Begell House Inc.
Multiphase Science and Technology
MST
0276-1459
3
1-4
1987
TWO-PHASE FLOW MODELS: THE CLOSURE ISSUE
3-30
10.1615/MultScienTechn.v3.i1-4.10
J. A.
Boure
Commissariat à I'Energie Atomique, Department des Reacteurs a Eau, Service d'Etudes Thermohydrauliques, Centre d'Etudes Nucleaires de Grenoble, Grenoble, France
The closure issue in two-phase flow modelling is discussed from various complementary points of view. It is shown that closure of engineering two-phase flow models implies several kinds of closure laws.
Some of these laws are of the same nature as the closure laws of single phase flow models. They include equations of state, correlation functions, field viscous and conduction laws and transfer laws. The other closure laws are specific to multiphase flow models. They account for the interface structure and the interface dynamics, and could be called topological laws. They play an essential part in the model behaviour. Except for the equations of state, none of the engineering closure laws is constitutive in the standard sense. The implication of the choice of closure laws are recalled and the closure methods are briefly discussed.
The most important topological laws are the void fraction and the interfacial area topological laws. The void fraction topological law, whose essential role is often overlooked, is discussed in some detail.
INTERFACIAL AREA MODELLING
31-61
10.1615/MultScienTechn.v3.i1-4.20
Mamoru
Ishii
Therma-Hydraulics and Reactor Safety Laboratory, School of Nuclear Engineering, Purdue University, 400 Central Drive, West Lafayette, IN 47907, USA
The interfacial transfer terms are one of the important aspects of the two-fluid formulation modelling. However, there exist considerable difficulties in this area both in terms of experiments and modelling. The first step in the right direction is to consider that the interfacial transfer terms are proportional to the interfacial area concentration and driving force. This approach effectively separates the first order geometrical effect on the interfacial transfers from the local driving forces. Hence more mechanistic modelling of the interfacial area concentration, the available experimental data, measurement methods and modelling are reviewed and new directions are indicated.
INTERFACIAL FRICTION MODELLING
63-83
10.1615/MultScienTechn.v3.i1-4.30
Graham B.
Wallis
Thayer School of Engineering, Dartmouth College, USA
Interfacial friction shows an important influence on the predictions of two-fluid model. This study discusses a few simple situations, such as one-dimensional and steady flow conditions for which correlations for interfacial frictions have been well established. Empirical terms derived from specified experiments are found necessary due to the impossibility of describing all the details of flows (e.g. interface geometry and phase interactions). More works are needed in future to develop more systematic methods and models representing the physics of the flows, and data banks containing measurements of enough parameters are then required.
INTERFACIAL MASS GENERATION RATE MODELLING IN NON-EQUILIBRIUM TWO-PHASE FLOW
85-127
10.1615/MultScienTechn.v3.i1-4.40
G.
Yadigaroglu
Swiss Federal Institute of Technology (ETH), Nuclear Engineering Laboratory ETH Zentrum, CH-8092 Zurich, Switzerland
A.
Bensalem
Nuclear Engineering Laboratory, Swiss Federal Institute of Technology (ETH), Zurich, Switzerland CH-8092
The various approaches used for modelling interfacial mass generation in two-phase flow are reviewed. The mass generation problem is strongly related to the questions of interfacial heat transfer and to the determination of the interfacial area. The available literature is discussed and a classification is made according to the nature of the heat transfer regimes or mechanisms (growth of bubbles, evaporation of droplets, post-dryout heat transfer, subcooled boiling, etc.) as well as according to the nature of the model used (mechanistic models, relaxation, profile-fit and empirical methods). Whenever possible, the discussion is directed toward the definition of the closure laws needed for the modern two-fluid thermal-hydraulics codes.
EXPERIMENTAL DATA SETS FOR VALIDATION OF PREDICTION METHODS IN MULTIPHASE FLOW: INTRODUCTION AND THE NATURE OF THE EVOLUTION OF MODELS
131-144
10.1615/MultScienTechn.v3.i1-4.50
Geoffrey F.
Hewitt
Department of Chemical Engineering & Chemical Technology, Imperial College of Science, Technology & Medicine, Prince Consort Road, London SW7 2B Y, England, UK
This paper describes the background to the selection of the experimental Data Sets given in this volume of Multiphase Science and Technology; the work on the selection and approval of the Data Sets was carried out under the aegis of the International Workshops on Two-Phase Flow Fundamentals (supported by the US Department of Energy). The evolutionary nature of modelling methods for multiphase flows is illustrated by considering models for droplet deposition in annular flow.
EXPERIMENTAL DATA SET NO. 1: PRESSURE DROP AND ENTRAINED FRACTION IN FULLY DEVELOPED FLOW
145-154
10.1615/MultScienTechn.v3.i1-4.60
D. G.
Owen
AERE, Harwell Laboratory, UK
Geoffrey F.
Hewitt
Department of Chemical Engineering & Chemical Technology, Imperial College of Science, Technology & Medicine, Prince Consort Road, London SW7 2B Y, England, UK
Experiments in a wide range of conditions are carried out at the Harwell LOTUS air-water vertical rig (31.8 mm ID) to provide data set for fully developed flows. Data for pressure drop and (in annular flow) the entrainment fraction are presented and compared with some published correlations. An interesting phenomenon, observed for the first time, is the existence of a region in which pressure gradient increases only slowly with increasing air mass flux in the annular flow regime. Clearly, the data presented here suggested that purely empirical correlations are incapable of predicting the nature of the pressure gradient and none of the correlations are particularly satisfactory for prediction of entrained flow fraction (or liquid film flowrate).
EXPERIMENTAL DATA SET NO. 2: CHARACTERISTICS OF BUBBLY FLOW
155-162
10.1615/MultScienTechn.v3.i1-4.70
Peter
Griffith
Department of Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139 USA
Bubbly flow experiments are conducted in a 25.4 mm ID pipe to investigate the behaviour of friction factor, radial void fraction and the momentum flux. The purpose of this work is to present experimental data on bubbly flow which can serve as a standard against the prediction models. Despite the fact that the homogeneous model produces reasonably good predictions of pressure drop, improved models are required taking into account the radial distributions of bubble velocity and void fraction.
EXPERIMENTAL DATA SET NO. 3: DEVELOPING ANNULAR FLOW
163-183
10.1615/MultScienTechn.v3.i1-4.80
Geoffrey F.
Hewitt
Department of Chemical Engineering & Chemical Technology, Imperial College of Science, Technology & Medicine, Prince Consort Road, London SW7 2B Y, England, UK
Studies of the nature of the flow development are performed for air-water vertical annular flow. Two forms of liquid injector (porous wall and central axial jet injectors, respectively) are employed in the experiments. Their effects on development of flow parameters (e.g. entrained fraction, film thickness and pressure gradient) are further analysed and compared with the published correlations. Poor agreement is observed which is ascribed to the difficulty in modelling the development of the interfacial structure, in particular the disturbance waves and the entrained droplets.
EXPERIMENTAL DATA SET NO. 4: PHASE DISTRIBUTION IN A TRIANGULAR DUCT
184-231
10.1615/MultScienTechn.v3.i1-4.90
Lateral phase distributions and phase velocities are measured for air-water vertical flows in a triangular duct (hydraulic diameter DH = 39.4 mm). The highest liquid velocity and largest void fraction are seen in the open region of the triangular section. The lateral 'void drift' phenomenon is clearly observed and found independent on liquid flowrate and/or wall induced turbulence. Experiments are also carried out for the various superficial liquid and air velocities to investigate the developing length (L/DH) required for a fully developed void distribution. The L/DH is found to increase as the superficial liquid velocity increases and decrease as the global void fraction increases.
EXPERIMENTAL DATA SET NO. 5: AIR-WATER COUNTERCURRENT ANNULAR FLOW IN VERTICAL TUBES
232-270
10.1615/MultScienTechn.v3.i1-4.100
S. C.
Lee
Yeungnam University, Kyungsan, Korea
S. George
Bankoff
Department of Chemical Engineering, Northwestern University, Evanston, Illinois 60201, USA
A wide range of experimental data is obtained from air-water countercurrent annular flows in vertical Lucite tubes of two sizes (25 and 51 mm, respectively). Parameters, such as the total pressure drop, pressure gradient within the tube, film thickness, liquid fraction and flooding point are presented and analysed. Also, a transition from a smooth film to a rough film and finally to a dry tube as the gas flowrate increases is illustrated.
EXPERIMENTAL DATA SET NO. 6: PRESSURE DROP IN HORIZONTAL FLOWS
271-284
10.1615/MultScienTechn.v3.i1-4.110
Dieter
Steiner
Institut fur Thermische Verfahrenstechnik, Universitat Karlsruhe (T. H.), Germany
The behaviour of frictional pressure gradient for horizontal flows of refrigerant 12 (R12) within a cirucular smooth tube (12 mm ID) is investigated. The range of parameter studies covers the flow structures with stratification, which provide a useful insight in the design of direct expansion evaporators for air conditioning equipment. Compared to the current experimental data, the predictions from available models are found to exhibit ± 30% relative deviations.
EXPERIMENTAL DATA SET NO. 7: STRATIFIED FLOW, PART I: LOCAL STRUCTURE
285-301
10.1615/MultScienTechn.v3.i1-4.120
J.
Fabre
Insitut National Polytechnique de Toulouse, Institut de Mechanique des Fluides (UA CNRS 005), Av. C. Soula 31400, Toulouse, France
C.
Suzanne
Institut de Mecanique des Fluides de Toulouse, France
Lucien
Masbernat
Institut de Mecanique des Fluides de Toulouse, France
This study gives a kinematic description of stratified gas-liquid flow in a horizontal or weakly inclined channel of rectangular cross-section and displays some specific behaviour occurring in wavy flows. Systematic measurements of the components of the mean velocities and Reynolds stresses are performed under careful controlled inlet conditions. The experimental data show an important mean secondary flow in the cross-section; the existence of transverse motion is also proved by the non-linearity of the shear stress profiles. In addition, the asymptotic behaviour of the mean velocity and Reynolds stress components near the wall and near the interface is also observed.
EXPERIMENTAL DATA SET NO. 8: STRATIFIED FLOW, PART II: INTERFACIAL AND WALL SHEAR STRESS
302-315
10.1615/MultScienTechn.v3.i1-4.130
Lucien
Masbernat
Institut de Mecanique des Fluides de Toulouse, France
C.
Suzanne
Institut de Mecanique des Fluides de Toulouse, France
J.
Fabre
Insitut National Polytechnique de Toulouse, Institut de Mechanique des Fluides (UA CNRS 005), Av. C. Soula 31400, Toulouse, France
The closure of the two-fluid models requires the knowledge of wall and interfacial frictions; however, limited understanding is available. This study therefore revisits this problem by conducing a variety of experiments in an air-water duct flows (see part I). The interfacial and wall shear stresses in gas and liquid phases, together with average value of the RMS wave height and wave spectra are measured and presented in detail. The wall shear stress in two-phase flows is found to be well predicted by the classical laws used in single-phase flow. However, the interfacial friction factor in a wavy-interface flow shows an increasing and then flattened behaviour as the gas velocity increases, which cannot be predicted by the previous models. A basic theory is then proposed based on the hypothesis on wave number spectra in internal flows to interpret this observation.
EXPERIMENTAL DATA SET NO. 9: DIVIDING FLOW IN A TEE JUNCTION
316-347
10.1615/MultScienTechn.v3.i1-4.140
The phase separation process is investigated based on the experiments of Seeger (1985), carried out in a very detailed and complete set of air/water and steam/water conditions in a horizontal tee. The effect of branch pipe’s orientation is also studied. Nearly identical trends are shown for the steam/water and air/water flows through a horizontal branch, which imply that nonequilibrium flashing phenomenon is of little significance during phase separation in a tee. Flow regimes and gravity effects are found to play an important role in phase separation in vertical branch pipes. Also, the data presented in this paper are the most comprehensive taken to data and indicate phenomena which represent a significant challenge to the predictive capability of current computer code.
EXPERIMENTAL DATA SET NO. 10: DOWNFLOW CONDENSATION
348-359
10.1615/MultScienTechn.v3.i1-4.150
D. G.
Owen
AERE, Harwell Laboratory, UK
Geoffrey F.
Hewitt
Department of Chemical Engineering & Chemical Technology, Imperial College of Science, Technology & Medicine, Prince Consort Road, London SW7 2B Y, England, UK
Condensation heat transfer results are presented derived from work carried out by NASA for the condensation of high velocity steam in downflow through a 7.44 mm ID copper tube. Overall parametric data, such as wall and coolant temperatures, local pressure, heat flux, quality and heat transfer coefficient are included and analysed in this paper. The flow development effects are shown to be significant in the determination of local characteristics (e.g. pressure and heat flux). Thus, a complete prediction of the flow may be necessary in order to achieve accurate estimate of local values.
EXPERIMENTAL DATA SET NO. 11: REFLUX CONDENSATION
360-367
10.1615/MultScienTechn.v3.i1-4.160
Q.
Nguyen
University of California, Santa Barbara, California, USA
Sanjoy
Banerjee
Department of Chemical Engineering University of California at Santa Barbara, Santa Barbara, California, 93106
Two-phase modelling of reflux condensation remains strongly dependent on knowledge of the wall and interfacial shear stresses. Detailed measurements of reflux condensation are conducted in an annular flow with large waves on the interface. Such experimental data as void fraction, pressure gradient and temperature are presented to contribute to the future modelling and fundamental understanding of reflux condensation near the flooding point.
EXPERIMENTAL DATA SET NO. 12: ANNULAR FLOW EVAPORATION
368-377
10.1615/MultScienTechn.v3.i1-4.170
Geoffrey F.
Hewitt
Department of Chemical Engineering & Chemical Technology, Imperial College of Science, Technology & Medicine, Prince Consort Road, London SW7 2B Y, England, UK
The experimental data of Bennett et al. (1966&1967), which studied the behaviour of the film flowrate over the full length of a channel with dryout occurring at the exit of the channel, are presented in this paper and are used to act as a basis in the sensitivity tests of the models. The annular flow model suggested by Whalley et al. (1972) give a rather accurate prediction of the dryout quality. However, there are large discrepancies in the prediction of the entrained flowrate in the upstream region of the channel, which is related to the inaccuracies in the prediction of entrainment rate by the correlations employed.
EXPERIMENTAL DATA SET NO. 13: FLASHING FLOW
378-389
10.1615/MultScienTechn.v3.i1-4.180
J. C.
Rousseau
Commissariat a I'Energie Atomique, Department des Reacteurs a Eau Serivice d'Etudes Thermohydrauliques, Centre d'Etudes Nucleaires de Grenoble, Grenoble, France
A series of experiments for critical steam-water two-phase flows under a large variety of pressure conditions are conducted to investigate the interphase mass and momentum transfers. Various experimental data, such as pressure, mass flux, void fraction, temperature and liquid superheating are included in this paper, which can be used for qualification of the flow models used in the advanced nuclear safety codes.
EXPERIMENTAL DATA SET NO. 14: POST DRYOUT HEAT TRANSFER—STEADY STATE DATA
390-394
10.1615/MultScienTechn.v3.i1-4.190
Peter
Griffith
Department of Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139 USA
This paper presents the post dryout steam water data obtained in two set of steady state experiments by Hull (1982). As an important factor, the sizes of drops within the heated section are controlled by a nozzle and measured photographically. The effect of drop size is found not to be as prominent as that suggested by the theory for post-CHF heat transfer.
EXPERIMENTAL DATA SET NO. 15: POST DRYOUT HEAT TRANSFER —QUENCH DATA
395-397
10.1615/MultScienTechn.v3.i1-4.200
Peter
Griffith
Department of Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139 USA
Experimental studies on post dryout heat transfer are conducted in an inconel-600 tube (2.44 m long and 10.16 mm ID). A set of unusual quench data are presented in this paper in that liquid N2 is used as the fluid and the absolute temperature ratio, wall to saturation, is as large as about 3. Theoretical modelling is shown to give good prediction by considering the whole boiling curve for liquid N2 after a superheat correction for the vapour is made.
EXPERIMENTAL DATA SET NO. 16: CONDENSATION IN STRATIFIED FLOW
398-422
10.1615/MultScienTechn.v3.i1-4.210
S. George
Bankoff
Department of Chemical Engineering, Northwestern University, Evanston, Illinois 60201, USA
S. C.
Lee
Yeungnam University, Kyungsan, Korea
Extensive work in countercurrent stratified flow is performed for determining the effect on condensation heat transfer. Parameters of heat transfer, interfacial wave structure, film thickness, pressure gradient and interfacial shear stress are obtained for various geometry conditions. Accordingly, improved empirical correlations are further postulated for better prediction in the condensation heat transfer coefficient.
EXPERIMENTAL DATA SET NO. 17: EVAPORATION IN HORIZONTAL TUBES
423-441
10.1615/MultScienTechn.v3.i1-4.220
Dieter
Steiner
Institut fur Thermische Verfahrenstechnik, Universitat Karlsruhe (T. H.), Germany
Perimeter-averaged heat transfer coefficients are investigated during flows of R12 in horizontal tube. The experimental data show that heat transfer is affected by the parameters in a different manner depending on the boiling region. The convective boiling region is mainly determined by mass velocity and flow quality, whereas the nucleate boiling is dominantly affected by heat flux. In particular in the developed nucleate region, the heat transfer coefficient is shown to have no dependence on the flow quality.
EXPERIMENTAL DATA SET NO. 18: BLOWDOWN
442-449
10.1615/MultScienTechn.v3.i1-4.230
J. C.
Rousseau
Commissariat a I'Energie Atomique, Department des Reacteurs a Eau Serivice d'Etudes Thermohydrauliques, Centre d'Etudes Nucleaires de Grenoble, Grenoble, France
Transient two-phase flow experiments are carried out in a horizontal pipe (4.09 m long and 206 mm ID) to investigate the variation of temperature and void fraction during the process of blowdown. Large pressure gradient is seen in the cross-section with higher temperatures at the bottom: the vapour at the top is saturated while the liquid at the bottom is superheated. Also, there is an evidence of a vapour annulus existing throughout the blowdown.
NUMERICAL BENCHMARK TESTS: INTRODUCTION
453-455
10.1615/MultScienTechn.v3.i1-4.240
Dudley Brian
Spalding
Concentration, Heat, and Momentum (CHAM), Limited, Bakery House, 40 High Street, Wimbledon Village, London SW19 5AU, England
This paper introduces the numerical benchmark tests developed as part of the activities of the Workshops on Two-Phase Flow Fundamentals sponsored by the US Department of Energy. The tests are grouped under 4 groups — one-dimensional (steady), one-dimensional (unsteady), two-dimensional (steady) and two-dimensional (unsteady).
NUMERICAL BENCHMARK TEST NO. 1.1: NOZZLE FLOW
456-458
10.1615/MultScienTechn.v3.i1-4.250
Dudley Brian
Spalding
Concentration, Heat, and Momentum (CHAM), Limited, Bakery House, 40 High Street, Wimbledon Village, London SW19 5AU, England
A test problem is outlined in this paper, which is to predict the axial distributions of pressure, volume fraction and velocity for flows of two fluids (one compressible and the other incompressible) through a convergent-divergent nozzle. The two-fluid model is to be used with specified inter-phase friction, initial and boundary conditions.
NUMERICAL BENCHMARK TEST NO. 1.2: MONOPROPELLANT ROCKET
459-461
10.1615/MultScienTechn.v3.i1-4.260
Dudley Brian
Spalding
Concentration, Heat, and Momentum (CHAM), Limited, Bakery House, 40 High Street, Wimbledon Village, London SW19 5AU, England
The test problem considered is that of the injection of monopropellant droplets through small holes in one end of a rocket motor, which burn at a rate dependent upon the droplet diameter. The task is to compute the distance from the injector at which the propellant is fully consumed, as functions of the injection velocity, the propellant and combustion-product properties and relative speed. A two-fluid model is to be used with allowance for mass transfer between phases.
NUMERICAL BENCHMARK TEST NO. 1.3: BOILING IN A PIPE
462-464
10.1615/MultScienTechn.v3.i1-4.270
Dudley Brian
Spalding
Concentration, Heat, and Momentum (CHAM), Limited, Bakery House, 40 High Street, Wimbledon Village, London SW19 5AU, England
A test problem is outlined in this paper for steady boiling flow in a duct, which is to predict the distributions of steam and water velocities and enthalpies, of pressure and volume fraction along the duct. A two-fluid model is to be used with full allowance for heat and mass transfer between phases.
NUMERICAL BENCHMARK TEST NO. 2.1: FAUCET FLOW
465-467
10.1615/MultScienTechn.v3.i1-4.280
V.H.
Ransom
Idaho National Engineering Laboratory, Idaho Falls, ID 83415, USA
The test problem is described which consists of a liquid stream entering a vertical solution space at the top and falling under the action of gravity to form a stream of uniformly decreasing cross-section. The objective is to test the interaction of the body force terms with the temporal and convective acceleration terms in the momentum formulation. The two-fluid model is to be used with allowance for the energy transfer.
NUMERICAL BENCHMARK TEST NO. 2.2: OSCILLATING MANOMETER
468-470
10.1615/MultScienTechn.v3.i1-4.290
V.H.
Ransom
Idaho National Engineering Laboratory, Idaho Falls, ID 83415, USA
A test problem is presented which consists of a 'U' tube manometer which contains two phases (vapour and liquid) and is connected at the top, so that a close system is formed. The objectives of this study are to test the ability of a solution scheme to preserve system mass, to model the period of oscillation, to test the ability of the numerical solution scheme to preserve the interface, and also to reveal any numerical damping. A two-fluid model is to be used with allowance for energy transfer between phases.
NUMERICAL BENCHMARK TEST NO. 2.3: EXPULSION OF STEAM BY SUB-COOLED WATER
471-473
10.1615/MultScienTechn.v3.i1-4.300
V.H.
Ransom
Idaho National Engineering Laboratory, Idaho Falls, ID 83415, USA
A test problem is described in this paper, consisting of sub-cooled water injected at a constant volume rate into a vertical tube, which is initially filled with superheated steam and connects at the top to a constant pressure source of superheated steam. The objective is to test numerical solution methods for anomalous numerical behaviour associated with mass transfer modelling of fixed node discretisation schemes. A two-fluid model is to be used with allowance for energy transfer between phases.
NUMERICAL BENCHMARK TEST NO. 2.4: SEDIMENTATION
474-476
10.1615/MultScienTechn.v3.i1-4.310
D. L.
Youngs
Atomic Weapons Research Establishment, Aldermaston, UK
The test problem considered is to simulate the gravitation-driven sedimentation process of a dense phase (ρ1=1.000) that rests on top of a light phase (ρ2=0.999). The purpose of this test is to provide a good examination of the numerical method, as the results, according to the model given in this paper, are sensitive to numerical diffusion in the volume fraction equations.
NUMERICAL BENCHMARK TEST NO. 2.5: KELVIN-HELMHOLTZ INSTABILITY
477-479
10.1615/MultScienTechn.v3.i1-4.320
D. L.
Youngs
Atomic Weapons Research Establishment, Aldermaston, UK
The test problem proposed in this work considers the effect of a small sinusoidal perturbation on countercurrent flows of two fluids (ρ1/ρ2=1 or ρ1/ρ2=1000). As this perturbation should grow with time in a manner similar to Kelvin-Helmholtz instability, the purpose of this test is therefore to compare numerical results with the growth rate derived from the linear perturbation analysis.
NUMERICAL BENCHMARK TEST NO. 2.6: SHOCK TUBE
480-482
10.1615/MultScienTechn.v3.i1-4.330
D. L.
Youngs
Atomic Weapons Research Establishment, Aldermaston, UK
A straight closed duct, of uniform cross-section, is divided by a diaphragm into two equal parts, which contain compressed air on the left and atmosphere-pressure air on the right, respectively. The test problem is thus to predict the transient process when the diaphragm breaks, as an examination on the numerical methods used.
NUMERICAL BENCHMARK TEST NO. 2.7: STRATIFIED FLOW
483-484
10.1615/MultScienTechn.v3.i1-4.340
Dudley Brian
Spalding
Concentration, Heat, and Momentum (CHAM), Limited, Bakery House, 40 High Street, Wimbledon Village, London SW19 5AU, England
A horizontal duct (square or circular cross-section) is divided into two equal length parts by a diaphragm, each of which contains both water and air. The depths of the water are different on two sides. The task is to simulate what happens within the next few seconds after the diaphragm suddenly breaks, as a test of employed numerical methods.
NUMERICAL BENCHMARK TEST NO. 3.1: IDEALIZED TEE-JUNCTION
485-487
10.1615/MultScienTechn.v3.i1-4.350
Dudley Brian
Spalding
Concentration, Heat, and Momentum (CHAM), Limited, Bakery House, 40 High Street, Wimbledon Village, London SW19 5AU, England
A test problem is described in this paper, which is to calculate the separation of a lighter from a heavier phase when a lateral pressure gradient is induced by the presence of an aperture in the side of a duct. The purpose of this problem is to provide a test of the ability of computational procedures. A two-dimensional two-fluid model is to be used with allowance for mass transfer between phases.
NUMERICAL BENCHMARK TEST NO. 3.2: BOILING IN A CHANNEL
488-491
10.1615/MultScienTechn.v3.i1-4.360
Dudley Brian
Spalding
Concentration, Heat, and Momentum (CHAM), Limited, Bakery House, 40 High Street, Wimbledon Village, London SW19 5AU, England
A test problem is outlined in this paper for steady boiling flow in a duct, which is to predict the distributions of steam and water velocities and enthalpies, of pressure and volume fraction along the duct. Taking into account the wall friction, a two-dimensional two-fluid model is to be used with full allowance for heat and mass transfer between phases.
NUMERICAL BENCHMARK TEST NO. 4.1: SEDIMENTATION
492-494
10.1615/MultScienTechn.v3.i1-4.370
D. L.
Youngs
Atomic Weapons Research Establishment, Aldermaston, UK
The test problem considered is to simulate the gravitation-driven sedimentation process of a dense phase (ρ1=1.000) that rests on top of a light phase (ρ2=0.999). The gravitational force is assumed not to be normal to the initial discontinuity in volume fraction, which leads to this problem to be two-dimensional. The purpose of this test is to provide a good examination of the numerical method.
NUMERICAL BENCHMARK TEST NO. 4.2: EXPULSION OF WATER BY AIR
495-497
10.1615/MultScienTechn.v3.i1-4.380
Dudley Brian
Spalding
Concentration, Heat, and Momentum (CHAM), Limited, Bakery House, 40 High Street, Wimbledon Village, London SW19 5AU, England
The problem presented in this paper involves a two-dimensional transient flow of two initially-separated fluids. The heavier fluid, initially filling a rectangular box, is pushed out through an exit at the bottom by the lighter fluid injecting into the box through the entrance at the top. The test is to simulate the development of the interface between phases. The purpose of this proposal is to test the ability of computer codes to preserve sharp interfaces, because the flow phenomena depend critically on the flow rate (Froude number) and on the aspect ratio of the box.