nureg/ia-0074, 'relap5/mod2 post-test calculation of the … · 2012-11-21 · nureg/ia-0074--...

NUREG/IA-0074- - ICSP-LP-SB-1-R

InternationalAgreement Report

RELAP5/MOD2 Post-TestCalculation of the OECDLOFT Experiment LP-SB-1

Prepared byJ. Perez, R. Mendizabal

Consejo de Seguaridad NuclearJusto Dorado, 4428040 Madrid, Spain

Office of Nuclear Regulatory ResearchU.S. Nuclear Regulatory CommissionWashington, DC 20555

April 1992

Prepared as part ofThe Agreement on Research Participation and Technical Exchangeunder the International Thermal-Hydraulic Code Assessmentand Application Program (ICAP)

Published byU.S. Nuclear Regulatory Commission

NOTICE

This report was prepared under an international cooperativeagreement for the exchange of technical information.' Neitherthe United States Government nor any agency thereof, or any oftheir employees, makes any warranty, expressed or implied, orassumes any legal liability or responsibility for any third party'suse, or the results of such use, of any information, apparatus pro-duct or process disclosed in this report, or represents that its useby such third party would not infringe privately owned rights.

Available from

Superintendent of DocumentsU.S. Government Printing Office

P.O. Box 37082Washington, D.C. 20013-7082

and

National Technical Information ServiceSpringfield, VA 22161

NUREG/IA-0074ICSP-LP-SB-1-R

InternationalAgreement Report

RELAP5/MOD2 Post-TestCalculation of the OECDLOFT Experiment LP-SB-1

Prepared byJ. Perez, R. Mendizabal



April 1992

Prepared as part ofThe Agreement on Research Participation and Technical Exchangeunder the International Thermal-Hydraulic Code Assessmentand Application Program (ICAP)

Published byU.S. Nuclear Regulatory Commission

NOTICE

This report documents work performed under the sponsorship of the Consejo De

Seguridad Nuclear of Spain. The information in this report has been provided

to the USNRC under the terms of an information exchange agreement between the

United States and Spain (Technical Exchange and Cooperation Agreement Between

the United States Nuclear Regulatory Commission and the Consejo De Seguridad

Nuclear of Spain in the field of reactor safety research and development,

November 1985). Spain has consented to the publication of this report as a

USNRC document in order that it may receive the widest possible circulation

among the reactor safety community. Neither the United States Government nor

Spain or any agency thereof, or any of their employees, makes any warranty,

expressed or implied, or assumes any legal liability of responsibility for

any third party's use, or the results of such use, or any information,

apparatus, product or process disclosed in this report, or represents that

its use by such third party would not infringe privately owned rights.

ABSTRACT

This document presents the analysis of the OECD LOFT LP-SB-1experiment performed by the Consejo de Seguridad Nuclear of Spainworking group making use of RELAP5/MOD2 in the frame of theSpanish LOFT Project.

LP-SB-1 experiment studies the effect of an early pump trip in asmall break LOCA scenario with a 3 inches equivalent diameterbreak in the hot leg of a commercial PWR.

iii

CONTENTSPAGE

SECTION 1.INTRODUCTION 12.DESCRIPTION OF THE LOFT INSTALLATION 13.RELAP5/MOD2 MODEL OF LOFT FACILITY 24.EXPERIMENT LP-SB-1 2

4.lSteady state calculation 3

4.2Transient boundary condition 34.2.lDecay heat data 34.2.2Pumps injection flow 34.2.3Auxiliary feedwater flow 34.2.4High pressure injection system 34.2.5Secondary side steam control valve 34.2.6Operational set-points 3

5.POST TEST CALCULATIONS 35.1Preliminary calculation (RUN A) 4

5.1.1Code performance 45.1.2Chronology of events 45.1.3Secondary side pressure 45.1.4Primary side pressure 45.1.5Temperatures 55.1.6Density distribution 55.1.7Break line density and break

massflow rate 55.1.8Primary system mass inventory 5

5.2Base calculation (RUN B) 65.2.1 Code performance 65.2.2Chronology 6f events 65.2.3Secondary side pressure 75.2.4Primary side pressure 75.2.5Temperatures 75.2.6Density distribution 75.2.7Break line density and break

massflow rate 85.2.8Primary system mass inventory 8

5.3Selected items 95.3.1 Loop flow and natural circulation 95.3.2 Hot leg flow stratification 9

6.CONCLUSIONS 117. REFERENCES 12TABLES 13FIGURES 16APPENDIX I -DISKETTE CONTAINING THE RELAP5/MOD2

INPUT DECKS FOR LP-SB-I 70APPENDIX II -TAITEL-DUKLER CRITERION FOR HORIZONTAL

FLOW STRATIFICATION 71APPENDIX III-HORIZONTAL STRATIFICATION ENTRAINMENT

MODEL IN RELAP5/MOD2 72

V

TABLES PAGE

1. INITIAL CONDITIONS FOR EXPERIMENT LP-SB-1 132. OPERATIONAL SETPOINTS FOR EXPERIMENT LP-SB-1 143. LP-SB-1 CHRONOLOGY OF EVENTS 15

vi

FIGURES PAGE

1. LOFT PIPING SCHEMATIC 162. AXONOMETRIC PROJECTION OF THE LOFT SYSTEM CONFIGURATION FOR

EXPERIMENT LP-SB-1 AND LP-SB-2 173. RELAP5 NODING DIAGRAM FOR LOFT LP-SB-1 CALCULATION (RUN A) 184. RELAP5 NODING DIAGRAM FOR LOFT LP-SB-1 CALCULATION (RUN B) 195. STEADY STATE CALCULATION - PRIMARY PRESSURE 206. STEADY STATE CALCULATION - S.G. & PZR LEVELS 217. STEADY STATE CALCULATION - RECIRCULATION RATIO 228. STEADY STATE CALCULATION - SURGE LINE MASS FLOW 239. C.P.U. TIME - RUN A 2410. SECONDARY SIDE PRESSURE - RUN A 2511. MAIN STEAM CONTROL VALVE STEM POSITION 2612. PRIMARY SIDE PRESSURE - RUN A 2713. HOT LEG TEMPERATURE - RUN A 2814. COLD LEG TEMPERATURE - RUN A 2915. HPIS MASS FLOW RATE - RUN A 3016. COLD LEG DENSITY - RUN A 3117. COLD LEG FLUID VELOCITY - RUN A 3218. HOT LEG DENSITY - RUN A 3319. S.G. U-TUBES LEVEL - RUN A 3420. BREAK LINE DENSITY - RUN5 A 3521. HOT LEG AND BREAK LINE DENSITIES - RUN A 3622. BREAK MASS FLOW RATE - RUN A 3723. BREAK VELOCITY - RUN A 3824. PRIMARY SYSTEM MASS INVENTORY - RUN A 3925. C.P.U. TIME - RUN B 4026. TIME STEP - RUN B 4127. BREAK LINE VELOCITY - RUN B 4228. PRIMARY SIDE PRESSURE - RUN B 4329. HOT LEG TEMPERATURE - RUN B 4430. PZR TEMPERATURE - RUN B 4531. COLD LEG LIQUID TEMPERATURE - RUN B 4632. COLD LEG VAPOR TEMPERATURE - RUN B 4733. S.G. PLENA TEMPERATURE - RUN B 4834. HOT LEG DENSITY - RUN B 4935. LOOP SEAL DENSITY - RUN B 5036. HOT LEG & BREAK LINE VOID FRACTION - RUN B 5137. S.G. U-TUBES COLAPSED WATER LEVEL - RUN B 5238. BRU7EAK MASS FLOW RATE - RUN B 5339. FLOW REGIME NUMBER IN THE BREAK LINE - RUN B 5439.(bis) FLOW REGIME NUMBER DIFF. IN THE BREAK LINE - RUN B 5540. PRIMARY SIDE MASS INVENTORY - RUN B 5641. S.G. PLENA VOID FRACTION - RUN B 5742. CORE VOID FRACTIONS - RUN B 5843. HOT LEG FLUID VELOCITY - RUN B 5944. EXPERIMENTAL MASS FLOW RATE BREAK LOCATION - SG. 6045. COLD LEG LEVEL - RUN B 6146. HOT LEG FLUID VELOCITY - RUN B 6247. U-TUBES FLUID VELOCITY - RUN B 6348. PCS - SG TUBES HEAT TRANSFER COEFFICIENT - RUN B 6449. FLOW REGIME NUMBER IN HOT LEG - RUN B 65:50. TAITEL-DUKLER LIMIT VELOCITY - RUN B 6651. DIFFERENCE ,Vg,-Vgl IN HOT LEG - RUN B 6752. DIFFERENCE fVg -Vgl IN HOT LEG - TIME STEP=O.01s. 6853. FLOW REGIME NUMBER IN HOT LEG - TIME STEP= 0.01 S. 69

vii

EXECUTIVE SUMMARY

Experiment LP-SB-1 was conducted on June 23, 1983 in the LOFTfacility at the Idaho National Engineering Laboratory.

The LP-SB-1 experiment simulated a 7.6 cm (3 inch) equivalentdiameter break in a hot leg pipe of a PWR plant. ExperimentLP-SB-1 addresses the analysis of a small break loss of coolantaccident with the break at the midplane of the intact loop hotleg. LP-SB-1 was one of a pair of experiments aimed a<t addressingthe effects of early and delayed pump trip on system behaviour.The primary coolant pumps were tripped early in experimentLP-SB-1.

The main objective of this calculation was to assess the code inthe challenging conditions of a small break scenario.

Our aim was to simulate the major physical phenomena of thetransient that took place until the beginning of the plantrecovery.

The code used to simulate the LP-SB-1 experiment was RELAP5/MOD2Cycle 36.04 installed on a CYBER 810.

The input data was based on that used in pretest calculations.Basically we have introduced the following changes:

(i)Use of an ideal steam separator.

(ii)Adjustment of the heat transfer from primary to secondarythrough an adjustment of the hydraulic diameter in the secondaryside.

(iii)A change of the nodalization in the upper part of the vesselby introducing crossflow junctions in the connections of thenozzles.

(iv)Introduction of crossflow junctions in the connections betweenthe break line and the surge line with the hot leg. The Fig 3shows the final nodalization of the preliminary calculation (RUNA).

A second calculation (called "Base Calculation") was run withthe following modifications in the input deck:

(v)HPIS piping was supressed and the injection was modelled by aTMDPJUN, in order to avoid the big amplitude flow oscillationsfound in RUN A.

(vi)The junction between the hot leg and the break line wasdefined as normal junction to make operative the offtake modelunder stratified flow conditions.

(vii)The break line was splitted in two volumes. The node close tothe break nozzle was made short (0.5 m), trying to eliminate thestepwise behaviour of the break mass flow.

ix

The preliminary calculation revealed discrepancies with theexperimental results. The most significant ones were:

-The very poor prediction of the break mass flow rate and thebreak line density. That did not show at all the break uncovery.The HZFLOW subroutine that accounts for offtake model in acrossflow junction under stratified flow conditions was not activein our code version.

-The strange, stepwise evolution of the break mass flow rate, dueto the occurrence of large differences for the phase velocities(slip) in the break. This deficiency has been identified byJRC-ISPRA as caused by the RELAP5/MOD2 interphase dragmodel.

-Strong instabilities of the HPIS mass flow rate, due to theentrance of vapor in the ECCS line from the cold leg.

The base calculation was performed using a normal junction in thebreak tee and splitting the break line in two nodes, a small oneclose to the break nozzle trying to improve the break flowbehaviour. The comparison with experimental results suggest thefollowing comments:

-The offtake model in the break tee was active in this run.However the splitting of the break line produced stronginstabilities in the break flow.

-The two-phase natural circulation finished in the experiment at-500 sec., and reflux condensation around 600 sec. later. In thesimulation this happened at -700 and -1200 sec respectively.

-In LP-SB-1, the intact loop hot leg flow stratified at -50 sec..In the simulation, the code detected stratified flow regime inILHL at -320 sec..

The major conclusions are:

i)The code could not account for the liquid entrainment andvapor pull-through in the break tee modelled as a crossflowjunction.

ii)It is necessary to improve the RELAP5/MOD2 chokedflow model to avoid large instabilities in the break phasevelocities and thermal disequilibrium effects.

iii)The Taitel-Dukler model is unable to describe properlyhorizontal flow stratification in the LP-SB-1 experiment.

iv)Natural circulation was correctly reproduced by the code.

X

FOREWORD

This report represents one of the assessment/applicationcalculations submitted in fulfilment of the bilateral -agreement for cooperation in" thermalhydraulic activitiesbetween the Consejo de Seguridad Nuclear of Spain (CSN) andthe United States Nuclear Regulatoy Commission (US-NRC) in -the form of Spanish contribution to the International CodeAssessment and Applications Program (ICAP) of the US-NRC whosemain purpose is the validation of the TRAC and RELAP systemcodes.

The Consejo de Seguridad Nuclear has promoted a coordinated -Spabish Nuclear Industry effort (ICAP-SPAIN) aiming to -satisfy the requirements of this agreement and to improve thequality of the technical support groups at the Spanish -Utilities, Spanish Research Establishments, Regulatory Staffand Engineering Companies, for safety purposes.

This ICAP-SPAIN national program includes agreements between

CSN and each of the following orgahizations:

- Unidad Elictrica (UNESA)

- Uni6n Iberoamericana de Tecnologia E16ctrica (UITESA)

- Empresa Nacional del Uranio (ENUSA)

- TECNATOM

- LOFT-ESPAIRA

The program is executed by 12 working groups and a generic codereview group and is coordinated by the "Comiti de Coordinaci6n".This committee has approved the distribution of this document -for ICAP purposes.

xi

ACKNOWLEDGEMENTS

Great appreciation is expressed to Mr. A. L6pez Lechas for hishelp in the understanding of RELAP5/MOD1 at the preliminary stepsof this work.

xiii

1. INTRODUCTION

Thermal-hydraulic research has required close interaction betweenexperimental and analytical work. A number of separate-effectexperiments have been performed to help in the validation of bestestimate computer codes. Analogously the overall results of codecalculations are assessed using data from integral testfacilities. The analyses show that the codes generally provideaccurate calculations of the Loss of Coolant Accident (LOCA).Areas where model's improvements are needed have also beenidentified by these tests. In particular the Loss of Fluid Test(LOFT) facility was adapted to study some small breaks. Themotivation of one of these is explained hereafter.

An analysis performed after TMI showed that one of the key factorsin the core damage was the tripping of the primary circuit pumps.The USNRC requested the reactor vendors to carry out an analysisof this problem. The conflict between the results of theseinvestigations led to a recommendation to carry out experiments onthis program in order to clarify the criteria for pump trip. Theexperiments LP-SB-1 and LP-SB-2 modelled small breaks in the hotleg. They differ in time of pump trip which is early in the formerand delayed in the later of these tests. In this paper the resultsobtained in a post test analysis of the experiment LP-SB-1 by theCSN working group, part of the Spanish LOFT project are set down.The calculations with RELAP5/MOD2 Cycle 36.04 were carried out ona CYBER 810 in Madrid.

2. DESCRIPTION OF THE LOFT INSTALLATION

The experimental LOFT installation simulates a four loop 1000 MWcommercial PWR. It has a thermal power of 50 MW. The installationconsists of a vessel scaled 1/47 in volume, an intact circuit withan active steam generator, a pressurizer, two pumps in paralleland a broken leg, connected by recirculation lines to the intactcircuit in order to maintain a temperature of this broken circuitnear to that of the coolant at core inlet at the beginning of theexperiment.More detailed information on the LOFT systemconfiguration is provided in Reference 1.

A LOFT piping schematic with instrumentation for experimentLP-SB-1 and LP-SB-2, and an axonometric projection of the LOFTsystem configuration are shown respectively in Figures 1 and 2.

1

3. RELAPS/MOD2 MODEL OF LOFT FACILITY

The code used for this calculation was RELAP5/MOD2 Cycle 36.04.

The input data was based on that used in pretest calculations.Basically we have introduced the following changes:

(i) Use of an ideal steam separator.

(ii) Adjustment of the heat transfer from primary to secondarythrough variation of the hydraulic diameter in thesecondary side.

(iii) A change of the nodalization in the upper part of thevessel by introducing crossflow junctions in theconnections of the nozzles [10].

(iv) Introduction of crossflow junctions in the connectionsbetween the break line and the surge line with the hot leg[10]. The Fig 3 shows the nodalization of the preliminarycalculation (RUN A).

A second calculation (called "Base Calculation", RUN B) was runwith the following modifications in the input deck:

i) HPIS piping was suppressed and the injection was modelledby a TMDPJUN, in order to avoid the big fl owoscillation found in RUN A.

ii) The junction between the hot leg and the break line wasdefined as normal junction to make operative the of ftakemodel under stratified flow conditions.

iii) The break line was splitted in two volumes. The node closeto the break nozzle was made short (0.5 m), trying toeliminate the stepwise behaviour of the break massflow.

Figure 4 shows the final nodalization.

4. EXPERIMENT LP-SB-1

Experiment LP-SB-l was conducted on June 23, 1983 in the LOFTfacility at thoe Idaho National Engineering Laboratory.

The LP-SB-1 experiment simulated a 7.6 cm (3 inch.) equivalentdiameter break in the midplane of the hot leg pipe of a PWR plant.LP-SB-l was one of a pair of experiments aimed to address theeffects of early and delayed pump trip on system behaviour. Theprimary coolant pumps were tripped early in experiment LP-SB-1.

A detailed description of the experiment is found in Reference 2.

2

4.1 Steady state calculations

To accelerate the achievement of steady-state conditions thefollowing variables were controlled.

(i) Liquid level in the "downcomer" of the steam generator.(ii) Primary mass flow.

In addition the upper part of the pressurizer was connected to adummy volume to maintain the desired pressure in the primary side.

Under these situation the code achieved steady-state conditions in100 secs. Then a calculation without controls for 25 secs wascarried out to demonstrate that a true steady state had beenreached. These stationary state conditions are compared with theinitial conditions of the plant in Table 1. The Figures 5 - 8 showsignificant parameters during the null transient (RUN A).

4.2 Transient boundary conditions

4.2.1 Decay Heat Data

Reactor power after scram was specified by means of a table.During the first 2 seconds of the transient, data were taken fromthe RELAP5/MOD1 input deck used for the pretest prediction ofLP-SB-l (3). After that, data contained in Reference 4 were useduntil the end of the transient.

4.2.2 Pumps injection flow

The pumps injection flow was simulated assuming a constant flow of0.0475 l/s, to each pump (2,5).

4.2.3 Auxiliary Feedwater Flow

An auxiliary feedwater flow of 0.5 1/s (5) was manually initiatedat 63.4 seconds and turned off at 1864.8 seconds.

4.2.4 High Pressure Injection System

The HPIS was initiated in experiment LP-SB-1 when the intact loophot leg pressure had fallen to 8.24 MPa (2).

4.2.5 Secondary Side Steam Control Valve

Descriptive data of the steam bypass valve were not available. Itsfunction was assumed by the steam control valve. After 80 secondsit was latched closed to a flow area of 0.0925% of its fullyopened value, throughout the transient.

4.2.6 Operational set-points

The operational set-points measured during the experiment, andthose used in the RELAP5/MOD2 calculation are given in table 2.

5. POST TEST CALCULATIONS

A preliminary calculation (RUN A) was carried out which led to

3

some important conclusions. To confirm these a number of detailedmodifications were incorporated and used for a second calculation(RUN B)

5.1 Preliminary calculation (RUN A)

The calculation was run for 2200 secs. This was considered to besufficient to obtain the most significant data. Table 3 shows theevent chronology.

5.1.1 Code Performance

Two thousand seconds of transient required about 100,000 cpuseconds (Fig. 9). This corresponds to a cpu/real time ratio ofabout 45. The user-specified minimum allowable time stepthroughout the calculation was 10-7 seconds and the maximun timestep was set to 0.1 seconds.

During the first 27 seconds the code reduced the time step to0.025 seconds. Afterwards and throughout the whole transient thestep was the specified 0.1 seconds.

The model consisted of 122 hydrodinamic volumes, 128 junctions and124 heat structures.

The grid time for this run was 34.6 ms per volume per advancement.

In a restart at 700 sec. with new control variables the c.p.u.time variable was reset to zero by the code (Fig. 9). There is notany reference at this behaviour in the code manual.

5.1.2 Chronology of events

The predicted timing of significant events is compared withmeasurements during the LP-SB-1 transient in Table 3.

5.1.3 Secondary side pressure

The isolation of the main feed water and the closure of the steamcontrol valve produced an increase on the secondary side pressure.In our simulation the rate of pressurisation was overpredicted(Fig. 10). Consequently the steam bypass valve started to open inadvance of the experiment (Fig. 11).

The energy removal from the steam generator was through the steamvalve leakage (around 3 x 10-2 kg/sec. from 500 seconds on ) andheat losses through the shell. The minimum flow area of the mainsteam valve was restricted to 0.0925% of its fully-open value.

5.1.4 Primary side pressure

The primary pressure was in agreement with the experiment duringthe subcooled blowdown (Fig. 12). The rate of depressurisation wasapproximately well predicted. However, the simulation did notaccount for the increase in the rate of depressurisation due tothe break uncovery. The result was that primary pressure wasoverpredicted from 1200 seconds on.

4

5.1.5. Temperatures

The measured and calculated fluid temperatures are shown inFigures 13 and 14 for the hot and cold leg, respectively.

The hot leg temperature, following the pressure trend, wasoverpredicted from 1200 seconds onwards.

Cold leg intact loop temperature is in good agreement with theexperiment until -400 seconds. At this time the cold leg suddenlyemptied through the vessel downcomer in coincidence with the endof natur al circulation (Fig. 17). Following this, the bigoscillations of HPIS flow rate (Fig. 15) produced analogous spikesin liquid temperature (Fig. 14).

The reactor vessel plena temperatures were overpredicted beyond

1000 seconds of transient.

5.1.6 Density distribution

The calculated cold leg mixture density is that of the liquiduntil -400 seconds (Fig. 16), when the pipe empties. Theexperimental density decreases less rapidly.

The hot leg density follows the experimental trend until -750 sec.After that the calculated density shows two big peaks. This is dueto the steam generator tubes depletion (fig. 18, 19).

5.1.7 Break line density and break mass flow rate

The break line density before the time of measured break uncovery(-700 sec.) was underpredicted (Fig. 20). In fact the calculatedhot leg and break line densities were nearly equal (Fig. 21). Thisshows that the offtake model with stratified flow conditions inthe main pipe does not work in our RELAP5/MOD2 version, when thecrossflow option is used. An obvious consequence is that thesimulation does not account for the break uncovery andoverpredicts density after -700 sec.

Analogously the break mass flow rate was underpredicted before andoverpredicted after the measured break uncovery time (Fig. 22).

The strange step-wise evolution of the calculated break mass flowrate is due to the occurrence of large differences between thephase velocities (slip) (Fig. 23).

5.1.8 Primary system mass inventory

The calculated HPIS plus pumps injection mass flow rates becameequal to the break mass flow rate (and so the inventory wasminimum) at around 2200 sec.. This minimum inventory was in goodagreement with the measured data (Fig. 24).

5

5.2 Base calculation (RUN B)

This calculation was run for 2500 secs. The grid time in this casewas 46.9 ms. per volume per advancement.

5.2.1. Code Performance

The c.p.u. time spent by the code [Fig. 25] was similar to RUN Auntil around 1400 seconds. After that, the code ran slowlier (Fig.26) because of the high velocity in the small node close to thebreak nozzle.(Fig. 27).

In this run the model consisted of 117 hydrodinamic volumes, 123junctions and 122 heat structures.

5.2.2. Chronology of events.

The predicted timing of significant events is compared withmeasurements during the LP-SB-1 transient in Table 3.

The opening of the valve in the ILHL break line was the beginningof the transient.

The reactor scram occured 0.65 seconds later than in theexperiment. The timing of the initial events was predicted, by theRELAP5/MOD2 calculation, to within -1 second.

Two seconds after the reactor scram, closure of steam controlvalve w as initiated. Isolation of the main feedwater took -3seconds.

The main steam control valve was fully closed at 17 seconds, 2.6seconds later than in the experiment. It is known (Ref. 7) thatthe steam flow bypass valve was open at around 30 seconds when thesecondary side pressure exceeded -6.7 MPa. To account for thisfact, in our calculation the main valve was let to reopen.

The primary coolant pumps trip occured at 26.8 sec. in thecalculation. The HPIS initiated at 44 sec., 3 seconds later thanin the experiment.

The coast down of both pumps was completed at 48 seconds.

The break line reached saturated conditions at 76 seconds. Thismarks the end of subcooled blowdown.

The auxiliary feedwater was initiated at 62.05 seconds and turnedoff at 1862.05 seconds.

In the experiment, the break line was uncovered at 715 sec. Thatdid not appear in the simulation. Until this time the break massflow rate was underpredicted and after that it was overpredicted.

Around 1650 seconds the primary coolant system pressure fellbellow the secondary system pressure (1077 seconds in theexperiment).

6

The minimun primary mass inventory was estimated to be reached atbetween 1800 and 2200 seconds in the experiment. That happened inthe simulation at 2050 seconds (when the HPIS+pump injection massflow rates exceeded the break mass flow rate).

The experiment finished at 3668 seconds.

Up to 50 seconds the chronology of events improved in thiscalculation. The break uncovery was poorly predicted because ofthe depletion of the S.G. U-tubes as it is explained later.

5.2.3. Secondary side pressure

The general behaviour is similar to that of RUN A.

5.2.4. Primary side pressure.

The primary pressure was well predicted until around 700 seconds.After that it was overpredicted, though the code detected thebreak uncovery at about 1400 seconds (Fig. 28).

5.2.5. Temperatures.

The hot leg temperature was closer to the experimental one than inRUN A (Fig. 29).

The pressurizer temperature history shows (Fig. 30) a sharpinitial decrease. After the emptying of the component, around 33seconds, the steam became superheated and its temperature began toincrease. That is due to the radiative heat transfer from thepressurizer wall.

Cold leg liquid temperature was right until 400 seconds (legdepletion). Afterwards it showed spikes due to HPIS injection andlevel oscillations (Fig. 31). From 400 seconds on the vaportemperature was in agreement with the data (Fig. 32 bis).

Steam generator plena temperatures are showed in Figure 33. Thereare not big discrepancies between simulated and measured valuesuntil their depletion . Hot wall radiation and thermal conductionfrom the wall to the thermocouples seem to have distorted themeasurement beyond 1100 seconds, Ref. 2.

5.2.6. Density distribution.

The cold leg density is similar to RUN A until 1400 seconds.After that there is a slight increase in density, in coincidencewith the beginning of a circulation loop between the vesseldowncomer and the vessel filler gap.

The hot leg density follows, again, the experimental trend, untilthe SG tubes depletion.

7

The subsequent density peaks are clearly smaller than in RUN A(Fig. 34). Also, the density does not fall, after 1500 sec, sorapidly as in RUN A.The agreement between the experimental and calculated loop seal

densities is very good (Fig.35).

5.2.7. Break line density and break mass flow rate.

The offtake model with stratified flow conditions and normaljunction option was active during RUN B. Now the break line voidfraction has a logical behaviour versus that of the hot leg (Fig.36). In spite of this, the calculated break line density is quitedifferent from the measured one, between 700 and 1600 sec. . Thisis another effect of the SG tubes water depletion (Fig. 37).

The break mass flow rate was again underpredicted before themeasured break uncovery time (Fig. 38). Now the simulation showsthe break uncovery, but much later than in the experiment (1400sec.).

The break mass flow step-wise evolution in RUN A is in RUN Breplaced by strong instabilities, likely due to unsimultaneousswitching between flow regimes in the adjacent volumes of thebreak piping (Fig. 39 and 39-bis). Perhaps a time step reductionshould eliminate some of these instabilities.

In both calculations, the code detected choked flow conditions inthe break as soon as it opened.

5.2.8. Primary system mass inventory.

Figure 40 compares the measured and calculated primary system massinventory. Although not accurately known from the experimentaldata minimum primary mass inventory was estimated to have occurredat between 1800 and 2200 seconds.

An obvious consequence of the break mass flow underprediction isthe overprediction up to 20 % of the mass inventory during 1500seconds.

Around 800 seconds began the S.G. tubes depletion (Fig. 37). Thisprocess culminated with the S.G. plena emptying between 1300seconds (cold side) and 1400 seconds (hot side) (Fig. 41).

The slow fall of the tubes liquid rised the hot leg level abovethe break line. This delayed the final break uncovery until 1650seconds in RUN A and 1400 seconds in RUN B.

Calculated core void fractions (Fig.42) reveal that the core wasnot uncovered, as observed in the experiment.

8

5.3 Selected items

5.3.1 Loop flow and natural circulation

In LP-SB-1 experiment, natural circulation was the only means forenergy transfer between the core and the steam generator.

The measured velocities (Fig. 43) indicate that naturalcirculation was established after the pumps coastdown (-50 sec.).At about 500 sec., the turbine meters showed zero velocity.

Figure 44 (2] indicates that break mass flow rate became largerthan thIGe hot leg mass flow rate at about 400 sec., and thissuggests some flow from the S.G. to the break after this time.This may be due to the blockage by vapor of the top of theU-tubes, and the resultant liquid draining from them.

The measured S.G. plena temperatures (Fig. 33) suggest that thethermocouples remained wet until -1100 sec. because of liquiddraining and from possible reflux condensate.

So, it appears that two-phase natural circulation finished at -500sec., due mainly to flow blocking through the U-tubes.

After this time, the system entered a reflux condensation mode,and the U-tubes, cold leg and hot leg piping successively drained.This cooling mode in the S.G. finished at about 1100 sec., whenthe secondary pressure became equal to primary pressure.

In our base calculation, the cold leg suddenly emptied at - 400sec., and the circulation ceased at about 700 sec. (Figures 45 and46). Inmediatly, the liquid velocity in the U-tubes indicated thatthe system entered the reflux condenKIsation mode (Fig. 47). Theprimary coolant-tubes heat transfer coefficient consistentlyincreased (Fig. 48). At about 1200 sec. the U-tubes emptied (Fig.37).

When the onset of reflux condensation was detected by the code theprimary coolant mass inventory was of 65% , very close to theactual inventory of 60%. These values are in the typical rangeencountered for experiments in several facilities (Semiscale, PKL,LSTF,...) [9].

5.3.2 Hot leg flow stratification.

In LP-SD-1, the intact loop hot leg fJlow stratified at -50 sec.(Fig. 34)(2].

In our RUN B simulation, the code detected stratified flowregime in ILHL at -320 sec.

9

The criterion defining horizontally stratified regime inRELAPS/MOD2 is that developed by Taitel and Dukler.[8]. It statesthat the flow is horizontally stratified when the vapor velocitysatisfies the condition that:

IVgl<Vgl

where

V ~~(Pf-Pg9)gag9A 2 (-OOVgl=-• ' P~)'(l-cosiO)

gl 2 pgDsin-a

The angle 0 is related to the liquid level h and the vaporfraction by:

h = D(l+cosO)2

rag= 0 - sin-a cosi

In Figure 43 we see the comparison between measured and calculatedhot leg fluid velocities. It is surprising to observe that, duringa short time around 50 sec., the calculated vapor velocity becamevery small, and even negative. However, the code did not predictany change to stratified flow regime during that lapse (Fig. 49).

Figure 50 shows the comparison between the calculated vaporvelocity and the Taitel-Dukler limit velocity [1] for RUN B. Asexpected, the limit velocity is higher than the vapor velocityfrom 320 sec. on. Around 50 sec., Vgl is very small. Figure 51plots the difference lVgl-Vgl and shows that the code neverdetected negative values of this magnitude.

We repeated the first 500 sec. of transient with a time step of0.01 sec. (rather than 0.1 sec.). Now, the above difference becamenegative during a few seconds, and,consistently, the code detectedstratified flow conditions (Fig. 52 and 53).

So, the code did not notice the change on flow regime because thetime step was too long.

In conclusion, the code predicted well a minimum of fluidvelocities at around 50 sec., and so, the onset of horizontalstratified flow in hot leg. After that, natural circulationestablished, and experimental and calculated velocities increased.The simulated flow changed back from stratified to bubbly, but theactual one did not.

Keeping in mind that hot leg density and velocity are reasonablyreproduced by the code, it is obvious that Taitel-Dukler model isunable to describe properly flow stratification in LP-SB-l.

10

6. CONCLUSIONS

The major conclusions are:

i) The code could not account for the liquid entrainment andvapor pull-through in the break tee modelled as acrossflow junction.

ii) It is necessary to improve the RELAPS/MOD2 choked flowmodel to avoid large instabilities in the break phasevelocities and thermal disequilibrium effects.

iii) The Taitel-Dukler model is unable to describe properlyhorizontal flow stratification in the LP-SPB-1experiment.

iv) Natural circulation was correctly reproduced by the code.

11

REFERENCES

1 READER, D L. "LOFT System and Test Description".NUREG/CR-0247, Tree-1208.

2 MODRO, S M et al. "Experiment Analysis and Summary Report onOECD LOFT Nuclear Experiments LP-SB-1 and LP-SB-2". OECD LOFTT-3205, May 1984.

3 CHEN, T H. MODRO, S M. "Best Estimate Prediction for OECDLOFT Experiment LP-SB-1", OECD LOFT T-3203, May 1983.

4 McPHERSON, G D. "Decay Heat Data for OECD LOFT Experiments",October 19E85 (Letter to OECD LOFT Programme Review GroupMembers).

5 MODRO, S M. CHEN, T H. "OECD LOFT Proyect ExperimentEspecification Document. Hot Leg Small Break ExperimentsLP-SB-l and LP-SB-2" OECD LOFT T-3201, April 1983.

6 PELAYO, F. "TRAC-PF1/MOD1 Post-test calculations of the OECDLOFT Experiment LP-SB-2" CSN/SEP/PROG/18/87. AEEW-R 2202,April 1987.

7 ALLEN, E.J. "TRAC-PF1/MOD1 Post-test calculation of the OECDLOFT Experiement LP-SB-I" AEEW-R 2254. August 1987.

8 RANSOM, V.H., et al. "RELAP5/MOD2 CODE MANUAL".NUREG/CR-4312.

9 "Compendium of ECCS Research for Realistic LOCA Analysis".NUREG-1230.

10 HALL, P.C. , BROWN, G. "RELAP5/MOD2 Calculations of the OECDLOFT test LP-SB-01". CEGB. GD/PE-N/544. November 1986.

12

TABLE 1

INITIAL CONDITIONS FOR EXPERIMENT LP-SB-1

MEASURED PREDICTED(RUN A)

PRIMARY COOLANT SYSTEM

Core T (K)

Hot leg pressure

Cold leg temperature (K)

Mass flow rate (kgs-1)

REACTOR VESSEL

Power level (MW)

STEAM GENERATOR SECONDARY SIDE

Liquid level (m)

Water temperature (K)

Pressure (MPa)

Mass flow rate (kgs-1)

18.5 ± 1.7

15.00 ± 0.08

557.2 ± 1.5

483.1 ± 3.2

48.8 ± 1.2

3.12 ± 0.01

535.2 ± 3.6

5.53 ± 0.05

25.79 ± 0.77

18.7

15.04

558.1

484.0

48.8

3.116

526.5

5.54

26.

PRESSURI ZER

Liquid volume (m3)

Steam volume (m3)

Water temperature (K)

Pressure (MPa)

Liquid level (m)

0.625 ± 0.001

0.377 ± 0.001

615.8 ± 8.2

15.06 ± 0.11

1.072 ± 0.002

0.598

0.403

615.1

15.0

1.126

BROKEN LOOP

Cold leg temperature (K) 555.7 ± 6.3 558.02

13

TABLE 2

OPERATIONAL SETPOINTS FOR EXPERIMENT LP-SB-1

Action ReferenceMeasuredSetpoint

Small-break valve Opened Time 0.

Reactor scrammed

Main feedwaterShut off

Main steam controlvalve started toclose

Primary coolantpumps tripped

HPIS Flow)initiated

Auxiliary feed-water initiated

Auxiliary feed-water terminated

ILHL Pressure (MPa)

ILHL Pressure (MPa)

Time after reactorscram (seconds)

ILHL Pressure (MPa)

ILHL Pressure (MPa)



14.57t 0.03

14.57t 0.03

2.± 0.2

11.12t 0.03

8.24± 0.03

62.±t 0.2

1864. 2 ± 0.8

14

TABLE 3

LP-SB-1 CHRONOLOGY OF El

EVENT

Small-break valve opened

Reactor scramed

Main feedwater shut off

Main steam control valve started to close

Main feedwater isolated

Main steam control valve fully closed

Primary coolant pumps tripped

Pressurizer liquid level below indicatingrange

HPIS flow initiated

Primary coolant pump 1 coastdown completed

Primary coolant pump 2 coastdown completed

Subcooled blowdown ended

Auxiliary feedwater initiated

Break started to uncover

Primary system pressure becomes less than

secondary system pressure

Auxiliary feedwater shut off

HPIS + pump injection flow rateexceeded break flow rate

HPIS flow rate exceeded break flow rate

IENTS

PLANT

0.0

1.4

1.4

3.4

3.8

15.

24.6

34.6

41.4

42.6

43.0

57.5

63.4

715.

1077.

1864.8

1998.0

RUN A

0.0

2.05

2.05

4.1

5.

4 17.

26.85

33.

44.

48.

48.

76.

62.05

1050. *1650.*

RUN B

0.0

1.9

1.9

3.95

4.3

13.95

25.9

32.

48.

49.

49.

55.

64.

880.*1440.

1450.

1865.4

* Collapsed level under the HL pipe midplane

15

I mA wdmaw

doU4o how

I

iacI.4awl sioEdoI,waive W.I.*

-4- ~ pm~ heneca

Figure 1. LOFT PiPing SChelidtic.

Intact loop Broken loop

-J

Figure 2. Axonometric projection of the LOFT system configuration forExperiments LP-SB-1 and LP-SB-2.

I--b

mI-

z00

0

0

0

I-

10

I-

r-1>04z

z

al~m M""MORAO

PRESSURIZER

4I1as1

IMEAKREACIOR

VESSEL

WEY'126 C0,11,4)"46

,MY- sRaNAlc Wilie JUmcciaON3fl MUCIOM On VALVEi

am GW SlRUCItAE-0- CROSS FLW JAiCIIO"

lso

C9~14

I

I!

In

A

m

Vs

z00ti

o0

70

Kr-

0

4

0

c-

z

w

!~PRESSURIZER

LisSIts

BREAKREACTOR

VESSEL

KEY&i2s COMPONIENT

Jim~ Compows 01

PMP

ImANCH WITH JUNCtIONS

[JUNCTION ORt VALVEa=HEAT STRUCTURE-4-- CROSS FLOW JUNCTION

trs

m

i00

18.80

0

(L

(1)(1)LLS

a.

13.00

ci ci ci ci ci ci ci ci ci ci a ciMP loseBIBoB

7.00

4.80

a 20 48 80 8B lea

TINE (SECONDS)

LP-SB-1 STEADY STATE. RUN A

FIG. 5

IV.'

"I

-A

L&I-J

Na.U.

C',

3.50

3.08

2.so

2.88

1.58

I .89

0.58

8.08

30 19 a0s

M VAR.I SD LEVEL£ VAR.2 PZR LEVEL

0 28 40 as 80

TIME CSECONDS)

LP-SB-! STEADY STATE. RUN A

Iee

FIG.6

I,,-.-•

' 8.880 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

J VAR 77

'-4o

' ' 5.88*1.-

*. I,,,,

! 8

ci

TIME CSECONDS)

LP-SB-1 STEADY STATE. RUN A. FIG.7

CA)

I 5.09

i II II II I

I I

I C

0 4

I ~a1 -J

I ,.0 II1~~

I I

; 1 ILVLVJ 4M1

1

!

too0 EsoTIME CSECONDS)

LP-SB- I STEADY STATE. RUN A. FIG. 8

78888.88I I I !, ] ! I i

6888.8 [ . - : . . - --. ..- '. .. ..

0PS- . RN1

4888.88 .... J.. -....... ..........I• J a I I I p

, I I I•, I ".,

0 208888.88 j ... . .7 r ]-. . .f • - . . i . ... f " ...

I, .e•l._.... . .. . . .... .. . * --J - l*-- -- - i

ti-. -. -.. 'II,., i _i ...... , . .. 4.. . .. ....

8826 226 425 625 862 1825 1225 1425 1625 1825 2025

TINE (SECON4DS)

LP-SB-1. RUN A

___ _ ___ ____FIG. 9 1

0.08

7.08

6. 8

'

la

cz

imc

6.08

4.08

* gm-384

3.06

2.08

1.88

8.00

2S 525 1l2o

TIME CSECONDS)

LP-SB-I. RUN A

1625 2825

FIG. 10

L.".W VLVSTZH "f0A, CV-U4-418X CV-f W4-OW

0...

0...II•3O•

I sIL

40

w

0.48

9.28

0. SO

-8.2626 45 es as I"5 i25

TIME CSECONDS)

LP-SB- I. RUN A FIG. ii

1S."9

I6.99

rI~

I.

at

9zCL

12.99

9.99

8.09

1M PE-Pt-4A P Ismiuo

28256.991

25 625 ton5 I525

TIME (SECONDS)

LP-SB- I. RUN A FIG. 12LP-SB-t. RUN A FIG. 12

. 668.86mdL0

e

TEHPF MOMgSS"TE-pc-ZATE-PC-28TE-PC-2CTE-PI 80-32-1

676. GO

$86..86II%) L&J

I.-

CDwL

6584.8

648.88

638.86

620.8825 526 1826 1625 2025

TIME (SECONDS)

LP-SB- I. RUN A FIG. 13LP-S -ti RUNA iGi 13

S9S..9M TEMPF 18M1 6l60A TE-PC-IAX TE-PC-tB0 TE-PC-ICI TE-P 189-28-2

566.66

(0

I-J

la

-s

CoC.)

566.61

456.66

498.16

25 525 1025 1525 2625

TIME (SECONDS)

LP-SB- I FRUN A FIG.14

1.6eM PTl2S,-1S4CIU-8lt8/,HL NMJO 6MUCKG/W

(A

0•I

IJi-

saw

08-i

as sas Ism5 2m2

TIMIE CSECOWDS)

LP-SB- I . RUN A FIG. 15LP.S.-i.iRUNlA FnG. i

609.99M R"O IMIwSA DE-PC-taS

I- ^ -'

z

_J0

us

26 225 425 825 625 1625 1225 1425 1026 1826 2625

TINE (SECONDS)

LP-SB- I. RUN A FIG. 16LP-SB-1. RUN A FIG. 16

3.U

r')

I

La

CD

C2

41-

2.W8

1.08

8.88

gZ VELS :IX F-PC-IA

FIG 17

-1.88e L25 525 102% 1526

TIME (SECONDS)

L_ LP- SB -I._R UN A

as.o"

o."11

I

CA)CA)I~ 40

s-s0-

C0

4".6S.

20..S.

No Imism

OtK-PC-IA

* K-.c-S* K~[.f-pc-a

252

jwiq"7wm7SG.

25 I6 ton 1625

TIME CSECONDS)

LP-SB- I. RUN A FIG. 18LP-SB-I. RUN A FIG. 18

3.08M QNKRLVA2 6SS,L gAIRLVAR gin

I2.58

Ik i

'S

2.88

(A) -J

LLI

Lii0o

a An

1.98

8.58

8.88

25 525 1826 1525 2825

TINE CSECONDS)

LP-SB-l. RUN A FIG. 19

eng."M no tuteiuiA Dtc-343s

em.g

I7t9.99

6".i09

(A)

I

(I,

s69.t.

I-14

wC

4s .09

83.99

23S.09

1l .. 9.

IS... I

2S 625 126 162S 2925

TIME (SECONDS)

LP-SB3-I. RUN A FIG. 20

69.8.6MUD 1041IUM,L Roo ImslSI

I889.68

I

I-'

480.80

'-4C0z'Al

290.60

0.88

25 525 1025 1525 2025

TIME (SECONDS)

LP-SB-1. RUN A FIG. 21

12.89

9.08

I

CA)-4

1

hign

0.80

3.09

A NLdJtawe

28250.00

2s 525 1S25 1525

TIME CSECONDS)

LP-SB-1 . RUN A FIG.22LP-SB-1. RUN A FIG.22

S2.08M VELFJ INsUse,L VELGJ INSSIMS

28.68

CAhLI

15.88

LI-CC0do

IS ."

S. so

S.8s25 225 425 825 825 1825 1225 1425 1026 1825 2825

TIME (SECONDS)

LP-SB-1. RUN A FIG. 23

188e8.88M VAR.71 R V HASSA VAR.72 PCS MASSX TIASS

toI

8098.80

>- 4999.084000. eS

0)')-

2000.00

8.88

25 225 425 825 825 1825 1225 1425 1825 1825 2825

TIME (SECONDS)

LP-SB-I . RUN A

FIG. 24

tsse88...

IS.....

I 4fl3B* B*I W m

I

CPUTZIME

I

I

0A

0Czahi0

hiz6-

0.

msBO.S so ~ -

Gem6. so;

46949. _ _

.1Oa sea tseS 1566

lIIE (SECONDS)

LP"SB-1. RUN B FIG. 25

4m

qjE

i .. I'~. ..- ~

I-

8.IaI-

1.1

0.V0

0.04

I.I II:Ii fli~

0.0o.i

'508

0.80

LOOOa see 1f0e S-' u ki

TIlL (SEtZ3NDS)

L P -- 3-I R U N.6FI. FIG.26

64."

48.88

I bi

I-

-AJ

44

bi

20. B0

* ~zt~pg 1:

-. ilom

2 1

Is...

S."

0 56" lees is"5

TINE CSECONDS)

LP-SB-I. RUN e FIG. 27

Is.".PE-PC-092

~~~~ 1099169______6

IS.8

II9.

3.6_________ ________ ________

S 501 tees 1566 n6 2560

TINE (SECONDS)

LP-SB-1. RUN B FIG. 28

suo0c0

Vi

AhI

rkmpf ;U9218086

It-?C--,IATE-P$;-RoIf -Pt -ecIE-e4 as -.32-1

V. 0 - IO

I

LAI

ILl

I.-

CD

LIJ

558.0 - -

540. 00

S38.00 -.

0 10f0 !588 2088 2M0

TIME (SECONDS)

LP--SB-I. RUN B FIG. 29

620.00

6ag.98

,n

I

co

Ico

I-

5688.9

568.90

a)TEMPS 4169S3S89A TEMPF 416039363X TE-PI3g-829

2s66

54.00

52S.00

0 5ee Igoe sees 2608

TIME (SECONDS)

LP-SB- I . RUN B FIG. 30LP-SB-1 . RUN B FIG. 30

600 G0&,L

x

rui'r .6001B8OaTE-Pr.-IATL-Pc- I j

TE-P139-29-2

i

O•

I Lai

w

-AJ

-j40

s!s000

r.060C

460 00

11

8/

'II'

1;

I'

480 00 0 ..0 1566sea t00$ 2080 2SOU

TIM E (SECON.)S)

L?~I3- RUN a3FG.3 FIG. 31

899.o9M rTo"e ieusisA TI-pt l-IS-2

-4I

Pi

bi9-

0.j

C31C.)

599.99

499.09S Soo logo 1609 2099 Mse9

TIME (SECONDS)

LP-SB-t. RUN B FIG. 32

586.88w

0

TEIPF 114818088TEKPF 11681BS8OYE-SB-ITE-$6-2A

570.88

568.88

I L&J

LuCL

Liu-jdi.

CD

550.08

540.80

538.00

520.08

a 508 1888 1s58

TINE (SECONDS)

LP-SB-I. RUN B

2880 2508

FIG.33

660.O0

A

0

RHtO ISvone6MAg-ru-26501-Pt-2ADE-PC-28

890.00

IMI-

lbý

3-

#a

4010.S9

201.99g

0.09S IOse 160 20e ZS6o

TIPiE (SECONDS)

LP-SB- I. RUN 0 FIG. 34... . SII-tI. RU B FIG. 34

988.08

608.80

600.00

I-

I I&IAJL ait.z - L. .~.L... - Li.m.J.

-A

RH :8204

.I

j~fl~~vwwr-qrwp

U _______________

I ~~'**~~* ~

w. .P

I0n0

w

C-

w

0)

-j

688.08

480.80 i-,.---.-.------.~~.-.I I..-.

1 I

$60.60 ~ . -. . -- ..

t i

a rho I I!

I.

~50ew *.•

Ut see

TIME (SECONDS)

LP-SB-I. RUN B

2000

FIG. 35

1 .88

8.88

U'1I z

i..

C2

8.68

8.48

A YODrOJ 1S23I19fl

8.29

8.88

8 588 1889 1588 2988

TIME (SECONDS)

LP-SB- I . RUN B FIG. 36LP-SB-1. RUN B FIG. 36

I

3.00

2.50

al CNTRLVARa SOOA CNTRLYAR Sig

I

ILaJ

-j

L*o

2.00

I .50

I1.00

0.50

0.00

a s00 1000 1588 2080 2580

TIME (SECONDS)

LP-SB--1 RUN B FIG.37

t2.O,

g.09

_____________________________________________ I ___________________________________ ________________________

PIVFLOIJ 1I3I1UBD0

I

CDbew~

09go

a."

LL I

S."

4"I

S. "

O I m Iis" a$" 2909

MEII (SECONDS)

LP-SB-t. RUN B FIG.38

(.n

cc

.4

lb.

12.88

19.98

9.98

8.80

7.80

4.88

2.80

1."8

8.889

a so a' a

MFL=9l 141616M~& FLOME IteasIew

9 sea 1888 1588

TIME CSECONDS)

LP-SB-I. RUN B

2888 2588

FIG. 39

LP-SB-1 run BLOFT CSN

4

3

-r ~- --

II

0rz4

2

1

0

-1

-2

Ift

tUILJI !Ii-

I

-3

-40 500

1000L500 2000

TIME (SECONDS) FIG. 39 MIS)

c88e. se4U A

RUN B

Ln)

s 5I6 186 1566 2660 2600

TIME (SECONDS)

FIG. 40Lr- o-i.

I .09SVOIDG 1149198I

A VOIDS IISSISSUSX VOZDO 116190051

0.80

Ifun z

0I-II-C.)

tA~

M0

8.68

ow

1.40

0.20

0.880 see lose Isla 2e9e 2598

TIME (SECONDS)

LP-SB-I GRUN B FIG. 41

1 00Vloz6 250818008V V012t 23824808

X V0Z10 2S008S800 VOZM6 2300408

V, vo129 230O8S808*VaZ)G 25008088

0 .80 -

I

I

bIP

Q.)

I-J

0

0.00 -.

0 40 ..

0.20 .

0.90

0 S00 1006 Iss0 2000 0 ou

rIML~ (SECOND~S)

LP-SB-1 R u 164 FIG. 42

S...M VII.? meetNg& VIEW ISWISIX FE-Pc-ZAo -e-a*FEt-PC-Ic

(0

wOil

0

0umb

'4

2.00

1.09

S...

-1 .09

0 lee 299 3ss 4.3 see

TIME CSECONDS)FIG.43

LP-SB- I . RUN B

Mass flow (kg/s)

I0 u'0 5=C

00

01

0

0C

-4

Lid-~~0

0

00

-oX

z-I

I--o

-n5.A4Pb

00

uso

Mass flow (Ibm/s)

9 30

0.25

q

'- 2:

~8IIJ

lAB-J

a&i-J

-l04,

0.29

0.1S

S. IS

0. eS

S. SO

-i. I.

-4

M01UffLVApi 4

|i

.I1L..~

1*

I

a1k I

FIG.45

a See oe80 1560 2090

TIME (SECONDS)

LP-SB-i . RUN B

3.88

0FIE~-PC -2j

r')

w

'-4

s-D

2.88

6 .888.68s

2588

FIG.46 1

-1.88 La sea less I1C88 288"

TIME (SECONDS)

L P- SB -1I. R UN B_____

6.00 __ _ _ _ _ _ _ _ _ _ _ _ _ _

Y ELP 11691090A EL6 116914MXYELP 11659408MVELS 11SO49M0YELP IlISQVlOO

4.0 __ ___ ___ ___ ___ ___ ___ __ ___ ___ ___ _ _ ___ ___ ___ * ELS 116 INN

CA)'

40

MA

!lob o

loeIrae29

ME(ECNS

LP-SB-1 . RUN B FIG. 47

0.28

I

Im

IA-

CD)

CDw)

t

4

8.15 -..

8.10 ..

8.88

HTHTC 6000180" ITHTC 6880288

X HTHTC 68803800 HTHTC 6808408

2580a see 1580 2880

TIME (SECONDS)FIG.48

LP-SB- R . RUN B

OR

I

of

2:

W

bi

:a

-4ba.

12.99

t t.se

1s. O9

* .68

e.g.

7.98

6.00S

6.g.

4.10S.

3...

2.68

1-.9

F- -; _ - -z _ -: _- .= - -Z -

~--1

-4

-4

~1

-i

-1

-4

6S.

MFLORES urn, US

.1 U El LI El CI CI C CI U CI U CI

I I I I

a too 299 39"

TIME (SECONDS)

LP-SB- I.

499

FIG.49

2.08

I I

d I

I i

M MITz,L VMS lessum

I

oy)

:0

1 .88

S...

-8.88

0 1I0 280 388 488 580

TINlE (SECONMDS

LP-SB-I. FIG. 50

LP-SB--1 run BLOFT CSN

I

Ell

q

0.01

0.009

0.008

0.007

0.006

0.005

0.004

0.003

0.002

0.001

0

-0.001

-0.002

-0.003

-0.004

-0.005

-0.006

-0.007

-0.008

-0.009

-0.01

0 200 400

TIME (SECONDS)TIME STEP =0.1 SEC. FIG. 51

LP-SB-1 run BLOFT CSN

I

I

P.p

1

0.80.6l

0.7

0.8

0.5

0.4

0.3

0.2

0.1

0

-0.1

-0.2

-0.3

-0.4

-0.5

-0.6

-0.7

-0.8

-0.9

-1

0 20 40 s0 80 100

TIME (SECONDS)TIME STEP =0.01 SEC FIG. 52

L U-SB- LJi.LOFT ('ST-

11

10'

I

"4

0

0•0

"4

rz0

a

7

5

4

3

2

1

0

-1.. . 1_ _;

0 20 40 60 8O 100

TIME (SECONDS)TIME STEP =0.01 SEC FIG. 53

APPENDIX I

DISKETTE CONTAINING THE RELAP5/MOD2 INPUT DECK FOR LP-SB-1

70

APPENDIX II

TAITEL-DUKLER CRITERION FOR HORIZONTAL FLOW STRATIFICATION

RELAP5/MOD2 employs a criterion developed by Taitel and Dukler todefine horizontally stratified flow regime. It applies tohydrodynamic volumes with an inclination angle s 150.

Taitel and Dukler proposed that the transition from stratifiedhorizontal to nonstratified flow regimes occurs as a result of theinstability of a solitary wave on the liquid layer. They derivedthis condition of instability as :

IVg I>Vg1

where

1

[ (P~) gaA 1Vgl C pgDsin J

and

C 1- (1-coso)

pf and pg are liquid and vapor densities, respectively. g is the

acceleration of gravity. A and D are, respectively, the crosssection area and the diameter of the horizontal volume.

The angle -5 is related to the liquid level, h, and the voidfraction a, by the relationships (Fig. 54)

h D2- (1+cos*)

an = O-sin~coso

If the horizontal stratification condition

IVg I<V g

is met, then the flow field undergoes a transition to horizontallystratified. If it is not met, the flow field undesgoes atransition to the bubbly, slug, or annular mist flow regime.

71

APPENDIX III

HORIZONTAL STRATIFICATION ENTRAINMENT MODEL IN RELAP5/MOD2

Under stratified conditions in horizontal components, the void

fraction of flow through a junction may be different from the

upstream volume void fraction [8]. Consequently, the regular

donoring scheme for junction void fraction is no longer appropriate

because vapor may be pulled through the junction and liquid may

also be entrained and pulled through the junction. The

correlations describing the onset of vapor pull through and liquid

entrainment for a centrally oriented junction are given hereafter.

The incipient liquid entrainment is determined by the criterion

that

Vg Wge

where V is the vapor velocity in the junction, and

V 3.25 - h g(Pf-P 9)(H - h)Vge d 3.2 9

pf and pg are liquid and vapor densities, respectively. g is theacceleration of gravity. D and d are, respectively, the horizontal

volume and junction diameters. h is the liquid level in the volume.

The condition for the onset of vapor pull-through is determined by

Vf >Vfp

where Vf is the liquid velocity in the junction, and

Ih D 5/2- ) d 1/2

Vfp = 3.25 d 2Pf P)

72

For liquid entrainment, the junction liquid fraction, afj, isrelated to the donor volume liquid fraction, afk' by theexpression

aj '•fk 1.-exp-C Vg/Vge - 0 2 /

where VgI is the Taitel-Dukler limit velocity (Appendix II). Forvapor pull-through, the junction void fraction is given by

agj = "gk 1 - exp -C2 Vf/Vfp - 10g2/ 2l

The constants C1 and C2 are obtained by comparisons of codecalculations with experimental data. Currently, C1 and C2 are both

equal to 1.

73

NRC FORM 335 U.S. NUCLEAR REGULATORY COMMISSION 1. REPORT NUMBER12-891 (AsleWmd by NRC. Add Vo.. S•P., F.,NRCM 1102. and Addendum Nsumerl, If amy.)3201,3202 BIBLIOGRAPHIC DATA SHEET NUREG/IA-0074

(See instructions on the reverse, ICSP-LP-SB-1 -R2. TITLE AND SUBTITLE

RELAP5/MOD2 Post-Test Calculation of the OECD 3. DATE REPORT PUBLISHED

LOFT Experiment LP-SB-1 MONTH YEAR

April. 19924. FIN OR GRANT NUMBER

A46825. AUTHOR(S) 6. TYPE OF REPORT

J. Perez, R. Mendizabal Technical7. PERIOD COVERED (nvlusee Dates)

8. PERFORMING ORGANIZATION - NAME AND ADDRESS (it NRC, provide Division. Office or Region, U.S Nuclear Regulatory Commission, and mailing adrerss If contractor, prvide

nwa and mailing addres.)


9. SPONSORING ORGANIZATION - NAME AND ADDRESS (If NRC. type -Sense as above"; if contractor, provide NRC Division. Office or Region, U.S Nuclea Regulatory Commission,&nd madlkn address.)


10. SUPPLEMENTARY NOTES

11. ABSTRACT (2w wond or k,•)

This document presents the analysis of the OECD LOFT LP-SB-1 Experiment performed bythe Consejo de Seguridad Nuclear of Spain working group making use of RELAP5/MOD2 inthe frame of the Spanish LOFT Project. LP-SB-1 experiment studies the effect of anearly pump trip in a small break LOCA scenario with a 3-inch equivalent diameterbreak in the hot leg of a commercial PWR.

12. KEY WORDS/DESCRiPTORS (List words orphrases that willssist reseachers in locating the repor.) 13. AVAILABILITY STATEMENT

Unl i mi tedICAP, RELAP5/MOD2, Post-Test, Calculation, OECD, Loft Experiment, 4, .SECURITYCLASSIFICATION

LP-SB-1 (iS Page)

Uncl assi fied(This Report

Unclassified15. NUMBER OF PAGES

16. PRICE

NRC FORM 335 42-891

THIS DOCUMENT WAS PRINTED USING RECYCLED PAPER

NUREG/IA-N74 RELAPS/MOD2 POST-TEST CALCULATION OF THE OECD LOFT EXPERIMENT LP-SB-1 APRIL 1M9

UNITED STATESNUCLEAR REGULATORY COMMISSION

WASHINGTON, D.C. 20555SPECIAL FOURTH-CLASS RATE

POSTAGE AND FEES PAIDUSNRC

PERMIT NO, G-67

OFFICIAL BUSINESSPENALTY FOR PRIVATE USE, $300

120555063572 4 1AN1CIUS NRC-RESDIV OF SYSTEMS PESEARCHBRANCH CHIEFREACTOR & PLANT SYSTEMS BRNL/N- 353WASHINGTON DC 20555