLY DATA (TYPICAL)

GENERAL. DJEL SPECIFICATIONS Fuel roc ciameter, ir. (Nom.) 0.430 Thimble tube diameten , in. (Nom.) 0.530 Instrument tube diameter, ia. (Nom.) 0.554 1 Fuel roa pitch, in. (Nom.) 0.5663 Fuel assembly pitch, in. (Nom.) 8.587 Fuel roc letth, in. (Nom.) 153.7 GENERAL FUEL CHARACTERISTICS M

. Grids: Material Quantity Location Inconel 2 Upper and Lower Non-mixing Vane Zircaloy 6 Ir.te rr4ediate Non-mixing Vane Fuel' rods: Material Quantity Zircaloy-4 208 Fuel C.cle Design Assembly Features Fuel Assy. Mark Math Mark Mark Designation: B42 B5Z B6 B7 3

Features: l l

l l

M

s sp f

s -

TABLE 4-1. OPERATING CONDITIONS r- ,

Inlet L

CASE

• Power Flow Pressure Temperature

%  % PSIA *F

~~

f ,

t 2 2.

3 4 <

• All cases were performed using a axial peak unless otherwise noted.

-i s

q M M M M1, M M M M M M W W W W W W W M M i

IAHLE 4-2 COMPAR1 SON OF 64 CilANNEL AND 8 CilANNEl MODEL STEADY-STATE RESULTS (TYPICAL) 1 f

EXIT QUALITY MASS VELOCITY MJDN

{MtBM/IM-FT J Ch. 1 Ch. 2 Ch.3 Ch.1 Ch 2 Ch 3 Ch. 1 Ch . 2 Ch. 3 0.114 0.102 CASE 1* 0.121 2.86 2.06 2.03 0.124 0.115 0.102 l 1.412 1.490 1.569 2.00 1.99 1.558 1.80 64 Channel Model 1.395 1.476 8 Channel Model 0.087 0.081 0.070 CASE 2 d

1.93 2.15 2.19 0.081 0.069 l 1.750 2.17 0.088 1.601 1.674 1.90 2.13

• 64 Channel Model 1.594 1.670 1.749 y 8 Channel Model 0.154 0 .1.18 1.53 0.163 CASE 3* 1.37 1.54 0.166 0.156 0.138 1.243 1.338 1.454 1.48 1.47 1.433 1.32 64 Channel Moden 1.216 1.315 8 Channel Model Table 4-1.  !

c.

3 are in reference to the operating conditions given in a) Cases l

l

s b

I, TABLE 4-3. COMPARISON OF 64 CHANNEL AND 8 CHANNEL MODEL TRANSIENT RESULTS (TYPICAL) r'; -

L 64 Channel Model 8 Channel Model Time Channel 1 Channel 1 (sec) MDNBR MDNBR-0.0 1.833 1.830 0.5 1.813 1.807 1.0 1.774 1.767 1.5 1.715 1.708 2.0 1.645 1.636 2.5 1.569 1.558 2.7 1.528 1.517 h

2.9 1.501 1.489 3.1 1.471 1,454 3.3 1.436 1.420

- 2. 4 - 1.396 1.376 3.5 1.356 1.333 3.6 1.331 1.308 3.7 1.306 1.285 3.8 1.280 1.260  !

3.9 1.261 1.238 4.0 1.241 1.221 1.216

(- 4.1 1.234 1.221 4.2 1.245 4.3 1.281 1.250

[

( ,

I l

i

ungr ums aux

.:r. ;

Mm aus -.i

.2

{

+;. -

. TABLE 4-4. COMPARISON OF 64 CHANNEL AND 9 CHANNEL TRANSITION CORE MODEL STEADY-STATE RESULTS / TYPICAL)

MASS VELOCITY EXIT QUALITY _

PON8R 2

(MLBM/HR-FT )

Ch.1 Ch.2 Ch.3 Cn. 1 Ch.-2 Ch.3 CASE 1* Ch. 1 Ch. 2 Ch. 3 1.86 2.06. 2.03 0.121 0.114 0.102 64 Channel Model 1.412 1.490 1.5fd 0.115 0.102 1.560- 1.80 2.00 -1.99 0.123 9 Channel Model 1.398 1.479 CASE 2' 1.93 2.15 2.19 0.087 0.081 0.070 64 Channel Model 1.600 1.674 1.750 0.069-L, 1.90 2.13 2.18 0.087 0.081 p 9 Channel Model 1.597 1.673 1.751 CASE 3"

'k.37 1.54 1.53 0.163 0.154 0.138 64 Channel Model 1.243 1.338 1.454 0.138 1.32 1.48 1.47 0.165 0.155 9 Channel Model 1.220 1.318 1.436 a) Denotes Cases 1, 2 and 3 are operating conditions frca Table 4-1.

m e

(

I TABLE 4-5. COMPARISONS OF 64 CHANNEL AND 9 CHANNEL k

.l. TRANSITION CORE MODEL TRANSIENT RESULTS (TYPICAL) 64 Channel Model  ? Channel Model Time .

Channel 1 Channel 1.

(sec)_ MONBR MONBR 0.0 1.834 1.831 0.5 1.814 1.808 1.0 1.775 1.769 1.5 1.715 1.710 2.G 1.646 1.639 2.5 1.571 1.562 1 h, 2.7 1.529 1.519 2.9 1.502 1.491 3.1 1.472 1.456 3.3 1.438 1.422 3.4 1.336 1.379 3.5 1.359 1.337 3.6 1.333 1.301 3.7 1.309 1.263 3.8 J.283 1.262 3.9 1.263 1.243 ,

4.0 1.244 1.229 4.1 1.238 1.224 4.2 1.249 1.230 4.3 ----- 1.256

{ 4.4 -----

1.302 c

n

z

.W' W g~

'E E E- E E E .E ~ E W .W 'M ' g g g g TABLE 5-1. 8 LHANNEL MODEL AXIAL N0DE' LENGTH SENSITIVITY STUGY (TYPICAL).

Node Axial Size Mode Study Chanr.cl 1 PON8R'S Operating X

-Ccnditions" Peak L (in.) Elevation (in.) PON8R Elevatinn (in.)

1.65 0.5 3 4.125-142.125 b 97.1-100.1 (CASE 1) 99.1-101.1 1.65 0.5 2 81.125-143.125 ,

(CASE 2) 1.65 0.5 3 4.125-142.125 b 'I-9#'I 1.65 0.5 2 81.125-143.125 95.1-97.1 1.70 0.1 3 4.125-142.125 64.1-67.1 (CASE 1) 66.1-68.1 1.70 0.1 2 32.125-94.125' 1 6

1.70 0.1 3 58.1-61.1 (CASE 2) 4.125-142.125" 58.1-60.1 i 1.70 0.1 2 32.125-94.125 ,

i Notes a) Operating conditions from Table 4-1.

b) 4.125-81.125 in. range modeled with'three-inch nodes.

c) 4.125-32.125 and 94.125-143.125 in. ranges modeled with three-inch nodes. ,

I i- - - - - . . . . .- . . . . . . . , _ _, . . . _ . , . , , ,_ ___, _ _ _ _ ,__ ,_ ___ _ _ _

)

TABLE 5-2. B CHANNEL MODEL INLET FLOW

. DISTRIBUTION SENSITIVITY STUDY (TYPICAL) n Operatino Condition Case 1 from Table 4-1.

i Percent Flow MONBR to Hot Assy. - Channel 1 Channel 2 Channel 3

~

O eh 2

(MLBM/HR-FT )

g W EXIT OUALITY

~

l

( .

L L

[- )

11

- - - - - ~ vv v

(

l TABLE S-3. 8 CHANNEL MODEL TWO-PHASE FLOW CORRELATION AND FRICTION MULTIPLIER SENSITIVITY STUDY (TYPICAL)

Sub- Two-Phase Exit Void Exit.

)perating Cooled Bulk Friction Mass-Velocity Fraction Quality Void. Multiplier PONBR

onditions# Vold (MLIM/HR-FTr)

ASE 1:

Ch. 1 Ch. 2 Ch. 3 Ch. 1 Ch. 2 Ch. 3 Ch. 1 Ch. 2 Ch. 3-Ch. 1 Ch. 2 Ch. 3 _

LEVY ZUBR EPRI

-LEVY ZUBR HOMO LEVY SMIT HOMO EPRI EPRI EPRI EPRI EPRI HOMO CASE 2:

LEVY ZUBR EPRI LEVY ZUBR HOMO LEVY SMIT HOMO EPRI EPRI EPRI EPRI EPRI HOMO a) Denotes operating conditions from Table 4-1.

m w v v; w. . .

O m. rm if 4

i e

Tat.le ' 5-4.. 8 CHANNEL MODEL TUR9ULENT MDMENTim FACTOR SENSITIVITY STUDY (TYPICAL)-

MASS VELOCITY Opercting S IqDNOR LOCATION EXIT QtMLITY MD888R Conditions FTM.

2 (ML8M/HR-FT )

Ch. 3 Ch. 1 Ch. 2 Ch. 3 Ch. 1 Ch.'2 Ch. 3 Ch. 1 Ch. 2 CASE 1: .

0.0 0.8 1.0 b

CASE 2:

0.0 0.8 1.0 .

W O

_ , . ~ , _ _ _

- -.n--..

. m Ms 5

m L

n . . , . .

L. ,

i. - 1 1: $ -

1 I-l. .

! I-

I l't

I- .  ! E l:

- i i .

ril 8

a .r3 .:r n -:1- .r . I g. . I

3. All Ji I ig i E

!':[

-  : : I

r-it-l$

4-r

a. .

3

Tr I

3-I "

3- -

l ' -

i=

t:

3-s.r  :

c 3,23 t-3 r si 5

a -:

r- l-8-1 c

1 tt

. .J.j a .- .  ;:  :

..: y .: =.-i. .-:,2I

a. !  :. 34 I

1

- t .

3 i ir f5.y :i it: fj: 58  !!  ! I Ig

::: : si  :

['4 4j E. t 4jj t 4}jg LlJ

& L t  : j, I ,

1! - -. .: 4: : 4 .I 4 af 41 ^

l m

5. .

. .r ~

I 4 I\ ~

E 8

$ l.1 i. :

w . .. g2 a.

g I f f. t

• l' 3 -2 2 r

-  : t t- t .

I  ;

a p!

t.. i.

1

. .i .f I. .i. j ,n i.

1.

i El

g. .s I

g -\

7 x.g L 8 a

m l.l 8 -l 13 1  :

1 l m ii  : ii  : j. .

5 2 I. I. =l  ;

e .

s. I. -

1

. -L Li & .: & ] g I x 1i ,

I 11 El I il I x s

. e ,

. I. : .

x;s t 1.. Is  !. : 11 t! x 88

- is s: g 43: .[e s.: g: s

.t:

xi s.I, aa:

se a::

sie as:

i l a: - a . a . a a a g..

N..

t-k-

w; . . '

FI ERE 4-1. 64 CHAMEL MODEL EIETH-CORE REPRESENTATION n ,

I l.

I- .

I a

I

. q a

I .

,ry- _,

e .

n 7 .. - - .n. . . - ~ ~ . . . , - . .

/($

j j ) t,s'd, er y;j ') e ,

.j.

is ,-

t ,

- 1 r y .* :J - 5

' 'f) ;- } l, *g -

s - ,

{#. ,- , ,

1

> 4 , *s! .

I,1 + p bk '? ('j - - -

W I ,I a M

l ('d. l l . ., . _ . f i '

( 13 l  ;

f' t

-j l

' j m ,

[.[ ,

- FIGullE '4-2. .

64-CWWelEL MODEL HOT ASSDSLY DETAIL i r.

1 .

s' . g .

9 wr e ,

i -j i

' j l '. .

-l l l

- l l s

{' ? : $ ,

I I ,-

L l ->-

I e

.-.l.'

1 I

1l l

.I

[

1 1

'w h , b

[? . , g

' A -p

/

4.,

e .

I T2 P I

1- 1 i, .3 5. ..wl:. . iW i... w m ,L.,, 4,,..,,,.,,,_,_..,_,,_,,. , , , . _ _ _ . _ , . , _ , , , _ . . _ , , , , , , . _ _ , _ . , _ , _ , , , . , , , , _ , _ . , , , _ , _ , _

y,'  ?.,' 4

( #

/. ~,

{' ,

. ) [v f

'p ' .

\' .

~'

.s

,25 .

l- ,-

I!

L ;-r i

FIGURE 4-3. 8 AND 9 CHAT *EL MODEL - HOT ASSDBLY DETAIL-e h

t I 1

l j

1 1

E -)

-.i k

h 4

{

i s

l l

45-

We '],

' ,f:

• e s,

FIGURE 4-4. 8 CHANNEL MODEL E;GHTH CORE REPRESENTATION >

?'

Hof ASSEMBLY

~ a< .

\

z l g i

.; 1 \

\

4 . l .\

l. N N 1  %

. s

=

l  %

g i ,

\

-- i .s \ i l

l \

[ i N i k' l

.\ _ ;

, 4 l k F i \

L- g i s

s

\

' N u' n l N%

I

\

[

1 a 's-%

I N

.- N l s; L s 1

' A E- , '

0,996 's i s t

'A I

I I

I~

I

. -)

e 1

e i

l i

i f

.{

l FIGURE 4-5. 9 CHANNEL MODEL EIGHTH CORE REPRESENTATION I.

tl  ; not , A ss t. km sa/t.

I __

,I tA x  :

v -

l \

l "'"

lI N _ ,

, s L

l f e AP 11 l h Yb h]b Vb // Yb LEGEND: 1

-85 ASSYS

!. 1 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ . . . _ _ _ . - - - . . ___ _...- .._ ... _ ..._..

m e m - - r m ja S w; w v,

-: - : .- - 7t; :

~

[~ ! . ,,

_ 3. 3 ~

~

- FIGURE 5-1:

~ m VIPRE-01 vs. IJNX2 M/P CHF BWC CHF CORRELATION .

i.4 .

y 1.2 -

k o Z

O a O ~

• S o

k i-m -O

- .U O

A 0.8 -

, , , , i 0.6 ,

i 0.8 0.9 I 1.1 1.2 f.3 1.4 0.6 0.7 VIPRE-01 M/P CHF

m-- um ums amm man n u m - m u m ! :: s u m umm-sum:umsiammEem a u s : ; a m m ~1 a u s : n m a n : s e e ; m u n e .

~

.. L..

l FIGURE 5-2 -

l VIPRE-01:vs. LYNX 2 MASS VE00C1TY AT CHF BWC CHF CORRiiLATION 4

l L Ex.

- Z

O E* 3- @

0-E- D o e

! Q

! 5 2-l A

2 .-

l.

a@

1

! O- ,0 , , _, ., , _, i j 0 0.5 ~l I.6 2 - 2.5 3 3.5 4 l ~ VIPRE-01 MASS VELDCITY AT CHF i

l i

d u . .. - -; .:.. - . _ _ _ _ __-.-_ _ . _ _ _ _ _ _ _ _

_ _ _ _ _ _ _ - - _ , _ _ _ _ _ _ _ - _ _ - - _ _ _ _ _ - -- ~

.+ . .

l -

FI6URE S-3 1

=

'VIPRE-01 vs. LYNX 2 QUALITY AT CHF-i -BWC CHF CORRELATION -

0.3 I 1

02- ..

O E-. 0.8 - .

g <; #

. ',s- s

. ,l

• t& f'

...s U

e: -0 ~

a

-O2 - ,.

- 0.3 - -

i i i i 1

-0.3 -02 -0.1 0.0 0.1 02 0.3 VIPRE-01 QUALITY AT CHF

, .g c -

fgifjQ?,n w. If'I;_[, \

m . .- -

x, v .,

g +

e FIGURE 5-88 MEASURED vs. PREDICrED CilF 1 VIPRE-01, BWC CORRELATION.

l

~

o O

'O o -o -

0.8 - o.

ggo o'00-O -

N ai Z 80o in O o

7 0 0.6 - o o O

e 5

M fo Bo

@ o o% E 0

< G.4 -

ff oo Z e 0 c

02-0.1 02 0.3 0.4 0.6 0.6 0.7 0.8- - 0.9 PREDICTED CHF

_mm .m .m

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

r - .

, m? ..-

i

'  % -r :.'

U

--- g i

.* x

, 1 l

x xx x xx x * '

xN M

% x x. o

. y*

• x*

(*x x*# *x y y

a xx xxx xx x

=

-i - QA #x x x x xg x x x O x x y

s. Y x% *h*, Nx h* # 1 y

, xx

• 4x x x#xx &x x x

=

n m m wx*x 48

• a 8 x-

_a _

sTa x X s O a- x.

x xx C i

. E

• _3 g

-x . x i 5 x 1

JM-x 7

j .h ,

i i

6

, i

.i 6 6 e

M a

m

- m d

3 d d

-- ~

dHD NTAID3Hd/G3HnSM

=

. _ _ _ ~

u w

l'.

+ -

4 y

x x%xx x @ -0

., u i ,

'z.

• g x *Iyx -e wE y 6

u xx

s , T *plik x** ,.gkg g W B

%= 5 g i5 8 x x =*xyp el*"

j g gH; i

=g oc man-x x

_n f

5 xt*M x # - x __ ,

3 x '

W x _g o

S 5 0 ~

3 8 9 dlO MTAIG3Hd/03HnSV3M

m -u- wg- . ,.

r.~- ..j..

F1GURE-5-7' MEASURED / PREDICTED CHF vs. PRESSURE VIPRE-01, BWC CORRELATION -

i.s q b -1.2- x l x x M- d  ; d, x

= x.

3, 6 18-

• x a

J x b E I x

[d g h x g t- R s

C x g g # -x*

l ri

• g 0.9- J g gx a *:  :

• 0.8 -

0.7 - , , , , ., ,

1400 1000 1000 2000 2000 PA00 2000 2000 PRESSURE (psia)

,3. = @

i4

c .p t

, ,g ' + a y y, >

- 's t, '

y -FIGURE 6-1 'RPS CORE PROTECTION-SAFETY LIMITS i a-

=

. 7, 3

~

6 s flierast Poser Levet, s- i

'l U54888P744L8 .]

+ , SPERAftet- , ;!

~l

! i 1 m.  !!$ I A48EPT48L8 s-Peep SPtta1108

. 108-I b lJQ 2

e. .

. gg \ [ '

s-k' At ttP 386 Ptf.A P BLE s GPER&Tl0B a- 0 $$

_- i

- .A88EF.Y&8t..t.P

.Pinar.ie.

'l r 4

g. .

. 2, ,

a s s 5 s 4 I 5 5 48 48 28' O 20 40 80 Reaster Pever leastaass. 5 l

J '#f ,'

r

': 'i )5y; ,

jp ,

~

-NJ;;

- n --

C {, - f

'!L 4

'4 I:

High Outlet-j W ennerature  ;$

Trio L 1

k c

-~.

/ -

Acceptable

/

t' Ocaration j

/

x /

N 'lariaele Lc.w Pressure Set-  !

E. point /

c-

/

m, E / -t-L A / Unacceptable i

/ Operation

/

. /

l /

7 Low Pressure ,: Limiting

• P-T Curve Trip ,

/

/

/

/

= >

Coolant Outlet Temperature

^} FIGURE-6-2. RPS P-T CORE PROTECTION ENVELCPE

~.f

,_.-b s

4,,

l -

I4 j[ # 1

.:TT. -

- t.y4 4

= 0

.I ,

$_ . { g. ,

FIGURE 6-2. HIGH TEMPERATUPI. TRIP MAPS.-

4

' l L h*-

1 = i r

i.j i I f,.

p> t j i

+

b

) i ..;

I i

a .e f

'l p

.. 0

\

(d-F a >

.I <

N_ q ' .

w K

F i

w=

W W;

x , .~4.

E" h

E.

=

e  : -

.ns-  ? i r g.

t r- ,

.. y ? hl 8 ~..

4 .

1-p ;;- ,

_~

.: v' <

r. -

Ji;

. n' r'h 3

-. m

.2. , '

t u):y. n ,

~' - ~ ' s: , . .

l' FIGURE 6-4. ~ L0ld PRESSURE TRIP MAPS ,

!I.y i I

.m!

6lla j

,i

.,<3 l

y ) ,*

.$+

e

?',

' 1, -

f i

.a a

Y h.

i

)

-i r

k i

)

s.6i 7 k'^, ' -'

.-}.- ~:D ' p ' j ' -

.5' i 4 . ._ w .e...': -

, , j e . ,

e'_

4

- 3, i. j ' % , 't 4 *

'i  %

Wm'

4. , ,_ g r

. . - a.'..t. '

l.

..k <

t

, d, 4 , . bh

i. A. . .,.

f F

6 e .u_ .:,

5 f

~ '

-FIGURE'6-5.:: FLUX-TO-FLOW MAPS.

, .;g f

.,K.{,=

r

• t_'

I

= . 1 }

_Y i

-b,

' L, k. '.L t

+

=

s ==

1

~j? ,

c 4

i

-4_

~,"

d . #./

R 5-w .

k

.A_

=

_^

- ) '

=

E Y

-E9-

~_

e >l, .I jtd cc e '

^i< r'

_, ;;j -l* '

^

^

zu;;

?N i

i t:

FIGURE 6 TYPICAL-2.PtW C0ASTDOWI TRANSIENT RESULTS-

,:l i

a

'li J .

a

l

.o

-c.

i i-s l

.. so 4 7

Q- 1

E

x. - q .

Limiting Pump . ,

, . Coastdown Statepoint  !

)F -

(

t: .

TIME, S EC, s

'- n 1

____ .. . ... . . . c 1-r)o

. . 9. s ,

~,.t ;c, q .-+,1 c

. ..a,. <

--y

+ ..-

r - .,+

/

' pu c,,

] *

>~ ,

-i i

i i^

Appendix A

, Safety Evaluation Report s ,

e i ,

^

=

I i -

4

= I

'=

4 i

N

'm

_= ~,y.

M A-1 4

_ 4 B

I

d

. . am; y '

,A

.0; p aC naq* %,- UNITED STATES

/" -

' t, .

' !I ' NUCLEAR REGULATORY COMMISSION

$-' I .4ASMG TON, D. C. 20698

'\v *...e July 19, 1989 Occret Nos: 50-269 50-270 ,

50-2871  ;

Mr. H. B.. Tucker, Vice'Presicent

, Nuclear' Proouction Department  !

Duke Power Coiopany.  !

2?2 South Church Street ,

Cnarlotte, North Carolina 28242

, Oear Mr. Tucker:

SUBJECT:

S AFETY EVALUATION REPORT ON GPC-ME-2003, " CORE THERMAL-HYDRAULIC METHODOLOGY USING VIFRE-01" (TACS 69377/69378/69379) he staff ano'its consultant, :nternational Technical Services, nave reviewed your Topical- Report CPC-NE-2003, " Core Thermal-Hydraulic Methocology Using VIPRE-01" submitted for application to the Oconee Nuclear Station, Units 1, t 2 and 3. We have found the teoical report to be acceptable for referencing i in the core thermal hydraulic analyses for the Oconee units with the following-- -

-j

' limitations: l l '(1) The acceptable DNBR limit is 1.18. Acceptance of. a ONBR limit less

/ :than 1.18 will require analysis of a broader CHF data base and detailed  :

i staff review. '

(2) The studies provided in the topical report were performee with the fuel t esently design currently used in the Oconee units. Although the approach is acceptable for future fuel assently Qsigns, you ,

should ensure that the seiected correlations are usea within their l applicability ranges.

A copy of our Safety Evaluatien Report is enclosed. This completes our action uncer TAC Nos. 69377, 69378 ano 69379. l Sincerely,

=

( &

Leonard A. Wiens, Project "anager Project Directorate 11-3 Division of Reactor Projects - I/II Office of Nuclear Peacter Regulation  ;

Enclosure:

As stated ec w/ encl:

See next page

" .A-2

, , . e .

Y~

cc:

Mr. : A. .V. Carr, Esq. Mr. Paul Guill Duke Power Company Duke Power Company P. O. Box 33189 Post Office Box 33189 422 Soutt Church Street 422 South Church Street Charlotte, North Carolina 28242 Charlotte, North Carolina 2B242 i i

J. Michael McGarry, III, Esq. ~

Bishop. Cook, Purcell & Reynolds Mr. Alan R. Herdt, Chief 1400 L Street, N.W. Project Branch #3 Washington, D.C. 20005 U.S. Nuclear Regulatory Commission j

-101 Marietta Street, NW, Suite 2900 '

Mr.. Robert B. Borsum Atlanta, Georgia 30323 Babcock & Wilcox Nuclear Power Division Ms. Karen E. Long Suite 525 .

Assistant Attorney General -

17C0 Rockville Fike N. C. Department of Justice '

Rockville, Marylano 20852 P.O. Box E29 ,

Raleigh, North Carolina 27602 Manager, LIS a

NUS . Corporation .

2536 Countryside Boulevard i Clearwater, Florida 34623-1693 Senior Resident Inspector U.S. Nuclear Regulatory Commission Route 2, Box 610 Seneca, South Carolina 29678 Regional Administrator, Region II U.S. Nuclear Regulatory Commission ,

101 Marietta Street, N.W., Suite 2900  ;

Atlanta, Georgia 30323 Mr. Heyward G. Shealy, Chief Bureau of Radiological Health South Carolina Department of Health and Environmental Control 2600 Bull Street Colurtia, South' Carolina 29201 Office of Intergovernn. ental Relations

. 116 Wes*, Jones Street Raleign, North Carolina 27603 I!onorable James M. Phinney '

County Supervisor of Oconee County Walhalla, South Carolina 29621 A-3 l

gj4r N

" ~~ ~ ~

~~ ~ ~ ~~ '~~ ^

l,, ~ ~ ~ ~ ~ ~ ~]

1 8

d..

M' ...,.'s.

UNITED STATES.

j' - > - T NUCLEAR REGULATORY COMMISSION

, y WA$mNGTON, D. C. 20$$$ .

{f ~  % , s .,1..... /

g ENCLOSURE 1 i SAFETY EVALUATION BY-THE OFFI.CE OF NUCLEAR REACTOR REGULATION l ,

RELATINC TO TOPICAL REPORT DPC-NE-2003, i  ! " CORE THERMAL-HYDRAULIC METHODOLOGY USING VIPRE-01" DUKE POWER COMPANY b- OC0 flee NUCLEAR STATION, UNITS 1, 2. AND_3 DOCKET fl05. 50-?69. 50-270 AND 50-287 d

?g;

1.0 INTRODUCTION

Duke Power Company (DPC) submitted Topical Report DPC-NE-2003, " Core Thermal-

/

1 Hyoraulic Methodology Using VIPRE-01," for . Nuclear Regulatory Commission staff review in a letter cated August 31,1988 (Ref.1) and amended by a letter of

May 3,1989 -(Ref. 2). This report cocuments DPC's use ~of the VIPRE-01' code

'(Ref. 3) in lieu of the currently used codes, CHATA and. TEMP (Rafs. 4 and. S),

for Oconee Nuclear Station licensing. core thermal-hydraulic methodology. . The.

Oconee core thermal-hycraulic analyses are routinely performed for fuel reloads to ensure that the departure from nucleate boiling ratio (DNBR) limit will not r

be violated during steady sta':e overpower condition and anticipated transients.

These analyses consist of (1) a steady state themel hyoraulic analysis to i.

-determine the allowable pressure-temperature operating limits and the power

-distribution limits, and (2) an analysis of the limiting two pump coastdown transient to determine a flux / flow reactor trip setpoint. Since the methodology.of determining these safety and operating limits has been reviewed

,g W- and approved (Ref. 6).previously, the staff review of the topical report

. concentrated on the use of the VIPRE-01 code in the core thennal hydraulic l

-h calculations.

VIPRE-01.is an open lattice subchannel core thermal-hydraulic code. In the open-lattice: analysis, the reactor core or fuel bundle is civided into a number i of quasi-one-dimensional channels that comunicate laterally by diversion f crossflow and turbulent mixing. This approach more realistically considers the

• A-4 I V _

- - - - .._ ......~ . ._ _-___ _ _ _ _ - _ _ , _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _

,y>

7

2 l

l flow redistribution effects in the open-latties core of a pressurized water

[ . reactor-(PWR) and resuits in less severe hot channel thermal hydraulic conditions thall that obtained from the closed-channel approach used in CHATA.

VIPRE-01 was' developeo by Battelle Pacific Northwest Laboratories uncer the i sconsorship of t'5e . Electric Power Research Institute. In December 1984, the  ;

. ility Group for Regulatory Application submitted the VIPRE-01 code for NRC i staf.fl review (Ref. 7). In approving VIPRE-01 for PWR licensing applications (Ref. 8), the staff required each VIPRE-01 user to submit separate j

_ documentation describing its intended use of VIPRE-01 and providing justification for its specific modeling assumptions, choices of particular models and correlations, and input values of plant specific data.

In a letter of June 19,1989 (Ref. 9), OPC. indicated that the VIPRE-01 models l 1

g- and methodology described in DPC-NE-2003 are related to the reload thermal

{ hydraulic' analyses, that the methodology of using VIPRE-01 model-for predicting I the' minimum DNBRs resulting from FSAR Chapter 15 transients, except for the two-pump coastdown, are described in Topical Report DPC-NE-3000, and that the VIPRE-01 methodology for transient analyses may be different from that used in I DPC-NE-2003. ' Therefore, the scope of the staff review of DPC-NE-2003 was p limited to the application of VIPRE-01 in the steady state and two-pump 1 coastdown analyses.

2.0 STAFF EVALUATION The staff review'and evaluation of DPC-NE-2003 included: (1) the nodal

! sensitivity studies to determine the radial noding details ano the axial node sizes, (2) the plant-specific core thermal-hydraulic parameters such as the crossflew parameters, grid loss coefficients, core bypass flow, inlet flow

_s distribution anc flove area reduction factor, power distributions, hot channel f actors, (3) the selected two-phase flow, heat transfer models and correlations, (4) the validation of the BWC critical' heat flux correlation (Ref.10) and the

' CNER limit in conjunction with VIPRE-01, and (5) the fuel pin heat conduction parameters.

A-5

-_---,--ua---,-----,----r_ _ _ _ .

_.__me, ,

TW

/9 3

l The . review was performed with technical assistance from International Technical

+

Services (ITS), and its review findings are contained in the technical-evaluation report (TER) which is attaphed. The staff has reviewed the ITS TER-and concurred with'its findings.

- 3.0 CO,NCLUSION )

=

The staff has reviewed the Topical Report DPC-NE-2003 and finds it acceptable- _;

'for referenciag in the Oconee reload thermal-hyoraulic analyses, subject' to '

the following-limitations:  ;

(1) The validation analysis with limited CHF data has demonstrated that the approved DNBR limit of 1.18 for the BWC CHF correlation, which was derived with the LYNX 2 thermal-hydraulic code, is conservative and acceptable for

_ use with VIPRE-01. Acceptance of a ONBR limit less than 1.18 will require i ar.alysis' of broader CHF data base and detailed staff review.

(2) The stuoits- provided in the topical report were performed with the Mark  :

~

BZ fuel assembly design currently.used in Oconee units, Though the

-approach. described is acceptable for future fuel assed61y designs, DPC should ensure-that the selected correlations be used within their

appiicability ranges.

4.0 FEFERENCES 5

1. -Letter from H. B. Tucker (DPC) to USNRC Document Control Desk, "Oconee Nuclear Station, Docket Nos. 50-269 -270, -287, Oconee Nuclear Station 3 Core Thermal-Hydraulic Methodology Using VIPRE-01, DPC-NE-2003," August 21, 1988.

2. Letter from H. B. Tucker (DPC) to USNRC Docu; ant Control Desk, "0:enee Nuclear Station, Docket Nes. 50-269, -270, -287, Topical Report

== CPC-NE-2003, ' Core Thermal-Hydraulic Methodology using VIPRE-Ol'; Response to Request For Additional Information," May 3,1989.

A-6

-= .. .

,c 7 g .

3

1. (( >

d! 4

3. EPRI-NP-2511-CCH-A, "VIPRE-01: A Thennel-Hydraulic Analysis Code for-Reactor Cores," 4 Voiumes, Electric Power Research Institute.  !"

-4 l BAW-10110,- Rev.1, "CHATA - Core Hyoraulic and Thermal Analysis," May.

1977..

S.: -BAW-10021, " TEMP - Thermal- Enthalpy Mixing Program," April 1970.

6. Letter from P. C. Wagner (USNRC) to W. O. Parker,' Jr. (DPC)..

Attachment:

' Safety Evaluatico Report on NFS-1001, "Oconee Nuclear Station Reload

>< Oesign. Methodology," July 29, 1981.

7. Letter from J. A. Blaisdell (Northeast Utilities Service Co.) to H. R. -

.Denton (USNRC), Subject related to UG.RA submittal < of the VIPRE-01 code, i December.17, 1984 -i

8. . Letter from C. E. Rossi (HRC) to J. A. Blaisdell, Chairman, UGRA Executive 4 I- Ccmittee, " Acceptance for Referencing of Licensing Topical Report,

EPRI-NP-2511-CCM, 'VIPRE-01: A Thermal-Hydraulic Analysis Code for Reactor Cores,' Volumes 1, 2, 3 and 4," May 1,1986.

9. . Letter from H. B. Tucker (DPC) to USNRC Document Control Desk, "0conee i Nuclear Station,< Docket Nos. '50-269 -270, -787, Response to Questions Regarding Differences Between Duke Topical Reports DPC NE-2003 and j DPC-NE-3000," June 19, 1989.

10. BAW-10143-A, "BWC Correlation of Critical Heat Flux," April 1985.

I A-7 ,

O I

• • b*@p #

IMAGE EVALUATION AkO

,,/[/4%f($9 fig, #

////7 tj., @// TEST TARGET (MT-3) 4

'W, 9 /4,,,,'k?

l.0 lff2 EM y @ EE i,l [m llillse h*

1.25 1.4 1.6 1

4 150mm

• 4 6" >

4k %,

, 3,,,>,,c .

/!b4

> ,;y%

f6 <g'Q x ~ ~

'r L 7

= AkO

• )I+ IMAGE EVALUATION ////p \ g($4h4 # ,

g$

g yp s $ @ .

TEST TARGET (MT-3) KY 4 p,,,9 //4,,,, '

l.0 'd # # N a yl[ EM ll L m EM -

l.8 1.25 1.4 1.6

[l; 4 150mm >

< 6" >

dI+ezz > y7///r

~

/# <>p;urp u%so <

L L ++a'r

- - - - - - Ns% _

, a ,

7%

?$+ '

///

• %+ >

NN'N/

im i eE Ev tu 1i e TEST TARGET (MT-3) pg%%k0 A

,j 1.0 E En E

'j y llO11 l,l  ? m R&e g

1.25 i.4 i.6 4 150mm >

< 6" >

4t[f.,p;; s

---

_____s -

su ......m.. . li

+3>@#

u IMAGE EVALUATION (Io

"" //gf

\N ,8

[ 4 #,

,8 gj, k///7 NNNI/ TEST TARGET (MT-3) y p#i4 f,7,,/ 7,,,

l.0 'd M E sll EM ii [= He M

1.25 1.4 1.6 l __

4 150mm >

• 6" >

+;k?%

/A 4%

>>,,;p>,$,p,=-

o p,, t; 777 4sg:y s

%inL _

.~

h A

i i g

h ATTl.C-D1E NT ITS/NRC/89 2  !

TECHNICAL EVALUATION.

OF TWE CORE TWEDMal Hv0RAULic METHODOLOGY USING VIPRE 01 4

TECHNICAL DEPORT DPC-NE 2003 FOR TWE -

DUKE POWER COMPANY OCONEE NUCLEAR STATION 1.0 INT:0D0CTION In Duke Power comoany (OPC) topical report OPC NE-2003, dated August 1988 (Ref. 1), OPC presented a description and qualification of their core thermal-hydraulic methodology using VIPRE-01 (Ref. 2) for steady-state and for two reactor coolant pump coastdown analyses of the Oconee Nuclear Station reload. VIPRE-01 has been previously reviewed. and approved for application to pressurized water reactor (PWR) planIs in. steady state and transient

analyses with heat transfer regimes up to critical heat flux. The NRC safety

- evaluation report (SER) on VIPRE-01 (Ref. 3) includes conditions requiring  ;

each user to document and submit .to the NRC for approval its procedure for -!

. using VIPRE-01 and provide justification for its specific' modeling  !

assumptions, choice of particular two phase flow models and correlations,  ;

heat ' transfer correlations, CHF correlation. and DNBR limit, input values of  !

plant specific data such as turbulent mixing coefficient and grid loss coefficient including defaults. This topical report was prepared to address these issues.

z

,~. ,

L The purpose of this review was to assure conformity of the DPC topical report {

and supplemental- information (Ref. 4, 5) to the VIPRE SER rect.irements, and to evaluate acceptability of OPC's intended use of the code as described in  ;

L the report.

[

i l I'n the past DPC used (Ref. 6) CHATA, a closed-channel (no energy or mass interchange among assemolies) computer code for core-wide analysis, and TEMP A-8 d

d

,(

L l'

l to determine the maximum permissible core power and distribution under various operating concitions for Oconee core thermal hydraulic design and licensing analyses. Althougn this approacn was conservative, these codes were unacle to realistically precict flow redistribution effects in an open lattice reactor core.

The VIPRE-01 computer coce (Ref. 2) is an open channel (permitting lateral communication among enannels by diversion crossflow and turbulent mixing) thermal-hydraulic computer code developed to evaluate nuclear reactor core safety limits. The code assumes the flow to be incompressible and homogeneous and incorporates mudels to reflect subcooled boiling and liquid / vapor slip. The input data to the VIPRE 01 code are the geometry of

( - the reactor core and coolant channels with thermal-hydraulic characteristics, and bouncary conditions. In addition, the user must select among certain correlations in the code for use in the particular analysis being performed.

The code calculates the core flow distributions, coolant conditions, fuel red temperatures and the minimum departure from nucleate boiling ratio. (MDNBR).

The OPC submittal, in fulfillment of VIPRE SER (Ref. 3) conditions, contains DPC's geometric representation of the core, its selection of thermal-hydraulic models and correlations, and a description of the methodology used

[

for steady state core reload design analysis and for a two-pump coastdown i

transient. These analyses are performed to determine the core thermal margin ano acceptable safety and operating limits and to analyze a two pump '

coastdown transient. It is not DPC's intent to use this methodology for FSAR Chapter 15 type licensing transient analysis.

2.0 EVALUATION.

I Acceptability of DPC's application of the VIPRE-01 computer code for thermal-hydraulic calculation of DNB for Oconee was evaluated with respect to the sensitivity of the comeuted steady-state operating conditions to inout selection, nodalization, thermal-hydraulic modeling, and correlations, by examination of the overall conservatism in the results.

l A-9

2.1 CORE NODALIZATION l

2.1.1 Radial Noding Sensitivity Since the VIPRE 01 code performs the thermal. hydraulic calculations simultaneously for all subchannels -(a single-pass approach) and permits flexibility in selection of channel sizes and shapes, a sensitivity study was performed to determine the sensitivity of predicted DNBR to the subchannel mooel sizes. The modeling of the reactor core uses the 1/8 core symmetry in wnich the hot assembly is located at the center of the core. The hot assembly includes the hot subenannel in which the minimum DNBR is expected to occur.

The thermal-hydraulic calculations were performed for three different core subchannel models; a 64 channel model, a 9 channel model, and an 8 channel model'. The 64 channel model consists of 36 subchannels representing the hot assembly and 28 subchannels individually modeling each of the remainder of assemblies in the 1/8-core segment. In the 8 channel model, 6 subchannels around the hot subchannel in the hot assembly are modeled individually. The rest of the subchannels in the het assembly and the remaining 28 assemblies in the core are lumped into 2 individual subchannels (Channels 7 and B). The 9 channel model, developed for evaluation of transitional mixed core effects,

. includes an additional subchannel to account for the different fuel assembly

esigns in the transition core.

The nadalization sensitivity studier used the same thermal hydraulic correlations and models which OPC intends to use in future reload licensing L analysis. Review of the particular correlations and thermal-h.udraulic models I

selected is provided in Section 2.2.

Steady-state and transient calculations using the previously approved RECIRC l umerical solution ootion were performed using these three different core mooels at four different operating conditions: the high temperature and the icw pressure safety limits, and two different sets of initial conditions for pump cuastdown transients including one representing the limiting MDN2R case.

l l

A-10 i

k.

I l

Sensitivity. to the core mooel size was studied by comparing the results of l using the 64 and 8 channel models. The 8 channel model was found to yield -

MDNBRs ranging from 0.447. to 2.2% lower than the 64 channel model. We .

therefore. find DPC's use of the 8 channel model acceptable for Oconee steady-state and 2 pump coastdown reload thermal-hydraulic analysis.

Sensitivity of the core models to transitional mixed core effects was examined using the 9 and 64 channel models in botti steady-state and -pump coastdown transient conditions. For steady-state ccnditions, the 9 channel transition core model preoicted 1.97. lower MDNBR than the 64 channel model.

For the transient analysis the MDNBR predicted by the 9 channel model was 1.6Y. lower than the 64 channel model. Based upon these sensitivity studies, DPC intenos to used the 9 channel model for steady-state and pump coastdown reload analyses involving transition cores of the Oconee Nuclear Station.

2.1.2 Axial Noding Sensitivity A steady-state sensitivity analysis for axial node length was performed with the 8 channel model using two sizes: a 3-inch node length applied uniformly-and a 2-inch node length applied where DNB is expected to occur.' The results indicated that the 3-inch axial nodes produced slightly more conservative MDNBR than did the 2-inch nodes. We, therefore, find that use of 3-inch uniformly spaced axial nodes is acceptable for Oconee reload steady-state and l

pumo coastcown thermal hydraulic analyses.

2.2 VIPRE-01 Input Data DPC's approach to generation of input to the VIPRE-01 code was reviewed for acceptability. No review was conducted of the input data in comparison to the actual physical geometry, i

2.2.1 *ctive Fuel Lengin Since power is distributed over the length of the active fuel, a shorter length yields higher power density, causing greater heat flux and is l

l A-11

j't L

.therefore conservative. DPC's choice for the active fuel length as described in Section 5.2 of Ref.1 is conservative when compared to hot conditions.

When a different assumption is used. DPC should justify its conservatism.

-2.2.2 Centroid Distance and Effective Crossflew Gaps The centroidal distance is used in the crossflow momentum equation to determine the lateral pressure gradient. The gap width is used in

! determination of the crossflow area. OPC calculates these parameters from t channel geometry following the code's prescription.

2.2.3- Spacer Grid Form Coefficients Pressure losses across the spacer grids impact the axial pressure distribution and therefore the axial location of DNB. The spacer grid form loss coefficients were obtained from a full size fuel assembly test conducted by the vendor (B&W) - To determine the. individual subchannel form loss coefficient, DPC stated (Ref. 4) in response to our question that the vendor used its computer code, GRIL. The input data to the GRIL coda are the individual subchannel geometry, drag areas and coefficients, and the coolant information. From this input, the code calculates individual subchannel loss coefficients, an overall grid loss coefficient and subchannel velocities L

based on single-phase flow input data by a iterative process. The calculated overall grid loss coefficient is matched with the measured value by adjusting the velocity field in the subchannel until consistency between the measured and predicted values is achieved. OPC has stated that the calculated velocity profiles were compared by the vendor with the experimental data and showed good agreement (Ref. 4).

2.2.5 Core Bypass Flow  !

l ,

1 i DNB is influenced by the aggregate flow rate past the location being examined, and therefore by the core bypass flow. Since the bypass flow depends on the number of control rod and burnable poison rod assemblies in the core, this is a cycle dependent . parameter. Therefore, the core bypass l l

A-12

. - - . . . - - . . . .. . _ . . _ . . . - . .- -- . - _ . - -. ._1

1 l

- flow data used in the analysis should ce based on a bounding value or on a cycle specific data.

'2.2.6 Inlet Flow Distribution CHF -is decreased and the precability of DNB is enhanced if flowrate is 1 reduced due to a flow maldistribution. The use of 5% inlet flow maldistribution to the hot assembly with all four reactor coolant pumps operating yielded slightly more conservative results than a uniform inlet flow distribution. This value is supported by a B&W 1/6-scale Vessel Model ,

Flow Test and was previously approved for Oconee reload analysis (Ref. 5).

For operation with less than four reactor coolant pumps operating, incre restrictive flow reduction factors are applied.

2. 2. 7. Flow Area Reduction Factor OPC reduced the het subchannel flow area..by a factor as stated in Section 5.12 of Ref.1 to account for variations in as built subchannel coolant flow area.

2.2.8 Reference Design Power Distribution The reference design- power distribution was developed using a radial-local hot pin peak of 1.714 which has- been previously approved for Oconee reload analysis (Ref. 5, 6). The corresponding assembly power was 1.6147, 2.2.9 Axial Power Distribution The axial power shape used in the analyses was a chopped cosine shape with a conservatively determined peaking factor. Although the axial power shape is

! cycle specific and transient dependent, the use of generic bounding axial power curves accounts for the effect on DNB of different axial shapes. This

, is discussed in Section 2.4.

l l

DPC added an optional new routine to VIPRE-01 to generate the axial power l

A-13

r. l 1

1;

} shapes using' a generalized power function. The currently defined function  !

can generate both symetric and s'kewed power shapes but cannot generate L

certain power shapes (such as double peaked) because of limitations of the L generalized function used. The axial power shapes calculated using this i routine agreed . with the symmetric axial shapes calculated using VIPRE-01

, symmetric cosine routine for axial peaks of 1.2 and 1.5 (Ref. 4).

1 DPC intends to maintain two options for power shape generation: one is to use.

this routine and the other is to use a user specified table. The use of this j routine is acceptable so long as the computed power shapes represent the true )

power shapes to be analyzed. l l

Although analyses in this report were performed using a higher axial peaking ]

factor, DPC will continue to use the reference axial peaking factor  :

consistent with the current FSAR Chapter 15 transient analysis in the reload licensing analysis (Ref. 5). I 2.2.10 Hot Channel Factor The power factor, Fq, used to account for variations in average pin power caused by differences in the fuel loading per rod was selected to be 1.0107 which has been previously approved for Oconee reload analysis (Ref. 6).

The local heat flux factor. F ", q used to account for the uncertainty in the manufacturing tolerances was selected to be 1.0137. In the determination of the maximum allowable peaking limits, two additional factors were used to increase the limit to 1.0371. These factors were 1.007 to account for power l spikes occurring, as a result of the flux depressions at the spacer grids, and l 1.016 to account for axial nuclear uncertainty (Ref. 6). All of these factors have been previously approved for Oconee reload analysis.

2.3 VIPRE 01 Correlations VIPRE-01 requires empirical correlations for the following models:

a. turbulent mixing

, A-14 l-t

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

i

b. friction pressure loss

c. two-phase flow correlations (subcooled and saturated void, and void-quality relation)

d. single-phase forced convection e, nucleate boiling heat transfer

f. critical heat flux .

2.3.1- Friction Pressure loss, Subcooled Void, Single-Phase and Two Phase Flow Correlations-For single-phase turbulent flow the Blasius smooth tube friction factor, a default option in VIPRE-01, will be used to calculate the friction pressure loss in the axial. direction. Crossflow resistance has a minimal effect on MONBR in transients where axial flow dominates. DPC's selection therefore has an inherent assumption of axial flow dominance. This choice is ac:eptable since we agree that in the analyses to be performed in the context of this topical report, the flows are expe.cted to be axially dominant.

~ For two-phase flow, subcooled and bulk void correlations, a sensitivity s,tudy using six different combinations of three subcooled and bulk void correlations was performed. for two operating conditions. The results indicated that the use of Levy subcooled void and Zuber-Findlay bulk void correlations, in conjunction with EPRI two-phase friction multiplier results in- conservatively predicted DNBR relative to other comeinations of correlations. DPC intends to use this combination in Oconee steady state and pumo coastdown reload analysis.

l This is consiste,nt with the VIPRE-01 SER findings.

l 2.3.2 Turbulent Mixing

'The lateral momentum ecuation recuires two parameters: a turbulent momentum factor and a turbulent mixing coefficient.

The turbulent momentum factor (FTM) describes the efficiency of the momentum A-15

~

mixing: -0.0 indicating that crossflow mixes enthalpy only; 1.0 inoicating that crossflow mixes enthalpy and momentum at the same strength. A sensitivity study using the 8 channel model was performed for two operating i.

conditions' and for three different values of FTM of 0.0, 0.8, and 1.0 and l

found little sensitivity in DNBR by different values of FTM. Conservative DNBR's were_ obtained with zero (Table 5 4 in Ref.1). However, in reality there will be always some momentum mixing. An FTM of 0.8 has been recommended by the VIPRE-01 coce developer.

Since the turbulent mixing coefficient determines the flow mixing rate, it is-an-important parameter. Based upon tests using a 5x5 heated bunale conducted by B&W, where the subchannel exit temperatures were measured, a mixing.

coefficient was conservatively determined for B&W Mark-B fuel (Ref. 4). This will be used in the Oconee core steady-state and pump coastdown reload thermal hydraulic analysis (Ref.1),

i 2.3.3 Single-Phased Forced Convection, Nucleate Boiling Heat Transfer DPC will use (for its steady-state and pump coastdown analyses) the default EPRI single-phased forced convection correlation and Thom subcooled and saturated nucleate boiling correlations, both of which were found to result in conservative MONBR for the two pump coastdown transient.

i 2.3.4 BWC Critical Heat Flux Correlation The BWC correlation (Ref. 7) was originally developed for 17x17 Mark-C fuel, and later used for 15x15 Zr grid Mark-BZ fuel. The use of BWC correlation with the LYNX 2, code (Ref. 8) for 15x15 Zr grid Mark-BZ fuel was previously approved by NRC with a design limit of 1.18 (Ref. 8, 9).

All Oconee thermal-hydraulic analyses using VIPRE 01 and the BWC correlation will use a design limit of 1.18. Since the BWC correlation is now being used with VIPRE-01, it is necessary for DPC to demonstrate that the CNBR limit of 1.18 for BWC CHF correlation used in VIPRE-01 can predict its date base of DNB occurrence with at least a 95% probability and a 95% confidence level.

A-16

_ _ _- . . . _ . ~ . _ , _ __ _ _ __

L In Section 5.13 of the topical report, DPC performed validation using more than 200 data point. Results show a 95%/95% limit of 1.16. Therefore use with -VIPRE 01 of the previously approved (with LYNX 2) value of 1.18 is conservative and acceptable. DPC agreed that when a lower DNBR limit becomes L desirable with use of BWC CHF correlation with' VIPRE-01, it will submit a separate topical report documenting analysis based on a broader CHF database for detailed NRC review and approval.

2.4 Oconee Thermal-Hydraulic Analyses Using the input, assumptions, and thermal-hydraulic correlations selected and justified in the subject topical report, OPC discussed its methodology to  ;

perform steady-state and generic two-pump coastdown analyses necessary to

define the core thermal margin or safety limits and acceptable operating limits.

The core safety limits that provide DNB pr.stection are pressure . temperature (P-T) envelope and power - power imbalance limits. The P-T envelope defines a' region of allowable operation in terms of reactor coolant system. pressure and coolant temperature (Ref. 6). ,

To ensure that the P-T envelope provides adequate DNB protection, P-T curves are determined for different numbers of RC pump operation. P-T curves are the comeinatiens of RCS pressure and vessel outlet temperature that yield the l design DNBR limit or the BWC correlation quality limit. The P-T envelope must be more restrictive than the most limiting P-T conditions. VIPRE-01 was used to generate the oa:.eric P-T curves using the 8 channel model.

The following are input to the code for generation of P-T curves:

1. a symmetric chopped cosine with a conservative axial peaking factor;

2. 112 Y. cf full power for 4-pump operation. and the power level for other modes of pump operation are based on trip setpoint plus margin for uncertainties;

3. 104% of design RCS flow for 4 pumps; appropriately lower for less A-17

. . . ~ . . . - - . - .

D.

L ]

l' 1 than 4-pump operation; 4 minimum coolant. temperature: and l 5 .' generic maximum allowable peaking (MAP) limit curves. l

l Having developed. the P-T curves, DPC, as part of its reload snalysis, l performs a two-pump coastdown transient to determine the flux / flow trip l setpoint.- This trip provides DNB protection during a loss of one or more reactor coolant pumps.

For this 2-pump coastdown analyses, the input to the fuel red heat conduction model in VIPRE were determined by sensitivity studies evaluating impact of pellet / clad gap, gas composition and pellet radial power profile to the DNBR, Results led to a censervative set of eight fuel parameters for the conduction model input. i The methodology described in the report is acceptable.

3.0 CONCLUSION

S de find that the subject topical report, together with DPC responses, contains sufficient information to satisfy.the VIPRE-01 SER requirement that

~

each VIPRE-01 usar submit a document describing proposed use, sources of input variables, and selection and justification of correlations as it relates to use by DPC for reload stead-state and pump coastdown analyses.

We further find that the manner in which the code is to be used for such analyses, selection of nodalization, models, and correlations provides, except as limited ,below, adecuate assurances of conservative results and is therefore acceptable.

l The-following items are limitations regarding application of DPC-NE-2003:

1. An MDNBR limit of less than 1.18 with the BWC CHF correlation, as l described in Section 5.13 of DPC-NE-2003, requires further justification based on broader CHF database for cetailed review.

A-18

l I I. '

t f

t

2. Studies presented in this report are performed using design data i s for Mark-BZ ' fuel assemblies, which are currently used in Oconee.

Although the approach described in this report..is acceptable, for future analysis of reloads which incorporate other fuel, DPC should-assure that .the VIPRE-01 computer code be used within the range of- l applicability. .

The scope of this review and the applicability of findings- are 3.

limited to OPC's use of VIPRE-01 for core reload steady state and,a two-pump coastdown transient analyses.

4.0 REFERENCES

1. " Core Thermal Hydraulic Methodology Using VIPRE 01," DPC NE-2003, August 1988.

2. "VIPRE-01: A Thermal-Hydraulic Code for Reactor Cores, EPRI NP-2511-CCM Revision 2, EPRI, July 1985.

3. Letter from C.E. Rossi (NRC) to J.A. Blaisdell (UGRA), (Transmittal of )

VIPRE-01 Safety Evaluation Report), M,ay 1,1986.

4. Letter from .H.B. Tucker (DPC) to USNRC, " Response to the Request for Additional Information," May 3, 1989.

5. Letter from H.B. Tucker (DPC) to USNRC, " Response to Questions Regarding Differences Between Duke Topical Reports OPC NE-2003 and DPC NE-3000,"

June 19, 1989. l

6. " Duke Power Company Oconee Nuclear Station Reload Design Methodology," I L DPC-NE-1001A, Rev. 4, April 1981.

7. "BWC Correlation of Critical Heat Flux," BAW-10143P-A, April 1985.

8. " LYNX 2-Subchannel Thermal Hydraulic Analysis Program." BAW-10130A, October, 1976.

l L 9. " Duke Power Company Oconee Nuclear Station Reload Design Methodology II," DPC-NE-1002A, October 1985.

l-l l

A-19 l

W~

g, 7, , - ,

~

1 ti tc; r.

4. O; i

q ,

l 4 . <

-1

+

l Appendix B Responses to Requests .for Additional ~Infonnation l

l l

s l

.v:

F r

l B-1

s

?c.i. .

'I k t Dt:KE POWER COMP.OiY '

P.O. BOK 33189 CHARLOTTE. N.C. 28949 R.id. B. Tt:CKER m,,

m e . .... .

mettaharamtcmpe 1

I

' flay 3, -1989

-U..S. Nuclear Regulatory C:mmission Washington, D.C.. 20555 Attention: Document' Control Desk

. Subject': Oconee Nuclear Station, Docket Numbers 50-269, -2'0, and -287

• Topical Report IPC-NE-2003, " Core Thermal-Hydraulic Methodology Using VIPRE-01"; Response To Request For Additional Information-IJsubmitted, by letter of August 31, 1988.the subject Topical Report for NRC 7 review. By letter dated March 22. 1989, the NRC staff requested additional information. Attached are responses to the staff's questions. Also attached are errata sheets, which correct various typographical errors and/or provida-additional clarifying infer =ation. Upon approval of the Topical Report the entire document will be reprinted with the corrected pages.

Please note that the original submittal was a proprietary document, and my August

.31, 1988. letter contained an affidavit attesting to that fact. The responses to the questions and the errata sheets should be considered part of-the-Topical Report, and should be withheld from public disclosure.

If. we may be of any further assistance, please call Scott Gewehr at (704)-

373-7581.

Very truly yours, 1

i k n H. B. Tucker SAG 163/lcs L

l xc: S. D. Sbneter, Regional Administrator U. S. Nuclear Regulatory Commission - Region II 101 Marietta Street, NW, Suite 2900  ;

Atlanta, Georgia 30323 P. H. Skinner

- Senior Resident Inspector

-Oconee Nuclear Stati:n Darl Hood, Project Manager Office of Nuclear Reactor Regulation U. S. Nuclear Regulatory Commission Washington, D.C. 20555 I

B-2 l . - -

p Cuestien 1.

Sect;:n 4.0 ot the topical report describes the nodali:ation sensitivity stucv rerformed to demonstrate that the simplified core models to be usec f'Or licensing calculations are conservative relative to the more

~

. detailed medel. (a) Was the study performed with the same thermal-hydrau;te models and/or correlations to be used for licensing calculations? If not, identify those mocels and correlations which are not une same. (b) Would the use of different correlations and/or models lead to different nodalization sensitivity study results? Iemonstrate the c:nservatism. of the simplified core model with the final T-H models '

for' licensing application. (c)-It is understood that only the BWC correlation will be used for critical heat flux calculation. What do you ntend to do if the core conditions are outside the ranges .of appl;; ability of the BWC :orrelation?

Restense

-(a) The nodali:ation study was performed using the same models and ec relations that will be used for licensing calculations.

(b) The use of different correlations and/or models would not lead to different nedalization sensitivity study results. Sensitivity studies (turrulent momentum factor, void models, etc.) have been performed for the M:Guire and Catawba Nuclear Stati0ns for Westinghouse optimized fuel using both a large (75 channel) model and a simplified (9 channel) model. Both models gave essentially identical sensitivity study results and the same conclusions were drawn from the 75 channel and 8 channel results.

The ::nservatism of the simplified model that will be used for licensing calculations is discussec . Sections 4.1 and 4.2 of the

-report.

(c) Tollowing the methodology discussed in Section 6.0 of the report, all :f the-core conditions analyzed as a part of the generic Oconeo thermal-hydraulic analysis are within the ranges of applicacility of the SWC :: relatten. If conditions must be analyzed that are cutside the range Of the BWC correlation the NRC will be informed of the CHF l correlation that will be used.

1 l

l s ,

l l

l l l I

l l

l l

l l

l B-3

i 1

.Cuestien 2.

Section 5.5 states that the spacer grid form loss coefficients for the individual subchannels are determinec analytically by the vendor from the cVerall grid form loss coefficient. Provide sufficient detail of the analytical determination of the individual subchannel form-loss c0efficients.' Are these values for single or two phase flow.

Rescense Spacer grid subchannel form loss coefficients are calculated by B&W Fuel Company using the grid loss evaluation program GRIL. The GRIL code is aole to determine subchannel form loss coefficients analytically based on-individual subchannel geometries and experimentally determined .

overall grid loss coefficients. Subchannels geometries are defined in  !

GRIL by inputting dimens: ens, drag areas, and drag coefficients for the different objects which obstruct flow in the individual subchannels.

These_ objects include such things as hard stops, spring stops, and spacer grid webbing. GRIL calculates grad loss coefficients based on single-phase flow with coolant flow information being input in the form of average coolant density, average kinematic viscosity, and average

.Reynolds number. Flow velocity in the rod gap is calculated by boundary

. layer eneory using a universal velocity profile which relates dimensionless velocity to wall distance parameters at different. flow recimes. Actual calculation of the subchannel loss coefficients in GRIL is'an iterative process. For the first iteration, the channel flow velocities are assumed to be equal to the average velocity in the channel. Using the individual subchannel geometry and drag information, 4 GRIL calculates individual subchannel loss coefficients, an overall grid i loss coefficient, and new subchannel velocities. The iterative process  !

continues until the calculated overall grid loss coefficient matches the experimental value. Comparisons made to laser doppler velocimeter (LDV) test results have shown that the subchannel velocity profiles calculated  ;

by GRIL agree well with experimental data.  ;

i I

f 4

B-4

q I l

4 U Question 3.

-Sect;en 5.B.2 discusses the cetermination of the value (preprietary) of the< turbulent mixing-coefficient to.be used for all Ccenee Nuclear Station core thermal-hycraulic analyses based on vencor prediction of the mixing test results. Explain the process of vender preciction of mixing test results and mixing c0 efficient, and explain hew.this JJ

crrelates to the Oconee c =putat:0n. ,

Rescense In subchannel crossflow c0 des such as VIPRE-01, the turbulent exchange between subchannels i and j is defined by wl3 =/s5n -

where 6 is the average mass flux of the adjacent subchannels, so is the wicth of the gap between succhannels, and / is the tur=ulent mixing i coefficient. The mixing coefficient is usually obtained by performing tests using.a heated buncle., A test specifically cesigned for B&W Mark-3 fuel was performed by Columbia University. Single-pnase subchannel mixing data were c=tainec from a 5x5 rod array by measuring subchannel.

exit temperatures for 57 tests covering the range of test concitions-shown below. A least-squares statistic based on exit temperature differences was calculated to determine, in con] unction with a

'subchannel.crossflew coce, an optimum value of the turculent mixing coefficient. The opth-um value of / was found to be [ ]with a j standard deviation of ] As a result of this test, B&W uses a value for /.ofi, Lor F

all Mark-B fuel crossflow analyses. Duke Power will'also use .s valu]e of( ] for all Oconee Nuclear Stction core thermal-hydraulic . analyses of Mark-B fuel.

Range Of Test Conditions System Pressure 2200 psia ,

Inlet Enthalpy 186.1 - 487.2 Stu/lbm Average Heat Flux 0.179 - 0.539 MBtu/hr-ft' i

Average Maas Flux 1. 072 - 3. 513 : Chm / hr-f t' i

l l

l l'

1 1

I l

l.

B-s

, 91 s 4

h- - @m 1

m

'Questien 4.

Se.ction' 5.8.2 also discusses the selection of the turbulent momentum if act:r l .(FTM) from-the sensitivity study. performed with the FTM of 0.0, b 0.82 and 1.0. .(a) Justify .the selected value which is not the most

! conservative"value as shown in Table 5-4. - (b) Explain how and why only the three values of FTM were selected for the sensitivity study, Resconse The turbulent momentum mixing between channels is included as a force.in the momentum balance. The total axial force on the control volume due to turbulent mixing, F,, is calculated as F, = -F IM a x _w' au kti where w'is the crossflow per unit length, u is the axial velecity difference between the control volume under consideration and an adjacent one, and FTM is a constant correction factor to account for the imperfect analogy between tur=ulent transport of thermal energy and momentum. As discussed in .the topical report, if the turbulent moment facter is 1.0, energy and momentum are mixed with equal strength. If FTM-is 0.0, only energy is mixed by the turbulent crossflow. These two extreme values and the value recc= mended in ref. 1, FTM = 0.8, were studied to determine the .effect that the turbulent momentum f actor has on the MDNBR. As expected, FTM = 0.0 (no momentum mixing) yields the ,

most conservative MDNBRs and FTM = 1.0 results in the least conservative MDNERs (see Table 5-4) . Battelle found in ref. 1 that DNBR is not sensitive to changes in the turbulent momentum f actor and the results in Jrdele 5-4 show that changing FTM f rom 0.0 to 0. 8 changes the MDNBR by less than 1.5 %. Using FTM = 0.8 reasonably assumes that thera is mcmentum mixing which benefits the het channel, but not by the maximum cossible amount. Thus, Duke Power has elected to ust.- FTM = 0.8 because

't i is a reasonable value, is the recommended value in ref. 1 (which has been approved by the NRC), anc the DNBR sensitivity to FTM is low as demonstrated by both Duke. Power and Battelle. As discussed in the respense to question 3, a conservatively low turbulent mixing coef ficient will be used in all Oconee thermal-hydraulic analyses, thus the amount of turbulent mixing (energy and momentum) will be conservatively predicted. ,

1 p Reference i

1. J. M. Cuta, et al., "VIPRE-01: A Thermal-Hydraulic Analysis Code for Reactor Cores", EPRI-NP-2511-CCM, Vol. 1-5, Battelle Pacific Northwest Laboratories, July 1, 1985.

1 l

i B-6

.- .r:

p

~

mi , ,

l )

t (Questien L., _

-[ LSGe'titnL5.10 states that:a new routine is added torthe VIPRE-01 code to 5

e gangratel axial power shapes with inlet, - symmetric, or outlet peaks.

, Provide sufficient details of this routine.

' ^

Jh nescense --.

$Thn-axial power shape routine added to VIPRE is based on the-

. foll: wing mathematical constraints on an axial power' shape: ,

z

+

1

' (1 ).. F (B) = PB

-(2): - F (M) =P-L(3); F(E) = PE ,

i4) Max F (x) = F(M) =P _

t!) F(x) is continuous from (B,E)

'(6) T' {x) is continuous from (B,E)

E 1

1(7)- E.3 F(x) dx = 1.0 B

, :whOre F(x) = axial power shape as a1 function of the axial location, x B,E = beginning and ending normalized location of the active length P = axial peak M = normalized axial location of the axie.1~ peak

'FB,PE = axial flux at the beginning and ending location of the active length, respectively i

Based L:n the constraints given above, the following generalized cxpression was. developed (8) F(x) = P + C(x - L) where C = ascenstant based on the axial peak (P) and the axial ,

flux at the beginning and ending location of the active length (PB,PE) and the respective axial locations (M, B, and E). Different expressions are used ts determine C based on the axial location (x). '

L=M, B, er E I = integer relationships based on the axial peak (P) and the beginning and ending flux values (PB and PE)

Symmetric axial power shapes calculated using the new routine are comoared with axial shapes calculated using th9 VIPRE-01 symmetric cosine routine for axial peaks of 1.2 and 1.5 in Figures 1 and 2. These B-7

1. __ _ . - _ . . . . _ _ _

~

.. .?-- '

u 1

P , .-,, ; --" i

L t;.a ii  ! .*igures clearly show the'. agreement between the two methods ofJgenerating

= symmetric axial' flux shapes.. The- new flux shape routine was added to

-cenerate skewed ~ axial 1 flux. shapes. As discussed in. Section 6.5 of the e r iopical report, Maximum Allowable-Peaking (MAP) limits are calculated. .

.for-a range of-axial-peaks with the location of the peak varied fr0m the i- . bottem to.the top of the core. As an example - of the flux shapes used to -

calculatelthe MAP limits, three axial shapes.are shown;in Figure 3 fort an axial pear.of l.3 at X/L = 0.3, 0.5, and'0.7.' -

(

Y

's ;-

g r

)

i i

L i

1 B-8

.x-3 -.

i . . . .

PIGURE 1. .

' AXIAL FLUX SHAPE COMPARISON - ^-

4 AXIAL PEAK = 1.2 .  ;

4 ,

! 1.2 -

~I 1

! VIPRE-01: C051140 .

l 1.I-  :

j ILUX SilAPE i400Tli1L.'

g_ ,

m X

[

b N

l a 0.9 --

4 -

I pl  :

i a 0.8 --

r l 0.7-e g

0 0.6 - , , , ,

0.8 0.9 i 0.2 0.3 0.4 0.6 0.6 0.7 0 0.1 AXIAL IDCATION, X/L u

-.,v < .w j - r ,..%, y . , _ . , , - . ,..% 4 .g. . - _ . , ,

.~.- . -

~

~

, ~,

w-

.' FIGUltE 2.

4

~

j. -

AXIAL FLUX SHAPE COMPARISON- .

AXIAL PEAK = 1.5 .

i 1.8-i

! 1.4 - .

viestt-ol cosint' ^

1.2 -

IL UX SilAPE l(OlliIHL

, x g_

! .'. D

o J

, k ='

a 0.8 -

l 4

l 0.8 -

l i

l '.

O.4 - / '

i l '

0.2 - .- t

, , i i i 0- , i i i i 0.3 0.4 0.5 0.8 0.7 0.8 0.9 I O 0.1 0.2 '

AXIAL IDCATION, X/L-1

._s. ,.  ;,_._ . . _ _ _ , _ . , _ _ . _ _ _ _ . _ , . , , . . _..,,s ,_ _, _ ,

1. ' s A s'O.a, aw,r. ., , . , , , ,_ . , _ , _ _ _ , ,

-i-. .

I s

5 I l' ' j f,)

5,e'1$:- ,

3'. }.j k ,

1 1

I r l l

-)

=

4 I

04

• O l

,,'g,7

.g  ;

• • ........ *,'.. ' 0 ..

a**.*g. *,e* ,

o O*,, '

r-q- .'

,,* ,o* =a

,o*, ,,=* ,*

0 g* * ,

b

• 4 g

'e*,9* ,#

l 4

• ,.- .** r.

l

• ,o

-o o,.

$p+ \ .. '

-o I.

ll

/g *, ,s' l ', '

>g. l. '. .,

• 9 c

..  ! ,4 -e m a ,' ' ,

e .y 8, ,# <

iz . , -,

p-C y- s*

N s,', -%

-o Y'< - ,',,s r =

l

/ .,

s s,

a

• g ,=*%

3 M i g's, g

• - N l

,- .,s ,

e o

s S b,  %

% 6*%" 4  %

%g

g. %g q

-o 7 .,,,,

p ..,- -

o , .., - ,

=

L

, . .o 4

., 's ,i,

%g g.

4

( ,

p 1 O

I I I I I i 4 r) " = 12 ID t= c d ]

• 6 6 6 6 Xf713 'lVIXV i

B-11

L' j

(

l

)

' t

- Question 6. -o

- In Section 6.6, the inputs for the fuel gap cot. duction model are selected through a sensitivity study performed by varying three input parameters, i.e., pellet-cladding gap size, gas composition, and pellet radial power profile. Explain how this study enables the selection of conservative, values of the eight parameters for input to the conduction model.

i' Resc_onse Conduction through-the gap between the fuel pellet and the clad is determined using the gap conductance model in VIPRE-01. This model is a

- simplified form of the models available in the FRAP and GAPCON codes.

The NRC stated in the VIPRE-01 SER, ref. 1, that " based on the use and qualification of the model in GAPCON and FRAP, we conclude that the fuel

' rod heat conduction model is acceptable for licensing analyses." To

. select the input for the conduction model (pellet diameter, gap width, etc.) sensitivity studies were performed using the 8 channel model discussed in Section 4.0 with a base set of conduction model input.

.n To investigate _the sensitivity of the input gap width on the DNBR during a pump coastdown transient, three cases were run using the nominal, maximum, and minimum pellet / clad gap. The dynamic gap conductance model was used for all of the sensitivity studies. The dynamic gap conductance model calculates any changes in the gap width due to fuel rod deformation and fuel pellet thermal expansion, but it does not cetermine any changes due to densification, swelling, cracking, or pellet relocation. The maximum cold gap studied was calculated based on a conservative pellet densification and on manufacturing data for the pellet and clad diameters. The input gap width can be varied axially, but all of the cases assumed a constant gap width. The different gap widths were studied using the nominal clad ID and varying the pellet diameter.

The maximum pellet / clad gap case yielded the lowest MDNBR during a 2 pump coactdown transient. The large gap resulted in a lower gap concuctance than that for the nominal gap, but the clad surface heat flux increased slightly when using the maximum gap (i.e., more energy was stored and then released at the time of MDNBR).

The gap width cases were run assuming that only nelium and nitrogen gases were in the gap. An additional case was run assuming that fission gas had been released into the gap. The fission gas composition was taken from a typical TACO 2, ref. 3, run at a burnup of 30,000 MWD /MTU.

l The VIPRE-01 sesults showed that the gap conductance and surf ace heat flux did not significantly change and the MDNBR did not change at all

! when assuming that fission gas was present in the pellet / clad gap.

Since the maximum gap resulted in the lowest MDNBR during the pump L coautdown transient and since the maximum gap would occur early in the burnup history of the fuel when peaking is highest, the generic pump coastdown analyses will assume that the gap is filled with only helium and nitrogen.

The base case for the sensitivity studies assumed that the power was uniformly distributed radially through the pellet. Cases were also run using a fuel pellet power profile from a typical TACO 2 run. VIPRE-01 integrates the input power prcfile over the width of each node in the pellet to define the 1ccal volumetric heat generation rate. The MDNBR B-12

'lY t 1 l' results assuming a uniform power distribution or a power profile frcm TACO 2 are essentially identical. The generic pump'coastdown analyses will be based on a uniform pellet power distribution.

l-One additional case was run assuming a maximum pellet / clad gap based en  ?

the nominal clad OD and nominal pellet diameter and a reduced clad thickness. The MDNBR results for this case were identical to the case '

with the maximum gap based on the nominal clad OD and ID and a reduced pellet diameter.

The conduction model input that will be used for the generic pump coastdown analyses was selected based on the sensitivity study results discussed above. The input that results in a conservative pump coastdown analysis is listed in Section 6.6 of the topical report.

References

1. Letter from C. E. Rossi (NRC) to J. A. Blaisdell (UGRA),

" Acceptance for Referencing of Licensing Topical Report, VIPRE-01:

A Thermal-Hydraulic Analysis Code for Reactor Cores", EPRI-NP-2511-CCM, Vol. 1-5, May 1, 1986.

2. J. M. Cuta, et al., " VIP RE-01 : A Thermal-Hydraulic Analysis Code for Reactor Cores", EPRI-NP-2511-CCM, Vol. 1-5, Battelle Pacific Northwest Laboratories, July 1, 1985.

3. Y. H. Hsii, et al., TACO 2 - Fuel Pin Performance Analysis, BAW-10141, August 1979.

i B-13

- 3

' Quest 10n 7 Section 6.6 also indicates that a sensitivity study shows very little difference in the pump coastdown results with regard to the choice of nucleate boiling correlation. Provide more detail of the sensitivity study. performed to select the nucleate boiling correlation Resoonse VIPRE-01 contains a number of heat transfer correlations for each of the '

four commonly recogni:ed modtis of heat transfer: single-phased forced convection, subcooled and s>torated nucleate boiling, transition boiling, and film boiling. Since only conditions up to the point of DNB are of interest during a pump coastdown transient, the code can be restricted to consider only convection and nucleate boiling heat

' transfer, speeding up the solution procedure.

To quantify the effect of different heat transfer correlations on the local coolant conditions and MDNBR during a pump coastdown transient, the following nucleate boiling correlations were studied using the default EPRI forced convection correlation:

Subcooled Saturated Nucleate Boiling Nucleate Boiling THOM THOM THSP* THSP CHEN CHEN

• Them plus the EPRI single-phased forced convection correlation The results given in Table 1 show that the choice of nucleate boiling correlatiens makes very little difference in the MDNBR during a two pump coastdown transient. The Thom subcooled and saturated nucleate boiling correlations, which yielded a conservative MDNBR, will be used along with the EPRI single-phased forced c0nvect10n correlat10n for the generic oconee pump-coastdown analyses.

i B-14

W ,

jl.

, , -t

- Table1. VIPRE-01 Nucleate Boiling Heat Transfer Correlation'.

Sensitivity Study ,.

~

t i

MDNBR BWC i

Saturated & Subcooled Nucleate Boiling Correlations

-Time THOM THSP CHEN ,

sec. THOM _THSP M 0,0 1.830 1.630' '1.830 fi 0.5 1.807 1.807 1.808 1.0 1.767- 1.768 1.770 '

1.5 1.708 1.709 1.712-2.0 1.636 1.637 1.642 2.5; 1.558 1.560 1.565 2.7 1.517 1.518 1.523 2.9 1.489 1.489 1.494 3.1 1.454 1.455 1.460-3.3 1.420 1.421 1.426 4 3.4 1.376 1.378 1.383 3.5- 1.333 1.335 1.341 3.6 1.308 1.310 1.315 3.7 1.285 1.286' 1.291 3.8 1.260 1.261 1.267 3.9 1.238 1.240 1.245 4.0 1.221 .1.223 1.228 .i 4.1 1.216 1.221 1.224 4.2 1.221- 1.234 1.235 4.3 1.250 1.259 1.258 I

B-15

%, ,t

. ]

gW e

.s k

@ Dmm Powen COMPMY ,

' P.Oi mOX 33189 C F A a8 a T12. M.C. 9 89 49 1 .

~

,, ', !L% 8. reseman

. . ice TUCKER - p 3pg _ ,

., . -cveta.s essocco.on a

- June 19. 1989 U. S. Nuclear-Regulatory Commission Washington, D. C. 20555 l Attention: Document Control Desk  ;

w I

Subject:

Oconee Nuclear Station.

' Docket Numbers 50-269 -270, and -287 Response to Questions Regarding Differences Between Duke Topical Reports DPC-NE-2003 and DPC-NE-3000

~

During a telecon on June 13, 1989, the NRC staff requested additional'information ,

to clarify the intnaded applications and other technical details regarding the

• j VIPRE-01 models for Oconee submitted in DPC NE-3000. Revision 1 and in DPC-NE-2003. This letter provides that information. In general, the VIPRE-01  ;

'models described in DPC-NE-2003 are applied in the thermal-hydraulic design of-each reload core. The VIPRE-01 models described in DPC-NE-3000 are applied in -

the_ prediction of the minimum DNBRs resulting from FSAR Chapter 15 transients. A more detailed description of the applications of these models follows.

DPC-NE-2003 describes the VIPRE-01.models and methodology to be used for reload thermal-hydraulic analyses. The steady-state analyses that determine the thermal-hydraulic limits that define the regions of safe operation in terms of -

. power'1evel, reactor coolt.nt temperature and pressure (Pressure-Temperature curves),1and power distribution (RPS Maximum Allowable Peaking'(MAP) limits) are -

described in:this report. -The steady-state' analyses, based on the limiting

'l two pump coastdown statepoint, that determine the-allowable power distribution

-during tho' limiting DNBR transient (Operational MAP limits) are also described.

The methodology for determining the limiting statapoint during the two pump coastdown transient is included. These analyses are routinely performed for a s

[; reload core to demonstrate that applicable safety criteria are met.

k As discussed in DPC-NE-2003, two additional hot channel f actors to account for power rpikea due to spacer grids, and axial nuclear uncertainty are applied to y

theilocal heat flux f actor, F ", only when calculating MAP limits. The two sets of MAP limits. RPS and operational MAP limits, are used to demonstrate 1 L

l that peaking will be acceptable during steady-state operation and during K anticipated transients. All other core thermal-hydraulic analyses (calculation of pressure-temperature curves, FSAR Chapter 15 analyses) are based on the reference design peaking given in the appropriate reports and F " without the additional hot channel factors. This approach is consistent 9 with the current application of hot channel factors in the NRC-approved methodology described in the Duke Power topical report NFS-1002. The use of the VIPRE-01 code has no impact on this approach.

B-16 .

c , - .. - - . , . . - -

l U. 3. Nuclear Regulatory Commsssion June 19, 1989 )

Page 2 ]

The reference axial peaking (1.50) used in the two-pump ceastdown transient is also used in the FSAR Chapter 15 transients to verify that the results arm acceptable. The higher reference axial peaking factor ( ) given in DPC-NE-2003 indicates the objective of using a higher value which results in less i limiting Operational MAP limits. A higher reference axial peaking factor yields I a lower two-pump coastdown MDNBR which results in higher allowable peaking. The  !

methodology described in DPC-NE-2003 is applicable to any axial peaking l assumption, provided that the resulting DNBRs and other peaking factor-related i aspects are addressed. The cutrent value of the reference axial peaking factor used in the MAP methodology is 1.50. Prior to increasing this value to, for example, , a complete evaluation of all potential safety concerns will be performed.

The VIPRE-01 SER states that "the use of VIPRE-01 with an approved CHF correlation and its safety limit should be justified by showing that, given the '

correlation data base, VIPRE-01, gives the same or a conservative safety limit."

VIPRE-01 was used to predict the BWC CHF test results as discussed in Section 5.13 of DPC-NE-2003. The VIPRE-01/BWC results yield a DNBR limit of 1.161: thus, it will be conservative to use the NRC approved BWC correlation limit of 1.18 for ,

all Ocones thermal-hydraulic analyses.

DPC-NE-3000 Section 2.3 describes the VIPRE-01 models to be used for predicting the minimus DNBRs resulting from FSAR Chapter 15 transients. The one exception ,

is the two pump coastdown described above, which is analyzed with the models described in DPC-NE-2003. The two pump coastdown is a unique transient in that it is an integral part of the reload thermal-hydraulic design methodology.

Therefore, the VIPRE model used for the two pump coastdown should be the same model used for all other reload design thermal-hydraulic analyses. As discussed in DPC-NE-3000, Section 2.3.4 the VIPRE methodology for transient analyses includes a few differences when compared to the DPC-NE-2003 methodology. These differences when compared to the DPC-NE-2003 methodology. These differences are either necessary for meeting the modeling requirements of transient analyses, or incorporate additional conservati2ms beyond those in the DPC NE-2003 methodology.

These additional conservatisms are desired in order to build margin into the transient DNBR results and avoid the need for reanalyzing transients in the future. It would be undesirable to use the DPC-NE-3000 VIPRE models as part of the normal reload thermal-hydraulic design process due to these differences.

In order to sypport the Oconee Unit 3. Cycle 12 reload licensing effort, an SER on DPC-NE-2003 is needed by August 15, 1989. If you have further questions regarding this matter, please contact Scott Gewehr (704/373-7581) or Gregg Swindlehurst (704/373-5176).

i Very truly yours.

W .

H. B. Tucker SAG 171/lcs L B-17

e h

'U S. Suelotr R:guletory Commission June 19, 1989 .

Page 3 cc L. A. Wrens, Project Manager ,

Office of Nuclear Reactor Regulation U. S. Nuclear Regulatory Commission Washington, D.C. 20555 Mr. S. D. Ebneter, Regional Administrator '

O. L. Nuclear Regulatory Commission - Region II 101 Marietta Street, NW - Suite 2900 Atlanta, Georgia 30323 Mr. P. H. Skinner Senior Resident Inspector Oconee Nuclear Station Mr. D. Katze Reactor Systems Branch U. S. Nuclear Regulatory Commission -

Washington, D.C. 20555 Mr. Y. H. Hsil Reactor Systems Branch U. S. Nuclear Regulatory Commission Washington, D.C. 20555 Mr. M. W. Hodges Reactor Systems Branch l U. S. Nuclear Regulatory Commission I

l Washington, D.C. 20555  ;

)

l l

I I

l s

l l

1 B-18 1

' *" W S 1

1 i

l l

l I  ?

i t

i i

i h

L t

)

I 6

I i

t

~

' i. . . . .

a

+v. n ..,,. , , - + -- .,,n.,-- - ..--- - ---. -, - ,-- -- ,