ML031570501

WCAP-14936, Code Qualification Document for Best Estimate Small Break LOCA Analysis, Revision 0, August 2001 and WCAP-14936-NP, Revision 0, April 2003, Sections 9 Through 23

ML031570501
Person / Time
Site:	Indian Point
Issue date:	05/07/2003
From:	Kemper R Westinghouse
To:	Office of Nuclear Reactor Regulation
References
FOIA/PA-2005-0108, LTR-LIS-03-210 WCAP-14936-NP, Vol 1, Rev 0
Download: ML031570501 (790)

Text

{{#Wiki_filter:SECTION 9 WCOBRAITRAC ONE-DIMENSIONAL COMPONENT MODELS 9-1 Introduction The one-dimensional components in WCOBRAtTRAC are modules derived from TRAC-PD2 to model the reactor primary system. These components provide models for accumulators, pressurizers, pipes, tees, pumps, steam generators, and valves. In addition, there are two modules that provide boundary conditions for parts of the system not modelled, consisting of either a pressure sink/source or a flow boundary. The conservation equations used for the one-dimensional components are discussed in Section 2-4. The following sections will describe the features of each of the one-dimensional components and elaborate on their unique characteristics. Many of the modules are virtually unchanged from their original TRAC-PD2 versions, so many of the descriptions are the same as those given by Liles et al. (1981). 9-2 PIPE Component Model Basis The PIPE component is used to model one-dimensional thermal-hydraulic flow in a duct or pipe. A PIPE can be used alone in a problem or can connect other components together to model a system. Area changes, wall heat sources and heat transfer across the inner and outer wall surfaces can be modelled in the PIPE component. Figure 9-1 shows a typical noding diagram for a PIPE containing a venturi and an abrupt area change. The numbers within the PIPE indicate cell numbers, and those above it are cell boundary numbers. The geometry is specified by providing a volume and length for each cell and a flow area and hydraulic diameter at each cell boundary. The junction variables JUN 1 and JUN2 provide reference numbers for connecting this PIPE to other components. Wall friction losses and form losses associated with bends, orifices, etc. are set where required at the appropriate node boundaries. Five options are available to determine the wall friction losses based on a variety of flow configurations and correlations. These options are described in Section 4-7. 4384-non\sec9.wpd-04203 9-1

Wall heat transfer from the inner and outer surfaces of the PIPE may be calculated as well as heat generation within the wall. The calculation of critical heat flux may be determined by the Biasi et al. (1967) correlation. Section 6-3 describes the selection of heat transfer coefficients in the one-dimensional components. The wall material properties are selected from stainless steel (304, 316, and 347), carbon steel A508, or Inconel 600. The PIPE component includes an option that allows the user to simulate the effect of a non-condensible gas on the condensation rate. This option is used to simulate the suppression of the condensation rates in the PIPE caused by nitrogen injection from the accumulator or from ingestion of air from the containment. Application of the condensation suppression factor to the interfacial heat transfer coefficients is described in Section 5-3-5. The numerical solution method used for the PIPE component is specified by the user. The semi-implicit method is adopted due to its increased computational efficiency. In components which can expect high flow velocities, the fully implicit solution method is used to avoid the restriction set by the low Courant limit. The junctions of the one-dimensional components are always solved semi-implicitly. Model as Coded No special models or correlations are applied in a PIPE component. The conservation equations are solved as described in Section 2, with the closure relations discussed in Sections 3 through 8, referring to one-dimensional components. The thermodynamic and material properties are described in Section 10. During the execution of a problem, the solution procedure is controlled by subroutines PIPEl, PIPE2, and PIPE3. At the beginning of each time step, PIPE 1 calls subroutine SLIP to obtain relative velocities, and subroutine FWALL for wall friction and irrecoverable loss coefficients to determine the interfacial drag coefficients and calculate the relative phase velocities. Subroutine HTPIPE is then called to determine the wall heat transfer coefficients. During the timestep iteration, PIPE2 calls DPlD, which is the controlling routine for the hydrodynamics solution. DF1D calls DFIDS or DFIDI depending on whether the semi-implicit or implicit solution scheme has been chosen. In these routines the interfacial mass and heat transfer, condensation suppression, and in the case of DFIDS, water packing logic are applied or calculated. The controlling routine PFICHK is called if the critical flow model has been selected. After a timestep is successfully completed, PIPE3 calls CYLHT and FPROP to determine the wall temperatures and calculate the new fluid properties, respectively. The boundary arrays are again 4394-non\sec9.wpd-04203 9-2

updated for the converged solution. If the time step fails to converge, the calculation is backed up to the previous time step values, and a new time step, half the size of the old one, is tried. 9-3 TEE Component Model Basis The TEE component models the thermal-hydraulics of three piping branches, two of which lie along a common line with the third entering at some angle ,Bfrom the main axis of the other two. The code basically treats a TEE component as two PIPEs, as indicated in Figure 9-2. The angle P is from the low-numbered end of PIPE 1 to PIPE 2. The low-numbered end of PIPE 2 always connects to PIPE 1. The straight PIPE segment is numbered from cell 1 to NCELL1, with the connection to PIPE 2 at cell JCELL. The branch PIPE segment is numbered from the cell immediately adjacent to JCELL, beginning with cell 1 and ending with cell NCELL2. The connection to PIPE 1 from PIPE 2 is treated with mass, momentum, and energy source terms. For PIPE 2 the conditions in cell JCELL of PIPE 1 form the inlet boundary conditions. The mass and energy terms associated with the side branch flow are added to the governing mass and energy equations representing the main branch flow. The losses at the junction are modelled in terms of the momentum change resulting from the combining or dividing flow. For the combining case an additional momentum source term is added to the main branch momentum equations. This term represents the momentum source or sink associated with the secondary flow in relation to the main branch flow. The time differencing and iteration procedures guarantee conservation of scalar quantities within a convergence tolerance. The levels of implicitness for the finite-difference equations applied to PIPE 1 and PIPE 2 can be specified independently using the input variables IHYD1 and IHYD2. Since the junction between PIPE 1 and PIPE 2 is always treated semi-implicitly, the velocity at that point is always included in the computation of the time step stability limit. Phase separation at the junction is calculated if the flag ISEP is set to one. Phase separation is computed if the void fraction in the junction cell JCELL exceeds the user-specified value ALSEP. Model as Coded Since the TEE is modelled as two connected PIPEs, the PIPE model description in Section 9-2 should be consulted for additional information. The calculational sequence for a TEE includes separate calculations of the primary and secondary sides. For the junction momentum source, an additional source term is calculated in subroutine ETEE and is incorporated in the momentum equation in DFlDS or DF1DI depending on the solution option chosen. This source term is set to zero when the TEE is a dividing tee. 4384-nonsec9.wpd-04203 9-3

9-4 PUMP Component LI Model Basis The pump model employed in WCOBRAITRAC describes the interaction of the system fluid with a centrifugal pump. The model calculates the pressure differential across the pump and its angular velocity as a function of the fluid flowrate and the fluid properties. The model is designed to treat any centrifugal pump and can include two-phase effects. The pump model is represented by a one-dimensional component with N cells, where N must be greater than 1. A typical noding diagram for the pump component is shown in Figure 9-3. The pump momentum is modelled as a source Q that is included between cells 1 and 2. The source is positive for normal operation with the pressure rise occurring from cell 1 to cell 2, so it is necessary to number the cells so that the cell number increases in the normal flow direction. The pump model is identical to the one-dimensional pipe model except that a momentum source is included in the mixture momentum equation written between cells 1 and 2: _U15 _ SUn5

                    -    .5  _(  (p1 _ 2)
                                      -P)     C" _+g - jUnI.(9-1)   U     K At             p/                             Dt~Cn_g                       (9-1)X where U is the mixture velocity, P is the pressure, C represents the convective terms, g is the gravity term, f is the friction factor, p is the fluid density, Ax is the cell length, Dh is the hydraulic diameter, the subscript 1.5 refers to the average value between cell 1 and cell 2, and the superscript n indicates that the parameter was evaluated at the previous timestep. Parameters without a superscript are the updated, new time values. The source term Q is taken to be:

2 pum n

                               +AC        g +    U1 .5 l U1 . 5 I
                                               ~~~Dh                                            (9-2) 43 84-non\sec9.wpd-04203                               9-4

where APpUmp is the pressure rise across the pump evaluated from the pump characteristic curves. With this definition of the momentum source, the steady-state solution of Equation 9-1 is P2 - = APpUmp. The model for APpUmp is described next. The Pump Characteristic Curves - The Homologous Curves It has been well known that for single-phase flow the characteristics of a pump can be quite accurately obtained from those of a geometrically similar scale-model using the similarity laws. Following these laws, the head and the torque of the pump can be represented in nondimensional forms which are independent of the scale of the pump model. The approach used to establish the so-called homologous curves is one of the methods that has utilized the similarity laws to nondimensionalize the variables involved in pump operations. In this approach, four homologous curve segments (one curve segment represents a family of curves) are established. These curves describe in a compact manner all the operating states of the pump. The following definitions are employed in the subsequent development: H = pump head = APP,,p p = fluid density at pump inlet Q = volumetric flow rate through pump co = pump impeller angular speed T = pump hydraulic torque To allow one set of curves to be used for a variety of pumps, the following normalized quantities are used:

               = QIQR AN =      / OR h = H/HR P = (TI TR)I(PR/P) 4384-non\sec9.wpd-04203                             9_5

where the subscript R denotes the rated conditions. Use of the pump similarity relations (Olson, 1974a) shows that h2) f (9-3) and

                   =

(9-4) a1N GIuN for I IUI I 1, and h

               )2 1L)    aV (9-5) and 1)2       ( V )

(9-6) for I N I 1-1* 4384-non\sec9.wpd-04203 9-6

Table 9-1 shows the resulting four segments of the homologous head and torque curves that represent the complete pump operational characteristics. Pump Single-Phase Head and Torque Homologous Curves Figures 9-4 and 9-6 show typical single-phase homologous head and torque characteristic curves for Westinghouse designed pumps. Pump Fully-Degraded Head and Torque Homologous Curves A basic assumption of the WCOBRA/IRAC pump model is that the same type of scaling laws, which are applied under single-phase conditions, can also be applied under two-phase conditions. It is assumed that there exists a condition at an intermediate range of void fractions in which the pump head and torque can be described by a set of homologous curves, similar to the single-phase curves. A typical set of curves is illustrated in Figures 9-5 and 9-7. The Head and Torque Multipliers To provide for a transition from single- to two-phase conditions, the following correlations are used: H = H - M(a) (H - H2) (9-7)(1) and T = T - N(a) (T1 - T 2) where M = head multiplier N = torque multiplier a = donor-cell vapor void fraction at pump inlet 4384-non\sec9.wpd-04203 9-7

and the subscript 1 denotes the single-phase value, the subscript 2 denotes the two-phase value, both calculated from the homologous curves, and the subscript

• denotes the derived value for a given two-phase condition.

Pump Impeller Speed The angular speed of the pump impeller is calculated from the equation I - d == T - (T + TFR + TE) (9-9) dt where I = moment of inertia of the pump rotor assembly TM = torque supplied by motor (after trip, T = QC) TFR = total friction torque (including all mechanical, bearing friction and windage loss) T = electric torque (caused by induced voltage after trip) The total friction torque is (Bordelon et al., 1974) [

                                                                                               ]a  (9-10) where

[

                       ]a,c 4384-non\sec9.wpd-04203                                  9-8

for the 93A pump, and is assumed to apply to other pumps of similar design. The pump hydraulic torque (T ) is evaluated from the homologous curves and Equation 9-8 as a function of the fluid density and flow rate as well as pump angular velocity. Pump Options and Limitations The wall heat transfer, wall friction, CHF calculation and implicit hydrodynamics options for the PUMP module are the same as for the PIPE module. In addition, the following options are specified: pump type, motor action, reverse speed option, two-phase option, and pump curve option. If the pump motor is energized, its angular velocity is assumed to be the constant value specified. If the motor is not energized, a pump coastdown calculation is performed using the specified initial pump speed. There are two pump options available. For pump option 1 (IPMPTY = 1) the pump speed variation is specified by input. The pump is initially energized at a constant speed specified by input (OMEGA). The pump motor may be tripped by a TRIP signal. If a pump trip has occurred, the pump speed is taken from a table of pump speed versus time-after-trip (array SPTBL). Pump option 2 (IPMPTY = 2) is similar to option I except that the pump speed is calculated from Equation 9-9 after a trip has occurred rather than from an input table. The electric torque TE is assumed to be zero. The relationships between the various pump input parameters as well E as the algorithm for the pump speed calculation are shown in Table 9-2. The value entered for IPMPTR is the TRIP identification number for pump trip initiation and NPMPTX is the number of pairs of points in the pump speed table (SPTBL). If IPMPTR = 0, the pump will maintain a constant speed. If the reverse speed option is specified (IRP = 1), the pump can rotate in both the forward and reverse directions. If reverse speed is not allowed (IRP = 0), the pump will rotate in the forward direction only. For this case, if a negative speed is calculated (after trip with option 2), the speed will be set to zero. 4384-non\sec9.wpd-04203 99

If the two-phase option is tumed on (IPM = 1), the degraded pump head and torque will be calculated from Equations 9-7 and 9-8. If the two-phase option is turned off (IPM = 0), only the single-phase head and torque homologous curves will be used. The user may either specify pump homologous curves in the input or use the built-in pump curves. The built-in pump curves are for the MOD-I Semiscale system pump and are based on the data of Olson (1974a, b) and Loomis (1974). For other types of PWR pumps their corresponding homologous curves and multiplier values would be specified. Since these homologous curves are dimensionless, they can be used to describe a variety of pumps by specifying as input the rated values for density, head, torque, flow, and angular velocity. There are several restrictions and limitations in the current version of the pump component. Since there is no pump motor torque-versus-speed model, the pump speed is assumed at the input value if the motor is energized. The pump momentum source must be located between cells 1 and 2 of the pump model. Finally, the head degradation multiplier M(a) and the torque degradation multiplier N(a) are assumed to apply to all operating states of the pump. The PUMP module input consists of the same geometric and hydrodynamic data and initial conditions that are required for the PIPE module. In addition, information specific to the PUMP is required. The speed table (SPTBL) as well as the homologous pump curve arrays must be input. Model as Coded For the new timestep, Equation 9-9 is evaluated explicitly: 0 = (On + (d)n t (9-11) The momentum source for a pump cell is evaluated once each timestep, and the source is applied only during the explicit pass in subroutine DF1DI or subroutine DFlDS. The mixture velocity and mixture density from the donor component (i.e., conditions at the upstream boundary of the pump component) are used to establish the volumetric flowrate through the pump. Standard curve fitting techniques are then used to compute the pump head. The pump source evaluation is performed by subroutine PUMPSR. 4384-non\sec9.wpd-04203 9-10

Scaling Considerations During blowdown and reflood periods, reactor coolant pumps will be under two-phase flow conditions, and both the pump head and the pump torque will be degraded. Although the physical mechanisms responsible for the performance degradation in two-phase flows are not well understood, analysis of tests on pumps (Kamath and Swift, 1982) revealed that "scaling down the size of the pump while maintaining the same design specific speed produces very similar performance characteristics both in single and two-phase flows." The study also indicated that effects due to size and operating speed were not discernible within the range of test conditions and within experimental uncertainties. The system pressure, however, appeared to affect the rate of degradation even for the same pump. Similar results were also observed in the scaled-pump experimental tests conducted by KWU (Kostner and Seeburger, 1983). These test results suggest that uncertainties due to scaling distortion from the pump are small compared to other contributors. The effect of scaling and other uncertainties is minimized in the WCOBRAITRAC model by using data from a 1/3-scale model similar in design to the Westinghouse pump (Snyder and Grigsby, 1982). Conclusions The pump model is constructed by combining the experimentally-established pump characteristic correlations and the WCOBRA/TRAC PIPE module based on a one-dimensional drift-flux formulation. The frictional torque correlation was also experimentally established. The pump model can handle all single- and two-phase operations (with or without phase separation) and provide accurate speed, flow, and head predictions during the transient (including coastdown). The options of the model provide the users with the flexibility to model a variety of system operating conditions. The WCOBRATIRAC pump model has been assessed against LOFT L2-5 test data (Bayless et al., 1982) with satisfactory results. The model can be utilized to simulate any PWR pump for which the homologous characteristic curves have been adequately established. 9-5 Steam Generator Component (STGEN) Model Basis In a PWR, the steam generators transfer energy from the primary coolant loop to the secondary coolant to produce steam. The STGEN module can model either "U-tube" or "once-through" steam generators; the basic operation is similar for both types. Primary coolant enters an inlet plenum, flows through a tube bank in which the primary coolant exchanges heat with a secondary coolant that flows over the exterior of the tube bank, and finally discharges into an outlet plenum. Figure 9-8 provides typical noding diagrams for U-tube and once-through steam generators. In both cases the tube bank is represented by a single effective tube that has heat transfer characteristics of the entire tube bank. 4384-non\sec9.wpd-04203 9-11

Model as Coded The number of fluid mesh cells is specified by NCELL1 on the primary side and by NCELL2 on the secondary side. There are some constraints imposed on the possible values for (NCELL1, NCELL2) combinations. For a once-through type, it is required that NCELL2=NCELL1-2. For a U-tube type, it is assumed that there is a one-to-one correspondence between two active primary cells and one active secondary cell (Figure 9-8). Thus for the fluid cells on the secondary side to reach the U-tube bundle top, it is required that NCELL2 2 (NCELLI-2)/2. The secondary-side cells that are greater than (NCELLI-2)/2 are treated adiabatically and are used to model possible area changes and volumes above the tube bank. In Figure 9-8, these are cells 6 through 8 on the secondary side. There is an inlet plenum (cell 1) and outlet plenum (last cell) on the primary side; these two cells are assumed adiabatic. The steam generator, primary-side, and secondary-side hydrodynamics are treated separately. Coupling between the two sides is achieved through wall heat transfer, which is modelled in a semi-implicit fashion. The calculational sequence for a steam generator is identical to that for a PIPE (component) except that it is performed twice, once for the primary side and once for the secondary side. It is possible to connect the secondary-side junctions to any TRAC component, but the most common arrangement is to connect the inlet to a FILL, specifying the secondary-side fluid inlet conditions and flow rate, and to a BREAK at the discharge, specifying the steam- , generator secondary discharge pressure. The cylindrical heat conduction equation for a typical tube is solved as described in Section 7-7. There must be at least one wall temperature node, but three are suggested, placing one at each tube surface and one at the tube wall center. The tube material is selected from the material options given in Section 10-5. Wall friction correlations and additional frictional losses for the primary and secondary sides can be specified as described in Section 4-7. Either fully implicit or seni-implicit hydrodynamics may be selected for the steam generator component. 9-6 Pressurizer Component (PRIZER) Model Basis The pressurizer in a PWR is used to control the primary coolant system operating pressure and accommodate any change in the coolant volume during normal operation. It consists of a pressure vessel connected to one of the hot legs by a surge line. Approximately half of the vessel is filled with water, which is pressurized by saturated steam above it. The pressure is maintained at the operating setpoint value by a system of heaters and sprays which regulate the energy input to the water. 4384-non\sec9.wpd-04203 9-12

Model as Coded The pressurizer is simulated by the PRIZER component. It can connect only to another one-dimensional component, and its nodes are numbered, 1 to NCELL, from the top (closed end) to the junction at the bottom as shown in Figure 9-9. The PRIZER component is treated in most respects as a PIPE; however, the drift velocities are not obtained from the slip routine, but are specified in subroutine PRIZR1, which imposes a sharp liquid/vapor interface during the pressurizer discharge. This is done by setting the relative velocity to a large value, [

                                                                                       ]a   (  l12)(2)

The negative sign is included to be consistent with the sign conventions used in the code. The controlling action of the heater/spray can be simulated in the PRIZER component. The heater/spray model is available as an input option and is used as a system pressure controller. If this option is used, the setpoint pressure and the pressure deviation DPMAX at which the heaters deliver their maximum power QHEAT are input. The calculated heater power is directly proportional to the difference between PSET and P(l), the pressure in node 1. Qpressurizer = QHEAT(PSET - P(l))IDPMAX This power (Qpressurizer) is limited to +/- QHEAT and is distributed to each node as a function of the node liquid fraction to total pressurizer liquid fraction. Power is not added if the collapsed liquid level falls below the input height ZHTR. The collapsed liquid level within the PRIZER component is given by the following equation: z = V,/A (9-13) where NCELIS VI i=l ( - xi) Vi(9-14) and V and V are the volume of the node i and the total volume of liquid in the pressurizer, respectively. A is the maximum flow area of nodes 1 and 2. 4384-non\sec9.wpd-04203 9-13

9-7 VALVE Component Model Basis The VALVE component is used to simulate the controlling action of a valve fitting. It comprises at least two fluid nodes. The flow area and hydraulic diameter at a given node boundary are used as the controlling parameters to model the valve operation. In all other respects, the VALVE component is identical to the PIPE component. Model as Coded The noding scheme is shown in Figure 9-10. Node IVPS defines the node boundary where the valve action is modelled. Five options are provided to describe the valve operation (Table 9-3). Options 1 through 4 open or close the valve with a trip. The action can be instantaneous or a function of time. Option 5 models a check valve with the open or closed condition determined by a pressure differential between the specified nodes (IVPS and IVPS-1) and a set point. For this option the valve opening and closing is damped to prevent pressure oscillations. 9-8 Accumulator Component (ACCUM) An accumulator is a pressure vessel partially filled with water and pressurized with nitrogen gas. The accumulator is isolated from the primary coolant system (RCS) by a check valve. If reactor coolant pressure falls below accumulator pressure, the check valve opens and the accumulator water is forced into the RCS. This flow continues until the accumulator is empty, after which the nitrogen cover gas is discharged. During a LOCA transient, the accumulators of a PWR will deliver ECC water to the cold legs. The accumulator injection period may be divided into two time intervals: Phase A: tAcc t to PhaseB: to t t, where tACc is the time when the accumulator starts to deliver ECC water, to is the time when the accumulator is empty of water, t is the time when the pressure in the accumulator is in equilibrium with that of the RCS, and no more flow issues from the accumulator. Although the core recovers during a small break LOCA event priorto the time at which the accumulator empties, Phase B is discussedfor completeness. 4384-non\sec9.wpd-04203 9-14

During phase A, only water enters the RCS. The nitrogen in the accumulator continues to expand in volume as the pressure in the accumulator decreases. The nitrogen cools as it expands. During this phase, accumulator water begins to fill the reactor vessel downcomer and core. Meanwhile, the reactor pressure falls to near the containment pressure. During phase B a water/nitrogen mixture, and finally only the-nitrogen gas, enters the RCS. Because of the width of the tank, the water-nitrogen interface is likely to be well-defined. Consequently, the time during which a water-nitrogen mixture flows from the tank is expected to be small. As the nitrogen flows into the vessel, the upper portion of the downcomer may be pressurized due to the presence of the nitrogen flow. This increase in pressure may affect the cooling flow entering or leaving the core. The way in which these phenomena are simulated in WCOBRA/TRAC is described below. Accumulator Model Basis (Phase A) The accumulator component is simulated in the ACCUM module in WCOBRA/TRAC. This component can only be connected at one junction to other WCOBRA/TRAC components. This connection is the highest number cell, and it is assumed that cell 1 is closed, as shown in the typical noding diagram in Figure 9-11. It is also assumed that the accumulator is not connected to a nitrogen pressure source. Therefore, the nitrogen pressure results from the expansion of the initial gas volume. The following additional assumptions are made for the ACCUM component during Phase A:

1. The vapor phase in the accumulator is an ideal gas with the properties of nitrogen.

2. The relative velocity between the vapor and liquid is set to a large value to create a sharp interface between the liquid and vapor. This assumption is made because the relatively large diameter of the tank leads to low fluid velocities and rapid phase separation.

3. The mixture properties at the last accumulator cell are controlled such that only pure liquid is discharged. This assumption is also a result of the expected sharp interface between liquid and vapor.

4. The wall friction factor for each accumulator tank cell is set to a constant value of 0.005. The accumulator is expected to represent a negligible portion of the overall resistance to flow.

4384-non\sec9.wpd-04203 9-15

5. The accumulator tank walls are assumed to be adiabatic. Heat transfer from the accumulator walls is not expected to be significant, due to the small surface area per unit volume.

Nitrogen Discharge Model Basis (Phase B) During the accumulator water injection period, a nitrogen gas field is assumed to exist in the accumulator, while steam is assumed everywhere else in WCOBRA/TRAC. While the nitrogen field can be extended (as an input option) to all other WCOBRA/TRAC components, a combined nitrogen-steam-water model is not available. To simulate the nitrogen discharge, the subcooled vapor model in WCOBRA/TRAC is used to provide similar pressure/flow characteristics to those obtained from a nitrogen model. In this model, the normal hydrodynamics package is used. However, the following additional assumptions are made:

1. Phase B is assumed to begin when the water level in the accumulator tank falls below [ ]C (the basis for this value is described in Section 16-2-5). At this point, a mixture of water and nitrogen is assumed to flow out of the tank.

2. During Phase B, heat transfer between liquid and vapor is suppressed in regions of the RCS expected to contain significant amounts of nitrogen. This is assumed to occur as long as the accumulator pressure remains significantly above the RCS pressure (implying significant flow of nitrogen).

The region over which the condensation suppression is assumed to occur is shown in Figure 9-12 and consists of the accumulator and line, the intact cold leg, the upper downcomer region, and the broken cold leg on the vessel side. The nitrogen influence is assumed to be limited to this region as discussed below. At the time nitrogen begins to inject, the lower plenum and downcomer are full of water, and the core has begun to reflood. Any steam generated in the core will flow up the core and out through the loops and upper head vent paths. In addition, the high downcomer water level provides a driving force for this flow. It is therefore unlikely that accumulator nitrogen flow will cause reverse flow in the loop or upper head. If it does, this flow would have to be sustained for a substantial period of time before the nitrogen will reach the upper plenum. 43 84-non\sec9.wpd-04203 9-16

In the reactor vessel, the accumulator water isolates the nitrogen from the core. The region of influence is assumed to extend to a point in the downcomer level with the bottom of the core. If the downcomer is full above this level, no steam will be available for condensation and the condensation suppression will make no difference.

3. During Phase B, the behavior of the nitrogen can be simulated using the subcooled vapor models in the code.

This assumption was checked by comparing two simple models of the accumulator, one in which the entire process takes place with nitrogen, and one where the nitrogen model is replaced during Phase B with a model using the one-dimensional component subcooled vapor equations. 3 ) In the nitrogen model, the pressure/temperature/density relationships are for a perfect gas. The simple models were used to predict pressure and flow, using a linear ramp for the pressure at the accumulator exit and representative accumulator dimensions. Accumulator and Nitrogen Model as Coded The procedures for data input, initialization of arrays, advancement of time-dependent variables, and editing are similar to those given for a PIPE component. The hydrodynamics are treated using the one-dimensional, semi-implicit drift-flux routine DFIDS. No metal heat transfer is permitted for the accumulator. In addition, the following special coding is employed for each of the phases. During Phase A:

1. Nitrogen properties are calculated in subroutine THERMO. The gas constant used is 287.12 Pa m 3 /kg K (53.4 ft lbf/lbm OR), which is consistent with standard values found in handbooks.

4384-non\sec9.wpd-04203 9-17

2. The liquid vapor interface is sharply defined by setting the relative velocity to a large value. This is set in ACCUMl and is [

                                                                                              ]a    (9-15)

The negative sign is included to be consistent with the sign conventions used in the code.

3. The discharge at accumulator exit during Phase A is limited to liquid only by setting the component boundary array elements representing the void fraction to zero. This is done in subroutine ACCUMBD.

4. Accumulator wall friction is set to 0.005 in ACCUM 1. User specified friction factors, input via the parameter FRIC, may be added to this value.

The end of Phase A is determined by the collapsed liquid level. The collapsed liquid level is calculated in subroutine ACCMIX by computing the total liquid volume in the accumulator tank and then determining the height of this volume at the bottom of the tank. This collapsed level is used to signal that the accumulator is nearly empty. The signal is set when the collapsed level falls below [ ]a.C The time when this occurs is to, and the code moves to Phase B. For Phase B (simulated nitrogen injection), additional special coding is required as described below.

1. In the accumulator, [

ja.c as described in Section 5-3.

2. The steam properties for [

                                                                                               ]a.c

3. Discharge from the accumulator becomes two-phase.

4384-non\sec9.wpd-04203 9-18

Scaling Considerations The model was tested in simulations of the accumulator in the integral test facility simultaneous, and against data obtained from in-plant tests. A description of the in-plant test simulation as it applies to the accumulator model is given in Section 22 of this report. Conclusions The basic assumptions which are made in the application of this model to the PWR and which introduce uncertainty into the calculation are:

1. The condensation is assumed to be suppressed in the intact cold legs, upper downcomer, and broken nozzle until all nitrogen has been exhausted from the accumulator and swept from the systems.

2. The nitrogen vapor properties are approximated by subcooled vapor flow.

These uncertainties are not relevant to a small break LOCA scenario, and do not need to be considered in the uncertaintymethodology. 9-9 BREAK and FILL Components These models differ from other components in that they do not model any system component per se, and no hydrodynamic or heat transfer calculations are performed for them. In all other respects, they are treated as any other component, with the same input, initialization, and identification procedures. A BREAK component is used to impose a pressure boundary condition adjacent to the one-dimensional component with which it connects (Figure 9-13). The boundary conditions specified by the BREAK are pressure, mixture temperature and node void fraction, all of which may be time dependent. Care is required when setting the mixture temperature and void fraction values, as these are used to determine the properties of the fluid if the flow is calculated to be in the reverse direction, i.e., into the system from the BREAK. In the normal mode of operation, where the fluid flows out through the BREAK, the mixture temperature and void fraction do not affect the calculation. The FILL component is used to impose a velocity boundary condition at the junction between the FILL and the adjoining one-dimensional component (Figure 9-14). The boundary velocity may be specified by one of the five different input options. The options define the velocity as a 4384-non\sec9.wpd-04203 9-19

constant, or as a function of time or pressure, or as a constant until a trip signal is reached, then -_l again as a function of time or pressure. The fluid properties within the FILL node are determined from the user input values of void fraction, mixture temperature, and pressure. 9-10 References Bayless, P. D., et al., 1982, "Experimental Data Report for LOFT Large-Break Loss-of-Coolant Experiment L2-5," NUREG/CR-2826, EGG-2210. Biasi, L., et al., 1967, "Studies on Burnout, Part 3: A New Correlation for Round Ducts and Uniform Heating and Its Comparison with World Data," Energia Nucleare, Vol. 14, pp. 530-536. Bordelon, F. M., et al., 1974, "SATAN VI Program: Comprehensive Space-Time Dependent Analysis of Loss-of-Coolant," WCAP-8302. Kamath, P. S. and Swift, W. J., 1982, "Two-Phase Performance of Scale Models of a Primary Coolant Pump," EPRI NP-2578, Final Report. Kostner, W. and Seeburger, G. J., 1983, "Pump Behaviour and Its Impact on a Loss of Coolant Accident in a Pressurized Water Reactor," Nuclear Technology, Vol. 60. Liles, D. R., et al., 1981, "TRAC-PD2, An Advanced Best Estimate Computer Program for Pressurized Water Reactor Loss-of-Coolant Accident Analysis," NUREG/CR-2054. Loomis, G. G., 1974, "Intact Loop Pump Performance During the Semiscale MOD-1 Isothermal Test Series," Aerojet Nuclear Company, Report-1240. Olson, D. J., 1974a, "Single- and Two-Phase Performance Characteristics of the MOD-I Semiscale Pump Under Steady-State and Transient Fluid Conditions," Aerojet Nuclear Company, Report ANCR-1 165. Olson, D. J., 1974b, "Experimental Data Report for Single- and Two-phase Steady-State Test of the 1 Loop MOD-1 Semiscale System Pump," Aerojet Nuclear Company, Report ANCR-1 150. Snyder, P. H., and Grigsby, J. M., 1982, "EVA Project on Two-Phase Reactor Coolant Pump Performance - Data Analysis and Model," Vol. 1-3, WCAP-10109. 4384-non\sec9.wpd-04203 9-20

9-11 RAI Listing

1. RAI1-233

2. RAI1-234

3. RAIS-21 4384-non\sec9.wpd-04203 9-21

Table 9-1 The Four Segments of Pump Homologous Curves Curve Homologous Homologous Variable Operating Segment Head Torque Range Condition 2 1 2__ _/'XN d I_ < ()o> 2 hAd fl/il It;I > Q>O 3 hl? I A/d I IN >1 Q <o Note: A fourth segment may also be input for negative pump roation ( < 0). This condition will not occur in Westinghouse PWR's due to locking devices on the pumps. 4384-nonsec9.wpd-04203 9-22

Table 9-2 Pump Control Input Parameter IPMPTY NPMPTX Pump IMIPPTR (SPTBL) Pump Speed Option Pump Trip I.D. Pair of Points Speed Table Algorithm 1 x = pump trip x x OMEGA before desired . tnp O= no pump trip 0 SPTBL after trip 2 x = pump trip x OMEGA before desired trip 0 = no pump trip 0 Code calculated after trip 4384-non\sec9.wpd-04203 9-23

Table 9-3 Valve Control Options

1. Valve is normally open and is closed instantly on a trip signal.

Controlling logic is as follows: Before trip, A(valve) = AVLVE Dh(valve) HVLVE After trip, A(vave) = 0.0 VM = 1.E-10 VR = 0.0 where, AVLVE equals completely open valve area HVLVE equals completely open valve hydraulic diameter VM equals mixture velocity of phases VR equals relative velocity of phases

2. Valve is normally closed and is opened instantly on a trip signal.

Controlling logic is as follows: Before trip, A(valve) 0.0 VM = L.E-10 VR = 0.0 After trip, A(valve) = AVLVE Dh(valve) = HVLVE

3. Valve is normally open and is closed on a trip signal according to a time-dependent valve table.

Controlling logic is as follows: Before trip, A(valve) = AVLVE Dh(valvc) = HVLVE After trip, A(valve) = AVLVE

• SCALE Dh(vlve) = HVLVE*SCALE where SCALE equals the linear interpolated multiplier from the user input forcing factor versus time table. If SCALE equals 0.0, VM= L.E-10 VR=0.0 X

43 84-non\sec9.wpd-04203 9-24

Table 9-3 (Cont'd) Valve Control Options

4. Valve is normally closed and is opened on a trip signal according to a time-dependent valve table.

Controlling logic is as follows: Before trip, A(valve) = 0.0 VM = .E-10 VR = 0.0 After trip, A(valve) = AVLVE

• SCALE Dh(valve) HVLVE
• SCALE where, SCALE has the same definition as given above.

5. Check valve is controlled by a static pressure gradient. If IVPG = 1, then DP = P(IVPS -

1) - P(IVPS); if IVPG = 2 then DP = P(IVPS) - P(IVPS-1)

If DP+PVS > 0, the valve opens. If DP+PVS < 0, the valve closes. For this option the valve opening and closing action is damped according to the following equations. Opening, A(valvc) = A(valve) *099 + 0.01

• AVLVE h(valve)= Dh(valve)
• 0.99 + L.0E-5 The above equations are applied at each timestep until the opening or closing action has been completed.

4384-non\sec9.wpd-04203 9-25

1 2 6 7 II 3 4 5 1 "I.,4 I a I I-1 1 2 1 3 4i 5 6 7 - f I~~~~~~~~

      / JUNi JUN 2 Figure 9-1. PIPE Component Noding 43 84-non\sec9.wpd-04203                   9-26

NCELL2 PIPE 2 I JCELL NCELL1 PIPE 1 Figure 9-2. TEE Component Noding 4384-nonkec9.wpd-04203 9-27

NORMAL FLOW DI RECTI ON 1 2 3 j44IJ lI l l l N-1l NJ SMOM Figure 9-3. PUMP Noding Diagram 4384-non\sec9.wpd-04203 9-28

a,c Figure 9-4. 93A Pump Single-Phase Homologous Head Curves a,c Figure 9-5. 93A Pump Two-Phase Homologous Head Curves 4384-non\sec9.wpd-04203 9-29

a,c Figure 9-6. 93A Pump Single-Phase Homologous Torque Curves a,c Figure 9-7. 93A Pump Two-Phase Homologous Torque Curves 4384-non\sec9.wpd-04203 9-30

U-TUBE TYPE ONCE THRU YPE FEEDWATER l l,-, INLET FEEDWATER t I NLET 11 PRIMARY PRIMARY INLET I NLET Figure 9-8. Steam Generator Noding Diagram 4384-non\sec9.wpd-04203 9-31

1 2 3 4 5 JUNCTION Figure 9-9. Pressurizer (PRIZER) Component Noding 4384-non\sec9.wpd-04203 9-32

JUN JUN 2 Vr Z ~~~~~~~~CELL < L l~~~~~~~~VPS< FLOW AREA CONTROLLED BY VALVE ACTION Figure 9-10. VALVE Component Noding 4384-non\sec9.wpd-04203 9-33

a,c Figure 9-11. Accumulator Noding Diagram 4384-non\sec9.wpd-04203 9-34

PA(N). TA(N) ACCUMUIATOR FLW FROM UPPER HEAD lACCLATOR IE X FLOWI FROM U_ , W.)TR(N) I IRTACT LOOP # RATCLDLCl> XNlT BROKEN COLD LEG iATER DOIdCOMER Figure 9-12. Condensation Suppression Region for Accumulator/Nitrogen Model 4384-non\sec9.wpd-04203 9-35

BREAK COMPONENT ADJACENT COMPONENT . I-I I '1-JUNCTI ON4 K s1 PRESSURE SPECIFIED AT THIS POINT Figure 9-13. Pressure Boundary Condition Using BREAK Component FILL COMPONENT ADJACENT COMPONENT VELOCtTY SECIFIEDt AT THE JUNCTION Figure 9-14. Velocity Boundary Condition Using FILL Component 43 84-non\sec9.wpd-04203 9-36

SECTION 10 THERMOPHYSICAL PROPERTIES 10-1 Introduction WCOBRA/TRAC includes a set of functional routines and individual correlations to calculate the thermal properties of water, air, nuclear rods and several common structural materials. This section describes the manner in which the thermal properties are calculated for the vessel and one-dimensional components. Section 10-2 describes calculation of the thermodynamic properties of water. Section 10-3 describes the WCOBRA/TRAC calculation of air thermal properties. Section 104 describes the thermal properties of materials used in nuclear fuel rods including mixed oxide fuel, clad materials, and fuel rod gap gases. WCOBRA/TRAC can also calculate the thermal properties of several common PWR structural materials such as stainless steel. These calculations are described in Section 10-5. 10-2 Thermophysical Properties of Water 10-2-1 Vessel Component Water Properties The thermal-hydraulic calculations performed by the WCOBRA/TRAC vessel component frequently require the thermal conductivity, specific heat, viscosity, Prandtl number, and surface tension for water as functions of the fluid pressure and specific enthalpy. This section describes the thermodynamic property calculations performed by WCOBRAtTRAC for saturated, superheated, and subcooled fluid conditions. 10-2-1-1 Saturated Fluid Properties Model Basis The saturated liquid and saturated vapor enthalpies are calculated as functions of the pressure. Values for the saturation temperature, densities of saturated liquid and vapor, thermal conductivities and viscosities of saturated liquid and vapor, saturated liquid specific heat, and the surface tension are interpolated from tables indexed by saturated liquid enthalpy. The saturated liquid and saturated vapor specific enthalpies are determined from polynomial representations of the saturation curve. This representation provides close agreement with ASME Steam Tables (1968) and the NBS/NRC Steam Tables (Haar, Gallagher, and Kell, 1984). 4384-non\secl O.wpd-04203 10-1

The tables of values at saturation for the other properties (conductivities, viscosities, etc.) are also in close agreement with the standard tables. The saturation enthalpies are calculated in BtuAlbm as functions of pressure based on expressions developed for EPRI (McFadden et al., 1980). The polynomial expansions for saturated liquid enthalpy are 9 H(P) = E A, [n(P)]Y-' (10-1) n=l if P < 2529.9 psia and 9 H/P) = E A[(3208.2 - p)041P-1 (10-2) n=1 for 2529.9 P < 3208.0 psia. The constants An for Equations 10-1 and 10-2 are shown in Table 10-1. The saturated vapor enthalpy is calculated using 5 8 Hg(P) B,[ln(P))n- + Z Bn[ln(P)]n 3 (10-3) n=l n=6 if 0.1 < P < 1467.6 psia, by 9 Hg(P) = I n[In(P)]"-' (10-4) n=l 4384-non\secIO.wpd-04203 10-2

if 1467.6 s P < 2586.0 psia and by 9 H(P) = E Bn[(3208.2 _ p)O41P-1 (10-5) n=1 if 2586.0 s P < 3208.0 psia. The constants Ln for Equations 10-3 through 10-5 are listed in Table 10-2. These expressions are compared to values from the ASME Steam Tables (1968, 1983) in Figures 10-1 and 10-2. Table 10-3 lists values of the saturation temperature, density, viscosity, thermal conductivity, specific heat, and surface tension that are used to represent the saturation curve for those properties. The saturation curves defined by these tables are compared to values from the standard tables in Figures 10-3 through 10-11. Model as Coded For a known pressure P the saturated liquid enthalpy is calculated using either Equation 10-1 or 10-2 in subroutine SAT. From that calculated value of saturated liquid enthalpy, the other properties are determined in subroutine PROP by linearly interpolating between the 90 values listed in Table 10-3. Scale Considerations Calculation of saturated water thermophysical properties is not dependent on scale. Conclusions The WCOBRAITRAC vessel component calculates saturated liquid and saturated vapor enthalpies as functions of pressure using polynomial representations, and then uses the saturated liquid enthalpy to determine the other thermal properties by linear interpolation. All of the saturated properties agree very closely with values found in the standardized Steam Tables. 4384-non\secl O.wpd-04203 10-3

10-2-1-2 Properties of Superheated Vapor Model Basis Vapor Enthalpv The enthalpy of superheated vapor as a function of pressure and temperature is calculated by the expression developed by Keenan and Keys (1936): H, = 0.43 [0.10129 (FOP + F+/- p2 + _2 p4 + 12 p13) + Fj (10-6) where, F, F, F3 , and F 2 are defined by Fk= a (Bk ), k = 0,1,3,12 (10-7) The coefficients Bk are defined as:

               = IT                                                                    (10-8)

Bo = 1.89 - 2641.62 10808702 (10-9) B, = B82.546 T2 - 1.6246(10)5 3

                                                  ,C  )                               (10-10)

B3 = B (0.21828 ' - 1.2697(10) 5 'T) (10-11) B = -B 13 (3.635(10)-4 -r2 - 6768(10)64Y36 ) (10-12) 4384-non\sec1O.wpd-04203 104

and F' is given by F' = 2502.36 + fT (1.472 + 0.00075566T + 478365)dT(10-13) 3.16 T In Equations 10-6 through 10-13, T is in K, P is in atmospheres, and HV is in J/g. Vapor Temperature Values for superheated vapor temperature as a function of pressure and enthalpy are calculated using an iterative method described by McClintock and Silvestri (1936). Estimates for T and C are computed from the expressions 2 3 2 T = Al + A 2 Hv + A3 H + A4 H, + A P + A 6P (10-14)

                   + A 7P 3 + P(A8Hv + Ag9 1        + AIOHv)

IICp = B + B 2 H + B 3 H 2 + B 4 H 3 + B 5 1nP + B6 (ln p)2 (10-15)

                     + B7 (ln P)3   + (InP)(B8 H + BH    2  + BIOH3 )

where T is in F, P is in psia, Hv is in Btu/bm, and Cp is in BtuAbm-°F. The constants An and B. depend on the range of pressure and enthalpy as shown in Table 10-4. The estimated temperature is then used to approximate the enthalpy as Hv?(P,T) = f(P,T) (10-16) where the function f(P,7) is described by Equations 10-6 through 10-13. A temperature correction is calculated as 4384-non\secl O.wpd-04203 10-5

AT = C (H, - H)(1-7 and a new estimated temperature is defined as T' = T + AT (10-18) A new enthalpy is calculated and the iteration is continued until IATI < 1.0°F or HV - HVfl < 0.5 Btullbm Vapor Density The vapor specific volume is calculated as a function of pressure and enthalpy using equations from Keenan and Keys (1936): u = = E + E 2P+ 3 + E4 H + E5 PH +E- 2 (10-19) where P is in psia, H, is in Btu/lbm, and u is in ft 3 Abm. The constants for these equations are El = -0.81735849E-03 E2 = 0.12378514E-04 E3 = -0.10339904E+04 E4 = -0.62941689E-05 E 5 = -0.872921608E-08 E6 = 0.12460225E+I01 Vapor Thermal Conductivity The thermal conductivity for superheated vapor is calculated as a function of temperature and density using equations given in the ASME Steam Tables (1968). The expression for thermal conductivity is: 4384-non\sec IO.wpd-04203 10-6

ky = k + (103.51 + 0.4198 T - 2.771(10)-5T 2 )p, + 2.1482(10)14 v (10-20) where: k = 17.6 + 5.87(10)-2T + 1.04(l0)-4 T 2 - 4.51(10)-8 T 3 (10-21) In Equations 10-20 and 10-21, T is in C, pv is in g/cm3 , and kv is in mW/m-°K. Vapor Viscosity The viscosity for superheated vapor is calculated as a function of temperature and density using equations given in the ASME Steam Tables (1968). The viscosity is given by p - p(1858 - 5.9Y) , if T <3400 C Pv l + 353p + 676.5p 2 + 102.1p 3 , if T > 3650C (1022) PI = 0.407T + 80.4 (10-23) For values of T between 340 0 C and 365°C the viscosity is interpolated between the values given by the two expressions in Equation 10-22. In Equations 10-22 and 10-23 temperature is in C, density is in g/cm3 , and viscosity is in micropoise. Values of superheated vapor enthalpy, temperature, density, thermal conductivity, and viscosity defined by the foregoing expressions are compared with the ASME tables (1968, 1983) and the National Bureau of Standards/National Research Council tables (Haar, Gallagher, and Kell, 1984) in Figures 10-12 through 10-16. Model as Coded The properties for superheated vapor represented by Equations 10-6 through 10-23 are coded as described above without modification in subroutines HGAS, TGAS, VOLVAP, and TRANSP. Properties are not calculated if P < 0.1 psia or if P > 3208.0 psia, in which cases an error message is printed and execution is terminated. 4384-non\secl O.wpd-04203 10-7

In the calculation of vapor temperature as a function of pressure and enthalpy, Equations 10-14 through 10-18 describe an iterative method. A maximum of 10 iterations are permitted. Scalin! Considerations The equations and methods used to calculate the properties for superheated vapor are independent of scale. Conclusions The WCOBRA/TRAC vessel component calculates superheated vapor enthalpy as a function of temperature and pressure, density as a function of pressure and enthalpy, and thermal conductivity as a function of temperature and density, using generalized polynomials. Temperature as a function of pressure and enthalpy is found iteratively using the enthalpy function. All of these properties agree closely with values found in standard steam tables. 10-2-1-3 Subcooled Liquid Properties Model Basis Subcooled liquid specific volume is calculated using the equation 3~ ~ ~ ~ V = exp [ ( Ccxi Pi4 H l (10-24) where H, is in Btu/Ibm, P is in psia, and the values of the coefficients Ccxij are given in Table 10-5. The liquid temperature at enthalpy (H,) is assumed to be equal to the saturation temperature at H,. The properties Cp, k, and p for subcooled liquid at temperature T are assumed to be equal to the saturated liquid properties at T. These properties are only weakly dependent on pressure in the low to moderate pressure range. The liquid Prandtl number is calculated as Pr = kP (10-25) kf 4384-non\sec1O.wpd-04203 10-8

Model as Coded The equation for subcooled liquid specific volume is programmed as shown in subroutine VOLLIQ. Other subcooled liquid properties are determined by linear interpolation of the saturation properties listed in Table 10-3. The liquid enthalpy is used as the index to determine the appropriate location in the table in which to perform the interpolation. Scalin2 Considerations The method in which subcooled liquid properties are determined is scale independent. Conclusions Subcooled liquid properties are estimated to be equal to the properties of saturated liquid corresponding to the liquid temperature. Since these properties are only weakly dependent on pressure, only a negligible error is introduced into the calculation. 10-2-2 One-Dimensional Component Water Properties The thermodynamic and transport properties used in the WCOBRATIRAC one-dimensional (D) components are based on polynomial fits to steam table data for water, and on ideal gas behavior for air.- The fits for transport properties were obtained from Coffrnan and Lynn (1966). 10-2-2-1 Saturated Fluid Properties Model Basis Saturation Temperature and Pressure Saturation temperature as a function of pressure, and saturation pressure as a function of temperature, are calculated using expressions recommended by Rivard and Torrey (1975). These are T23 - 255.2 0 (10-26) 117.8 ) and Tsa = 117.8 (10-5Psa,)223 + 255.2 (10-27) 4384-non\secl 0.wpd-04203 10-9

The derivative of saturation temperature with respect to pressure is given by 0.223 (sat - 255.2) (10-28) Psat Saturated Vapor Internal Enera and Enthalpy Two main pressure regions are used in the calculation of water vapor internal energy and enthalpy. The low pressure range is P < 2.0 x 106 Pa and the high pressure range is 2.0 x 106 Pa < P, where P is the pressure and Tsat is its corresponding saturation temperature. Low Pressure Region: The internal energy of saturated vapor and its derivative with respect to pressure are eg = AVE(1) + BVE(1)T, (10-29) deg = -BVE(1) T' (10-30) dP where: T = 1I(P + 3.403E5) (10-31) High Pressure Region: eg = AVE(2) + BVE(2) P + CVE(2) p2 (10-32) deg = BVE(2) + 2 CVE(2) P (10-33) dP 4384-non\sec10.wpd-04203 10-10

The values of the constants AVE(i), BVE(z), and CVE(i) are listed in Table 10-6. All pressures: The ratio of specific heats, saturated vapor enthalpy, and derivatives with respect to pressure are calculated from: yg = AVG(i) + BVG(z) P + CVG(z) p 2 (10-34) dy_ dPl = BVG(z) + 2 CVG(i) P (10-35) Hg = yg e (10-36) dH8 = deg (10-37) dP g dP The values of the constants AVG(i), BVG(i), and CVG(i) are listed in Table 10-7. Saturated Liquid Internal Energy and Enthalpy A series of polynomials in T is used to calculate the internal energy of saturated liquid and its derivative with respect to saturation temperature. These are given by: ef = ALE(i) + BLE(I) T + CLE(i) T + DLE(i) T + ELE(I) T (10-38) and f0 ==t BLE(i) + 2 CLE(i) Tsa, + 3 DLE(i) T + 4 ELE(t) T . (10-39) 4384-non\sec IO.wpd-04203 10-1 1

where: i =1 for Tsat < 548.15 K, i=2 for 548.15 T < 611.15 K, i=3 for 611.15 Tt Table 10-8 lists the constants ALE(i), BLE(i), CLE(i), DLE(O), and ELE(i) for the given temperature ranges. Saturated liquid enthalpy is calculated using the definition Hf = ef + P (10-40) Pf and its derivative by dHf def dTsat I P p( + P a7 t (10-41) dP dT, dP pf 2 ap T,,U taTsa)p dP J where ef and its derivative are evaluated as shown earlier, and where p = p (P, Tsa) and its derivatives are evaluated using the equations in Section 10-2-24 with T equal to 7'sa, Saturated Vapor Specific Heat Capacity The heat capacity of saturated steam at constant pressure is also calculated using a polynomial representation in Tsat. The saturated vapor specific heat and its derivative are given by CPdpg = ACP

                    = 2 0&2 + BCP Cr       Er   + CCP + DCP e), +r ECP E)2                (1042) dCpg        [2AcP      3 +      C      -     CPrEPEjT           d~t, dP                 r                rPO;    c    - E'       r  dp            (10-43) 43 84-non\sec IO.wpd-04203                              10-12

where: E)= I1TI.t /Tcnt T'l,= 647.3 K and ACP = 8.349824 BCP = 349.519444 CCP = 2996.018036 DCP = -8448.077393 ECP = 9700.016602 Model as Coded Subroutine THERMO supplies thermodynamic properties for WCOBRA/TRAC one-dimensional components. The input variables are the pressure and the liquid- and vapor-phase temperatures. The output variables include the saturation temperature, saturated liquid, and saturated vapor enthalpies corresponding to the pressure, and their derivatives with respect to pressure. These variables also include the internal energies and densities of the liquid and vapor phases, and their partial derivatives with respect to pressure (at constant temperature) and with respect to temperature (at constant pressure). THERMO supplies thermodynamic properties valid for temperatures and pressures within the following ranges: 280 K T*s 697 K and 1000 Pa P 19.0 X 106 Pa. If THERMO is provided with a temperature outside this range, the calculation stops. Given a pressure outside this range, it adjusts the data to the corresponding limit and issues a warning message. 4384-non\secl O.wpd-04203 10-13

Subroutine RHOLIQ calculates liquid densities and density derivatives used in THERMO. Saturation pressure, and phasic densities and enthalpies as calculated are compared with NBS/NRC tables (Haar, Gallagher, and Kell, 1984) in Figures 10-17 through 10-21. Scaling Considerations Not applicable. Conclusions The saturation conditions for the WCOBRAIRAC one-dimensional components are calculated using polynomial expressions that provide a close approximation to the Steam Table values. The error introduced by the WCOBRA/TRAC routines is small and is not considered a major contributor to the overall code calculational uncertainty. 10-2-2-2 Properties of Superheated Vapor Model Basis Specific Heat at Constant Pressure The constant pressure specific heat of steam at temperature Tv is approximated as c 'm_pvideal I ____ Cpv = a( ) 8T 2 l (T2 ) ] (1044) where: [ ( 2 Cpg 1)2l (1045) pv. ideal The term Cpg is calculated as defined in Equation 10-42 and CpV, ideal is defined by ideal gas behavior, such that cpv.ideal RvYideal ideal 1 (10-46) 7 ideal 4384-non\sec I O.wpd-04203 10-14

where RV is the gas constant for steam (461.7 J/kg-K) and ideal = 1.3 is the ratio of ideal specific heats for steam. Internal Enerav The internal energy is obtained by integrating the expression for C along a line of constant pressure P. Integrating Equation 10-44 gives hv= hg+ CPideal (T~ T+ 2. _ 3) 1/2 (10-47) 2v The internal energy of vapor is therefore ee + Cpv ideal Tsa,) + (Tv )2 Pv Pg) (10-48) The definitions of enthalpy and internal energy allow the density of the water vapor to be written such that P p P = P_ = h - e [ hg + Cidea1 (T - Tsat) ] - [e + Cwideal (T - Ttaf) p (10-49) (hg - eg) + (ideal - 1) (e - e) 4384-non\secI O.wpd-04203 10-15

Substitution of P, and P, as defined by the preceding equation, into the equation for the internal IL energy of the vapor, gives e eg + Cideal (T T 1) + (T2 _ _ Ta ) (10-50) 2 Cpg -1 6 CPV~~~~~~~~~~~ideal where C idcal is the constant volume specific heat for steam as defined by ideal gas behavior given by Cv ideal -- I (10 -51) Yideal - The partial derivatives are given by I' aev) c w,ideal av P - p) (10-52) C JJ (aevl aP T,

                        =       1 2

ae) aTg) p 1 Ij K2

                                                               / +
                                                                  +K do dP (10-53) where:

2 I K = (e, - eg) + T7a, I + 2Cpg w,ideal _1 (10-54)

                                                                   -1 CPV,ideal 4384-non\sec IO.wpd-04203                                 10-16

and 2 de dT, g+ I + (ap) T w,ideal dP dP (10-55) dCpg dP and I dp 2 dT, 2 Tsat IdCpg dP 2 Cpg - dP Ts.t Cpvmideal dP (10-56) Cpv.ideal

                                                          -               )

Superheated Vapor Density The vapor density is calculated as p P = (10-57) (yg - 1) eg + (idea - 1) (e - eg) Therefore, the partial derivatives are calculated by (T apv a7

                       = -    aev aT

[ VJp j('(g - ('Yidea!

1) eg + (ideal

1) Pv

                                                                      -  1) (ev  -  e )]

(10-58) 43 84-non\sec O.wpd-04203 10-17

and (_] ap) TV

                         =        -     [e_dP +ide)Tg -             ddP (10-59) 1(^yg - )  ) +(d/          ( av)p,   ae)
          .(yg  -   l)eg + (ided!    -  1) (e     - e) )e               P T where:

( apv ) aev ) (Yideal 1

                                                )Pv (Yg- )eg+(yideal-l)(ev-eg)

(10-60) Enthalpy The enthalpy of superheated vapor is calculated using the definition of enthalpy, h = p (10-61) where ev is calculated from Equation 10-50, and pv is calculated using Equation 10-57. Model as Coded Thermodynamic properties for superheated water vapor are calculated in subroutine THERMO as described in this section. For superheated vapor, however, minimum and maximum limits are placed on the calculated values of the density and its partial derivatives. In low pressure regions where the above equations may predict a negative density, and near the critical point, it is necessary to impose the following limits on the density ratio 0 < PV - 0.9 (10-62) Pt 4384-non\secIO.wpd-04203 10-l 8

to avoid singularities when calculating certain parameters. If the calculated value of p, is outside these bounds, the vapor density and its derivatives are superseded by Pv= 0 . 9 PI (10-63) Capv '1 aTv) p 0 (ap2 J p (10-64) (ap = (ap T (10-65) Scaling Considerations Not applicable. Conclusions The thermodynamic properties for superheated vapor in WCOBRA/TRAC one-dimensional components are calculated from thermodynamic first principles. The calculated values are in good agreement with those found in the Steam Tables. The error introduced by the WCOBRA/TRAC routines is small and thus is not considered a major contributor to the overall code calculational uncertainty. 10-2-2-3 Subcooled Vapor Properties Model Basis WCOBRA/TRAC calculates internal energy, density, and enthalpy in one-dimensional components in the following manner when the vapor is subcooled. Internal Energ The internal energy and its derivatives for subcooled vapor are calculated as e, = eg+(Tv-7,) Pg (10-66) Yideal 4384-non\secl O.wpd-04203 10-19

( aTv)p Cpg Yideal (10-67) aev - deg (ev-eg dCpg aev) dTsat (10-68) aPJ T) dP Cpg ) dP p dP where 7'af is the saturation temperature corresponding to the vapor pressure (v) . The subcooled vapor density is calculated using Equation 10-57. If this value falls outside of the range 0 < Pv < 0 9P, tI then the intemal energy and its derivatives are recalculated and used in subsequent density recalculations. A new value of constant-pressure specific heat for vapor at the saturation condition is estimated: C = 958.75 (1 - 7T)-08566 (10-69) pg 7~ric and its derivative is T -1.8566 1 d, dCpg = (958.75) (0.8566) _ sat (10-70) dP T7t Tcn dP

                                                                                                   'X' 4384-non\secl O.wpd-04203                            10-20

Vapor internal energy and its derivatives are ev = eg + ( - wt) Cpg lyideal (10-71) (aT') aeV) pg T da (10-72) ( ae,) deg (ev - eg) dCpg ( ae, dTsat (10-73) ap T dP Cpg dP t aTy) aY) PdP Density Subcooled vapor density and its derivatives are deternined using the same method of calculation as in the case of superheated vapor, as described in Section 10-2-2-2. If the subcooled vapor density calculated with Equation 10-57 falls outside the range 0 < P < 0.9 Pt then the vapor internal energy is recalculated using Equations 10-69 through 10-73 and the density and its derivatives are recalculated: p = P / ((yg - 1) e) (10-74) fapv ep, j aTe (10-75) aATv p eV tdv p 4384-non\secl .wpd-04203 10-21

8pV = PV p dyg P ae,) 1- (10-76) ap) T P (yg-1) dP e, aP Tv Enthalpy The enthalpy of subcooled vapor is calculated using the definition of enthalpy, h = e+ P (10-77) PV where ev is calculated from Equation 10-66 or 10-71, and pv is calculated using Equation 10-57 or 10-74. Model as Coded The thermodynamic properties for subcooled vapor are calculated directly as described in this section, in subroutine THERMO. The enthalpy is calculated in subroutine FPROP. Scaling- Considerations Not applicable. Conclusions The thermodynamic properties for subcooled vapor in WCOBRAITRAC one-dimensional components are calculated in a manner consistent with calculations for superheated vapor, which are derived from thermodynamic first principles. Subcooled vapor occurs only infrequently during a LOCA transient. As such, the error introduced by WCOBRA/TRAC subcooled vapor property calculations is assumed minor and is not considered a contributor to the code uncertainty. 10-2-2-4 Subcooled Liquid Properties Model Basis Internal Energy For a liquid at a subcooled temperature T, and pressure P the liquid internal energy associated with that state is calculated starting with the internal energy of the saturated liquid state described by T and Psa(T,), which is the saturation pressure corresponding to T, 4384-non\secl O.wpd-04203 10-22

and adding an additional term which represents the change in internal energy from the state (Tf, Psai(Ti)) to the state (T,, P). That is, e,(T1 , P) = ef(T) + OI(P,T,) (10-78) The additional term 01, which represents the change in energy required to move along the isotherm at T between two different pressure values, namely Pat(TI) and P, is represented as 01 = (P'Psat(Ti)) ~a (10-79) where: rT - 255.2 1/0.223 Pa(T)=(10)t 117.8 J(10-80) The partial derivative with respect to pressure of the internal energy is ( =CROK + C 2 Psar (TI) (10-81) where: CKO = -8.33544 x 104 CK2 = -2.24745 x 10-17 Therefore the partial derivative with respect to T of the internal energy increment is calculated as ( aj -CKO - CK2[2 Psa,(T,)P -3 P,1at -)8 aTj ) (I117.8)(0.223) P(Ti) 0767 (10-82) lo, Lo,5 4384-non%secI0.wpd-04203 10-23

and the derivative of internal energy is ( f) =d-e +( 1+ (10-83) 0T, dT0~, aT,74 The saturated liquid intemal energy and its derivative with respect to temperature are detemnined over three temperature domains as was previously described (Equations 10-38 and 10-39). Density Liquid density is also calculated over three temperature domains. Defining PBAR=(10-5)P and T,, = T -273.15, density and its derivatives are as follows. For T,>525.15: p = 1.43 +100[C, +Ct2PBAR+Ct3 PBAR+PJTt,+P2 Tk2]' (10-84) (ap) = -(pt - 1.43)2 (108)[C 2 +2 C 3 PBAR + Tc(C5 +2 6 BAR) (10-85)

                                      + TIC(CI 8 +2c 9 PBAR)]

ap - (p, 1.43)2(10-3)(p, + 22 Ttc) (10-86) where:

               =  C14  + CS PBAR + C     PBAR                                                 (10-87) 4384-non\sec IO.wpd-04203                                 10-24

and 2 = C7 + CI8 PBAR + C9 PBAR (10-88) For T<521.15:

p. = wOOO[dtl + 12 PBAR + d. 3 PBAR + 3 T + Tc]-2.0 (10-89)

( J ap1

                       = -(pl+2.01)2(108)[dt2+ 2dt3 PBAR +Tk(dt 5 + 2dt6 PBAR)

(10-90)

                                     + T(di8 + 2ds PBAR)]

(ap 2 aT)P

                       = -(p1 +2.01)2(103)(P3 +204 T')                         (10-91) where:

13 = dt 4 +d5PBAR +dt6PBAR (10-92) and J34 = d7 +dt8PBAR +d. PAR (10-93) 4384-non\sec O.wpd-04203 10-25

For 521.15 T

• 525.15 a linear interpolation of the above two ranges is used. Representing the values from Equations 10-84 through 10-88 by Pta and those from Equations 10-89 through 10-93 by PtbI then pt Fb Ptb Fa Pt. (10-94) apl =_P b aptb , a aPta aPb+ aP,. (10-95) ap ~ ap ajp ap1 = Fb aP'b +Fa aP- + Pi. - Plb 5T bT, aT, (10-96)

4.0 where

Fa = (T - 521.15)/4 (10-97) Fb = 1 -Fa (10-98) The coefficients used in Equations 10-84 through 10-93 are: Cil = 2.25262 d, = 1.00213623 C12= 0.014859 dj2 = -5.632785E-05 C13 = -7.15488E-05 dj3 = -8.971304E-09 C14 = -0.0104588 dj4 = -2.28287459E-05 C15 = -1.02962E-04 dt5 = 4.76596787E- 07 CC6 = 5.09135E-07 dt6 = 5.021318E- 10 C17 = 2.59266E-05 dt7 = 4.10115658E-06 cis = 1.7241E-07 dI8 = -3.803989E- 09 C,9 = -8.98419E- 10 de9 = -1.42199752E- 12 4384-non\sec O.wpd-04203 10-26

Model as Coded The thermodynamic properties for subcooled liquid are calculated in subroutine THERMO as described in the previous paragraphs. For subcooled liquid, however, the density and its derivatives are corrected to reflect a residual void fraction. The correction is shown below. In the following, the liquid values calculated in the previous section are denoted by a tilde (). For PŽ0.4xlO6 Pa pi 1 1000) p (10-99) and af), = 1- iooo ap, (10-100) aTj p p aTj p I apt = 1I- -oooap + ap Te P ap Tr p2 (10-101) For P<0.4x 106Pa P = (0.995+6.25xlO-9P)P, (10-102) and, (ap, ap I -Y = (.9 .5x1-P (10-103) 4384-non\secl O.wpd-04203 10-27

( a = (0.995+6.25x1O-9P) ( ap +6.25x 10-9PI (10-104) Scaling Considerations Not applicable. Conclusions The TRAC-PD2 subcooled water thermodynamic property routines used in WCOBRA/TRAC for one dimensional components have been compared by Rivard and Torrey (1975) with steam table data. The agreement is good in the region for 373 K < T < 523 K and 0.4178 x 106 JlKg < e < 1.0808 x 106 JlKg. Comparison with the WATER package (Coffman and Lynn, 1966) over a wider range also showed good agreement except for very extreme cases not expected in a PWR LOCA. 10-2-2-5 Transport Properties Model Basis This section describes the WCOBRA/TRAC calculations performed to obtain the specific heat, fluid viscosity, thermal conductivity, and surface tension for one-dimensional components. The equations used for these quantities are polynomial fits to data. Specific Heat The constant pressure specific heat for liquid water is given in J/kg°K as a function of enthalpy and pressure by CP1 = {H1[H1(DO +D,,P)+(COI + C1 P)1 +B 0 1+B 1IP}' (10-105) For vapor, the constant pressure specific heat is given by: Cpv = Cl,+C2 ,Tv+ C v,I 3 + C 4 vP3 (10-106) (c 5 wTi-c 6 i (c 5 T-c 6 ar where the coefficients of Equations 10-105 and 10-106 are listed in Table 10-9. 4384-non\sec IO.wpd-04203 10-28

Liquid Viscosity Calculation of liquid viscosity is divided into three different ranges based on the liquid enthalpy. For H1 s 0.276xl 06 J/kg, liquid viscosity in N-s/rM2 Pt= KO+AIlX+A2 X2+A 3 ,X3+A 4Ix4} - Bot+BjIIt+B 2t12 +B3 erI3] (P-Po) (10-107) where: X (HI-CO)HO (10-108) and Tj = (Hl-ecOf)hO (10-109) For 0.276x106 J/kg <H,* 0.394x 106 J/kg the liquid viscosity is, Pt = [EOe+EHj+E2tH2 +E3 H] +[FO,+FlHt+F 21H +F3 H3] (-P) (10-110) and for H1 > 0.394x10 6 J/kg Pt N +DItZ+D2CZ2+D RZ3+D41Z41 (10-111) z = (HI-c.)uoo (10-112) The coefficients for the liquid viscosity equations are found in Table 10-10. Vapor Viscosity Calculation of the viscosity of vapor is divided into three different temperature ranges. The ranges and expressions used for vapor viscosity are: 4384-non\sec I O.wpd-04203 10-29

For T,< 280 K,

               = 17.08 x 10-6 + 5.927 x 10-8 (Tv-273.15)
          -8.14xl'      '(T-273.15)2                                              (10-1 13a)

For 280 K T573.15 K, V = [Blv(Tv-273. 15)+Cl] - pv[DIv-E 1v(Tv-273.15)] (10-113b) For 573.15 K < Tv < 648.15 K, Pv = Blv(Tv-273. 15)+Clv+pv[Fo-Flv(Tv -273.15) (10-114)

          +  F2V(TV-273.15)2 +F 3 ,(TV-273.15)3]
          +  PV[Gov+Glv(T,-273. 15)+G2,(Tv-273.15)2]
          +G 3 v(Tv 273.15)$J(4Ov AivPv +A2Pv) and for Tv         648.15 K, Pv = B              15)+Clv+Pvov+AlvPv+A2vpv 1V(Tv0-273.                                    (10-1 15)

The coefficients for Equations 10- 113 through 10- 115 are listed in Table I10-I1. 4384-non\seclO.wpd-04203 10-30

Liquid Thermal Conductivity The liquid thermal conductivity is given W/m-K by (10-116) where: HI (10-117) At4 Vapor Thermal Conductivity If 280.0 K TV

             =  Xl +PtX2+                                                          (10-1 18)

(27315)4.2 where: x = AvO+Avl(T -273.15)+A2(T-273.15)2 +Av3(Tv -273.15) 3 (10-119) and X2 = BVo+BI(Tv-273.15) +B2(Tv-273.15)2 (10-120) The coefficients used in Equations 10-116 through 10-120 are listed in Table 10-12. If Tv < 280.0 K, the vapor conductivity is kv = 0.0228 (10-121) 4384-non\sec O.wpd-04203 10-31

Surface Tension The surface tension is calculated in N/m as, C( = (a 2 + 1 + a3 a4 + a504 (10-122) where: 0 = 647.3 - sat (10-123) The coefficients for Equation 10-122 are given in Table 10-13. Model as Coded Subroutine FPROP is used to obtain transport properties for liquid and vapor water. The input variables for this subroutine are the saturation temperature corresponding to the total pressure, the internal energies, densities, and temperatures of the liquid and vapor phases and the total pressure. The output transport variables include the constant pressure specific heats, viscosities, and thermal conductivities of the liquid and gas phases, and the surface tension of the liquid. The transport property calls are function calls within the FPROP subroutine. Function CPLL calculates the constant pressure specific heat of the liquid, while function CPVV 1 determines the value of the constant pressure specific heat of the vapor. Function THCL evaluates the liquid thermal conductivity, and function THCV calculates the steam thermal conductivity. Similarly, functions VISCL and VISCV determine viscosity values. Finally, function SIGMA calculates the surface tension. The equations shown are coded directly. Sample curves of liquid and vapor specific heat, viscosity, thermal conductivity, and the surface tension calculated by these routines along the saturation line are shown in Figures 10-22 through 10-28. 43 84-non\sec IO.wpd-04203 10-32

In some instances, upper and lower limits are maintained on the calculated values of the transport properties. These limits are summarized as follows: Specific Heat The maximum permitted value for the liquid specific heat is C =4.Ox 104. If the calculation of C,, by Equation 10-105 performed by function CPLL yields a value greater than this, Cp, is reset to 4.0x10 4 . No limits are placed on the calculation of the vapor specific heat. Viscosity The minimum permitted value of vapor viscosity is p,= 10-7. If the calculation of p, by Equations 10-113 through 10-115 yields a value less than this in function VISCV, p, is reset to 10-7. No limits are imposed on the liquid phase viscosity. Thermal Conductivity The minimum permitted value of the liquid thermal conductivity is k,=0.09. If, in function THCL, Equation 10-116 yields a value lower than this, k is reset to 0.09. The minimum permitted value for vapor thermal conductivity is k=10-4 . If Equation 10-118 in function THCV calculates a value less than 10 4, kv is reset to 104. Surface Tension If Tsa, > 647.3, the surface tension is set to a = 0.0. Scalin2 Considerations Not applicable. Conclusions In NUREG/CR-2054, it was reported that the thermodynamic and transport property fits used in TRAC-PD2 were compared by Rivard and Torrey (1975) with steam table data over a wide range of parameters. The agreement is satisfactory in the saturation region and in the superheated steam region for 1.0 x 105 Pa < P < 100.0 x 105 Pa and 423.0 K < T, < 823.0 K. The agreement also is good in the subcooled water region for 373.0 K < T < 523.0 K and 0.417 8 x 10 6J/kg < e < 1.080 8 x 106 J/kg. Further verification was performed by comparing the TRAC-PD2 polynomial fits with the WATER package (Coffman and Lynn, 1966) over a wider range of nonequilibrium (99 K of both superheat and subcooling) for a pressure variation of 1.0 x 105 Pa to 2.0 x 107 Pa. The 4384-non\secO.wpd-04203 10-33

comparisons showed good agreement for both the thermodynamic and transport properties throughout the saturation and nonequilibrium regions except for very extreme cases, which are not expected in a PWR LOCA. The WCOBRA/TRAC property package for one-dimensional components is identical to the TRAC-PD2 package. Therefore, for most WCOBRA/TRAC applications, the thermodynamic and transport property routines will provide realistic values over a wide range. The simplified polynomial fits provide an efficient and low-cost method compared to other approaches such as steam table interpolation. 10-3 Thermophysical Properties of Air 10-3-1 Vessel Component Model Basis WCOBRA/TRAC can perform calculations for conditions in which there is air in the vessel component. This section describes the thermodynamic properties which are defined for air in the WCOBRA/TRAC vessel component. Enthalpv The enthalpy of air is calculated as Hair =prA air, rf/iT ref (10-124) where the reference values are Tref = 40.0 0 F, Hrej = 188.49 BtuAlbm, and Cpref = 0.249 Btu/ Ibm-°F. Densitv The density of air is calculated from the ideal gas law with the gas constant for air assumed to be Rair 0.37042 psi, °R (Iblft3) . Thus, the density of air is given by p Pair= R(T +4596) (10-125) 4384-non\secl O.wpd-04203 10-34

Gas Temperature The air temperature is estimated from the enthalpy using the inverse of Equation 10-124. Specific Heat The specific heat for air in BTU/lbm-°F is determined in two different temperature ranges. If T 600K, C 0.244388 +ATj +AT 3

                = p~~~~~~i            2 air +AT 3 air                                   (10-126) and if Tair > 600K,
                            +B~~T ir+BT 2 Cp = 0.208831 p                  2Tair +3Tir
                                     ~~~I     3ai                                     (10-127) where the coefficients Ai and Bi are listed in Table 10-14, and Tair is in degrees K.

Model as Coded The equations used to calculate the thermodynamic properties of air, Equations 10-124 through 10-127, are coded as shown without modification. No upper or lower limits are imposed on the values calculated. Calculations are performed in Subroutines HGAS and TGAS. Scaling Considerations Not Applicable Conclusions The WCOBRA/TRAC vessel component can perform calculations to estimate the thermodynamic properties of air. This option, however, is not used in a LOCA analysis. 10-3-2 One-Dimensional Components Model Basis This section describes the calculation of thermodynamic and transport properties in WCOBRA/TRAC one-dimensional components for air. 4384-non\secl O.wpd-04203 10-35

Internal Energy The intemal energy and its derivatives for air are given by eail = Ci,Tir (10-128) (T aair air) Cvair (10-129) and aeair (P Tair

                           = 0.0                                                       (10-130)

The constant volume specific heat (Cvir) is L Cvair = 714.9 Jlkg-K (10-131) Density The density and its derivatives are based on the Ideal Gas Law and are given by p Pai = T (10-132) Rair air ( aPair) i (10-133) aair, Tir Rair azir C aPair aTatirp

                       =    Ra,r,ar aPair) ap I

T./ (10-134) 4384-non\seclO.wpd-04203 10-36

where Rai = 287.12 Jlkg-K (10-135) Enthalpv The enthalpy of air is calculated using the definition of enthalpy: H.air air Pr (10-136) Pair where eair is deterrnined by Equation 10-128, and p,,i, is given by Equation 10-132. Viscosity Two different temperature ranges are used to calculate the viscosity of air. If T^, <502.15 K, pair = a+a 1 (T -273. 15)+aa (Tair273.15)2 (10-137) and if Tair>502.15 K, Pair abo+abl(T.s-273.15) +ab2 (Tair-273.15)2 (10-138) where the coefficients aai and abi are listed in Table 10-15. Thermal Conductivity The thermal conductivity of air is assumed to be constant, ki = 0.0228 Wlm-K (10-139) 4384-non\secl O.wpd-04203 10-37

Model as Coded The internal energy and its derivatives and the density and its derivatives for air are calculated in subroutine THERMO. Subroutine FPROP calculates the enthalpy. The transport properties viscosity and thermal conductivity are determined in subroutines VISCV and THCV, respectively. Scaling! Considerations Not applicable. Conclusions The WCOBRA/TRAC one-dimensional components calculate thermodynamic properties for air assuming it behaves as an ideal gas. The transport properties are based on polynomial fits to data. The correlations approximately calculate properties for air at low temperatures. 10-4 Thermal Properties of Nuclear Fuel Rod Materials A typical nuclear fuel rod is composed of uranium-dioxide fuel pellets and a zirconium based clad material. The gap between the fuel pellets and the clad is filled with the initial backfill gas and fission gas. As part of the WCOBRAITRAC default nuclear fuel rod model, the material properties of uranium-dioxide, Zircaloy-4, ZIRLOm, and of gas mixtures are included. This section describes the calculation of the thermal properties for these fuel rod materials. 10-4-1 Uranium Dioxide Model Basis The material properties of uranium dioxide are based on MATPRO-9 (MacDonald et al., 1976) and on MATPRO-1 1, Rev. 1 (Hagrman, Reymann, and Mason, 1980) calculations. Density The (cold) density for uranium-dioxide is assumed to be Puo = 684.86 fD (10-140) where fD is the fraction of theoretical density and is input by the user. The density puo2 has units of Ibm/ft3 . 4384-non\seclO.wpd.04203 10-38

Thermal Conductivity The UO2 thermal conductivity is computed from the MATPRO-9 correlation instead of the more complex version in MATPRO-1 1 to reduce computer time. Both correlations have the same error band (0.2 W/m-°K) and give very nearly the same conductivity over the expected operating range of 500-3000° K. The thermal conductivity in BtuJhr-ft-°F is determined from kUo2 = [max (0.0191, ( 464)) +1.216x104exp(I.867x1l-3T]C (10-141) where T, is the temperature in Celsius and C = (0.5779)100[1.0 -(1.0 -fD)]/(l.O-.0 O 5 P) (10-142) and 3= 2.58-(5.8x10-4 )T (10-143) Specific Heat The specific heat in Btu/lbm-°F for uranium dioxide is given by K 02exp(O/TK) Fom K3ED CpU = (2.388x10 4 ) 2e12 IT +K2 TK+ 2 3 Jexp(-EDRTK) (10-144) where TK is the temperature in K and 0 = Einstein temperature(535.285°K) R = 8.3143 (Jlmol-°K) K, = 296.7(Jlkg-°K) 2 K 2 = 2.43x10-2(J/kg-'K ) K3 = 8.745x10 7 (J/kg) ED= 1.577xlO5 (J/mol) FOM = oxygenlmetal ratio(2.0) 4384-non\seclO.wpd-04203 10-39

Model as Coded The equations representing the density, thermal conductivity and specific heat for uranium dioxide are coded into WCOBRA,TRAC as described by Equations 10-140 through 10-144 without modification. Calculations for uranium dioxide density are performed in Subroutine SETUP, those for thermal conductivity in subroutines SSTEMP and TEMP, and those for specific heat in Subroutines TEMP and MOVE. Values of conductivity and specific heat versus temperature are shown in Figures 10-29 and 10-30. Scalin2 Considerations Not applicable. Conclusions The WCOBRA/TRAC correlations for U0 2 density, specific heat and thermal conductivity are based on MATPRO-9 and MATPRO-1 1. The models and correlations for these properties were used in simulations of NRU and LOFT. Therefore, the uncertainty and reliability of these models is accounted for in the overall code bias and uncertainty. 10-4-2 Zircaloy-4 Model Basis The material properties of Zircaloy-4 are based on MATPRO-9 and MATPRO-1 1 calculations. Density The (cold) density of Zircaloy-4 clad material is assumed to be p=409.0 Ibm/ft 3 . Thermal Conductivity The thermal conductivity in Btulhr-ft-°F for Zircaloy-4 clad is given by kz = 0.5779[7.51 +0.020 9 TK-(1.45x 10-5)T2+ (7.67xlo-90TK3] (10-145) where TK is temperature in Kelvin. Specific Heat WCOBRAITRAC calculates the specific heat for Zircaloy-4 by linearly interpolating between values from a built-in table. Table 10-1 6 lists the values used to determine the specific heat of Zircaloy4. 4384-nonsecl O.wpd-04203 10-40

Model as Coded The equations for the density, thermal conductivity and specific heat of Zircaloy4 are coded into WCOBRA/TRAC as described above without modification. Density is calculated in Subroutine SETUP and HEAT, conductivity in Subroutines STEMP, TEMP, and HEAT and specific heat in Subroutines TEMP, HEAT, and MOVE. Curves of conductivity and specific heat versus temperature are shown in Figures 10-31 and 10-32. Scaling Considerations Not applicable. Conclusions The WCOBRAITRAC correlations for the density, therrnal conductivity, and specific heat of Zircaloy-4 are based on MATPRO-9 and MATPRO-1 1. These property relations were used in simulations of NRU and LOFT. 10-4-3 ZIRLOTM Model Basis The ZIRLO' alloy developed by Westinghouse represents a modification to Zircaloy-4 which was achieved by reducing the tin and iron content, eliminating the chromium, and adding a nominal one percent niobium. Table 10-17 shows a comparison of the two alloys. Since tin is an alpha phase stabilizer and niobium is a beta phase stabilizer, the reduction in tin and the addition of niobium result in reductions in the temperatures at which the ZIRLO' alloy undergoes the alpha to beta phase change, relative to Zircaloy-4. Measurements performed by Westinghouse show that the ZIRLO alloy starts the transformation at 1023 °K and ends at 1213 OK. Since the ZIRLOTm and Zircaloy-4 alloys are both about 98 percent zirconium, it should not be expected that the material properties are significantly different, except to the extent that they are affected by the differences in the phase change temperatures. Density, thermal expansion, thermal conductivity, and specific heat of both alloys have been measured by the Properties Research Laboratory using samples cut from Westinghouse production tubing (Taylor, Groot, and Larimore, 1989). Evaluation of the test results indicated that the materials are sufficiently similar that the Zircaloy-4 material properties can be used for the ZIRLO6 alloy, with the exception of the specific heat (Davidson and Nuhfer, 1990). The specific heat of the ZIRLOM alloy is based on an adjustment to Table 10-16, which considers the difference in phase change temperatures. 4384-non\sec .wpd-04203 10-41

Density The (cold) density of the ZIRLO Zircaloy-4 (409.0 Ibm/ft 3 ). cladding material is taken to be identical to that of L." Thermal Conductivity The thermal conductivity of the ZIRLO' cladding material is taken to be identical to that of Zircaloy-4, given by Equation 10-145. Specific Heat The specific heat shown in Table 10-16 for Zircaloy-4 includes both the true specific heat and the alpha to beta phase heat of transformation. The specific heat for the ZIRLO T ' cladding material was obtained by adjusting Table 10-16 to account for the difference in phase change temperatures, assuming both the true specific heat and the heat of transformation are the same for the two alloys. The true specific heat is taken to be equal to the total specific heat in Table 10-16 for T 1090°K, 0.085 BtuAlbm-°F for T 1213°K, and[ (10-146)

                                                                                       ]    (10-147) where: [
                                                                                       ]" (10-148)

WCOBRAJTRAC calculates the specific heat for the ZIRLO' cladding material using the resulting total specific heat values, shown in Table 10-18. Model as Coded The density, thermal conductivity, and specific heat of the ZIRLOQ cladding material are coded into WCOBRA/TRAC as described above, without modification. Figure 10-33 shows a comparison of specific heat for ZIRLO with that of Zircaloy-4. 4384-non\sec O.wpd-04203 10-42

Scaling Considerations Not applicable. Conclusions Comparisons of the material properties for the ZIRLO and Zircaloy-4 cladding materials have shown that the Zircaloy-4 relations for density and thermal conductivity can also be applied to the ZIRLOTI alloy. The difference in the phase change temperatures of the two alloys requires that different specific heat correlations be used. The specific heat correlation for the ZIRLO alloy is based on an adjustment to the Zircaloy-4 correlation, which accounts for the different phase change temperature range. This correlation will be used for analyses of nuclear reactors which utilize the ZIRLO cladding material. 10-4-4 Fuel Rod Gas Mixtures Model Basis For the gas mixture in the fuel-clad gap, only the thermal conductivity is calculated. The fill gas in the WCOBRA/TRAC fuel rod model assumes that the gas is a mixture composed of helium, xenon, argon, krypton, hydrogen, and nitrogen. The thermal conductivity of the gas mixture as a function of temperature is determined, as described in MATPRO- 11 Rev. 1 (Hagrman, Reymann, and Mason, 1980), from the relation N k kgas =Y- i=kI N nI_ (10-149) j=1in j*i where N = number of component gases, and where (+. -M.1M-0.142.) Tij = siD +2.41M,.\ (10-150) 4384-non\sec I O.wpd-04203 10-43

and ( 2 1 (10-151) 23/ 1+ i M where: Mi = molecular weight of gas species i n = mole fraction of gas species i ki = thermal conductivity of gas species I The thermal conductivities of the six component gases are evaluated in Btu/hr-ft-°F as a function of temperature from the following relations: Gas k(Btu/hr-ft-°F) Helium (1.314x1O-)T°6 6 8 (10-152) Argon (1.31x1O-3)T,j 01 (10-153) Krypton (1.588x1 9 23 3 ' Ts s)T (10-154) Xenon (1.395x1 72 as)T° (10-155) Hydrogen (5.834x10-4 )T°gas8213 (10-156) 4384-non\sec I O.wpd-04203 10-44

Nitrogen (7.35xl 0-5) T 46 (- (10-157) where Tgas = gas temperature (R). Model as Coded Equations 10-149 through 10-151 for gap gas thermal conductivity are coded in WCOBRATRAC as described without modification in subroutine GTHCON. Scaling Consideration Not applicable. Conclusions Thermal conductivity for the gas mixture in the fuel-clad gap is calculated using the equations in MATPRO-1 1 Rev. 1 (Hagrman, Reymann, and Mason, 1980). 10-5 Thermal Properties of Structural Materials 10-5-1 Vessel Component Structural Material Properties Model Basis The density, specific heat, and thermal conductivity for structural materials within the vessel are specified by the user for a range of temperatures. Values for each material are obtained from standard references for thermal properties such as Touloukian (1967). When available, material properties provided by the material supplier are used. Model as Coded Values for the material specific heat and thermal conductivity are linearly interpolated with temperature. A warning message is printed if the temperature is outside of the range supplied by the user. Scaling Considerations Not applicable. Conclusion Material thermal properties are supplied by the user. This permits the representation of the material properties by the actual measured values and minimizes uncertainty. 4384-non\secl O.wpd-04203 1045

10-5-2 One-Dimensional Component Structural Material Properties Model Basis A library of temperature-dependent material properties is incorporated in WCOBRAiTRAC for the one-dimensional components. There are five sets of material properties that make up the library. Each set supplies values for the density, thermal conductivity, specific heat, and spectral emissivity for use in heat transfer calculations. The material sets are for Types 304, 316, and 347 Stainless Steel, Medium Carbon Steel, and Inconel 600. In the following expressions, p density (k3) C = specific heat ( 1 k = thermal conductivity W m-K TK = temperature(K) TF = temperature (F) Stainless Steel. Type 304 The density is given by p(TF) = 8 054 . 6 5-0. 2 595TF (10-158) Specific heat is given by CP(TF) = 426.17+0.43816TF-(6.3759x104 )TF2 +(4.4803x 10 7 )TF3 -(1.0729x 10 ')TF 4 (10-159) Thermal conductivity is calculated by k(TF) = 14 . 7 9 +0.0071 4 TF (10-160) 43 84-non\sec IO.wpd-04203 10-46

Stainless Steel, Type 316 Density is given by p(TK) = 8084.0-0.4209TK-(3.894x10- 5 )TK2 (10-161) Specific heat is given by Equation 10-159 and thermal conductivity is given by k(TK) = 9 . 24 8 +0.01571TK (10-162) Stainless Steel, Tve 347 The density is assumed constant at p = 7913 kg (10-163) m The specific heat is given by CP(TF) = 502.416+0. 0 9 84 (TF- 240) (10-164) and the thermal conductivity is k(TF) = 14.1926+(7.269x10-3)TF (10-165) Carbon Steel The density for carbon steel is assumed constant: p = 7855.23 kg (10-166) m3 The specific heat is given by CP(TF) = 400.48+0. 4 5 8 2 TF-( 6 . 5 5 3 2 x 10 4 )TF2 +(5.3706xlO 7 )TF3 (10-167) 4384-non\secI O.wpd-04203 10-47

and the thernal conductivity is given by k(TF) = 4 8 .4 3 -0.011 3 6 6 TF (10-168) Inconel 600 The density for Inconel 600 is assumed constant, p = 8409.45 kg (10-169) m3 The specific heat is given by CP(TF) = 4184.[0.1014456+(4.378952x 10 5)TF-(2.046138x 10 8)TF2+ (10-170) (3.41811 1x0'I')TF3 -(2.0603 18x10 3)TF4+(3.682836x 10-16)T 5 (2.458648x10 ' 9)TF 6 +(5.59757 IX 1023)TF7] and thermal conductivity is given by k(TF) = 1.730[8.011332+(4.64371 9X1O- 3)TF+( 1.872857x 10-6 )TF2 - (10-171) (3.914512x10- 9)TF3 +(3.475513x 1012)TF4-(9.936696x10I6)TF5] Model as Coded The correlations described by Equations 10-158 through 10-171 are programmed as shown without modification in subroutine MSTRCT. Curves of specific heat and thermal conductivity as functions of temperature calculated with this subroutine are shown in Figures 10-34 through 10-42. 4384-non\sec IO.wpd-04203 10-48

Scalin2 Considerations Not applicable. Conclusions The WCOBRAITRAC code uses built-in correlations to calculate the thermal properties of common structural materials modeled by one dimensional components. Comparisons to data show that these correlations provide a good estimate of the properties at low temperature. Since the one-dimensional components generally remain at low temperature during a LOCA transient, use of these correlations introduces only a small uncertainty into the transient calculation. 10-6 Conclusions WCOBRAITRAC routines provide appropriate means for calculation of thermodynamic and transport properties of liquid water, steam, and air for the vessel component and for one-dimensional components. Routines to calculate properties of fuel rod materials, i.e., fuel, cladding, and gap gas, are also included. Properties of structural materials in the vessel component are interpolated from user-provided tables. For one-dimensional components, routines to calculate properties of common structural materials are included. The routines generally calculate properties in the form of equations, for example as functions of temperature and pressure, or by linear interpolation in built-in tables. These property calculations have been compared with standard references and found to agree satisfactorily over the range of conditions expected for PWR LOCA calculations. No scaling uncertainty is required for the use of these models in reactor analysis. 10-7 References ASME Steam Tables, 1968, American Society of Mechanical Engineers, 2nd Edition. ASME Steam Tables, 1983, Thermodynamic and Transport Properties of Steam, 5th ed., The American Society of Mechanical Engineers, New York. Coffman, W. A. and Lynn, L. L., 1966, "WATER: A Large Range Thermodynamic and Transport Water Property FORTRAN-IV Computer Program," Bettis Atomic Power Laboratory Report WAPD-TM-568. 4384-non\sec O.wpd-04203 10-49

Davidson, S. L. and Nuhfer, D. L. (Eds.), 1990, "VANTAGE+ Fuel Assembly Reference Core Report," WCAP-12610, Westinghouse Electric Corporation, Pittsburgh, Pennsylvania. [PROPRIETARY). Haar, L., Gallagher, J. S., and Kell, G. S., 1984, NBS/NRC Steam Tables, Hemisphere Publishing Corporation, New York. Hagrman, D. L., Reymann, G. A., and Mason, R. E., 1980, "MATPRO - Version 11 (Revision 1): A Handbook of Materials Properties for Use in the Analysis of Light Water Reactor Fuel Rod Behavior," USNRC Report NUREG/CR-0497, TREE-1280, Revision 1. Keenan, J. H. and Keyes, F. G., 1936, Thermodynamic Properties of Steam, John Wiley & Sons, New York. MacDonald, P. E., et al., 1976, "MATPRO - Version 9: A Handbook of Materials Properties for Use in the Analysis of Light Water Reactor Fuel Rod Behavior," Idaho National Engineering Laboratories, TREE-NUREG-1005. McClintock, R. B. and Silvestri, G. J., 1936, Formulations and Iterative Procedures for the Calculation of Properties of Steam, American Society of Mechanical Engineers. McFadden, J. H., et al., 1980, "RETRAN-02 A Program for Transient Thermal-Hydraulic Analysis of Complex Fluid Flow Systems, Volume 1: Equations and Numbers," Report NP-850, Electric Power Research Institute, Palo Alto, California. Rivard, W. C. and Torrey, M. D., 1975, "Numerical Calculation of Flashing from Long Pipes Using a Two-Field Model," Los Alamos Scientific Laboratory Report LA-6104-MS. Taylor, R. E., Groot, H., and Larimore, J., 1989, "Thermophysical Properties of ZR-4 and ZIRLO," PRL-820, Properties Research Laboratory, West Lafayette, Indiana. [PROPRIETARY]. Touloukian, Y. S., 1967, Thermophysical Properties of High Temperature Materials, Thermophysical Properties Research Center, Purdue University, The Macmillan Co., New York. 4384-non\secl O.wpd-04203 10-50

Table 10-1 Constants for Saturated Liquid Enthalpy Pressure: A,, 0.1 s P < 898.7 898.7 s P <2529.9 2529.9 3208 I 0.6970887859E+02 0.8408618802E+06 0.9060030436E+03 2 0.3337529994E+02 0.3637413208E+06 -0.1426813520E+02 3 0.2318240735E+01 -0.4634506669E+06 0.1522233257E+01 4 0.1840599513E+00 0.1 130306339E+06 -0.6973992961E+00 5 -0.5245502294E-02 -0.4350217298E+03 0.1743091663E+00 6 0.2878007027E-02 -0.3898988188E+04 -0.2319717696E-01 7 0.1753652324E-02 0.6697399434E+03 0.1694019149E-02 8 -0.4334859620E-03 -0.4730726377E+02 -0.645477171OE-04 9 0.3325699282E-04 0.1265125057E+01 0.1003003098E-05 Table 10-2 Constants for Saturated Vapor Enthalpy Pressure: B P < 1467.6 1467.6 s P < 258.6.0 2586.0 s 3208.0 1 0.1105836875E+04 0.5918671729E+06 0.9059978254E+03 2 0. 1436943768E+02 -0.2559433320E+06 0.5561957539E+01 3 0.8018288621E+00 0.3032474387E+05 0.3434189609E+01 4 0.1617232913E-01 0.4109051958E+01 -0.6406390628E+00 5 -0.1501147505E-02 0.3475066877E+00 0.5918579484E-01 6 -0.1237675562E-04 -0.3026047262E+00 -0.2725378570E-02 7 0.3004773304E-05 -0.1022018012E+02 0.5006336938E-04 8 -0.2062390734E-06 0.1591215116E+01 0.0 9 0.0 -0.6768383759E-01 0.0 43 84-non\sec I Oa.wpd-04203 10-51

WESTINGHOUSE PROPRIETARY CLASS 2 Table 10-3 Vessel Component Saturated Water Thermal Properties P. r,,, P Ps H, H, /* kg k, Cp, Cp, _ (psia) (°F) (lbmlft3 ) (Ibm/fe) (Btu/lbm) (Btu/lbm) (Ibm/hr/ft) (Ibm/hr/ft) (Btu/hr/ft/F) (Btu/hr/ft/F) (Btu/lbm/F) (Btu/lbm/F (Ibf/ft) 0.1 41.97 62.42 0.000 10.00 1079.83 3.61570 0.02262 0.33023 0.01002 1.00440 0.44426 0.00513 0.2 51.93 62.40 0.001 20.00 1084.18 3.06850 0.02295 0.33627 0.01022 1.00320 0.44477 0.00508 0.3 61.91 62.36 0.001 30.00 1088.55 2.64160 0.02331 0.34218 0.01041 1.00140 0.44542 0.00502 0.4 71.90 62.29 0.001 40.00 1092.92 2.30190 0.02368 0.34791 0.01062 0.99975 0.44623 0.00496 0.5 81.91 62.20 0.002 50.00 1097.28 2.02710 0.02406 0.35338 0.01083 0.99851 0.44723 0.00491 0.7 91.93 62.09 0.002 60.00 1101.62 1.80170 0.02445 0.35848 0.01105 0.99776 0.44844 0.00484 1.0 101.95 61.97 0.003 70.00 1105.94 1.61440 0.02485 0.36334 0.01128 0.99743 0.44988 0.00478 1.4 111.98 61.83 0.004 80.00 1110.23 1.45700 0.02526 0.36765 0.01152 0.99745 0.45157 0.00472 1.8 122.00 61.68 0.005 90.00 1114.49 1.32340 0.02568 0.37183 0.01177 0.99774 0.45353 0.00465 2.3 132.02 61.52 0.007 100.00 1118.70 1.20900 0.02611 0.37530 0.01203 0.99823 0.45577 0.00459 3.0 142.04 61.34 0.009 110.00 1122.86 1.11020 0.02654 0.37863 0.01230 0.99888 0.45832 0.00452 3.9 152.04 61.15 0.011 120.00 1126.97 1.02440 0.02698 0.38146 0.01258 0.99965 0.46117 0.00445 5.0 162.04 60.95 0.014 130.00 1131.04 0.94915 0.02742 0.38403 0.01287 1.00050 0.46435 0.00438 6.3 172.02 60.75 0.017 140.00 1135.03 0.88297 0.02787 0.38624 0.01318 1.00150 0.46786 0.00432 7.9 182.01 60.53 0.021 150.00 1138.98 0.82425 0.02832 0.38814 0.01349 1.00270 0.47172 0.00424 9.7 191.96 60.31 0.025 160.00 1142.85 0.77208 0.02877 0.38984 0.01381 1.00390 0.47591 0.00417 12.0 201.92 60.07 0.031 170.00 1146.66 0.72533 0.02923 0.39115 0.01415 1.00530 0.48047 0.00410 14.7 211.84 59.83 0037 180.00 1150.39 0.68345 0.02969 0.39236 0.01449 1.00690 0.48538 0.00403 439 non\seclOa.wpd.04203 10O-52

( WESTINGHOUSE PROPRIETARY CLASS 2 Table 10-3 (Cont'd) Vessel Component Saturated Water Thernal Properties P T, p pg H He f 'U kf k, Cpf Cp, a (psia) (OF) (bni/f) (Ibm/fe) (Btu/lbm) (Btullbm) Obmlhr/ft) (Ibmhrlft) (Btu/hr/ftlF) (Btulhr/ftll) (BtuiAbn/F) (Btu/Ibm/F (Ibf/ft) 17.8 221.78 59.58 0.045 190.00 1154.05 0.64561 0.03015 0.39320 0.01486 1.00860 0.49067 0.00396 21.4 231.66 59.32 0.053 200.00 1157.62 0.61149 0.03061 0.39397 0.01523 1.01050 0.49633 0.00388 25.7 241.55 59.05 0.063 210.00 1161.12 0.58043 0.03107 0.39444 0.01561 1.01260 0.50239 0.00381 30.6 251.39 58.78 0.074 220.00 1164.50 0.55228 0.03153 0.39481 0.01601 1.01490 0.50882 0.00373 36.2 261.22 58.50 0.087 230.00 1167.79 0.52655 0.03199 0.39496 0.01642 1.01740 0.51569 0.00366 42.5 271.02 58.21 0.101 240.00 1170.98 0.50302 0.03245 0.39498 0.01684 1.02010 0.52299 0.00358 49.8 280.80 57.92 0.117 250.00 1174.05 0.48145 0.03291 0.39485 0.01727 1.02300 0.53075 0.00351 58.0 290.54 57.61 0.135 260.00 1177.01 0.46163 0.03337 0.39456 0.01772 1.02610 0.53899 0.00343 67.2 300.26 57.30 0.155 270.00 1179.84 0.44339 0.03383 0.39418 0.01817 1.02940 0.54775 0.00335 77.6 309.93 56.99 0.177 280.00 1182.54 0.42656 0.03429 0.39358 0.01864 1.03290 0.55706 0.00327 89.1 31958 56.66 0.202 290.00 1185.10 0.41101 0.03474 0.39293 0.01912 1.03670 0.56696 0.00320 101.8 329.19 56.34 0.229 300.00 1187.53 0.39661 0.03520 0.39205 0.01962 1.04060 0.57748 0.00312 116.0 338.76 56.00 0.260 310.00 1189.82 0.38325 0.03565 0.39113 0.02012 1.04470 0.58869 0.00304 131.6 348.28 55.66 0.293 320.00 1191.95 0.37083 0.03610 0.39000 0.02064 1.04910 0.60063 0.00296 148.6 357.77 55.31 0.329 330.00 1193.94 0.35927 0.03655 0.38882 0.02116 1.05380 0.61336 0.00288 167.4 367.21 54.95 0.368 340.00 1195.77 0.34849 0.03699 0.38743 0.02170 1.05870 0.62693 0.00280 187.8 376.61 54.59 0.411 350.00 1197.44 0.33842 0.03744 0.38597 0.02225 1.06390 0.64141 0.00272 4384-non\seclOa.wpd-04203 10-53

WESTINGHOUSE PROPRIETARY CLASS 2 Table 10-3 (Cont'd) Vessel Component Saturated Water Thermal Properties

    -. 4        T..,         P1       pg      Hr           H,         p1            p,            kr          k,           C,f          C1,       a (psia)       (OF)      (ibm/fe) (IbmIW)  (Btu/lbm)    (Btu/lbm) (bmnlhr/ft)    Obm/hr/ft) (Btulhr/ltF) (Btu/hr/ftlF) (Btu/Ibin/F) (Btu/Tbm/F (Ibflft) 210.0       385.96       54.22     0.458  360.00       1198.96    0.32898       0.03788      0.38435      0.02281       1.06940     0.65687  0.00264 234.0       395.26       53.85     0.508  370.00       1200.30    0.32014       0.03832      0.38265      0.02338       1.07530     0.67338  0.00256 260.0       404.50       53.47     0.563  380.00       1201.48    0.31182       0.03876      0.38078      0.02396       1.08150     0.69101  0.00248 288.0       413.69       53.08     0.622  390.00       1202.49    0.30399       0.03920      0.37881      0.02455       1.08820     0.70984  0.00240 318.1       422.83       52.69     0.686  400.00       1203.32    0.29660       0.03964      0.37667      0.02516       1.09540     0.72995  0.00232 350.4       431.90       52.29     0.755  410.00       1203.97    0.28961       0.04008      0.37441      0.02578       1.10300     0.75144  0.00224 384.9       440.91       51.88     0.828  420.00       1204.44    0.28299       0.04052      0.37199      0.02642       1.11130     0.77439  0.00216 421.6       449.86       51.47     0.907  430.00       1204.71    0.27670       0.04095      0.36946      0.02707       1.12010     0.79891  0.00208 460.7       458.73       51.05     0.992  440.00       1204.79    0.27072       0.04139      0.36679      0.02773       1.12970     0.82510  0.00200 502.1       467.53       50.62     1.082  450.00       1204.67    0.26501       0.04183      0.36401      0.02841       1.13990     0.85307  0.00192 546.0       476.26       50.18     1.178  460.00       1204.34    0.25954       0.04227      0.36106      0.02912       1.15100     0.88295  0.00185 592.2       484.91       49.74     1.281  470.00       1203.79    0.25431       0.04271      0.35800      0.02984       1.16290     0.91488  0.00177 641.2       493.51       49.29     1.391  480.00       1203.02    0.24926       0.04315      0.35472      0.03059       1.17590     0.94916  0.00169 692.1       501.94       48.84     1.507  490.00       1202.04    0.24444       0.04359      0.35138      0.03136       1.18980     0.98549  0.00161 745.9       510.35       48.37     1.631  500.00       1200.81    0.23976       0.04404      0.34782      0.03218       1.20490     1.02460  0.00154 802.0       518.65       47.90     1.762  510.00       1199.35    0.23524       0.04450      0.34426      0.03300       1.22130     1.06650  0.00146 860.5       526.84       47.42     1.901  520.00       1197.64    0.23086       0.04495      0.34043      0.03390       1.23890     1.11140  0.00139 43PA    on\scc1Oa.wpd-04203                                               .-   54

( ( WESTINGHOUSE PROPRIETARY CLASS 2 Table 10-3 (Cont'd) Vessel Component Saturated Water Thermal Properties P.. T. pH, H I, P p k ks Cp, CPS a (psia) (OF) (lbm/fte) (Ibmft3) (Btu/lbm) (Btu/ibm) (Ibm/hr/ft) (Ibm/hr/ft) (Btu/hr/ft/F) (Btulhr/ftlF) (Btu/lbm/F) (BtuIbm/F (Ibf/ft) 921.3 534.91 46.93 2.048 530.00 1195.69 0.22662 0.04541 0.33660 0.03480 1.25800 1.15940 0.00132 984.4 542.86 46.44 2.204 540.00 1193.50 0.22250 0.04588 0.33255 0.03581 1.27870 1.21100 0.00124 1050.0 550.72 45.93 2.368 550.00 1191.03 0.21846 0.04636 0.32846 0.03684 1.30120 1.26690 0.00117 1117.8 558.47 45.42 2.543 560.00 1188.31 0.21452 0.04685 0.32427 0.03795 1.32560 1.32710 0.00111 1187.8 566.10 44.89 2.727 570.00 1185.33 0.21067 0.04734 0.31999 0.03915 1.35220 1.39240 0.00104 1259.9 573.61 44.36 2.922 580.00 1182.08 0.20689 0.04785 0.31571 0.04038 1.38120 1.46330 0.00097 1334.0 580.98 43.82 3.127 590.00 1178.57 0.20318 0.04837 0.31134 0.04174 1.41290 1.54050 0.00091 1410.0 588.22 43.27 3.344 600.00 1174.80 0.19953 0.04891 0.30694 0.04324 1.44770 1.62510 0.00084 1487.8 595.33 42.71 3.574 610.00 1170.75 0.19594 0.04946 0.30255 0.04486 1.48590 1.71800 0.00078 1567.2 602.29 42.14 3.816 620.00 1166.42 0.19239 0.05004 0.29817 0.04663 1.52810 1.82060 0.00073 1648.2 609.11 41.56 4.072 630.00 1161.76 0.18889 0.05063 0.29382 0.04854 1.57490 1.93450 0.00067 1730.4 615.77 40.96 4.343 640.00 1156.76 0.18543 0.05125 0.28954 0.05069 1.62710 2.06150 0.00061 1813.8 622.28 40.36 4.629 650.00 1151.40 0.18200 0.05190 0.28531 0.05307 1.68570 2.20410 0.00056 1898.2 628.62 39.74 4.931 660.00 1145.66 0.17859 0.05258 0.28115 0.05565 1.75180 2.36520 0.00051 1983.9 634.84 39.11 5.253 670.00 1139.49 0.17518 0.05330 0.27709 0.05848 1.82770 2.55010 0.00046 2069.5 640.84 38.47 5.593 680.00 1-132.96 0.17181 0.05405 0.27314 0.06173 1.91440 2.76120 0.00042 2155.4 646.65 37.81 5.953 690.00 1126.01 0.16845 0.05485 0.26926 0.06527 2.01490 3.00550 0.00037 4384-non\seclOa.wpd-04203 10-55

WESTINGHOUSE PROPRIETARY CLASS 2 Table 10-3 (Cont'd) Vessel Component Saturated Water Thermal Properties P. T. PP S H, HI Of - k, kg Cp C (psia) (OF) (Ibn/fP) Obnife) (Btu/lbm) (Btu/lbm) Qbn/llr/ft) Obm/hr/f) (Btu/hr/ftlE (Btu/hr/ftlIF) (Btu/lbm/F) (Btu/lbm/F (Ibf/ft) 2241.1 652.28 37.14 6.335 700.00 1118.63 0.16509 0.05570 0.26545 0.06919 2.13280 3.29110 0.00033 2326.3 657.69 36.46 6.740 710.00 1110.82 0.16173 0.05661 0.26186 0.07374 2.27250 3.62830 0.00029 2410.6 662.89 35.76 7.170 720.00 1102.59 0.15837 0.05758 0.25836 0.07854 2.44030 4.03110 0.00025 2494.0 667.89 35.03 7.630 730.00 1093.91 0.15497 0.05863 0.25520 0.08401 2.64650 4.52210 0.00022 2575.2 672.62 34.29 8.117 740.00 1084.90 0.15156 0.05975 0.25212 0.08983 2.90110 5.12290 0.00019 2653.8 677.08 33.53 8.635 750.00 1075.52 0.14813 0.06096 0.24935 0.09704 3.22290 5.87350 0.00016 2729.8 681.29 32.75 9.190 760.00 1065.74 0.14464 0.06228 0.24673 0.10465 3.64070 6.14500 0.00013 2801.8 685.18 31.95 9.777 770.00 1055.59 0.14113 0.06371 0.24493 0.11447 4.19150 6.14500 0.00011 2869.6 688.77 31.12 10.403 780.00 1045.00 0.13755 0.06526 0.24313 0.12429 4.94400 6.14500 0.00008 2931.9 692.01 30.27 11.066 790.00 1034.06 0.13393 0.06694 0.24418 0.13890 5.99630 6.14500 0.00007 2988.5 694.90 29.39 11.771 800.00 1022.75 0.13024 0.06878 0.24576 0.15442 6.14500 6.14500 0.00005 3038.4 697.40 28.48 12.513 810.00 1011.17 0.12649 0.07076 0.24734 0.16993 6.14500 6.14500 0.00004 3081.4 699.53 27.55 13.293 820.00 999.32 0.12268 0.07291 0.25288 0.19158 6.14500 6.14500 0.00002 3116.7 701.26 26.60 14.101 830.00 987.40 0.11885 0.07520 0.26470 0.22270 6.14500 6.14500 0.00002 3144.7 702.62 25.63 14.927 840.00 975.50 0.11500 0.07761 0.29237 0.27220 6.14500 6.14500 0.00001 3165.7 703.63 24.65 15.750 850.00 963.98 0.11119 0.08008 0.81017 0.80644 6.14500 6.14500 0.00000 3180.5 704.34 23.68 16.541 860.00 953.38 0.10744 0.08253 2.55507 2.55265 6.14500 6.14500 0.00000 Mon\sec I Oa.wpd-04203 ¢-56

( ( Table 10-3 (Cont'd) Vessel Component Saturated Water Thennal Properties P,, Tp 1 pg Hr H Ps k kg C, C, a (psia) (0 F) Obm/fe) ObnW/) (Btullbm) (Btu/Ibm) Ibm/hrft) Obmhr/ft) (Btu/hr/ftIF) (BtuAhr/ftfF) (Btu/Ibm/F) BtuAbmnF (Ibf/ft) 3190.3 704.81 22.72 17.243 870.00 944.22 0.10382 0.08475 4.29997 4.29886 6.14500 6.14500 0.00000 3196.0 705.08 21.78 17.759 880.00 937.47 0.10034 0.08642 10.00000 10.00000 6.14500 6.14500 0.00000 3198.3 705.19 20.87 17.987 890.00 934.25 0.09704 0.08717 50.00000 50.00000 6.14500 6.14500 0.00000 3206.4 705.39 20.16 19.244 900.00 917.46 0.09704 0.08717 100.00000 100.00000 6.14500 6.14500 0.00000 43 84-non\sec I Oa.wpd-04203 10-57

WESTINGHOUSE PROPRIETARY CLASS 2 Table 104 Superheated Vapor Temperature Constants P 1000 psia or P > 1000 psia and Term P > 1000 psia and h 1280 Btu/lbm L, < 1280 Btu/1bm A -1.0659659E+04 4.5298646E+03 A2 2.0110905E+01 1.5358850E+01 A.; -1.250954E-02 -1.5655537E-02 A4 2.8274992E-06 5.2687849E-06 [ ffi 4.9815820 4.4185386E-01 A6 -7.7618225E-06 -9.1654905E-06 A7 2.4391612E-10 2.7549766E-10 A8 -9.8147341E-03 -1.1541553E-03 A9 6.5824890E-06 1.2384560E-06 A0 -1.4749938E-09 -4.1724604E-10 B, -2.8557816 1.2659960E+02 B,2 1.3250230E-02 -2.5611614E-01 B3 -1.0521514E-05 2.2270593E-04 B4 2.5007955E-09 -5.9928922E-08 BS -3.4620214 -2.1818030E+01 B6 -3.6261637E-02 1.3424036 B7 7.3529479E-04 4.9110372E-02 Bs 5.7703098E-03 2.7966370E-02 B9 -2.9972073E-06 -2.4665012E-05 tB10 5.2037300E-10 6.7723080E-09 43 84-non\sec Oa.wpd-04203 10-58

WESTINGHOUSE PROPRIETARY CLASS 2 Table 10-5 Subcooled Water Density Constants 1= 1 2 3 4 5 jl -0.413450E1 0.13252E-4 0.15812E-5 -0.21959E-8 0.21683E- 1 j=2 -059428E-5 0.63377E-7 -0.39974E-9 0.69391E-12 -0.36159E-15 j=3 0.15681E-8 -0.4071 lE-10 0.25401E-12 -0.52372E-15 0.32503E- 18 Table 10-6 Saturated Steam Internal Energy Constants i P l AVE(i) BVE(i) CVE(i) 1 l s2E+6 T 2.619410618E+6 -4.995E+10 T - . 2 l > 2E+6 I 2.5896E+6 I 6.350E-3 I -1.0582E-9 4384-non\seclOa.wpd-04203 10-59

WESTINGHOUSE PROPRIETARY CLASS 2 Table 10-7 Saturated Steam Enthalpy Constants i P AVG(i) BVG(i) CVG(i) 1

• 2E+6 1.06655448 1.02E-8 -2.548E-15 2 > 2E+6 1.0764 3.625E-10 -9.063E-17 JJ Table 10-8 Saturated Liquid Internal Energy Constants i ALE(i) BLE(i) CLE(i) DLE(i) ELE(i)

I 1.75880E+4 3.7402E+3 4.02435 -0.0157294 3.1301E-5 2 6.18527E+6 -8.14547E+4 4.46598E+2 -1.04116 9.26022E-4 3 2.283789029E+9 -2.62215677E+7 1.12948667E+5 -2.16233985E+2 0.155283438 4384-non\seclOa.wpd-04203 10-60

WESTINGHOUSE PROPRIETARY CLASS 2 Table 10-9 Constants for Specific Heat C1v= 1.68835968 x 103 Bo= 2.394907 x 1 B,,= -5.196250 x 10713 C2 v,= 0.6029856 CO,= 1.193203 x 10-" C 1 ,= 2.412704 x 10-18 C3 ,= 4.820979623 x 102 Dot= -3.944067 x 10717 D,,= -1.680771 x 10e C4 v= 2.95317905 x Id' Cs,= 1.8 ___ ___ ___ ___ _ _ _ _ _ _ _ _ _ __. C6V=4.60x 102 43 84-non\sec Oa.wpd-04203 10-61

WESTINGHOUSE PROPRIETARY CLASS 2 Table 10-10 Liquid Viscosity Constants Aot = 1.299470299x10- 3 BO, =-6.5959x 10-' 2 A1l =-9.264032108x10- 4 Bit = 6.763x1'- 2 A 21 = 3.81047061x10- 4 B2, =-2.88825x10-' 2 A 31 = -8.219444458x 10-5 B 31 = 4.4525xlr- 13 A 41 = 7.022437984x 10-6 Do, = 3.026032306x10- 4 Eol = 1.4526052612x10-3 DU =-1 .836606896x 10- 4 Ell =-6.9880084985xlO-9 D21 = 7.567075775x10-5 E2t = 1.5210230334x1-1 4 D31 = -1 .647878879x 10-5 E 3 = -1.2303194946x 1W-2 0 D4t = 1.416457633x10-6 Fo =-3.8063507533x10- 1 l Ho = 8.581289699x1T-6 Fit = 3.9285207677x1-1 6 Con= 4.265884x10 4 F2 1 =-1.2585799292x10-21 PO = 6.894575293x10 5 F3 t = 1.2860180788x1- 27 H 3.892077365x1- 6 ehO = 6.484503981x1-6 00 = ecO= 5.53588xl Cn = 4.014676x10 5 4384-non\sec IOa.wpd-04203 10-62

WESTINGHOUSE PROPRIETARY CLASS 2 Table 10-11 Vapor Viscosity Constants Aov = 3.53xlo 8 Bl = 0.407x10-7 A Iv= 6.765x1O-" Civ = 8.04x 10-6 A2v = 1.021x10-14 Div = 1.858x10-7 Eiv = 5.9xlO-10 Fov =-0.2885x10- 5 Gov = 0.176x10 3 F1 V = 0.2427x10-7 Giv =-1.6 F2V =-0.67893333333x10 -l0 G2V = 0.0048 F3 =

        =3  0.63 7037037x10-13 7307                                GV3v =-0.47407407407xlO-5 Table 10-12 Liquid and Vapor Thermal Conductivity Constants A  =  0.573738622              Avo   = 1.76x 10-2             Bv   = 1.035 1X1cV 4 A,,   =  0.2536103551             Av    = 5.87x10 5            l Bvo  = 0.4198x 10  6 A12 = 0.145468269                 AV2 = 1.04x10-7                BV2 =-2.771x1-"

A,3 = 0.01387472485 Av=4.1l-A 4 = 5.815x10 5 Av3 = .1x15 5 ____ _____ ____ Av 4 = 2.1482x10 4384-non\seclOa.wpd-04203 10-63

WESTINGHOUSE PROPRIETARY CLASS 2 Table 10-13 Surface Tension Constants a, = 1.160936807E-04 a 2 = 1.12140468E-06 a3 = -5.752805180E-09 a 4 = 1.286274650E- 11 a5 = -1.149719290E-14 Table 10-14 Constants for Specific Heat of Air i ~~Ai B, 1 4.20419E-05 7.71027E-05 .L 2 9.61128E-08 -8.56726E-09 7 3 -1.16383E-11 4.75772E-12 4384-non\seclOa. wpd-04203 10-64

WESTINGHOUSE PROPRIETARY CLASS 2 Table 10-15 Constants for Viscosity of Air i apj abi 0 1.708x10-5 1.735x10-5 1 5.927x10- 8 4.193x1c- 8 2 -8.14x10-" -1.09XT1" 43 84-non\sec Oa.wpd-04203 10-65

WESTINGHOUSE PROPRIETARY CLASS 2 Table 10-16 Specific Heat of Zircaloy4 T(0 K) C, (Btubm - 'F) 300.0 0.0671 400.0 0.0721 640.0 0.0790 1090.0 0.0896 1093.0 0.1199 1113.0 0.1409 1133.0 0.1469 1153.0 0.1717 1173.0 0.1949 1193.0 0.1839 1213.0 0.1478 1233.0 0.1120 1248.0 0.0850

                        >1248.0                                      0.0850 4384-non\sec Oawpd-04203                         10-66

WESTINGHOUSE PROPRIETARY CLASS 2 Table 10-17 Chemical Composition of ZIRLOTI and Zircaloy4 Alloys Element (wt %) ZIRLOTNi' Alloy Zircaloy-4 Alloy Sn 0.8-1.2 1.2-1.7 Fe 0.09-0.13 0.18-0.24 Cr 0.07-0.13 Fe+Cr 0.28-0.37 Nb 0.8-1.2 Zr Balance Balance Table 10-18 Specific Heat of ZIRLO"' Alloy 11 [ I _ ____11 1*

                                                     +

4. 4. 1* _] 4384-non\secl Oa.wpd-04203 10-67

1000-900-800-l--% E

  -o        700-
   -0       600-500-z I      400-D
   \

300-200 i I o ASMF STFAMF TARLE

                                                      -    WCOBRA/TRAC 100 0     400 800 1200  1600 2000    2400 2800      3200 PRESSURE (psic)

Figure 10-1. WCOBRA/TRAC Vessel Component Saturated Liquid Enthalpy Function 4384-non\seclOb.wpd-04203 10-68

 .- 1% .- - -
 -       1100 Z      1000 Li 0
  >        900-o  ASME STEAM TABLE 8-  WCOBRA/TRAC 800-                                                    I 0       400 800 1200 1600  000  2400   2800    3200 PRESSURE (psia)

Figure 10-2. WCOBRA/TRAC Vessel Component Saturated Vapor Enthalpy Function 4384-non\seclOb.wpd-04203 10-69

7001 LL- 600-0 w La D 500-Li 0L 400 Li z 300-0

200 (I) 100-o ASME STEAM 'ABLE

                                                         -    WCOBRA/TRAC 0      100 200   300 400   500   600     700   800    90 0 LIQUID ENTHALPY (btu/Ibm)

Figure 10-3. WCOBRAJTRAC Vessel Component Saturation Temperature 4384-non\seclOb.wpd-04203 10-70

100 80 4-E

  -0        60 LO) z LU 0        40-0 5:

CY 20- t o ASME STEAM TAE3,LE 010WCOBRA/TRAC 0 400 800 1200 1600 2000 2400 2800 3200 PRESSURE (psia) Figure 10-4. WCOBRA/TRAC Vessel Component Saturated Liquid Density 4384-non\seclOb.wpd-04203 10-71

20-16-I4.* E

     -o     12-(I) z bJ 8-0 0

4-o ASME STEAM TABLE

                                                       -     WCOBRA/TRAC a II                  -  -  -   -             .      . 0. 0    .

b 100 200 300 400 500 600 700 800 900 I SATURATED LIQUID ENTHALPY (btu/Ibm) Figure 10-5. WCOBRA/TRAC Vessel Component Saturated Vapor Density 4384-non\seclOb.wpd-04203 10-72

1.0-0.8-E

   -n       0.6-0 O )

t) 0.4-DD

   -L       0.2 o  NRC STEAM TABLE 0.0-                                         -    WCOBRA/TRAC o       400   800     1200     1600        2000     2400 PRESSURE (psia)

Figure 10-6. WCOBRA/TRAC Vessel Component Saturated Liquid Viscosity 4384-non\seclOb.wpd-04203 10-73

0.100-e--, 0.080-

  '4-L.

E

  -o      0.060-                                                   0 o

o 0cn 0 C) 0.040-0 a_ c) 0.020-o ASME STEAM TABLE

                                                         -   WCOBRA/TRAC 0.00n Il 0    400   800      1200      l 600     2000      2400 PRESSURE (psia)

Figure 10-7. WCOBRA/TRAC Vessel Component Saturated Vapor Viscosity 4384-non\secl Ob.wpd-04203 10-74

0.500-Ii-0 0.400-L- F- 0.300-

D z

0 0.200-0

   -j 0.100-LUJ I-o   ASME STEAM TABLE
                                                        -     WCOBRA/TRAC 0.000          I     260  I   4I    560         I   . I  I  I 0    100    200 300  400   500 600   700     800   900 SATURATION TEMPERATURE (F)

Figure 10-8. WCOBRAITRAC Vessel Component Saturated Liquid and Vapor Thermal Conductivity 43 84-non\sec Ob.wpd-04203 10-75

3.00-oL 2.50 E 2.00 co 0z 1.50_ LL 1.00 - n CY 0.50-oNRC STEAM TABILE

                                                            -WCOBRA/TRAC o.0   460 860       1200     1600     2000       2400 PRESSURE (psia)

Figure 10-9. WCOBRA/TRAC. Vessel Component Saturated Liquid Speciic Heat 4384-nori\sec IOb.wpd-04203 10-76

LiL 5.000/ 0

   -o D/

F- 4.000 I 3.000/ LU Q. 2.000-0 1.000-o ASME STEAM TABLEE

                                                         -   WCOBRA/TRAC 0        1000            2000               3000 SATURATION PRESSURE (psia)

Figure 10-10. WCOBRA/TRAC Vessel Component Saturated Vapor Specific Heat 43 84-non\sec IOb.wpd-04203 10-77

                                                        -     WCOBRA/TRAC o   ASME STEAM TABLE 0.400-CIA 0

x 0.300-z 0 z

  '-      0.200-Li c*,     0.100 0.000 0     100 200   300     400  500        600    700 TEMPERATURE (OF)

Figure 10-11. WCOBRAITRAC Vessel Component Saturated Liquid Surface Tension 4384-non\secl Ob.wpd-04203 10-78

2000-1900 tP 0-- ----GL-------E-----0----- Cl --- E]----- Ea 1800-1700-

   -D 1600-
   -0 A-  .  -A_-    A-    -  _A._     _A   _     _

C- 1500-I z 1400-Li 1300- oi 1700°F NRC TABLES

                                                                       --   -    1 700°F WCOBRA/TRAC A&      1000°F ASME TABLES 1200-                                                        -         1000°F WCOBRA/TRAC o       600°F ASME TABLES 1linn-
                                                                       -         600°F WCOBRA/TRAC I I  iuu -T       I       I I '             I'   I    '-   I       I    I    I   I 0      400    800     1200  1600     2000      2400        2800      3200 PRESSURE (psia)

Figure 10-12. WCOBRAITRAC Vessel Component Superheated Vapor Enthalpy 4384-non\sec I Ob.wpd-04203 10-79

2000-0 O-cO O) O O), O 1800-1600-1-S% 1400-LL-0 1200-LUJ 0:: D Q-LUJ 800-LUJ 600-400-o NRC TABLE 200- - WCOBRA/TRAC h = 2000

                                                           -   WCOBRA/TRAC  h = 1500
                                                           --  WCOBRA/TRAC  h of sat. vap.

0 - - - - - - - - - - - - - - 6 ' 460 860 1200 1600 2000* 2400* 2800 PRESSURE (psia) Figure 10-13. WCOBRAITRAC Vessel Component Superheated Vapor Temperature 4384-non\sec IOb.wpd-04203 10-80

6.00-E

  -0 1ef     °                          o 0   NRC TABLE Ir                                    -    WCOBRA/TRAC h = 2000
                                                               -    WCOBRA/TRAC h = 1500 0.00-           ~  4 v.uu     ~. I ,

II , nI r I

                                                               -    WCOBRA/TRAC I       I h = 1200 400   800   1200    1600   2000    2400    2800 PRESSURE (psia)

Figure 10-14. WCOBRA/TRAC Vessel Component Superheated Vapor Density 4384-non\secI Ob.wpd-04203 10-81

0.100-IL 0 L P=3000 psia m P=2000 psic 0.050-IL 0 P=1000 psia 9j z 0

     -j P=14.7 psia w

H o ASME

                                                           -   WCOBRA/TRAC 0.000 I                500          1000              1500 TEMPERATURE (OF)

Figure 10-15. WCOBRAJTRAC Vessel Component Superheated Vapor Thermal Conductivity 4384-non\sec I Ob.wpd-04203 10-82

0.100-I 0.080-0'

                                                               -9' 0.060-                                           la E                                                 00 [
  -o 0.040-0 C

c,) 0.020-o ASME TABLES

                                                                     - -   P = Psat(T-500)
                                                                     -     AT SATURATION 0.000      l   g i I  I    I

• I. I I I . . , I . . . I 6 200 400 600 800 1000 1200 TEMPERATURE (OF)

Figure 10-16. WCOBRA/TRAC Vessel Component Superheated Vapor Viscosity 4384-non\secl Ob.wpd-04203 10-83

15-10-0 w D U) U) bi 5 O NRC TABLES o p

• I I - WCOBRA/TRAC C

4.00 450 500 550 600 650 700 TEMPERATURE (OK) Figure 10-17. WCOBRA/TRAC 1-D Component Saturation Pressure 43 84-non\sec I Ob.wpd-04203 10-84

100. 80 E In z Lu 0 0 0-o NRC TABLES 450 500 600 650 TEMPERATURE (OK) Figure 10-18. WCOBRA/TRAC 1-D Component Saturated Vapor Density 4384-non\secl Ob.wpd-04203 10-85

1000-e4,) 900-E 800 z 0

                                                                                ,ti 0

ar 700 o NRC TABLES

                                                           -    WCOBRA/TRAC 600-1_                 5                               I    .      I 400      450     500       550      600         650         700 TEMPERATURE (OK)

Figure 10-19. WCOBRA/TRAC 1-D Component Saturated Liquid Density 4384-non\seclOb.wpd-04203 10-86

3.OOE+06-2.80E+06-0-1 2.60E+06-I- z 2.40E+06-0 2.20E+06-o NRC TABLES 2.00E+06-j_ - WCOBRA/TRAC

                            . . I  T
                                      * . I  I*

1 200 300 400 500 I600 700 TEMPERATURE ( K) Figure 10-20. WCOBRA/TRAC 1-D Component Saturated Vapor Enthalpy 43 84-non\sec Ob.wpd-04203 10-87

cn

      -    2.OOE+06-z LJ 0

D 1.00E+06-C NRC TABLE! 3 0.00- g° - WCOBRA/TR 200 300 400 500 600 TEMPERATURE ( K) Figure 10-21. WCOBRAITRAC 1-D Component Saturated Liquid Enthalpy 4384-non\seclOb.wpd-04203 10-88

8000-7000 N 6000 N 5000 4000 C) LX

0) 3000 2000 i

o NRC TABLES 1000-I ~ ~~~~~~- WCOBRA/TRAC

                         .        I               I              I.

400 500 600 700 TEMPERATURE ( K) Figure 10-22. WCOBRA/TRAC 1-D Component Saturated Vapor Speciic Heat 4384-non\seclOb.wpd-04203 10-89

9000-8000-7000-C) ii 6000-ai-Co 5000-

                          . o - v                             ASME TABLES
                                                         -    WCOBRA/TRAC 4000-          ,     .r            . I  I               I 400               500            600                   700 TEMPERATURE (   K)

Figure 10-23. WCOBRA/TRAC 1-D Component Saturated Liquid Specific Heat 43 84-non\sec Ob.wpd-04203 1 0-90

3.OE 2.OE z cri 0 U) 1.OE o NRC TABLES I - WCOBRA/TRAC C 0% I I.4 40 560 600 700 TEMPERATURE ( K) Figure 10-24. WCOBRA/TRAC 1-D Component Saturated Vapor Viscosity 4384-non\sec I Ob.wpd-04203 10-91

3.OE '> 2.OE U) z Cl) 0 U) 1.OE o NRC TABLES 0- WCOBRA/TRAC 400 500 600 700 TEMPERATURE ( K) Figure 10-25. WCOBRAITRAC -D Component Saturated Liquid Viscosity 4384-non\seclOb.wpd-04203 10-92

0.20- _e E 0.15-

      "I-,

0CD 0.10-0 z 0 CD I

0.05-o NRC TABLES 0-o I_ - WCOBRA/TRAC 40 500 600 70 TEMPERATURE ( K)

Figure 10-26. WCOBRA/TRAC 1-D Component Saturated Vapor Thermal Conductivity 43 84-non\sec Ob.wpd-04203 10-93

1.0-0.8-Z: 0~ '0 0 D E 0.6- 1-- 0 C) D C] 0.4-L I- 0.2-o NRC TABLES

                                                           -   WCOBRA/TRAO n_. X-
                   .I I --

X I - w T w I - 40 500 600 700 TEMPERATURE ( K) Figure 10-27. WCOBRA/rRAC 1-D Component Saturated Liquid Thermal Conductivity 4384-non\seclOb.wpd-04203 10-94

100-

                                                               -   WCOBRA/TRAC o NRC TABLES 80-z rn) 0 60-0 z

40-C-) D 20-Cf) 4 ff X l w w w w w w w s)bw l 460 500 600 700 TEMPERATURE ( K) Figure 10-28. WCOBRA/TRAC 1-D Component Surface Tension 4384-non\seclOb.wpd-04203 10-95

v o 4.0-4- 1 3.5-L.

    Ž        3.0-m
    ~>       2.5-
    >        2.0-C)\

D 1.5-. z o 1.0-

   <         0.5-F-            5(00     1000 1500 2000 2500 3000 3500 4000 4500 5000 TEMPERATURE (OF)

Figure 10-29. WCOBRA/TRAC U0 2 Thermal Conductivity (95% of Theoretical Density) 4384-non\sec lOc.wpd-04203 10-96

0-Lu IL Lu 0LJ 600 120(0 l00 2400 3000 TEMPERATURE (K) Figure 10-30. WCOBRAITRAC UO2 Specific Heat 4384-non\seclOc.wpd-04203 10-97

1I 0 20-

i. 1 18-

  -D 16-14-c)     12-D       10-8-

z 6-0 4-

   -J       2-I 0-   I       l                              llll 0              500       1000          500        2000       2500 TEMPERATURE (OF)

Figure 10-31. WCOBRA/TRAC Zircaloy-4 Thermal Conductivity 4384-non\seclOc.wpd-04203 10-98

 - c)      -           3    ZIRCALOY E                                                                t
   -a mn      0.10-H LU C-)

LU C/) 0.00- . . . 300 500 700 900 1100 13( TEMPERATURE (deg K) Figure 10-32. WCOBRA/TRAC Zircaloy-4 Specific Heat 4384-non\seclOc.wpd-04203 10-99

a,c L 0.20 LL-

                       *-*    ZIRCALOY
                          - E ZIRLO U)
  -ci                                                         IP
  -0 f

4- 0.10 w 0 (I) 0.00

                                          '0       900     1100        1300      1500 TEMPERATURE (deg K)

Figure 10-33. Comparison of ZIRLOTm and Zircaloy-4 Specific Heat 4384-non\secl Oc.wpd-04203 10-100

20.0 0 I1 4- J1 i 4-

   -C       15.0-D m

10.0-C z 0

   -J        5.0-4 LUJ 2:

H-0.0 X l

                                  .f.   .   . . .  .  . .  .   . .  .I 0     I 200       400         600        *80       1000 TEMPERATURE (OF)

Figure 10-34. WCOBRA/TRAC 1-D Component 304 Stainless Steel Thermal Conductivity 4384-non\sec Oc.wpd-04203 10-101

0.200-LL o-0 0.150-E m 0.100-CD i 0 0 0.050-u/) 0.000 -I 0 I 200 I0 400 600 , I 800

                                                                         . 1000 I

TEMPERATURE (OF) Figure 10-35. WCOBRA/TRAC 1-D Component 304 and 316 Stainless Steel Specific Heat 4384-non\secl Oc.wpd-04203 10-102

I00-0 1 4-L. 15.0-m O 10.0-F5 C) 0C] 5.0-

2 LI I

0.0 l . . . . . . . . . . . . . . . . . I 200 400 600 800 doo TEMPERATURE (OF) Figure 10-36. WCOBRAITRAC 1-D Component 316 Stainless Steel Thermal Conductivity 4384-non\secl Oc.wpd-04203 10-103

20.0-0 L& 15.0-m H-10.0-C: 0 z 0 C-: 5.0-

  -IJ I

F-- 0.0 I I I I I r I I 0 200 400 600 800 1000 TEMPERATURE (OF) Figure 10-37. WCOBRA/TRAC 1-D Component 347 Stainless Steel Thermal Conductivity 4384-non\sec Oc.wpd-04203 10-104

0.200-LL 0 0.150-E m 0.100-0 i Cl) 0.050-0.000 -I 4I0 ,I I 200 400 600 800 1 000 TEMPERATURE (F) Figure 10-38. WCOBRA/TRAC 1-D Component 347 Stainless Steel Specific Heat 4384-non\sec Oc.wpd-04203 10-105

40.0-LL 0-

 '4-LC     35.0-z
     ~125.0 Lii H

20.0 0 200 400 600 800 1000 TEMPERATURE (OF) Figure 10-39. WCOBRAITRAC 1-D Component Carbon Steel Thermal Conductivity 4I84-nnn\,cI Oc-wnd-04203 10-106

0. 0 0.150 E m 0.100 0 Lii 0.050 CL)

                                                     -  7 -                  ooo 200      400          600          800     1000 TEMPERATURE (F)

Figure 10-40. WCOBRAITRAC 1-D Carbon Steel Speciflc Heat 4384-non\secl Oc.wpd-04203 10-107

20.0-1 0 L 15.0-i__ m 4 0 0 0

  -j          5.0 Li H-0.0 4             400       16801I        II I   I 6      200     400         600
                                                    .I,*       I Boo I  . I 1000 TEMPERATURE (OF)

Figure 10-41. WCOBRA/TRAC 1-D Component Inconel 600 Thermal Conductivity 4384-non\sec I Oc.wpd-04203 10-108

0.200-0.150-L. 0

D m

0.100-0 LU-0 0.050-UI) 0.000+- 0 200 400 600 800 1000 TEMPERATURE (F) Figure 10-42. WCOBRA/TRAC 1-D Component Inconel 600 Specific Heat 4384-non\secl Oc.wpd-04203 10-109

I 4384-non\seclOc.wpd-04203 10-- 10

SECTION 11 SMALL BREAK LOCA-RELATED CAPABILITIES 11-1 Introduction To enable modelling of small break LOCA events in a CSA U methodology, certainfeatures to facilitate WCOBRA/IRAC execution were added. This section describes these features, added to enable pertinentmodel sensitivity studies and to increase modellingflexibilities. In addition, the models and correlationsadded to permit accuratepredictionsof small break LOCA phenomena which have been previously describedare summarized; consult Volume 2 for validation results of the small break LOCA code version, WCOBRAITRAC-SB. 11-2 WCOBRA/TRAC-SBAdditional Features WCOBRA/TRAC Mod7A is the computer code developed by Westinghouse (Bajorek et al., 1998) for the best estimate analysis of large break LOCAs and approved by the NRC stafffor that purpose (Jones, 1996). Some extra features have been added during development of the WCOBRA/ITRAC-SB code version. 11-2-1 FeaturesPreviously Developed The following WCOBRA/ITRAC modificationpreviously introducedto enable modelling of certainAP600 components (Garneret al., 1998) has been placed into the small break code version. Check Valve Option, Type 6 The MOD7A code hasfive control optionsfor the VALVE component. These options are described in Section 9-7. Option 5 is usedfor check valve simulation. For this option, a 4384-non\secl Il.wpd-04303 11-1

specified ramp opens or closes the valve, taking several hundred timesteps. Under certain circumstances, the valve stays partiallyopen and allowsflow in both directions. A Type 6 check valve option has been added to the WCOBRA/TRAC code versionfor small break LOCA analysis. The Type 6 valve is eitherfully opened orfully closed in one timestep. The check valve opens when the pressure gradientacross the valve reaches a user-specifled value and closes when the pressuregradientis less than the specified value, or when reverseflow is detected. 11-2-2 FeaturesIntroducedto Enable ParameterRanging In order to perform response surface and/oruncertainty analyses, parametersimportant to the small break LOCA analysis must be ranged. To facilitate this, thefollowing variablesare introduced into WCOBRA/TRAC-SB; the default value for each is 1.0. 11-2-2-1 Variable "YDRAG" This modification enables the user to apply a multiplier to the vertical interfacialdrag coefficient. This multiplier is applied to the verticalflow regimes. User-suppliedinput allows specificationof multipliersforindividual channels. 11-2-2-2 Variable "XCNDSB" This modification enables the user to apply a multiplierto the interfacialcondensation heat transfercoefficient. User-suppliedinput allows specification of multipliersforindividual channels. 4384-non\sec l.wpd-04303 11-2

11-2-2-3 Variable "XSHASB This modification enables the user to apply a multiplierto the wall condensation heat transfer coefficientfor unheated conductors. User-suppliedinput allows specificationof multipliersfor individual channels. 11-2-3 Multiple Regions in VESSEL Channels As described in the simulations presented in Volumes 2 and 3 of this report, VESSEL channels are used to model almost the entireprimary coolant loop. Because their use is more widespread than in the large break LOCA WCOBRAITRAC model, moreflexibility in specifying channel input is necessary. One of the restrictionsassociatedwith the use of VESSEL channels is that one set of cell height variationsis applied to all channels in a section. To relax this restriction, an update has been implemented into WCOBRAITRAC-SB making multiple sets of cell height variationtables availablefor channels in a section. Figure11-1 shows the primary circuit of the PWR and the approximateelevation of major components, i.e., the pressure vessel, cold and hot legs, steam generator, cross-overpipe and RCP. To model a loop using VESSEL channels, the cell heights in the steam generatorplenum and tubes would previously have been restrictedto the same value as the VESSEL upperplenum nodes. Similarly, the cross-over leg [ Jac Channels are [

                                                                                                ]a 4384-nonsec l.wpd-04303                         11-3

must match between the two connecting regions. Regions can also be connected through a gap. In this case, IL) JaC to assist debugging of input. 11-2-4 Hydraulic Cell Level TrackingforHeat Transfer Computations The WCOBRA/TRAC code does not contain an explicit mixture level tracking model. Rather, level tracking is accomplished by nodalization, andprediction of the axial void gradientbetween hydraulic cells. In most regions of the vessel and reactorcoolant system (RCS), nodalization is sufficient to track the mixture level, as the structures do not have a quench front. A detailed tracking of the mixture level is not necessary. In the reactorcore, however, an accurateassessment of the mixture level is vital in the prediction of the peak cladding temperature (PCT). Local voidfractionsfor use in heat transfer IL calculationsare typically linearly interpolatedbetween adjoiningaxial hydrauliccells. The detailed nodalization in the core [ Jac helps to resolve the axial voidfraction gradient. However, the linear interpolationof voidfraction does not allow the location of a sharp interface to be identified. Therefore, WCOBRAITRAC-SB includes logic to detect the possiblepresence of a sharp voidfraction gradientin the vicinity of a quench front. This new logic is used in conjunction with the linearinterpolation logic so that the calculation affects only the hydrauliccell in which the sharp gradientis assumed to occur; it is employed in fuel rod heat transfercomputations. Figure 11-3 shows an example of the linearinterpolationscheme. Continuity cell 'j' is at the boundary [ ac' 4384-non\sec I l.wpd-04303 11-4

I Ia.c I If a sharp gradientis detected I

                                      ]" (11-2-1)

[ (11-2-2)

             ]apc 4384-non\sec I l.wpd-04303        11-5

With this example of level sharpening logic,

                          ]aC 11-2-5 Tmin Definition A variable (ITMINHN) is introduced to enable the user to define Tmin, the minimum stablefilm boiling temperature, to be the homogeneous nucleation temperature. If chosen by the user, the calculation of TUn previously describedherein is not performed; Tw, is always calculated to be the homogeneous nucleation temperature, which is generally regardedas the lowest value of T.jnw accordingto Equation 6-97 in this volume. Setting ITMINHN to equal 1 allows a conservative approachto be taken in PWR calculations.

11-2-6 Momentum Transfer at Pipe Elbows When WCOBRAITRA C-SB VESSEL component channels are used to model the loop piping in a PWR, pipe elbows in the hot leg and at the loop seal in the crossover leg (two elbows) are considered in which theflow turnsfrom a horizontal to a vertical direction (or vice-versa). Because the flow velocity is stopped in one direction, then restartedin anotherwith the staggeredmesh used in the WCOBRAITRAC-SB model, inappropriatepressuredrop conditions arepredicted. This situation has been corrected by introducing a VESSEL component momentum scheme for a U-bend region. Figure 11-5 presents the Momentum Cell A case, when a verticalflow turns into a horizontal channel. In this case, the momentum convection is calculatedasfollows: jac 4384-non\sec I l.wpd-04303 11-6

I Figure11-6 presents the Momentum Cell B case, when a gap between horizontalflow channels is stopped at a wall. In this case, the cell momentum in Cell B is pVAgap but the associated

                                             ]n.c Figure 11-7 presents the Momentum Cell C case, when the horizontalflow is redirectedinto a vertical channeL In this case, the cell momentum in Cell C is pUA, the associatedconvection velocity is U. The axial momentum     [

a] 4384-nonkecl .wpd-04303 11-7

I jac 11-2-7 EnhancedReactivity Insertion Model In a small break LOCA event negative reactivity is introduced into the core both by voiding of thefluid and by the insertion of the control rods. WCOBRAIfRAC-SB has been programmed with the capability to use the internalreactivityfeedback models described in Section 8 together with an input reactivity table. The user can specify a negative reactivity insertion as affunction of time to model the action of the control rods. 11-3 Summary of Identified Improvements Necessary to Model Small Break LOCAs The capability of WCOBRA/TRAC MOD7A to model high ranked small break LOCA processes was assessed. Modelling of thefollowing processes wasjudged to requirenew and/or upgraded capability in the WCOBRAITRAC-SB code version used in small break LOCA analysis: a HorizontalStratifiedFlow Phenomena a Break Flow a Steam GeneratorTube Condensation a Heat Transfer to Uncovered Fuel The models/correlationsimplemented in WCOBRAITRAC-SB for each of these phenomena is identifed in preceding sections of Volume 1. Other modelfeatures which are unique to WCOBRA/ITRAC-SB are also presented. Overall, this volume presents a complete description of the models and correlationscontained in the WCOBRA/TRAC-SB code version. 4384-nonsec I l.wpd-04303 11-8

114 References Bajorek, S. M. et al., 1998 "Code QualificationDocumentfor Best Estimate LOCA Analysis, " WCAP-12945-P-A, Revision 2, Volumes 1 through 4. Jones, R. C. Jr. (USNRC) letter to N. J. Liparulo (&, "Acceptancefor Referencing of the Topical Report, WCAP-12945(P), Westinghouse Code QualificationDocumentfor Best-Estimate Loss-of-Coolant Analysis, " June 28, 1996. Garner, D. C. et al., 1998 "WCOBRAITRAC OSU Long-Term Cooling Final Validation Report, " WCAP-14776, Revision 4. 4384-nonsecl l.wpd-04303 11-9

Region II

                                                \    1D Pump Component Region III Region I Figure11-1. PWR PrimaryCircuit 4384-non\sec I .wpd-04303              1 1-10

                                                                     -q a,c Figure11-2. Channel and Cell Identification 4384-non\secl 1.wpd-04303                 11-11

a,c Figure 11-3. WCOBRAITRAC Void Fraction Interpolation for Rod Heat Transfer Calculations a,c Li Figure 11-4. WCOBRA/TRAC Level Sharpening for Rod Heat Transfer Calculations 4384-nonsecl l.wpd-04303 11-12

a,c Figurel1-5. Momentum CellA 4384-non\secl I .wpd-04303 11-13

a,c L; Figure11-6. Momentum Cell B L 4384-non\secl l.wpd-04303 11-14

a,c Figure11-7. Momentum Cell C 4384-non\secl .wpd-04303 1 1-15

4384-non\secl l.wpd-04303 11-16 ATTACHMENTA SBLOCA PIRT Independent Review 4384-non\secl 1.wpd-04303 A-1

PMX INCORPORATED P M x) 209 MAIN STREET P.O. BOX 153 _____._________ NORTHPORT. NY 11768-0153 516-754.4721 FAX 516.754-4727 November 17, 1998 Mr. Bob Kemper J Westinghouse Electric Corporation P.O. Box 355 Pittsburgh, PA 15230-0355

Dear Bob:

Please find enclosed a mini-report that contains a Table of Contents, discussions by the evaluators and the PIRT. This work was perfonned independently by Drs. Dan Speyer, Yassin Hassan, Peter Griffith and Tom Fernandez. They evaluated various events for Small Break LOCA and determined 'X' their rankings for the PIRT. Please note that some of the items, discussed by the evaluators, need analysis and/or evaluation by Westinghouse. Mini-Report Prepared by Approved By

    / P0A~              A-4(-

Paul Malik Dr. Arthur Ginsberg Chairman 4384-non\secl 1.wpd-04303 A-2

SBLOCA PIRT Evaluations Rankings Prepared for Consolidated Edison Nuclear Safety & Licensing Section Indian Point Station Buchanan, New York 10511 November 1998 Compiled by Paul Malik PMX Incorporated 209 Main Street Northport, NY 11768 4384-non\secl I .wpd-04303 A-3

TABLE OF CONTENTS ITEM EVALUATOR PAGE EVALUATIONS CORE Fonner Plate Region Speyer 1 Heat transferpatterns Hassan 4 Rewetl T,,,> Griffith 6 3D Power Distribution Fernandez 7 Top Nozzle/7Te Plate CCFL Fernandez 8 UPPER PLENUM Hot Leg-Downcomer Gaps Fernandez 8 UPPER HEAD Metal Heat Release & Initial Fluid Temp Speyer 1 FUEL ROD Decay Heat Fernandez 7 Local Power Fernandez 7 STEAM GENERATOR AD VSRV Mass Flow & Energy Release Speyer 2 Primary Side Heat Transfer Hassan 4 Secondary Side Heat Transfer Hassan 4 Primary side two-phase *P Griffith 6 PRESSURIZER Interfacial Heat Transfer/Metal Heat Release Speyer 1 PUMP On/Off (Manual Operation) Speyer 2 Mixing Hassan 4 ACCUMULATOR Interfacial Heat Transfer & Metal Heat Release Speyer 2 BREAK CriticalFlow in Complex Geometries Hassan 5 COLD LEG Water Hammer Griffith 6 APPENDIX RANKING TABLES 4384-non\secl I.wpd-04303 A-4

Dan Spever CORE Former Plate Region The former plate region was added since it contains a potentially significant source of water (about 25 % of that residing in the core) and has small drain holes distributed axial (and radially). The modeling of this volume, as regards the draining and refilling processes, may be of greatest (relative) importance during the period after natural circulation and prior to core recovery. UPPER HEAD Metal Heat Release & Initial Fluid Temperature The upper head metal release and initial fluid temperature were added as these effect when the upper head reaches saturation and (thus) acts as a pressurizer. This has importance in larger breaks (due to core stagnation and reverse flow as it flashes), and may also have greater impact on smaller breaks. For small breaks it can effect the system pressure and thus break flow rate, and thus time to cessation of natural circulation. The upper head temperature is established (at steady state) by the inflow (downcomer-to-upper head and peripheral upper plenum-to-upper head flows), and the outflow (central upper head-to-upper plenum flow); and this temperature in turn establishes when the region will flash. Depending on the degree of complexity/3D modeling in the WCOBRA/TRAC code and IP2 model, the upper head-upper plenum region hydraulics may establish a initial temperature between the TCOLD and THOT. The reasonableness of this temperature (compared to plant data and/or other sources) could be part of the code/model V&V and, if not reasonable, require model "adjustment." Of somewhat lesser importance, but similar in effect, is the modeling of the metal mass heat release which will tend to maintain the upper head temperature, as it flashes and cools. This is potentially effected by CRDM fans, etc. PRESSURIZER InterfacialHeat Transfer & Metal Heat Release The metal heat release and interphase heat transfer were added. Although of lesser importance than for pressurizer insurge (increasing pressure) transients, the pressurizer heat transfer model is of greater importance for the slower small break transients, than for large break LOCA in which the pressurizer steam space is essentially subject to an adiabatic expansion. This pressurizer heat transfer includes both the liquid-steam interface and metal-to-liq-uid/steam heat transfer. These will effect the RCS pressure and thus will effect the blowdown phase break flow. 4384-non\secl 1.wpd-04303 A-5

STEAM GENERATOR ADV/SRV Mass Flow & Energy Release (Equipment Available/Operator Action Considerations) The operator actions on AFW (and equipment available) were added as they can effect the small break LOCA and because the emergency operating procedures (EOPs) also include directions for operator actions on SG level. The assumptions as regards M/D and S/D (motor and steam driven) AFW pumps included the number of pumps, and presumably 2 M/D and 1 S/D pump is best estimate, and operator actions to control SG level. The other assumption on equip-ment relates to the SG relief valves, and the best estimate condition would presumably be relief valve available. The latter (SG relief valves) would have an effect on the SG pressure, and thus the RCS pressure and break flow from approximately end of blowdown to loop seal clearing. The assumptions on AFW flow also may effect the results, e.g., the effect on SG fluid temperature, etc. PUMP Pumps ON/OFFwas an area in which the post TMI Westinghouse Owners Group (WOG) analyses established (and NRC accepted), as I recall, about 2 minute operator action. These results, from the time period 1980 to 1982, may be different than current assumptions ( Is best estimate about 2 minutes?). Running the pumps longer maintains a liquid or two-phase mixture at the break, and (thus) when punps are tripped the subsequent core uncovery can be severe; but, on the other hand, other transients are significantly benefitted by RCPs ON--notably SGTR, in which RCP(s) ON allow use of normal PZR spray, which is a significant help for operators to rapidly terminate the SG filling, and lessen likelihood of filling steam lines with water. Is there a possibility that using WCOBRA/TRAC could allow loss of RCPs at longer times and still yield acceptable results? If this were the case (and credit could be taken by allowing RCPs to operate) what would be the RCP trip criteria, and would the calculation of the 95th percentile PCT value be a problem in this case, as it would also consider the mean and variance for operator action times? ACCUMULATOR InterfacialHeat Transfer & Metal Heat Release Although the accumulators are probably of lesser importance, these and other areas (such as pressurizer discussed above) are subject to essentially adiabatic expansion in large LOCAs, where the 4384-non\sccl 1.wpd-04303 A-6

transient time is small; but small LOCAs occur over considerably longer duration and heat transfer is more important. Specifically, where the two phases are a non-condensing gas and subcooled water (as in accumulators) how realistic is the code? Perhaps control model(s) are, or can, be used to describe the differential equation for gas pressure--both for the adiabatic case, and with heat transfer. (I have developed such an approach for RETRAN, but never actually implemented it!) 4384-non\secl 1.wpd-04303 A-7

Yassin Hassan CORE Heat Transfer Patterns can be organized in the order as they are experienced during the accident scenario. It should start with single-phase flow. It is followed by nucleate boiling heat transfer and so on. Then, plant responses are listed. As an example, the table can be listed in this order: Heat Transfer to covered Core. DNB Post-CHF Heat Transfer Radiation Heat Transfer Entraimnent/De-Entrainment Rwet/Tn Mixture Level (This is a plant response) Etc. STEAM GENERATOR PrimarySide Heat Transfer BLD Period During the blowdown period, the tube heat transfer from the primary to the secondary system is important and is ranked high (H). It represents the main mechanism by which the core power is removed from the core. The accurate estimation of this convection heat transfer is important. LSC Period The primary side heat transfer can be divided into two subdivisions: heat transfer and condensation heat transfer. Or, the brackets (condensation) can be omitted. Secondary Side Heat Transfer The secondary side heat transfer is ranked Medium during blowdown and natural circulation periods. The secondary side is a heat sink during these periods. The direction of heat transfer reverses (i.e., secondary-to-primary) following loop seal clearance. Consequently, low ranking is assigned for boiloff and recovery periods. PUMP Mixing Pump mixing is ranked Medium' during the blowdown period. This is due to turbulence induced by flow through reactor coolant impellers following pump coastdown. This influence is limited by the short duration of the coastdown. 4384-non\secl l.wpd-04303 A-8

BREAK Critical Flow in Complex Geometries Break mass flow is ranked High. Choking in complex geometries is also ranked High. Due to the difficulty of estimating the break flow, it is recommended to perform a sensitivity study with bounding calculations. Several parameters should be tested and varied as:

                    - Flow resistance
                    - Upstream flow conditions
                    - Break quality
                    - Spectrum of flow locations
                    - Various critical flow models.

The sensitivity calculations may reduce the ranking during certain phases. 4384-non\secl l.wpd-04303 A-9

Peter Griffith CORE Rewetl T, It would appear that T..,h should be added to this item. However, for the small break LOCA of concern, the process in which T,UC,-h appears is never expected to arise. For quenching a fuel rod from below the rewet temperature or T. ,D are sufficient. Inverted annular flow uses the rewet temperature while dispersed flow film boiling the TnD. T,,C is appropriate for rewetting from above. This never occurs during this transient as the core cannot overheat as long as there water in the upper plenum and there isn't any at this point in the transient. The core always rewets from below in this transient. STEAM GENERATOR Primary side two-phase P The key to allowing the liquid in the core to come to the same level as in the downcommer is to clear at least one of the loop seals. This can only happen when the pressure drop in the steam generators is small enough. A large part of this pressure drop is due to gravity so it is very important to get the pressure in the risers of the steam generators correct. For this reason primary side two phase

          *P is rated High for the loop seal clearing part of the transient.

COLD LEG Water Hammer This concern is rated low for this transient because its effect on the average flows and temperatures into and out of the cold leg are low and fleeting. A condensation induced water hammer occurs over in a small fraction of a second and does not reoccur (if it ever does) for a relatively long time afterwards. To explore the possible consequences of a condensation induced water hammer would entail altering the scenario for this transient. This is beyond the scope of this review. 4384-non\secl I .wpd-04303 A-10

Tom Fernandez FUEL ROD Decay Heat refers to the decay heat model and its uncertainties (range and distribution type) used for the fuel rods. This affects the local heat generation rate for all fuel rods throughout the core, including the hot rod and PCT location. The decay heat model and associated uncertainties are well defined in the 1979 ANS Stand for Decay Heat, independent of other phenomena discussed under Local Power. Decay heat is a primary driver for fuel rod thermal response. Therefore, it is considered an important effect throughout the accident, and is assigned a high (H) ranking for all periods. Local Power (Local Peaking & Relocation) phenomena refer to the axial power shape, linear heat generation rate (especially the Peak LHGR), and potential fuel relocation after accident initiation. The modeling techniques and uncertainties are considered to be distinctly different from decay heat. The first two phenomena affect the initial power distribution in the hot assembly, hot rod, and PCT location. During the accident, the axial power shape affects the mixture level in the two-phase region, and the vapor superheat in the deficient cooling region. The PLHGR affects the magnitude of the PCT. These two phenomena are considered to have medium (M) importance during the Blowdown, Natural Circulation and Loop Seal Clearing periods when the core is generally well cooled for long periods. They are considered to have (H) importance during the Boiloff and Recovery periods when higher peak clad temperature, including the PCT occur. Relocation refers to the potential for fuel to relocate inside the cladding toward the rupture zone after accident initiation. This is considered to be unlikely for SBLOCA conditions since even if cladding rupture occurs, the local cladding strain is expected to be relatively small, asymmetrical, and localized ("warts"). CORE 3D Power Distribution referee to the combined radial and axial -power distribution in the core. First, this affects the initial stored energy in the fuel roads and core internal structures. This alone is expected to have low (L) importance during the Blowdown and Natural Circulation periods for SBLOCAs; however it is considered to have medium importance for IBLOCAs (0.1 At

• 1.0 ftZ where core recovery may occur sooner. Thus, it is ranked a L for these periods to indicate the rank (*)

depends on the scenario. Second, this increasingly affects the core internal 3D circulation, two-phase level, and vapor superheat as the accident progresses through the Loop Seal Clearing, Boiloff, and Recovery periods. Therefore, it is ranked medium (M) during the LSC and High (H) during BO and REC phases. 4384-non\secl l.wpd-04303 A-1lI

Top Nozzle/Tie Plate CCFL refers to the Counter Current Flow Limitation (Liquid down flow limited by vapor upflow) than occur at the top of the core during two-phase conditions. This affects the ability of Liquid to drain gravity back down into the core region to retain a well cooled core as RPV inventory is depleted through the break. This is considered to have low (L) importance during SBLOCA Blowdown and medium (M) importance during the subsequent SBLOCA periods as tow-phase flow conditions and core uncovery become more manifest. It is expected to become more important during the BLD, NC and LSC periods for IBLOCAs since two-phase conditions and core uncovery will occur sooner. Therefore it is ranked L and M' for BLD, NC and LSC periods. UPPER PLENlM Hot Leg-Downcomer Gaps refer to the leakage paths that exist between the hot leg nozzles and upper downcomer region during all operating modes. Physically these represent the small residual radial gaps between the core barrel hot leg nozzle tips and the reactor vessel hot leg nozzle inner surfaces. Their presence, by design, allows the upper plenum shroud/core barrel to be installed and removed. These gaps exist even after differential thermal expansion of the core barrel, relative to the RPV, has occurred at rated operating conditions. These gaps can account for on the order of 1% leakage flow directly the upper downcomer to the hot legs during normal operation. These gaps open up as the reactor is shut down and brought to cold conditions. The radial gap is on the order of 0.1 inch for cold conditions, and about 0.01 to 0.02 inches for hot operating conditions. The hot leg circumference is about 94 inches (w times 30 inches) for each leg. The leakage associated with these gaps can occur during all accident periods; the leakage direction is controlled by the pressure difference between the upper downcomer and inner region of the hot leg nozzles. These leakage paths are expected have a small affect (L) during Blowdown when the system and core flow rates are dominated by other stronger forces (RCP, SG heat sink, break). They are considered to have medium importance during the Natural Circulation periods (as small sneak circuits) that short circuit flow otherwise headed to the core region. They are expected to have High (H) importance during the Loop Seal Clearing period when they provide alternative paths from the upper plenum to the break location to vent some hot core fluid enthalpy and relive some two-phase level depression. Thereafter, they are considered to have increasingly diminished importance of medium (M) during the Boiloff and low (L) during the Recovery periods. 4384-non\secl .wpd-04303 A-12

APPENDIX This appendix contains the PIRT Ranking Tables for Small Break LOCA processes. 4384-non\secl l.wpd-04303 A-13

Ranking Table - PIRT for Small Break Processes Period Process BLD lNC LSC J BO J REC NOTES FUEL ROD I Stored Energy L- L L L L Oxidation L L L H_ H _ Decay Hcat H H H H H TF Local Power (Local Peaking, Relocation) M M M H H Added: TF Clad Deformation (Burst Strain, Temp.) L L L M M Added Gap Conductance L L L L L CORE DNB L L L L L Post-CHF Heat Transfer L L M H H RewetT,, L L M H H PG Heat Transfer to Covered Core M L L L L Radiation Heat Transfer L L L M M Mixture Level M M H H H 3-D Flow/Core Natural Circulation L L L M M Entrainment De-Entrainment L L L M M Flow Resistance L M M L L 3-D Power Distribution L L M H H Added: TF Top Nozzle/Tie Plate CCFL L M M M M Added: TF Former Plate Region L L M M L Added: DS UPPER HEAD _ Draining/Mixture Level - M M L L L Metal Heat Release L L L L L Initial Fluid Temperature M L L L L Added: DS UPPER PLENUM _ Hot Assembly Location LL L L Entrainnient/De-Entrainmrent L L L M L DrainingoFallback/CCFL L W MF M M Mixture Level M M M L L Horizontal Stratification L M M L L Phase Separation at Pressurizer Tee L L L L L Counter-Current Flow & CCFL L H H L L Hot Leg - Downcomer Gap Flow L M H M L Added: TF Condensation N/A L L L L Added Metal Heat Release L L L L L Added 4384-non\sec I .wpd-04303 A-14

R:.k., f -T S r Ranking Table - PIRT for Small Break Processes Period Process BLD NC LSC BO REC NOTES PRESSURIZERISURGE LINE (CL Break) _ = ___ Level Swel/Flashing M L L L L Surge Line Flow/Flooding L L L L L Entrainment/De-Entrainment L L L L L Interface Heat Transfer M L L L L Metal Heat Release (including PZR Heater) M L L L L Added Interface Heat Transfer M L L L L Added: DS STEAM GENERATOR Primary Side Heat Transfer (condensation) H M H M M YH Non-condensable Gas Effects L L L L L CCFL/Tube Voiding L M H L L Prirnary Side 2-Phase AP L M H L L PG Multi-tube Behavior L M M L L Secondary Side Stratification & Recirc. L M L L L Secondary Side Level L M L L L ADVSRV Mass Flow & Energy Release L M L L L DS Tube PlugginglSGTP asymmetry L M M M M Secondary Side Heat Transfer M M L L L YH Metal Heat Release L L L L L Added HOT LEG EntrainmentJDe-Entrainrnent L L L L L PUMP SUCTION PIPINGILOOP SEAL CCFL L L L L L WEST. Entraimnent/Flow Regime/Interfacial Drag L L H M L Horizontal Stratification L L H M L Flow Resistance L L M L L Metal Heat Release L L L L L Added PUMP Mixing M NL N N N 2-Phase Performance M N/A N/A N/A N/A Flow Resistance L M M L L Coastdown Performance M' N/A N/A N/A N/A Friction/Windage Losses M M M L L Added Pump CCFL N/A L M M M Metal Heat Release L L L L L Added 4384-non\secl 1.wpd-04303 A-15

Ranking Table - PIRT for Small Break Processes Period Process BLD NC I LSC BO I REC NOTES ACCUMULATOR Injection Flow Rate N/A N/A N/A N/A M Line Resistance N/A N/A N/A N/A T - Nitrogen Effects N/A N/A N/A N/A L Check Valve Hysteresis N/A N/A N/A N/A L' Dissolved Nitrogen Effects N/A N/A N/A N/A IT Added Interfacial Heat Transfer N/A N/A N/A N/A M Added: DS Metal Heat Release L L L L L Added: DS COLD LEG Condensation (Stratified) N/A L M H H Non-Condensable Effects N/A L L L L Horizontal Stratification/Flow Regine L L H H H Flow Resistance L L L L L Added Water Hammer L L L L L Added: PG Metal Heat Release L L L L L Added SAFETY INJECTION Condensation/ Jet Efficiency N/A L L L L DOWNCOMERfLOWER PLENUM Condensation N/A L L L L Non-Condensable Effects N/A L L L L 3-D Effects M L L L L Mixture Level/FlashingfVoid Fraction M M H H H Enterainment/De-Enterainment L L L L L Flow Resistance L L L L L Added Vessel Metal Wall/RPV Int Heat Release L L L L L Effects _ BREAK =_=_=_= Critical Flow In Complex Geometries H H H H H YH Upstream Flow Regime & Break Quality H H H H H Non-condensable Effects L L L L L Note:

• means that the ranking is "Break Size" dependent.

4384-non\sec I .wpd-04303 A-16

Ranldng Table - PIRT for Small Break Processes Period Process BLD NC I LSC BO REC NOTES PRESSURIZER/SURGE LINE (CL Break) = = =_=_= Level Swell/Flashing -. M L L L L Surge Line Flow/Flooding. L L L L L EntrainmentfDe-Entrainment L L L L L Interface Heat Transfer M L L L L Metal Heat Release (including PZR Heater) M L L L L Added Interface Heat Transfer M L L L L Added: DS STEAM GENERATOR Prirnary Side Heat Transfer (condensation) H M H M M YH Non-condensable Gas Effects L L L L L CCFLlTube Voiding L M H L L Primary Side 2-Phase AP L M H L L PG Multi-tube Behavior L M M L L Secondary Side Stratification & Recirc. L M L L L Secondary Side Level L M L L L ADV/SRV Mass Flow & Energy Release L M L L L DS Tube Plugging/SGTP asymmetry L M M M M Secondary Side Heat Transfer M M L L L YH Metal Heat Release L L L L L Added HOT LEG Entrainment/De-Entrairnent L L L L L PUMP SUCION PIPING/LOOP SEAL _ CCFL L L L L L WEST. Entrainment/Flow Regine/Interfacial Drag L L H M L Horizontal Stratification L L H M L Flow Resistance L L M L L Metal Heat Release L L L L L Added PUMP Mixing . NL N N N 2-Phase Performaance - M N/A N/A N/A N/A Flow Resistance L M M L L Coastdown Performance MW NIA N/A N/A N/A Friction/Windage Losses M M M L L Added Pump CCFL N/A L M M M Metal Heat Release L L L L L Added 4384-non\secl l.wpd-04303 A-17

Ranking Table - PIRT for Small Break Processes _ _ __ __ _ __ __ _ _ _ __ __ _ __ _ _ _ _ _ _ _ _ _P Process BLD NC eriod_ LSC BO REC NOTES ACCUMULATOR Injection Flow Rate N/A N/A N/A N/A M? Line Resistance N/A N/A N/A N/A Nitrogen Effects N/A N/A N/A N/A L Check Valve Hysteresis N/A N/A N/A N/A LW Dissolved Nitrogen Effects N/A N/A N/A N/A L Added Interfacial Heat Transfer N/A N/A N/A N/A M Added: DS Metal Heat Release L L L L L Added: DS COLD LEG Condensation (Stratified) N/A L M H H Non-Condensable Effects N/A L L L L Horizontal Stratification/Flow Regime L L H H H Flow Resistance L L L L L Added Water Hammner L L L L L Added: PG Metal Heat Release L L L L L Added SAFETY INJECTION _ _ Condensationl Jet Efficiency N/A L L L L DOWNCOMER/LOWER PLENUM L Condensation N/A L L L L Non-Condensable Effects N/A L L L L 3-D Effects L L L L Mixture Level/FlashingfVoid Fraction M M H H H Enterainment/De-Enterainment L L L L L Flow Resistance L L L L L Added Vessel Metal Wall/RPV Int Heat Release L L L L L Effects BREAK _ Critical Flow In Complex Geometries H H H H H YH Upstream Flow Regime & Break Quality H H H H H Non-condensable Effects L L L L L Note:

• means that the ranking is "Break Size" dependent.

I,) 4384-non\sec I l.wpd-04303 A-18

WESTINGHOUSE NON-PROPRIETARY CLASS 3 WCAP-14936-NP Volume 2, Rev. 0 Sections 12-23 Code Qualification Document for Best Estimate Small Break LOCA Analysis Volume 2: Small Break Code Validation R. M. Kemper S. M. Bajorek D. C. Golden K. Ohkawa D. J. Shimeck A. K. Muftuoglu D. Paramonov M. Dzodzo M. Y. Young April 2003 Westinghouse Electric Company LLC P.O. Box 355 Pittsburgh, PA 15230-0355 C 2003 Westinghouse Electric Company LLC All Rights Reserved o:\4384-non\Yol2-ftmLwpd:lb-04043

o\4384-non\Vol2-frmLwpd:lb-04033 ii ABSTRACT The document "Code Qualification Document for Best Estimate Loss of Coolant Accident Analysis" (WCAP-12945-P-A) discussed the WCOBRA/TRAC computer code and the methodology used to determine the 95 percentile peak cladding temperature (PCT) for a large break loss of coolant accident (LOCA) scenario. Westinghouse has reviewed the large break code and methodology to determine if the same principles could be adapted to reliably predict the processes that occur in a small break LOCA lasting from several hundred to several thousand seconds. This document, "Code Qualification Document for Best Estimate Small Break LOCA Analysis," (WCAP-14936), describes the WCOBRAITRAC small break LOCA code version, the code validation performed and a methodology to determine the 95' percentile PCT for small break LOCA transients. Volume 1 describes the features, models and correlations contained in the small break LOCA version of the WCOBRAITRAC computer code. First, the small break processes considered to have the greatest effect during a small LOCA event are identified and ranked in the phenomena identification and ranking table (PIRT). The sufficiency of the large break WCOBRAITRAC models and correlations for small LOCA analysis is then evaluated. A comprehensive presentation of the WCOBRAITRAC-SB models and correlations follows. Volume 2 documents simulations of a large number of separate and integral effects tests using this small break version of the code. The simulations provide, at different scales, predicted transients in which all of the important processes are compared with experimental data. The information obtained from the simulations is used to assess errors within the code. The test simulations and subsequent comparison to experimental data determine the bias and uncertainty of major model packages as they apply to small break LOCA thermal-hydraulic conditions. Volume 3 reviews the operator actions pertinent to a small break loss-of-coolant accident (LOCA) event using Indian Point Unit 2, a four-loop pressurized water reactor (PWR), as the reference. Sources of uncertainty in the plant condition and the limiting accident analysis assumptions are identified. The effects of various assumptions on small break LOCA transient behavior are investigated through numerous calculations using ECOBRAfIRAC-SB. The calculations examine the sensitivity of the results to the break size, location, orientation, and offsite power availability. o:\4384-non\vol2-frmtwpd:lb04O33 ill

I Volume 4 presents calculations that are performed to determine the sensitivity of results to the plant core power distribution, the initial and boundary conditions, and code modelling assumptions. These studies, in which parameters are varied one at a time, are performed for Indian Point Unit 2 to quantify the sensitivity of plant behavior to changes in plant initial conditions and accident modelling. An uncertainty methodology consistent with the application of the Code Scaling, Applicability, and Uncertainty (CSAU) methodology is identified to define the overall plant analysis uncertainty and is applied to determine the 95' percentile PCT for the Indian Point Unit 2 small break LOCA analysis. Volume 4 also demonstrates the compliance of the Westinghouse best estimate large break LOCA methodology with U.S. Nuclear Regulatory Commission (NRC) Regulatory Guide 1.157 and with 10CFR50.46. o\4384-non\Vol2-fmnt.wpd:lb-04033 iV

TABLE OF CONTENTS Volume 1 Models and Correlations Volume 2 Small Break Code Validation Volume 3 PWR Uncertainties and Sensitivities for Small Break LOCA Volume 4 Small Break Uncertainty Methodology Section Title Paae ABSTRACT ................................................... iii LIST OF TABLES ................................................... xv LIST OF FIGURES ................................................... xvii LIST OF ACRONYMS .......................................... xxxiii COMMONLY USED EQUATION NOMENCLATURE ...... ........... xxxvii 12 CORE HEAT TRANSFER DURING A SMALL BREAK LOCA 12-1 Introduction .. 12-1 12-2 Physical Processes .. 12-2 12-3 WCOBRAtTRAC-SB Heat Transfer Model .. 12-3 12-3-1 Convective Heat Transfer.12-3 12-3-2 Drop-Wall Contact .12-5 12-3-3 Radiation From Wa to Vapor. .12-5 12-3-4 Mixture Level Sharpening .12-6 12-4 Assessment of WCOBRAIRAC-SB Heat Transfer Model for Small Break LOCA Application . .12-7 12 4-1 ORNL-THTF DFFB Test Simulations .12-8 124-2 INEL Single Tube Heat Transfer Experiments .12-10 124-3 ORNL Uncovered Bundle Heat Transfer Test Simulation .12-11 12-5 Summary and Conclusions .. 12-12 12-6 References .. 12-13 13 ASSESSMENT OF BREAK FIJOW MODEL 13-1 Introduction .............................. 13-1 13-2 Critical Flow in Small Break LOCA .............................. 13-1 o:\4384-non\Vol2-frmtwpd:lb-04033 v

I TABLE OF CONTENTS (Cont'd) Section Title Page 13-2-1 Subcooled Liquid Discharge ............................... 13-2 13-2-2 Stratified Entrainment at Break .......... ................... 13-2 13-2-3 Correlation for Onset of Liquid and Vapor Entrainment .. ....... 13-3 13-24 Correlation for Break/Branchline Quality . . 134 13-3 Assessment of Horizontal Stratified Entrainment Model ............... 13-10 13-3-1 Branchline Quality/Mainline Liquid Level Comparison Using TPFL ......................... 13-10 13-3-1-1 Description of Test Facility ...... ................ 13-11 13-3-1-2 Test Ranges ........... ....................... 13-11 13-3-1-3 WCOBRAJIRAC Model ....... ................ 13-11 13-3-14 Comparison of WCOBRA/TRAC Prediction to Horizontal Data ......... ...................... 13-11 13-3-1-5 Comparison of WCOBRA/TRAC Prediction to Downward-Vertical Data ....... ................ 13-12 13-3-1-6 Comparison of WCOBRA/TRAC Prediction to Upward-Vertical Data ....... ................... 13-12 134 Assessment of WCOBRA/TRAC Break Flow Model .................. 13-19 134-1 Assessment Objective ................................... 13-19 134-2 Assessment Test Matrix ................................. 13-19 134-3 Assessment Results ..................................... 13-42 134-3-1 Bias and Uncertainty ........................... 1342 134-3-2 Model Prediction with Respect to Pressure ..... .... 1345 13-4-3-3 Model Prediction with Respect to Quality ..... ..... 1345 1343-4 Model Prediction Trend with Respect to Channel Length ...................................... 1348 134-3-5 Model Prediction Trend with Respect to Hydraulic Diameter ................. ................... 13-50 134-3-6 Model Prediction Trend with Respect to LID ..... ... 13-51 134-3-7 Influence of Upstream Void Fraction .............. 13-54 134-3-8 Influence of Two-Phase Multiplier ....... ......... 13-54 134-3-9 Influence of Mesh Size ......................... 13-55 134-3-10 Influence of Fraction Factor/Entrance Effect .. 13-55 o:\4384-non\Vol2-frmLwpd:lb-04033 vi

TABLE OF CONTENTS (Cont'd) Section Title Page 13-4-3-11 Critical Flow Predictions for Individual Dataset ..... 13-56 13-5 Scaling Consideration .......................................... 13-84 13-5-1 Pressure, Subcooling, and Quality .......... ............... 13-84 13-5-2 Break Flow Area ........................ 13-84 13-5-3 Break Geometry ....................................... 13-84 13-5-4 Pressure Effect on the Onset of Entrainment and Branchline Quality ............................................... 13-84 13-5-5 Mainline Pipe Diameter Variation on the Onset of Entrainment and Branchline Quality .......... .............. 13-84 13-6 Conclusions ................................................. 13-85 13-7 References .................................................. 13-85 14 SAFETY INJECTION JET CONDENSATION: COSI EXPERIMENTS 14-1 Introduction ................ .................. 14-1 14-2 Description of COSI ................ .................. 14-2 14-2-1 Facility Description .................................. 14-2 14-2-2 Key Phenomena . .................................. 14-2 14-2-3 Applicable Tests and Parameter Ranges ......... ............. 14-3 14-3 Description of WCOBRA[IRAC Model ............................. 14-3 14-4 Simulations ................ .................. 14-4 14-4-1 Summary of Experimental Results ............ .............. 14-4 14-4-2 WCOBRAJfRAC-SB Results .............................. 14-5 14-5 Conclusions ................ .................. 14-7 14-6 References .. 14-7 15 MIXTURE LEVEL SWELL 15-1 Introduction . 15-1 15-2 Physical Processes . 15-2 15-3 WCOBRAJTRAC Determination of the Mixture Level . 15-2 154 Assessment of WCOBRA/TRAC Mixture Level Predictions . 15-3 oA4384-non\Vo2-frm.Lwpd:1b-44033 ..

TABLE OF CONTENTS (Cont'd) Section Title Page 15-4-1 Introduction .......................... 15-3 15-4-2 ORNL-THTF Small Break Tests ........................... 15-4 15-4-2-1 Introduction ................................ 15-4 15-4-2-2 WCOBRA/TRAC Model of the ORNL-THTF ....... 15-4 15-4-2-3 Test Matrix for ORNL-THTF Simulations ..... ...... 15-5 154-2-4 Simulation of ORNL-THTF Tests ....... .......... 15-5 15-4-2-5 Summary and Conclusions ........... ............ 15-6 154-3 Simulation of G-1 Core Uncovery Tests ..................... 15-29 15-4-3-1 Introduction ............................ 15-29 15-4-3-2 WCOBRA/TRAC Model of G-1 Test Facility ....... 15-29 15-4-3-3 Test Matrix for G-1 Uncovery Tests . 15-30 15-4-3-4 Simulation of G-1 Core Uncovery Tests . 15-30 154-3-5 Discussion of Results. 15-31 154-3-6 Summary and Conclusions. 15-32 15-4-4 GE Vessel Blowdown Tests . 15-45 -1, 15-44-1 Introduction. 15-45 15-44-2 WCOBRAITRAC Model for GE Vessel Blowdown Tests . 15-46 15-44-3 Test Matrix for Simulations. 15-46 15 44 Simulation of GE Vessel Blowdown Tests . 15-46 15-4-4-5 Effect of Interfacial Drag Multiplier . 15-47 154.4-6 Summary and Conclusions. 15-47 15-5 Summary and Conclusions. 15-71 15-6 References. 15-71 16 LOOP SEAL CLEARANCE 16-1 Introduction ......................... ............ 16-1 16-2 Important Physical Process and Scaling Laws ....... .................. 16-2 16-2-1 Westinghouse Loop Seal Tests . . .16-3 16-2-1-1 Test Facility Description .16-3 16-2-1-2 Test Procedures .16-4 o:\4384-non\Vol2-frmLwpd:lb-04033

TABLE OF CONTENTS (Cont'd) Section Title Page 16-2-1-3 Analysis of 1/3-Scale Test Results ................. 16-6 16-2-1-4 Effect of Scale .......... ............. 16-10 16-2-1-5 Full-Scale Steam-Water Tests .................... 16-11 16-2-1-6 U-Tube Oscillation Effects ...................... 16-11 16-3 WCOBRA[IRAC Modelling of Loop Seal Clearing Process ............ 16-12 16-4 WCOBRA/IRAC Simulation of the UPTF 3-Bar and 15-Bar Tests ...... 16-12 16-5 Conclusions ................................................ 16-14 16-6 References ................................................ 16-15 17 STEAM GENERATOR REGION HYDRAULICS MODELLING 17-1 Introduction .. .. 17-1 17-2 Physical Processes .. 17-2 17-3 CCFL Modelling in WCOBRATRAC-SB ........................... 17-3 17-3-1 Introduction ........................................... 17-3 17-3-2 CCFL in a Vertical Channel ............................... 17-3 17-3-2-1 Vertical WCOBRAJIRAC Channel Model ..... ..... 17-3 17-3-2-2 Predicted CCFL at High Pressure ...... ............ 17-6 17-3-3 CCFL in a Perforated Plate ................................ 17-6 17-3-3-1 Correlations and Scaling for CCFL in a Perforated Plate ......... ................... 17-7 17-3-3-2 WCOBRAfIRAC Results ....... ............... 17-10 17-3-3-3 WCOBRAtTRAC MOD7 Results ...... .......... 17-12 17-3-3-4 Comparison With Data ........ ................. 17-12 17-3-4 CCFL in a Horizontal Channel . . ............ 17-12 17-3-4-1 WCOBRAJTRAC Simulation of Horizontal Flow .... 17-13 17-3-4-2 Relation of Flooding Correlations to Slug Flow Regime Transition Models ....... ............... 17-13 17-3-4-3 Predicted Horizontal CCFL at High Pressure ..... ... 17-17 17-3-4-4 Predicted Water Level ......... ................. 17-18 174 CCFL in Hot Leg-to-Steam Generator Flow Path ..................... 17-18 174-1 Introduction .......................... 17-18 o:\4384-nonwVol2-frmLwpd:lb-04033 1X

I TABLE OF CONTENTS (Cont'd) Section Title Page 174-2 Physical Processes .................. ............. 17-19 174-2-1 Small-Scale Tests ............................. 17-19 174-2-2 Large-Scale Tests ............................. 17-20 174-3 Conclusion .......... ..................... 17-20 17-5 WCOBRA/TRAC-SB Modelling of Wall Condensation ..... .......... 17-21 17-6 Steam Generator Tube Condensation ............ .................. 17-21 17-7 Simulation of Semiscale MOD-2A NC Test Series Experiments ......... 17-22 17-7-1 Introduction ............................... 17-22 17-7-1-1 Natural Circulation Phenomena ........ .......... 17-22 17-7-1-2 Applicable Tests .............................. 17-23 17-7-2 Description of WCOBRA/TRAC-SB Model ....... .......... 17-23 17-7-3 Simulation Results . ............................... 17-24 17-74 Conclusions ............................... 17-25 17-8 References ... ... 17-26 18 HORIZONTAL STRATIFIED FLOW BENCHMARKS 18-1 Introduction ................... ...................... 18-1 18-2 Physical Processes ......................................... 18-1 18-3 WCOBRA/TRAC-SB Horizontal Stratified Flow Models ..... .......... 18-2 18-3-1 Interfacial Drag ......................................... 18-2 18-3-2 Entrainment ......................................... 18-2 18-3-3 Condensation ......................................... 184 184 Assessment of WCOBRAIIRAC-SB Horizontal Stratified Flow Models .. 18-6 184-1 Test Facility Description and Modelling ....... ............... 18-6 184-2 Calculational Results ..................................... 18-8 18-5 Conclusions ................. ........................ 18-10 18-6 References ................ ......................... 18-10 19 ROSA TEST SIMULATIONS 19-1 Introduction . 19-1 o:\4384-non\VoI2-finwpd:lb.04033 x

TABLE OF CONTENTS (Cont'd) Section Title Page 19-2 WCOBRAITRAC Model of the LSTF .................. ............ 19-2 19-3 Steady-State Simulation .............................. ............ 19-4 19-4 Simulation of SB-CL-05, 5-Percent Cold Leg Break ....... ............ 19-4 19-5 Simulation of SB-CL-09, 10-Percent Cold Leg Side Break .. ............ 19-7 19-6 Simulation of 2.5-Percent Cold Leg Breaks .............. ............ 19-8 19-7 Summary of Results ................................. ........... 19-11 19-8 References ........................................ ........... 19-12 20 LOFT SIMULATIONS USING WCOBRA/TRAC-SB 20-1 Introduction ................................................... 20-1 20-2 LOFT Facility Description ........................................ 20-2 20-3 WCOBRA/IRAC-SB Model for Simulation of LOFT Small Break LOCEs L3-1, L3-7, and L3-5 ...................................... 20-7 20-3-1 Reactor Vessel Modelling ................................. 20-7 20-3-2 Active Loop Hot Leg, Pressurizer, and Steam Generator Inlet Piping Modelling .......................... 20-8 20-3-3 Active Loop Steam Generator Modelling ............ 20-8 20-34 Active Loop Pump Suction Piping and RCP Modelling ......... 20-3-5 Active Loop Cold Leg Modelling .................. 20-10 20-3-6 Accumulator and ECCS Modelling ................ 20-10 20-3-7 Inactive Loop Modelling ......................... 20-10 20-3-8 Break Modelling ............................... 20-11 20-4 Steady-State Simulations for LOFT Small Break LOCEs L3-1, L3-7, and L3-5 . 20-21 204-1 Steady-State Simulation for LOFT LOCE 13-1 . 20-21 20-4-2 Steady-State Simulation for LOFT LOCE L3-7 . 20-22 204-3 Steady-State Simulation for LOFT LOCE 13-5 . 20-23 20-5 Transient Simulations for LOFT LOCEs L3-1, L3-7, and L3-5 . 20-27 20-5-1 Transient Simulation for LOFT LOCE L3-1 . 20-27 20-5-2 Transient Simulation for LOFT LOCE 13-7 . 20-38 20-5-3 Transient Simulation for LOFT LOCE 13-5 . 20-46 o:\4384-non\Vol2-frmLwpd:lb-04033 xi

TABLE OF CONTENTS (Cont'd) Section Title Pa2e 20-6 Conclusions ............ 20-57 20-7 References ............ 20-57 21 SIMULATION OF SEMISCALE SMALL BREAK LOCA EXPERIMENTS 21-1 Introduction .21-1 21-2 Key Phenomena .21-1 21-3 Applicable Tests .21-2 214 Facility Configuration for Tests S-LH-1 and S-LH-2 .21-3 21-5 Description of WCOBRA,TRAC-SB Model .214 21-6 Steady-State Simulations .21-7 21-7 Transient Simulations .21-7 21-7-1 S-LH-1 Simulation Results .21-8 21-7-2 S-LH-2 Simulation Results .21-11 21-8 Conclusions .21-13 21-9 References .21-13 22 NUCLEAR ROD AND COMPONENT MODEL ASSESSMENT 22-1 Nuclear Fuel Rod Model .22-1 22-1-1 Introduction .22-1 22-1-2 Fuel Rod Model Assessment .22-2 22-1-3 NRU Test Description .224 22-14 NRU Test Bundle Description .22-5 22-1-5 WCOBRAfRAC Model of NRU .22-6 22-1-6 Simulation of NRU Test MT-3.06 .22-7 22-1-7 Simulation of NRU Test PTH-1 10 .22-9 22-1-8 Summary and Conclusions .22-9 22-2 Accumulator Component .. 22-25 22-2-1 Introduction .22-25 22-2-2 Indian Point Unit 2 Accumulator Test .22-25 22-2-3 WCOBRAITRAC Model .22-25 o:\4384-non\Vo12-frxnt.wpd:Ib-O4033

TABLE OF CONTENTS (Cont'd) Section Title Page 22-2-4 WCOBRAITRAC Model With PWR Line Noding ....... 22-26 22-2-5 Nitrogen Model Switching and Accumulator Noding ..... 22-26 22-3 Pump Component Model ................................... 22-37 22-3-1 Westinghouse Pump Data ........................... 22-37 22-3-1-1 Single-Phase Data ........................ 22-38 22-3-1-2 Two-Phase Data ......................... 22-39 22-3-2 Pump Model Comparison to Data ..................... 22-42 22-4 References .............................................. 22-55 23 CODE ASSESSMENT

SUMMARY

AND CONCLUSIONS 23-1 Introduction .......................................... 23-1 23-2 Separate Effects Test Simulations ......................... 23-2 23-3 Integral Test Facility Simulations ......................... 23-3 23-4 Nodilization Consistency ................................ ......... 23-5 Conclusions .......................................... 23-9 o-\4384-non\Vol2frint.wpd:lb-04033 Xlll

o\4384-non\Vol2-frznLwpd:lb-04033 XiV LIST OF TABLES Table Title Pa2e 12-1 ORNL-THTF Steady-State DFFB Tests . ..... .12-14 12-2 Summary of ORNLTHTF Driver-Plotter Comparison, Forslund/Rohsenow Model ..................................... ..... 12-15 12-3 ORNL Uncovered Bundle Test Matrix . ..... 12-16 13-2-1 Experimental Results of C, and C2 . .......................... ......... 13-7 13-4-1 Selected Dataset and Input Variables ........ 13-20 13-4-2 Critical Flow Data Considered for Model Evaluation ........ 13-21 13-4-3 Marviken Test Matrix ........ 13-35 13-4-4 Critical Flow Data Comparison for WCOBRA-TRAC Critical Flow Model .... 13-43 13-4-5 Prediction Sensitivity to the Initial Void Fraction ........ 13-54 13-4-6 Prediction Sensitivity to Mesh Size ........ 13-55 14-1 Summary of Applicable COSI Experiments ........ 14-8 14-2 Comparison of the Experimental and Numerical Results for the Case With Break Diameter = 0.22 Inches and Weir HID = 0.5 .14-9 14-3 Comparison of the Experimental and Numerical Results for the Case With Break Diameter = 0.90 Inches and Weir H/D = 0.5 . 14-10 14-4 Comparison of the Experimental and Numerical Results for the Case With Break Diameter = 0.90 Inches and Without Weir (H/D = 0.0) .14-11 15-4-2-1 ORNL-THTF Test Simulation Matrix ................................ ... 15-8 15-4-2-2 Summary of ORNL-THTF Simulation Results ......................... ... 15-9 15-4-2-3 Summary of ORNL-THTF Simulation Results With YDRAG = 0.8. .. 15-10 15-4-24 YDRAG Values to Match ORNL-THTF Data ......................... .. 15-11 15-4-3-1 Core Uncovery Test Matrix ........................................ .. 15-33 15-4-3-2 G-1 Simulation Results Summary ................................... .. 15-34 15-4-3-3 YDRAG Values to Match G-1 Level Swell Data ....................... .. 15-35 15-4-4-1 Summary of Test Parameters for Small Blowdown Vessel Steam Blowdown Tests ................................................ .. 15-49 15-4-4-2 Characterization of WCOBRAIrRAC-SB Results Versus Test Data. .. 15-50 o\4384-non\Vol2-frmtwpd:lb-04033 xv

I LIST OF TABLES (Cont'd) Table Title Page 17-1 Water Injection Rates ....................................... ....... 17-30 17-2 Steam Injection Rates ....................................... ....... 17-30 17-3 Predicted Water Levels in WCOBRAITRAC Horizontal Channel Compared With Weir Flow Theory ............................. ....... 17-31 18-1 WCOBRAITRAC-SB 3-D Interfacial Heat Transfer Models ......... ....... 18-12 18-2 Text Matrix Parameters ...................................... ....... 18-13 19-1 Major Design Characteristics of LSTF and PWR .................. ....... 19-13 19-2 Steady-State Parameter Checklist .............................. ....... 19-14 19-3 Operational Setpoints for Run SB-CL-05 ........................ ....... 19-15 19-4 Transient Results Summary for 5-Percent Cold Leg Side Break ...... ....... 19-16 19-5 Operational Setpoints for Run SB-CL-09 ........................ ....... 19-17 19-6 Chronology of Events for Run SB-CL-09, 10-Percent Cold Leg Side Break . . .. 19-18 19-7 Chronology of Events for Run SB-CL-01, 2.5-Percent Cold Leg Side Break .. 19-19 19-8 Chronology of Events for Run SB-CL-02, 2.5-Percent Cold Leg Bottom Break ................................................... .. 19-20 19-9 Chronology of Events for Run SB-CL-03, 2.5-Percent Cold Leg Top Break . . .. 19-21 19-10 2.5-Percent Cold Leg Break Loop Seal Venting Times .................. .. 12-22 20-4-1 Comparison of LOFT LOCE L3-1 Steady-State Calculation to L3-1 Data 20-24 20-4-2 Comparison of LOFI LOCE L3-7 Steady-State Calculation to L3-7 Data 20-25 2043 Comparison of LOFT LOCE L3-5 Steady-State Calculation to L3-5 Data 20-26 20-5-1-1 Sequence of Events for LOFT LOCE L3-1 (Bayless, et al., 1980) ...... 20-29 20-5 1 Sequence of Events for LOFT LOCE L3-7 (Gillas and Carpenter, 1980) 20-40 20-5-3-1 Sequence of Events for LOFT LOCE L3-5 (Dao and Carpenter, 1980) .. 20-49 21-1 Comparison of Steady-State Conditions: Test S-LH-1. .. 21-15 21-2 Comparison of Steady-State Conditions: Test S-LH-2. .. 21-16 22-1-1 NRU Test MT-3.06 Rod Failure Data Comparison ..................... .. 22-10 o:\4384-non\Vol2-fritwpd:Ib-04163 xvi

LIST OF FIGURES Figure Title Page 12-1 Heat Transfer Regime Map for Vessel Component ....................... 12-17 12-2 WCOBRAITRAC Heat Transfer Driver-Plotter Routine ..... ............. 12-18 12-3 ORNL-THTF Rod Bundle Cross Section .............................. 12-19 12-4 Comparison of Large Break LOCA WCOBRAfIRAC Code Logic Predicted and Measured HTCs for ORNL-THTF Steady-State Tests - Bundle Average .................................................. 12-20 12-5 Comparison of Large Break LOCA WCOBRAJIRAC Code Logic Predicted and Measured HTCs for ORNL-THTF Steady-State Tests - All Thermocouples .. 12-21 12-6 Comparison of Predicted and Measured INEL Film Boiling HTCs as a Function of Vapor Re Using WCOBRA/TRAC-SB Models ... 12-22 12-7 Comparison of Predicted and Measured HTCs for Combined ORNL and NEL Data Using WCOBRAJTRAC-SB Heat Transfer Models .. 12-23 12-8 Comparison of Predicted and Measured HTCs for ORNL-THTF Uncovered Bundle Data Using WCOBRAfTRAC-SB ........ ..................... 12-24 13-2-1 Diagram of Subcooled Break Flow ........... ......................... 13-8 13-2-2 Diagram of Two-Phase Upstream Conditions ....... ..................... 13-8 13-2-3 Vapor Pull-Through and Liquid Entrainment Phenomena ..... ............. 13-9 13-3-1 DiagramofTPFL ................................................. 13-13 13-3-2 Schematic View of TPFL Test Section ......... ....................... 13-14 13-3-3 WCOBRAITRAC Noding for TPFL Branchline Quality Test Simulation ..... 13-15 13-3-4 Branchline Quality Versus Mainline Liquid Level for Horizontal Configuration .................................................... 13-16 13-3-5 Branchline Quality Versus Mainline Liquid Level for Downward-Vertical Configuration ................................................... 13-17 13-3-6 Branchline Quality Versus Mainline Liquid Level for Upward-Vertical Configuration ................................................... 13-18 13-4-la Upstream Condition in Ardron-Ackerman ........ ..................... 13-23 13-4-lb Upstream Condition in Ardron-Ackerman ........ ..................... 13-23 13-4-2a Upstream Conditions in Boivin ............. ......................... 13-24 o:\4384-non\Vol2-frmt.wpd:lb-04033 xvii

LIST OF FIGURES (Cont'd) Figure Title Page 134-2b Upstream Conditions in Boivin ......................  ; 13-25 134-3a Upstream Conditions in Fincke-Collins .................. 13-26 13-4-3b Upstream Conditions in Fincke-Collins .................. 13-26 1344a Upstream Conditions in Jeandy ........................ 13-27 134-4b Upstream Conditions in Jeandy ........................ 13-28 13-4-5a Upstream Conditions in Neusen ....................... 13-29 13-4-5b Upstream Conditions in Neusen ....................... 13-29 134-6a Upstream Conditions in Reocreux ...................... 13-30 13-4-6b Upstream Conditions in Reocreux ...................... 13-31 134-7a Upstream Conditions in Seynhaeve ..................... 13-32 13-4-7b Upstream Conditions in Seynhaeve ..................... 13-32 13-4-8a Upstream Conditions in Sozzi-Sutherland ................ 13-33 13-4-8b Upstream Conditions in Sozzi-Sutherland ................ 13-34 13-4-9a Upstream Conditions in Marviken Tests 6, 7, 23, and 24 .... 13-36 13-4-9b Upstream Conditions in Marviken Tests 6, 7, 23, and 24 13-36 ., 13-4-10a Upstream Conditions in Amos-Schrock ............ 13-37 13-4-lOb Upstream Conditions in Amos-Schrock ............ 13-38 134-1 la Upstream Conditions in TPFL .................... 13-39 134-1 lb Upstream Conditions in TPFL .................... 13-39 134-12a Upstream Condition in Test Matrix ................ 13-40 134-12b Upstream Condition in Test Matrix ................ 13-41 134-13 Predicted and Measured Critical Flows ............. 13-44 134-14 Prediction Trend in Pressure Variation ............. 13-45 13-4-15a Prediction Trend in Quality Variation .............. 1346 ... 13-4-15b Prediction Trend in Quality Variation .............. 13-47 ... 134-16a Prediction Trend in Channel Variation in Linear Scale. 1348 ... 134-16b Prediction Trend in Channel Length Variation in Log Scale .......... 13-49 134-17a Prediction Trend in Channel Diameter in Linear Scale .............. 13-50 134-17b Prediction Trend in Channel Diameter in Log Scale ................ 13-51 134-18a Prediction Trend in Channel LID Variation - Linear Scale ........... 13-52 134-18b Prediction Trend in Channel LID Variation - Log Scale ............. 13-53 134-19 Prediction Comparison with Ardron-Ackerman Data ............... 13-57 o:\4384-nonXVol2-frmt.wpd:1b04033 xviii

LIST OF FIGURES (Cont'd) Figure Title Page 13-4-20 Prediction Comparison with Boivin Data .13-58 13-4-21 Prediction Comparison with Fincke Data .13-60 13-4-22 Prediction Comparison with Jeandey Data .13-62 13-4-23 Prediction Comparison with Neusen Data .13-64 134-24 Prediction Comparison with Reocreux Data .13-65 13-4-25 Prediction Comparison with Seynhaeve Data .13-67 13-4-26 Prediction Comparison with Sozzi-Sutherland Data .13-75 13-4-27 Prediction Comparison with Marviken Data .13-79 134-28 Prediction Comparison with Amos-Schrock Data .13-81 13-4-29 Prediction Comparison with TPFL Data .13-83 14-1 COSI Facility Arrangement .14-12 14-2 Test Section Arrangement .14-13 14-3 COSI WCOBRARAC Model Component Layout .14-14 14-4 COSI Main Test Section and Downcomer WCOBRA/TRAC Model .14-15 14-5 COSI Cold Leg Pipe Vertical Cell Nodalization .14-16 14-6 Depiction of Flow Patterns in the Test Section as Deduced from Data .14-17 14-7 Condensation Efficiency for Small Injection Pipe (d = 0.22 Inches) and Weir (H/D = 0.5) .14-18 14-8 Condensation Efficiency for Large Injection Pipe (d = 0.90 Inches) and Weir (H/D = 0.5) .14-19 14-9 Condensation Efficiency for Large Injection Pipe (d = 0.90 Inches) and Without Weir (HID = 0). 14-20 15-4-2-1 Cross Section of the ORNL-THTF Test Bundle ............. 1............ 5-12 15-4-2-2 Axial View of the ORNL-THTF Test Bundle ................. .......... 15-13 15-4-2-3 WCOBRAfIRAC Model of the ORNLTHTF ............... ........... 15-14 15-4-2-4 Comparison of Predicted and Measured Void Profiles for ORNL-THTF Test 3.09.10 .15-15 15-4-2-5 Comparison of Predicted and Measured Void Profiles for ORNL-THTF Test 3.09.1OJ .15-16 o.\4384-non\Vol2-frmt.wpd:lb-04033 XIX

I LIST OF FIGURES (Cont'd) Figgre Title Page 154-2-6 Comparison of Predicted and Measured Void Profiles for ORNL-THTF Test 3.09.10K ..................................................... 15-17 154-2-7 Comparison of Predicted and Measured Void Profiles for ORNL-THTF Test 3.09.10L . ................................................... 15-18 154-2-8 Comparison of Predicted and Measured Void Profiles for ORNL-THTF Test 3.09.1OM ......... 15-19 154-2-9 Comparison of Predicted and Measured Void Profiles for ORNL-THTF Test 3.09.1ON ......... 15-20 154-2-10 Comparison of Predicted and Measured Void Profiles for ORNL-THTF Test 3.09.10AA ......... 15-21 154-2-11 Comparison of Predicted and Measured Void Profiles for ORNL-THTF Test 3.09.1OBB . .................................................. 15-22 154-2-12 Comparison of Predicted and Measured Void Profiles for ORNL-THTF Test 3.09.10CC .................................................. 15-23 154-2-13 Comparison of Predicted and Measured Void Profiles for ORNL-THTF Test 3.09.1ODD . .................................................. 15-24 154-2-14 Comparison of Predicted and Measured Void Profiles for ORNL-THTF Test 3.09.1OEE . .................................................. 15-25 154-2-15 Comparison of Predicted and Measured Void Profiles for ORNL-THTF Test 3.09.1OFF ................................................... 15-26 1542-16 Comparison of Predicted and Measured Mixture Levels for ORNL-THTF Tests, YDRAG = 0.8 ............. ................................. 15-27 1542-17 Comparison of Predicted and Measured Collapsed Liquid Levels for ORNL-THTF Tests, YDRAG = 0.8 ................................... 15-28 1543-1 Westinghouse ECCS High Pressure Test Facility (G-1 Loop) ..... ......... 15-36 154.3-2 G-1 Core Uncovery Test Heater Rod Bundle ........................... 15-37 154.3-3 WCOBRABIRAC Model of the G-l Test Bundle ........................ 15-38 154.34 Collapsed Liquid Level and Predicted Cladding Temperatures at the 8- and 10-foot Elevations, G-1 Run 63 ................................ 15-39 154.3-5 Comparison of Predicted and Measured Uncovery Times, YDRAG = 0.8 ..... 1540 154.3-6 Comparison of Predicted and Measured Level Swells at Uncovery, YDRAG=0.8 . .................................................. 15-41 o:.4384-nonNVoI2-frmLwpd:lb4033 X)C

LIST OF FIGURES (Cont'd) Figure Title Page 15-4-3-7 WCOBRA,TRAC-SB YDRAG Value to Match Measured Level Swell Versus Bundle Power .............................................. 15-42 15-4-3-8 WCOBRA/TRAC-SB YDRAG Value to Match Measured Level Swell Versus Pressure .............. .................................... 15-43 154-3-9 WCOBRA/TRAC-SB YDRAG Value to Match Measured Level Swell Versus Bundle Elevation . 15-44 154-4-1 Small Blowdown Vessel . 15-51 154-4-2 Small Blowdown Vessel Instrumentation. 15-52 154-4-3 WCOBRA/TRAC Model of the GE Vessel Blowdown Facility . 15-53 154-4-4 Comparison of Predicted and Measured Vessel Pressure, Test 8-21-1. 15-54 15-4-4-5 Comparison of Predicted and Measured Vessel Level, Test 8-21-1. 15-55 154-4-6 Comparison of Predicted and Measured Vessel Pressure, Test 8-25-1. 15-56 154-4-7 Comparison of Predicted and Measured Vessel Level, Test 8-25-1. 15-57 1544-8 Comparison of Predicted and Measured Vessel Pressure, Test 8-28-1. 15-58 154-4-9 Comparison of Predicted and Measured Vessel Level, Test 8-28-1 ... ....... 15-59 15-44-10 Comparison of Predicted and Measured Vessel Pressure, Test 9-1-1 ... ...... 15-60 15-44-11 Comparison of Predicted and Measured Vessel Level, Test 9-1-1 ... ........ 15-61 154-4-12 Comparison of Predicted and Measured Vessel Pressure, Test 9-15-1 ... ..... 15-62 1544-13 Comparison of Predicted and Measured Vessel Level, Test 9-15-1 ... ....... 15-63 1544-14 Comparison of Predicted and Measured Vessel Pressure, Test 1004-3 ... ..... 15-64 154-4-15 Comparison of Predicted and Measured Vessel Level, Test 1004-3 ... ....... 15-65 15-4-4-16 Comparison of Predicted and Measured Vessel Pressure, Test 1004-2 ... ..... 15-66 15-44-17 Comparison of Predicted and Measured Vessel Level, Test 1004-2 ... ....... 15-67 1544-18 Effect of YDRAG Multiplier on the Pressure Prediction, Test 8-28-1 ... ..... 15-68 1544-19 Effect of YDRAG Multiplier on the Pressure Prediction, Test 9-1-1 ... ...... 15-69 1544-20 Effect of YDRAG Multiplier on the Pressure Prediction, Test 1004-3 ... ..... 15-70 16-1 Loop Seal Clearing and Refilling ............................... ...... 16-17 16-2 Loop Seal Clearing Process ................................... ...... 16-18 16-3 1/3-Scale U-Tube Test Facility ................................ ...... 16-19 16-4 Taitel-Dukler Flow Regime Map, Comparing 1/3-Scale Pipe at 14.7 psia and Full-Scale Pipe at 1000 psia (Taitel and Dukler, 1976) ... ...... 16-20 oA4384-non%Vol2-frnLwpd:I b-04033

LIST OF FIGURES (Cont'd) Figure Title Page 16-5 1/3-Scale U-Tube Residual Water Level Remaining After Test as a Function of Test Gas Flowrate ......... ...................... 16-21 16-6 1/3-Scale U-Tube Flow Regimes Observed Under the Limit Line ..... 16-22 16-7 113-Scale U-Tube Horizontal and Vertical Leg Average Void Fractions During Test .................. 16-23 16-8 1/3-Scale U-Tube Horizontal Average Void Fraction During Test Compared with Average Void Fraction After Test .16-24 16-9 Pressure Difference Across the 1/3-Scale U-Tube .16-25 16-10 1/3-Scale U-Tube Normalized Level and Lirit Lines ..................... 16-26 16-11 Hysteresis in Loop Seal Limit Line .16-27 16-12 Effect of Increased Geometric Scale on Limit Lines .16-28 16-13 Effect of Increased Pressure and Scale on Limit Lines .16-29 16-14 IVO Full-Scale Final Void Fraction and Limit Lines .16-3.0 16-15 UPTF Facility and Single Loop Seal (Liebert and Emmerling, 1998) .16-31 16-16 Lines of Constant Gas Velocity Compared to UPTF Data for 3-Bar and 15-Bar Loop Seal Tests (Liebert and Emmerling, 1998) ..... .......... 16-32 16-17 UPTF and PWS Compared to the Ishii Correlation and Data Base (Ishii and Grolmes, 1975) ................. 16-33 16-18 WCOBRAITRAC Model of the UPTF Separate Effects Loop Seal Clearing Tests .16-34 16-19 Calculated Residual Levels versus UPTF 3-Bar Data .16-35 16-20 Calculated Level for 3-Bar Test at a Gas Superficial Velocity of 5.7 ft/s .16-36 16-21 Calculated Liquid Level in Steam Generator Downhill Pipe and Sump Suction Pipe for a Superficial Gas Velocity of 5.7 ft/s .16-37 16-22 Comparison of WCOBRA/TRAC-SB Calculations and UPTF Data for the 3-Bar Tests .16-38 16-23 Calculated Final Residual versus UPTF 15-Bar UPTF .16-39 16-24 Comparison of WCOBRAfrRAC-SB Calculations and UPTF Data for the 15-Bar Tests .16-40 16-25 Calculated Residual Levels for 1015 psia .1641 o\434-non\Vol2-ftmLwpd:lb-04033 xxii

LIST OF FIGURES (Cont'd) Figure Title Page 16-26a Measured Pressure Drop for UPTF 3-Bar and 15-Bar Loop Seal Tests (from Libert and Emmerling, 1998) ........... ........................ 16-42 16-26b Calculated Loop Seal Pressure Drop for 3-Bar, 15-Bar and 1000 psia .... .... 1642 16-27 Calculated Pressure Drop for 15-Bar and Vapor Superficial Gas Velocity of7ftls .. 1643 16-28 Calculated Loop Seal Pressure Drop for 15-Bar and Superficial Gas Velocity of 17 ft/s .. 1643 16-29 Calculated Loop Seal Pressure Drop for 15-Bar and Superficial Gas Velocity of 29 ft/s ............................................... 16-44 17-1 Flooding Model for a Vertical WCOBRAiTRAC Channel ..... ............ 17-32 17-2 Liquid Injection Rate for Case 3 (1000 psi) ........... ................... 17-33 17-3 Vapor Mass Flowrate at Middle of Pipe for Case 3 (1000 psi) ..... ......... 17-34 17-4 Liquid Mass Flowrate at Middle of Pipe for Case 3 (1000 psi) ..... ......... 17-35 17-5 Vapor Flow (Wg) Versus Liquid Flow (Wf) for Case 3 (1000 psi) ..... ...... 17-36 17-6 jg* Versus j; for Case 3 (1000 psi) .............. ...................... 17-37 17-7a Typical Flooding Results for Vertical Pipe (ID = 1.6 inches) ..... .......... 17-38 17-7b Countercurrent Flow Map Predicted by WCOBRATRAC ..... ............ 17-39 17-8 Flooding Model 1 for a Perforated Plate .......... ..................... 17-40 17-9 Flooding Velocities for Saturated Liquid and Vapor at 35 psia and j, = 3.3 ft/s Compared With Northwestern Flooding Limit (WCOBRA/TRAC MOD7A) .. 17-41 17-10 Flooding Velocities for Saturated Liquid and Vapor at 35 psia and j, = 8.0 ft/s Compared With Northwestern Flooding Limit (ECOBRAFRAC MOD7A) .. 17-42 17-11 Flooding Velocities for Saturated Liquid and Vapor at 1000 psia and j, = 3.3 ft/s Compared With Northwestern Flooding Limit (&COBRAJ1RAC MOD7A) .. 17-43 17-12 Liquid Mass Flowrates Through Perforated Plate at 35 psia andj,= 8.0 ft/s (&COBRARAC MOD7A) .. 17-44 17-13 Vapor Mass Flowrates Through Perforated Plate at 35 psia and j, = 8.0 ft/s (COBRA/TRAC MOD7A) .. 1745 o\434-non\Vol2-frmt.wpd:lb-4033 xxiii

LIST OF FIGURES (Cont'd) Figure Title Page 17-14 Predicted Flow Conditions and CCFL for Perforated Plate at High Pressure ............................................... 17-46 17-15 Predicted Flow Conditions and CCFL of a Perforated Plate at Low Pressure ......... 17-47 17-16 CCFL Data for Perforated Plate and Air/Water at Atmospheric Conditions (Hsieh, 1980) .17-48 17-17 A 3-D Model for a Horizontal Pipe .17-49 17-18 Drift Flux in a Horizontal Stratified Flow and Flooding Curves .17-50 17-19 Flooding Curves for Horizontal Stratified Flow .17-51 17-20 Computed Horizontal Flow State and Flooding Curves .17-52 17-21 PWR Hot Leg-to-Steam Generator Inlet Plenum Connection .17-53 17-22 Semiscale Mod-2A System for NC Tests .17-54 17-23 WCOBRAITRAC-SB Inventory Versus Time, Semiscale Mod-2A NC Test ... 17-55 17-24 Comparison of WCOBRA/TRAC-SB Prediction and Semiscale NC 60-kW (3-Percent Power) Data ............... 17-56 17-25 Predicted Void Fraction in the Steam Generator Tubes Uphill Region, Semiiscale NC 60-kW Test ................. 17-57 17-26 Predicted Void Fraction in the Steam Generator Tubes Downhill Region, Semiscale NC 60-kW Test ................. 17-58 17-27 Predicted Condensation Rate in the Steam Generator Tubes Uphill Region, Semiscale NC 60-kW Test ................. 17-59 17-28 Predicted Condensation Rate in the Steam Generator Tubes Downhill Region, Semiscale NC 60-kW Test .17-60 17-29 Comparison of Semiscale and PKL Natural Circulation Flowrates .17-61 18-1 WCOBRAITRAC-SB Representation of Interfacial Heat Transfer .18-17 18-2 Schematic Diagram of the Experimental System (Lim, et al., 1981) .18-18 18-3 WCOBRA/TRAC Noding .18-19 18-4 Measured Water Thickness Versus Axial Position for Various Liquid Flowrates and Inlet Water Layer Thickness of 1.583 cm ...... ............. 18-20 18-5 Measured Water Thickness at 0.157 m From the Channel Inlet Versus Liquid and Steam Flowrates .................. 18-21 o:\4384-non\Vol2-frMLwpd:lb04033 xxiv

LIST OF FIGURES (Cont'd) Figure Title Page 18-6 Calculated Liquid Level (Run 275) .......... ......................... 18-22 18-7 Calculated and Measured Liquid Levels Versus Axial Position (Run 275) .... 18-23 18-8 Calculated Steam Pressure (Run 275) .............. ................... 18-24 18-9 Calculated and Measured Steam Pressure Versus Axial Position (Run 275) ... 18-25 18-10 Calculated Steam Flowrate (Run 275) ............ .............. 18-26 18-11 Calculated and Measured Steam Flowrate Versus Axial Position (Run 275) ... 18-27 18-12 Comparison of Condensation Heat Transfer Correlations ....... ........... 18-28 18-13 Predicted Versus Measured Liquid Level at Various Axial Locations ........ 18-29 18-14 Predicted Versus Measured Steam Flowrate at Various Axial Locations ...... 18-30 18-15 Predicted Versus Measured Liquid Temperature at the Channel Exit .. ....... 18-31 18-16 Predicted Versus Measured Steam Pressure Drop at Various Axial Locations ................................................. 18-32 19-1 Schematic Diagram of the LSTF .............. ....................... 19-23 19-2 WCOBRAfTRAC-SB Model of the LSTF Pressure Vessel ..... ........... 19-24 19-3 WCOBRAIRAC-SB Model of the LSTF Hot and Cold Legs ..... ......... 19-25 19-4 WCOBRA[IRAC-SB Model of the LSTF Loop A Steam Generator ......... 19-26 19-5 WCOBRAITRAC-SB Model of the LSTF Loop B Steam Generator ......... 19-27 19-6 WCOBRA/TRAC-SB Model of the LSTF Loop Seals ...... .............. 19-28 19-7 WCOBRA/TRAC-SB Model of LSTF Safety Injection ...... ............. 19-29 19-8 Nodalization of the LSTF Break Unit .......... ....................... 19-30 19-9 Comparison of Predicted and Measured Primary System Pressure, ROSA 5-Percent Cold Leg Side Break .. 19-31 19-10 Comparison of Predicted and Measured Break Flowrates, ROSA 5-Percent Cold Leg Side Break .. 19-32 19-11 Predicted Intact Loop Seal Steam Flowrate, ROSA 5-Percent Cold Leg Side Break .. 19-33 19-12 Predicted Broken Loop Seal Vapor Flowrate, ROSA 5-Percent Cold Leg Side Break .. 19-34 19-13 Comparison of Predicted and Measured Core Collapsed Liquid Levels, ROSA 5-Percent Cold Leg Side Break .. 19-35 o:M4384nonNo2-Entwpd:b-O04033 xxv

LIST OF FIGURES (Cont'd) Figure Title Page 19-14 Comparison of Predicted and Measured PCTs, ROSA 5-Percent Cold Leg Side Break .............................................. 19-36 19-15 Predicted Intact Loop Uphill Steam Generator Tube Collapsed Liquid Level, ROSA 5-Percent Cold Leg Side Break ........................... 19-37 19-16 Comparison of Predicted and Measured Broken Loop Uphill Steam Generator Tube Collapsed Liquid Level, ROSA 5-Percent Cold Leg Side Break ..... 19-38 19-17 Predicted Primary System Pressure, ROSA 10-Percent Cold Leg Side Break . . 19-39 19-18 Predicted Break Flowrate, ROSA 10-Percent Cold Leg Side Break .19-40 19-19 Predicted Core Collapsed Liquid Level, ROSA 10-Percent Cold Leg Side Break ....... 1941 19-20 Predicted Intact Loop Seal Steam Flowrate, ROSA 10-Percent Cold Leg Side Break ..... 19-42 19-21 Predicted Broken Loop Seal Steam Flowrate, ROSA 10-Percent Cold Leg Side Break ....... 19-43 19-22 Intact Loop Uphill Steam Generator Tube Collapsed Liquid Level Prediction, ROSA 10-Percent Cold Leg Side Break ........ .............. 19-44 19-23 Broken Loop Uphill Steam Generator Tube Collapsed Liquid Level Prediction, ROSA 10-Percent Cold Leg Side Break .19-45 19-24 Predicted PCT, ROSA 10-Percent Cold Leg Side Break .19-46 19-25 Break Orientation in LSTF 2.5-Percent Cold Leg Break Tests (Koizumi, et al., 1987) .............. 1947 19-26 Comparison of Predicted and Measured Primary System Pressure, ROSA 2.5-Percent Cold Leg Side Break ...................... 19-48 19-27 Comparison of Predicted and Measured Break Flowrates, ROSA 2.5-Percent Cold Leg Side Break ...................... 19-49 19-28 Comparison of Predicted and Measured Core Collapsed Liquid Levels, ROSA 2.5-Percent Cold Leg Side Break ........................... 19-50 19-29 Comparison of Predicted and Measured PCTs, ROSA 2.5-Percent Cold Leg Break Tests ............. 19-51 o:\4384-nonWVol2-frmLwpd:lb-04033 xxvi

LIST OF FIGURES (Cont'd) Figure Title Page 19-30 Comparison of Predicted and Measured Broken Loop Uphill Steam Generator Tube Collapsed Liquid Level, ROSA 2.5-Percent Cold Leg Side Break ..................................................... 19-52 19-31A Comparison of Experimental Break Flowrate for 2.5-Percent Top, Side, and Bottom Cold Leg Breaks (Koizurni, et al., 1988) ...... ............... 19-53 19-3 1B Comparison of Predicted Break Flowrate for 2.5-Percent Top, Side, and Bottom Cold Leg Breaks ............. ........................... 19-53 19-32A Comparison of Experimental Core Collapsed Liquid Levels for 2.5-Percent Top, Side, and Bottom Cold Leg Breaks (Koizumi, et al., 1988)... 19-54 19-32B Comparison of Predicted Core Collapsed Liquid Levels for 2.5-Percent Top, Side, and Bottom Cold Leg Breaks ........ ....................... 19-54 19-33A Mixture Levels in Broken Cold Leg Measured for Side, Bottom, and Top Break Experiments (Koizumi, et al., 1988) ....... .................. 19-55 19-33B Two-Phase Mixture Level Prediction in Broken Cold Leg for 2.5-Percent Break Cases ..................................................... 19-55 19-34A Cladding Temperature of B-20 Rod at Position 7 (8.67-ft Elevation) for Side, Bottom, and Top Break Experiments (Koizumi, et al., 1987) ....... 19-56 19-34B Clad Temperature Predictions at 8.66-ft. Elevation for 2.5-Percent Side, Bottom, and Top Break Experiments .19-56 20-2-1 Diagram of LOFT Major Components (Bayless, et al., 1980) .20-4 20-2-2 Diagram of LOFT Reactor Vessel and Flowpaths (Reeder, 1978) .20-5 20-2-3 Cross Section of LOFT Core Arrangement (Bayless, et al., 1980) .20-6 20-3-1 Reactor Vessel Modelling for LOFT LOCEs L3-1, 13-7, and L3-5 .20-12 20-3-2 Active Loop Hot Leg, Pressurizer, and Steam Generator Inlet Piping Modelling for LOFT LOCEs L3-1, 13-7, and L3-5 .20-13 20-3-3 Active Loop Steam Generator Modelling for LOFT LOCEs .3-1, L3-7, and L3-5 ........ 20-14 20-3-4 Active Loop Pump Suction Piping and RCP Modelling for LOFT LOCEs L3-1, L3-7, and L3-5 .20-15 20-3-5 Active Loop Cold Leg Modelling for LOFT LOCEs L3-1 and L3-7 .20-16 20-3-6 Inactive Loop Modelling for LOFT LOCEs L3-1 and L3-7 .20-17 o:\4384-non\Vol2-fnt.wpd:lb 04033 xxvii

I LIST OF FIGURES (Cont'd) Figure Title Page 20-3-7 Break Orifice Assembly for LOFT LOCEs L3-1, and L3-5 (Bayless, et al., 1980) .............................................. 20-18 20-3-8 Break Orifice Assembly for LOFT LOCE L3-7 (Gillas and Carpenter, 1980) .. 20-19 20-3-9 LOFT LOCE L3-5 Break Unit Configuration .20-20 20-5-1-1 Comparison of Calculated and Measured Temperature Upstream of Break Orifice for LOFT LOCE L3-1 .20-30 20-5-1-2 Comparison of Inactive Loop Cold Leg, Inactive Loop Hot Leg, and Saturation Temperature for LOFT LOCE L3-1 From Test Measurements .20-31 20-5-1-3 Comparison of Calculated Liquid Temperature Upstream of Break Orifice, Liquid Temperature in RABL, and RABL Mass Flowrate .20-32 20-5-1-4 Comparison of Calculated Inactive Loop Cold Leg Liquid Temperature and Saturation Temperature .20-33 20-5-1-Sa Measured Mass Flowrate in Inactive Loop Cold Leg (Qualified) (Bayless, et al., 1980) .............................................. 20-34 20-5-1-Sb Calculated Break Mass Flowrate and Upstream Void Fraction for LOFT LOCE L3-1 ............................................. 20-35 20-5-1-6 Comparison of Calculated and Measured Primary System Pressure for LOFT LOCE L3-1 ............................................. 20-36 20-5-1-7 Calculated Void Fractions Upstream of Break Orifice and in 3-D Channel Connected to Break Unit for LOFT LOCE L3-1 ..... ......... 20-37 20-5-2-la Measured Inactive Broken Loop Cold Leg Temperature for LOFT LOCE L3-7 (Gillas and Carpentar, 1980) ........................................ 20-41 20-5-2-lb Calculated Liquid Temperature Upstream of Break Orifice for LOFT LOCE L3-7 ............ .................................... 20-41 20-5-2-2a Comparison of Measured Break Flow and ECCS Flow (Not Qualified) (McCreery, 1980) ............. .................................... 20-42 20-5-2-2b Calculated Break Mass Flowrates for LOFT LOCE L3-7 ..... ............. 20-42 20-5-2-3a Measured Inactive Loop Cold Leg Pressure for LOFT LOCE L3-7 .... ...... 20-43 20-5-2-3b Calculated Inactive Loop Cold Leg Pressure for LOFT LOCE L3-7 .... ..... 20-43 20-5-2-4a Measured Steam Generator Inlet and Outlet Temperatures for LOFT LOCE L3-7 ........... 20-44 o:\4384-non\Vol2-frimLwpd:lb-04033 xx.)Vii

LIST OF FIGURES (Cont'd) Figyure Title Page 20-5-2-4b Calculated Steam Generator Inlet and Outlet Temperatures versus Calculated Active Loop Mass Flow Rate for LOFT LOCE 13-7 ...... ............... 20-44 20-5-2-5 Calculated Steam Generator Heat Rejection versus Decay Heat for LOFT LOCE L3-7 ................................................ 20-45 20-5-3-1 Comparison of Calculated and Measured Hot Leg Pressures for LOFT LOCE L3-5 ................................................ 20-50 20-5-3-2 Comparison of Liquid and Steam Mass Flowrate at Steam Generator Inlet for LOFT LOCE L3-5 . ............................................. 20-51 20-5-3-3 Comparison of Calculated Liquid and Steam Mass Flowrate at Steam Generator Outlet for LOFT LOCE L3-5 ....... ................... 20-52 20-5-3-4 Comparison of Calculated Steam Generator Outlet Flow versus Active Loop Hot Leg and Cold Leg Temperatures for LOFT LOCE L3-5 ..... 20-53 20-5-3-5 Comparison of Calculated and Measured Break Mass Flowrates for LOFT LOCE L3-5 ................................................. 20-54 20-5-3-6 Comparison of Calculated and Measured Loop Seal Level for LOFTLOCE L3-5 ................................................ 20-55 20-5-3-7 Calculated Primary System Coolant Inventory for LOFT LOCE L3-5 ....... 20-56 21-1 Semiscale Mod-2C System ............... .......................... 21-17 21-2 Cold Leg Break Assembly . ......................................... 21-18 21-3 Vessel Upper Head Configuration .................................... 21-19 214 Core Heater Rod Axial Power Profile ......... ........................ 21-20 21-5 WCOBRA/TRAC Model of Semiscale Mod-2C Component Layout .... ..... 21-21 21-6 WCOBRA/IRAC Model of Semiscale Reactor Vessel ...... ............. 21-22 21-7 WCOBRAI'RAC Model of Semiscale Intact Loop Steam Generator .... .... 21-23 21-8 WCOBRAIIRAC Model of Semiscale Broken Loop Steam Generator ....... 21-24 21-9 Semiscale Reactor Coolant Loop Noding ........ ...................... 21-25 21-10 Core Power Versus Time Curve ............ ......................... 21-26 21-11 Safety Injection Rates as a Function of RCS Pressure ...... ............... 21-27 21-12 Pressurizer Pressure, Test S-LH-1 ........... ......................... 21-28 21-13 Primary and Secondary Predicted Pressures, Test S-LH-1 ..... ............ 21-29 21-14 Broken and Intact Loop Secondary Pressure Predictions, Test S-LH-1 ....... 21-30 o\434-non\Vol2-fmntwpd:lb-04033 xxix

LIST OF FIGURES (Cont'd) Figure Title Pave 21-15 Break Mass Flowrates, Test S-LH-1 . 21-31 21-16 Core Collapsed Liquid Levels, Test S-LH-1 . 21-32 21-17 Calculated Void Fractions in the Intact Loop Pump Suction Piping, Test S-LH-1 ............................................. ........ 21-18 Calculated Void Fractions in the Broken Loop Pump Suction Piping, Test S-LH-1 ............................................. ........ 21-19 Core Heater Rod Temperature Response, Test S-LH-1. ........ 21-20 Integrated Break Mass Flow Comparison, Test S-LH- 1. 21-36 21-21 Integrated Break Mass Flow Comparison, Test S-LH-2 . ........ 21-22 Pressurizer Pressure, Test S-LH-2. ........ 21-23 Primary and Secondary Predicted Pressures, Test S-LH-2. 21-39 21-24 Broken and Intact Loop Secondary Pressure Predictions, Test S-LH-2 . 21-40 21-25 Break Mass Flowrates, Test S-LH-2 . 21-41 21-26 Core Collapsed Liquid Levels, Test S-LH-2 . 21-42 21-27 Calculated Void Fractions in the Intact Loop Pump Suction Piping, Test S-LH-2 . 21-43 21-28 Calculated Void Fractions in the Broken Loop Pump Suction Piping, Test S-LH-2 . 21-44 21-29 Core Heater Rod Temperature Response, Test S-LH-2. 21-45 22-1-1 Vertical Test Train Configuration for NRU Reflood Experiments . 22-11 22-1-2 NRU Test Bundle Cross Section (Test PTH-110 Bundle Shown) . 22-12 22-1-3 WCOBRA/TRAC Model of NRU . 22-13 22-1-4 NRU Test M'T-3.06 Injection Flowrate. 22-14 22-1-5 Comparison of Rod 1 Predicted and Measured Inner Cladding Temperatures at Level 15 for NRU Test MT-3.06 ...... .................. 22-15 22-1-6 Comparison of Rod 2 Predicted and Measured Inner Cladding Temperatures at Level 15 for NRU Test MT-3.06 ...... .................. 22-15 22-1-7 Comparison of Rod 1 Predicted and Measured Inner Cladding Temperatures at Level 17 for NRU Test MT-3.06 ...... .................. 22-16 22-1-8 Comparison of Rod 2 Predicted and Measured Inner Cladding Temperatures at Level 17 for NRU Test MT-3.06 ...... .................. 22-16 o:\4394-non\Vol2-frmLwpd:lb-04033 XXX

LIST OF FIGURES (Cont'd) Figure Title Page 22-1-9 Comparison of Rod 2 Predicted and Measured Inner Cladding Temperatures at Level 18 for NRU Test MT-3.06 ......... .......................... 22-17 22-1-10 Comparison of Rod 2 Predicted and Measured Pellet Temperatures at Level 15 for NRU Test MT-3.06 .......... ........................... 22-17 22-1-11 Comparison of Rod 1 Predicted and Measured Pellet Temperatures at Level 17 for NRU Test MT-3.06 .......... ......................... 22-18 22-1-12 Comparison of WCOBRA,TRAC Predicted Quench Front Elevations with MT-3.06 Data from NUREG/CR-2528 ................................ 22-19 22-1-13 Comparison of Rod 1 Internal Pressure to the Measured Plenum Pressure of NRU Rod 2C for NRU Test MT-3.06 ............................... 22-20 22-1-14 NRU Test PTH-1 10 Injection Temperature ............................. 22-21 22-1-15 Comparison of Rod 1 Predicted and Measured Inner Cladding Temperatures at Level 17 for NRU Test PTH-110 .......... ......................... 22-22 22-1-16 Comparison of Rod 2 Predicted and Measured Inner Cladding Temperatures at Level 17 for NRU Test PTH-1 10 .......... ......................... 22-22 22-1-17 Comparison of Rod 1 Predicted and Measured Inner Cladding Temperatures at Level 18 for NRU Test PTH-1 10 .......... ......................... 22-23 22-1-18 Comparison of Rod 1 Predicted and Measured Pellet Temperatures at Level 17 for NRU Test PTH-1 IO ..................................... 22-23 22-1-19 Comparison of Rod 2 Predicted and Measured Pellet Temperatures at Level 17 for NRU Test PTH-110 .......... ......................... 22-24 22-2-1 Indian Point Unit 2 Loop 21 Accumulator Line Schematic Diagram .... ..... 22-28 22-2-2 WCOBRAfTRAC Model of Accumulator and Safety Injection Line in a PWR ... 22-29 22-2-3 Predicted Accumulator Pressure (Solid Line) Compared with Measured Test Data (Dashed Line) . .................................... 22-30 22-2-4 Predicted Accumulator Flowrate .................................... 22-31 22-2-5 Predicted Gas Temperature at Top of Accumulator ......... ............. 22-32 22-2-6 Comparison of Predicted Pressure/Volume Relationship with Adiabatic Assumptions .. 22-33 22-2-7 Comparison of Detailed Noding with Simplified PYYR Noding Prediction of Accumulator Pressure .. 22-34 o:\4384-nonWVol2-frtwpd:lb04033 xxxi

LIST OF FIGURES (Cont'd) Figure Title Page 22-2-8 Basis for Transition from Water to Nitrogen Flow From Accumulator (Andreychek, et al., 1988) ................ .......................... 22-35 22-2-9 Predicted Void Fraction at Accumulator Line Exit ...... ................. 22-36 22-3-1 Cross-Sectional View of the Westinghouse Scale Model Pump ..... ........ 22-43 22-3-2 Scale Model Homologous Head Single-Phase Data in the Pumping Mode, Forward and Reverse Flow .. 22-44 22-3-3 Scale Model Homologous Head Single-Phase Data in the Dissipation Mode, Forward Flow .................................................... 2245 22-3-4 Data Scatter for Dissipative Mode 1/3-Scale Pump Data (Cudlin, 1977) ...... 22-46 22-3-5 Schematic Diagram of the Air-Water Test Facility ...... ................. 2247 22-3-6 Homologous Head Curves and Westinghouse Air-Water Data ..... ......... 22-48 22-3-7 Single-Phase and Fully Degraded Pump Head Curves Compared With Two-Phase Data .. 22-49 22-3-8 Pump Single-Phase and Fully Degraded Torque Curves, Compared With Two-Phase Data ............ ...................... 22-50 22-3-9 Two-Phase Multiplier and Pumping Mode Data ............ ............. 22-51 22-3-10 Two-Phase Multiplier and All Two-Phase Data ........... .............. 22-52 22-3-11 M(a) for Pump Torque (Referred to as N(a) in Equation 9-8 in This Document) .. 22-53 22-3-12 Westinghouse Pump Head Curves Compared with LOFT Pump Head Curves .. 22-54 o:\4384-non\Vol2-fmt.wpd:lb.04033 xxxii

LIST OF ACRONYMS AND ABBREVIATIONS A. O. Axial Offset ACRS Advisory Committee on Reactor Safeguards AFLUX Core Average Heat Flux ANS American Nuclear Society ANSI American National Standards Institute BE-SBLOCA Best Estimate Small Break LOCA BLD Blowdown BO Boil-off BOL Beginning of Life CAOC Constant Axial Offset Control CCFL Counter-current Flow Limitation CD Discharge Coefficient for Two-phase Break Flow CE Combustion Engineering CHF Critical Heat Flux COLR Core Operating Limits Report COSI Condensation On Safety Injection CP Conditional Probability CQD Code Qualification Document CSAU Code Scaling Applicability and Uncertainty DFFB Dispersed Flow Film Boiling DNB Departure from Nucleate Boiling ECCS Emergency Core Cooling System EOP Emergency Operating Procedure FAC Final Acceptance Criteria FEM Entrained Droplet Flowrate FLM Continuous Liquid Flowrate GEDM Generalized Energy Deposition Model H High (Importance Level in Los Alamos PIRT Ranking Scheme) HAFLUX Hot Assembly Average Power HAPHR Hot Assembly Peak Heat Rate HHSI High Head Safety Injection HRFLUX Hot Rod Average Power HTC Heat Transfer Coefficient IADF Inverted Annular Dispersed Flow o:\4384-non\Vol2-frmt.wpd:lb-04033 xvcxiii

LIST OF ACRONYMS AND ABBREVIATIONS (Cont'd) IAFB Inverted Annular Film Boiling INEL Idaho National Engineering Laboratory IP2 Indian Point Unit 2 JAERI Japan Atomic Energy Research Institute L Low (Importance Level in Los Alamos PIRT ranking scheme) LOCA Loss of Coolant Accident LOCE Loss of Coolant Experiment LOFT Loss of Fluid Test LOOP Loss of Offsite Power LSC Loop Seal Clearance LSTF Large Scale Test Facility M Medium (Importance Level in Los Alamos PIRT ranking scheme) MSSV Main Steam Safety Valve MS1V Main Steam Isolation Valve MTC Moderator Temperature Coefficient N/A Not Applicable NC Natural Circulation NPP Nuclear Power Plant NRC Nuclear Regulatory Commission NRU National Research Universal NSSS Nuclear Steam Supply System NUCL Saturated Nucleate Boiling OPA Offsite Power Available ORNL Oak Ridge National Laboratory PCT Peak Cladding Temperature PIRT Phenomena Identification and Ranking Table PLHGR Peak Linear Heat Generation Rate PLHR Peak Linear Heat Rate PLOW Low Power Region Relative Power PORV Pressure-operated Relief Valve PWR Pressurized Water Reactor RABL Reflood Assist Bypass Line RAI Request for Additional Information RAOC Relaxed Axial Offset Control o:\4384-non\Vol2-frmtwpd:lb-04033 XX7UV

LIST OF ACRONYMS AND ABBREVIATIONS (Cont'd) RCP Reactor Coolant Pump RCS Reactor Coolant System REC Core Recovery RHR Residual Heat Removal ROSA Rig-of-Safety Assessment RSIC Radiation Shielding Information Center RWST Refueling Water Storage Tank SBLOCA Small Break Loss of Coolant Accident SCNB Subcooled Nucleate Boiling SG Steam Generator SGTP Steam Generator Tube Plugging SI Safety Injection SIS Safety Injection Systems SPL Single-phase Liquid Convection SPV Single-phase Vapor Convection THTF Thermal Hydraulic Test Facility TPFL Two-Phase Flow Loop TRAN Transition Boiling TS Technical Specifications TSI Safety Injection Water Temperature UHI Upper Head Injection UPTF Upper Plenum Test Facility o:\4384-non\Vol2-frnt.wpd:lb-04033 xxxv

I o.A4384-nonWVol2-frmt.wpd:lb-04033 XXXVi

COMMONLY USED EQUATIONNOMENCLATURE a sonic velocity h heat transfer coefficient ar grid blockage ratio h normalized pump head (Ch. 9) a, vapor absorption coefficient hi interfacial heat transfer coefficient a, liquid absorption coefficient H enthalpy A area Hfg enthalpy of vaporization Ax axial flow area H. Meyer hardness A2, lateral flow area I grid rewet index (Ch. 5,6) A,, wall heat transfer area I pump moment of inertioi (Ch. 9) A, intercell friction area k thermal conductivity Ai interfacial area K loss coefficient (Ch. 2,4) B mass transfer number K conductance (Ch. 7) CO slip distribution parameter vertical interfacial drag coefficient K,x CD drag coefficient transverse interfacial drag Cp specific heat at constant pressure KWX coefficient C, specific heat at constant volume vertical wall drag coefficient KWZ L D diameter transverse wall drag coefficient KXz Dh hydraulic diameter axial flow form loss coefficient D deformation tensor z transverse flow form loss coefficient e specific energy L length fl" wall friction factor L9 gap width f; interfacial friction factor Lg° orthogonal gap width f theoretical density fraction (Ch. 7) Lb mean beam length F ramping function momentum mixing length F turbulence anisotropy tensor energy mixing length mass flowrate F gray body factor (Ch. 6) m M momentum (Ch. 2) FcHEN Chen convective boiling multiplier M molecular weight (Ch. 7) g- force g gravitational acceleration n pump head multiplier (Ch. 9) N mole fraction gc gravitational conversion constant N number density g gravitational acceleration vector N pump torque multiplier (Ch. 9) G mass flux p viscosity number Gx axial mass flux p pressure Gz transverse mass flux Pw wetted perimeter o.\4384-non\Vol2-fmt.wpd:lb-04043 xxxvii

COMMONLY USED EQUATIONNOMENCLATURE (Cont'd) Pr Prandtl number vc mesh cell volume Prod fuel rod pitch w transverse velocity component, wall-liquid heat transfer rate Cartesian coordinates q.v wall-vapor heat transfer rate W transverse velocity, subchannel q. interface-liquid heat transfer rate coordinates qno interface-vapor heat transfer rate 1f orthogonal transverse velocity, Q.f wall-liquid heat transfer subchannel coordinates Q. wall-vapor heat transfer We Weber number r bubble/drop radius x quality radial coordinate x vertical direction, Cartesian R internode resistance (Ch. 7) coordinates (Ch. 2) R radiation resistance (Ch. 6) X vertical direction, subchannel R gas constant (Ch. 10) coordinates Re orifice hole radius X axial direction, 1D components Re Reynolds number y transverse direction, Cartesian s specific entropy coordinates S net rate of entrainment z transverse direction, Cartesian SCNiEN Chen building suppression factor coordinates SE rate of entrainment Z transverse direction, subchannel SDE rate of de-entrainment coordinates St Stanton number t time Greek T temperature T pump torque (Ch. 9) a void fraction T stress tensor aN normalized pump speed Tr Reynold stress tensor 13 volumetric coefficient of expansion F, net rate of mass transfer u vertical velocity component, 3 film thickness Cartesion coordinates Kronecker delta U vertical velocity component, sij subchannel coordinates & thermal emissivity v transverse velocity component, s strain Cartesian coordinates 'q fraction of vapor generation coming V volume from entrained liquid o-\4384-non\Vol2-fimLwpd:lb-04043 xxxviii

COMMONLY USED EQUATION NOMENCLATURE (Cont'd) iiNR de-entrainment efficiency CHEN Chen correlation

'K         thermal diffusivity                     CHF   critical heat flux churn churn flow regime characteristic wave length CT    chum-turbulent flow regime Pt          viscosity d      drop u'         turbulent viscosity dcht   direct contact heat transfer p          density DD     dispersed droplet flow regime E          absorption cross section DE     de-entrainment a          surface tension                        dfib   dispersed flow film boiling a          stress (Ch. 2, 7)                      DFFB   dispersed flow film boiling a          fluid-fluid stress tensor              e      entrained field vsB      Stephan-Boltzmann constant             E      entrainment f       saturated liquid r         shear stress fb      film boiling T          viscous drag force fr      flow regime vi         interfacial drag force                fric    friction loss form form loss v           specific volume FC     forced convection v           normalized pump volumetric flow FD     film/drop flow regime x           Martinelli-Nelson factor FF     falling film flow regime N'a        absorption efficiency g      saturated vapor Q           source term                            gas    gas co          specific speed                         gv     grid to vapor Gr    Grashof number Subscripts                                         h      hydraulic Henry Henry correlation am          annular-mist flow regime                i     interfacial ACC         accumulator                            IVA    inverted annular flow regime b           bubble                                 11s    inverted liquid slug flow regime br          bubble rise                            k      phase k bubbly      bubbly flow regime                            liquid field Brom        Bromley correlation                    liq    liquid crit        critical                               LB     large bubble cwv         convection wall-vapor                  m      mixture o:\4384-non\Vol2-frmt.wpd: Ib-05053         xxxix

I COMMONLY USED EQUATION NOMENCLATURE (Cont'd) MIN minimum film boiling point v vapor field nc natural convection vap vapor Inc laminar natural convection ve between vapor and entrained fields Ifc laminar forced convection v1 between vapor and liquid fields N normalized w wall NB nucleate boiling wb wall to fluid as latent heat o onice W, wall to liquid p pipe wv wall to vapor QF quench front x vertical direction, Cartesian r relative coordinates r radial (Ch. 7) X vertical direction, subchannel rwe radiation wall-entrained field coordinates rwg radiation wall to grid X axial direction, 1D components rw,O radiation wall-liquid field y transverse direction, Cartesian rwv radiation wall-vapor field coordinates s drop formation z transverse direction, Cartesian sat saturation coordinates .X: slug slug flow regime Z transverse direction, subchannel s slug - coordinates SB small bubble flow regime Zr Zirconium SCL subcooled liquid 20 two-phase SCNB subcooled nucleate boiling F phase change SCV subcooled vapor SNL superheated liquid Superscripts SLV superheated vapor SLB small to large bubble flow regime i interfacial surface average SPL single-phase liquid n old time value SPV single-phase vapor Ft donor cell old time value sup suppression T turbulent TB transition boiling t transpose TD top deluge flow regime II per unit area tnc turbulent natural convection It, per unit volume TQ top quench U0 2 uranium dioxide o:\4384-non\Vol2-frt.wpd:lb-04033 xi

SECTION 12 CORE HEAT TRANSFER DURING A SMALL BREAK LOCA 12-1 Introduction The small break LOCA transient is characterized by the draining of the initial Reactor Coolant System (RCS) inventory to the break location. Five distinct periods have been identified during a small break LOCA event: blowdown, natural circulation, loop seal clearance, boiloff, and core recovery. The duration of each period is break-size dependent. Each small break LOCA period is described in the detailed discussion of the small break LOCA PIRT (Volume 1, Section 1-4, of this document). In Westinghouse-designed pressurized water reactors (PWRs), core uncovery and fuel rod heatup occur in the boiloff period and termiinate in the recovery period. During the boiloff period, the vessel mixture level reaches a minimum value as the liquid inventory gradually boils away. If this two-phase mixture level is low enough, core uncovery occurs. The recovery phase begins when the RCS is depressurized to the point where boiloff is exceeded by the delivery of safety injection to the vessel. The fuel rod heatup transient is terminated once the entire core is quenched and the safety injection flow from the safety injection pumps and/or accumulators exceeds the break flow. The core flowrates during the fuel rod heatup period of a small break LOCA are lower than those associated with large break LOCAs. In general, the core flowrates in all phases of the large break LOCA are large enough for the convective flow to be turbulent. This is not always the case during a small break LOCA. For example, in a 3-inch cold leg break analysis of Indian Point Unit 2, similar to the case reported in Section 27 in Volume 3 of this document, the steam flow in the hot assembly at the time of PCT is less than 1 lbm/s; the Reynolds number (Re) based on film temperature is below the value for fully developed turbulent flow (Re = 10,000) throughout virtually all of the core uncovery transient, and it falls below 3000 near the time of minimum inventory. Moreover, for such low Re steam flows, the steam velocity is not sufficient to cause significant entrainment of droplets. For the 3-inch break case at Indian Point Unit 2, the entrained field flow is nonexistent for the core uncovery transient. In general, the small break LOCA transients show steam flow only above the mixture level for the smaller break sizes. At larger break sizes (8-inch break and greater), an entrained field is predicted, but the steam Re in o.M4384-non/sec12.wpd: b-04033 12-1

I the hot assembly for a 10-inch break transient for Indian Point Unit 2 is approximately 10,000 for much of the time that the core is uncovered. 12-2 Physical Processes During the boiloff period, the fuel rods above the core mixture level are cooled by steam flowing (with or without entrained droplets) at a low rate. Review of the Indian Point Unit 2 cases emphasizes the importance of having heat transfer models that are valid for the laniinar-turbulent transition range of Re, defined as [ ]" in this work. Therefore, the important physical processes are those associated with heat transfer in the single-phase vapor (SPV) and dispersed droplet regimes. In WCOBRA/TRAC modelling in the best estimate small break LOCA version of the code, validation is needed that the code provides reasonable predictions of the heat transfer coefficient (HTC) in low Re steam flows, with and without droplets, at high pressure conditions typical of the boiloff period in a small break LOCA. The important physical processes in predicting heat transfer involving low Re flows, consisting of either SPV or a high quality [ P.C dispersed droplet flow, are as follows:

• Convective HTC in the laminar-turbulent transition range dependence on the steam Re
• Drop-wall contact heat flux in the dispersed flow film boiling (DFFB) regime dependence on Re In the large break LOCA scenario, drop-wall contact depends solely on properties, local void fraction, and wall superheat because the effectiveness of drop-wall contact is comparable regardless of the mass flux. However, the drop-wall contact heat flux decreases to zero if the flow becomes laminar.
• Wall-to-steam thermal radiation in the DFFB and the SPV regimes
• Void fraction gradient defining a mixture level The hydraulics and the fluid condition at a sharp void fraction gradient differ markedly between the mixture and vapor phases.

oA4384-non/sec12.wpd:1b-4033 12-2

12-3 WCOBRA/TRAC-SB Heat Transfer Model The WCOBRA/TRAC heat transfer regime map presented as Figure 6-3 in Volume I of this document is repeated here as Figure 12-1 for ease of reference. Section 6 of this document provides the details of the models and correlations of the WCOBRAITRAC-SB heat transfer package; some of the Section 6 equations are included in the discussion that follows. 12-3-1 Convective Heat Transfer The SPV regime HTC is selected from among four correlations: Dittus-Boelter (Dittus and Boelter, 1930), Wong-Hochreiter (Wong and Hochreiter, 1981), laminar flow heat transfer (Nu = 10), and turbulent natural convection. The SPV HTC is used when the void fraction is The wall HTC h is selected according to: [a The wall HTC h,,sPv is selected according to:[ Ia.c I Ia.c In the SPV regime, liquid phase HTCs are set to zero: hWlAPV= 0 hwb,sPv 0 o:\4384-non/secl2.wpd:Ib-04033 12-3

I The selection logic ensures a smooth and continuous transition in HTC from low Re (laminar N. flow) to high Re (turbulent). In ensuring that correlations are used within the range of Re that is appropriate for each, the convective HTC for vapor is calculated as: h FC = (1 - RHTCV)hwvlow RHTCVh.,high (6-2-7) where, RHTCV acts to linearly ramp the HTC in the laminar-turbulent transition regime. This term is calculated as: [

                                                                              ]a,c            (6-2-8)

The low and high Re convective HTCs are selected as: h Z = maximum { (if Re < []ac) (6-2-9) and hwvhigh = maximum {h (6-2-10) Thus, if Rev < [ ]", the maximum of the HTC from a constant Nusselt number (Nu = 10) and turbulent natural convection is selected. For Re [ a.c, the maximum of the Dittus-Boelter and Wong-Hochreiter correlations is used. (The Wong-Hochreiter predicts a larger value up to Re = 25,000.) o:\4384-non/secl 2.wpd: b-04033 12-4

This selection logic retains the possibility of natural convection if flows are appropriately low (Regulatory Guide 1.157, 1989) and prohibits the use of turbulent flow correlations at Re below their validity. 12-3-2 Drop-Wall Contact The direct contact heat transfer for the dispersed droplet field is calculated using a model originally proposed by Forslund and Rohsenow (Forslund and Rohsenow, 1968) with modifications suggested by Bajorek and Young (Bajorek and Young, 1998) to improve performance at low Re. The direct wall contact term hdC,h is discussed in Section 6-2-8 of this document. 12-3-3 Radiation From Wall to Vapor The SPV heat transfer includes the thermal radiation from wall to steam. Thermal radiation occurs at void fractions up to 1.0. The SPV HTC is calculated as: hwvSPv = F,y,dF wc + hrwv (6-2-11) where the radiation HTC from the wall to vapor is calculated by Equation 6-156 in Section 6 of this document. The grid enhancement term Fg,id applies only to the convective term. Likewise, the two-phase enhancement multiplier should not be applied to the thermal radiation term in the DFEB regime. Therefore, F 2, is restricted to the convective term and is not applied to the radiation term. oA\4384-nonlsecl2.wpd:Ib-04033 12-5

I The HTCs for the DFFB regime are calculated as follows: hwvDFFB = FgZi 2#h'FC + hrw (6-2-15) hw4DFFB = hwe (6-131) hwb,DFFB = hdc, (6-132)

where, h,,,Fc is the convective HTC as described in subsection 12-3-1 h,w, is given by Equation 6-156 h.e is given by Equation 6-157 hd,, is given by Equation 6-2-14 The two-phase enhancement term (F2 ,,) is calculated by Equation 6-125 and is limited to values within the range 1.0 < F2 9,< [

                                                          ]a.c 12-34 Mixture Level Sharpening In a core uncovery situation during the boiloff period of a small break LOCA event, fuel rod heat transfer is orders of magnitude less in the SPV and/or DFFB regimes which prevail above the nixture level elevation than in the nucleate boiling regime which exists in the two-phase mixture region. Thus, for the calculation of fuel rod heatup, it is imperative that the mixture level interface be accurately defined. Section 11-2-4 of this document describes in detail the WCOBRAfTRAC-SB level sharpener coding.

The level sharpening logic is applied to void fractions from COBRA channel cells; it initially assumes that no sharp level exists. The parameter ISHARP is an indicator of where the mixture level is located. The void gradient is defined as being sharp when [ Ia.c If a sharp gradient is detected in the bottom half of the cell, ISHARP is set to 1; if the sharp gradient is in the top half of the cell, the value of ISHARP is set to 2. The upper and lower lirnits of the void gradient search are [ ]ac respectively. [ Ia'c corresponds to the void fraction for transition between the oA4384non/sec2wpd:1b04033 12-6

inverted annular dispersed flow (IADF) and DFFB heat transfer regimes. Thus, when ISHARP = 1, the physical picture of the cell is one in which a transition from inverted annular film boiling (IAFB) occurs in the bottom half of the cell and the transition from IADF to DFFB is at the top edge. Likewise, [ ]6 corresponds to the transition from inverted annular film boiling (IAFB) to IADF. The coding of the level sharpener logic has been verified via a series of standalone calculations, which assume a uniform cell height. 12-4 Assessment of WCOBRAITRAC-SB Heat Transfer Model for Small Break LOCA Application A driver-plotter progran (COBRAHT), similar to that used in large break LOCA heat transfer assessment, was used to examine the performance of WCOBRAfIRAC-SB in the SPV and DFFB regimes important to small break LOCA PCT deternination and to determine the bias and uncertainty of these models. In COBRAHT, convective heat transfer correlations are selected per the Re dependence. In testing the performance, HTCs predicted from the WCOBRAflRAC heat transfer package were compared to experimental test data. The test data for comparison contained a complete set of local measurements so that the HTC could be assessed without uncertainty due to compensating error. The HTC calculation in COBRAHT can be altered to perform assessments and investigate sensitivities. Figure 12-2 provides pictorially the calculational procedure used with COBRAHT. The local conditions for each (steady-state) test-wall temperature, vapor temperature, quality, pressure, and total mass flux were used as input to the driver-plotter routine. This driver-plotter routine consisted of the WCOBRAIRAC-SB heat transfer package, property routines, and drop size correlations. The input hydraulic conditions were used to estimate the film boiling HTCs (based on XT = T,,,-T,,), which were then compared to the measured HTCs for each test. The driver-plotter routine calculates an overall HTC in terms of the local heat flux and wall superheat, defined as: h q-(T. T) o-4384-nonlsecI2.wpd:b-04033 12-7

I The SPV and/or DFFB HTCs are then compared with the HTC values reported by the experimenters to validate the performance of WCOBRAJrRAC-SB. One of the problems in quantifying the accuracy of heat transfer relations in a large thermal-hydraulic systems code such as WCOBRA/TRAC is that few experimental tests provide a sufficient amount of simultaneous local information on void fraction, phasic flowrates, and phase temperatures. While modelling an entire separate effects test facility and simulating experiments can provide useful information on overall code performance, the predicted results are subject to compensating errors. That is, inaccuracies in one model package can compensate for the inaccuracies in another package producing a fortuitously correct result. An example is an accurate prediction of wall heat flux when HTCs are underpredicted, while (TW-T,) was overpredicted because of errors in the hydraulics package. If sufficient local information is available, it is possible to separate the calculation of the HTCs from the calculation of the fluid conditions and provide an assessment of the heat transfer prediction alone. 12-4-1 ORNL-THTF DFEB Test Simulations The Oak Ridge National Laboratory Thermal Hydraulic Test Facility (ORNL-THTF) blowdown tests are one source of data for validating the heat transfer predictions of WCOBRA/TRAC-SB in the DFFB/SPV regimes of interest for small break LOCA. A series of high-pressure steady-state upward DFFB tests in a rod bundle was performed in the ORNL-THTF and is discussed by Yoder (Yoder, et al., 1982). Tests were conducted for pressures ranging from 23 bar (635 psia) to 132 bar (1908 psia) at flowrates from 226 kgm 2 -s (166,300 lbm/ft 2 -hr) to 713 kg/m2 -s (525,300 lbm/ft2 -hr). The test section was composed of 64 full-length (3.66 m) rods in an 8x8 bundle geometry typical of PWR designs. A cross section of the ORNL-THTF bundle is shown in Figure 12-3. Sixty of the rods were electrically heated fuel rod simulators (that is, heated rods) with flat axial and radial power distributions. One rod (rod 32) failed to function properly during certain tests (B, C, D, and E). Four of the other rods (rods 19, 22, 36, and 46) were inactive (unheated rods). Six spacer grids were evenly spaced along the bundle. Because the axial power shape in the bundle was uniform, critical heat flux (CHF) occurred at the bundle exit first, then moved down the bundle. There were two fluid flow and temperature measurement sites at both the bottom and top ends of the test section. The heated rod surface temperatures were measured by the thermocouples at 30 axial levels and different circumferential locations. Some of the thermocouples were installed to measure the in-bundle fluid temperature. Local information was provided at the bundle exit, oA4384-nonlsecl2.wpd:Ib-G4033 12-8

which was designated "Level G." During steady-state operation of the ORNL-THTF, the inlet flow at the bottom of the test section was established and the loop was adjusted to provide the desired inlet fluid temperature and inlet quality. The bundle power was then increased until the dryout (CHF) point was obtained. The steady-state point was assumed to be reached when both pressure and rod surface temperatures stabilized. The results of both rod surface conditions and local equilibrium fluid conditions were then reported as cross-sectional average values for each level. Table 12-1 lists the thermal-hydraulic conditions of the 10 selected ORNL-THTF steady-state film boiling tests used to evaluate the WCOBRAITRAC-SB heat transfer package. An initial assessment was performed with a version of COBRAHT that used the Forslund and Rohsenow (Forslund and Rohsenow, 1968) drop-wall DFFB contact model used in the large break LOCA version of WCOBRA/TRAC (Bajorek, et al., 1998) in place of the modified correlation of Equation 6-133. Comparisons between predicted and experimental HTC values for the ORNL-THTF simulations are shown in Table 12-2 for this version of COBRAHT. The measured local heat flux and wall surface temperature were reported for each thermocouple at different levels for individual rods and as a cross-sectional average value of all thermocouples at each level. In this validation, the HTC data at Level G are used for comparisons to determine the spread of the data relative to the prediction. This level is at 143 inches above the beginning of the heated length, and 1 inch below the top of the active bundle. Each test was screened for HTC "outliers." Figure 12-4 shows a comparison of the predicted and average measured HTCs for the ORNL-THTF DFFB tests. For each test, the COBRAHT film boiling calculation based on the large break LOCA code version underpredicted the bundle average experimental HTC. Figure 12-5 also compares the predicted and measured HTCs, but in this figure, all 235 valid thermocouples are shown. On average, the experimental HTCs are underpredicted by 21.7 percent. This bias in the predicted heat transfer led to the work performed by Bajorek and Young (Bajorek and Young, 1998). 12-4-2 INEL Single Tube Heat Transfer Experiments The COBRAHT version with Forslund and Rohsenow (Forslund and Rohsenow, 1968) driver-pIotter predictions of the ORNL-THTF film boiling data show that the heat transfer package tends to underpredict experimental data. Increasing the direct contact heat transfer improves the predictions, but simply increasing this term according to Equation 6-133 may cause the heat transfer package to overpredict the HTCs for a different range of thermal-hydraulic conditions. o.\4384-nonfsec12.wpd:1b-04033 12-9

Therefore, in validating the WCOBRA/TRAC-SB heat transfer modelling for the small break LOCA regimes of importance (SPV and DFFB), an additional experiment that provides film boiling data, the Idaho National Engineering Laboratory (INEL) single tube test, was also simulated. Post-CH film boiling tests were performed at INEL and are reported by Gottula (Gottula, et al., 1985). Steady-state film boiling tests were conducted in a 15.7-mm inside diameter vertical tube for water flowing upward. The experiments included tests at pressures up to 7 Mpa at mass fluxes ranging from 12 to 70 kgm 2 -s. The test section inlet quality ranged from apprQximately 7 to 47 percent. Steam temperature, and thus the thermal nonequilibrium, was measured using differentially aspirated microthermocouple probes located at various axial positions. Data points for the COBRAHT driver-plotter were selected at four different pressures: 7.0, 3.6, 0.5, and 0.3 Mpa. Of the tests conducted, information for 198 points was available which provided local conditions for pressure, mass flow, quality, and steam temperature at the tube exit. For the INEL film boiling test simulations, a version of COBRAHT containing the WCOBRAITRAC-SB code SPV and DFFB heat transfer package with the drop-wall contact term (per Equation 6-133), was used. Figure 12-6 shows a comparison of predicted HTCs for the INEL film boiling tests using the COBRAHT version containing the heat transfer models in the small break LOCA version of WCOBRA/TRAC. In this comparison, the positive values are underpredictions of the HTC; the negative values are overpredictions. The good agreement indicates the effectiveness of the Re-dependence in the convective and drop-wall contact heat transfer terms in predicting SPV and DFFB heat transfer. Figure 12-7 shows a comparison of COBRAHT results using the WCOBRAITRAC-SB heat transfer model for both the ORNL DFFB and the INEL data. In this comparison, average bias in the HTC is small, with the predicted HTC for all data shown underpredicting the measured HTC by approximately 5 percent. The heat transfer multiplier E is defined as E. = ,, hqi e code where l o:\4384non/sec12.wpd:1b04033 12-10

q," e,p is the experimental heat flux at elevation i code is the predicted heat flux at elevation i in order to ascertain the predictive capability of the WCOBRAflRAC-SB computer code. For each of these tests, the minimum and maximum values of E for any individual data point were determined, as were the average value and the standard deviation of the distribution for the experimental facility simulation results as a whole. For the ORNL DFFB tests, the values are: E =0859, E=,,,l1.332, and Eave=1.031 with standard deviation a=0.09. For the INEL experiment, Emi,n0.571, E,,1.886, and Eave=1.065 with a0.29. 12-4-3 ORNL Uncovered Bundle Heat Transfer Test Simulators A series of experiments investigating small break LOCA phenomena were performed in the ORNL-THTF high pressure rod bundle thermal-hydraulics loop, as reported in NUREG/CR-2456 (Anklam, et al., 1982). The test facility and the WCOBRAITRAC-SB representation of it are described in more detail in subsection 15-4-2 of this report. The ORNL-THTF series of uncovered bundle heat transfer tests provide another set of heat transfer data for small break LOCA model validation. The six uncovered bundle tests (I through N) were simulated with WCOBRAfTRAC-SB. The uncovered bundle tests were steady-state experiments with electrically heated rods in which the inlet liquid mass flow was approximately equal to exiting steam mass flow. Rod temperatures and heat transfer coefficients in the stean-cooling region of the rod bundle were determined. Once the steady-state condition was established, thermal and hydrodynamic data were collected at several heights in the uncovered portion of the rod bundle. Table 12-3 summarizes the steady-state test conditions in terms of system pressure, linear rod power, and inlet mass flux. The tests were characterized by low (580-650 psia) or high (1010-1090 psia) pressure, and low (0.10 and 0.14 kW/ft), medium (0.31 and 0.33 kW/ft), or high (0.66 and 0.68 kW/ft) linear power. The WCOBRAfrRAC-SB model described in Section 15 was used for these ORNL test simulations, using an appropriate value of the interfacial drag multiplier (YDRAG). This multiplier adjusts the interfacial shear calculated between rising bubbles and liquid, and its use allows the separation of WCOBRA/RAC-SB's heat transfer and hydrodynamic packages. Values of YDRAG to alter WCOBRAITRAC-SB's two-phase level to better match the experimental two-phase level were used in the WCOBRAITRAC-SB input decks that simulated oM384-nonlsecI2.wpd:Ib-04033 12-1 1

I these tests. With this approach the difference between the predicted heat transfer and the data can be attributed to the WCOBRAITRAC-SB heat transfer models alone. Overall, rod temperatures were under-predicted for the uncovered bundle tests, while vapor temperatures were slightly over-predicted. WCOBRA/TRAC-SB vapor heat transfer coefficients were generally greater than the experimental values. With all 10 ORNL-THTF experimental levels included (the top 2' of the rod bundle), the average ratio of experimental to WCOBRA/TRAC-SB vapor heat transfer coefficients (Ei) was 0.7769, with a standard deviation of 0.18 (see Figure 12-8). The E. and En,, values are 1.314 and 0.452, respectively. The ORNL uncovered bundle data are in the same Reynolds number range and exhibit heat transfer coefficients of similar magnitude to the INEL test data. There appears to be a wide range on the WCOBRA/TRAC-SB heat transfer multipliers for these datasets, from 0.452 to 1.886, or over a factor of 4. However, this is not surprising. As noted by Anklam (Anklam et al., 1982), the flow regimes present in these tests can vary between forced, mixed and free convection. This uncertainty about flow regime means that a significant range of results may be expected to occur in the WCOBRAIRAC-SB predictions of the ORNL-THTF uncovered bundle tests. Anklam further notes that convective heat transfer under the high pressure uncovered bundle conditions can be very complex because of the number of possible flow regimes and flow transitions that may occur. 12-5 Summary and Conclusions The WCOBRAfrRAC-SB heat transfer modelling of the SPV and DFFB regimes important to small break LOCA analysis has been assessed. The effect of using the drop-wall contact expression of Bajorek and Young (Bajorek and Young, 1998) in place of that of Forslund and Rohsenow (Forslund and Rohsenow, 1968) is shown to markedly improve predictions of the ORNL film boiling tests. The implementation of the Bajorek and Young drop-wall contact term and a Re-dependent laminar/turbulent flow transition convective heat transfer term has allowed the WCOBRAITRAC-SB heat transfer package to predict the ORNL DFFB and uncovered bundle tests and INEL film boiling test data well. The statistical treatment of core heat transfer in the Westinghouse uncertainty methodology is based on these results as described in Volume 4 of this document.

                                                                                                  .LI o:\4384-non/secI2.wpd:lb-04033                  12-12

12-6 References Anklam, T. M., et al., 1982, "Experimental Investigations of Uncovered Bundle Heat Transfer and Two-Phase Mixture Level Swell Under High Pressure Low Heat Flux Conditions," NUREG/CR-2456. Bajorek, S. M., et al., 1998, "Code Qualification Document for Best Estimate LOCA Analysis, Volume I: Models and Correlations," WCAP-12945-P-A, Vol. 1. Bajorek, S. M. and Young, M. Y., 1998, "Assessment and Quantification of WCOBRA(IRAC-MOD7A Heat Transfer Coefficients for Blowdown Dispersed Flow Film Boiling." Proc. International Conf. On Nuclear Engineering (ICONE-6), Paper 6184. Dittus, F. W. and Boelter, L. M. K., 1930, "Heat Transfer in Automobile Radiators of the Tubular Type," Publications in Engineering, 2, Univ. of California, Berkeley, pp. 443461. Forslund, R. P. and Rohsenow, W. M., 1968, "Dispersed Flow Film Boiling," J. Heat Trans., Vol. 90, No. 6, pp. 399-407. Gottula, et al., 1985, "Forced Convective, Non Equilibrium Post-CHF Heat Transfer Experiment Data and Correlation Comparison Report," NUREG/CR-3193. U.S. NRC Regulatory Guide 1.157, 1989, "Best Estimate Calculations of Emergency Core Cooling System Performance." Wong, S. and Hochreiter, L. E., 1981, "Analysis of the FLECHT-SEASET Unblocked Bundle Steam Cooling and Boiloff Tests," NRClEPRI/Westinghouse-8. Yoder, et al., 1982, "Dispersed Flow Film Boiling in Rod Bundle Geometry-Steady State Heat Transfer Data and Correlation Comparisons," NUREG/CR-2435, ORNL-5822. o:\4384-nonfsec12.wpd:Ib-04033 12-13

I Table 12-1 ORNL-THTF Steady-State DFFB Tests Pressure Mass Flux Inlet Power Test (bar) (kg/mi-s) Temperature (C) (kW/m) B 127.6 713 310 2.52 C 124.5 334 293 1.58 D 127.5 518 303 1.92 E 131.7 593 304 1.99 K 43.8 226 213 1.24 L 23.0 527 276 2.17 M 85.7 657 284 2.47 P 60.3 520 267 2.26 Q 65.3 325 261 1.58 X 60.1 344 268 1.64 o:\4384-nonlsec12.wpd:1b-04033 12-14

Table 12-2 Summary of ORNL-THTF Driver-Plotter Comparison, Forslund/Rohsenow Model Data HTCdd,C/ Test Samples HTC&tia,mzn HTCda,ae HTCdatrar HTCcoie HTCCOdc B 24 451.5 498.5 529.9 405.45 1.229 C 24 221.2 247.4 276.7 185.91 1.331 D 24 342.1 379.3 400.5 275.69 1.376 E 24 381.2 427.7 447.6 317.11 1.349 K 20 145.9 156.1 165.8 116.73 1.337 L 26 298.7 331.1 356.1 291.57 1.136 M 26 350.4 391.7 415.8 353.07 1.109 P 22 280.5 308.5 327.0 280.35 1.100 Q 26 196.0 213.1 232.3 192.00 1.110 X 19 205.4 221.1 236.3 200.47 1.103 o:\4384-non/sec12.wpd:1b04033 12-15

Table 12-3 ORNL Uncovered Bundle Test Matrix

                                                                                 .t,.

I J K L M N System Pressure (psia) 650 610 580 1090 1010 1030 Linear Power (kW/ft) 0.68 0.33 0.10 0.66 0.31 0.14 2 0.23E-4 2.15E-4 0.93E4 0.34E-4 Mass Flux (bm/h ) 2.19E-4 0.94E-4 o:\4384nontsecl2Zwpd:b-04033 12-16

a,c Figure 12-1. Heat Transfer Regime Map for Vessel Component o:\4384-nonlsecl2.wpd:Ib-04033 12-17

I, Known Known experimental values HTC for a, Twai, Tvap, mvap, mljq "COBRAHT.F" WCOBRA/TRAC heat transfer package + property routines

1. Select HT regime

2. Calculate hwl, hw, q"

3. Re-define HTC as hw-r I

zare < I l Predicted HTC l 4 Quantifiable Measure of HTC Modell Figure 12-2. WCOBRAJTRAC Heat Transfer Driver-Plotter Routine oA4384-nonIsec12.wpd:1b04033 12-18

0.104m A 6 C) E I~~~Hae o Daee .5c Unheated Rod 1.27 cm Heated Rod Diameter - 0.95 cm Unreated Rod Diameter - 1.02 cm Figure 12-3. ORNL-THI F Rod Bundle Cross Section o:M4384-nonJsecl2.wpd:Ib-04033 12-19

                                                                                          'Li D r i v e r-P I o t t e r      Comp a r i son         of WCOBRA/TRAC Heat       Trdnsf6r      Ru t i ne    f 6r ORNL       Film Boil ing Dtb a          * ?THCOO 1O         9        0         0 HTC   Code (W/m2-K) 3000
      "   2500 E

3 2000 C-)

     ~ 1500
     -V O     500 co Q- 0
                   *-        o   .i AJLL     4L           14L        JLL      JL .LJ u          Uu       1  uUU      1 5uu        zuuu     4.3uU      Ju u Measu red HT C                ( W/m2-K Figure 12-4. Comparison of Large Break LOCA WCOBRAITRAC Code Logic Predicted and Measured HTCs for ORNL-TiTF DFFB Tests - Bundle Average o\4384-nonfsecI2.wpd:Ib-04033                    12-20

B I ow down Cooling Heat Tra ns f e r ORNL Heat Transfer Test Compa r i sons __ 3500

       "   3000 3:_   2500
                                -~~~~~~~~~~~~~~~T             -  mw C-,

2000 1 500 f i l l-1 000. C.) 500 ta,

      -L 0

0 500 1 00 1500 2000 2500 3 0 350 0 Me a s u red HTC. W/m2-K Figure 12-5. Comparison of Large Break LOCA WCOBRAfTRAC Code Logic Predicted and Measured HTCs for ORNL-THTF DFFB Tests - All Thermocouples o:\4384-non/sec12.wpd:1b-04033 12-21

Dr i ve rLPI o t ter Compar i so n ot WCUBKA/IKALG Heat Transfer Rou tine to INEI L Data - w/Revisions a n M- 0 O 9 . 9 .1 O H3TC Code 1 a- 5 L~~~~ 0 ur I-,0

  -oC.)
                             -             --                    9-O 0    -1 . 5
                               - ---    Is    l l    s-i--     t   l    I
                                                                      - b-l l       l-O        -2 I
         -2 5 0                  5V000           1 0000          15000          20000 Va p o r Reyno I d         s    Numbe r Figure 12-6.        Comparison of Predicted and Measured INEL Film Boiling HTCs as a Function of Vapor Re Using WCOBRA/TRAC-SB Heat Transfer Models o:4384-nontsec12.wpd:1b-44033                    12-22

B I owd o w n Coo I i ng He at Transfer O R N L a n d I NEL D a t a Comparison 4 10 cI 10 z~~~~~~~~~~~~~~~~~~~~~~ 2 10

   =                              INEL     K       D   --            ORNL C-,

I- 1 10 la 0 10o a- -1 0 0M3 0 4 10 10 10 10 10 10 Measured HTC B t u/h f t2-F Figure 12-7. Comparison of Predicted and Measured HlTCs for Combined ORNL DFFB and INEL Data Using WCOBRA/TRAC-SB Heat Transfer Models o\4384-nontsecI2.wpd:Ib-04033 12-23

WCT HTCV vs. Experimental HTCV (10 levels for each test) I-0 0 10 20 30 40 S0 60 70 Expenmental HTCV (Blt2.hr) ORNL-THTF Uncovered Figure 12-8. Comparison of Predicted and Measured HTCs for Bundle Data Using WCOBRA/TRAC-SB 33 12-24 o\4384-nOn/secl2-wpd:1O40

SECTION 13 ASSESSMENT OF BREAK FLOW MODEL 13-1 Introduction During a small break LOCA, the break flowrate determines the depressurization rate as well as the mass inventory of the primary system of a PWR. These parameters in turn influence the timing of various engineered safeguard system responses, such as reactor trip and safety injection. Early in a small break LOCA, the fluid condition upstream of the break location is subcooled. This results in a high discharge flowrate and a fast depressurization. As the pressure drops to the saturation pressure corresponding to the coolant liquid temperature upstream of the break, the discharge becomes two-phase and a relatively low discharge rate and a slow depressurization result. As the system mass depletes and the flow in the main pipe stratifies, the break location (typically a branch pipe) begins to uncover. This results in the void fraction upstream of the break changing from 0.0 (saturated liquid) to 1.0 (saturated vapor) as the liquid level in the main pipe drops. As the stratified surface lowers in the vicinity of the break, the quality at the break is greatly influenced by the entrainment of vapor/liquid off the stratified surface upstream of the break. Although the size, location, and shape of the break are not known for the postulated small break LOCA, the best estimate code needs to predict consistent responses relative to experimental data over a range of pressure, subcooling, and upstream fluid states, as well as the break flow area variations, so that accurate sensitivity to small break LOCA responses can be obtained. In this section, an assessment is made of the break flow model in the WCOBRAJTRAC-SB version described in Section 3, Volume 1, of this document. This version was created for the small break LOCA application from the WCOBRAtRAC-MOD7A, Rev. 4. 13-2 Critical Flow in Small Break LOCA A fluid system contained in a reactor vessel with a pipe break is in communication with the containment atmosphere, which is at a lower pressure through the break flow path. Under critical flow conditions, the discharge flowrate from the high pressure system becomes independent of the containment conditions, which are at the lower pressure. o-W384-non4384-13.wpd:1b-4303 13-1

ff lI WESTINGHOUSE PROPRIETARY CLASS 2 13-2-1 Subcooled Liquid Discharge Early in a small break LOCA, the fluid condition upstream of the break is subcooled. As the fluid accelerates through the path leading to the break, the static pressure decreases and the liquid flashes at the throat as seen in Figure 13-2-1. In subcooled liquid discharge, the degree of subcooling thus greatly influences the break flowrate. At the onset of a small break LOCA, Ps: (T,) is substantially lower than the primary system pressure. This results in a relatively high break flow. As the system depressurizes, the liquid subcooling decreases and the break flow lowers accordingly. Even slightly subcooled liquid going through a sudden depressurization does not flash at P.,, (T,) due to the underpressure at flashing inception (or nucleation delay). As a result, the throat pressure is lower than P,,,, (T,) and the break flow is still higher than the two-phase break flowrate. The nucleation delay (or nonequilibrium effect) dominates the break flowrate for the subcooled liquid in a geometry in which the fluid accelerates into a short flow path to an opening. The fluid going through a rapid expansion is not able to flash instantaneously; it remains as superheated liquid until at or near the throat, where the static pressure becomes lower than the underpressure required for the nucleation and the fluid becomes two-phase. 13-2-2 Stratified Entrainment at Break As the system mass depletes, the break location becomes two-phase and eventually becomes stratified following an RCP trip. The two-phase break flowrate is a strong function of the upstream quality. Stratified flow conditions near the break may lead to vapor and liquid entrainment into the break path. The entrainment amount depends on the velocity of the fluid near the break and the height of the stratified liquid level in the pipe relative to the break elevation as seen in Figure 13-2-2. The location of the branchline leading to the break, relative to the liquid level, determines the quality of the two-phase mixture in the branchline and at the break. In these conditions, the vapor pull-through and liquid entrainment may become important as seen in Figure 13-2-3. oA4384-nonM4384-13.wpd:1Wb4303 13-2

WESTINGHOUSE PROPRIETARY CLASS 2 13-2-3 Correlation for Onset of Liquid and Vapor Entrainment A correlation for the onset of liquid and vapor entrainment to the branchline was suggested by Zuber (Zuber, 1980). This correlation is based on earlier investigations by Craya (Craya, 1949) and Lubin (Lubin, 1967). The liquid entrainment off the side orifice was derived by Craya as: Fr = C* ( h!Jt) where U (13-1) Frg= _ d g - pg Vapor pull-through off the bottom orifice was derived by Lubin as: Fr, = dC where UF (13-2) Fr, d g- AP pi where Fr. and Fr,are the Froude numbers for vapor and liquid U8 and U, are phasic velocities in the break path d is the break diameter hait is the distance between the stratified level and the break elevation at the onset of entrainment AP = P-Pg Pl, p8 are the liquid and vapor densities g is the gravitational acceleration o:W384-non\4384-13.wpd:b-4303 13-3

I WESTINGHOUSE PROPRIETARY CLASS 2 In both cases, the constants are theoretically derived to be C, = 3.23 and C2 = 2.5. Other researchers have experimentally deternined these constants, and the results are tabulated in Table 13-2-1. The following are the selected correlation constants for WCOBRAIrRAC-SB:

                                                ]a 13-2-4 Correlation for BreakBranchline Quality The following correlations were selected for WCOBRA/TRAC-SB:

Upward-Vertical Branch Schrock (Schrock, et al., 1986) proposed the following correlation for this orientation: x R3z2 (I R)2 (13-3) where R = tl and where h is the distance between the break and the liquid surface. o\4384-non\4384-13.wpd:lb-41603 13-4

WESTINGHOUSE PROPRIETARY CLASS 2 Downward-Vertical Branch Smoglie and Reimann (Smoglie and Reimann, 1986) suggested the following correlation for the bottom branch: X = X * [1 0.5 R (1 + R) X6 ] (13-4) where R= 1.15 x0 = 1 Pf N Pg

• Horizontal Side Branch Smoglie and Reimann also suggested the following correlations for the side branch:

Horizontal above the midplane x= 1.09 [1 - 0.5- R (1 + R) x R)]05 (13-5) where R= h hcrit 1.15 xO =

                               +       Pf Pg oA4384-non\4384-13.wpd:1b-4303                         13-5

WESTINGHOUSE PROPRIETARY CLASS 2 Horizontal below the midplane

                        =   oR)       [ 1 - 05 R   (1 + R) x(                       (13-6) where 1.15 X =

1 + Pf Pg Section 13-3 shows the result of an assessment of the ability of WCOBRA/TRAC to predict the branchline quality as a function of the mainline liquid level. o:\4384non\4384-13.wpd:1b-4303 13-6

WESTINGHOUSE PROPRIETARY CLASS 2 Table 13-2-1 Experimental Results of C, and C2 Yonomoto Smoglie and Ardron Item and Tasaka Reimann Schrock Anderson Maciaszek and Brycela) Maximum 0.7 0.5 1.1 6.0 2.0 pressure (MPa) Main pipe 190 x 190 206 102 102 284 135 diameter square (mm) Branch pipe 10,20 6-20 3-10 3 34 20 diameter (mm) Fluid Air/water Air/water Air/ Steam/ Stean/water Steam/water - water water Top break 3.22b) 0.35/ 0.4/2.5 0.4/2.5 2.17/1.5 0.35/2.5 2.5° 2.5 _ Side break 4.29/2.5 3.22/2.5 3.25/ - 4.21/2.5 4.21/2.5 3.22/2.5 liquid 2.5 entrainment Side break 2.61/2.5 2.61/2.5 2.2/2.0 1.19/2.0 2.0912.5 4.21/2.5 2.61/2.5 vapor entrainment Bottom break 1.27/2.5 0.94/2.5 1.47/ 0.78/2.0 1.27/2.5 1.27/2.5 0.46/2.5

. _ _____ ____ _ __ _______ ______ ______ ___       2.0   I

a. No experiments were performed for this work.

b. Constant C,

c. Constant C2

References:

Yonomoto and Tasaka, 1988 Anderson and Benedetti, 1985 Smoglie and Reimann, 1986 Maciaszek and Memponteil, 1986 Schrock, et al., 1986 Ardron and Bryce, 1990 o:\4384-non\4384-13.wpd:lb-41603 13-7

I WESTINGHOUSE PROPRIETARY CLASS 2 Psys TL

                                                                         'O PTHROAT Figure 13-2-1. Diagram of Subcooled Break Flow Vapor            2 phase mixture Figure 13-2-2. Diagram of Two-Phase Upstream Conditions o:\4384-non\4384-13.wpd:lb-4303                    13-8

WESTINGHOUSE PROPRIETARY CLASS 2 Vapor pull-through Vapor

                                                °o0     0           0    0 Uquod    n      0    00 Liquid entrainment Figure 13-2-3. Vapor Pull-Through and Liquid Entrainment Phenomena o-.\4384-non\4394-13.vwpd:Ib-4303                 13-9

I WESTINGHOUSE PROPRIETARY CLASS 2 13-3 Assessment of the Horizontal Stratified Entrainment Model 13-3-1 Branchline Quality/Mainline Liquid Level Comparison Using TPFL The break flowrate is a function of the pressure and the flow quality at the break and the size of the break. For two-phase fluid conditions encountered in a small break LOCA, this quality is, in turn, a function of the stratified level in the mainline and the liquid entrainment/vapor pull-through behavior. The break flow experiment performed at the two-phase flow loop (TPFL) in INEL specifically examined a break flow from the branchline off of a simulated hot/cold leg where the two-phase flow is horizontally stratified. This experiment was motivated by the apparent inability to predict the break flow by RELAP5 and TRAC (Condie, 1980), when they were used to simulate LOFT test L3-5 (small break LOCA experiment) (Doa, 1980). The break flow predictions by the two codes were substantially higher than the experiment. The overprediction of the break flowrate and the depressurization rate was thought to be the consequence of the inability of the codes to pull through the vapor when the stratified level is above the branchline entrance. The objective of the TPFL experiment is to develop a reliable and accurate experimental data base for critical flow through small pipe breaks in which stratified two-phase flow is prevalent. Recent experiments with an air-water mixture in small pipes have indicated that liquid entrainment and vapor pull-through significantly influence discharge rates and that these phenomena are functions of stratified level and break azimuthal location. The objective of these experiments is to obtain accurate data on critical discharge rates of steam-water mixtures as a function of break orientation and stratified level. More specifically, the objectives of the experiments are to establish the following:

• An experimental data base on critical flow through small breaks for two different break orientations: bottom and side
• A data base relating discharge rates to stratified level and thermal-hydraulic conditions in the mainline
• An experimentally measured data base relating the discharge rate and level in the mainline to the conditions in the branchline oA4384-non\4384-13.wpd:1b.4303 13-10

WESTINGHOUSE PROPRIETARY CLASS 2 13-3-1-1 Description of Test Facility The tee/critical flow experiments were performed in the TPFL at the INEL Thermal Hydraulics Laboratory (Figure 13-3-1). The loop consists of 28.4-cm diameter mainline pipe which drains into a separator tank, and 3.44-cm diameter branchline which has a 1.62-cm diameter critical flow nozzle providing a known choke point. The mainline pipe is 7.9 meters long measured from the steam/water mixer to the separator. The branchline inlet tees off from the mainline pipe 2.7 meters from the separator. The branchline could be attached to the mainline either at the side (horizontal configuration) or at the bottom (vertical configuration). The schematic view of the facility is shown in Figure 13-3-2. A six-beam gamma densitometer was used 0.4 meters upstream of the branchline entrance in the mainline to determine the level in the mainline. A single-beam gamma densitometer was used to measure the density 0.3 meters upstream of the nozzle in the branchline. 13-3-1-2 Test Ranges In the experiment, for each configuration (horizontal and vertical), the mainline liquid level varied from 0 cm (all vapor) to 24 cm (85 percent of pipe diameter) at the system pressures of 900 psia, 640 psia, and 500 psia. 13-3-1-3 WCOBRA/TRAC Model The vessel component of WCOBRA[IRAC uses the COBRA-TF formulation and can model multidimensional flows, such as the flow in the vessel of the RCS. Figure 13-3-3 shows the WCOBRAITRAC noding diagram for TPFL simulation. This vessel component [ Ia,c 13-3-1-4 Comparison of WCOBRAtI'RAC Prediction to Horizontal Data Figure 13-3-4 shows the comparison of the WCOBRATRAC-SB prediction for the branchline quality as a function of the mainline liquid level for the horizontal configuration. WCOBRAIRAC predicted the liquid level for the onset of vapor pull-through at D. x 0.75, which compares well with the experimental observation. WCOBRAJIRAC predicted the onset o:\4384-non\438413.wpd:lb-4303 13-1 1

I WESTINGHOUSE PROPRIETARY CLASS 2 of the liquid entrainment at DH x 0.22, which also compares well with the experiment. WCOBRAITRAC slightly overpredicted the vapor pull-through. In the TPFL horizontal configuration, the prediction and the data compared well. The trend of quality variation relative to the liquid level is also correct though higher than the data when the liquid level is above the mid-elevation in the mainline pipe. 13-3-1-5 Comparison of the WCOBRAITRAC Prediction to Downward-Vertical Data Figure 13-3-5 shows the comparison of the WCOBRAIRAC-SB prediction and the experimental data of the branchline quality as a function of the mainline liquid level for the downward-vertical configuration. Predictions at 900 and 500 psia are plotted against the data taken at 900, 640, and 500 psia Both an onset of vapor pull-through and the quality as a function of liquid level are well predicted by WCOBRATRAC-SB. The prediction did not show a significant trend relative to the pressure change, which is in agreement with the data. 13-3-1-6 Comparison of the WCOBRA/TRAC Prediction to Upward-Vertical Data Figure 13-3-6 shows the comparison of the WCOBRA/TRAC-SB prediction for the branchline quality as a function of the mainline liquid level for the upward-vertical configuration. Because there. are no TPFL data for this configuration, the WCOBRAIRAC prediction for this configuration with TPFL geometry and pressure was compared against data taken at much lower pressures. This set of data was used to benchmark the correlation given by Equation 13-3, in Ardron and Bryce (Ardron and Bryce, 1990). o:\4384-non\4384-13.wpd:1b-4303 13-12

WESTINGHOUSE PROPRIETARY CLASS 2 Figure 13-3-1. Diagram of TPFL o:\4384-non\4384-13.wpd:Ib-4303 13-13

WESTINGHOUSE PROPRIETARY CLASS 2 DE-1A, 19, 1C DE-2A, 2B, 2C' PDE-310 Steam tee 301/ DE.BL-IA, 1, 1C PE.305 302 1 f-t TE.300 W01 Luj 50 cm Xl PE-300 DE-3 BA, 3 163 cm _ PDE-341 3.4 cm branchline TE-402 DEE' DE--4 Critical flow nozzle PDE-45

                                         <                ~~86 cm a
                                                -t  _      Catch tank Figure 13-3-2. Schematic View of TPIFL Test Section o:\4384-non\4384-13.wpd:1b 4 303                      13-14

WESTINGHOUSE PROPRIETARY CLASS 2 Figure 13-3-3. WCOBRAJTRAC Noding for TPFL Branchline Quality Test Simulation o.\4384-non\43&4-13.wpd:lb-4303 13-15

WESTINGHOUSE PROPRIETARY CLASS 2

                  -~         QUAL I TY        10      1       0 WC/T AT 900psia D        O Quo I i ty         1      0       0  900 psia Horiz o        OQua I  i ty         1      0       0  640 psia Horiz A        A Qua I i ty         1      0       0  500 psia Horiz 1

C:J

              .)

I-m 0 Normalized Mainline Liquid Level (HL/D) Figure 13-34. Branchline Quality Versus Mainline Liquid Level for Horizontal Configuration o:\4384-non\4384-13.wpd:1b-4303 13-16

WESTINGHOUSE PROPRIETARY CLASS 2 OUAL I TY 10 29 0 WC/T at 900 psia OUAL I TY 10 29 0 WC/T at 500 psia O OOual i ty 1 0 0 900 psia Vertical o OQua I i ty 1 0 0 640 psio Vertical A ACua, i ty 1 0 0 500 psia Vertical I

               =3 -\~     0 (D0.5-C)1 0

1.. Figure 13-3-5. Branchline Quality Versus Mainline Liquid Level for Downward-Vertical Configuration o:4384-non\4384-13.wpd:1b4303 13-17

WESTINGHOUSE PROPRIETARY CLASS 2 WCOBRA/TRAC at TPFL scale at 900 psia a a Air/Water UCB (Schrock et. al.. 1986) 0 C Steom/Woter uCB (Schrock et. ol.. 1986) A A Air/Water KfK (Smoglie et. l.. .1986)

          - - - -Equation          13-3 (Schrock et. al.. 1986) 1-
           .8 C.)
           .4 0

Quality Figure 13-3-6. Branchline Quality Versus Mainline Liquid Level for Upward-Vertical Configuration oA4384-non\4384-13.wpd:1b-4303 13-18

WESTINGHOUSE PROPRIETARY CLASS 2 13-4 Assessment of the WCOBRAITRAC Break Flow Model 13-4-1 Assessment Objective In this section, the break flow model in WCOBRAIRAC-SB is assessed relative to the following effects on the break flow:

• Break path length
• Break flow area variation
• Upstream pressure variation
• Variation in degree of subcooling during liquid discharge
• Upstream void fraction/quality variation
• Break entrance geometry The critical flow model's bias and uncertainty will be determined by comparing the critical flow model prediction implemented in WCOBRAflTRAC-SB with selected data from the qualified break flow dataset by V. fllic et. al. (1986), the Marviken full scale critical flow test (EPRI-NP-2370, 1982, and Amos and Schrock, 1983) and the Two Phase Flow Loop (TPFL)

(Anderson and Benedetti, 1985). 13-4-2 Assessment Test Matrix Dataset mentioned in V. Illic (1986) was further examined for selection for comparison and bias/uncertainty evaluation. Data without well defined stagnation condition or upstream condition were excluded at this time. Dataset by Cruver (1963), Fauske (1962), Henry (1990), Isbin (1957) and Zaloudek (1964) do not report stagnation pressure. Dataset by Guizovarn (1975) contains superheated liquid upstream of the nozzle, which is contrary to the description in V. Illic (1986) which states subcooled inlet condition. Dataset by Bryers and Hsieh (1966) contains highly subcooled stagnation condition contrary to the description. The dataset by Ogasawara (1969) did not contain the reservoir temperature or the quality. Datasets by Danforth (1941) and Schrock (1977) are suspect with regard to achieving the critical condition according to Illic. Dataset by Morrison (1977) was felt to be inconsistent with other similar data. The dataset mentioned above need to be further investigated for the use in the bias and uncertainty study since as-reported upstream condition is suspect. Table 134-1 is a summary of all selected dataset for this assessment. The dataset represents more than 1400 points from 40 geometries containing data from 13 to 2500 psia. The geometry ranges from 0 < L < 2300mm, 0.464 < DH< 500mm. o:\4384-non\4384-13.wpd:1b-4303 13-19

WESTINGHOUSE PROPRIETARY CLASS 2 Table 134-1 Selected Dataset and Input Variables No. of Data Set Pressure Data Length Dhyd No. Reference (psia) Upstream Condition Points (mm) (mm) 1 Ardron (1978) 22-55 Subcooled 32 1015 26.3 2-4 Boivin (1979) 200-1500 Subcooled 21 500-1830 12-50 5 Fincke (1981) 13-45 Subcooled 92 79.72 18.28 6-7 Jeandey (1981) 100-2100 Subcooled 88 463 20.13 8-9 Neusen (1962) 100-600 Subcooled 37 0 6.4-11.125 10 Reocreux (1974) 30 50 Subcooled 28 2335 20 11-12 Seynhaeve (1980) 40-150 Subcooled 57 221-306 12.5 13-33 Sozzi (1975) 400-1100 Subcool and Saturated 667 4.7-1822.5 12.7 34-37 Marviken (1982) 400-750 Subcooled and Saturated 252 150-300 300-500 38-39 Arnos (1983) 500-2300 Subcooled 44 63.5 0.464-0.748 40 Anderson (1985) 500-900 Saturated Liquid up to 109 54 16.2 Saturated Vapor TOTAL 13-2300 Subcooled Liquid to 1427 0-2335 0.418-500 Saturated Vapor The next table, Table 134-2, is the complete assessment test matrix used to evaluate the accuracy of the WCOBRA-TRAC break flow model; it describes in detail all 40 nozzle geometries and orientations. o:\4384-non\4384-13.wpd:1b-4303 13-20

WESTINGHOUSE PROPRIETARY CLASS 2 Table 13-4-2 Critical Flow Data Considered for Model Evaluation Data Set L D No. Reference (mm) (mm) cosO N-Data Counents I Ardron, K H. &Ackerman, M. C. (1978) 1015 26.3 0 33 One superheated upstream condition was not used 2 Boivin (1979) 500 12 0 10 D=50 (z<0); <z<50rounded entrance; D=12 (50<z<500); D=12+19(z- 500) (500<z<700); D=50 (z>700 rmm) 3 Boivin (1979) 1600 30 0 5 D=150 (z<0); (kz<30 rounded entrance; D=30 (130<z<1730); D=30+0.12(z-1730) (1730<z<2305); D=100 (z>2305 um) 4 Boivin (1979) 1700 50 0 6 D=150 (z<0); 0cz<130 rounded entrance; D=50 (130<z<1 830); D--50+0.12(z-1830) (1830<z<2240); D=100 (z>2240 mm) S Fincke & Collins (1981) 13 44 0 92 D=1 8.28 (54.7cz<79.7); D-18.28+0.12(z-79.7), (z215.9 mm) 6 Jeandey et al. (1981) 463 20 1 15 D=66.7-0.54z (0<z<86.9); D-20.1 (z>86.9 rmm) 7 Jeandey etal. (1981) 463 20 1 73 see Appendix C.7.1 for(z<100); D--20.13 (100.cz<463); D-20.13+0.12(z- 463) (z<900); D=737 _ 4>(900 mu) 8 Neusen (1962) 0 11 0 7 D=11.12 nm atthroat; D=11.12+0.425z _______ (0<z<35.91 un) 9 Neusen (1962) 0 6 0 5 D=16.4 mm at throat; D=6.4+0.425z (0<z<59.81 mnm) 10 Reocreux (1974) 2335 20 1 28 D-20 (Oczc2335); D=20+0.12(z-2335) (z<2662 nn) 11 Seynhaeve (1980) 306 13 1 26 D=12.5 (Dczc306);D=12.5+0.245(z-306) (z>541); ______ . -70 (z>541 mn) 12 Seynhaeve (1980) 306 13 1 31 D=12.5 (0czc221); D=12.5+0.245(z-221) (z>541); ______ _ D70 (z>541 rn) 13 Sozzi & Sutherland (1975) 45 12.7 0 129 D=43.2 (z=0); rounded convergent (0cz<44.5); _______ D=12.7+0.105(z-44.5) (z<1585 mm) (Nozzle 1) 14 Sozzi & Sutherland (1975) 45 12.7 0 13 D=43.2 (z=O); rounded convergent (0<z<44.5 mm) ___________ _ _ _ (Nozzle 2) 15 Sozzi & Sutherland (1975) 57 12.7 0 47 D=43.2 (z=0); rounded convergent (0<z<44.5 mnm) I (Nozzle 2) 16 Sozzi & Sutherland (1975) 362 12.7 0 19 D=43.2 (z=0); rounded convergent (0<z<44.5 mrm) _ _____________________________ _______ (Nozzle 2) 17 Sozzi & Sutherland (1975) 83 12.7 0 17 D=43.2 (z=O); rounded convergent (0<z<44.5 rmnm) (Nozzle 2) 18 Sozzi & Sutherland (1975) 553 12.7 0 13 D=43.2 (z=O); rounded convergent (0<:z<44.5 rmn) (Nozzle 2) 19 Sozzi & Sutherland (1975) 108 12.7 0 23 D=43.2 (z=O); rounded convergent (0<z<44.5 rmm) (Nozzle 2) 20 Sozzi & Sutherland (1975) 679 12.7 0 96 0=43.2 (z=0); rounded convergent (0<z<44.5 mm)

                                                                   .___ _ _ _ _ _ _ _(Nozzle

_ 2) 21 Sozzi & Sutherland (1975) 159 12.7 0 15 D=43.2 (z=O); rounded convergent (0<z<44.5 rnm) (Nozzle 2) 22 Sozzi & Sutherland (1975) 1823 12.7 0 81 D=43.2 (z=O); rounded convergent (0<z<445 mm) _ __________________________ _ _(Nozzle 2) 23 Sozzi &Sutherland (1975) 235 12.7 0 12 D=43.2 (z=O); rounded convergent (0<z<44.5 runm) I__ (Nozzle 2) o:\4384-non\4384-13.wpd:1b-4303 13-21

WESTINGHOUSE PROPRIETARY CLASS 2 Table 134-2 (Cont'd) Critical Flow Data Considered for Model Evaluation set L D No. Reference (mn) (mm) cosO N-Data Comments 24 Sozzi &Sutherland (1975) 273 12.7 0 22 D=43.2 (z=0); rounded convergent (O<z<44.5 Iu) 25______&_Sutnerland_(1975)_5_12.7_o_58 (Nozzle 2) 25 Sozzi &Sutherland (1975) 5 12.7 0 58 Nozzle No. 3 (Sharp entrance) 26 Sozzi & Sutherland (1975) 322 12.7 0 24 Nozzle No. 3 (Sharp entrance) 27 Sozzi & Sutherland (1975) 513 12.7 0 24 Nozzle No. 3 (Sharp entrance) 28 Sozzi & Sutherland (1975) 640 12.7 0 17 Nozzle No. 3 (Sharp entrance) 29 Sozzi & Sutherland (1975) 195 12.7 10 23 Nozzle No. 3 (Sharp entrance) 30 Sozzi & Sutherland (1975) 45 19 0 23 D=43.2 (z=o); rounded convergent (0<zc44.5 mn) 31 Sozzi &Sutherland (1975) 732 54 0 4 D-260.0.39(z-202) (202<z<732); D-54+0.263(z.732) (z<1112 un) 32 Sozzi &Sutherland (1975) 696 76 0 3 D-260-0.39(z-223) (223<2<696); D=54+0.263(z-696) (z<1076 mnm) 33 Sozzi &Sutherland (1975) 63 28 0 5 D-72.6 (z=O); munded elliptical sec. (0<z<63.5); Dl=28+0.246(z-63.5) (z<228.5) 34 Marviken Test 6 (1982) 300 300 -l 84 Rounded entrance 35 Marviken Test 7 (1982) 300 300 -l 84 Rounded entrance 36 Marviken Test 23 (1982) 150 500 -l 44 Rounded entrance 37 Marviken Test 24 (1982) 150 500 1 39 Rounded entrance 38 Amos &Schrock (1983) 63.5 0.747 -1 18 Rec. Slit 0.381x63.5 mn with known entrance losses 39 Amos &Schrock (1983) 63.5 0.418 -1 26 Rec. Slit 0.254x63.5 rmn with known entrance losses 40 Anderson & Benedetti (1985) 31.9 16.2 0 109 Rounded entrance ( at 500. 640 and 900 psia) The following material contains a brief description and graphical presentation of the upstream conditions of experiments for selected data sources. The stagnation condition of each dataset such as Pressure/Temperature and Pressure/Quality are shown graphically in the following figures. The Pressure/Temperature trajectories of the primary system of LOFT and ROSA during small break LOCA experiments along with the saturation line are shown for comparison. Ardron and Ackerman Ardron and Ackerman conducted critical flow experiments by discharging subcooled water from a pressure vessel through a horizontal test section. The test section consisted of a straight cylindrical pipe 0.0263 m in diameter and 1.015 m long. Instrumentation included measurement of stagnation pressure and temperature with reported uncertainties of 7.0 kPa and 0.1 °C, respectively, mass flux with uncertainty of 200 kg/m 2-s, and differential pressure measurements, the roughness of pipe was estimated to be 2.5E-06 m. As seen in Figures 13-4-la and 134-lb, the range of stagnation pressure tested was from 150 to 370 kPa with subcooling from 0 to 7°C. All tests were conducted with demineralized and degassed water. oA4394-nonN438413.wpd:lb-4303 13-22

WESTINGHOUSE PROPRIETARY CLASS 2 Copu of PesrTempera of UDte Conditon in Critcal Flaw Test Matrin end ROSL X Break Test ad LOFT 1.-5 26X Small Break Test O OATA 2 0 O Ardo 6. Aeheteo

                                                .ROSA         3    0     O SB-Ct-05 (60)

LorT 2 0 a tOTT L3-5 (2.5X) TSAT(P) 0 0 0 ISAT GSO e I!0 Figure 134-la Upstream Condition in Ardron-Ackerman Quality/Pressure of Upstream Conditions in Critical Flow Test Maxtrix Quality=(V-VF)/VFG O D DATA 3 0 0 Criti ow ovie riew rrc 0- 13 013

                                                -    03               CP 0         0 9--am i EC                         0 0&

a 0 5 Pr essIs) Figure 134-lb Upstream Condition in Ardron-Ackerman o:\4384-non\4384-13.wpd:Ib-4303 13-23

I WESTINGHOUSE PROPRIETARY CLASS 2 Boivin Boivin conducted critical flow experiments by discharging water through long, horizontal nozzles. Three nozzles were tested. Each nozzle had a rounded inlet, a long cylindrical smooth pipe, and a diffuser having a small expanding angle. In the three cases, the LID ratio is greater than 30 to minimize 2-D effects. The first nozzle had a pipe diameter of 0.012 m, 0.45 m long with a diffuser angle of 11 degrees. The second nozzle had a pipe diameter of 0.030 m, 1.6 m long with a 7 degree diffuser. The diameter of the third nozzle was 0.050 m, 1.7 m long with a diffuser of 7.7 degree. Measurements reported include inlet (stagnation) pressure and temperature, mass flux, and throat pressure. No measurement uncertainties were reported. Stagnation pressure conditions ranged from 1960 to 10100 KPa with inlet water somewhat subcooled. The upstream conditions in Pressure/Temperature and Pressure/Quality planes, along with (P, T) trajectories observed in LOFT and ROSA small break tests, are shown in Figures 13-4-2a and 13-4-2b. Comparso of Pressre/?em tr of %ntream Comin fa Crcal law Test Mab and floSL 5X Break Test and LO I3-5 2.5 Sm Break Test n ODATA 2 0 0 SelvIm I- 0 RDSA 3 0 0 55-CL-06 (5) tOFT 2 0 0 LOFT 13-5 (2.5T)

                                  -      TSAT(p)    0     0    0 TSAT Cf IaM Figure 13-4-2a Upstream Conditions in Boivin o:\4384-non\4384-13.wpd:1b-4303                            13-24

WESTINGHOUSE PROPRIETARY CLASS 2 Quality/Pressure of Upstream Conditions in Critical Flow Test Maxtrix Quality=(V-VF)/VFG a O DATA 3 a O Crttlel Flow Dt D. v _.. OLA a I a a L a D C> 1:-?- a

                               .2XH 0
                                    -ww-w   4.

I I I I... .. ... ... 46 46 a6 icb d* t4 t 0 Pm (oi) Figure 134-2b Upstream Condition in Boivin Fincke and Collins Fincke and Collins performed critical flow experiments by flowing subcooled water through a loop and test section. Mass flow rate was controlled by a flow control valve upstream of the test section and back pressure was controlled by a valve downstream of the test section. The test section consisted of a 1.8 m long, 0.0444 m diameter Lexan cylindrical tube followed by a convergent-divergent Lexan nozzle with a minimum diameter of 0.01828 m. Degassed water was used for all experiments. Instrumentation included upstream temperature (reported uncertainty of 0.1 C), volumetric flowrate (uncertainty of 0.1 uls), pressure just upstrearn of the nozzle (no uncertainty given), and differential pressure measurements along the nozzle (uncertainty ranging from 0.5 to 2.5 kPa). The differential pressure measurements were used to determine the throat pressure that is included in this data base. The upstream pressure ranged from 90 to 300 kPa, inlet temperatures were 50 to 40°C subcooled. The upstream conditions in Pressure/Temperature and Pressure/Quality planes, along with (P, T) trajectories from LOFT and ROSA small break tests, are shown in Figures 13-4-3a and 13-4-3b. o:\4384-non\4384-13.wpd:Ib-4303 13-25

WESTINGHOUSE PROPRIETARY CLASS 2 Comparso of P e="/tFMP'atwe of Upr Codtn Critica1 71w Test Katrix and ROSM -5 Break Tet and LOFT L9-5 26 }maD Break Test 'Li O ODATA 2 0 0 Ttaek. god Col l1g ItOSA 3 O O SS-CL-Db (OX)

                                    -         LOrT          2          a       0 LOFT L-5       (2.5)
                                    -         TSAT(P)       D          0       0 TSAI 0.

E I-Figure 134-3a Upstream Condition in Fincke-Collins Quality/Pressure of Upstream Conditions in Critical Flow Test Maxtrix Quality=(V-VF)/VFG Ct ODATA 3 D 0 Critiel Flow Oila 1E-00 D O Azl  :~~~a 0~~~~~ n I--4 a a I0 - U,-4a 0 C6

                                                                                             %lao OD 11.11.111111....                    ..    ....

I 1 t b z Presm" S) Figure 13-4-3b Upstream Condition in Fincke-Collins o:\4384-6on\4384-13.wpd:l1b4303 13-26

WESTINGHOUSE PROPRIETARY CLASS 2 Jeandey Jeandey performed critical flow experiments by flowing subcooled, demineralized and degassed water through a vertical test section. The test section consisted of a smoothly convergent entrance followed by a straight cylindrical pipe 0.02013 m in diameter followed by a diverging section with a divergent angle of 7 degrees. Flow was vertically upward for all the experiments. Stagnation conditions ranged from pressures of 900 to 12000 kPa and temperatures of 148.5 to 324.6 C. The resulting critical mass fluxes ranged from 14500 to 62000 kglm-s. The throat pressure was measured along with many other pressures along the test section. In addition, for 21 of the experiments, axial and radial void fraction profiles were obtained using an X-ray densitometer. The upstream conditions in Pressure/Temperature and Pressure/Quality planes, along with (P, T) trajectories observed in LOFT and ROSA small break tests, are shown in Figures 13-4-4a and 13-4-4b. Compeiso of Preesure/ p mm of Upstrem Cvn&tos tn (rteg F Test NarI d ROS1Z Bre Test and Itl I3-5 26% Sml Broak Tet O DATA 2 0 0 Jad.y _ .OSA 3 0 0 -CL-06 (6X) LOFT 2 0 0 LOFT 3-5 (2.51)

                                 -       TSAT(P)      0     0    0 TSAT Figure 13-4-4a Upstream Condition in Jeandey o.\4384-non\4384-13.wpd:lb-4303                             13-27

WESTINGHOUSE PROPRIETARY CLASS 2 Quality/Pressure of Upstream Conditions in Critical flow Test Maxtrix Quality= (V-VF)/VFG a ODATA 3 0 Ctiel Fto Dts 2-I, 0* C-2E02 I~4*~~8 0 D D IZ0%- (I) a 0O 0 a

                                  -IE-01 SI         dO th        1ib mt        30 P=     (ps;a)

Figure 13-4-4b Upstream Condition in Jeandey Neusen Neusen performed experiments to determine design criteria for convergent-divergent nozzles. Critical flow occurred during these experiments, and the data are included in this data base. Neusen flowed saturated water through two convergent-divergent nozzles with minimum diameters of 0.0064 and 0.011 m. Reported stagnation conditions ranged from pressures of 840 to 5540 kPa and qualities of 0.0028 and 0.228. Stagnation conditions for these experiments were determined by measuring subcooled temperature and pressure upstream of a throttling valve. The throttling process was assumed to be isentropic, and pressure was measured downstream of the throttling valve (reported uncertainty of 1%). Reported uncertainties for mass flux and calculated enthalpy were less than 2.5% and 0.5%, respectively. The upstream conditions in Pressure/Temperature and Pressure/Quality planes, along with (P, T) trajectories observed in LOFT and ROSA small breaks, are shown in Figures 13-4-5a and 13-4-5b. o:\4384-non\434-13.wpd:1b4303 13-28

WESTINGHOUSE PROPRIETARY CLASS 2 Comparin -atr of UPStram CoriAOns in Crl Flow Test Matim ard RM&6%Brek Tet' and LI I-5 2.6% Small }rek Test O ODATA 2 0 O N.....

                                   -0SA                         3         0       0 R-CL-O5 (Z)

LOrT 2 0 0 LOFT L3-5 (2.5%) TSAT(M) 0 0 0 TSAT

&5w, e

t. 0-19 40DI Figure 13-4-5a Upstream Condition in Neusen Quality/Pressure of Upstream Conditions in Critical Flow Test Maxtrix Quality=(V-VF)/VFG 0 DATA 3 a 0 Ctleel Fig. .te

                                        .40   I
                                          .2-0 9,

a 5.1 I a 13 o 19 4 a O D a iE o 0a a

                                     .E-Cl-4 a          a           a
                                                . . f .   .   .  .   . .i                  . . .  . .

i aD Press *) Figure 134-Sb Upstream Condition in Neusen o:\4384-non\4384-13.wpd:1b4303 13-29

I WESTINGHOUSE PROPRIETARY CLASS 2 Reocreux Reocreux performed critical flow experiments by flowing subcooled degassed, demineralized water upwards through a vertical test section. The test section consisted of a straight, cylindrical section 2.335 m long and 0.020 m in diameter, followed by a divergent section 0.327 m long. Stagnation pressures ranged from 212 to 340 kPa, and stagnation temperatures ranged from 115.9 to 121.8 C. Pressure were measured along the test section at many locations, most concentrated near the choking point (at the entrance to the divergent section). The critical or throat pressures were determined from these measurements. In addition, the void fraction at the choking point was measured for most of the tests using X-ray attenuation method. The upstream conditions in Pressure/Temperature and Pressure/Quality planes, along with (P, T) trajectories observed in LOFT and ROSA small break tests, are shown in Figures 134-6a and 13-4-6b. Comparisn of PnreftemperattUr oUpstrea condtu= hL Critica Flov Test Ma_i and ROSX Break Test and LOFr I3-5 2% Rma Break Tes D VATA 2 0 0 R..cr.x ROSA 0 0 D-cL-05 (5X) LOT 2 O 0 LOFT L3-5 (2.5X) TSAT(P) a 0 O TSAT Figure 13(4-6a Upstream Condition in Reocreux o:\4384-non\4384-13.wpd:1b-4303 13-30

WESTINGHOUSE PROPRIETARY CLASS 2 Comparison of Presurve/Tempeztre of Upstrnam Conftons in Citica no Test Matr and RS5X Break Test and LOFT s-5 26% s m Brek Test O ODATA 2 O 0 Ssy.h.., i- 1 ROSA 3 O O S-CLOS (6X) LOrT 2 O0 LOrT 3-5 (2.5X)

                                -       TSAT(P)    0     a    0 TSAT Figure 13-4-6b Upstream Condition in Reocreux Seynhaeve Seynhaeve performed critical flow experiments by flowing subcooled, demineralized water upwards in vertical test sections. Two test sections were employed. Each section consisted of a straight, cylindrical pipe 0.0125 m in diameter followed by a divergent section. One section had the straight pipe 0.306 m long, and the other 0.221 m long. Stagnation conditions for these experiments range from 280 to 1015 kPa in pressure and 1 to 166.8 C in temperature. Critical pressure was measured near the choking plane. Measurement uncertainties are not known.

The upstream conditions in Pressure/Temperature and Pressure/Quality planes, along with (P, T) trajectories observed in LOFT and ROSA small break tests, are shown in Figures 13-4-7a and 13-4-7b. o:\4384-non\4384.13.wpd:b-4303 13-31

WESTINGHOUSE PROPRIETARY CLASS 2 Comparlo of Presure/Tempratmr. of Uam Cm&ticw im tUcal Flav Test Mahtz and RS e Test and LFE I:-5 24% Sml Break Test O O DATA 2 0 0 S.y.h.. ROSA 3 0 O sI-eL-O5 (6X) L OFT 2 a I LOFT t 3-S 2.SX) TSAT(P) 0 0 0 TSAT S: 5m - 2

                                =3
                               'E A   400-050       id:o         1*           2E0        2 a Presse b6ni)

Figure 13-4-7a Upstream Condition in Seynhaeve Quality/Pressure of Upstream Conditions in Critical Flow Test Maxtrix Quality=(V-VF)/VFG 0 1DATA 3 0 0 Critcel Flow DOt. D,. com1

                                  & -AE09*                a Ba             %E          a
                                                                                     ~00 I-*E04 E                        aU aO
                                       -IE-M 4 4i)    lb        ad      i       tD       AO    I0 Presstr (e)

Figure 13-4-7b Upstream Condition in Seynhaeve o:\4384-nonN4384-13.wpd:1b-4303 13-32

WESTINGHOUSE PROPRIETARY CLASS 2 Sozzi and Sutherland Sozzi and Sutherland conducted a series of critical flow experiments with subcooled and low quality water. The water for each experiment was demineralized and degassed. Water from a large vessel was blown-down through test nozzles. Data from 21 different nozzle shapes and configurations have been taken with more than 650 individual data points. Stagnation pressure ranged from 3000 to 7000 kPa, and stagnation qualities ranged from approximately -0.04 to 0.007 (based on the specific volume) The upstream conditions in Pressure/ITemperature and Pressure/Quality planes, along with (P, T) trajectories observed in LOFT and ROSA small break tests, are shown in Figures 13-4-8a and 13-4-8b. Comperlsn of Presre/Tempeate of Upstrmm Condit in CriMl Fl Test Matrix and RMOSAX B Drkest

                                           <<d IT I3-5 2ZX Small Dre Test D      '.DATA     2    0     0  S.1        th i.d Id
                                 -      ROSA       a    O     0 SB-CL-OS    )(5 LOFT       2    0     0 LOFT L.-5 (2.6Z)
                                 -      TSAT(P)    a    0     D TSAT Figure 13-4-8a Upstream Condition in Sozzi-Sutherland o:'4384-non\4384-13.wpd:lb-4303                        13-33

WESTINGHOUSE PROPRIETARY CLASS 2 Quality/Pressure of Upstream Conditions in Critical Flow Test Maxtrix Quality=(V-VF)/VFG 0 ODATA 3 0 0 C,ltelI Flow Date A&MS. U U 0 im06q _ 0 O la S 0 to

                                        . 0          0 a

0 AS 5b 0 7es 8 ) d6 1tied PM (*) Figure 13-4-8b Upstream Condition in Sozzi-Sutherland Marviken Tests 6, 7, 23 and 24 Marviken tests provide very large diameter downflow data typically considered full scale. The Marviken facility was used for full-scale critical flow tests between mid-1977 and December 1979. During this time, 27 tests were conducted by a downward discharge of water and steam mixtures from a full-sized reactor vessel through a large diameter vertical discharge pipe that supplied the flow to a test nozzle. There were 9 nozzles tested; all had rounded entrances followed by a noninal 20, 30 and 50 cm constant diameter straight section. Table 134-3 below shows the characteristic dimensions for the tests. As seen in the table, tests selected for the model evaluation are datasets taken with two of the shortest nozzles in the Marviken test series. o:\4384-non\4384-13.wpd:Ib-4303 13-34

WESTINGHOUSE PROPRIETARY CLASS 2 Table 13-4-3 Marviken Test Nozzles Nozzle Number Diameter (mm) Length (mm) Used in Tests 1 200 590 13,14 2 300 300 6,7 3 300 510 25,26 4 300 895 1,2,12 5 300 1110 17,18,19 6 500 166 23,24 7 500 730 20,21,22,27 8 500 1809 15,16 I9 1 509 1589 3,4,5,,8,9,10, 11 The discharge pipe that connects the vessel to the nozzle is 6283 mm long and is geometrically complex. It is made up of several pieces: nozzle, permanently attached to the vessel with a 752 mm diameter, a 1980 mm long drift tube of the same diameter, a 1778 mm long global valve with a 780 mm diameter and a 1000 nun long with 752 mm diameter section to which the nozzle is attached. Besides these there were two 120 mm long instrument rings inserted on either end of the 1980 mm'drift tube. It is quite clear that with this degree of geometric complexity, the question of establishing a consistent set of complete inlet conditions is not simple. For this study, only the nozzle is modelled by the critical flow model. Thus the inlet condition to the nozzle was taken from 004M109 for pressure (0.7 m upstream) and 003M404 for temperature (2.8 m upstream). Probable error Pressure - 7 kPa, Temperature - 0.6°C. The upstream conditions in Pressure/Temperature and Pressure/Quality planes, along with (P, T) trajectories observed in LOFT and ROSA small break tests, are shown in Figures 13-4-9a and 13-4-9b. o.W384-nonN4384-13.wpd:1b-4303 13-35

WESTINGHOUSE PROPRIETARY CLASS 2 Comparbson of Pre fTempertre of Upstream Conditions Jn Critical Flo Test Matri and RS05%Bregk Test and LOFT L-5 2.5X Sal Break Test n ODATA 2 0 aryl.

                                        -    RtOSA              3      O      0 SD-CL-05 (61)

OFT 2 0 0 LOF7 L-5 (2.51) TSAT(P) 0 0 O TsAT 7W so_ > 50 le I&* 200 26 Pressre (psa) Figure 134-9a Upstream Condition in Marviken Tests 6, 7, 23 and 24 Quality/Pressure of Upstream Conditions in Critical Flow Test Maxtrix Quality=(V-VF)/VFG a O DATA S O O CtS r D ZE- - 0-~~~~~~~ tEk-M41313 U,~~~~~~~~~~fr ° ,Le Figure UpstreamOCondition in Marviken °3-4-9b Tests 6,7,23 and 24 13 _ , _ , , . Pen (ia) Figure 13^4-9b Upstream Condition in Marviken Tests 6, 7, 23 and 24 . o:\4384-non4384-13.wpd:1b-4303 13-36

WESTINGHOUSE PROPRIETARY CLASS 2 Amos and Schrock Amos and Schrock's break flow data cover a wide range of pressure from 4000 to 15500 kPa, and subcooling from 0 to 60°C which is suited for evaluating a performance of the break model for small break LOCA analyses. The configuration of the break is thin rectangular slit with the nominal width of 0.381 and 0.254 mm. These set of tests are two of larger slit size of the three of their experiments. Although the break flow area is rectangular and small (equivalent hydraulic diameter = 0.748 and 0.464 mm), the data is valuable since the phenomena which governs the critical condition appeared to be the same for breaks of all sizes. This may be why the D flow model is sufficiently accurate to describe the break flows. The upstream conditions in Pressure/Temperature and Pressure/Quality planes, along with (P, T) trajectories observed in LOFT and ROSA small break tests, are shown in Figures 13-4-10a and 134-lOb. Comparo of Pmre/remparatum of tfteeam U CodtosW hn Crid=a Flo Test Ias and osA freak test aud IF .S 26X San Bra Test ODATA 3 2 0 0 Am.. *md S.hr

                                 -       It0SA       5    O    0 SB-CL-D6 (S)

LOFT 2 0 0 LOFT L3-5 (2.5) Ts ATP) 0 0 0 SAt S Figure 134-10a Upstream Condition in Amos-Schrock o:\4384-non\4384-13.wpd:1b.4303 13-37

I WESTINGHOUSE PROPRIETARY CLASS 2 Quality/Pressure of Upstream Conditions in Critical Flow Test Maxtrix Quality= (V-VF)/VFG O O DATA 3 a 0 Critical i. Dt

                                    .Cl4- .r-I               8                    0
                                                       °       a a       a o       -

g B3 C oIt: 0 0 0 0

                                                                                  *0.

een, -*ts-0 1cho t _ I Pre ~*) Figure 13-4-lOb Upstream Condition in Amos-Schrock Anderson and Benedetti (TPFL) Anderson and Benedetti conducted critical flow tests at Two Phase Flow Loop (TPFL) located in INEL, for purpose of investigating the entrainment at the break off the stratified upstream flow under saturated condition. Two phase mixture of known phasic mass flow rate flowed through a branch line pipe of 1.63 m long, 34 mm diameter attached to a simulated cold leg pipe, to the nozzle which is 54 mm long and has a diameter of 16.2 mm. The pressure just upstream of the rounded entrance nozzle as well as the void fraction was measured by a gamma attenuation method. Their experiments are well instrumented critical flow tests with saturated upstream conditions at 900, 640 and 500 psia. The flow quality in the tests were varied from 0 to 1 at all three pressures. The upstream conditions in Pressure/Temperature and Pressure/Quality planes, along with (P, T) trajectories observed in LOFT and ROSA small break tests, are shown in Figures 134-1 la and 13-4-1 lb. o:\4384-non\4384-13.wpd:1b-4303 13-38

WESTINGHOUSE PROPRIETARY CLASS 2 Comperison of Presue/Teratmre of Upat Condhim in Critical Flo Test Matrix and R1O&5% Break Test and LOFT 1-5 26Z Rl Break Test Oi O DATA 2 0 0 TPFL

                                        -__"ROSA           3      0      0    S-CL-06 (5X)

LOFT 2 0 0 LOFT 3-6 (2.1Z) TSAT(r) D 0 0 TSAT 7 amn E C4W Figure 13-4-la Upstream Condition in TPFL Quality/Pressure of Upstream Conditions in Critical Flow Test Maxtrix Quality= (V-VF)/VFG O0 DATA 3 0 0 Crtilcel Flow Dlet u Li u 0 a na 0 o X- lo~~~~

A c.

                                .,                  oi                                       B O                                        D3 4                                                          0 Prsu     (Psa)

Figure 13-4-llb Upstream Condition in TPFL o:\4384-non\4384-13.wpd:1b-4303 13-39

WESTINGHOUSE PROPRIETARY CLASS 2 Overall The test matrix selected covers from 13 psia to 2300 psia, and quality of -0.039 to 1.0. The coverage of upstream condition is graphically shown in Figures 13-4-12a and Figure 134-12b below. Cowparbon of PressreAemperatmre of Upstream Cond an in CriUCkl flon Test Matrix aid ROS X Break Test and LOFr -5 2.5% Small Break Test D OOATA 2 0 0 Criticul Fw Dla

                            -       ROSA              3       0       0 SB-CL-05 (5X)

LOFT 2 0 0 LOFT L-5 (2.5X) TSAT(P) a 0 0 TSAT 7OD eoo _Vf 1 1Z*~~~~~~~~~~~~~1 5~~~~~~~O~I 2m. tmIO I, I* I . . . . ,t ' ,40 ' . . _s Pe, (psio) .- - - Figure 13-4-12a Upstream Condition in Test Matrix o.\4384-non\4384-13.wpd:1b-4303 13-40

WESTINGHOUSE PROPRIETARY CLASS 2 Quality/Pressure of Upstream Conditions in Critical Flow Test Maxtrix Qua&ity=(V-VF)/VFG 0 DATA 3 0 0 Crtical Flow Dta 1 Figure 134-12b Upstream Condition in Test Matrix o.%4384-non\4384-13.wpd:lb-4303 13-41

WESTINGHOUSE PROPRIETARY CLASS 2 13-4-3 Assessment Results A total of 1427 data points from 40 nozzle geometries were used for the determination of bias and uncertainty associated with the critical flow model prediction used in WCOBRA[IRAC-SB. The following results were obtained through the comparison to data. 134-3-1 Bias and Uncertainty A valid range of the bias and uncertainty estimate given here is based on selected experimental data. A comparison was made for 0 < L < 2335 mm, and 0.418 < DH < 500 mm. Overall (-0.039 < Quality < 1.0) Predictions for all selected data are shown in Table 134.4. The mean error v Gcaic-G.) 1was found to be -8.2% and the standard deviation N a(£) = N-1 was found to be 19.8%. Subcooled Liquid Region (-0.039 < Quality < 0) Bias=4.3% Standard Deviation = 14.9% Saturated Two Phase Region (0 < Quality < 1.0) Bias = -13.40% Standard Deviation = 23.7% o:\4384-non\4384-13.wpd:1b4303 13-42

WESTINGHOUSE PROPRIETARY CLASS 2 Table 13-4-4 Critical Flow Data Comparison for WCOBRAITRAC Critical Flow Model Mean Error (%) Data (Gcalc G a Set L D No. Reference (nun) (nm) cosO N-Data Gmas a() (%) I Andron, K. H. &Ackerman, M.C. (1978) 1015 26.3 0 32 1.1 12.7 [10] 2 Boivin (1979) [12] 500 12 0 10 9.2 2.8 3 Boivin (1979) 12] 1600 30 0 5 0.6 23.5 4 Boivin (1979) [12] 1700 50 0 6 -13.8 7.3 5 Fincke &Colins (1981) [15] 13 44 0 92 5.3 3.4 6 Jeandey et al. (1981) [20] 463 20 - 15 -1.0 9.2 7 Jeandeyetal.(1981)[20] 463 20 1 73 -10.8 11.6 8 Neusen (1962) [22] 0 11 0 25 16.3 18.8 9 Neusen (1962) [221 0 6 0 12 0.4 10.5 10 Reocreux (1974) [24] 2335 20 1 28 -2.6 6.5 11 Seybhaeve (1980) [25] 306 13 1 26 -11.0 1.9 12 Seybhaeve(1980)[25] 306 13 1 31 -9.4 4.2 13 Sozzi &Sutherland (1975) [27] 45 12.7 0 128 -34.6 11.8 14 Sozi &Sutherland (1975) [27] 45 12.7 0 13 -45.8 6.5 15 Sozzi &Sutherland (1975) [27] 57 12.7 0 47 -37.1 6.6 16 Sozzi &Suthedrand (1975) [27] 362 12.7 0 19 -5.1 9.9 17 Sozzi &Sutherland (1975) [27] 83 12.7 0 17 -23.4 12.5 18 Sozzi &Sutherland (1975) [27] 553 12.7 0 13 -2.9 7.7 19 Sozzi &Sutherland (1975) [27] 108 12.7 0 23 -2.4 6.4 20 Sozzi &Sutherland (1975) [27] 679 12.7 0 96 6.1 14.6 21 Sozzi &Sutherland (1975) [27] 159 12.7 0 15 -16.6 9.5 22 Sozzi &Sutherland (1975) [27] 1823 12.7 0 81 7.2 14.1 23 Sozzi & Sutherland (1975) [27] 235 12.7 0 12 -12.9 6.0 24 Sozzi & Sutherland (1975) [27] 273 12.7 0 22 -14.4 7.3 25 Sozzi & Sutherland (1975) [27] 5 12.7 0 58 -25.7 13.5 26 Sozzi &Sutherland (1975) [27] 322 12.7 0 24 -4.4 6.4 27 Sozzi & Sutherland (1975) [27] 513 12.7 0 24 -4.3 8.0 28 Sozzi & Sutherland (1975) [27] 640 12.7 0 17 -2.3 7.7 29 Sozzi & Sutherland (1975) [27] 195 12.7 0 23 -14.0 5.2 30 Sozzi & Sutherland (1975) [27] 45 19 0 23 -27.8 6.5 31 Sozzi & Sutherland (1975) [27] 732 54 0 4 -17.7 2.5 32 Sozzi & Sutierland (1975) [27] 696 76 0 3 -8.8 3.3 33 Sozzi & Sutheland (1975) 127] 63 28 0 5 -23.9 8.3 34 Marviken Test 6 (1982) [6] 300 300 -1 85 -10.3 9.0 35 MarvikenTest7 (1982) [6] 300 300 -1 84 -16.5 8.1 36 MarvikenTest23 (1982) [6] 150 500 -1 44 -2.1 11.1 o:4384-non\4384-13.wpd:1b4303 13-43

WESTINGHOUSE PROPRIETARY CLASS 2 Table 134-4 (Cont'd) Critical Flow Data Comparison for WCOBRAITRAC Critical Flow Model .XS Mean Error E (%) Data Gca]c Gn ) S'et L D fGCC Geas CFE No. Reference (mn) (nun) cosO N-Data (E) (%) 37 Marviken Test 24 (1982) 6] 150 500 .1 39 -19.2 14.3 38 Amos&Schrock (1983) [7 63.5 0.748 -1 18 0.7 7.2 39 Arnos &Schrock (1983) [7] 63.5 0.464 -1 26 -0.3 9.5 40 Anderson Benedeti (1985) [8] 31.9 16.2 0 109 15.2 26.9 TOTAL 1427 -8.2 19.8 Figure 134-13 below shows the comparison of all points in the test matrix with +10% lines above and below the 450 line. WCOBRA/TRAC Model Prediction vs. ALL_DATA Data Mean Error is -8.2% Standard Deviation is 19.8% n ElWCT 4 0 D PREDICTION 1 E

                                      ~~~~

10'~~~~~~~~~1 a~~~~~~~~~~~~~q C,4 Measured Mass Flow Flux (kg/m2-s) Figure 134-13 Comparison of Predicted and Measured Critical Flows o:\4384-non\4384-13.wpd:1b-4303 13-44

WESTINGHOUSE PROPRIETARY CLASS 2 13-4-3-2 Model Prediction Trend with Respect to Pressure In this section, a possible model trend with respect to the upstream pressure is examined. Figure 13-4-14 below shows the error vs. pressure of all data points. The figure does not show global trend relative to the upstream pressure, although it does show that there is a larger spread in the lower pressure points (p < 1000 psia). Model Prediction Trend w. r. . Pressure O OERROR S 0 0 Error (X) 60.' 40 - 0 20 - _ I0 0O 0 I 5I P iI 9D I& 2kO Pressure (sia) Figure 13-4-14 Prediction Trend in Pressure Variation 13-4-3-3 Model Prediction Trend with Respect to Quality In this section, a possible model trend with respect to the upstream quality is examined. Figure 13-4-15 below shows the error vs. quality of all data points. The figure shows global trend relative to the upstream quality. The model tends to underpredict the critical mass flux for saurated liquid, X =0, and overpredict in the two-phase region. At or near single phase vapor region, the model's overprediction becomes substantially smaller. o:\4384-nonW4384-13.wpd:1 b4303 13-45

I WESTINGHOUSE PROPRIETARY CLASS 2 Model Prediction Trend w. r. t. Quality oC OERROR S 0 0 Error (%) 0 0 40-0 a0 El0 a 0 1m

                                                                                '33 Eb El ImD~~~~~~~E 0

a-o OB El 0

                                                      -0
                          -  W I           .                    . .  .

26 a a0:4 Stognotion Quoalty (V-V)/VFG  :>L: Figure 13-4-15a Prediction Trend in Quality Variation A significant uncertainty is seen at or near the saturation. However, in the subcooled region no bias is seen whereas in the low quality two-phase region (0 < X < 0.02), a bias of about -20% is observed.

                                                                                                ,LX, o:\4384-non\4384-13.wpd:1b-4303                               13-46

WESTINGHOUSE PROPRIETARY CLASS 2 Model Prediction Trend w. r. t. Quality O O ERROR S 0 0 Error (X) J T 40- C b0~~~ 20-0 a 0 0 0-0 ~~~~~~1 7~~~~ . . . .. *.* . . Fw 0 w4-Ol -JE1 -lE01 6 JE!O1 _E 4O StagnatUon Qualy (V-VF)/FG Figure 134-15b Prediction Trend in Quality Variation o:.4384-non\4384-13.wpd:1b-4303 13-47

I WESTINGHOUSE PROPRIETARY CLASS 2 13-4-34 Model Prediction Trend with Respect to Channel Length In this section, a possible model trend with respect to the channel length is examined. Figures 134-16a and 134-16b below show the error vs. channel length of all data points. The figures do not show global trend relative to the channel length, although they do show that there is larger spread in the short length nozzle predictions. Model Prediction Trend w. r. t. Channel Length O O ERROR 6 0 0 Error () 40 0 20 Channel Length (m) Figure 134-16a Prediction Trend in Channel Length Variation in Linear Scale o:\4384-non\4384-13.wpd:1b-4303 13-48

WESTINGHOUSE PROPRIETARY CLASS 2 Model Prediction lfzend w. r. t Channel Length 0 0 ERROR e 0 0 Error () I 40-0~~~ 1~~~~~~ I aa 0-I HJ C3i

                             -2D -                                           0 I                         ~~~~0 0

j~El o 0 ~06 1D 10 1 10 10 10 Channel Length (in) A . Figure 13-4-16b Prediction Trend in Channel Length Variation in Log Scale o:N4384-non\4384-13.wpd:1b-4303 13-49

I WESTINGHOUSE PROPRIETARY CLASS 2 134-3-5 Model Prediction Trend with Respect to Hydraulic Diameter In this section, a possible model trend with respect to the hydraulic diameter is examined. Figure 134-17a below shows the error vs. hydraulic diameter of all data points. The figure does not show global trend relative to the hydraulic diameter variations. Model Prediction Trend w. r. t. Channel Hydraulic Diameter 0 DERROR 5 0 0 Error () W.

                                       -0
                                       -0 40-   I3 I

I 20-Ii 0-fr0 C[O 0 Ia 0 B

                                                                                                    .,/
                                                                   -m         .  .  .     . I   I   .  .  . .     . .    . . . I .

i 0:3 05 Channel Hydrouric Diameter (m) Figure 13-4-17a Prediction Trend in Channel Diameter in Linear Scale There is a slight tendency to underpredict for large diameter nozzles, which can be seen in Figure 13-4-17b. o:\4384-non\43&4-13.wpd:1b-4303 13-50

WESTINGHOUSE PROPRIETARY CLASS 2 Model Prediction Trend w. r. t. Channel Hydraulic Diameter 0 OERROR e 0 a Error (X) EO . 09 0 40-0 a a 20- 3 0- 0 C 0 0 I&

                             -20  -        B
                                           .0                       -               -

1 0~~~~~~ 03 0 8

                                                    .. I .... .                  0    I0
                             -W    I  s s X w w wwr    l    w  r r s s el         w   r  s  E XXwz   E l  l l l l wl s v f w we s I             I         -~~~~~~~-

10 10 l 10 it - - 1D 10 Channel Hydraurc Diometer (m) Figure 13-4-17b Prediction Trend in Channel Diameter in Log Scale 13-4-3-6 Model Prediction Trend with Respect to IJD Figures 13-14-18a and 13-14-1Sb show the relative errors vs. I/D of the break path in linear and log scale. There is no global trend observed relative to LID variations. o:"4384-non\4384-13.wpd:1b-4303 13-51

I WESTINGHOUSE PROPRIETARY CLASS 2 Model Prediction Trend w. r. t. Channel L/D 0 O ERROR e 0 0 Error () Chonnel L/D Figure 13-4-18a Prediction Trend in Channel L/D Variation - Linear Scale o:\4384-non\4384-13.wpd: lb-4303 13-52

WESTINGHOUSE PROPRIETARY CLASS 2 Model Prediction Trend w. r. t. Channel L/D 0 0 ERROR 6 0 a Error (X) a0 0 0 40-E 0 0 20 - 0~~ p Ela 0-6 La D; 1 1':l

                          -2   -

O~~~~1 i 10 to 1 1a la Channel (iD I, Figure 13-4-18b Prediction Trend in Channel L/D Variation - Log Scale o:%4384-nonW4384-13.wpd:1b4303 13-:53

I WESTINGHOUSE PROPRIETARY CLASS 2 13-4-3-7 Influence of Upstream Void Fraction The model prediction's sensitivity to the initial void fraction assumed in the critical flow module was investigated in this section. The minimum void fraction is set to ALMIN. The code [

    ]aC model prediction. The result is shown in Table 13-4-5 below.

Table 134-5 Prediction Sensitivity to the Initial Void Fraction BiastStandard Deviation (%) ALMIN Subcooled Saturated Total L.OE-03 -4.6/14.9 -14.0122.2 -8.7/19.0 1.0E-04 -4.3114.9 -14.0/22.2 -8.5/19.0 1.OE-08 4.0/15.0 -14.0/22.2 -8.4/19.1 1.OE-12 4.0/15.0 -14.0/22.2 -8.3/19.1

                   .OE-15             -2.8/18.7             -14.0/22.2             -7.6/21.0 0.0               -1.6122.3             -14.0/22.2             -7.0/23.0 It is interesting to note that for subcooled regions, if [
                                                            ] C   Judging from the ALIIN sensitivity results, the critical flow calculation is insensitive to the magnitude of residual void.

13-4-3-8 Influence of Two-Phase Multiplier The current model uses Levy's model. For a sensitivity study, Richardson's model was used. The model is expressed as: q0 = 1(lcr7)5 The results shown below indicated that the critical mass flow prediction was relatively insensitive to the choice of two-phase multiplier. For 1427 Data Points, Average Error = -8.59 %, STD = 19.04 % For 807 Subcooled Data Points, Average Error = -4.28 %, STD = 14.95 % For 620 Saturated Data Points, Average Error = -14.20 %, STD = 22.09 % o:\4394-non\4384-13.wpdlb-4303 13-54

WESTINGHOUSE PROPRIETARY CLASS 2 13-4-3-9 Influence of Mesh Size The model prediction's sensitivity to a number of axial nodes used in the critical flow module was investigated. The number of axial nodes, NMAX, is set [

                                    ]aC The result is shown in Table 134-6 below.

Table 13-4-6 Prediction Sensitivity to the Mesh Size Bias/Standard Deviation (%) NMAX Subcooled Saturated Total 21 -3.6/15.1 -13.5/22.1 -7.9/19.1 51 -4.1/15.1 -13.8122.2 -8.3/19.1 101 4.3/15.0 -14.0/22.2 -8.5/19.0 401 4.4/15.0 -14.0/22.2 -8.6/19.0 801 4.4/14.9 -14.1/22.2 -8.6/19.0 As seen in the table, the model prediction is relatively insensitive to mesh size. [ Iac 134-3-10 Influence of Friction Factor/Entrance Effect The entrance and friction factors were found to be very important for predicting the low pressure experiments such as those of Ardron and Ackerman (1978). This is the reason the reported friction factors were used for simulation of Ardron and Ackerman. o:\4384-non\4384-13.wpd:1b4303 13-55

I WESTINGHOUSE PROPRIETARY CLASS 2 134-3-11 Critical Flow Predictions for Individual Dataset The charts below show the comparison between the predicted and the measured data for individual datasets. A figure below each chart compares WCOBRA/TRACs predicted performance in relation to each data set. Ardron and Ackerman Run No. Pressure TesVerature Quality Predicted Gc Measured Gc Error L D (Gp-GM) /GM (Pa) (IK) lKg/m2-s) (Kg/m2-s) (in %) (mm) mm) 1 220500.0 391.75 -0.000005 13025.00 13300.00 -0.02 1015.0000 26.3000 2 216000.0 392.05 -0.000004 11910.00 10500.00 0.13 1015.0000 26.3000 3 207700.0 391.25 -0.000003 11207.00 10900.00 0.03 101S.0000 26.3000 4 205400.0 391.95 -0.000002 9797.90 9510.00 0.03 1015.0000 26.3000 5 203300.0 391.55 -0.000002 9869.60 9510.00 0.04 1015.0000 26.3000 6 199000.0 392.45 0.000000 7105.00 7080.00 0.00 1015.0000 26.3000 7 190000.0 391.65 0.000000 5359.10 5670.00 -0.05 1015.0000 26.3000 8 2640D0.0 399.85 -0.000003 10931.00 11800.00 -0.07 1015.0000 26.3000 9 255200.0 398.95 -0.000002 10492.00 10700.00 -0.02 1015.0000 26.3000 10 250000.0 398.85 -0.000001 9421.40 9300.00 0.01 1015.0000 26.3000 11 247300.0 399.65 0.000000 6719.20 6880.00 -0.02 1015.0000 26.3000 12 299000.0 405.05 -0.000001 9463.80 9980.00 -0.05 1015.0000 26.3000 13 299900.0 406.05 0.000000 7054.50 7740.00 -0.09 1015.0000 26.3000 14 203000.0 391.25 -0.000002 10220.00 9700.00 0.05 1015.0000 26.3000 15 354700.0 412.25 0.000000 5573.80 6760.00 -0.18 101S.0000 26.3000 16 206600.0 392.85 -0.000001 8684.50 9010.00 -0.04 101S.0000 26.3000 17 177000.0 385.15 -0.000004 11804.00 10800.00 0.09 1015.0000 26.3000 18 173000.0 384.35 -0.000004 11801.00 10900.00 0.08 1015.0000 26.3000 19 166000.0 383.95 -0.000003 10822.00 9160.00 0.18 1015.0000 26.3000 20 159000.0 384.75 -0.000001 8171.50 7740.00 0.06 1015.0000 26.3000 21 155400.0 385.05 0.000000 6480.50 6510.00 0.00 101S.0000 26.3000 22 154000.0 384.75 0.000000 6511.20 5900.00 0.10 1015.0000 26.3000 23 190200.0 384.55 -0.000007 14478.00 13800.00 0.05 1015.0000 26.3000 24 190600.0 384.85 -0.000007 14317.00 13600.00 0.05 1015.0000 26.3000 25 215100.0 391.75 -0.000004 12087.00 11600.00 0.04 1015.0000 26.3000 26 254400.0 399.55 -0.000001 9234.50 9610.00 -0.04 1015.0000 26.3000 27 306000.0 406.25 0.000000 8504.40 9630.00 -0.12 1015.0000 26.3000 28 365200.0 413.15 0.000000 6195.70 7730.00 -0.20 1015.0000 26.3000 29 360400.0 412.75 0.000000 5896.20 7790.00 -0.24 101S.0000 26.3000 30 191900.0 385.05 -0.000007 14363.00 13900.00 0.03 1015.0000 26.3000 31 220300.0 392.75 -0.000004 11898.00 11800.00 0.01 1015.0000 26.3000 32 366600.0 411.45 -0.000003 11530.00 7730.00 0.49 1O1S.0000 26.3000 For 32 Data Points, Average Error 1.09 , STD - 12.68 % For 26 Subcooled Data Points, Average Error 4.35 %.STD 11.18 For 6 Saturated Data Points, Average Error -13.06 %,STD - 8.75 % Minizmu quality is 0.0000. Maximum quality is 0.0000 o3.434-non\4384-13.wpd:1b-4303 13-56

WESTINGHOUSE PROPRIETARY CLASS 2 WCOBRA/TRAC Model Prediction vs. Ardron Ackerman Data Mean Eror is 1.1% Standard Devation is 12.7%. U WcT 4 a 0 PREDICTION en 12000

E 4E ooo.

                    £0
                         . 6000 Li.

(-). o C 400 Figure 13-4-19 Prediction Comparison with Ardron-Ackerman Data oA4384-non\4384-13.wpd:1b-4303 13-57

I WESTINGHOUSE PROPRIETARY CLASS 2 Boivin Run No. Pressure Teoperature Quality 1 Predicted Cc Measured Ge Error L (Gp-MI /GM (Pal (K) (Kg/=2-s) (Kg/I2-s) (in  %) (mmn)

...................................................................................................................                                                           (....

1 2300000.0 489.15 -0.000063 17738.00 16900.00 4.96 500.0000 50.0000 2 1960000.0 477.95 -0.000099 21103.00 19500.00 8.22 500.0000 50.0000 3 6140000.0 544.45 -0.000629 30785.00 28800.00 6.89 500.0000 50.0000 4 3610000.0 507.35 -0.000351 32061.00 28600.00 12.10 500.0000 50.0000 5 7170000.0 555.05 -0.000865 31850.00 30300.00 5.12 500.0000 50.0000 6 4050000.0 512.05 -0.000525 36140.00 32600.00 10.86 500.0000 50. 0000 7 5140000.0 529.25 -0.000644 35535.00 32700.00 8.67 500.0000 50.0000 8 7220000.0 552.55 -0.001313 37864.00 34200.00 10.71 500.0000 50.0000 9 3840000.0 505.95 -0.000566 38831.00 34700.00 11.90 500.0000 50.0000 10 3580000.0 500.85 -0.000550 38919.00 34700.00 12.16 500.0000 50.0000 11 6110000.0 546.15 -0.000414 23411.00 26600.00 -11.99 1730.0000 30.0000 12 10100000.0 567.15 -0.005431 49852.00 35100.00 42.03 1730.0000 30.0000 13 5400000.0 538.85 -0.000254 21124.00 24500.00 -13.78 1730.0000 30.0000 14 9320000.0 577.15 -0.000536 27822.00 31000.00 -10.25 1730.0000 30.0000 15 6280000.0 541.65 -0.001104 33478.00 34500.00 -2.96 1730.0000 30.0000 16 3740000.0 504.15 -0.000556 35526.00 37800.00 -6.02 1830.0000 50.0000 17 9040000.0 564.15 -0.003129 44200.00 50000.00 -11.60 1830.0000 50.0000 18 6730000.0 538.65 -0.002114 46664.00 50000.00 -6.67 1830.0000 50.0000 19 8460000.0 567.15 -0.001118 30307.00 38000.00 -20.24 1830.0000 50.0000 20 3240000.0 503.15 -0.000245 25573.00 29800.00 -14.18 1830.0000 50.0000 21 3050000.0 504.15 -0.000101 18061.00 23800.00 -24.11 1830.0000 50.0000 For 21 Data Points. Average Error - 0.56 , SD 15.04 For 21 Subcooled Data Points, Average Error - 0.56 , S . 15.04 WCOBRA/TRAC Model Prediction vs. Boivin Data Mean Error is .56% Standard Deviation is 15% a *W CT 4 0 0 PREDICTION U) EI 0

                                                          ,n ou 2COO U)

CQ-Figure 134-20 Prediction Comparison with Boivin Data o:A4384-non\4384-13.wpd: Ib-4303 13-58

WESTINGHOUSE PROPRIETARY CLASS 2 Fincke and Collins Run No0. Pressure Tesperature Quality Predicted c Measured Cc Error L D (Op-Cm) 1CM (Pa) (K) CKgIm2 -a) CKta2-s) (in %) (mm) (mm) 1 99020.0 342.21 -0.000012 11908.00 9936.70 29.84 79.720D 28.2800 2 100510.0 342.68 -0.000012 11973.00 11038.00 8.47 79.7200 18.2800 3 134970.0 343.05 -0.000021 14400.00 13516.00 6.54 79.7200 18.2800 4 108150.0 343.37 -0.000014 12505.00 13072.00 -4.34 79.7200 18.2800 5 173570.0 344.47 -0.000033 16633.00 15572.00 6.81 79.7200 18.2800 4 173960.0 344.66 -0.000033 16621.00 15581.00 6.67 79.7200 18.2800 7 174080.0 344.74 -0.000033 16626.00 15564.00 6.82 79.7200 18.2800 8 174830.0 345.20 -0.000033 16658.00 15570.00 6.99 79.7200 18.2800 9 174830.0 345.55 -0.000033 16626.00 15579.00 6.72 79.7200 18.2800 10 217810.0 348.89 -0.000046 18654.00 18063.00 3.27 79.7200 18.2800 11 219620.0 348.77 -0.000047 18733.00 18176.00 3.06 79.7200 18.2800 12 220400.0 348.82 -0.000047 18798.00 18173.00 3.44 79.7200 18.2800 13 221900.0 348.88 -0.000048 18861.00 18230.00 3.46 79.7200 18.2800 14 223110.0 348.89 -0.000048 1891.3.00 18287.00 3.42 79.7200 18.2800 15 277580.0 349.00 -0.000071 21424.00 20647.00 3.76 79.7200 18.2800 16 269220.0 349.01 -0.000067 21062.00 20414.00 3.17 79.7200 18.2800 17 269100.0 349.01 -0.000067 21057.00 20394.00 3.25 79.7200 18.2800 18 270270.0 348.90 -0.000068 21104.00 20460.00 3.15 79.7200 18.2800 19 270460.0 349.01 -0.000068 21111.00 20430.00 3.33 79.7200 18.2800 20 94950.0 359.05 -0.000005 9145.50 8139.80 12.36 79.7200 18.2800 21 104030.0 358.69 -0.000007 10108.00 9541.80 5.93 79.7200 18.2800 22 131530.0 358.33 -0.000013 12447.00 12013.00 3.61 79.7200 18.2800 23 185940.0 358.10 -0.000028 16057.00 15443.00 3.98 79.7200 18.2800 24 187470.0 358.09 -0.000029 16124.00 15572.00 3.54 79.7200 18.2800 25 188010.0 357.90 -0.000029 16178.00 15592.00 3.76 79.7200 18.2800 26 187750.0 357.86 -0.000029 16167.00 15592.00 3.69 79.7200 18.2800 27 188040.0 357.74 -0.000029 16210.00 15667.00 3.47 79.7200 18.2800 28 242600.0 358.57 -0.000047 19007.00 18311.00 3.80 79.7200 18.2800 29 234200.0 358.46 -0.000044 18610.00 17907.00 3.93 79.7200 18.2800 30 234320.0 358.45 -0.000044 18615.00 17871.00 4.16 79.7200 18.2800 31 233270.0 358.51 -0.000044 18540.00 17833.00 3.96 79.7200 18.2800 32 232190.0 358.46 -0.000044 18498.00 17760.00 4.16 79.7200 18.2800 33 287670.0 358.45 -0.000065 21104.00 20339.00 3.76 79.7200 18.2800 34 287770.0 358.46 -0.000065 21107.00 20266.00 4.15 79.7200 18.2800 35 286350.0 358.48 -0.000065 21055.00 20155.00 4.47 79.7200 18.2800 36 287550.0 358.50 -0.000065 21099.00 20228.00 4.31 79.7200 18.2800 37 289320.0 358.57 -0.000066 21163.00 20301.00 4.25 79.7200 18.2800 38 91150.0 363.83 -0.000003 7501.80 6460.70 16.11 79.7200 18.2800 39 106030.0 343.50 -0.000005 9242.10 8775.40 5.32 79.7200 18.2800 40 194810.0 363.47 -0.000027 15843.00 15348.00 3.23 79.7200 18.2800 41 192230.0 363.35 -0.000026 15710.00 15356.00 2.31 79.7200 18.2800 42 194580.0 363.24 -0.000027 15866.00 15256.00 4.00 79.7200 18.2800 43 193190.0 363.11 -0.000027 15811.00 15213.00 3.93 79.7200 18.2800 44 192570.0 363.12 -0.000027 15756.00 15204.00 3.63 79.7200 18.2800 45 243920.0 363.11 -0.000043 18569.00 17942.00 3.49 79.7200 18.2800 46 244530.0 363.00 -0.000044 18594.00 17959.00 3.54 79.7200 18.2800 47 216830.0 362.88 -0.000034 17163.00 18052.00 -4.92 79.7200 18.2800 48 247730.0 362.87 -0.000045 18784.00 18074.00 3.93 79.7200 18.2800 49 248730.0 362.88 -0.000045 18822.00 18146.00 3.73 79.7200 18.2800 50 248690.0 362.87 -0.000045 18820.00 18146.00 3.71 79.7200 18.2800 Si 294400.0 362.99 -0.000063 20956.00 20165.00 3.92 79.7200 18.2800 52 293350.0 362.90 -D.000063 20920.00 20130.00 3.92 79.7200 18.2800 53 292830.0 362.87 -0.000062 20902.00 20130.00 3.84 79.7200 18.2800 54 293000.0 362.87 -;0.000062 20908.00 20057.00 4.24 79.7200 18.2800 55 294860.0 362.88 -0.000063 20974.00 20203.00 3.82 79.7200 18.2800 56 96420.0 348.90 -0.000009 10970.00 9936.00 10.41 79.7200 18.2800 57 98620.0 349.01 -0.000010 11157.00 10269.00 8.65 79.7200 18.2800 58 98770.0 348.89 -0.000010 11188.00 10752.00 4.06 79.7200 18.2800 59 101450.0 348.77 -0.000010 11426.00 11197.00 2.05 79.7200 18.2800 60 101940.0 348.72 -0.000011 11454.00 10753.00 6.52 79.7200 18.2800 61 112130.0 348.54 -0.000013 12288.00 11459.00 7.23 79.7200 18.2800 62 112140.0 348.42 -0.000013 12312.00 11904.00 3.43 79.7200 18.2800 63 112430.0 348.30 -0.000013 12328.00 12165.00 1.34 79.7200 18.2800 64 121160.0 348.06 -0.000015 13013.00 12389.00 5.04 79.7200 18.2800 65 121160.0 348.75 -0.000015 12937.00 12162.00 6.37 79.7200 18.2800 66 131130.0 348.66 -0.000018 13657.00 12681.00 7.70 79.7200 18.2800 67 141630.0 348.54 -0.000021 14374.00 13350.00 7.67 79.7200 18.2800 68 151540.0 349.01 -0.000024 14944.00 13902.00 7.50 79.7200 18.2800 69 162580.0 349.01 -0.000027 15617.00 14569.00 7.19 79.7200 18.2800 70 173260.0 348.90 -0.000030 16265.00 15238.00 6.74 79.7200 18.2800 71 204430.0 348.90 -0.000041 17978.00 17351.00 3.61 79.7200 18.2800 72 185490.0 348.79 -0.000035 16949.00 15980.00 6.06 79.7200 18.2800 73 93730.0 358.57 -0.000005 9130.60 7954.60 14.78 79.7200 18.2800 74 98900.0 358.69 -0.000006 9611.80 8815.60 9.03 79.7200 18.2800 75 109890.0 358.58 -0.000008 10647.00 9991.60 6.56 79.7200 18.2800 76 118730.0 358.56 -0.000010 11405.00 10827.00 5.34 79.7200 18.2800 77 131330.0 357.86 -0.00001.3 12490.00 11988.00 4.19 79.720D 18.2800 78 140650.0 358.45 -0.000015 13098.00 12115.00 8.11 79.720D 28.2800 79 132360.0 358.33 -0.00001.3 12511.00 11523.00 8.57 79.720D 18.2800 80 120370.0 358.33 -0.000011 11579.00 10617.00 9.06 79.7200 18.2800 81 107770.0 358.34 -0.000008 10512.00 9525.20 10.36 79.7200 18.2800 82 99560.0 358.81 -0.000006 9653.70 8656.90 11.51 79.7200 18.2800 83 144700.0 358.94 -0.000016 13307.00 12360.00 7.66 79.7200 18.2800 84 158190.0 358.93 -0.000020 14210.00 13302.00 6.83 79.7200 18.2800 85 159770.0 358.81 -0.000020 14336.00 13483.00 6.33 79.7200 18.2800 86 170500.0 358.81 -0.000023 15033.00 14309.00 5.06 79.7200 18.2800 87 186890.0 358.49 -0.000028 16034.00 15354.00 4.43 79.7200 18.2800 88 198590.0 358.70 -0.000032 16680.00 16028.00 4.07 79.7200 18.2800 o:\4384-non\4384-13.wpd:1b-4303 13-59

I WESTINGHOUSE PROPRIETARY CLASS 2 89 213390.0 358.69 -0.000037 17510.00 16836.00 4.00 79.7200 18.2800 90 224860.0 358.57 -0.000041 18105.00 17381.00 4.17 79.7200 18.2800 91 241190.0 358.59 -0.000047 18952.00 18219.00 4.02 79.7200 18.2800 92 256940.0 358.70 -0. 000053 19698.00 18938.00 4.01 79.7200 18.2800 ...................................................................................................................... Bs For 92 Data Points, Average Error 5.33 %,ST - 3.37 t For 92 Subcooled Data Points, Average Error 5.33 t, STD 3.37 t WCOBRA/TRAC Model Prediction vs. Fincke Data Mean Error is .3% Standard Deviation is 3.37%

                                                              *

• WCT 4 0 0 PREDICTION 500 -  ;

10000-CX Figure 13-4-21 Prediction Comparison with Fincke Data o:\4384-non\4384-13.wpd:1 b-4303 13-60

WESTINGHOUSE PROPRIETARY CLASS 2 Jeandey Run N~o. Pressure Temperature Quality Predicted Ge Measured Gc Error L D (GP-Gm) /GM 1 899000.0 408.15 -0.000233 30458.00 29200.00 4.31 463.0000 20.1300 2 1000000.0 433.15 -0.000136 24666.00 23500.00 4.96 463.0000 20.1300 3 1996000.0 438.15 -0.000715 44856.00 43300.00 3.59 463.0000 20.1300 4 2001000.0 473.15 -0.000199 26780.00 25500.00 5.02 463.0000 20.1300 5 4202000.0 486.35 -0.001658 57104.00 54900.0 4.01 463.0000 20.1300 6 4209000.0 518.35 -0.000400 30014.00 29400.00 2.09 463.0000 20.1300 7 601S000.0 525.15 -0.002002 53040.00 49900.00 6.29 463.0000 20.1300 8 6005000.0 542.55 -0.000640 30549.00 33300.00 -8.26 463.0000 20.1300 9 7999000.0 548.75 -0.003434 53684.00 50500.00 6.30 463.0000 20.1300 10 8006000.0 563.15 -0.001013 30998.00 37200.00 -16.67 463.0000 20.1300 11 10001000.0 563.15 -0.006150 59477.00 55500.00 7.17 463.0000 20.1300 12 9989000.0 579.15 -0.001601 33754.00 40700.00 -17.07 463.0000 20.1300 13 12001000.0 580.35 -0.007971 57579.00 55300.00 4.12 463.0000 20.1300 14 12010000.0 593.55 -0.002128 36432.00 44100.00 -17.39 463.0000 20.1300 15 13995000.0 597.55 -0.008258 50977.00 52700.00 -3.27 463.0000 20.1300 16 2003000.0 421.65 -0.000931 49142.00 48800.00 0.70 463.0000 20.1300 17 2000000.0 424.25 -0.000896 48514.00 48200.00 0.65 463.0000 20.1300 18 2001000.0 438.65 -0.000712 44796.00 44800.00 -0.01 463.0000 20.1300 19 2004000.0 440.25 -0.000693 44345.00 44100.00 0.56 463.0000 20.1300 20 2004000.0 450.75 -0.000545 40445.00 40400.00 0.2.1 463.0000 20.1300 21 2003000.0 460.25 -0.000403 35875.00 36100.00 -0.62 463.0000 20.1300 22 2006000.0 466.25 -0.000312 32266.00 32700.00 -1.33 463.0000 20.1300 23 2009000.0 471.65 -0.000227 28298.00 29000.00 -2.42 463.0000 20.1300 24 2008000.0 475.55 -0.000163 24701.00 25400.00 -2.75 463.0000 20.1300 25 2008000.0 477.55 -0.00013D 22516.00 23400.00 -3.78 463.0000 20.1300 26 2004000.0 479.35 -0.000097 20180,00 21600.00 -6.57 463.0000 20.1300 27 2005000.0 481.65 -0.000059 16903.00 18800.00 -10.09 463.0000 20.1300 28 1997000.0 483.25 -0.000028 13753.00 17300.00 -20.50 463.0000 20.1300 29 2003000.0 484.55 -0.000008 11614.00 16200.00 -28.31 463.0000 20.1300 30 2003000.0 485.45 0.000001 10298.00 14500.00 -28.98 463.0000 20.1300 31 6004000.0 521.55 -0.002279 56069.00 55900.00 0.30 463.0000 20.1300 32 6005000.0 525.05 -0.001995 52987.00 52900.00 0.16 463.0000 20.1300 33 6001000.0 528.25 -0.001788 49866.00 50000.00 -0.27 463.0000 20.1300 34 6009000.0 530.95 -0.001615 47105.00 47500.00 -0.83 463.0000 20.1300 35 6009000.0 534.55 -0.001344 42863.00 43800.00 -2.14 463.0000 20.1300 36 5994000.0 537.85 -0.001052 38125.00 39700.00 -3.97 463.0000 20.1300 37 5999000.0 539.95 -0.000874 34979.00 37500.00 -6.72 463.0000 20.1300 38 6006000.0 542.95 -0.000603 29797.00 35100.00 -15.11 463.0000 20.1300 39 5998000.0 545.95 -0.000293 23865.00 33000.00 -27.68 463.0000 20.1300 40 6001000.0 548.65 -0.D00012 21502.00 31000.00 -30.64 463.0000 20.1300 41 11999000.0 579.05 -0.008492 59503.00 62000.00 -4.03 463.0000 20.1300 42 12008000.0 581.15 -0.007670 56537.00 58400.00 -3.19 463.0000 20.1300 43 12003000.0 585.35 -0.005863 49573.00 54200.00 -8.54 463.0000 20.1300 44 12003000.0 588.75 -0.004347 43190.00 51300.00 -15.81 463.0000 20.1300 45 12000000.0 592.85 -0.002428 37004.00 48200.00 -23.23 463.0000 20.1300 46 12006000.0 597.75 -0.000050 33195.00 - 44500.00 -25.40 463.0000 20.1300 47 4820000.0 507.15 -0.001476 52050.00 52100.00 -0.10 463.0000 20.1300 48 4539000.0 507.15 -0.001174 48047.00 48200.00 -0.32 463.0000 20.1300 49 4292000.0 507.15 -0.000936 44210.00 44500.00 -0.65 463.0000 20.1300 50 4011000.0 507.15 -0.000680 39393.00 40100.00 -1.76 463.0000 20.1300 51 3850000.0 507.15 -0.000529 36280.00 37200.00 -2.47 463.0000 20.1300 52 3666000.0 507.15 -0.000397 32430.00 33500.00 -3.19 463.0000 20.1300 53 3460000.0 507.15 -0.000258 27447.00 29100.00 -5.68 463.0000 20.2300 54 3326000.0 507.15 -0.000174 23692.00 26300.00 -9.92 463.0000 20.2300 55 3246000.0 507.25 -0.000122 20949.00 24500.00 -14.49 463.0000 20.1300 56 3108000.0 507.15 -0.000046 16106.00 22200.00 -27.45 463.0000 2D.1300 57 3070000.0 507.15 -0.000025 14777.00 21500.00 -31.27 463.0000 20.1300 58 3024000.0 507.25 0.000002 2.3568.00 20400.00 -33.49 463.0000 20.1300 59 6255000.0 525.15 -0.002379 56117.00 56000.00 0.21 463.0000 20.1300 60 5812000.0 525.15 -0.001707 50280.00 50100.00 0.36 463.0000 20.1300 61 5385000.0 525.15 -0.001151 43891.00 44300.00 -0.92 463.0000 20.1300 62 5087000.0 525.25 -0.000810 38756.00 39400.00 -1.63 463.0000 20.1300 63 4778000.0 525.15 -0.000509 32822.00 34300.00 -4.31 463.0000 20.1300 64 4594000.0 525.15 -0.000348 28594.00 31500.00 -9.23 463.0000 20.1300 65 4448000.0 525.15 -0.000230 24753.00 29300.00 -15.52 463.0000 20.1300 66 4334000.0 525.25 -0.000140 21239.00 27700.00 -23.32 463.0000 20.1300 67 4235000.0 525.15 -0.000075 18618.00 26400.00 -29.48 463.0000 20.1300 68 4151000.0 525.15 -0.000019 17098.00 25400.00 -32.69 463.0000 20.1300 69 8385000.0 553.25 -0.003631 52996.00 53600.00 -1.13 463.0000 20.13D0 70 7885000.0 553.15 -0.002511 46380.00 47600.00 -2.56 463.0000 20.13D0 71 7570000.0 553.15 -0.001871 41617.00 43400.00 -4.11 463.0000 20.1300 72 7357000.0 553.15 -0.001471 38052.00 40800.00 -6.74 463.0000 20.1300 73 7158000.0 553.25 -0.001106 34189.00 38700.00 -11.66 463.0000 20.1300 74 6732000.0 553.15 -0.000442 25883.00 34900.00 -25.84 463.0000 20.1300 75 6515000.0 553.15 -0.0001.34 23336.00 33100.00 -29.50 463.0000 20.1300 76 6429000.0 553.15 -0.000019 22536.00 31900.00 -29.35 463.0000 20.1300 77 2008000.0 465.45 -0.000326 32833.00 33300.00 -1.40 463.0000 20.1300 78 2005000.0 477.35 -0.000132 22675.00 23500.00 -3.51 463.0000 20.1300 79 2006000.0 484.95 -0.000003 11093.00 15300.00 -27.50 463.0000 20.1300 80 4003000.0 500.65 -0.000918 45012.00 45200.00 -0.42 463.0000 20.1300 81 4002000.0 513.65 -0.000418 31802.00 32800.00 -3.04 463.0000 20.1300 82 4003000.0 522.55 -0.000045 17100.00 25200.00 -32.14 463.0000 20.1300 83 8000000.0 549.65 -0.003302 52608.00 53400.00 -1.48 463.0000 20.1300 84 7995000.0 559.75 -0.001623 37469.00 42400.00 -11.63 463.0000 20.1300 85 8000000.0 567.75 -0.000091 26131.00 36700.00 -28.80 463.0000 20.1300 86 12006000.0 578.85 -0.008602 59878.00 62200.00 -3.73 463.0000 20.1300 87 11995000.0 592.35 -0.002648 37497.00 48200.00 -22.21 463.0000 20.1300 88 11992000.0 597.75 -0.000006 33142.00 40900.00 -18.97 463.0000 20.1300 Por 88 Data Points. Average Error - -9.07 %.STD = 11.76 % Por 86 Subcooled Data Points. Average Error = -8.56 %,STD 11.38 % For 2 Saturated Data Points, Average Error - -31.23 %.STD 3.19 % o-:\4384-non\43S4-13.vpd:lb~4303 13-61

WESTINGHOUSE PROPRIETARY CLASS 2 WCOBRA/TRAC Model Prediction vs. Jeandey Data Mean Error is -9.1% Standard Deviation is 11.8% U

• WCT 4 0 0 PREDICTION U

t~~~~~~~~ -> soooo CO 40000 m m . 1 C-4 co F= C VA 300 20000 aR; .~~~ U U UK g........... ..... .... 1OOO ti I' l I I ' l l l I l I . , . I ' lI a io6oo 20600 30600 406oo 506o 60600 -A6 Measured Moss Flow Flux (kg/m2-s) _N A.

                                                                                                                        -r Figure 134-22 Prediction Comparison with Jeandey Data o:A4384-non\4384-13.wpd:1b4303                                         13-62

WESTINGHOUSE PROPRIETARY CLASS 2 Neusen Run No. Pressure Temperature Quality Predicted Cc Measured Gc Error L D (Gp-Cm) tG (Pa) (K) (Rg/m2-s) 4T.g/m2-s) (in ) (en) i) 1 2654500.0 499.93 0.094000 12069.00 10265.00 17.57 1.0000 11.1250 2 2813000.0 503.12 0.108000 12304.00 10406.00 18.24 1.0000 11.1250 3 2875100.0 504.33 0.110000 12451.00 10546.00 18.06 1.0000 11.1250 4 2930200.0 505.39 0.121000 12420.00 10124.00 22.68 1.0000 11.1250 5 1744400.0 478.06 0.041700 9817.10 10617.00 -7.53 1.0000 11.1250 6 2606200.0 498.93 0.086400 12050.00 10617.00 13.50 1.0000 11.1250 7 2861300.0 504.06 0.100000 12586.00 10617.00 18.55 1.0000 11.1250 8 3461100.0 514.85 0.167000 13106.00 10617.00 23.44 1.0000 11.1250 9 1406500.0 467.61 0.021300 8713.40 10607.00 -17.85 1.0000 11.1250 10 2316600.0 492.61 0.063000 11514.00 10607.00 8.55 1.0000 11.1250 11 3261200.0 511.43 0.138000 13055.00 10617.00 22.96 1.0000 11.1250 12 1199700.0 460.21 0.043700 7669.10 6939.40 10.52 1.0000 11.1250 13 1516800.0 471.22 0.085100 8424.80 7030.80 19.83 1.0000 11.1250 14 2020100.0 485.47 0.142000 9314.10 7241.70 28.62 1.0000 11.1250 15 2282100.0 491.82 0.184000 9477.00 7171.40 32.15 1.0000 11.1250 16 841150.0 444.61 0.020400 6205.80 7171.40 -13.46 1.0000 11.1250 17 1489300.0 470.34 0.017000 9106.70 11741.00 -22.44 1.0000 11.1250 18 1096300.0 456.13 0.111000 6460.70 4865.30 32.79 1.0000 11.1250 19 1310000.0 464.27 0.161000 6607.50 4788.00 38.00 1.0000 11.1250 20 1572000.0 472.95 0.228000 6616.50 4009.10 465.04 1.0000 11.1250 21 2254600.0 491.18 0.041700 11648.00 11882.00 -1.97 1.0000 11.1250 22 2840600.0 503.66 0.070800 13050.00 11882.00 9.83 1.0000 11.1250 23 3550800.0 516.34 0.122000 14189.00 11812.00 20.12 1.0000 11.1250 24 3840300.0 520.94 0.163000 14199.00 11741.00 20.94 1.0000 11.1250 25 917000.0 448.29 0.072400 6191.60 4795.00 29.13 1.0000 11.1250 26 1599600.0 473.79 0.024600 9467.00 11952.00 -20.79 1.0000 6.4010 27 2330400.0 492.93 0.050300 11764.00 12023.00 -2.15 1.0000 6.4010 28 2985400.0 506.43 0.089800 13145.00 11812.00 11.29 1.0000 6.4010 29 3557700.0 516.45 0.157000 13543.00 11812.00 14.65 1.0000 6.4010 30 2109800.0 487.71 0.010900 19363.00 20460.00 -5.36 1.0000 6.4010 31 3240500.0 511.07 0.015600 22874.00 20671.00 10.66 1.0000 6.4010 32 4274700.0 527.37 0.034700 18359.00 20530.00 -10.57 1.0000 6.4010 33 5205500.0 539.59 0.058300 20323.00 20530.00 -1.01 1.0000 6.4010 34 2447600.0 495.55 0.002800 24324.00 27280.00 -10.84 1.0000 6.4010 35 3495600.0 515.43 0.005700 26635.00 27139.00 -1.86 1.0000 6.4010 36 5536400.0 543.53 0.018500 28625.00 26928.00 6.30 1.0000 6.4010 37 6515500.0 554.19 0.034200 28466.00 27139.00 4.89 1.0000 6.4010 For 37 Data Points. Average Error .

                                                 ~~~~~~~~~~~~~~~~~...............                       10.88 *, LTD .        18.22 t
                                                                                                                                                                     ===s=;===a;   =;===   =641 s;**=w=;=

For 37 Saturated Data Points, Average Error - 10.88 %,SD - 18.22 % o\4384-non4384-13.wpd:lb-4303 13-63

WESTINGHOUSE PROPRIETARY CLASS 2 WCOBRA/TRAC Model Prediction vs. Neusen Data Mean Error is 11% Standard Deviation is 18.2% N *WCT 4 0 0 PREDICTION . 3KDO Co

                      -z; C 2D000

_'C 15000 ir ir C Mo MeUsurea MOSS Iow iux Kg/mLZ-SJ

                                             -          -                      . It Figure 13-4-23 Prediction Comparison with Neusen Data o:\4384-non\4384-13.wpd:1b-4303                      13-64

WESTINGHOUSE PROPRIETARY CLASS 2 Reocreu.x Run No. Pressure Tererature Quality Predicted Cc Measured Gc Error L D (Gp-G) /M (Pa) (R) (Kg/m2-s) (Kg/m2-s? (in %) (=) (mm) l t ... ....................................................................... 1 246500.0 389.86 -0.000013 6637.10 6526.00 1.70 2335.0000 20.0000 2 246700.0 389.75 -0.000013 6705.80 6465.00 3.72 2335.0000 20.0000 3 246100.0 389.82 -0.000013 6631.90 6495.50 2.10 2335.0000 20.0000 4 212300.0 389.87 -0.000005 4049.70 4192.50 -3.41 2335.0000 20.0000 5 211800.0 389.87 -0.000005 4007.00 4164.90 -3.79 2335.0000 20.0000 6 274100.0 389.43 -0.000021 8313.60 8708.90 -4.54 2335.0000 20.0000 7 289600.0 389.44 -0.000025 9078.20 8717.80 4.13 2335.0000 20.0000 8 253900.0 389.22 -0.000016 7312.80 8681.60 -15.77 2335.0000 20.0000 9 324100.0 387.02 -0.000039 11093.00 10291.00 7.79 2335.0000 20.0000 10 330400.0 389.09 -0.000038 10915.00 10309.00 5.88 2335.0000 20.0000 11 329700.0 389.07 -0.000038 10885.00 10311.00 5.57 2335.0000 20.0000 12 329900.0 389.20 -0.000037 10868.00 10324.00 5.27 2335.0000 20.0000 13 273200.0 394.47 -0.000013 6548.10 6518:60 0.45 2335.0000 20.0000 14 273200.0 394.14 -0.000013 6687.30 6558.90 1.96 2335.0000 20.0000 15 272700.0 394.23 -0.000013 6612.10 6499.20 1.74 2335.0000 20.0000 16 243300.0 395.05 -0.000005 3960.30 4382.70 -9.64 2335.0000 20.0000 17 242400.0 394.91 -0.000005 3969.60 43 56.90 -8.89 2335.0000 20.0000 18 241800.0 394.84 -0.000005 3948.80 4355.40 -9.34 2335.0000 20.0000 19 241700.0 394.91 -0.000005 3899.40 4359.70 -10.56 2335.0000 20.0000 20 241900.0 394.91 -0.000005 3901.00 4345.40 -10.23 2335.0000 20.0000 21 242000.0 394.95 -0.000005 3901.70 4330.80 -9.91 2335.0000 20.0000 22 287200.0 394.25 -0.000017 7471.80 8474.20 -11.83 2335.0000 20.0000 23 313700.0 394.21 -0.000024 8858.10 8529.30 3.85 2335.0000 20.0000 24 298300.0 394.04 -0.000020 8136.50 8537.60 -4.70 2335.0000 20.0000 25 298600.0 394.12 -0.000020 8140.10 8508.20 -4.33 2335.0000 20.0000 26 298900.0 393.90 -0.000020 8217.60 8536.80 -3.74 2335.0000 20.0000 27 339500.0 394.00 -0.000032 10079 .00 10111.00 -0.32 2335.0000 20.0000 28 327000.0 394.01 -0.000028 9S34 .40 10131.00 -5.89 2335.0000 20.0000 For 28 Data Points. Average Error -2.60 t. STD 6.45 For 28 Subcooled Data Points. Average Error - -2.60 t. STD - 6.45 WCOBRA/TRAC Model Prediction vs. REOCREUX Data Mean Error is -2.6% Standard Deviation is 645%

                                                *

• WCT 4 0 0 PREDICTION 12OW c-C>

c, C-, 0) 0!! M Figure 134-24 Prediction Comparison with Reocreux Data o\4394-non\4384.13.wpd:1b4303 13-65

I WESTINGHOUSE PROPRIEIARY CLASS 2 Seynhaeve Run No. Pressure Temperature Quality Predicted Ge Measured Ge Error L D (Gp-GM) /GM (Pa) _ ) (gI/m2-s) (Xg/m2-s) (in %) (mm)=m 1 674900.0 433.17 -0.000011 8750.30 10105.00 -13.41 306.0000 12.5000 2 683200.0 433.34 -0.000013 9089.30 10103.00 -10.03 306.0000 12.5000 3 431200.0 413.52 -0.000013 9098.20 10138.00 -10.26 306.0000 12.50 00 4 430200.0 413.26 -0.000014 9195.70 10182.00 -9.69 306.0000 12.5000 5 471400.0 407.72 -0.000042 13712.00 15266.00 -10.18 306.0000 12.5000 6 470400.0 407.81 -0.000041 13629.00 15328.00 -11.08 306.0000 12.5000 7 474400.0 408.14 -0.000041 13683.00 15098.00 -9.37 306.0000 12.5000 8 608900.0 421.34 -0.000041 13475.00 15277.00 -11.80 306.0000 12.5000 9 602900.0 420.94 -0.000041 13414.00 15221.00 -11.87 306.0000 12.5000 10 606600.0 420.76 -0.000043 13657.00 15230.00 -10.33 306.0000 12.5000 11 418400.0 401.60 -0.000040 13625.00 15472.00 -11.94 306.0000 12.5000 12 420400.0 401.36 -0.000042 13788.00 15367.00 -10.28 306.0000 12.5000 13 874600.0 439.91 -0.000040 13080.00 14851.00 -11.93 306.0000 12.5000 14 871200.0 439.65 -0.000041 13117.00 14912.00 -12.04 306.0000 12.5000 15 873000.0 439.83 -0.000040 13069.00 14683.00 -10.99 306.0000 12.5000 16 431500.0 384.15 -0.000085 17867.00 20770.00 -13.98 306.0000 12.5000 17 430000.0 384.53 -0.000084 17765.00 20775.00 -14.49 306.0000 12.5000 18 575500.0 404.92 -0.000089 18111.00 20499.00 -11.65 306.0000 12.5000 19 578200.0 405.73 -0.000087 17991.00 20509.00 -12.28 306.0000 12.5000 20 577000.0 405.92 -0.000086 17895.00 20507.00 -12.74 306.0000 12.5000 21 1005500.0 439.93 -0.000091 17656.00 19908.00 -11.31 306.0000 12.5000 22 998500.0 439.53 -0.000091 17634.00 19653.00 -10.27 306.0000 12.5000 23 1014200.0 439.72 -0.000096 18021.00 19751.00 -8.76 306.0000 12.5000 24 999500.0 439.19 -0.000094 17869.00 18736.00 -4.63 306.0000 12.5000 25 847000.0 430.01 -0.000089 17702.00 19836.00 -10.76 306.0000 12.5000 26 838400.0 429.27 -0.000090 17772.00 19780.00 -10.15 306.0000 12.5000 27 563000.0 424.67 -0.000014 9632.60 10236.00 -5.89 221.0000 12.5000 28 563700.0 425.19 -0.000012 9274.50 10235.00 -9.38 221.0000 12.5000 29 568400.0 425.77 -0.000012 9103.30 10348.00 -12.03 221.0000 12.5000 30 407200.0 410.68 -0.000015 9670.20 10170.00 -4.91 221.0000 12.5000 31 401200.0 410.30 -0.000014 9530.40 10081.00 -5.46 221.0000 12.5000 32 400200.0 409.95 -0.000014 9665.00 10131.00 -4.60 221.0000 12.5000 33 274500.0 396.73 -0.000010 8461.90 10481.00 -19.26 221.0000 12.5000 34 283700.0 397.32 -0.000011 8798.60 10584.00 -16.87 221.0000 12.5000 35 282700.0 397.93 -0.000010 8456.50 10674.00 -20.77 221.0000 12.5000 36 510700.0 420.47 -0.000014 9615.00 9963.00 -3.49 221.0000 12.5000 37 504700.0 420.30 -0.000013 9378.00 10016.00 -6.37 221.0000 12.5000 38 500200.0 420.13 -0.000012 9235.80 10114.00 -8.68 221.0000 12.5000 39 580400.0 418.57 -0.000042 14000.00 15137.00 -7.51 221.0000 12.5000 40 571400.0 418.65 -0.000039 13599.00 15032.00 -9.53 221.0000 12.5000 41 574900.0 418.64 -0.000040 13738.00 14875.00 -7.64 221.0000 12.5000 42 381700.0 397.29 -0.000038 13792.00 15366.00 -10.24 221.0000 12.5000 43 367400.0 397.15 -0.000034 13215.00 15325.00 -13.77 221.0000 12.5000 44 358600.0 396.72 -0.000032 12962.00 15267.00 -15.10 221.0000 12.5000 45 828600.0 436.31 -0.000046 14102.00 14889.00 -5.29 221.0000 12.5000 46 824900.0 436.70 -0.000042 13692.00 14785.00 -7.39 221.0000 12.5000 47 812900.0 436.40 -0.000039 133 98.00 14691.00 -8.80 221.0000 12.5000 48 657900.0 425.90 -0.000039 13448.00 14800.00 -9.14 221.0000 12.5000 49 659400.0 425.81 -0.000040 13568.00 14733 .00 -7.91 221.0000 12.5000 50 659700.0 426.06 -0.000039 13438.00 14771.00 -9.02 221.0000 12.5000 51 943800.0 436.13 -0.000091 18160.00 19718.00 -7.90 221.0000 12.5000 52 942000.0 435.89 -0.000092 1823 9.00 19743.00 -7.62 221.0000 12.5000 53 768000.0 424.77 -0.000085 17931.00 20180.00 -11.14 221.0000 12.5000 54 758500.0 424.22 -0.000084 17858.00 20281.00 -11.95 221.0000 12.5000 55 849200.0 429.68 -0.000092 18346.00 19726.00 -7.00 221.0000 12.5000 56 839500.0 429.14 -0.000091 18280.00 19463.00 -6.08 221.0000 12.5000 57 815500.0 428.05 -0.000087 18039.00 19901.00 -9.36 221.0000 12.5000 Por 57 Data Pointr. Average Error - -10.11 - STD . 3.40 For 57 Subeooled Data Points Average Error - - 10.11 . St - 3.40 o:\4384-non\4384-13.wpd:1b-4303 13-66

WESTINGHOUSE PROPRIETARY CLASS 2 v: .£ i i, WCOBRA/TRAC Model Prediction vs. Seynhaeve Data Mean Error is -10% Standard Deviation is 3.4%

                         *       *WCT              4        0    0 PREDICTION XAM.U m 1500 -

cI

                                                                    'SO E
                     'C
                      '  1WDOO-C V>

V:7 a) C-, c)I=F co5= ft.

                             -     1      5D I.I 1000      15600         206o     256 Measured Mass Flow Flux (kg/m2-s)

_.. I ... Figure 13-4-25 Prediction Comparison with Seynhaeve Data oA\4384-non\4384-13.wpd:1b-4303 13-67

WESTINGHOUSE PROPRIETARY CLASS 2 Sozzi and Sutherland Run No. Pressure Teuperature Quality Predicted Gc Measured Gc Error D 1-," (Gp-Gm) IGM (Pa) (X) (Kg/m2-s) (Kg/m2-s0 Iin %) (nm) (nu)} ....................................................................................................................... sss 1 5377900.0 541.66 0.003500 22225.00 32371.00 -31.34 44.5000 12.7000 2 5274400.0 540.43 0.003600 21986.00 31034.00 -29.16 44.5000 12.7000 3 5963900.0 528.77 -0.001700 53730.00 65030.00 -17.38 44.5000 12.7000 4 5839800.0 533.41 -0.001200 46511.00 54191.00 -14.17 44.5000 12.7000 5 5619200.0 533.64 -0.000900 42351. 00 53581.00 -20.96 44.5000 12.7000 6 5102100.0 536.96 -0.000100 25617.00 48322.00 -46.99 44.5000 12.7000 7 4095200.0 524.78 0.003000 19652.00 32190.00 -38.95 44.5000 12.70 00 8 4860800.0 535.28 0.003000 21258.00 30438.00 -30.16 44.5000 12.70 00 9 4757300.0 533.94 0.003000 21037.00 30691.00 -31.46 44.5000 12.7000 10 6756800.0 540.25 -0.002000 52283.00 69390.00 -24.65 44.5000 12.7000 11 6894700.0 557.98 0.003000 25345.00 36521.00 -30.60 44. 5000 12.7000 12 6825800.0 557.31 0.004000 25088.00 36521.00 -31.31 44. S000 12.7000 13 6791300.0 556.97 0.004200 25004.00 36521.00 -31.54 44. 5000 12.7000 14 6722300.0 556.28 0.004400 24848.00 36521.00 -31.96 44. 5000 12.7000 15 6653400.0 555.59 0.004700 24669.00 36521.00 -32.45 44. 000 12.7000 16 6446500.0 553.48 0.005000 24218.00 35300.00 -31.39 44.5000 12.7000 17 6274200.0 551.68 0.005000 23868.00 35300.00 -32.39 44.5000 12.7000 18 6170800.0 550.59 0.005000 23640.00 35300.00 -33.03 44.5000 12.7000 19 6963600.0 558.66 0.005000 25290.00 36131.00 -30.00 44.5000 12.7000 20 6894700.0 557.98 0.005500 25105.00 36131.00 -30.52 44.5000 12.7000 21 6860200.0 557.65 0.005700 24995.00 35447.00 -29.49 44. 5000 12.7000 22 6791300.0 556.97 0.005900 24836.00 35447.00 -29.93 44. 5000 12.7000 23 6722300.0 556.28 0.006100 24676.00 34422.00 -28.31 44.5000 12.7000 24 6584400.0 554.89 0.006800 24318.00 33689.00 -27.82 44.5000 12.7000 25 4619400.0 532.12 0.000900 21371.00 36619.00 -41. 64 44.5000 12.7000 26 4550500.0 531.19 0.001600 21016.00 35794.00 -41.29 44.5000 12.7000 27 4412600.0 529.30 0.002100 20579.00 34080.00 -39. 62 44.5000 12.7000 28 4205800.0 526.38 0.002800 19901.00 30272.00 -34.26 44.5000 12.7000 29 6618900.0 510.38 -0.004100 76434.00 75671.00 1.01 44.5000 12.7000 30 6274200.0 512.64 -0.003300 71165.00 73726.00 -3.47 44.5000 12.7000 31 6136300.0 534.89 -0.001500 49414.00 66832.00 -26.06 44.5000 12.7000 32 5998300.0 546.85 -0.000200 27904.00 49802.00 -43.97 44.5000 12.7000 33 5688100.0 545.27 0.000600 23497.00 44333.00 -47.00 44.5000 12.7000 34 5377900.0 541.66 0.000800 22878.00 43357.00 -47.23 44.5000 12.7000 35 5102100.0 538.32 0.001200 22214.00 40915.00 -45.71 44.5000 12.7000 36 6481000.0 522.39 -0.003001 66302.00 71284.00 -6.99 44.5000 12.7000 37 6343100.0 529.97 -0.002200 57808.00 66402.00 -12.94 44.5000 12.7000 38 6274200.0 536.17 -0.001600 50039.00 66402.00 -24.64 44. 5000 12.7000 39 5929400.0 547.97 0.000500 23987.00 46530.00 -48.45 44.5000 12.7000 40 5791500.0 546.44 0.000900 23565.00 45603.00 -48.33 44.5000 12.7000 41 5653700.0 544.88 0.001000 23280.00 43747.00 -46.78 44.5000 12.7000 42 5515800.0 543.29 0.001100 22990.00 42478.00 -45.88 44. 5000 12.7000 43 5377100.0 541.66 0.001200 22725.00 40915.00 -44.46 44. 5000 12.7000 44 6274200.0 551.68 0.000300 24665.00 48469.00 -49.11 44.5000 12.7000 45 6136300.0 550.22 0.000300 24424.00 47141.00 -48.19 44. 5000 12.7000 46 5977700.0 548.50 0.002400 23608.00 33802.00 -30.16 44. 5000 12.7000 47 5060500.0 537.81 0.004400 21406.00 32029.00 -33.17 44. 5000 12.7000 48 5777800.0 546.28 0.004000 22950.00 32029.00 -28.35 44. 5000 12.7000 49 5757100.0 546.05 0.003700 22973.00 32337.00 -28.96 44. 5000 12.7000 50 5639900.0 544.72 0.003500 22741.00 32337.00 -29.67 44.5000 12.7000 51 5536400.0 543.53 0.003500 22536.00 32029.00 -29.64 44.5000 12.7000 52 5495100.0 543.05 0.003600 22432.00 30940.00 -27.50 44. 5000 12.7000 53 5446800.0 542.48 0.003600 22355.00 30940.00 -27.75 44. 5000 12.7000 54 5343400.0 541.25 0.003700 22117.00 30940.00 -28.52 44. 5000 12.7000 55 5240000.0 540.01 0.003700 21905.00 30940.00 -29.20 44. 5000 12 .7000 56 5171000.0 539.17 0.003000 21865.00 30940.00 -29.33 44.5000 12.7000 57 5102100.0 538.32 0.003800 21608.00 29866.00 -27.65 44.5000 12.7000 58 4998700.0 537.03 0.003800 21386.00 29329.00 -27.08 44.5000 12.7000 59 6756800.0 556.62 0.002700 25101.00 36814.00 -31.82 44.5000 12.7000 60 6550000.0 554.54 0.004000 24540.00 34178.00 -28.20 44.5000 12.7000 61 6412100.0 553.12 0.004400 24216.00 33494.00 -27.70 44.5000 12.7000 62 6170800.0 550.59 0.004500 23698.00 33494.00 -29.25 44.5000 12.7000 63 5929400.0 547.97 0.004400 23214.00 33006.00 -29.67 44.5000 12.7000 64 5688100.0 545.27 0.004600 22701.00 31240.OD -27.33 44.5000 12.7000 65 5515800.0 543.29 0.004600 22331.00 30320.O -26.35 44.5000 12.7000 66 6770600.0 556.76 0.003100 25096.00 33640.00 -25.40 44.5000 12.7000 67 6674100.0 555.80 0.004100 24785.00 33396.00 -25.78 44.5000 12.7000 68 6584400.0 554.89 0.004400 24557.00 33299.00 -26.25 44.5000 12.7000 69 6481000.0 553.84 0.004400 24364.00 33152.00 -26.51 44.5000 12.7000 70 6446500.0 553.48 0.004400 24290.00 33152.00 -26.73 44.5000 12.7000 71 6329300.0 552.26 0.004500 24022.00 32029.O0 -25.00 44.5000 12.7000 72 6253500.0 551.47 0.004400 23868.00 32029.00 -25.48 44.5000 12.7000 73 6170800.0 550.59 0.004400 23720.00 30125.O0 -21.26 44.5000 12.7000 74 6067300.0 549.48 0.004600 23474.00 30125.00 -22.08 44.5000 12.7000 75 5963900.0 548.35 0.004600 23271.00 29539.00 -21.22 44.5000 12.7000 76 5860500.0 547.21 0.004600 23039.00 29539.00 -22.00 44.5000 12.7000 77 6288000.0 547.53 -0.000500 33013.00 62955.00 -47.56 44.5000 12.7000 78 6088000.0 547.89 -0.000200 27898.00 55084.00 -49.35 44.5000 12.7000 79 5743200.0 545.89 0.000000 24061.00 51149.00 -52.96 44.5000 12.7000 80 5626000.0 544.56 0.000004 23890.00 47575.00 -49.78 44.5000 12.7000 81 5722600.0 545.66 0.001500 23312.00 39704.00 -41.29 44.5000 12.7000 82 5639100.0 544.71 0.003000 22845.00 31477.00 -27.42 44.5000 12.7000 83 5605400.0 544.32 0.003500 22691.00 32190.00 -29.51 44.5000 12.7000 84 5515800.0 543.29 0.003500 22511.00 33089.00 -31.97 44.5000 12.7000 85 5446800.0 542.48 0.003500 22354.00 32371.00 -30.94 44.5000 12.7000 86 5412300.0 542.07 0.003500 22276.00 30760.00 -27.58 44.5000 12.7000 87 6136300.0 500.23 -0.003800 77294.00 82270.00 -6.05 44.5000 12.7000 88 6067300.0 514.93 -0.002800 67212.00 72993.00 -7.92 44. 5000 12.7000 89 5653700.0 544.88 0.000500 23485.00 45896.00 -48.83 44. 5000 12.7000 90 5377900.0 541.66 0.000500 23003.00 44968.00 -48.85 44. 5000 12.7000 91 5033100.0 537.46 0.000900 22194.00 39646.00 -44.02 44. 5000 12.7000 92 5240000.0 540.01 0.000600 22710.00 44187.00 -48.60 44. 5000 12.7000 93 6481000.0 543.64 -0.001200 43290.00 63814.00 -32.16 44.5000 12.7000 o:\4384-non\4384-13.wpd:1b-4303 13-68

WESTINGHOUSE PROPRIETARY CLASS 2 94 6205200.0 550.95 0.0 00500 24449.00 47751.00 -48.80 44.5000 12.7000 95 5791500.0 546.44 0.001300 23454.00 42917.00 -45.35 44.5000 12.7000 96 5653700.0 544.88 0.001500 23166.00 39792.00 -41.78 44.5000 12.7000 97 5446000.0 542.47 0.001600 22753.00 38328.00 -40.64 44.5000 12.7000 98 4826300.0 517.33 -0.001000 47925.00 59078.00 -18.88 44.5000 12.7000 99 4688400 .0 524.01 -0.000500 37541.00 56149.00 -33.14 44.5000 12.7000 100 4550500.0 531.19 0.000500 21455.00 41111.00 -47.81 44.5000 12.7000 101 4205800.0 526.38 0.001000 20557.00 39792.00 -48.34 44.5000 12.7000 102 4067900.0 524.37 0.001300 20147.00 39402.00 -48.87 44.5000 12.7000 103 3447400.0 514.62 0.000002 20165.00 51755.00 -61.04 44.5000 12.7000 104 3309500.0 496.25 -0.000500 40683.00 47311.00 -14.01 44.5000 12.7000 105 3171600.0 509.84 0.000002 19605.00 41990.00 -53.31 44.5000 12.7000 106 3033700.0 507.33 0.000100 18897.00 39988.00 -52.74 44.5000 12.7000 107 3447400.0 514.62 0.000002 20165.00 40818.00 -50.60 44.5000 12.7000 108 3447400.0 514.62 0.000100 19787.00 39744.00 -50.21 44.5000 12.7000 109 3447400.0 514.62 0.000400 19391.00 38328.00 -49.41 44.5000 12.7000 110 3378400.0 513.45 0.000450 19203.00 35935.00 -46.56 44.5000 12.7000 1ll 3309500.0 512.27 0.000500 19011.00 35545.00 -46.52 44.5000 12.7000 112 3171600.0 509.84 0.000900 18394.00 31248.00 -41.14 44.5000 12.7000 113 3171600.0 509.84 0.001000 18330.00 30272.00 -39.45 44.5000 12.7000 114 6150100.0 550.37 0.000300 24445.00 49313.00 -50.43 44.5000 12.7000 115 5805300.0 546.59 0.000800 23645.00 45310.00 -47.82 44.5D00 12.7000 116 6715400.0 544.29 -0.001500 46322.00 65426.00 -29.20 44.5000 12.7000 117 6467200.0 553.69 0.000004 25205.00 52863.00 -52.32 44.5000 22.7000 118 6225900.0 551.17 0.002000 24181.00 42282.00 -42.81 44.5D00 12.7000 119 5908800.0 547.74 0.001500 23639.00 41.208.00 -42.63 44.5000 12.7000 120 6749900.0 542.06 -0.001800 49895.00 68843.00 -27.52 44.5000 12.7000 121 6501700.0 548.40 -0.000700 35879.00 62740.00 -42.81 44.5000 12.7000 122 6356900.0 552.55 0.000500 24741.00 49802.00 -50.32 44.5000 12.7000 123 6177700.0 550.66 0.001200 24227.00 44919.00 -46.07 44.5000 12.7000 124 5784700.0 546.36 0.001800 23350.00 40051.00 -41.70 44.5000 12.7000 125 7074000.0 559.72 0.002000 25825.00 42478.00 -39.20 44.5000 12.7000 126 6998100.0 558.99 0.003000 25552.00 39548.00 -35.39 44.5000 12.7000 127 6481000.0 553.84 0.004000 24395.00 36570.00 -33.29 44.5000 12.7000 128 6219000.0 551.10 0.005000 23742.00 34910.00 -31.99 44.5000 12.7000 129 6743000.0 556.49 0.002900 25064.00 48044.00 -47.83 44.5000 12.7000 130 6584400.0 554.89 0.004300 24578.00 45607.00 -46.11 44.5000 12.7000 131 6300700.0 551.96 0.005200 23902.00 42590.00 -43.88 44.5000 12.7000 132 5998300.0 548.73 0.005300 23241.00 41667.00 -44.22 44.5000 12.7000 133 6687100.0 539.06 -0.002000 52713.00 75825.00 -30.48 44.5000 12.7000 134 6550000.0 543.70 -0.001300 64437.00 75825.00 -41.40 44.5000 12.7000 135 6481000.0 549.83 -0.000500 32716.00 64937.00 -49.62 44.5000 12.7000 1.36 6343100.0 552.41 0.000004 25020.00 61031.00 -59.00 44.5000 12.7000 137 5915700.0 547.82 0.002100 23536.00 48473.00 -51.45 44.5000 12.7000 138 5722600.0 545.66 0.002200 23149.00 45993.00 -49.67 44.5000 12.7000 139 6756000.0 556.62 0.004900 24875.00 44709.00 -44.36 44.5000 12.7000 140 6481000.0 553.84 0.006000 24194.00 42859.00 -43.55 44.5000 12.7000 141 6190300.0 550.80 0.006500 23527.00 41540.00 -43.36 44.5000 12.7000 142 6653400.0 538.47 -0.002000 52408.00 75010.00 -30.13 57.2000 12.7000 143 6618900.0 539.93 -0.001800 50114.00 67847.00 -26.14 57.2000 12.7000 144 6481000.0 545.49 -0.001000 39799.00 61368.00 -35.15 57.2000 12.7000 145 6412100.0 553.12 0.000300 24407.00 58195.00 -58.06 57.2000 12.7000 146 6839500.0 557.44 0.002000 25014.00 43698.00 -42.76 57.2000 12.7000 147 6777500.0 556.83 0.002500 24829.00 40720.00 -39.03 57.2000 12.7000 148 6743000.0 556.49 0.002800 24726.00 39890.00 -38.01 57.2000 12.7000 149 6722300.0 556.28 0.003000 24666.00 39890.00 -38.16 57.2000 12.7000 150 6504400.0 554.08 0.003000 24189.00 39250.00 -38.37 57.2000 12.7000 151 6481000.0 553.84 0.003800 24057.00 38035.00 -36.75 57.2000 12.7000 152 6446500.0 553.48 0.004400 23914.00 38035.00 -37.13 57.2000 12.7000 153 6329300.0 552.26 0.004500 23673.00 36492.00 -35.13 57.2000 12.7000 154 6239700.0 551.32 0.004500 23473.00 36687.00 -36.02 57.2000 12.7000 155 6205200.0 550.95 0.004500 23398.00 36004.00 -35.01 57.2000 12.7000 156 6122500.0 550.07 0.004800 23183.00 34617.00 -33.03 57.2000 12.7000 157 5904600.0 547.70 0.004000 22797.00 34265.00 -33.47 57.2000 12.7000 158 6025800.0 549.02 0.003500 23109.00 39592.00 -41.63 57.2000 12.7000 159 6812000.0 557.17 0.004000 24742.00 39709.00 -37.69 57.2000 12.7000 160 6756800.0 556.62 0.004000 24622.00 39021.00 -36.90 57.2000 12.7000 161 6722300.0 556.28 0.004000 24550.00 38035.00 -35.45 57.2000 12.7000 162 6632700.0 555.38 0.004700 24300.00 38088.00 -36.20 57.2000 12.7000 163 6584400.0 554.89 0.005000 24171.00 36101.00 -33.05 57.2000 12.7000 164 6515500.0 554.19 0.005000 24022.00 36101.00 -33.46 57.2000 12.7000 165 6412100.0 553.12 0.005500 23751.00 35344.00 -32.80 57.2000 12.7000 166 6329300.0 552.26 0.005000 23623.00 35344.00 -33.16 57.2000 12.7000 167 6308700.0 552.05 0.005200 23558.00 34885.00 -32.47 57.2000 12.7000 168 6205200.0 550.95 0.005200 23329.00 34441.00 -32.26 57.2000 12.7000 169 6081100.0 549.62 0.005500 23009.00 33011.00 -30.30 57.2000 12.7000 170 6032900.0 549.10 0.006000 22851.00 33753.00 -32.30 57.2000 12.7000 171 6067100.0 549.47 0.002900 23265.00 39744.00 -41.46 57.2000 12.7000 172 6818900.0 557.24 0.003300 24820.00 39744.00 -37.55 57.2000 12.7000 173 6777500.0 556.83 0.003500 24713.00 39744.00 -37.82 57.2000 12.7000 174 6681000.0 555.87 0.004000 24480.00 39744.00 -38.41 57.2000 12.7000 175 6543100.0 554.47 0.004000 24163.00 37839.00 -36.14 57.2000 22.7000 176 6300700.0 551.96 0.004500 23598.00 36497.00 -35.34 57.2000 12.7000 177 6067300.0 549.48 0.004300 23111.00 35764.00 -35.38 57.2000 12.7000 178 6770600.0 541.44 -0.001900 50451.00 72993.00 -30.88 57.2000 12.7000 179 6667200.0 545.37 -0.001300 43313.00 62740.00 -30.96 57.2000 12.7000 180 6577500.0 549.33 -0.000700 34968.00 62740.00 -44.27 57.2000 12.7000 181 4150100.0 550.37 0.001200 23693.00 43088.00 -45.01 57.2000 12.7000 182 6012200.0 548.88 0.001300 23386.00 42331.00 -44.75 57.2000 12.7000 183 5798400.0 546.52 0.001700 22884.00 42331.00 -45.94 57.2000 12.7000 184 6701600.0 537.24 -0.002200 54504.00 71529.00 -23.80 57.2000 12.7000 185 6591300.0 541.46 -0.001600 47639.00 66500.00 -28.36 57.2000 12.7000 186 6391300.0 552.91 0.000004 24519.00 56783.00 -56.82 57.2000 12.7000 187 5770900.0 546.21 0.001300 22884.00 40891.00 -44.04 57.2000 12.7000 188 5605400.0 544.32 0.001500 22520.00 40891.00 -44.93 57.2000 12.7000 189 6694800.0 556.00 0.004000 22318.00 23802.00 -6.23 362.0000 12.7000 190 6577500.0 554.82 0.004000 22048.00 23802.00 -7.37 362.0000 12.7000 191 6460300.0 553.62 0.004000 21775.00 23802.00 -8.52 362.0000 12.7000 192 6350000.0 552.48 0.004300 21505.00 23802.00 -9.65 362.0000 12.7000 193 6867100.0 557.71 0.003500 22727.00 25291.00 -10.14 362.0000 12.7000 194 6832600.0 557.38 0.003000 22705.00 25291.00 -10.22 362.0000 12.7000 o:X384-non\4384-.13.wpd:lb-4303 13-69

WESTINGHOUSE PROPRIETARY CLASS 2 195 6694800.0 556.00 0.003000 22393.00 23802.00 -5.92 362.0000 12.7000 196 6556900.0 554.61 0.003000 22105.00 23802.00 -7 .13 3 62.0000 12.7000 197 6412100.0 553.12 0.003000 21761.00 23802.00 -8.57 3 62.0000 12.7000 198 6274200.0 551.68 0.003000 21441.00 23802.00 -9.92 3 62.0000 12.7000 199 6577500.0 533.71 -0.002300 51293.00 46872.00 9.43 362.0000 12.7000 200 6474100.0 537.45 -0.001800 45776.00 41990.00 9.02 3 62.0000 12.700D 201 6411000.0 540.63 -0.001400 40848.00 37595.00 8.65 3 62.0000 12.7000 202 6370700.0 543 .9t -0 .001000 35173.00 33933.00 3.65 3 62.0000 12.7000 203 6329300.0 543 .40 -0 .001000 3533 8.00 30760.00 24.88 3 62 .0000 12.7000 204 6253500.0 547.11 -0 .000500 27129.00 28026.00 -3 .20 362.0000 12.7000 205 6074200.0 549.55 0.000004 21306.00 25780.00 -17.35 362.0000 12.7000 206 5095000.0 538.23 0.000200 18959.00 24657.00o -23 .11 362.0000 22.7000 207 5715700.0 545.58 0.000500 20406.00 23 802.00 -14 .27 362.OOOD 12.7000 208 6674100.0 534.50 -0 .002400 56455.00 57125.00 -1.17 82.600D 12.7D00 209 6598200.0 536.40 -0.002100 53236.00 57208.00 -6 .94 82. 6000 12.7000 210 6481000.0 536.46 -0.001900 51501.00 54293 .00 -5.14 82. 6000 12.7000 211 6425900.0 542.84 -0 .001200 4203 6.00 52609.00o -20 .0 82. 6000 12.7000 212 6288000.0 551.83 0.000004 23669.00 43762.00 -45.91 82. 6000 12.7000 213 6853300.0 557.58 0.002700 24593.00 31936.00 -22 .99 82.6000 12.7000 214 6784400.0 556.90 0.002700 24427.00 31936.00 -23 .51 82.6000 12.7000 215 6701600.0 556.07 0.003 700 24154. 00 31936.OD -24 .37 82.600D 12.70D0 216 6584400.0 554.89 0.003 400 23919.00 31936.00 -25.10 82.6000 12.7000 217 6392400.0 552.91 0.003900 23434.00 29686.00 -21 .06 82.6000 12.7000 218 6198300.0 550.88 0.004000 22992.00 29686.00 -22 .55 82.6000 12.7000 219 6853300.D 541.85 -0 .002000 503 54.00 58102.00 -13 .34 82. 6000 12.7000 220 6777500.D 543.41 -0.001700 47149.00 58102.00 -18 .85 82.6000 12.7000 221 6618900.D 551.43 -0 .000500 30639.00 51022.00 -39.95 82.6000 12.700D 222 6315500.D 552.12 0.001090 23527.00 38669.00 -39.16 82.6000 12.7000 223 6026000.D 549.03 0.001500 22843.00 35398.00 -35.47 82.6000 12.7000 224 5736400.0 545.82 0.001900 22140.00 32225.00o -31 .30 82.6000 12.7000 225 6790200.0 556.96 0.001500 21917.00 23241.00 -5 .70 552.5000 12.7000 226 6687100.0 555.93 0.002000 21632.00 22313.00 -3 .05 552.5S000 12.7000 227 6577500.0 554.82 0.002000 21.393.00 22704.00 -5 .77 552.5S000 12.7000 228 6467200.0 553 .69 0.002000 21153.00 22323 .00 -5 .20 552.5000 12.7000 229 6350000.0 552.48 0.002000 20913.00 22483 .00 -2 .6S 552.5000 12.7000 230 6715400.0 539.55 -0.002000 443 82.00 41111.00 7.96 552.5000 12.7000 231 6694800.0 539.19 -0.002000 44474.00 38328.00 16.04 552.5000 12.7000 232 656380D.0 546.60 -0 .001000 32634.00 35740.00 -8.69 552.5000 12.7000 233 6508600.0 545.86 -0.001000 32808.00 33592.00 -2 .33 552.5000 12.7000 234 6460300.0 545.21 -0.001000 32976.00 32225.00o 2.33 552.5000 12.7000 235 6419D00.0 549~.lo -0.000500 25487.00 28319.00 -10 .00 552.500D 12.7000 236 5991500.0 548.65 0.001000 20218.00 22704.00 -10 .95 552.5D00 12.7000 237 58122DD.0 546.67 0.001000 19796.OD 21776.00 -9 .09 552.500D 12.7000 238 68120D0.0 5 57 .17 0.0 023 00 29447.OD 3 1492 .00 -6.49 10.8000 12.7000 239 6777500.0 556.83 0.002500 29295.00 3 1492 .0 -6.98 10.8000 12.7000 240 6729100.0 556.35 0.0029C00 29092.00 31492.00 -7.62 10.8000 12.7000 241 6667200.0 555.73 0.003200 28948.00 30428.00 -4 .83 10. 8000 12.7000 242 6563800.0 5 54 .68 0.00320 0 28768.00 30418.00 -5 .42 10. 8000 12.7000 243 6391400.0 552.91 0.003500 28437.00 28465.00 -0 .10 10. 8000 12 .7000 244 6232800.0 551.25 0.003600 28152.00 28465.00o -1.10 10. 8000 22.7000 245 6039800.0 549.18 0.003500 27881.00 28465.00 -2 .05 10. 8000 12.7000 246 6867100.0 S57.71 0.003700 29089.00 29881.00 -2 .65 10. 800D 12.7000 247 6805100.0 557.1D 0.004400 28788.00 29881.00 -3 .66 10. 8000 12.7000 248 6708500.0 5 56.14 0.004500 28600.00 29882.00 -4 .29 10.8000 12.7000 249 6639600.0 5 55.4 5 0.004500 28510. 00 28953.00 -1.53 10.8000 12.7000 250 6570600.0 554.75 0.004500 28420.00 28953.00 -1.84 10.8000 12.7000 251 6474100.0 553.76 0.004500 28281.00 28953.00 -2 .32 10.8000 12.7000 252 6301800.0 5 51.97 0.004 500 27995.00 27830.00 D.59 10.8000 12.7000 253 6736100.0 536.81 -0.002300 59147.00 55172.00 7.20 10.8000 12.7000 254 6660300.0 537.54 -0.002100 57399.00 50778.00 13.04 10.8000 12.7000 25S 6584400.0 539.35 -0. 001800 54312.00 50778.00 6.96 10.8000 12.7000 256 6508600.0 541.15 -0. 001500 51059.00 47849.00 6.71 10.8000 12.7000 257 6432800.0 545.11 -0.000970 44576.00 47849.00 -6.84 10.8000 12.7000 258 6363000.0 544.16 -0.000970 44790.00 46304.00 -3 .27 10.8000 12 .7000 259 6225900.0 547.59 -0 .000410 37000.00 43454.00 -14 .85 10.8000 12.7000 260 5708800.0 545.51 0.000600 28827.00 33201.00 -13 .17 10.8000 12.7000 261 6722300.0 556.28 0.003000 21091.00 22362.00 -5.68 679.5000 12.7000 262 6460300.0 5 53 .62 0.004 000 20423.00 22362.00 -8 .67 679 .5000 12.7000 263 6743000.0 556.49 0.003000 21140.00 22557.00 -6 .28 679.5S000 12.7000 264 6501700.0 5 54.05 0.004 000 20523.00 22557.00 -9 .02 679.5S000 12.7000 265 6867100.0 557.71 0.003000 21404.00 22S57.00 -5 .11 679.5000 12.7000 266 6474100.0 535.19 -0.002000 44324 .00 41211.00 7.82 679.5000 12.7000 267 6260400.D 542.42 -0 .001000 32S72.00 27391.00 18.91 679.5000 12.7000 268 6956800.0 558.59 0.001000 21796.00 2543 8.00 -14 .32 679.5000 12.7000 269 6515500.0 554.19 0.002500 20698.00 22826.00 -9 .32 679.5000 12.7000 270 6246600.D 551.39 0.003000 20064.00 22582.00 -11 .15 679.5000 12.7000 271 6874000.0 557.78 0.004000 221331.00 21971.00 -2 .91 679.5000 12.7000 272 6798200.0 557.03 0.004000 21160.00 21971.00 -3 .69 679.5S000 12.7000 273 6687900.0 555.94 0.004000 20916.00 21971.00 -4 .8D 679.5000 12 .7000 274 6453400.0 553.55 0.004000 20424.00 21971.00 -7.04 679.5000 12 .7000 275 6660300.0 526.40 _0.003000 54114.00 46481.00 16.42 679.5000 12.7000 276 6515500.0 529.82 -0.002500 49780.00 41745.00 19.25 679.5D00 12.7000 277 6274200.0 542.162 -0 .001000 32498.00 27098.00 19.93 679.5000 12 .7000 278 6094900.0 545.13 -0.000500 25119.00 22948.00 9.46 679.500D 12.7000 279 5908800.0 545.8D -0.000200 21232.00 20995.00 1.13 679.5000 12.7000 280 7129100.0 56D.25 0.001000 22138.00 23485.00 -5 .74 679.5000 12.7000 281 7080100.0 559.78 0.002000 21950.00 23485.00 -6 .54 679.500D 12.7000 282 6936100.0 55R.39 0.002500 21609.00 23485.00 -7 .99 679.500D 12.7000 283 6704400.0 556.10 0.003000 21069 . 0 2294S.00 -8 .19 679. S000 12.7000 284 663960D.0 555.45 0.003000 20923.00 22948.00 -8 .82 679.5S00D 12.7000 285 6487900.0 553 .91 0.003000 20608.00 22948.00 -10 .20 679. S00D 12.7000 286 7018800.0 559.19 0. 002000 21808.00 22826.00 -4 .46 679.5S000 12.7000 287 6977400.0 558.79 0.002000 21741.00 22826.00 -4 .75 679.5S00D 12.7000 288 6832600.0 557.38 0.002000 21435.00 22626.00 -5 .26 679.5S000 12.7000 289 6694800.0 556.00 0.002500 21084.00 22626.00 -6 .82 679.5000 12.7000 290 6556900.0 5 5 4.6 1 0.002500 20796.00 22826.00 -8.89 679.5S000 12.7000 291 6412100.0 5 53 .12 0.003 000 20436.00 22 826.00 -10 .47 679.5000 12.7000 292 6260400.0 551.54 0.003000 20088.00 22826.00 -12 .00 679. 500D 12.7000 293 7039500.0 559.39 0.003000 21759.00 23924.00 -9.05 679.5000 12.7000 294 6998100.0 558.99 0.003000 21663 .00 23924.00 -9 .45 679.5000 12.7000 295 6308700.0 549.53 -0.000300 22659.00 22118.00 2.45 679. 5000 12.7000 o:A4384-nonW4384-1 3.wpd:1b-4303 13-70

WESTINGHOUSE PROPRIETARY CLASS 2 296 680.0 524.24 '.i03200 55937.00 42527.00 ?~ 3liS3 679.5000 12.7000 297 6632700.0 525.64 -0.003000 54422.00 39939.00 36.26 679.5000 12.7000 298 6570600.0 531.08 -0.002500 49369.00 38669.00 27.67 679.5000 12.7000 299 6508600.0 535.84 -0.002000 44148.00 35252.00 25.24 679.5000 12.7000 300 6460300.0 540.38 -0.001500 38385.00 32078.00 19.66 679.5000 1.2.7000 301 6425900.0 541.86 -0.001300 36016.00 32078.00 22.28 679.5000 12.7000 302 6743000.0 509.69 -0.004400 65374.00 48581.00 34.57 679.5000 12.7000 303 6667200.0 508.84 -0.004300 65195.00 48581.00 34.20 679.5000 12.7000 304 6584400.0 509.34 -0.004100 64242.00 51266.00 25.31 679.5000 12.7000 305 6536200.0 510.94 -0.003900 62931.00 51266.00 22.75 679.5000 12.7000 306 6460300.0 519.37 -0.0D3200 57068.00 45505.00 25.41 679.5000 12.70800 307 6749900.0 552.24 -0.000590 25624.00 27147.00 -5.61 679.5000 12.70800 308 6701600.0 552.14 -0.000530 25007.00 24168.00 3.47 679.5000 12.7000 309 6646500.0 552.06 -0.000460 24330.00 23582.00 3.17 679.5000 12.7000 310 6584400.0 551.99 -0.000380 23592.00 22752.00 3.69 679.5000 12.7000 311 6522400.0 552.09 -0.000280 22726.00 21971.00 3.44 679.5000 12.7000 312 6474100.0 552.27 -0.000190 22003.00 21971.00 0.15 679.5000 12.7000 313 6439600.0 552.54 -0.000110 21455.00 21971.00 -2.35 679.5000 12.7000 314 6825800.0 557.31 0.001300 21504.00 20897.00 2.90 679.5000 12.7000 315 6805100.0 557.10 0.001800 21389.00 20897.00 2.35 679.5000 12.7000 316 6701600.0 532.72 -0.002600 49424.00 39060.00 26.53 679.5000 12.7000 317 6715400.0 533.01 -0.002600 49337.00 39060.00 26.31 679.5000 12.7000 318 6667200.0 533.18 -0.002500 48578.00 38035.00 27.72 679.5000 12.7000 319 6584400.0 532.64 -0.002400 48124.00 36668.00 31.24 679.5000 12.7000 320 6515500.0 534.83 -0.002100 45219.00 36375.00 24.31 679.5000 12.7000 321 6446500.0 539.12 -0.001600 39706.00 34910.00 13.74 679.5000 12,7000 322 6384500.0 542.23 -0.001200 34894.00 30516.00 14.35 679.5000 12.7000 323 6805100.0 557.10 0.001000 21470.00 25682.00 -16.40 679.5000 12.7000 324 6818900.0 557.24 0.002800 21335.00 23338.00 -8.58 679.5000 12.7000 325 6805100.0 557.10 0.003500 21222.00 21044.00 0.85 679.5000 12.7000 326 6777500.0 556.83 0.003400 21160.00 21044.00 0.55 679.5000 12.7000 327 6743000.0 556.49 0.003300 21101.00 21044.00 0.27 679.5000 12.7000 328 6708500.0 556.14 0.003600 21017.00 21044.00 -0.13 679.5000 12.7000 329 6853300.0 557.58 0.001500 21522.00 22167.00 -2.91 679.5000 12.7000 330 6825800.0 557.31 0.001900 21423.00 22167.00 -3.36 679.5000 12.7000 331 6874000.0 557.78 0.000520 21668.00 22167.00 -2.25 679.5000 12.7000 332 6874000.0 557.78 0.000790 21640.00 22167.00 -2.38 679.5000 12.7000 333 6805100.0 557.10 0.002000 21362.00 22167.00 -3.63 679.5000 12.7000 334 6770600.0 556.76 0.002000 21291.00 20897.00 1.89 679.5000 12.7000 335 6743000.0 556.49 0.002200 21219.00 20897.00 1.54 679.5000 12.7000 336 6715400.0 538.53 -0.002100 43989.00 35935.00 22.41 679.5000 12.7000 337 6749900.0 538.12 -0.002200 44824.00 35935.00 24.74 679.5000 22.7000 338 6722300.0 538.65 -0.002100 43917.00 34910.00 25.80 679.5000 12.7000 339 6646500.0 538.35 -0.002000 43267.00 32078.00 34.88 679.5000 12.7000 340 6570600.0 539.12 -0.001800 41423.00 31004.00 33.61 679.5000 12.7000 341 6501700.0 541.04 -0.001500 38212.00 30467.00 25.42 679.5000 12.7000 342 6446500.0 544.09 -0.001100 33304.00 29686.00 12.19 679.5000 22.7000 343 6391400.0 545.48 -0.000870 30188.00 25731.00 17.32 679.5000 12.7000 344 6350000.0 548.63 -0.000460 24293.00 23680.00 2.59 679.5000 12.7000 345 6736100.0 542.77 -0.001700 39475.00 34861.00 13.24 679.5000 12.7000 346 6708500.0 542.34 -0.001700 39571.00 32957.00 20.07 679.5000 12.7000 347 6646500.0 542.32 -0.001600 38720.00 31785.00 21.82 679.5000 12.7000 348 6584400.0 542.32 -0.001500 37807.00 31004.00 21.94 679.5000 12.7000 349 6515500.0 543.19 -0.001300 35613.00 30516.00 16S.70 679.5000 12.7000 350 6460300.0 545.21 -0.001000 31835.00 28660.00 11.08 679.5000 12.7000 351 6405200.0 547.90 -0.000620 26429.00 27879.00 -5.20 679.5000 12.7000 352 6074000.0 549.55 0.004000 19595.00 22069.00 -11.21 679.5000 12.7000 353 6756000.0 556.62 0.004000 21062.00 22069.00 -4.56 679.5000 12.7000 354 6625800.0 555.31 0.004000 20794.00 22069.00 -5.78 679.5000 12.7000 355 6508600.0 554.12 0.004000 20549.00 22069.00 -6.89 679.5000 12.7000 356 6377600.0 552.77 0.004000 20249.00 22069.00 -8.25 679.5000 12.7000 357 6756800.0 556.62 0.002700 23746.00 27342.00 -13.15 158.8000 12.7000 358 6722300.0 556.28 0.002700 23676.00 27342.00 -13.41 158.8000 12.7000 359 6632700.0 555.38 0.003000 23431.00 27342.00 -14.30 158.8000 12.7000 360 6536200.0 554.40 0.003100 23195.00 28074.00 -17.38 158.8000 12.7000 361 6474100.0 553.76 0.003200 23058.00 28074.00 -17.87 158.8000 12.7000 362 6315500.0 552.12 0.003200 22690.00 27342.00 -17.01 158.8000 12.7000 363 6150100.0 550.37 0.003300 22272.00 26610.00 -16.30 158.8000 12.7000 364 6839500.0 543.46 -0.001800 46547.00 47604.00 -2.22 158.8000 12.7000 365 6770600.0 545.09 -0.001500 43114.00 46872.00 -8.02 158.8000 12.7000 366 6701600.0 544.08 -0.001500 43450.00 43454.00 -0.01 158.8000 12.7000 367 6639600.0 546.74 -0.001100 38137.00 43454.00 -12.24 158.8000 12.7000 368 6536200.0 552.08 -0.000300 25856.00 38572.00 -32.97 158.8000 12.7000 369 6205200.0 550.95 0.000800 22629.00 31248.00 -27.58 158.8000 12.7000 370 5991500.0 548.65 0.001000 22106.00 31004.00 -28.70 158.8000 12.7000 371 5764000.0 546.13 0.001100 21572.00 29783.00 -27.57 158.8000 12.7000 372 6800900.0 557.06 0.000300 18501.00 20165.00 -8.25 1822.5000 12.7000 373 6867100.0 557.71 0.002000 18401.00 -19286.00 -4.59 82.00 12.7000 374 6032600.0 549.10 0.002800 16721.00 18358.00 -8.92 1822.5000 12.7000 375 6791300.0 556.97 0.003100 18124.00 17919.00 1.14 1822.5000 12.7000 376 6756000.0 556.62 0.002500 18149.00 17919.00 1.28 1822.5000 12.7000 377 6722300.0 556.28 0.003500 17948.00 17528.00 2.40 1822.5000 12.7000 378 6681000.0 555.87 0.003300 17918.00 17528.00 2.23 1822.5000 12.7000 379 6646500.0 555.52 0.003400 17836.00 17528.00 1.76 1822.5000 12.7000 380 6605100.0 555.10 0.003300 17771.00 17528.00 1.39 1822.5000 12.7000 381 6743000.0 543.79 -0.001600 30276.00 29441.00 2.84 1822.5000 12.7000 382 6736100.0 543.14 -0.001660 30804.00 29441.00 4.63 1822.5000 12.7000 383 6708500.0 542.90 -0.01640 30759.00 29441.00 4.48 1822.5000 12.7000 384 6625800.0 543.42 -0.001450 29268.00 24559.00 19.17 1822.5000 12.7000 385 6584400.0 544.20 -0.001300 27916.00 24559.00 13.67 1822.5000 12.7000 386 6550000.0 545.53 -0.001100 25994.00 23875.00 8.88 1822.5000 12.7000 387 6508600.0 546.67 -0.000910 24214.00 23875.00 1.42 1822.5000 12.7000 388 6674100.0 543.12 -0.001560 30137.00 25682.00 17.35 2822.5000 12.7000 389 6777500.0 527.53 -0.003140 43450.00 33933.00 28.05 1822.5000 12.7000 390 6756800.0 526.39, -0.003180 43959.00 33933.00 29.55 1822.5000 12.7000 391 6729200.0 526.37 -0.003130 43710.00 33933.00 28.81 1822.5000 12.7000 392 6681000.0 -526.67 -0.003020 43126.00 32810.00 31.44 1822.5000 22.7000 393 6618900.0 528.06 -0.002810 41724.00 32078.00 30.07 1822.5000 12.7000 394 6570600.0 531.85 -0.002440 38677.00 32078.00 20.57 1822.5000 22.7000 395 6522400.0 535.76 -0.002030 35203.00 31541.00 11.61 1822.5000 12.7000 396 6474100.0 539.27 -0.00163D 31599.00 28856.00 9.51 1822.5000 12.7000 o:\4384-non\4384-13.wpd:lb-4303 13-71

I WESTINGHOUSE PROPRIETARY CLASS 2 397 6591300.0 0894 -0.04140 51579.00 39304.00 31.23 1822.5000 12.7000 398 6556900.0 5050 -0.004130 51711.00 39304. 00 31.57 1822.5000 12.7000 399 6460300 .0 508.5D -0. 003910 50705.00 36863.00 37.55 1822.5000 12.7000 400 6384500.0 511.25 -0.003590 48956.00 36668.00 33.51 1822.5000 12.7000 401 6515500.0 508.15 -0.004040 51313.00 39304.00 30.55 1822.5000 12.7000 402 6322400.0 517.74 -0.003060 45242.00 35935.00 25.90 1822.5000 12.7000 403 6260300.0 523.42 -0.002530 41401.00 34226.00 20.96 1822.5000 12.7000 404 6212100.0 530.12 -0.001980 36414.00 30320.00 20.10 1822.5000 12.7000 405 6936100.0 558.39 0.001880 18532.00 19335.00 -4.15 1822.5000 12.7000 406 6915300.0 558.19 0.003010 18369.00 19335.00 -5.00 1822.5000 12.7000 407 6901600.0 558.05 0.004050 18238.00 19335.00 -5.67 1822.5000 12.7000 408 6880900.0 557.85 0.004280 18165.00 18309.00 -0.79 1822.5000 12.7000 409 6853300.0 557.58 0.004300 18110.00 17772.00 1.90 1822.5000 12.7000 410 6825800.0 557.31 0.004550 18032.00 17577.00 2.59 1822.5000 12.7000 411 6598200.0 552.53 -0.000330 20079.00 21385.00 -6.11 1822.5000 12.7000 412 6550000.0 554.54 0.000270 18044.00 21385.00 -15.62 1822.5000 12.7000 413 6407900.0 551.64 -0.000180 18847.00 20458.00 -7.87 1822.5000 12.7000 414 6653400.0 553.44 -0.000290 19885.00 21385.00 -7.01 1822.5000 12.7000 415 6432800.0 552.39 -0.000120 18515.00 20067.00 -7.73 1822.5000 12.7000 416 6377600.0 552.28 -0.000060 18075.00 19237.00 -6.04 1822.5000 12.7000 417 6336200.0 552.01 -0.000040 17909.00 19237.00 -6.90 1822.5000 12.7000 418 6208000.0 550.98 0.000150 17420.00 18846.00 -7.57 1822.5000 12.7000 419 6246600.0 551.39 0.000290 17466.00 18846.00 -7.32 1822.5000 12.7000 420 6212100.0 551.03 0.000510 17365.00 18358.00 -5.41 1822.5000 12.7000 421 6170800.0 550.59 0.000790 17272.00 17577.00 -1.74 1822.5000 12.7000 422 6129300.0 550.14 0.000980 17156.00 17577.00 -2.40 1822.5000 12.7000 423 6674100.0 535.63 -0.002300 36818.00 30320.00 21.43 1822.5000 12.7000 424 6694800.0 536.03 -0.002300 36730.00 30320.00 21.14 1822.5000 12.7000 425 6681000.0 535.76 -0.002300 36809.00 30320.00 21.40 1822.5000 12.7000 426 6610900.0 535.53 -0.002200 36269.00 29588.00 22.58 1822.5000 12.7000 427 6556100.0 533.26 -0.002300 37468.00 28056.00 33.55 1822.5000 12.7000 428 6501700.0 533.36 -0.002200 36903.00 28074.00 31.45 1822.5000 12.7000 429 6453400.0 541.29 -0.001400 29367.00 27537.00 6.65 1822.5000 12.7000 430 6825800.0 557.31 0.000370 18522.00 20067.00 -7.70 1822.5000 12.7000 431 6874000.0 557.78 0.000770 18584.00 20067.00 -7.39 1822.5000 12.7000 432 6860200.0 557.65 0.001310 18480.00 19401.00 -4.75 1822.5000 12.7000 433 6832600.0 557.38 0.001620 18380.00 18944.00 -2.98 1822.5000 12.7000 434 6812000.0 557.17 0.001720 18351.00 18944.00 -3.13 1822.5000 12.7000 435 6784400.0 556.90 0.001780 18283.00 18554.00 -1.46 1822.5000 12.7000 436 6756000.0 556.62 0.001770 18240.00 18554.00 -1.69 1822.5000 12.7000 437 6708500.0 556.14 0.001970 18129.00 18554.00 -2.29 1822.5000 12.7000 438 6708500.0 556.14 0.002090 18114.00 18554.00 -2.37 1822.5000 12.7000 439 6749900.0 546.53 -0.00130D 27443.00 27879.00 -1.56 1822.5000 12.7000 440 6743000.0 544.24 -0.001550 29822.00 27879.00 6.97 1822.5000 12.7000 441 6729200.0 544.31 -0.00152D 29576.00 26707.00 10.74 1822.5000 12.7000 442 6687900.0 544.24 -0.001460 29156.00 25504.00 14.32 1822.5000 12.7000 443 6639600.0 544.35 -0.00137D 28452.00 25243.00 12.71 1822.5000 12.7000 444 6591300.0 544.66 -0.001260 27495.00 24559.00 21.95 1822.5000 12.7000 445 6550000.0 546.06 -0.001040 25391.00 24120.00 5.27 1822.5000 12.7000 446 6501700.0 510.83 -0.003840 50018.00 38474.00 30.00 1822.5000 12.7000 447 6007800.0 548.83 0.000290 17006.00 20116.00 -15.46 1822.5000 12.7000 448 6880900.0 557.85 0.001000 18555.00 20116.00 -7.76 1822.5000 12.7000 449 6860200.0 557.65 0.002400 18347.00 19335.00 -5.11 1822.5000 12.7000 450 6825800.0 557.31 0.002400 18274.00 17870.00 2.26 1822.5000 12.7000 451 6791300.0 556.97 0.002800 18173.00 17870.00 1.70 1822.5000 12.7000 452 6756800.0 556.62 0.002900 18092.00 17577.00 2.93 1822.5000 12.7000 453 6825500.0 557.31 0.002200 23446.00 28231.00 -16.95 235.0000 12.7000 454 6687900.0 555.94 0.003000 23055.00 26795.00 -13.96 235.0000 12.7000 455 6543100.0 554.47 0.003000 22716.00 26795.00 -15.22 235.0000 12.7000 456 6908400.0 558.12 0.004000 23476.00 26224.00 -10.48 235.0000 12.7000 457 6812000.0 557.17 0.004400 23221.00 26224.00 -11.45 235.0000 12.7000 458 6632700.0 555.38 0.004400 22805.00 26224.00 -13.04 235.0000 12.7000 459 6501700.0 554.05 0.004900 22477.00 25150.00 -10.63 235.0000 12.7000 460 6336200.0 552.33 0.004800 22091.00 26224.00 -15.76 235.0000 12.7000 461 6618900.0 543.78 -0.001400 41362.00 42722.00 -3.18 235.0000 12.7000 462 6584400.0 548.58 -0.000800 32712.00 36585.00 -10.59 235.0000 12.7000 463 6812000.0 543.93 -0.001700 44321.00 47390.00 -6.48 235.0000 12.7000 464 6377600.0 552.77 0.000004 22643.00 31150.00 -27.31 235.0000 12.7000 465 6867100.0 557.71 0.000005 23551.00 35642.00 -33.92 273.1000 12.7000 466 6770600.0 551.63 -0.000700 30221.00 35642.00 -15.21 273.1000 12.7000 467 6674100.0 552.06 -0.000500 27285.00 35642.00 -23.45 273.1000 12.7000 468 6632700.0 552.37 -0.000400 25931.00 33933.00 -23.58 273.1000 1.2.7000 469 6550000.0 554.54 0.000500 22773.00 31492.00 -27.69 273.1000 12.7000 470 6356900.0 552.55 0.001000 22284.00 27830.00 -19.93 273.1000 12.7000 471 6170800.0 550.59 0.001000 21850.00 27098.00 -19.37 273.1000 12.7000 472 5991500.0 548.65 0.002000 21320.00 24657.00 -13.53 273.1000 12.7000 473 6922300.0 558.25 0.002000 23467.00 26854.00 -12.61 273.1000 12.7000 474 6874000.0 557.78 0.002700 23288.00 25877.00 -10.01 273.1000 12.7000 475 6784400.0 556.90 0.003000 23078.00 25877.00 -10.82 273.1000 12.7000 476 6722300.0 556.28 0.003000 22934.00 26121.00 -12.20 273.1000 12.7000 477 6563800.0 554.68 0.003000 22571.00 25145.00 -10.24 273.1000 12.7000 478 6405200.0 553.05 0.003500 22172.00 24901.00 -10.96 273.1000 12.7000 479 6246600.0 551.39 0.003700 21763.00 24266.00 -10.31 273.1000 12.7000 480 6984300.0 558.86 0.004500 23379.00 25145.00 -7.02 273.1000 12.7000 481 6936100.0 558.39 0.004200 23303.00 26854.00 -13.22 273.1000 1.2.7000 482 6887800.0 557.92 0.004500 23161.00 25145.00 -7.89 273.1000 12.7000 483 6805100.0 557.10 0.004500 22993.00 25145.00 -8.56 273.1000 12.7000 484 6667200.0 555.73 0.004500 22674.00 25145.00 -9.83 273.1000 12.7000 485 6529300.0 554.33 0.004800 22330.00 24413.00 -8.53 273.1000 12.7000 486 6391400.0 552.91 0.004700 22021.00 23729.00 -7.20 273.1000 12.7000 487 6894700.0 557.98 0.00000 28581.00 45075.00 -36.59 4.7000 12.7000 488 6825800.0 557.31 0.003500 28342.00 43737.00 -35.20 4.7000 12.7000 489 6756800.0 556.62 0.004000 28101.00 42970.00 -34.60 4.7000 12.7000 490 6687900.0 555.94 0.004000 28014.00 42600.00 -34.24 4.7000 12.7000 491 6584400.0 554.89 0.004300 27828.00 41872.00 -33.54 4.7000 12.7000 492 6550000.0 554.54 0.004500 27729.00 40271.00 -31.14 4.7000 12.7000 493 6481000.0 553.84 0.004600 27583.00 40271.00 -31.51 4.7000 12.7000 494 6260400.0 551.54 0.004700 27248.00 39836.00 -31.60 4.7000 12.7000 495 6274200.0 551.68 0.004700 27299.00 39836.00 -31.47 4.7000 12.7000 496 6205200.0 550.95 0.004700 27149.00 39304.00 -30.93 4.7000 12.7000 497 6101800.0 549.85 0.004700 27003.00 38787.00 -30.38 4.7000 12.7000 o-:\4384-non\4394-13.wpd:lb-4303 13-72

WESTINGHOUSE PROPRIETARY CLASS 2 498 6607900.0 537.66 U.g.b2 000 48396.00 58492.00 .i.64.7000 12.700D 99 6628900.0 537.86 -0.002000 48378.00 58492.00 -17.29 4.7000 12.7000 500 58400.0 539.35 -0.001800 46701.00 5411.00 -13.82 4.7000 12.7000 501 6550000.0 544.62 -0.001200 41393.00 54586.00 -24.17 4.7000 12.7000 502 6377600.0 552.77 0.000004 29739.00 52267.00 -43.10 4.7000 12.7000 503 6791300.0 523.24 -0.003500 60194.00 57130.00 5.36 4.7000 12.7000 504 6687900.0 527.14 -0.003000 56859.00 56930.00 -0.12 4.7000 12.7000 505 6618900.0 528.20 -0.002800 55516.00 56930.00 -2.48 4.7000 12.7000 506 6584400.0 539.86 -0.001750 46285.00 55831.00 -17.10 4.7000 12.7000 507 6412100.0 539.61 -0.001500 44589.00 53795.00 -17.11 4.7000 12.7000 508 6308700.0 552.05 0.000004 29674.00 48971.00 -39.40 4.7000 12.7000 509 6825800.0 557.31 0.005000 27991.00 39621.00 -29.35 4.7000 12.7000 510 6805100.0 557.10 0.004500 28093.00 39836.00 -29.48 4.7000 12.7000 511 6756000.0 556.62 0.004500 28000.00 41404.00 -32.37 4.7000 12.7000 512 6722300.0 556.28 0.04500 27955.00 44397.00 -37.03 4.7000 12.7000 513 6618900.0 555.24 0.005100 27663.00 39206.00 -29.44 4.7000 12.7000 514 6570600.0 554.75 0.005400 27567.00 38689.00 -28.75 4.7000 12.7000 515 6515500.0 554.19 0.005500 27417.00 38689.00 -29.13 4.7000 12.7000 516 6412100.0 553.12 0.005600 27271.00 38181.00 -28.57 4.7000 12.7000 517 6343100.0 552.41 0.005700 27171.00 38181.00 -28.84 4.7000 12.7000 518 6294900.0 552.90 0.005700 27069.00 38181.00 -29.10 4.7000 12.7000 519 6205200.0 550.95 0.005900 26914.00 38186.00 -29.52 4.7000 12.7000 520 6088000.0 549.70 0.005900 26706.00 38181.00 -30.05 4.7000 12.7000 521 6067300.0 549.45 0.005900 26706.00 36668.00 -27.17 4.7000 12.7000 522 5963900.0 548.35 0.005900 26496.00 36668.00 -27.74 4.7000 12.7000 523 5860500.0 547.21 0.006000 26332.00 36668.00 -28.19 4.7000 12.7000 524 6563800.0 479.18 -0.005900 76335.00 61227.00 24.68 4.7000 12.7000 525 6460300.0 507.08 -0.004000 65607.00 59029.00 11.14 4.7000 12.7000 526 6412100.0 511.90 -0.003600 62893.00 59029.00 6.55 4.7000 12.7000 527 6274200.0 524.06 -0.002500 54862.00 57662.00 -4.86 4.7000 12.7000 528 6205200.0 536.12 -0.001500 45259.00 53600.00 -15.56 4.7000 12.7000 529 6053S00.0 541.54 -0.000800 38310.00 51359.00 -25.41 4.7000 12.7000 530 5963900.0 543.46 -0.000500 35060.00 49460.00 -29.11 4.7000 22.7000 532. 5791500.0 544.40 -0.000200 31482.00 47243.00 -33.36 4.7000 12.7000 532 5667400.0 545.03 0.000100 28532.00 45632.00 -37.47 4.7000 12.7000 533 6515500.0 554.19 0.004400 27683.00 42966.00 -35.57 4.7000 12.7000 534 6205200.0 550.95 0.005200 27032.00 40266.00 -32.87 4.7000 12.7000 535 4743600.0 533.76 0.001800 25720.00 39499.00 -34.88 4.7000 12.7000 536 4743600.0 533.76 0.001800 25720.00 39499.00 -34.88 4.7000 12.7000 537 4688400.0 533.03 0.001800 25640.00 39499.00 -35.09 4.7000 12.7000 538 4688400.0 533.0 0.002000 25542.00 38323.00 -33.35 4.7000 12.7000 539 4633200.0 532.30 0.002100 25380.00 37585.00 -32.47 4.7000 22.7000 540 4598800.0 531.84 0.002200 25270.00 37585.00 -32.77 4.7000 1.2.7000 541 4440200.0 529.658 0.002400 24856.00 36961.00 -32.75 4.7000 12.7000 542 4412600.0 529.30 0.002600 24730.00 36961.00 -33.09 4.7000 12.7000 543 4357500.0 528.53 0.002600 24577.00 34636.00 -29.04 4.7000 12.7000 544 4274700.0 527.37 0.002600 24448.00 33162.00 -26.28 4.7000 12.7000 545 6887800.0 557.92 0.001000 21915.00 24046.00 -8.86 322.2000 12.7000 546 6846400.0 557.51 0.002000 21765.00 24046.00 -9.49 322.2000 12.7000 547 6853300.0 557.58 0.002500 21722.00 23045.00 -5.74 322.2000 12.7000 548 6860200.0 557.65 0.003000 21708.00 22704.00 -4.39 322.2000 12.7000 549 6453400.0 553.55 0.003000 20839.00 22704.00 -8.21 322.2000 12.7000 550 6322400.0 552.19 0.003200 20540.00 22704.00 -9.53 322.2000 12.7000 551 6681000.0 555.87 0.002000 21410.00 22948.00 -6.70 322.2000 1.2.7000 552 6550000.0 554.54 0.002000 21122.00 22948.00 -7.96 322.2000 12.7000 553 6419000.0 553.20 0.002000 20860.00 22948.00 -9.10 322.2000 12.7000 554 6281100.0 551.76 0.002500 20520.00 22948.00 -10.58 322.2000 12.7000 555 6880900.0 557.85 0.003000 21756.00 22826.00 -4.69 322.2000 12.7000 556 6846400.0 557.51 0.003000 21685.00 22826.00 -5.00 322.2000 12.7000 557 6777500.0 556.83 0.0D3000 21514.00 22826.00 -5.75 322.2000 12.7000 558 6729200.0 556.35 0.00000 21419.00 22826.00 -6.16 322.2000 1.2.7000 559 6612000.0 555.17 0.003000 21179.00 22826.00 -7.22 322.2000 12.7000 560 6384500.0 552.84 0.00350D 20645.00 22826.00 -9.55 322.2000 1.2.7000 561 6687900.0 539.07 -0.002000 39671.00 40525.00 -2.11 322.2000 12.7000 562 6639600.0 542.22 -0.001600 35764.00 34666.00 3.17 322.2000 12.7000 563 6598200.0 542.53 -0.001500 34888.00 31248.00 11.65 322.2000 12.7000 564 6563800.0 546.60 -0.001000 29191.00 28563.00 2.20 322.2000 12.7000 565 6501700.0 545.77 -0.001000 29395.00 25633.00 14.68 322.2000 -12.7000 566 6350000.0 549.14 -0.000400 22952.00 23924.00 -4.06 322.2000 12.7000 567 6191400.0 548.62 -0.000250 21783.00 22948.00 -5.08 322.2000 12.7000 568 6026000.0 548.11 -0.000100 20670.00 22215.00 -6.95 322.2000 12.7000 569 6943000.0 558.46 0.002000 21140.00 22752.00 -7.09 512.7000 12.7000 570 6894700.0 557.98 0.002000 21047.00 22752.00 -7.49 512.7000 12.7000 571 6756800.0 556.62 0.002500 20730.00 22752.00 -8.89 512.7000 12.7000 572 6610900.0 555.16 0.002500 20417.00 22069.00 -7.49 512.7000 12.7000 573 6401000.0 553.01 0.003000 19942.00 22069.00 -9.64 512.7000 12.7000 574 6343100.0 552.41 0.003000 19819.00 22069.00 -10.20 512.7000 1.2.7000 575 6904300.0 558.08 0.002500 21012.00 22362.00 -6.04 512.7000 1.2.7000 576 6943000.0 558.46 0.002500 21109.00 22362.00 -5.60 512.7000 12.7000 577 6825800.0 557.31 0.003500 20781.00 22362.00 -7.07 512.7000 12.7000 578 6694800.0 556.00 0.003500 20516.00 22362.00 -8.26 512.7000 12.7000 579 6570600.0 554.75 0.004000 2021.2.00 22362.00 -9.61 512.7000 12.7000 580 6439600.0 553.41 0.004000 19940.00 22069.00 -9.65 512.7000 12.7000 581 6556900.0 530.77 -0.002500 43824.00 41013.00 6.85 512.7000 12.7000 582 6453400.0 534.80 -0.002000 39380.00 39792.00 -1.04 512.7000 12.7000 583 6343100.0 538.48 -0.001500 34569.00 31736.00 8.93 512.7000 1.2.7000 584 6301800.0 537.78 -0.001500 34738.00 29539.00 17.60 512.7000 12.7000 585 6232800.0 542.02 -0.001000 29031.00 25487.00 13.91 512.7000 12.7000 586 6060400.0 545.67 -0.000400 21937.00 21971.00 -0.15 512.7000 12.7000 587 6867100.0 557.71 0.001000 21070.00 23045.00 -8.57 512.7000 12.7000 588 6825800.0 557.31 0.001500 20940.00 23045.00 -9.13 512.7000 12.7000 589 6687900.0 555.94 0.00200D 20622.00 23045.00 -10.51 512.7000 12.7000 590 6543100.0 554.47 0.00250D 20271.00 21776.00 -6.91 512.7000 12.7000 591 6405200.0 553.05 0.00250D 20007.00 21776.00 -8.12 512.7000 12.7000 592 6253500.0 551.47 0.00300D 19620.00 21776.00 -9.90 512.7000 12.7000 593 6556900.0 536.74 -0.002000 37892.00 38816.00 -2.38 639.7000 12.7000 594 6460300.0 534.93 -0.002000 38387.00 35642.00 7.70 639.7000 12.7000 595 6288000.0 542.81 -0.001000 28107.00 23558.00 19.31 639.7000 12.7000 596 6239700.0 546.01 -0.000600 23367.00 22411.00 4.27 639.7000 12.7000 597 5977700.0 543.64 -0.000500 22431.00 21117.00 6.22 639.7000 12.7000 598 6901600.0 558.05 0.003000 20515.00 21776.00 -5.79 639.7000 12.7000 oA\4384-non\4384-13.wpd:Ib4303 13-73

WESTINGHOUSE PROPRIETARY CLASS 2 599 6687900.0 555.94 0.003000 20084.00 21776.00 -7.77 639.7000 12.7000 600 6474100.0 553.76 0.004000 19555.00 21776.00 -10.20 639.7000 12.7000 601 6929200.0 558.32 0.000509 20786.00 21117.00 -1.57 639.7000 12.7000 602 6887800.0 557.92 0.001000 20661.00 21117.00 -2.16 639.7000 12.7000 603 6632700.0 555.38 0.002000 20052.00 21117.00 -5.04 639.7000 12.7000 604 6377600.0 552.77 0.002000 19549.00 21117.00 -7.43 639.7000 12.7000 605 6901600.0 558.05 0.002000 20598.00 21776.00 -5.41 639.7000 12.7000 606 6887800.0 557.92 0.002000 20575.00 21776.00 -5.52 639.7000 12.7000 607 6763700.0 556.69 0.002000 20313.00 21776.00 -6.72 639.7000 12.7000 608 6632700.0 555.38 0.002000 20052.00 21776.00 -7.92 639.7000 12.7000 609 6384500.0 552.84 0.002000 19547.00 21483.00 -9.01 639.7000 12.7000 610 6867100.0 557.71 0.001500 22476.00 26121.00 -13.95 195.2000 12.7000 611 6818900.0 557.24 0.002000 22336.00 25389.00 -12.02 195.2000 12.7000 612 6736100.0 556.42 0.002500 22125.00 24657.00 -10.27 195.2000 12.7000 613 6681000.0 555.87 0.002700 21975.00 24657.00 -10.88 195.2000 12.7000 614 6543100.0 554.47 0.002800 21672.00 24657.00 -12.11 195.2000 12.7000 615 6398300.0 552.98 0.003000 21351.00 24168.00 -11.66 195.2000 12.7000 616 6253500.0 551.47 0.003000 21031.00 23436.00 -10.26 195.2000 12.7000 617 6846400.0 557.51 0.004000 22253.00 24901.00 -10.63 195.2000 12.7000 618 6812000.0 557.17 0.003000 22245.00 24901.00 -10.67 195.2000 12.7000 619 6694800.0 556.00 0.004000 21913.00 24412.00 -10.24 195.2000 12.7000 620 6577500.0 554.82 0.004000 21669.00 24412.00 -11.24 195.2000 12.7000 621 6460300.0 553.62 0.004000 21424.00 23436.00 -8.59 195.2000 12.7000 622 6336200.0 552.33 0.004500 21099.00 23436.00 -9.97 195.2000 12.7000 623 6853300.0 546.27 -0.001500 34904.00 42722.00 -18.30 195.2000 12.7000 624 6784400.0 545.29 -0.001500 35161.00 40036.00 -12.18 195.2000 12.7000 625 6729200.0 546.24 -0.001300 33156.00 38816.00 -14.58 195.2000 12.7000 626 6667200.0 547.11 -0.001100 31106.00 36619.00 -15.06 195.2000 12.7000 627 6612000.0 550.56 -0.000600 25374.00 34861.00 -27.21 195.2000 12.7000 628 6570600.0 550.88 -0.000500 24491.00 31346.00 -21.87 195.2000 12.7000 629 6529300.0 552.00 -0.000300 23204.00 29539.00 -21.45 195.2000 12.7000 630 6494800.0 553.98 0.000004 21811.00 29100.00 -25.05 195.2000 12.7000 631 6136300.0 550.22 0.000700 20965.00 24022.00 -12.73 195.2000 12.7000 632 5950100.0 548.20 0.000900 20515.00 23094.00 -11.17 195.2000 12.7000 633 6625800.0 555.31 0.006100 24554.00 32469.00 -24.38 44.5000 19.0000 634 6439600.0 553.41 0.006700 24107.00 32469.00 -25.75 44.5000 19.0000 635 6343100.0 552.41 0.007100 23871.00 30760.00 -22.40 44.5000 19.0000 636 6253500.0 551.47 0.007600 23636.00 30760.00 -23.16 44.5000 19.0000 637 6157000.0 550.44 0.007600 23423.00 30760.00 -23.85 44.5000 19.0000 638 5963900.0 548.35 0.007900 22977.00 30760.00 -25.30 44.5000 19.0000 639 5757100.0 546.05 0.007700 22526.00 30760.00 -26.77 44.5000 19.0000 640 6632700.0 555.38 0.006300 24541.00 33933.00 -27.68 44.5000 19.0000 641 6550000.0 554.54 0.006600 24357.00 32469.00 -24.98 44.5000 19.0000 642 6356900.0 552.55 0.008300 23787.00 32469.00 -26.74 44.5000 19.0000 643 6157000.0 550.44 0.009000 23290.00 29783.00 -21.80 44.5000 19.0000 644 6060400.0 549.40 0.009200 23053.00 29783.00 -22.60 44.5000 19.0000 645 5957000.0 548.27 0.009900 22759.00 29783.00 -23.58 44.5000 19.0000 646 5853600.0 547.13 0.009000 22618.00 29783.00 -24.06 44.5000 19.0000 647 6577500.0 552.54 -0.000300 29542.00 51266.00 -42.38 44.5000 19.0000 648 6419000.0 553.20 0.001500 24698.00 46628.00 -47.03 44.5000 19.0000 649 6288000.0 551.83 0.003000 24203.00 36619.00 -33.91 44.5000 19.0000 650 6157000.0 550.44 0.004000 23813.00 36619.00 -34.97 44.5000 19.0000 651 6032900.0 549.10 0.004400 23498.00 33933.00 -30.75 44.5000 19.0000 652 5929400.0 547.97 0.005000 23207.00 33933.00 -31.61 44.5000 19.0000 653 5819100.0 546.75 0.006900 22766.00 30760.00 -25.99 44.5000 19.0000 654 5708800.0 545.51 0.008700 22318.00 29295.00 -23.82 44.5000 19.0000 655 5590500.0 544.15 0.008700 22057.00 29295.00 -24.71 44.5000 19.0000 656 6894700.0 557.98 0.005000 23771.00 27781.00 -14.43 732.0000 54.0000 657 6584400.0 554.89 0.006100 22948.00 27781.00 -17.40 732.0000 54.0000 658 6446500.0 553.48 0.006200 22601.00 27781.00 -18.65 732.0000 54.0000 659 6260400.0 551.54 0.006450 22133.00 27781.00 -20.33 732.0000 54.0000 660 6722300.0 556.28 0.008250 23472.00 24803.00 -5.37 696.0000 76.2000 661 6377600.0 552.77 0.009250 22569.00 24803.00 -9.01 696.0000 76.2000 662 6101800.0 549.85 0.009900 21844.00 24803.00 -11.93 696.0000 76.2000 663 6929500.0 552.01 -0.000900 36884.00 43757.00 -15.71 63.5000 28.0000 664 6232800.0 545.46 -0.000650 34857.00 43923.00 -20.64 63.5000 28.0000 665 6812000.0 557.17 0.000005 25274.00 40671.00 -37.86 63.5000 28.0000 666 6798200.0 557.03 0.003000 24859.00 32298.00 -23.03 63.5000 28.0000 667 6770600.0 556.76 0.003000 24814.00 31932.00 -22.29 63.5000 28.0000 For 667 Data Points, Average Zrror -14.71 %,STD - 20.32 For 212 Subcooled Data Points, Average Error - 0.54 t, S 21.06 t For 455 Saturated Data Points, Average Error . -21.81 %,S - 15.51 % o:4384-non'4384-13.wpd:lb-4303 13-74

WESTINGHOUSE PROPRIETARY CLASS 2 WCOBRA!1RAC Model Prediction vs. Sozzi-Sutherland Data Mlean Error is -i6% Standard Deviation is 20.3%

• U WCT 4 0 0 PREDICTION 8000

                                -/     t.          t.

• U 40W-:-.

2o D - 4=^ NM x o 1 Measured Mss Flow Flux (kg/rn2-s) Figure 13 4-26 Prediction Comparison with Sozzi-Sutherland Data o\4384-non\4384-13.wpd:1b4303 13-75

I WESTINGHOUSE PROPRIETARY CLASS 2 Marviken Tests 6,7,23 and 24 Run No. Pressure TeDerature Quality Predicted G Measured Ge Error L D (Gp-m) /Gm (P&) (K) (Xglm2-s) (Kg/m2-s (in % ) (no) (mm) 1 4664000.0 503.45 -0.001469 56961.00 57833.00 -1.51 300.0000 300.0000 2 4847000.0 502.85 -0.001706 60041.00 59828.00 0.36 300.0000 300.0000 3 4830000.0 502.85 -0.001686 59808.00 60451.00 -1.06 300.0000 300.0000 4 5009000.0 503.45 -0.001878 61843.00 59800.00 3.42 300.0000 300.0000 5 4907000.0 503.55 -0.001746 60353.00 59107.00 2.11 300.0000 300.0000 6 4938000.0 503.15 -0.001803 61095.00 59955.00 1.90 300.0000 300.0000 7 4822000.0 503.65 -0.001640 59116.00 58498.00 1.06 300.0000 300.0000 8 4889000.0 503.85 -0.001711 59895.00 58442.00 2.49 300.0000 300.0000 9 4800000.0 503.95 -0.001601 58510.00 58852.00 -0.58 300.0000 300.0000 10 4847000.0 503.95 -0.001656 59244.00 57749.00 2.59 30D.0000 300.0000 11 4742000.0 504.55 -0.001507 57221.00 57324.00 -0.18 300.0000 300.0000 12 4772000.0 504.35 -0.001550 57799.00 56306.00 2.65 300.0000 300.0000 13 4683000.0 505.45 -0.001401 55599.00 56037.00 -0.78 300.0000 300.0000 14 4654000.0 505.55 -0.001365 55106.00 57409.00 -4.01 300.0000 300.0000 15 4702000.0 505.65 -0.001413 55759.00 56829.00 -1.88 300.0000 300.0000 16 4689000.0 506.35 -0.001367 54938.00 55103.00 -0.30 300.0000 300.0000 17 4677000.0 506.45 -0.001349 54672.00 54410.00 0.48 300.0000 300.0000 18 4582000.0 506.45 -0.001248 53185.00 54382.00 -2.20 300.0000 300.0000 19 4540000.0 506.85 -0.001188 52172.00 54551.00 -4.36 300.0000 300.0000 20 4699000.0 507.15 -0.001342 54433.00 54268.00 0.30 300.0000 300.0000 21 4526000.0 506.75 -0.001178 52037.00 54127.00 -3.86 300.0000 300.0000 22 4487000.0 507.15 -0.001122 51049.00 52061.00 -1.94 300.0000 300.0000 23 4409000.0 507.55 -0.001028 49424.OD 54325.00 -9.02 300.0000 300.0000 24 4507000.0 508.25 -0.001094 50410.00 52514.00 -4.01 300.0000 300.0000 25 4446000.0 508.45 -0.001026 49203.00 52415.00 -6.13 300.0000 300.0000 26 4460000.0 509.25 -0.001004 48617.00 52557.00 -7.50 300.0000 300.0000 27 4326000.0 511.45 -0.000787 43978.00 52458.00 -16.17 300.0000 300.0000 28 4385000.0 515.65 -0.000658 40200.00 50449.00 -20.32 300.0000 300.0000 29 4386000.0 516.55 -0.000620 39082.00 49458.00 -20.98 300.0000 300.0000 30 4472000.0 517.25 -0.000663 40022.00 47690.00 -16.08 300.0000 300.0000 31 4407000.0 517.65 -0.000589 38102.00 46332.00 -17.76 300.0000 300.0000 32 4405000.0 519.35 -0.000512 35688.00 45780.00 -22.04 300.0000 300.0000 33 4184000.0 519.65 -0.000326 29691.00 44733.00 -33.63 300.0000 300.0000 34 4260000.0 520.35 -0.000353 30567.00 44323.00 -31.04 300.0000 300.0000 35 4364000.0 516.95 -0.000584 38132.00 46714.00 -18.37 300.0000 300.0000 36 4383000.0 517.55 -0.000573 37708.00 44521.00 -1S.30 300.0000 300.0000 37 4213000.0 518.25 -0.000407 32753.00 44677.00 -26.69 300.0000 300.0000 38 4175000.0 518.85 -0.000353 30789.00 44097.00 -30.18 300.0000 300.0000 39 4360000.0 517.75 -0.000545 36947.00 40093.00 -7.85 300.0000 300.0000 40 4071000.0 518.95 -0.000266 27752.00 32850.00 -15.52 300.0000 300.0000 41 4225000.0 518.95 -0.000386 31954.00 29964.00 6.64 300.0000 300.0000 42 4208000.0 520.35 -0.000314 29110.00 32411.00 -10.18 300.0000 300.0000 43 4181000.0 519.25 -0.000341 30274.00 30063.00 0.70 300.0000 300.0000 44 4102000.0 520.05 -0.000247 26644.00 31859.00 -16.37 300.0000 300.0000 45 4230000.0 520.35 -0.000330 29707.00 31746.00 -6.42 300.0000 300.0000 46 4180000.0 520.45 -0.000289 28174.00 30331.00 -7.11 300.0000 300.0000 47 4183000.0 523.05 -0.000156 23098.00 29737.00 -22.33 300.0000 300.0000 48 4175000.0 523.05 -0.000151 22819.00 29384.00 -22.34 300.0000 300.0000 49 4109000.0 522.85 -0.000115 21058.00 29186.00 -27.85 300.0000 300.0000 50 4064000.0 523.35 -0.000052 18696.00 26427.00 -29.25 300.0000 300.0000 51 4077000.0 522.65 -0.000099 20419.00 25309.00 -19.32 300.0000 300.0000 52 4068000.0 522.95 -0.000076 19509.00 25337.00 -23.00 300.0000 300.0000 53 4132000.0 522.65 -0.000143 22263.00 25295.00 -11.99 300.0000 300.0000 54 4116000.0 522.75 -0.000127 21506.00 24856.00 -13.48 300.0000 300.0000 55 4097000.0 522.45 -0.000127 21525.00 25295.00 -14.90 300.0000 300.0000 56 4028000.0 522.35 -0.000075 19436.00 23626.00 -17.73 300.0000 300.0000 57 4066000.0 522.35 -0.000106 20685.00 24319.00 -14.94 300.0000 300.0000 58 4042000.0 521.95 -0.000108 20753.00 24956.00 -16.84 300.0000 300.0000 59 4042000.0 521.55 -0.000129 21615.00 24022.00 -10.02 300.0000 300.0000 60 4039000.0 521.75 -0.000116 21063.00 24234.00 -13.08 300.0000 300.0000 61 4008000.0 521.55 -0.000102 20457.00 24687.00 -17.13 300.0000 300.0000 62 3994000.0 521.05 -0.000116 21058.00 24404.00 -13.71 300.0000 300.0000 63 3955000.0 521.05 -0.000084 19726.00 23696.00 -16.75 300.0000 300.0000 64 3967000.0 520.95 -0.000098 20346.00 23484.00 -13.36 300.0000 300.0000 65 3956000.0 520.65 -0.000101 20606.00 22876.00 -9.92 300.0000 300.0000 66 3954000.0 521.15 -0.000079 19497.00 23017.00 -15.29 300.0000 300.0000 67 3967000.0 520.95 -0.000098 20346.00 22621.00 -10.06 300.0000 300.0000 68 3954000.0 520.35 -0.000111 21197.00 23230.00 -8.75 300.0000 300.0000 69 3954000.0 520.35 -0.000111 21197.00 23215.00 -8.69 300.0000 300.0000 70 3910000.0 520.35 -0.000076 19689.00 22494.00 -12.47 300.0000 300.0000 71 3895000.0 519.75 -0.000088 20446.00 22692.00 -9.90 300.0000 300.0000 72 3898000.0 519.95 -0.000082 20115.00 21772.00 -7.61 30D.0000 300.0000 73 3890000.0 519.75 -0.000084 20267.00 21051.00 -3.72 30D.0000 300.0000 74 3889000.0 519.75 -0.000083 20236.00 22466.00 -9.93 30D.0000 300.OOOD 75 3883000.0 519.45 -0.000090 20674.00 22338.00 -7.45 30D.0000 300.0000 76 3847000.0 519.65 -0.000054 19026.00 21574.00 -11.81 300.0000 300.0000 77 3837000.0 519.45 -0.000056 19090.00 21985.00 -13.17 300.0000 300.0000 78 3826000.0 519.25 -0.000057 19123.00 21136.00 -9.52 300.0000 300.0000 79 3830000.0 519.05 -0.000067 19690.00 21546.00 -8.61 300.0000 300.0000 80 3827000.0 518.35 -0.000092 21081.00 21192.00 -0.52 300.0000 300.0000 81 3790000.0 518.45 -0.000065 19549.00 21376.00 -8.55 300.0000 300.0000 82 3809000.0 518.65 -0.000069 19812.00 20952.00 -5.44 300.0000 300.0000 83 3780000.0 518.15 -0.000071 19861.00 22169.00 -10.41 300.0000 300.0000 84 3765000.0 518.15 -0.000061 19329.00 21192.00 -8.79 300.0000 300.0000 85 3737000.0 517.85 -0.000056 18966.00 21150.00 -10.33 300.0000 300.0000 86 4630000.0 517.55 -0.000794 42779.00 52698.00 -18.82 300.0000 300.0000 87 4934000.0 516.45 -0.001158 49473.00 53632.00 -7.75 300.0000 300.0000 88 5014000.0 515.65 -0.001288 51581.00 54141.00 -4.73 300.0000 300.0000 89 4980000.0 517.45 -0.001159 49188.00 54127.00 -9.12 300.0000 300.0000 90 4827000.0 517.45 -0.000995 46554.00 56122.00 -17.05 300.0000 300.0000 91 4926000.0 516.45 -0.001149 49307.00 54622.00 -9.73 300.0000 300.0000 92 4868000.0 517.45 -0.001038 47252.00 53080.00 -10.98 300.0000 300.0000 93 4888000.0 517.15 -0.001074 47935.00 52401.00 -8.52 300.0000 300.0000 o:\4384-non\4384-13.wpd:1b-4303 13-76

WESTINGHOUSE PROPRIETARY CLASS 2 94 4951000.0 517.45 0O.001127 48665.00 52443.00 -,7.'20 300.0000 300.0000 95 4922000.0 517.85 -0 .001075 47734.00 5282S.00 -9 .64 300.0000 300.0000 96 4767000.0 518.05 -0 .000904 4473 8.00 51255.00 -12 .71 300.0000 300.0000 97 4775000.0 517.95 -0 .000917 45029.00 51255.00 -12 .15 300.0000 300.0000 98 4847000.0 518.35 -0 .000972 45903.00 50477.00 -9 .06 300.0000 300.0000 99 4653000.0 519.35 -0.000732 41005 .00 51326.00 -20 .11 300.0000 300.0000 100 4768000.0 518.65 -0.000877 44085.00 49897.00 -11 .65 300.0000 300.0000 101 4702000.0 519.05 -0.000793 42342.00 49501.00 -14 .46 300.0000 30D.0000 102 4637000.0 519.15 -0.000726 40946.00 50519.00 -18 .95 300.0000 300.0000 103 4647000.0 519.25 -0.000731 40984.00 49472.00 -17 .6 300.OOOD 300.0000 104 4613000.0 519.25 -0.000699 40288.00 50562.00 -20 .32 300.0000 300.0000 105 4633000.0 519.35 -0.000713 40612.00 49600 .00 -18 .12 300.OOOD 300.0000 106 4637000.0 519.85 -0.000693 3 9992.00 48454 .00 -17 .46 300.0000 300.0000 107 4633000.0 519.95 -0.000685 3 9787.00 49331. 00 -19 .35 300.0000 300.0000 108 4596000.0 520.15 -0.000642 38769.00 49034.00 -20 .93 300.000D 300.0000 109 4437000.0 520.25 -0.000498 3 5117.00 47605.00 -26 .23 300.0000 30D.OOOO 110 4558000.0 520.15 -0.000607 37946.00 47251.00 -19 .69 300.0000 300.0000 Ill 4600000.0 520.55 -0.000626 38280.00 47379.00 -19 .20 300.0000 300.0000 112 4623000,0 520.85 -0.000633 38358.00 46233 .00 -17 .03 300.0000 300.0000 213 4427000.0 520.85 -0.000463 33921.00 47251.00 -28 .21 300.0000 300.0000 214 4429000.0 521.75 -0 .000415 32572.00 47605.00 -31 .58 300.0000 300.0000 115 4532000.0 522.15 -0.000477 34417.00 45426.00 -24 .24 300.0000 300.0000 126 4482000,0 523.15 -0.000375 31550.00 44365.00 -28 .89 300.0000 300.0000 117 4509000.0 523.85 -0.000356 31068.00 45271.00 -31 .37 300.0000 300.0000 118 4445000.0 523.95 -0.000299 29190.00 44648.00 -34 .62 300.0000 300.0000 119 4290000.0 524.25 -0 .000165 23974.00 42654.00 -43 .79 300.0000 300.0000 120 4417000.0 524.45 -D.000248 27456.00 41225 .00 -33 .40 300.0000 300.0000 121 4339000.0 524.65 -0D.000178 24702.00 42003.00 -41 .19 300.0000 300.0000 122 4541000.0 525.35 -0D.000295 29218.00 37065.00 -21 .17 300.0000 300.0000 223 4424000.0 525.25 -0 .000207 26058.00 33034.00 -21 .12 300.0000 300.0000 124 4369000.0 525.25 -0 .000166 24352.00 27941.00 -12 .84 300.ODOO 300.0000 125 4388000.0 524 .95 -0 .000196 25586.00 26993 .00 -5 .21 300.0000 300.0000 126 4344000.0 525.05 -0.000158 24000.00 29171.00 -17 .73 300.0000 300.0000 127 4346000.0 524 .95 -0.000165 24271.00 30855.00 -212.34 300.ODOO 300.0000 228 4400000.0 525.05 -0.000200 25733.00 29638.00 -13 .8 300.ODOO 300.0000 129 4362000.0 525.25 -0 .000161 24128.00 29865.00 -19 .21 300.0000 300.0000 130 4322000.0 526.05 -0.000095 21124.00 29086.00 -27 .37 300.0000 300.0000 131 4304000.D 525.85 -0.000092 20971.00 28549.00 -26 .54 300.0000 300.0000 132 4307000.0 525.85 -0.000094 21065.00 26681.00 -21 .05 300.0000 300.0000 133 4305000.0 525.75 -0 .000097 21237.00 26144.00 -18 .77 300.0000 300.0000 234 4294000.0 525.85 -0.000084 20658.00 25790.00 -19 .90 300.0000 300.0000 135 4273000.0 525.85 -0.000070 20032 .00 25408.00 -21 .16 300.0000 300.0000 136 4267000,0 525.05 -0.000103 21SI2.00 25960.00 -17 .13 300.0000 300.0000 137 4221000.0 525.05 -0.000071 20050.00 24757.00 -19 .01 300.0000 300.0000 138 4246000.0 524.75 -0 .000106 21471.00 24927.00 -13 .86 300.0000 300.0000 13 9 423 1000 .0 524.65 -0.00D01 21208 .00 24927.00 -14 .92 300.0000 300.0000 14D 4234000.0 524.35 -0 .000120 21946 .00 24475.00 -10 .33 300.0000 300.0000 241 4219000.0 524.75 -0 .000087 20601.00 24319.00 -15 .29 300.0000 300.0000 142 4243000.0 524.55 -0.00DI15 21817.00 24687.00 -11 .63 300.0000 300.0000 143 4208000.0 524.75 -0 .000080 20259 .00 24291.00 -16 .60 300.0000 300.0000 144 4212000.0 523 .95 -0 .000127 22078.00 23838.00 _7 .3 8 300.0000 300.0000 145 4154000.0 524.05 -0 .000082 19986.00 24050.00 -16 .90 300.0000 300.0000 24 6 4161000.0 524.05 -0.000087 20202.00 23074.00 -12 .45 300.0000 300.0000 147 4158000.0 523.65 -0 .000107 20951.00 23513.00 -10 .90 300.0000 300.0000 148 4263000.0 523.65 -0 .000110 21106.00 23484.00 -10 .13 300.0000 300.0000 149 4150000.0 523.75 -0 .000096 20471.0 0 23583.00 -13 .2 0 300.0000 300.0000 150 4142000.e 523.85 -0.000085 20021.OD 23696.00 -15 .51 300.0000 300.0000 151 4119000.0 523.25 -0 .000102 20528 .0 0 23668.00 -313 .27 300.0000 30O.C00 152 4139000.0 523.45 -0 .000105 20769.00 23626.00 -12 .09 30D.0000 300.000D 153 4092000.0 523.35 -0.000074 19467.00 23795.00 -18 .19 300.0000 300.0000 154 4092000.0 523.05 -0.000090 20072.00 22593.00 -11 .16 300.0000 300.0000 155 4065000.0 523.05 -0.000068 19245.00 22451.00 -14 .28 300.0000 30O.OOD0 156 4103000.0 522.85 -0.000110 20844.00 21475.00 -2 .94 300.0000 300.0000 157 4022000.0 522.95 -0 .000039 18276.00 23456.00 -22.08 300.0000 300.0000 158 4056000.0 522.75 -0 .000077 19545.00 23046.00 -15 .19 300.O000 300.0000 159 4042000.0 522.35 -0 .000087 19892.00 23385.00 -14 .94 300.0000 300.0000 160 4050000.0 521.95 -0 .000115 20999.00 23187.00 -9.44 300.0000 300.0000 161 4016000,0 521.95 -0 .000087 19871.00 22947.00 _13 .40 300.OOOD 300.0000 162 4022000,0 521.55 -0.000113 20915.00 22352.00 -6 .43 300.0000 300.0000 263 3989000.0 521.95 -0 .000065 19021.00 21956.00 -13 .37 300.0000 300.0000 264 3989000.0 521.65 -0 .000081 19620.00 21589.00 -9.12 300.0000 300.0000 26S 3973000.0 521.35 -0 .000084 19704.00 223 95.00 -12 .02 300.0000 300.0000 166 3965000.0 521.05 -0.000092 20058.00 22239 .0 -9 .81 300.0000 300.0000 167 3935000.0 520.55 -0 .000088 20109.00 22635.00 -11 .6 300.0000 300.0000 268 3949000.0 519.95 -0.000123 21895.00° 21857 .00 0.17 300.0000 300.0000 169 3859000.0 520.25 -0 .000039 18265.00 19268 .00 -5 .21 300.0000 300.0000 270 4693000.0 526.25 -0 .000379 32227 .00 28948.00 11.33 150.0000 500.0000 271 4926000.0 527.15 -0 .000547 36124.00 30986.00 16.58 150.0000 500.0000 172 4872000.0 530.45 -0 .000311 292 87.00 35085.00 -16 .3 150.OOOD 500.0000 173 4964000.0 530.75 -0.000379 31058.00 33639.00 -7 .67 150.0000 500.0000 174 4731000.0 531.45 -0.000129 23345.00 30670.00 -23 .8 8 150.0000 500.0000 175 500500D00 531.85 -0.000351 30093.00 29081.00 3.48 150.0000 500.0000 176 4503000.0 531.85 0.000003 16860.00 26107.00 _35 .42 150.0000 SOD.O000 177 4953000.0 531.65 -0.000314 29132.00 24675.00 18.06 150.0000 500.0000 178 4832000.0 531.65 -D.000204 25830.00 24584.00 5.07 150.0000 500.0000 179 4850000.0 531.65 -0.000220 26334.00 24080.00 9.36 150.0000 500.0000 ISO 4794000.0 531.65 -0.000171 24731.00 24706.00 0.10 150.0000 500.0000 181 4661000.0 531.35 -0 .000077 21597.00 25164.00 -14 .18 150.0000 500.0000 182 4703000.0 531.05 -0 .000129 23382.00 24523 .00 -4 .65 150.0000 500.0000 183 4603000.0 530.65 -0.000071 21340.00 24706.00 - 13 .62 150.0000 500.0000 184 4519000.0 530.65 -0.00D006 19440.00 23871.00 -18 .56 150 .000 5 00.00 00 185 4704000.0 530.55 -0.000159 24454.00 24375.00 0.32 150.0000 500.0000 186 4670000.0 530.15 -0.000153 24292.00 25460.00 -4 .59 150.0000 500.0000 187 4704000.0 530.05 -0.D00187 25465.00 24278.00 4.89 150.0000 500.0000 188 4693000.0 529.75 -0 .000194 25777 .00 23779.00 8 .40 150.0000 500.0000 189 4541000.0 529.55 -0.000083 21736.00 23723.00 -8 .3 8 150.0000 500.0000 190 4664000.0 529.35 -0 .000192 25744.00 23703.00 8.61 150.0000 500.0000 191 4720000.0 529.35 -0.000231 27044.00 23448.00 15.34 150.0000 500.0000 192 4532000.0 529.05 -0 .000102 22499.00 22898.00 -1 .74 150.0000 500.0000 193 4578000.0 528.85 -0 .000149 24281.00 23591.00 2.92 150.0000 500 .00D00 194 4541000.0 528.65 -0.000130 23600.00 23570.00 0.13 150.0000 500.0000 o:\4384-nonX4384-13.wpd:l1W303 13-77

WESTINGHOUSE PROPRIETARY CLASS 2 195 4541000.0 528.25 -O.OOOlSl 24413.00 23540.00 3.71 150.0000 500.0000 196 4358000.0 528, OS -0 .000025 19580.00 23815.00 -17 .78 150.0D00 50D.0000 197 4419000.0 527. 9 -0 .000074 21382.00 23901.00 -10 .54 150.0000 50D.0000 198 4430000.0 527.75 -0.000092 22116.00 23489.00 -S .85 150.0000 50D.0000 199 4508000.0 527.3S -0.000171 25274.00 23596.00 7.11 150.0000 500.0000 200 4343000.0 526.8S -0.000072 21350.00 24288.00 -12 .10 150.0000 5 00.00 00 201 4325000.0 526.25 -0 .000088 22020.00 22241.00 -O0.99 15 0.000 0 50 0.000 0 202 4249D00.0 525.75 -0.000058 2072<.00 23204.00 -10 .69 150.0000 500.0000 203 4292000.0 525.25 -0 .000110 23067.00 23041.00 O .11 lS0 .0000 500.0000 204 4284000.0 524.85 -0.000126 23639.00 22322.00 5.90 150.0000 500.0000 205 4139000.0 523. 9 -0 .000078 20947.00 21635. 00 -3 .18 150.0000 500.0000 206 4139000.0 523.55 -0.000099 21759.00 22384.00 -2 .79 150.0000 S00 .0000 207 4142000.0 523. 1 -0 .000123 22666.00 22710.00 -0.19 150.0000 500.0000 208 4003000.0 522.0OS -0 .000071 20479 .0 21762.00 -5 .90 250.0000 500,0000 209 4046000.0 521.35S -0 .000143 23299.00 23056.00 1. 05 150.0000 500.0000 210 4003000.0 520.85 -0 .000131 22942.00 22414.00 2.36 150.0000 500,0000 211 3845000.0 519. SS -0 .000041 19591.00 22221.00 -11. 84 150.0000 S00 .0000 212 3881000.0 519.2 5 -0 .000096 22199.00 21217.00 4. 63 150.0000 S00.0000 213 3842000.0 518.35 -0.0001D1 22725.00 20418.00 11.30 150.000 0 500.0000 214 4157000.0 502.05 -0.001019 50025.00 56491.00 -11.45 150.000 0 500.000 0 21S 4273000.0 502.35 -0.001114 51658.00 55798.00 -7.42 150.000 0 500.0000 216 3932000.0 502.5SS -0 .000776 45682.00 53639.00 -14 .83 150.0000 500.0000 217 4256000.0 503.65 -0.001046 50264.00 5273 8.00 -4 .69 15 0.0000 500.0000 218 4124000.0 504.25 -0 .000904 47554.00 52850.00 -10 .02 15 0.000 0 500.0000 219 3938000.0 504.25 -0 .000719 44172.00 5lS10 .00 -14 .25 15 0.0000 500.0000 220 3871000.0 504.55 -0 .000644 42639.00 49717.00 -14 .24 15D0.00 00 500.0000 221 3905000.0 504, SS -0 .000676 43282.00 47691.00 -9.24 15 0.00 00 500. 0000 222 3 829000.Q 504.65 -0 .000605 41722.00 47023.00 -11 .27 150.0000 500.0000 223 3495000.0 504.85 -0 .000357 3 4295.OD 45587.00 -24 .77 15 0.00 00 5 00.0000 224 3 632000.0 504.65 -0 .000460 37609.OD 45944.00 -18 .14 150.0000 500.0000 225 3537000.0 504,75 -0.000389 35373.00 44360.00 -20 .26 150.0000 50D.0000 226 3311000.0 504.85 -0 .000237 29554.00 42664.00 -30.73 150.0000 500.0000 227 3483000.0 504.75 -0.000353 34080.OD 42745S.00 -20.27 15 0.0O000 500. O00 0 228 3317000.0 504.75 -0.000244 29847.OD 45974.00 -35 .08 150.0000 500.0000 229 3051000, 0 504.95 -0.000078 21204.0O0 40474.00 -47 .61 15 0.0000 5 00. O0000 230 3121000.0 S04.85 -0 .000121 23837.00 39832.00 -40 .16 15 0.00 00 5 00.0000 231 3253000.0 504,95 -0.000198 27784.00 38676.00 -28.16 150.0000 500.0000 232 3426000.0 505.05 -0.000305 32346.00 37337 .00 -13 .37 lS0.0000 500.0000 233 3186000.0 505. 1 -0 .OOOlSl 25434.00 35259.00 -27 .87 150.0000 500.0000 234 3103000.0 505.25 -0 .000099 22514.00 34113.00 -34 .00 150.0000 500.0000 235 2792000.0 505.25 0.000002 12664.00 28979.00 -56 .30 lS0.0000 500.0000 236 2812000.0 SOS5.25 0.000002 12943.00 27894.00 -53 .60 lS0.0000 500.0000 237 3115000.0 505.55 -O0.000097 22395.00 25862.00 -13 .41 150.0000 500.0000 238 2984000.0 SOS .SS -0 .000024 17478.00 22786.00 -23 .30 150.0000 500,0000 239 3143000.0 SOS5.45 -0 .000116 23544.00 22898.00 2.82 150.00 00 500.0 00 0 24D 2900000.0 505.35 0.000002 14800.00 21059.00 -29 .72 150.0000 500,0000 241 297D000.0 505.25 -0 .000025 27519.00 20464.00 -14 .39 150.0000 500.0 00 0 242 2895000.0 504.85S 0.000 002 15425.00 19486.00 -20 .84 150.000 0 S00 .0000 243 3004000.0 SOS.15 -0.000046 19041.00 18966.00 0.40 150.0000 500.0000 244 2S80000.0 504.75 0.000002 15074.00 17953 .00 -16 .04 150.000 0 500. O0000 245 2868000.0 504.35 0.000002 15315.00 17820.00 -14 .06 150.000 0 S00 .0000 246 2870000.0 503 ,45 -0.000021 17095.0°O 20122.00 -15 .04 150.000 0 500.0000 247 2782000.0 503.05 0.000002 14496.00 19328.00 -25 .00 15 0.000 0 500. O0000 248 2838000.0 502.45 -0.000031 17729.00 19027.00 -6.82 15 0.000 0 500.0000 249 2718000.0 501.75 0.000002 14378.DO 17647.00 -18 .52 15 0.000 0 500.0000 250 2756000.0 501.45 -0 .000014 16369.00 18406.00 -11 .07 15 0.00 00 5 00.0000 251 2730000.01 S00. SS -0 .000023 17054.00 17708.00 -3 .69 150.0000 500.0000 252 2722000.0 499.65 -0 .000042 18461.00 16848.00 9.57 15D.0000 500.0000 ....................................................................................................................... s For 252 D)ata Points, Aerage Error -12.24 , S . 11.56 For 243 Subc:ooled Data Points. Average Error . -11.58 %,STD 10.87 % For 9 Satulrated Data Points, Average Error -29.94 *, STD 1S .68 % o-.4384nonW934-13.wpdlb4303 13-78

UH 1 2t -'.' WESTINGHOUSE PROPRIETARY CLASS 2 WCOBRA/TRAC Model Prediction vs. Marviken Data Mean Error is -12%

                                        ,Standard Deviation is 11.6%

U EWCT" 4 0 0 PREDICTION 70000. C3O Co 0 I I t

• I I I4 1 0 1
• 1 9
• 1 1 1
• 1 1 1 Measured Mass Flow Flux (kg/m2-s)

                -*. _%    -                   A . I         r                                   A  . I Figure 134-27 Prediction Comparison with Marviken Data oA4384-non\4384-13.wpd:1b4303                                      13-79

WESTINGHOUSE PROPRIETARY CLASS 2 Amos and Schrock Run No. pressure Teaperature Quality Predicted Gc Measured Gc Error D (GP-Gm) /GM (Pa) (K) (g/m2-s) (Kg/m2-s) (in %) I=) (u) 1 7073000.0 530.11 -0.003522 47138.00 44160.00 6.74 63.5000 0.74 80 544.15 -0.002142 37030.00 33980.00 8.98 63.5000 0.7480 2 7077000.0 553.69 -0.000931 24547.00 25220.00 -2.67 63.5000 0.7480 3 7091000.0 500.31 -0.005817 59298.00 57810.00 2.57 63.5000 0.7480 4 7093000.0 521.74 -0.011554 59922.00 57920.00 3.46 63.5000 0.7080 5 9595000.0 541.80 -0.017580 58551.00 57830.00 1.25 63 .000 0.7080 6 11580000.0

                                                                                                              -0.007416          45819.00        43940.00          4.28    63 .5000 0.7080 7                 9584000.0                                        551.16 34961.00        33230.00          5.21    63.5000  0.7080 8                 9619000.0                                        566.22                           -0.004212 566.68                           -0.011275          45903.00        44480.00          3.20    63.5000  0.7080 9              11608000.0 498.59                           -0.001212          33778.00        32560.00          3.74    63.5000  0.7470 10                   4220000.0 512.21                           -0.000638          26058.00        25260.00          3.16    63.5000  0.7470 11                   4187000.0 468.02                          -0.002387           40811.00        40990.00        -0.44     63.5 000 0.7470 12                   4271000.0 16499.00        14330.00        15.14     63.5000  0.7470 13                   4134000.0                                        522.54                          -0.000151 562.80                           -0.037795          68735.00        69690.00        -1.37     63.5000  0.7470 14               15398000.0 580.74                              0.000008        21910.00        26630.00       -17.72     63.5000  0.7470 1S                  9542000.0                                                                                                                                               0.7470 16               11672000.0                                          583.90                           -0.005217          35246.00        38600.00        -8.69     63.5000 591.18                           -0.021448          52698.00        52960.00        -0.49     63.5000  0.7470 17               15452000.0 593.60                           -0.000980          26333.00        30580.00       -13.89     63.5000  0.7470 18               11672000.0 467.97                           -0.002411          35559.00        35000.00          1.60    63.5000  0.4640 19                  4289000.0 497.31                           -0.001362          28951.00        28460.00          1.73    63.5000  0.4640 20                   4320000.0 22778.00        21600.00          5.45    63.5000  0.4640 21                   4281000.0                                        512.56                           -0.000702 17007.00        16870.00          0.81    63.5 000 0.4640 22                  4272000.0                                        522.53                           -0.000249 23                   7117000.0                                        498.94                           -0.005968          47691.00        42710.00        11.66     63.5000  0.5020 24                   7050000.0                                        499.19                          -0.005790           41809.00        40910.00          2.20    63.5000  0.4180 25                   7055000.0                                        530.34                          -0.003467           3 8811.00       36410.00          6.59    63.5000  0.5020 26                   7055000.0                                        541.34                          -0.002420           33988.00        31310.00          8.55    63.5000  0.5020 27                   7000000.0                                        555.71                          -0.000494           21242.00        20270.00          4.80    63.5000  0.5020 28                   9553000.0                                        521.72                          -0.011399           44577.00        44050.00          1.20    63.5000  0.4530 29                   9600000.0                                        553.08                          -0.007095           33601.00        32470.00          3.48    63.5000  0.4530 30                  9667000.0                                        559.29                           -0.005999          31164.00        27480.00         13.41    63 .5000 0.4530 31                  9602000.0                                        577.30                           -0.001168          20847.00        21980.00        -5.15     63.5000  0.4180 32                   9774000.0                                        524.45                           -0.011858          48123.00        48830.00        -1.45     63.5000  0.5020 33                   9609000.0                                        581.25                              0.000008        18729.00        22540.00       -16.91     63.5000  0.5020 34                11728000.0                                          536.16                           -0.019396          50162.00        49240.00          1.87    63.5000  0.4180 35                11601000.0                                          540.34                           -0.017970          42448.00        44130.00        -3 .81    63.5 000 0.4530 36                11696000.0                                          565.86                           -0.011934          39474.00        38520.00          2.48    63.5000  0.4180 37                12420000.0                                          538.95                           -0.022870          51640.00        51370.00          0.53    63.5000  0.5020 38                11642000.0                                          569.11                           -0.010627          38094.00        39010.00        -2.35     63.5000  0.5020 39               11838000.0                                          583.88                           -0.005855          30077.00        32240.00        -6.71     63.5000  0.5020 40               11675000.0                                          593.52                           -0.001027          22281.00        20580.00          8.27    63.5000  0.5020 41               15601000.0                                          564.26                           -0.038999          56678.00        50190.00         12.93    63.5000  0.4640 42               15798000.0                                          607.48                           -0.011009          32940.00        42310.00       -22.15     63.5000  0.5020 43               15747000.0                                          607.22                           -0.010916          31214.00        37030.00       -15.71     63.5000  0.5020 44               15703000.0                                          616.79                           -0.002074          26350.00        32920.00       -19.96     63.5000  0.5020 For           44 Data Points, Average Error                                                                             0.13 t, S    -       8.78 For           42 Subcooled Data Points, Average Error                                                                             0.96 t, STD -          8.08 %

For 2 Saturated Data Points. Average Error -17.32 %.SD . 0.58 t o:.4384-non\4384-13.wpd:1b-4303 13-80

WESTINGHOUSE PROPRIETARY CLASS 2 WCOBRA/TRAC Model Prediction vs. Amos Schrock Data Mean Error is .13% Standard Deviation is 8.78%

• U WCT 4 0 0 PREDICTION I,)

C C g) 30000 C1

                     -o
                     -o
                     -o     20000 a

Measured Mass Flow Flux (kg/m2-s)

                                            -              -                  .. A I Figure 13-4-28 Prediction Comparison with Amos-Schrock Data o:4384-non4384-13.wpd:1b-4303                          13 WESTINGHOUSE PROPRIETARY CLASS 2 TPFL (Anderson and Benedetti)

Run No0. Piessure Teupersture Quaity Predicted Gc Measuired Ge Error L. D (Gp-Gn) 1Gm 1 3450000.0 514.67 0.990000 4723.60 4269.40 2.0.64 54.0000 16.2000 3450000.0 514.67 0.990000 4230 4511.90 4.69 54.0000 16.2000 3 3490000.0 515.33 0.000002 9951.00 29012.00 -3.354.0000 16.2000 4 3450000.0 514.67 0.001100 18656.00 27072.00 -31.09 54.0000 16.2000 5 3430000.0 514.33 0.001400 18462.00 22899.00 -19.38 54.0000 16.200D 6 3460000.0 514.83 0.007100 17107.00 17466.00 -2.06 54.0000 16.200D 7 3460000.0 514.83 0.002D00 18292.00 17466.00 4.73 54.0000 16.2000 8 3440000.0 514.50 0.029D00 15910.00 13778.00 15.47 S4.0000 16.2000 9 3440000.0 514.50 0.026D00 16001.00 13439.00 19.06 54.0000 16.2000 10 3440000.0 514.50 0.050000 15353.00 12954.00 18.52 54.0000 16.2000 11 3440000.0 514.50 0.045000 15475.00 12177.00 27.08 54.0000 16.2000 12 3450000.0 514.67 0.056000 15249.00 12663.00 20.42 54.0000 16.2000 13 3450000.0 514.67 0.810000 5946.90 4754.50 25.08 54.0000 16.2000 14 3470000.0 515.00 0.760000 6281.80 4900.10 28.20 54.0000 16.2000 15 3470000.0 515.00 0.690000 6752.90 4803.00 40.60 54.0000 16.2000 16 3460000.0 514.83 0.890000 5533.50 4609.00 20.06 54.0000 16.2000 17 3450000.0 514.67 0.001200 18600.00 27072.00 -31.29 54.0000 16.2000 18 3450000.0 514.67 0.001100 18656.00 15622.00 19.42 54.0000 16.2000 19 3440000.0 514.50 0.04400D 15500.00 17320.00 -10.51 54.0000 16.2000 20 3450000.0 514.67 0.021000 16196.00 16883.00 -4.07 54.0000 16.2000 21 3470000.0 515.00 0.001300 18596.00 17320.00 7.37 54.0000 16.2000 22 3470000.0 515.00 0.067000 15062.00 18824.00 -19.99 54.0000 16.2000 23 3470000.0 515.00 0.001300 18596.00 17514.00 6.18 54.0000 16.2000 24 3450000.0 514.67 0.001300 18569.00 19164.00 -3.10 54.0000 16.2000 25 3440000.0 514.50 0.008000 16908.00 17708.00 -4.52 54.0000 16.2000 26 3470000.0 515.00 0.020000 15605.00 15185.00 2.77 54.0000 16.2000 27 3440000.D 514.50 0.026000 16001.00 14361.00 11.42 54.0000 16.2D00 28 3470000.0 515.00 0.054000 15351.00 12954.00 18.50 54.0000 16.2D00 29 3450000.0 514.67 0.280000 11190.00 7374.40 51.74 54.0000 16.2000 30 3450000.0 514.67 0.530000 8052.90 5821.90 38.32 54.0000 16.2000 31 3460000.0 514.83 0.220000 12179.00 8102.10 50.32 54.0000 16.2000 32 4470000.0 530.09 0.990000 6030.40 5724.80 5.34 54.0000 16.2000 33 4470000.0 530.09 0.990000 6030.40 5433.70 10.98 54.0000 16.2000 34 4450000.0 529.82 0.000003 21705.00 30856.00 -29.66 54.0000 16.2000 35 4410000.0 529.27 0.002600 20221.00 26101.00 -22.53 54.0000 16.2000 36 4440000.0 529.68 0.017000 18416.00 20134.00 -8.53 54.0000 16.2000 37 4440000.0 529.648 0.025000 17783.00 17369.00 2.38 54.0000 16.2000 38 4440000.0 529.68 0.023000 17943.00 17417.00 3.02 54.0000 16.2D00 39 4440000.0 529.68 0.045000 18524.00 14652.00 26.43 54.0000 16.2000 40 4440000.0 529.68 0.042000 18818.00 15331.00 21.44 54.0000 16.2D00 41 4420000.0 529.40 0.093000 17211.00 12517.00 37.50 54.0000 16.2000 42 4420000.0 529.40 0.089000 17304.00 12614.00 37.18 54.0000 16.2D00 43 4420000.0 529.40 0.100000 17050.00 12371.00 37.82 54.0000 16.2000 44 4420000.0 529.40 0.096000 17141.00 12614.00 35.89 54.0000 16.2000 45 4430000.0 529.54 0.074000 17696.00 13390.00 32.16 54.0000 16.2000 46 4430000.0 529.54 0.630000 9064.80 6210.00 45.97 54.0000 16.2000 47 4400000.0 529.13 0.086000 17817.00 13633.00 30.69 54.D000 16.2000 48 4440000.0 529.68 0.094000 17242.00 13293.00 29.71 54.0000 16.2000 49 4470000.0 530.09 0.900000 7734.80 5918.90 30.68 54.0000 16.2000 50 4540000.0 531.05 0.670000 8902.80 6258.50 42.25 54.0000 16.2000 51 4450000.0 529.82 0.640000 9008.80 6695.10 34.56 54.0000 16.2000 52 4400000.0 529.13 0.860D00 7212.10 4754.50 51.69 54.0000 16.2000 53 4440000.0 529.68 0.000450 21072.00 27848.00 -24.33 54.0000 16.2000 54 4400000.0 529.13 0.018000 18228.00 29285.00 -35.56 54.0000 16.2000 55 4470000.0 530.09 0.011000 19057.00 20231.00 -5.80 54.0000 16.2000 56 4420000.0 529.40 0.000230 21220.00 18339.00 15.71 54.0000 16.2000 57 4480000.0 530.23 0.030000 19147.00 17902.00 6.95 54.0000 16.2000 58 4410000.0 529.27 0.000260 21178.00 18630.00 13.68 54.0000 16.2000 59 4450000.0 529.82 0.017000 18431.00 17660.00 4.37 54.0000 16.2000 60 4400000.0 529.13 0.001400 20556.00 15088.00 36.24 54.0000 16.2000 61 4450000.0 529.82 0.043000 18616.00 16059.00 15.92 54.0000 16.2000 62 4410000.0 529.27 0.110000 16799.00 10819.00 55.27 54.0000 16.2000 63 4450000.0 529.82 0.130000 16479.00 10770.00 53.01 54.0000 16.2000 64 4430000.0 529.54 0.350000 12491.00 8732.80 43.04 54.0000 16.2000 65 6150000.0 550.36 0.990000 8142.60 7422.90 9.70 54.0000 16.2000 66 6260000.0 551.53 0.990000 8279.40 7762.50 6.66 54.0000 16.2000 67 6260000.0 551.53 0.990000 8279.40 7908.00 4.70 54.0000 16.2000 68 6260000.0 551.53 0.990000 8279.40 7859.50 5.34 54.0000 16.2000 69 6160000.0 550.47 0.000004 24575.00 33379.00 -26.38 54.0000 16.2D00 70 6170000.0 550.58 0.000004 24599.00 32942.00 -25.33 54.0000 16.2D00 71 6210000.0 551.00 0.001900 24070.00 32991.00 -27.04 54.0000 16.2000 72 6240000.0 551.32 0.000004 24703.00 29546.00 -16.39 54.0000 16.2D00 73 6200000.0 550.90 0.000068 24544.00 27848.00 -11.86 54.0000 16.2D00 74 6240000.0 551.32 0.015000 22915.00 24888.00 -7.93 54.0000 16.2D00 75 6270000.0 551.64 0.030000 22098.00 22075.00 0.10 54.0000 16.2D00 76 6220000.0 551.11 0.025000 22269.00 20037.00 11.14 54.0000 16.2000 77 6200000.0 550.90 0.058000 23041.00 18533.00 24.32 54.0000 16.2000 78 6200000.0 550.90 0.054000 23184.00 18921.00 22.53 54.0000 16.2D00 79 6170000.0 550.58 0.091000 21916.00 17223.00 27.25 54.0000 16.2000 80 6170000.0 550.58 0.093000 21858.00 17029.00 28.36 54.0000 16.2000 81 6210000.0 551.00 0.120000 21227.00 16689.00 27.19 54.0000 16.2000 82 6280000.0 551.74 0.088000 22284.00 17320.00 28.66 54.0000 16.2000 83 6260000.0 551.53 0.090000 22175.00 17417.00 27.32 54.0000 16.2000 94 6230000.0 551.22 0.110000 21540.00 16641.00 29.44 54.0000 16.2000 85 6230000.0 551.22 0.700000 11699.00 8829.80 32.49 54.0000 16.2000 86 6250000.0 551.43 0.710000 11630.00 8781.30 32.44 54.0000 16.2000 87 6240000.0 551.32 0.160000 20292.00 17805.00 13.97 54.0000 16.2000 88 6240000.0 551.32 0.290000 17603.00 16544.00 6.40 54.0000 16.2000 89 6230000.0 551.22 0.960000 9413.90 7859.50 19.78 54.0000 16.2000 90 6220000.0 551.11 0.970000 9326.10 7762.50 20.14 54.0000 16.2000 91 6220000.0 551.11 0.999990 8229.70 7568.40 8.74 54.0000 16.2000 92 6190000.0 550.79 0.000550 24313.00 32117.00 -24.30 54.0000 16.2000 93 6220000.0 551.11 0.046000 23526.00 19843.00 18.56 54.0000 16.200D o:A4384-noon4384-13.vTd:11-4303 13-82

WESTINGHOUSE PROPRIETARY CLASS 2 94 6180000.0 550.69 $!0.009500 23222.00 30759.00 ) V24'.50 54. 0000 16.2000 95 6170000.0 550.58 0.028000 21968.00 19891.00 1 10.44 54.0000 16.2000 96 6220000.0 551.11 0.011000 23192.00 30371.00 -23.64 54.0000 16.2000 97 6220000.0 551.11 0.024000 22328.00 21347.00 4.60 54.0000 16.2000 98 6230000.0 551.22 0.015000 22909.00 32117.00 -28.67 54.0000 16.2000 99 6220000.0 551.11 0.036000 21686.00 20522.00 5.67 54.0000 16.2000 o00 6270000.0 551.64 0.036000 21792.00 20279.00 7.46 54. 0000 16.2000 101 6180000.0 550.69 0.038000 21478.00 19940.O0 7.71 54.0000 16.2000 102 6200000.0 550.90 0.021000 22456.00 22317.00 0.62 54.0000 16.2000 103 6190000.0 550.79 0.012000 23045.00 22705.00 1.50 54.0000 16.2000 104 6210000.0 551.00 0.038000 21555.00 21056.o 2.37 54.0000 16.2000 105 6210000.0 551.00 0.089000 22075.00 17563.00 25.69 54. 0000 16.2000 106 6230000.0 551.22 0 .140000 20775.00 15768.00 31.75 54.0000 26.2000 107 6230000.0 551.22 0 .220000 18964.00 14312.00 32.50 54.0000 16.2000 108 6220000.0 551.11 0.270000 17939.00 13390.00 33.97 54.0000 16.2000 109 6190000.0 550 .79 0.580000 13003.00 9751.60 33.34 54.0000 16.2000 For 109 Data Points, Average Error 11.96 %,STD = 22.20 For 109 Saturated Data Points, Average Error

• 11.96 II. 5T - 22.20 WCOBRA/TRAC Model Prediction vs. TPFL Data Mean Error is 12%

Standard Deviation is 22.2%

                           *

• CT 4 0 0 PREDICTION 2=m 2CV1 C.'1 E

9-C en Eooa> CL C) 0w Figure 13-4-29 Prediction Comparison with TPFL Data o:\4384-non\4384-13.wpd:Ib-4303 13-83

WESTINGHOUSE PROPRIETARY CLASS 2 13-5 Scaling Consideration An observation relative to the scalability of the model is addressed in this section. 13-5-1 Pressure, Subcooling, and Quality For the subcooled break flow model, a pressure range of 13 to 2300 psia and a quality range of -0.039 to 1.0 were examined. The results indicated that the model is scalable relative to pressure and subcooling with reasonable accuracy. The results showed that the model adequately accounts for the pressure and the quality variations. 13-5-2 Break Flow Area The break flow comparisons showed that the present model predicted both small diameter tests such as Amos and Schrock for 0.0295 inch (Amos and Schrock, 1983), and Sozzi and Sutherland for 0.5-inch (Sozzi and Sutherland, 1975), as well as the large diameter (19.7-inch) data obtained in the Marviken tests (EPRI-NP-2370, 1982) with adequate accuracy. The WCOBRA/TRAC break model was able to simulate both small and large diameter nozzles adequately. 13-5-3 Break Geometry The entrance effects, such as the roundness/sharpness of the orifice are accounted for in the present model although the flow area variation along the axis is neglected. 13-54 Pressure Effect on the Onset of Entrainment and Branchlne Quality The horizontal stratified entrainment model was validated from 500 up to 900 psia, close to the full pressure at which the entrainment is an important factor in small break LOCAs. Thus, no significant distortion in the PWR calculation is expected relative to the pressure. 13-5-5 Mainline Pipe Diameter Variation on the Onset of Entrainment and Branchline Quality This parameter may be important for stratified entrainment behavior. The correlation used in WCOBRA/TRAC is validated up to 28.4 cm. This is roughly one-third of the cold leg diameter, o:4394-non\438413.wpd:1b-4303 13-84

WESTINGHOUSE PROPRIETARY CLASS 2 where a postulated small break LOCA is assumed to occur. There may be a scale distortion at a full PWR diameter of 99 cm. However, the original correlations were derived by assuming an infinite diameter tank; it is likely that the size distortion is small. 13-6 Conclusions The break flow comparisons showed that the present model predicted both small diameter tests such as Amos and Schrock at 0.0295 inch, and Sozzi and Sutherland at 0.5-inch as well as the large diameter (19.7-inch) data obtained in the Marviken tests (EPRI-NP-2370, 1982) with acceptable accuracy. The onset of vapor pull-through and liquid entrainment, in addition to break quality, are well simulated by WCOBRA/TRAC-SB. However, it is possible that the entrainment is strongly influenced by the presence of waves and vortex; the uncertainty may be higher than estimated in this assessment in other applications. 13-7 References Alamgir, M. D. and Lienhard, J. H., 1981, "Correlation of Pressure Undershoot During Hot-Water Depressurization," Trans. ASME, J. Heat Transfer, Vol. 103, pp. 52-55. Amos, C. N. and Schrock, V. E., 1983, "Critical Discharge of Initially Subcooled Water Through Slits," NUREG/CR-3475. Anderson, J. L. and Benedetti, R. L., 1985, "Critical Flow Through Small Pipe Break," NUREG/CR-4532. Ardron, K H. & Ackerman, M. C., "Studies of the Critical Flow of Subcooled Water in a Pipe," CEGB Report: RD/BlN4299, 1978. Ardron, K. H. and Bryce, W. M., 1990, "Assessment of Horizontal Stratification Entrainment Model in RELAP5/MOD2 by Comparison With Separate Effects Experiment," Nuclear Engineering and Design 122, pp. 263-271. Bajorek, S. M., et al., 1992, "Westinghouse Code Qualification Document for Best Estimate Loss of Coolant Accident Analysis Volume 1: Models and Correlations," WCAP-12945-P-A, Vol. 1, pp. 4-124. o:4384-nonW4384-13.wpd:Ib4303 13-85

WESTINGHOUSE PROPRIETARY CLASS 2 Boivin, J. V., 'Two-phase Critical Flow in Long Nozzles," Nuclear Technology, 46, pages 540-545, 1979. Bryers, R. W., Hsieh, W. W., Hunter, J. A. & Sieder, E. N., "Study of Two-phase Metastable Flow," U. S. Department of Interior, Office of Saline Water, R&D Progress Report No. 234, November, 1966. Condie, K. G., 1980, "LOFT LOCE L3-5/L3-5A Results and Analysis." Paper presented at the LOFT Review Group Meeting, Idaho Falls, Idaho. Craya, A., 1949, "Experimental Research on the Flow of Nonhomogeneous Fluids," LeHouille Blanche, January-February, pp. 56-64. Cruver, J. E., "Metastable Critical Flow of Steam Water Mixtures," Ph.D. Dissertation, Department of Chemical Engineering, University of Washington, 1963 (University Microfilms Inc., Ann Arbor, Michigan). Danforth, J. L., "Flow of Hot Water Through a Rounded Orifice," M. S. Dissertation, Department of Mechanical Engineering, Massachusetts Institute of Technology, 1941. Doa, L. T. C and Carpenter, J. M., 1980, "Experiment Data Report for LOFT Nuclear Small-Break Experiment L3-5/L3-5A," NUREG/CR-1695, EGG-2060. EPRI-NP-2370, 1982, "The Marviken Full-Scale Critical Flow Tests," Vol. 1, Summary Report. Fauske, H. K., "Contribution to the Theory of Two-phase, One Component Critical Flow," Argonne National Laboratory Report ANL-6633, 1962. Fincke, J. R. & Collins, D. R., "The Correlation of Two-dimension and Non-equilibrium Effects in Subcooled Choked Nozzle Flow," NUREG/CR-1907, EGG-2081. Guizovarn, L., Pinet, B., Barriere, G. & Pietri, D., "Etude Expermentale des Debits Critiques en Ecoulement Diphasiques eau Vapeur a Faible Titre Dans un Canal Avec Divergent de 70 a des Transports Thermiques," Note TT No. 501, 1975. Henry, R. E., "An Experimental Study of Low Quality, Steam-Water Critical Flow at Moderate Pressures," Argonne National Laboratory Report ANL-7740, 1970. o:A384-non\4384-13.wpd:1b-4303 13-86

WESTINGHOUSE PROPRIETARY CLASS 2 Illic, V., Banerjee S. and Behling S., "A Qualified Database for the Critical Flow of Water," EPRI-NP4556, May, 1986. Jeandey, C., Gros D'Aillon, L., Bourgin, R. & Barriere, G., "Auto Vaporisation d'Ecoulements EaulVapeur," Departement des Reacteurs a Eau Service des Transferts Thermiques (Centre D'Etudes Nucleaires de Grenoble) Report T. T. No. 163, 1981. Jones, 0. C., Jr., 1980, "Flashing Inception in Flowing Liquids," Trans. ASME, J. Heat Transfer, Vol. 102, pp. 439-444. Lubin, B. T. and Springer, G. S., 1967, "The Formation of a Dip on the Surface of a Liquid Draining from a Tank," Journal of Fluid Mechanics, 29, pp. 385-390. Maciaszek, T. and Memponteil, A., 1986, "Experimental Study on Phase Separation in a Tee Junction for Steam-Water Stratified Inlet Flow," Paper C2 presented at European Two-phase Flow Group Meeting, Munich, June 10-13. Morrison, A. F., "Blowdown Flow in the BWR BDHT Test Apparatus," GEAP-21656, 1977. Neusen, K. F., "Optimizing of Flow Parameters for the Expansion of Very Low-quality Steam," University of California, Lawrence Radiation Laboratory, UCRL-6152, 1969. Reocreux, M., 1974, "Contributions a 'Etude das Debits Critique en Econlement Diphasique Eau-Vapeur," Ph.D. Thesis, Universite Scientifique et Medicale de Grenoble, France. Schrock, V. E., et al., 1986, "Small Break Critical Discharge - The Roles of Vapor and Liquid Entrainment in a Stratified Two-phase Region Upstream of the Break," NUREG/CR-4761. Schrock, V. E., Starkman, E. S. & Brown, R. A., "Flashing Flow of Initially Subcooled Water in Convergent-divergent Nozzles," Journal of Heat Transfer 99 (2), 1977. Seynhaeve, J. M., "Etude Experimentale des Ecoulements Diphasiques Critiques a Faible Titre," Doctoral thesis, Department Thermodynamique et Turbomachines, Universite Catholique de Louvain, 1980. o:43&4-non\4384-13.wpd:Ib4303 13-87

WESTINGHOUSE PROPRIETARY CLASS 2 Seynhaeve, J. M., et al., 1976, "Nonequilibrium Effects on Critical Flow Rates at Low Qualities," Vol. 2, pp. 672-688, Proceedings of the CSNI Specialist Meeting on Transient Two-phase Flow, Toronto, August 3-4. Smoglie, C. and Reimann, J., 1986, "Two-phase Flow Through Small Branches in a Horizontal Pipe with Stratified Flow," Intemational Journal of Multi-Phase Flow, pp. 609-626. Sozzi, G. L. and Sutherland, W. A., 1975, "Critical Flow of Saturated and Subcooled Water at High Pressure," NEDO-13418. Yonomoto, T. and Tasaka, K, 1988, "New Theoretical Model for Two-phase Flow Discharge from Stratified Two-phase Region Through Small Break," Journal of Nuclear Science and Technology, 25[5], pp. 441-455. Zaloudek, F. R., 1964, "Steam-Water Critical Flow From High Pressure Systems," HW-80535. Zuber, N., 1980, "Problem in Modelling of Small Break LOCA," NUREG-724. o:4384non\4384-13.wpd:1b-4303 13-88

SECTION 14 SAFETY INJECTION JET CONDENSATION: COSI EXPERIMENTS 14-1 Introduction The phenomenon of direct-contact condensation, the condensation of vapor by subcooled liquid, takes place in the cold leg piping during a small break LOCA transient once any voiding has occurred and the relatively cold safety injection water is being injected. Steam condensation results in volume shrinkage, and this in turn affects the pressure globally throughout the RCS, with ultimate implications for how much water the centrifugal safety injection pumps can inject against the RCS backpressure. If left within the RCS indefinitely, the injected cold water eventually mixes with the hot liquid and steam and reaches equilibrium conditions; a simple mass and energy balance should be adequate to describe and predict the process. However, because the water is injected into the cold leg piping, many more complex effects can occur from the localized condensation process. In the broken loop cold leg, the degree of condensation affects the enthalpy of fluid at the break plane. This can affect the break flowrate and thus the system pressure and mass loss. Also, the volume reduction from the steam condensed in the cold leg tends to be replaced by steam flowing from other parts of the RCS, and this generates flow pressure drops that affect coolant distribution. Finally, the degree to which complete mixing and condensation occur in the cold leg affects the water conditions in the vessel downcomer, which in turn can affect the gravity head for coolant distribution in the reactor vessel. To investigate the ability of the WCOBRA/RAC-SB code to correctly predict condensation phenomena, a model was constructed of a series of experiments which were performed in the Condensation On Safety Injection (COSI) facility. The COSI facility is an approximately 1:100 scale model of the cold leg and safety injection lines of a Westinghouse-type nuclear power plant (NPP), constructed specifically for investigating the interaction of steam and cold safety injection water in a prototypical NPP configuration and at typical NPP fluid conditions encountered during a small break LOCA. A description of the facility, the experiments, and the modelling follow. o:\4384-non\4384-14.wpd:lb-040403 14-1

I 14-2 Description of COSI "Li 14-2-1 Facility Description The COSI facility is a 1:100 scale model of the cold leg and safety injection ports of a Westinghouse-type NPP. It is capable of operating at pressures [ ]ac and at appropriately scaled flowrates to cover nearly the full range of injection conditions expected in an NPP transient, during which condensation on the safety injection water is an identified phenomenon. The main scaling philosophy followed in designing the system was [ Iax Figure 14-1 illustrates the arrangement of the main components of the test facility. The main pipe (cold leg simulator) is [ as seen in Figure 14-2. The experiments simulated herein [

                                                                        ]'   Instrumentation in the facility was state-of-the-art in the mid-1980s, and measurement accuracies are extremely good.

Measurements are available for steam and liquid flowrates in and out of the test assembly, for temperatures of all fluid entering and exiting, for pressures, and for differential pressures. Within the test section, a series of thermocouple rakes provides information concerning stratification of the liquid. 14-2-2 Key Phenomena Information obtained from the tests provides a data base for assessing models for steam condensation on cold safety injection jets and with varying levels of water in the main pipe simulating the cold leg. Previous evaluations of the data (Shimeck, 1988 and Jonicot and Bestion, 1993) concluded that there was significant condensation for nearly all test conditions which, when measured in terms of condensation efficiency, approached values of 100 percent in some cases. Condensation efficiency is defined as the ratio of the steam mass condensation which occurs to the mass of the steam condensation that would raise the enthalpy of the injected safety injection water to saturation. There was clear evidence of stratification in the main pipe, and other evidence pointed to the conclusion that the majority of the condensation was occurring o:\4384-non\4384-14.wpd:lb-040403 14-2

directly within the relatively small jet mixing zone. The jet mixing zone is defined as encompassing the jet of water flowing out of the injection port together with the area a short distance immediately upstream and downstream of the injection point in which water in the cold leg is turbulently mixed by the impact and spreading ofthejet. Investigations were conducted for a range of pressures, injection rates, and weir heights to determine to what degree any of these affected the results. 14-2-3 Applicable Tests and Parameter Ranges A large matrix of tests was conducted over the course of the program by both Westinghouse and Framatome, and some reconfiguration of the facility test section was performed with regard to the length of the main pipe in the test assembly and the angle and size of the injection piping. The experiments of most interest and applicability to code validation for small break LOCA transients were the steady-state points with flows simulating pumped high head injection. A core series of 11 tests, with 55 individual data points, is identified in WCAP-1 1767 (Shimeck, 1988). The key parameters of interest from these tests, which are useful for sensitivity studies, are shown in Table 14-1. 14-3 Description of WCOBRA/TRAC Model Figure 14-3 shows the component layout of the WCOBRAfIRAC-SB model of the COSI facility. The main test section, which consists here of the cold leg pipe and the downcomer, is modelled [

                                                                                 ]3$ The test points to be simulated were established to be steady-state with constant pressure and constant flowrates of steam and water. The two BREAK components on either end of the assembly allow for controlling the system at a given pressure. The FILL component provides a constant source of safety injection water.
                                                      ]'   Figure 14-4 shows more detail o:\4384-non\4384-14.wpd:lb-040403               14-3

I concerning the modelling of the test section [ ],a and Figure 14-5 is a cross-sectional view of [ so that tests with the weir, which was one-half of the pipe diameter, can be more properly simulated. The/[channel size was selected to provide a reasonable degree of detail, but without an excessive number of channel§]"T} 14-4 Simulations 14-4-1 Summary of Experimental Results A set of test points, as obtained from the experiments, corresponds to a given configuration of injection line size and weir installation (in or out). Two injection line diameters (d) were tested [ x1, Information from measurements made with the thermocouple rakes was reviewed to deduce the phenomena involved in the condensation process along the cold leg when a weir is present. Significant amounts of temperature stratification were observed, and combined with examination of the actual temperatures, the conclusion was that the overall behavior depicted in Figure 14-6 was taking place. Fluid temperatures upstream and downstream of the injection point were stable and indicated that a countercurrent flow pattern was in place on the upstream side. On the downstream side, it is not clear whether the flow pattern was cocurrent or countercurrent, but again stratification was noted. In the immediate vicinity of the injection port, the thermocouple measurements exhibited a significant standard deviation, indicating turbulent conditions. The downward impingement of the safety injection jet, combined with the significant influx of steam to this point, supported a turbulent jet mixing zone, with rather complex flow and heat transfer patterns. Any safety injection water that leaves the jet mixing zone and is not saturated will support further condensation on the pool surface both upstream and downstream. Additionally, the small waterfall that occurs in the downcomer region accentuated the condensation in this region. Although this waterfall effect may be argued to possibly be somewhat prototypical of water in an NPP falling into the downcomer from the cold leg, it was not the intent of the o:\4384-non\4384-14.wpdtlb040403 14-4

experiment to simulate this effect. Therefore, a set of experiments was conducted in the facility in which the downcomer water level was varied to obtain an adjustment to the data that removed this factor. The conclusion from analysis of the data was that there is a strong condensation mechanism in the jet mixing zone which must be modelled in a small break LOCA simulation. Any nonsaturated water that exits this zone will then simply interact in a more quiescent fashion. Condensation beyond the jet mixing zone is calculated with other models (see Section 18). The experimental results are best described in terms of the condensation efficiency, which is defined as the ratio of the mass flowrate of steam to the mass flowrate of steam that raises all of the safety injection water to saturation. The black symbols in Figures 14-7, 14-8, and 14-9 show the experimental results for various configurations. The results are presented in Tables 14-2, 14-3, and 14-4 as well. The presented condensation efficiencies were corrected so that condensation in the downcomer region is not taken into account. [ I" In general, condensation efficiencies were high for all conditions that were examined at all pressures, injection rates, and injection configurations. [

                                                                                     ]a.c 14-4-2 WCOBRAfRAC-SB Results The calculations with the WCOBRA/IRAC-SB model were performed [
                                                                              ]^* In this model, no attempt is made to modify or increase the heat transfer due to the entrainment or droplet production.

The effects of condensation in the downcomer region were not taken into account (by deactivating condensation in that region) to make the code results comparable with the corrected experimental results. Only condensation on the water-steam interfacial surface in the horizontal pipe and condensation due to the presence of the subcooled jet water were applied in the model. o\434-non\4384-14.wpd:lb040404 14-5

The condensation due to the jet presence was [

                               ]aC in agreement with the analysis of the experiments.

[ ]ac Figure 14-7 shows calculated condensation efficiency (white symbols). The results are presented in Table 14-2 as well. The comparison of numerical and experimental results (white and black symbols) shows that calculated condensation efficiency is in good agreement with the measured results for the [

                      ]' c Comparison of the calculated condensation efficiencies for various pressures leads to the conclusion that they have similar values regardless of the pressure. That could be expected because the developed correlation for the subcooled liquid condensation heat transfer coefficient does not take into account pressure effects. Calculated condensation efficiencies have a range of the possible numerical values (presented as variation bars in the Figure 14-7 or in the last column in Table 14-2) due to small oscillations of the calculated results.

The results for the case [

                                                                                               ]ac The case [

l As for the previous cases, the smallest difference is for the lowest injection water flowrates. The WCOBRA/TRAC-SB small break LOCA computer code contains a safety injection condensation model to simulate that phenomenon. The condensation in the jet mixing zone is dominant, and an increase of the condensation due to the jet presence is necessary to achieve better agreement with the experimental results, particularly at high safety injection flows. The present model does not take into account detailed effects of the entrainment, disturbance of the water surface due to the jet impingement, droplet formation, and consequently, the increase of heat transfer area with the increased jet Reynolds number. Agreement between experimental and o:\4384-non\4394-14.wpd: lb-040403 146

numerical results is closest at the lowest COSI safety injection water flowrate. This flowrate is the closest to the scaled injection rate that corresponds to the analyzed single failure condition for the PWR. For instance, [

                                                          ]axc 14-5 Conclusions A WCOBRA.'RAC-SB model of the COSI safety injection condensation separate-effects experiments has been used to simulate the phenomena. The comparison with experimental results shows that the code was able to predict condensation rates within a reasonable range for the lower safety injection flowrates. In the range of the higher safety injection flowrates, the code underpredicts the condensation efficiencies. The lowest COSI injection flowrate is the most representative of the as-analyzed PWR flowrates during the pressure range of interest for small break LOCA events. Therefore, the jet condensation efficiency in the WCOBRA/FRAC-SB predictions is judged to be acceptable for integral test facility and PWR simulations.

14-6 References Jonicot, A. and Bestion, D., 1993, "Condensation Modelling for ECC Injection," Nuclear Engineering and Design, 145, pp. 37-45. Shimeck, D. J., 1988, "COSI SI/Steam Condensation Experiment Analysis," WCAP-1 1767, Proprietary. o\4384-non\4384-14.wpd:lb-040403 14-7

Table 14-1 Summary of Applicable COSI Experiments Item Parameter Pressures [ ]ac COSI Full-Scale Flowrates [ Safety injection line diameters Safety injection temperature Simulated pump weir Ia.c

                                                                                    .,X, o-\4384-non\4384-14.wpd:lb-040403                  14-8

a,c Table 14-2 o:.4384-non\4384-14.wpd:lb-040403 14-9

a,c Table 14-3 7 7 r

                                                        .L,
                         .5.                  .5. 1 o:\4384-non\4384-14.wpd:lb-040403     14-10

II a,c Table 14-4 r I P i o:\4384-non\4384-14.wpd:1b 040403 14-1 1

I a,c Figure 14-1. COSI Facility Arrangement 4 o:4384-non\4384-14.wpd.lb-040 03 14-12

a,c Figure 14-2. Test Section Arrangement 0 o.\4384-non\4384-14.wpd:Ib-0404 3 14-13

ax Figure 14-3. COST WCOBRAJTRAC Model Component Layout o\4384-non\4384-14.wpd:lb-040403 14-14

a,c Figure 14-4. COSI Main Test Section and Downcomer WCOBRAJTRAC Model o:4384non\4384-14.wpd:lb-040403 14-15

I a,c IL, Figure 14-5. COSI Cold Leg Pipe Vertical Cell Nodalization o:\4384-non\4384-14.wpd:1b-O404O3 14-16

( ( - 0 SI Injection 9 0 I tzI Vapor Flow / , ,\SI Jet I ~0 I"-W

            ~
        , ^ I~ t% ,~ .M~, s~ r -,                  ~_$  _b_l 1  _.

0 Counter-Current Turbulent Jet Upstream Flow Mixing Zone Downcomer

a,c Figure 14-7. Condensation Efficiency for Small Injection Pipe (d = 0.22 Inches) and Weir (ID = 0.5) o:\4384-non\4384-14.wpd:lb-040403 14-18

a,c Figure 14-8. Condensation Efficiency for Large Injection Pipe (d = 0.90 Inches) and Weir (HID = 0.5) o:\43S4-non\4384-14.wpd:lb-040403 14-19

I a,c - Figure 14-9. Condensation Efficiency for Large Injection Pipe (d = 0.90 Inches) and Without Weir (H/D =0) oA4384-non\4384-14.wpd:lb-O40403 14-20

SECTION 15 MIXTURE LEVEL SWELL 15-1 Introduction Early in a small break LOCA, voids are generated in the primary RCS by flashing and boiling in the core. Because of the small break size, flows in the RCS are primarily gravity-driven. Following the initial rapid depressurization stage of the LOCA, distinct liquid levels are formed at several locations. Below this liquid or two-phase mixture level, the fluid is a low quality two-phase mixture; while above the level, it is primarily single-phase vapor. Liquid levels initially occur in the pressurizer, in the upper head, and in the uphill and downhill steam generator tubing. Eventually, the RCS drains so that the level in the reactor vessel reaches the hot leg. At this point, the rate of system depressurization is low and vapor generation results from boiling in the core, from power produced by decay heat. Because the vapor generated by this decay heat can be high, regions in the vessel can achieve a significant void fraction. The two-phase mixture level depends on the interfacial shear exerted by the vapor on the liquid, and as a result, the mixture level can be significantly higher than the collapsed liquid level. The difference between the two-phase mixture level and the collapsed level is a measure of the "mixture level swell," which is defined as: s= Z2 (5-1) ZCLL where Zcu is the collapsed liquid level and Z.¢, is the two-phase mixture level. Prediction of the mixture level swell and tracking of the mixture level are important in the later stages of a small break LOCA. As more liquid is boiled away, the mixture level can eventually drop into the core. While good cooling can be maintained below the mixture level, dryout occurs above the mixture level. Heat transfer above the mixture level is by convection and thermal radiation to steam. These relatively poor modes of heat transfer cause the cladding temperature above the mixture level to increase rapidly. Thus, prediction of the two-phase mixture level in the active core is vital to an accurate prediction of the PCT in a small break LOCA. o:\4384-non\4384-15awpd:b-{4033 15-1

I 15-2 Physical Processes As described in Section 15-1, mixture level swell is the process that determines the vertical position of the two-phase interfaces in the system; below the interface the mixture is low quality and above the interface the mixture is essentially single-phase vapor. The mixture level swell depends on several processes: the interfacial drag between the vapor and liquid (film), wall drag, bubble rise velocity and bubble size, entrainment of droplets at the two-phase interface, and transition point between bubbly and other vertical flow regimes. In general, the liquid and vapor flowrates are low, which make wall drag due to form and friction losses negligible compared to the interfacial drag. In small break LOCA scenarios, the steam velocities are too low to entrain droplets at the two-phase interface, and thus entrainment is negligible. Therefore, mixture level swell is most directly affected by processes that deternine the interfacial shear and the relative velocity between the phases. Several experimental tests have been run under small break LOCA thermal-hydraulic conditions to measure the effects of various parameters on mixture level swell. Bundle power, or more accurately the vapor generation rate, had the most dominant effect on the measured mixture level and void fraction distributions. Transition to dryout did not occur until the void fraction exceeded a value of approximately 0.85. Based on these observations, factors considered important in the assessment of predictions of mixture level swell include:

• Mixture level as a function of bundle power and inlet flowrate
• Collapsed liquid level as a function of bundle power and inlet flowrate
• Void fraction distribution 15-3 WCOBRAITRAC Determination of the Mixture Level The models and correlations for wall and interfacial drag are described in Section 4, Volume 1, of this document. Flow regime transitions are described in Section 3 of this document. These models are used to determine the void fraction distribution within a region.

WCOBRA[TAC-SB does not include a specific model or pointer to identify the mixture level. Thus, mixture level tracking is accomplished [ aC o:\4384-non\4384-15a.wpd:lb-04033 15-2

[ Iac While the WCOBRATRAC interface logic prevents the use of an unrealistically low void fraction in a cell, it does not uniquely determiine the mixture level elevation. Therefore, the ability of WCOBRA/TRAC to track a mixture level is dependent upon the axial noding. In the core, [

            ]ax 15-4 Assessment of WCOBRA/TRAC Mixture Level Predictions 154-1 Introduction There are several separate effects experimental tests that provide data on the mixture level and mass inventory distribution in a rod bundle under small break LOCA thermal-hydraulic conditions. Three such experimental facilities were modelled with WCOBRAITRAC-SB, and several experimental tests were simulated to determine the predictive capability of the code.

The tests were as follows:

• The ORNL-THTF Uncovered Bundle Tests by Anklam (Anlam, et al., 1982)
• The Westinghouse G-1 Core Uncovery Tests, WCAP-9764 (WCAP-9764, 1980)
• The General Electric (GE) Vessel Blowdown Tests by Findlay and Sozzi (Findlay and Sozzi, 1981)

Each of these tests, run at pressures typical of those in a small break LOCA (1100 to 400 psia), provides information on the mass distribution in a vessel for various thermal-hydraulic conditions. The ORNL-THTF and G-1 tests provide mixture level and mass inventories for uncovered rod bundles, and the GE tests provide mass inventory in a vessel during a rapid depressurization. o:"4384-non\4384-15a.wpd:lb-04033 15-3

I The following sections discuss each test, the WCOBRAITRAC-SB simulation, and the comparisons between the measured and predicted results. 154-2 ORNL-THTF Small Break Tests 154-2-1 Introduction The ORNL-THTF performed a series of experimental tests pertinent to small break LOCA model validation. The ORNL-THTF was a high pressure rod bundle thermal-hydraulics loop. The bundle was full height and contained 64 electrically heated rods with internal dimensions typical of a 17x17 PWR fuel bundle. Figure 154-2-1 shows a cross section of the ORNL-THTF test bundle. Four of the rods were unheated to represent control rod guide tubes in a nuclear fuel assembly. Figure 154-2-2 shows an axial profile of the ORNL-THTF bundle. The bundle had a heated length of 12 feet (3.66 m) and contained six spacer grids. Thermocouples were located at 25 different axial elevations. Two types of experiments were conducted in the ORNL-THTF. One series consisted of several uncovered bundle heat transfer tests. In these tests, the experiment was continued until a steady-state condition was reached in the uncovered part of the bundle and rods were heated to a high temperature. The second type of tests did not have bundle uncovery. The bundle remained covered, and a void profile over the entire axial length was obtained. Additional information on the ORNL-THTF test bundle, and on the tests conducted in the facility, is in NUREG/CR-2456 (Anklam, et al., 1982). 154-2-2 WCOBRA/1RAC Model of the ORNL-THTF Figure 154-2-3 shows the WCOBRA/TRAC model of the ORNL-THTF. The heated length is modelled [

                                                         ].

o:\4384-non\4384-15a.wpd:lb-04033 15-4

154-2-3 Test Matrix for ORNL-THTF Simulations Simulations of small break LOCAs in PWRs generally show that there are two periods in which the core can possibly be uncovered. The first occurs during the loop seal clearance period. During this uncovery, the primary system pressure is high (approximately 1150 psia) and the two-phased mixture level can drop below the top of the core. The second uncovery occurs if the break flow exceeds the pumped SI flow during the boil-off period. The system pressure during this uncovery is low, typically 600 to 650 psia, which is just below the accumulator gas pressure. Table 15-4-2-1 lists tests selected for simulation by WCOBRAITRAC-SB. Six of the tests are bundle uncovery tests. Three are at relatively low pressure (580 to 650 psia), and three are at high pressure (1010 to 1090 psia). All six had roughly one-half the bundle uncovered. Six other tests are from the level swell test series. Again, three were at low pressure (520 to 590 psia), and three were at high pressure (1090 to 1170 psia). These tests span the expected range of conditions for uncovery in PWR calculations leading to the most limiting PCTs. 15-4-2-4 Simulation of ORNL-THTF Tests Each test was simulated by imposing a flow and enthalpy boundary condition at the bottom of channel 1, and a constant pressure boundary condition at the top of channel 3. The simulation was continued until the calculation reached a steady-state condition. The parameter YDRAG and the tests are discussed below. Interfacial Drag Multiplier YDRAG The parameter YDRAG has been introduced to facilitate WCOBRAIRAC-SB ranging of interfacial drag. YDRAG is a multiplier on the interfacial drag value that is computed according to the vertical flow regime map. It is specified on an individual cell basis, and 1.0 is the default value. The results obtained for some of the ORNL-THTF tests with a value of 1.0 are presented in Table 154-2-2. The series of ORNL-THTF tests was executed using a variety of YDRAG values in the simulated core region. Table 15-4-2-3 contains the results obtained when a YDRAG value of 0.8 is specified. With YDRAG equal to 0.8, the amount of level swell is reduced. In order to quantify the appropriate YDRAG for each test level swell prediction to match the data, further simulations were performed at YDRAG o\4384-non\4384-15awpd:lb-4033 15-5

                                                                                  -                     l~~~~~~~~

values of 1.2, 0.65, and 0.5. Based on these results, the YDRAG value for each test that enables the prediction to match the data is shown in Table 154-2-4. The average YDRAG for the set is 0.79. Bundle Uncovery Tests (YDRAG = 0.8) Figures 154-24 to 154-2-9 show the results of WCOBRA/TRAC-SB simulations with YDRAG set to 0.8 of the bundle uncovery tests, I to N, in which one-third to one-half of the core is uncovered. Results for the bundle uncovery tests are shown in comparison with the predicted values of axial void fraction profiles; in the predictions, some variation in the void fractions was observed. In general, WCOBRA/TRAC predicts the void fraction reasonably well in the lower half of the bundle. In the upper half, the void fraction tends to be overpredicted. The cladding temperatures predicted and the heat transfer/void fraction relationship as modelled in WCOBRA/TRAC-SB are discussed in Volume 4 of this document. Level Swell Tests (YDRAG = 0.8) Figures 154-2-10 to 154.2-15 show the results of WCOBRAIRAC simulations with YDRAG set to 0.8 of the six level swell tests, AA to FF, in which the mixture level is at the top of the bundle. Because these tests had no uncovery, a cladding heatup did not occur. In general, the predicted and measured void profiles were in good agreement for this test series. As in the bundle uncovery test simulations, WCOBRA/TRAC-SB tends at times to overpredict the void fraction in the upper half of the bundle. Also, at times the void fraction profile is not smooth in the upper part of the bundle (tests AA and BB). Overall, however, agreement between the predictions and the measured profiles is reasonable. 15-4-2-5 Summary and Conclusions A WCOBRATRAC-SB mixture level was determined for each of the 12 ORNL-THTF steady-state tests simulated. The mixture level was defined as the elevation where a = 0.9 at a sharp gradient in the predicted void fractions, based on a linear interpolation between two continuity o:\4384-non\4384-15a.wpd:lb-04033 15-6

cells. Table 154-2-3 lists the mixture level and the collapsed liquid level at steady-state for each test with YDRAG = 0.8. Figure 15-4-2-16 shows a comparison of the predicted and measured mixture level at YDRAG = 0.8; Figure 154-2-17 compares the collapsed liquid level. In general, the agreement for two-phase mixture level is good. The poorest agreement is for test 3.09.1ON, which is a test at the lowest power at high pressure. Taken together, the summary figures show that WCOBRAJTRAC-SB tends to predict mass in the rod bundle (at YDRAG = 0.8) well compared with the experimental data. The average misprediction is small, and there is not a great deal of scatter. This indicates that the correlations affecting mixture level have a small bias and the uncertainty is also small. The predicted and measured void profiles are in reasonable agreement for both the uncovery test series and the level swell series. This supports the premise that the models for interfacial drag and bubble rise are well behaved. The simulation of ORNL tests at different values of YDRAG was conducted to produce a set of results that could be used to determine heated core interfacial drag multipliers for a small LOCA calculation in a PWR. The range of YDRAG values was sufficient to bound the data for low mixture level swell values. No modification to the core interfacial drag (YDRAG = 1.0) was found to overpredict the level swell; the multiplier of YDRAG = 0.8 produced improved results on average. YDRAG = 0.8 is the reference value for PWR core calculations. The set of multipliers that forces the code to match the data level swell was also identified. The minimum value for this set is YDRAG ,, = 0.503, and the maximum is YDRAG,,, = 1.169; the average value was YDRAG,,, = 0.79. Because the number of tests simulated is low (12), the multipliers in these simulations are combined with those from other level swell tests (G-1) to obtain a YDRAG distribution for application to the core region in the small break LOCA analysis of a PWR. o.4384-non\4384-15a.wpd:lb-04033 15-7

I Table 154-2-1 ORNL-THTF Test Simulation Matrix Data Data Mixture Collapsed Pressure Rod Power Level Liquid Level Test No. (psia) (kWlft) (ft) (ft) Bundle uncovery tests 3.09.10I 650 0.68 8.60 4.39 3.09.1OJ 610 0.33 8.10 5.31 3.09.10K 580 0.10 6.98 5.31 3.09.1OL 1090 0.66 9.02 5.77 3.09. IOM 1010 0.31 8.60 6.20 3.09.10N 1030 0.14 6.98 6.10 Level swell tests 3.09.10AA 590 0.39 11.23 6.56 3.09.10BB 560 0.20 10.85 7.61 3.09.10CC 520 0.10 11.80 9.45 3.09.10DD 1170 0.39 10.61 7.84 3.09.10EE 1120 0.19 11.40 9.35 3.09.10FF 1090 0.098 10.61 9.51 o:\4384-non\4384-15a.wpd:b-4033 15-8

I Table 15-4-2-2 Summary of ORNL-THTF Simulation Results Data Code Data Code Rod Mixture Mixture Collapsed Collapsed Pressure Power Level Level Liquid Level Liquid Level Test No. (psia) (kW/ft) (ft) (ft) (ft) (ft) 3.09.10J 610 0.33 8.10 7.62 5.31 4.92 3.09.10K 580 0.10 6.98 6.97 5.31 5.23 3.09.10AA 590 0.39 11.23 11.54 6.56 6.12 3.09.1OBB 560 0.20 10.85 10.10 7.61 6.88 3.09.10DD 1170 0.39 10.61 10.11 7.84 7.35 o:\4384-non\4384-15a.wpd:1b-04033 15-9

Table 15-4-2-3 .Li Summary of ORNL-THTF Simulation Results With YDRAG = 0.8 Data Code Data Code Rod Mixture Mixture Collapsed Collapsed Pressure Power Level Level Liquid Level Liquid Level Test No. (psia) (kW/ft) (ft) (ft) (ft) (ft) 3.09.10I 650 0.68 8.60 8.61 4.39 4.36 3.09.1OJ 610 0.33 8.10 7.70 5.31 5.07 3.09.10K 580 0.10 6.98 6.97 5.31 5.34 3.09.1 OL 1090 0.66 9.02 9.42 5.77 5.58 3.09.1OM 1010 0.31 8.60 8.42 6.20 6.52 3.09.10N 1030 0.14 6.98 6.16 6.10 5.21 3.09.1OAA 590 0.39 11.23 10.97 6.56 6.51 3.09.10BB 560 0.20 10.85 10.11 7.61 7.34 3.09.10CC 520 0.10 11.80 >12.0 9.45 9.40 3.09.1ODD 1170 0.39 10.61 10.21 7.84 7.64 3.09.10EE 1120 0.19 11.40 >12.0 9.35 9.73 3.09.10FF 1090 0.098 10.61 >12.0 9.51 10.33 o:\4384-non'4384-15a.wpd:lb04033 15-10

Table 15-4-2-4 YDRAG Values to Match ORNL-THTF Data Pressure Rod Power Test Number (psia) (kW/ft) YDRAG 3.09.10AA 590 0.390 0.827 3.09.10BB 560 0.200 0.908 3.09.10CC 520 0.100 0.698 3.09.10DD 1170 0.390 0.881 3.09.10EE 1120 0.190 0.752 3.09.10FF 1090 0.098 0.635 3.09.10I 650 0.680 0.779 3.09.10J 610 0.330 0.840 3.09.10K 580 0.100 0.871 3.09.10L 1090 0.660 0.503 3.09.10M 1010 0.310 1.169 3.09.10N 1030 0.140 0.61 o:\4384-non\4384-15a.wpd:Ib-04033 15-11

ORNL-DWG 77-5718D 14.08 in.) IL .0.104 m- 4 I F-- I I 0. 501 i n.)

                     @    UNHEATED RODS HEATED ROD DIAMETER - 0.95 cm (0.374 in.)  /0.501 in.)

UNHEATED ROD DIAMETER - 1.02 cm (0.401 in.) Figure 15-4-2-1. Cross Section of the ORNL-THTF Test Bundle o:W384-nonX438415a.wpd:lb-04033 15-12

OML-VG 3n-288 ETD SPACER GRID T/C ROD T/C DESIGNATiON LEVEL LEVEL in.) 144-942 POWER 3f4 DISTRIBUTION F8- -4 f7 138 F6- -136 134 F5 TE296X F4-  ::z - 32 F3 129

r-. -126 F2- 123 F1 .Z.

F- - 19 I ES 1?7

                                                                      -     114 E7-                                    112 E6 TE295X                            C==Z--          -   t08
                                   £3                                           -905
                                                                      -102
                                            £2-        --

Z.=_ I1-E4 -- r-- 99 0

                                                                      -95 0

TE294 X _ _ _84 m ~ 4

0. uJ

_-,-72 Y~ __-,j -~ 64 6 2 H - d-TE293X - 60

                                                                      -58 U   _   _                 -                   56 C -              =--    -    48 TE292X                                                         36 8-~~~~-1
                                                                      -25

_ _ _ _ _ _ _a - == -a TE291x 12 A-

                                                                      -1       12               0-Figure 154-2-2. Axial View of the ORNL-THTF Test Bundle o:\4384-non\4384-15awpd:1b04033                               15-13

I a,c Figure 154-2-3. WCOBRA/TRAC Model of the ORNL-THTF o:W384-non\4384-15a.wpd:lb-04033 15-14

ORNL Test II Results for YDRAG = 0.8 AL 413 0 0 WCOBRA/TRAC-SB 3 F YVALUE 1 0 0 Test II Doto 1

                  .8 I

0 c L- .4

                  .2 0

Elevation Along Heated Rod (ft) Figure 15-4-2-4. Comparison of Predicted and Measured Void Profiles for ORNL-THITF Test 3.09.101 o.\4384-non\438415a.wpd:lb-04033 15-15

I ORNL Test JJ Results for YDRAG = 0.8 AL 392 O O WCOBRA/TRAC-SB a H YVALUE O O Test JJ Data

                   .8 C .6-0
              -)

LL

             .5
             >4-Elevation Along Heated Rod (ft)

Figure 15-4-2-5. Comparison of Predicted and Measured Void Profiles for ORNL-THTF Test 3.09.10J o:\4384-non\4384-15a.wpd:lb-04033 15-16

ORNL Test KK Results for YDRAG = 0.8 AL 599 0 0 WCOBRA/TRAC-SB p C YVALUE I 0 0 Test KK Data

                    .6 0

1* I _ I

                    .2 0          2        4            6        8        10     12 Elevation Along Heated Rod (ft)

Figure 15-4-2-6. Comparison of Predicted and Measured Void Profiles for ORNL-THTF Test 3.09.10K o:\4384-non\4384-15a.wpd:lb04033 15-17

I ORNL Test LL Results for YDRAG = 0.8 AL 307 0 O WCOBRA/TRAC-SB I

• YVALUE I 0 0 Test LL Data I

I - c .6 0 0 Li-

                           *2
                                ~~I        _ I  I      I  I   I   I  I I    I  I I     I I I 0                    4            6        8         lo        12 Elevation Along Heated Rod (ft)

Figure 15-4-2-7. Comparison of Predicted and Measured Void Profiles for ORNL-THTF Test 3.09.10L o:\43S4non\4384-15a.wpd:lb-04033 15-18

ORNL Test MM Results for YDRAG = 0.8 AL 475 O O WCOBRA/TRAC-SB I H YVALUE O O Test MM Data 1*

                       .8-c .6-0
              >0.      . -
                       .2 -

0-4 6 8 Elevation Along Heated Rod (ft) Figure 15-4-2-8. Comparison of Predicted and Measured Void Profiles for ORNL-THTF Test 3.09.10M o:\4384-non\4384-15a.wpd:lb-04033 15-19

ORNL Test NN Results for YDRAG = 0.8 AL 599 O O WCOBRA/TRAC-SB pU YVALUE O O Test NN Data 1 c- .6 o. CI.) 0

              -a. -
              >     .4
                    .2 0

Elevation Along Heated Rod (ft) Figure 154-2-9. Comparison of Predicted and Measured Void Profiles for ORNL-THTF Test 3.09.10N o:\4384-non\4384-15a.wpd:lb-04033 15-20

ORNL Test AA Results for YDRAG = 0.8 AL O O WCOBRA/TRAC-SB d d YVALUE 4919 O O Test AA Data o .6-0 0 L. Figure 154-2-10. Comparison of Predicted and Measured Void Profiles for ORNL-THTT Test 3.09.10AA o:\4384-non\4384-15awpd:lb4O4033 15-21

ORNL Test BB Results for YDRAG = 0.8 AL 499g O O WCOBRA/TRAC-SB

• I YVALUE O O Test BB Data I1-

                   .8 0

C .6 0 L.

                                                                                          .L LL-
              > .4
                   .2 0

Elevation Along Heated Rod (ft) Figure 15-4-2-11. Comparison of Predicted and Measured Void Profiles for ORNL-THTF Test 3.09.10BB o:\4384-non\4384-15a.wpd:1b04033 15-22

ORNL Test CC Results for YDRAG = 0.8 AL 307 0 0 WCOBRA/TRAC-SB a P YVALUE 0 0 Test CC Data 1.

                ,.8 C

0

                    .6-0 O.-',
                    .2 Figure 154-2-12.           Comparison of Predicted and Measured Void Profiles for ORNL-THTF Test 3.09.10CC o:%43S4-non\4384-15awpd:Ib-04033                  15-23

I ORNL Test DD Results for YDRAG = 0.8 AL 346 O O WCOBRA/TRAC-SB I H YVALUE O O Test DD Data 1

                    .8 a    .6 -

0 0 0 L. L.

              >     .4-
                    .2-O-   ,

4 6 8 Elevation Along Heated Rod (ft) Figure 15-4-2-13. Comparison of Predicted and Measured Void Profiles for ORNL-THTF Test 3.09.10DD o:\4384-non\4384-15a.wpd:lb-04033 15-24

ORNL Test EE Results for YDRAG = 0.8 AL 499 O 0 WCOBRA/TRAC-SB d P YVALUE O 0 Test EE Data I. z.- 0 0

              .n .6 L-
              > .4
                    .2-O- a I       ,o Elevation Along Heated Rod (ft)

Figure 15-4-2-14. Comparison of Predicted and Measured Void Profiles for ORNL-THTF Test 3.09.10EE o:\4384-non\4384-15a.wpd:lb-04033 15-25

I ORNL Test FF Results for YDRAG = 0.8 AL 49 9 0 O WCOBRA/TRAC-SB

• H YVALUE 1 0 O Test FF Data 1*

c .6 - 0 C-) U-Li-

              ~>.4
                     .2 O-      ~     I      I  I Elevation Along Heated Rod (ft)

Figure 15-4-2-15. Comparison of Predicted and Measured Void Profiles for ORNL-THTF Test 3.09.10FF o:\4384-non\4384-15a.wpd:lb-04033 15-26

ORNL Small Break LOCA Tests Steady State 3.09.10 I-N and AA-FF Tests l4 12-10-U)8-~ b EL6 4 / 2n m 4- / Measured Mixture Level, ft. Figure 154-2-16. Comparison of Predicted and Measured Mixture Levels for ORNL-THTF Tests, YDRAG=0.8 o:\4384-non\43S4-15a.wpd:lb-04033 15-27

ORNL Small Break LOCA Tests Steady State 3.09.10 I-N and AA-FF Tests 10~~~~~~~~~ Q0 cr A Ql) co C)

             -     4-,

2-0 2 4 6 8 10 Measured Collapsed iquid Level, ft. Figure 15-4-2-17. Comparison of Predicted and Measured Collapsed Liquid Levels for ORNL-THTF Tests, YDRAG=0.8 o\4384-nonX4384-15awpd:lb-04033 15-28

15-4-3 Simulation of G-1 Core Uncovery Tests 15-4-3-1 Introduction A series of core uncovery experiments was conducted in the Westinghouse Emergency Core Cooling System (ECCS) High Pressure Test Facility. Figure 15-4-3-1 shows a schematic of this facility. [

                                                       ]ax Figure 15-4-3-2 shows a schematic of the heater rod bundle.

The bundle was instrumented [ Ia,c The tests were performed for a range [ Ia.c Additional information on the test facility and the data for the G-l Core Uncovery Tests are in WCAP-9764 (1980). 15-4-3-2 WCOBRAITRAC Model of G-1 Test Facility Figure 154-3-3 shows the WCOBRA/TRAC model for the G-1 test bundle and loop. The heated bundle is modelled with [

                                                                                              ]a.c o:\4384-non43S4-15b.wpd:lb-04033               15-29

I [

                                     ]ax 15-4-3-3 Test Matrix for G-1 Uncovery Tests Eight of the tests run in the G-1 Uncovery Test series fall into the range of conditions expected in a small break LOCA in a typical PWR. Table 154-3-1 lists these tests. The tests selected [

154-34 Simulation of G-1 Core Uncovery Tests Each simulation was started with the initial water level covering the top of the test bundle (i.e., the initial liquid fraction in the test section was 1.0); channels above the top elevation of the test bundle were initiated as void. The initial water temperature was based on available fluid temperature measurements and was generally a few degrees subcooled in core and downcomer channels at the start of the test. The test was started when the power to the bundle was turned on at 0.0 seconds and simulation continued for several hundred seconds, depending on the length of the experiment. The main parameters of interest in the simulations are the uncovery times of both the 10-foot and the 8-foot elevations, and the amount of water present in the bundle when the uncovery occurred. The uncovery time is readily identified in the test data as the time when the thermocouples at a particular elevation began to rapidly increase in temperature. The test report listed the average void fraction below the measurement elevation as the parameter representative of the amount of mass in the bundle. This value was deduced from DP cell measurements. o:\4384-non'4384-15b.wpd:Ib-04033 15-30

15-4-3-5 Discussion of Results Typical simulation results are shown for run 63 in Figure 15-4-34. Run 63 [

                                 ]a.c The predicted void fraction in the test bundle increased rapidly at the start of the (YDRAG = 1.0)

WCOBRA/TRAC-SB run. While the lower half of the bundle was predicted to remain at relatively low void fraction, boiloff at the top of the bundle progressively caused the void fractions to increase and eventually become single-phase vapor. Early in the simulation, the [

       ]a    The collapsed liquid level predicted in the test section falls as indicated in Figure 154-34.

After 210 seconds, the void fraction in the channel at the 10-foot bundle elevation became single-phase vapor. The predicted cladding temperature at the 10-foot elevation for run 63, as shown in Figure 154-34, increases rapidly at this time. At the end of the simulation, the mixture level approaches the 6-foot elevation. The uncovery times at the 8- and 10-foot elevations are easily identifiable in Figure 154-34; uncovery at the 10-foot elevation occurs at 210 seconds and at the 8-foot elevation, 320 seconds. The heatup rate is greater at the 8-foot elevation because of the higher local power. The summary of results and the comparison to the data for run 63 and the other G-1 simulations are listed in Table 154-3-2. The uncovery times in the WCOBRATfRAC-SB predictions executed with YDRAG = 0.8 in the test bundle are gleaned from the review of the cladding temperature figures, as described above. Table 154-3-2 compares the predicted and measured uncovery times for the eight G-1 test simulations. The predicted uncovery time is less than that measured for some points, but the prediction is greater for others. The trend is for the low pressure cases to exhibit predicted uncovery times beyond the measured values and for the high pressure cases to underpredict the uncovery times. The table also compares the level swell predictions to the data at the time of core uncovery for the 10-foot and 8-foot elevations. Figures 154-3-5 and 154-3-6 present the data in Table 154-3-2 graphically. o.M4384-non\4384-15b.wpd:lb-04033 15-31

I The level swell is both overpredicted and underpredicted by ECOBRA/TRAC-SB with YDRAG = 0.8 specified for the test bundle. Therefore, cases with increased and reduced interfacial drag were investigated by analyzing the tests using YDRAG values of 1.0, 1.2, 0.65, 0.5, 0.4, and 0.3. WCOBRA/IRAC-SB predictions of uncovery time and level swell decrease as YDRAG is decreased. The YDRAG values at which WCOBRArfRAC-SB matches the G-1 test data level swell are shown in Table 154-3-3. Figure 15-4-3-7 compares the YDRAG values to match the measured level swell as a function of total bundle power for the tests. Consistent with the data, WCOBRATRAC-SB shows [

                                                                        ]ac 15-4-3-6 Summary and Conclusions In general, WCOBRA/TRAC-SB predicted the uncovery times and major trends in the G-1 Core Uncovery Test data with reasonable accuracy; predicted level swell values were both greater and less than the test data. The effect of variations in YDRAG were explored, and the set of multipliers that forces the code to match the G-1 test data level swell was identified. The minimum value for this set is YDRAG,,,, = 0.353, and the maximum is YDRAG,, = 1.121. The average value is YDRAG,,,, = 0.693.

The bias observed in the WCOBRAfIRAC-SB prediction of the G-1 tests is similar but somewhat higher than that observed in the ORNL-THTF simulations. Taken together, the ORNL-THTF and G-1 simulations provide the YDRAG distribution for the core region level swell in the small break LOCA uncertainty analysis of a PWR. o:\4384-non\4384-15b.wpd:lb-04033 15-32

Table 15-4-3-1 Core Uncovery Test Matrix ac 4 4 4- 4 4 4. 4- 4 4 4 4 4 4 4 4 4

                     £          I               .L1           I   L o\4384-non\4384-15b.wpd:lb-04033               15-33

I Table 15-4-3-2 \.Li G-1 Simulation Results Summary, YDRAG = 0.8 a,c I T 7 1 7 I I I I I 7 7 t I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ l~~ l - l (a) The specific elevations reported in the WCOBRAITRAC-SB runs are 10.03 ft. and 7.90 ft., respectively. o:\4384-nonW4384-lb.wpd:Ib-04033 15-34

Table 15-4-3-3 YDRAG Values to Match G-1 Level Swell Data a,c I + 1

4. 4 4.

4. 4. 9

4. 4. 9

4. 4. 9

4. 4. I o:\4384-non\4384.-15b.wpd:Ib-04033 15-35

a,c Figure 15-4-3-1. Westinghouse ECCS High Pressure Test Facility (G-1 Loop) o:43S4-non4384-15b.wpd:Ib-04033 15-36

a,c Figure 154-3-2. G-1 Core Uncovery Test Heater Rod Bundle o\4384-non\4384-15b.wpd:b-04033 15-37

I a,c Figure 1543-3. WCOBRA/TRAC Model of the G-1 Test Bundle o-\4384-non\434-15b.wpd:lb-04033 15-38

                              ***G-l     UNCOVERY TEST 63 Temperoture (F)

TCLAD 1 67 312 ELEV. 10. FT.

         - -- - TCLAD                    1         54      309 ELEV.       8. FT.

Collopsed liquid Level (ft) L0-LEVEL 1 0 0 COLLAPSED lI0. LEVEL 900 8 15

                                . ~~ ~~~      I         .      . I                    -J0D
     <-'700                           /.        "s.                    I 0~

Q)  %~ ~ ~ "  : H-

                                                     . I     .       I   .

6.5 (In

                                                                                       -Q 60 I

250 300 350 Time (s) Figure 154-34. Collapsed Liquid Level and Predicted Cladding Temperatures at the 8- and 10-Foot Elevations, G-1 Run 63 o\43&4-non\4384-15b.wpd:1b4033 15-39

I 800 600-C', a) D 40-C200- / Q~ 0 200- 0 a 0 200 400 600 Measured Heat Up Time, sec. Figure 15-4-3-5. Comparison of Predicted and Measured Uncovery Times, YDRAG=0.8 o:\4384-non\43S4-15b.wpd:lb-04C33 15-40

                   .7
                   .6 S
                   .5 (I)
            - a, .4 CD
            -j au
                   .2
                   .1 0

Measured Level Swell Figure 15-4-3-6. Comparison of Predicted and Measured Level Swells at Uncovery, YDRAG=0.8 o:4384-non'4384-ISb.wpd:1b-04033 15-41

a,c Figure 15-4-3-7. WCOBRA/TRAC-SB YDRAG Value to Match Measured Level Swell Versus Bundle Power 0 4384-non\4384-15b.wpd:Ib-04033 1542

a,c Figure 154-3-8. WCOBRA/TRAC-SB YDRAG Value to Match Measured Level Swell Versus Pressure o:\4384-non4384-15b.wpd:lb4033 15-43

I 1.2 I 1-I s8-I 4 4 re .6-0 4

                   .4-4
                   .2-0            Il    I  I  1 I ~~  I
                                                       ~I  I
                                                               ~ I
                                                                   ~~~~

I I I I u 10 Elevation, ft. Figure 15-4-3-9. WCOBRAJTRAC-SB YDRAG Value to Match Measured Level Swell Versus Bundle Elevation o:\4384-non\4384-15b.wpd:b-04033 154

15-4-4 GE Vessel Blowdown Tests 154-4-1 Introduction The GE Vessel Blowdown Facility is designed to study basic phenomena such as void fraction distribution and transient liquid-vapor level swell during blowdown. A description of all the tests performed is given in NUREG/CR-1899 (Findlay and Sozzi, 1981). The blowdown tests were performed in a cylindrical carbon steel vessel. The vessel was a two-piece unit that could be separated at a pair of flanges located near the center of the vessel. The cylindrical portion of the vessel was constructed from Schedule 80 pipe, 12 feet long with an inside diameter of 1 foot. Elliptical heads were welded onto the ends of the pipe to create the vessel. The total vessel volume was 10 cubic feet, and the total height was 14 feet. There were five calorimetric heater rods, 1 inch in diameter and 2 feet high, in the bottom of the vessel to heat the water. The steam exhaust was located at the 13-foot elevation with an orifice that was captured in a flange. The orifices used to control the tank blowdown rate were plates with the prescribed hole machined without a chamber. The orifice was located close to the vessel in a 2-inch Schedule 80 pipe. Figure 15-4-4-1 is a scaled drawing that shows the vessel, its penetrations, the blowdown line, and a suppression pool where the blowdown effluent was discharged. A 3/4-inch thick perforated plate (containing 109 holes, 9/16-inch diameter), designed to provide an internal flow restriction, was installed between the main vessel flanges at the mid-elevation during some of the tests. The resistance of the plate was varied by plugging a selected number of holes. Orifice plates with different flow areas were used in the blowdown line to limit the blowdown flowrate and vary the vessel depressurization rate. Figure 15-44-2 shows the instrumentation arrangement used to measure three basic parameters: pressures, pressure differences, and temperatures. Vessel pressure and differential pressures were measured using strain-gauge pressure transducers, and temperatures were measured using Iron-Constantan thermocouples. The transient void fraction and the mixture level were calculated from differential pressure measurements. The vessel was initially filled with demineralized water and boiled at atmospheric pressure for approximately 30 minutes to liberate any dissolved gas in the supply water. A vent at the top of the vessel was then closed, and the water was heated to establish the initial conditions (which o:\4384-non\4384-1Sb.wpd:lb-04033 15-45

I were a nominal pressure of 1000 psi and 545°F). Actual initial conditions for each test are given in the test matrix in Table 15-44-1. With the facility initially heated and pressurized, several top-break blowdown tests were conducted using different-sized orifice plates to vary the blowdown transient. The tests also varied the open area of the resistance plate at the vessel mid-plane. 15-4-4-2 WCOBRAJTRAC Model for GE Vessel Blowdown Tests The WCOBRA/TRAC model of the GE Vessel Blowdown Facility is shown in Figure 1544-3. The test vessel itself is modelled [

                                                                     ].C Actual dimensions of the orifice were used in the modelling of the flowpath to the break.

15-4-4-3 Test Matrix for Simulations Table 15-44-1 lists the seven tests in the small break test series. Each was simulated with WCOBRAJTRAC-SB. 1544-4 Simulation of GE Vessel Blowdown Tests The results of the WCOBRAIIRAC-SB simulations of the Vessel Blowdown Tests are compared to experimental data in Figures 1544-4 to 15-4417. For each test, a comparison between the predicted and measured vessel pressure is presented; the two-phase level predicted and measured values are also presented. For the WCOBRAITRAC-SB prediction, the two-phase -LI o:\4384-non\43S4- 1Sb.wpd:I b-04033 15-46

level was defined as the elevation where a void fraction of 0.95 defines a sharp gradient between adjacent hydraulic cells. In general, the comparison between predicted and measured vessel pressure shows the code underpredicted the measured pressures. Table 154-4-2 compares the WCOBRA/TRAC-SB model prediction of several parameters with test data. The predicted mixture levels for the tests vary from being overpredicted to being underpredicted. The variation of mixture level with time is generally well predicted. Overall, the void fraction in the two-phase mixture tends to be overpredicted by WCOBRAJTRAC-SB. 15-44-5 Effect of Interfacial Drag Multiplier The simulations of the GE Vessel Blowdown Tests were also rerun with WCOBRAIRAC-SB to investigate the impact of the interfacial drag multiplier (YDRAG) on the prediction of the two-phase level and pressure. Simulations were made with YDRAG = 0.65 to compare the effect of interfacial drag on the results using WCOBRA/TRAC-SB. The results indicate a small pressure effect of using a lower YDRAG value. Figures 15-4-4-18, -19, and -20 compare the YDRAG = 1.0 (solid curve) to YDRAG = 0.65 (dashed curve) results. In all cases, the depressurization to 400 psi is quicker with YDRAG = 0.65 because steam is passing to the exit BREAK more efficiently. These effects are expected because with a lower interfacial drag coefficient, there is less force acting to raise the two-phase level in the vessel. For void fraction predictions, the difference is again small; in some cases, the void fraction is shifted lower with a YDRAG = 0.65 at a given elevation, and in others, it shifts higher. Overall, there is little difference between the simulation results of void fraction for a given test caused by varying YDRAG. 154-4-6 Summary and Conclusions The results of the GE Vessel Blowdown Test simulations tend to confirm the results for the ORNL-THTF Uncovered Bundle Tests and the G-1 Core Uncovery Test simulations, which showed that WCOBRA/TRAC tends to underpredict the amount of mass present in a given test. The results of the GE Vessel Blowdown Tests using WCOBRA/TRAC-SB indicate that varying YDRAG appears to make little difference relative to the pressure and void fraction predicted. Therefore, the blowdown flashing behavior predicted by the code as YDRAG is varied within the core during plant sensitivity studies is judged to be minimally affected; the YDRAG values in the o:\4384-non\4384-15b.wpd:IbO4033 15-47

I core may be specified based on the effect on the fluid condition at the time of core uncovery 'Ij during the boiloff phase of a small break LOCA, in the knowledge that the blowdown phase behavior is not significantly impacted by the selection. In particular, a core YDRAG value of 0.8 is employed for both the integral test and PWR simulations. o\4384-non\4384-15b.wpd:lb-04033 15-48

Table 15-4-4-1 Summary of Test Parameters for Small Blowdown Vessel Steam Blowdown Tests Restriction Plate Initial Conditions (9/16 in. diameter Test No. Orifice Size (in.) holes) Pressure (psia) Level (ft) 8-21-1 3/8 109 holes 1015 8.89 8-25-1 /2 109 holes 1020 8.82 8-28-1 1 109 holes 1015 8.76 9-1-1 3/8 77 holes 1014 8.75 9-15-1 3/8 55 holes 1015 8.74 1004-3 3/8 No plate 1011 10.4 1004-2 7/8 No plate 1011 10.5 o:\4384-non\4384-15b.wpd:1b-04033 15-49

I Table 15-4-4-2 Characterization of WCOBRA/TRAC-SB Results Versus Test Data Test P(t) L(t) ALP(1) ALP(2) AILP(3) ALP(4) ALP(5) ALP(6) 8-21-1 Low High High OK High NC OK OK 8-25-1 Low OK Low OK High NC OK OK 8-28-1 Low NC OK High High Low High OK 9-1-1 OK High High OK Low None High OK 9-15-1 OK Low High Low High NC High OK 1004-3 Low Low High OK OK High High OK 1004-2 Low Low Low High High High OK OK Key: High = Code overpredicts data. None = No comparison. Data are not available. Low = Code underpredicts data. P(t) = Pressure transient prediction OK = Prediction agrees well with data for most of the transient. L(t) = Mixture level transient prediction NC = Not clear. No consistent trend found in the comparison. ALP(N) = Void fraction prediction at elevation N of Figure 15-4-3 o:\4384-non\4384-l5b.wpd:lb-04033 15-50

PRESSURE VESSEL TRUCTURE LINE SUPPORT VN VALVE 5CASSEMBLY e4EATER CONNECTIONS SUPPRESSION TANX Figure 15-4-4-1. Small Blowdown Vessel o\4384-non\4384-15b.wpd:1b-04033 15-51

14 -_ BLOWDOWN 12t - P ORIFICE loh - A 8ft - A APU- T VESSELRESTRICTION SLOWDOW

          ,_tt-Ott  -

20t-0~ ~~ _ _{ PRESSURE VESSEL SFta ESSW )N POOL Figure 15-4-4-2. Small Blowdown Vessel Instrumentation o:\4384-non\4384-15b.wpd:lb 04033 15-52

a,c Figure 1544-3. WCOBRAITRAC Model of the GE Vessel Blowdown Facility o\4384-non\4384-15b.wpd:lb-04033 15-53

                                                                                     ~ ~~~ ~ ~ ~ ~

Thu.. April 13. 2000; 08:38:03 AM EDT UNCONFIGURED!

                                  *** GE Test 8-21-1            ***

p 4 3 0 PRESSURE

      *

• YVALUE 1 0 0 DATA PRESSURE 1200-10 00 . ................ .. . .......... ...........

c1800 - . . . . . . . . . . . .. . ....... 00 .. .. . 23 00- . O. 400-

                     ....... ..    ~     .~     ...

0 0 50 100 150 200 250 Time (s) Figure 15-4-4-4. Comparison of Predicted and Measured Vessel Pressure, Test 8-21-1 o:4384-nonX4384-15b.vTd:1bW4O33 15-54

                                            *** GE Test 8-21-1              ***

YVALUE 1 0 0 COLUMN 00002 U

• YVALUE 1 0 0 DATA SWELL 12-r 11 - - _ ~~~~~~~. . . . . . . . . .....

10 - 9-CD ,..  :" .. . . . . . . . . . . . . . B-CD 7-6 - - - - --- - . . . . . 6 20 40 60 10 i4o T ime . sec Figure 1544-5. Comparison of Predicted and Measured Vessel Level, Test 8-21-1 o:\4384-non\4384-15b.wpd:1b04033 15-55

Tue.. May 30. 2000; 11:29:36 AM EDT UNCONFIGURED!

                                          *$* GE Test 8-25-1 ***

p 4 3 0 PRESSURE

      *

• YVALUE 1 0 0 DATA PRESSURE 1200 100 0 . . . . . . . . ... . .. . .. . .. . .. ..I. . ... .. .. .. . . .. . ... . .. .. .. .. . ..

    ' 800 -        .    . . .. ..     . . . . . . . . ........ .............. ....... .I ..................

Q)600 - . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I. . . . . . . . . . . CD 00400-. ....... .. . . . . . .. . . . ................... 200 0- , , , , I 0 50 100 150 200 Time (s) Figure 15-4-4-6. Comparison of Predicted and Measured Vessel Pressure, Test 8-25-1 o.W384-nonW384-15b.wpd:1b4033 15-56

                                                 ***         GE Test 8-25-1 ***

YVALUE 1 0 0 COLUMN 00002 a

• YVALUE 1 0 0 DATA SWELL 12 11 > ~~ ~ ~ ~ ~ ~~ . . . . . .

10 -~~~~~~ 9 a) 8 a) 7 I ~ 6

                                  - .--. -. -. -. -. -.-. - -. -. . .. . . . . .                                   . . .   . . . .  . . . . . I. . . . .~~~~~I 5

0 20 40 60 100 120 T ime . sec Figure 15-4-4-7. Comparison of Predicted and Measured Vessel Level, Test 8-25-1 o-\4384-non\4384-15b.wpd:b-44033 15-57

Tue.. May 30. 2000; 12:31:45 PM EDT UNCONFIGURED!

                                           *** GE Test 8-28-1                  ***

p 4 3 0 PRESSURE

      *

• YVALUE 0 0 DATA PRESSURE 1200 1000 - . . . . . . . . . . . . . ..

C. 0-800 -

                                 ~..    .    . .  . .  . . . . . . . .   .

600 - a) En

                            ....          .            \~.....         ...........   . .   . . . . . . .

cn 400 - CD 200 - I . . . 0 -p 0 10 io 30 40 50 Time (s) Figure 15-4-4-8. Comparison of Predicted and Measured Vessel Pressure, Test 8-28-1 o:\4384-non\4384-15b.wpd:b- 0403 3 15-58

                                       *** GE Test 8-28-1                       ***
                         'YVALUE                 I        0          0 COLUMN 00002 X

• YVALUE 1 0 0 DATA SWELL 13 -

12 - 11 -

                     ~~~~~.       .    . . . . . . . .     . . .

10 - 9- I .. ~~~~. . . . . . . ...... ... 8- .U~~ _ 6I

                                                                                          .a 5 2    _    I      I  *

• I T- r I

r

                                               !      I X

I

                                                                            '                  S 0                    10                    io                  30                   40 T i me.        sec Figure 15-4-4-9. Comparison of Predicted and Measured Vessel Level, Test 8-28-1 o:\4384-nonX4384-15b.wpd:1bO4033                           15-59

Tue., May 30. 2000; 11:36:39 AM EDT UNCONFIGURED!

                                     *** GE Test 9-1-1         ***

p 4 3 0 PRESSURE

      *

• YVALUE 1 0 0 DATA PRESSURE 1200 1000 - . . . . . . . . . . . . . ... . . . ...

U5 800 - tO ..... 600 . ......... .............................................;.

   'n   400-. ... ... .. ..      ...........  ....... ...............................

20 0 . ...... .. ........ 0 5 10 15 20 25 30 Time (s) Figure 154-4-10. Comparison of Predicted and Measured Vessel Pressure, Test 9-1-1 oA4384-non\434-15b.wpd:Ib-04033 15-60

                                              ***   GE Test 9-1-1               ***

YVALUE 1 0 0 COLUMN 00002 U YVALUE 1 0 0 DATA SWELL 11.5

         *11 10.5 10
 -a aD    9.5         ....     . ~             p.           .       ...      ....               ..

9 _.. . . . . . . . . . . . . . . . .  : 8.5 . . . 0~~~~~~~~~~~~ 10 15 20 25 30 Time. sec Figure 15-4-4-11. Comparison of Predicted and Measured Vessel Level, Test 9-1-1 o:\4384-non\4384-15b.wpd:b-4033 15-61

I Tue., May 30. 2000; 11:35:02 AM EDT UNCONFIGURED!

                               *** GE Test 9-15-1         **

p 4 3 0 PRESSURE

        *

• YVALUE 1 0 0 DATA PRESSURE 1200 -

1000 - ... ... ............... ..............

      'Q 800- -                  .....                                    . ..     . .

600- ... .................. . .. .......................... L. 400 . ....................... ............ ........................ 200-. ......................................... ... . .... 0 ~~10 20 304 Time (s) Figure 15-4-4-12. Comparison of Predicted and Measured Vessel Pressure, Test 9-15-1 33 o:\4384-nonW4384-15b.wpd:b-0 15-62

                                     ***   GE Test 9-15-1     ***

YVALUE I 0 0 COLUMN 00002 U U YVALUVE 1 a 0 DATA SWELL 11-5 - 11 - 10.5 _ 10 . . .

                                     ~~~....        .                         ...

Ci) 9.5 .

                                     ~~....         .                ........

IJ 0 40 ime sec Figure 15-4-4-13. Comparison of Predicted and Measured Vessel Level, Test 9-15-1 o:\4384nloIA43S4-15b.wpd:lbO04033 15-63

I Tue., May 30. 2000; 11:32:48 AM EDT UNCONFIGURED!

                                           *** GE Test 1004-3 ***

p 4 3 0 PRESSURE

        .

• YVALUE 1 0 0 DATA PRESSURE 1200 1000

                       ...  ~...            ..    :.      ..  ...        ... :.  .. ...    . .:.

1-800 0n 600 en U)

                                  .\.........................

Q 400 r2) 200-I I I 0- I l { 0 50 100 1.50 200 300 350 Time (s) Figure 15-4-4-14. Comparison of Predicted and Measured Vessel Pressure, Test 1004-3 o:\4384-non\4384-15b.wpd:l b-04033 15-64

                                      ***GE     Test 1004-3 ***

YVAL UE 1 0 0 COLUMN 00002

          *

• YVALUE 1 0 0 DATA SWELL 11.5 - -

11- .......... . . . . . . . . . . ..... I --  :~~~. 10 ........... W4 10 . ...........

                                                                                               .U.~~~~~~~~~~~~~~~

9.5 - .I 0) I I- _ I) 9.. 8-........ 8- F t ~,_

                                                       ~ ~ ~~~~

I I

• t , * -

0 20 40 do 10 T ime . se c Figure 154 4-15. Comparison of Predicted and Measured Vessel Level, Test 1004-3 ooS4384onon\4384-15b.wpd:lb-04033 15-65

Wed.. April 12 2000; 10:34:30 AM EDT UNCONFIGURED!

                                 *** GE Test 1004-2 ***

p 4 3 0 PRESSURE

      *

• YVALUE 1 0 0 DATA PRESSURE 1200-:

1000 . .. ...........  : .: ............:.' .

    *' 800 -       \.                     .

60 0 - ........ ...... ... ....... .... 0 a>40 0 - . .. .. . .. . . .. ... .. .. . .. .. .. . . .. .. . .. . . ... . . 200-.;...........  ;..........;;.. I3 BOO20 40 80 100 Time (s) Figure 15044-16. Comparison of Predicted and Measured Vessel Pressure, Test 1004-2 o:u4384-nonW4384-I5b.wp:lbNO33 1 5-66

                                          *** GE Test 1004-2 ***

YVAL UE 1 0 0 COLUMN 00002

         .

• YVALUE 1 0 0 DATA SWELL 10 -_

                         ~~~~~~.                         . . . . . . . . . ..           .  . . .  .

CD _- __ 78- - 7- _ .. .\ ....... .... .......... . . . . . . . . . . . U) I U) 5.- . . . . . .. . . . . . . 1 . - 4--

                     .     .        ,       . .                                    I      I      I      ~     I 0                    20                 40                          6                                                i6O Time.                  sec Figure 15-4-4-17. Comparison of Predicted and Measured Vessel Level, Test 1004-2 o:\4384-non\4384-15b.wpd:lb-04033                          15-67

see GE Test 8-28-1 *c p 4 3 0 Press.(YDRAG=1.0)

                - - - - p                    4       3      0 Press.(TDRAG=.65) 1 200
                         -~~~~~~~~~~
       .       1 000 V)

I--800 600 C/) 400

                                                                                        .1,.

(I, a) 200

                      - I -             II                                   II 0                                                            _

0 10 20 30 40 50 60 T ime ( s) Figure 15-4-4-18. Effect of YDRAG Multiplier on the Pressure Prediction, Test 8-28-1 "I o:\4384-non\4384-15b.wpdlb-04033 15-68

C.* GE Test 1 *.'

                     ~~~

p 4 3 0 Press.(YDRAG-1.0)

          ----       P                    4       3         0   Press.(YDRAG-.65) 1 200
  ._   1 000

_ 800

                              -  ~ ~~~~~~~~~                  -     --- - --            - ----

600 L. CD, 400 C- 200 0 - -- u

                       -- e- L  l I  1 50v
                                       -l t l 1 t l   -- r  1 tI Zuud.~~~~~~~~~

sa _.t* S-I I I ZDU T ime ( E;) Figure 1544-19. Effect of YDRAG Multiplier on the Pressure Prediction, Test 9-1-1 o:X4384-non\4384-15b.%pd:lb04033 15-69

see GE Test 1004-3 ...

                 ~~P                 4       3         0 Press.(YDRAG=1.0)
                 -    P              4       3         0 Press.(YDRAG=.65) 1 200
       -   1000

_800 600 U) U) 400 Ca) 200 0 -- I U I

                               -Xx~~

zu Itu a I I Iu I

                                                                   -l u I IIC I n1   I_
                                                                        -tuI1ou T ime         (

Figure 154-4-20. Effect of YDRAG Multiplier on the Pressure Prediction, Test 1004-3 o:\4384-non\438415b.wpd:1b0433 15-70

15-5 Summary and Conclusions This section considers mixture level swell tests in three different facilities. The ORNL-THTF simulations showed that WCOBRATRAC-SB tends to predict both the collapsed liquid level in the test bundle and the mixture level fairly well but in general overpredicts the level swell. The G-1 Core Uncovery Tests simulations showed the WCOBRA(rRAC-SB predictions of boiloff to a given elevation ranged both earlier than and later than the data. The code predictions of bundle level swell are both less and greater than that reported in the data, but most often are overstated. Finally, the simulations of the GE Vessel Blowdown Tests showed WCOBRArfRAC-SB, in general, overpredicts the reported voiding in the test vessel. The interfacial drag model contains a bias so that the mixture level is not "frothed" to the appropriate level. The simulated test facilities, on the average, overpredicted the void fraction in the bundle during flashing and/or boiloff, suggesting that the interfacial drag is too high.

                                                                                        ]ac The use of a nominal value for YDRAG of 0.8 in the integral test facility simulations and in PWR calculations is supported by the WCOBRA[IRAC-SB predictions of G-1 and ORNL-THTF.

15-6 References Anklam, T. M., et al., 1982, "Experimental Investigations of Uncovered Bundle Heat Transfer and Two-Phase Mixture Level Swell Under High Pressure Low Heat Flux Conditions," NUREG/CR-2456. Bajorek, S. M., et al., 1998, "Code Qualification Document for Best Estimate LOCA Analyses Volume I: Models and Correlations," WCAP-12945-P-A, Vol. 1, Proprietary. Findlay, J. A. and Sozzi, G. L., 1981, "BWR Refill-Reflood Program - Model Qualification Task Plan, NUREG/CR-1899. WCAP-9764, 1980, "Documentation of the Westinghouse Core Uncovery Tests and the Small Break Evaluation Model Core Mixture Level Model," Proprietary. o\4384-non'4384-15b.wpd:1b-04033 15-71

o\4384-non4384-15b.wpd:lb-04033 15-72 SECTION 16 LOOP SEAL CLEARANCE 16-1 Introduction The small break LOCA PIRT in Volume 1 of this document identifies the loop seal behavior as an important process affecting the evolution of the transient. This component, and its effect on the transient, is discussed in more detail below. The following sections assess the important phenomena occurring in the loop seal, the available experiments which quantify the phenomena, and the performance of WCOBRA/TRAC-SB in predicting the phenomena. During a small break LOCA, mass is slowly depleted from the system. Early in the transient, the pumps continue to run and the flow through the pump suction piping remains single-phase. After generation of a trip signal, the reactor automatically trips and subsequently the pumps. The system then enters a natural circulation phase. Pressures have fallen sufficiently to cause boiling in the fluid entering the hot leg, but the steam generator acts as a heat sink and the fluid entering the pump suction pipe is still nearly single-phase. Any bubbles that enter the pump suction pipe are carried through by natural circulation as illustrated in Figure 16-1(a). When the primary pressure approaches the secondary pressure, voids remain in the fluid as it enters the steam generator. As the loop mass flowrate decreases further, liquid begins to drain down both the uphill and downhill sides of the steam generator tubes. Natural circulation is terminated, and mixture levels form on both the uphill and downhill sides of the tubes. The levels then move downward as liquid drains and vapor rises as shown in Figure 16-1(b). Because there is no escape path for the steam generated in the core, except for some small bypass paths such as the upper head, the pressure in the region above the core (the upper plenum, the hot legs, and the steam generator tubes) rises and depresses the level in both the core and the downhill sides of the pump suction pipe. Eventually, the downhill side level reaches the top of the horizontal portion of the pump suction pipe, as shown in Figure 16-1(c), and vapor begins to escape into the pump and flow toward the break. At the onset of clearing, the fluid pressure in the downhill leg of the loop seal is about 3 psi higher than on the uphill side, due to the column of water from the horizontal leg to the pump outlet as shown in Figure 16-1(c). Because the volume of steam at this pressure is significant - in the steam generator tubes, hot legs, vessel upper plenum, and upper head - the steam flowing o.M4384-nonW4384-16.wpd:1b-04043 16-1

I through the pump suction becomes significantly greater than the core steam generation rate for a period of time (Kukita, 1990). This causes the loop seal to clear completely, not resealing until much later in the transient. As the steam flows through the pump suction, the flow regime is first a slug regime with significant amounts of water being entrained from the pump suction pipe as seen in Figure 16-2 and described by Tuomisto and Kajanto (Tuomisto and Kajanto, 1988). Eventually, a residual level of water will remain in the pump suction pipe. As the pressure in the system is relieved, the steam flow decreases to the core steam generation level. If this steam flow is low enough, water in the cold leg may begin to drain back through the pump and begin to fill the pump suction again as shown in Figure 16-1(d). Another potential source of loop seal refilling is the draining of condensed steam from the downhill side of the steam generators. Because there is no pressure driving force, the steam flow through the loop seal is quickly terminated when the water level reaches the top of the horizontal section and plugs the loop seal. The system pressure increases, and core and loop seal levels change once again as the loop seal plugging and clearing cycle is repeated (Kukita, 1990). 16-2 Important Physical Processes and Scaling Laws The onset of loop seal clearing is a function of the pressure difference across the loop seal, which depresses the level to the bottom of the loop seal and depends on the following factors:

• Core steam generation rate
• Bypass steam flow rate through vent paths
• Rate of accumulation of water in the pump suction pipe These factors are the result of processes that occur elsewhere in the system and are accounted for in other components (for example, the core steam generation rate is accounted for by ranging core power and core mixture level).

The loop seal clearing and refilling process is a function of the interfacial drag between the vapor and the water. The initial steam flow surge and the interfacial drag determine the rate at which water is expelled. The steam flowrate, in turn, depends on the loop pressure drop, of which the loop seal is a part. This determines how quickly the venting process takes place and the final water level in the horizontal section. The residual water and degree to which water is held up by steam flowing out of the pump suction pipe determine the rate at which the pump suction refills o:\4384-non\43B4-16.wpd:Ib-04043 16-2

and replugs. Based on these considerations, the following factors are considered to be important in the assessment of predictions of loop seal behavior:

• Overall loop seal pressure drop as a function of steam flow
• Liquid distribution in the loop seal as a function of steam flow Various experiments have shown that the basic physical process is controlled by two factors: the extent to which a stratified flow regime can be maintained in the horizontal leg of the loop seal and the degree to which liquid pushed into the downstream vertical leg can be entrained out of the loop seal. Figure 16-2 illustrates these processes.

Scaled loop seal experiments are discussed in the following sections to gain a better understanding of the loop seal behavior. These tests are used to highlight important physical and scaling features and are then compared with larger scale tests to confirm the indicated scaling trends. Finally, these tests are predicted using MYCOBRAJTRAC-SB to assess the models and correlations in the code. 16-2-1 Westinghouse Loop Seal Tests Westinghouse performed scaled U-tube experiments designed to examine the hydraulic behavior of a U-tube under conditions similar to those encountered during a small break LOCA. The vapor flow required to "clear" the U-tube was a specific focus of the tests. 16-2-1-1 Test Facility Description The tests were run in a plexiglass facility with air and water at atmospheric pressure. The facility, illustrated schematically in Figure 16-3, consists of a blower, a run of horizontal piping from the blower, a U-tube, and a catch tank. The pipe diameter chosen for the facility was 25 cm or 0.82 feet. This corresponds to approximately 1/3-geometric scale compared with a PWR, which has a pipe diameter of 2.58 feet. The air and water flowrates were scaled so that approximate similitude was maintained for the Froude number, shown to define the flow regime transition from stratified to intermittent and annular flow by Taitel and Dukler (Taitel and Dukler, 1976). Figure 16-4 shows the predicted flow regime transition using the Taitel and Dukler flow regime map for atmospheric pressure, 1/3-scale geometry, compared with the transition for 1000 psia, full-scale geometry. This figure indicates that the transition occurs at a higher vapor flux in the air-water o:\4384-non\43S4-16.wpd:lb-04043 16-3

tests. While better similitude could have been obtained with a smaller pipe, the chosen diameter also assures that the vertical pipes of the U-tube are sufficiently large so that any countercurrent flow limits (CCFL) that occur will not be affected by the pipe diameter. According to Richter (Richter, 198 1), the critical vapor flux for CCFL in pipes larger than approximately 2 inches in diameter depends only on pressure, not on pipe diameter. Pressure drop across the U-tube was measured. In the horizontal and in the downstream vertical sections, several independent measurements of void fraction were made using pressure drops, optical probes, and gamma densitometers. 16-2-1-2 Test Procedures Several tests series were performed, as described below: Limit Line Tests These tests were designed to obtain the water level in the horizontal portion of the U-tube, which produces significant water entrainment for a given air flowrate. This is equivalent in some ways to the CCFL limit and is termed the U-tube limit line. The tests were performed as follows:

1. Begin with an empty U-tube.

2. Start the air flow at the desired value.

3. Add water at the bottom of the horizontal pipe until significant entrainment into the catch tank is observed.

4. Terminate the water flow, and continue the test until the entrainment becomes negligible.

5. Measure flows and pressure drops.

6. Stop the air flow and measure the quiescent water level remaining in the pipe.

o:\4384-non\4354-16.wpd:lb-04043 16-4

          *    . Within Limit Line Tests These tests were performed at air and water flows inside the limit line established in the first phase with little or no entrainment. The tests primarily examined the interaction, if any, between the gas and the liquid at nonlimiting flows. The tests were run as follows:

1. Set the desired water level in the horizontal pipe with the air flow at zero.

2. Incrementally increase the air flow and take measurements.

3. Confirm that final level is approximately the same as the initial level.

In addition to flow and delta-p measurements, the appearance of the water level was observed. At low air flows, the water was either quiescent or small ripples were observed. At higher flows, the water began to drop near the upstream side of the U-tube. Next, droplets were observed forming and reaching the downstream elbow of the U-tube. At still higher flows, droplets began to reach the top of the downstream pipe, and finally, water was observed to stream upward in the downstream pipe.

• Optical Probe Tests These tests were performed similarly to the test series at conditions inside the limit line. Optical probes were used to measure the water level. These tests confirmed the delta-p measurements, later used to derive vapor fraction.
• Complementary Tests In some of the tests with high initial water level, oscillatory flow was observed.

These oscillations consisted of movements of water back and forth between the upstream and downstream elbows. Slugs of water momentarily filled the pipe, increasing the pressure drop across the U-tube. These slugs were then ejected from the U-tube. The tests were similar to the limit line tests except that continuous readings were taken during the oscillations and until steady-state was reached. o:\43&4-non\4384 16.wpd:1b4043 16-5

I

• Gamma Densitometer Tests These tests used a gamma densitometer to measure the mixture density inside the horizontal portion of the U-tube. The tests confirmed void fraction measurements based on delta-P.

16-2-1-3 Analysis of 1/3-Scale Test Results Figure 16-5 plots the normalized residual water level in the loop seal (H/D) as a function of the vapor volumetric flux (ii). The loop seal was completely cleared when gas velocities exceeded about 70 ft/s. At low gas flows, some hysteresis was observed; i.e., the residual water level remaining after the test depended on how the test was performed. The lower levels shown in Figure 16-5 were obtained when the test was started with an initial water level above the top of the horizontal leg. This configuration introduced level oscillations between both vertical legs. These oscillations caused additional water to be entrained from the loop seal. Figure 16-6 shows the results of tests performed under the limit line. The water level in the horizontal leg was relatively unaffected by gas flow until flows near the limit line were reached. The residual water level is an indication of the overall liquid mass contained in the loop seal as a function of gas flow, but does not represent the liquid distribution within the U tube during the tests. Figure 16-7 shows the average void fraction at the midpoint of the horizontal leg and in the downstream vertical leg during the test. At low gas flowrates, there is wide scatter in the measured void fraction in the horizontal leg and indications of significant liquid content in the downstream vertical leg. The void fraction in the vertical leg was inferred from a delta-P cell spanning the vertical leg (P, 5 in Figure 16-3). The flow regime in the vertical leg is, therefore, likely to be that depicted in Figure 16-2C with a low void fraction region at the bottom of the pipe, and a high void fraction at the top of the pipe. The measured void fraction in the horizontal leg is more representative of the void fraction in the elbow and bottom of the vertical leg. Figure 16-8 compares the measured void fraction in the horizontal leg with the void fraction calculated from the residual water level. This plot shows that at higher gas flows, nearly all the water retained in the loop seal resides in the horizontal leg. At lower gas flows, the same is still generally true, although the storage in the vertical leg is more evident due to the higher measured void fraction during the test. o:\4384-non\438416.wpd:1b04043 16-6

Figure 16-9 shows the measured pressure difference between the upstream and downstream exits of the U-tube. As water collects in the downstream vertical leg, the pressure difference increases. The basic processes occurring during these tests can be explained in terms of several hydrodynamic limits applied to both the horizontal and vertical legs. Figure 16-10 shows the horizontal leg average void fraction as a function of i*, defined as: Jg = (16-1) (pj-pg)gD Pg where D is the pipe diameter. The loop seal behavior can be explained in terms of three regimes, bounded by the limit lines shown in Figure 16-10. These regimes are described in the following paragraphs. Regime III: Droplet Entrainment Ishii and Grolmes (Ishii and Grolmes, 1975) describe entrainment in horizontal cocurrent flow as the stripping of drops from the tops of waves. They describe four mechanisms: 1) the shearing off of the top of roll waves by the turbulent gas flow, 2) the undercutting of the liquid film by the gas flow, 3) the gas bubbles bursting at the liquid-vapor interface, and 4) the liquid impingement on the liquid-vapor interface. The only mechanism of the four expected to be of significance to loop seal clearing is the shearing off of the tops of the waves, which Ishii and Grolmes state is valid for liquid Reynolds numbers greater that 160 in horizontal cocurrent flow. Assuming for the moment that the WCOBRA/TRAC-SB calculations reported herein are reasonably valid, the horizontal (GAP) liquid velocities during entrainment vary from less than 1 ft/s to several ft/s. The liquid film thickness is then estimated to be from about 0.5 inch at 1000 psia to about 1 inch at 43 psia, and about 1.5 inches for air-water at atmospheric pressure. For roll wave entrainment, Ishii and Grolmes provide two correlations based on the Reynolds number. o\4384-non\4384-16.wpd:lb-04043 16-7

I For Reynolds numbers greater than [ ]a,c the following applies: ____g 0 8 fUgt, 2 N 1 for N < - Pu 15(16-2) IIUg' g Ž0.1146 for N > 1 Inequality 16-2 is valid in the rough turbulent regime. For Reynolds numbers below [ a (laminar-turbulent transition regime), Ishii suggests the following correlation: PjUg .2 Ž11.78N- Re, for N ( p1 Pi 15 (16-3) plug 2Ž1.35Re, 3 for N,> - The Reynolds number is calculated based on the liquid film thickness where ug is the minimum gas velocity for entrainment to occur. The gas velocity can be represented in terms of a more easily measured velocity, the superficial gas velocity (): ig =aU (164) As droplets are entrained into the downstream vertical leg, they are ejected out of the loop seal because the gas flow exceeds the CCFL in the vertical pipe. This limit is described further in Regime II. o:\4384-nonW384-16.wpd:1b-04163 16-8

Regime II: Wave Instability and Vertical CCFL Regime In this regime, the water level in the horizontal leg is govemed by the stability of waves on the stratified interface. If these waves grow, they could span the pipe, as illustrated in Figure 16-2(b), and cause a slug of water to be pushed into the downstream vertical leg as seen in Figure 16-2(c). The water level or void fraction at the onset of wave instability was characterized by Taitel and Dukler (Taitel and Dukler, 1976). They proposed the following criterion: 2 3

            .*2   (1-h)
           -*2    -     )a
                    -da/dh H=                                                                            (16-5)

D a = ![cos1(2A-l)-(27-1) I-(2h-1)2] where the relationship between void fraction and level is deternined by the pipe geometry. The third line of Equation 16-5 describes the relationship for a circular pipe. The limit line shown in Figure 16-10 is Equation 16-5, solved for a as a function of j. The data follow this linit in Region II. If the gas flow in the downstream vertical leg still exceeds the flooding linit, then any water pushed into the vertical leg by wave instabilities will be ejected from the loop seal. The CCFL limit line shown in Figure 16-10 is based on the critical velocity for liquid holdup (known as the Kutateladze number) developed by Pushkina and Sorokin (Pushkina and Sorokin, 1969) for large diameter pipes: UCCFL = 3.2 g (16-6) Pg Thej, versus a relationship is determined as in Equation 16-4. o:\4384-non\4384.16.wpd:1b-04043 16-9

Regime I: Slug/Oscillatory Regime When gas velocities are reduced below the CCFL, water pushed into the vertical leg collects there and can fall back. This leads to a low void fraction, chaotic regime in which there is wide scatter in measured void fraction at constant gas flow as seen in Figure 16-11. Hysteresis is also observed in this regime with variations in residual water level depending on how the tests are performed. This hysteresis is caused by U-tube oscillations, which are the result of intermittent holdup and fallback in the vertical leg as the flow regime changes from slug to chum-turbulent. 16-2-14 Effect of Scale An important question which must be answered is what distortions the scaled geometry and low pressure used in these tests has introduced relative to the PWR. Having explained the data in terms of the limit lines above, we can examine the effect of scale by seeing how these limit lines change with scale (Figure 16-12). Here, the limit lines at 1/3-scale are compared to the limit lines at full-scale at the same (atmospheric) pressure and with air-water. The wave limit line is constant with respect to . For the same , however, the entraining and CCFL limit lines move to the left, and flow regime II[ becomes more important over a wider range ofj;. Figure 16-13 shows what happens when the full-scale pressure of 1000 psia is also introduced. Although it is. expected that both the critical entrainment and critical CCFL velocities will decrease with pressure, the entrainment velocity calculated from Equation 6-2 becomes very small (less than 1 ft/s), indicating this correlation may not be valid at high pressure. In Figure 16-13, it has been assumed that the entrainment limit is the same as the CCFL limit, calculated by Equation 16-5. The limit line has moved even further to the left, and the droplet and CCFL lines have effectively merged. This would indicate that a full pressure and full-scale, the most dominant regimes are I and IIM This implies that over a fairly narrow range of steam flows, the liquid in the loop seal will be almost completely expelled, and that the wave instability limit does not play an important role in determining the amount of water contained in the loop seal. Full-scale air/water experiments were carried out at the IVO test facility simulating PWRs of Russian design (Tuomisto, 1988). These tests were performed in a manner similar to the Westinghouse tests. It was found that the Taitel-Dukler unstable wave theory predicted the onset of slugging in the U-tube (in the Westinghouse tests, this would correspond to the point at which significant water entrainment occurred; that is, at the limit line), particularly when the water level was adjusted to account for the increased level at the downstream elbow of the U-tube. o:\4384-nonW438 4 -16.wpd:Ib-04043 16-10

Figure 16-14 plots the full-scale U-Tube data against the corresponding limit lines. It can be seen that the data lies entirely to the left of the CCFL limit line. This behavior is different from the 1/3-scale test, and is believed due to the large U-tube oscillations which occurred in most of the large scale tests. These oscillations, which were attributed to undesirable air blower operating characteristics (see next section), pushed water into the downstream vertical leg, which was then ejected from the loop seal. This process continued until sufficient water was ejected to allow the gas velocity to fall below the CCFL limit, at which point the oscillations stopped and no more mass was lost out of the loop seal. These results indicate that if the loop seal behavior is oscillatory, the dominant mechanism controlling the amount of water in the loop seal is vertical leg CCFL. 16-2-1-5 Full-Scale Steam-Water Tests Tests were performed at full-scale for a typical four-loop PWR in the Upper Plenum Test Facility (UPTF) at pressures of 3 bar (43.5 psia) and 15 bar (217.5 psia). The separate effects tests (Liebert and Emmerling, 1998) were conducted by blocking three of the four loops as seen in Figure 16-15, partially filling the loop seal in the open loop, injecting steam into the reactor vessel simulator, and measuring the residual level once entrainment had completed, but before the steam flow was terminated. The published data from the two test series are shown in Figure 16-16 (Liebert and Emmerling, 1998 and Ohvo, et al., 1998). A line is drawn through the data that represents a constant average gas velocity as seen in Figure 16-16. This velocity is the best estimate of the minimum velocity at which entrainment from the liquid surface will take place within the horizontal section of the loop seal and is independent of the level in the horizontal run. Also shown is the Taitel-Dukler line for transition from slug to entrained flow. Liebert and Emmerling note that slugging was observed only at the lowest Froude number in each test series. Otherwise, the flow was observed to be stratified. The calculated critical gas velocities are 60 ft/s and 32 ft/s for the 3-bar and 15-bar test series, respectively. Using the above critical velocities and calculated viscosity numbers and the critical velocity from the Westinghouse air-water tests (Figure 16-10), the results can be compared to Ishii's correlation as shown in Figure 16-17 (Ishii and Grolmes, 1975). The UPTF and Westinghouse data lie well below the data base upon which Ishii's correlation was constructed. While the loop seal data lie close to the correlation, the data do not quite fit. If Ishii's correlation is used to deterrine the critical gas velocity, the lines of constant velocity are as shown in Figure 16-16. For this case, the lines of constant velocity collapse to what is in effect a single line and are not representative o:\43S4-non\43S4-16.wpd:lb-04163 16-1 1

                                                          -                                            l~~~~~~~~~~

of the data. However, as shown in Figure 16-17, the UPTF data lie on approximately a line of constant Ishii parameter equal to 0.0033. The deviation in the UPTF data from the Ishii and Grolmes correlation might be explained as a scale effect that is not included in the correlation. Because the UPTF data are at full geometric scale, these data are believed to be more reliable. The viscosity number (6.3 x 10') for full pressure (about 70 bar) lies between the viscosity numbers at 3 bar and 15 bar. Therefore, the Ishii correlation is modified to become constant for viscosity numbers less than [ Using [

                   ]a.C* The UPTF data may lie in the transition regime and not the rough turbulent regime. This may point to the reason for the wider spread in the data for the 3-bar test series as compared to the spread in the 15-bar test series.

16-3 WCOBRA/TRAC Modelling of Loop Seal Clearing Process The objective of this assessment is to confirm that the WCOBRA/TRAC-SB loop seal model adequately calculates the loop seal clearing process for a PWR. The assessment will be performed as follows: Model-scaled experiments to confirm that the interfacial drag models adequately predict the liquid distribution in the loop seal for various flowrates

• Examination of the effect of changes in scale on the predicted results 16-4 WCOBRAITRAC Simulation of the UPTF 3-Bar and 15-Bar Tests The two UPTF full-scale steam-water tests are worthy of simulation. The separate effects tests were conducted by blocking three of the four loops and injecting steam into the reactor vessel simulator as shown in Figure 16-15 (Liebert and Emmerling, 1998). The WCOBRAIIRAC model for the simulations has [

                                                       ]. The noding in this model is sufficient for simulation of the UPTF tests.

o:\4384-non\4384-16.wpd:lb-04163 16-12

Initial conditions for the tests appear to have been subcooled liquid in the loop seal and a superheated steam supply (Ohvo, et al., 1998). It is expected that [ a.c Each of the test simulations is run separately, starting from the same initial conditions. The steam flowrate is increased from zero to the specified flowrate [

                                                                ]ax The results of the 3-bar simulations are shown in Figure 16-19. The solid squares are the data values. At low vapor velocities, WCOBRAITRAC-SB over predicts the residual level. While there is a tendency to under predict the residual level at j* > 0.1. The calculations for the solid square cases to the left of the Ishii's entrainment limit line predict a sudden blowout of fluid from the pump suction leg as shown in Figures 16-20 and 16-21 at approximately 270 seconds. This behavior appears to result from the oscillation in the pump suction by building sufficient momentum to clear the liquid into the hot leg.

The results of the 3-bar calculations are summarized in Figure 16-22. WCOBRAfTRAC seriously over predicts the residual level for j < 0.1 and under predict the residual level for j > 0.1. The predicted behavior for the 15-bar tests compared to the data and limit line is shown in Figure 16-23. As with the 3-bar tests, WCOBRAfRAC-SB over predicts the data jg < 0.1 and under predicts the data for j > 0.1 with some variance in the transition region. However, there is less variance in the trend of both the data and calculations for 15-bar than for the 3-bar tests and calculations, as shown in both Figures 16-23 and 16-24. o:\4384-non\4384- 16.wpd:lb04163 16-13

I Although no data are known to exist for full-scale and approximately 1000 psia, WCOBRAIRAC-SB calculations were made using the UPTF model. As shown in Figure 16-24, the calculations follow the trend of the 15-bar calculations. It is reasonable to conclude that for 1000 psia WCOBRA/TRAC-SB will over predict residual level for j less than approximately 0.1 and underpredict residual levels forj >0.1. Measured pressure drops across the UPTF loop seal are shown in Figure 16-26a. The highest pressure drops occur forj <0.1 and then become approximately constant with increasing stean velocity. Also the magnitude of the observed differential pressure oscillations is significantly greater forj <0.1. The pressure drop calculated by WYCOBRAfRAC-SB is shown in-Figure 16-26b. The calculated pressure drops shown in Figure 16-26b represent averages over a 300 second slice of the transient. The calculations generally show the same trend with the pressure drops increasing below = 0.1 and approximately constant above j 0.1. However there are several significant exceptions to the general trend. The 3-bar calculations (open squares) designated 1, 2 and 3 represent cells where the liquid in the loop seal was overpredicted at low steam velocity and correspond to cases where the liquid level was underpredicted by WCOBRA/TRAC-SB. Calculations for 15-bar (solid triangles) marked 4 and 5 represent cases where significant liquid is retrained in the pump suction leg, which increases the flow resistance. Solid triangles designated 6 and 7 represent single phase steam flow. As with points 4 and 5 the calculated resistance is greater than implied by the data. The described variances in the calculations may well be attributed to [

             ]',. The calculations for 1000 psia (open diamonds) do not exhibit the variances noted for the 3-bar and 15-bar calculations. As shown in the data there is little effect of system pressure on the pressure drop, and that trend is observed in the calculations.

16-5 Conclusions Assessment of the experimental data indicates the following:

• Hysteresis and flow oscillations are likely to occur during the clearing process.

These oscillations will result in continued inventory loss from the loop seal until conditions fall below the CCFL in the vertical leg. Uncertainty in the remaining mass inventory of liquid in the loop seal will result in a corresponding uncertainty in the time at which the loop seal will possibly replug.

• Full-scale low pressure steam-water test data are consistent in trend with low pressure air-water test data, both for residual water level and pressure drop.

o:\4384-non\4384-16.wpd:Ib-04163 16-14

Assessment of WCOBRA/TRAC-SB relative to the experiments indicates the following:

• WCOBRA/IRAC-SB does not reproduce the observed flow oscillations or residual water level exactly. This deficiency can be accounted [

                        ]a.C

• WCOBRArIRAC-SB overpredicts the quantity of liquid cleared from the loop seal for high vapor flows. This may be attributed to [

                          ]a.c

• WCOBRArrRAC-SB adequately predicts the trend of the residual level data in full-scale tests.
• WCOBRA/TRAC-SB adequately predicts the trend of the pressure difference data.

16-6 References Ishii, M. and Grolmes, M. A., 1975, "Inception Criteria for Droplet Entrainment In Two-Phase Concurrent Film Flow," AIChE J., Vol. 21, No 2. Kukita, Y. et al., 1990, "Loop Seal Clearing and Refilling During a PWR Small Break LOCA," Nuclear Eng. and Design, 121, 431-440. Liebert, J. and Emmerling, R., 1998, UPTF Experiment Flow Phenomena During Full-Scale Loop Seal Clearing of a PWR, Nuclear Engineering and Design, Vol. 179, pp. 51-64. Ohvo, J., et al., 1998, Simulation of Full-Scale UPTF Loop Seal Experiments with APROS, CATHARE and RELAP, 6th International Conference on Nuclear Engineering, ICONE6-6090. Pushkina, 0. L. and Sorokin, Y. L., 1969, "Breakdown of Liquid Film Motion in Vertical Tubes," Heat Transfer-Soviet Research, Vol. 1, No. 5. Richter, H. J., 1981, "Flooding in Tubes and Annuli," Int. J. Multiphase Flow, Vol. 7, No. 6, pp. 647-658. oA4384-non\4384-16.wpd:1b041 63 16-15

Taitel, Y. and Dukler, A. E., 1976, "A Model for Predicting Flow Regime Transitions in Horizontal and Near Horizontal Gas-Liquid Flow," AIChE J., Vol. 22, No 1. 'J Tuomisto, H. and Kajanto, P., 1988, "Two Phase Flow in a Full Scale Loop Seal Facility," Nuclear Eng. and Design, 107, 295-305. o:\4384-non\4384-16.wpd:lb-4043 16-16

a) Initial Phase: Natural Circulation b) Natual Circulation Broken; Phase Separation c) Loop Seal Clearing d) Loop Seal Refilling Figure 16-1. Loop Seal Clearing and Refilling o4384-non\4384-16.wpd:lb-04043 16-17

a) Quiet Level b) Plug Formation c) CCFL in Vertical Pipe d) Drop Entrainment Figure 16-2. Loop Seal Clearing Process o\4384-non\4384-16.wpd:1b-04043 16-18

Separator Plenum AP, Apa A; A; Figure 16-3. 1/3-Scale U-Tube Test Facility o-\4384-non\4384-16.wpd:lb-04043 16-19

I 100 1-1 9 LT-4 u Ir: w 10 9 0

    'IO cr bi I

0.1 1 10 100 1,000 Vapor Volumetric Flux (ft/s) stOm tpda AfW& Dl= ft D-.S a Figure 16-4. Taitel-Dukler Flow Regime Map, Comparing 1/3-Scale Pipe at 14.7 psia 12 and Full-Scale Pipe at 1000 psia (Taitel and Dukler, 1976) o:.4384-non\4384-16.wpd:lb-04043 16-20

0.8 0.7 U.s 0.6

                        - U  U~~~~~~~~~~~
                                          'i.~~~~~~~~

0.5 *~~~~~~~~* - _~~~~ U 0.4 U.~~~~~~. U... 0.3 II I

• I j I  ! U l l t w l I 0.2 0.1 0 10 20 30 40 50 60 70 Jg (FtISec)

Figure 16-5. 1/3-Scale U-Tube Residual Water Level Remaining After Test as a Function of Test Gas Flowrate o.\4384-non\4384-16.wpd:1b-04043 16-21

0.7

                                                             -S~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
                        *               .5Q-0.6             -      *                ~~~~~~~~~~~S
                    -                                           s~~~~~~~~~~~~5 0.5    _             00 0   ~bO   0     *        -+

B-+ - .5" ss %

                                                                                 -. JIMIT       LINE
                                                                                           ""s~~~~~~~~~~~~~~-

0.4 o o o 00000-,.-...-***#***KXXX  %~~~~~~~~.

                                                                                             .5*O                   -

M 0.3 0.2 0 . 0.1 . O o0 0 0 . * *

• X OK*
• C> o0 S 0

0 10 20 30 40 50 60 70 Jg (Ft/Sec) Small Ripples Small Waves LevelDepression Droplets in Droplets Reach Water Encrained Liquid Entained Limit Point at Upstrear Elbow Downsucam Pipe Top of Pipe into Vertcal Pipe Out of Pipe

• 0
• X * +

Figure 16-6. 113-Scale U-Tube Flow Regimes Observed Under the Limit Line o:\4384-non\4384-16.wpd: b-04043 16-22

1 ,~o V VUCD WC) OU UO 0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Oo o

0) 0 00 0 CZ o 0.8 0 0 o w w "~~~~~~~~~~~~~~~~~~~~~

2~~~~~~~~~~~~~~~~~~~~ o 2 0 B~~~~~~~~~~~

• U

                                                    *                   .

• 9~~~~~~~~~~~

t:0.6 U

• E

                     *~~~~~         U
                           *

• U 0.4 0.2 0 10 20 30 40 50 60 70 Jg (Ft/Sec)

Horizontal Leg Veical Leg O 0 Figure 16-7. 13-Scale U-Tube Horizontal and Vertical Leg Average Void Fractions During Test o-\43S4-non\4384-16.wpd:1bO4043 16-23

Li 1

                                                                                                   &0
                                                                                                 *0 0.8                                                0                                    80    0 0                                                           Vo 0 0                                    Qs ,# Oo 0                                             0 00                                       . eo t 0.6               0          o 3        Oo                           0
                            ~00                        0 . . e      0 10 0!     0 oL)

OOge* 0 .0 0 0.4 80 0m 0 0o .... s e a

                    .33a,a 0.2           .

I l I I I I I I I I I 0 0 10 20 30 40 50 60 70 Jg (Ft/Sec) Afta During 0 Figure 16-8. 1/3-Scale U-Tube Horizontal Average Void Fraction During Test Compared With Average Void Fraction After Test o:\4384-non\4384-16.wpd:lb-04043 16-24

0.3 a

                         . 8 0.25 U

U 0.2 0 mm

• U*

"f 0.15 g U.

• mu~~~~~~~~

0.1 I n

• U ~~

a~~~~~~~~~

                                                              ~ ~ ~ s         a  a U~~ ~ ~ ~ ~ ~                2 0:

II I I I I 1 1 1 1 1 i IU 0.05 0 0 10 20 30 40 50 60 70 Jg (Ft/Sec) Figure 16-9. Pressure Difference Across the 1/3-Scale U-Tube o:\4384-non\4384-16.wpd:1b-4043 16-25

1 0.8 o-

 ~0.6
       >~~~~~~~~
                         'Ue6I.                                          Oscilons,     slug flow, failback in vertical leg.

0.4

• II. Liquid pushed into vertical

                                        .. * // /.                       leg by waves, caried out b3
                                .    /*     /                            CCFL in venical leg.
                                                                  / / /. Droplets entrained into verical leg, carried out by CCFL in vericaL leg.

0.2'dI 0 0.1 0.2 0.3 0.4 0.5 Jg Dw Figure 16-10. 13-Scale U-Tube Normalized Level and Limit Lines o-\4384-non\4384-16.wpd:1b-04043 16-26

0.8 Limit Line Approached 0.7 * . . From Above 0.6 3~~~~ I

• U~~

0.5 _ a~~~~ M 0.4 _ ~* Limit Line Approached* U~~~~~~~~~~: From. Below*. I I , I I . I I I 0.3 0.2 0.1 0 10 20 30 40 50 60 70 Jg (Ft/Sec) Figure 16-11. Hysteresis in Loop Seal Limit Line o:\4384-non\4384-16.wpd:1b04043 16-27

I 0. a

0.4 0.2 0 . 1 0.2 03 0.4 0.5 Jge Figure 16-12. Effect of Increased Geometric Scale on Limit Lines o:.4384-non\4384-16.wpd:lb-04043 16-28

1 0.6

      ~0.4
      '5 0            0.1           0.2          0.3           0.4        0.5 Jg*

Figure 16-13. Effect of Increased Pressure and Scale on Limit Lines o\4384-non\4384-16.wpd:lb-04043 16-29

Tests dominated by u-tube oscilladons caused by blower feedback. Water retained in loop seal only when Jg < CCFL. 0.8

     ~0.6 0
                             -  J.~     ~ ~       J
                 .                  D.l     0.2              0.3            0.4           0.5 j9*

Figure 16-14. IVO Full-Scale Final Void Fraction and Limit Lines o:\4384-non\4384-16.wpd:lb-04043 16-30

                                                       -   lo cntirnent sxnulator Reactr pressure vesel Steam generator Reactr coolant pDur Pressureer Break Vave Loop sal Sleam flow from Steam generator srnulato D

3330 Figure 16-15. UPTF Facility and Single Loop Seal (Liebert and Emmerling, 1998) o\4384-nonN4384-16.wpd:lb-04043 16-31

I 0.9 tWT? ZS bar 0.8

                                                                        ....... Taute.Duklw 0.7                                                                cr",-al-&svetocky-60 Nsec 0.6                             Ps                                Crkicaig   vcky-32 ftl/sec 0O.50.4~~~-                       **\X"""'-

0.4~~~~~~ 0.3 * ' 0.2 0.3 *^ -. - 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 J. Figure 16-16. Lines of Constant Gas Velocity Compared to UPTF Data for 3-Bar and 15-Bar Loop Seal Tests (Liebert and Emmerling, 1998) o:\4384-nond \ 0384'16.wpd:lb04043 16-32

0.0001 0.001 0.01 0.1 I 1.0000

                     +    Ishii UPTF & Westinghouse
                       -~~Ishii Correlation Data v       A         . 0.1000 A  Ishii equation 22 A
                                            ^iX
                                            .r                                       .. 0.0100
                               /               ~~I=0.0033 0.0010 0.0001 N

Figure 16-17. UPTF and PWS Compared to the Ishii Correlation and Data Base (Ishii and Grolmes, 1975) o:\4384-non\4384-16.wpd:lb-04043 16.33

I a,c Figure 16-18. WCOBRAITRAC Model of the UPTF Separate Effects Loop Seal L Clearing Tests o:\4384non\4384-16.wpd:lb-04163 16-34

( ( ( w 1.2 - 0.8 3 WCT 3bar

                             -   Taitel-Dukler a sO0.6 0.4 0.2 0-0 0.1 0.2       0.3            0.4  0. 5 Jg*

I

             ~~~~ Av g             1       0         0 COLLAPSED      LIO. LEVEL 1.1 1
          .9
   -=
          .8 C>

E C>

          .7 V

c: CS 'L.'

          £6
          .5
          .4 Figure 16-20. Calculated Level for 3-Bar Test at a Superficial Gas Velocity of 5.7 ft/s o.43&4-non43B4-16.wpd:1b-04043              16&36

L Leve I O 0 0 Pump Suction

           - -   --    Leve               O       0         0  SG   Downhill 6-5 s3-~3I a,>       I CY,
         -     I

• 1 I

Figure 16-21. Calculated Liquid Level in Steam Generator Downhill Pipe and Pump Suction Pipe for a Superficial Gas Velocity of 5.7 ft/s o.4384-nonX4384-16.wpd:Ib-04043 16-37

1 0.9 0.8 o0 07 3 0.6 0 O 0.5 0 0.4 I 0.3 0.2 0.1 0 0 0.2 0.4 0.6 0.8 1 h/D UPTF Data Figure 16-22. Comparison of WCOBRAITRAC-SB Calculations and UPTF Data for the 3-Bar Tests o:\4384-non\4384- 6.wpd:1b-04043 16-38

1.2 1- 0°

• UPTF 15bar Data o WCT 15bar 0.8 - . -- *.Taltel-Dukler 0.6 - c

                                       -4, 0.4 -                                                     --          -    ..

0.2-. 0~~~~~~~~~~ _* , I 0 0.1 0.2 0.3 0.4 0.5 Jg* Figure 16-23. Calculated Residual Levels versus UPTF 15-Bar Data o:\4384-non\4384-16.wpd:lb-04043 16-39

1 0.8 0 75Q0.6 U C.) C-'0.4 co 0.2 0 0 0.2 0.4 0.6 0.8 h/d UPTF Data Figure 16-24. Comparison of WCOBRA/TRAC-SB Calculations and UPTF Data for the 15-Bar Tests o:\4384-non\4384-16.wpd:1b-04043 16-40

1.2 1 00 LA I VUU 1 Je,i v~~~~~~~~~~~tiJ level

                                                       - --    Taitel-Dukler 0.8 -'

0.6 -

0. -

0.2 0 0 0 0.1 0.2 0.3 0.4 0.5 Jg* Figure 16-25. Calculated Residual Levels for 1015 psia o\43S4-non\4384-16.wpd:lb-04043 16-41

1 3

                                .1 0         0.1        2      03    0. 05 Figure 16-26a. Measured Pressure Drop for UPTF 3-Bar and 15-Bar Loop Seal Tests (from Liebert and Emmerling, 1998)

IL 10 0 O-a 9- A

a. a3 bar 8- A 15 bar co 7-

• o 4 6 AA ° 1000psi 0

6-to O, 0 0 1- 5-U 0 5 A a 4-0 o~ 0 3-0 A 0 2- A' ° a ° a 1-a 1 2 33 0 0 0.1 0.2 0.3 0.4 0.5 Jg* Figure 16-26b. Calculated Loop Seal Pressure Drop for 3-Bar, 15-Bar, and 1000 psia o:\4384-nonW438416.wpd:Ib 04043 16-42

                       ~~DP                    4      4        0 PRESSURE 0n-2 C',

500 550 600 650 700 750 800 Time (s) Figure 16-27. Calculated Pressure Drop for 15-Bar and Vapor Superficial Gas Velocity of 7 ft/s DP 4 4 0 PRESSURE 1.75 1.7 1.65-

         .21.6    5
              ~1.5 1.45 1.A      I I   I      I  1 I      I                            1 500          550         600         650       700  750  80 Time (s)

Figure 16-28. Calculated Loop Seal Pressure Drop for 15-Bar and Superficial Gas Velocity of 17 ft/s o:4384-non\4384-16.wpd:Ib-04043 16-43

DP 4 4 0 PRESSURE 7.295 - CL- 7.29 - U, 7.285 I.- a) L., a) 0 7.28 - 0 C' 7.275 - 7.27-650 Time (s) Figure 16-29. Calculated Loop Seal Pressure Drop for 15-Bar and Superficial Gas Velocity of 29 ft/s o.43S4-non\4384- 16.wpd:lb-04043 16-44

SECTION 17 STEAM GENERATOR REGION HYDRAULICS MODELLING 17-1 Introduction Steam generator hydraulics is identified as one of the major processes affecting the small break LOCA transient (See Section 1, Volume 1, of this document). Early in a postulated small break LOCA event, the power generated in the core is removed to the secondary systems by heat transfer through the steam generator tubes. This primary to secondary heat transfer initiates several processes that occur in the steam generator and which are ranked high in the PIRT because they affect the small break LOCA transient. These important processes include steam generator primary side heat transfer; steam generator tube voiding and CCFL, which may occur in the tubes themselves andlor in the steam generator inlet plenum and hot leg piping; and primary side pressure drop associated with the two-phase mixture in the tubes. Validation of the WCOBRAI`RAC-SB computer code for the small break LOCA application should include a demonstration that the code adequately predicts steam generator hydraulic phenomena as follows:

• Flooding in the vertical tubes and horizontal tubes
• Possible flooding in the hot leg steam generator inlet plenum connection
• Condensation within the steam generator tubes for primary to secondary heat transfer
• Steam generator primary mass inventory
• Associated pressure drop as the RCS inventory decreases during a small break LOCA event Therefore, in this section, the performance of WCOBRAIRAC-SB is qualified against pertinent single-effects experimental data and analytic solutions for benchmark problems, and against the Natural Circulation (NC) test series experiments performed in the Semiscale facility.

o:\4384-non\4384-17.wpd:lb-04043 17-1

17-2 Physical Processes The natural circulation period of small break LOCA scenarios is in large measure determined by the steam generator hydraulics. The steam generator processes affect the rate of RCS depressurization, particularly for the smaller small break LOCAs, which in turn affects the break flowrate, the rate of safety injection flow into the RCS, the state of the RCS at the time of loop seal clearance, and ultimately the core mixture level depression. Condensation in the steam generator tubes not only removes the core decay power but also generates liquid mass within tubes. For the smaller small break LOCAs, an extended quasi-steady-state condition of nearly constant pressure occurs during which the core power is closely matched by the steam generator heat transfer to the secondary side. As system inventory decreases, eventually the steam generator tubes drain and the steam generator heat transfer reverses. During the natural circulation period, condensate in the uphill portion of the steam generator tubes may possibly be carried into the steam generator outlet plenum in cocurrent flow or drain countercurrently into the steam generator inlet plenum. The prediction of CCFL in the vertical steam generator tubes determines whether any liquid from the steam generator uphill can proceed to drain into the steam generator inlet plenum. Furthernore, CCFL in the hot leg connection at the steam generator inlet could potentially prevent the steam generator liquid from drawing back into the hot legs, where it may contribute to the vessel upper plenum inventory. The steam generator hydraulic behavior affects not only the pressure drop in the steam generator but also the timing and nature of the loop seal clearing. Consideration of steam generator hydraulics in the context of the overall mass distribution in the RCS is important. The Semiscale NC test series experiments are similar in design and execution to those conducted in other test facilities; they covered a range of single- and two-phase conditions with a regulated system coolant inventory. Predictions of test results from the Semiscale NC series are compared to such parameters as flow versus system mass inventory as the RCS is drained and transitions between two-phase flow and heat transfer modes. o:\4384-non\4384-17.wpd:Ib-04043 17-2

17-3 CCFL Modelling in WCOBRA/TRAC-SB 17-3-1 Introduction The CCFL is associated with the process of restricting liquid flow by counterflowing vapor due to interphasic drag forces. For example, liquid downflow in a pipe under the influence of gravity becomes unstable with increasing vapor upflow and eventually flows together with the vapor. Thus, stable countercurrent conditions can exist only within a certain range. The boundary of this range is recognized as the CCFL. This type of phenomenon can also exist in a horizontal stratified flow. CCFL can occur in several locations in the PWR during the small break LOCA. CCFL may occur in the U-tubes of a steam generator. Inside the reactor vessel, the liquid inside the upper head can be prevented from downflow by steam flow inside the guide tubes. The liquid in the upper plenum may be held up at the upper core plate or upper fuel tie plates by upflowing steam from the core. The focus of this section is the predictive capability of the multidimensional vessel hydrodynamics models for CCFL. In Section 17-3-2, the CCFL in a vertical pipe is evaluated with saturated liquid and steam at 1000 psia. In Section 17-3-3, the CCFL on a perforated plate is evaluated with saturated liquid and steam at 1000 and 35 psia. The geometry of the plate (perforation ratio and thickness) simulates, at small scale, the upper tie plates in a PWR. The computed results are compared with Northwestern test data (Hsieh, et al., 1980). In Section 17-3-4, CCFL of horizontal stratified flow is computed and the results are compared with available correlations. 17-3-2 CCFL in a Vertical Channel 17-3-2-1 Vertical WCOBRA,TRAC Channel Model Flooding in a pipe has been studied with a TRAC 1-D component (Takeuchi and Young, 1983 and Takeuchi, et al., 1992). In this section, the prediction of pipe flooding with the WCOBRArRAC 3-D vessel fluid models is assessed. The purpose is to evaluate the interphase flow model, which is applied to several locations in the vessel such as support columns and guide tubes, and to compare predictions to classical flooding relationships. o:\4384-non\4384-17.wpd:lb-04043 17-3

I Figure 17-1 illustrates the WCOBRAfIRAC model for pipe flooding computations. [

                                ]aC All of the computations use saturated steam/water mixtures and saturated steam/subcooled water.

The computational experiment to generate the flooding curve proceeds as follows. Liquid is injected into channel 9 at a constant rate. After a steady liquid downflow into the bottom tank is established, steam is injected from channel 4 at a gradually increasing rate. At the beginning, a countercurrent condition is observed with steam flowing up through channels 4, 5, and 6, and separated via channel 7 into the pressure boundary condition in channel 8. By increasing steam flowrate and by maintaining a constant liquid injection rate, the magnitude of the liquid downflow through channels 5 and 4 is reduced. Eventually, liquid downflow is prevented by steam upflow. This experiment is repeated for several constant liquid flowrates. The cases studied included the following conditions:

• Liquid is injected from the top of the test section at constant flowrates as shown in Table 17-1.
• For each case in Table 17-1, after a steady-state was established for 80 seconds with falling water but no vapor counterflow, vapor is injected and gradually increased to maintain quasi-steady conditions. Vapor is injected from the bottom of the test section with a linearly increasing rate as shown in Table 17-2.

o:\4384-non\4384-17.wpd:lb-04043 17-4

Computed steam (g) flowrates and liquid (f) flowrates through channel 5 are plotted in a ( J)1/2 versus ( )If2 coordinate system, where the dimensionless volumetric fluxes are defined by: jg j [Pg/g Dh AP] 1 (17-1) and j; is similarly defined with subscript g replaced by f. All the calculations exhibited similar behavior. A typical calculation (Case 3 in Table 17-1) is described as follows.

                                             ]a.c Figure 17-5 shows the vapor versus liquid mass flows at several time intervals. These flows are then converted to            and if, using Equation 17-1, and plotted along with the Wallis flooding curve (Wallis, 1969), Equation 17-5 in Figure 17-6. This comparison shows that the flooding limit is not violated even in the presence of flow oscillations.

The results for several cases at high pressure are shown in Figure 17-7a, which plots the square root of the fluxes. Each symbol represents a jf I jg pair taken from the WCOBRAfIRAC run. The fanily of symbols (Figure 17-7b) indicates how the liquid downflow is reduced as the vapor upflow is increased (some of the oscillatory points have been removed for clarity). A straight line drawn through the upper bound of these points is regarded as the CCFL predicted by WCOBRAJTRAC. o:\4384-non\4384-17.wpd:lb4O4043 1775

I 17-3-2-2 Predicted CCFL at High Pressure For the pipe model evaluation, the system pressure is 1000 psia, a representative pressure for the draining of steam generator tubes during a small break LOCA event. Both liquid and steam are assumed to be saturated. The countercurrent flow conditions at various points are plotted as: 1 2 Y =[jg]I (17-2) x = [*]2 / C (17-3) in Figure 17-7b. In this coordinate system, the Wallis flooding correlation (Wallis, 1969) becomes: x + y =1 (174) the straight line shown in the plots. The figures clearly indicate the existence of a flooding limit approximately defined by: (j)12 + ( )l/2 =C (17-5) with C = 1. In the case of the TRAC 1-D pipe model, which was previously studied by Takeuchi and Young (Takeuchi and Young, 1983), the predicted flooding curve was at C = 0.726. In the vessel component pipe model, the interphasic drag force is somewhat weaker and leads to a higher flooding limit. However, pipe flooding data typically lie between 0.9 < C < 1.0 (Wallis, 1969). 17-3-3 CCFL in a Perforated Plate CCFL in a perforated plate has been tested and analyzed by Hsieh (Hsieh, et al., 1980). The geometry of the perforated plates approximately simulated the geometry of a fuel assembly top nozzle (or upper tie plate). The tests were conducted with air/water and steam/water systems. The air/water experiment was designed to investigate the effects of geometric factors on CCFL. The steam/water tests investigated subcooling effects on the CCFL. Air/water test data are o:\4384-non\4384-17.wpd:lb04043 17-6

analyzed in this section with a 15-hole perforated plate that most closely simulated the upper tie plate. The analyses were performed with the WCOBRAIIRAC code for a saturated steam/ saturated water system at high pressure (1000 psia) and at low pressure (35 psia). The low pressure condition approximates the vapor density in the air/water test at atmospheric conditions so that the computed results can be compared with the test data. In Section 17-3-3-1, the CCFL at a perforated plate is described. The computed CCFL at high pressure is developed in Section 17-3-3-2. The computed CCFL at 35 psia is compared with the test data in Section 17-3-3-3. 17-3-3-1 Correlations and Scaling for CCFL in a Perforated Plate Various scaling methods and the correlations associated with them are described below: Northwestern (H) Scaling Hsieh (Hsieh, et al., 1980) developed a scaling parameter similar to the one used by Wallis (Wallis, 1969) to define a nondimensional volumetric flux, which is referred to here as Northwestern scaling: h ig[Pg / g W Ap] 1 2 (17-6) where: W8 D [aig Ap]2 (17-7) and where Dh is the hole diameter and is defined as: a - tanh{k Dh Ah /AT} (17-8) for the perforation ratio Ah IAT (hole area divided by total plate area) and a wave number defined by: k -2,t (17-9) where t is the thickness of the plate and h for the liquid phase is similarly defined. o-.14384-non\4384-17.wpd:lb-04043 17-7

With these dimensionless volumetric fluxes, the test data for CCFL in the perforated plates were correlated by Hsieh to yield: .t,, h; 1/2 + h 1/2 = C (17-10) where: C = min (2.0, 1.07 + 0.004332 LI) LI = n7r D [ gp/a]1/2 and n is the number of holes. One way to examine Northwestern scaling is to compare it to other scaling methods as discussed below. Wallis (J*) Scaling

                            *= jg [plg Da Ap]" 2                                             (17-11)

• Kutateladze (K*) Scaling k = j [pg2 lg a Ap]" 4 (17-12)

                                = j[D      ]1 2 where use has been made of the dimensionless diameter:

D * - D [g Ap/ C]" 2 (17-13) o:\4384-non\4384-17.wpd:lb-04043 17-8

• Westinghouse (L) Scaling Takeuchi and Young (Takeuchi and Young, 1983) proposed a generalized scaling approach that combined J and K scaling as follows:

                             -    (D /K2)12                                                    (17-14) where K is the critical Kutateladze number that approaches KOVbF at small diameters, and K,, at large diameters, respectively. That is:

19 - jKo as D - O (17-15)

                            -   kg1K     as D   -

where Ko = 0.645 and K = 3.2. Similar relationships hold for the liquid phase. For a given plate thickness (t), the Northwestern scaling approaches the following linits:

               -      For Dh - 0, it approaches the Wallis number:

h - (17-16) hf - f

               -      For Dh - -, on the other hand:

h; - k; (17-17) o:4384non\438417.wpd lb-04043 17-9

I 17-3-3-2 WCOBRAITRAC Results A WCOBRAITRAC analysis of the test data was performed; the test case selected was the 15-hole plate with the following dimensions ffCOBRA/TRAC-SB produces approximately the same results): Dh = 0.413 in. t = 0.787 in. AT = 4.726 in2 Ah = 2.013 in2 Al/AT = 0.4260 The WCOBRA/TRAC analysis was initially performed for the large break LOCA application (Bajorek, et al., 1992). Given these dimensions, C 2.0 is used for both 1000 and 35 psia. This approximates the typical dimensions in a PWR or LOFT fuel assembly (tie plate). The WCOBRA/TRAC model used to predict the CCFL for the perforated plate is shown in Figure 17-8. Channels 1 through 3 simulate a large tank receiving falling water. Channel 4 is a large diameter pipe section where vapor is injected. Channel 5, representing the region directly above the plate holes, has the area of the holes in the plate, the hydraulic diameter of the hole, and the gap geometry obtained by assuming that the holes are projected as channels into the space above the plate. This modelling is similar to that used in the CCFL and upper plenum regions of the PWR. Channel 6 is the space above the solid portion of the plate. Channel 7 is a bridge for the countercurrent flow carried into the constant pressure boundary condition in channel 8. Cells 2 and 3 of channel 7 are isolated and inactive. Channel 9 is the top of the test section where liquid is uniformly injected. Gaps for lateral flow are also indicated with numbers enclosed by circles. The lower half of gaps 4 and 5 are blocked. The computational experiment is performed in the same manner as the previous study. The saturated liquid is injected at a constant rate over an entire transient. After 30 seconds or so, when a steady-state is reached, steam is gradually injected at a linearly increasing rate. When the steam injection rate is low, all the falling liquid passes through the perforated plate and settles in the large tank at the bottom. The injected steam flows up through the perforated plate channel 5, and then through channels 6 to 8 into the constant pressure sink. As the steam injection rate increases, the amount of falling liquid becomes less and the residual liquid accumulates in channel 6, which then overflows into channel 8. Eventually, falling liquid is shut off, and the o:4384-nonM4384-17.wpd: lb-04043 17-10

accumulated liquid flows into channel 8 together with injected steam. The computed liquid and steam flowrates are expressed in terms of the Northwestern dimensionless flowrates (h; and h and then the values are plotted in the coordinate system: g versus h C C The results are shown in Figures 17-9 and 17-10 for different liquid injection rates under a system pressure of 35 psia. Saturated steam and liquid in 35 psia simulate air/water under the atmospheric condition. For both cases, the computed flow states are bounded by the Northwestern flooding limit. For a higher system pressure, such as 1000 psia, similar results can be seen in Figure 17-11, but the flooding linit seems to be slightly more severe than lower pressure. P.c o:.4384-nonW43S4-17.wpd:Ib-04043 17-1 1

In the comparisons that follow, the results are presented in terms of the Northwestem scaling parameters. 17-3-3-3 WCOBRAJTRAC MOD7 Results The predicted countercurrent fluxes are shown in Figure 17-14 in nondimensional form for 1000 psia as calculated using WCOBRA/TRAC MOD7. Similarly, the results at 35 psia are shown in Figure 17-15. The countercurrent flow conditions at various points are plotted in an x-y coordinate system as: Y = [h;]112IC (17-18) x = [h *]" 2 /C As described previously, each point on the figure represents conditions at a point in time as steam flow is gradually increased. The CCFL predicted by WCOBRAITRAC is approximately located by the tangent drawn as the dashed line. [c 17-3-3-4 Comparison with Data The MOD7 and MOD7A results can be compared with the test data (Hsieh, et al., 1980) for air/water at 14.7 psia (Figure 17-16). The Northwestem flooding curve represents the midpoints of all the test data of all the cases. For the 17-hole case studied here (n = 15), the test data lie above the flooding curve as shown by the open squares in Figure 17-16. In conclusion, the WCOBRA/TRAC MOD7A predicted CCFL is slightly conservative relative to the test data. The WCOBRAITRAC MOD7A method is included in the WCOBRA/TRAC-SB version. 17-3-4 CCFL in a Horizontal Channel In a small break LOCA, substantial stratification of two-phase flows occurs. These and other problems in predicting two-phase flow phenomena have been summarized by Zuber (Zuber, 1980). Entrainment of liquid and/or vapor pull-through at a break located at the top, the side, or o:\4384-non\4384-17.wpd:lb-04043 17-12

the bottom of a horizontal pipe cannot be determined without a stratification model. The same problem exists in the hot leg at the pressurizer surgeline junction and at the steam generator inlet elbow, where flow stratification and CCFL in the hot leg must be predicted. In a large break LOCA, conditions in the loops are much more homogeneous due to the high fluid velocities. However, some stratification is likely in some regions of the reactor vessel, such as the upper core plate. In this section, the vessel model capability to predict liquid levels and flow transitions in horizontal flow is compared against flow regime transition correlations and weir flow models. 17-34-1 WCOBRITRAC Simulation of Horizontal Flow A horizontal 18-foot long pipe of a 3x3-foot square cross section was modelled with six horizontal channels and three cells in each channel, as illustrated in Figure 17-17. These dimensions approximate a PWR hot leg. The channels are connected laterally by gaps. Channel 10 has a dead-end at the left and a constant pressure boundary condition at the top. A pipe component is connected at the bottom of channel 9 to supply liquid. Steam is injected at the top of channel 5. The right end of channel 5, as shown in Figure 17-17, is connected to a large volume simulating a vessel region. The CCFL is obtained with saturated steam and liquid flowing initially in a countercurrent state. As steam flowrate is gradually increased, the liquid flow direction (initially from the injection point to the vessel) is eventually reversed. 17-34-2 Relation of Flooding Correlations to Slug Flow Regime Transition Models The CCFL in a horizontal pipe was originally considered by Wallis (Wallis, 1969). He defined the flooding curve as a bounding condition of Long's equation for existence of a solitary wave (Long, 1956). Using a similar basis of an instability condition of a solitary wave, Taitel and Dukler (Taitel and Dukler, 1976) and also Wallis and Dobson (Wallis and Dobson, 1973) studied the flow regime transition from a stratified flow to a slug flow. The conditions for these two events are mathematically the same, but the expressions given for the phase transition are different from those used to define the flooding curves. In this section, the two events are related and the formula for the flow regime transitions are translated to the flooding curves for the stratified flow in a circular pipe and a square channel. These translated flooding curves and the Long-Wallis flooding curve are also compared in the next section. The predicted CCFL transitions computed with the WCOBRAtRAC-SB horizontal pipe model are developed and o:\4384-non\4384-17.wpd:Ib-D4043 17-13

compared in Section 17-34-3. The predicted water levels in the channel are also compared with results from the weir flow model (Wallis and Dobson, 1973) in Section 17-3-4-4. L; Wallis derived a flooding model for a horizontal stratified flow by identifying the bounding condition of Long's equation for a stratified wave in a channel. In accommodating other correlations and theories, the more general form is assumed:

               . *2         *2 if        + jg      =   2                                                       (17-19)

(1 - a)' a" as the drift flux relation, where dimensionless volumetric flux for steam is defined by: ig = jg [pg/g Dh Ap]lf2 (17-20) wherejg is the volumetric flux, Dh is the hydraulic diameter, g is the acceleration of gravity, P is the steam density, and ap = pf - Pg. The dimensionless volumetric flux (f ) for liquid is similarly defined with the liquid density (pf). Equation 17-19 with c = 1 and v = 3 becomes the Long-Wallis equation which represents a family of ellipses in the ( ji;)-plane. The derivative of this equation with respect to void fraction a:

                . *2            *2 if             j_     = 0                                                     (17-21)

(-a)" I1 a+ yields a family of curves tangent to the above family of ellipses. The flooding curves are obtained from these two equations by eliminating the void fraction (a). Eliminating jf from Equations 17-19 and 17-21 yields the expression: g= c (17-22) o:\4384-non\4384-17.wpd:lb-04043 17-14

The flooding curve is:

             ) 2 /(v+ 1) + (j)2/(v+1) = C2(v1)                                                     (17-23)

The drift flux relation for a vertical flow in the (j; j) coordinate system is a straight line for a given void fraction. Equation 17-19 gives the drift flux relation for a horizontal flow, which forms a farmily of ellipses as a is varied, as shown in Figure 17-18. The first quadrant is for the cocurrent flow, and the second quadrant, for countercurrent flow. The envelope in the first quadrant defines the flow regime transition. The flow regime outside the envelope can no longer be a horizontal stratified flow but a slug, intermittent, or annular dispersed flow. The envelope in the second quadrant is the flooding limit beyond which no countercurrent flow state exists. Both conditions for the flow regime transition and the flooding point have been derived mathematically based on the same process of wave instability. The flow regime transitions are expressed in the form of Equation 17-22. Once the flow regime transition is identified, the coefficients c and v are determined for the flooding relationship of Equation 17-21. Specific applications of the above equations are described in the following examples: Taitel-Dukler Flow Regime Transition in a Circular Pipe The Taitel-Dukler transition from horizontal stratified flow to intermittent and annular dispersed liquid flow regimes (Taitel and Dukler, 1976, Equation 23) can be expressed as: i > C2 a3 n [ 4 dAL (17-24) where C2 = 1 - hL/D and AL and hL are the liquid flow area and the liquid level, respectively. For a circular pipe, this is a complicated expression of a, which can be approximated to yield: [

                                                                                             ]aC  (17-25) o:\4384-non\4384-17.wpd:lb-04043                      17-15

Therefore, in Equation 17-22, c = I and v = 5. The flooding curve of Equation 17-21 becomes: (jJ)11 + ( = 1 (17-26) Taitel-Dukler Flow Regime Transition in a Circular Pipe with C2 = 0.5 In the Taitel-Dukler transition formula, the factor C2 can be set to 0.5 as in Wallis and Dobson (Wallis and Dobson, 1973), which is discussed below. In this case, the transition condition can be approximated by: g = 0.55 a2 (17-27) therefore, c = 0.55 and v = 3, and the flooding curve of Equation 17-23 becomes: (jf)1 /2 + (j') 1 1 2 = 0.742 (17-28)

• Wallis-Dobson Transition in a Square Channel After a series of tests, Wallis and Dobson derived the flow regime transition formula in a similar expression:

jg = 0.5a 15 (17-29) In this case, therefore, c = 0.5 and v = 2 and the flooding curve of Equation 17-23 becomes: (Jj) 2 ' 3 + j) 2 '3 = 0.707 (17-30) These flooding curves are shown in Figure 17-19. The three flooding curves are approximately the same, especially at the middle point, a = 0.5. o:\4394-non\4384-17.wpd:lb-04043 17-16

The transition and the flooding curves are symmetric, consistent with the Taitel-Dukler approximation that the interfacial force is dependent only on the steam velocity, ignoring the interface velocity. The Long-Wallis flooding curve (Wallis, 1969) is shown as curve 4 in Figure 17-19. Their flooding curve is derived from Long's wave equation, which is based on the velocity potential theory for a stratified flow. The Taitel-Dukler phase transition correlations takes into account the interphasic forces as well as fluid wall forces. Therefore, the two phases are more strongly coupled in the Taitel-Dulder correlation than they are in the Long-Wallis correlation. The Wallis-Dobson correlation in curve 3 is derived from the test data, and obviously, these forces are in effect. 17-3-4-3 Predicted Horizontal CCFL at High Pressure The CCFL predicted by WCOBRA/TRAC-SB for the model shown in Figure 17-17 was obtained as follows. Initially, the entire system is filled with saturated steam at 1000 psia. A transient calculation begins with injection of saturated liquid at a constant rate. After a steady-state of liquid flowing into the container is established, saturated steam is injected. The injected steam flows out of the system at the pressure boundary. This forms a countercurrent state. The steam injection rate is gradually increased so that a quasi-steady-state is maintained throughout the computation. With a constant liquid injection rate ranging from 70 to 650 lbm/sec, the predicted countercurrent conditions are shown by circles in Figure 17-20. For a given liquid injection flowrate, the predicted conditions are linked by lines as steam flow is increased. Although the lowest point shows a fnite steam flowrate, the steam injection rate is zero; the steam flow shown is the steam in the large vessel volume displaced by the liquid flowing into the vessel. As the steam injection rate increases, the circles move upward and turn along the flooding curve. Eventually, the liquid flow direction is reversed, and both liquid and steam flow out of the system through the pressure boundary at the top of channel 10. The predicted flow states agree well with the flooding constraints defined by the transition curves. At low values ofjf (high void fraction), the flooding linit is underpredicted, possibly because the 3-cell model cannot resolve the liquid level accurately. o:\4384-non\4384-17.wpd:lb-04043 17-17

I A sensitivity study was conducted to find stability in the above results by varying the number of cells per channel, by changing the length of the pipe, and then by changing the channel length. The computed results were quite stable as long as the number of cells is greater than one and the channel length (or horizontal cell spacings) remains approximately the size of the pipe diameter. 17-3-4-4 Predicted Water Level Before steam was injected in the computational experiments described in the previous section, a steady-state liquid flow was established through the square channel into the vessel. The liquid fractions computed in each cell are shown in Table 17-3. The equivalent water levels are also shown. It is evident that realization of horizontal stratified flow is predicted. The predicted water levels are compared with the weir flow level (Shames, 1982), no, given by: qT = Dg 2 (2/3 n0 )3 /2 (17-31) for the total volumetric flowrate (qT), the width of the crest (D), and the acceleration of gravity (g). The weir flow level calculations and results for each flowrate are shown in Table 17-3. The water levels predicted with the WCOBRA/TRAC code agree reasonably well near the channel exit (channel 6) with the weir flow calculations over the entire range of flowrates considered. 17-4 CCFL in Hot Leg-to-Steam Generator Flow Path 17-4-1 Introduction During a small break LOCA, the amount of countercurrent flow in the hot legs of a PWR is important in determining the inner reactor vessel mixture level response. One factor influencing the countercurrent flow in the hot leg is the potential for countercurrent flow at the hot leg-to-steam generator inlet plenum connection. If the amount of steam generated in the core by decay heat is large enough, liquid flow from the steam generator inlet plenum to the hot leg may be inhibited. The potential for liquid holdup (flooding) in the hot leg-to-steam generator inlet plenum connection during a small break LOCA is considered in this section. o:\4384-non\4384-17.wpd:lb-04043 17-18

174-2 Physical Processes Figure 17-21 depicts the PWR geometry in the hot leg-to-steam generator inlet plenum connection region. Countercurrent flow at the steam generator tube inlet due to reflux condensation results in liquid falling from the tubes to the bottom of the steam generator inlet plenum where it will collect. The liquid then attempts to flow down the 45-degree incline which joins the steam generator inlet plenum to the horizontal section of the hot leg. Steam flow in the hot leg impinges upon the 45-degree incline and must change direction to flow into the steam generator inlet plenum. The interaction between the steam and liquid tends to restrict countercurrent flow for some steam flow conditions. The specific geometry of the hot leg and the steam generator inlet plenum can influence the countercurrent flow and flooding phenomenon. For instance, the inclination of the hot leg-to-steam generator inlet plenum connection has a complicated effect on flooding, as indicated by Tien and Liu (Tien and Liu, 1979), Hewitt (Hewitt, 1977), and Lee and Bankoff (Lee and Bankoff, 1982). Moreover, liquid entry effects described by Dukler and Smith (Dukler and Smith, 1979) in the steam generator inlet plenum may also influence the flooding phenomena in this region of a PWR. 17-4-2-1 Small-Scale Tests There have been several small-scale experiments (Wongwises, 1996; Ghiaasiaan, et al., 1994; Ohnuki, 1986; Ohnuki, et al., 1988; Wan, 1986; Siddiqui, et al., 1986; Kroleswki, 1980; and Richter, et al., 1978) that have studied countercurrent flow phenomena and flooding in geometric configurations, consisting of horizontal pipes connected to inclined (at various angles) vertical pipes or elbows as in the hot leg-to-steam generator geometry of a PWR. In these small-scale tests, the onset of flooding appears to coincide with interfacial wave instability and growth, which leads to water slug formation in the horizontal piping section, usually near the bend or elbow region. At lower liquid flowrates, the slugging occurs simultaneously with formation of a hydraulic jump near the bend or inclined region where the liquid flow transitions from super-critical to subcritical. At higher liquid flowrates, the water slug formation moves away from the bend (in the horizontal section) and the hydraulic jump is usually weaker or not present at all. This flooding pheonema has also been confirmed by Choi and No (Choi and No, 1995) in experimental studies of nearly horizontal pipes and Kawaji (Kawaji, et al., 1991) in experimental studies of flooding in vertical to inclined pipes. It was observed in these small-scale tests that the geometric configuration of the bend and vertical pipe section strongly influences the location of the hydraulic jump and the water slugging. Gas velocities associated with flooding in these o:\4384non\4384-17.wpd:b-04043 17-19

small-scale geometric configurations were also found to be well below those expected for countercurrent flow in vertical pipes alone. Based upon the small-scale test experience, a special CCFL model would seem to be needed for the hot leg-to-steam generator geometry. However, as discussed in the next section, this is not the case for the full-scale PWR geometry based upon the experimental results from full-scale (geometric) tests. 17-4-2-2 Large-Scale Tests There have been few large- or full-scale experiments for studying countercurrent flow and flooding in hot leg-to-steam generator geometries. While the experimental facility used by Richter was of a larger scale than the numerous small-scale experimental facilities, the UPTF is the only full-scale (geometric) experiment to study the countercurrent flow and flooding prototypic of the hot leg-to-steam generator geometry. The results of the full-scale UPTF test run 37, which is prototypical of PWR steam and liquid flowrates during reflux condensation during a small break LOCA, indicate that CCFL does not occur in the hot leg-to-steam generator flow path. All the water injected in the steam generator drained back into the hot leg unimpeded by the steam injected via the hot leg. Test run 38 of UPTF, which has twice the steam flowrate and three times the liquid injection rate as test run 37, also shows virtually no impedence of water drainback either because 98 percent of the water injected in the steam generator during this test run drains into the hot leg. It takes test run 39, which has nearly three times the steam and three times the liquid injection rate as test run 37, to approach the flooding limit (84 percent of the water drains into the hot leg). Analysis performed by Wang and Mayinger (Wang and Mayinger, 1995) for UPTF test results also supports that no CCFL occurs during reflux conditions in the PWR, as margin to the flooding limit is shown when data from test runs 37 and 38 are plotted against the Richter flooding correlation for the hot leg-to-steam generator geometry. Work by de Bertodano (de Bertodano, 1994), in which a flooding correlation was developed from scaled test data, further supports that flooding is precluded in the hot leg-to-steam generator flow path for a less than 4-percent decay heat power condition in a four-loop PWR. 174-3 Conclusion Based upon the full-scale UPTF test results, CCFL is not expected to occur in the hot leg-to-steam generator geometry during the reflux condensation phase of a small break LOCA, when it o\4384-non\4384-17.wpd:lb-04043 17-20

might possibly be important. Therefore, no special flooding models (beyond those discussed in this section for horizontal and vertical piping flooding) are needed to handle the hot leg-to-steam generator geometry in WCOBRAfl'RAC-SB. 17-5 WCOBRA/TRAC-SB Modelling of Wall Condensation The WCOBRATRAC-SB code calculates wall condensation heat transfer, based on the void fraction of the fluid cell in contact with the wall, as discussed in Section 4 of this document. The correlation of Shah (Shah, 1979) is applied at void fractions [ ]'. For void fractions [ ]C, the heat transfer coefficient computed using Shah is ramped into the heat transfer coefficient from the EPRI correlation (EPRI, 1988). At void fractions exceeding [ . 17-6 Steam Generator Tube Condensation During a smaller size small break LOCA event, the RCS primary depressurizes to an equilibrium pressure slightly above the steam generator secondary side pressure. The primary RCS pressure equilibrates for a time at the pressure at which the primary fluid volume swells due to decay heat and pumped safety injection equals the primary fluid volume shrinkage due to mass lost through the break and through primary to secondary heat transfer. Initially, a small cold leg break is incapable of compensating for the safety injection and decay heat induced fluid volume swell. Primary to secondary heat transfer results and the steam generator secondary side conditions determine the equilibration pressure. Three modes of primary to secondary heat transfer that occur depending on the primary fluid conditions are as follows:

• Subcooled convection heat transfer, which could be forced convection heat transfer or natural circulation convection heat transfer
• Two-phase condensation heat transfer
• Steam condensation heat transfer Condensation will occur at the steam generator tube walls when the tube walls are at a lower temperature than the primary side vapor. Condensation heat transfer as calculated by WCOBRAITRAC-SB during two-phase flow conditions has been validated by simulation of the Semiscale Mod-2A NC tests, as discussed in Section 17-7.

o\4384-non\4384-17.wpd lbO4043 17-21

I The Semiscale NC 60-kW power test is a suitable experiment to simulate to validate the performance of the WCOBRAITRAC-SB computer code in predicting condensation in the steam generator tubes over the void fraction range for small break LOCA conditions. In the high void fraction range, the EPRI correlation (EPRI, 1988) for cocurrent/countercurrent flow film condensation heat transfer is used as described in Section 6, Volume 1, of this document. At low void fractions, the Shah correlation (Shah, 1979) two-phase multiplier to liquid heat transfer coefficients is used. The mass inventory of the primary is continuously reduced during the Semiscale NC tests; therefore, the effect of steam generator tube condensation is observed over a wide range of fluid qualities. In the latter stages of the NC test simulation, a high void fraction (a>0.980) condition exists in the steam generator tube primary fluid. This provides the opportunity to assess the performance of WCOBRA/TRAC-SB in predicting condensation heat transfer coefficients in the high void fraction range. The WCOBRA/TRAC-SB predicted heat transfer coefficients for the high void fraction condition at the top of the SG tubes are found to be in the range of 2500 to 25,000 Btulhr/ft2 , which is in agreement with the values predicted by the Nusselt film condensation theory. 17-7 Simulation of Semiscale Mod-2A NC Test Series Experiments 17-7-1 Introduction The Semiscale facility is a small scale (1:1700) replica of a Westinghouse RCS including all of the major components. Figure 17-22 shows the layout of the major components, as configured for the NC tests. Although there are two loops in the facility, the one scaled as a single loop was blocked during these tests; the reactor vessel upper head was also blocked. The Mod-2A facility modifications focused on small break LOCA phenomena, and extensive instrumentation was installed to measure key phenomena. The Semiscale reactor vessel houses an electrically heated bundle consisting of 25 heater rods; the tests simulated herein are at a power of 60 kW. The overall scaling philosophy used in designing the facility is the maintenance of the power-to-volume ratio, coupled with a 1:1 elevation scaling criteria (Larson, et al., 1980 and Loomis, 1987). 17-7-1-1 Natural Circulation Phenomena The Semiscale NC test series experiments provided information concerning the overall flow and the qualitative interaction of phenomena that occur throughout the various stages of a small break LOCA in a simulated integral RCS. The NC test series consisted of three individual (30-kW, 60-kW, and 100-kW core power) experiments with multiple points each. In all, approximately o:\4384-non\4384-17.wpd:lb-04043 17-22

15 discrete data points were available at given primary system inventories that were established for a fixed core power and secondary mass by draining fluid from the reactor vessel lower plenum. Each discrete inventory was maintained until steady-state (or nearly steady-state) conditions were established. The NC tests were tabulated by Loomis and Soda (Loomis and Soda, 1982), and they provide valuable data for validating the predictive capability of WCOBRA/TRAC-SB for the single-phase, two-phase, and reflux condensation natural circulation modes. One important phenomenon that influences the severity of small break LOCA transients is steam generator tube liquid holdup. This holdup phenomenon was first identified experimentally in a Semiscale small break LOCA experiment (Leonard, 1982). It has since been duplicated in other facilities such as ROSA (Osakabe, et al., 1987) and has been discussed extensively in the open literature (Leonard, 1983 and Loomis, 1985a). Steam generator liquid holdup is the result of condensation due to natural circulation flow in the upflow side of the tubes relatively early during a small break LOCA transient. This holdup is unable to gravity-drain back through the hot leg because it is impeded by high upward steam flowrates; the pressure drop induced by this holdup affects the hydrostatic head balances throughout the RCS. Therefore, the liquid present in the steam generator tubes as a function of total system inventory is an important phenomenon in small break LOCA performance. 17-7-1-2 Applicable Tests Information from the 60-kW core power test conducted in the NC test series at the Semiscale Mod-2A configuration is available in the literature (Loomis, 1985b and Loomis, 1987). Among the NC experiments, the 60-kW experiment is chosen for simulation because its core power, corresponding to 3-percent full power, is closest to the decay power during the natural circulation period for a PWR 3-inch break event. A series of individual test points was generated at various inventories while heat generated in the core was being removed by heat transfer to the steam generator secondary side. Temperatures and pressures are typical of PWR conditions during small break LOCA events. 17-7-2 Description of WCOBRAITRAC-SB Model The WCOBRAflRAC-SB model used to simulate the Semiscale NC tests is described in detail in Section 21 of this document. The same nodalization was used in the NC test prediction with these exceptions: [ o:\4384-non43S4-17.wpd: lb-04043 17-23

I

                                        ].a c 17-7-3 Simulation Results A series of inventory points in the 60-kW NC test (at mass inventories decreasing from 100 percent of nominal until the reflux condensation heat transfer mode was established) were simulated with WCOBRAJTRAC-SB. The results are presented in several ways.

Figure 17-23 shows the system inventory percentage as a function of time to which the WCOBRA/TRAC-SB simulation was set. During the first 50 seconds of each time interval, the inventory was reset downward using a boundary condition. The code was run long enough at each inventory to reach a steady-state prediction of loop flow at that inventory condition. Figure 17-24 shows the natural circulation mass flow rate predicted as a function of system inventory by WMCOBRA/TRAC-SB compared with the data as reported by the experimenters (Loomis and Soda, 1982). The code prediction of peak flowrate agrees well with the data. However, drain of the steam generator tubes occurs earlier in the prediction than in the test. Figures 17-25 and 17-26 show the void fraction in the uppermost nodes of the steam generator tube uphill and downhill, respectively. The draining predicted by COBRA/TRAC-SB as mass inventory decreases takes place when the void fraction exceeds 0.5 in these nodes. At this point, [

                                                                          ]a. Figures 17-27 and 17-28 present condensation rate as a function of time in the upper nodes of the steam generator tube uphill and downhill sides. Condensation is almost constant at inventory levels approaching the draining; therefore, draining is not triggered by a lack of condensation. This is true even though the condensation is underpredicted by WCOBRAfTRAC-SB; not quite all of the steam entering the steam generator is predicted to condense in the code, as it did in the experiment.

Because the draining predicted by WCOBRAfTRAC-SB is a function of void fraction in the steam generator tubes, [ Iax o:A4384-nonW4384-17.wpd:1b-04043 17-24

I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~aI

                                                                            ]a.

Overall, the qualitative prediction of WCOBRA/TRAC-SB shows fair agreement with the Semiscale Mod-2A NC test data. The peak flowrate is matched closely, and the characteristic shape is predicted by the code. The data are similar to that observed in comparable experimental facilities. Figure 17-29 compares the 30-kW Semiscale natural circulation result with that obtained in the PKL facility at a similar power level. The PKL facility (Hein, et al., 1980) is similar to the Semiscale facility except the scaling factor is approximately 134:1 compared with the Semiscale scaling factor of 1705.5:1. As reported in Loomis and Soda (Loomis and Soda, 1982), PKL conducted natural circulation experiments similar to those in Semiscale, but at lower pressure (approximately 3 to 4 MPa versus 6.9 MPa). The results of the experiments were similar. Figure 17-29 compares the loop mass flowrate versus primary system mass inventory for the two systems, indicating similar trends. (The PKL mass flowrate shown was reduced by the ratio of the volume scaling factors used for each facility, and both experiments represent core decay powers of approximately 1.5 percent.) Not only is the overall trend between the two experiments similar, but also the peak two-phase mass flowrate agrees well quantitatively. Furthermore, Loomis and Soda state that the fact that PKL entered the reflux mode at about 80-percent system mass inventory and Semiscale, at about 70-percent inventory constitutes only a slight difference (the uncertainty is +5 percent). The variance between the WCOBRA/TRAC-SB prediction and the Semiscale Mod-2A data for natural circulation and for the time at which the reflux mode begins is similar to that observed between the two comparable test facilities and is characterized as slight by the Semiscale experimenters. Therefore, the WCOBRAfrRAC-SB prediction is judged to be acceptably accurate for small break LOCA analysis. 17-7-4 Conclusions The following conclusions about WCOBRAITRAC-SB are drawn from the Semiscale NC tests simulations:

• The peak natural circulation flowrate is well predicted.
• Draining is a function of the void fraction predicted in the steam generator tubes.

o:\4384-non\4384-17.wpd:lb-04043 17-25

• The difference between the code prediction and the Semiscale data is similar to the difference observed between the Semiscale and PKL data.

Therefore, the two-phase natural circulation and steam generator drain phenomena are adequately predicted. 17-8 References Bajorek, S. M., et al., 1992, "Code Qualification Document for Best Estimate LOCA Analysis Volume I: Models and Correlations," WCAP-12945-P, Vol. 1. Choi, K. and No, H., 1995, "Experimental Studies of Flooding In Nearly Horizontal Pipes," Int. Journal of Multiphase Flow, Vol. 21, pp. 419-436. de Bertodano, M., 1994, "Countercurrent Gas-Liquid Flow in a Pressurized Water Reactor Hot Leg," Nuclear Science and Engineering, Vol. 117, pp. 126-133. Dukler, A. E. and Smith, L., 1979, "Two-Phase Interactions in Countercurrent Flow: Studies of the Flooding Mechanism," NUREG/CR-0617. EPRI Report NP-5700, 1988, "Condensation Inside Tubes," Research Project 1160-3 Final Report. Ghiaasiaan, S., et al., 1994, "Countercurrent Flow Limitation in Inclined Channels with Bends," Nuclear Engineering and Design, Vol. 152, pp. 379-388. Hein, D., et al., 1980, "PKL Experiment Data Report," Report Number R 513/58/80, KWV. Hewitt, G. F., 1977, "Influence of End Conditions, Tube Inclination and Fluid Physical Properties on Flooding in Gas-Liquid Flows," HTFS-RS222, Heat Transfer and Fluid Flow Service. Hsieh, C., et al., 1980, "Countercurrent Air/Water and Steam/Water Flow Above a Perforated Plate," NUREG/CR-1808. o:\4384-non\4384-17.wpd:lb-04043 17-26

Kawaji, M., et al., 1991, "Countercurrent Flooding in Vertical to Inclined Pipes," Exp. Heat Transfer 4, pp. 95-110. Krowlewski, S., 1980, "Flooding Limits in a Simulated Nuclear Hot-Leg," B.Sc. Thesis, MIT. Larson, T. K., et al., 1980, "Scaling Criteria and an Assessment of Semiscale Mod-3 Scaling for Small Break Loss Of Coolant Transients," EGG-SEMI-5121. Lee, S. C. and Bankoff, S. G., 1982, "Stability of Steam-Water Countercurrent Flow In An Inclined Channel," ASME Paper No. 82-WAIHT-6. Leonard, M. T., 1982, "Vessel Coolant Mass Depletion During a Small Break LOCA," EGG-SEMI-6010. Leonard, M. T., 1983, "An Analytical Study of a Small Break Loss-Of-Coolant Accident with Upper Head Injection," Nuclear Technology, Vol. 62. Long, R. R., 1956, "Solitary Wave in the One- and Two-Fluid Systems," Tellus, Vol. 8. Loomis, G. G. and Soda, K., 1982, "Results of the Semiscale Mod-2A Natural Circulation Experiments," NUREG/CR-2335 (EGG-2200). Loomis, G. G., 1985a, "Semiscale Liquid Holdup Investigations: A Comparison of Small Break LOCA Tests Performed in the Semiscale Mod-2A and Mod-2C Facilities," 13th Annual Water Reactor Safety Research Information Meeting, Gaithersburg, MD. Loomis, G. G., 1985b, "Summary of Semiscale Small Break Loss-of-Coolant Accident Experiments (1979 to 1985)," NUREG/CR-4393 (EGG-2419). Loomis, G. G., 1987, "Summary of the Semiscale Program (1965-1986)," NUREG/CR4945 (EGG-2509). Ohnuki, A., 1986, "Experimental Study of Counter-Current Two-Phase Flow in Horizontal Tube Connected to Inclined Riser," Joumal of Nuclear Science and Technology, Vol. 23, pp. 219-232. Ohnuki, A., et al., 1988, "Scale Effects on Countercurrent Gas-liquid Flow in a Horizontal Tube Connected to an Inclined Riser," Nuclear Engineering and Design, Vol. 107, pp. 283-294. o:\4384-non\4384-17.wpd:lb-04043 17-27

I Osakabe, M., et al., 1987, "Core Liquid Level Depression due to Manometric Effect during PWR Small Break LOCA," J. of Nuclear Science and Technology 24(2), pp. 103-110. Richter, H., et al., 1978, "De-entrainment and Countercurrent Air-Water Flow in a Model PWR Hot-leg," USNRC Report NRC-0193-9. Shah, M. M., 1979, "A General Correlation for Heat Transfer During Film Condensation Inside Pipes," Int. J. Heat and Mass Transfer, V22 N4. Shames, L H., 1982, "Mechanics of Fluids," McGraw-Hill, New York, N. Y. Siddiqui, H., et al., 1986, "Flooding in an Elbow Between a Vertical and a Horizontal or Near-horizontal Pipe," Int. Journal of Multiphase Flow, Vol. 12, pp. 531-541. Taitel, Y. and Dukler, A. E., 1976, "A Model for Predicting Flow Regime Transitions in Horizontal and Near Horizontal Gas-Liquid Flow," AIChE Journal, 22. Takeuchi, K. and Young, M.Y., 1983, "Generalized Drift Flux Correlation for Steam Generators," First Proceedings of Nuclear Thermal Hydraulics, American Nuclear Society,

p. 102.

Takeuchi, K., et al., 1992, "Generalized Drift Flux Correlation for Vertical Flow," Nuclear Science & Engineering, 112, 170. Tien, C. L. and Liu, C. P., 1979, "Survey on Vertical Two-Phase Countercurrent Flooding," EPRI NP-984. Wallis, G. B., 1969, One Dimensional Two Phase Flow, McGraw-Hill. Wallis, G. B. and Dobson, J. E., 1973, "The Onset of Slugging in Horizontal Stratified Air-Water Flow," Int. J. Multiphase Flow, Vol. 1. Wan, P., 1986, "Countercurrent Steam-Water Flow in an Upright 90° Elbow," Proc. 8h International Heat Transfer Conference, pp. 2313-2318. o:\4384-non\4384-17.wpd:1b-04043 17-28

Wang, M. and Mayinger, F., 1995, "Simulation and Analysis of Thermal-Hydraulic Phenomena in a PWR Hot Leg Related to SBLOCA," Nuclear Engineering and Design, Vol. 155, pp. 643-652. Wongwises, S., 1996, "Two-Phase Countercurrent Flow in a Model of a Pressurized Water Reactor Hot Leg," Nuclear Engineering and Design, Vol. 166, pp. 121-133. Zuber, N., 1980, "Problems in Modelling of Small Break LOCA," NUREG-0724. o:4384non\4384-17.wpd:lb04043 17-29

I Table 17-1 Water Injection Rates Liquid Flow jf (fts) Case 1000 (psia) 35 (psia) 1 0.0239 0.0238 2 0.0718 0.0713 3 0.1196 0.1188 4 0.1675 0.1664 5 0.2153 0.2139 6 0.248 Table 17-2 Steam Injection Rates 1000 psia 35 psia Time (sec) jg(ftIs) jg(ftls) 0-80 0.0 0.0 80-300 0-7.123 0-74.02

                                                                           'I o:\4384-non\43&4-17.wpd:lb-04043         17-30

Table 17-3 Predicted Water Levels in WCOBRAJTRAC Horizontal Channel Compared With Weir Flow Theory Channel 10 9 8 7 6 5 WL=70 bm/sec , = 0.168 ft/s) aL Cell 2 0.0 0.0 0.0 0.0 0.0 0.0 Cell 1 0.674 0.617 0.494 0.424 0.359 0.160 Water level (ft.) 0.674 0.617 0.494 0.424 0.359 0.160 Weir flow level = 0.30 ft. WL=1 7 0 lbm/sec (j, = 0.408 ft/s) aL Cell 2 0.0 0.0 0.0 0.0 0.0 0.0 Cell 1 0.993 0.923 0.988 0.876 0.763 0.321 Water level (ft.) 0.993 0.923 0.988 0.876 0.763 0.321 Weir flow level = 0.54 ft. WL=3 2 5 lbm/sec (I = 0.780 ftls) aL Cell 2 0.030 0.104 0.030 0.0 0.0 0.0 Cell 1 0.992 0.992 0.994 0.942 0.892 0.499 Water level (ft.) 1.022 1.096 1.024 0.942 0.892 0.499 Weir flow level = 0.83 ft. WL= 6 5 0 lbm/sec (j1 = 1.559 ftls) aL Cell 2 0.733 0.988 0.525 0.351 0.222 0.0 Cell 1 0.992 0.999 0.989 0.991 0.980 0.581 Water level (ft.) 1.725 1.987 1.514 1.342 1.202 0.581 Weir flow level = 1.32 ft. WL= 9 7 5 lbm/sec (I = 2.339 ft's) aL Ce 3 0.134 0.016 0.040 0.023 0.0 0.0 Cell 2 0.992 0.990 0.846 0.683 0.528 0.296 Cell 1 0.995 1.0 0.993 0.992 0.998 0.746 Water level (ft.) 2.121 2.006 1.879 1.698 1.520 1.042 Weir flow level = 1.72 ft. o:\4384-non\4384-17.wpd:lb-04043 17-31

I Liquid EL------- 71 ----------- I5I I. Steam

                                                       ===---------------

2 ~ ~ 11 31 I I I Pipe Model Figure 17-1. Flooding Model for a Vertical WCOBRAJTRAC Channel o:\4384-non\4384-17.wpd1 b-04043 17-32

3D PIPE F I o o d i n g T e s t s L i qu i d i n j e c t i o n Ro t e MTHO0007 2 2 0 TOTAL UQIJD FLOW 8 E-O 1 6E-01 E

        -0
             . 4E-0 1 0
        ~C
             . 2E-01 0

0 n A e% I n C f7 ' flfl n Ju u u IU U I JU LuLl 4 in T ime (s) Figure 17-2. Liquid Injection Rate for Case 3 (1000 psi) o:\4384non\4384-17.wpd:1b404043 17-33

U I Figure 17-3. Vapor Mass Flowrate at Middle of Pipe for Case 3 (1000 psi) o\434-non\4384-17.wpd:b-04043 17-34

ax, Figure 17-4. Liquid Mass Flowrate at Middle of Pipe for Case 3 (1000 psi) o:43S4-non\4384-17.wpd:lb-O4043 17-35

3D P I PE F I ood i ng Tes t s Wg v s Wf

           *

• FGM 5 6 0 VAP AXIAL MASS FLOW

              .4 U

3 I~~~~~~~~~~~~~

             .2 ci I~~~~~~~~~

E

     -0
                                                                      *~~ 0                U U~~
             .1 I-I~~~~~~

n I m U I I~~~~~ U~~~~~ 0 I

           -. 1   -

II T l [ I l I l I I [ l I I z I l f r I r I l Il I Il I l l I I I I l l r

                -. 25          -. 2         _ , 15     -.      +OO   - . 5 .- O 1                . 5 E O1     . 1 E+0o Wf (Ibm/sec)

Figure 17-5. Vapor Flow (Wg) Versus Liquid Flow (Wf) for Case 3 (1000 psi) o:\4384-non\4384-17.wpd:lb-04043 17-36

3D PIPE Ft 0 0 d i n g T e s t s FLOOD I N CU RV a

• MTHOO006 5 6 0 VAP AXIAL MASS FLOW 1.5 5

0

            -. 5
                                                            -j f Figure 17-6. ig Versus j; for Case 3 (1000 psi) o:\4384-non\4384-17.wpd:lb-04043                    17-37

3D PIPE Flooding Tests sqrt(jg*) vs. sqrt(j fo) a a .U.l O, . 5 AP A L AS P rtOW

1. ,.

      =      -5                                                   -

3D P.IPE Flooding Tests 3D PIPE Flooding Tests sqr t( jgs ) vs. sqrt(j f) sqrt(jg*) vs. sqrt(jf*) U *M1M eSS I S S o vAp ARiAS MAss rLow a UMYXO GIII S I O VAP AXIAL OASS FLOs 1.5 - T* I -

      =-.5-                                                       -

Figure 17-7a. Typical Flooding Results for Vertical Pipe (ID = 1.6 inches) o:\4384-non\4384-17.wpd:1b-4043 17-38

1.1 0.9 0.8 0.7 0.6 0.5 0.4-0.2 0.1 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 jf a ifin = 0.0239 & f,in = 0.1675

                         +       f,  = 0.0718          x    ifin = 0.2153
                          °     ifi =0.1196            v    if,i = 0.248 Figure 17-7b. Countercurrent Flow Map Predicted by WCOBRA/TRAC o:\4384-non\4384-17.wpd:lb-04043                       17-39

Perforated Plate Steam Injection Figure 17-8. Flooding Model 1 for a Perforated Plate oA4384-non\43S4-17.wpd:1b-04043 17-40

Pe r f o r a t ed P I a t e F ood ng Analyses MODEL 1 ( 3 5 p s i a ) sq r t ( Hg )/C5 v s. sqr t (H f * ) /C U J MTHOOO16 1 0 WCOBIBRArRC 1.5 - 1 C, L. Cr c,,

                       - 5 sqrt( Hf*)/C Figure 17-9. Flooding Velocities for Saturated Liquid and Vapor at 35 psia and j, = 3.3 ft/s Compared With Northwestern Flooding Limit (ECOBRAfIRAC MOD7A) o:\4384-non\4384-17.wpd:lb-04043                 17-41

Pe r f ora ted P I a t e Flooding Analyses MODEL (35 psi a) s q r t (Hg )/C 5 V s. sqrt(Hf *)/c II NMTHOO016 5 I ° WCOBRATRAC

           .5 I

I-

         .5                                                                                 .j" c) v 0
            -_5                                   .5 sqrt(         Hf*)/C Figure 17-10. Flooding Velocities for Saturated Liquid and Vapor at 35 psia andj, = 8.0 ft/s Compared With Northwestern Flooding Limit (COBRAITRAC MOD7A) o\4384-non\4384-17.wpd:lb-04043              17-42

Perforated P I ot e F I ood i ng Analyses Mode I 1 ( 1000 ps i sqrt (Hg*) '/C v s . s q r t ( H f ) /C a *MTHOO016 5 1 0 \ WCOBRNTRAC l.5 - 1-Q

          . 5 -.

ci, 0-

       -. 5--
             -. 5 sqrt( Hf*)/C Figure 17-11. Flooding Velocities for Saturated Liquid and Vapor at 1000 psia and j, = 3.3 ft/s Compared With Northwestern Flooding Limit fXCOBRAMTRAC MOD7A) o:\4384-non\438417.wpd:Ib.04043                 17-43

I

                                                                                        ^LI Figure 17-12. Liquid Mass Flowrates Through Perforated Plate at 35 psia andj, = 8.0 ft's COBRATRAC MOD7A) o:\4384-non\4384-17.wpd:lb-04043           17-44

AILC Figure 17-13. Vapor Mass Flowrates Through Perforated Plate at 35 psia andj, = 8.0 ft/s QyCOBRAIRAC MOD7A) o:W384-nonV$384-17.wpd:Ib-04043 17-45

1.1 0.9 0.8 0.7 L) 0.6 I 0.5 0.4-0.3 0.2 0.1 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 Northwestern Correlation

                    - - - - - WCOBRATRAC Prediction Figure 17-14. Predicted Flow Conditions and CCFL for Perforated Plate at High Pressure o:\4384-non\4384-17.wpd:lb-04043                      17-46

C.) 0 0.1 0.2 0.3 0.4 0.S 0.6 0.7 0.8 0.9 1 1.1

                                                      +j-c Northwestern Correlation
                    - - - - -WCOBRAIRAC Prediction Figure 17-15. Predicted Flow Conditions and CCFL of a Perforated Plate at Low Pressure o:\4384-non\4384-17.wpd:lb-04043                      17-47

1.1 1

                         .1 0.9
                                      \

0.8 1 C-) 0.7 [l 0.6 00> 0.5 0 0.4 0 0 0.3 1 o0 0.2 0.1 0 1 1 p Ip I 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 Northwestem Correlation

                                           -WCOBRAiTRAC Prediction Figure 17-16. CCFL Data for Perforated Plate and Air/Water at Atmospheric Conditions (Hsieh, 1980) o-\4384-non\4384-17.wpd:Ib-04043                         17-48

( ( (C P 0 I i PTI I L CHANNELS 9

      -A    CONSTANT 41 3

PRESSURE -O-- GAPS 14 STEAM INJECTION w I 6 -11 4 I... . I I I I I - - I. I - I~~~~~~~~~~~~ - - - - -II \0 M. _____ ____ _____ __------- ------- ____- - _ _ _ l _ _ _ _ _ 51 --------- I I I E CONTAINER 0 VESSEL m 0 R. . - A - -J . L _ J 00 E LIQUID INJECTION

    .w.m

ig COUNTER-CURRENT CO-CURRENT

                                -p Figure 17-18. Drift Flux in a Horizontal Stratified Flow and Flooding Curves 4 4 o.\4384-non\4384-17.wpd:lb-0 0 3           17-50

1 LL ui 0

        -J LL w

I-0 cn

        -e 0

CD co

        -j z

0 zw co U) 0 0 1 SqR OF DIMENSIONLESS VOLUMETRIC UQUID FLOW RATE. Figure 17-19. Flooding Curves for Horizontal Stratified Flow o:\4384non\438417.wpd:lb 04043 17-51

1 0 LL co C) a 0

3

     ; 9 z

0 QO z w cr O ~ 1-a/;E 1 SqR OF DIMENSIONLESS VOLUMET RIC UQUID FLOW RATE. Figure 17-20. Computed Horizontal low State and Flooding Curves o-\4384-non43S4-17.wpd:b-04043 17-52

HOT LEG Figure 17-21. PWR Hot Leg-to-Steam Generator Inlet Plenum Connection o.4384-non\434-17.wpd:Ib-4043 17-53

loop aImospheric dump valve (ADV) itnlact loop maln steam isolation valve (MSIV)

                                                                  -Intact loop Type It steam generator
                                                                  "'Pressurfzer INTACTLOOP
                                                                  % Intact loop valve colls Figure 17-22. Semiscale Mod-2A System for NC Tests o:\4384-non\4384-17.wpd:lb-04043                   17-54

500 2250 ~ ~ 75 27g0

                                  %~~~~~~?

0 60 ~~~~~~~9 o 70 3000 320 3buo 76, 490D 7 o?0

                                                                      'Y2pD 9Pv 7D%

5200

                                                           &2 7 54DD Figure 17-23. WCOBRAITRAC-SB Inventory Versus Time, Semiscale Mod-2A NC Test o:\4384-nonM4384-17.wpd:Ib-04043                   17-55

SEMISCALE NC 60 KW

• E DATA 4 0

             ----              FLM                    8            0 2

1 u/u I

1. 5

_______________________ ______________________ I cn I E

  -D
                                                                                           .1 I

I I

  .0

______ I - I I 0 U I U-cn U, I 0 5 __________________ __________________ II

                                                                                        *1 I

I U U I  ; ...i I I I ______ ______ ______ 0

                .)U                       bU              7U                      du                            9U            I 0a Inventory                             (7)

• Data

                      ---         WCOBRArTRAC Figure 17-24. Comparison of WCOBRAJTRAC-SB Prediction and Semiscale NC 60-kW (3-Percent Power) Data o:\43S4-non\4384-17.wpd: I b-04043                           17-56

1

       .8 c

C,

  .    .6
.5
       .4
        .2 0

Vapor fraction, lower middle elevation Vapor fraction, middle elevation Vapor fraction, upper middle elevation Vapor fraction, top elevation Figure 17-25. Predicted Void Fraction in the Steam Generator Tubes Uphill Region, Semiscale NC 60-kW Test o:\4384-non\4384-17.wpd:lb-04043 17-57

         .8 . . . . . . . . . . . . . . . . . . .        .   . . . .. . .   .... . ..                  .. . . .

o .- . .

         .4 .    .     .           ....            .....            ...                 ......

0 2 .. ..... . . . . . . . .. . . . . . . . 500 1000 1500 2000 2500 3000 3500 4 Time (s) Vapor fraction, lower middle elevation - - - - - -- - Vapor fraction, middle elevation Vapor fraction, upper middle elevation - - - - - - Vapor fraction, top elevation Figure 17-26. Predicted Void Fraction in the Steam Generator Tubes Downhill Region, Semiscale NC 60-kW Test o:\4384-non\4384-17.wpd:lb-04043 17-58

1) .2 '

uI~~~~I

             .15 -.                                   . .     . . . .. . . . . . .       .  . .. . .  .
               ~
               -   ~~ ~ ~ ~ ~ ~ ~ ~           .....               .  .    .. .I      I...I E24
        -5E-DO 1.. . . . . . . ...                  .   .
              .0 50           1000      1500            2000       2500        3000       3500      4000 0^                                                      Time (s) z 0

Lower middle elevation rate - - - - ---- Middle elevation rate Upper middle elevation rate

            - - Top elevation rate Figure 17-27. Predicted Condensation Rate in the Steam Generator Tubes Uphill Region, Semiscale NC 60-kW Test o:%4384-nonW384-17.wpd:Ib-04043                          17-59

   '44 U
              .2 U
            .15 0

0: .1 A U) 5E-01 E. 4 E4 N 0 z 2000 2500 4000 Time (s) 0 U Lower middle elevation rate

---

Middle elevation rate Upper middle elevation rate Top elevation rate Figure 17-28. Predicted Condensation Rate in the Steam Generator Tubes Downhill Region, Semiscale NC 60-kW Test o:\4384-non\438417.wpd:1b04043 17-60

M=b

                        -v   °- r-r%6%ww rsVJ A   $%Is   I u.JJ
        .0.5       _

0.4 _- *:

        -f   00                                                       /^

0O.3-0.2 _ 0.1 _.  ! 0 t *ck/ 50 55 60 65. 70 75 80 85 90 95 Sysaem mass inventory (%) Figure 17-29. Comparison of Semiscale and PKL Natural Circulation Flowrates o:\4384-non\4384-17.wpd:lb-04043 17-61

1" o:A43S4-non\4384-17.wpd: lb-04043 17-62

SECTION 18 HORIZONTAL STRATIFIED FLOW BENCHMARKS 18-1 Introduction The predicted performance of a PWR during a small break LOCA transient is to a large extent determined by the two-phase flow regime present in the horizontal pipes of the RCS. The duration of the natural circulation period, the loop seal clearing process, and the break flow composition are a consequence of the flow regime(s) in the hot leg, pump suction leg, and cold leg horizontal sections, respectively. In the WCOBRAfTRAC-SB computer code, the Taitel and Dukler flow regime map (Taitel and Dukler, 1976) is used to define the horizontal pipe flow regime. At the relatively low flowrates associated with the break size range of small break LOCA, the horizontal two-phase flow is in the stratified wavy and/or stratified smooth flow regimes most of the time. Therefore, the prediction of small break LOCA phenomena in the stratified flow regimes is of central importance for the horizontal RCS piping. Within WCOBRAIIRAC-SB logic, the horizontal flow regime is identified [

                           ]ac using the Taitel and Dukler regime map. If the path is determined to be stratified, the Jensen and Yuen model (Jensen and Yuen, 1982) is applied to calculate the interfacial drag and condensation that occurs; entrainment at the interface between gas and liquid is calculated according to the Kataoka and Ishii model (Kataoka and Ishii, 1983). Because the interfacial drag, condensation, and entrainment modelling for horizontal stratified flow are basic processes that are directly related to high-ranked items in the small break LOCA PIRT in Volume 1 of this document, individual validation of each of these models is needed to confirm their accuracy. This is accomplished using the experimental WCOBRA/TRAC-SB simulations presented in the following sections.

18-2 Physical Processes In the condition of a smooth, equilibrium-stratified flow, the wall resistance of the liquid is similar to that for open-channel flow and that of the gas is similar to closed-duct flow. Because the gas phase velocity is much larger than the velocity at the gas-liquid interface, the gas side interfacial shear stress is evaluated using the equation for gas wall shear. The interfacial drag is thus easily defined theoretically. o:\4384-non\4284-1S.wpd:lb.040403 18-l

I Entrainment from the liquid film at the stratified flow two-phase interface is important in determining the mass distribution in the RCS during a small break LOCA. This is particularly true in establishing the liquid fraction of flow through the break once the break location is uncovered. Liquid that exits the RCS through the break is no longer available to possibly contribute to core liquid inventory and to maintain the core in a covered state. Also, entrainment from the stratified two-phase interface in the hot leg affects the natural circulation period in the small break LOCA, and entrainment from the residual liquid in the horizontal leg of the pump suction leg affects the loop seal clearing. Condensation of steam by the subcooled water of the safety injection jet has been separately addressed in Section 14 in this volume. Condensation of steam in the cold leg remains important in the depressurization of the RCS to the point where safety injection exceeds break flow and the recovery period of the small break LOCA begins. 18-3 WCOBRA/TRAC-SB Horizontal Stratified Flow Models Important phenomena in the production of horizontal, stratified flow during a small break LOCA - the interfacial drag, entrainment, and condensation - are discussed in this section. 18-3-1 Interfacial Drag The models and correlations used to calculate interfacial drag in horizontal stratified flow are described in Section 4-6 in Volume 1 of this document. In particular, the work reported by Jensen and Yuen (Jensen and Yuen, 1982) is used. 18-3-2 Entrainment Section 4-6 describes the models and correlations in WCOBRAJIRAC-SB that are used to calculate the entrainment and de-entrainment processes. Entrainment is the result of interfacial shear between vapor and liquid film. In WCOBRATAC-SB, liquid is moved from the continuous liquid field to the entrained field when the interfacial shear forces acting on the liquid o4384-non\4284-1 8wpd:1b-040403 18-2

are sufficient. In de-entrainment, liquid is moved from the entrained field to the continuous liquid field. A summary of the applicable models in WCOBRAITRAC-SB is as follows:

• Entrainment in Film Flow WCOBRAITRAC determines film entrainment rates by comparing the entrainment rate based on a stable film flow to an empirical entrainment rate based on the work of Walley (Walley, et al., 1973).
• Entrainment in Bottom Reflood The model for entrainment in the core near the quench front is based on a model by Kataoka and Ishii (Kataoka and Ishii, 1983) assuming vapor bubbling through a liquid pool.
• Entrainment at a Horizontally Stratified Surface In small break LOCA events, if the vapor velocity is sufficient, entrainment can occur from a horizontal interface of vapor and liquid.
• De-entrainment in Film Flow The model to estimate the de-entrainment of entrained drops into the continuous liquid field uses an empirical model by Cousins (Cousins, et al., 1965).
• Crossflow De-entrainment Entrained liquid in the upper plenum can de-entrain on structures there as the two-phase mixture flows from the vessel into the hot legs. WCOBRAITRAC uses a model based on experiments by Dallman and Kirchner (Dalman and Kirchner, 1980) to determine the amount of de-entrainment in the upper plenum and other regions of the reactor vessel.

o:4384-non\4284-1 8.wpd:b-040403 18-3

I

• De-entrainment at Area Changes De-entrainment occurs as a two-phase mixture encounters a flow restriction such as a tie plate. WCOBRAfIRAC uses a simple area ratio to de-entrain a fraction of the droplet field where an area reduction occurs in the reactor vessel.

De-entrainment at Solid Surfaces and Liquid Pools Drops are assumed to de-entrain when the drops flow into a cell with a solid surface at the opposite face or when the drops flow into a cell which is in a bubbly flow regime. The small break LOCA PIRT presented in Volume 1 identified entrainment as a high-ranked phenomenon only during loop seal clearing. Therefore, WCOBRAiTRAC-SB simulations of loop seal clearing in small break LOCA scenarios presented in Section 16 in this volume provide the necessary validation for code prediction of the entrainment from a stratified interface. 18-3-3 Condensation Section 14 in this volume presents the validation of the model for condensation on the safety injection jet in a small break LOCA transient. When this location is not being modelled, WCOBRAJRAC-SB uses a model for interfacial heat and mass transfer similar to other best estimate codes. As described in Section 5 in Volume 1 of this document, four components are evaluated to calculate interfacial heat and mass transfer; they may be described as HASCL(T,-T) FSCL = H-H Vf

                  =   HASHL(TI-Ti) rSIL  =       Hg-H (18-1)

HAscv(T -T.) rs = H-Hf HASHBV(Tv-Td) r'SV= _ Hg-H_ o:\4384-non\4284-1 8.wpd:lb040403 18-4

where: FscL = condensation to subcooled liquid

         "'sHL = evaporation from superheated liquid rscv = condensation from subcooled vapor
           'sN = evaporation to superheated vapor Figure 18-1 provides a pictorial representation of the WCOBRAJIRAC-SB approach. [
            ]a' This term is described in Section 6-4 of this document.

Overall, condensation of vapor in WCOBRAITRAC-SB [

                                                                          ]axc Table 18-1 is a summary, according to the Equation 18-1 terminology, of the interfacial heat transfer models used in the vessel component of WCOBRAfTRAC-SB. [
                                                    ]a o:4384-nonW284-18.wpd:Ib-040403                    18-5

I [

                                                      ]a,c In the following sections, an additional set of experiments is examined which focuses on condensation in a geometry similar to the PWR cold leg.

18-4 Assessment of WCOBRA/TRAC-SB Horizontal Stratified Flow Models The performance of the horizontal stratified flow models in WCOBRAJTRAC-SB must be established in predicting interfacial drag and condensation heat transfer for a pertinent single-effect test to demonstrate that the models are adequate for small break LOCA applications. The interfacial drag predictive capability is validated against relevant experimental data (Lim, et al., 1981); these data are also used to validate the interfacial condensation heat transfer. 184-1 Test Facility Description and Modelling The test facility of Lim (Lim, et al., 1981) used a rectangular channel to measure condensation of steam in cocurrent, horizontal flow. The channel was constructed of stainless steel with pyrex glass windows; its dimensions were 160.1 cm long, 6.35 cm high, and 30.48 cm wide. Data were taken in the course of 35 runs. Controlled parameters in the experiments included water and steam inlet temperatures, mass flowrates, and water layer thickness at the inlet. The range of steam (maximum velocity 18 m/s) and water (maximum velocity 41 cmls) flowrates were restricted by either the initiation of bridging phenomena or the occurrence of a hydraulic jump. Inlet steam pressure was approximately 1 atmosphere. Steam velocity, static pressure (for some experiments), and water layer thickness were measured at five locations along the channel. The water inlet temperature was also measured. Figure 18-2 is a schematic diagram of the experimental system. Figure 18-3 presents the WCOBRAfI`RAC noding of the test facility. [

                                                                                        ]ac o-.\4384-non\4284-18.wpd:lb.040403                18-6

I

                                 ]a,c As shown in Figure 18-3, the experimental channel is modeled axially [

laC. This was considered sufficient to provide enough resolution to compare with experimental measurements, which are available at only five axial locations. The experimental channel is divided [

                                                       ]ac The experimental report (Lim, et al., 1981) offers no data on liquid level in the discharge tank during the experiments and on the tank dimensions. Because it is impractical to simulate a constant liquid level in the tank due to condensation in the channel, the liquid level in the tank was allowed to rise during the simulation, but it was always kept below the liquid level in the channel. Condensation was tumed off [
                    ' to minimize the effect of the discharge tank on the channel flow.

Ia,c The liquid level at the channel inlet [

                                                 ]:C As shown in Figures 18-4 and 18-5, the liquid profile away from the channel inlet is determined only by the steam and water flowrates. The "line" in Figure 18-5 is a linear correlation plane oriented in parallel to the reader's line of sight.

Because essentially all of the variation in the liquid water thickness in the experimental channel can be attributed to the variations in steam and water flowrates, the effect of the initial water layer thickness on the flow pattern away from the inlet can be ignored. o:\4384-non\4284-18.wpd:lb-040403 18-7

I The experimental results used in this analysis are reported to be at steady-state. That is, the water level, pressure, temperature, and steam flow in the channel were stable and not varying significantly. The WCOBRA/TRAC-SB simulations were run [ Ia.c. 18-4-2 Calculational Results A total of 35 tests are reported in Lim (Lim, et al., 1981). Those tests in which the horizontal two-phase flow is fully within the wavy or stratified flow regimes (32 in number) were simulated. The experimental results and test conditions for the tests simulated with WCOBRAtrRAC-SB are shown in Table 18-2. Steam density and steam and water velocities were input as boundary conditions in the model's steam and liquid fill components, respectively. In Table 18-2, steam flowrate and water layer thickness data at locations 1, 2, 3, 4, and 5 correspond to 6.18, 12.05, 23.08, 34.18, and 48.14 inches from the experimental channel inlet. Static pressure difference measurements at 4.88, 10.75, 21.77, 32.87, and 47.24 inches are listed as being at locations 1 through 5. Nomenclature is provided on the table. Steam density input is calculated using NISTIASME steam properties for given values of the steam inlet temperature and constant pressure of 16 psi. Due to small variations in the liquid temperature and density among the tests and along the experimental channel, a constant liquid density corresponding to the average liquid temperature of 148.6 0 F is assumed. Steam and water inlet velocities in the model fill components (Figure 18-3) are calculated using a constant flow area of 0.2083 ft2. Figures 18-6 through 18-11 provide, for a typical case (run 275), the predicted results and the comparison between experimental data and the WCOBRAtRAC-SB predictions. For calculated quantities, stable or periodic (at one or two axial locations) behavior is observed over the duration of the test (Figures 18-6, 18-8, and 18-10). There is a reasonably good agreement between the measured and predicted average values of liquid level and pressure drop' in the channel as seen in Figures 18-7 and 18-9. While the liquid level at 47.27 inches is significantly underpredicted, the observed trend of the liquid level to recover toward the channel outlet is well Note that the pressure actually increases as the steam flow proceeds through the channel. o:\4384-non\4284-18.wpd:1b-040403 18-8

reproduced by WCOBRAHrRAC-SB (Figure 18-7). WCOBRAITRAC-SB overpredicted the steam flowrate axially as seen in Figure 18- 1; underpredicting the steam condensation rate is the cause. This matter was investigated further; condensation heat transfer correlations used in WCOBRAJIRAC-SB (Jensen and Yuen, 1982), and one derived from the experimental data, were compared to each other for typical flow conditions in the channel. This comparison is presented in Figure 18-12. The altemative correlation for a smooth interface based on this test data (Lim, et al., 1981) is given by: Nu.,: =A0.631 * (Re )0 58

                                                       . (Re )009 (Pr,                        (18-1) where:

NUMW = is the Nusselt number (Nu), equals 1344 for case 275 The principal difference between the correlations is that the Nu value in WCOBRAIRAC-SB is [

                                 ]c The cumulative results of all tests simulated are shown in Figures 18-13 through 18-16, which show scatter plots of predicted versus measured quantities of the liquid level, steam mass flowrate, liquid temperature at the channel exit, and the pressure drop in the channel, respectively. For most of the cases, liquid level predictions are within +/- 0.2 inches of the measurements. The steam flowrate is overestimated almost everywhere in the test section, particularly near the channel exit. As a result, the liquid temperature at the channel exit is underpredicted by 20° to 40°F. The large majority (approximately 80%) of the pressure drop predictions are within + 33 percent of the experimental data, as shown in Figure 18-16.

o:\4384-non\4284-18.wpd:1b4)40403 18-9

18-5 Conclusions WCOBRAITRAC-SB predictions of two-phase flow in a horizontal channel were verified against data of steam condensation in a rectangular channel with cocurrent water flow at atmospheric pressure. A model of the experimental channel, consisting of [ Ia,c was developed. The pertinent cases among the 35 test cases reported in Lim (Lim et al., 1981) were simulated. For most of the cases, liquid level predictions are within + 0.2 inches of the measurements. Depending on the axial position, steam flowrate can be overestimated by a factor of 2 or more (near the channel exit). As a result, the liquid temperature at the channel exit is underpredicted by 20 to 40°F. To address this, values of the condensation heat transfer coefficient calculated by the code were compared with those given by the correlation used in WCOBRA/TRAC-SB and one derived from the experimental data. The difference in the condensation heat transfer coefficient is determined to be due to the correlation used in the code. Condensation heat transfer in horizontal stratified flow will be ranged in the PWR sensitivity study to address this discrepancy and considered in the uncertainty methodology. Most of the pressure drop predictions are within + 33 percent of the experimental data, and the number of points for which the pressure drop is underpredicted is approximately the same as the number for which it is overpredicted. Inasmuch as steam velocities are low when horizontal stratified flow conditions exist in PWR loop pipes during a small break LOCA event, the pressure drop prediction uncertainties of this model are judged to be unimportant relative to the total hydraulic pressure drop of the RCS. Nevertheless, interfacial drag variations in the horizontal stratified flow regime are considered in the uncertainty methodology. 18-6 References Cousins, L. B., et al., 1965, "Liquid Mass Transfer in Annular Two-Phase Flow," Paper C4, Symposium on Two-Phase Flow, Vol. 2, Exeter, England. Dallman, J. C. and Kirchner, W. L., 1980, "De-Entrainment Phenomena on Vertical Tubes in Droplet Cross Flow," NUREG/CR-1421. Lim, I S., et al., 1981, "Cocurrent Steam-water Flow in a Horizontal Channel," NUREG/CR-2289. o.4384-non\4284-18.wpd:lb040403 18-10

Jensen, R. J. and Yuen M. C., 1982, "Interphase Transport in Horizontal Stratified Cocurrent Flow," NUREG/CR-2334. Kataoka, I. and Ishii, M., 1983, "Mechanistic Modelling and Correlations for Pool Entrainment Phenomena," NUREG/CR-3304. Taitel, Y. and Dukler, A. E., 1976, "A Model for Predicting Flow Regime Transitions in Horizontal and Near Horizontal Gas-Liquid," AIChE Journal, Vol. 22. Walley, P. B., et al., 1973, "Experimental Wave and Entrainment Measurements in Vertical Annular Two-Phase Flow," AERE-R7521, Atomic Energy Research Establishment, Harwell, England. o:\4384-non\4284-18.wpd:b-040403 18-11

0 Table 18-1 WCOBRA/TRAC-SB 3-D Interfacial Heat Transfer Models [ w 00 0-I IC

Table 18-2 Test Matrix Parameters Location W, TG" TL" TL-No. Units(a Inlet 1 2 3 4 5 (Ib/s) (°F) (OF) (OF) WG (ib/s) 0.09 0.083 0.077 0.069 0.065 0.064 6 211 L(in) 0.623 0.534 0.393 0.223 0.222 0.241 0.866 281 76.7 160 AP(psi) 0 7E-05 lE-04 2E-04 3E-04 3E-04 WG (bis) 0.09 0.082 0.074 0.063 0.06 0.059 231 EL (in) 0.623 0.626 0.487 0.317 0.293 0.317 0.896 271 33.8 118 zlP(psi) 0 1E-04 2E-04 3E-04 4E-04 5E-04 WG (Ibis) 0.09 0.077 0.072 0.06 0.055 0.054 251 &L(in) 0.623 0.624 0.55 0.349 0.403 0.436 1.17 272 33.8 98.1 4P(psi) 0 3E-04 5E-04 7E-04 7E-04 7E-04 WG (lb/s) 0.143 0.129 0.12 0.086 0.063 0.039 5 253 L (in) 0.623 0.569 0.444 0.3 0.417 0.484 1.447 281 70.88 156 JP(psi) 0 7E-04 1E-03 0.002 0.002 0.002 WG (b/s) 0.204 0.188 0.167 0.113 0.081 0.061 255 vL (in) 0.623 0.411 0.291 0.208 0.218 0.433 1.57 278 72.68 175 z1P(psi) 0 0.001 0.002 0.004 0.004 0.004 WG (Ib/s) 0.275 0.248 0.222 0.163 0.128 0.101 257 6L (in) 0.623 0.298 0.208 0.173 0.178 0.23 1.573 287 72.86 190 AP(psi) 0 0.002 0.004 0.006 0.007 0.007 WG (lb/s) 0.144 0.119 0.096 0.061 0.042 0.025 5 273 L (in) 0.623 0.783 0.643 0.525 0.591 0.642 2.253 280 77.54 144 JP(psi) 0 7E-04 0.001 0.002 0.002 0.002 WG (lb/s) 0.202 0.169 0.14 0.097 0.069 0.047 6 275 L (in) 0.623 0.623 0.51 0.403 0.352 0.622 2.244 285 79.7 163 AP(psi) 0 0.001 0.002 0.004 0.004 0.005

a. Definitions for all units are listed at the end of this table.

oMW384-non\4284-18.wpd:lb-040403 18-13

Table 18-2 (Cont'd) Test Matrix Parameters Location WLLa TG'" TL' TL-No. Units Inlet 1 2 3 4 5 (b/s) (OF) ( 0 F) (OF) WG (lb/s) 0.277 0.24 0.212 0.156 0.117 0.08 277 AL (in) 0.623 0.427 0.334 0.307 0.283 0.314 2.289 287 76.1 175 JP(psi) 0 0.002 0.004 0.006 0.007 0.008 WG (Ib/s) 0.144 0.106 0.084 0.05 0.033 0.019 293 AL (in) 0.623 0.956 0.819 0.658 0.702 0.754 3.17 279 76.82 126 JP(psi) 0 7E-04 0.002 0.002 0.002 0.003 WG (b/s) 0.199 0.155 0.127 0.08 0.055 0.034 295 AL (in) 0.623 0.869 0.693 0.551 0.652 0.726 3.148 284 78.44 144 JP(psi) 0 5E-04 0.002 0.004 0.004 0.005 WG (IbIs) 0.276 0.224 0.193 0.141 0.101 0.064 297 AL (in) 0.623 0.605 0.444 0.446 0.389 OA 19 3.165 287 79.34 161 JP(psi) 0 0.001 0.004 0.006 0.007 0.008 WG bIs) 0.144 0.132 0.127 0.09 0.067 0.043 353 AL (in) 0.873 0.653 0.528 0.309 0.242 0.451 1.5 281 76.73 160 WG Obs) 0.274 0.255 0.231 0.173 0.138 0.109 357 AL (in) 0.873 0.493 0.303 0.203 0.173 0.213 1.489 288 77 192 W (b/s) 0.141 0.125 0.114 0.077 0.049 0.03 373 AL (in) 0.873 0.828 0.665 0.453 0.363 0.585 2.233 281 75.92 139 WG (/s) 0.272 0.246 0.218 0.155 0.112 0.074 377 6L (in) 0.873 0.653 0.456 0.316 0.282 0.302 2.236 288 76.1 175 WG(lbls) 0.141 0.118 0.102 0.06 0.042 0.024 393 (in) 0.873 0.931 0.776 0.562 0.606 0.711 3.143 280 78.62 127 o:l4384-non\4284-18.wpd:1b440403 18-14

II Table 18-2 (Cont'd) Test Matrix Parameters Location WLM TG' 1 TL' TL No. Units Inlet 1 2 3 4 5 (bIs) (OF) (OF) (0F) WG bIs) 0.277 0.233 0.201 0.144 0.104 0.067 397 6L (in) 0.873 0.688 0.638 0.441 0.367 0.393 3.095 288 77.36 161 W (lb/s) 0.146 0.13 0.117 0.071 0.05 0.031 153 JL (in) 0.375 0.568 0.524 0.414 0.541 0.573 1.5 221 73.04 165 WG (Ib/s) 0.285 0.254 0.227 0.169 0.135 0.124 5 157 c L(in) 0.375 0.306 0.279 0.196 0.241 0.484 1.463 241 75.74 194 WG (IbIs) 0.147 0.128 0.105 0.063 0.043 0.041 173 aL (in) 0.375 0.779 0.71 0.546 0.663 0.681 2.311 220 73.4 144 W. (IbIs) 0.285 0.262 0.217 0.159 0.115 0.086 177 L (in) 0.375 0.503 0.438 0.335 0.36 0.381 2.315 241 80.06 177 W. (IbIs) 0.142 0.131 0.123 0.099 0.08 0.063 453 6L (in) 0.623 0.6 0.544 0.43 0.535 0.567 1.504 280 122.2 182 WG (bls) 0.207 0.193 0.176 0.138 0.119 0.108 455 6L (in) 0.623 0.445 0.361 0.299 0.305 0.507 1.5 284 119.5 190 WG bIs) 0.282 0.261 0.238 0.199 0.179 0.165 457 L (in) 0.623 0.407 0.293 0.257 0.252 0.263 1.496 287 118.4 197 WG (Ib/s) 0.344 0.315 0.294 0.254 0.236 0.223 459 6 (in) 0.623 0.329 0.257 0.227 0.214 0.249 1.562 288 125.8 201 WG ObIs) 0.141 0.125 0.112 0.084 0.064 0.045 473 3 (in) 0.623 0.766 0.663 0.526 0.61 0.675 2.344 280 123.8 172 W. (ib/s) 0.199 0.176 0.156 0.119 0.094 0.079 475 6 (in) 0.623 0.635 0.53 0.444 0.367 0.632 2.286 284 119.5 180 W0 (Ibls) 0.285 0.256 0.233 0.187 0.158 0.132 477 JL (in) 0.623 0.491 0.367 0.336 0.298 0.333 2.337 287 117.9 189 o:\4384-non\4284-18.wpd:lb-040403 18-15

I Table 18-2 (Cont'd) Test Matrix Parameters Location WL' TG" TL TL No. Units Inlet 1 2 3 4 5 (Ib/s) (OF) (OF) (OF) WG bIs) 0.143 0.118 0.102 0.072 0.056 0.037 6 493 L (in) 0.623 0.906 0.825 0.665 0.728 0.77 3.002 278 119.7 164 WG ObIs) 0.2 0.17 0.149 0.109 0.083 0.064 495 AL (in) 0.623 0.812 0.735 0.546 0.451 0.721 3.007 285 119.8 172 WG (Ib/s) 0.282 0.252 0.225 0.178 0.142 0.11 497 5 L (in) 0.623 0.622 0.458 0.426 0.392 0.426 3.156 287 119.3 181 WG = steam mnass flowrate

              = water layer thickness LP      = differential pressure WL      = inlet liquid mass flowrate TG      = inlet vapor temperature
              = inlet liquid temperature I,

TL TL" = outlet liquid temperature o\4384-non\4284-18.wpd.lb-040403 18-16

a;c Figure 18-1. WCOBRAITRAC-SB Representation of Interfacial Heat Transfer o:4384-non\4284-18.wpd:1b-40403 18-17

                                       =          - l~~~~~~~~~~~'

tIR M PRESSURE

                           -O             ~GAGE GATE   er           GLOZ EY T            I    9~~~ENTURI                  p
    -- a              F-PASRAIO                           I                               STEAM STEAM           I WATER PLENUM WATER SUPPLY DRAINAGE WATER 11         It coaOLIN; WATER Figure 18-2. Schematic Diagram of the Experimental System (Lim, et al., 1981) o:4384-non4284-1 8.wpd: b-040403                          18-18

aX4 Figure 18-3. WCOBRAJTRAC Noding o:\4384-non\4284-18.wpd:lb-040403 18-19

2.5. WI (kg/sec) E 20.288 2 - ------ --------- ---------- -------- - - 0.407

                                                                                                -~0.531 U                                       \                     ~       X-                  -    0.657 1.5                                                                                   --     0.713
                                                                                                -o-0.714 Y15'- 0.7651 o                      \Xg<=-10231 a-                                                                     ~~~~~~~~~1.019
                                                                                                --- 1.0391
                                                                                           -o-11. 043 0.5    -------------                       ...                                                1.4391
                                                                                           .- x-- 1.429!
                                                                                           -3. -. 1.437 0              0.157        0.306         0.586         0.868            1.233 Axial position (m)

Figure 184. Measured Water Thickness Versus Axial Position for Various Liquid (WI) Flowrates and Inlet Water Layer Thickness of 1.583 cm o\4384-non\4284- 18.wpd: 1b440403 18-20

I Liquid level f (cm) ' Stea flow rate Water flow (kgfsec) rate (kg/sec) Figure 18-5. Measured Water Thickness at 0.157 m From the Channel Inlet Versus Liquid and Steam Flowrates o\384-non\4284-18.wpd:1bO40403 18-21

0.6 la0.4 .... =iJ> . -- -- -- - A 0 0.2_ O SO 100 150 200 250 300 350 Time (s) Figure 18-6. Calculated Liquid Level (Run 275) A38A-non\4

   ._ # __ .. w..

2S41 R18.wDd:lbO40-3 w_ _ ^-. r- s ' 18-22

0.6 U 0.5

                         \               ~x 3

s~~~ I 0.4 0.3 0 10 20 30 40 50 60 Axial position (in) XXX Experiment

                    -       Calculations Figure 18-7. Calculated and Measured Liquid Levels Versus Axial Position (Run 275) o:\4384-non\4284-18.wpd:lb-040403                 18-23

I 14.72 14.715 - 14.71 o 14.705 14.69 14.685 0 50

                                    ' 100 1SO

_ _ _ _ _ I_ 200 250 300 350 Time (s) Figure 18-8. Calculated Steam Pressure (Run 275) o\4384-non\4284-1 8.wpd:lb-40403 18-24

0.008 0.006 0. D1 a I-c) 0.004 0.002 / i~~~~

                                / ____________  _____________      __________

A

         'J0                   10              20                30           40 50 Axial position (in)

XXX Experiment Calculations Figure 18-9. Calculated and Measured Steam Pressure Versus Axial Position (Run 275) oM4384non\4284418.wpd:1b-040403 18-25

0.025 0

S S B

              . 0.015 C-,

50 100 150 200 250 300 350 Time (s) Figure 18-10. Calculated Steam Flowrate (Run 275) o:\4384-non\4284-18.wpd:1bO4O4 -03 18-26

0.25 0.2

         .0 E                     x i   0.1 0.05 n

0 10 20 30 40 so 60 Axial position (in) XXX Experiment

                  -      Calculations Figure 18-11. Calculated and Measured Steam Flowrate Versus Aial Position (Run 275) o\4384-non\4284-18.wpd:lb-040403          18-27

3.10 2.5 -104 2 104 r-% 1.5.104 co U' 0 0 Ca) I.104 1, 5000 A-0 It - ---- - I 2 3 4 5 Distance from inlet (m) _____ Rough surface (per Lim et al., 1981) Smooth surface (per Lim et al., 1981) WCOBRA/TAC-SB (per Jensen and Yuen, 1982) Figure 18-12. Comparison of Condensation Heat Transfer Correlations o:\4384-non\4284-18.wpd:1b-040403 18-28

I 0.9 A: 0.8 . .. ... ......... .. .. .... .. . ... ... ... . ....... ...... .. . .... ... . ... . .. . A :U  : AXX  :: A :A 5 0.7 A A X; X U' C c u 0.6 A*A I-C

  'U  0.5 3                                           x          Ax                       X C

0'U 0.4 41 ,..X....................

• 6.18 in

 .5        ........................ .................                                       ,.

A a 12.05 in

 *C, 0.3                                                                                                                 ;.......      A 23.08 in x34.18in Ix48.14 In 0.2  .......       ..........                 ......                                                         *                     ...... :......
                   ...............                        X        -'.
                                                                    .... X.....

0.1 /~ ~ ~ ~ ~~~------- -St 0 I 0 0.2 0.4 0.6 0.8 I 1.2 Measured water layer thickness (in) Figure 18-13. Predicted Versus Measured Liquid Level at Various Axial Locations oA4384-non\4284-18.wpd:1b440403 18-29

0.35 0.3 0.25 0 2 0 Cu a it E 0 0.15 0

        -                                                                 *6.18in m12.05 in 23.08 in VI0.1                                                             x 34.18 in ..............

48.14 in 0.05 0

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 Measured steam flow rate (Ib/sec) I, Figure 18-14. Predicted Versus Measured Steam Flowrate at Various Axial Locations o:\4384-non\42&4-18.wpd:Ib440403 18-30

i. 200 180 . .. , 7.

                                                       .~                   ~            ~~          .      ~~    ,

4 as16 0 - -------- ------- - .. ... . .. .... - -- - - ----------- .......... .. .. . .. ... .. . ..... .... ... .. . C) ' / . + +. + *

                                                                                                                 *,~~~*

U-120 ...-,  ;  ;  ; 'j ......... . .' 80 80 10D 120 140 160 180 200 Measured liquid exit temperature (F) Figure 18-15. redicted Versus Measured Liquid Temperature at the Channel Eit o:X4384-non\4284-1 wpd:lb040403 18-31

0.009 0.0 08 . . .. ............... .............. ............... . ....................... ;. x 0.007 . ................. ........... ............ . ........... ... ..........- sL0.0 06 . ..... . ........ .................. ............... ..................... .............. ...... ........" 0.005-. ... ,, . .... ,;' S~~~~~~~ 0.00 . .... . . . .. ... - -- - --- - - -- ---- - - - ... .. .. . ------.... ..................... ....... ... ..... . . . . .. a0.00-~ ~ + 4 ~ 0 .0 0 . .E....... .. .. .. . . ........... . . . ....~~~~~~A:

                                                                               ;A . . ........................... ............                    . . . ....... I. .. .. . . .

0.003 . ............... 0O . . .' ,,. .

                                                                                                                                                                                 ., X, 000.00130.002                                0.003             0.004            0.005             0.006                    0.007            m0.08            ...

A 21.77 in x 32.87 in 0.001 ~ ~ ~~~~~~~~~~~~~~~~~~x47.24 in 0 0001I0002 0.003 0.004 0.005 0006 0.007 0.008 0.009 253, 25,27x7,25 7,23 9,ad27 Measured steam pressure drop (psi) (Measurements were taken only during the following runs: 211, 231, 251, 253, 255, 257, 273 275, 277, 293, 295, and 297) Figure 18-16. Predicted Versus Measured Steam Pressure Drop at Various Axial Locations o:\4384-non\4284-1 8.wpd:1b.040403 18-32

SECTION 19 ROSA TEST SIMULATIONS 19-1 Introduction The Rig-of-Safety Assessment Number 4 (ROSA-IV) program conducted a series of experiments to investigate the thermal-hydraulic behavior of a Westinghouse-designed four-loop PWR during small break LOCAs and operational transients using the Large Scale Test Facility (LSTF). The LSTF is a 1/48 volume scale representation of a Westinghouse four-loop 3423 MW PWR. Figure 19-1 is a schematic diagram of the facility. The LSTF consists of two equal volume loops, A and B, with the pressurizer attached to loop A. The elevations of the major components are full-scale to preserve natural circulation phenomena important to core cooling. The hot and cold legs, with a diameter of 8.15 inches, are sized to conserve volume scaling and the ratio of length to the square root of the pipe diameter (LIVD) of the reference PWR. Table 19-1 compares the major design characteristics of the LSTF and the PWR. The core contains 16 square 7x7 and 8 semicrescent heater rod assemblies. The heater rods are 0.374 inches in diameter and 12 feet in length. The maximum power in the facility is 10 MW, which is equivalent to 14 percent of the scaled steady-state core power of the reference PWR. The secondary coolant system consists of the steam generator, main and auxiliary feedwater pumps, and condensing system. The height of the LSTF steam generator is the same as in the reference PWR. The downcomer in the steam generators consists of four pipes located outside the steam generator vessel. The pipes are sized to give a representative volume and width. Each steam generator contains 141 U-tubes with 0.772 inches ID and 1.0 inches OD. Primary and secondary steam separators are included in each steam generator vessel. The LSTF ECCS consists of a high pressure charging system, a high pressure injection system, a low pressure injection system, an accumulator system, and a residual heat removal system. A detailed description of the facility is contained in the Japan Atomic Energy Research Institute (JAERI) document (JAERI-M 84-237, 1985). o\4384-non\4384-19.wpd:lb-04043 19-1

I This section describes the modelling and simulation of five ROSA tests using WCOBRA/TRAC-SB. The tests simulated were the 2.5-, 5-, and 10-percent cold leg breaks, each with the break at the middle of the pipe. This group of tests provides a break size sensitivity. Two other 2.5-percent cold leg break tests were also simulated: one with the break at the top of the pipe and the other with the break at the bottom. These provide a break orientation sensitivity. Comparisons between predicted and measured results are used to validate the WCOBRAITRAC-SB code version and to identify possible compensating errors. 19-2 WCOBRA/TRAC Model of the LSTF The WCOBRA/TRAC-SB nodalization of the LSTF uses a similar amount of detail in the vessel, steam generators, and loops as used in the PWR model. Figure 19-2 shows the WCOBRA/TRAC-SB noding of the LSTF pressure vessel. Because the LSTF is a two-loop facility, the vessel is modelled [ la; The upper plenum modelling of the LSTF facility includes [

         ]ac Modelling of the LSTF hot and cold legs is shown in Figure 19-3. Each hot leg is modelled

[ Iag o:\4384-non\4384-19.wpd:1b04043 19-2

I The LSTF steam generators are shown in Figures 19-4 and 19-5. Primary flow enters the steam generator [

                                                                                  ]a,c The steam generator secondary side includes sufficient detail to model recirculation in the downcomer and separation in the vapor dome region. [
                                                    ]ac During steady-state simulation, and prior to reactor trip, steam leaving the generators passes through a TEE component and VALVE component to a constant pressure BREAK. At reactor trip, the main steam isolation valve (MSIV) is closed and flow goes through a VALVE component representing the main steam safety valve (MSSV) to a second BREAK component that provides a constant pressure boundary condition at the MSSV setpoint pressure.

Figure 19-6 shows the loop seal nodalization. Flow from the steam generator outlet passes through [

                                              ]a.c The safety injection system is shown in Figure 19-7. Charging and high pressure safety injection flows to each loop are modelled [
                                                                                           ]a.c o:t4384-non\4384-19.wpd:lb-04043                 19-3

accumulator setpoint of 647.5 psia. The combined safety injection from the pumps and accumulators entered the cold legs through TEE components 10 and 19 to loops A and B respectively. 19-3 Steady-State Simulation Verification that the WCOBRAfTRAC model of the LSTF adequately represented the facility was accomplished through a full-power, 100-second steady-state simulation. At the end of this 100-second simulation, predicted and measured flow parameters were compared to ensure reasonably good agreement by the model. Table 19-2 summarizes initial conditions used in the steady-state comparison. Agreement was obtained between the simulated initial system pressure and initial test pressure. The comparison shows that the total core flow is underpredicted due to too high of an estimate of the core and loop flow resistance. As a result, Th,, is several degrees high. A good energy balance was achieved between the primary and secondary sides. The steady-state parameters are considered close enough for reliable transient calculations based on the WCOBRATRAC model of the facility. 194 Simulation of SB-CL-05, 5-Percent Cold Leg Break Experiments as part of ROSA-IV were conducted for several different break areas. Test SB-CL-05 simulated a 5-percent cold leg break, which corresponds to approximately a 6-inch break in a PWR. The break was located in loop B and had a horizontal orientation. Safety injection flowrates corresponding to a single failure in the safety injection system were assumed. Experimental results are discussed by Kawaji (Kawaji, et al., 1986) and Tasaka (Tasaka, et al., 1988). Operational setpoints for this test are listed in Table 19-3. The core power was scrammed once the primary pressure decreased below 1862 psia At scram, the primary coolant pumps began to coast down, the main feedwater was stopped, and the steam generator secondaries were isolated by closure of the main steam isolation valves (MSIV). In this test, the primary system rapidly depressurized to a pressure slightly higher than the secondary pressure, approximately 1150 psia, until the loop seal cleared at 140 seconds. o\4384-non\4384-19.wpd:1b-04443 19-4

After loop seal clearance, the break quality changed from a low quality mixture to primarily vapor and the primary system continued to depressurize. Test SB-CL-05 had a core depression during loop seal clearance that was considerably below the elevation of the bottom of the loop seal piping. Osakabe (Osakabe et al., 1987) attributed this to a large liquid holdup in the uphill steam generator tubes. During this core level depression, the cladding temperature increased by approximately 180°F reaching a maximum cladding temperature of approximately 830°F. After loop seal clearance, the core level recovered quickly. Accumulator injection began at 417 seconds and prevented a second core uncovery. The WCOBRA/TRAC-SB simulation of the LSTF 5-percent cold leg break was initiated by attaching a PIPE component to the middle level of the loop B cold leg. The break unit in this test was aligned horizontally. Figure 19-8 shows the nodalization for modelling the break unit in the LSTF. Table 19-4 compares the sequence of events in the simulation and the experiment. In the current modelling, the pressurizer pressure decreased more slowly than the experiment. This caused a delay in the generation of reactor trip and safety injection signal generation. As the primary system continued to drain, a manometeric balance was set up between the core and downcomer, and the uphill and downhill sides of the loop seal piping. A deep depression in the core collapsed liquid level occurred as steam slipped through the loop seal piping toward the cold leg. The core level became depressed nearly to the bottom of the core, while liquid remained in the uphill side of the loop seal. At this time, the heater rods heated up rapidly. While most liquid had drained from the steam generator tubes, a level remained in the plenum and bottom of the uphill side. After steam slipped through the loop seals, the core level recovered and most of the water was pushed out of both loop seals. In this simulation, no core uncovery occurred following the loop seal clearance period. By 420 seconds, the primary system pressure decreased below 647.5 psia and accumulator injection began. At this time, the core remained covered with a low void fraction mixture and no heatup was predicted. o:W384-nonW4384-19.wpd:1b4043 19-5

I Figures 19-9 through 19-16 compare predicted and measured results for the 5-percent cold leg break test. Figure 19-9 compares predicted and measured primary system pressure. For the first 75 seconds of the simulation, the pressurizer pressure remains higher than the measurement while the pressure in the upper head is predicted lower than the data. Overprediction of pressurizer pressure for this period causes a delay in reactor trip and safety injection signals. The system depressurization slows down briefly near the hot leg saturation pressure; it then continues until both the primary and secondary pressures equilibrate at approximately 1200 psia in both the prediction and the test data. From about 100 seconds until loop seal clearance begins at 197 seconds, the predicted pressure increases approximately 50 psi. This increase is not observed in the data. The cause of this reduction in the primary to secondary heat transfer may be due to excessive liquid holdup/fluid stagnation in the steam generator tubes. As the liquid film hangs, the convective heat transfer from the primary side water to the tube is diminished. The primary pressure increases by an amount necessary to continue heat transfer with a higher primary to secondary temperature difference. After loop seal venting, the predicted pressure remains higher than the measurement. The accumulator setpoint pressure of 646 psia was reached 417 seconds into the test. The WCOBRA/TRAC-SB simulation depressurized to this value at 420 seconds. Break flow is compared in Figure 19-10. Early in the transient, flow out of the break is subcooled, that is, single-phase liquid. During this period, with no discharge coefficient applied, WCOBRAtIRAC-SB slightly overpredicts the break flow. Between 50 and 150 seconds, the break flow is underpredicted in the simulation. During this period, break flow becomes two-phase in the calculation, some 100 seconds earlier than the test data. This mismatch is attributed partly to early overprediction of the break flow, which depleted the inventory faster, and to a higher core outlet initial temperature. Once the flow quality turns two-phase, the venturi meter used in the experiment becomes significantly unreliable and has great uncertainty at low mass flowrates. For that portion of the comparison, break mass flowrate derived from the catch-tank level is used. The derived flowrate is not responsive to the rapid changes in flow at the beginning of the transient. After 150 seconds, the break flow is slightly overpredicted compared to the data. Loop seal venting occurs at approximately 140 seconds in the test. WCOBRATRAC-SB predicts loop venting initially at 195 seconds in the intact loop and continually through both loop seals after 200 seconds, as seen in Figures 19-11 and 19-12. The data also show venting through both loop seals. This delay relative to test data is likely due to the mismatch in break flow prediction. o:\4384-non4384-19.wpd:lb-04043 19-6

As the loop seals vent, the collapsed liquid level in the core is depressed. Figure 19-13 compares the core liquid levels. During the initiation of loop seal venting, the predicted level is depressed nearly to the bottom of the core. The data also show a deep core level depression with the level decreasing well below the bottom of the loop seal piping to within 2 feet of the bottom of the core. After loop seal clearance, the core level recovers to approximately the level observed in the data. Core heatup occurs during the loop seal clearance period while the core remains uncovered. Figure 19-14 compares the PCT predicted by WCOBRAITRAC to the maximum cladding heatup observed in the data. The core uncovery period during the loop seal clearance event is overpredicted in the simulation. This results in a higher PCT compared to measured data. The core uncovery during the loop seal clearance period depends upon the manometric balance between the core and downcomer, and the sum of pressure drops through the loop and uphill side of the loop seal piping. An important static head exists on the uphill side of the steam generator tubes, where water condensed in the tubes collects because of CCFL and flooding in the steam generator up-hill tubes. Figures 19-15 and 19-16 show collapsed liquid levels in the uphill steam generator tubes. The apparent high resistance in the bypass flow modelling between the downcomer and the upper head and excessive liquid holdup in the steam generators result in extended core level depression. Following loop seal clearance, the steam generator tubes drain briefly and then retain a small collapsed level. Table 19-4 summarizes the predicted and recorded results for the 5-percent cold leg test. 19-5 Simulation of SB-CL-09, 10-Percent Cold Leg Side Break One of the integral shakedown tests performed in the LSTF was a 10-percent cold leg break, which was the maximum break size for the facility design. This is a relatively large break size, corresponding to approximately a 9-inch break in a PWR, which could be considered more of an intermediate break as opposed to a small break LOCA. This break size is relevant, however, because limiting small break sizes may shift to larger equivalent diameters in other plants. The break was located in loop B, which was the loop without the pressurizer, and was oriented horizontally from the middle of the cold leg. The operational setpoints for this 10-percent break are listed in Table 19-5. o.\4384-non\4384.19.wpd:lb-04043 19-7

I The 10-percent cold leg break test simulated using WCOBRAITRAC-SB is the ROSA-IV experiment designated as run SB-CL-09. This test uses a conservative core power transient curve, which is significantly higher than the realistic power curve for the first 300 seconds of the transient. While the same experiment with realistic power curve, SB-CL-14, showed no significant rod heatup, this case exhibits large rod heatup during the loop seal clearing (Koizumi and Tasaka, 1988). In addition, lack of high pressure charging injection and high pressure safety injection resulted in the simulation predicting a core boiloff with rod heatup after the loop clearance event. Predicted results for the 10-percent break are shown in Figures 19-17 to 19-24. Figure 19-17 depicts the primary system pressure. The prediction was found to depressurize rapidly during the first few seconds of the transient. By 30 seconds, the primary pressure equilibrates at the secondary side pressure until loop seal clearance. Accumulator injection begins at 198 seconds in the prediction. The break flow is shown in Figure 19-18. Figure 19-19 shows the predicted core collapsed liquid level. The initial rapid depressurization of the primary side in the prediction causes the core to flash quickly. This is seen in the initial deep drop in the collapsed liquid level 12 seconds following the break. The broken side loop seal begins to clear at 80 seconds, and the intact side loop seal clears at 71 seconds as seen in Figures 19-20 and 19-21. Liquid holdup on the uphill side of the steam generator tubes is shown in Figures 19-22 and 19-23. Both broken and intact side steam generators drain quickly because of the relatively large break size. The PCT prediction is shown in Figure 19-24; the calculated value does not reach the experimental value of 1135°F. Table 19-6 summarizes the predicted results for the 10-percent cold leg test. 19-6 Simulation of 2.5-Percent Cold Leg Breaks A set of three experiments - SB-CL-01, 02, and 03 - was conducted in the LSTF to investigate the effect of break orientation. All three tests simulated a 2.5-percent break in the cold leg, which approximates a 3-inch break in a PWR. In these experiments, the break was oriented at the side, bottom, and top of the loop B cold leg. Experimental results are summarized in the data report by Koizumi (Koizumi, et al., 1987). The test results showed that break orientation had only a small effect on system parameters such as pressure and core collapsed liquid level. o:4384-non\4384-19.wpd:lb-04043 19-8

Figure 19-25 provides a description of the break geometry for these tests. These tests provide a useful means of evaluating the break flow model in WCOBRAtTRAC-SB for the effects of vapor pull-through and liquid entrainment near the break orifice. Operational setpoints for the 2.5-percent cold leg break tests were the same as those shown in Table 19-3, with two exceptions. The charging and high pressure safety injection was delayed in these tests until 1200 seconds to force boiloff to occur, instead of the normal 12- and 17-second delays for these system flows. In addition, core power trip control turned the power off once the heater rod temperatures reached 1196°F in the experiment. Figure 19-8 shows the break modelling used in the 2.5-percent cold leg break simulations. Results for the 2.5 percent cold leg side break are compared to data in Figures 19-26 through 19-30. Figure 19-26 compares the predicted and measured primary system pressure. Over the first 200 seconds of the transient, WCOBRA/TRAC-SB tends to slightly underpredict the pressure. By 200 seconds, however, both the predicted and measured pressures have equilibrated with secondary side pressure at approximately 1200 psia. Between 200 and 400 seconds, WCOBRA/TRAC-SB overpredicts the system pressure. Furthermore, WCOBRA'RAC-SB predicted the pressure to slightly increase, which is not observed in the data. After loop seal clearance, the predicted and measured pressures are in good agreement. Figure 19-27 shows a comparison of predicted and measured break flows. Early in the transient, until 350 seconds, the break flow is underpredicted, after which it is overpredicted for a time. The code underprediction of the SB-CL-01 breakflow leads to an overprediction of the time to loop seal clearance. A comparison of the vessel collapsed liquid level is shown in Figure 19-28. The agreement is good when comparing the times of loop seal clearance (462 seconds in the simulation versus 380 in the test). Neither the data nor the prediction shows enough of a core level depression to cause a significant core heatup at this time, as seen in Figure 19-29; one does occur later as the boiloff period begins. The reason for the lack of a severe loop seal clearance level depression in the core can be explained by Figure 19-30. Because of the small break size, the transient proceeds at a relatively slow rate. Liquid held up in larger breaks in the uphill steam generator tubes drains, and the tubes are nearly voided by the time loop seal venting occurs. The additional static head due to liquid in the steam generator tubes is not present, and less of a core level depression is required to maintain a manometric balance with the loop seal. The characteristic deep core uncovery during loop seal clearance in other ROSA-IV tests does not occur. o:\4384-non\4384-19.wpd:lb-04043 19-9

The boiloff period is turned around with the accumulator injection. In the experiments, the core power was tripped once the heater cladding temperatures reached almost 1200°F. The core remains covered with the help of safety injection, which was manually delayed until 1200 seconds. In the simulations, the power is tripped once the PCI reached 1196°F, stopping a further increase in the cladding temperatures. The power trip also reverses the core boiloff, and the core collapsed liquid level starts to recover. The transients for the top and bottom breaks were terminated before a full level recovery was observed for these cases. The timing of key events for experiments with side, bottom, and top orientation is given by Tables 19-7 through 19-9, respectively. Test results for the three 2.5-percent breaks showed relatively little difference in break flowrate (Koizumi, et al., 1988). The experimental break flowrates are shown in Figure 19-31A. A comparison of the predicted break flowrates is shown in Figure 19-3 1B. Similar to the experimental data, the break orientation had only a small effect on the predicted break flow. The test data for all three orientation breaks show no difference in break flowrate until 100 seconds. However, the test data show that the break flows diverge from each other when the break flow quality turns two-phase. First, the top break and side break discharge becomes two-phase, and the discharge flowrate reduces abruptly. The last to become two-phase is the bottom break. Change in the break flow from single-phase subcooled discharge to high-void two-phase discharge is predicted in a consistent manner for top and bottom break orientations. The bottom break takes longer to become two-phase because the level in the cold leg needs to drop to break location. However, because most of the liquid is exhausted, the break flowrate then reduces below the top orientation flow. This detail is predicted in the simulations. The side break case did not produce a consistent prediction with the data; it exhibits an increase in break flow after the initial abrupt drop. While the test data showed that break orientation had only a minor effect on the break flowrate, the orientation did affect the timing of the loop seal clearance and core uncovery. In the test data, of Figure 19-32A, the side break orientation had the earliest loop seal clearance time, then the top break and the bottom break vented at almost the same time. Also, the bottom break produces an earlier and deeper uncovery in the boiloff period. In the simulations, as in the tests, the side break was the first one to vent; the bottom and top breaks vent later as shown in Figure 19-32B. The sequence in which loop seal clearing occurs as a function of break orientation is well predicted by WCOBRA/IRAC-SB, although the timing is delayed from the data in every case. o:.4384-non\4384-19.wpd:lb-04043 19-10

Figures 19-32A and 19-32B provide a comparison of measured and predicted core levels for the top, side, and bottom 2.5-percent breaks. In the test data, the start of the boiloff period for the bottom break occurs first, and the side break boiloff began prior to the top break. The timing of the onset of the core boiloff period in the predictions shows the top and side break begin at almost the same time, later than for the bottom break; as in the test data, the bottom break exhibits the most rapid core uncovery. Figures 19-33A and 19-33B indicate that the WCOBRAfIRAC-SB simulations show the general two-phase level characteristics of the experiments. Top and side breaks maintain a higher mixture level in the broken cold leg compared to the bottom-oriented break. The core heatup rate during the boiloff period is adequately predicted. Figure 19-34A shows the cladding temperature for an 8.67-foot rod elevation. Figure 19-34B depicts the code-predicted temperatures at the same elevation for 2.5-percent break cases. Simulations were ended at 975 seconds for the side-oriented case and 1114 seconds for the bottom-oriented case. 19-7 Summary of Results This section describes the predicted results for five different ROSA-IV/LSTF small-break tests. The simulations provide an adequate representation of the test data. The loop seal clearance behaviors are predicted with acceptable accuracy, and core uncovery predictions that closely agree in magnitude with the test data produce cladding heatup predictions comparable to those in the data. Liquid holdup in the steam generator tubes appears to be predicted with good accuracy in the SB-CL-01 simulation, although in the 5-percent cold leg break simulation, the code prediction of liquid holdup in the steam generators appears to lead to a repressurization of the primary side. WCOBRA.TIRAC-SB results for the 5-percent cold leg break case show the discharge of single-phase subcooled liquid at high pressures is overpredicted slightly in the initial 50 seconds when compared against the available data, then underpredicted. The break flowrates in the 2.5-percent series of test simulations trend well with the data with the exception of an early surge in the side orientation case (SB-CL-01) result.- In both the 5-percent and 2.5 percent break size predictions, the underprediction of break flow causes a delay in the predicted time of loop seal clearance. The timing of the initiation and the turnaround of the boiloff period in the 2.5-percent bottom and side break cases are in general well predicted, although in the top break case the boiloff is o\4384-non\4384-19.wpd:lb-04043 19-1 1

predicted to begin earlier than in the SB-CL-03 data. Because the power was tripped at a preset heater temperature, the PCTs are exactly predicted. 19-8 References JAERI-M 84-237, 1985, "ROSA-IV Large Scale Test Facility (LSTF) System Description." Kawaji, M., et al., 1986, "ROSA-IV/LSTF 5% Cold Leg Break LOCA Experiment Data Report, Run SB-CL-05," JAERI-memo 61-056. Koizumi, Y., et al., 1987, "ROSA-IV/ILSTF 2.5% Cold Leg Break LOCA Experiment Data Report for Runs SB-CL-01, 02 and 03," JAERI-memo 62-399. Koizumi, Y., et al., 1988, "Investigation of Break Orientation Effect During Cold Leg Small-Break LOCA at ROSA-IV LSTF," NucI. Sci. and Tech., 25. Koizumi, Y. and Tasaka, K., 1988, "Quick Look Report for ROSA-IV/LSTF 10% Cold Leg Break LOCA Test, SB-CL-14," JAERI-memo 62-262. Osakabe, M., et al., 1987, "Core Liquid Level Depression due to Manometric Effect during PWR Small Break LOCA," J. of Nucl. Sci. and Tech., 24. Tasaka, K., et al., 1988, "The Results of 5% Small Break LOCA Tests and Natural Recirculation Tests at the ROSA-IV LSTF," Nucl. Eng. Des., 108. o:\4384-non\4384-19.wpd:lb-04043 19-12

Table 19-1 Major Design Characteristics of LSTF and PWR Characteristic LSTF PWR PWR/LSTF Pressure (psia) 2250 2250 1 Temperature (°F) 617 617 1 Number of fuel rods 1064 50,952 48 Core height (ft) 12 12 1 Fluid volume (ft 3 ) 255.2 12,254.2 48 Core power (MW) 10 3423(t) 342 Power density (kW/ft 3 ) 39.64 280.34 7.1 Core inlet flow (lbmls) 97.6 33,400 342 Downcomer gap (in.) 2.09 10.24 4.9 Hot leg Diameter (D) (ft) 0.679 2.418 3.56 Length (L) (ft) 12.1 22.93 1.89 L1vD (ftl/2) 14.76 14.76 1.0 XCD 2 L (m 3 ) 0.124 2.98 24.0 4 Number of loops 2 4 2 Number of tubes in steam generator 141 3382 24.0 Length of steam generator tube (average) (ft) 66.3 66.3 1.0 o:\4384-non\43S419.wpd:1bO4043 19-13

I Table 19-2 Steady-State Parameter Checklist Parameter Target Predicted Pressurizer pressure (psia) 2262 2257.9 Hot leg fluid temperature (F) 619 626.2 Cold leg fluid temperature (°F) 557.6 553.7 Core power (MW) 10 10.0 Core inlet flowrate (lbm/s) 108.7 94.1 Pressurizer water level (ft) 8.53 8.64 Pump speed (rpm) 800 800.0 Pressure vessel top-bottom AP (psi) 10.76 10.99 Hot leg A AP (psi) 0.5 1.4 Steam generator loop A inlet to tube top AP (psi) 10.88 11.57 Cold leg A AP (psi) 0.1 0.04 Upper plenum - downcomer AP (psi) 0.6 3.85 Steam generator secondary pressure (psia) 1055 1055.0 Steam generator secondary level (ft) 33.78 31.23 Steam generator feedwater temperature (F) 431.6 431.6 Steam generator feedwater flowrate (lbmls) 5.95 6.09 Steam generator steam flowrate (lbm/s) 5.95 5.91/5.88 o:\4384nonM4384-19.wpd:Ib-04043 19-14

Table 19-3 Operational Setpoints for Run SB-CL-05 Event Setpoint Reactor scram signal (psia) 1862 Initiation of RCP coastdown With reactor scram Safety injection signal (psia) 1761.5 High pressure charging 12 s after safety injection signal Safety injection 17 s after safety injection signal Accumulator injection (psia) 647.5 Low pressure injection (psia) 185.2 Main feedwater termination With reactor scram Turbine throttle valve closure With reactor scram Auxiliary feedwater initiation 28 s after reactor scram o:M4384-non\4384-19.wpd:lb-04043 .19-15

I Table 19-4 Transient Results Summary for 5-Percent Cold Leg Side Break Event Data Prediction Break (s) 0 0 Reactor trip (s) 12 24.0 MSIV closure (s) 15 25.0 Safety injection signal (s) 17 32.1 Steam generator feedwater stop (s) 18 23.0 Charging injection ON (s) 31 44.1 High pressure safety injection ON (s) 34 49.1 Auxiliary feedwater ON (s) 40 51.0 Core uncovery (s) 120 to 155 135 to 210 PCT (F) 830 850 Minimum vessel collapsed liquid level (ft) 1.4 1.9 Loop seal clearing (s) 140 197 Accumulator injection ON (s) 417 420 Accumulator injection OFF (s) 1447 > 500(a)

a. Transient calculation terminated at 500 seconds.

o\4384-non\4384-19.wpd:lb-04043 19-16

Table 19-5 Operational Setpoints for Run SB-CL-09 Event Setpoint Reactor scram signal (psia) 1862 Initiation of RCP coastdown With reactor scram Safety injection signal (psia) 1761.5 High pressure charging Not actuated Safety injection Not actuated Accumulator injection (psia) 647.5 Low pressure injection (psia) 185.2 Main feedwater termination With reactor scram Turbine throttle valve closure With reactor scram Auxiliary feedwater initiation Not actuated o:\4384-non\4384-19.wpd:1b-04043 19-17

I Table 19-6 Chronology of Events for Run SB-CL-09, 10-Percent Cold Leg Side Break Events Predicted Time (s) Break 0.0 Reactor trip 10 Main steam line valve close 11 Loop seal clearing 71 Primary to secondary pressure reversal 104 Accumulator injection on (ACC-cold) 198 o:\4384-non\4384-19.wpd:lb-04043 19-18

Table 19-7 Chronology of Events for Run SB-CL-01, 2.5-Percent Cold Leg Side Break Measured Predicted Events Time (s) Time (s) Break 0 0.0 Reactor trip 15 13 Safety injection signal 19 21 Main steam line valve close 20 14 RCPs stop 272 -280 Loop seal clearing 380 462 Primary to secondary pressure reversal (loop B) 380 464 Primary to secondary pressure reversal (loop A) 460 464 Core dryout. 600 to 980 608 to 975a) Accumulator injection on (ACC-cold) 835 882 Core power trip 872 919 High pressure charging injection on 1199 >975(a) High pressure safety injection on 1200 Low pressure injection on 1446 Accumulator injection off 1460 Experiment terminated (break unit close) 2429

a. Transient calculation is terrninated at 975 seconds.

o:.4384-nonW4384-19.wpd:1b-04D43 19-19

I Table 19-8 Chronology of Events for Run SB-CL-02, 2.5-Percent Cold Leg Bottom Break Measured Predicted Events Time (s) Time (s) Break 0 0 Reactor trip 16 13 Safety injection signal 21 20 Main steam line valve close 21 14 RCPs stop 273 -280 Loop seal clearing 446 468 Primaiy to secondary pressure reversal (loop B) 450 471 Primary to secondary pressure reversal (loop A) 550 471 Core dryout. 600 to 970 587 to 1093 Core power trip 846 877 Accumulator injection on (ACC-cold) 853 856 High pressure charging injection on 1201 >1114.3(a) High pressure safety injection on 1201 Low pressure injection on 1464 Accumulator injection off 1471 Experiment terminated (break unit close) 2409 a Transient calculation is terminated at 1114.3 seconds. o.\4384-non'4384-19.wpd:lb-04043 19-20

Table 19-9 Chronology of Events for Run SB-CL-03, 2.5-Percent Cold Leg Top Break Events Measured Predicted Time (s) Time (s) Break 0 0 Reactor trip 16 13 Main steam line valve close 20 14 Safety injection signal 21 20 RCPs stop 272 -280 Loop seal clearing 430 474 Primary to secondary pressure reversal (loop B) 430 474 Primary to secondary pressure reversal (loop A) 490 474 Core dryout 670 to 1030 618 to 1015 Accumulator injection on (ACC-cold) 914 889 Core power trip 957 1006 High pressure charging injection on 1201 1200 High pressure safety injection on 1201 1200 Accumulator injection off 1479 1396 Low pressure injection on 1507 N/A b Low pressure injection system off 1961 N/A(b) Experiment terminated (break unit close) 2731

a. Transient calculation is terminated at 2000 seconds.

b. This was not modelled.

o:\4384-non\4394-19.wpd:Ib-04043 19-21

Table 19-10 2.5-Percent Cold Leg Break Loop Seal Venting Times Break Orientation Measured (s) Predicted (s) Top 450 474 Side 380 462 Bottom 450 468 o:\4384-non\4384-19.wpd:lb-04043 19-22

( 0 Loop - A ------

                                       .~~~~~~~~~-

Loop-B r I I I I "0 STEAM GENERATOR r l_m 0 ACCUMULATOR O. Cj W n Pa 0 FLOW CONTROL VALVE

Figure 19-2. WCOBRAITRAC-SB Model of LSTF Pressure Vessel o4384-non\4384-19a.wpd:Ib-040403 19-24

a.g Figure 19-3. WCOBRAITRAC-SB Model of LSTF Hot and Cold Legs o-A4384-non\4384-19awpd:lb-40403 19-25

Figure 19-4. WCOBRAJTRAC-SB Model of LSTF Loop A Steam Generator o\4384-non\4394-19a.wpd:lb-040403 19-26

a; Figure 19-5. WCOBRA/TRAC-SB Model of LSTF Loop B Steam Generator o\4384-non\4384-19a.vwpd:b-0403 19-27

Figure 19-6. WCOBRAJTRAC-SB Model of LSTF Loop Seals o:\4384-non\4384-19a.wpd:lb-040403 19-28

a; Figure 19-7. WCOBRAITRAC-SB Model of LSTF Safety Injection o-\4384-non\4384- 19a.wpd:lb-040403 19-29

Figure 19-8. Nodalization of LSTF Break Unit oA4384non\4384-19awpd:lb-040 403 19-30

Measured

       -_-______   Predicted 2500 2000 U>

c,) 1500 v) v) CL. L-, r)- 1000 U) a-) 500 0 Figure 19-9. Comparison of Predicted and Measured Primary System Pressure, ROSA 5-Percent Cold Leg Side Break o:4384-non\4384-19awpd:1b4403 19-31

I Measured

      - - - - -_ -Predicted Measured (catch-tank) 100 -

80 - .................... C-E

   -o 60 a.)

a r 40 M C/) U 20 0 Time (s) Figure 19-10. Comparison of Predicted and Measured Break Flowrates, ROSA 5-Percent Cold Leg Side Break o\4384-non\4384-19a.wpd:lb-040403 19-32

6 5- . . . . . . . . . 4 - .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . - .l *. a, 3- . . .. .. .....

                        . . .   ~.. . . .. .. ...          I                            afe0:N #

C,, 2-

                                         ~.. . . . ..        .         .

1- .. . . . . . . . . . . . . 1. ...........

                                                     .1 0-                                                                 . . . . .  . . . . . . . . . . . . .  .
                      .   :   ;       i   . I    I      I     I I   I       I       1I   I           :   I   !             i       I
        -1 0                  100                 200                 300                 400                 500                   600 Time (s)

Figure 19-11. Predicted Intact Loop Seal Steam Flowrate, ROSA 5-Percent Cold Leg Side Break o:\4384-non\4384-19awpd:lb-040403 19-33

I 5 4-. 4

                . . . t . \. l. lg..
                                 . .. . .. . .. .. ...          J.        .     .                  .
    -E        i                     *                    ,      !1t',             -

0. .. . . . . . . . . . . . . . . . . . . . . . .. . . . . . . .

                    -1t   ~~       ~      ~    ~    .1        -                      I 0'                              010                0             0            0                   0 Time (s) v)                            .       v                        a             v...

Figure 19-12. Predicted Broken Loop Seal Vapor Flowrate, ROSA 5-Percent Cold .III Leg Side Break oaM 4-non\438419a.wDd:1b 040403 19-34

Measured __Predicted 14-12 ............................................................... .....

                .2
            -Cl                         ~~I       . ... . .   .   .   .   .   .    .  .   .   .   .   .   .   .   .   .     .   .   .

2-8.... ....... . ......... .......... ............ .. C-) 6O- . ..... ,. 0r 0 200 400 600 80 100C Time (s) Figure 19-13. Comparison of Predicted and Measured Core Collapsed Liquid Levels, ROSA 5-Percent Cold Leg Side Break o.\4384-non\4384-19awpd:lb-040403 19-35

I Measured Predicted 900 800 I . 700 co E 600 500 400 Figure 19-14. Comparison of Predicted and Measured PCTs, ROSA 5-Percent Cold Leg Side Break o:\4384-non\438419a.wpd:lb-040403 19-36

         ----              P r ed i c t e d 40 -
               ~1
                -I

__- 30 - -__I . . . . . . . . . . . . . . . . . . . . . . . .

               -*1 a)
   -J                   I 20 -1 ....    -  II..........

I I

                -            I
                                            .\I I

CD 10 - . . . . . . I . . . . . . . . . . . .. . . . . . . . . . . . . . .

                                 \_,I, 5

I 1 .

A I I I . I a  : u 0 100 200 360 400 500 600 Time (s) Figure 19-15. Predicted Intact Loop Uphill Steam Generator Tube Collapsed Liquid Level, ROSA 5-Percent Cold Leg Side Break o\4384-non\4384-19a.wpd:lb-040403 19-37

Measured

        -_-___--  _    Predicted 50 40
    -E 30
    -J
    -o 0J C,'

c 20

    -U C

co 0D

    -5 0-10 0

Time (s) Figure 19-16. Comparison of Predicted and Measured Broken Loop Uphill Steam Generator Tube Collapsed Liquid Level, ROSA 5-Percent Cold Leg Side Break o:\4384-non\4384-19a. .wpd:b440403 19-38

Pr edi c ted 2500 - 2000 . .......... ....... ..................... .. ...................

   .: 1500 -         ...... .......

L 1000 . ..... . 0~ 500-. .... ....... ... ........ ...... .. ....... .. ... .. ... 100 .200 300 400 .500 Time (s) Figure 19-17. Predicted Primary System Pressure, ROSA 10-Percent Cold Leg Side Break o\4384-non\4384-19a.wpd:lb-040403 19-39

        -~           Predicted
       *160-140 . .................. . . . . . .                         .. ............ ..............

120-. . .... ... ...... ... .. .. .. .. ... ....... .. ..... ...... ..... ... E 1 00-. .. ........ ... ......... . . .. . .. . .. . . .. . .. . . cS80-. . .. ........:.............. cn 60 - .. . . . . .-. .- 20- . . ... .. .. .. . . .. . . . . ..- 0 -  ; ' I I I t 0 I,l 100 200 300 40 Time (s) Figure 19-18. Predicted Break Flowvrate, ROSA 10-Percent Cold Leg Side Break o:\4384-non\4384-19a.wpd:lb040403 19-40

            -~        Predicted 12-cin Y.-                 . . ....          . .;. . . . . . .           .a.. ....

2 .. 100 200 300 400 Time (s) Figure 19-19. Predicted Core Collapsed Liquid Level, ROSA 10-Percent Cold Leg Side Break o:\4384-non\4384-19a.wpd:lb040403 19-41

Predicted 8-6 . .. . . .. . . .. . . .. -. .. .. .. .. .. . .............. E 4 . ....... .J.. .... .. ............ .V. . 2......... O-. . 0

     -2 100                      200             300                   400              50 Time (s)

Figure 19-20. Predicted Intact Loop Seal Steam Flowrate, ROSA 10-Percent Cold Leg Side Break oA\4384-non\4384-19awpd: Ib-040403 19-42

Pr ed ic ted 10 - E

                          .        V 4        ..

Cfo co U 2 . . . .. . . . . . . . . ........ . . . . . . . . . . . . . . . . . . . . . .4 02 ............ .... .....

     -2 100                     200                300                400                    500 Time (s)

Figure 19-21. Predicted Broken Loop Seal Steam Flowrate, ROSA 10-Percent Cold Leg Side Break oMS4384-non4384-19a.wpd:lb-040403 19-43

Pr ed ic ted 40 - 30

  -5
       -       I
  -cU
 'S 20-          .

a,C) Cl

   -o 0

10 - . 0 O-100 Time (s) Figure 19-22. Intact Loop Uphill Steam Generator Tube Collapsed Liquid Level Prediction, ROSA 10-Percent Cold Leg Side Break o:\4384-non\434-19awpd:lb-040403 194

Pred ic ted 40 30

  -J
  -o 20-     -

_D _ . cn 0~ 10- . . 100 Time (s) Figure 19-23. Broken Loop Uphill Steam Generator Tube Collapsed Liquid Level Prediction, ROSA 10-Percent Cold Leg Side Break o-.M4384-non438419a.wpd:lb-040403 1945

I Predicted 1100 - 1000 . . . . . . . . . . . .. ................................... 1000~~~~ 9Q0- . ..... <...... ....... ............. ........ ............ LL800-

  -     700 -                          ........                       ...

5 00-. ...................... ..... ....... .............. ..... 400 . .. . . .. . . .. . . .. . . .. . . .. . . . ........... 300 f I I 100 200 300 400 Time (s) Figure 19-24. Predicted PCT, ROSA 10-Percent Cold Leg Side Break o:\4384-non\4384-19a.wpd:lb-040403 19-46

                                             'v  I 600   T  Venturi 60 03 Top SB-CL-03 Venturi Side SB-CL-01 Bottom SB-CL-02 Orifice Venturi Figure 19-25. Break Orientation in LSTF 2.5-Percent Cold Leg Break Tests (Koizumi, et al., 1987) o\4384-non\438419awpd:lb040403                    19-47

I Measu red P---Pred i cted 2500 - 2000 .............

 . 1500-      S.

c) CM_ co 1000 .............. 0 Time (s) Figure 19-26. Comparison of Predicted and Measured Primary System Pressure, ROSA 25-Percent Cold Leg Side Break o-\4384-non\4384-19a.wpd:lb-040403 19-48

Measured Pred i c ted 100 - 80 . ... 1-co E 60 . ...

 -o Cif 0

i) 40 ...... 0 u) co) 40 . . . .. 20 *1' vt0J 0-Time (s) Figure 19-27. *Comparison of Predicted and Measured Break Flowrates, ROSA 2.5-Percent Cold Leg Side Break o:4384-non\4384-19a.wpd:lb-040403 19-49

Measured

         ---      -   Predi c ted 14 12
   '4-10 cv
    -C.
     -5 C-,

Time (s) Figure 19-28. Comparison of Predicted and Measured Core Collapsed Liquid Levels, ROSA 25-Percent Cold Leg Side Break o:\4384-nonN4384-19a wpd:lb-040403 19-50

Measured

         ----         Pr ed i c ted 1400 1200 1000 11 L-800 a-E 600 400 200 1000 Time (s)

Figure 19-29. Comparison of Predicted and Measured PCTs, ROSA 2.5-Percent Cold Leg Side Break o:\4384-non\4384-19a.wpd:lb-040403 19-51

Measured Pred i c ted 40 30- 6........... J I

        ' 20 - -J II      . . .. . .l .*
   -o              'I1 cn        _        1.

C) _ I _. IX 0~10- .. ,.....-.... 10' tt .- __ Time (s) Figure 19-30. Comparison of Predicted and Measured Broken Loop Uphill Steam Generator Tube Collapsed Liquid Level, ROSA 25-Percent Cold Leg Side Break o:\4384-non\4384-19a.wpd:1b-040403 19-52

30 Side Break

                                                   ----        Botom Break 20 - --             Top Break 0k9ge.

u-Uiceriaint Is 3 l C>

                                          =0 0              500           1000 TIME (s)

Figure 19-31A. Comparison of Experimental Break Flowrate for 2.5-Percent Top, Side, and Bottom Cold Leg Breaks (Koizumi, et al., 1988)

                               -            Side
                                      ---     Bottom
                               ---          Top 60
                                                        .                ,               ,.     -25 50-

0. . . . ..

20

                             .40-E
                           -2 15.

D3 - 'I - U, 10 ; ,10 1 20204060 0 Time (s) Figure 19-31B. Comparison of Predicted Break Flowrate for 2.5-Percent Top, Side, and Bottom Cold Leg Breaks o:\4384-non\4384-19a.wpd:lb-040403 19-53

E 4.0 SIDE BREAK Xj ---- BOTTOM BREAK

                             >-              ---    T~~OP BREAK uj .0                              \       V

a. 1.0 'I 00 0 500 1000 1500 TIME (s)

Figure 19-32A. Comparison of Experimental Core Collapsed Liquid Levels for 2.5-Percent Top, Side, and Bottom Cold Leg Breaks (Koizumi, et al., 1988)

            -         SiSTde Bottom Top 12-                                                  .            .
      -J
     'On 6-.        .....                                                 ..
                                                             \.................        ........

0 200 400 600 800 1000 1200 Time (s) Figure 19-32B. Comparison of Predicted Core Collapsed Liquid Levels for 2.5-Percent Top, Side, and Bottom Cold Leg Breaks n-\.43RLnnnU3R4-l gw-n-lb-40403 19-54

200 Cold Lq Top I uRt/wned to wo-cse c,dxtm Pow W.1 Loap Seal Clearin turned ICSle ClxlO O DOWiw#tw Dle A Side Broo b - Sid Brook o Boltai Bred --- llau Break P \s~ ~ , f _op

                                                                  -       Break
     -                       Toap Woek
     > 100
    -J              K          Branc Il Pipe areak 10 cr 0

A Co*-Leq UtlmI - . {.

             -0                        100             200                300        4                500 Time     ls)

Figure 19-33A. Mixture Levels in Broken Cold Leg Measured for Side, Bottom, and Top Break Experiments (Koizumi, et al., 1988) S i de

                --     - - Bottom

__- Top 0.7 200 0.6 _ 0.5 150

         >- 0.4
         -J Q)     0-3 100 E I-x      0.2 50 0.1 0                                                                           0 Time (s)

Figure 19-33B. Two-Phase Mixture Level Prediction in Broken Cold Leg for 2.5-Percent Break Cases o:\4384-non\4384-19a.wpd:lb-040403 19-55

RUNS SB-CL-01. 02 N 03

               .&           MI TL4  377 CD2 Tu   377 -3    T       377 RUN       01 Ct                            02         n03 L,

gr - - - Sn - w .ww awu lZW 34oe l *Ou I Bio TIU S Figure 19-34A. Cladding Temperature of B-20 Rod at Position 7 (8.67-ft Elevation) for Side, Bottom, and Top Break Experiments (Koizumi, et al., 1987) S i de

        -  -     -  -    Bo t tom T  op 1200
                                                                                                        -  900 1000                                                                                             - 800 i  S 800                                                                                            - 700 0)
                    -     ¢   z.z       *                       -I c>

tE - 600 Q) 600 _~~ . . I

                                                                                                        - 500 400
                                                                                                        -  400 200 0           200       400       600         Tim (s)     0 0    12      14C)         I 100 Ti me          (s)

Figure 19-34B. Clad Temperature Predictions at 8.66-ft Elevation for 2.5-Percent Side, Bottom, and Top Break Experiments o4384-non\4384-19a.w pd:1b-040403 19-56

SECTION 20 LOFT SIMULATIONS USING WCOBRA/TRAC-SB 20-1 Introduction Other integral-systems tests that were simulated using WCOBRA/TRAC-SB are based on experiments conducted at the Loss-of-Fluid Test (LOFT) facility. The LOFT loss-of-coolant experiments (LOCEs) have been widely used for validation of PWR computer models due to the relatively large scale of the facility (1:60 volume scaling of a commercial four-loop PWR) and the use of a nuclear core designed to have the same physical, chemical, and metallurgical properties as a PWR core (Reeder, 1978). The large scale of the facility enables multidimensional effects which allow assessment of the ability of the code to predict these effects. Also, because LOFT is the only integral facility to use a nuclear core, the experiments are considered to be an essential part of the validation package for any PWR computer model. The LOFT facility is designed to provide thermal-hydraulic data representative of a large rupture of a main coolant pipe. Consequently, the facility design and instrumentation are oriented toward fulfilling these goals. The LOFT facility contains a number of atypicalities to a large-scale PWR for large break LOCA simulations; for small break LOCAs, the facility contains even more atypicalities, some of which were not recognized until after some small break LOCEs had been completed. Nevertheless, the facility remains a valuable benchmark for model assessment, provided the atypicalities are recognized and do not overshadow the thermal-hydraulic behavior of interest. In general, LOFT fluid volumes were scaled according to the ratio of LOFT core power to PWR core power of a large plant. If practical, flow areas were scaled by the same ratio. In this section, simulations of LOFT small break LOCEs L3-l, L3-7, and L3-5 using WCOBRAtIRAC-SB are presented and compared to various data acquired during the experiments. L3-1 and L3-5 simulate a 4-inch equivalent diameter break. The L3-1 break is located at the centerline of the inactive loop cold leg. For LOCE 13-5, the break is located in the active loop cold leg. The L3-1 experiment is of interest for model validation due to the influence of accumulator injection on the primary system response during the test; by comparison, LOCE L3-5, because of its location in the active loop, is more typical of the break geometry expected for a small break LOCA in a full-scale PWR. L3-7 simulates a 1-inch equivalent diameter break also at the centerline of the inactive loop cold leg. This experiment is o:\4384-non\438420.wpd:lb440403 20-1

of interest for model validation due to the extended period of natural circulation that was established and maintained during the test. 20-2 LOFT Facility Description The following text describing the LOFT facility is summarized from NUREG CR-1 145 (Bayless, et al., 1980) with additional information from NUREG CR-0247 (Reeder, 1978) and changes for readability where necessary. Figure 20-2-1 (Bayless, et al., 1980) illustrates the layout of the LOFT facility. LOFT consists of five major components: the reactor vessel, the active loop, the inactive loop, the blowdown suppression system, and the emergency core cooling system (ECCS). A reflood assist bypass line (RABL) was also included in the inactive loop to provide additional safeguards capability in an emergency. The LOFT reactor vessel is similar to a PWR reactor vessel in that it includes a nuclear core and an integral annular downcomer. However, the LOFT downcomer contains large metal filler blocks not found in a standard PWR downcomer to maintain volume scaling. Also, the LOFT vessel does not have an upper head typical of a PWR vessel. Figure 20-2-2 based on Reeder (Reeder, 1978) illustrates the LOFT reactor vessel and shows the various flowpaths that are available for coolant that enters through the vessel inlet nozzle. The 5.5-foot LOFT nuclear core consists of nine fuel assemblies designed for a thermal output of 50 MW. As shown in Figure 20-2-3 (Bayless, et al., 1980), five assemblies have a 5xl5 square cross section and the remaining four assemblies have a triangular cross section that represents a portion of the square cross-sectional design. The square assemblies have 225 pin locations, 21 of which are occupied by guide tubes except for the center assembly; the center guide tube is not installed to allow for additional instrumentation. The triangular assemblies have 78 pin locations, 8 of which are occupied by guide tubes. In all, the 9 LOFT assemblies contain 1,300 fuel rods, 136 guide tubes, and 1 open hole for instrumentation. The LOFT active loop is similar to a PWR main coolant loop in that it includes a hot leg, an active steam generator (inverted U-tube and shell design), pump suction piping, and a cold leg. However, the LOFT active loop uses two coolant pumps in parallel, rather than a single coolant pump typical of a PWR loop, and the LOFT steam generator tubes are not full height. The LOFT o:\4384-non'4384-20.wpd:Ib-040403 20-2

secondary side steam flow is controlled on a pressure hysteresis following steam generator trip and is, therefore, also different from the PWR. The LOFI' inactive loop contains a hot leg, a steam generator simulator to represent the steam generator resistance, a reactor coolant pump (RCP) simulator to represent the pump resistance, and a cold leg. The hot and cold legs are connected on one side to the reactor vessel and on the other side to the quick-opening blowdown valves of the blowdown suppression system. The hot and cold legs are also connected by the RABL, normally closed during the LOCEs, but which provides additional safeguards capability by allowing steam generated in the core to be vented directly to the break in an emergency. The LOFT blowdown suppression system consists of header pipes from the quick-opening blowdown valves in the inactive loop, connected to a blowdown suppression tank with a spray system for steam condensation. This system provides the backpressure to the RCS for the LOCEs and, therefore, simulates the containment in a PWR. The LOFT ECCS consists of two accumulators; a high-pressure injection system (HPIS), consisting of two high-pressure injection pumps and a low-pressure injection system (LPIS), consisting of two low-pressure injection pumps. Generally, only one of each is active during a given experiment. o:\4384-non\4384-20.wpd:lb-040403 20-3

0 IntactA loop w 0 0e S f0 w CZ 0 4 PI 0 0

UPPER CORE SUPPORT STRUCTURES REACTOR VESSEL FILTER 025 D IN. FILTER GAP

                                                             -TOP OF FUEL ASSEMBUES

_ CORE BARREL AND FLOW SKIRT

                                                               .2 IN. ANNULAR DOWNCOMER CENTER FUEL
                                                             -MODULE
                                                             -  CORNER FUEL MODULES

_LOWER CORE SUPORT STRUCTURE Figure 20-2-2. Diagram of LOFT Reactor Vessel and Flowpaths (Reeder, 1978) o:\4384-non\4384-20.wpd:lb-040403 20-5

0 Broken loop 'QJ99UUt9QUUUQJQJQ9Q9( Broken loop IO / 2 K cold leg 3'0 0 0 03fi000 0 0 0 003f63 hot leg 00000000@@0E( 3 I4 5 I 6 84888890083i0()33EX63 , w -w '0@0000000080e B,~ GE)

a eI 0600006000000( 93306

                                   'N tJ8y Core key 08981 83000664 1800000080000E B,I 0300300 0 00888810             SG 06@@00           80S(

rrt~ N a88088(01es t o e e eOO Io@ee.e ee@

                                                            -a I@e.ee@eo8ecee@sss GSGG3i3f00000G0 @(

C 0 t a a' an a 080808001O nan a in in inn v 9 in in an r nmn 1WXbYXYYY-YYYYa_ Identification key 3C300 00

00880MC18808i C396 0' 0

     -I                                           C i@oce3E3ecieoe                               0080000000000(

30000@0000000 eeee0oeGO G G @@G00f36E00 06 CD

• Guide tube Ci0006000000i Instrumented C S0@00 0000000E3( oeeeeeeee I

9-CD is Pin Pin 1000000688000i 10O@0

                                                    ' o e e 00808@OO 8

e0e9 eo

                                                                                               '800600000008w0
                                                                                               '003000000000(

00080*0880000000C Unnstrumented - Iwe w C3 w w C3 an w C3 0 0 A eee@eeoe@eoe 1 Guide tube teoeecaeecinee@e

                                                                                                  @eeeee8eeeeeec B,- ee336ee6eeee 332e

_tssasssasssya_ v AAAAAAAAAAAAIARAAAA

                                                                                               ' & @ Ge3  A _e3  iC Y R

• iE Yee(3 Y( ._ y y y yyyyyyyyS-C

                                                                                               'teeoeeee eeeeeeeeeeeee                    30603 30000*'3
     '0                                                                                        e@ee@eee@ee@e                    )8ec3OG0000803/

0 eeeee@eeeeee 30w 000@9

                                                                          .@eGcE30e         *eeeeeeeeeoeee c 36@3G000(lti
                                                                             / 000636e         e.eeeeeeeeese CN                                                                                           se0000e@00000                     30608 S@cheeGsoeec,ea 0000000000S 0@0G000000000 eeeeceooe Intact loop                                  @,G@G00600000                                             Intact loop
                                                                                               @@G@00000000&

hot leg @000000000& cold leg L3-1 INEL-A-6901-2

20-3 WCOBRAITRAC-SB Model for Simulation of LOFT Small Break LOCEs L3-1, L3-7, and L3-5 Section 14-1 of WCAP-12945-P-A (Bajorek, et al., 1998) describes the WCOBRAfIRAC model that was used to simulate LOFT large break LOCEs L2-2, L2-3, L2-5, and LB-1. For large break, the LOFT reactor vessel is modelled [

                                          ]axc This section discusses the WCOBRAIRAC-SB modelling for LOFT small break LOCEs L3-1, L3-7, and L3-5 using the large break modelling described in Section 14-1 as a basis. For small break, LOFT is modelled [
                    ]a,C 20-3-1 Reactor Vessel Modelling In the LOFT large break model, the reactor vessel is represented [

a.c Figure 20-3-1 illustrates the reactor vessel modelling for the WCOBRARAC-SB simulations of LOCEs L3-1, L3-7, and L3-5. o\4384-non'4384-20.wpd:lb-040403 20-7

                                                                                     -               l~~~~~~~~

The core power as a function of time for the LOFT small break LOCEs is supplied to WCOBRATRAC-SB as a boundary condition, based on Figure 21 of NUREG CR-1 145 (Bayless, et al., 1980) for L3-1; Figure 4-2 of NUREG CR-1570 (Gillas and Carpenter, 1980) for L3-7; and Figure 2-2 of NUREG CR-1695 (Dao and Carpenter, 1980) for L3-5. Use of these best estimate curves in place of the WCOBRAiTRAC-SB kinetics and decay heat models ensures that the thermal-hydraulic predictions are not influenced by known differences in core power behavior between the code modelling and the experiments. 20-3-2 Active Loop Hot Leg, Pressurizer, and Steam Generator Inlet Piping Modelling In the LOFT large break model, the active loop hot leg and pressurizer surge line are modelled [

                                                                       ]a. Figure 20-3-2 illustrates the active loop hot leg, pressurizer, and steam generator inlet piping modelling for the WCOBRAIRAC-SB simulations of LOCEs L3-1, L3-7, and L3-5.

20-3-3 Active Loop Steam Generator Modelling In the LOFT large break model, the active loop steam generator is modelled [

                                                                                        ]a,c o:\4384-non4384-20.wpd:1b-040403                20-8

[ Ja' Figure 20-3-3 illustrates the active loop stearn generator modelling for the WCOBRA/TRAC-SB simulations of LOCEs L3-1, L3-7, and L3-5. The LOFT steam control valve operates on a pressure hysteresis following steam generator trip and is, therefore, different from the PWR. For L3-1, L3-7, and L3-5, a nontrivial amount of leakage through this valve affected the experimental results. [ Iaxc 20-34 Active Loop Pump Suction Piping and RCP Modelling In the LOFT large break model, the active loop pump suction piping was modelled [ Iaxc Figure 20-3-4 illustrates the active Ioop pump suction piping and RCP modelling for the WCOBRArRAC-SB simulations of LOCEs L3-1, L3-7, and L3-5. The pump coastdown for the LOFT small break LOCEs is supplied to WCOBRAITRAC-SB as a boundary condition, based on Figures 59 and 60 of NUREG CR-1 145 (Bayless, et al., 1980) for L3-1; Figure 5S-1 of NUREG CR-1570 (Gillas and Carpenter, 1980) for L3-7; Figures 3S-45 and 3S-46 of NUREG CR-1695 (Dao and Carpenter, 1980) for L3-5. Use of these experimentally obtained curves in place of the WCOBRAITRAC-SB pump coastdown calculations ensures that the thermal-hydraulic predictions are not influenced by known differences in RCP behavior between the code modelling and the experiments. o:\4384-non\4384-20.wpd:lb-040403 20-9

I 20-3-5 Active Loop Cold Leg Modelling In the LOFT large break model, the active loop cold leg was modelled [

                                                      ]aC Figure 20-3-5 illustrates the active loop cold leg modelling for the WCOBRAfJRAC-SB simulations of LOCEs L3-1 and L3-7.

The LOFT pumped injection enters the cold leg at a location near the reactor vessel, while the PWR injection point is typically further upstream. This results in distortion between the flow regimes observed in the LOFT cold leg and the flow regimes observed in a PWR cold leg and must be considered before using LOFT cold leg behavior to draw conclusions regarding the PWR small break model. 20-3-6 Accumulator and ECCS Modelling In the LOFT large break model, the accumulator and ECCSs were modelled using: [

                                                                               ]ac For LOFT LOCEs L3-5 and L3-7 accumulators were valved out during the period of interest - only in LOCE L3-1 is the recommendation active in the simulation.

20-3-7 Inactive Loop Modelling In the LOFT large break model, the inactive loop was modelled [

                                                  ]^. For small break, the inactive loop modelling is as illustrated in Figure 20-3-6. The hot leg is modelled [
                                                                                                   ]a'c o:\4384-non\43S4-20.wpd:lb-040403               20-10

I

                 ]C The RABL connecting the inactive loop hot and cold legs was designed to remain closed during the experiments. For L3-1 and L3-7, however, a nontrivial amount of leakage through the RABL affected the overall system bypass, which must be modelled in the WCOBRAJ1RAC-SB simulations. The RABL is modelled for the small break simulations, [
               ]' Also for 13-5, safety and accumulator injection are moved to the downcomer.

20-3-8 Break Modelling L3-1 and L3-7 simulated single-ended breaks and used the same break units shown in Figures 20-3-7 and 20-3-8 of the inactive loop cold leg. For the WCOBRAtRAC transient simulations, the break assembly is modelled, as shown in Figure 20-3-6, [

                                                                                    ]aC For L3-5, the break unit is located in the intact loop cold leg through an instrument as shown in Figure 20-3-9 (Dao and Carpenter 1980). In this case, [

IaC o:.4384-non\4384-20.wpd:lb-040403 20-11

Figure 20-3-1. Reactor Vessel Modelling for LOFT LOCEs L3-1, L3-7, and L3-S o\4384-non\4384-20.wpd:lb-040403 2012

Figure 20-3-2. Active Loop Hot Leg, Pressurizer, and Steam Generator Inlet Piping Modelling for LOFT LOCEs L3-1, L3-7, and L3-5 o\4384-non\4384-20.wpd:lb-040403 20-13

a4 Figure 20-3-3. Active Loop Steam Generator Modeling for LOFT LOCEs L-, L3-7, and L3-5 o\4384-non\4384-20.wpd:lb-040403 20-14

asq Figure 20-34. Active Loop Pump Suction Piping and RCP Modelling for LOFT LOCEs L3-1, L3-7, and L3-5 o:W4384-nonW4384-20.wpd:lb-040403 20-15

I a4 3/4L Figure 20-3-5. Active Loop Cold Leg Modelling for LOET LOCEs L3-1 and L3-7 o\4384-non\4384-20.wpd:lb-040403 20-16

a4 Figure 20-3-6. Inactive Loop Modelling for LOFT LOCEs L3-1 and L3-7 o\43B4-non\4384-20.wpd:lb-040403 20 17

w g o 0 w DE-BL-1A,B,C PE-BL-1 04 TE-BL-1A,B,C TTE-BL-lA,B,C 00 0 0 I To QOBV thJ orifice L3-1 INEL-A-13 099-1

o C7

                                                         -0 C                 J~~~C E          c 0~~

0 _~~~~~~~~~~~ E Li ~ ~~~~~~~;CD co ICED

                                                      ~~~~~~~~~~~~~

U,: 0 C>~~~~~~~~~~~~~~~~~~~~~~~~.

• m m -m Figure 20-3-8. Break Orifice Assembly for LOFT LOGE L3-7 (Gillas and Carpenter, 1980) o:\4384-non\4384-20.wpd:Ib-040403 20-19

                         .Exlstlng 14-in. Sch 160 pipe Intact loop cold log at PC-Instrument locatlon
                                               ,DE-PC-SI&B.C
                                                             .PE-rS1.2 TE-PC-SI A,AC
                                                             \,    Break orifice PdE PCS1X Figure 20-3-9. LOFT LOCE L3-5 Break Unit Configuration o:\4384-non\4384-20.wpd:lb-040403                              20-20

20-4 Steady-State Simulations for LOFT Small Break LOCEs L3-1, L3-7, and L3-5 Prior to the transient simulations, a 200-second steady-state was run for each LOCE to ensure stable system behavior prior to break initiation. Parameters that are normally varied in large break LOCA test simulations to obtain satisfactory steady-state conditions are varied in these small break LOCE calculations. System coolant mass (i.e., enthalpy) is varied to obtain a primary system pressure within the stated data uncertainty. Pump speed is varied to obtain the desired primary system flow. Secondary system pressure is varied to obtain active loop hot and cold leg temperatures within specified limits. The average linear heat generation rate is varied to obtain the correct core power. Feedwater temperature is varied to obtain the appropriate secondary fluid temperature. With these parameters controlled, other parameters (such as the inactive loop hot and cold leg temperatures) may be within or outside of the limits specified by the data uncertainty. 20-4-1 Steady-State Simulation for LOFT LOCE L3-1 Table 20-4-1 compares the results of the 200-second WCOBRATRAC-SB steady-state simulation for LOCE L3-1 to the initial conditions identified in NUREG CR-1 145, pages 28 and 29 (Bayless, et al., 1980). The following summarizes the results of the steady-state simulation for L3-1:

• The calculated active loop hot leg temperature predicts hot leg streaming. The temperature at the top of the hot leg is calculated to be greater than that calculated for the cells below. The numerical average of the calculated temperatures lies within the data uncertainty, as does the mixed liquid temperature for channel 88 at the steam generator unit.
• The calculated RABL flow is 1.5 percent of total active loop flow. This is below the estimated 3-percent bypass for the RABL.
• The calculated inactive loop cold leg and hot leg temperatures lie within the data uncertainty.
• The steam generator collapsed liquid level is approximately 9.4 feet; this is approximately 0.9 feet below the data minus uncertainty.

o:\4384-non\4384-20.wpd:lb-040403 20-21

• The steam generator feedwater and steam flowrates are within the data and uncertainty. The steaming rate is approximately 1 percent higher than the feed water flowrate. This is consistent with a calculated liquid temperature at the bottom of the boiler approximately 5 °F greater than the data for the steam generator liquid temperature.
• The steam generator pressure was adjusted to give the correct primary side temperatures. [

                                         ]a.

20-4-2 Steady-State Simulation for LOFT LOCE L3-7 The results of the L3-7 steady-state calculation, as summarized in Table 204-2, are within the limits of the data and uncertainties identified in NUREG/CR-1570 (Gillas, 1980) except as noted in the following. The steady-state values outside the limits of the data and uncertainty are judged to be minor and have no significant effect on the overall transient calculation.

• The calculated active loop hot leg temperature predicts hot leg streaming. The temperature at the top of the hot leg is calculated to be greater than that calculated for the lower cells. The numerical average of the calculated temperatures lies within the data uncertainty, while the mixed fluid temperature in channel 88 is 0.1 °F below the data and uncertainty.
• The calculated RABL flow is 1.5 percent of total active loop flow; this is below the estimated 3-percent bypass for the RABL.
• The calculated inactive loop cold leg temperature lies within the data uncertainty.

Even with the lower RABL bypass flow, the inactive loop hot leg temperature is calculated to be approximately 1.3°F below the data minus uncertainty.

• The steam generator level is approximately 9.3 feet and approximately 1 foot below the data minus uncertainty.

o:\4384-non\4384-20.wpd:lb-440403 20-22

Steam generator pressure is 29.3 psi below the data and uncertainty. As with L3-l, the secondary side pressure is adjusted to obtain desired primary side temperature. [ a.c 204-3 Steady-State Simulation for LOFT LOCE L3-5 The results of the L3-5 steady-state calculation, as summarized in Table 204-3, are within the limits of the data and uncertainties identified in NUREG/CR-1695 (Dao and Carpenter, 1980) except as noted in the following. The steady-state values outside the limits of the data and uncertainty are judged to be minor and have no significant effect on the overall transient calculation.

• The calculated active loop hot leg temperature predicts hot leg streaming. The temperature at the top of the hot leg is calculated to be greater than that calculated for the lower cells. The numerical average of the calculated temperatures lies within the data uncertainty, as does the mixed fluid temperatures in channel 88.
• The calculated RABL flow is 1.5 percent of total active loop flow; this is below the estimated 3-percent bypass for the RABL.
• The calculated inactive loop hot leg temperature is greater than the data plus uncertainty by 1.4°F. The inactive loop cold leg temperature lies within the data and uncertainty.
• The steam generator level is approximately 9.3 feet and approximately 1 foot below the data minus uncertainty.
• The pressurizer temperature is approximately 0.4 F below the data minus uncertainty, which is consistent with the calculated pressurizer pressure.
• As in the L3-1 steady-state calculation, the secondary side pressure is adjusted to obtain the desired primary side temperatures, and is 19.4 psi below the data minus uncertainty. This translates to 4 °F in secondary side saturation temperature -

[

                                  ]a.c.

o:\4384-non\4384-20.wpd:lb- 040403 20-23

Table 20-4-1 Comparison of LOFT LOCE L3-1 Steady-State Calculation to L3-1 Data Calculated Value Uncertainty',' Parameter (190 sec) Datal'2 (+) Reactor Power 48.9 MW 48.9 MW 1 Peak Linear Heat Generation Rate 16.099 kW/ft 15.75 kWM 0.305 Intact Loop Flow Rate 1065 lb/sec 1067.0 lb/sec 13.9 Hot Leg Pressure 2149.6 psia 2153.25 psia 5.8 Hot Leg Temperature Top Cell 577.2 0 F Average 573.9 0 F SG Entry Pipe 573.8 0 F 573.5-F 1.8 Cold Leg Temperature 540.2 0 F 537.5-F 5.4 Pressurizer Liquid Volume 21.9 ft3 21.9 ft3 0.28 Pressure 2146.5 psia 2147.5 psia 5.8 Temperature 6445.9 0F 650.9-F 5.4 Broken Loop Cold Leg Temperature 540.3 OF 543.5*F 9.0 Hot Leg Temperature 4 554.8°F 552.0-F 9.0 Steam Generator Secondary Secondary Flowrate 55.1 lb/sec 0.88 Feedwater Flowrate 55.6 lb/sec Steam Flowrate 3 55.8 lb/sec Pressure 760.9 psia 787.4 psia 16.0 Water Temperature 505.4-F 7.0 Downcomer 497.1 F Boiler 510.7F Liquid Level 9.4 ft 10.34 ft 0.03

1. NUREG CR-1 145 (Bayless, et al., 1980), see Appendix B for a copy of SS data

2. Converted from SI units

3. Value in the table represents value at 190 sec.

4. Average of 549.8, 554.6, and 560.4 o:\4384-non\4384-20.wpd:lb-040403 20-24

Table 20-4-2 Comparison of LOFT LOCE L3-7 Steady-State Calculation to L3-7 Data Calculated Value Uncertainty',' Parameter (200 sec) Data2 () Reactor Power 49 MW 49 MW I Peak Linear Heat Generation 16.132 kW/ft 16.093 kW/ft 1.128 Rate Intact Loop Flow Rate 1064.2 lb/sec 1061.1 lb/sec 13.9 Hot Leg Pressure 2162.0 psia 2160.5 psia 36.3 Hot Leg Temperature Top Cell 579.6 0 F Average 576.4°F SG Entry Pipe 576.3°F 577.3 0.9 Cold Leg Temperature 542.80 F 541.1`F 5.4 Pressuizer Liquid Volume 22.6 ft 22.25 ft' 1.77 Pressure 2158.7 psia 2160.5 psia 5.8 Temperature 646.9 0 F 647.3°F 0.54 Broken Loop Cold Leg Temperature 542.9 F 544.2-F 4.5 Hot Leg Temperature' 556.7 F 550.9 0 F 4.5 Steam Generator Secondary Secondary Flowrate 61.7 lb/sec 0.88 Feedwater Flowrate 60.7 lb/sec Steam Flowrate 60.7 lb/sec Pressure 777.5 psia 808.5 psia 1.74 Water Temperature 519.5-F 0.36 Downcomer 511.06°F Boiler 515.5°F Liquid Level 9.3 ft 10.50 ft 0.20

1. NREG CR-1570 (Gillas, et al., 1980)

2. Converted from SI units

3. Average of 551.0, 556.4,562.6 o:\4384-non434-20.wpd:lb-0404032 20-25 5

I Table 20-4-3 Results of LOFT LOCE L3-5 Steady-State Calculation to L3-5 Data LI Calculated Uncertainty'; Parameter Value Data'2 (+) (190 sec) Reactor Power 49 MW 49 MW 1 Peak Linear Heat Generation 16.132 kW/ft 16.032 kWM 1.128 Rate _ _ _ _ _ _ _ _ Intact Loop Flow Rate 1048.4 lb/sec 1050.3 b/sec 13.9 Hot Leg Pressure 2160.0 psia 2154.7 psia 20.3 Hot Leg Temperature Top Cell 580.4F Average 577.1F SG Entry Pipe 577.0-F 577.1F 3.6 Cold Leg Temperature 543.1F 544.F 1.8 Pressurizer Liquid Volume 24.0 ft3 24.0 ft3 2.1 Pressure 2156.9 psia 2157.6 psia 2.9 Temperature 646.6-F 647.23-F 0.18 Broken Loop Cold Leg Temperature 543.2-F 541.1F 4.5 Hot Leg Temperature 3 557.8-F 551.9-F 4.5 Steam Generator Secondary Secondary Flowrate 58.2 lb/sec 2.2 Feedwater Flowrate 57.7 lb/sec Steam Flowrate4 57 lb/sec Pressure 781.0 psia 809.1 psia 8.7 Water Temperature 517.7-F 1.8 Downcomer 504.7F Boiler 515.0-F Liquid Level 9.3 ft 10.30 ft 0.13

1. NUREG CR-1695 (reference 8, copy of reference table in Appendix B)

2. Converted from SI units

3. Average of 564.5, 557.4, 551.4

4. Slight oscillation about the steady-state feed flow, see figure, average about 190 sec = 57.0 lb/sec.

o:\4384-non\4384-20.wpd:lb-040403 20-26

20-5 Transient Simulations for LOFT Small Break LOCEs L3-1, L3-7, and L3-5 20-5-1 Transient Simulation for LOFT LOCE L3-1 The LOFT LOCE L3-1 is a 4-inch equivalent break in the inactive loop cold leg. This experiment started from initial conditions similar to L3-5 with the same break orifice as described in this section. The reactor was tripped several seconds prior to opening the blowdown valves to initiate the break. The sequence of events is listed in Table 20-5-1-1 (Bayless, et al., 1980). The RCPs were tripped shortly after the break opening, and safety injection was initiated on low primary system pressure. Safety injection was directed into the active loop cold leg. The intent is to simulate L3-1 with WCOBRAfTRAC-SB based on the initial conditions described in Section 20-4-1 and the appropriate boundary conditions. The physical arrangement of the inactive loop cold leg with the RABL connecting the inactive loop cold and hot legs coupled with the lack of a complete loop means that L3-1, like L3-7, is atypical of a full-scale PWR geometry. Leakage through the RABL equalizes pressure between the hot and cold legs and acts like loop seal clearing. While L3-1 and L3-7 may not represent typical behavior of a small break LOCA in a PWR, they remain useful in evaluating WCOBRAJIRAC-SB. A comparison of the calculated and measured temperatures upstream of the break, as seen in Figure 20-5-1-1, shows that the calculated temperature (dashed curve) initially decreases several degrees, while the measured temperature indicates an initial increase of approximately 10°F. A comparison of the measured inactive loop hot and cold leg temperatures, as seen in Figure 20-5-1-2, equalizes within approximately 30 seconds into the transient. This indicates the possibility that flow reversal through the RABL occurred by 30 seconds into the transient. The calculated time of flow reversal through the RABL is approximately 30 seconds into the transient as seen in Figure 20-5-1-3. If this is later than the actual time of flow reversal, or the calculated RABL flow is less than the actual flow, these errors allow the calculated temperature to remain low. After the calculated RABL flow reversal, the calculated cold leg temperature increases only slightly. The hot and cold leg temperatures equalize when the hot leg temperature has fallen to the cold leg temperature. Calculated saturation of the inactive loop cold leg is delayed by more than 200 seconds compared to the data as seen in Figures 20-5-1-2 and 20-5-1-4. o-\4384-non\434-20a.wpd:lb-040403 20-27

A comparison of the measured inactive loop mass flowrate and the indicator of the actual break flow, and the calculated break mass flowrate is shown in Figures 20-5-1-5a and 20-5-1-5b. The calculated mass flow is significantly greater than that indicated by the inactive loop flowrate after 200 seconds. This is consistent with the extended period of subcooled flow and lower than measured temperature. Even with an unrealistically high calculated mass ejection through the break, the calculated pressure stays higher than the data after saturation is reached as seen in Figure 20-5-1-6. This may be the result of an underprediction of the quality upstream of the break, as seen in Figure 20-5-1-7, which causes the break to remain plugged. Both the calculated break mass flowrate and system pressure remain higher than the data. The extent to which the assumed RABL leakage area is a contributor to the errors in the simulation of L3-1 is not determinable without explicit knowledge of the RABL leakage during L3-1. At best, the sensitivity of the simulation to RABL leakage could be determined by additional calculations that vary the leakage area. o:\4384-non\4384-20a.wpdlb-040403 20-28

Table 20-5-1-1 Sequence of Events for LOFT LOCE L3-1 (Bayless, et al., 1980) l Measured Data

                                                                    ~~~~~~~~~~~~L3-1 Event                                    (seconds)

LOCE initiated 0.0 Primary coolant pumps tripped 0.04 0.01 HPIS injection initiated 4.6 ++/-0.5 Pressurizer emptied 17.0 +/- 1.0 Upper plenum reached saturation 24.4 +/- 0.5 Auxiliary feed pump started 75.0 +/- 1.0 Accumulator injection initiated 633.6 +/- 0.5 Accumulator empty 1741.0 +/- 1.0 Auxiliaiy feed pump tripped 1875.0 +/- 1.0 o:\4384-non\4384-20a.wpd:lb-040403 20-29

I CHANNEL 104 0 0 TE-BL-OO1B

         - --      -     TLN              25       2     0 LIQUID TEMPERATURE 600  -  _______

550- > 500 - V i 450 a-E 00 _ 350 - 300 - I I Time (s) Figure 20-5-1-1. Comparison of Calculated and Measured Temperature Upstream of Break Orifice for LOFT LOCE L3-1 o:\4384-non\4384-20a.wpd:lb-040403 20-30

TSATBLCL 65 0 0 PE-BL-001 BLHL 105 0 0 TE-BL-002B

            ~~BLCL                     104        0      0 TE-BL-OOIB 650-600 -

550-L _ 500 - W _

   - a) ta         _

E 49 ca- 450 E _ 3D 400- - 350-300 wOo Time (s) Figure 20-5-1-2. Comparison of Inactive Loop Cold Leg, Inactive Loop Hot Leg, and Saturation Temperature for LOFT LOCE L3-1 From Test Measurements o:\4384-non\4384-20a.wpd:lb-040403 20-31l

I Mass Flow Rote (Ibm/s) RMVM 19 2 0 MASS FLOWRATE Temperature (F) TLN 19 2 0 LIQUID TEMPERATURE

        -------- TLN                     25       2      O LIQUID TEMPERATURE 20 -                                                                     550
                                                                                 -540 15 530 U)

E ' 10

   -Q                                                       >NN.                   520 'i N                     I.-

a a) 0 5- -% %% -510 - N IN, L NC-0

                                                                                 -500E en V)     0-

_ve -490

         -5
                                                                                 -480
                           ,                                   ,   ,        -470 400 Time (s)

Figure 20-5-1-3. Comparison of Calculated Liquid Temperature Upstream of Break Orifice, Liquid Temperature in RABL, and RABL Mass Flowrate o:4384-non\4384-20a.wpd:lb)40403 20-32

T STSAT 83 3 0 SATURATION TEMP.

         - - - -TL                      83       3        O LIQUID TEMPERATURE 650 600

• 550 a,

00 E C, 500 c ~ ~ - ~ I II 450 1 0 200 400 600 800 1200 1400 1600 Time (s) Figure 20-5-1-4. Comparison of Calculated Inactive Loop Cold Leg Liquid Temperature and Saturation Temperature o:\4384-non\4384-20a.wpd:b-040403 20-33

                        ~

I 0 ~ ~ ~ -~ O. 000. t

                                         -- x ZOOO0.
                                                -rX.ME AFTER 3000.

RUPTURE Cs) 5000 DOO0. Figure 20-5-1-5a Measured Mass lowrate in Inactive Loop Cold Leg (QualiRed) (Bayless, et al? 1980)

                .I_<OR o:wj 84-non4 542iw sXl

_1t _A^\A:^,.^n

                                   -s 2034

Moss Flow Rote ( k g/s ) RMVM 25 3 O MASS FLOWRATE Void Fraction AL PN 25 2 O VOID FRACTION 25 - 20 - .8 15- .6 C-) 0

                                                                                        -4.

U-

                                                                                         -X0 o                                                                                    a i::   10-co        _                I 5-                                                                         .2 Time (s)

Figure 20-5-1-5b. Calculated Break Mass Flowrate and Upstream Void Fraction for LOFT LOCE L3-1 o:\4384-non\4384-20awpd:1t b443 20-35

CHANNEL 65 0 0 PE-BL-001

         - - - - PN                  25         2       0 PRESSURE 2500-2000-
   .2   1500 co Q>

en CO a> 1000 0 500 0 800 Time (s) Figure 20-5-1-6. Comparison of Calculated and Measured Primary System Pressure for LOFT LOCE L3-1 o-.\4384-non\4384-20awpd:1b40403 20-36

ALPN 25 2 0 VOID FRACTION

         -      - -   AL                  83         2      0 VAPOR FRACTION
               -      AL                  83         3      0 VAPOR FRACTION
               --    AL                   83         4      0 VAPOR FRACTION
         .8 c    .6 0

C-)

   >.1o
      .2 .4                   r
         .2                A I

0 200 Figure 20-5-1-7. Cal culated Void Fractions Upstream of Break Orifice and in 3-D) Channel Connected to Break Unit for LOFT LOCE L3-1 o:\4384-non\4384-20a.wpd:lb-040403 20-37

I 20-5-2 Transient Simulation for LOFT LOCE L3-7 LOFT LOCE L3-7 is a 1-inch equivalent break (13.2 mm2 ) using the spoolpiece depicted in Figure 20-3-8. The objective of L3-7 was to establish conditions conducive to long-term natural circulation. However, a geometric distortion exists in the form of the RABL upstream of the break plane. The RABL provides a direct leakage path between the hot and cold legs that does not exist in a full-scale PWR. This may influence the calculations favorably or unfavorably as discussed later in this section. For L3-7, the facility was operated similarly to a full-scale PWR, unlike the other small break LOCEs. The reactor and RCP trips occurred on low pressure. The reactor trip occurred at 36 seconds, and the pump trip occurred at 39.3 seconds, as shown in the sequence of events in Table 20-5-2-1. A comparison of the measured and calculated inactive loop cold leg temperatures upstream of the break, as seen in Figure 20-5-2-1, indicates the code overestimates the temperature during the first 800 seconds. Overall, the WCOBRAITRAC-SB calculations follow the trend of the data, although the calculation predicts the decrease in temperature to occur approximately 200 seconds early. The final calculated temperature is under predicted by approximately 14K (25°F). As shown in Figure 20-5-2a and -2b, the break/mass flow rate predicted by WCOBRA,TRAC-SB follows the unqualified measurement. WCOBRA/TRAC-SB under predicts the unqualified measurement of break flow by approximately 0.2 kg/s by the end of the transient calculation. Since the data are unqualified only the trend can be judged, and the trend is judged to be adequate. The WCOBRAITRAC-SB prediction of inactive loop cold leg pressure adequately matches the data to 400 seconds when both the prediction and measurement are approximately 8 MPa (1100 psia). At 800 seconds the prediction and measurement are approximately the same. After 800 seconds the calculated depressurization rate increases and WCOBRA/TRAC-SB under predicts the data. By the end of the transient calculation the inactive loop cold leg pressure is under predicted by approximately 0.7 MPa (102 psia). The cause of the under prediction appears to be steam generator heat rejection. o:434-non\4384-20a.wpd:lb-040403 20-38

As shown in Figure 20-5-2-4b, natural circulation is predicted to be established between 100 and 200 seconds into the transient. The predicted natural circulation flow is predicted to be between approximately 15 and 22 kg/s. This is less than half of the unqualified measurement (Gillas and Carpenter 1980). As noted previously for unqualified data, only the trend can be judged and the trend is judged to be adequately predicted. Also, shown in Figure 20-5-2-4a and -4b, are the measured and predicted steam generator inlet and outlet temperatures. The predicted temperature difference is greater than measured. Thus the primary system average temperature is significantly under predicted starting at approximately 800 seconds. The error in calculated temperature is a result of overestimation of the auxiliary feedwater flowrate. The result of the under prediction of the loop average temperature is underprediction of the primary system pressure as shown in Figure 20-5-2-3b. Figure 20-5-2-5 compares decay heat and heat rejected to the secondary system. At approximately 700 seconds predicted heat rejection increases from approximately 200 KW to approximately 500 KW. This is consistent with the above observation on steam generator primary side temperature difference. o:\4384-non\4384-20a.wpd:lb-040403 20-39

Table 20-5-2-1 Sequence of Events for LOFT LOCE L3-7 (Gillas and Carpenter, 1980) L3-7 Measured Data Event (seconds) LOCE initiated 0.0 Reactor scrammed 36.0 +/- 0.1 Control rods reached bottom 38.1 +0.1 Primary coolant pumps tripped 39.3 +/- 0.5 HPIS injection initiated 65.6 ++/-0.1 Auxiliary feed initiated 75.0 + 3.0 Pressurizer emptied 264.0 7.0 / Upper plenum reached saturation 382.0 + 6.0 End of subcooled break flow 1037.0 + 10.0 Auxiliary feed terminated 1800.0 +/- 5.0 HPIS flow terninated 1805.3 +/- 0.1 SCS steam bleed initiated 3603 1 HPIS flow reinstated 5974.2 + 0.1 Accumulator injection initiated 6028 +/-5 Break isolated 7302.0 + 0.1 o:\4384-non\4384-20a.wpd:lb-040403 20-40

Ma~~~--1. w hJ I-. I I t- -- I tt - 4 0-J I§ -%.M

                                                                 .                                  L
                          %M's I
                                                                      . _ _ _. _          t i      I _ I M
                            -R   -                                        _~_~          _I I     _I   I I
                               -AN            a         400        fi0              00                     2000 TIME ArTER RUPTUR            (s)

Figure 20-5-2-la. Measured Inactive Broken Loop Cold Leg Temperature for LOFT LOCE L3-7 (Gillas and Carpenter, 1980) TLN 25 2 0 LIQUID TEMPERATURE 565

                    ..-     555 0)

I.- c0 550 E 545 Figure 20-5-2-lb. Calculated Liquid Temperature Upstream of Break Orifice for LOFT LOCE L3-7 o:\4384-non\4384-20a.wpd:1b-040403 20-41

I 2.0 PRESSUR IZER6 l l l l_ _ DATA 5-

   -CC 1.0 =        .RF1 v            -A    1rzL LLU
    -3 Ck) cn U:

0.5 DATA _j ____r~~~'lA - 0.0 0 500 1000 I1500. 2000 TIME AFTER RUPTURE s) Figure 20-5-2-2a. Comparison of Measured Break Flow and ECCS Flow (Not Qualified) (McCreery, 1980) RMVM 25 3 0 MASS FLOWRATE 1.6 1.4 U, 1.2 -1

              .4 0

0 500 1000 1500 2000 Time (s) Figure 20-5-2-2b. Calculated Break Mass Flowrates for LOFT LOCE L3-7 o:\4384-non\4384-20awDd:lb-040403 2042

            -1
            .9n 0V

_1_f_ _I _ _ =1 lB _IIII1: _ _~_ _ _1 _ _:_: ._ -: w.~ ---- - - AR* ;a0 1uo -o 2Uoo TIME AVrER ALPME (*) Figure 20-5-2-3a. Measured Inactive Loop Cold Leg Pressure for LOFT LOCE L3-7 (Gillas and Carpenter, 1980) p 8. P 3 0 PRESSURE 16* 14 12 -

                  & 10 a-8 6t 6+

4 Figure 20-5-2-3b. Calculated Inactive Loop Cold Leg Pressure for LOFT LOCE L3-7 o-\4384-non\4384-20awpd:b-040403 20-43

I

            -     U -       S      U S   *- S -  S - *- U   - S-          S-    S-   I
                           .  *v            i   I   *t1       I' I      tI   I I   i  -   t 1 ,         @   s l              *-t TLF-          E  1   1 tt                       o T-SC-00i o TSC002     O 175                                 I     I  I    I   I  I    I     I
                                                . t  I     I   I  I    I    !       * '    t 4-    I I t-              i MN",

0% Sa tL I _. . . _. . I.

                 -4W               0               400                  s0                     4W         WQQ I ME AFTER RUPTURE (s)

Figure 20-5-24a. Measured Steam Generator Inlet and Outlet Temperatures for LOFT LOCE L3-7 580 - 100 80 O el560 -t_ 6O1 60 o: r" Q) I 40 0-0 500 - o 20 gn a 1000 Time (s) Figure 20-5-24b. Calculated Steam Generator Inlet and Outlet Temperatures Versus Calculated Active Loop Mass Flow Rate for LOFT LOCE L3-7 o:\4384non\4384-20a.wpd:1b-040e 403 20-44

QHTR 47 6 0 SG HTR

       - --     -   POWERF               0       0      0 RELATIVE       CORE POWER 50000  - __     ____

40000 - 30-0 CL 2000 10000 - I O- , , Time (s) Figure 20-5-2-5. Calculated Steam Generator Heat Rejection versus Decay Heat for LOFT LOCE L3-7 o-4384-non\4384-20a.wpd:lb-040403 20-45

I 20-5-3 Transient Simulation for LOFT LOCE L3-5 LOCE L3-5 is one of three tests in the LOFT facility that represent 4-inch diameter equivalent breaks in a full-scale PWR. It and the other tests, L3-1 and L3-6, all start from approximately the same boundary conditions. LOCE L3-5 was set up as a break in the active loop cold leg using the break unit depicted in Figure 20-3-9. The reactor was tripped several seconds prior to the initiation of the break as shown in the sequence of events in Table 2-5-3-1 (Dao and Carpenter, 1980). Once the break was opened, the RCPs were tripped and safety injection was initiated on low pressure directly to the downcomer. The accumulators were valved out to allow evaluation of the safety injection system. LOCE L3-5 was terminated at 2309 seconds with closure of the break and the safety injection system. LOCE L3-5 is the most typical of the three tests in its representation of a small break LOCA in a full-scale PWR. The WCOBRAflRAC-SB model used for the transient analysis is as previously described except that the break unit (PIPE 25) is connected to the center cell of channel 66. Safety injection is directed to the upper plenum bypass channel 33. In the WCOBRA/IRAC-SB LOFT model, channel 33 is in the downcomer segment opposite the active loop cold leg. This was done to best represent the injection path without increasing the number of downcomer segments.(') Initial conditions for the analysis are as stated in Section 2043. Calculated and measured hot leg pressures are depicted in Figure 20-5-3-1. At approximately 125 seconds, both the data and the calculation show an increase in primary system pressure. The calculated repressurization ends at approximately 300 seconds. At approximately 300 seconds, the calculated pressure falls below the data and starts to parallel the data at approximately 750 seconds. The calculation ended at 1600 seconds. The potential causes of the calculated behavior are examined in the following paragraphs. The repressurization probably results from a steam generator stall as the cold leg approaches saturation coupled with timing of the pump coastdown. A comparison of the calculated liquid flow at the steam generator inlet and calculated system pressure indicates that the repressurization starts when the liquid flow drops from approximately 65 lb/s to an average of

1. The actual injection point is midway between the active loop cold and hot legs. To connect the safety injection to the cell above the active loop cold leg will artificially inject cold water too close to the break o:A4384-non\4384-20a.wpd: lb-040403 20-46

approximately 10 lb/s as shown in Figure 20-5-3-2. Also, natural circulation appears to start at approximately 85 seconds as indicated by the increase in liquid flow. A near mirror image of this behavior is shown for the outlet of the steam generator in Figure 20-5-3-3 with liquid flow out of the steam generator ceasing at approximately 220 seconds. Thus by approximately 120 seconds, the steam generator is stalled and the active hot and cold legs are isolated from each other. The stall occurs as the last RCP reaches zero rotational speed at 120 seconds. Natural circulation decreases as the hot and cold sides of the active loop approach the same temperature as shown in Figure 20-5-3-4. Because the cold side is still calculated to be below the saturation temperature (that is, hot leg temperature), break flow remains slightly subcooled as seen in Figure 20-5-3-5. Data from the experiment show that saturated conditions at the break started at approximately 93 seconds. This limited the pressure rebound in the data. Because the break remains subcooled in the WCOBRArlRAC-SB calculation, the rate at which vapor is being generated in the core exceeds the volume flowrate out of the primary system. Primary system pressure rebounds, and the rebound is greater than the pressure rebound experienced in LOCE L3-5. The measured hot leg pressure shows a decrease in the rate of depressurization from approximately 130 seconds to 150 seconds. WCOBRAfTRAC-SB is correctly modeling the phenomena of the pressure rebound as indicated by the pressure rebound present in the data. The magnitude of the pressure rebound may primarily be the amount of RABL leakage. As shown in Figure 20-5-3-6, the measured liquid level in the steam generator side of the loop seal shows a level depression starting at approximately 400 seconds. The level depression, as measured, does not reach the horizontal pipe run, and the loop seal is not cleared. Leakage through the RABL that results in pressure equalization between the hot and cold legs prevents loop seal clearing. Also, RABL leakage will bring fluid approaching the upper plenum temperature to the cold side of the facility bypassing the active loop steam generator. The effect of the leakage is an increase in active loop cold leg temperature sufficient to cause early voiding and saturation in the active loop cold leg, which results in an early transition to saturated break flow. WCOBRAJIRAC-SB is performing well and within the uncertainties associated with RABL leakage. Because the break flow is calculated to remain subcooled during the pressure rebound, the break flow increases significantly and more mass is removed from the system than is indicated by the data as shown in Figure 20-5-3-5. Calculated break flow remains above the data until single o:\4384-non\4384-20a.wpd:lb-040403 47

phase vapor reaches the break. Because of the greater volume and mass flow out of the primary system, the primary system pressure drops below the data at approximately 400 seconds. A comparison of the calculated collapsed liquid level and the measured liquid level between the centerline of the loop seal and the steam generator inlet is shown in Figure 20-5-3-6. The WCOBRA/TRAC-SB simulation predicts the level to decrease to the top of the elbow, bottom cell of channel 58, at approximately 260 seconds; and the loop seal to pass steam at approximately 300 seconds. Because pump injection was not modeled, level recovery in the loop seal is not indicated in the WCOBRAfIRAC-SB simulation. Primary system coolant inventory decreases to a minimum of approximately 3500 lb at 1800 seconds as seen in Figure 20-5-3-7. The overall influence of pump injection on the transient simulation is minor except for loop seal and total primary inventory. o:\4384-non\4384-20a.wpd:lb-040403 20-48

Table 20-5-3-1 Sequence of Events for LOFT LOCE L3-5 (Dao and Carpenter, 1980) L3-5 Measured Data Event (seconds) Reactor scrammed -4.8 +/-0.1 Control rods reached bottom -2.8 +0.1 LOCE initiated 0 Primary coolant pumps tripped 0.8 + 0.2 HPIS injection initiated 4.0 + 0.2 Core natural circulation first indicated 17 3 Primary coolant pump coastdown completed 17.7 +/-0.2 Pressurizer emptied 22.2 +/- 0.5 Upper plenum reached saturation pressure 28.4 +/- 0.4 Active loop hot leg voiding began 30+/-5 Secondary coolant system (SCS) auxiliary feed initiated 63 3 Active loop cold leg voiding began 80+/-5 End of subcooled break flow 92.9 +/- 0.2 SCS pressure exceeded PCS pressure 745 20 Primary coolant system (PCS) mass at a minimum 1480 +/- 100 SCS auxiliary initial feed terminated 1800 5 Reactor vessel mass at a minimum 2125+/-t 180 Break isolated 2309.1 0.5 o\4384-non\4384-20a.wpd:lb-040403 20-49

                                                                                       -- I~~~~~~~~

CHANNEL 147 0 0 PE-PC-002

               - - - _ p                41       4     0 PRESSURE C)

Q a) cn U) LQ n~ Time (s) Figure 20-5-3-1. Comparison of Calculated and Measured Hot Leg Pressures for LOFT LOCE L3-5 o:\43S4-non\438420a.wpd:1b440403 20-50

Moss Flow Rote (I bm/s)

         -~            FLM                88        3     0 LIO AXIAL MASS FLOW Moss Flow Rote               ( I bm/s )

FGM 88 3 O AP AXIAL MASS FLOW 1200 .7 1000 6

   .1-1800 E
    -Q E

500 0 IIiI CD 0 III 400 1 1Co 0L 0 u) co

                                                                 '~~~k'~~~ "I    ~      c 200 II     -0
        -200                                                                          -1 Time (s)

Figure 20-5-3-2. Comparison of Liquid and Steam Mass Flowrate at Steam Generator Inlet for LOUI LOCE L3-5 o:\4384-non\4384-20a.wpd:lb-040403 20-51

I Moss Flow Rate ( I bm/s) FLM 61 3 OL AXIAL MASS FLOW Moss Flow Rote ( I bm/ s ) FGM 61 3 0 VAP AXIAL MASS FLOW 200 - -6 0-I I

     ,,I  * -200 -

I 4. E

    -o                                                                                     l-
             -400 -

0 J

                                                                                      -2Dcl 0      -600 -

L-CD C',

            -500 W  ,,     l   , ,       -     0
           -1000
           -1200 Tme (s)

Figure 20-5-3-3. Comparison of Calculated Liquid and Steam Mass Flowrate at Steam jL. Generator Outlet for LOFT LOCE L3-5 o-\4384-non\4384-20a.wpd:lb-040403 20-52

Mcss Flow Rate (Ibm/s) FLM 61 3 0 LIQ AXIAL MASS FLOW Tempercture (F)

          - - - -TV                     88          3     0 VAPOR TEMPERATURE
          -------- TV                   61          3     O VAPOR TEMPERATURE
               --- TL                   88          3     O LIQUID TEMPERATURE
          -I- - TL                      61          3     0 LIQUID TEMPERATURE 200                                                                       650 0

600

           -200 E                                                                                    LL
    -D 550
           -400 L-.

a)

                                                                                          -I.

0 C), 0 a)

        -So en cn 0     -800 450
         -1000
         -1200                                                                        400 lime (s)

Figure 20-5-3-4. Comparison of Calculated Steam Generator Outlet Flow versus Active Loop Hot Leg and Cold Leg Temperatures for LOFT LOCE L3-5 o:\4384-non\4384-20a.wpd:lb-040403 20-53

I CHANNEL 40 0 0 FR-PC-St21 RMVM 25 3 0 MASS FLOWRATE 50 - 40 30-E

   -a 0

0 20T\,t 20 o_ 10

       - 10   - g       I       I  I    I   I     I      I  I  I  II         I  !   I 0                    500               1000             1500             200 Time (s)

Figure 20-5-3-5. Comparison of Calculated and Measured Break Mass Flowrates for LOFT LOCE L3-5 o:4384-non\4384-20awpd:lb-040403 20-54

             ~~~~CHAN NE l               8        0     0 LEPDE-PC-027 Leve            53         2     0 LS downhill 6-5-           g 4~~~~~~~~

2-

         -         X,I 10 lime (s)

Figure 20-5-3-6. Comparison of Calculated and Measured Loop Seal Level for LOFT LOCE L3-5 o:\4384-nonU384-20a.wpd1b-040403 20-55

PR'I MARY

                     ~~                 0        0        0 MASS 12000 10000-E 6000-4000-2000-                  1 I     I     -

0 500 1000 1500 20 Time (s) Figure 20-5-3-7. Calculated Primary System Coolant Inventory for LOFT LOCE L3-5 o:\4384-non\4384-20awpd:1b-040403 2056

20-6 Conclusions Overall, WCOBRA/TRAC-SB simulated the phenomenology of the LOFT LOCEs fairly well notwithstanding the probable influence of unknown experimental conditions important to accurate simulation of the experiments. WCOBRAfTRAC-SB simulated LOCE L3-5 well except for the transition from forced to all natural circulation flow that resulted in an extended period of subcooled break flow. The simulation of LOCE L3-1 is initially similar to LOCE 13-5 in that a distinct pressure rebound occurs following the end of forced flow. The effect of the RABL appears to have been underestimated leading to overestimation of system pressure. WCOBRAITRAC-SB well simulated the natural circulation in LOCE L3-7, but underpredicted the depressurization that occurred during the experiment, which may also have resulted from an error in modeling the RABL leakage. 20-7 References Adams, J. P., 1979, "Quick-Look Report on LOFT Nuclear Experiment L3-1," EGG-LOFT-5057. Bajorek, S. M., et al., 1998, "Code Qualification Document for Best Estimate LOCA Analysis - Volume m: Hydrodynamics, Components, and Integral Validation," WCAP-12945-P-A, Vol. 3. Bayless, P. D., et al., 1980, "Experiment Data Report for LOFT Nuclear Small Break Experiment L3-1," NUREG CR-1 145/EGG-2007. Condie, K. G., et al., 1981, "Four-Inch Equivalent Break Loss-of-Coolant Experiments: Posttest Analysis of LOFT Experiments L3-1, L3-5 (Pumps Off), and L3-6 (Pumps On)," EGG-LOFT-5480. Czapary, L. S., 1980, "LOFT L3-1 Preliminary Comparison Report," EGG-CAAP-5255. Dao, L. T. L. and Carpenter, J. M., 1980, "Experiment Data Report for LOFT Nuclear Small Break Experiment L3-51L-35A," NUREG CR-1695tEGG-2060. Gillas, D. L. and Carpenter, J. M., 1980, "Experiment Data Report for LOFT Nuclear Small Break Experiment L3-7," NUREG CR-1570tEGG-2049. o:\4384-non\4384-20a.wpd:lb-040403 20-57

I McCreery, G. E., 1980, Quick Look Report on LOFT Nuclear Experiment L3-7, EGG-LOFT-5792. Reeder, D. L., 1978, "LOFT System and Test Description (5.5-ft. Nuclear Core 1 LOCEs)," NUREG CR-02471TREE-1208. o:\4384-non\4354-20a.wpd:lb-040403 20-58

SECTION 21 SIMULATION OF SEMISCALE SMALL BREAK LOCA EXPERIMENTS 21-1 Introduction The Semiscale facility is a small scale (1:1705) replica of a Westinghouse RCS which includes all of the major components. Figure 21-1 shows the layout of the major components in the Mod-2C configuration. There are two loops in the facility, with one scaled as a single loop, and the other scaled as a combined three loops. The facility evolved through several major modifications over the course of approximately a decade of testing and was used for both large and small break experiments. Later modifications to the facility focused on small break LOCA phenomena, and extensive instrumentation was installed to measure key phenomena such as liquid levels and break discharge rates. The simulated reactor vessel houses an electrically heated bundle consisting of 25 heater rods with a total power of 2 MW. The overall scaling philosophy used in designing the facility is the maintenance of the power-to-volume ratio, coupled with a 1:1 elevation scaling criteria (Larson, et al., 1980 and Loomis, 1987). The facility is capable of operating at actual nuclear power plant pressures and temperatures, and therefore, a full range of pressures and fluid states occurs during a transient. 21-2 Key Phenomena Due to the small scale, combined with 1:1 elevation scaling, the corresponding pipe and vessel sizes used to construct the facility are generally characterized as exhibiting 1-D fluid flow behaviors. Therefore, some scaling distortion is expected to be evident in comparisons with larger facilities or full-scale plants. However, the purpose of the experiments is to provide information concerning the overall flow behaviors and qualitative interaction of phenomena that occur throughout the various stages of a small break LOCA in a complete integral RCS. Comparisons of calculated results to the experiments will be focused on general phenomena such as the relative timing of events and the factors which influence fluid distributions within the RCS. One particular phenomenon that the facility can be used to address is the integral effects nature of loop seal clearing. Due to size considerations, Semiscale is argued to exhibit a strong 1-D loop seal clearing behavior with liquid in the piping being moved and expelled in a plug-like fashion. While separate effects facilities address the 3-D phenomena given fixed fluid conditions, they do o:4384non\sec21.wpd:1b-1 12000 21-1

I not provide information on the general aspects of how the loop seals behave over the course of a transient in relation to the fluid distribution in the entire RCS. The accuracy with which the code is able to calculate loop seal formation and clearing when the 3-D aspects are unimportant will be important in establishing its capability to model more complex system interactions, including phase separation and other phenomena for the nuclear power plant (NPP) calculations. Another important phenomenon that influences the severity of small break transients is steam generator tube liquid holdup. This holdup phenomenon was first identified experimentally in a Semiscale small break LOCA experiment (Leonard, 1982); it has since been duplicated in other facilities such as ROSA (Osakabe, et al., 1987) and has been discussed extensively in the open literature (Leonard, 1983 and Loomis, 1985a). Steam generator liquid holdup is the result of liquid being condensed in the upflow side of the tubes relatively early during a small break LOCA transient. This liquid is unable to gravity-drain back through the hot leg because it is impeded by high upward steam flowrates. The large pressure drops induced by this holdup, in turn, affect the hydrostatic head balances throughout the RCS. Whether the pump suction seal clearing phenomenon discussed above results in core uncovery is significantly affected by the amount of liquid holdup. 21-3 Applicable Tests Information from more than 40 small break LOCA tests, which were conducted in the various Semiscale configurations, is available in NUREGICR-4393 and NUREG/CR-4945 (Loomis, 1985b and Loomis, 1987). Because the purpose of the comparisons is to study the more general behaviors during a small break transient, the experiments used for comparison are selected based upon data quality. Given this, the experiments conducted later in the program are better choices because the instrumentation and test procedures at that point were better refined for small breaks. The following are two experiments conducted in the final Semiscale Mod-2C configuration (Loornis and Streit, 1985a): S-LH-1: A 6-inch equivalent cold leg break with downcomer-to-upper head bypass set to 0.9 percent at steady-state S-LH-2: A 6-inch equivalent cold leg break with downcomer-to-upper head bypass set to 3 percent at steady-state o:A4384-non\sec21.wpd:1b-1 12000 21-2

Both of these transients exhibited core uncoveries, which allow investigation of in-bundle mixture level swell and rod heat transfer (Loomis and Streit, 1985b and Shaw and Loomis, 1985). The purpose of varying the vessel upper head bypass was to investigate the influence of this relief path on the core liquid level depressions that occur from the manometric balances that form among the various sections of the RCS: the core/downcomer, pump suction crossover legs, and steam generator tubes. The transients exhibited a notable difference in the amount of core uncovery and rod heatup as a function of upper head bypass flow. The difference was attributed to the timing of upper head drain, which clears a relief path between the upper plenum and the downcomer/cold leg. 21-4 Facility Configuration for Tests S-LH-1 and S-LH-2 In conducting the small pipe break experiments, a tee spool piece is inserted in the loop that is scaled to represent a single NPP loop and is designated as the broken loop. The loop arrangement is shown in Figure 21-1, and a detail of the break spool is shown in Figure 21-2. The break orifice is scaled to represent 5 percent of the cold leg area of a full-size cold leg (27.5-inch inside diameter) and has a diameter of 0.1488 inches, which is equivalent to a 6.37-inch inside diameter break at fll-scale. It is positioned at the horizontal centerline of the cold leg piping, so the break flow will be susceptible to flow regimes in the piping, such as stratification. All of the break effluent is passed through a system of condensing coils and is collected in a catch tank. This arrangement not only allows for accurate measurements of the integral break flow, but also allows for measurement of instantaneous break flow, other than the delay during the initial opening of the break at the start of the experiment. The layout of the vessel upper head area is depicted in Figure 21-3. A bypass line connects the top of the downcomer to the upper head at elevations representative of the reference NPP. A replaceable orifice is inserted in the bypass line and used to adjust the bypass flow ratio to the desired value for the particular experiment. A single tube is scaled to represent the flow area and elevations for the aggregate of all the control rod guide tubes in an NPP. There are also two tubes representing upper internals support columns, which connect the upper head to the upper core plate. These were originally included for experiments modelling the internals configuration of upper head injection plants and were, therefore, plugged off for the experiments to be examined here. The core in the Semiscale facility is composed of 25 electrically heated rods, each of which are geometrically similar to nuclear fuel rods in an NPP with 0.422-inch outside diameter cladding. o:A4384-non\sec21.wpd:lb1A 12000 21-3

                                                                                                  - -- -- - --L_

The rods are capable of operating at the full steady-state power of a PWR nuclear rod. The resistive element windings are sized such that a stepped cosine axial power profile results, as shown in Figure 21-4, with a peak linear power of approximately 11.2 kW/ft at full power. The steam generators are scaled to a full 1:1 elevation matching the reference NPP, and each individual steam generator tube is made from the same tube stock: 0.776-inch inside diameter, 49.5-mil thick. There are a total of six tubes in the intact loop and two tubes in the broken loop steam generator in order to conserve the scaled heat transfer areas. Elevation scaling is preserved 1:1 relative to the reference NPP. The Semiscale hot legs join together the vessel and the steam generator inlet plenums. This maintains the correct elevations relative to one another. The pump suction piping does not have a horizontal run, as in a full-size NPP, because the volume would have become excessive. In general, due to the use of standard piping sizes, the loop volumes are somewhat overscaled relative to the ideal, but this is not a significant distortion relative to the overall volume of the other major components. 21-5 Description of WCOBRAITRAC-SB Model Figure 21-5 shows the component layout of the WCOBRAITRAC-SB model of the Semiscale Mod-2C system. The reactor vessel, the primary loop piping, and the steam generators are [ The break is modelled as proceeding from the middle cell of the broken cold leg, consistent with the break assembly elevation in the test facility. The break model as described in Section 13 of this volume is used as shown in Figure 21-5 with no discharge coefficient or flow area multiplier I o:W384-non\sec21.wpdIb-1 12000 21-4

                                                             ]aC The WCOBRAITRAC-SB model, therefore, represents the Semiscale Mod-2C test facility break geometry without any adjustments.

Due to the small scale of the facility, the general thermal-hydraulic behaviors are assumed to be close to 1-D. In particular, the formation and blowout of the pump suction loop seals tend to behave in a plug-like fashion. Experimental measurements in the horizontal piping, including video probe information, verified that there can be significant stratification even at this scale. Therefore, the modelling of the primary loops reflects the same approach as used on the other integral facilities. Information used to compile the facility geometries, and the like, was obtained largely through the review of an available drawing. Where feasible, some parameters were compared to the facility description and RELAP model document of Leonard (Leonard, 1981). Additionally, personnel at INEL were contacted to reconcile some of the more ambiguous items, operating conditions, and procedures. Figure 21-6 shows the nodalization of the simulated reactor vessel using the VESSEL component. The WCOBRA/IRAC-SB noding used for the Semiscale facility is consistent with the nodalizations of the other integral test facility simulations and the nodalization used in the PWR computations. The number of each section in the vessel is shown, together with the number of cells within each section, in parentheses. Figure 21-6 shows the elevation of the vessel for WCOBRAfIRAC-SB analysis. The section boundary heights are relative to the inside of the bottom of the vessel. Values within squares are channel numbers, and values within circles are gap numbers. WCOBRAfIRAC-SB assumes that a flow path exists between vertically connected channels, unless otherwise specified in the input. Transverse flow between channels in the same section only exists if the channels are specified as connected by gaps. The volume, axial flow area, and wetted perimeter of each channel is specified in the code input. There is also the capability to vary these quantities within a channel if the geometry warrants. As in the facility, the downcomer is a stand-alone pipe with a short annulus region at the top (Sections 7 and 8) where the cold leg piping is connected. The cold legs and hot legs [

                           ]a.c o:\4384-non\sec21.wpd:1b-1 12000                  21-5

I Figures 21-7 and 21-8 show the models of the two steam generators for the intact and broken loops, respectively. Because the steam generator tubes remain mostly covered throughout the transients, the amount of axial detail shown is judged to be adequate for modelling the phenomena of interest. [

                                                                                                 .t,l
                                          ]a,c A set of reference points and four quadrant homologous curves are supplied for the primary recirculation pumps. However, in general, because the pumps are tripped relatively early in the transient, the contribution of the pumps is made during periods in which they are pumping a positive head of single-phase liquid. Because the Semiscale pumps physically do not have a scaled moment of inertia, they are controlled on a powered coastdown curve after the trip. This behavior is replicated in the input to WCOBRA/TRAC-SB.

Similarly, the electrical power to the core heater rods is controlled by a computer program to simulate a normalized decay heat curve, and this is duplicated as shown in Figure 21-10. The pumped safety injection system used in the Semiscale facility uses positive displacement pumps. The pumps are controlled by a computer which uses a pressure measurement from the RCS as input to vary the injection rate as a function of pressure to follow a prescribed curve to simulate the performance of a centrifugal pump. Figure 21-11 shows the injection rates for the intact and broken loop safety injection pumps as a function of pressure, as actually derived by oA4384-non\sec21.wpd:1b-1 12000 21-6

INEL personnel during post-experimental data reduction. The curve for the intact loop pump shows an anomalous behavior over the range of pressures from about 400 to 800 psi. Because this is the best estimate of the actual safety injection pump characteristics, it has been modelled in the WCOBRA/TRAC-SB analysis as shown. 21-6 Steady-State Simulations Steady-state operating conditions are attained in the WCOBRA/TRAC-SB model by running the code with no break in the system until conditions have stabilized. Table 21-1 compares the key parameters from test S-LH-1 to those obtained by the model; Table 21-2 provides the same information for test S-LH-2. All of the parameters are within acceptable tolerances for conducting validation simulations. The vessel upper head bypass flow ratios in the WCOBRA/ TRAC-SB steady-state simulation match the S-LH-1 and S-LH-2 experiments well. Because upper head bypass was the critical parameter varied between the two tests in the LH test series, it is important in these simulations to obtain good agreement of the bypass flow to the experimental value. The steady-state secondary masses do not agree well with the reported values from the experiment. However, this parameter has a large uncertainty associated with it; a review of related Semiscale documentation (Shimeck, 1983) shows that secondary mass has a large range of possible values. Also, in the description of the INEL simulation of the experiments with the RELAP5 code, it is stated that it was necessary to run the model with lower masses to maintain stable operation (Loomis and Streit, 1985b). The amount of secondary inventory in the WCOBRATRAC-SB model allows for substantial coverage of the tube bundles. Therefore, the effect on the transient predictions of any discrepancies in secondary mass (if in fact they exist) are judged to be insignificant. 21-7 Transient Simulations The steady-state model conditions, as described in the previous section, were used to perform simulations of tests S-LH-1 and S-LH-2. This set of 5-percent small break LOCA experiments in the Semiscale MOD-2C facility addressed the sensitivity to vessel upper head bypass. Through review of the calculational results in comparison to the test data, it was determined that the holdup of liquid in the steam generator tubes, and associated phenomena in the hot legs and loop o:\4384-non\sec21 .wpd:Ib1 12000 21-7

seal piping, was a dominant factor with regard to the ability to replicate the experimental results. The following discussion will compare the code predictions to data from tests S-LH-1 and S-LH-2. 21-7-1 S-LH-1 Simulation Results Figure 21-12 shows the pressurizer pressure transient from the WCOBRAJTRAC-SB test S-LH-l calculations (dashed line) compared to test data (solid line). At the opening of the break, the model predicts the pressure to drop to near the hot leg saturation value a bit more rapidly than the experimental data indicate. Once the calculation reaches hot leg saturation (approximately 1700 psi) agreement exists between the data and prediction as primary fluid begins to flash. Once the system depressurizes to cold leg saturation (approximately 1100 psi), the depressurization rate slows appreciably both in the experiment and the calculation. After this point, WCOBRArRAC-SB overpredicts, than underpredicts pressure. Overall, the WCOBRAITRAC-SB calculation predicts the primary pressure well versus the test data all the way through the accumulator actuation. The time of accumulator injection is predicted to occur approximately 20 seconds earlier than in the test. Figure 21-13 overlays the predicted primary pressure, as a solid line, along with the predicted steam generator secondary side pressures as dashed lines, with the intact loop steam generator - exhibiting a higher value than the broken loop value most of the time. Figure 21-14 overlays the broken and intact loop steam generator pressures against the test S-LH-1 data. In general for small break LOCAs, during the initial portion of the transient, the primary pressure hovers above the secondary pressure because the break energy removal is supplemented by continuing heat transfer to the steam generator secondary side fluid. Once the loop seals clear (approximately 190 seconds in the prediction), the break uncovers, the primary depressurizes, and the secondaries became heat sources. After loop seal clearance occurs, the steam generator secondary pressure becomes unimportant. During the early portion of the transient, the secondary pressures predicted by WCOBRA/rRAC-SB are higher than in the experiment. The first stage safety valve setpoint on the steam generator secondaries is set at 1047 psia. In the experiment, neither secondary pressure reached the setpoint, while in the calculation, the broken loop pressure is predicted to reach the setpoint. As discussed in Section 21-6 on the steady-state parameters, the amount of initial secondary inventory is suspected to be low in the calculation relative to the experimental conditions. Due to scaling distortions, the Semiscale steam generators have a large amount of structural mass relative to fluid volume, including some large "fillers" in the intact loop generator. Comparing the behavior of the predicted secondary oA434-non\sec21.wpd:1b-1 12000 21-8

pressures to the experiment, it is speculated that the misprediction is due at least in part to the liquid mass of the steam generator secondary in WCOBRAflRAC-SB being too small relative to the metal mass. Focusing on the broken loop steam generator, once the initial peak has passed, the predicted and experimental pressures agree well until beyond the time of loop seal clearance. Then, heat transfer to the cooler secondary side metal causes the pressure in WCOBRAITRAC to fall below the data. Toward the end of the transient, reverse heat transfer from the metal to the secondary side liquid holds up the predicted pressure versus the data. Figure 21-15 compares the measured (solid line) and calculated break mass flowrates. The experimental data, obtained from a catch tank system, have good accuracy for integral flow but do not always reflect sudden flowrate changes. For instance, at the opening of the break, the measured break flow may lag somewhat because of the transit time and buffering effects of the condensing coil system. In general, it is seen that the WCOBRAITRAC-SB calculation is in good agreement with the experimental measurement during the first 30 seconds of the transient. For the next 120 seconds of the transient, the predicted break flow is somewhat low (on the average by about 20 percent) as its trend follows the data. This leads to a collapsed liquid level prediction above the data in this time interval in Figure 21-16. At the time of loop seal clearance in the calculation, the predicted break flow drops suddenly. The same phenomenon occurs in the test data, but 20 seconds earlier. Following loop seal clearance, WCOBRAITRAC-SB predicts a significantly larger break flow than the data for 60 - 70 seconds. Figure 21-16 compares the collapsed liquid level predicted in the core region to the test data as provided by Loomis and Streit (Loomis and Streit, 1985b) and shown as the solid line. The notable differences between the data and the calculation are that the calculation predicts a depression and recovery of the level in the core approximately 20 seconds late during loop seal clearance and then predicts a collapsed liquid level that is approximately 2.5 ft low in the ensuing period until accumulator actuation. In the test data, the core level depression observed to bottom out at approximately 170 seconds is relieved by the intact loop pump suction seal blowout. In WCOBRAfRAC-SB a very similar loop seal behavior is predicted. This indicates that there is a similar liquid holdup in the steam generator tubes in the prediction as in the test. Later on in the transient, the predicted core level is well below the measured value because WCOBRAfTRAC-SB predicts too high a break flow post-loop seal clearing. Figures 21-17 and 21-18 show the calculated void fractions in the top two nodes in the pump suction piping for the intact and broken loops. The solid and dashed lines in the figures are the top and middle nodes, respectively. In test S-LH-1, once the intact loop seal cleared, the pressure o:\4384-non\sec21.wpd:1b-I 12000 21-9

I relief temporarily removed the driving heads for clearing the broken loop seal, which eventually blew out at approximately 270 seconds. In the simulation, as in the test, the flow through the cleared intact loop is inadequate to prevent the broken loop from clearing. After the initial clearance of each loop seal, the loop seal refills partially. The broken loop loop seal middle node does refill completely for a short time and then reclears later on. The upflow sides of each of the pump suction legs were swept out gradually, which is in good agreement with the type of behavior observed in all Semiscale small break LOCA experiments. Figure 21-19 compares the core heater rod temperature response of the lead rod in the test S-LH-1 data with the peak temperature predicted by WCOBRAITRAC-SB indicated by the dashed curve. WCOBRA/TRAC does not predict a modest heatup of the heater rods during the loop seal clearance even though the severity of the core depression is predicted well. This indicates that the two-phase mixture level is overpredicted by IYCOBRA/rRAC-SB. The code-predicted heatup above saturation temperature during the core boiloff portion of the transient is approximately 540'F, versus approximately 400°F in the test. As is evident in Figure 21-19, the predicted heater rod temperature excursion is of longer duration than in the test. This is a consequence of the code underprediction of collapsed level. Figure 21-19 also indicates that there may be another overprediction of nixture level swell in the Semiscale heated rod bundle; the collapsed level at which the core boiloff excursion in clad temperature begins is about 0.2 feet lower than in the experiment. The heat transfer prediction in the uncovered core situation, as shown in Section 12 of this document, is close to the data. The WCOBRA/TRAC-SB underprediction of collapsed liquid level and overprediction of level swell in the core region compensate somewhat but produce a PCT value above the data. Figure 21-20 provides a comparison of the integrated break mass flow between experiment and prediction. For the transient overall, the agreement is within 1 percent; the low break flow that occurs in WCOBRAJIRAC-SB at low subcoolinglsaturated liquid conditions before the clearance of the intact loop loop seal at 190 seconds, defines the largest point of departure in the WCOBRA/TRAC prediction from the data. The excessive break flow predicted by WCOBRA/TRAC-SB for two-phase flow after loop seal clearance brings the code's integrated total back to the data, and by the time the pressure decreases to the accumulator setpoint (630 psi), the integrated break flows are a close match. The error in overpredicting the two-phase flow compensates for the previous underprediction of the break flow. o:\4384-non\sec21 .wpd:1b- 112000 21-10

21-7-2 S-LH-2 Simulation Results Figure 21-21 shows that the integrated break mass flow comparison for the test S-LH-2 prediction and data follows the same trend as the test S-LH-1 result. Figure 21-22 shows the pressurizer pressure transient from the WCOBRATRAC-SB test S-LH-2 calculations (dashed line) compared to test data (solid line). As was the case for S-LH-1 at the opening of the break, the model over-predicts the pressure drop to near the hot leg saturation value. The initial depressurization of the system is totally dependent upon the draining of the pressurizer, which is restricted by the surge line to the hot leg. Once the calculation reaches hot leg saturation (approximately 1700 psi), agreement exists between the data and prediction as primary fluid begins to flash. Once the system depressurizes to cold leg saturation (approximately 1100 psi), the depressurization rate slows appreciably both in the experiment and the calculation. Then, the WCOBRArRAC-SB calculation over-predicts the primary pressure versus the test data for a time, but the predicted value decreases so that the predicted accumulator actuation time is approximately on the mark. Figure 21-23 overlays the predicted primary pressure (solid line) along with the predicted steam generator secondary side pressures (dashed lines) with the intact loop steam generator - exhibiting a higher value than the broken loop unit most of the time. Figure 21-24 overlays the broken and intact loop steam generator pressures against the test S-LH-2 data. In general for small break LOCAs, during the initial portion of the transient, the primary pressure hovers above the secondary pressure because the break energy removal is supplemented by continuing heat transfer to the secondaries. Once the intact loop loop seal clears (approximately 200 seconds), the break uncovers and the primary side can depressurize below the secondaries. During the early portion of the transient, the secondary pressures are predicted to be higher than in the experiment. The first stage safety setpoint on the steam generator secondaries is set at 1047 psi. Contrary to the experiment, where neither secondary reached the setpoint, WCOBRAfIRAC-SB predicts the broken loop steam generator secondary pressure to reach the setpoint for a few seconds. As discussed in Section 21-7-1, the amount of initial secondary inventory is suspected to be low in the calculation relative to the experimental conditions and to the steam generator metal mass. The WCOBRATRAC predicted steam generator secondary pressures exceed the test values during the transient until after loop seal clearance occurs. Figure 21-25 compares the measured (solid line) and calculated break mass flowrates. The experimental data, obtained from a catch tank system, have good accuracy for integral flow, but do not always reflect sudden flowrate changes. In general, it is seen that the o:4384-non\sec21 .wpd:Ib-1 12000 21-1 1

I WCOBRArIRAC-SB calculation is in good agreement with the experimental measurement during the first 25 seconds of the transient. For the next 90 seconds of the transient, the predicted break flow is somewhat low (about 35-40 percent) in the low subcooling/saturated liquid region. Clearing of the intact loop seal in the prediction (within a few seconds of the time that it cleared in the experiment) affects the break mass flow comparison thereafter, in the same way as in S-LH-1. The overprediction of two-phase break flow compensate for the previous underprediction of the break flow. Figure 21-26 compares the collapsed liquid level predicted in the core region to the test data. The notable difference between the data and the calculation is the greater decrease in the predicted level after 350 seconds, before which the agreement is good. The calculation predicts a depression and recovery of the level in the core, and, consistent with the data, the pump suction seal depression and blowout causes no core uncovery. Not until the longer term boiloff begins at approximately 300 seconds into the experiment, causing a depletion of inventory, does any core heatup occur in test S-LH-2. The S-LH-2 prediction shows an increase in core collapsed level in the time interval between 485-510 seconds. This is caused by the sudden draining of liquid that has been held up in the intact loop hot leg back into the reactor vessel upper plenum. Once this draining is complete, the core region collapsed liquid level of WCOBRAfIRAC-SB agrees very < well with the test value. The error in predicting the core collapsed level is compensated for by the draining of liquid that WCOBRA/TRAC-SB has held up in the hot leg due to its overprediction of CCFL. Figure 21-27 shows the predicted void fractions in the intact loop pump suction piping middle and top nodes. Figure 21-28 provides void fractions in nodes in the pump suction piping for the broken loop as predicted by WCOBRA/TRAC-SB. The solid and dashed lines are the top and middle nodes, respectively, in these two figures. The WCOBRAiTRAC-SB model predicts no clearing of the broken loop loop seal, which is consistent with the test result. The steam relief flow path through the vessel upper head is sufficient, together with the cleared intact loop, to vent steam to the break. Figure 21-27 shows the middle node in the intact loop pipe tends to replug after clearing. Figure 21-29 compares the core heater rod temperature response of the lead rod in the test S-LH-2 data with the peak temperature predicted by WCOBRATRAC-SB. In contrast to the test, WCOBRAITRAC-SB predicts two periods of heatup of the heater rods during the transient. The mass addition from draining of the intact hot leg causes the code-predicted initial heatup l above saturation temperature during the core boiloff portion of the transient to terminate. In o:\4384-non\sec2I.wpd:Ib-I 12000 21-12

Figure 21-29, the second predicted heater rod temperature excursion begins at about the same time as the test excursion, but at a collapsed liquid level about 0.4 lower than in the test. This behavior is consistent with that previously noted in the test S-LH-1 discussion. 21-8 Conclusions The S-LH-1 and S-LH-2 experiments of the Semiscale Mod-2C configuration have been simulated with the WCOBRAIRAC-SB model using the boundary conditions from these small break LOCA experiments. A comparison of calculated steady-state conditions from the model to experimental data is generally in good agreement, and all parameters were within acceptable tolerances for performing transient simulations to ascertain the general ability of the code to predict the major thermal-hydraulic phenomena. A review of the calculation results indicates that the code is generally doing a good job of predicting the key small break LOCA parameters. Notably, the expected top-down drain of the system and the formation of quasi-equilibrium hydrostatic balances associated with liquid inventories in the vertical components, particularly the core/downcomer and pump suction loop piping, are reasonable. Key transient parameters of depressurization rate and break mass discharge are predicted adequately. The predicted heater rod temperature excursions differ somewhat from the data; the lower collapsed liquid level in the core region in the S-LH-1 simulation leads to a higher PCT than was observed in the test, while a delayed hot leg draining leads to a dual temperature excursion in the test S-LH-2 simulation. Consistent with the findings reported in Chapter 13, in the Semiscale simulations WCOBRA/TRAC-SB underpredicts the critical mass flux for saturated liquid, then overpredicts critical flow for two-phase conditions. 21-9 References Larson, T. K., et al., 1980, "Scaling Criteria and an Assessment of Semiscale Mod-3 Scaling for Small Break Loss Of Coolant Transients," EGG-SEMI-5121. Leonard, M. T., 1981, "RELAP5 Standard Model Description for the Semiscale Moda System," EGG-SEMI-5692. Leonard, M. T., 1982, "Vessel Coolant Mass Depletion During a Small Break LOCA," EGG-SEMI-6010. o:\4384-non\sec2.wpd:Ib-1 12000 21-13

Leonard, M. T., 1983, "An Analytical Study of a Small Break Loss-Of-Coolant Accident with Upper Head Injection," Nuclear Technology, Vol 62. Loomis, G. G., 1985a, "Semiscale Liquid Holdup Investigations: A Comparison of Small Break LOCA Tests Performed in the Semiscale Moda and Mod-2C Facilities," 13th Annual Water Reactor Safety Research Information Meeting, Gaithersburg, MD. Loomis, G. G., 1985b, "Summary of Semiscale Small Break Loss-of-Coolant Accident Experiments (1979 to 1985)," NUREG/CR-4393 (EGG-2419). Loomis, G. G., 1987, "Summary of the Semiscale Program (1965-1986)," NUREG/CR-4945 (EGG-2509). Loomis, G. G. and Streit, J. E., 1985a, "Quick Look Report for Semiscale Mod-2C Experiments S-LH-1 and S-LH-2," EGG-SEMI-6884. Loomis, G. G. and Streit, J. E., 1985b, "Results of Semiscale Mod-2C 5% Small Break LOCA Experiments S-LH-1 and 2," NUREG/CR-4438 (EGG-2424). Osakabe, M., et al., 1987, "Core Liquid Level Depression Due to Manometric Effect During PWR Small Break LOCA," J. of Nuclear Science and Technology 24(2), pp. 103-110. Shaw, R. and Loomis, G. G., 1985, "Vessel Coolant Mass Depletion During a 5% Small Break LOCA in the Semiscale Mod-2C Facility," Specialists Meeting on Small Break LOCA Analysis in LWRs, 2, Piza, Italy, pp. 159-175. Shimeck, D. J., 1983, "Analysis of Semiscale Moda System UHI/SBLOCA Experiments," NUREG/CR-3195. o:\4384-non\sec21 .wpd:1b-1 12000 21-14

Table 21-1 Comparison of Steady-State Conditions: Test S-LH-1 WCOBRAMTRAC Parameter Measured Model Pressurizer pressure (psia) 2244 2214 Core power (kW) 2019 2019 Cold leg temperature (F) Intact loop 552.1 550 Broken loop 555.6 554.3 Core AT (°F) 67.8 66 Primary flowrates (lbm/s) Intact loop 15.7 16.0 Broken loop 5.2 5.2 Upper head temperature (F) 545 543 Upper head bypass (%) 0.90 0.957 Primary leakage rate (lbm/s) 0.004 0 Steam generator secondary pressures (psia) Intact loop 830 830 Broken loop 882 881 Steam generator secondary masses (lbm) Intact loop 421 269 Broken loop 95 84 o:\4384-non\scc21.wpd:1b-1 12000 21-15

Table 21-2 Comparison of Steady-State Conditions: Test S-LH-2 WCOBRAITRAC Parameter Measured Model Pressurizer pressure (psia) 2237 2204 Core power (kW) 2019 2019 Cold leg temperature (F) Intact loop 552 549.4 Broken loop 556.2 553.2 Core AT (F) 66.9 67 Primary flowrates (lbm/s) Intact loop 16.2 16.1 Broken loop 4.4 5.2 Upper head temperature (F) 546 547 Upper head bypass (%) 3.0 3.1 Primary leakage rate (lbm/s) 0.004 0 Steam generator secondary pressures (psia) Intact loop 827 827 Broken loop 864 865 Steam generator secondary masses (bm) Intact loop 421 269 Broken loop 106 89 o:4384nonsec21.wpd:1 b 12000 21-16

Broken loop atmospheric dump valve (ADV) Recirculaton Intact loop main steam Isolatilon valve (MSIV) Type I steam II Intact loop generation Type It steam downcomer - generator Pressurizer Pressure vessel - BROKEN LOOP INTACT LOOP Figure 21-1. Semiscale Mod-2C System o:\4384-non\sec21.wpd:1b-1 12000 21-17

Flow from F pump

                                                                                *Optical Flow to         ~~7 break flow condensing system               Filter ring flow disperser.

vessel Flow All dimensions are in mm Figure 21-2. Cold Leg Break Assembly o:\4384-non\sc21 .wpd:l b-l 12000 21-18

Upper head Bypass line -_ Support columns Hot leg (2) owncomer 0

                                                                   - Top of core Figure 21-3. Vessel Upper Head Configuration o:\4394-non\sec21.wpd:b-112000                     21-19

I 1.6 1.4 1.2 _1.21 2.43 o*1.0 0.89 274 0.6 0.91 Q061 3~~~~03 0.4 0 0.30 0.2 0 2 3 Oistance from bottom of heated length Im) Figure 21-4. Core Heater Rod Axial Power Profile o:4384-non\sec2I.wpd:Ib-1 12000 21-20

ax Figure 21-5. WCOBRAITRAC Model of Semiscale Mod-2C Component Layout oA4384-non\sc21.wpd:Ib-1 12000 21-21

a,c Figure 21-6. WCOBRAITRAC Model of Semiscale Reactor Vessel oA43S4-non\sec21.wpd:b-1 12000 21-22

ac Figure 21-7. WCOBRAJTRAC Model of Semiscale Intact Loop Steam Generator o\4384-non\sec21.wpd:1b-1 12000 21-23

I Figure 21-8. WCOBRAITRAC Model of Semiscale Broken Loop Steam Generator oA4384-non\scc2.wpd:1b-1 12000 21-24

Figure 21-9. Semiscale Reactor Coolant Loop Noding o:W384-non\sec21.vpd:Ib-1 12000 21-25

I 2500 2 000 1500 a) 1000 3 0 0-II I l -I I-0 0 . 20 40 60 80 100 Time (s) Figure 21-10. Core Power Versus Time Curve o:\4384-non\sec21.wpd:lb-1 12000 21-26

IL HP I S

          - --      -   BL     HP I S 35E-O 3          I -

U, E

   -o
            *251>

r-a 0 o C 0 500 1000 1500 I0 Figure 21-11. Safety Injection Rates as a Function of RCS Pressure o:\4384-r non\sec2l-kwnd:Ib4112000 21-27

MTHOO006 10 9 0 0 PPRZ+632

       ----          TMTHOOO 1 3     35         7       0 PRESSURE 2500 - _

2000 - -- - .. 0

   ,c 1500-         .. .

a-CD cn 1000 - a) 0-500 - .. 0 I I Time (s) Figure 21-12. Pressurizer Pressure, Test S-LH-1 o:4384-nons21-A.wpd:tb-112000 21-28

P 24 2 O PRESSURE

        -   - -  -    p            60       2      O PRESSURE

_ _ __ __ - p 76 2 O PRESSURE 2000- - 1800-

   ,   1600-     -

Mo1400-a, 1200- - co 1000 - j 800-600 400-Time (s) Figure 21-13. Primary and Secondary Predicted Pressures, Test S-LH-1 o4384-non\sec21 -Awpdlb-12000 21-29

I MTHOO01 7 1 19 0 0 PBG+1 186 76 2 0 PRESSURE 60 2 0 PRESSURE

        ---          MTHOO015                 1 10               0        0  P I S+     17 1U3U 1000 950
   .Fn 0-   900
                                   <-t~~~~~~~~~~~~~~~~~~~~~~~                            - -:: -----

850 cn 800 a) . . . . ... . . . . . . . ..- I-0l- 750 700 ... ... . ...... . ..... .. ~~~~~~~~~~~~~~~ 650 0 200 400 600 800 1GO Time (s)

                                             - -          DDtas IL SC
                                             ........ WCOBRA/TRACS          IL SC Data' BL SO
                                             - - --       WCOBRA/TRAC'      DL SG Figure 21-14. Broken and Intact Loop Secondary Pressure Predictions, Test S-LH-1 o:\4384-non\s=c2I-Awpd:lb-1 12000                                21-30

MTHOO020 13 0 0 MDOT*BREAK

         - - - -       MTHOO019               5       2      0 MASS FLOWRATE 3 -3.

2.5 - . .. E

   -D Ci) 1.5      0        .        .

0 U, 0

           .5 50 'ti         L--         -

Time (s) Figure 21-15. Break Mass Flowrates, Test S-LH-1 o:\4384-non%sec21-A.wpd:lb-1 12000 21-31

I p 76 2 0 PRESSURE

        - - - -LO-LEVEL                          0       0 COLLAPSED       LIQ. LEVEL 20

_-/ 15 Q> J a,

    ':  10 crI C) 0 Figure 21-16. Core Collapsed Liquid Levels, Test S-LH-1 o:W384-non\sec21-kwpd:Ib-1 12000                 21-32

AL 96 4 O VAPOR FRACTION AL 96 3 0 VAPOR FRACTION 1 _

        .8 - -

0= 0 .6 - ..... 70

        .4 -     .- . .
        .2-      .

I Ii 0* Time (s) Figure 21-17. Calculated Void Fractions in the Intact Loop Pump Suction Piping, Test S-LH-1 o:43S4-non\sec2l-Avwpd:lb-112000 21-33

I

         -     ~       AL               1 13  4       0 VAPOR   FRACTION
           - --        AL               1 13  3       0 VAPOR   FRACTION
          .8 .....

0

          .6 -    .*

I C-) I 0

   . _   .4-   .
          .2- -      . 1' III 0            1 Figure 21-18. Calculated Void Fractions in the Broken Loop Pump Suction Piping, Test S-LH-1 o:\4384-non\s=c21-A.wpd:lb-1 12000            21-34

MTHOO036 26 3 0 0 THV*C2+231 TCLAD 60 715 ELEV. 8.96 FT. 1100 - 1000 ------- U- 900 - - -- - - - - C.) 800- ........ a) 700 - . .. ..... Q-600 .. ......... 500 .... 400- I 400 Time (s) Figure 21-19. Core Heater Rod Temperature Response, Test S-LH-1 o:\4384-non\sec21 -A.wpd: b-1 12000 21-35

I MTHOO022 13 0 0 MDOT*BREAK

                      -- - O 2 1 MT               5       2      0  MASS  FLOWRATE 400 300 E
   -C 200 CI)

U)

ER 100 0

Time (s) Figure 21-20. Integrated Break Mass Flow Comparison, Test S-LH-1 o:.4384-non\sec21-Awpd:1b-1 12000 21-36

MTHOO03 1 12 0 0 MDOT*BREAK

      ---      -     MTHOO030        5       2       0 MASS FLOWRATE 350 -

300 - 250 -

 -5%  200 -
 -o co U3  150 -             .     .

C> 0 100 - . 50 .... 0o Time (s) Figure 21-21. Integrated Break Mass Flow Comparison, Test S-LH-2 oA\4384\sec2l-B.wpd:lb- 12000 21-37

I UTHOOO IS 1 22 0 0 P*PRZ+632 UTHOO006 35 7 0 PRESSURE 2500 2000 C) g 1500 0 Dj U) 1000 co 500 0 Time (s) Figure 21-22. Pressurizer Pressure, Test S-LH-2 o\384scc21-B.wpd:lb-1 12000 21-38

p 24 2 O PRESSURE

        - - - - p                                  60            2               O PRESSURE
            -- ---    p                            76            2               O PRESSURE 2000 1800          .... ........                                            ......................                     '..

c' 1600

.2_

en

  -    1400           ~~~~.               . . ....                            .

1 1200-cn 1000 a) CL 800

                   ... . .. . . . . ...        .          .                        z   --    - - -   -  - - -    - -  -

600 l~ ~~~ ~ ~ ~ ~~~~ . . . . . . . . . . . . . . ., . 400 - 0 200 400 600 800 Tim e (s) Figure 21-23. Primary and Secondary Predicted Pressures, Test S-LH-2 o:\4384\sec21-B.wpd:lb-1 12000 21-39

MTHOO01 9 1 33 0 0 PBG+1 186

           -_--         P           76          2        0  PRESSURE 60          2        0  PRESSURE

__- TH00025 12 3 0 0 PIS+11 1 7 105 - 1000 - C' 950 - I C~- I-, 900-850 - an v) 800 - Q 750 - 700 - 4. - 0 200 400 600 800 1000 Time (s)

                                     ---        Date  IL SC
                                     .-__.,._   WCOBRA/TRAC#   IL SC Dotal  L SC
                                     -  -  - -  WCOBRA/TRACe   Bt SC Figure 21-24. Broken and Intact Loop Secondary Pressure Predictions, Test S-LH-2 o:\4384\sec2-B.wpd:l b- 12000                   21-40

       -~              MTHOO029         12       0      0 MDOT*BREAK MTHOO028           5      2      0 MASS FLOWRATE 3-CI) 2.5 -                 .
  -E 2-         ......

2 1.5 - --... - I _II *I cn 0 .5 - - I-Time (s) Figure 21-25. Break Mass Flowrates, Test S-LH-2 o:A43&4se21-B.wpd:1b-I 12000 2141

          - ---        LO-LEVEL                   v 20 I-15 a)

J

  '3=3   10 C)

C0 I 5 CL, 0 Time (s) Figure 21-26. Core Collapsed Liquid Levels, Test S-LH-2 o:\4384\sec21-B.wpd: lb-I 12000 2142

AL 96 4 0 VAPOR FRACTION

         - -     -   AL                 96  3        0 VAPOR     FRACTION II t=     .4    .   .   .
  --r-   .4-........
                     'I It 0-400 Time (s)

Figure 21-27. Calculated Void Fractions in the Intact Loop Pump Suction Piping, Test S-LH-2 o:W384\s21-B.wvpd:1b1 12000 2143 S

AL 113 4 0 VAPOR FRACTION

          - - - - AL                    113        3             0 VAPOR           FRACTION 1-_            -

8 ...... ... . . . . ... . . . . . . . . .. . . . . . . . . . . . ... .. ... .*4 0 C-) .6- .... . 0 I - 70

          .4-         g W.:- ....                    ............                              .
          .2       X       I" 0*             -

0 200 400 600 800 Time (s) Figure 21-28. Calculated Void Fractions in the Broken Loop Pump Suction Piping, Test S-LH-2 o:\4384\sec21-B.wpd:1b112000 2144

i MTHOO041 2 51 0 0 THV*C5+254

          - --       T CL A D        1     59      715  ELEV. 8.81    FT.

800 -_ 750 . 700 - .. .. Icv 650 - - . . . c-v 600 -  : cv 550 - . Q 500 450 400 600 Time (s) Figure 21-29. Core Heater Rod Temperature Response, Test S-LH-2 o:\384\se21 -B.vTd:1b112000 21A45

                                    .,LI oaN4384\sec21 -B.wpd:lb- 12000 21-46

SECTION 22 NUCLEAR ROD AND COMPONENT MODEL ASSESSMENT 22-1 Nuclear Fuel Rod Model 22-1-1 Introduction The fuel rod model in WCOBRAfLRAC is used to predict the following quantities: a) Fuel Initial Stored Energy During normal operation, the fuel average temperature in the high power rod is approximately 2000°F, controlled primarily by the relatively low conductivity of the fuel and the thermal resistance of the gap between the fuel and the cladding. During the LOCA, this stored energy is conducted to the cladding and is the primary contributor to the degree to which the cladding heats up early in the LOCA transient. b) Fuel Rod Thermal Conduction During the LOCA, residual power and stored energy accumulates in or is removed from the fuel. The change in fuel and cladding temperature is controlled primarily by the fuel conductivity and specific heat and by the changing thermal resistance in the fuel-clad gap. c) Cladding Swelling and Burst During the LOCA, internal pressure and high cladding temperature may cause the cladding to deform. This deformation is controlled primarily by the predicted cladding temperature and the burst and strain rate models used. d) Cladding Rewet Rewet may occur during the LOCA. This process may involve the rewetting of large areas of cladding surface as a result of the cladding cooling to temperatures o:4384-non\4384-22.wpd:lb-04043 22-1

I below the minimum film boiling temperature, or the slower quenching process during reflood resulting from axial conduction in the fuel rod. e) Cladding Reaction Reaction between the zirconium in the cladding and the steam environment during the LOCA deposits additional energy into the cladding which must be removed. The reaction rate is controlled primarily by the cladding temperature and by the cladding surface area presented to the steam. f) Residual Fission and Decay Heat Throughout the LOCA, additional energy is deposited into the fuel. The rate of energy generation is controlled early in the transient by the rate of void generation in the core, which is the primary cause of shutdown of the fission process during the LOCA, and later in the transient by the assumed composition of the fuel as fission product decay continues. The models used in WCOBRA/TRAC-SB to predict the above quantities are described in Volume 1, Sections 7 and 8, of this document. In the following sections, the ability of WCOBRAlIRAC-SB to predict the above quantities is assessed. Uncertainties in these models are considered in Section 25 of WCAP-12945-P-A (Bajorek, et al., 1998). 22-1-2 Fuel Rod Model Assessment The following paragraphs discuss the fuel rod model: a) Fuel Initial Stored Energy The fuel temperature during normal operation depends, in complex ways, on fuel pellet condition and its properties. One of the most complex processes that affects this parameter is the relocation of the fuel pellet within the cladding and the resulting asymmetric fuel-clad gap. Complex fuel rod computer models have been developed to predict the fuel temperature as a function of power and burnup. For Westinghouse fuel, the fuel performance code PAD (Weiner, et al., 1985) is used. Incorporating all the detailed models necessary into WCOBRAflRAC is not o:\4384-non\4384-22Zwpd:lb-04043 22-2

practical. The predicted WCOBRAIIRAC fuel temperature, during normal operation, is compared with values predicted from the PAD code for the fuel design being considered. Agreement between PAD and WCOBRAfIRAC predicted fuel temperature is obtained by adjusting the fuel-cladding gap width, which is an input quantity in WCOBRAITRAC. In predicting the LOFT experiments, as described in Section 14-1 of WCAP-12945-P-A (Bajorek, et al., 1998a), the initial fuel temperature for the LOFT core was obtained by using the PAD information for fuel of similar design and composition and using the same approach as in the PWR. No further adjustments were made. This, therefore, amounts to a "blind" prediction of the initial fuel temperature for LOFT. The reasonable agreement between the predicted and measured cladding temperatures for LOFT is evidence that the method for predicting this quantity in WCOBRAJTRAC is also reasonable. b) Fuel Rod Thermal Conduction Two experiments were used to assess the ability of WCOBRAfrRAC to predict the transient temperature of a fuel rod. The LOFT tests include comparisons with cladding and fuel temperatures. The composition of the gas in the LOFT fuel rods is similar to that of a PWR fuel rod; consequently, these comparisons provide evidence that the gap conductivity model is working properly. The LOFT fuel rods were not highly pressurized. Therefore, the cladding did not deform significantly. Two National Research Universal (NRU) reflood tests were simulated to examine the gap conductance predictive capability of the code with deformed cladding. NRU test MT-3.06 used pressurized nuclear fuel rods, while test PTH-1 10 used unpressurized rods. These simulations are described in Sections 22-1-3 through 22-1-7 of WCAP-12945-P-A (Bajorek, et al., 1998). c) Cladding Swelling and Burst NRU test MT-3.06 is used to assess the models in WCOBRAIRAC for cladding swell and burst. This test and the WCOBRA1 TRAC simulation are described in Sections 22-1-3 through 22-1-6 of WCAP-12945-P-A. o:\4384-non\4384-22.wpd:lb-04043 22-3

d) Cladding Rewet The cladding rewet model in WCOBRA/TRAC consists of two parts: the heat transfer coefficient model, described in Section 6-2-6 of this document, and the axial conduction model, described in Section 7-3-1 of this document. These combined models are extensively assessed in WCAP-12945-P-A, Volume II (Bajorek, et al., 1998b), and in the integral tests in Section 14 (Bajorek, et al., 1998a). e) Cladding Reaction The cladding reaction model in WCOBRA/TRAC is described in Section 7-5 of this document. Few tests other than those laboratory experiments used to derive the reaction rate formula are available for verification of this quantity. A few tests early in the FLECHT program were performed with zircaloy cladding. However, the contribution to cladding heatup resulting from the reaction could not be identified. The NRU tests did not reach cladding temperature high enough to cause the reaction rate to become significant. Therefore, the reaction rate equations are based directly on the oxidation test results reported in ORNLINUREG-17 (Cathcart, et al., 1977) and WCAP-12610 (Burman, 1990). f) Residual Fission and Decay Heat The reactor kinetics and decay heat model are described in Section 8 of this document. This model was used in a manner analogous to the approach in the PWR, using values of the moderator temperature coefficient and initial boron concentration existing in the LOFT facility at the time of the experiment. The resulting predicted core power is compared with measured values in Section 14-1 of WCAP-12945-P-A and shows that the model adequately predicts residual power. 22-1-3 NRU Test Description The NRU reactor at the Chalk River Nuclear Laboratories in Canada was used to conduct a series of experimental tests to investigate the therrmal-hydraulic and mechanical deformation behavior of nuclear rods in a LOCA. The test bundles were made of full-length, 3-percent enriched, 17x17 fuel rods powered by low level nuclear fission to simulate decay heat. Two NRU tests o:4384-non\4384-22 wpd:Ib44043 22-4

were modelled and simulated with WCOBRA/TRAC: test PTH-110, which was primarily a thermal-hydraulic test, and test MT-3.06, which was a mechanical deformation test. The PTH-1 10 test is described in NUREG/CR-1882 (Mohr, et al., 1981) and the MT-3.06 test in NUREG/CR-2528 (Mohr, et al., 1983). 22-14 NRU Test Bundle Description The NRU test train was approximately 30 feet long and consisted of six major sections: the inlet region, the test bundle, the shroud, the outlet region, the hanger, and the closure head. The entire test train was inside a pressure tube inside the NRU reactor. The closure region provided the pressure boundary between the test train and the NRU pressure tube. Figure 22-1-1 shows a configuration of the NRU test train and the NRU pressure tube. The hanger tube suspended the test bundle and shroud from the closure head. The 14-foot stainless steel shroud, constructed from two halves clamped together at 7-inch intervals, supported the test bundle. The PTH-1 10 fuel bundle consisted of a 6x6 segment of a 17x17 PWR assembly with the four corner rods removed for easier insertion in the shroud. Figure 22-1-2 shows a cross section of the PTH-1 10 test section. The outer ring of 16 rods plus the corner rods of the next inner ring served as guard rod heaters during the tests. The central 11 rods with the instrument thimble (inside the dotted line in Figure 22-1-2), arranged in a cruciform pattern, were the test rods of interest. The nuclear fuel rods had a cladding outside diameter of 0.379 inches and a pellet diameter of 0.325 inches. The rod-to-rod pitch was 0.502 inches, and the chopped cosine power profile had a peak power of 0.55 kW/ft at the 6-foot elevation. None of the rods in test PTH-1 10 were prepressurized. The MT-3.06 fuel bundle also consisted of a 6x6 segment of a 17x17 assembly with the four corner rods removed. In MT-3.06, however, the instrument tube was replaced by a fuel rod. Thus, the cross section of the MT-3.06 test section is the same as that shown in Figure 22-1-2 with a fuel rod in place of the instrument tube. In addition, the 12 central rods in the MT-3.06 bundle (inside the dotted line in Figure 22-1-2) were pressurized to 550 psia. All 12 of these rods ruptured during test MT-3.06. Horizontal movement and/or bowing was restricted by seven typical PWR grid spacers at 21-inch intervals, starting at the beginning of the heated length. The shroud was insulated on the outer surface to reduce the amount of heat loss to the environment. o:\4384-non\4384-22.wpd:lb-04043 22-5

The experimental test conditions were obtained in two steps: a steady-state phase and a transient phase. During the steady-state phase, the rod power was slowly increased to the desired value for the particular test while the dry steam coolant flowrate was decreased to produce a peak cladding temperature of 800°F. The steady-state conditions were maintained at these values until the thermocouple readings stabilized. The transient phase was then initiated. The steam coolant flow was stopped as quickly as possible; then reflood was started at the desired flowrate. The time period between steam shutoff and reflood initiation was an adiabatic heatup period, which continued until the specified maximum peak cladding temperature was reached. 22-1-5 WCOBRA/TRAC Model of NRU The NRU rod bundle is modelled using the VESSEL component. Boundary conditions are applied to the top and bottom VESSEL cells. The PIPE and zero-velocity FILL components are attached to the VESSEL solely to fulfill the requirement that the model contain at least one one-dimensional component. Figure 22-1-3 shows the WCOBRAITRAC noding diagram of NRU which was used for the MOD7A analysis. The VESSEL component is composed of four channels. Two channels are used to model the rod bundle section: channel 2 for the inner region of the bundle, with a cross-sectional flow area of 1.67 square inches, and channel 3 for the outer region, which includes the shroud wall with a cross-sectional flow area of 3.912 square inches. Channels 1 and 4 are used to represent entrance and exit regions, respectively. The axial lengths are shown in Figure 22-1-3 and are typically 10.5 inches. For the simulation of test MT-3.06, the inner region modelled by channel 2 contains 11 test rods and an instrument tube. The test rods are represented by rod 1. For the simulation of test PTH-1 0, rod 1 represents the 12 interior test rods. The 20 guard rods in the outer region are modelled by rod 2 for both test simulations. Both rods are given the WCOBRAITRAC default material properties of U0 2 fuel and Zircaloy-4 cladding. The shroud wall is modelled as an unpowered conductor with a tube geometry connected to channel 3. The initial temperatures for the rods and the shroud were determined from the thermocouple measurements for each test. The cold gap size between the fuel pellet and the cladding was based on the initial undeformed dimensions. In the initialization and during the transient simulations, the WCOBRAITRAC dynamic gap conductance model was used to predict the effective heat transfer coefficient across the fuel pellet-clad gap. o:\4384-non\4384-22.wpd:lb-04043 22-6

22-1-6 Simulation of NRU Test MT-3.06 Test MT-3.06 was simulated for the first 310 seconds of the transient using WCOBRA/TRAC-MOD7A. The flowrate into the bundle and the fluid temperature are shown in Figure 22-1-4. The initial flowrate was approximately 0.42 bm/s, which decreased steadily after 50 seconds to a rate of just less than 0.1 bm/s for the remainder of the transient. The predicted and measured inner cladding temperatures at NRU level 15 are compared in Figure 22-1-5 for the inner rods (rod 1) and in Figure 22-1-6 for the guard rods (rod 2). This elevation was 97.3 inches from the bottom of the active fuel. The predicted temperatures for the inner rods track the data for about the first 40 seconds of the transient; then they drop below the data. This point in time corresponds to the first sharp reduction in flooding rate (Figure 22-1-4) and reflects the resulting increase in entrainment. The predicted temperatures for the outer guard rods, which are at a higher average power, track the data for about 120 seconds before dropping below the data as the flooding rate is reduced to the minimum. At NRU level 17, the predicted inner cladding temperatures exceed the data after about 140 seconds for the inner rods and after about 80 seconds for the guard rods. These comparisons are shown in Figures 22-1-7 and 22-1-8, respectively. At NRU level 18, which was located near the bundle exit (139.3 inches above the bottom of the active fuel), the predicted inner cladding temperature for the guard rods exceeds the data after about 50 seconds, as shown in Figure 22-1-9. Comparisons of predicted and measured pellet centerline temperatures are shown in Figure 22-1-10 for the guard rod at level 15 and in Figure 22-1-11 for the inner rod at Level 17. These comparisons show the same general trends as the corresponding inner cladding temperatures, indicating that the gap conductance and fuel conductivity are reasonably predicted.

                                                                      ]a o:\4384-non\4384-22.wpdlb-04043                22-7

Cladding deformation and burst information predicted by WCOBRAITRAC are compared with the experimental data in Table 22-1-1. The information provided under the "Data" column is the averages of the data provided in Tables 6 and 7 of NUREG/CR-2528 (Mohr, et al., 1983). The burst times, temperatures, and strains predicted by MOD7A are in good agreement. The burst elevations require some additional discussion, which follows. WCOBRA/TRAC predicts rupture of Zircaloy-4 cladding as a function of heatup rate and engineering hoop stress, as described in Section 7-4-1 of this document. This typically results in the burst elevation coinciding with the heat transfer node with the highest temperature at the time of rupture. This was true in the NRU MT-3.06 simulation. [

                                             ]a.c Figure 22-1-13 shows a comparison of the predicted rod 1 internal pressure to the measured plenum pressure of rod 2C. [
                                                             ]axc The pressure increase at 110 seconds in Figure 22-1-13 is due to heatup of the fuel above the quench front and dryout of the cladding at the top of the fuel rod. The MT-3.06 test used a variable (decreasing) flooding rate, which resulted in stagnation of the quench front. The resulting heatup of the gas in the pellet-cladding gap, the fuel stack, and the plenum lead to the calculated pressure increase.

The early underprediction of the rod pressure transient in Figure 22-1-13 raised questions whether a tendency exists to not calculate burst and blockage in a PWR transient, when burst realistically should occur. To resolve this issue, Westinghouse committed to increasing the hot assembly rod initial pressure until burst and blockage are achieved if the nominal calculation results in a hot assembly rod reflood PCT greater than 1600°F without burst. In these unlikely cases, the most limiting of the burst and nonburst cases will be used as input to the uncertainty evaluation. o:\4384-non\4384-22.wpd:b-04043 22-8

22-1-7 Simulation of NRU Test PTH-110 Test PTH-1 10 was simulated using WCOBRAIIRAC-MOD7A. The inlet temperature transient used in the simulation was taken from the data report and is shown in Figure 22-1-14. The inlet flowrate was set to the nominal value of 1.9 in/s, based on Table 2 of NUREG/CR-1882 (Mohr, et al., 1981). Initial rod temperature data were reported at levels 13, 15, and 17 only for PTH-1 10. The initial temperature distribution in the lower regions of the bundle were, therefore, estimated using the more detailed initial distributions measured in the MT-3.06 tests. Figures 22-1-15 through 22-1-17 compare the predicted and measured inner cladding temperatures at levels 17 and 18. Figures 22-1-18 and 22-1-19 compare the predicted and measured pellet centerline temperatures for the inner and guard rods, respectively. Each of these comparisons shows reasonable agreement with the data. 22-1-8 Summary and Conclusions Simulations of LOFT and NRU were made using WCOBRAfrRAC to validate the nuclear rod models. Predictions of cladding temperatures in LOFT were in good agreement with the data as were the predictions of cladding and pellet temperatures in NRU test PTH-1 10. The simulation of NRU test MT-3.06 showed a tendency to [

                                                                         ] Even so, the simulation showed reasonable agreement with the test data for rupture and blockage. The underprediction of the measured rod pressure transient early in MT-3.06 raised questions regarding whether a tendency exists to not calculate burst and blockage in a PWR transient when burst realistically should occur. To resolve this issue, Westinghouse committed to increasing the hot assembly rod initial pressure until burst and blockage are achieved if a WCOBRAIRAC calculation results in a hot assembly rod reflood PCT greater than 1600°F without burst. In these unlikely cases, the most limiting of the burst and nonburst cases will be used as input to the uncertainty evaluation.

o:A4384-non\4384-22.wpd: lb-04043 22-9

Table 22-1-1 NRU Test MT-3.06 Rod Failure Data Comparison Parameter WCOBRA/TRAC Data Range Burst time (s) 135.0 133.0 109 - 182 Burst elevation (in.) 113.1 104.3 102 - 106 Burst temperature ( 0f) 1456.0 1463.0 1430 - 1500 Burst strain 0.568 0.586 0.522 - 0.736 i oA4384-non\4384-22 wpd:lb-04043 22-10

APPROXIMATELY NRU ELEV (42'-W'Y -PRESSURE TUBE CLOSURE TEST HANGER TUBE OULET TEMPERATURE INDI CATORS NRU ELEV(47 ) SHROWD FUEL I BLINDLE S \ I ~~~~CLADDING ThEMOCOUPLES I. L t S AT VARIOUS'X LOCATIONS ON FLB. RODS MRU'JELEV(4i5-Q)I , ,B INLETTEMPERATURE

• l*  :~k 1CAT.R '

APPO MATELY .XI NRU EEV (452'-') Figure 22-1-1. Vertical Test Train Configuration for NRU Reflood Experiments o:\4384 _non\4384-22wrdlb-04043 22-11

PRESSURE TUBE TEST FUEL ROD eSHROUD GUARD -THIMBLE. FUEL ROD Figure 22-1-2. NRU Test Bundle Cross Section (Test PTH-110 Bundle Shown) o:\4384-non438422.wpd:lb-04043 22-12

20 168.00 O1 F p 12

                                                                                                  .19 150.00 15             0                               10 CD 15 145.44                                            14                                             17 It 0                                                              14 134.76                          ----- -
                                  ---                13 16

_o 134.0 13 m 12 124.32 15 IQ o 12 Il I 113.76 4-- ----F-1 14 U, 113.0 0 11 10 --- 103.32 1--- 13 Lr. uz cli z 6 6 10 92.76 -z- - - - II - --- 9---L z_ -- _j= --- 12 0 92.0 0 ir 9

                 ---                                  8 ao 82.32                                                                            ED cI)
                                                                                     .(

l--- 11 IQ) 0 0- 8 ct. 71.76 0 --- - - - - - - F -7 I--- 10 71.0 7 Ito 6 61.32 0 1--- 9 6 b-- 50.76 o 5 --- 1--- 8 50.0 5 a . 4 40.32 0 7 l--- V1-

                          -------     F           F   ~~3- -  -

4 0 29.76 1 6 29.0 3 2 --- 17.88

                                   - - ©---- - - -                       @2 5

1 6.0 s 3 2 0.0 1 1 0 Figure 22-1-3. WCOBPRA/fRAC Model of NRU o:4384-non\4384-22.wpd:I b-04043 22-13

I NRU Test MT-3 450 - 400 -

                  -o 350 -

300 - 0~ 250 - cn 200 - 150 - 100-I- 0 C) 50 - 0- 4 C I I 0 100 200 300 Time (sec) IL, 0.60 a-0

                  =    0.40 q,

0 cz 0 0.20 0 C._ c~ 0.00 0 100 200 300 vvsunt-3 Time (sec) Figure 221-4. NRU Test MT-3.06 Injection Flowrate oA4384-non\4384-22wpd:lb-04043 22-14

- Figure 22-1-5. Comparison of Rod 1 Predicted and Measured Inner Cladding Temperatures at Level 15 for NRU Test MT-3.06 Figure 22-1-6. Comparison of Rod 2 Predicted and Measured Inner Cladding Temperatures at Level 15 for NRU Test MT-3.06 o:\4384-non\4384-22.wpd:1b04043 22-15

I ac Figure 22-1-7. Comparison of Rod 1 Predicted and Measured Inner Cladding Temperatures at Level 17 for NRU Test MT-3.06

                                                                           ,'X' Figure 22-1-8. Comparison of Rod 2 Predicted and Measured Inner Cladding Temperatures at Level 17 for NRU Test MT-3.06 o\4384-non\4384-22.wpd:lb-04043                 22-16

Figure 22-1-9. Comparison of Rod 2 Predicted and Measured Inner Cladding Temperatures at Level 18 for NRU Test MT-3.06 Figure 22-1-10. Comparison of Rod 2 Predicted and Measured Pellet Temperatures at Level 15 for NRU Test MT-3.06 o:\4384-non\4384-22.wpd: b-04043 22-17

I a,c Figure 22-1-11. Comparison of Rod 1 Predicted and Measured Pellet Temperatures at Level 17 for NRU Test MT-3.06 o:\4384-non\4384-22.wpd:1b04043 22-18

a; Figure 22-1-12. Comparison of WCOBRAJTRAC Predicted Quench Front Elevations with MT-3.06 Data from NUREG/CR-2528 o:\4384-non\4384-22-1 awpd:lb-04043 22-19

1-", Figure 22-1-13. Comparison of Rod 1 Internal Pressure to the Measured Plenum Pressure of NRU Rod 2C for NRU Test MT-3.06 IL o\4384-non\4384-22-1a.wpd:lb-04043 22-20

NRU Test PTH-1 10 450 0 L. 300 L. C-E 150 a, C 0

  -W-C, a)         0 C

0 100 200 300 Time (sec) Figure 22-1-14. NRU Test PTH-11O Injection Temperature o:\4384-non\4384-22-1a.wpd:lb-04043 22-21

ax I "L/ Figure 22-1-15. Comparison of Rod 1 Predicted and Measured Inner Cladding Temperatures at Level 17 for NRU Test PTH-110 t.sL,' Figure 22-1-16. Comparison of Rod 2 Predicted and Measured Inner Cladding Temperatures at Level 17 for NRU Test PTH-110 o:\4384-non\4384-22-la.wpd:lb04043 22-22

Figure 22-1-17. Comparison of Rod 1 Predicted and Measured Inner Cladding Temperatures at Level 18 for NRU Test PTH-110 Figure 22-1-18. Comparison of Rod 1 Predicted and Measured Pellet Temperatures at Level 17 for NRU Test PTH-110 o:\4384-non\4384-22-lawpd:1b-04043 22-23

a,c Figure 22-1-19. Comparison of Rod 2 Predicted and Measured Pellet Temperatures at Level 17 for NRU Test PTH-110 o\4384-non\4384-22-la.wpd:lb-04043 22-24

22-2 Accumulator Component 22-2-1 Introduction The accumulator component model is described in Volume 1, Section 9-8, of this document. Section 9-8 also describes the phases of accumulator water injection, emptying, and accumulator nitrogen discharge. In verifying the application of the model to the PWR, WCOBRAJTRAC calculations were perforned and the results were compared to available separate effects accumulator test data. In addition, nitrogen discharge effects were assessed. 22-2-2 Indian Point Unit 2 Accumulator Test An accumulator blowdown test was performed at Indian Point Unit 2 in 1971 during startup testing. The initial gas pressure in the accumulator was about 100 psig, the gas volume was about 400 cubic feet, and the water volume was 700 cubic feet. Test runs were performed at ambient temperature (80°F) with an RCS back pressure of 0 psig. The cold legs were empty, and the water level in the vessel was well below the cold leg nozzle elevation. The control valves used to initiate the test runs were set to open from 0 to 100 percent in 10 seconds. Test runs were performed for the four accumulators that had various accumulator line lengths. The test runs would terminate when the pressure in the accumulator reached approximately 20 psig while the accumulator line was still in single-phase liquid flow. The measured pressure responses of the four accumulators were all similar. Pressure response for one of the accumulators was selected for WCOBRAIRAC model verification. Figure 22-2-1 shows the layout of the accumulator piping. 22-2-3 WCOBRAflRAC Model A WCOBRA/TRAC model was constructed to simulate the accumulator test. A typical PWR model of the accumulator and its piping consists of up to four WCOBRAfTRAC model components: an accumulator, one or two valves, and a pipe, as shown in Figure 22-2-2. In this model, the RCS is simulated by a BREAK component, supplying a constant back pressure. The volume, height, length, and hydraulic diameter for the accumulator and the accumulator line are all preserved. The various levels and elevation changes in the actual line are simplified to some extent as shown by the dotted lines in Figure 22-2-1. The line resistance in the accumulator line is simulated in two ways: using the WCOBRA/TRAC built-in friction model (NFF = 4) and using the hydraulic resistance (fLlD) value obtained from measurement and uniformly o:\434-non\4384-22-2wpd.b-O4043 22-25

distributed over the pipe length. Little difference is observed in the two approaches. The initial and boundary conditions are the same as those used in the 1971 Indian Point Unit 2 test. A L, steady-state run of 20 seconds was first performed, followed by a blowdown run initiated by opening a control valve in the accumulator line. The valve reached 100-percent opening within the first 10 seconds of the blowdown run. The accumulator pressure predicted by WCOBRA/TRAC is compared to measured test data (the only data available) in Figure 22-2-3. WCOBRA/TRAC predicts a [

                                         ]a,c 22-2-4 WCOBRAITRAC Model With PWR Line Noding Although the accumulators are of the same design for all loops, the lines connecting the accumulator and the cold leg may vary from loop to loop. In the WCOBRAIRAC PWR plant model, the accumulator and the connecting line in all loops are [
                           ]ac 22-2-5 Nitrogen Model Switching and Accumulator Noding WCOBRAiTRAC calculates the accumulator blowdown before and after the accumulator becomes empty. With this continuous blowdown model, the accumulator will switch from the o:\4384-non\4384-22-2.wpd:lb-04043              22-26

normal nitrogen model to the subcooled vapor model when the [ ac The model described above predicts the behavior shown in Figure 22-2-9 for the Indian Point Unit 2 test. Nitrogen fills the pipe from the accumulator end, and a two-phase mixture appears at the pipe end at 100 seconds. o:\4384-non\4384-22-2.wpd:1bO4043 . 22-27

Elevation (FT) 70 r- Length: A to B: 326.48 FT B to C: 50.75 FT & HHSI C to D: 58.71 Fr I-65 435.94 Fr C.L Valve ! 60 I-55 i-t RHR 50 I- (Honzontal Distance Not to Scale) 45 L Figure 22-2-1. Indian Point Unit 2 Loop 21 Accumulator Line Schematic Diagram oA\4394-non\4384-22-2.wpd:lb-04043 22-28

Accumulator 1 3 pI SI Source N-1 PiporValValve Cold Leg TEE N 9 Cells Equal Length -* Figure 22-2-2. WCOBRAIRAC Model of Accumulator and Safety Injection Line in a PWR o:\4384-nonX4384-22-2.wpd:1b04043 22-29

Figure 22-2-3. Predicted Accumulator Pressure (Solid Line) Compared With Measured Test Data (Dashed Line) o:\4384-non\4384-22-2.wpd:lb.04043 22-30

a.q Figure 22-2-4. Predicted Accumulator Flowrate o\4384-non\4384 Zwpd:lb-O4043 22-31

Figure 22-2-5. Predicted Gas Temperature at Top of Accumulator o43S4-non\4384-22-Zwpd:db-04043 22-32

Figure 22-2-6. Comparison of Predicted Pressure/Volume Relationship With Adiabatic Assumptions o.\434-non\4384-22-2.wpd:lb-04043 22-33

I Figure 22-2-7. Comparison of Detailed Noding With Simplified PWR Noding Prediction of Accumulator Pressure o:A4384-non\438422-2.wpd:lb-04043 22-34

ax Figure 22-2-8. Basis for Transition from Water to Nitrogen Flow From Accumulator (Andreychek, et al., 1988) o:A4384-non\438422-2.wpd:lb-04043 22-35

Fu 2L Figure 22-2-9. Predicted Void Fraction at Accumulator Line Exit

                                                                                     ;>L,,

o:\4384-nonz4384-22-2.wpd:lb-04043 22-36

22-3 Pump Component Model The pump component model is described in Volume 1, Section 94, of this document. It is an empirical model in which the pressure differential generated by the pump, and the corresponding torque applied to the pump during single- and two-phase flow, is derived from single- and two-phase flow data in scaled pumps. In particular, the pump head and torque during two-phase flow is assumed to vary as a function of void fraction from the single-phase value to a "fully degraded" or minimum value which occurs at intermediate void fractions. For the pump head: H = H, + M(i') * (H 2 - H,) (22-3-1) where: H = pump head H, = single-phase pump head H2 = fully degraded pump head M(a) = two-phase multiplier A similar equation is used for the pump torque (Equation 9-8 from Section 94) with the multiplier defined as N(a). This is clearly an approximate description of the actual variation of the pump head. As described in NUREGICR-5249 (Rohatgi, et al., 1989), the uncertainty associated with such a model is relatively large and needs to be considered in the code uncertainty. This section describes the basis for the empirical model used in the LOCA analysis of the PWR, establishes the basis for its uncertainty, and relates it to the pump model used in LOFT. Comparisons with LOFT data of the predicted pump head then serve as validation that the empirical model adequately predicts pump head for both LOFT and a PWR. 22-3-1 Westinghouse Pump Data The Westinghouse pump model is based on the air and water data obtained from a scale model of a 93A model pump, designed to operate at a pump head of 92.6 feet, a flow of 7420 gpm, and an impeller speed of 1799 rpm. Figure 22-3-1 shows a scale model used to obtain single- and two-o:4384-non\4384-22-3.wpd.lblO4043 22-37

phase data. The model is designed to be geometrically similar to a full-scale Westinghouse model 93A pump with an equivalent specific speed. The specific speed of a centrifugal pump is defined as: N = N Q"21113 4 (22-3-2) where: N is in rpm Q is in gpm H is in feet of water Specific speed is a convenient parameter in distinguishing the performance characteristics of different pumps. The specific speeds of Westinghouse pumps range from 5000 to 7000 rpm. In contrast, the specific speed of the LOFT pumps is 3300 rpm. 23-3-1-1 Single-Phase Data Figures 22-3-2 and 22-3-3 show some of the test data used to determine the single-phase homologous curves for forward and reverse flow through the pump. The data consist of water data from the scale model of the 93A pump, as well as air data from the same scale model and test facility where two-phase data were obtained (Howland and Lamers, 1973). The air and water data agree well, indicating that the change in test fluid and test facility had little effect on the test results. The uncertainty of the single-phase data was determined by evaluating two data sources. The first source was from the Westinghouse single-phase data cited above. A band can be drawn about the data in Figure 22-3-3 (the normalized head data are plotted against the inverse of the normalized flow in this figure). [

                                                                                     ]ax The second source examined was from data developed by Cudlin (Cudlin, 1977). The normalized head in the forward flow, dissipative quadrant for a 1/3-scale model pump is shown in Figure 22-3-4. [                                                                      Iac o:\4384-non\4384-22-3.wpd:lb/04043             22-38

23-3-1-2 Two-Phase Data The two-phase data were obtained by running air-water mixtures through the pump (Howland and Muench, 1975). The test facility is illustrated in Figure 22-3-5. Water was drawn from a large basin using a diesel-powered pump, mixed with air in a mixing chamber, and pushed through the scale model pump. Inlet line venturi meters and orifices were used to measure inlet flowrates. Pump pressure differential, impeller speed, and impeller torque were also measured. The inlet void fraction was not measured but was inferred from the flowrates. A correlation was used to estimate the void fraction from the flowrates. In addition, a homogeneous void fraction was used. The basic nature of the data was not affected by the choice of void fraction. In the following discussion, the homogeneous (zero slip) void fraction is used. Typically, homologous head data are plotted using two x-axes: normalized flow divided by normalized speed (Figure 22-3-2), and normalized speed divided by normalized flow (Figure 22-3-3). An altemative way to plot the head data is as a function of normalized head divided by normalized speed squared, versus normalized flow divided by normalized speed, for all forward flow conditions. This results in the data in Figure 22-3-6, which more clearly shows the transition, as flow increases from a positive head or pumping mode, to a negative head or energy dissipation mode. As seen in Section 25-3 of WCAP-12945-P-A (Bajorek, et al., 1998), the intact loop pumps are operating in the pumping mode during the initial stages of a cold leg break LOCA, while the broken loop pump is operating in an energy dissipation mode during the entire transient. The two-phase data are also shown on this figure and indicate that the pumping mode data shows relatively little scatter, while the dissipation mode data show more scatter. The increased scatter may be due to the fact that when the downstream pressure is lower, the upstream conditions are no longer as accurate a representation of conditions within the pump. Also plotted on this figure are the single-phase head curve and a fully degraded head curve drawn through the lower bound data. The method for determining the two-phase multiplier M(a) and N(a) in Equations 9-7 and 9-8 (in Section 9 of this document) from the pump data is as follows:

1. Determine single-phase homologous head and torque. The pressure difference across the pump and the torque applied to the pump impeller are measured under a variety of flow conditions. Homologous head and torque curves are derived by dividing these data by the appropriate quantities (rated flow, rated speed, and the like). Each pump model (designated 93, 93A, 100, and so on) designed by o:\4384-non\4384-22-3.wpd:lbl04043 22-39

Westinghouse has a set of homologous curves derived from scale model single- I> phase tests using both air and water.

2. Measure the pump pressure difference and torque under two-phase conditions over a range of void fractions. The lower boundary of the data, when converted to homologous form, is defined as the "fully degraded" homologous head and torque.

These data were obtained from a 1/3-scale model pump with the same specific speed as the model 93A pump. The pump head data are shown in Figure 22-3-7, and the pump torque data in Figure 22-3-8. The single-phase and "fully degraded" curves constructed from these data are also shown in Figure 22-3-8 and Figures 9-4 to 9-7 in Volume 1, Section 9, of this document. The fully degraded curves are always drawn below the single-phase curves and bound nearly all the data. The two-phase data indicate that the amount of full degradation in head or torque is approximately a constant. That is, the fully degraded curve is offset from the single-phase curve by a constant. This is more easily seen in Figure 22-3-6. This observation allows the fully degraded curve to be extended into areas where data are sparse or lacking.

3. Assume that the homologous head and torque go from single-phase to fully degraded, back to single-phase values, as the pump inlet void fraction ranges from 0 to 1.0. Use Equation 22-3-1 in the following form to calculate M(a;) for each pump head data point:

H(a,) - H(single-phase) H(degraded) - H(single-phase) Use the M(a;) data to define the appropriate shape of the M(a) function, as in Figures 22-3-9 and 22-3-10. Figure 22-3-9 includes only the pumping mode data, while Figure 22-3-10 includes all the data. Perform a similar exercise for the pump torque (Figure 22-3-11). Data are lacking for void fractions greater than 70 percent. [

                                                                                     ]ac This assumption is supported by test data from other design pumps, for example, Figure 2.1 on page L-9 of the code scaling, applicability, and uncertainty (CSAU) report (Boyack, et al., 1989).

o:\4384-non\438422-3.wpd:1b/04043 22-40

The simple form of the M(a) function results in considerable scatter in the data in the dissipative, or turbine mode of pump operation. The effect of this uncertainty was examined in the PWR scoping studies by defining a new multiplier drawn through the lower bound of the data. The multiplier resulted in a relatively small effect due to the relatively short time that the pump is in the fully degraded, low void fraction two-phase regime. This result is consistent with results obtained in the CSAU report. Most studies of pump model uncertainties focus only on the two-phase characteristics. The Westinghouse methodology examines and accounts for the uncertainty present in the single-phase head curves as well as discussed in Section 25-3-2 of WCAP-12945-P-A (Bajorek, et al., 1998).

                                                                                                ]a.C o\4384non4384-22-3.wpd:lb/04043                  22-41

                                                                                   ]axc 22-3-2 Pump Model Comparison to Data The only test that contains a powered pump is the LOFT test. Although the pumps in LOFT are of a different design than PWR pumps, they exhibit similar overall performance as can be seen from Figure 22-3-12. The pump model used in the LOFT simulations, described in Section 14-1 of WCAP-12945-P-A (Bajorek, et al., 1998a), is the same as that used in the PWR, except that the homologous curves and the two-phase multiplier used were the LOFT specific curves obtained from tests on the Semiscale pump (Reeder, 1978). Another difference was that the pump speed was input from the LOFT data, rather than calculated. This was done to specifically examine the pump head prediction. The resulting prediction for LOFT test L2-5 is shown in Figure 14-1-38 (Bajorek, et al., 1998a). These comparisons show that the predicted pressure difference across the pumps in the intact loops compares well with the measured pressure difference during blowdown.

The comparisons indicate that the relatively simple pump model in WCOBRAiTRAC adequately predicts pump behavior during a LOCA. Because pump performance is not of high importance in a small break LOCA, and the pumps are tripped early in the event, uncertainty is not considered. o:\4384-non\4384-22-3.wpd:lb/04043 22-42

Figure 22-3-1. Cross-Sectional View of the Westinghouse Scale Model Pump oA4384-nonN4384-22-3.wpd:1b/04043 2243

ac I_I Figure 22-3-2. Scale Model Homologous Head Single-Phase Data in the Pumping Mode, Forward and Reverse Flow o-\4384-non\4384-22-3.wpd:1b/04043 22-44

Figure 22-3-3. Scale Model Homologous Head Single-Phase Data in the Dissipation Mode, Forward Flow o:\4384-non438422-3.wp&-1bl043 2245

I ac \L; Figure 22-3-4. Data Scatter for Dissipative Mode 1/3-Scale Pump Data (Cudlin, 1977) o:\4384-non\4384-22-3.wpd:1b/04043 2246

Figure 22-3-5. Schematic Diagram of the Air-Water Test Facility o:\4384-non\4384-22-3.wpd:1b/04043 22-47

93A PUMP HEAD CURVES PUMP (0) AND DISSIPATION (*) MODE DATA 2 1 Cl4 a 0 (L 0 -1 N

  -j        -2 0

z -3

            -4 a
            -5 0
            -6 z
            -7
            -8 0                   1                    2                   3       4 NORMALIZED FLOW/NORMALIZED SPEED
        -        (UPPER)SINGLE PHASE                            -     (LOWER) DEGRADED Figure 22-3-6. Homologous Head Curves and Westinghouse Air-Water Data o:\4384-non\4384-22-3.wpd: 1b04043                 2248

Figure 22-3-7. Single-Phase and Fully Degraded Pump Head Curves Compared With Two-Phase Data o-A4384-non\4384-22-3.wpd:1b/04043 2249

ac Figure 22-3-8. Pump Single-Phase and Fully Degraded Torque Curves, Compared With Two-Phase Data o:\43 84-non\43 84-22-3.wpd: I b/04043 22-50

a.c Figure 22-3-9. Two-Phase Multiplier and Pumping Mode Data o.\4384-non\4384-22-3.wpd:1b/04043 22-51

Figure 22-3-10. Two-Phase Multiplier and All Two-Phase Data o-\4384non\4384-22-3.wpd:1bJ04043 22-52

Figure 22-3-11. M(a) for Pump Torque (Referred to as N(a) in Equation 9-8 in This Document) o\4384-non\4384-22-3.wpd:1b104043 22-53

Figure 22-3-12. Westinghouse Pump Head Curves Compared With LOFT Pump Head Curves o-.\4384-non\4384-22-3.wpd:lb/04043 22-54

22-4 References Andreychek, T. A., et al., 1988, "Loss of RHRS Cooling While the RCS is Partially Filled," WCAP-1 1916. Bajorek, S. M., et al., 1998, "Code Qualification Document for Best Estimate LOCA Analysis Volume V: Quantification of Uncertainty," WCAP-12945-P-A, Vol. 5. Bajorek, S. M., et al., 1998a, "Code Qualification Document for Best Estimate LOCA Analysis Volume HI: Hydrodynamics, Components and Integral Validation," WCAP-12945-P-A, Vol. 3. Bajorek, S. M., et al., 1998b, "Code Qualification Document for Best Estimate LOCA Analysis Volume II: Heat Transfer Model Validation," WCAP-12945-P-A, Vol. 2. Boyack, et al., 1989, "Quantifying Reactor Safety Margins," NUREG/CR-5249. Burman, D. L., 1990, ZIRLOTm High Temperature Oxidation Tests," WCAP-12610, Appendix E. Cathcart, J. V., et al., 1977, "Zirconium Metal-Water Oxidation Kinetics IV - Reaction Rate Studies," ORNIUNUREG-17. Cudlin, J. J., 1977, " 1/3 Scale Air-Water Pump Program, Analytical Pump Performance Model," EPRI NP-160. Howland, G. R. and Larners, R. P., 1973, "Air Test Program to Establish the Complete Pump Characteristics of WEMD 93A Model Reactor Coolant Pump," Westinghouse Research Report 73-7E9-TAPSC-Rl. Howland, G. R. and Muench, R. A., 1975, "Air/Water Mixed Flow Testing of the WEMD 93A Model Reactor Coolant Pump," Westinghouse Research Report 75-7E9-CORCL-RI. Mohr, C. L., et al., 1981, "Prototypic Thermal-Hydraulic Experiment in NRU to Simulate Loss-of-Coolant Accident," NUREG/CR-1882. o:\4384-non\4384-22-3.wpd:1b/04043 22-55

Mohr, C. L., et al., 1983, "LOCA Simulations in the National Research Universal Reactor Program" NUREG/CR-2528. Reeder, D. L., 1978, "LOFT System and Test Description," NUREG/CR-0247. Rohatgi, et al., 1989, "Quantifying Reactor Safety Margins," NUREG/CR-5249. Weiner, R. A., et al., 1988, "Improved Fuel Performance Models for Westinghouse Fuel Design and Safety Evaluations," WCAP-10851-A (Proprietary), WCAP-1 1873-A (Non-Proprietary). o:\4384-non\4384-22-3.wpd:1b/04043 22-56

SECTION 23 CODE ASSESSMENT

SUMMARY

AND CONCLUSIONS 23-1 Introduction Sections 12 through 22 in this volume provide comparisons of WCOBRA/TRAC-SB predictions to experimental test results involving small break LOCA processes. This section summarizes the code performance in simulating the separate and integral effects tests necessary to demonstrate that WCOBRAIRAC-SB produces satisfactory results for small break LOCA processes. These simulations provide a consistent set of information that can be used to determine the bias and uncertainty for the WCOBRA1tRAC-SB computer code. These test simulations also provide a comprehensive validation of the code capability. A wide range of test facilities and conditions was selected for simulation, not only to establish code applicability, but also to provide a means of isolating and assessing individual model and correlation packages. Tests in the assessment matrix were selected to validate the ability of the code to model the important phenomena that occur during a small break LOCA in a PWR. At the beginning of the development effort, Westinghouse generated a small break LOCA PIRT (Section 1-4, Volume 1, of this document). To a great extent, the important processes identified by an independent expert panel were in agreement with the Westinghouse table. In fact, Westinghouse technical experts revised their original ranking process to coincide with the rankings assigned by the independent panel (Attachment A, Volume 1, of this document). This final PIRT will be used as a checklist to demonstrate that WCOBRAJTRAC-SB is capable of performing best-estimate small break LOCA analysis. 23-2 Separate Effects Test Simulations The separate effects tests simulated provide insight into the capability of WCOBRATRAC-SB to predict high-ranked PIRT phenomena as discussed below. Section 12 presents simulations of heat transfer experiments relevant to post-CHF core uncovery conditions using the WCOBRArFRAC-SB heat transfer package. During small break LOCA events, this process is ranked high in the PIRT. A comparison between measured and predicted heat transfer coefficients is available in the Reynolds number regime of interest from the ORNL o:"4384-nonN4384-23.wpd:b/414103 23-1

and INEL experiments. The minimum and maximum values of the multiplier to adjust the code-predicted coefficient to match the test value [

                                                                                         ]ac Section 13 compares the WCOBRA,TRAC-SB break model predictions against data for a wide range of experimental critical flow geometries and conditions to validate the code for this highly ranked process. Uncertainty in the break flow model is accommodated in the uncertainty methodology by varying the break size to identify the limiting size in the plant break spectrum.

[

        ]axc Section 14 describes the implementation of the safety injection jet heat transfer correlation from the COSI experiment into WCOBRAJIRAC-SB. The agreement of the code prediction for this low-ranked process with COSI data is adequate.

In Section 15, the mixture level swell in the core (a high ranked PIRT item) is examined for the ORNL and G-1 facilities, with heater rod bundles, and the GE facility test vessel. The interfacial drag multiplier is adjusted to enable the predicted WCOBRA/TRAC-SB level swell to match the ORNL and G-1 experimental data, and the median YDRAG value to match the experimental data from both facilities is 0.78. This supports the use of YDRAG = 0.8 in the core in integral test simulations and the reference PWR studies. [

                    ]axC For the loop seal clearing period of the small break LOCA transient, the PIRT identifies a number of high-ranked processes. The capability of WCOBRAJFRAC-SB to predict loop seal clearance phenomena is established by simulating the full-scale UPTF test facility in Section 16.

Overall, the code predicts the trends in behavior observed in the UPTF experiment. Section 17 considers a number of phenomena pertinent to upper plenum/hot leg/steam generator region hydraulics during a small break LOCA event. The CCFL prediction of WCOBRAfRAC-SB in the relevant flow situations matches the data well. The condensation o:\4384non\4384-23.wpd:lb/414/03 23-2

heat transfer modeling of the code has been examined by simulating the NC test series 60-kW core power test in the Semiscale Mod-2A facility. The agreement in these natural circulation two-phase flow and heat transfer predictions is judged to be adequate based on commentary from the experimenters. Chapter 18 validates the WCOBRAfTRAC-SB computer code for the prediction of horizontal stratified flow phenomena in the PWR loop piping; horizontal flow is a high-ranked process in the PIRT. WCOBRA/TRAC-SB was used to simulate a test facility that used a rectangular channel to measure condensation of steam in concurrent, horizontal flow. The channel is constructed of stainless steel with Pyrex glass windows. The 32 experiments simulated with WCOBRA/TRAC-SB included a range of steam and water flowrates and temperatures, and water layer thickness at the inlet. Inlet steam pressure was approximately 1 atmosphere. Steam velocity, static pressure, and water layer thickness as measured at five locations along the channel were compared with the WCOBRA/TRAC-SB predictions. The code was shown to predict pressure variation reasonably well and to underpredict condensation occurring at the interface. [

                                                                        ]a,c 23-3 Integral Test Facility Simulations Facilities of varying scales and characteristics were simulated using consistent nodalization to evaluate the capability of WCOBRAfRAC-SB to predict experimental small break LOCA transients. The small scale Semiscale Mod-2C facility provides single-effect sensitivity tests (S-LH-l and S-LH-2) which assess the impact of upper head flow bypass on the scaled equivalent of a 6-inch diameter cold leg break. The depressurization rate is predicted well for these tests. The break flow is underpredicted and then overpredicted at different times, and the S-LH-2 core heater rod temperature transient is influenced by compensating errors in the mass inventory and hot leg draining predictions. The predictions mirror the test results in identifying the significant impact of the upper head bypass flow on transient behavior.

The LOFT breaks simulated with WCOBRA/IRAC-SB range from a small break size (13-7, the scaled equivalent of a 1-inch break) to the 4-inch equivalent diameter break series of cases: 3-1 in the broken loop, and L3-5 in the intact loop cold leg. The L3-7 predictions are adequate, demonstrating the capabilities of the code for small break LOCA cases. The set of 4-inch equivalent break cases reinforces the importance of the high-ranked PIRT item regarding o:\4384-non\4384-23.wpd:lb/4/4103 23-3

conditions upstream of the break; the code performance is judged based on detailed evaluation to be adequate when the atypicalities of the LOFT facility are taken into account. The ROSA tests analyzed provide a set of 2.5-percent break size cases that examine the impact of break orientation (top, side, and bottom locations). WCOBRA/TRAC-SB predicts the general trends observed in the different results for the three orientations. The 5-percent ROSA test is the larger scale equivalent of the Semiscale cases in break size, and it illustrates the effect of scale on the adequacy of code predictions to be minor. The 10-percent break ROSA test provides an intermediate size small break LOCA to illustrate WCOBRA/TRAC-SB capabilities for the larger breaks in the small break LOCA spectrum. Among the 6-inch equivalent diameter break simulations, both of the Semiscale cases and the ROSA-IV SB-CL-05 break flowrate predictions exhibit flowrates less than the data values in the saturated liquid flow regime prior to loop seal clearance. This is followed by critical flowrate predictions that exceed the data once two-phase flow is established after loop seal clearance. The global model sensitivity cases performed as part of the PWR uncertainty methodology consider this variability by ranging the two-phase break flow discharge coefficient after loop seal clearance. 23-4 Nodalization Consistency The performance of the code in the ROSA, LOFT, and Semiscale simulations supports its use in PWR calculations in a best estimate methodology for small break LOCAs. The PWR nodalization scheme is consistent with the schemes used in the integral test facility simulations. The nodalizations used in the integral test facility simulations and the PWR calculation are reviewed by region or component. Hot Leg and Cold Leg

                                                ]axc o:\4384-non\4384-23.wpd:lb/44103                 23-4

I 3ac Steam Generators I

                                             ]a,c Horizontal Crossover Leg Pipin, I

Ia.c o:.4384-non\438-23.wpd:1b414/03 23-5

[ Ia,c RCPs Ia,c [ Lower Plenum [ I c Core [

                        ]aC The PWR model is [

Ia,c o-\4384-nonW384-23.wpd:lb/4/4/03 23-6

Upper Core Plate I Ia,c Upper Plenum - Below the Hot Leg [

                                ]a,c Upper Plenum - Above the Hot Leg I

Iac o:\4384-non\4384-23.wpd:lb/4/403 23-7

I Ia,c Upper Head I Iac Downcomer [ Iac o:\4384-non\4384-23.wpd:lb/4/4/03 23-8

Downcomer-to-Upper Head Bypass

                                                                                                    ]a,c Summar The only notable differences among the various facility models are those that are forced by some of the facility-specific requirements, and in the lower upper plenum region where there is a combination of necessary differences caused by the core nodalization and other differences related to modeling of jet channels.

23-5 Conclusions Based on the separate effects test and integral test simulations, Westinghouse has made the following conclusions about the performance of WCOBRA/TRAC-SB for small break LOCA processes:

1. The model for critical break flow was validated in separate effects test simulations. It provides acceptable results for a wide range of break sizes, geometries, and inlet conditions. The code-predicted break flow shows reasonable agreement with the data in the integral effects tests over a range of scales.

2. The WCOBRA/TRAC-SB correlations for heat transfer in highly voided conditions show good agreement with ORNL and INEL data. The application of the WCOBRAJIRAC-SB heat transfer package to integral effects tests simulations was successful.

3. The models affecting mixture level swell provide agreement with a bias and a range of uncertainty for mixture level swell tests over a series of tests conducted under small break LOCA conditions. Interfacial drag in the core will be ranged according to these results in PWR global model sensitivity studies.

4. Mass retention in the loop seal during loop seal clearance is adequately predicted, and the loop seal clearing predictions of integral effects test simulations were found to agree well with the data.

o:\4384-non\4384-23.wpd:lbt4/4/03 23-9

5. The horizontal stratified flow model was incorporated [

                          ]C   The option to cause the code to identify the horizontal flow regimes was used in both separate and integral effects test simulations successfully.

6. The COSI model for condensation in the cold leg on the jet of safety injection water in WCOBRA/TRAC-SB shows suitable agreement with the test data.

7. Simulations of various facilities were performed to assess the WCOBRAIRAC-SB prediction of upper plenum/hot leg/steam generator hydraulic phenomena. The code performance is shown to be adequate for important small break LOCA processes.

8. The integral test facility simulations investigated and affirmed the capability of WCOBRAIrRAC-SB to predict experimental small break LOCA transients at different scales:

• The code exhibited the capability to predict the trends associated with the different upper head bypass configurations of Semiscale tests S-LH-1 and S-LH-2. Predicted results were consistent with the code biases observed in the separate effects tests.
• WCOBRAITRAC-SB predictions of the suite of LOFT small break LOCA tests exhibit agreement ranging from good to marginal with the data. Specific reasons for misprediction have been discussed.
• The ROSA series of 2.5-percent break test simulations shows the general ability of the code to distinguish between top, bottom, and side break locations. The code is also used to predict larger small break LOCA tests (5-percent and 10-percent breaks). As is true for the Semiscale predictions, the predicted core behavior during the loop seal clearance and boiloff periods is satisfactory.

o:\4384-non\4384-23.wpd:lb1/41403 23-10

Volume 3 of this document reports the results obtained using the WCOBRAfIRAC-SB computer code to perfonn small break LOCA ana]yses for Indian Point Unit 2. The Indian Point Unit 2 noding is consistent in principle with the nodalizations used to model the test facilities. Volume 4 of this document describes the bias and uncertainty characterizations of the WCOBRAfIRAC-SB code for the small break LOCA application. An uncertainty methodology is developed and applied to Indian Point Unit 2. o\4384-non\438423.wpd:lb/414103 23-11

                                       -L) o:\4384-non\4384-23.wpd:lb/4/4/03 23-12}}