0

Flows in Manufacturing Processes

2002;():3-11. doi:10.1115/FEDSM2002-31375.

A fundamental study of welding nozzle flow under cold flow conditions is presented. The aim is to examine the shielding gas flow characteristics for several Gas Metal Arc Welding (GMAW) flow conditions. Experimental investigations and numerical modeling are used to predict the flow behaviour of the gas shielding the weld pool. Results are presented for generic GMAW nozzle configurations at typical welding situations under cold flow. Flow visualization and Particle Image Velocimetry (PIV) reveal the various flow characteristics that are crucial to optimization of weld pool protection by the shielding gas. Numerical modeling of the flow is performed at conditions similar to the PIV experiments using the k-ε turbulence model in a commercial CFD package. Numerical predictions of the mean velocities agree reasonably well with PIV experiments, particularly in the radial wall jet region of the flow field. However, the turbulent kinetic energy is greatly over-predicted due to the eddy-viscosity stress-strain approximation in the k-ε model.

Commentary by Dr. Valentin Fuster
2002;():13-19. doi:10.1115/FEDSM2002-31445.

The paper describes a study of convective heat transfer in a multiple-jet systems composed of straight and inclined slot nozzles. The application concerned is the fast cooling of moving strip. The experimental approach involves the application of infrared thermography associated with the steady-state heated foil technique. Three-dimensional numerical simulations performed with the code FLUENT compare agreeably with the IR data. The study aims to determine the effect on the average heat transfer coefficient of the slot Reynolds number up to the value of 100000, the nozzle spacing normalised by the slot hydraulic diameter in the range 6 ≤ W/S ≤ 18, the normalised nozzle emergence length, E/S, from 5 to 17 and the normalised nozzle to strip standoff distance Z/S from 3 to 10. The geometrical arrangements tested include perpendicular (90°) and tilted (60°) nozzles. A thermal entrainment phenomenon is found for cooling system of small width. A corrective factor is derived to account for this effect. The experimental findings are compared with existing correlation; deviations, which are observed at high values of the Reynolds number may reach 25%. The numerical simulation emphasises the benefit to use H2 /N2 gas mixture to enhance significantly the cooling rate.

Topics: Jets , Convection
Commentary by Dr. Valentin Fuster
2002;():21-24. doi:10.1115/FEDSM2002-31446.

In this paper we propose a new relationship between the opposing mechanical torque and the electric conductivity of a rotating liquid specimen in a permanent external magnetic field of constant induction, which includes the effect of fluid flow. The proposed relationship was applied to describe the experimental data for a conductive specimen rotating in a permanent magnetic field.

Commentary by Dr. Valentin Fuster
2002;():25-30. doi:10.1115/FEDSM2002-31447.

The infiltration process is one of the techniques employed to work out polymer matrix composites (PMC) or metal matrix composites (MMC) and with continuous reinforcement. It consists in injecting the liquid (resin or metal) through a fibrous reinforcement placed in a mould. In the case of the MMC, the fibres and the mould are initially preheated at temperatures lower than the temperature of metal solidification. These thermal conditions induce phenomena of phase change of metal when the metal is in contact with the fibres. These phenomena disturb the flow of metal through the fibrous network. In order to follow the displacement of metal in the preform and take into account of the phase change, a two-dimensional numerical model based on a finite volume formulation was developed, on a very simple geometry.

Commentary by Dr. Valentin Fuster
2002;():31-40. doi:10.1115/FEDSM2002-31448.

The laminar isothermal mixing of a viscoplastic fluid is considered in the absence of any free surface and when the material can be modelled using a Bingham constitutive equation. The material is mixed by a rotating cylindrical tank and a fixed anchor impeller. A two-dimensional numerical simulation is performed with a finite elements method. The effects of viscoplasticity and inertia are related to an Oldroyd (Od) and to a Reynolds (Re) number. Stream lines, vortices, yield surfaces and power consumption are shown and analysed depending on the value of Od and Re parameters.

Commentary by Dr. Valentin Fuster
2002;():41-46. doi:10.1115/FEDSM2002-31449.

During the cooling of an injected polymer, a very strong coupling exists between the thermal and mechanical phenomena. In order to fill in the mold and to compensate the effects of reduction of the thermal shrinkage, the pressure in the molding cavity is maintained to a very high level. The purpose of this paper is to show the essential importance of thermomechanical coupling present during the cooling of a hot polymer in a cold mold via a comparison between experiments and numerical simulation. Experiments show that the polymer-mold contact evolves under the effect of the internal contraction caused by the lowering of temperature. On the one hand, the air gap width, depending on the shape of the part, modifies the Thermal Contact Resistance (TCR) between the mold and the part. On the other hand, shrinkage is divided between the two sides of the part depending on the internal stresses, so that the geometry of the part is also linked to changing temperature. The resolution of the coupled thermal and mechanical problems allows us to predict the shape of the part, the stresses, the temperature field in the polymer and in the surrounding mold, and the air gap widths on both sides.

Commentary by Dr. Valentin Fuster
2002;():47-54. doi:10.1115/FEDSM2002-31450.

The results of an experimental study and 3D numerical simulations of resin bonded sand/air flow in a square corebox with an H-shape insertion and passage between upper and lower pockets of the pattern are presented. A computer controlled electronic system was designed and built to measure pressures and flow rates inside the corebox during mold filling, gassing and purging cycles of Phenolic Urethane Amine (PUA) process. Contour maps of the pressure distributions inside the corebox were created based on barometric measurements. A good agreement between experimental results and numerical simulations was found.

Commentary by Dr. Valentin Fuster
2002;():55-62. doi:10.1115/FEDSM2002-31451.

When bubbles are continuously released from a located source at the bottom of a fluid layer initially at rest, a plume is produced. The motion of the carrier fluid is initiated and driven by buoyancy of the bubble cloud. In the present study, a detailed analysis of the bubble plume transition is investigated. The continuous phase flow is obtained by direct numerical resolution of Navier-Stokes equations forced by the presence of bubbles. Collective effects induced by the presence of bubbles are modelled by a spatio-temporal distribution of momentum. Time evolution of the dispersed phase is solved by lagrangian tracking of all the bubbles. Focused on the description of plume transition, several configurations (plume widths, fluid viscosity, injection rate) are investigated. During the laminar ascension of the plume, fluid velocity profiles can be non-dimensionalised on a single auto-similar evolution. Dimensional analysis provides a prediction of the limit rising velocity of the plume top. This prediction has been confirmed by our numerical simulations. Furthermore, our first results point out the symmetry breaking induced by plume instability which appears beyond a critical transition height. Various data show that the Grashof number based on injection conditions is the key parameter to predict the transition of the plume. Our results agree very well with recent experimental data. Comparison with experiments on thermal plumes in air shows that the bubble plume is more unstable. This feature should be related to the lack of diffusion in the lagrangian transport of density gradient by the bubble cloud and to the slip velocity between the two phases.

Commentary by Dr. Valentin Fuster
2002;():63-70. doi:10.1115/FEDSM2002-31452.

Recent numerical simulations of the wake of a fixed sphere have confirmed that hydrodynamic forces are likely to have a significant impact on the trajectory of a freely falling (or ascending) sphere. An ideally spherical body ceases to follow a straight vertical trajectory at the Reynolds number (based on its velocity U and diameter d) corresponding to the onset of the primary instability responsible for the breaking of axisymmetry in a fixed sphere wake, i.e. at Re = 212. This instability has been shown to generate a steady non axisymmetric flow with a symmetry plane containing the asymptotic flow velocity, the orientation of which is arbitrary, i.e. selected by any small perturbation at the instability onset. In this communication, we present further work focussed on the experimental investigation of the effect of instabilities on the trajectory of a free sphere. The axisymmetry breaking results in a lift and torque, the vector of lift lying in the symmetry plane and the torque being normal to this plane. This leads to the conclusion that a free-falling (ascending) sphere will be deviated from its vertical trajectory as soon as its Reynolds number reaches the threshold of 212. Moreover, the trajectory will be deflected in an arbitrarily selected vertical plane. An experimental setup has been implemented to investigate this effect. It consists of a 2.5 m high water tank with a .5 times .5 m cross section placed in an air-conditioned chamber allowing to control finely the asymptotic Reynolds number of small spheres (on the order of a mm in diameter) by varying the water temperature. Spheres of densities close to that of water, both lighter and heavier, are considered. The trajectories are investigated fully in three dimensions by processing of images of two cameras following the sphere movement. The preliminary results, presented here for polypropylene spheres lighter than water, confirm the numerically and theoretically predicted effect. After a short acceleration phase roughly in vertical direction the primary instability deflects the trajectories each time in a different vertical plane. The investigation of the fixed sphere wake showed the onset of a secondary Hopf-type instability at Re ≈ 275. The same type of instability develops clearly for free spheres. Unlike for the fixed sphere, the secondary instability is observed to dominate and to yield a wavy trajectory with a vertical mean direction.

Commentary by Dr. Valentin Fuster
2002;():71-76. doi:10.1115/FEDSM2002-31453.

The purpose of this paper is to simulate numerically the impact of droplets onto a substrate to give a better knowledge of coating manufacturing. A 2D model is proposed, with symmetry conditions to simulate the impact of an infinity of identical droplets. The dynamic behavior is studied, and a thermal study of superimposed splats is added separately to give a better understanding of each phenomenon.

Commentary by Dr. Valentin Fuster

Experimental and Numerical Flow Visualization and Laser Anemometry

2002;():77-85. doi:10.1115/FEDSM2002-31159.

The three-dimensional flow field in the tip region of an isolated axial flow fan rotor with two different tip clearances are investigated using a three-color, dual-beam PDA system (Particle Doppler Anemometer, DANTEC Measurement Technology). The global performance is also obtained, and is compared favorably with CFD (Computational Fluid Dynamics) modeling of this fan flow at a zero tip clearance. The detailed flow field measurements are taken at 15 axial locations upstream, inside and at the exit of the rotor. In the radial direction, 15 measurement locations are arranged from 50% of the blade span to the casing wall, mainly focusing on the tip region from 90% of the blade span location to the casing wall (about 10 measurement locations). The PDA data has provided a quantitative understanding of the flow phenomena in the tip region of the fan rotor. For both tip clearances, it has been observed that the tip leakage flow rolls up into a tip leakage vortex. Due to the rotation of the rotor, this tip leakage vortex moves away from the suction surface of the fan blade. A reverse flow is induced in the main flow passage because of the tip leakage vortex. The depth and extent of the tip leakage vortex grow noticeably with the increase of the tip clearance.

Commentary by Dr. Valentin Fuster
2002;():87-93. doi:10.1115/FEDSM2002-31160.

This paper reports PIV measurements made at three locations in an axisymmetric, confined jet that is approaching a free surface from below. The apparatus consists of a tank 40.5 cm × 40.5 cm at its base and 61 cm high. A 9 mm diameter nozzle is centered in the base of the tank and directs a jet of water upwards. The jet produced has a top-hat velocity profile with a maximum deviation of 0.32% of the mean and an axial relative turbulence intensity of 0.60%. The water is removed from the tank by an overflow around the perimeter of the tank. The PIV measurements achieved a spatial resolution of between 0.425–1.08 mm. The measurements show details of the velocity field in three regions of the flow; at the jet exit, near the surface on the centerline of the jet, and near the top corner of the tank. The centerline velocity remains at the exit velocity until ≈5D from the exit. The axial confinement of the jet begins to significantly influence the centerline velocity at ≈13D from the free surface. All entrained fluid is deflected downward from the horizontal surface flow as it approaches the overflow around the perimeter of the tank. This creates a large recirculation region in the upper region of the tank driven by the downward flow along the wall and the upward flow of the jet itself at the center of the tank.

Topics: Measurement
Commentary by Dr. Valentin Fuster
2002;():95-101. doi:10.1115/FEDSM2002-31161.

This experimental study is concerned with the effect of density variations on the development of a turbulent jet. Our interest concerns the large turbulent scales in order to understand how density variations affect these structures involved in the mixing process. The integral scale evolutions are estimated from laser velocimetry measurements (PIV or LDV). Comparisons between the results obtained in an air jet and a helium jet are given in the development zone (X/Dj < 20), so that the effect of strong density variations (ratio 1 to 7) on these scales is deduced.

Topics: Density , Lasers , Turbulence , Jets
Commentary by Dr. Valentin Fuster
2002;():103-113. doi:10.1115/FEDSM2002-31162.

The unsteady flow field due to blade passing at the pump/turbine interface of a torque converter was studied. The current geometry is wide and has a large outer to inner radius ratio. A laser velocimeter was used to measure the periodic velocity components at four operating conditions determined by the speed ratios between the turbine and pump of 0.065 (near stall), 0.600, 0.800, and 0.875 (coupling point). The flow fields at the pump exit and turbine inlet planes were visualized and are presented. Using instantaneous pump and turbine blade positions with the velocity data, animations (“slow-motion movies”) are generated to effectively visualize and understand the unsteady behavior. The turbine inlet flow was markedly periodic due to the exiting jet/wake from the upstream pump passage; however, the pump exit flow field showed little dependence on the turbine blade positions. The highest unsteadiness was seen for the highest speed ratios. Four “shots” from the sequences of one cycle for all speed ratios and each plane are presented herein. The results are also compared to unsteady results for a previously examined torque converter with a small radius ratio to determine the effect of parametric geometric changes on the flow field. Generally, the unsteady velocity fields show no significant difference for the two geometries — the trends are the same.

Commentary by Dr. Valentin Fuster
2002;():115-122. doi:10.1115/FEDSM2002-31163.

Recent aero-engine combustors have been of a large flame dome. The combustor diffuser wit h deep flame dome results in the large total pressure loss. It is important to obtain both better aerodynamics performance of the total pressure and exhaust gas characteristics of NOx in the combustor. Installation of an inclined wall within the dump diffuser combustor with a split flow at the flame dome head is suggested in order to improve the aerodynamic performance. The cold flow experiment in a model channel of the combustor diffuser conducted by the Laser Doppler Velocimetry system shows that the inclined wall is effective in improvement of the diffuser performance regardless of the ratio of the split flow rate. The key technology factor of inclined wall installation is the selection of the throat area ratio between the inclined wall and the flame dome.

Commentary by Dr. Valentin Fuster
2002;():123-133. doi:10.1115/FEDSM2002-31164.

Micro-scale fluidic devices are now being designed and manufactured for a host of new applications, and it is certain that new applications will emerge. There is, therefore, a need for increased understanding of the momentum transport phenomena at this scale to aid in the understanding, and design and optimization of such devices. This need is behind the development of new techniques for making flow measurements at the micro-scale. Molecular Tagging Velocimetry (MTV) is a laser-based non-intrusive technique for obtaining detailed measurements of velocity profiles. This paper reports on the extension of the method to microtubes of inside diameter of order 100 µm. Fully developed velocity profile measurements are reported here for a Reynolds number of about 140 in a capillary of inside diameter 148 µm. Two image analysis techniques were compared—the line center method, and the correlation method. It was found that the correlation method produced better results and smaller overall uncertainty. Volumetric flow rate determined from integration of the measured velocity profiles agree with accumulation measurements made over a specified time interval to within 3%; the agreement was usually 1–2%. This work reports on the difficulties encountered in applying MTV at these physical scales, the influence of measurement and analysis parameters on results, and the uncertainty associated with measurements.

Commentary by Dr. Valentin Fuster
2002;():135-146. doi:10.1115/FEDSM2002-31165.

An extensive experimental study on sprays from an injector for gasoline direct injection (GDI) engines has been performed. Spatial and temporal evolution measurements of a large cone-angle jet, emerging from a high pressure swirled injector, have been carried out in an optically accessible vessel. The spray has been lightened, both along the spray axis and in cross sections perpendicular to it, by a 532 nm Nd-YAG pulsed laser sheet, 80 mm thickness and 12 ns duration. The scattered light has been collected at 90° with respect to the sheet direction by a digital CCD camera with a frame grabber synchronized with the injection command and the laser pulse. A digital delay system has provided a fine temporal shift (up to microseconds range) of the images acquisition with respect to the start of the injection (SOI). Finally, a digital image processing system has provided analysing the images collected by the CCD camera. The emerging spray has been acquired with three spatial scales, providing both the global and local spray behaviour, and with a detailed temporal resolution to characterize the early stage of the jet formation. The initial phase of the spray is characterized by a strong axial component of the velocity with respect to the radial one, resulting in a pre-spray or slug phase. It produces a cylindrical shape of the jet with the characteristic mushroom shape. Large droplets with high momentum are produced, travelling downstream in advance to the main spray. At later time the radial velocity component controls the process and it gives up to the classical hollow-cone shape with a strong interaction with the gas in the vessel. The images give evidence in time of the collapse of the hollow-cone structure hence resulting in a full cone spray. This behaviour is confirmed by the cross section measurements through the spray, carried out in the range 10–40 mm from the nozzle tip. These measurements make evidence of the refilling of the cone with the presence of strong vortexes on the boundary of the jet. The effects of the fuel injection pressure, injection duration and air-flow field interaction on the structure and evolution of the spray have been studied in details. The digital image processing system also has allowed to reconstruct the spray profile and to determine a refilling index.

Commentary by Dr. Valentin Fuster
2002;():147-154. doi:10.1115/FEDSM2002-31166.

The study of the recovery of an open channel boundary flow in the presence of increased freestream turbulence (FST) generated in the wake region of a surface mounted flat plate is presented. Detailed LDA velocity measurements were obtained upstream and downstream of the flat plate, which is 3 mm in thickness and has a thickness-to-chord ratio of 0.12. The chord is placed parallel to the flow direction. The characteristics of the mean velocity, turbulence intensity, and the velocity skewness and flatness factors were investigated. The skin friction was increased while the strength of the boundary layer wake parameter decreased in the wake region. The turbulence intensity profiles in the wake region increasingly deviated significantly from the upstream profile. Generally, the increased FST noticed in the near-wake region was observed to decay with downstream distance. As a result, the mean velocity and turbulence intensity profiles showed a general sense of recovery towards the state of the approaching flow.

Commentary by Dr. Valentin Fuster
2002;():155-160. doi:10.1115/FEDSM2002-31167.

A theoretical study is performed to investigate unsteady, two-dimensional, incompressible thermal-fluid flow over both sides of a slot-perforated flat surface, which is placed in a pulsating channel flow. The roles of both the pulsating Strouhal number and the ratio of the channel width to the plate thickness, W/δ, on the velocity and thermal fields are disclosed. It is found from the study that: (i) when the channel stream is pulsated, the alternating change in the fluid flow disturbs the thermal boundary layer formed along the plate and induces mixing of the upper and lower streams of the plate downstream from the slot, resulting in an amplification of heat transfer performance, and (ii) heat transfer performance at the plate is attenuated with a decrease in W/δ.

Commentary by Dr. Valentin Fuster
2002;():161-168. doi:10.1115/FEDSM2002-31168.

Turbulent wedges induced by a 3D surface roughness placed in a laminar boundary layer over a flat plate were visualised for the first time using both shear-sensitive and temperature-sensitive liquid crystals. The experiments were carried out at three different levels of favourable pressure gradients. The purpose of this investigation was to examine the spreading angles of the turbulent wedges indicated by their associated surface shear stresses and heat transfer characteristics and hence obtain further insight about the difference in the behaviour of transitional momentum and thermal boundary layers when a streamwise pressure gradient exists. It was shown that under a zero pressure gradient the spreading angles indicated by the two types of liquid crystals are the same, but the difference increases as the level of favourable pressure gradient increases. The result from the present study could have an important implication to the transition modelling of thermal boundary layers over gas turbine blades.

Commentary by Dr. Valentin Fuster
2002;():169-175. doi:10.1115/FEDSM2002-31169.

Mixing vessels are widely used for blending and chemical reactions. Although much has been done on mixing processes, the complex, three-dimensional flow phenomena are still not well understood. The purpose of our first step in this research is the simulation and validation of time-resolved, three-dimensional velocity vector data. Such results are an essential part of the design of mixing systems, but are generally not available to the engineers. The computational work involves direct numerical simulation (DNS) and large eddy simulation (LES) of the Navier-Stokes equations. Later, modeling of the Reynolds averaged Navier-Stokes (RANS) equations will be undertaken as a simplified approach. Simulations and modeling are being validated by experiments. Two flow mixing systems are under investigation. First and most important for validation is an opposed jet flow system that offers some unique characteristics that can be used for validation of DNS/LES simulations. It also has applications in the injection molding of plastics. Second, simulations of impeller driven mixing vessels that are more commonly used in processing are under development. Here the moving mesh system adds complexity. In addition, visualization of both numerical and experimental results, 3-D particle tracking velocimetry (PTV) techniques have been developed. The proposed paper will address the problems in the modeling of chemical mixing and discuss the results of simulation and validation.

Commentary by Dr. Valentin Fuster
2002;():177-181. doi:10.1115/FEDSM2002-31170.

High attack-angle vanes have been proposed as a means of maintaining sediment in suspension in rivers and streams. Laboratory and computer studies have, up to now, been unable to describe how the flow mechanism actually works in the sediment-suspension mode. A turbulent-flow tracer has been used to study model vanes in a small water channel at the University of Bristol. The sediment-suspension capabilities of such vanes becomes evident from their highly swirling non-symmetric Kármán vortex-type wakes. The wake patterns have been analyzed using video recordings and frame-counting software. Using the video frame rate as a time base, Strouhal numbers and velocities in the vane wake can be extracted by following the movement of particular tracer strands.

Commentary by Dr. Valentin Fuster
2002;():183-192. doi:10.1115/FEDSM2002-31171.

Continuing advances in digital imaging technology stimulate greater interest in applying particle image velocimetry (PIV) over increasingly larger fields of view. Unfortunately when larger fields of view are analyzed, velocity gradients in the image become more localized. In addition, non-uniformities in image illumination and particle number density become more prevalent. These factors, coupled with the requirement that large areas of interest (AOIs) must be employed to measure the full range of velocity, cause degradation of correlation results (i.e. broadening and/or splintering of the cross correlation peak) which leads to positional bias errors in the measured velocity field. More advanced super resolution strategies that employ an iterative AOI reduction process inherently reduce positional bias in PIV results but these strategies can break down in complex flows where velocity gradients are steep and particle dispersion does not remain uniformly random. To mitigate these problems a simple but effective technique is presented that enables individual velocity vectors to be placed within an AOI at locations toward which the cross correlation plane is biased. The method involves analysis of the correlation plane to extract the dominant features that are matched in two successive AOIs. To demonstrate the utility of the methodology results obtained from synthetic images are compared against results obtained using the conventional PIV approach.

Commentary by Dr. Valentin Fuster
2002;():193-198. doi:10.1115/FEDSM2002-31173.

PIV (Particle Image Velocimetry) is a new technique of flow measurement. In this experiment PIV technique is used to measure the flow conditions in the pump sump model, which are very complicated, especially in the surrounding of the pump bell where there are very complex vortexes. These will affect the flow characteristics in the pump system broadly and result in vibration and cavitation, and decrease the efficiency of the system. And these will harm the performance of pump station, and even lead it out of order. The vortex surrounding the pump bell has the direct relationships with the structure of the pump suction sump and the pump bell. In this experiment the pump suction sump and the pump bell with particularly structure are used to research the flow situation in the surrounding of the bell. The PIV method is used in the experimental to study the flow conditions surrounding the pump bell. Using this technique the flow velocities, the streamlines and the turbulent kinetic energies in the full range can be obtained. Moreover, other important parameters can be gotten after being analyzed the base experiment data further. Here, the results of the PIV experiment on flow in pump suction sump are analyzed further, and the following parameters are obtained: circulation vector of the inlet flow of bell, three-dimensional velocity and turbulent kinetic energies. And through analyzing the experiment’s photos, the air core’s conditions in the bell also can be gotten.

Commentary by Dr. Valentin Fuster
2002;():199-203. doi:10.1115/FEDSM2002-31174.

Previous studies dealing with sprays have used a variety of techniques to determine spray droplets and spray penetration. In particular, the sedimentation tower method and the liquid immersion sampling technique were most popular. However, in these techniques sampling is done after spray formation is complete. The completion time of spray formation appears to vary with the conditions of injection and ambient factors, thus making measurements under transient conditions during injection difficult. A pulsed Malvern drop-size analyzer, based on Fraunhofer diffraction, was utilized to determine spray penetrations of diesel fuels under different conditions of injection, along with the effects of fuel properties. In these study, the spray is formed by injecting a calibrated amount of fuel into air. A two mm diameter collimated beam illuminated a cylindrical volume perpendicular to the axis of the fuel spray, and its attenuation was recorded and stored on the oscilloscope. With the optical measurement being synchronized to the needle lift of the injector, the output of the needle lift transducer and the optical signal was recorded simultaneously. Thus, the arrival and the duration of the spray at various positions along its axis were measured. A spray penetration correlation is obtained, and is compared to other existing correlations in the literature.

Topics: Sprays
Commentary by Dr. Valentin Fuster

Erosion Processes

2002;():205-209. doi:10.1115/FEDSM2002-31283.

When slurry pumps are considered for procurement for a solids transport requirement, decisions have often been made on initial price rather than the total cost of ownership (TCO). Depending on the type of solids to be transported, the design (and selection) of the pumps and the way they are operated; the wear, part replacement, and downtime can be a significant portion of the overall operating cost. Wear lives are here estimated with recently developed numerical simulation procedures for pump impeller and suction liner wear together with an established wear modeling schema for shells. White iron material is considered and the resulting modeled wear does not include any uncertainty related to the inherent stochastic nature of wear. It is shown in an example how shell, impeller and suction wear life and relative costs are related to different pump size and rotary speed designs. The results form a basis to find the best compromise between pump wear, power and capital cost in an overall TCO-perspective including the users practice and maintenance strategy.

Topics: Wear , Pumps , Slurries
Commentary by Dr. Valentin Fuster
2002;():211-222. doi:10.1115/FEDSM2002-31284.

Transport of solid-liquid slurries in pipeline transport over short and medium distances is very important in many industries, including mining related processes. The particle image velocimetry technique was successfully utilized to investigate the velocities and kinetic energy fluctuations of slurry particles at the tongue region of an optically-clear centrifugal pump. The experiments were conducted using 500 micron glass beads at volumetric concentrations of 2.5% and 5% at pump speeds of 725 and 1000 rpm. The fluctuation kinetic energy increased approximately 200% to 500% as the pump speed was increased from 725 rpm to 1000 rpm. The directional impingement mechanism is more significant at the pressure side of the blade, tongue and the casing. This mechanism becomes more important as the speed increases. This suggests that the impeller, tongue and the casing of the slurry pump can wear out quickly, especially with an increase in speed. In this paper the emphasis is on the tongue region. The random impingement mechanism caused by the fluctuation kinetic energy of the solids can play an important role on the erosion of the tongue area.

Commentary by Dr. Valentin Fuster
2002;():223-230. doi:10.1115/FEDSM2002-31285.

The Department of Energy is sponsoring the River Protection Project, which includes the design of a facility to stabilize liquid radioactive waste that is stored at the Hanford Site. Because of its experience with radioactive waste stabilization, the Savannah River Technology Center of the Westinghouse Savannah River Company is assisting in the development and testing of parts of the waste treatment process. One part of the process is the separation of highly radioactive solids from the liquid wastes by cross-flow ultrafiltration. For the projected forty-year life of the filtration facility, wear will occur from a combination of erosion and corrosion due to the flow of slurries. A scaled cross-flow filter facility will be tested with simulated waste to quantify the wear rate so that an effective maintenance schedule can be developed. This paper discusses the application of computational fluid dynamics (CFD) methods to ensure that the test facility design would capture the erosion phenomena expected in the full-scale cross-flow ultrafiltration facility. An initial literature survey helped identify the principal drivers of erosion for a solids laden fluid. These were the solids content of the working fluid, the regions of recirculation and particle impact with the walls, and the regions of high wall shear. A series of CFD analyses was then designed to characterize slurry-flow profiles, wall shear, and particle impingement distributions in key pipe bends and fittings representative of the plant. Pipe diameters, lengths, the locations of pipefittings, and slurry velocities were scaled with the CFD calculations to ensure that the erosion drivers were appropriately represented in the test facility. This resulted in a validation of the theoretical determination of those drivers, and allowed the test results to be applied to a prediction of wear in the full-scale filtration facility.

Commentary by Dr. Valentin Fuster
2002;():231-236. doi:10.1115/FEDSM2002-31286.

Pipe fittings used in the production of oil and gas can experience erosion damage from solid particles such as sand present in the produced fluid. The fittings disturb the flow, which often results in erosion. Certain geometries are more susceptible to erosion damage than others; for example, a sudden expansion can experience severe erosion under certain operating conditions. A sudden expansion can promote erosion for two reasons. First, a radial velocity develops downstream of the expansion. This velocity component drives particles toward the wall. Second, the sudden expansion causes a zone of elevated turbulent kinetic energy. The high level of turbulent kinetic energy results in large turbulent fluctuations, and these fluctuations can also force particles to the wall. A comprehensive model for predicting sand erosion has been developed by utilizing a commercially available computational fluid dynamics (CFD) code. This procedure involves flow modeling, particle tracking, and applying erosion equations. Due to the low sand concentrations a one-way coupling is used between the fluid and the particles. The goal of this study is to use the procedure to calculate erosion in sudden expansions and determine if modifications to the existing procedure are necessary. Simulation results for sand in air flowing through sudden expansions with different diameter ratios (1.25 to 2.00) are compared with experimental data. Both the simulations and data show that the maximum erosion rate evaluated in thickness loss per mass of sand passing through the geometry decrease with increasing expansion ratio.

Topics: Erosion
Commentary by Dr. Valentin Fuster
2002;():237-243. doi:10.1115/FEDSM2002-31287.

Swirl inducing pipes are proposed for the alleviation of problems of poor particle distribution and sliding wear, particularly at downstream bends and elbows. In a well-designed conventional pipeline, the mean axial velocity to assure good dispersion of particles is much greater than the velocity required to merely transport the slurry. This gives the impetus to design swirl-inducing pipes which allow for reduced pumping power, and reduced erosion, while efficiently maintaining suspension at strategic points. This paper covers research that has been aimed at producing good distribution of particles at relatively low velocities, by applying swirl induction. Computational models for the impact velocity and impact angle in a bend have been successfully applied to the flow field and validated by experiments in a perspex flow loop including electrical resistance tomography (ERT) to confirm the placement of particle burdens. Particle impact parameters from this work have been used as inputs to erosion models to predict wall wastage rates in bends and the location of damage from well distributed and swirling particulate flows.

Topics: Erosion , Pipes
Commentary by Dr. Valentin Fuster
2002;():245-246. doi:10.1115/FEDSM2002-31288.
FREE TO VIEW

The ingestion of suspended particles by high performance turbomachinery reduces engine efficiency and life. New developments in blade coating materials have contributed to compressor and turbine design improvements. A knowledge of the important phenomna associated with material erosion by particulate flow is required in the design. This paper gives an overview of the experimental studies of the erosion characteristics of coated superalloy blades conducted at the University of Cincinnati’s test facilities for blade and coating materials erosion will be described. Results will be presented and discussed for the erosion characteristics of various blade materials and coatings. The investigated coatings are produced via plasma spray, detonation guns, chemical vapor deposition (CVD), and physical vapor deposition (PVD).

Commentary by Dr. Valentin Fuster
2002;():247-254. doi:10.1115/FEDSM2002-31289.

Solid particle erosion is a complex phenomenon that depends on many factors such as particle and fluid characteristics, type of material being eroded, and flow geometry. Fittings used in the oil and gas industry such as elbows are susceptible to erosion when solid particles are present in the flow. The momentum of particles carries them across streamlines and the particles impinge the outer wall of the elbow resulting in erosion damage. In an erosive environment, plugged tees are commonly used instead of elbows to reduce the erosion especially where space considerations are important and long-radius elbows can not be used. However, it is unclear how much of a reduction in erosion occurs by replacing an elbow with a plugged tee. In order to compare the erosion in an elbow and a plugged tee exposed to the same flow conditions, a CFD-based erosion prediction model is applied. The model has three primary steps: flow modeling, particle tracking, and applying erosion equations. The results from the model agree with experimental findings for the elbow geometry. However, the simulation results for erosion rate generated for the plugged tee requires a stochastic approach. Results obtained with the erosion prediction model before and after this modification are shown.

Topics: Erosion
Commentary by Dr. Valentin Fuster
2002;():255-264. doi:10.1115/FEDSM2002-31290.

A contoured plug of stainless steel #316 inserted in a low CV valve has been subjected to erosion tests at a pressure reduction of 20MPa. The test duration on a single plug has extended over 300 hours until the shut-off function becomes incomplete. The fundamental flow pattern inside the valve is almost unchanged at the inlet pressure up to 20MPa in a sense that the pressure reduction has been almost completed at the throat formed by the seat-ring and the characteristic surface of the plug head. The kinetic energy of the flow at the throat describes the energy that is dissipated in the valve. Simulated pressure distribution and cavitation pictures obtained at a low inlet pressure condition indicate that the plug seat-joint-taper is the part that suffers impact of the high kinetic energy flow and collapse of the cavitation bubbles. Then this part is specified as the part of the most heavy erosion. In the erosion test, the process of the erosion development shifts from a phase of pit formation to a phase of rooting out the inside at a stage that the mass loss has reached a certain critical level. In the initial phase, the time averaged loss rate is proportional to a power of the dissipated flow energy. In the second stage of rapid growth, the cumulative mass loss increases with fourth power of the working duration. By using values of the exponent and factor obtained at the full opening test, that provides the shortest test periods, mass loss at an arbitrary condition of CV -value and cavitation number is estimated. Then, a procedure to estimate the valve lifetime is suggested based on a criterion that the seat-leakage flow must not exceed the controllable minimum flow of the valve.

Commentary by Dr. Valentin Fuster

Fluid-Structure Interaction and Flow-Induced Noise in Industrial Applications

2002;():267-274. doi:10.1115/FEDSM2002-31420.

The objective of this study is to characterize the velocity, vorticity, wall pressure fluctuations and resulting structural vibrations caused by injection of a round, turbulent jet into a turbulent boundary layer. The experiments are performed in a quiet water channel with back ground noise well below the local pressure fluctuations. One of the channel walls is replaced by a vibration isolated, 1m long, aluminum plate from which the 1cm-diameter jet is injected. The cross flow velocity is fixed at 2 m/s and the velocity ratio, r (ratio of mean jet velocity to the cross flow), varies from 0.5 to 2.5 and Re based on cross flow and jet diameter is 20,000. High-resolution PIV is used to measure the flow field and high sensitivity, low-noise pressure sensors are used for the wall pressure measurements. The flush-mounted transducers are installed at several locations ranging from 2–15 diameters behind the jet. Auto-spectra of the pressure signals show that the effect of the jet is in the 15–100Hz range, and increase the wall pressure levels by 25dB for r = 2.5. The fluctuations increase with velocity ratio and decrease with distance from the jet, although there is only a 6dB increase in overall levels at r = 2.5 as compared to r = 1. Hilbert-Huang “amplitude” spectrum shows the frequency content of the signal as it evolves in time, and is found to be a useful tool to characterize such unsteady phenomena. Velocity and pressure measurements have been performed simultaneously and thousands of frames have been recorded. Analysis of these frames demonstrates the relationship between the pressure fluctuations and the vortical structures. Several striking differences in the flow structure between high and low velocity ratios are described in the paper.

Commentary by Dr. Valentin Fuster
2002;():275-280. doi:10.1115/FEDSM2002-31421.

Sloshing liquid in a container may be used to suppress excessive structural vibrations. The principle of control of sloshing absorbers, is very similar to that of the conventional tuned vibration absorbers. The fluctuating pressure forces of sloshing liquid is employed to provide the control effect, instead of the intentionally resonated auxiliary oscillator of a tuned absorber. This paper presents the results of an experimental study to investigate the effectiveness of the sloshing of the raw contents of a hen’s egg. The objective is to explore the possibilities of borrowing design characteristics to enhance the effectiveness of sloshing absorbers for structural control. To this end, experimental observations are presented in the form of design charts.

Commentary by Dr. Valentin Fuster
2002;():281-288. doi:10.1115/FEDSM2002-31422.

Aerodynamic flow, thermal flow, and stress analyses of a proposed 2.5-in braided composite ablative nozzle were conducted. The nozzle was made up of PAN based carbon fiber and Primaset Cyanate Ester resin. The designed thermal and structural layers’ thicknesses were 0.62-in and 0.42-in, respectively. The objective of this paper is to establish the thermal stability and structural integrity of the nozzle. A ballistic profile of 17 seconds burn at 740 psi maximum pressure was used in the analysis. The combustion gas flow was subsonic, sonic, and supersonic, respectively, in the inlet, throat, and exit sections of the nozzle. The exit plane Mach number was 2.32. Three thermochemical states were considered: 100% efficiency, 72% efficiency and no aluminum combustion and 72% efficiency and 100% aluminum combustion. Flow analysis was conducted by NAT92 Code, thermal analysis by CMAFLOW92 Code, and structural analysis by ANSYS. Ablation rates for the three cases were 0.023, 0.047, and 0.01-in/sec, respectively. The recession life of thermal layer was 27, 13, and 62 sec for the three cases. Stress analysis of the nozzle due to aerodynamic pressure and temperature resulted in low strains and presents no concern.

Commentary by Dr. Valentin Fuster
2002;():289-294. doi:10.1115/FEDSM2002-31423.

In supersonic nozzles, changing pressure ratio varies not only flow conditions but also noises. In order to clarify the relationships between noises and flows (such as shock waves and separation), the present study was conducted using the Fourier Analysis of aerodynamic noises and the Schlieren method. From the investigation, it was found that when the flow was separated from one side of the nozzle, the peak of the spectrum of the frequency response was at around 2000Hz. However, when the flow was in the whole area, the peak of the spectrum was at around 5000Hz.

Commentary by Dr. Valentin Fuster
2002;():295-302. doi:10.1115/FEDSM2002-31424.

Sloshing is the low frequency oscillation of a liquid in a partially full container. Only a limited number of attempts is reported in the literature to control sloshing. Amongst these limited attempts, no effort can be found to employ the flexibility of liquid containers to suppress sloshing. Here, progress is reported of an ongoing research program at Victoria University to determine the critical design parameters of flexible containers to achieve effective control. Numerical predictions as well as experimental observations are presented.

Topics: Containers , Design , Sloshing
Commentary by Dr. Valentin Fuster

Numerical Methods for Multiphase Flows

2002;():303-318. doi:10.1115/FEDSM2002-31212.

The accurate prediction of void fraction profiles near solid walls using the two-fluid model is still an unresolved issue. These profiles result from the combination of two factors: a) forces acting on the bubbles, and, b) geometric constraints imposed by their shape. We propose herein a two-fluid model that involves a particle-center-averaging procedure for the disperse phase, and a reinterpretation and postprocessing of the results obtained. This center-averaged approach averages the disperse phase (bubbles) based on a particle center indicator function, while using the standard phase indicator averaging for the continuous phase (liquid). The solution fields obtained are then postprocessed to introduce the geometry of the bubbles in order to recover the values that should be representative of measured fields. The key idea here is to separate the geometric aspect from the dynamic aspect of the problem into two independent, successive steps. The new model may be easily incorporated into existing two-fluid model codes. Results obtained with the new model showing agreement with experimental data (Moursali et al, 1995) are also presented.

Commentary by Dr. Valentin Fuster
2002;():319-326. doi:10.1115/FEDSM2002-31213.

In this paper we present the results of the application of a Lagrangian particle dynamics (LPD) method and a random flow generation technique (RFG) developed by the authors used in conjunction with large eddy simulations (LES) applied to the case of a two-phase bubbly mixing layer. The objective is to validate the present hybrid Lagrangian-Eulerian approach within the context of LES. The dynamics of the vortices and bubble concentrations reproduced in simulations are in a good agreement with experimental data.

Commentary by Dr. Valentin Fuster
2002;():327-332. doi:10.1115/FEDSM2002-31214.

A Large-Eddy Simulation (LES) with a two-way coupling is used to study bubble-liquid two-phase confined jets in a two-dimensional channel. The results show the large-eddy vortex structures of both liquid flow and bubble motion, the shear-generated and bubble-induced liquid turbulence. For comparison, the second-order moment (SOM) modeling was also carried out for the same case. Both LES and SOM results indicate much stronger bubble fluctuation than the liquid fluctuation, the enhancement of liquid turbulence by bubbles even for the higher velocity case. Both shear production and the production due to bubble-liquid interaction are important for the liquid turbulence generation in the case studied. The LES statistical results and the SOM simulation results are in qualitative agreement with each other.

Commentary by Dr. Valentin Fuster
2002;():333-340. doi:10.1115/FEDSM2002-31215.

Heat transfer in a gas-solid suspension flowing upwards in a heated pipe has been addressed numerically using an Eulerian-Lagrangian approach. Following the basic idea of available experimental studies, several configurations have been simulated and compared, varying the relative size of the particles, i.e. the ratio dp /D. Numerical results show that this ratio has a significant influence upon the fluid and particle velocities and temperatures. However, the resulting suspension Nusselt number is found to be only weakly modified by the parameter dp /D at moderate loading ratio m. This is in accordance with available measurements (e.g. Farbar and Depew, 1963) which do not show a clear dependence of the Nusselt number upon dp /D in the range 0 < m < 3. Associating the influence of this relative size parameter with effects related to inter-phase exchange area, other dimensionless parameters like the flow and particle Reynolds numbers, some Froude number, as well as the heat capacity ratio, density- and loading ratios, would of course be required in order to achieve full similarity of the suspension flow.

Commentary by Dr. Valentin Fuster
2002;():341-347. doi:10.1115/FEDSM2002-31216.

This paper describes a new approach to the direct numerical simulation of particle flows and illustrates its performance on several examples. The method uses a fixed regular orthogonal grid and combines an analytic solution, valid in the neighborhood of each particle, with a finite-difference solution away from the particle. In effect, the analytic solution is used to ‘transfer’ the boundary conditions at the particle surface to the neighboring grid nodes, thus eliminating the complexity arising from the incompatibility of the particle geometry with that of the underlying fixed grid. Furthermore, the hydrodynamic force and couple on the particle are computed directly with no need for the error-prone extrapolation or interpolation procedures necessary with other methods.

Commentary by Dr. Valentin Fuster
2002;():349-354. doi:10.1115/FEDSM2002-31217.

The effect of bubbles on vortical flows near walls is examined by direct numerical simulations. A simple model problem, consisting of two slightly perturbed counter rotating vortex tubes plus a parabolic pressure driven velocity profiles is used. Initially, the flow is seeded with several bubbles near the wall. As the flow evolves, the bubbles are entrained into the vortices, where they accelerate the breakdown of the vortices. The flow rate and the wall shear is monitored and the result show that the presence of bubbles leads to a slight decrease in drag and thus an increase in flow rate.

Topics: Bubbles , Vortex flow
Commentary by Dr. Valentin Fuster
2002;():355-364. doi:10.1115/FEDSM2002-31218.

This paper describes an extension and validation of the Euler/Lagrange approach for three-dimensional time-dependent calculations of the flow in a bubble column. The fluid phase was calculated based on the Euler approach solving the unsteady Reynolds-averaged Navier-Stokes equations in a time-dependent way. The conservation equations were closed using the standard k-ε turbulence model. The coupling between the phases is considered through the momentum source terms and source terms in the k- and ε-equations. The usage of the Consistent term for the k-equation and taking into account an additional dissipation due to the presence of small bubbles yielded a reasonable agreement of the predicted turbulent kinetic energy level with measurements. Bubble motion was calculated by solving the equations of motion taking into account drag force, pressure, added mass force, transverse lift force, buoyancy and gravity. Numerical calculations are presented providing information on the sensitivity of the results on several boundary conditions, such as disturbed aeration. The computational results are validated based on available measurements in a laboratory-scale bubble column.

Commentary by Dr. Valentin Fuster
2002;():365-371. doi:10.1115/FEDSM2002-31219.

A new approach and numerical method for study gas-liquid two-phase flows in elastic pipes is suggested. “A nonlinear wave dynamical model for liquid containing gas bubbles” is applied to derive governing equations for two-phase flow-filled pipelines. On assuming the hydraulic approximation the continuity and momentum equations of two-phase flow in a pipe are obtained for the first time. From these equations the inhomogeneous wave equation of Lighthill-type for two-phase flow in pipelines is derived. The shear stress at the tube surface, deformation of the tube cross-section, and liquid’s phase compressibility are taken into account. A high effectively and accurate finite difference technique for the exact solution of the basic equations in the case of Neumann boundary conditions is developed. Based on the proposed algorithm various numerical experiments have been carried out to investigate the major fluid dynamical features of hydraulic shocks and shock waves in the horizontal pipes. Comparisons with both the experimental data and computational results obtained with a second-order accurate predictor-corrector method support our numerical technique as well as the model.

Commentary by Dr. Valentin Fuster
2002;():373-380. doi:10.1115/FEDSM2002-31220.

An efficient algorithm for solving the equations of multi-phase flow has been developed and implemented in the software CFX-5. The motivation for this method is to obtain scalable behaviour for solving industrial multiphase flow problems. This is achieved using a finite volume discretization of the equations on unstructured meshes, preserving important inter-equation coupling at the algebraic level, and solving the algebraic equations using an efficient multigrid solver. The method also features a high resolution advection scheme for the volume fraction equations, which is particularly important for obtaining accurate solutions on tetrahedral meshes. The algorithm is also parallelized so that fast turn-around times can be achieved. The method is validated for an Enichem airlift loop reactor studied in an ADMIRE project. The results compare well with experimental data and with previous CFX-4 calculations. Solutions are obtained in the same number of iterations independent of the mesh size and with near-linear parallel speedup.

Commentary by Dr. Valentin Fuster
2002;():381-384. doi:10.1115/FEDSM2002-31221.

Effect of turbulence on drag force in gas-particle two-phase flow had been investigated using numerical simulation. In order to select an accurate turbulence model, some promising models, such as standard k-ε model, RNG k-ε model and Realizable k-ε model, had been examined through calculating the flow over a backward-facing step. RNG k-ε model performing better than others had been used to simulate the turbulence flow over a spherical particle. In computation, the turbulence intensity was ranged from 10% to 80%, and the turbulence length scale from 10−5 m to 4m. Results show that the turbulence length scale had a small effect on the drag force, except at small length scale. Comparing with the drag on a particle in laminar flow, the turbulence intensity enhances it comparatively, especially at small particle Reynolds numbers, which differs from the previous publications.

Commentary by Dr. Valentin Fuster
2002;():385-398. doi:10.1115/FEDSM2002-31222.

An efficient algorithm of numerical simulation of two-way coupled viscous flows of a dusty gas with collisions between particles is described. The flow of a carrier gas which is treated as a continuum is simulated by solving the modified Navier-Stokes equations using a CFD-method. The reverse effect of particles on a gas flow is modelled by the source terms entered into the momentum and energy equations. A dispersed phase is treated as a discrete set of particles which move in the carrier gas and can collide with each other. The particle drag force, the Magnus lift force, the damping torque and the heat exchange are taken into account in gas-particle interaction. Particles are assumed to collide inelastically and frictionally. A modified majorant frequency scheme of the Direct Simulation Monte Carlo (DSMC) method is proposed for computations of flow fields of a collisional “gas” of particles. The developed combined CFD / DSMC method is applied to the study of the supersonic gas-particle flow over a blunt body.

Commentary by Dr. Valentin Fuster
2002;():399-407. doi:10.1115/FEDSM2002-31223.

Interaction of solid particles with shock and expansions in supersonic flows is analyzed. In this analysis, a dense cloud of solid particulates is modeled by using a fully Eulerian approach. The dispersed flow and the gas flow were considered in the Eulerian frame whereby most of the physical aspects of the gas-particle flow can be incorporated. In addition to the momentum and energy exchanges in the form of source terms appearing in the governing equations, the two phases were strongly coupled by considering the volume fraction of the particulate phase in the equations. The simulation performed for a High Velocity Oxy-Fuel (HVOF) process under typical operating conditions in which the powder loading is high and the two-phase flow is not dilute near the injection port. The simulations showed large variations in the flow regime in the region that most of the particles exist. Unlike the results corresponding to the Lagrangian approach, the flow becomes subsonic near the centerline and the drag force decreases significantly since the relative Mach number is small. The validation experiments showed that the variation of flow regime by changing the relative Mach number could significantly change the particle drag force, and consequently process efficiency.

Commentary by Dr. Valentin Fuster
2002;():409-422. doi:10.1115/FEDSM2002-31224.

An efficient numerical method developed recently in Russia for calculations of dilute particle phase flow fields in two-phase gas-particle flows is presented in this paper. The method is based on the full Lagrangian approach (FLA) to the description of the regular motion of the particle phase. The collisionless “gas” of particles is treated as a continuum for which all basic equations, namely, the continuity equation, the momentum equation and the energy equation are written in the Lagrangian coordinates. The traditional Lagrangian approach (LA) widely used in the West does not deal with the explicit form of the continuity equation for the particle phase, and this is the key difference between FLA and LA. The FLA-method has some advantages over the traditional Lagrangian technique. Examples of numerical calculations of two-phase flows with a complex particle phase flow structure illustrate this method.

Topics: Flow (Dynamics)
Commentary by Dr. Valentin Fuster
2002;():423-430. doi:10.1115/FEDSM2002-31225.

An approach to parallel solution of an Eulerian-Lagrangian model of dilute gas-solid flows is presented. Using Lagrangian treatments for the dispersed phase, one of the principal computational challenges arises in models in which inter-particle interactions are taken into account. Deterministic treatment of particle-particle collisions in the present work pose the most computationally intensive aspect of the simulation. Simple searches lead to algorithms whose cost is O(N2 p ) where Np is the particle population. The approach developed in the current effort is based on localizing collision detection neighborhoods using a cell-index method and spatially distributing those neighborhoods for parallel solution. The method is evaluated using simulations of the gas-solid turbulent flow in a vertical channel. The instantaneous position and the velocity of any particle is obtained by solving the equation of motion for a small rigid sphere assuming that the resulting force induced by the fluid reduces to the drag contribution. Binary particle collisions without energy dissipation or inter-particle friction are considered. The carrier flow is computed using Large Eddy Simulation of the incompressible Navier-Stokes equations. The entire dispersed-phase population is partitioned via static spatial decomposition of the domain to maximize parallel efficiency. Simulations on small numbers of distributed memory processors show linear speedup in processing of the collision detection step and nearly optimal reductions in simulation time for the entire solution.

Commentary by Dr. Valentin Fuster
2002;():431-438. doi:10.1115/FEDSM2002-31226.

This paper is dedicated to Lagrangian approach for modelling particle-particle interactions in gas-solid turbulent flows. This approach is based on a joint fluid-particle pdf equation solved using Monte Carlo method which has been developed to take into account the correlation between colliding particles induced by the fluid. The modification of the turbulence by the particles is not included in the simulations. The Lagrangian results are presented in comparison with LES simulations (Laviéville et al., 1995) and continuum model.

Commentary by Dr. Valentin Fuster
2002;():439-446. doi:10.1115/FEDSM2002-31227.

A new mesh adaptation technique for two dimensional deforming domains has been combined with an Arbitrary-Lagrangian-Eulerian, spectral/hp finite element solver. This combination provides a powerful simulation tool for two-fluid flows; With the spectral/hp finite element method, high accuracy solutions can be obtained efficiently, but by itself this technique is limited to problems with moderate interface deformation. By combining it with a mesh adaptation strategy for deforming domains, large deformation problems can be simulated. In addition, solution-based mesh refinement can be performed which is critical for resolving multi-scale two-fluid problems. The capabilities of these techniques together are demonstrated for some preliminary examples.

Commentary by Dr. Valentin Fuster
2002;():447-457. doi:10.1115/FEDSM2002-31228.

The investigation of three-dimensional transient propagations of dry-out fronts within a nuclear fuel rod bundle is performed, based on experimental and numerical simulations. The spreading of critical heat flux (CHF) fronts across a bundle, caused by gradual decrease of coolant mass flow rate without power increase is predicted, and the locus of dry patches is shown. Obtained results show that a simultaneous occurrence of CHF and rewet multi-fronts in simulated reactor coolant flow transient are less pronounced than in power transients. Due to a possible building of a vapour zone, the CHF front spatial propagation has to be carefully analysed in transient conditions.

Commentary by Dr. Valentin Fuster
2002;():459-466. doi:10.1115/FEDSM2002-31229.

We have developed a cubic interpolated propagation (CIP) code combined with a finite volume method using unstructured grid spaces. The CIP code, which can solve both compressible and incompressible flows simultaneously, was used to calculate gas-liquid flows — in this case, waterspouts — in an oil separator. We found that waterspouts raise the oil at the bottom of the separator’s chamber and lower the efficiency of oil separation remarkably. We also found that the waterspouts can be classified as circulatory or non-circulatory and that they are caused by a low-pressure core in the discharge pipe. Furthermore, we predicted the critical conditions under which the waterspouts occur, and these predictions agree with measurements taken with a test separator.

Topics: Space , Two-phase flow
Commentary by Dr. Valentin Fuster
2002;():467-472. doi:10.1115/FEDSM2002-31230.

We present a review of the CIP method, which is a kind of semi-Lagrangian scheme and has been extended to treat incompressible flow in the framework of compressible fluid. Since it uses primitive Euler representation, it is suitable for multi-phase analysis. The recent version of this method guarantees the exact mass conservation even in the framework of semi-Lagrangian scheme. Comprehensive review is given for the strategy of the CIP method that has a compact support and sub-cell resolution including front capturing algorithm with functional transformation.

Topics: Flow (Dynamics)
Commentary by Dr. Valentin Fuster
2002;():473-483. doi:10.1115/FEDSM2002-31231.

Computer simulations of multiphase flows are currently limited to two distinct regimes: Fluid elements that are much larger than the computational grid cells can be modeled using fully-resolved calculations, and fluid elements that are much smaller than the grid cells can be modeled using point-particle approximations. However, many flows involve fluid elements for which neither of these limits is appropriate. For simulations of spray atomization, in particular, the grid size needs to be small enough to resolve the details of the injected fluid stream, yet resolving the smallest droplets produced would require an impossibly large number of grid points. To simulate spray atomization or other flows with a wide range of fluid element sizes, then, it is necessary to develop a computational framework that can bridge the fully-resolved and point-particle regimes. The development of such a framework is the topic of this paper. We begin by borrowing the mathematical framework of “large-eddy simulations” of turbulence—the equations for the flowfield are mathematically filtered, producing a set of equations which are smooth above a certain size and contain a set of terms which model the effects of the smaller scales. Because the filtering provides a mathematically valid smoothing of the fluid interfaces, it is also a natural way of simulating fully-resolved fluid elements on a fixed grid. The subfilter-scale terms can be shown to be zero in the fully-resolved limit, and so fully-resolved simulations provide an appropriate test of the basic filtered-multiphase method, independent of the complex subgrid modeling that would be necessary for unresolved elements. In anticpation of future calculations of droplets, we demonstrate the method with two-dimensional calculations of fluid cylinder oscillations, and compare the results to an analytical solution.

Topics: Multiphase flow
Commentary by Dr. Valentin Fuster
2002;():485-494. doi:10.1115/FEDSM2002-31232.

This paper presents a model dedicated to the numerical simulation of two-component liquid-vapor flows with phase-change. This model is an extension to binary mixtures of the second gradient method, initially dedicated to pure fluids. It is a diffuse interface method: by assuming that the free energy of the mixture depends on its density gradient, liquid-vapor interfaces are described as volumetric transition regions across which physical properties vary continuously. The corresponding governing equations of the mixture are derived and the thermodynamic closure relation is established in order to recover the equilibrium properties of a mixture. As a first validation of this model, it is applied to study a one-dimensional isothermal phase-change problem. When the system reaches a stationary state, an asymptotic analysis shows that this model is in good agreement with sharp interface models, as well as numerical calculations.

Commentary by Dr. Valentin Fuster
2002;():495-501. doi:10.1115/FEDSM2002-31233.

The context of this study is the prediction of re-ignition for turbojet engines after in-flight extinction at high altitude. Experiments have been performed on a simple geometry of a combustion chamber to test ignition at ambient conditions for three positions of the spark plug. Then, the two-phase flow corresponding to the experimental configuration has been simulated with the eulerian-lagrangian code used at ONERA. In parallel, a time dependent 0-dimensional model has been developed to predict the ignition of a cluster composed of fuel droplets when it is submitted to the spark inside the combustion chamber. This model has been applied on the two-phase flow computation in three elementary volumes located close to different spark plug positions. Ignition has been tested numerically for these clusters of drops, whose characteristics are dependent of their location in the combustion chamber, as well as, of the two-phase flow configuration in the geometry. Comparisons between experimental and numerical results are presented in this paper.

Topics: Sprays , Ignition , Flight
Commentary by Dr. Valentin Fuster
2002;():503-510. doi:10.1115/FEDSM2002-31234.

A three-dimensional (3-D) model of spreading and solidification was used to investigate the sequential deposition of two tin droplets for different offset landing. Numerical simulations predicted the shape and size of the landing tin droplet as it spread over a previously landed splat. The model applies a fixed-grid Eulerian control volume to solve the fluid dynamics and energy equations. The Volume of Fluid (VOF) algorithm is used to track the free surface deformation. The comparison of the simulated images and experimental photographs validated the prediction of the model.

Commentary by Dr. Valentin Fuster
2002;():511-522. doi:10.1115/FEDSM2002-31235.

We consider the axisymmetric flow of a Newtonian fluid associated with the spreading of a thin liquid film on a rotating annular disk. The effects of surface tension and gravity terms are included. The asymptotic solution for the free surface of the thin film is found using an expansion for the film thickness in powers of a small parameter characterizing the thickness of the film and applying the method of matched asymptotic expansion. This solution can be used to calculate the thickness of the film, the velocity field, and the pressure at any point on the disk with good accuracy. Numerical results are presented for a specific initial distribution of the film thickness. Many features of the spin-coating thinning process are captured by our asymptotic solution. We also produce results which are in excellent agreement with the experimental findings of Daughton and Givens (6) and Hwang and Ma (9).

Commentary by Dr. Valentin Fuster
2002;():523-529. doi:10.1115/FEDSM2002-31236.

This study aims to investigate the effect of topology change on the rise velocity of bubbly flows and the phase distribution in a channel at a moderate Reynolds number. A front tracking/finite difference method is used to solve the momentum equation inside and outside deformable bubbles. It is found that bubble/bubble coalescence enhances the average rise velocity of the bubbles dramatically and also increases the fluctuations of the liquid velocity. Examination of the pair distribution function shows that the flow becomes more non-homogeneous as a result of topology change.

Topics: Buoyancy , Bubbly flow
Commentary by Dr. Valentin Fuster
2002;():531-536. doi:10.1115/FEDSM2002-31237.

Direct numerical simulations are used to examine the effect of electric fields on the behavior of suspension of drops in dielectric fluids. The effect of electric field is modeled using the “leaky dielectric” model, coupled with the full Navier-Stokes equations. The governing equations are solved using a front-tracking/finite volume technique. The interaction of the drops is strongly dependant on the conductivity and the permittivity ratio, but fibration, where drops line up into long columns, takes place over a wide range of these parameters. The hydrodynamic interaction due to fluid circulation induced by the electric field has a strong influence on the drop distribution and the rate of fibration.

Commentary by Dr. Valentin Fuster
2002;():537-538. doi:10.1115/FEDSM2002-31238.
FREE TO VIEW

Numerical simulation continues to evolve as an important tool in the analysis and prediction of two-phase turbulent flows. Computations are playing an increasingly important role as both a means for study of the fundamental interactions governing a process or flow, as well as forming the backbone for engineering predictions of physical systems. At a practical level, computations for engineering applications continue to rely on solution of a statistically-averaged equation set. Many of the statistical correlations requiring closure in Reynolds-averaged models are often difficult or impossible to measure in experimental investigations of two-phase flows. Computational techniques that directly resolve turbulent eddies are an important component in evaluating closure models, while at the same time offering a useful approach for basic studies of fundamental interactions. The focus of the lecture is on numerical prediction and study of turbulent two-phase flows using computational techniques such as Large Eddy Simulation (LES) that directly resolve the large, energy-containing scales of the turbulent motion. Within this broad class, the subset of two-phase flows in which a dispersed phase is comprised of small particles and is present at low volume fractions is of primary interest, using Lagrangian computational techniques for the prediction of trajectories of a large ensemble of discrete particles. The scope of such an approach considered is on systems in which the ensemble comprising the particulate phase is large enough that direct resolution of the flow in the vicinity of each particle is not feasible and, consequently, models on fluid-particle interfacial transfer and particle-particle interaction must be imposed. The focus of the lecture is on numerical prediction and study of turbulent two-phase flows using computational techniques such as Large Eddy Simulation (LES) that directly resolve the large, energy-containing scales of the turbulent motion. Within this broad class, the subset of two-phase flows in which a dispersed phase is comprised of small particles and is present at low volume fractions is of primary interest, using Lagrangian computational techniques for the prediction of trajectories of a large ensemble of discrete particles. The scope of such an approach considered is on systems in which the ensemble comprising the particulate phase is large enough that direct resolution of the flow in the vicinity of each particle is not feasible and, consequently, models on fluid-particle interfacial transfer and particle-particle interaction must be imposed. The advantages and limitations of such a technique are first considered and its accuracy is evaluated by comparison with discrete particle simulations coupled with fluid turbulence predictions obtained using DNS (understood in the present context as solution of the carrier-phase flow without the use of explicit subgrid turbulence models). An overview and examples of the application of LES to prediction and scientific study of dispersed, turbulent two-phase flows is then presented for several representative flow configurations: statistically stationary and decaying particle-laden isotropic turbulence, homogeneous shear flow, fully-developed turbulent channel flow, and turbulent particle-laden round jet. In such flows, the detailed description possible using LES enables in-depth evaluations of statistical and structural features. In particular, the role of inter-particle collision in turbulent channel flow and more recent efforts focused on exploration and analysis of the spatial structure of the particle concentration and velocity fields in homogeneous turbulence are discussed.

Commentary by Dr. Valentin Fuster
2002;():539-547. doi:10.1115/FEDSM2002-31239.

In wall-bounded gas-solid flows the wall collision process plays an important role and may be strongly affected by wall roughness and particle shape. The modelling of the particle-wall collision mostly relies on the assumption of spherical particles. To extend such models appropriately for non-spherical particles, two-dimensional kinetic simulations were performed for different particle shapes. This implies, that the particle translational and angular motion is calculated by considering the particle shape, however neglecting fluid dynamic effects. The change of the particle velocities during the impact and rebound process was calculated by solving the impulse equations together with Coulombs law of friction. The simulations were performed for a given initial particle velocity by varying impact angle and initial angular velocity. The results for 2000 particle wall collisions allowed us to derive the distribution functions of the impact parameters required to describe the wall collision process for non-spherical particles correctly. Moreover, other wall collision properties, such as rebound angle and velocity ratios could be determined. Finally also a comparison with measurements was possible.

Commentary by Dr. Valentin Fuster
2002;():549-555. doi:10.1115/FEDSM2002-31395.

Three-dimensional simulations of bidisperse bubble swarms rising in a liquid have been carried out. This article describes the microstructure of a swarm of mostly spherical bubbles representing 6% void fraction. The swarm consists of an equal number of large and small bubbles with volume ratio 2. While the behavior of the large bubbles is similar to that in a comparable monodisperse system, the behavior of the small bubbles is different.

Topics: Bubbles
Commentary by Dr. Valentin Fuster

Numerical Developments in CFD

2002;():557-578. doi:10.1115/FEDSM2002-31367.

Multidimensional numerical simulation of the atmospheric saturated pool boiling is performed under high heat fluxes, near to and at the occurrence of burnout conditions. Heat flux through the vessel bottom wall is varied and its influence on the pool boiling dynamics is analysed. Dynamics of vapour generation on the heating wall is modelled through the density of nucleation sites and the bubble residence time on the wall. The nucleation sites are determined by a random function. The applied numerical grid is able to represent the nucleation sites on the heating wall for both fresh (polished) and aged (rough) heaters at the atmospheric pool boiling conditions. Results are presented for short time period after the initiation of heat supply and vapour generation on the heating surface, as well as for quasi steady-state conditions after two seconds from pool boiling initiation. The results show a replenishment of the heating surface with water and partial surface wetting for lower heat fluxes, while heating surface dry-out is predicted for high heat fluxes. The influence of the density of nucleation sites and the bubble residence time on the wall on the pool boiling dynamics is investigated. Numerical simulations show that decrease of the density of nucleation sites and increase of bubble residence time on the heating surface (characteristics pertinent to fresh-polished heaters) lead to the reduction of critical heat flux values. Obtained results are in excellent agreement with the recent experimental investigations of the upward facing burnout conditions on the horizontal heated plate. Details of the developed numerical procedure are presented. The introduced method of random spatial and temporal generation of the vapour at the heated wall is a new approach. It enables the macroscopic representation of the population of microscopic vapour bubbles, which are generated at nucleation sites on the heater wall, and which burst through liquid micro-layer in thermal-hydraulic conditions close to the burnout. The applied numerical and modelling method has shown robustness by allowing stable calculations for wide ranges of applied modelling boiling parameters (density of nucleation sites and bubble residence time).

Commentary by Dr. Valentin Fuster
2002;():579-586. doi:10.1115/FEDSM2002-31368.

A domain decomposition method for the Stokes problem using Lagrange multipliers is described. The dual system associated with the Lagrange multipliers is solved based on an iterative procedure using the two-level finite element tearing and interconnecting (FETI) method. Numerical tests are performed by solving the driven cavity problem. An analysis of the number of outer iterations and an evaluation of the cost of the inner iterations are reported. Comparison with the well-known Uzawa algorithm shows a reduction in the floating point operations count of the inner iterations while achieving the same number of outer iterations.

Commentary by Dr. Valentin Fuster
2002;():587-596. doi:10.1115/FEDSM2002-31369.

The formulation, testing, and application of some mathematical models and numerical solution methods for tracking electrically non-conducting particles in laminar flow of molten aluminum in straight, constant cross-section ducts subjected to steady, uniform electric and magnetic fields are presented. The separation chamber is a straight duct of constant rectangular cross-section and electrically non-conducting walls. The effects of the induced electric field in the molten aluminum are considered negligible. Attention is focused on the fully-developed flow regime. In the calculation of the inclusion trajectories, consideration is given to drag, gravity, virtual mass, Basset, lift, and electromagnetic forces. A control-volume finite element method is used for the calculation of the fluid flow. A fourth-order Runge-Kutta method and a particle-location and interpolation algorithm are used to calculate the particle velocities and trajectories. Results obtained for a test problem and a demonstration problem are presented and discussed in this paper.

Commentary by Dr. Valentin Fuster
2002;():597-604. doi:10.1115/FEDSM2002-31370.

A grid-free method for the LES of turbulent incompressible flow is presented. The computational engine is based on the Lagrangian Vortex Element Method to account for the dynamics of the resolved vorticity field and the Vorticity Redistribution Method (VRM) to account for turbulent diffusion. Turbulence is modeled here using the standard Smagorinsky subgrid-scale model but work is underway to implement a dynamic version. In this paper, the accuracy and convergence rate of VRM are presented for the case of diffusion with variable viscosity. Furthermore, the adaptivity as well as the robustness of the proposed method in capturing large-scale vortex structures typical of vortex dominated turbulent flow are demonstrated using preliminary results from the LES of the collision of a pair of coaxial vortex rings, perturbed initially by the wave-number of their most unstable mode.

Topics: Flow (Dynamics)
Commentary by Dr. Valentin Fuster
2002;():605-613. doi:10.1115/FEDSM2002-31371.

Since the goal of racing is to win and since drag is a force that the vehicle must overcome, a thorough understanding of the drag generating airflow around and through a race car is greatly desired. The external airflow contributes to most of the drag that a car experiences and most of the downforce the vehicle produces. Therefore, an estimate of the vehicle’s performance may be evaluated using a computational fluid dynamics model. First, a computer model of the race car was created from the measurements of the car obtained by using a laser triangulation system. After a computer-aided drafting model of the actual car was developed, the model was simplified by removing the tires, roof strakes, and modification of the spoiler. A mesh refinement study was performed by exploring five cases with different mesh densities. By monitoring the convergence of the computed drag coefficient, the case with 2 million elements was selected as being the baseline case. Results included flow visualization of the pressure and velocity fields and the wake in the form of streamlines and vector plots. Quantitative results included lift and drag, and the body surface pressure distribution to determine the centerline pressure coefficient. When compared with the experimental results, the computed drag forces were comparable but expectedly lower than the experimental data mainly attributable to the differences between the present model and the actual car.

Commentary by Dr. Valentin Fuster
2002;():615-626. doi:10.1115/FEDSM2002-31372.

The goal of this paper is to develop a dynamic algorithm that can be used in conjunction with computational fluid dynamics (CFD) simulation codes to quantify the discretization error in a selected process variable. The focus is on fluid dynamics applications where conservation equations are solved for primitive variables using finite difference and/or control volume approach. A transport equation for the error is formulated and solved along with a localized residual estimation based on modified equation concept. Spatiotemporal evolution of the error distribution is mapped and compared to exact error for various cases. A new method is suggested for deriving the modified equation specifically aimed at using it with commercial CFD codes which use finite volume approach.

Topics: Equations , Errors
Commentary by Dr. Valentin Fuster
2002;():635-639. doi:10.1115/FEDSM2002-31374.

In this paper, a full three-dimensional inverse method for the design of mixed-flow pump runner is described. The three-dimensional flow in the runner is decomposed into a tangential mean flow and a tangential periodic flow. The blades are represented by superposition of vortices and sources on the blade mean surface, and the blade mean surface is determined by the inverse method. In this method, the distribution of the circumferentially mean swirl VRθ on the meridional geometry of the runner is prescribed and the corresponding blade shape is computed iteratively. The new method is applied to the design of a mixed-flow pump runner and the result is satisfactory.

Commentary by Dr. Valentin Fuster

Non-Invasive Measurement in Multiphase Flows

2002;():641-648. doi:10.1115/FEDSM2002-31376.

The effect of liquid properties on axial development of gas-volume-fraction profiles in bubble-column flows is investigated. Experiments are conducted in a cylindrical vessel with an inner diameter of 0.48 m and a height of 3 m. The liquids examined include water and two lightweight mineral oils. A cross sparger with 96 holes is used to inject air into the column with all the holes facing either upwards or downwards. The superficial gas velocity ranges from 5 to 30 cm/s, and the absolute column pressure ranges from 0.1 to 0.5 MPa. Gamma-densitometry tomography (GDT) is used to measure radial distributions of gas volume fraction at eight axial locations. The development length of the gas-volume-fraction profile is shown to increase with gas velocity and column pressure for all three liquids. The development of the cross-sectionally averaged gas-volume fraction for the air/water flow is remarkably different from that for the air/oil flows.

Commentary by Dr. Valentin Fuster
2002;():649-654. doi:10.1115/FEDSM2002-31377.

An implementation of resistive electrical-impedance tomography (EIT) for measuring material distributions of multiphase flows in vessels with electrically conducting walls is presented. Seven ring electrodes are equally spaced on a thin nonconducting rod that is inserted into the vessel. The vessel wall is grounded and serves as the ground electrode. Voltage distribution measurements are used to numerically reconstruct the time-averaged impedance distribution within the vessel, from which the material distributions are inferred. Experimental results for the case in which the rod is inserted coaxially into a liquid-filled vertical standpipe containing beds of different heights of nonconducting solid particles are presented. Agreement between the direct measurement and the numerical reconstruction of the particle-bed height is good. Application of this technique to a pilot-scale slurry bubble-column reactor is discussed.

Commentary by Dr. Valentin Fuster
2002;():655-662. doi:10.1115/FEDSM2002-31434.

A new reconstruction method, which is called sampled pattern mating method (SPM), has been applied to an ill-posed inverse problem of a capacitance-computed tomography for solid air two-phase flow. As a result, the accuracy of the reconstructed image is improved as compared with a conventional Newton-Rampson iterative method. Moreover, the particles volume ratio in solid air two-phase flow calculated by SPM method is more accurate than the conventional method.

Topics: Two-phase flow
Commentary by Dr. Valentin Fuster
2002;():663-670. doi:10.1115/FEDSM2002-31435.

A particle image velocimetry (PIV) technique developed for application to two-phase flows is presented and validated. The technique is capable of simultaneously measuring carrier and bubble phase velocities on a plane. The validation experiments have been conducted in a vertical upwards, two-phase (water-air) bubbly jet flow at a Reynolds numbers of 5,673 and 11,345 and low bubble concentration matching the experiments of Stanley and Nikitopoulos (1998). Comparisons with measurements obtained by Stanley and Nikitopoulos (1998) using Phase Doppler Analysis (PDA) experiments indicate that the agreement between the two techniques is very satisfactory (deviations of the order of 5%) for both single-phase and two-phase jet carrier-flow velocities. In addition, bubble phase velocity measurements obtained from backlit visualizations of the bubbly jet flow using the bubble-tracking method of Fiedler et al. (2001) are successfully compared to those obtained from PIV. The PIV study confirms that bubbles experience a substantial deceleration in the unmixed core of the jet near field and illustrates carrier-phase mean-flow modification consistent with past point-wise measurements.

Commentary by Dr. Valentin Fuster
2002;():671-676. doi:10.1115/FEDSM2002-31436.

Two different three-dimensional reconstruction techniques for the shape of gas bubbles flowing in a liquid are presented. The first technique is based on the Dynamic Generalized Hough Transform Algorithm, and the second on the Metaball Model. These techniques are suitable for analysis of turbulent two-phase bubbly flows. Both techniques require at least two views of the bubble intended for three-dimensional reconstruction, and can be used in either stereoscopic or orthogonal camera setups. Once the reconstruction is accomplished, the bubble images can be accurately removed from the images acquired during Particle Image Velocimtery or Shadow Image Velocimetry measurements. After removing the bubble images from PIV images, a typical analysis of the liquid phase can be performed. This improves the accuracy of the statistical analysis of the parameters of each phase.

Commentary by Dr. Valentin Fuster
2002;():677-684. doi:10.1115/FEDSM2002-31438.

Aerothermal properties in a fuel spray is a central problem in the field of the design of the combustion chambers of automobile engines, turbojets or rocket engines. Heat and mass transfer models are necessary in the predictive calculation schemes used by the motorists. Reliable experimental data must be obtained for both the validation and development of new physical models linked to heat transfer and evaporation in sprays, where aerodynamic interactions has a key role. This paper proposes an experimental study of the energetic budget of a monodisperse ethanol droplet stream, injected in the thermal boundary layer of an heated plate. The droplet size reduction is measured using a light scattering technique (interferential method) in order to characterize the evaporation, as the droplet mean temperature is monitored using the two colors laser-induced fluorescence technique. The convection heat transfer coefficient and the Nusselt number are inferred from the overall energetic budget, as a function of the inter-droplet distance, characterizing the interaction regime. The results are compared to physical models combined with numerical simulations available in the literature, for moving, evaporating isolated droplets and for three droplets arrangement in linear stream.

Commentary by Dr. Valentin Fuster
2002;():685-691. doi:10.1115/FEDSM2002-31455.

The identification of two-phase flow patterns has been widely studied, and the diagnostic procedures are traditionally based on statistical or spectral signal analysis, while the spatial information related with the geometrical topology of the phase distribution in the pipe is never taken into account. The aim of this study is to demonstrate how the exploitation of both spectral and spatial information leads to an unambiguous identification of the flow patterns. Experiments are performed on a 30 meters long horizontal air-water loop. By simultaneously analyzing the power spectral density of the signals delivered by a multi-electrode impedance sensor, we obtain a space-frequency representation which exhibits particular features of the different flow regimes. They can be characterized by a set of 3 scalar parameters, quantifying respectively the localization in space, in frequency and the shape of the spectral content. The final demonstration of this space-frequency characterization is provided by the use of a multi-layer neural network, trained on a 80 tests database. This net exhibits a successful identification rate above 80% when used in blind real-time tests.

Commentary by Dr. Valentin Fuster
2002;():693-699. doi:10.1115/FEDSM2002-31456.

This paper describes the development and validation of a high spatial resolution X-ray tomograph devoted to investigate air-water two-phase flows. The device hardware is mainly composed of a 60 keV X-ray source, a detector and an accurate mechanical bench. This paper concentrates on accuracy quantification and emphasis is given on the reconstruction procedure. It is well known that absorption gradients induce reconstruction artifacts when using standard algorithms based on uniform regularization. In the particular case of two-phase flows in pipes, this leads to a poor measurement accuracy in the vicinity of the walls. To overcome such effects, improved algorithms have been developed in this study, involving different spatially adaptative regularization methods. A first calibration performed on static phantoms clearly exhibits the benefit brought by such advanced reconstruction algorithms. A validation procedure has been carried out on an air-water bubble column, equipped with an optical probe which can be translated in order to explore the 80 mm × 80 mm square cross section. Comparisons of local void fraction measurements have been performed pixel by pixel, and demonstrate the accuracy improvement induced by the advanced reconstruction algorithms.

Commentary by Dr. Valentin Fuster

Advances in Numerical Modeling of Aerodynamics and Hydrodynamics in Turbomachinery

2002;():703-708. doi:10.1115/FEDSM2002-31175.

This paper describes the use of a rotating all mirror image derotator system in collaboration with Particle Image Velocimetry (PIV) to visualise and quantitatively examine the flow patterns between the blades of a centrifugal impeller. The authors have been able to obtain the relative velocities in a centrifugal impeller at rotational speeds between 300 and 600 revs per minute with water flow rates between 450 and 900 litres per hour. Velocity contours and vector maps of the relative flow field within a blade passage are presented for an impeller speed of 350 rpm and a flow rate of 510 l/h. The data are compared with the results of a computational fluid dynamics (CFD) model.

Commentary by Dr. Valentin Fuster
2002;():709-719. doi:10.1115/FEDSM2002-31176.

Advances in Computational Fluid Dynamics have made it possible to visualize some of features of flow through rotating machines, that are if not impossible, difficult to measure experimentally. Now Industry standard CFD codes are commercially available and are being used by leading industries as a design verification tool. Most of the commercial codes are general purpose, keeping in mind vast application needs. Various turbulence models, numerical schemes and grid types are given as an option for the end user. Appropriate choice of grid, numerical scheme, interfaces and turbulence models are the key parameters that influence the results from CFD software. A centrifugal pump was designed and developed using empirical and optimization tools developed in-house. The flow passages were checked for smooth flow pattern using CFD tools. The pump was manufactured and tested for its performance. The tested performance matched well with predictions within 3% at best efficiency point as regards to efficiency and head. This model was taken to carry out numerical experiments using CFD tools. The entire performance including flow-head-efficiency and cavitation performance in terms of NPSHr requirement was available. Numerical experiments were carried out to study impact of different parameters like numerical scheme, turbulence models, type of interface between stationary and rotating parts. The results from these numerical experiments are presented and discussed in the light of actual test results from laboratory testing.

Commentary by Dr. Valentin Fuster
2002;():721-727. doi:10.1115/FEDSM2002-31177.

A CFD method, previously developed by the authors for compressible flows, has been modified through a preconditioning technique to account for purely incompressible flows. Such a code is used to compute three-dimensional flows in a mixed flow pump impeller at design and off-design conditions. The results of the inviscid flow approach are critically discussed by comparison to available experimental data.

Commentary by Dr. Valentin Fuster
2002;():729-734. doi:10.1115/FEDSM2002-31178.

For the centrifugal compressor aerodynamic design of a turbocharger, first of all, the works for system matching to the engine specification must be preceded. Then, mean line design together with performance prediction should be carried out for preliminary design. In the mean line prediction, a slip factor is adopted as a function of flow coefficient and geometry instead of Wiesner’s equation, and it is found that the predicted result of slip magnitude is more accurate than that of conventional slip factor. Also, three-dimensional blade profile shape is generated on the basis of the preliminary design. The Navier-Stokes Equation solver with a turbulent model is used to find whether three-dimensionally designed geometry is reasonable by analyzing loading distribution of the blade. By investigating diffuser flow field of the simulated result, the diffuser inlet and exit angles were modified for the flow to move smoothly along the diffuser geometry. Modified performance prediction results shows better than those of original specification. Consequently, off design performance prediction results and numerical simulation result show good agreement with the experimental data. The modified design results show more increased compression ratio and efficiency than those of previous design results. The increased choke margin has made a stable operating range larger.

Commentary by Dr. Valentin Fuster
2002;():735-741. doi:10.1115/FEDSM2002-31179.

The effects of casing shape on the performance and the interaction between the impeller and casing in a small-size turbo-compressor are investigated. Numerical analysis is conducted for the compressor with circular and single volute casings from inlet to discharge nozzle. In order to predict the flow pattern inside the entire impeller, vaneless diffuer and casing, calculations with multiple frames of reference method between the rotating and stationery parts of the domain are carried out. For compressible turbulent flow fields, the continuity and three-dimensional time-averaged Navier-Stokes equations are employed. To evaluate the performance of two types of casings, the static pressure and loss coefficients are obtained with various flow rates. Also, static pressure distributions around casings are studied for different casing shapes, which are very important to predict the distribution of radial load. To prove the accuracy of numerical results, measurements of static pressure around casing and pressure difference between the inlet and outlet of the compressor are performed for the circular casing. Comparison of these results between the experimental and numerical analyses are conducted, and reasonable agreement is obtained.

Commentary by Dr. Valentin Fuster
2002;():743-749. doi:10.1115/FEDSM2002-31180.

Flow analysis was carried out for a double-suction centrifugal pump. An impeller-only model and a full pump model were used to simulate the velocity and the pressure field of the pump. Pump head and efficiency were calculated with flow rate in order to obtain general performance of the pump. The calculation results were compared to the experimental data, and satisfactory results were obtained. Also, the velocity and the pressure field of this pump were analyzed for the rated point and off-design points. Changes of the velocity and the pressure field with flow rate were investigated at impeller eye and impeller exit.

Commentary by Dr. Valentin Fuster
2002;():751-760. doi:10.1115/FEDSM2002-31181.

The 2-D vortex method and the commercially available CFD software are applied to calculate unsteady hydrodynamic forces on a diffuser pump impeller and the pressure fluctuations caused by the interaction between the impeller and the diffuser vanes. Calculated pressure and fluid forces on the impeller are compared with measured ones. The numerical analysis yields fairly accurate predictions of the fluid forces and the pressure fluctuations in diffuser passages and the pipe systems. It has been demonstrated that the fluid forces caused by the interaction between the rotor and stator vanes are small when the number of vanes on impeller and diffuser is identical. In this case, however, the local pressure fluctuations are larger in diffuser passages and the pipe systems.

Commentary by Dr. Valentin Fuster
2002;():761-768. doi:10.1115/FEDSM2002-31182.

Both experimental and numerical studies of the unsteady pressure field inside a centrifugal pump have been carried out. The unsteady patterns found for the pressure fluctuations are compared and a further and more detailed flow study from the numerical model developed will be presented in this paper. Measurements were carried out with pressure transducers installed on the volute shroud. At the same time, the unsteady pressure field inside the volute of a centrifugal pump has been numerically modelled using a finite volume commercial code and the dynamic variables obtained have been compared with the experimental data available. In particular, the amplitude of the fluctuating pressure field in the shroud side wall of the volute at the blade passing frequency is successfully captured by the model for a wide range of operating flow rates. Once the developed numerical model has shown its capability in describing the unsteady patterns experimentally measured, an explanation for such patterns is searched. Moreover, the possibilities of the numerical model can be extended to other sections (besides the shroud wall of the volute), which can provide plausible explanations for the dynamic interaction effects between the flow at the impeller exit and the volute tongue at different axial positions. The results of the numerical simulation are focused in the blade passing frequency in order to study the relative effect of the two main phenomena occurring at that frequency for a given position: the blade passing in front of the tongue and the wakes of the blades.

Commentary by Dr. Valentin Fuster
2002;():769-776. doi:10.1115/FEDSM2002-31183.

A screw-type centrifugal pump with a wide flow passage has been widely used for drainage of rainwater as well as slurries and mud to avoid the flow passage blockage with the congestion of solids. Due to the complicated configuration of this pump, the design method of this pump has not been established yet. The authors succeeded in predicting the internal flow numerically by using our own grid-generation system and a commercial 3-D N-S code, TASCflow, as a solver. In this study, the internal flow has been predicted numerically at the design point on the five impellers with different hub cone shapes in order to clarify the influence of meridian shape on the pump performance. In particular, the relationships among the pump characteristics, the back flow at the blade pressure surface and the back flow from the volute casing to the impeller exit have been discussed in the numerical results on the velocity and pressure distributions.

Commentary by Dr. Valentin Fuster
2002;():777-789. doi:10.1115/FEDSM2002-31184.

In a recent advanced aerodynamic design of turbomachinery, the physical interpretation of three-dimensional flow field obtained by a numerical simulation is important for iterative modifications of the blade or impeller geometry. This paper describes an approach to the physical interpretation of the tip clearance flow in turbomachinery. First, typical flow phenomena of the tip clearance flow are outlined for axial and radial compressors, pumps and turbines to help comprehensive understanding of the tip clearance flow. Then, a vortex-core identification method which enables to extract the vortical structure from the complicated flow field is introduced, since elucidation of the vortical structure is essential to the physical interpretation of the tip clearance flow. By use of the vortex-core identification, some interesting phenomena of the tip clearance flows are interpreted, especially focussing on axial flow compressors.

Commentary by Dr. Valentin Fuster
2002;():791-798. doi:10.1115/FEDSM2002-31185.

The rotating passages of turbomachinery contain some very interesting and complex fluid flow phenomena. This paper presents the three-dimensional turbulent flow through the impeller passages and surroundings of a mixed-flow pump. The model has five impeller blades mounted on a conical hub and nine stator blades in a diffuser which brings the diagonally outward flow back to the axial direction. This pump was tested with air, giving a nominal flow-rate of 1.01 m3 /s and 250 Pa at 1200 rpm. Temporal discretization has second order accuracy and this is in line with the discretization of convection which is also second order. For turbulence closure the standard k-e model has been applied with conventional wall functions employed at solid surfaces. For this transient, three-dimensional computation, the numerical grid has been decomposed into five separate regions in order to process these in a parallel cluster of five individual PC’s. The results show entirely reasonable correlations with published experimental data as detailed in the flow rate-head comparisons and the numerical / experimental flow fields. These outcomes allow us to confirm that such a complex transient phenomenon may be reasonably captured by employing a commercial CFD code.

Commentary by Dr. Valentin Fuster
2002;():799-806. doi:10.1115/FEDSM2002-31186.

A three-dimensional Navier-Stokes analysis was performed to investigate the tip clearance flows in a highly forward-swept axial flow fan operating at design condition. The numerical solution was based on a fractional step method, and two-layer k-ε model was used to obtain the eddy viscosity. The tip leakage vortex decayed very quickly inside the blade passage and, thus, no distinct leakage vortex appeared behind trailing edge. The main reason was the severe decrease of the streamwise velocity of the vortex. Also the interaction of the vortex with the casing boundary layer and the through-flow were other possibilities of the fast decay of the vortex. Comparison between the numerical results and LDV measurements data indicated that the complex viscous flow patterns inside the tip region as well as the wake flow could be properly predicted, but more refinement in numerical aspects are needed.

Commentary by Dr. Valentin Fuster
2002;():807-813. doi:10.1115/FEDSM2002-31187.

The open water model tests technique is well known and commonly used to predict propellers performance. In this paper, a quite different approach is intended and the main propeller variables are numerically modelled using a finite volume commercial code. Particularly, a fishing-boat propeller is numerically treated using a three-dimensional unstructured mesh. Mesh dependency and different turbulent models are considered together with an sliding technique to account for the rotation. Typical turbomachinery boundary conditions for a volume containing the propeller are imposed (inlet velocity and outlet static pressure). In order to get the open water test performance coefficients for the considered propeller (KT , KQ , η), different advance coefficient (J) are imposed as boundary conditions for the numerical model. The results of such simulations are compared with experimental data available for the open water tests of the propeller. Once the model is validated with the experimental data available, a wake field simulation would be possible and would lead to the definition of the fluid-dynamic variables (pressure, iso-velocity maps, etc.) which are needed during any design process. Also some comparisons with real scale thrust measurements are intended.

Commentary by Dr. Valentin Fuster
2002;():815-824. doi:10.1115/FEDSM2002-31188.

The quasi-steady cavitating behavior of three pumps was investigated by 3D unsteady viscous computations. The numerical model is based on the commercial code FINE/TURBO™, which was adapted to take into account the cavitation phenomenon. The resolution resorts to a time-marching algorithm initially devoted to compressible flows. A low-speed preconditioner is applied to treat low Mach number flows. The vaporization and condensation processes are controlled by a barotropic state law that links the void ratio evolution to the pressure variations. A radial pump, a centrifugal pump, and a turbopump inducer were calculated and the cavitating behaviors obtained by the computations were compared to experimental measurements and visualizations. A reliable agreement is obtained for the two pumps concerning both the head drop charts and the extension of the vapor structures. A qualitative good agreement with experiments is also observed in the case of the turbopump inducer. The accuracy of the numerical model is discussed for the three geometries. These simulations are a first attempt to simulate the complete 3D cavitating flows in turbomachinery. Results are promising, since the quasi-steady behaviors of the pumps in cavitating condition are found quantitatively close to the experimental ones. A continuing effort is pursued to improve the prediction accuracy, and to simulate unsteady effects observed in experiments, as, for example, rotating cavitation.

Commentary by Dr. Valentin Fuster
2002;():825-833. doi:10.1115/FEDSM2002-31189.

This paper presents a comparison between a coupled and a non-coupled approach for the prediction of the cavitation development in pumps. The coupling is defined here as the influence of the development of the cavitation on the main flow. Commercial CFD (computational fluid dynamic) software having a cavitation module and in-house developed code are used for this comparison. The intention of the authors is to evaluate these methods and their capabilities in predicting cavitating performance of pumps from an industrial point of view. In a first part, the two methods used are introduced and developed. In a second part, the results of these two approaches are compared for two impellers of the same specific speed having small geometrical differences leading to significant differences in the cavitation development. The ability and the benefits of the use of these different cavitation prediction approaches in the design process of a pump are finally discussed.

Topics: Cavitation , Pumps
Commentary by Dr. Valentin Fuster
2002;():835-841. doi:10.1115/FEDSM2002-31190.

Analyses on the characteristics of flow fields in the plug valve are required to improve performance and safety at severe operating conditions such as high-pressure and high-temperature. In this study, numerical and experimental analyses are carried out with various opening rates of the valve and the shapes of the plug port: circular, triangle, and modified triangle shapes. Especially, the distributions of the static pressure, velocity vector and streamlines are investigated to calculate the flow coefficient (Cv ) and the resistance coefficient (K) in each opening rate and shape. For the case of full open, the static pressure passed through the valve port has almost been recovered. However, in case of other opening rates, pressure has not permanently recovered due to pressure drop leading to loss. This phenomenon in each shape of the valve shows the different behaviors. Result of calculation show that the modified triangle shape is good on flow characteristics and the triangle one is good for the flow rate control.

Commentary by Dr. Valentin Fuster
2002;():843-850. doi:10.1115/FEDSM2002-31191.

In order to get the details of flow fields in steam turbines, three-dimensional turbulent flow calculations are useful. However in a design procedure, three-dimensional flow calculations are only possible in the last design stage, because they need in-depth boundary conditions of both geometries and flows. At such a late time in the procedure, it is difficult to go back and change main design parameters, such as flow area and stage load. Both three-dimensional flow patterns and non-equilibrium condensation caused by rapid expansions of steam have important roles with respect to steam turbine performance particularly in low-pressure sections. The steam turbine internal efficiency can be improved by taking account of these effects in the early design stage, especially in flow pattern design. This paper describes a multi-stage through-flow calculation technique including both three-dimensional flow efffects and phase changes from vapour to small droplets. To compute the high-speed two phase steam flow, a flux-splitting procedure including non-equilibrium homogeneously condensation is introduced. Three-dimensional blade forces are calculated by using angles of both blade camber and radial lean. The blade camber lines can be decided without in-depth blade geometries. Therefore this computational technique is applicable in the flow pattern design. The calculation results agree well with fully three-dimensional flow calculation and the calculation can predict supersaturating states and Wilson lines which are defined as the maximum supercooling.

Commentary by Dr. Valentin Fuster
2002;():851-858. doi:10.1115/FEDSM2002-31192.

This paper describes a new design method of blade geometry for a Francis turbine runner by using a three-dimensional inverse design method and the Computational Fluid Dynamics (CFD) technique. The design objectives are the suppression of cavitation by reducing the area in which static pressure is lower than the vapor pressure while keeping the efficiency high. In the inverse design method, it is possible to optimize the static pressure distribution in the runner by controlling blade loading parameters and/or stacking condition, which is related to a blade lean angle, for the same design specification. A Francis turbine runner was re-designed by the inverse design method for different blade loading and stacking conditions, and the flow fields were evaluated by applying CFD. It was confirmed that the present design method is very practical and effective to control low pressure region and achieve high efficiency for Francis turbine runners.

Commentary by Dr. Valentin Fuster
2002;():859-865. doi:10.1115/FEDSM2002-31193.

The numerical simulations of three types of two-phase flow in centrifugal pump impellers are described. First, the liquid-solid particle flow is modeled by an Euler-Lagrangeian approach assuming a mass concentration less than 5% and particle diameters being less than 1000 microns. The empirical erosion model to predict the local and total wear is calibrated by measurements. Second, the influence of the relative air contents on the head-drop is simulated assuming a relatively small volume fraction and applying a simple one-fluid model. The mixture is characterized by a common density depending on the flow field. Finally, the cavitating flow is studied by implementing the Rayleigh equation into the numerical procedure describing the transient process of bubble growth and collapse. The developed simulation tools are applied to predict the three types of two-phase flows in impellers. Within the defined ranges of application the simulation results agree fairly well with the experimental data.

Commentary by Dr. Valentin Fuster
2002;():867-883. doi:10.1115/FEDSM2002-31194.

Numerical predictions of three-dimensional flow and heat transfer are presented for non-rotating and rotating turbine blade cooling passages with or without the rib turbulators. A multi-block Reynolds-averaged Navier-Stokes method was employed in conjunction with a near-wall second-moment closure to provide detailed velocity, pressure, and temperature distributions as well as Reynolds stresses and turbulent heat fluxes in various cooling channel configurations. These numerical results were systematically evaluated to determine the effect of blade rotation, coolant-to-wall density ratio, rib shape, channel aspect ratio and channel orientation on the generation of flow turbulence and the enhancement of surface heat transfer in turbine blade cooling passages. The second-moment solutions show that the secondary flow induced by the angled ribs, centrifugal buoyancy, and Coriolis forces produced strong nonisotropic turbulent stresses and heat fluxes that significantly affected flow field and surface heat transfer coefficients.

Commentary by Dr. Valentin Fuster
2002;():885-891. doi:10.1115/FEDSM2002-31196.

The three-dimensional turbulent flow in a compact hydraulic machine elbow draft tube is numerically investigated for several operating conditions, covering an extended range around the best efficiency point. Comparisons with the experimental data are presented as validation. The interest is focused on the experimentally observed pressure recovery drop occurring near the best efficiency point. The flow is first analyzed locally by means of a topological analysis, then globally with an energetic approach. The study provides evidence for the role played by a Werlé-Legendre separation originating in the bend. The separation is due to the contrasting flow angles imposed by the blades, and the angle resulting from the secondary flow.

Commentary by Dr. Valentin Fuster
2002;():893-897. doi:10.1115/FEDSM2002-31197.

In this paper, authors use the parallel calculation methods to solve the incompressible turbulent flow through a pump-turbine runner. The calculation aims at probing the road using parallel calculation methods to simulate complex flow field. The simulation is conducted based on the N-S equations, by using the k-ε model. SIMPLEC algorithm [1] is adopted in the numerical procedure with body-fitted coordination [2] and staggering grid system. The calculation is carried out on the THTF “Explore 108” Cluster Computer, where Solaris8.0 plays the roles of operating system and MPI1.2 as message-passing interface. The results of parallel simulation agree well with those of serial simulation, which shows that the parallel algorithms are feasible and useful to numerical simulation.

Commentary by Dr. Valentin Fuster
2002;():899-908. doi:10.1115/FEDSM2002-31198.

Steady flow over rounded and ducted tip hydrofoils has been studied computationally using the CFDRC-ACE(U) flow solver and a k-ε turbulence model. The flow domains were gridded with a combination of C-H, H-H, tetrahedral and prism grid blocks and mesh sizes ranged from 350,000–550,000 cells. A good agreement in flow pattern was achieved between the numerical solutions and available experimental data. The computations show that the ducted tip hydrofoil sheds less bound circulation over the majority of the wing span than does a rounded tip hydrofoil with the same cross section and aspect ratio. Observation of the streamwise component of vorticity immediately downstream of the different hydrofoils shows that the vorticity from the ducted tip hydrofoil is shed in the shape of a duct instead of the concentrated circular vortex shed by the rounded tip hydrofoil.

Commentary by Dr. Valentin Fuster
2002;():909-916. doi:10.1115/FEDSM2002-31199.

Effects of tip clearance on through flows and performance of a centrifugal compressor impeller with six different tip clearances were numerically studied using CFX-TASCflow. The flow structures inside the impeller of a centrifugal compressor were visualized observing streamlines starting the leading edge of blade tips. The calculated results at the impeller exit were circumferential averaged for quantitative discussion. Flow, pressure and entropy contours at the impeller exit were largely influenced by the tip leakage flow. Tip clearance effect on the performance was decomposed into inviscid and viscous components using one-dimensional relations expressed in terms of the specific work reduction and the additional entropy generation. Both inviscid and viscous effects affected performance to similar extent, while efficiency drop was mainly influenced by viscous loss of the tip leakage flow. Performance reduction and efficiency drop due to tip clearance was proportional to the ratio of tip clearance to blade height. A simple model suggested in the present study predicts performance and efficiency drop quite successfully.

Commentary by Dr. Valentin Fuster
2002;():917-923. doi:10.1115/FEDSM2002-31200.

A computational evaluation of impeller blade moment during the impeller-diffuser interaction is performed. The moment calculated is based on the axial moment generated by fluid forces acting on a control volume surrounding an impeller blade passage. Included in the computations are unsteady flow field contributions to the unsteady impeller moment. Results are compared to numerical predictions of the impeller blade moment based on pressure and shear stress terms integrated over the blade surface. A comparison is also made to the Euler turbomachinery equation often used for calculating blade moment and energy transfer in simplified design procedures.

Topics: Impellers , Blades
Commentary by Dr. Valentin Fuster
2002;():925-934. doi:10.1115/FEDSM2002-31201.

Numerical and experimental investigations were performed to study the effects of blade loading on pump inducer performance and flow fields. To compare the performance of inducers with different blade loadings, a three-dimensional inverse design method was applied to control the blade loading distribution of inducers. Firstly, a conventional helical inducer was designed. The blade number is three and the blade angle at the tip was chosen by the conventional design method. Then, two inducers were designed using a three-dimensional inverse design method with different blade loading distributions. One inducer was designed with fore-loading and the other was designed with aft-loading, but both inducers were designed with no leading edge loading. These two inducers have the same design specification as the conventional helical inducer. The CFD (Computational Fluid Dynamics) analyses and water model tests were performed on these three inducers. Both results showed that the inlet backflow characteristics of the 3-D inverse design inducers are improved from those of the conventional inducer. It was also found that the inlet backflow characteristics of inducers that have no leading edge loading are almost same despite different blade loading distributions. The inducer designed with fore-loading showed almost the same suction performance as the conventional inducer. Cavitation visualization and FFT analysis of unstable phenomena were also performed in this study.

Commentary by Dr. Valentin Fuster
2002;():935-942. doi:10.1115/FEDSM2002-31202.

The effects of the parameters of inlet distortions on the trend of downstream flow feature in axial compressor are simulated using an integral method. Other than the ratio of drag-to-lift coefficients of the blade and the angle of incidence, the value of distorted inlet velocity is found to be another essential parameter to control the distortion propagation. With this in mind, a distortion propagation line and corresponding distortion propagation factor are proposed to express the effect of the two main inlet parameters: the angle of incidence and the distorted inlet velocity, on the propagation of distortion. From the viewpoint of compressor efficiency, the distortion propagation is further described by a compressor critical performance. The results provide a physical insight of compressor axial behavior and asymptotic behavior of the propagation of inlet distortion, and confirm the active role of compressor in determining the velocity distribution when compressor responds to an intake flow distortion.

Commentary by Dr. Valentin Fuster
2002;():943-948. doi:10.1115/FEDSM2002-31203.

Three-dimensional numerical simulation of turbulent flow and of convective heat transfer are becoming integral to the complex procedure of gas turbine blade design. The application to heat transfer presents new problems in flow prediction by Reynolds averaged methods. Many fundamental and practical developments are needed before full three-dimensional computational analysis becomes reliable. This article reviews some of our work in this area. Fundamental studies of turbine flow by Direct Numerical Simulation are also surveyed. Reynolds numbers encountered in parts of the turbine are such that DNS is practicable. Our initial simulations were of a flat plate geometry, addressing the topic of wake-induced transition. More recently flow flow in a linear, low pressure turbine cascade has been simulated, examining wake-induced transition, wake distortion, vortex formation and turbulence distortions.

Commentary by Dr. Valentin Fuster
2002;():949-954. doi:10.1115/FEDSM2002-31204.

The description of mesh evolution during a transient computation with moving walls and mesh adaptation has to respect many rules. Using a good mesh for each computational time step is important for accuracy of results. The complexity of geometry can make this objective more complex. A method has been developed to obtain a good moving mesh description with complex boundary geometry. It is based on a local observation of boundary movement and can be resumed by two main ideas: • Add cells where the volume of solution domain increases. • Slide the mesh where the boundary has a tangential displacement.

Topics: Computation , Geometry
Commentary by Dr. Valentin Fuster
2002;():955-962. doi:10.1115/FEDSM2002-31205.

This paper describes large eddy simulation (LES) of the internal flows of a high-specific-speed mixed-flow pump at low flow-rate ratios over which measured head-flow characteristics exhibits weak instability. In order to deal with a moving boundary interface in the flow field, a form of the finite-element method in which overset grids are applied from multiple dynamic frames of reference has been developed. The method is implemented as a parallel program by applying a domain-decomposition programming model. The predicted pump heads reproduce the instability and agree quantitatively well with their measured equivalents although the predicted stall takes place at somewhat lower flow-rate ratio than in the measurements. The phase-averaged distributions of the meridional- and tangential-velocity components at the impeller’s inlet and exit cross-sections were also compared with those measured by a Laser-Doppler velocimetry (LDV). Reasonably good agreements have been obtained between the computed and measured profiles. The developed LES program thus seems to be a promising design tool for a high specific-speed mixed-flow pump particularly for off-design evaluations.

Commentary by Dr. Valentin Fuster
2002;():963-972. doi:10.1115/FEDSM2002-31206.

A computational study is presented which investigates the predictive performance of two non-linear turbulence closures in simulating the physics pertinent to decelerating turbomachinery flows. The compared approaches are a cubic non-linear k-ε model and an algebraic Reynolds stress model. They have been considered as promising closures for improving the industry CFD state-of-the-art accounting for non-equilibrium effects. The authors adopt a parallel multi-grid algorithm, which is developed with a finite element formulation based on a highly accurate stabilized Petrov-Galerkin method. The finite element formulation is here applied on equal-order Q1-Q1 as well as mixed Q2-Q1 element pairs, and the accuracy of the latter approximation is assessed on near-wall flows simulation. The parallel solution algorithm for Reynolds Averaged Navier-Stokes modeling exploits an overlapping domain decomposition technique based on an “inexact explicit non linear Schwarz method”. The compressor flow considered for model benchmarking is highly challenging because of the transitional nature of the flow and the existence of significant leading- and trailing-edge separations. The potential of non-isotropic closures has been investigated. The algebraic stress model is shown to provide a better base-line for non-equilibrium effects simulation with respect to the cubic k-ε model. As it is shown for the studied compressor cascade, the cubic eddy-viscosity model exhibits some predictive weaknesses, among them an excessive turbulence attenuation that results in un-realistically delayed transition to turbulence.

Commentary by Dr. Valentin Fuster
2002;():973-979. doi:10.1115/FEDSM2002-31207.

A Wells turbine for wave power conversion has hysteretic characteristics in a reciprocating flow. The hysteretic loop is opposite to the well-known dynamic stall of an airfoil. In this paper, the mechanism of the hysteretic behavior was elucidated by an unsteady 3-dimensional Navier-Stokes numerical simulation. It was found that the hysteretic behavior was associated with a streamwise vortical flow appearing near the blade suction surface. In the accelerating process of axial flow velocity, the vortex is intensified to enlarge the flow separation area on the blade suction surface. In the decelerating flow process, the flow separation area is reduced because of the weakened vortex. Therefore, the aerodynamic performance in the accelerating flow process is lower than in the decelerating flow process, unlike the dynamic stall. Based on the vortex theorem, the mechanism to vary the intensity of the vortex can be explained by the trailing vortices associated with change in the blade circulation.

Topics: Wells , Turbines
Commentary by Dr. Valentin Fuster
2002;():981-985. doi:10.1115/FEDSM2002-31208.

Retaining rings in a generator are used to constrain the centrifugal force of the field winding end turns. A conventional retaining ring may cause a high pressure drop at the stator-rotor gap entrance and thus result in a large windage loss. The present work relates to the design of a retaining ring with a spline profile near its inboard end to thereby improve the axial cooling flow fluidity and enhance generator cooling efficiency. By altering the outer profile of the retaining ring, it has been found that approximately 30–60% of the pressure drop across the stator-rotor gap entrance can be reduced. As a result, the overall generator efficiency may increase about 0.005–0.01%.

Topics: Splines , Design , Generators
Commentary by Dr. Valentin Fuster
2002;():987-992. doi:10.1115/FEDSM2002-31209.

Numerical predictions of fan noise have not been studied extensively. This is due to the scattering effect of the fan casing, duct and the difficulty in obtaining aerodynamic acoustic source. New method to predict the fan noise and performance is developed and used to calculate various fan noise problems. A vortex method is used to model the fan and to calculate the flow field. Acoustic pressures are obtained from the unsteady force fluctuations of the blades using an acoustic analogy. But the acoustic analogy can be applied only in the free field in general. In order to consider the solid boundary effects of the casing, the newly developed Kirchhoff-Helmholtz BEM (Boundary Element Method) is introduced. With the above-mentioned method, the flow field and sound field of centrifugal and axial fan were calculated. Reasonable results are obtained not only for the peak frequencies but also for the amplitudes of the tonal sound. Also, in the predicted sound field, we can see the scattering effect of duct and casing.

Topics: Noise (Sound)
Commentary by Dr. Valentin Fuster
2002;():993-1000. doi:10.1115/FEDSM2002-31210.

In this work, a numerical study about the aerodynamic tonal noise generation in an industrial centrifugal fan with backward curved blades has been carried out. A three-dimensional numerical simulation of the complete unsteady flow on the whole impeller-volute configuration has been performed. Special attention has been focused on the impeller-volute interaction phenomena, analysing the influence of the distance between the impeller and the volute tongue. The numerical results have been contrasted using previous experimental investigations carried out in the same machine.

Topics: Noise (Sound)
Commentary by Dr. Valentin Fuster

Fluid Power

2002;():1005-1007. doi:10.1115/FEDSM2002-31244.

It is commonly perceived that turbulent flows yield turbulent wall fluxes, while laminar flows yield correspondingly laminar wall fluxes. Experiments support a recent theory that turbulent flows can yield laminar wall fluxes if the flow is “persistent.” Adding strong, stationary vortices to a turbulent boundary layer lowers the wall heat flux to a laminar value.

Topics: Vortices , Heat flux
Commentary by Dr. Valentin Fuster
2002;():1009-1014. doi:10.1115/FEDSM2002-31245.

The World Solar Challenge is a race from Darwin to Adelaide which attracts purpose-built vehicles from around the world. Using only power from the sun, the vehicles reach speeds of over 100 km/hr and the current record holder averaged a speed of over 90 km/hr. In this paper the background to the race and some of the technology used is described. Since aerodynamic drag is the major resistance to motion, this is examined in detail, including the testing and design principles applied to the Australian “Aurora” vehicle which won the race in 1999 and came second in 2001. A highly efficient electric motor, developed specifically for the Aurora is also described.

Commentary by Dr. Valentin Fuster
2002;():1015-1021. doi:10.1115/FEDSM2002-31247.

Wind towers are architectural designs employed for natural ventilation and passive cooling of buildings. In this study, it is shown that the performance of these towers can be improved appreciably by incorporating evaporative cooling in them. Two designs, called wetted columns and wetted surfaces, were employed, and their performances were evaluated and compared with those of the conventional towers. It is found that both designs can deliver air to the building they serve at higher flow rates and at temperatures very near the ambient air wet-bulb temperatures. In general, the wind tower with wetted columns performs better in areas with relatively high wind speeds, whereas the designs with wetted surfaces performs better in areas with no winds, or with very low wind speeds.

Commentary by Dr. Valentin Fuster
2002;():1023-1029. doi:10.1115/FEDSM2002-31249.

A remote area power supply using cold thermal storage and wind as the energy source is proposed. The primary objective is to provide a renewable energy remote area power supply with cheaper and more robust storage than lead-acid batteries. The proposal amalgamates a vapour compression refrigeration system with a Rankine cycle engine, both using the same working fluid. A tank of freezing brine acts as the condenser in the Rankine cycle and as the evaporator in the refrigeration cycle but also provides the “energy storage”. Analysis of the system indicates that it is practical and that its performance is comparable with existing battery based systems.

Commentary by Dr. Valentin Fuster
2002;():1031-1041. doi:10.1115/FEDSM2002-31250.

An experimental investigation was conducted to determine the effect of the vaneless diffuser width on the unsteady flow performance of a centrifugal compressor stage. Two compressor configurations with different vaneless diffuser width were investigated at four different impeller speeds and compared in the frequency and time domain. Only one diffuser rotating stall but different types of impeller rotating stalls were detected. The experiments show that the diffuser has a strong influence on the flow in the impeller including in areas way upstream. Analysis of the results indicated: • With increasing diffuser width the onset of impeller rotating stall was shifted to lower flow rates. • With increasing diffuser width the frequencies of the rotating stalls decreased. • There is a common tendency in most of the experiments to lower numbers of rotating cells with increasing relative speed. The impeller rotating stalls can be subdivided in a slow pattern with a relative speed to the impeller of 0.21 to 0.29 and a fast pattern with a relative speed of 0.50 to 0.56. This occurrence of two rotating pressure patterns confirms analytical results presented in previous investigations.

Commentary by Dr. Valentin Fuster
2002;():1043-1046. doi:10.1115/FEDSM2002-31251.

Turbulent boundary layers on turbomachine blades are three-dimensional and highly unsteady. The flow becomes periodically unsteady due to the relative motion of the rotor and stator blades in a stage. A large velocity defect within the wake of an upstream blade generates a varying incidence angle, and a fluctuating velocity at entrance to the next stage. The velocity defect, or wake, travels downstream with a finite speed and intermittently perturbs the boundary layer on downstream blades.

Topics: Wakes , Turbomachinery
Commentary by Dr. Valentin Fuster
2002;():1047-1056. doi:10.1115/FEDSM2002-31252.

This is Part I of a two-part paper describing a new compressor volute design system, which is applied to an interactive program that completely describes volute geometry and its interactions with CFD. Part I goes into the theory behind the program and the design of the volute geometry by application of Bezier polynomials while Part II fully describes the actual design system approach and its application to the program. Based on the developed mathematics of B-spline and Bezier polynomials, the present paper further develops the extent of Bezier polynomials by application of Bezier to volute design. Bezier polynomials will be used to describe and define volute geometry for centrifugal compressors with the objective of serving as an aid for volute design and performance analysis. The current research aims to develop an algorithm for volute design that will enable the volute geometry, to be fully described and viewed, in two-dimensional (2D) and three-dimensional (3D) modes. A program based on this algorithm will deliver output data for grid generation and CFD flow simulation and eventual manufacture of the volute. These procedures will be developed for a family of centrifugal compressors.

Topics: Compressors , Design
Commentary by Dr. Valentin Fuster
2002;():1057-1063. doi:10.1115/FEDSM2002-31253.

Volutes are widely used in industrial process, refrigeration system, small gas turbines and gas pipeline centrifugal compressors as the transition from the impeller-diffuser to the pipings, because of their simple structure, ease of production and wide operating range. This paper illustrates a new design tool that incorporates a new volute design system that integrates and automates geometry generation, grid generation and aerodynamic analysis. In optimizing the available technology in terms of grid generation, CFD, and computer graphics, the program will utilize existing technology used by industry to generate a powerful volute design tool. The design tool is programmed in a way that integrates the features and methods a designer would use for volute design. This is fundamentally by means of geometrical constraints and/or functional relationships. Grids can be generated in minutes accommodating geometrical changes thus reducing the bottlenecks associated with geometry/grid generation for CFD applications. Prior to most CFD analysis work, a structured grid must be produced ensuring high quality such that convergence is assured and the time to convergence of the solution is minimized. However, there are usually only a few people that have the required skills to produce the geometry and generate a high quality structured grid. In essence, the tool provides a sidestep around both the geometry generation and the grid generation process. It automates the process such that anybody can produce a high quality grid from the geometry and move straight to the CFD component of the work and hence can incorporate CFD as part of the design process. The volute design tool will enable the user to generate a family of volutes and display 2D volute cross sections and 3D solid models of the scroll, diffuser inlet, discharge conic, and connecting channel. Separate interfaces will be written to accommodate the different operating systems. The geometry generation will be written in windows however, a separate interface will be written to produce the grid being compatible in NT, Unix, and Linux platforms.

Topics: Compressors , Design
Commentary by Dr. Valentin Fuster
2002;():1065-1074. doi:10.1115/FEDSM2002-31254.

The Engineering Business Ltd. (EBL) have developed a patented system for generating electricity from oscillating hydroplanes driven by flowing water. The concept started development with the award of a Smart Award from the UK governments Department of Trade and Industry (DTI) in 1998, which led to the development of the Active Water Column Generator (AWCG). The AWCG was a simple, sea / air interface device which demonstrated, through the actuation of hydroplanes in moving water, that power could be generated. However, this concept suffered from inherent size and vulnerability limitations. It was decided to alter the basic form such that it is now seabed mounted, and therefore protected from the worst storm and wave effects. This concept, known as Stingray, is being developed under the New and Renewable Energy Research and Development program run by the DTI. Following the successful conclusion of the first phase of work, a fast-track 15 week feasibility study undertaken in the last quarter of 2001, phase two of the program is well under way. This involves a full design study, for 2002/3, to design, install, operate, decommission and evaluate a 150kW demonstrator — the first of its kind internationally. This paper describes the progression from concept to demonstrator design, focussing on the modelling work undertaken in the initial investigation and plans for its subsequent validation. It outlines mathematical and physical modelling techniques used and then gives a brief summary of the findings to date.

Topics: Generators , Tides
Commentary by Dr. Valentin Fuster
2002;():1075-1081. doi:10.1115/FEDSM2002-31255.

This paper examines the impact on the power generation capacity of a wind turbine as a result of the modification of the shape of the blades of an existing wind turbine. The modification involves curving the blades in the direction of rotation resulting in an increase in generated lift and therefore an increase in the power output of the wind turbine. Two three-bladed models were tested in a wind tunnel, one original straight-bladed model and one modified model both of which were 0.84 m in diameter. A study of the methods of flow visualization for a wind turbine in a wind tunnel was investigated. The corresponding results are presented. It was discovered that the china clay method of flow visualization in conjunction with a strobe light gave a good indication of the direction of the airflow over the turbine blades as did condensed oil droplets from a smoke wand which presented a very clear indication of the span-wise flow. It was concluded from the investigation that curving the blade into the direction of rotation on a wind turbine produced a greater power output at the same wind speed as an unmodified wind turbine.

Commentary by Dr. Valentin Fuster
2002;():1083-1090. doi:10.1115/FEDSM2002-31256.

During development of the updraft free-exit-flow hydropower turbine system, serious attention was paid to impact of the traditional turbine system to environment. The paper presents results of experimental research and three-dimensional turbulent flow simulation of the runner. The experiment demonstrated that, on one hand, the system can enhance exit flow aeration and downstream water quality; on the other hand, it gives good results in terms of its power production efficiency, about 85%. Based on the time-averaged Navier-Stokes equations and standard k-ε model, the SIMPLEC algorithm was applied for the numerical analysis. To guarantee credibility of the calculation, FLUENT5.5 code was used, which can provide distribution of pressure and velocity respectively. With comparison between the experimental data and calculating results, it can be concluded that flow simulation can be an effective tool for predicting performance and optimizing design of new turbine runner.

Commentary by Dr. Valentin Fuster
2002;():1091-1093. doi:10.1115/FEDSM2002-31257.

A short synopsis pertaining to the concept, analysis and development of the new advanced zero head HYdro-Propulsive System (HYPS) has been drawn. Within this system, of particular importance are the implemented revolutionary feedback control systems. These systems provide the HYPS with a unique ability to efficiently receive and capture the kinetic energy embedded in multi-directional, random and turbulent fluid flow motions. The system is equipped with specially designed bladings (propulsors) that are attached through the feedback controllers to the extending radial arms, allowing the blades to adjust their orientations according to the oncoming flow. It has been demonstrated, through implicit analysis and extensive tests, that this system can effectively maximise the favourable aero-hydrodynamic forces on its impellers and, in turn, achieve a much higher energy conversion rate compare to those of the conventional devices.

Topics: Feedback
Commentary by Dr. Valentin Fuster

Fluid Flow in Micro Systems, Measurements, Analysis and Application

2002;():1095-1100. doi:10.1115/FEDSM2002-31299.

This paper aims to apply experiment and computer simulation to investigate heat transfer characteristics of different fluids through microchannels with different geometries. An experimental apparatus with multi-functions, low uncertainties, high precision and temperature control was constructed. The experimental results are compared with the results of computer simulations and theoretical predictions from the conventional laminar flow theory. The flow rate, temperature and pressure drop were measured at steady states to analyse flow characteristics. It was concluded that the effect of allowance for machining of microchannels on heat transfer characteristics could be neglected under lower flow rate. The fluid mass flow rate, the inlet temperature of fluid, and the geometric parameters, particularly the hydraulic diameter and aspect ratio H/Wm in rectangular microchannels were found to have a significant influence on fluid flow and heat transfer characteristics.

Commentary by Dr. Valentin Fuster
2002;():1101-1107. doi:10.1115/FEDSM2002-31300.

Results from the application of the spatial filtering velocimetry on capillary flow are presented. The type of the spatial filter is a differentail fibreoptical filter made by 16 step index fibres. Using tracer particles, the flow velocities along the capillary diameter and the volume flow were measured. The capillary is a transparent glass tube with inner diameter ranging from 0.15 mm to 0.5 mm. Using a gravitational measuring system the uncertainty of the flow measurement by the optical system was determined. For the laminar flow of newtonian fluids, the volume flow determination is based on the equation of Hagen-Poiseuille. The range of volume flow was 0.5 ml/h up to 50 ml/h.

Commentary by Dr. Valentin Fuster
2002;():1109-1114. doi:10.1115/FEDSM2002-31301.

In microscopic particle image velocimetry (microPIV) experiments, the entire volume of a flowfield is illuminated, resulting in all of the particles in the field of view contributing to the image. Unlike in light-sheet PIV, where the depth of the measurement volume is simply the thickness of the laser sheet, in microPIV, the measurement volume depth is a function of the image forming optics of the microscope. In a flowfield with out-of-plane motion, the measurement volume (called the depth of correlation) is also a function of the magnitude of the out of plane motion within the measurement volume. Equations are presented describing the depth of correlation and its dependence on out-of-plane motion. The consequence of this dependence and suggestions for limiting its significance are also presented. Another result of the out-of-plane motion is that the height of the PIV signal peak in the correlation plane will decrease. Because the height of the noise peaks will not be affected by the out-of-plane motion, this could lead to erroneous velocity measurements. An equation is introduced that describes the effect of the out-of-plane motion on the signal peak height, and its implications are discussed.

Commentary by Dr. Valentin Fuster
2002;():1115-1123. doi:10.1115/FEDSM2002-31302.

The oscillatory free-surface displacement in an orifice periodically driven at the inlet is studied theoretically and experimentally. A potential flow analysis is applied using two different sidewall boundary conditions: no flow through the wall, and no slip at the wall. Both analyses result in surface shapes defined by Bessel functions, but with significantly different frequencies associated with the discrete modal shapes. Experimental results for two orifice diameters, 1180 µm and 794 µm, verify the general shape of the surface displacement but show a significant downward shift of the measured modal frequencies from the predicted values. This shift is shown to be a function of mode number and is thought to be a consequence of orifice edge effects.

Topics: Displacement
Commentary by Dr. Valentin Fuster

The Measurement and Modeling of Large-Scale Turbulent Structures

2002;():1125-1133. doi:10.1115/FEDSM2002-31408.

Recent applications of the ‘slice POD’ to the axisymmetric turbulent wake and jet are reviewed, and the results used to provide a critique of commonly held views about these flows. It is argued that the so-called ‘coherent structures’ are simply artifacts of the source conditions, and have little to do with the far down-stream development of these flows. Also, experimental evidence is presented for the possible presence of Townsend’s large eddies, eddies whose primary role is to warp the mean motion. Finally, classical linear stability analysis which ‘predicted’ that only azimuthal mode-1 could be unstable for these flows is shown to be deficient, with the result that at very least modes-0, 1 and 2 could be important, consistent with the experiments.

Commentary by Dr. Valentin Fuster
2002;():1135-1142. doi:10.1115/FEDSM2002-31409.

The effects of various combinations of vortex-generating tabs on the turbulence structures in the near-field of a round jet are investigated using LES simulation and flow visualization techniques. The visualization methods include stream-wise and non-streamwise vorticity, and a variety of methods that use the invariant of the velocity gradient tensor (the discriminant, Q value and the second eigenvalue condition). Integration of the LES data sets suggest that the structural changes as a result of introducing tabs is significant; however, the methods used to deduce these changes are not always consistent with one another. In some cases, one scheme will produce large amounts of background “noise” while others are less prone to this effect. It is concluded that qualitatively, the four tab case produces the greatest amount of small scale structure.

Commentary by Dr. Valentin Fuster
2002;():1143-1150. doi:10.1115/FEDSM2002-31410.

Multipoint instantaneous measurements of the streamwise velocity component were obtained in the far field region of an axisymmetric turbulent jet from 20 to 69 diameters downstream for jet exit Reynolds numbers ranging from 40,000 to 84,700. The Proper Orthogonal Decomposition (POD) was then applied to a double Fourier transform in time and azimuthal direction of the two-point velocity correlation tensor. The first eigenspectrum, which contains more than 62% of the resolved streamwise energy, has two peaks: one at azimuthal mode-2 at near zero frequency, and another at mode-1 at a local Strouhal number (fx/Uc ) of approximately 1. When checked against two-point statistics similarity analysis from Ewing (1995) extended to the POD, the results compare favorably and also give some hints into the resolution necessary to cover the field.

Topics: Turbulence
Commentary by Dr. Valentin Fuster
2002;():1151-1157. doi:10.1115/FEDSM2002-31411.

This paper presents the findings of three experiments using multi-point hot-wire arrays in the high Reynolds number axisymmetric turbulent wake behind a disk. The purpose of the multiple experiments was to validate earlier and less extensive experiments. The ‘slice POD’ was applied to all sets to examine the effects of array coverage and the disk support system. The Reynolds number based on the free stream velocity and disk diameter was kept constant at 28,000. The investigated region spanned from 10 to 60 disk diameters downstream. These results confirm the earlier findings. In particular, the eigenvalues integrated over frequency show a azimuthal mode-1 dominance at x/D = 10 which evolves to a mode-2 dominance by x/D = 50. For all downstream positions, two distinct peaks were found in the first eigenspectrum: one for azimuthal mode-2 at near zero frequency, and another for azimuthal mode-1 at a Strouhal number (fd/U∞ ) of 0.126. Both peaks decrease in magnitude as the flow evolves downstream, but the peak at the Strouhal number 0.126 decrease more rapidly then the one at near-zero frequency, leaving the latter to eventually dominate.

Topics: Wakes , Disks
Commentary by Dr. Valentin Fuster
2002;():1159-1171. doi:10.1115/FEDSM2002-31412.

Physics based low dimensional approaches are playing an increasingly important role in our understanding of turbulent flows. They provide an avenue for us to understand the connection between coherent structures and the overall dynamics of the flow field. As such these approaches are fundamental to the implementation of physics based active control methodologies. In this paper we review applications of various low dimensional approaches (including Proper Orthogonal Decomposition (POD), Linear Stochastic Estimation (LSE), Conditional Averages and Wavelets) to turbulent shear layers and connect the results to simulation tools. The applications of all these methods to the 2D shear layer suggest a kind of universal behavior of both the large scale structure extracted and the background turbulence, irrespective of the technique (filtering method) used. A review of the application of POD and LSE to the axisymmetric jet at Reynolds numbers between 100,000 and 800,000 and Mach numbers ranging from very low to 0.6 suggest a universal behavior where the dynamics can be described with relatively low dimensional information (1 POD mode and 5 or 6 Fourier azimuthal modes) over the Reynolds/Mach number range studied. These results provide physical justification for simulation tools such as VLES, LES and SDM since such computational methods involve different levels of low-dimensional modeling.

Commentary by Dr. Valentin Fuster
2002;():1173-1178. doi:10.1115/FEDSM2002-31413.

Both time-dependent and time-invariant Proper Orthogonal Decompositions are performed on LES and PIV data as an initial step in a study of tumble breakdown in in-cylinder flows. Evidence of tumble instability during compression is found in the time-dependent POD of both data sets. Time-invariant POD modes are presented which will be used later in low-dimensional models of these systems.

Commentary by Dr. Valentin Fuster
2002;():1179-1186. doi:10.1115/FEDSM2002-31414.

The development of the large-scale structures in the intermediate region of the three-dimensional wall jet was examined using measurements of the mean streamwise vorticity and the two-point, two-time correlations of the streamwise fluctuating velocity in the vertical and lateral directions. It was found that a dominant large-scale double horse-shoe structure persisted to 40 diameters downstream of the jet exit. It was also found that the structure continued to evolve throughout the intermediate field. In particular, the inner horse-shoe vortex was induced toward the wall, while the position of the outer horseshoe vortex relative to the half-width moved toward the centerline. The inclination of this structure relative to the wall also increased as the flow evolved downstream. These changes in the structure combined to cause the lateral spread rate of the jet to aprroximately double in the intermediate field.

Commentary by Dr. Valentin Fuster