ST-HL-AE-1767, Comparison of Treat & Notrump Small Break LOCA Transient Results

Comparison of Treat & Notrump Small Break LOCA Transient Results

ML20215C043
Person / Time
Site:	South Texas
Issue date:	09/22/1986
From:	Frantz E, Huang P, Ofstun R WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP.
To:
Shared Package
ML19292F992	List: ML20210Q647 ML20210Q685 ML20215B637 ML20215C022 ML20215C032 ST-HL-AE-1767, Comparison of Treat & Notrump Small Break LOCA Transient Results
References
ST-HL-AE-1767, WCAP-11297, NUDOCS 8610100056
Download: ML20215C043 (120)

Text

. - _ _ _ _ _ _ _ _ .

WESTINGHOUSE CLASS 3 WCAP-11297 COMPARISON OF THE TREAT AND NOTRUMP SMALL BREAK LOCA TRANSIENT RESULTS t

E. R. Frantz P. H. Huang R. P. Ofstun WESTINGHOUSE ELECTRIC CORPORATION Nuclear Technology Systems Division P. O. Box 355 Pittsburgh, Pennsylvania 15230 .

e 8610100056 860930 PDR ADOCK 05000498 E PDR 9734Q.1D/092286 1

TABLE OF CONTENTS Section Title Page Summary _

8 1 Introduction 10 2 Description of Principal TREAT Models 11 3 Description of the South Texas Plant TREAT and NOTRUMP Model 38 3-1 South Texas TREAT Model and Unique Plant Features 3-2 South Texas NOTRUMP Model 4 ' Comparison of the TREAT and NOTRUMP Small Break LOCA Transients 44 4-1 Modeling Assumptions and Initial Conditions Comparison 4-2 Comparison of the Analysis Results

. 5 TREAT Compliance with 10CFR50, Appendix K 53 6 Conclusion 59

. 7 References 61 Tables 63 Figures 72 Appendix A - NOMENCLATURE 116

-9734 0:10/092286 3

~

List of Tables Table 2-11-1. Reactor Protection System Actions Modeled Table 3-1-1. South Texas Plant Data used in TREAT Primary System Table 3-1-2. South Texas Plant Data used in TREAT Secondary System Table 4-1-1. TREAT /NOTRUMP Initial Condition Comparison for 1.5 Inch Cold Leg Break LOCA Analysis Table 4-2-1. Time Table of Events STP 1.5 Inch Cold Leg Break with Cooldown Table 5-1. TREAT Model Compliance Summary List of Figures Figure 3-1-1. South Texas Safety Injection System Figure 3-1-2. TREAT RCS Noding Diagram

~

Figure 3-1-3. TREAT Secondary Noding Diagram Figure 4-1-1. Comparison of Minimum and Best Estimate SI Figure 4-2-1. TREAT RCS Pressure and NOTRUMP Pressurizer Pressure Figure 4-2-2. TREAT RCS Pressure and NOTRUMP Cold Leg Pressure Figure 4-2-3. Steam Generator No. 1 Pressure Figure 4-2-4. Steam Generator No. 2 Pressure Figure 4-2-5. Upper Plenum Mixture Level ,'

97340:10/091286 4

Figure 4-2-6. Upper Head Mixture Level Figure 4-2-7. Pressurizer Mixture Level Figure 4-2-8. SG No. 1 Inlet Side Mixture Level Figure 4-2-9. SG No. 1 Outlet Side Mixture Level Figure 4-2-10. RCP No. 1 Inlet Mixture Level Figure 4-2-11. SG No. 2 Inlet Side Mixture Level Figure 4-2-12. SG No. 2 Outlet Side Mixture Level Figure 4-2-13. RCP No. 2 Inlet Mixture Level Figure 4-2-14. SG No. 1 Secondary Side Mixture Level Figure 4-2-15. SG No. 2 Secondary Side Mixture Level Figure 4-2-16. SG No. 3 Secondary Side Mixture Level Figure 4-2-17. Upper Plenum Void Fraction figure 4-2-18. Core Inlet Flow Figure 4-2-19. Upper Head Flow l Figure 4-2-20. Pressurizer Surge Line Flow l

I Figure 4-2-21. Loop No. 1 Hot Leg Flow l Figure 4-2-22. Loop No. 2 Hot Leg Flow l -

j 97340;1D/091286 5 l

Figure 4-2-23. Loop No. 3 Hot Leg Flow Figure 4-2-24. Loop No. 4 Hot Leg Flow Figure 4-2-25. Fluid Break Flow 1

Figure 4-2-26. Vapor Break Flow Figure 4-2-27. TREAT Integrated Break Flow Figure 4-2-28. Safety Injection Flow Figure 4-2-29. SG No. 1 Secondary Steam Flow Figure 4-2-30. SG No. 2 Secondary Steam Flow Figure 4-2-31. SG No. 2 Auxiliary Feedwater Flow Figure 4-2-32. Core Exit Temperature Figure 4-2-33. Pressurizer Mixture Temperature Figure 4-2-34. Cold Leg No. 1 Temperature Figure 4-2-35. Cold Leg No. 2 Temperature Figure 4-2-36. Cold Leg No. 3 Temperature Figure 4-2-37. RCS Downcomer Temperature Figure 4-2-38. TREAT Loop 3 RCP Inlet and Outlet Temperatures Figure 4-2-39. TREAT Inner, TREAT Outer, and NOTRUMP Lumped Upper Head Metal Temperature -

~

Figure 4-2-40. TREAT Core Boron Concentration 97340:1o/091286 6

97340:10/091286 7

Summary The applicability of the TREAT code for analyzing small break LOCA and long term cooling recovery without core uncovery is presented in this report.

A detailed description of the TREAT models is presented. TREAT-is a real-time interactive, two phase, non-equilibrium, non-homogeneous, thermal-hydraulic system code. The code has been used in the past for developing and addressing emergency response strategy, i.e., evaluating operator recovery actions. ~A description of the South Texas specific TREAT model and special features associated with this model is also presented.

To demonstrate the applicability of TREAT for the proposed analysis, the TREAT predictions are critically compared against the NOTRUMP code (NRC approved small break evaluation model) predictions for a 1.5 inch diameter cold leg break. (a small break that does not result in core uncovery). A detailed review of the compliance status of the TREAT models against the 10CFR50 Appendix X requirements is also presented.

The comparison shows that TREAT correctly predicts all the trends throughout the transient. All the important parameters predictions by TREAT, including; RCS pressure, core exit fluid temperature, core level, pressurizer response,

~

loop flow, and steam generator response, are in good agreement with the NOTRUMP results. RCS inventory depletion is accurately predicted (integrated break flow is within 3% at the end of one hour). Loop asymmetric effects are closely simulated and the codes show good agreement of the RCS response to cooldown caused by operator action to dump steam from two steam generator PORVs.

97340:1D/091286 8

l The results of the Appendix K compliance review show that TREAT is in compliance with the Appendix K modeling requirements in all areas that impact the thermal-hydraulic response of a small break LOCA without core uncovery.

Areas in Appendix K that deal with the modeling of large break LOCA response and the modeling of small break LOCAs that cause significant core uncovery are not necessary for the proposed application and are not covered by this compliance review.

Based on the results from the Appendix K compliance evaluation and the TREAT /NOTRUMP transient comparison, it is concluded.that TREAT has the necessary and required models and can adequately model the plant response including operator recovery actions for a small break LOCA transient that does not uncover the core. TREAT is, therefore, adequate and suitable for analyzing the long term cooling recovery issue for the South Texas Plant.

9734Q:1D/091286 9

Section 1 Introduction The purpose of this report is to demonstrate the applicability of TREAT, a real-time, interactive, thermal-hydraulic code for performing small break LOCA and long term cooling recovery analyses which do not cause core uncovery.

To demonstrate the applicability of TREAT for the proposed analysis, a small break LOCA transient generated by the TREAT code and input model was compared with the same transient generated by the NRC approved NOTRUMP code small break LOCA evaluation model (References 1-1 and 1-2). A TREAT input model, consistent with the evaluation model, was prepared for the South Texas Project (STP). A 1.5 inch small break LOCA was chosen for the transient comparison since it provides a good example of the natural circulation and two phase flow predictions from both codes. Loop asymmetry consistent with system design was also considered. This involved feed to and subsequent cooldown with two of the four steam generators plus safe.ty injection (SI) flow delivery to only one loop. Although substantial reactor coolant system voiding occurred (due to the evaluation model's conservative assumptions with regard to SI flow), core uncovery does not occur.

~

A general description of the TREAT code is presented in Section 2 of this report. The STP TREAT and NOTRUMP plant models are described in Section 3. A comparison of the principal parameters of the small break LOCA transient is presented in Section 4. Finally, Section 5 contains a summary which indicates the degree the TREAT code and input model complies with the Appendix K requirements.

O 9734o10/091286 10

Section 2 Description of Principal TREAT Models 2-1 Introduction TREAT is a real-time, interactive, two phase, nonequilibrium, nonhomogeneous, thermal-hydraulic network code. The network consists of a system of nodes connected by flow links. Each node may contain two separate regions

a steam region and a mixture region. The regions are separated by a moving interface. Properties in each region are solved independently by using two mass and two energy conservation equations. The mass equations are solved explicitly, and the energy equations are solved using a predictor / corrector method. In conjunction with the corrector part of the solution, a global pressure is found which conserves global volume. Although overall system volume is conserved, the fluid in an individual node might not occupy the same volume as the physical node. This local volume error is coupled to the momentum equation using ( ']a,c to obtain the.

volumetric flow in each flow link. By applying drift flux correlations, the total volumetric flow is separated into vapor and liquid flows.

The TREAT pressurized water reactor (PWR) system includes models for neutronics, heat transfer, automatic controllers, plant protection systems, boundary flows, and reactor coolant pumps (RCPs). The neutronic models compute the axial flux, power, and fission product distributuions in the core. The heat transfer models compute core and steam generator (SG) heat transfer, as well as conduction-limited, thick metal heat transfer. The simulated controllers for reactor power, pressurizer pressure, pressurizer

level, steam / feed flow, and SG level all operate automatically in response to changes in load demand; they may also be placed under manual control. The Reactor Protection System (RPS) monitors the reactor trip, SI actuation, turbine trip, steamline isolation, feedwater isolation, letdown isolation, and auxiliary feedwater actuation setpoints. The RCP model uses a four quadrant homologous curve to compute the pump head. The boundary flow models either automatically control or allow the user to manually adjust the SI, charging, letdown, pressurizer or SG power-operated relief valve (PORV), and spray flows.

97340:1D/091286 11

l 2-2 MODEL COMPONENTS The TREAT model components are interior fluid nodes, boundary fluid nodes, j conduction limited metal nodes, SG tube-type metal nodes, interior flow links, j boundary flow links, and heat links. A specific representation of the RCS or the SG can be constructed by using the components to form a network of multiple fluid nodes which are appropriately interconnected by flow links.

An interior fluid node is defined as a fixed control volume containing a variable amount of fluid mass and energy. No flow (only mass and energy inventories) is associated with a fluid node. An interior fluid node may be connected with other fluid nodes by flow links.

A boundary fluid node is defined as a control volume containing fluid at a specified pressure and enthalpy. A boundary fluid node has no volume or mass associated with it. It may be connected with interior fluid nodes by boundary flow links.

A boundary flow link is defined as a path for fluid flow where the net mass flow rate is a specified function. A boundary flow link always connects an interior and a boundary fluid node.

An interior flow link is defined as a path for fluid flow between interior fluid nodes. A momentum conservation equation is solved for each interior flow link. No mass and energy inventories (only flow) are associated with a flow link.

A conduction limited metal node is defined as a control volume containing a metal mass and energy. A 1-D heat conduction equation is solved to determine the temperature distribution within each metal node. Each metal node is connected directly with fluid node via a heat link.

A heat link is defined as a path for energy transfer from a conduction limited metal node (thick metal slab) to a fluid node. A maximum of eight SG tube metal nodes are available for modeling SG tubes. A single temperature is used 97340.1o/092286 12

l to represent the lumped tube metal. These tube metal nodes are connected to one primary and one secondary fluid node via the special SG tube heat links.

2-3 REACTOR CORE THERMAL-HYDRAULICS MODEL The reactor core model is made up of three separate modules. T'hese are the neutron kinetics module, the fuel rod energy module, and the core fluid thermal-hydraulics module. This section describes the fuel rod and core fluid modules. The neutron kinetics module is described in Section 2-4.

2-3-1 Fuel Rod Energy Module A fuel rod has a temperature distribution which is dependent upon both the local fluid thermal-hydraulic conditions and the local power production. The fuel rod energy module models an average fuel rod in the core to obtain this distribution.-

In order to determine the axial temperature distribution, the fuel rod is divided into ( la,c nodes. The radial temperature distribution within each axial node is also computed. Energy exchange between and within the nodes is by conduction. A convective heat transfer boundary condition is applied at the rod surface.

The two-dimensional conduction equation (for an average fuel red) written in cylindrical coordinates is given below. Azimuthal symmetry has been assumed.

h=a( +fh+)+h (2-1)

To solve this equation, the radial temperature distribution is assumed to be a polynomial of the following form:

( Ja,c (2-2)

Substituting equation (2-2) into (2-1) results in an equation with the three unknown coefficients a, b, and c for each of the axial nodes. -

97340.1D/091286 13

l l

The heat transfer boundary condition at the rod surface supplies a second equation to solve for the coefficients.

l

-K fhr=R =f h,ff(T - T,yg) = q (2-3)

~ ~

where 1

h,ff = 1 ar h

p K

C eh I fq) ,

[ la,c (Reference 2-1), which is basically an error minimization technique, is applied to equations (2-1) and (2-2) to obtain three linearly independent equations for the solutions of the coefficients a, b, and c.

Once the coefficients are determined, the temperature distribution, T(r,z),

within the rod can be calculated by using ( ]a,c ,

2-3-2 Core Fluid Thermal-Hydraulics Module The core fluid is represented by [ la,c in the TREAT network model. However, within this region the fluid properties are not homogeneous,

~

thus a model is needed to compute'the fluid enthalpy rise up the core. To

. simulate this, [ Ja.c fluid subnodes, corresponding directly to the

[ la,c fuel rod nodes, are used to model the core fluid.

~

The core fluid thermal-hydraulic module computes the average enthalpy, quality, specific volume, and temperature for each of these ( la,c fluid subnodes in the core. This is accomplished by coupling the mass con-servation equation to the transient fluid energy equation. The resulting equation is transformed to an ordinary differential equation using the method of characteristics.

= q + alongpAh=W p

(2-4) 97340.1D/091286 14

This, in turn, is written as a finite difference equation in both the implicit and explicit forms and solved for each fluid region in the core.

The fluid in the core is subdivided into three regions when applicable. These are the subcooled water, saturated mixture, and superheated steam regions.

Moving interfaces separate the three regions.

In the subcooled water region, the [ Ja,c calculated values of the mass flow rate are used to compute the characteristic line. The mass flow rate in the saturated region is equal to (

Ja,c . Similarly, a mass conservation equation relates the mass flow rate in the superheated region to the saturation flow rate and [ la,c ,

A core reflux model has been incorporated in TREAT to model the multidimensional flow effects of the reactor core during two phase, low-flow conditions (i.e., steam generator reflux cooling). The core reflux model simulates the liquid flow returning to the lower core and downcomer via the peripheral of the core. Based on the computed core exit component flow rates, the upper plenum void fraction and the counter current flooding limit, the model transports a fraction of liquid in the upper plenum into the lower plenum through [ la,c ,

2-3-3 Reactor Core Heat Transfer The fuel rod and core fluid modules are coupled through heat transfer. The effective heat transfer coefficient at the fuel surface, which is used as a boundary condition for the rod conduction equation, is a combination of the convective and gap heat transfer coefficients.

The core convective heat transfer coefficients, which are dependent upon the axial fluid properties and clad temperatures, are determined using the following correlations:

o Forced convection -- (Reference 2-2) ,'

97340.lo/092286 15

Nu = (0.0333 Fluid + 0.0127) Re.8 Pr.33 (2-5)

Cell where Re = Reynolds number .

Pr = Prandtl number o Nucleate boiling -- Jens/Lottes (Reference 2-3) q = 278[

g sat) ,P*/900 34 (2-6) o DNB -- Rohsenow and Griffith (Reference 2-4) q = (143) ( - 1)0.6 (2-7)

g f The convective heat transfer coefficient is used along with the clad surface temperature and axial fluid temperature to calculate the heat removal rate from the fuel.

The heat transfer coefficient used to calculate the temperature difference j ,

across the gap between the clad inner surface and the fuel pellet surface is as follows (Reference 1-2):

l a,c (2-8) e l

I O

e 9734Q:10/092286 16

l l

l Because of the axial fuel subnode structure of the core node, thick metal heat l transfer to the core node (via metal nodes) is not modeled.

2-4 NEUTRON KINETICS MODEL i

l The TREAT neutron kinetics calculations are handled by three subroutines, FLUX, RCSBOR, and DECAY. The axial neutron flux and power distribution are computed in FLUX, the nodal boron concentration is computed in RCSBOR, and the fission decay product concentrations are computed in DECAY.

. 2-4-1 Neutron Kinetics Module A coupled space-time neutron kinetics equation is used to model the reactor's total power and core spatial power distribution. The neutron population in the core is represented by the following equation:

hh=f(1-6)vI4+ f biCj g + S - v+J - I,# [ Ja,c (2-9)

. i=1 This equation relates the change in neutron concentration to the production, leakage, and absorption rates. The production rate is further broken down into a term for direct fission production, a term for delayed neutron precursor decay, and a constant source term.

The delayed neutron precursors are modeled with (six groups]a,c in TREAT.

The delayed neutron precursor concentration is represented with a first-order differential equation as follows:

dC dt Oi " f' - ii [ ]> ( -10)

O e

9734Q:10/091286 17

The leakage term is modeled by applying Fick's law of diffusion.

l J = -DW (2-11)

The kinetics model uses an average [ Ja,c cross-section data at each space node for the space-time neutron kinetics calculations. [

neutron cross ja,c ,

Therefore, the average cross-section data calculated by this method includes the effects due to [ la,c in the core. The fast and thermal neutron macroscopic absorption cross sections (at each space node) are calculated with the following equations:

a,c e

where a,c 9734Q.1D/091286 18

a,c The input data contains macroscopic cross sections evaluated at a reference water density and resonance-effective fuel temperature. These input macro-scopic cross sections are modified for the following:

. a,c The neutron kinetics equations (2-9 through 2-11) are integrated over space and time by the finite difference technique. The nuclear core is represented by [ Ja,c fuel nodes and [ la,c reflector nodes. Both radial and azimuthal symmetry are assumed; this simplifies the equations considerably.

The simplified set of difference equations results in a tridiagonal matrix which is solved by forward elimination and back substitution.

~

2-4-2 Nodal Boron Module The change in nodal boron concentration is modeled with a simple first order differential equation.

hg[(M)nnB g l

• I INf )tB n - I INf)t Bn (2-14) l IN OUT where B = upstream nodal boron concentration l n l- Bn = nodal baron concentration This equation is solved for each node in the RCS and SG models.

1 l

97340:10/091286 19

2-4-3 Decay Heat Module The calculation of the decay heat power in TREAT is based on the data provided in ANS Standard 5.1 (Reference 2-5). The decay heat contribution from residual fission is not included in this calculation since it is modeled by the neutron kinetics equation.

The fission decay product concentrations are updated in subroutine DECAY. An

[ la,c decay heat model is used to simulate the heat release from fission product decay. An additional [ ] model is used to simulate the actinide decay heat.

dX.

g=1[ 3

-X]3

[ Ja,c (2-15) where X

3 = 1 C /Y3 3 3 n(o)

1. = disintegration rate for j group C = concentration for the j th group Y = fission yield of the j th group n(o) = steady-state total fission rate The fission product poison concentrations (xenon and samarium) are modeled with first order differential equations of the following form:

dX dt = Yy,Ife + 1;Xg - (1Xe + '2 Xe2) XXe (2-16)

The fission product poisons are formed from dircet fission or from the decay of other fission products. The first term represents the direct yield, the second term represents the decay of a parent atom, and the third term represents the decay of the poison.

2-5 PRIMARY / SECONDARY N0DAL AND NETWORK MODELS The RCS fluid network and the SG fluid network are separate subsystems in TREAT. Each network consists of a series of nodes connected by flow links.

9734Q 10/092286 20 l

l

TREAT employs nonequilibrium, nonhomogeneous flow assumptions on both the l

i primary and secondary subsystems. Each node may contain two separate regions: a steam region and a mixture region. Fluid properties in each region are solved independently through mass and energy conservation equations.

i In order to reduce the complexity of this set of equations, some simplifi-cations are made. All fluid properties within a given subsystem (primary or secondary) are computed using a single global pressure for that system. The

advantages of using global compressibility are a large reduction in computing time and an increase in the stability of the system of equations. The disadvantage of using global compressibility is that acoustic waves cannot be computed without some auxiliary equations.

The momentum equation is written in terms of volumetric flow rates that use local volume errors between adjacent nodes, along with calculated frictional loss and gravitational forces, to compute the flow rate in an interior flow link. By writing the momentum balance in terms of volumetric flows, the squations are much more stable than when they are expressed as mass flow rates.

Within each node, correlations are applied to compute heat and mass transfer between the vapor and mixture regions (see Section 2-5-4). There is no need to develop special numerics to treat the pressurizer or upper head.

When coupled with the volumetric momentum equations for the flow links, the nodal mass and energy equations form a system of five conservation equations.

The nodal mass and energy conservation equations are detailed on the following pages.

In addition to these equations, several constitutive equations are imposed on the system. In particular, drift flux correlations are used to convert the total volumetric flux into the vapor and mixture mass flow rates used by the mass equations. Heat transfer correlations are used to compute heat transfer rates to and from the fluid. These are described in Sections 2-6 and 2-7.

D e

97340:1o/091286 21

2-5-1 Mass Conservation Equations The mixture and vapor mass conservation equations solved by TREAT are given in equations (2-17) and (2-18).

~

o Mixture mass conservation  !

l d INM )n = a dt I

!=1 nt - ((1-Dnt)INf )t +C nt

• INg )tl

+WCD , gC ,yCI _ gel _ gBR ,yDF  ; n = 1,...,N; (2-17) n o Vapor mass conservation d ING )n , _ Il

• ng * [Dnt(Nf ) t + (1 - C3g) - (Wg )tl t=1

-W Cn -N -N +NEI ,gBR _ gDF  ; n = 1,...,N g (2-18) n where atn = on ,u(t) - 6n,d(t)

' ~

. 1 if i = j o

g,3 - (0 if i = j C

nt represents (

3a,c ,

a,c (2-19) l 97340;1o/092386 22

l l

l l

l b nt represents the (

Ja c in fluid node n and is given by equation (2-20).

a,c (2-20) where (Etop)nt =6 n,u(t)

• I op)2 +6 n,d(t)

• I op)t and (Ebot)nt =6 n,u(t) * ( bot)t +6 n,d(t)-(Efot)t D

nt represents the (

ja,c ,

""~

a,c All liquid which enters a fluid node through a flow link is assumed to flow directly into the mixture region.

The terms of the form W** are the interface mass transfer rates. The rates are computed by using empirical correlations (see Section 2-5-4) before the mass conservation equations are integrated. The mass conservation equations are integrated by an explicit numerical scheme.

2-5-2 Energy Conservation Equations and Global Pressure Calculation @G The energy conservation equations (written in terms of enthalpy) solved by

~

TREAT are given in equations (2-21) and (2-22).

973401D/091286 23

o Mixture energy conservation d(Hg )n _ 144 y dP* ,_

• nt - {(1-Dnt) (Nf )t h,

+Cnt (Ng )t h3 e

i

+W C C 0F n (hg)n + N n hf+W n (hG)n

• W n hf -WEI (h gg)n _yBR (h gg)n C

+Of0 -Q n +QEI + QM  ; n = 1, . . .,N g (2-21) o Vapor energy conservation d(HG )n _ 144 y dP* =-I h dt J G dt t=1 a"E* (D"E (WI )*

+ (1 - Cnt) ( g)t*hl e

-W Cn C (hg)n _ y n (hg)n - W h +WfI(hgM)n ,yBR (h gy)n g

, -W DF h f

-Q Cn +Q CI -OfI+0 6

n = 1, . . .,Ny (2-22) e nt represents the (

ja,c ,

a,c nt represents the [

3a,c ,

O I

9734Q:10/092386 24

a,c Coupled to these equations is the constraint equation given by the following:

Ng I ((Mg)n v ((hy)n,P*] + (MG)n v ((hG)n,P*] - Vn) = 0 (2-23) n=1 This is the requirement that the global volume error be zero.

Equations (2-17), (2-18), (2-21), (2-22), and (2-23) will form a set of 2Ng+1 cquations and with the same number of unknowns in g(H ); . . .,.N y , (HG )l' . . .,N y '

and P*.

In the energy equation, the node enthalpies are based on the [ ]a,c cnthalpies. The internode heat and mass transfer rates are computed using correlations (see Section 2-5-4) before the energy equations are integrated.

a,c 2-5-3 Volumetric Momentum Conservation Equation l

The momentum equation solved by TREAT is as follows:

N dq L I C

=

(I a P

~

IN!9 + D ) g ; e = 1, . . .,L dt e n=1 nt n t t e N . (2-24) 9734Q:1o/092386 25

g_-

1 l

, where

. 1449 c l

t " (L/A)t In addition to this equation, there are two sets of constraints that must be

atisfied along with the equation. The first requirement is that the change in pressure around any cycle be zero. A cycle is a path or a set of links connecting a node with itself. The second requirement is the volume error in a node must equal the sum of the volumetric flows into and out of the node.

_ By applying these constraints, the original set of L m N mentum equations is reduced to ( Ja,c . These equations are then solved by explicit integration.

Equation (2-24), together with its constraints, forms a system of simultaneous equat, ions for calculating the net volumetric flow rates in all interior flow links. The original system of simultaneous equations is reduced to a simpler set of equations by the network theory through the introduction of a different

, set of network variables. The [ Ja,c method is used to solve this reduced set of equations. This method avoids the need to compute the pressure drops in the links explicitly. These terms can be obtained by substituting the calculated link flow rates from equation (2-24). Details of

~

the general method are given in References 2-8 and 2-9.

2-5-4 Interface Mass and Heat Transfer Models The mass and heat transfer rates for a nonequilibrium, stratified interior fluid node given in equations (2-17), (2-18), (2-21), and (2-22) are defined l

as follows:

l l

O e

97340.lo/092386 26

a,c (2-26)

(2-27)

(2-28)

(2-29)

(2-30)

(2-31)

(2-32)

(2-33)

(2-34)

E (Agg)n is the heat transfer area between the vapor region and the wall of node n. (AMG)n is the surface area of the interface bet m <. the mixture and vapor regions of node n. They are defined as f'et in . of the volume fraction of the two regions.

O O

9734Q:10/092386 27

2-6 HEAT TRANSFER MODELS The heat transfer between metal surfaces and interior fluid nodes is computed by four subroutines in TREAT: RCSHTX, GECHTX, RCSHTR, and SECHTR. The first l

two subroutines compute the convective heat transfer coefficients for the

~

primary and secondary heat links. The last two subroutines compute the heat flow rates.

The following heat transfer correlations are used by TREAT:

o Forced convection -- Dittus/Boelter o Nucleate boiling -- Jens/Lottes o DNB -- Rohsenow and Griffith o Condensation -- Westinghouse proprietary To determine the heat transfer coefficient, the average flow thrcugh the fluid node is required. From this information, the Reynolds number and the forced convection heat transfer coefficient are computed. If the surface temperature is greater than saturation, the nucleato boiling heat transfer coefficient and DNB heat flux are computed. The minimum of the two is then compared with the forced convection heat transfer coefficient, and the maximum is used to compute the heat flow rate.

G O

l l

l 97340:1o/091286 28

2-7 DRIFT FLUX MODELING The total volumetric flow rate for each flow link is computed by equation (2-24) in subroutine RCW. Drift flux and flooding correlations are then applied to each flow link to compute the component flow rates when two phase conditions exist. For a vertical flow link, the following drift flux correlations are used in TREAT:

o Bubbly flow a,c (2-35) where

-~~

a,c o Slug flow a,c (2-36)

. o Annular flow a,c (2-37)

A simplified link void fraction flow-regime-dependent map, based on the NOTRUMP flow regime map (Ref. 1-2) is used to determine the proper correlation to apply.

For horizontal flow links, the user may select a relative velocity which will then be used to compute the component mass flow rates. A value of 0.0 corresponds to a homogeneous link. .

4 97340:10/092286 29

The link average void fraction for concurrent flow is determined as follows:

( ja,c (2-38)

For countercurrent flow, a weighted combination of the upstream and downstream link void fractions calculated with equation (2-38) is used for the link void fraction.

To ensure that the flooding limit is not violated by the component fluxes in countercurrent flows, a flooding correlation is applied. The TREAT flooding correlation is the same as the correlation used by NOTRUMP (Ref. 1-2, Appendix W). The component flux determined by the drift flux model is modified as necessary according to this correlation. After the component fluxes have been determined, the link specific volume is used to compute the component and total mass flow rates.

o O

9 07340.10/091286 30

2-8 BOUNDARY FLOW LINKS AND BREAK FLOW MODELS Boundary flow links allow the user to simulate charging and letdown, SI, accumulator injection, pressurizer spray flow, break flow, PORV and s'fety valve flows, etc. Models for these functions can either be incorporated by

~

the user or the existing models can be used.

The charging pump head versus flow curves is input for the charging / letdown.

Next, the valve opening is computed based on the level deviation from the pressurizer level program. The letdown orifices are not modeled; instead, a critical flow link exists so that a user-specified letdown flow rate may be input.

The SI model is similar to the charging pump model. The pump head versus flow curves are input, and the SI flow rate is computed based on the RCS pressure.

Links- to each-cold leg are modeled and may be individually failed. Also, each pump may be individually failed to simulate the assumptions used for an Appendix K design basis or more realistic E0P-type analysis.

The cold leg SI accumulators are modeled as boundary nodes. An ideal gas,

[ Ja,c model is used to simulate accumulator blowdown.

l ,

Links to each cold leg are modeled and may be individually failed.

The pressurizer spray flow is controlled by the RCS pressure controller. When the compensated RCS pressure exceeds the setpoint, spray is automatically initiated. The user also has manual control of the spray flow rate.

Both auxiliary feedwater and main feedwater are modeled. The main feedwater flow rate is controlled by the feedwater flow control model. Auxiliary feedwater flow rate is user controlled. Separate links for each are included.

The steam dump flow rate calculations are performed in subroutine STMDUMP.

The steam dumps may be placed in either the T,yg control mode or steam header pressure control mode. In the pressure control mode, the user sets the e

9734o;10/091286 31

desired steam header pressure. A single link from the steam header is used to model steam flow through the steam dumps.

The main steam flow to the turbine is also modeled. Basically, the turbine demand power determines the main steam flow rate. Currently, a [ ]a,c relationship between the demand power and steam flow rate is used.

Break flow links are used to model the pressurizer PORVs and safety valves, SG PORVs and safeties, tube ruptures, and LOCAs. The following critical flow correlations are used in TREAT:

o Subcooled flow -- Modified Zaloudek (Reference 2-6) o Saturated flow -- Moody (Reference 2-7) o Superheated flow -- Moody The critical flow rate is compared to the flow rate derived from the orifice equation, and the minimum is used. The change in node mass is computed based on the calculated break flow. If the node mixture mass is predicted to reach zero, the break flow rate is adjusted to account for more vapor flow.

Break flow links may be placed anywhere in the system. Also, multiple-break flow paths may be used. For example, a LOCA coincident with a tube rupture and a stuck open PORV may be placed in any or all loops in the model.

l l

1 -

l l

9734Q:10/091286 32 i

2-9 REACTOR COOLANT PUMP MODEL The RCP model determines the pump head and speed during normal operation and transient coastdown. These characteristics are evaluated from normalized pump homologous curves for the pump head and hydraulic torque and from a frictional torque characteristic equation.

~

When the RCP is on, the pump speed is manually controlled by the user. The pump head is then calculated from'the characteristic head-speed-flow homologous curves. The heat addition to the primary coolant from the pump work is also modeled when the RCP is on.

When the RCP is off, the pump torque and change in pump speed are calculated along with the pump head, as described above. The pump hydraulic torque is determined from the torque-speed-flow homologous curves. The frictional torque is calculated from the following:

( Ja,c (2-39) where Tfric = frictional torque in Ibf -ft fRCP1 = frictional torque' coefficient

. a = RCP speed / rated speed fRCP2 = windage torque coefficient The motor torque is neglected. Given the total torque and pump inertia, the change in the RCP speed is determined for each time interval.

e O

e l

9734Q.1D/091286 33 i

l 2-10 STEAM GENERATOR MODELS 2-10-1 Recirculation and Separation The recirculation and separation models are closely linked. To simplify the modeling,[ la,c is assumed. A two phase mixt'ure leaves the tube region and is separated into its component flows by a drif t flux correla- tion. The liquid phase of the fluid enters the mixture region of the steam dome (i.e., it flows to the bottom of the steam dome node). The steam phase of the fluid enters the steam region of the steam dome directly. Since the exit from the steam dome is at an elevation higher than the mixture level, all fluid leaving the steam dome and entering the steamline is saturated steam.

2-10-2 U-Tube Region Because of large heat additions in the U-tube region node, the average node properties and exit properties can be significantly different. To more accurately calculate the U-tube region inlet and exit flow rates and the total tube region mass, a U-tube region quality profile is modeled. This model calculates the exit quality based on the node average quality and inlet quality assuming a ( ]a,c quality profile. Subcooled, saturated, and superheated inlet and exit flow conditions are all considered.

e 9

0 97340:1D/091286 34 I

l l

t

.Because of the design of the quality profile and recirculation models, the U-tube region is modeled as [ Ja,c ,

The heat transfer from the tubes is computed using the following correlations:

~

o Dittus/Boelter for single phase or two phase forced convection -

(Ref. 2-2) o Jens/Lottes for nucleate boiling (Ref. 2-3) o A Westinghouse proprietary correlation for condensation (Ref. 1-2) b O

9734Q.10/092286 35

2-11 CONTROL AND PROTECTION SYSTEMS' MODELS The Reactor Control Systems modeled in TREAT include the following:

o Pressurizer Pressure Control System l o Pressurizer Level Control System ~

l 1

o Reactor Rod Control System o Steam Generator Level Control System o Steam Dump Control System All the time-dependent control systems are modeled by linearizing the governing control system equations and integrating the resulting equation over the TREAT model timestep (typically 0.25 second).

2-11-1 REACTOR PROTECTION SYSTEM The Reactor Protection System monitors and actuates the safeguards systems of the reactor plant. The Reactor Protection System monitors the safeguards setpoints for reactor trip, SI actuation, auxiliary feedwater actuation, turbine trip, main feedwater trip and isolation, steamline isolation, and letdown isolation. Table 2-11-1 lists the protection systems modeled and the

~

parameters on which these systems actuate. All the plant instrument signals are compared directly with the setpoint inputs except for reactor trip on pressurizer pressure and SI actuation on steamline pressure. These signals are lead / lag compensated.

The main steamline header pressure is used for SI actuation and steamline isolation. Since TREAT normally models all SGs by a single global pressure subsystem up to the time of steamline isolation, each individual SG pressure is not monitored.

The overtemperature delta T setpoint is a nominal setpoint corrected for deviations in pressure from the referer.ce pressure, deviations in T,yg from the full power T,yg, and deviations in the axial offset from the operat.ing band. The T,yg correction is lead / lag compensated. ,

97340:1o/091286 36 l

The overpower delta T setpoint is a nominal setpoint corrected for deviations in T,yg above the full power T,yg, increases in T,yg compensated by a high pass filter, and deviations in the axial offset from the operating band (whenapplicable).

I e

e e

S 6

97340.lo/091286 37

. . . . . .~.m.~. . _ , .. . . . . . . . _ _ .. . . , _m. __._m._...m.__, _._m_.._ _ __. .._.m_,_.-.,_ . . . _ , _ . ,

7_

s a

1 r

i i

f-i .

4 a i 1 4

n Y

f. .

k. -

i.

1 4  :

} -1 3

e 4 -

0 4

i 3

I -

f.

I t

J e

a a

i 4

l i

4 t

b I

f i

1 i

i i,

(

i

+

4 e , - - - -- , - - . , . - . - - - _. ..- .._.__.---..m., . - - - -

Section 3 l

Description of the South Texas Plant TREAT and NOTRUMP Models i 3-1 South Texas TREAT Model and Unique Plant Features The South Texas Project (STP) consists of essentially identical Unit 1 and 2 Westinghouse 1250 MWe pressurized water reactors (PWRs), owned and to be operated by Houston Lighting and Power Company. The STP reactor and nuclear steam supply system (NSSS) differs from a W four-loop standard 1150 MWe plant (e.g., Callaway, Wolf Creek, Seabrook, or Millstone 3) in a number of important respects. These include the following: .

1. The active fuel length is 14 feet instead of 12 feet. This accounts for the higher core power (3800 versus 3411 MWt).

2. The higher power results in approximately 10 F higher temperatures.

Secondary steam pressures are also higher by approximately 100 psi (1100 psia versus 1000 psia at full power,1300 psia versus 1200 psia at the lowest safety valve set pressure).

3. The safety injection system of STP is also different.'

In the standard

~

plant, there are 2 charging /SI pumps, 2 high-head SI pumps, and 2 low-head SI

. pumps. These pumps have shutoff head pressures of approximately 2500, 1500 and 200 psig, respectively. The standard plant's charging /SI pumps also 4

provide normal charging and the low-head SI pumps also serve as residual heat removal (RHR) pumps. In the standard plant, there are two emergency busses and diesels. One of each pump type gets loaded on one bus upon receipt of an SI signal. The SI flow is also headered; if one diesel fails, SI flow will i

still be delievered to all four loops. The standard plant also has 4 accumulators (one per loop), which start to inject when RCS pressure is approximately 600 psig.

4 i

97340:1D/092286 38 i

In the STP SI system (see Figure 3-1-1), there are 3 high-head SI pumps and 3 low-head SI pumps with shutoff head pressures of 1500 and 300 psig respectively. There are also 3 emergency busses and diesels. One high-head and one low-head SI pump get loaded on each bus upon receipt of the SI signal. Unlike the standard plant the SI flow in STP is not headered. Loop 4 (or Loop D) plus any loop associated with a failed power train ~will not recieve SI flow. The accumulators start to inject at approximately 600 psig and there is no accumulator on Loop 4, which is also the pressurizer loop.

Other unique differences include 3 separate RHR pumps which share heat exchangers with the low-head pumps. High-head recirculation flow is not cooled. Finally, upon loss of offsite power and SI, any charging pump operating gets strjpped from its associated bus.

4. The auxiliary feedwater (AFW) system for STP is also unique. There are 3 motor driven 550-675 gpm capacity AFW pumps, one per bus. Each AFW pump delievers flow to one SG and one of these trains (Train A) is permitted to be out of service per tech specs. There is also a turbine driven AFW pump with 550-675 gpm capacity. This pump feeds SG D and the steam supply valves are controlled by DC power. Normally closed fail closed cross connects between SGs can be opened to permit flow from any pump to any SG.

5. The SG power-operated relief valves (PORVs) for STP are electrohydraulically operated'(see FSAR, Section 7.4). They are DC controlled, but AC power is required to operate the hydraulic pumps. Train A controls the pumps for the SG A and D PORVs. Trains B and C control the pumps for the SG B and C PORVs, respectively.

Based on these features, the following equipment will remain operable upon loss of a single emergency bus. Train A AFW pump is assumed out of service, as permitted by the tech specs.

G S

97340.lo/092286 39

Bus A Bus B Bus C Failure Failure Failure SGs with AFW Flow B,C,D C,D B,D SGs with Operable PORVs B,C A,C,0 A,B,0 Loops Receiving SI Flow B,C A,C A,B For these and all other single failure caes, there will be two SGs with both AFW and operable SG PORVs for cooldown. A minimum of two SI trains will also be available.

In view of these features, the small LOCA calculation for STP will consider these loop asymmetries and incorporate the single failure assumption in the following manner:

Loop 1 (or A) - No SI, no AFW and no SG PORV Loop 2 (or B) - Break location, no SI due to spill, Minimum AFW capability SG PORV operation for cooldown Loop 3 (or C) - SI flow, AFW flow, and SG PORV operation Loop 4 (or D) - no SI due to design, no AFW and no SG PORV due to Bus A failure i The above set of assumptions is a limiting one resulting in loss of 2 AFW pumps, 2 SG PORVs, and 2 SI pumps (including spill). Failure of the "A" protection system actuation icgic could cause this scenario. It is similar to the Bus A failure case above except that the turbine driven AFW pump is not considered operable.

The TREAT model noding diagrams for South Texas are shown in Figures 3-1-2 and 3-1-3. Four explicit loops are modeleled in both primary and secondary 97340:1D/091286 40

systems. Tables 3-1-1 and 3-1-2 summarize some of the relavent plant parameters modeled for the full power steady state. In the small break analysis, a 102% power steady state with 120% ANS 1971 decay heat was used.

The initial conditions used in this " Appendix K" analysis are compared with the NOTRUMP initial conditions in Section 4-1.

e e

e 9

97340.1D/091286 41

3-2 South Texas NOTRUMP Model Description i

NOTRUMP (Ref. 1-2) is a general one-dimensional network code which is used for the analysis of thermal-hydraulic transients. The name NOTRUMP is an acronym i of N0dal Transient U, M, P. (U, M, P are the important nodal parameters: total

~

internal energy, total mass, and pressure, respectively.

Thermal hydraulic effects are modeled in the code. Flow correlations model the effects of pressure drop and phase separation. Heat transfer correlations represent all regimes from liquid convection, through nucleate and transition boiling, to stable film boiling or forced convection vaporization and, i

finally, to steam forced convection. The spatial detail of a problem is modeled by elemental control volumes (nodes) appropriately interconnected by paths (links). The spatial-temporal solution is then governed by the integral forms of the conservation equations in the nodes and links. The NOTRUMP computer code.has received approval as the new Westinghouse Small Break Loss Of Coolant Accident (LOCA) analysis emergency core cooling system (ECCS)

evaluation model.

The NOTRUMP small break LOCA ECCS evaluation model (Ref. 1-1) addresses all of the NRC Appendix K model related concerns. appearing in NUREG-0611 and NUREG-0623. The NOTRUMP model has the capability of accurately calculating

~

the detailed thermal-hydraulic system response during the highly voided

) -

conditions characteristic of a small break LOCA. This includes thermal nonequilibrium in all fluid volumes, regime dependent heat transfer correlations, and counter-current flow limited flow regime dependent drift flux correlatio'ns for the accurate calculation of system mass distribution, j A 2-loop noding scheme is used to represent a standard Westinghouse non-UHI l

plant in the NOTRUMP evaluation model. The broken loop is modeled seperately and the non-faulted loops are lumped together into a single intact loop. This classical 2-loop noding scheme can not model asymmetric AFW and ECCS I delivery. Due to the Tech Specs allowance of one train AFW pump unavailable and the single failure assumption of loss of one emergency bus, only two AFW O

en40:10/os12ss 42 1

.~..-e,- ~,., ,,-------,,o-wawsw -- w- m -

m~ ~ - -w~~s ~~<-w -----=vsn-v-----------=---~w

pumps remain available. In order to model this asymmetric AFW delivery, an additional third loop is added to the standard NOTRUMP evaluation model. The final noding scheme includes a broken loop with AFW delivery, a lumped loop l

for the 2 loops without AFW delivery, and an intact loop with AFW and SI delivery.

I a

9 6

97340.1D/091286 43 l _ _ . -___ - . .

Section 4 Comparison of the TREAT and NOTRUMP Small Break LOCA Transients 4-1 Modeling Assumptions and Initial Conditions Comparison To model the small LOCA with Appendix K assumptions and to better match the NOTRUMP initial conditions, the following adjustment to TREAT was made. A new TREAT steady state snapshot was obtained from the 100% power snapshot described in Tables 3-1-1 and 3-1-2 by changing the steam demand to correspond to 102% power (PDEMAND=1.02) and changing the decay heat multiplier to 120%

(APPKMULT=1.2). The SG tube resistance fouling factor was also increased to lower the SG pressure by approximately 100 psi (consistent with the NOTRUMP l

steady state) and the reference pressure was redefined as 2280 psia (to be 4

consistent with NOTRUMP). After a new steady state was obtained, the charging and letdown flows were stopped and pressurizer heaters were turned off. The press'urizer mixture and vapor masses, enthalpies, and specific volumes were 4

also reset to better match NOTRUMP steady state conditions (approximately 5%

higher level than program level). These pressurizer changes were made in a consistent fashion so as not to alter other the steady state parameters. A i summary of the new TREAT " Appendix K" steady state conditions and a comparison to NOTRUMP is shown in Table 4-1-1.

For the transient analysis, minimum AFW flow is assumed to SG-2 and SG-3. A 60 sec delay following reactor trip and loss of offsite power is assumed before AFW delivery in TREAT and a similar delay is modeled in NOTRUMP.

Minimum SI assumptions are made in the analysis. For a cold leg break, SI to the broken loop is assumed to spill and SI to the remaining loop (s) is derated l in pressure. In the non-headered SI design of STP (see Section 3-1), credit

! is taken for flow to only one loop. The min SI flow as a function of RCS l pressure is illustrated in Figure 4-1-1. For comparison, the best estimate SI flow to one loop is also shown.

i 97340:10/091286 44 l

1

1 1

The NOTRUMP and TREAT breakflow models are both based on the modified Zaloudek correlation for subcooled flow and Moody for saturated flow (see Appendix M of Ref. 1-2). [

la,c In both the NOTRUMP and TREAT analyses, the break area was assumed to be .0123 sq-ft in.the middle of the cold leg.

Since the NOTRUMP evaluation model neglects some of the protective functions in the plant, several of the TREAT protection system models had to be frozen (or input modified) to match the plant system simulated by NOTRUMP. First, if the break was assumed to occur at t=0 and all protection systems were modeled, the reactor would have tripped on over-temperature delta-T in approximately 10 sec. The small break evaluation model ignores the over-temperature delta-T trip. The OTDT setpoint in TREAT was therefore increased to allow a better comparison with NOTRUMP.

Secondly, prior to the delayed trip on low RCS pressure (modeled at 44 see in both analyses), the TREAT rod controller, if running, would cause the rods to step into the core. Doppler and moderator feedback would also cause the power level to decrease. Consequently, the RODS, FLUX, and DECAY models in TREAT were frozen prior to trip to maintain constant power.

Third, at the time of reactor trip, a number of failures or trips were modeled in TREAT to simulate coincident loss of offsite power. These include RCP trip and loss of steam dump to condenser. The SG PORV controllers were also assumed not functional, so secondary pressure increased to the safety valve set pressure (1300 psia).

Finally a number of minor post-trip modifications were also made to make the TREAT secondary model more consistent with the NOTRUMP single-node SG secondary model. These changes included turning off the secondary metal heat (except in the SG tubes) and the recirculation model flow.

G 97340.1D/091286 45

During the early post-trip portion of the transient, only the SI and AFW systems were assumed operational for accident mitigation. Starting at 1500 sec (25 min), it is assumed that the operator starts a 100 F/hr cooldown using the two active SG PORVs (powered by busses B and C). This cooldown is in accordance with emergency procedures for small LOCA recovery and the timing

~

assumed is conservatively long in comparison to the operator response time during the ERG validation and verification (V&V) tests of a similar nature.

For example, ERG validation for an intermediate size LOCA (5000 gpm LOCA of Ref. 4-1) had the following action times (after reactor trip) for cooldown related actions:

170 see for cooldown to no-load conditions 794 see for start of 100 F/hr cooldown The time assumed in this analysis is conservatively long in comparison to the above, the LOCA is of comparable size (initial break flow roughly 3000 gpm),

and the E0P guidance would be similar to that assumed in the V&V tests.

Comparison of the NOTRUMP and TREAT predictions of the 1.5 inch cold leg break analysis for STP are presented in the following section.

'1 l

l 97340:10/091286 46

I 4-2 Comparison of the Analysis Results The TREAT and NOTRUMP small LOCA analysis comparisons are described in this section. The time table of events for the 1.5 inch cold leg break scenario for STP is given in Table 4-2-1. In the comparison plots which follow, fluid

~

node and flow link numbers correspond with the TREAT noding diagrams of Figures 3-1-2 and 3-1-3.

Figure 4-2-1 shows the TREAT RCS global pressure versus the NOTRUMP pressurizer pressure. The initial depressurization slows at 1600 psia .due to flashing in the upper head, upper plenum, and pressurizer. The depressurization then slows again above 1300 psia as the top of the SG U-tubes begin to uncover. Pressure remains nearly constant and slightly higher than the secondary pressure (1300 psia) until the cooldown starts. The RCS pressure then trends down with the SG 2 and 3 pressures for the duration of the transient. Figure 4-2-2 compares the TREAT RCS global pressure with the NOTRUMP cold leg pressure at the break location. There is approximately 100 psi difference between the two (NOTRUMP) pressures prior to RCP trip.

Following RCP trip, there is very little difference between these two pressures. In both figures the agreement between TREAT and NOTRUMP is very close.

Figure 4-2-3 compares SG1 pressures. In both codes, pressure is predicted to reach the safety valve set pressure (1300 psia) soon after trip and remain at or near this value until after the cooldown is started at 1500 sec. Both pressures fall below the safety valve setpoint soon after the cooldown begins due to the reduced heat load. The 30 psi mismatch in pressures at the end of the transient is actually quite small when converted to temperature (3*F).

The pressure in SG1 is similar to SG4, so the latter is not presented here.

The active SG2 pressures are compared in Figure 4-2-4. This pressure is similar to the pressure in SG1 up until the cooldown is started. Due to heat removal by the AFW flow, this SG does not steam as much as SG1. Consequently, TREAT shows cycling of the safety valves during the initial 1500 see of- this O

l 97340:10/091286 47

transient. NOTRUMP predicts a small, slowly varying, steam flow during this period. During the subsequent cooldown, NOTRUMP ramps the SG pressure down to simulate the cooldown whereas TREAT models the cooldown by ramping the SG PORV controller pressure setpoint. The pressure in SG3 is similar to that in SG2 and is not presented. The agreement is very good.

Upper plenum mixture levels are compared in Figure 4-2-5. TREAT agrees well with NOTRUMP until roughly 2000 sec; at this time, NOTRUMP predicts higher break flow. Near the end of the hour transient, the TREAT level starts to catch up as the break flow catches up. At the end of this transient, the level is more than five feet above the top of the active fuel region.

The upper head levels are compared in Figure 4-2-6. The trend in TREAT closely follows NOTRUMP until the top of the upper head spray nozzles are uncovered (at 34.2 ft in TREAT). The NOTRUMP spray nozzles are modeled at approximately.one foot lower elevation and this accounts for the difference in upper head levels after 600 sec.

The pressurizer mixture level comparison is shown in Figure 4-2-7. The agreement on the predicted pressurizer level response is extremely good for the entire transient. .

~

Theuphill'sideleveloftheSG1d-tubesalsocompareswellasshowninFigure 4-2-8. This is the RCS mixture level on the hot leg side of the SG U-tube.

The NOTRUMP level represents the stacked level of the tube and SG inlet plenum nodes. Note that 34 ft is the top of the tube sheet in TREAT. The mixture level continues to decrease into the SG inlet plenum and hot leg as illustrated in the plot.

The downhill side mixture level of the SG1 U-tube and outlet plenum comparison is shown in Figure 4-2-9. This is the RCS mixture level on the cold leg side of the SG U-tube. The agreement is quite reasonable for the entire transient.

e 97340.1D/091286 48

Figure 4-2-10 compares the mixture level in the pump suction node in loop 1.

TREAT predicts a slight void prior to 1500 sec. After this time, the level decreases to below the top of the cold leg, in reasonable agreement with NOTRUMP.

The level trend in the steam generator U-tubes of the active loop 2 are shown in Figures 4-2-11 and 4-2-12. These levels show similar voiding as SG1 prior to 1500 sec, then recover as steam is condensed due to the cooldown. This recovery causes the rate of level decrease in the inactive loop to speed up (see Figures 4-2-8 and 4-2-9). After the SG tubes in the inactive loop drain

(~ 2000 sec), the active loop SG levels again start tc fall. TREAT predicts the level to fall faster on the co-side than NOTRUMP and slower on the down-side. NOTRUMP predicts a faster level decrease on the downside of the SG tubes because of the higher break flow between 2000 to 3000 seconds of the transient. The overall trend is in reasonable agreement with NOTRUMP.

The RCP inlet comparison for loop 2 is shown in Figure 4-2-13. Both codes show draining to the top of the cold leg (28.4 ft).

The secondary side mixture level comparisons for SG1, SG2 and SG3 are shown in Figures 4-2-14, 4-2-15, and 4-2-16 respectively. In NOTRUMP a single node is used to represent the secondary side of a SG. The level shown for TREAT is the mixture level for the low (or zero) void fraction downcomer region. This is expected to be below the single-node SG mixture level in NOTRUMP. In the inactive SG1, the level continues to decrease until approximately 1600 sec.

Af ter this time, the pressure is below the safety valve set pressure and steam relief no longer occurs. The active SGs (2 and 3). recover due to the ,

introduction of AFW flow. Good agreements are shown on the predicted levels and trends.

Figure 4-2-17 presents the upper plenum void fraction compcrison. The agreement is very good for the entire transient.

O e

97340.lo/091286 49

I l

Core flow following RCP trip is compared in Figure 4-2-18. Both codes predict single phase followed by two phase natural circulation flow during the first 1000 sec. This flow then slows down as the SG U-tubes uncover. After the cooldown is started at 1500 sec, flow in the two active loops recovers. The resulting core flow subsequently reduces to less than 1000 lbm/sec at the end of the transient due to the eventual uncovery of the top of the'SG U-tubes in

.the active loops.

Figure 4-2-19 shows a comparison of the upper head spray nozzle flow. This is the flow from the upper head to the downcomer through the spray nozzle defined as positive downward. The initial flow into the head from the downcomer reverses after RCP trip. The sudden drop in flow rate (to 7 lbm/sec) after 600 sec indicates uncovery of the top of the spray nozzles and transition to vapor flow. Both codes predict similar flow reversal and transition to vapor flow.

Pressurizer surge line flow (positive out of the pressurizer) is compared in Figure 4-2-20. The TREAT trend agrees reasonably well with NOTRUMP for the entire of transient.

Figures 4-2-21 through 4-2-24 show the hot leg flows in the four loops. As previously explained, flows in loops 1 and 4 (inactive loops) reduce and

~

eventually stagnate after 2000 sec. The active loops recover and remain predominantly positive for the duration of the transient. Note that NOTRUMP lumps both inactive loops while TREAT models both loops explicitely. A comparison of the TREAT prediction for the two inactive loop (Figures 4-2-21 and 4-2-24) reveals only a slight difference due to the pressurizer. At 400 sec, for example, surge line flow is on the order of 50-100 lbm/sec. This explains why the loop 4 flow is smaller than loop 1 flow at this time.

Otherwise, the two loops behave almost identically. This demonstrates that the lumped loop model assumed in the NOTRUMP analysis is adequate.

The break flow comparisons for the liquid and vapor component are shown in Figures 4-2-25 through 4-2-26. Up until 500 sec, TREAT and NOTRUMP are.in O

97340.1D/091286 50

very good agreement and the integrated break flows are very close (105,100 lbm

- TREAT versus 103,800 lbm - NOTRUMP). The integrated break flow for TREAT is shown in Figure 4-2-27. Only NOTRUMP numerical valves were readily available. The trends are similar out to 2000 sec, although the NOTRUMP break flow is slightly higher due to slightly higher RCS pressure. At 2000 sec, the integrated break flow for TREAT is 4% lower than NOTRUMP (283,200 lbm - TREAT versus 295,000 lbm - NOTRUMP). Between 2300 to 2800 seconds, high subcooling predicted at the break location causes the NOTRUMP break flow to increase until saturated conditions at the break result in a reduced flow rate (less than 100 lbm/sec) for the remainder of the transient. The effect of high subcooling at the break occurs at a later time in TREAT (after 3000 sec).

Higher break flow averaging over 125 lbm/sec is then predicted in TREAT for the remainder of the transient. At 3600 sec, the integrated break flows are within 3% (441,400 lbm - TREAT versus 470,000 lbm - NOTRUMP). Therefore, total inventory depletion is in reasonable agreement. Throughout the i

transient, the vapor component of break flow remains small in both TREAT and NOTRUMP (less than 0.1 lbm/sec), and both ccdes predict similar behavior.

Safety injection flows are compared in Figure 4-2-28. The TREAT SI flow t

during the 700 - 2000 see time period is slightly larger than NOTRUMP due to slightly lower RCS pressure predicted. Otherwise, the two results.are very t

similar.

Predictions of secondary steam flow after turbine trip are shown in Figures 4-2-29 (SG1, similar to SG4) and 4-2-30 (SG2, similar to SG3). Both

analyses assumed SG isolation prior to AFW delivery. The steam flows for the inactive SGs (Figure 4-2-29) agree well. This flowrate is roughly 50 lbm/sec during the first 500 see and gradually decreases to zero by 2000 sec, after the cooldown is started with the active SGs. The reduced steam flow in an active SG with AFW average only 5 lbm/sec during the first 500 sec (Figure 4-2-30). In an averaged sense, the steam flows out the cycling safety valves in TREAT agrees with NOTRUMP. Prior to and during the cooldown, the two steam flows are in reasonable agreement.

4 4

S 97340.1D/091286 51

4 The AFW flow in SG2 (or SG3) is shown is Figure 4-2-31. This flow was

~

manually controlled in TREAT to approximately match the NOTRUMP flow, which is '

controlled based on SG 1evel. ,

Comparisons of the temperatures in the core exit and pressurizer mixture are shown in Figures 4-2-32 to 4-2-33. The TREAT predictions for these parameters agree well with NOTRUMP For the entire transient. After 400 sec, these temperatures are all at saturation.

Cold leg temperatures for Loops 1, 2, and 3 and downebmer temperatures are shown in Figures 4-2-34 through 4-2-35. The addition of cold SI water.into the loop during low flow conditions and the downcomer to loop interactions strongly affect the cold leg temperature predictions. Out to 2500 sec, TREAT and NOTRUMP generally agree well. After 2500 sec, TREAT is si'ghtly higher <

, due to more stangnant flow in NOTRUMP. After this time, due to the low and fluctuating loop flows predicted by TREAT (in the active loops), the cold leg temperatures also fluctuate accordingly. Although the cold leg in loop 3 is predicted to cooldown because of cold SI, the downcomer (Figure 4-2-37) and upstream node (Figure 4-2-38) remain considerably warmer.

Figure 4-2-39 shows the upper head metal temperature. NOTRUMP models a single lumped metal temperature which is shown to fall between the inner and outer temperatures predicted by TREAT.

Finally, the core boron concentration due to 2500 ppm SI water predicted by TREAT is presented in Figure 4-2-40. NOTRUMP does not calculate the boron concentration. The amount of boron delivered in this one hour transient is sufficient to assure adequate cold shutdown margin, even if all Xe were to decay away.

6 e

e e

'4 97340:1D/091286 52

I' 4

Section 5 TREAT Compliance with 10CFR50, Appendix K The following section summarizes the degree to which TREAT complies with the requirements of 10CFR50 Appendix K. It also provides justifications why TREAT is applicable for small break LOCA without core uncovery and long term cooling recovery analysis. A table summarizing the compliance status of TREAT is presented in Table 5-1.

A. Sources of heat during the LOCA TREAT input variable PDEMAND represents the turbine demand power. The user may set this variable to 1.02 and establish a steady state condition at 102%

power.as required.

The TREAT axial power shape is computed based upon a 1-D neutron diffusion model. The power shape can be modified by adjusting the axial absorption or

= fission cross sections until a previously defined limiting power shape is attained. However, the decay heat is not a function of the axial position in the TREAT model.

Since core uncovery transients are not involved in the proposed application and since the code is not intended to be used to compute peak clad temperatures or model fuel damage, a limiting power shape has not been determined and is not necessary for this application.

b< 1. Initial Stored Energy in the Fuel l

l .The TREAT core model computes a temperature dependent thermal conductivity for both the fuel (UO2) and clad (Zircalloy). It does not account for the effect of burnup on those properties. An empirical gap conductance model allows the user to specify the gap dimension and pressure. The change in composit-ion of the gases with burnup is not modelled. Cladding creep is not modelled.

9734Q:10/091286 53

l' j Since core uncovery transients are not involved and since the code is not intended to be used to compute peak clad temperatures or model fuel damage, the importance of burnup on the fuel and clad properties is diminished and not necessary for this application.

Fission Heat.

~

2.

The TREAT core model solves a 10 neutron diffusion equation to determine the axial neuten flux and power distribution. The neutron absorption cross section-is modified to account for changes in the boron concentration, the axial fluid density, doppler resonance broadening absorption as a function of the fuel temperature, the change in fission product poison concentration with fluy lavel and changes in control rod position. Control rod cross sections have been reduced in this application to reduce the shutdown margin and thereby increase the post-trip fission heat.

3. Decay of Actinides ,

A( Ja.c actinide decay model is used by TREAT to simulate the decay

of U-239 and Np-239.

4. Fission Product Decay The 1971 ANS standard is used in TitEAT. A user specified multiplier, APPKMULT has been used in this application to increase the computed decay heat by 20%.

4

5. Metal-Water Reaction Rate Since core uncovery transients are not involved and since the code is not intended to compute peak clad temperatures or model fuel damage, the metal water reaction has not been modeled and is not necessary for this application.

6. Reactor Internals Heat Transfer Heat transfer from the reactor vessel and piping is computed with a conduction limited heat transfer model in TREAT.

97340:1o/091286 54

7. Primary to Secondary Heat Transfer l l

This is modelled (in both directions) in TREAT.

B. Swelling and Rupture of the Cladding and Fuel Rod Thermal Parameters .

Since core uncovery transients are not involved, cladding swelling and rupture will not occur and since the code is not intended to compute peak clad temperatures or model fuel damage, the swelling and rupture of the cladding is not modeled and is not necessary for this application.

l 1

l C. Blowdown Phenomena

1. Break Characteristics and Flow

a. TREAT is not capable of modeling full double ended pipe breaks. The analyses are limited to small break LOCAs.

b. The Moody break flow correlation is used for two phase flow conditions. This is the same break flow correlation which is used in the NOTRUMP evaluation model.

A constant discharge coefficient of 1.0 is used in the calculation of the break flow rate.

c. End of blowdown does not apply for the small break LOCA application.

D. TREAT allows flexible noding of the RCS. Reliable analysis of the thermodynamic history in the regions near the break and the ECCS injection points can be made for a small break LOCA. Blowdown will not be modeled.

2. Frictional Pressure Drops -

l TREAT has a Reynolds number dependent friction factor and uses the Thom two phase multiplier on the computed friction factor. This model is similar l

to an available model in the NOTRUMP code.

I l

9734 0:1o/092386 55

l

3. Momentum Equation TREAT does not model the area change momentum flux term and the momentum change due to compressibility in the momentum equation. For the proposed small break LOCA application, the change in momentum flux due to area and compressibility are negligible. The contribution of these term's is expected to be significantly smaller than the convection, friction loss, and gravitational terms. This observation is verified by the close agreement between the TREAT and NOTRUMP predicted RCS flows (see Figures 4-2-18 through 4-2-24).

4. Critical Heat Flux TREAT computes a steady state CHF using the Rohsenow-Griffith correlation. A transient CHF correlation is not included in the code. Since core uncovery transients are not involved, the core is not expected to experience CHF.

5. Post CHF Heat Transfer Correlations Since core uncovery transients are not involved and since the code is not intended to compute peak clad temperatures or model fuel damage, post CHF transition and film boiling correlations are not included in TREAT and are not necessary for the proposed application.

6. Pump Modeling TREAT uses homologous curves to determine the pump coastdown characteristics similar to the NOTRUMP model.

7. Core Flow Distribution During Blowdown Since core uncovery transients are not involved and since the code is not intended to compute peak clad temperatures or model fuel damage, cross flow and flow blockages are not modeled with TREAT. TREAT models an average fuel channel, ie. a separate hot assembly is not modeled. ,

97340:10/091286 56 ,

l

D. Post-Blowdown Phenomena; Heat Removal by the ECCS

1. Single Failure Criterion TREAT is capable of modeling the effects of damage to the ECCS. ECCS flow is computed based upon the number of operating pumps, the number of lines capable of injecting and the pumped ECCS flow vs. head curves input by the user.

2. Containment Pressure Containment pressure is fixed at 15 psia in TREAT. This is a standard assumption for all small break LOCA ECCS performance analyses.

3. Calculation of Reflood Rate Since core uncovery transients are not involved and since the code is not intended to compute peak clad temperatures or model fuel damage, TREAT does not contain a specific model for the reflood period.

Accumulator injection is modeled with an ideal gas law expansion. Since this application (sna11 LOCA with recovery actions) will not empty the accumulators, non-condensible gases are not expected in the RCS and have not been modeled.

4. Steam Interaction with Emergency Core Cooling Water Since core uncovery transients are not involved, steam interaction with ECCS water during refill and reflood is not modeled. TREAT does not limit steam flow in unbroken loops while the accumulators are injecting during a small break LOCA. However, interfacial coefficients can be adjusted during run time to simulate a conservative steam water mixing condition.

e l

97340:1D/092386 57

5. Refill and Reflood Heat Transfer Since core uncovery transients are not involved and since the code is not intended to compute core refill and reflood conditions, TREAT has no special refill and reflood core heat transfer models. These models are not necessary for this application. -

m 8

9734(11D/091286 58

l l

l Section 6 Conclusion A detailed review of the compliance status of the TREAT code, used for the South Texas Plant long term cooling analysis, against the 10CFR50 Appendix K requirements is presented. It is found that TREAT is in compliance with the Appendix K modeling requirements in all areas that have impact on any small break LOCA transient that does not result in core uncovery. The areas of Appendix K thst TREAT does not strictly comply with deal with either thermal-hydraulics phenomena that, 1) occur only in large break LOCA transient, including phenomena such as end of blow down, post-CHF heat transfer calculations, core flow distribution during blowdown, refill and reflood heat transfer; or 2) are not important for small break LOCA that does not uncover the core during the transient. These include effects such as metal.-water reaction, swelling and rupture of the cladding, momentum flux due to area change, and momentum change due to compressibility.

The adequacy of the TREAT code predictions for small break LOCA transients that do not uncover the core, is demonstrated in a comparison of TREAT against the NRC approved small break LOCA evaluation model (NOTRUMP) using the South Texas Plant NOTRUMP evaluation input model. A 1.5 inch diameter cold leg break was considered because it provides a critical test for the major T/H

- phanomena of interest, i.e., natural circulation, two phase flow, and loop asymmetry effects. The comparison is carried out for one hour. After this time, NOTRUMP predicts that the safety injection flow will match or exceed the break flow and the plant is recovering. The TREAT prediction is expected to be reliable past this point since no new T/H phenomenon is expected as the plant slowly refills and recovers.

The transient comparison clearly shows that TREAT correctly predicts all the trends throughout the transient. All the important parameters predictions by TREAT, including; RCS pressure, core exit fluid temperature, core level, pressurizer response, loop flow, and steam generator response, are in good agreement with the NOTRUMP results. RCS inventory depletion is closely predicted (integrated break flow is within 3% at the end of one hour). Loop 97340:1D/091286 59

asymmetric effects are closely simulated and the codes show good agreement of the RCS response to cooldown caused by operator action to dump steam from two steam generator PORVs.

Based on the results from the Appendix K compliance evaluation, and the LOCA transient comparison it is concluded that TREAT has the necessary and required models, can adequately model the plant response including operator recovery actions, and is in compliance with Appendix K requirements for application to a small break LOCA transient that does not uncover the core. TREAT is, therefore, adequate and suitable for analyzing the long term cooling recovery issue for the South Texas Plant.

e i

l l

O e

97340.1D/o92286 60

I REFERENCES ,

1 1-1 Lee, N., Rupprecht, S. D., Schwartz, W. R., and Tauche, W. D.,

Westinghouse Small Break ECCS Evaluation Model Using the NOTRUMP Code, WCAP-10054-P-A (Westinghouse Proprietary, Class 2), August 1985.

1-2 Meyer, P. E., NOTRUMP, A Nodal Transient Small Break and General Network Code, WCAP-10079-P-A (Westinghouse Proprietary, Class 2), August 1985.

2-1 Mikhlin, S. G., " Variational Methods in Mathematical Physics," 0xford, London Pergamon Press, 1963.

2-2 Tong, L. S. and Weisman, J., " Thermal Analysis of Pressurized Water Reactors", Hinsdale, Ill.: American Nuclear Society, 1970 pg. 195.

2-3 .Jens, W. H. and Lottes, P. A., " Analysis of Heat Transfer, Burnout, Pressure Drop, and Density Data for High-Pressure Water," USAEC Report ANL-4627, 1951.

2-4 Rohsenow, W. and Griffith, P., " Correlation of Maximum Heat Flux Data for Boiling of Saturated Liquid," Reprint, Heat Transfer Symposium, Am.

Int. Chem. Engineers, Louisville, KY., March 1955.

5 American Nuclear Society Proposed Standard, ANS 5.1, " Decay Energy Release Rates following Shutdown of Uranium Fueled Thermal Reactors,"

October 1971, Revised October 1973.

2-6 Zaloudek, F. R., " Steam-Water Critical Flow from High-Pressure Systems t Interim Report," HW-80535, UC-38, tid-4500, 29th Ed., January 1964.

2-7 Moody, F. J., " Maximum Flow Rate of Single Component, Two-Phase Mixture," ASME, Paper No. 64-HT-35.

l D

e 1

9734Q:1D/092286 61 1

2-8 Hall, C. A., Prosching, T. A., Dougall, R. S. et. al., " Numerical Methods for Thermally Expandable Two-Phase Flow - Computational Techniques for Steam Generator Modeling," EPRI-NP-1416, 1980.

2-9 Amit, R., Hall, C. A. and Porsching, T. A., "An Application of Network

~

Theory to the Solution of Implicit Navier-Stokes Difference Equations,"

Institute for Computational Mathematics and Applications, Department of Mathematics and Statistics, University of Pittsburgh, Technical Report ICMA-79-09, 1979.

4-1 Meyer, C. E., Emergency Response Guidelines Validation Program Final Report, WCAP-10599 (Westinghouse Non-Proprietary, Class 3), June 1984.

m 9

1 l

l 97340:1D/091286 62

i TABLE 2-11-1 REACTOR PROTECTION SYSTEM ACTIONS MODELED Reactor Trip on:

High power flux Overtemperature delta T Overpower delta T Compensated low pressurizer pressure -

High pressurizer pressure High pressurizer water level Low primary coolant flow Low-low SG level Turbine trip SI actuation Manual SI Actuation on:

Low pressurizer pressure (and level)

Compensated low steamline pressure Manual Auxiliary Feedwater Actuation on:

Low-low SG level SI actuation Main feedwater trip Manual G

e 9734&1D/091286 63

TABLE 2-11-1 (Cont) i REACTOR PROTECTION SYSTEM ACTIONS MODELED Turbine Trip on:

High-high SG level Reactor trip (if applicable)

SI actuation Manual Main Feedwater Isolation on:

High-high SG level SI actuation Low T,yg and reactor trip Manual Steamline Isolation on:

Low steamline pressure Manual

~

Safety Injection (if applicable)

Letdown Isolation on:

Low pressurizer level e

9 e

97340:10/091286 64

l l

l TABLE 3-1-1 South Texas Plant Data used in TREAT Primary System Reactor Coolant System:

Number of Coolant Loops and RCPs 4 4

RCS Core Flow Rate 3.93x10}bm/sec Total RCS Volume (excluding pressurizer) 11200 ft Core Tavg - 100% power 596.6 F Tavg - no-load 567 F Thot/Tupper head - 100% power 629.2 F/622.7 F Upper Head Flow - 100% power 153 lbm/sec Licensed Core Power 3800 MWt ]

(3817 MWt-NSSS) J Pressurizer:

Volume 2100 ft 3 Total Heater Capacity 2100 kW Level - 100% Power 60%

Level - no-load 25% '

Nominal pressurizer pressure 2235 psig Pressurizer PORVs:

Number and Flow Rate at 2335 psig setpoint 2 at 58.3 lbm/sec (per valve)

. Pressurizer Safety Valves:

Nunter and Flow Rate at 2485 psig setpoint 3 at 116.7 lbm/sec (per valve)

Charging and Letdown Flows Normal Letdown Flow 100 gpm Best Estimate Normal Charging to RCS 1 pump at 2235 psig (max flow, excludes 145 gpm approx. 12 gpm seal return flow)

Reactor Protection System:

Low Compensated Pressurizer Pressure Trip 1870 psig Low Pressurizer Pressure SI Actuation 1850 psig Low-low SG NR Level Trip 33%

Other Setpoints modeled: OT delta-T, OP delta-T Low SL Pressure, etc.

97340;1o/091286 65

TABLE 3-1-2 l South Texas Plant Data used in TREAT Secondary System Steam Generators:

Number / Type 4 U-tube Model E 4 Steam Pressure - 100% power 1085 psig Volume, each SG 8012 ft 3 Narrow Range Level - 100% power 58.6 %

Steam Flow Rate, each SG 1200 lbm/sec Number of U-tubes, each SG 4864 (Unit 1) 4851 (Unit 2)

Tube I.D. 0.0553 ft Tube Plugging Assumed 5% (Unit 1 modeled)

Steam Line:

Steamline Volume to MSIV 1485 ft 3 (calculated for one SL, used for all)

Atmospheric Steam Dump Valves:

Number per SL/ Capacity at 1285 psig 1 at 250 lbm/sec Flow Rate at 85 psig 18.9 lbm/see Safety Valves:

Number / Lowest to Highest Setpoints 5 / 1285 to 1325 psig Flow Rate per valve at 1285 psig 287 lbm/sec Condenser Steam Dump and Bypass Capacity 40% of full power (all 12 valves)

. Auxiliary Feedwater System: .

Number of Motor-Driven AFW Pumps / Capacity 3 at 550-675 gpm/ pump (MD AFW feeds A,B,and C SGs)

Number of Turbine-Driven AFW Pumps / Capacity 1 at 550-675 gpm (TD AFW feeds D SG)

Note: One AFW pump is dedicated to one SG.

Cross-connects can be opened to supply two or more SGs from one AFW pump.

S 4

1 97340:10/092286 66

TABLE 4-1-1 TREAT /NOTRUMP Initial Condition Comparison for 1.5 Inch Cold Leg Break LOCA Analysis TREAT NOTRUMP RCS Pressure (psia) 2286.25 2280.54 SG Pressure (psia) . 997.43 984.48 Reactor Power (MWt) 3903 3876 Core Flow (1bm/sec) 39172.3 38673.9 Upper.Haad Flow (lbm/sec) 152.241 154.305 Total Steam Flow (1bm/sec) 4805.2 4753.7 Pressurizer Level (ft) 58.53 (65.8%) 59.61 Cold Leg Temperature (F) 563.4 559.5

. Core Exit Temperature (F) 629.6 626.4 Upper Head Temperature (F) 621.8 622.3 Break Area (sq-ft) 0.0123 0.0123 O

O l

[

l 97340:1D/091286 67 1

TABLE 4-2-1 Time Table of Events ,

STP 1.5 Inch Cold Leg Break with Cooldown ,

1 NOTRUMP TREAT Event Time (sec) Time (sec) 1.5" Cold Leg Break on Lo.op 2 0 0 102% Constant Power Ope-ation 0-44 0-46

. Reactor Trip with Loss of Offsite Power 44 46 AFW Injection at 120 F (SG 2 and 3) 94 106 SI Starts Injecting to CL 3 400 (approx) 400 (No.1 fails, No.2 spills)

Start 100 F/hr Cooldown using SG PORVs 1500 1500 on SG 2 and 3 (SG PORV controller pressure

. was ramped from 1300 psia to 560 psia over 3600 see)

End of Comparison (SI approx. matches 3600 3600 Break Flow of 100 lbm/sec) 97340:1o/092386 6B

h Table 5-1 TREAT Model Compliance Summary APP. K PARAGRAPH / SUBJECT TREAT MODEL COMPLIANCE WITH APP. K I.A. Source of Heat a) Initial core power User inputs initial In compliance power at 1.02 b) Distribution of Power 1-D neutron diffusion In compliance is used to compute the distribution Decay heat not a Not needed for this function of axial application position I.A.1 Initial Stored TREAT fuel properties Not needed for Energy are functions of the this application fuel temperature.

The burnup effects are not modeled I.A.2 Fission Heat 1-0 neutron diffusion In compliance equation accounts for changes in fluid props.,

fuel temp., boron conc.,

and rod position.

. I.A.3 Decay Heat from TREAT uses a ( Ja,c In compliance Actinides exponential decay model I.A.4 Fission Product TREAT uses an [ Ja,c In compliance Decay Heat exponential decay model based on the 1971 ANS standard.

I.A.5 Zr-H2O Reaction Not modelled Not needed for this I

application I.A.6 Heat From Reactor TREAT includes models In compliance i Vessel and Piping for conduction limited heat transfer from the thick metal walls I.A.7 RCS to SG Heat Flow Modelled both ways in In compliance TREAT ,

I.B. Clad Swelling Not modelled Not needed for this application 97340:1D/091286 69

l Table 5.1 (Continued)

APP. K PARAGRAPH / SUBJECT TREAT MODEL COMPLIANCE WITH APP. K I.C Blowdown Phenomena Not modelled Not needed for this application I.C.1 Break Characteristics I.C.1.a Break size and Variable, but TREAT is Not needed for this Location not capable of modeling application full double ended breaks I.C.1.b Break Flow Model TREAT uses the modified In compliance Zaloudek and Moody break flow correlations (same as NOTRUMP)

I.C.1.c End of Blowdown Not applicable Not needed for this application I.C.1.d Break Noding Flexible Noding allowed In compliance I.C.2 Frictional Pressure Correlations based on In compliance Drop Models the Moody Friction Factor chart are used.

I.C.3 Momentum Equation The momentum flux Not strictly in terms are neglected. compliance Momentum flux change is negligible for

. . proposed application.

I.C.4 Critical Heat Flux TREAT uses the Rohsenow- Limited by application Griffith correlation to .

determine DNB.

I.C.5 Post CHF Film Not applicable Not needed for this Coefficients application I.C.6 Pump Model Homologous curves are In compliance used to determine the pump coastdown.

I.C.7 Core Flow During Average fuel channel is Not needed for this Blowdown modeled, 1-D flow, no application flow blockages or cross flow is modeled.

O e

97340:1D/091286 70

~

Table 5.1 (Continued)

APP. K PARAGRAPH / SUBJECT TREAT MODEL COMPLIANCE WITH APP. K I.D Post-Blowdown Phenomena l

I.D.1 Single Failure TREAT can model ECCS In compliance failures or other single failure 1

I.D.2 Containment Pressure Conservatively fixed In compliance at 15 psia I.D.3 Reflood Rate No special reflood model Not needed for this is available. The SI application accumulators are modeled with an ideal gas law expansion I.D.4 Steam / Water No special model for Not needed for this Interaction refill and reflood application I.D.5 Refill /Reflood Heat No special model is Not needed for this Transfer available application l

l l .

D 97340:1D/091286 71

WO "-y- yIe* m$$ .

ag' m s r_

S '

W DR 9__ 9 ~

~

y g S R R ~

4 l O

h ]

B-A A 3 C. C' b ' -

r I 1

g g 3 C

Il - - S j 5 i~' S f 3

y

+I S C R H H H H g H R 6 H L N t L O

! Il lgg C ' ' .~ ^

R c

R ~ -

1 1

P P m M u U s S A B ~~

C s R P RP RP HM HM g

SR P O 2 URu P p

ohm Ru RB- gC P R-A a-X HX RH IN I

HX R H m gH INI E l

% R o

- R R

)  ;  ; -

A a c L- L- t- -

H w n -

i l

[

9: ?-: 8 8 R~' p_ -

' 9 9~

l

'c 9f .

. { _

" n n '

l l -

l l

A A B C c L- - E-L L t- L- L-H H c H c -

( (c ( ( ( (

u~

1 i

- Figure 3-1-2. TREAT RCS Noding Diagram if 8,C i

t I.

1 l

l . e t

t l

)

t 73

_ l l

~

Figure 3-1-3. TREAT Secondary Noding Diagram a,C 4

a 1 e e

i 4

3 i

i i

i i

\

i 74 M

S

_ _ _ - - , - - - - - - - - - - - - - ---,-,v--- '-' '

Figure 4-1-1. Comparison of Minimum and Best Estimate Safety Injection tese.

t?se.

] 1-Minimum SI Flow to One Loop floo.  %

s

\

2-Best Estimate HHSI Flow to One Loop isso.

130s.

\

1200. g 9

=v 4 9 leo .

5 1 \

3 im. \

\

000.

E *- \

m e t2 y fee. g 8.

a i s r

- sn. s

\

aos. t

\

m.

\

too. \

t tee . \

\

e .

t e Se too tSe Ice 2Se 3ee 3Se ese eSo See SSe See Safety Injection Flow (lbm/sec)

Note: These flows are consistent with recent test results, typically to within 44. ,'

75

o 8

, 5 k

// o u I I 8

w i

• o. U.

CE j o 8

< T o #

CL / o k Z m De o

a O.

CL 9

- / 8 m

a a 2

o

=

s e %

W*

i M

/

(k, F*

o 8

w

= / d 8 8-w+-g o 5 o = d. e a - o u

m m I w U I I m

. y I  !;

I o o o E

- , S t

a.

O. ,

x, o N u

o -

A 9 9 o o o o o o ed l

8 8 8 8 8 6 o

6 o

6 e

6 o

• m o o e N o o N CU N w w [ m (VISd) 3BDSS3Bd BZBd IN-2 SDB Bi-;

9 6

76

o. L

? e u

D C $

e d n Z

8 m

a.

- O b

o. J d O

v x 0 O o M U U

z .

3 O N 1 O H

>- ( D O Z

v i b

H O e 4 Z C W O. O g m l O U W O w .

& f o $ w x $ lh $ N

< m = @

w e t Js x .

E g-J/-

sO g U

z w h o 1, m Z l 8 H O. se:

I O D O sia M

J l

,> S B 0.

b O N 4 5 A O I D 4

0

} k

_m .

5 O.

o 9 "

O O 9

O 9

O 9

O O

. 9 O

9 O

O O O O O O O O O O h N w w w w w (VISd) 380SS38d NY388 IN-2 SOU Hi-T -

i l

77 l

l

O.

O O

O V

O.

A l

- OO in z @ l 8

• O m e I d u

$ $li 5

• m 5 m i

O O

=

e i O k m A u i

~

a 1 0, 5

c i O o H I O 2 l 1D E "

s I E o H I O

Z v 4 O 4 IM  ! O. 9 O U e E O W

+- $ C O e g m o 4 W k

m O h d

d "

=

g- '

O M Z O I

~

d O  ?,

sn. O w w g 1

e o 3 O N O

l 4 -__ _____._m__,___ ._ ._

_ O O. O. O. O. O. O. O. O. O. O O O O O O O O O O I

lO O in O in O lD o in

• l (T) m Cu CU e w O O 0) w -e w w w w w w IN-2 81-T (VISd) 3BnSS3Bd T OS .

78

o.

O o

o V

o.

/ o e r o u U 3 g m 8

m e -

8 m

a 8 O l

/ o l

b P-2 o

M o

cu

$ 8 8

4 W e

tu / 9 E

• s , o . .

U r m v l 5 8 -* 8

• Z gg 4 N Z N h 8

H O- 6 O.

o .

m. f o N

- E &

o. -

>- N U

o.

o o

O

'u~ - ----.~ i o

o. o. o. o. o. o. o. e. o. o. o. d O o o o o o o o o o o O o O o m o o o e o 7 e w w [ [ o w g m m m IN-2 81-T (VISd) 380SS3Bd 2 SS -

e 79

STP 1.5 INCH CL BREAX TREAT /NOTRUMP COMPARISON 31.0 1

30.0 7 u s 29.0 \

TREAT

{ 28.0 ' - ~ ~ ~- -ggy- ,yg

^

, ymy

[ 27.0 "'^~'" '

~

NOTRUMP 26.0 "i

8 x 25.0 S

y 24.0 E

E' 23.0 5

22.0 -

21.84 ft - Top of Active Fuel 21.0 0.0 500.0 1000.0 1500.0 2000.0 2500.0 3000.0 3500.0 ' 4000.0 l TIME (SECONDS)

Figure 4-2-5.- Upper Plenum Mixture Level l

s

O O

1 o.

/ O

/ o

/ e

'% m a

5  ; /

a.

/ 8 3 o

h / g .

i

• N u a 5 D k h o.

o t 0

N D

z a 5 '"

N r i, % a 2, 1 o

4 E I Ua C

- o m o

- } O w u.

m x

4  ;

< 8 ~

w =

= ,/

O. 3 r .

J d i O

e n

z 's s  ?

z

~ N O e

. I 5

/ 8 1 /

/

k.

(A '

O.

O s' O e /7

'O O o.

O, C, C, o OO f d E E E m

• 1N-2 B1-3 (13) 13A37 OV3H 83ddQ D

9 81

h O

O

O l

8 m -

=

E m

1 >

g I 9 3

• ' 8 .

l O u M D U l M M

1 O +

1 &

• O

- I m u.

k 3 o M s I u k ( R E g W < g S =

• • s )a " O $ $

m

• W5 x <

a i g s= e O W e

m le b

• 7 d s I~ 8

- 7 I I M < O 8

- ~

/ b O

g o -I S m I

, a. I r g- O m '

e f b O

J

_e -

_- - O, ,

O O O O O O O OO b b b b b v

b m

b m

b m

e o o y I IN-2 81-T (1.-1) 73A37 3Bn1XIH BZUd e

e e

82

STP 1.5 INCH CL BREAK TREAT /NoTRUMP COMPARISON 75.0 l l

^'7 TP f U-Tubes 70.0 @.

/ i A 65.0 E

4 60.0 -

i 4

p 55.0 ~

lb a 50.0 i

, I

"' Y U 45.0 NOTRUMP d \'

E 40.0 -

s Bottom of U-Tubes

- 35.0 _ . _ ~ ~ . __ TREAT Hot Leg s i.

e __%

u) _%

t g#\

v ,

30.o 25.0 1

g

--- '} Bottom of Hot Leg l l l

' 00 500.0 1000.0 1500.0 2000.0 2500.0 3000.0 3500.0 4000 o TIME (SECONOS)

Figure 4-2-8. SG No. 1 Uphill Mixture Level

1 0

0 0

0

,4 0 l e

w 0 v s q 0 5

e L

e Q 3 N b $ e O

S u / r I

T f 0 u

- t R U "1 A

0 0 i x

P f 0 M o ~ 3 M O

C  % l m h l P o L 0 i M t h U t ' 0 n R o 0 5 w T B \

2 ) o O ,

S D N O

/ N 1 T 0 O A C .

E 0 E o R

h 0 S T 0 ( N s T 2 K e A k E G

S A b E v\ M E u R 7 0 I R

B T T g

\

T .

- 0 0 9 L U P 5 -

C H

f o \,

$ M U

1 2 6 C R 4 N p l' T 0 o O e I

5 T

i^& N -

. 0 0

r u

g 0

1 K A

1 i F

s E P \

T R 0 S T 0

0 5

0 0 0 0 0 0 0 00

- 0 0 0 0 0 0 0 8 7 6 5 4 3 2 sYm $lN ; Pb d d

_. O*a .

g

)

o o

o I d

[

l 8

m -

m t

$ i

• 3 E i 8 .

2 --

m o u 5

u g i

" E a I M M

b E 5 9 E }1 8 a

$ s

$ O g '

th C

>4 )

o - o a kn = -

s j 8 e .

e -r- 'T, 8 5 e x w Q g- -

- p a.

= o J

u 8'

- e x y y- o y m

8 o .

4

" 4 l I

o.

• 6--

m o. fa.

O o

lD l

l C

m o in 9 m od b

d m

m' m @ $ g W m

4 m

IN-2 81-I (13) 73A3I 137NI i d38 O

e 85

. 8 -

I?

o.

% . o s . I._~ E 5, D

< 8 8

o o o t d

0 2: 3 s-

=

g d 9

s 7

~ ms g

o "a e o o

a c.

E o N i

a 8 -d E k g

h e e W *=

n o

z hu g f1 j 8

=

A

~ (> o

& L m

=>t o 4 )

O a / a N :1 o. $

g o

l o, U. U. U. O. O. U. O. O. O. O. U m m - o m e b LD m v b b b b U3 LD LD LD ED LD IN-2 W1-T (13) 13A37 llIHdn 2 OS O

e

! 86

s STP 1.5 INCH CL BREAK TREAT /NOTRUMP COMPARISON]

75.0 ,

Top of U-Tubes j

g 70.0  ! l-[

• [ (

65.0

\ ll k,

g s 3 60 0 If 8 lll f l

d 55'0 I N i y

1. '

O a INI J I h h 'NOTRUMP

$ g 50.0  !

g 1 .

I x

E jl 8

45.0

" ths,$h% .

S 40.0 35.0 3500.0 4000.0 0.0 500.0 1000.0 1500.0 2000.0 2500.0 3000.0 TIME (SECONOS)

Figure 4-2-12. SG No. 2 Downhill Mixture Level

l 8

o -

N  ;> d a H Y g

• W v h e o ,

m

. _ __--j:. d o

v u

a. - -- ,

Q f R 4 e EI' r a MI o e

% o M C

a 2l o H g o H

b l e s

c o d 8

• W i o w a r l 8 E E x

< g o h d

o O

i a m n U " i g-b o e

~ 3 M

d $ E m

h m

I o

I -

i o

I o

i 1 1 1

( i o, 9 9 9 0 o mo 9 m m m i o o a m m m m m m m IN-2 81-T (13) 13A31137NI 2 d38 O

e 88

STP 1.5 INCH CL BREAK TREAT /NOTRUMP COMPARISON 45.0 ,

5 40.0 Ss cc

\

l

{ 35.0 g p \

b 30.0 \

E! N 25.0 '

! e' 'N s NOTRUMP

@ 20.0 - ---- - - -- --~--'-~~

E 10.0

\ -

h TmT 5'0 0.0 500.0 1000.0 1500.0 2000.0 2500.0 3000.0 3500.0 4000.0 TIME (SECONOS)

Figure 4-2-14. SG No. 1 Secondary Side Mixture Level

i

?

I 8

m 3

z l m o O w b O 5  ! d 3

% m' g =

8 Eli 4 m e

g -

=

g i e 9 m =! = e

- I m e l G 1O g l O C

o

< E

& I 1 d 8 0

• . / 8 e n x m l

0 m I . O .E 2

. b o d 5 1

8 m

I g .

~

$ 1 9 0 m i s A E

\

s

\ a 4

\ .

E

\ 8

- , O O O O O O O Ob

$ $ $ N $ $ d 1

IN-2 B1-T (13) 13A37 3801XIH 'O'O 2 SS l

O 1

l l 90

o E

I o .

i)

\ E g E I a

$ 1 o y

$ 1 6 -

% $1 I e 8 a

8 hI I 4 a El # o 5 g =1 6

- I 8 m

e.

E s

I a1

i. {

% o @

o 8 g l 1

< & B

• 8 en C

- i I

h W d o - =

o d i

• 1 \ .

M H

\ o  ?

\ d kI o

\

a \ .

w E \ o o

(

d o

3A N L o o a o o 7 ,

o. . .

f d N N E E E 1s_a et-; tu) 7am sentnw a a e es O

e 91

l l

l o.

O o

o T

o.

m h m -

g f.>

0 a a r t

]~ g b.e 5 5 s m 3O U gz; 5'm. 4 o

O.

E $

z " ~ m, ,__ --

c

__ _ , :=-=

8 d S

g c

E _-

m - e o --

u) M Z & a b

% Z

& o. o k o & L n -

8 m

e o a

g 14

< s:,

K e laJ o >.a

~

W

_____" =- o' d =C- 8

~

A

=r. _

4 I __<

T e

.z .J o. w

. o P O* x o Ut 3

o H

W g b G. h-r - o U3 N_ d o

3

% N i o.

I o o o oo I m. N. . o i

o o o -

l l

IN-2 81-1 NOI13VBd OI0A HAN37d 83ddA 1

l 92

O.

O O

O V

O 7.7 b m N '% O g M ( in m .n.

g 5  !

O c f S m

U a

-a=

= -- . J u ~c _  !

x -

cl m

) v -- M=- --M- g

• 5 i=M EfM*T- c y e

y ^ M "Yq M --- ss e

_L5 O.

z O

E 2. O O 9

>- T i O td 4

• % l O M O

M N O 4 ,

W W y E .

s -

E "I'1== r:n .

~

y -(

a A = =7_ _ O I g Z@~=_ *

?,

z a ~; _E2:-

- M e e O .

m N-f se O O

+

M

s w \p O

" N L , 4 c.s F-(t)

'  % g 6 O.

\ O (s O

)

s

) D JJ O

O.

O O.

O O.

O O.

O O. O. O.d O O O O O O O O O O O O O O O O T M N w .-9 l I

l IN-2 B1-1 (33S/H87) M0ld 137NI 3803 l

93 1

8 O

V i O O

I O

(T)

- O O

Y >

u 1

1. 3 b

O c~

d O

C

- .e E

g =

N 4 a La U. k

$$S

~

W 0

I $

4

= -

J d

O i

U p 4 O T

• s i W

g a O

=

E z d R e

1 8

e A

1 a

TiJ" O, o

c' t ,/ O U

1. ' ~.

O. O. O. O. O. O. O O O O O O O w

i w

i ki IN-2 81-T (33S/H97) M073 OY3H 83ddn k

O O

O O.

O i

' D 3 z M S

~

8

• g O. O

< ~ O C 1 O M 1 O A O M tn 3

o. e E J O in

& O

> D k k

s

• 3  !

N < O.

9 O

W U $

E ' O W e

• o g x

O E 9

< y N 4 a

w g

=

pm- O k

~

a .

d O

5 n

^? w I

u O g .

ig3 L- _ _-- - - - -

O y a

o. 5 a:

3 P O lt) 4

$ 2.

/e_ _ -_ __

O O.

O O.

O O.

O O.

O O.

O O. O. O. O.d O O O O O O O O O O O O O

G3 ED v N N T (D l 1 l IN-2 81-T (33S/H91 M073 3NI7 3980S l

95

O O

o a

5

" o s 3 O

M o M D

U O

I , d O

o O

G in lr.

M k E a

a s Z .

H o. o o 4

[ I k* L

$O $ o,

1. ~

C\1 O

.k l LM

]

y --

, o 5 J =>

= d .

~ .

-_- O -

a U M [p%1 h 7 I E=F2C o ,7 M -=-

m ,

x-- &

O u

" to r

o.

h A

e e-i

- ~

o r

CD o N 2

s

.>. ~ $

O

( ,

o

/

N O.

o o. o. o. o. o. o. o. o. o.

o o o O o o O O O o O O O O O O O O O N O CD LO T C\1 N T LO e e I i i l

IN-2 B1-3 (33S/H91) M073 937 10H T dOO7 .

O 6

96

O.

o

' o j

O v

l

( -

o.

s x 8 O

z s

• M O

CD - ,

g O 3 Q.

] ,,

L o o

O e-f k

Z ',

O O 4, U

O O

Z

~4_W.

• y w

', - w.

@ O O f=ME-MWh =

m

• E = g C= 0 $ a a O

W -

sp- o.

O Z

O O' E -

- o U 2 r q% , 1 o W

% > o

• Cn Q.

g > N O W

st O _ W b a

~

8 O

+ =___ g - .

o U 2 m---

O c

n n

, ZKM -

n w 3

= Z -- .d ~r ~

(J ==== eg .D an=. .

Z ,,.b ,

O w , ,- - O e S o u 4 g > ._ ,

O D

tp k N W g , #D ,

b (n \  % O 2 \ A l'3 b 8 O

O O Ob O o O O d 6 8 % 8 8 8 8

=

8 8 O g g O v m IN-2 Efi-; (33gfggl) MOl.d 931 10H 2 dOOl .

O 97

O.

O i

O l - O Y

O

1. % b O

s

/ O b

gn -

l

%] 3 O

p-4 A O cr i 'rf O

.k.#

l h O O

U

/ D b M ~Slm 48 CL g

s E

X

- -m _ z I . 4~l l

~ - w ---

'2 O.

O g

n O

O g -_ _

W "z..r_>_____  ! cu -

g O

N- C H

< Y_ 9 o.

Z O

E N J O @ o x

'-y~- i 8

cu a3 4 35.=. _, . Id

$ - - -x

^ _=_ _= A O h n*

m g ' 1 '- b ~ n a .

O o U I= @is*,, "-

z- -

T _I_'

% _ =__ - - -

,e 1

O "l N _~,, %.,

I g

Z -

O b

w

', -l w

$ w A M

& g N l

• x I F\/. N i o

O I

s. B l m O.

O. O. O. O. O. O. O. O. O. O t O O O O O O O O O l O O O O O O O O i T cu o w

G3 (D v CV Cu e ge l IN-2 81-T (33S/H87) M073 937 10H E dOOl -

h 98

l l

l C.

O o

o v

9 O

o D

Z M O

tt)

M O 3 E

• O 4 , o e4 Q. O rn, o

f M m G- b Z <

O. p U '

O O h e E e

N a a*

c O Z O o

W ~ d e y '

-(

e.,

8 N

e *&

y 3 E t o

'___T (I) ,- _

J T =1'A o  %

1 O

z

&_ _W^

g__, _ ._

o 4

, s- - %

d :g

. a. w o h

g >#

a E U) O N ,,,W O

• iT o

/ o

, N N D e s l

B o

O.

O O.

O O. O. O. O. O. O. O.b o

O O o o O O O o o o o o o o N

o CD LO v N N v

" l l IN-2 B1-1 (33S/H97) M0ld 037 10H P dOO7 r

l

I l l l

l 99 l l

8 A 5 o

- 'g S a:

S m

i l

E (n

' y 1 g 3 Oz O.

o l

< I ko s m .S o  : g_<

m A

-w \ 8 u E *r &

93 f

l k<

I

o. o a

$ W $ )

W p o 8 M W m h

o hN

= d d - A a e 8 4 d g - ..

I g '

- 'e 8 R $

M z )1 o -

i "

R g 8

a. )

/ o G b Y

8 9

• o R * **

• . R R . .

8 8 8 8 8 8 8 8, m m m m - -

IN-2 81-1 (33S/H81) M0ld >lV388 OInld .

e 100 i

C.

O 4 O O

E '

= -

B O C.

2N ~

%^-_.

_-=== 0

.s

. _ = =

• m (T) i O

w -

_ . - - = j f s

~"___

g j = =sg ._

_ _ _ _ g j

} '

% x 0

N

- e O

k O M j O

>- ru - k W O k " O a s

t-O

- h O

4

< O w w

g O cf3 O -

• t- Cu e W N M Z l 4 C M N W

• p- I G O V ID O x o e

$, o.

g

= t sO b -

(/)

C.

O O

O i

h O O O o o. o.

O 6 6 6 d O O o IN-2 81-1 (33S/H87) M01.-l >lV389 80dVA

[  :

l .

l l 101

0 0

- 0

- - 0 4

0 r

=

0 0

5 3

N w O o S l 0 F I

R f 0 k

A 0 P 0 a M -

3 e O r C B 0

P d M - - 0 e U f 0 t R 5 a T

O

-- - 2 )

S r

g N D e

/

T - 0 N O

t n

A E

R 7 - 0 0

C E

S I

0 T T / 2

(

A E

K A

j -

E M R E 0 I T R T B 0 0

L C

- 5 1

7 2

H C -

- 0 2

N - 4 I -

0 e

0

. 5 - r 1

1. 0 1 u g

i P

T S

/ 0 0

0 5

F

- 0 5 5 5 5 5 5 5 5 5 0 00 0 0 0 0 0 0 0 0 0

+ + + + + + + + + 0 O E E E E E E E E E 0 0 5 0 5 0 5 0 5 0 0 0

5 4 4 3 3 2 2 1 1 5 Ym E gE"OhOE:$$

l <! '  :

e o

I .

8 a

h - N o

5 d B a -

x o N 8 m

g i a N c 1

o O E

5 k 6 t g N> 8 m

=

s a a

e s

4 o $ >

w g a o a

• %. o w e s o E  %

4 h w "

W a i o 5 d

h g -

s 5i S 8 7 b <5

• 7

• 5 llb>$ 1o

• e

[.y 2 o k

" \ o i E E s

= i

, o.

. N o l NNN o o

o o o o o e o N O o a v N IN-2 81-1 (33S/H87) E dOO7 M073 IS O

O O

103

l

{

o

. o I i o, I .

I o.

o =

o O l N C 8

m o a

-

• g cc

< _ g o a

a ,

o m

6 o B' E o 5O h

o un cu -

u o mm z -

y o c ~.

z

s. a g

s o 1.0 O H o Sx x C g < ~ e g a e s W=

W m M y 4 o. - g -

a g' g -

- o S 2

" l 2 -

~

m x i r  ?

i a  % *

- M

~ S =s f o.

• s < o m o 54

~ 5 n L- t

$ a e -

\, p- l g I )

__ ___.___ _.__b_e

' I l

o.

o o o o o oo N N N 5

$ $ $ $ N 1N-2 81-T (33s/Hgl) M073 HY31S T 95 e

104

STP 1.5 INCH CL BREAK TREAT /NOTRUMP COMPARISON 120.0 t

= l A 100.0 g 5 4 4

di 4 80.0 ,'l - -

U w

ll SG PORV FI4W (TREAT) h 60.0 gl SAFETY VAINE FIDW (TREAT)

I 5

8 i g d 40 0 - ' k --- '

I ll g STEAM FLOW f

we

't glI II /I ,A I s, ,7 ,

M l ' '

'"(#

m J ' l' , , [NOTRUMP)

• 20'0 '

e lj ' } f f sfl/

15 ) I ,

f W ll1 ~1'- I b a 0.0 ti IIl A l 0.0 500.0 1000.0 1500.0 2000.0 2500.0 3000.0 3500.0 4000.0 TIME (SECONDS)

Figure 4-2-30. SG No. 2 Secondary Steam Flow

o o

9 3 O

o -

l o h O

I " u 8

/ o S

/ -

a l o V 5

e I o o 0 x f m e o "

u I t

9 D

& t o 4 6

S o M L W o

Z f " m o

1a h< ,I I o.

Z o

o 84 n

o k& '

g N

o o

Lu

$ O'

) N g

N. k e g \

s o

• H M C \ o.

o .

J < O g p  ?

I l n, t s k 7

~ b /

• O g- 'I. 8 g

c' 5 a.

4% = a e

4- g o e 0;

p- = o o

I- LD I

I l i t i ,

9 o o oo 9 9 9 o' o' e

o 0

m E  ? 8 g 2 OS OJ. N073 MdV 1

e 106

3 i .

l I l l i

t I

i l  ?

j I  !,  ! l l 1 ,

s o I i l l i' @

l i i t/ m i f

- I e - -

i I  ! o

< , I I }, l l/ ,

fi o e

/

j

! O e Y,/ l 0 Q

- / 8 e h.

s

< N G

/ o a E o. 8 Y W

1. g S "

x o

E e s o 0 Y U e

O

,/ o. -

i

/  ! $ "

+

d o. 'l s 8

- i I g )i ,

N N {

- l i o.

m X l--le 6 l o e o u 4 , I l

o.

u t o

/ I 5

i i e o I / O h

bs

_ <- i o,

. . . . . O. . . . .

8 e

8

. .S 8 8 8 R 8 8  % 8  ;

m o o o o o o ,

IN-2 Ell-T (.d) dH31 IIX3 3tf03

~

e 107 l

o

. I i o g .

I

! I '

l  !

- '  ! l o l l i l i  ! i i f d o

i l

l. l  !  !  ! .'/ 8 2 E I . i

!  ! o i

$2  : I j 5 i l i i i  !  !

I d o

g 2,

Iu

!  ! l 8 5 I 1. e l o

& , l t .

e g i I  !

I d e

- g a g

s

> m a4 gx l o u 4 o.

E l # g S :

- o a x

! ls '

- N o +$

a l

W l/ o E r Iw E d "

d

a. l 8

1 l

5t a ,mg a "

A W g1 o 7

~

zI ,e d n

• i i 8 4

• l  ! ]j #

e 1  : u

$ I l o 5 i  !

z i i  !

i

! d o

g j .

i

/ l o

/ I.

1 l (! I Mi  !  ! I o.

. . . . c! . . .

8 8 S 8 8 8 R 8 o

8 o

?

o 8

o o o o o o o o IN-2 81-T U) dH31 3801XIH BZBd .

O 108

M STP 1.5 INCH CL BREAT TREAT /NOTRUMP COMPARISON 590.0 580.0 } __ , _2 _;zg; H l z 570.0 -- - - - - - - - - - - - - - --

m TREAT y* 560.0 1 -

-3 i

E 550.0 - -

g N

o. wy" I

uJ 540.0 --- - - - -\- - --

" NOTRUMP \

E" o - g SJ 530.0 - - - - - - -

\--\ -

\

o \

l d 520.0 \-

)

o (

l

\

510.0 - - - - -- - -- -

---g --

i I i

500.0 0.0 500.0 1000.0 1500.0 2000.0 2500.0 3000.0 3500.0 4000.0 i

TIME (SECONDS)

Figure 4-2-34. Cold Leg No. 1 Temperature

1 m 8 o

5 x

b O.

x >  ; o

(

/ c-

,e 8 m

/C. f e 5 ( -

a o y 0

2

[ 8 o

Y "

5 2.5n=. ~-  ?

U gm -- o g a

6' s d o

n e <

_ v -- m .

H .M - N g O B e a x

@ l y / o b &

< p / d M 3 y is a w

-  ? a- 8 -

3o x / w y / o. 3 o a -

w m F x 8 g a S n l

U z

M hg 8

• S S &

c.

+ (\,\ o S.

m  : &

1 o I m I

I

-- - ,-. o.

o o o o o. o. o. .

d d d d 8 o 8 8 S g a S 8 m o m o o IN-E U1-1 (d) *dH31 2 937 0703 .

9 O

110

a a o.

o o

o 64 .T aC N

M o A C.

o D- in z m e o

W = M. y - u g

a f -a , s .

/ ~w- _-= e it m

8 5 a .

g a

• a a f v. f -_ 2 d a h

o

'"~

- :t _,, -C=

@ o m

z -- -

m z

% [v - o

• Q W

e a 2 o o

- g3 a

m M I $

N m C

~

H x )

4 4 W h E e o E

= =g _- -

g -

J & =cs-a o h ","""1 m = -

5 z

gi W _ $

N

?

w H o g

M o.

8 o e O o 4

• e l

a- E

~

}\ o o

( D o

o.

o o.

o o.

o o.

o o.

o o.d .

o o o

gn W

In in o l

O O V V m 1

IN-2 til-T (d) *dH31 E 9310703 111

o

) e I I I o i

I t T I ' l l,

I i  ! o

, i  ! -

/ l 1 L- d

- 8

, i i, /

j  ! l m z ' i I

/

o I /

u) i / '

e M -

l \ o.

9 / o 1

l I I

I p'

/ '

o l 1 o b  !

d /[ o &

0 T ~

\@rw f De a

o tn i

e o / @ G e R ,i =

o g =

q / 8 o 8 W

E

, = o o ec m 3

$ ~

b W

W i

i O

W m U

M l  ! -

b o

a j m .

u - n

. z 7 W

M i o. 7 i

) o y a

o o e

- i - u l

D Q. l tp G l * . E i,l, 1 i i i l o i  ! o I

o

,1 I

o.

o. o. o. o. o. o. o. o. o. o o o o o o o o o o o CD LD v OJ o 03 LD T m W O W O O T T T IN-2 81-T (d) dH31 83H00NM00 SOU -

O 112

1 0

0 0

_ 0

- 2 s e

r 0 u t

\ 0 a

' 0 r h

. l

- 5 e 3 p T

m N

O N i1 h1 1

1 y~

~

' 0 T e

S b)sl 0 t I

R A

N h'

^

0 0

e l

P M gm ~ 3 t u

O ~ - O C t \ 0

\

e Y d P l q 0 0 n M n. ,

- 5 a U -

2 )

R \' S t T P \ t - D e O C s e N l N R l 0 O n

/ - t C I u 0 T

,' E A o 0 S P E I 0 (

C R s 2 T P -

R E

K A

p ^~fi I C

R lI il F1 1\

\U, 0

M I

T 3

P E 0

.nI li o

{

R _ -

0 B _

j 5 o

- I

.]

l 1 L L E , gtg gg

. C 31" 1 g g - T 0 A H 1I ] E C i 1

0 R N _- 0 T I .

0 1

5 - .

1 8

- 3 P -

0 -

T - 2

~

0 S -

_ 0 -

5 4 e

r u

.- 0 i g

7 - -

F 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 8 6 0 6 4 2 4 4 4 0 8 5 5 5 6 5 5 b?m cha1 o$" m d rgf .

="

ll

}

l O

, y e o E

/ r / o' b

/ j '

O m g / / $

" ' # o s N j

E / '

/

h O

$ / i 8 E, u / j =u u e O o G / ~9

} b D

T

$O

/

l l l O e?

9G W I D gg

/

• G os E i - e B< / a

• / O. o k y i/: -

I g &

m N2 80

+ , o 4 / i w u' N l' m

a e

/ ,

I

/ O.

o s-

- c$ E.

-=

im u u eu I /

f Igm .

E di W

/

/ '8

' 2 8

9 o

se

. / I o *

~ i t e  ?

E > /

O 4

m / /

/

• I

/

.o / / o m .

I i u I / &

o Ar./

1-o.

t

. 9 9 9 9 . . 9 .

8 m

8 o

2 m

8 a

8 w

8 m

R o

8 o

8 o

o 8

o IN-E U1-2*i (3) 'ld 'dH31 7V13H HQ O

e 114

m o.

O O

O V

o

\ $

\ 8 M

Z o

V)

E . c 4 o

• O 1 3 O

z O M o (v) 4 O S4 9

1 C Z i o. g D si O O E O C H in O N U Z

N O e k o Z o O

& b M "

E

- 8 x

N m e .

< gg be tu z O E q o w U

% b -

E4 U

J O <

10 N

5 e 5 9 .

9 8  ?

- e i c.

Q. I H v M

o. .

o L4 O D in m 1 -4 Ese o

o. o. o. o. e. o. o. b O o O O O O O O O O O O O O e w w (Hdd) NOI1VH1N3DNO3 NOBO8 3803 IV381 O

e 11C

l Appendix A NOMENCLATURE NOTATIONS _

a nt Incidence matrix defined as follows:

-1 if link i enters node n a

nt

= 0 if link t does not connect to node n

+1 if link i exits node n (Agg)n Heat transfer area between vapor region and wall in node n

-- Interface surface area between mixture and vapor region in (AMG)n node n C --

Link contact coefficient; this is the fraction of the vapor nt flow in a link which comes from the mixture region of the donar node

. D --

Link contact coefficient; this is a fraction of the liquid nt flow in a link which comes from the vapor region of the donar node E

Fluid elevation

-- Elevation of the top of node n (Etop)n (Ebot)n

-- Elevation of the bottom of node n (Eu op)t -- Elevation of the top of upstream end of flow link t d -- Elevation of the top of downstieam end of (E top)t ,

flow link t 9734ce1D/092286 116

JKr NOMENCLATURES (Cont)

(E" bot)t -- Elevation of the bottom of upstream end of flow link t d

(E bot)t -- Elevation of the bottom of downstream end of flow link t (Egg)n -- Elevation of the mixture region in node n h --

Specific enthalpies h

f

-- Global liquid saturation enthalpy h

g

-- Global vapor saturation enthalpy (hM)n -- Enthalpy of mixture region of node n (hG)n -- Enthalpy of vapor region of node n (hgM)n -- Enthalpy of vapor component in mixture region of node n (h,g)n -- Enthalpy of liquid component in vapor region of node n

~

h,ff -- Effective heat transfer coefficient hg ,p Gap heat transfer coefficient for fuel rod h f$), -- Convective film heat transfer coefficient l

H -- Total enthalpy of a node I =

(HM ) (hM)n * (MM)n

=

(HG )n (hG )n - (MG)n Kg ,p -- Gap thermal conductivity L

N Total number of noncritical flow links S

9734&1D/092286 117

l NOMENCLATURES (Cont)

M -- Mass of fluid in a node

-- Mixture region fluid mass (MM )n

-- Vapor region fluid mass -

(Mg )n Ng Total number of interior fluid nodes Nu Nusselt number P* --

Global pressure P

Gap pressure (psia) gap P

L cal pressure for node N N

q t

V lumetric flow rate in link t q --

Heat flux Q

Heat transfer and/or heat generation rate G

Qn Wall to vapor region heat transfer Q

Wall to mixture region heat transfer C

Qn Heat transfer to condensing droplets C

Qn Heat transfer across interface due to condensation Qn

-- Heat transfer across interface due to evaporation t -- Time D

e 97340:1D/092386 118

NOMENCLATURES (Cont)

T --

Temperatures sat -- Fluid saturation temperature T .

T, -- Wall surface temperature T

fric Frictional torque for the pump v --

Specific volumes v

t

-- Link average specific volume (v,)n -- Mixture region specific volume (vG )n -- Vapor region specific volume V --

Total volumes V

n Total node volume (VG )n Vapor volume of node n (VM)n Mixture volume of node n

- V gj Flow-regime-dependent drift velocity &

e O

9 97340:1D/092286 119

l 0 --

Mass flow rates (Wf )t Liquid mass flow rate in link t (Wg ), -- Vapor mass flow rate in link t, Droplet condensation rate g CDn CW W

n Wall condensation rate CI W

n

-- Interface condensation rate EI --

Interface evaporation rate W

n BR W

n

-- Bubble rise rate DF W -- Droplet fall rate n

a --

Void fractions (aM)n Mixture region void fraction of r. ode n (ag)n -- Vapor region void fraction of node n a

g

-- Average void fraction in link t p -- Fluid densities pf Saturation liquid density p -- Saturation vapor density g

p -- Average density in link t t -- Time constants ar c --

Thickness of the fuel rod clad O

O 9734Q:1D/092286 120

_.