ALBERT

Description: This paper presents new trends in Particle Image Velocimetry and practical aspects relevant to the application of the technique to large scale wind tunnel testing. The various problems and their solutions to the operation of PIV in large scale wind tunnels are discussed. Application of the technique in mapping complex flows are also presented.

Description: A 3D CFD code, SCISEAL, has been developed and validated. Its capabilities include cylindrical seals, and it is employed on labyrinth seals, rim seals, and disc cavities. State-of-the-art numerical methods include colocated grids, high-order differencing, and turbulence models which account for wall roughness. SCISEAL computes efficient solutions for complicated flow geometries and seal-specific capabilities (rotor loads, torques, etc.).

Description: Pressure and temperature sensitive paints have been utilized for the measurement of blade surface pressure and temperature distributions in a high speed axial compressor and an Allied Signal F109 gas turbine engine. Alternate blades were painted with temperature sensitive paints and then pressure sensitive paint. This combination allows temperature distributions to be accounted for when determining the blade suction surface pressure distribution. Measurements were taken and pressure maps on the suction surface of a blade were obtained over a range of rotational speeds. Pressure maps of the suction surface show-strong shock waves at the higher speeds.

Description: The oil film interferometry skin-friction technique is described and applied to flows in some of the NASA Ames large wind tunnel facilities. Various schemes for applying the technique are discussed. Results are shown for tests in several wind tunnels which illustrate the oil film's ability to measure a variety of flow features such as shock waves separation, and 3D flow.

Description: The flow in a hydrostatic pocket is described by a mathematical model that uses the three dimensional Navier-Stokes equations written in terms of the primary variables, u, v, w, and p. Using a conservative formulation, a finite volume multi-block method is applied through a collocated, body fitted grid. The flow is simulated in a shallow pocket with a depth/length ratio of 0.02. The flow structures obtained and described by the authors in their previous two dimensional models are made visible in their three dimensional aspect for the Couette flow. It has been found that the flow regimes formed central and secondary vortical cells with three dimensional corkscrew-like structures that lead the fluid on an outward bound path in the axial direction of the pocket. The position of the central vortical cell center is at the exit region of the capillary restrictor feedline. It has also been determined that a fluid turn around zone occupies all the upstream space between the floor of the pocket and the runner, thus preventing any flow exit through the upstream port. The corresponding pressure distribution under the shaft presented as well. It was clearly established that for the Couette dominated case the pressure varies significantly in the pocket in the circumferential direction, while its variation is less pronounced axially.

Description: An experimental program was conducted at NASA Langley Research Center that included development and evaluation of an operational facility for wall drag measurement of potential scramjet fuel injection or wall cooling configurations. The facility consisted of a supersonic tunnel, with one wall composed of a series of interchangeable aluminum plates attached to an air bearing suspension system. The system was equipped with load cells that measured drag forces of 115 psia (793 kPa). This flow field contained a train of weak, unsteady, reflecting shock waves that were produced in the Mach 2 nozzle flows, the effect of reflecting shocks (which are to be expected in scramjet combustors) in internal flows has not previously been documented.

Description: The wall drag test tunnel at NASA Langley Research Center was used to evaluate simulated scramjet fuel injection into a wall cavity. In this tunnel, one wall consists of interchangeable aluminum plates attached to an air bearing suspension system. The plates were equipped with load cells to measure drag forces and static taps to determine pressure distributions. The plates were exposed to a Mach 2 air stream at a total pressure of 115 psia (793 kPa). This flow field contained a train of weak unsteady, reflecting shock waves that were produced in the nozzle assembly located upstream of the test section.

Description: Minuscule amounts (e.g., 1 part in 10,000) of a surface-active material in a liquid can drastically affect the surface behavior of the liquid, influencing how the material flows and mixes with other liquid and solid materials. In many respects, the science of surfactants has been empirical, with trial and error dominating over the ability to predict how surfactant type and concentration influence surface behavior. A program for the modeling of surfactant behavior has been established at Yale. This program combines experimental work performed both on the ground and in space, and theoretical and numerical modeling. By levitating a drop of liquid in air, away from solid container surfaces, and by manipulating the drop with acoustic radiation forces, we have been able to establish idealized conditions for surface behavior studies. The primary experiments involve the study of the free oscillations of initially deformed drops. In STS-73, the USML-2 mission of the Space Shuttle, we performed the following measurements: 1) the oscillation of a spherical drop in its quadrupole mode; 2) the oscillation of a drop about a deformed (oblate) shape; 3) the slow static squeezing of the drop from spherical to nearly flat; and 4) the superoscillations of drops when the radiation forces maintaining the drop in a flattened state are suddenly reduced. Analytic and numerical studies have enabled us to understand the physics of these oscillations and to extract material properties such as the dynamic surface tension and the surface viscosities (shear and dilatational). The relation to ground-based studies is essential, because the knowledge and understanding gleaned from our space studies enable us to interpret ground-based data.

Description: The flow through a plane asymmetric diffuser is a good test case for assessing the capability of LES since it contains features such as large scale unsteady separation and strong intermittency which are difficult to capture using conventional modeling approaches. Previous attempts to simulate this flow (Kaltenbach, 1994) have significantly underpredicted the extent of separation. The objective of the present research is to understand why the previous simulations did not predict the flow separation correctly. This study focuses on mesh refinement and matching of the inlet velocity profile. In order to perform this study, the flow solver of Kaltenbach (1994) was modified to increase its accuracy and efficiency. The improved algorithm allows for better resolution at affordable CPU cost.

Description: The objective of the present research is to assess the usefulness of large-eddy simulation (LES) methodology for flows in complex geometries. Flow past a circular cylinder has been calculated using a central-difference based solver, and the results have been compared to those obtained by a solver that employs higher-order upwind biased schemes (Beaudan & Moin, 1994). This comparison allows us to assess the suitability of these schemes for LES in complex geometry flows.

Description: Accurate SGS models must be capable of correctly representing the energy transfer between GS and SGS. Recent direct assessment of the energy transfer carried out using direct numerical simulation (DNS) data for wall-bounded flows revealed that the energy exchange is not unidirectional. Although GS kinetic energy is transferred to the SGS (forward scatter (F-scatter) on average, SGS energy is also transferred to the GS. The latter energy exchange (backward scatter (B-scatter) is very significant, i.e., the local energy exchange can be backward nearly as often as forward and the local rate of B-scatter is considerably higher than the net rate of energy dissipation.

Description: An efficient scheme for the direct numerical simulation of 3D transitional and developed turbulent flow is presented. Explicit and implicit time integration schemes for the compressible Navier-Stokes equations are compared. The nonlinear system resulting from the implicit time discretization is solved with an iterative method and accelerated by the application of a multigrid technique. Since we use central spatial discretizations and no artificial dissipation is added to the equations, the smoothing method is less effective than in the more traditional use of multigrid in steady-state calculations. Therefore, a special prolongation method is needed in order to obtain an effective multigrid method. This simulation scheme was studied in detail for compressible flow over a flat plate. In the laminar regime and in the first stages of turbulent flow the implicit method provides a speed-up of a factor 2 relative to the explicit method on a relatively coarse grid. At increased resolution this speed-up is enhanced correspondingly.

Description: The role of unsteady flow processes in establishing the performance of axial flow turbomachinery was investigated. The development of the flow model, as defined by the time average flow equations associated with the passage of the blade row embedded in a multistage configuration, is presented. The mechanisms for unsteady momentum and energy transport is outlined. The modeling of the unsteady momentum and energy transport are discussed. The procedure for simulating unsteady multistage turbomachinery flows is described.

Description: A mathematical model for closing or mathematically completing the system of equations is proposed. The model describes the time average flow field through the blade passages of multistage turbomachinery. These average-passage equation systems govern a conceptual model useful in turbomachinery aerodynamic design and analysis. The closure model was developed to insure a consistency between these equations and the axisymmetric through-flow equations. The closure model was incorporated into a calculation code for use in the simulation of the flow field about a high-speed counter rotating propeller and a high-speed fan stage.

Description: While turbulence models are being developed which show promise for simulating the transition region on a turbine blade or vane, it is believed that the best approach with the greatest potential for practical use is the use of models which incorporate the physics of turbulent spots present in the transition region. This type of modeling results in the prediction of transition region intermittency which when incorporated in turbulence models give a good to excellent prediction of the transition region heat transfer. Some models are presented which show how turbulent spot characteristics and behavior can be employed to predict the effect of pressure gradient and Mach number on the transition region. The models predict the spot formation rate which is needed, in addition to the transition onset location, in the Narasimha concentrated breakdown intermittency equation. A simplified approach is taken for modeling turbulent spot growth and interaction in the transition region which utilizes the turbulent spot variables governing transition length and spot generation rate. The models are expressed in terms of spot spreading angle, dimensionless spot velocity, dimensionless spot area, disturbance frequency and Mach number. The models are used in conjunction with a computer code to predict the effects of pressure gradient and Mach number on the transition region and compared with VKI experimental turbine data.

Description: The objective of this study is to provide careful qualification and detailed measurements in a re-creation of the Obi experiment. The work will include extensive documentation of the flow two-dimensionality and detailed measurements required for testing of flow computations. Also important to this study is the close interaction of the experimental and computational groups to improve the utility of the data obtained and the accuracy of computation.

Description: The control of vortical flows is gaining significance in the design of aeronautical and marine structures. While passive devices have been used effectively in the past, active control strategies have the potential of allowing a leap in the performance of future configurations. The efficiency of control schemes is strongly dependent on the development of accurate flow models that can be devised using information that is available not only from numerical solutions of the governing Navier-Stokes equations but also can be measured experimentally. In that context it is desirable to construct adaptive control schemes using information that can be measured at the wall.

Description: Stability of the elliptic vortex attracted interest in the past decade. Cambon (1982), and Cambon, Teissedre and Jeandel (1985) have studied the stability of such flows with spatially uniform velocity gradient, and have provided RDT solutions for a wide range of the parameter S/Omega (where the strain rate S and the vorticity 2 Omega define the velocity gradient matrix). The range studied included those of hyperbolic streamlines (strain dominated, S/Omega is greater than 1), linear streamlines (simple shear, S/Omega = 1), and elliptical streamlines (vorticity dominated, S/Omega is less than 1). The latter class has more recently attracted interest and several studies appeared (Pierrehumbert 1986, Bayly 1986, Craik and Criminale 1986). These studies will be collectively referred to as PBCC.

Description: The objective is the development of a new theory which enables the algorithmic computation of all self-similar mean velocity profiles. The theory is based on Liegroup analysis and unifies a large set of self-similar solutions for the mean velocity of stationary parallel turbulent shear flows. The results include the logarithmic law of the wall, an algebraic law, the viscous sublayer, the linear region in the middle of a Couette flow and in the middle of a rotating channel flow, and a new exponential mean velocity profile not previously reported. Experimental results taken in the outer parts of a high Reynolds number flat-plate boundary layer, strongly support the exponential profile. From experimental as well as from DNS data of a turbulent channel flow the algebraic scaling law could be confirmed in both the center region and in the near wall region. In the case of the logarithmic law of the wall, the scaling with the wall distance arises as a result of the analysis and has not been assumed in the derivation. The crucial part of the derivation of all the different mean velocity profiles is to consider the invariance of the equation for the velocity fluctuations at the same time as the invariance of the equation for the velocity product equations. The latter is the dyad product of the velocity fluctuations with the equation for the velocity fluctuations. It has been proven that all the invariant solutions are also consistent with similarity of all velocity moment equations up to any arbitrary order.

Description: Nonpremixed combustion is the process whereby fuel and oxidizer species, which are each nonflammable in isolation, concurrently mix to burn a flammable mixture, and chemically react in the flammable mixture. In cases of practical industrial interest, the bulk of nonpremixed combustion occurs in a turbulent mixing regime where enhanced mass transfer rates flow the maximum power density to be achieved in any given thermochemical device.

Description: Direct numerical simulation of turbulent reacting flows places extreme demands on computational resources. At the present time, simulations can be performed only for greatly simplified reaction systems and for very low Reynolds numbers. Direct simulation of more realistic cases occurring at higher Reynolds number and including multiple species and numerous chemical reactions will exceed available computational resources far into the future. Because of this, there is a clear need to develop the technique of large eddy simulation for reacting flows. Unfortunately this task is complicated by the fact that combustion arises from chemical reactions that occur at the smallest scales of the flow. Capturing the large-scale behavior without resolving the small-scale details is extremely difficult in combustion problems. Thus LES modeling for turbulent combustion encounters difficulties not present in modeling momentum transport, in which the main effect of the small scales is to provide dissipation. The difficulty is more pronounced in premixed combustion, where detailed chemistry plays an essential role in determining the flame speed (or overall burning rate); in nonpremixed combustion infinite rate chemistry can be assumed, eliminating the small scale features to a first approximation.

Description: Many industrial applications in such fields as aeronautical, mechanical, thermal, and environmental engineering involve complex turbulent flows containing global separations and subsequent reattachment zones. Accurate prediction of this phenomena is very important because separations influence the whole fluid flow and may have an even bigger impact on surface heat transfer. In particular, reattaching flows are known to be responsible for large local variations of the local wall heat transfer coefficient as well as modifying the overall heat transfer. For incompressible, non-buoyant situations, the fluid mechanics have to be accurately predicted in order to have a good resolution of the temperature field.

Description: In simple flows, where the mean deformation rates are mild and the turbulence has time to come to equilibrium with the mean flow, the Reynolds stresses are determined by the applied strain rate. Hence in these flows, it is often adequate to use an eddy-viscosity representation. The modern family of kappa-epsilon models has been very useful in predicting near equilibrium turbulent flows, where the rms deformation rate S is small compared to the reciprocal time scale of the turbulence (epsilon/kappa). In modern engineering applications, turbulence models are quite often required to predict flows with very rapid deformations (large S kappa/epsilon). In these flows, the structure takes some time to respond and eddy viscosity models are inadequate. The response of turbulence to rapid deformations is given by rapid distortion theory (RDT). Under RDT the nonlinear effects due to turbulence-turbulence interactions are neglected in the governing equations, but even when linearized in this fashion, the governing equations are unclosed at the one-point level due to the non-locality of the pressure fluctuations.

Description: Direct numerical simulation (DNS) was used to study homogeneous turbulence with a mean flow having elliptic streamlines. This flow combines the effects of rotation and strain on the turbulence. There are many important reasons for studying the elliptic streamline flow. This flow contains the effects of both rotation and strain and is therefore similar to the mean flow in a vortex strained in the plane perpendicular to its axis. Such flows provide insight into the fundamental vortical interactions within turbulence. A strained vortex also occurs in airplane wakes, in which each wingtip vortex induces a strain field on the other. The strain field can affect the stability of these vortices and thereby their turbulent structure downstream. The ability to understand and predict the turbulent structure of the vortices is important to the wake hazard problem which is of major concern for the safety of commercial aircraft.

Description: A combined experimental and computational effort to test the performance of turbulence models in prediction of three-dimensional separation is under way. The current phase pf the project is the acquisition of flow field data using laser Doppler velocimetry. Accomplishments for summer 1996 include development of the instrument and signal processing system and acquisition of mean velocity profiles.

Description: The mean and phase averaged pressure and wall shear stress distributions were measured on the stator wall of a 50% eccentric annular seal which was whirling in a circular orbit at the same speed as the shaft rotation. The shear stresses were measured using flush mounted hot-film probes. Four different operating conditions were considered consisting of Reynolds numbers of 12,000 and 24,000 and Taylor numbers of 3,300 and 6,600. At each of the operating conditions the axial distribution (from Z/L = -0.2 to 1.2) of the mean pressure, shear stress magnitude, and shear stress direction on the stator wall were measured. Also measured were the phase averaged pressure and shear stress. These data were combined to calculate the force distributions along the seal length. Integration of the force distributions result in the net forces and moments generated by the pressure and shear stresses. The flow field inside the seal operating at a Reynolds number of 24,000 and a Taylor number of 6,600 has been measured using a 3-D laser Doppler anemometer system. Phase averaged wall pressure and wall shear stress are presented along with phase averaged mean velocity and turbulence kinetic energy distributions located 0.16c from the stator wall where c is the seal clearance. The relationships between the velocity, turbulence, wall pressure and wall shear stress are very complex and do not follow simple bulk flow predictions.

Description: Computational programs developed for the thermal analysis of tilting and flexure-pad hybrid bearings, and the unsteady flow and transient response of a point mass rotor supported on fluid film bearings are described. The motion of a cryogenic liquid on the thin film annular region of a fluid film bearing is described by a set of mass and momentum conservation, and energy transport equations for the turbulent bulk-flow velocities and pressure, and accompanied by thermophysical state equations for evaluation of the fluid material properties. Zeroth-order equations describe the fluid flow field for a journal static equilibrium position, while first-order (linear) equations govern the fluid flow for small amplitude-journal center translational motions. Solution to the zeroth-order flow field equations provides the bearing flow rate, load capacity, drag torque and temperature rise. Solution to the first-order equations determines the rotordynamic force coefficients due to journal radial motions.

Description: In recent years the techniques of computational fluid dynamics (CFD) have been used to compute flows associated with geometrically complex configurations. However, success in terms of accuracy and reliability has been limited to cases where the effects of turbulence and transition could be modeled in a straightforward manner. Even in simple flows, the accurate computation of skin friction and heat transfer using existing turbulence models has proved to be a difficult task, one that has required extensive fine-tuning of the turbulence models used. In more complex flows (for example, in turbomachinery flows in which vortices and wakes impinge on airfoil surfaces causing periodic transitions from laminar to turbulent flow) the development of a model that accounts for all scales of turbulence and predicts the onset of transition is an extremely difficult task. Fortunately, current trends in computing suggest that it may be possible to perform direct simulations of turbulence and transition at moderate Reynolds numbers in some complex cases in the near future. This presentation will focus on direct simulations of transition and turbulence using high-order accurate finite-difference methods. The advantage of the finite-difference approach over spectral methods is that complex geometries can be treated in a straightforward manner. Additionally, finite-difference techniques are the prevailing methods in existing application codes. An application of high-order-accurate finite-difference methods to direct simulations of transition and turbulence in a spatially evolving boundary layer subjected to high levels of freestream turbulence will be presented.

Description: Turbulent flows that exhibit inhomogeneities in the streamwise direction pose a particular challenge to numerical simulation approaches due to the need to prescribe time-dependent turbulent inflow conditions. In most cases the flow downstream is more or less 'driven' by the conditions at the inlet, making it necessary to specify realistic turbulent fluctuations that are in equilibrium with the assumed mean flow. This requirement often dictates that the inflow data should satisfy the Navier-Stokes equations, which in turn implies that an independent simulation be used to generate them. Detailed simulations for the purpose of creating inflow conditions can be costly and thus certain levels of approximation are desirable. In this paper we shall focus on an approximate yet accurate method for generating inflow conditions for spatially-developing boundary layer simulations. The proposed method is essentially a simplification of the method of Spalart and Leonard (1985), who devised an ingenious transformation that allows for the calculation of spatially evolving boundary layers in conjunction with periodic boundary conditions applied in the streamwise direction. While this method is elegant and highly accurate, it is more complicated than is necessary for the purpose of generating inflow data. A few key approximations are used in this work to arrive at a 'modified Spalart method' that is very easy to implement and efficient to use. The new method is shown to yield results that compare well with the computations of Spalart (1988). When used as a means of generating inflow data, the modified Spalart method is shown to be superior to existing approaches.

Description: Accurate large eddy simulation (LES) of a wall-bounded flow generally requires a near-wall resolution comparable to that in direct numerical simulation (DNS). As much as 50% of the total grid points and computational costs are expended in the near-wall regions in a typical simulation. This limits LES to fairly low Reynolds numbers on current computers. To perform practical flow applications at realistically high Reynolds numbers, such as flow over an airfoil, it is desirable to replace very thin, near-wall regions in the LES with easily and inexpensively computed wall models to specify the near-wall boundary conditions.

Description: Reducing the infinite degrees of freedom inherent in fluid motion into a manageable number of modes to analyze fluid motion is presented. The concepts behind the center manifold technique are used. Study of the Blasius boundary layer and a precise description of stability within the flow field are discussed.

Description: A key ingredient in the simulation of flow in porous media is the accurate determination of the velocities that drive the flow. The large scale irregularities of the geology, such as faults, fractures, and layers suggest the use of irregular grids in the simulation. Work has been done in applying the finite volume element (FVE) methodology as developed by McCormick in conjunction with mixed methods which were developed by Raviart and Thomas. The resulting mixed finite volume element discretization scheme has the potential to generate more accurate solutions than standard approaches. The focus of this paper is on a multilevel algorithm for solving the discrete mixed FVE equations. The algorithm uses a standard cell centered finite difference scheme as the 'coarse' level and the more accurate mixed FVE scheme as the 'fine' level. The algorithm appears to have potential as a fast solver for large size simulations of flow in porous media.

Description: The effect of the unsteady flow field in a multistage compressor on the time-averaged performance was assessed. The energy transport by the unsteady deterministic flow field was taken into account. The magnitude of the body force resulting from the aerodynamic loading on a blade row was compared to the gradient of the stress tensor associated with the unsteady time-resolved flow field generated by the blade row. The magnitude of the work performed by these forces was compared to the divergence of the energy correlations produced by the unsteady time-resolved flow field. The stress tensor and the energy correlations are non-negligible in the end wall regions. The results suggest that the turbulence is the primary source of flow mixing in the midspan region.

Description: During hypersonic flight, high temperatures and high heat fluxes are generated. The Flight Loads Laboratory (FLL) at Dryden Flight Research Center (DFRC) is equipped to calibrate high heat fluxes up to 1100 kW/sq m. There are numerous uncertainties associated with these heat flux calibrations, as the process is transient, there are expected to be interactions between transient conduction, natural and forced convection, radiation, and possibly an insignificant degree of oxidation of the graphite cavity. Better understanding, of these mechanisms during the calibration process, will provide more reliable heat transfer data during either ground testing or flight testing of hypersonic vehicles.

Description: The Geophysical Fluid Flow Cell (GFFC) experiment performed visualizations of thermal convection in a rotating differentially heated spherical shell of fluid. In these experiments dielectric polarization forces are used to generate a radially directed buoyancy force. This enables the laboratory simulation of a number of geophysically and astrophysically important situations in which sphericity and rotation both impose strong constraints on global scale fluid motions. During USML-2 a large set of experiments with spherically symmetric heating were carried out. These enabled the determination of critical points for the transition to various forms of non-axisymmetric convection and, for highly turbulent flows, the transition latitudes separating the different modes of motion. This paper presents a first analysis of these experiments as well as data on the general performance of the instrument during the USML-2 flight.

Description: An overview of the NASA Lewis Low-Pressure Turbine (LPT) Flow Physics Program will be presented. The program was established in response to the aero-engine industry's need for improved LPT efficiency and designs. Modern jet engines have four to seven LPT stages, significantly contributing to engine weight. In addition, there is a significant efficiency degradation between takeoff and cruise conditions, of up to 2 points. Reducing the weight and part count of the LPT and minimizing the efficiency degradation will translate into fuel savings. Accurate prediction methods of LPT flows and losses are needed to accomplish those improvements. The flow in LPT passages is at low Reynolds number, and is dominated by interplay of three basic mechanisms: transition, separation and wake interaction. The affecting parameters traditionally considered are Reynolds number, freestream turbulence intensity, wake frequency parameter, and the pressure distribution (loading). Three-dimensional effects and additional parameters, particularly turbulence characteristics like length scales, spectra and other statistics, as well as wake turbulence intensity and properties also play a role. The flow of most interest is on the suction surface, where large losses are generated as the flow tends to separate at the low Reynolds numbers. Ignoring wakes, a common flow scenario, there is laminar separation, followed by transition on the separation bubble and turbulent reattachment. If transition starts earlier the separation will be eliminated and the boundary layer will be attached leading to the well known bypass transition issues. In contrast, transition over a separation bubble is closer to free shear layer transition and was not investigated as well, particularly in the turbine environment. Unsteadiness created by wakes complicates the picture. Wakes induce earlier transition, and the calmed regions trailing the induced turbulent spots can delay or eliminate separation via shear stress modification. Three-dimensional flow physics and geometry will have strong effects. Altogether a very complex and challenging problem emerges. The objective of the program is to provide improved models and physical understanding of the complex flow, which are essential for accurate prediction of flow and losses in the LPT. Experimental, computational and analytical work as complementing and augmenting approaches are used. The program involves industry, universities and research institutes, and other government laboratories. It is characterized by strong interaction among participants, quick dissemination of results, and responsiveness to industry's needs. The presentation will describe the work elements. Highlighting some activities in progress are experiments on simulated blade suction surface in low-speed wind tunnels, on curved wall, and on a flat-plate, both with pressure gradient. In the area of computation, assessment of existing models is performed using RANS (Reynolds Averaged Navier Stokes) simulations. Laminar flow DNS was completed. Analytical studies of instability and receptivity in attached and separated flows were started. In the near future the program is moving to include wake effects and development of improved modeling. Experimental work in preparation stages are: (1) Addition of wakes to the curved tunnel experiment; (2) Low-speed rotating rig experiment on GE90 engine LPT; and (3) Transonic cascade. In the area of computation, it is expected to move from model assessment towards development of improved models. In addition, a new project of Large Eddy Simulation (LES) of LPT is to begin and will provide numerical data bases. It is planned to implement the emerging improved models in a multistage turbomachinery code and to validate against the GE90 engine LPT.

Description: Turbulent wakes are known to develop self-similarly sufficiently far downstream from obstacles that generate them. It has long been assumed that the spreading rate of the wake in the self-similar regime is independent of the details of the body generating the wake, being dependent only on the total drag (or momentum deficit). This assumption seems to be in contradiction with some recent experiments. In this study we attempt to complement these experimental investigations through a numerical study of a time-developing wake. A numerical study has the advantage of eliminating many of the uncontrolled factors present in experiments and allowing precise control of initial conditions. Large-eddy simulations employing the recently developed dynamic localization model are used to extend previous results from direct numerical simulations. The large-eddy simulation results are compared to the direct numerical simulation database, wherever such comparisons are feasible, as a check of the method. Like the experiments, the large-eddy simulations suggest that non-unique self-similar states, characterized by different spreading rates and turbulent statistics, are possible and that they can be maintained for significant time periods. The study also demonstrates the predictive capability of the dynamic localization subgrid model.

Description: Turbulence modeling plays an important role in the study of high-speed flows in engineering and aerodynamic problems; they include flows in supersonic combustion engines and over hypersonic transport aircraft. The enhancement of the kinetic energy dissipation by the dilatational terms is one of the typical compressibility effects. Zeman (1990) and Sarkar et al. (1991) proposed that the dilatation dissipation is proportional to the solenoidal dissipation and is a function of the turbulent Mach number. Sarkar (1992) also modeled the pressure-dilatation correlation using the turbulent Mach number. Zeman (1991) related the correlation to the rate of change of the pressure variance.

Description: Combustion is generally categorized as either premixed, where flames propagate into homogeneous mixtures of reactants, or as nonpremixed, where initially separated reactants diffuse into the reaction zones. Although these approaches are applicable to many combustion devices, there are cases not in either of these two limiting regimes. Under such circumstances, one must consider partially premixed combustion. In partially premixed combustion, mechanisms from both premixed and nonpremixed regimes coexist and, as a result, some interesting phenomena arise. One such phenomenon is flame stabilization in laminar mixing layers by triple flames.

Description: A new concept, Multiple Scale Simulation (MSS), is presented in this paper. The basic idea is that the flow is decomposed into several component groups according to spatial and temporal length scales. Each group has its own subdomain, governing system, mesh size, and discretization method. The simulation is then performed groupwise. This approach has been successfully applied in combination with the intergrid dissipation technique for simulation of transitional and turbulent flow in 3-D boundary layers, and it is feasible for 3-D airfoils and other more complex configurations. MSS should prove to ameliorate the scale problems associated with conventional direct numerical simulation.

Description: The flow in a hydrostatic journal bearing (HJB) is described by a mathematical model that uses the three dimensional non-orthogonal form of the Navier-Stokes equations. Using the u, v, w, and p, as primary variables, a conservative formulation, finite volume multi-block method is applied through a collocated, body fitted grid. The HJB has four shallow pockets with a depth/length ratio of 0.067. This paper represents a natural extension to the two and three dimensional studies undertaken prior to this project.

Description: A steady, three dimensional average-passage equation system was derived. The purpose was to simulate multistage turbomachinery flows. These equations describe a steady, viscous flow that is periodic from blade passage to blade passage. Moreover, these equations have a closure problem that is similar to that of the Reynolds-average Navier-Stokes equations. A scaled form of the average-passage equation system could provide an improved mathematical model for simulating the flow in the design and in the off-design conditions of a multistage machine.

Description: Rarefaction effects are important for hypersonic applications for a wide spectrum of conditions ranging from low-density (high altitude) situations to relatively high-density flows where the characteristic dimension is small. The present chapter concentrates on two hypersonic flow problems at flow conditions that produce a significant range of rarefaction effects: corner flow with jet interaction and blunt body flow with special emphasis on the near wake, These problems were chosen because they involve complex flow interactions that have significant implications for both spacecraft and re-entry vehicles. In an effort to clarify issues associated with these two general flow problems and to enhance their respective databases, both experimental and computational contributions were executed by an international group of researchers. In some cases, multiple data sources for both experimental and computational contributions are achieved.

Description: In this chapter recent activity in real-gas database definition and code validation will be summarized. In the Phase I report of the Working Group (WG) 181, aerothermodynamic problems were classified, for purpose of discussion, into seven types: aerodynamic parameters, viscous/shock interaction, boundary-layer transition, forebody-heating/heat-transfer, radiation and ablation, lee and base-region flow, and low-density flow. Several of these problem types were the subject of various chapters of the Phase 1 report describing real-gas effects and ground test facility issues. In this chapter some background and objectives outlined in the real-Gas effects Chapter V of the Phase 1 report will be reviewed. The results of the blunt cone test campaign developed under the auspices of the WG18 activity to study real-gas phenomena will be summarized, including the experimental and computational programs, issues and questions, and recommendations. Further, recent progress in other real-gas areas beyond the blunt cone test campaign will be discussed. Finally, a summary in which the present status of our understanding of real-gas issues will be presented.

Description: CFD has played a major role in the resurgence of hypersonic flight, on the premise that numerical methods will allow us to perform simulations at conditions for which no ground test capability exists. Validation of CFD methods is being established using the experimental data base available, which is below Mach 8. It is important, however, to realize the limitations involved in the extrapolation process as well as the deficiencies that exist in numerical methods at the present time. Current features of CFD codes are examined for application to propulsion system components. The shortcomings in simulation and modeling are identified and discussed.

Description: Implementation issues associated with the application of Krylov subspace iterative methods, such as Newton-GMRES, are presented within the framework of practical computational fluid dynamic (CFD) applications. This paper categorizes, evaluates, and contrasts the major ingredients (function evaluations, matrix-vector products, and preconditioners) of Newton-GMRES Krylov subspace methods in terms of their effect on the local linear and global nonlinear convergence, memory requirements, and accuracy. The discussion focuses on Newton-GMRES in both a structured multi-zone incompressible Navier-Stokes solver and an unstructured mesh finite-volume Navier-Stokes solver. Approximate versus exact matrix-vector products, effective preconditioners, and other pertinent issues are addressed.

Description: Ultrasonic and electrostatic levitation techniques have allowed the experimental investigation of the nonlinear oscillatory dynamics of free droplets with diameter between 0.1 and 0.4 cm. The measurement of the resonance frequencies of the first three normal modes of large amplitude shape oscillations in an electric field of varying magnitude has been carried out with and without surface charges for weakly conducting liquids in air. These oscillations of nonspherical levitated drops have been driven by either modulating the ultrasonic field or by using a time-varying electric field, and the free decay from the oscillatory state has been recorded. A decrease in the resonance frequency of the driven fundamental quadrupole mode has been measured for increasing oblate deformation in the absence of an electric field. Similarly, a decrease in this frequency has also been found for increasing DC electric field magnitude. A soft nonlinearity exists in the amplitude dependence of the resonant mode frequencies for freely decaying as well as ultrasonically and electrically driven uncharged drops. This decrease in resonance frequency is accentuated by the presence of free surface charge on the drop. Subharmonic resonance excitation has been observed for drops in a time-varying electric field, and hysteresis exists for resonant modes driven to large amplitude. Mode coupling from lower-order resonances to higher-order modes has been found to be very weak, even for fairly large amplitude shape oscillations. Most of these results are in general agreement with predictions from recent analytical and numerical investigations.

Description: The Liquid Propellant Program (LPP) computer code is a super-set of the industry standard Two Dimensional Kinetics (TDK) computer code. The TDK code uses a two dimensional method of characteristics solution with fully coupled finite rate kinetics for axially symmetric nozzles. The chemical reactions are modeled with a generalized reaction package that include three dimensional body efficiencies and four reaction rate forms. The code performs optional solutions for frozen or equilibrium flow. TDK evaluates discrete shocks, both attached or induced. The Transonic module models variable mixture ratio profiles from the combustion chamber injector. The Mass Addition Boundary Layer module (MABL) calculates the boundary parameters with the same chemistry options, and includes transpiration or tangential slot injection of gas at the wall. The LPP upgrades include: planar nozzle, scarfed nozzles, plug nozzles, and scramjet nozzle configurations. The code evaluates both upper and lower wall flow simulation, and includes the interaction with the external flow. The MABL module evaluates equilibrium radiation heat transfer for both upper and lower walls. In addition, LPP code models combustion effects due to injector inefficiencies with the Spray Combustion Analysis Program (SCAP) module. The LPP package provides extensive post plotting capabilities for flow visualization. The LPP is sufficiently fast and robust to provide performance predictions for extensive parametric studies and sufficiently accurate to provide flow field and performance solutions for detailed studies.

Description: To better understand the Space Shuttle Main Engine (SSME) startup/shutdown tansients, an airflow test of a two dimensional nozzle was conducted at Marshall Space Flight Center's trisonic wind tunnel. Photographic and other instrumentation show during an SSME start large nozzle shell distortions occur as the Mach disk is passing through the nozzle. During earlier develop of the SSME, this startup transient resulted in low cycle fatigue failure of one of the LH2 feedlines. The two dimensional SSME nozzle test was designed to measure the static and fluctuating pressure environment and color Schlieren video during the startup and shutdown phases of the run profile.

Description: The objective of the present research is to develop improved turbulence models for the computation of complex flows through turbomachinery passages, including the effects of streamline curvature, heat transfer and secondary flows. Advanced turbulence models are crucial for accurate prediction of rocket engine flows, due to existance of very large extra strain rates, such as strong streamline curvature. Numerical simulation of the turbulent flows in strongly curved ducts, including two 180-deg ducts, one 90-deg duct and a strongly concave curved turbulent boundary layer have been carried out with Reynolds stress models (RSM) and algebraic Reynolds stress models (ARSM). An improved near-wall pressure-strain correlation has been developed for capturing the anisotropy of turbulence in the concave region. A comparative study of two modes of transition in gas turbine, the by-pass transition and the separation-induced transition, has been carried out with several representative low-Reynolds number (LRN) k-epsilon models. Effects of blade surface pressure gradient, freestream turbulence and Reynolds number on the blade boundary layer development, and particularly the inception of transition are examined in detail. The present study indicates that the turbine blade transition, in the presence of high freestream turbulence, is predicted well with LRN k-epsilon models employed. The three-dimensional Navier-Stokes procedure developed by the present authors has been used to compute the three-dimensional viscous flow through the turbine nozzle passage of a single stage turbine. A low Reynolds number k-epsilon model and a zonal k-epsilon/ARSM (algebraic Reynolds stress model) are utilized for turbulence closure. An assessment of the performance of the turbulence models has been carried out. The two models are found to provide similar predictions for the mean flow parameters, although slight improvement in the prediction of some secondary flow quantities has been obtained by the ARSM model. It's found that the wake profiles inside the endwall boundary layers are predicted better than those near the mid-span.

Description: The effective performance of modular thrusters in an aerospike configuration is difficult to determine. Standard analytical tools are applicable to conventional nozzle shapes, but are limited when applied to an aerospike nozzle (An aerospike nozzle is an altitude compensating external nozzle). Three baseline nozzle shapes are derived using standard analytical procedures. The baseline nozzle sizes are restricted to fill a volume envelope. The three shapes are an axi-symmetric round nozzle, a two dimensional planar square exit nozzle, and a super elliptic round to nearly square nozzle. The integrated (thruster/aerospike) performance of the three nozzles is determined through the use of three dimensional viscous computational fluid dynamic (CFD) calculation where complex features of the flow field can be accurately captured. The resulting installed performance is then used to evaluate the efficiency of these nozzle shapes for aerospike applications. The determination of effective performance of a thruster nozzle integrated into an aerospike nozzle require the solution of the three dimensional turbulent Navier-Stokes equations. The model used in this study consisted of two zones; one of the upstream thruster cowl surface so freestream conditions can be accurately predicted, and two, the aerospike surface beginning with the thruster outflow and extending to the end of the aerospike surface. The numerical grid consisted of over 120,000 nodes and used symmetry on the thruster centerline and edge. A two species non-reacting chemistry model was used to capture the variation of fluid properties between the hot plume base and freestream air. From the results of the three baseline nozzle aerospike calculations, the effictive performance of the nozzle was determined. The flow fields of these calculations do show some variation between the cases. Recirculation zones on the cowl surface is predicted for the two dimensional planar nozzle and a smaller one for the super elliptic nozzle. The recirculation is caused by the strong pressure gradient between the plume and freestream flows. The axi-symmetric nozzle results indicate recirculation zones on the thruster face. These recirculation zones smooth the pressure gradient between the plume and freestream flow limiting the formation of recirculation on the cowl surface. Thruster to thruster interaction is evident for the axi-symmetric and supper elliptic calculation while the two dimensional planar nozzle did not have any lateral expansion in the nozzle, so thruster to thruster interaction is limited. The integrated performance results, at the altitude choosen, show very little variation between the three thruster shapes. This result allows for nozzle shape determination based on additional considerations (thermal, structural, weight) besides performance.

Description: The Hydroblast Research Cell at MSFC is both a research and a processing facility. The cell is used to investigate fundamental phenomena associated with waterjets as well as to clean hardware for various NASA and contractor projects. In the area of research, investigations are made regarding the use of high pressure waterjets to strip paint, grease, adhesive and thermal spray coatings from various substrates. Current industrial methods of cleaning often use ozone depleting chemicals (ODC) such as chlorinated solvents, and high pressure waterjet cleaning has proven to be a viable alternative. Standard methods of waterjet cleaning use hand held or robotically controlled nozzles. The nozzles used can be single-stream or multijet nozzles, and the multijet nozzles may be mounted in a rotating head or arranged in a fan-type shape. We consider in this paper the use of a rotating, multijet, high pressure water nozzle which is robotically controlled. This method enables rapid cleaning of a large area, but problems such as incomplete coverage (e.g. the formation of 'islands' of material not cleaned) and damage to the substrate from the waterjet have been observed. In addition, current stripping operations require the nozzle to be placed at a standoff distance of approximately 2 inches in order to achieve adequate performance. This close proximity of the nozzle to the target to be cleaned poses risks to the nozzle and the target in the event of robot error or the striking of unanticipated extrusions on the target surface as the nozzle sweeps past. Two key motivations of this research are to eliminate the formation of 'coating islands' and to increase the allowable standoff distance of the nozzle.

Description: The operation of Capillary Pumped Loops (CPL's) in low gravity has generally been unable to match ground-based performance. The reason for this poorer performance has been elusive. In order to investigate the behavior of a CPL in low-gravity, an idealized, glass CPL experiment was constructed. This experiment, known as the Capillary-driven Heat Transfer (CHT) experiment, was flown on board the Space Shuttle Columbia in July 1997 during the Microgravity Science Laboratory mission. During the conduct of the CHT experiment an unexpected failure mode was observed. This failure mode was a result of liquid collecting and then eventually bridging the vapor return line. With the vapor return line blocked, the condensate was unable to return to the evaporator and dry-out subsequently followed. The mechanism for this collection and bridging has been associated with long wavelength instabilities of the liquid film forming in the vapor return line. Analysis has shown that vapor line blockage in present generation CPL devices is inevitable. Additionally, previous low-gravity CPL tests have reported the presence of relatively low frequency pressure oscillations during erratic system performance. Analysis reveals that these pressure oscillations are in part a result of long wavelength instabilities present in the evaporator pores, which likewise lead to liquid bridging and vapor entrapment in the porous media. Subsequent evaporation to the trapped vapor increases the vapor pressure. Eventually the vapor pressure causes ejection of the bridged liquid. Recoil stresses depress the meniscus, the vapor pressure rapidly increases, and the heated surface cools. The process then repeats with regularity.

Description: Heinzl et al. (1985) reports that experiments in ink-jets to produce drawings or signals occurred as early as 1930. Various companies such as IBM and Pitney-Bowes have conducted extensive studies on these devices for many years. Many such reports are available in such journals as the IBM Journal of Research and Development. While numerous articles have been published on the jetting characteristics of ink and water, the literature is rather limited on fluids such as waxes (Gao & Sonin 1994) or non-water based fluids (Passow, et al. 1993). This present study extends the knowledge base to determine the performance of molten waxes in "ink-jet" type printers for rapid prototyping. The purpose of this research was to qualitatively and quantitatively study the droplet formation of a drop-on-demand ink-jet type nozzle system for rapid prototyping.

Description: Accurate prediction of all physical phenomena in a combustion chamber is essential for better understanding of the system performance. Atomization, evaporation, combustion, chemical kinetics, and turbulence are those processes of great importance that need to be well understood. Processes involving the liquid phase in a combustion chamber will be further complicated under supercritical conditions. More advanced and accurate numerical techniques are required to extend our understanding of the above phenomena. A computer program for multi-species/multi-phase flow was developed for NASA/MSFC in 1992. This code, called Liquid Thrust Chamber Performance (LTCP) program takes an Eulerian- Eulerian approach and is based on the Total Variation Diminishing (TVD) technique with Lax-Friedrichs upwind method. Under the NASA/ASEE SFFP the LTCP code was used to predict the performance characteristics of several engines that were of particular interest to NASA. This code was also successful in a combustion detonation study. Converting the program to the PC platform was accomplished which extends usability and makes it available to a wider range of users. The Eulerian formulation of the liquid phase provides a suitable model that can be extended to include combustion modeling under supercritical conditions. The results have been compared against the ones of other codes and available measured data. The algorithm proved to be robust and efficient for problems with stiff source terms.

Description: This is a preliminary study devoted to verifying whether or not direct simulations of turbulent Magneto-Hydro-Dynamic (MHD) flows in liquid metals reproduce experimental observations of drag reduction. Two different cases have been simulated by a finite difference scheme which is second order accurate in space and time. In the first case, an external azimuthal magnetic field is imposed. In this case, the magnetic field acts on the mean axial velocity and complete laminarization of the flow at N(sub a) = 30 has been achieved. In the second case, an axial magnetic field is imposed which affects only fluctuating velocities, and thus the action is less efficient. This second case is more practical, but comparison between numerical and experimental results is only qualitative.

Description: Direct numerical simulations are performed for the elliptic streamline flow, which is a homogeneous turbulent flow that combines the effects of solid body rotation and strain. Simulations are run over a range of parameters in order to determine the effect of changing rotation and strain separately. For early times the nonlinear cascade is suppressed, but then is re-established at later times. The growth rate of turbulent kinetic energy agrees at early times with the trends from linear theory, but at later times the flow seems to approach an asymptotic state that is independent of the ratio of mean flow rotation rate to strain rate. A comparison with standard Reynolds stress turbulence models is made. It is found that for strong rotation rates, the models predict decay of the turbulence, while the simulations show exponential growth. Close examination of the simulation results shows that they are affected by excessively low Reynolds numbers. Suggestions for reducing low Reynolds number effects in future simulations is given.

Description: We report preliminary results obtained from the large eddy simulation of a backward facing step at a Reynolds number of 5100. The numerical platform is based on a high order Legendre spectral element spatial discretization and a least squares time integration scheme. A non-reflective outflow boundary condition is in place to minimize the effect of downstream influence. Smagorinsky model with Van Driest near wall damping is used for sub-grid scale modeling. Comparisons of mean velocity profiles and wall pressure show good agreement with benchmark data. More studies are needed to evaluate the sensitivity of this method on numerical parameters before it is applied to complex engineering problems.

Description: A wall pressure frequency spectrum model (Blake 1971 ) has been evaluated using databases from Direct Numerical Simulations (DNS) of a turbulent boundary layer (Na & Moin 1996). Good agreement is found for moderate to strong adverse pressure gradient flows in the absence of separation. In the separated flow region, the model underpredicts the directly calculated spectra by an order of magnitude. The discrepancy is attributed to the violation of the model assumptions in that part of the flow domain. DNS computed coherence length scales and the normalized wall pressure cross-spectra are compared with experimental data. The DNS results are consistent with experimental observations.

Description: A variable explicit/implicit characteristics-based advection scheme that is second-order accurate in space and time has been developed recently for unstructured deforming meshes (O'Rourke & Sahota 1996a). To explore the suitability of this methodology for Large-Eddy Simulation (LES), three subgrid-scale turbulence models have been implemented in the CHAD CFD code (O'Rourke & Sahota 1996b): a constant-coefficient Smagorinsky model, a dynamic Smagorinsky model for flows having one or more directions of statistical homogeneity, and a Lagrangian dynamic Smagorinsky model for flows having no spatial or temporal homogeneity (Meneveau et al. 1996). Computations have been made for three canonical flows, progressing towards the intended application of in-cylinder flow in a reciprocating engine. Grid sizes were selected to be comparable to the coarsest meshes used in earlier spectral LES studies. Quantitative results are reported for decaying homogeneous isotropic turbulence, and for a planar channel flow. Computations are compared to experimental measurements, to Direct-Numerical Simulation (DNS) data, and to Rapid-Distortion Theory (RDT) where appropriate. Generally satisfactory evolution of first and second moments is found on these coarse meshes; deviations are attributed to insufficient mesh resolution. Issues include mesh resolution and computational requirements for a specified level of accuracy, analytic characterization of the filtering implied by the numerical method, wall treatment, and inflow boundary conditions. To resolve these issues, finer-mesh simulations and computations of a simplified axisymmetric reciprocating piston-cylinder assembly are in progress.

Description: Subgrid-scale models for Large Eddy Simulation (LES) in both the velocity-pressure and the vorticity-velocity formulations were evaluated and compared in a priori tests using spectral Direct Numerical Simulation (DNS) databases of isotropic turbulence: 128(exp 3) DNS of forced turbulence (Re(sub(lambda))=95.8) filtered, using the sharp cutoff filter, to both 32(exp 3) and 16(exp 3) synthetic LES fields; 512(exp 3) DNS of decaying turbulence (Re(sub(Lambda))=63.5) filtered to both 64(exp 3) and 32(exp 3) LES fields. Gaussian and top-hat filters were also used with the 128(exp 3) database. Different LES models were evaluated for each formulation: eddy-viscosity models, hyper eddy-viscosity models, mixed models, and scale-similarity models. Correlations between exact versus modeled subgrid-scale quantities were measured at three levels: tensor (traceless), vector (solenoidal 'force'), and scalar (dissipation) levels, and for both cases of uniform and variable coefficient(s). Different choices for the 1/T scaling appearing in the eddy-viscosity were also evaluated. It was found that the models for the vorticity-velocity formulation produce higher correlations with the filtered DNS data than their counterpart in the velocity-pressure formulation. It was also found that the hyper eddy-viscosity model performs better than the eddy viscosity model, in both formulations.

Description: A general discussion on the structure of the eddy viscosity tensor in anisotropic flows is presented. The systematic use of tensor symmetries and flow symmetries is shown to reduce drastically the number of independent parameters needed to describe the rank 4 eddy viscosity tensor. The possibility of using Onsager symmetries for simplifying further the eddy viscosity is discussed explicitly for the axisymmetric geometry.

Description: In numerical simulation of complex flows, it is important to identify different length scales of the flow and treat them differently. In this report, we introduce a new multilevel scheme for simulating turbulent channel flows. Two different versions of the scheme, namely the spectral and finite difference versions, are presented. The spectral version of the scheme is based on a spectral-Galerkin formulation which provides a natural decomposition of the flow into small and large wavelength parts, and which leads to linear systems that can be solved with quasi-optimal computational complexity. In the finite difference version, the Incremental Unknown (IU) is used to separate the length scales. Preliminary numerical results indicate that the scheme is well suited for turbulence computations and provides results which are comparable to that by Direct Numerical Simulation (DNS) but with significantly less CPU time.

Description: The possibility of using the information from simultaneous equivalent Large Eddy Simulations (LAS) for improving the subgrid scale modeling is investigated. An ensemble average dynamic model is proposed as an alternative to the usual spatial average versions. It is shown to be suitable independently of the existence of any homogeneity directions, and its formulation is thus universal. The ensemble average dynamic model is shown to give very encouraging results for as few as 16 simultaneous LES's.

Description: The equations describing the first two terms of an asymptotic expansion of the solution of the planar turbulent mixing layer for values of the velocity ratio close to one are obtained. The first term of this expansion is the solution of the well-known time-evolving problem and the second, which includes the effects of the increase of the turbulence scales in the stream-wise direction, obeys a linear system of equations. Numerical solutions of these equations for a two-dimensional reacting mixing layer show that the correction to the time-evolving solution may explain the asymmetry of the entrainment and the differences in product generation observed in flip experiments.

Description: A four-equation model of turbulence is applied to the numerical simulation of flows with massive separation induced by a sudden expansion. The model constants are a function of the flow parameters, and two different formulations for these functions are tested. The results are compared with experimental data for a high Reynolds-number case and with experimental and DNS data for a low Reynolds-number case. The computations prove that the recovery region downstream of the massive separation is properly modeled only for the high Re case. The problems in this case stem from the gradient diffusion hypothesis, which underestimates the turbulent diffusion.

Description: The beta-assumed-pdf approximation of (Cook & Riley 1994) is tested as a subgrid model for the LES computation of nonpremixed turbulent reacting flows, in the limit of cold infinitely fast chemistry, for two plane turbulent mixing layers with different degrees of intermittency. Excellent results are obtained for the computation of integrals properties such as product mass fraction, and the model is applied to other quantities such as powers of the temperature and the pdf of the scalar itself. Even in these cases the errors are small enough to be useful in practical applications. The analysis is extended to slightly out of equilibrium problems such as the generation of radicals, and formulated in terms of the pdf of the scalar gradients. It is shown that the conditional gradient distribution is universal in a wide range of cases whose limits are established. Within those limits, engineering approximations to the radical concentration are also possible. It is argued that the experiments in this paper are essentially in the limit of infinite Reynolds number.

Description: A Second Moment Closure computation (SMC) is compared in detail with the Direct Numerical Simulation (DNS) data of Le and Moin for the backstep flow at Re = 5,000 in an attempt to understand why the intensity of the backflow and, consequently, the friction coefficient in the recirculation bubble are severely underestimated. The data show that this recirculation bubble is far from being laminar except in the very near wall layer. A novel 'differential a priori' procedure was used, in which the full transport equation for one isolated component of the Reynolds stress tensor was solved using DNS data as input. Conclusions are then different from what would have been deduced by comparing a full simulation to a DNS. One cause of discrepancy was traced back to insufficient transfer of energy to the normal stress by pressure strain, but was not cured. A significant finding, confirmed by the DNS data in the core region of a channel flow, is that the coefficient that controls destruction of dissipation, C epsilon(sub 2), should be decreased by a factor of 2 when production is vanishing. This is also the case in the recirculation bubble, and a new formulation has cured 25% of the backflow discrepancy.

Description: New models are proposed for the 'slow and 'rapid' parts of the pressure diffusive transport based on the examination of DNS databases for plane mixing layers and wakes. The model for the 'slow' part is non-local, but requires the distribution of the triple-velocity correlation as a local source. The latter can be computed accurately for the normal component from standard gradient diffusion models, but such models are inadequate for the cross component. More work is required to remedy this situation.

Description: A new approach to Reynolds averaged turbulence modeling is proposed which has a computational cost comparable to two equation models but a predictive capability approaching that of Reynolds stress transport models. This approach isolates the crucial information contained within the Reynolds stress tensor, and solves transport equations only for a set of 'reduced' variables. In this work, Direct Numerical Simulation (DNS) data is used to analyze the nature of these newly proposed turbulence quantities and the source terms which appear in their respective transport equations. The physical relevance of these quantities is discussed and some initial modeling results for turbulent channel flow are presented.

Description: A second order effect associated with high intensity sound field, acoustic streaming has been historically investigated to gain a fundamental understanding of its controlling mechanisms and to apply it to practical aspects of heat and mass transfer enhancement. The objectives of this new research project are to utilize a unique experimental technique implementing ultrasonic standing waves in closed cavities to study the details of the generation of the steady-state convective streaming flows and of their interaction with the boundary of ultrasonically levitated near-spherical solid objects. The goals are to further extend the existing theoretical studies of streaming flows and sample interactions to higher streaming Reynolds number values, for larger sample size relative to the wavelength, and for a Prandtl and Nusselt numbers parameter range characteristic of both gaseous and liquid host media. Experimental studies will be conducted in support to the theoretical developments, and the crucial impact of microgravity will be to allow the neglect of natural thermal buoyancy. The direct application to heat and mass transfer in the absence of gravity will be emphasized in order to investigate a space-based experiment, but both existing and novel ground-based scientific and technological relevance will also be pursued.

Description: It is well known that radiative heat transport influences many types of buoyant flows due to its effect on the temperature and thus density field in the fluid medium. It is of interest to study gaseous flows driven solely by radiation in the absence of buoyancy, particularly because of its application to astrophysical flows that are well known from astronomical observations and numerical simulation. However, no laboratory-scale experiments of this phenomenon have ever been conducted. To study the possibility of obtaining such flows in the laboratory, an apparatus was built to produce large temperature differences (Delta (T)) up to 300 K in a gas confined between flat parallel plates. SF6 was used as the radiatively-active gas because its Planck absorption length is much shorter than that of any other common non-reactive gas. The NASA-Lewis 2.2 second drop tower was used to obtain reduced gravity in order to suppress buoyancy effects. To image the resulting flows, a laser shearing interferometer was employed. Initial results indicate the presence of flow that does not appear to be attributable to the residual flow resulting from buoyancy influences before the drop. For Delta(T) greater than 70 K, slight deformations in the interferometer fringes seen at lower Delta(T) became large unsteady swirls. Such behavior did not occur for radiatively-inactive gases, suggesting that a flow driven solely by radiation was obtained in SF6 and to a lesser extent in CO2. This was more pronounced at higher pressures and plate spacings, consistent with our scaling predictions.

Description: The first objective of this experiment is to build an experimental system in which, in analogy to a geophysical system, a compressible fluid in a spherical annulus becomes radially stratified in density through an A.C. electric field. When this density gradient is demonstrated, the system will be augmented so that the fluid can be driven by heating and rotation and tested in preparation for a microgravity experiment. This apparatus consists of a spherical capacitor filled with critical fluid in a temperature controlled environment. To make the fluid critical, the apparatus will be operated near the critical pressure, critical density, and critical temperature of the fluid. This will result in a highly compressible fluid because of the properties of the fluid near its critical point. A high voltage A.C. source applied across the capacitor will create a spherically symmetric central force because of the dielectric properties of the fluid in an electric field gradient. This central force will induce a spherically symmetric density gradient that is analogous to a geophysical fluid system. To generate such a density gradient the system must be small (approx. 1 inch diameter). This small cell will also be capable of driving the critical fluid by heating and rotation. Since a spherically symmetric density gradient can only be made in microgravity, another small cell, of the same geometry, will be built that uses incompressible fluid. The driving of the fluid by rotation and heating in these small cells will be developed. The resulting instabilities from the driving in these two systems will then be studied. The second objective is to study the pattern forming instabilities (bifurcations) resulting from the well controlled experimental conditions in the critical fluid cell. This experiment will come close to producing conditions that are geophysically similar and will be studied as the driving parameters are changed.

Description: We are beginning laboratory experiments using magnetically active ferrofluids to study surface waves in novel geometries. Terrestrial gravity is eliminated from the dynamics, and the magnetic body force felt by ferrofluid in the presence of a magnetic field gradient is used to create a geopotential field which is a section of or an entire sphere or cylinder. New optical, electromagnetic and ultrasonic diagnostic techniques are under development to initially study capillary-gravity wave propagation and interaction in such geometries.

Description: In a recent review it is said that free-surface flows 'represent some of the difficult remaining challenges in computational fluid dynamics'. There has been progress with the development of new approaches to treating interfaces, such as the level-set method and the improvement of older methods such as the VOF method. A common theme of many of the new developments has been the regularization of discontinuities at the interface. One example of this approach is the continuum surface force (CSF) formulation for surface tension, which replaces the surface stress given by Laplace's equation by an equivalent volume force. Here, we describe how CSF formulation might be made more useful. Specifically, we consider a derivation of the CSF equations from a minimization of surface energy as outlined by Jacqmin (1996). This reformulation suggests that if one eliminates the computation of curvature in terms of a unit normal vector, parasitic currents may be eliminated. For this reformulation to work, it is necessary that transition region thickness be controlled. Various means for this, in addition to the one discussed by Jacqmin (1996), are discussed.

Description: The technique of Fluorescence Recovery After Photobleaching (FRAP) has been applied to the measurement of interfacial transport in two-phase systems. FRAP exploits the loss of fluorescence exhibited by certain fluorophores when over-stimulated (photobleached), so that a two-phase system, originally at equilibrium, can be perturbed without disturbing the interface by strong light from an argon-ion laser and its recovery monitored by a microscope-mounted CCD camera as it relaxes to a new equilibrium. During this relaxation, the concentration profiles of the probe solute are measured on both sides of the interface as a function of time, yielding information about the transport characteristics of the system. To minimize the size of the meniscus between the two phases, a photolithography technique is used to selectively treat the glass walls of the cell in which the phases are contained. This allows concentration measurements to be made very close to the interface and increases the sensitivity of the FRAP technique.

Description: Analytical and numerical studies are to be carried out to examine time-averaged thermal effects which are induced by the interaction of strong acoustic fields with a rigid boundary (thermoacoustic streaming). Also of interest is the significance of a second-order thermal expansion coefficient that emerges from this analysis. The model problem to be considered is that of a sphere that is acoustically levitated such that it is effectively isolated in a high-intensity standing acoustic field. The solution technique involves matched asymptotic analysis along with numerical solution of the boundary layer equations. The objective of this study is to predict the thermoacoustic streaming behavior and fully understand the role of the associated second-order thermodynamic modulus.

Description: The motion of a fluid-fluid-solid contact line on a rough surface is well known to display hysteresis in the contact angle vs. velocity relationship. In order to understand the phenomenon at a fundamental microscopic level, we have conducted molecular dynamics computer simulations of a Wilhelmy plate experiment in which a solid surface is dipped into a liquid bath, and the force-velocity characteristics are measured. We directly observe a systematic variation of force and contact angle with velocity, which is single-valued for the case of an atomically smooth solid surface. In the microscopically rough case, however, we find (as intuitively expected) an open hysteresis loop. Further characterization of the interface dynamics is in progress.

Description: We will pursue an improved physical understanding and mathematical model for the boundary condition at an oscillating contact line at high Reynolds number. We expect that the body force is locally unimportant for earth-based systems, and that the local behavior may dominate the mechanics of partially-filled reservoirs in the microgravity environment. One important space-based application for this contact-line study is for Faraday-waves. Oscillations in the direction of gravity (or acceleration) can dominate the fluid motion during take-off and reentry with large steady-state accelerations and in orbit, where fluctuations on the order of 10(exp -4)g occur about a zero mean. Our experience with Faraday waves has shown them to be 'cleaner' than those produced by vertical or horizontal oscillation of walls. They are easier to model analytically or computationally, and they do not have strong vortex formation at the bottom of the plate. Hence many, if not most, of the experiments will be performed in this manner. The importance of contact lines in the microgravity environment is well established. We will compare high resolution measurements of the velocity field (lO micro-m resolution) using particle-tracking and particle-image velocimetry as the fluid/fluid interface is approached from the lower fluid. The spatial gradients in the deviation provide additional means to determine an improved boundary condition and a measure of the slip region. Dissipation, the size of the eddy near the contact line, and hysteresis will be measured and compare to linear and nonlinear models of viscous and irrotational but dissipative models.

Description: The oscillation frequency and damping rate for small-amplitude axisymmetric shape modes of a gas bubble in an ideal liquid are obtained, in the limit when the bubble interface possesses Newtonian interfacial rheology with constant surface shear and dilatational viscosities. Such results permit the latter surface properties to be measured by analyzing experimental data on frequency shift and damping rate of specific shape modes of suspended bubbles in the presence of surfactants.

Description: Single bubble dynamics are investigated using acoustic techniques for isolation and manipulation. The goal of the investigations is to understand the dynamic origin of the various phenomena that bubbles exhibit: light emission, enhanced mass transport, chaotic and quasiperiodic oscillations, and translations. Once understood, acoustically manipulated bubbles can serve as platforms for materials effects on free surfaces, using surfactants to alter surface rheology and observing how that affects both dynamics and also mass transport. The effects of gravity on the problem will be shown to be significant. The first set of observations from 1g experimentation are presented. These observations are of the onset conditions for instability of the spherical shape of the bubble. For the size range 55-90 microns in diameter we observe instability governed by resonant mode coupling, which is significantly affected by the buoyant force and its effects.

Description: The proposed research is divided into three components concerned with molecular structure, molecular orientation, and continuum averages of discrete systems. In the experimental program, we propose exploring how changes in interfacial molecular structure generate contact line motion. Rather than rely on the electrostatic and electrokinetic fields arising from the molecules themselves, we augment their interactions by an imposed field at the solid/liquid interface. By controling the field, we can manipulate the molecular structure at the solid/liquid interface. In response to controlled changes in molecular structure, we observe the resultant contact line motion. In the analytical portion of the proposed research we seek to formulate a system of equations governing fluid motion which accounts for the orientation of fluid molecules. In preliminary work, we have focused on describing how molecular orientation affects the forces generated at the moving contact line. Ideally, as assumed above, the discrete behavior of molecules can be averaged into a continuum theory. In the numerical portion of the proposed research, we inquire whether the contact line region is, in fact, large enough to possess a well-defined average. Additionally, we ask what types of behavior distinguish discrete systems from continuum systems. Might the smallness of the contact line region, in itself, lead to behavior different from that in the bulk? Taken together, our proposed research seeks to identify and accurately account for some of the molecular dynamics of the moving contact line, and attempts to formulate a description from which one can compute the forces at the moving contact line.

Description: Self-diffusion coefficients that describe cross-flow migration of non-Brownian drops in a dilute sheared emulsion were obtained by trajectory calculations. A boundary integral formulation was used to describe pairwise interactions between deformable drops; interactions between undeformed drops were described with mobility functions for spherical drops. The results indicate that drops have large anisotropic self-diffusivities which depend strongly on the drop viscosity and modestly on the shear-rate. Pairwise interactions between drops in shear-flow do not appreciably promote drop breakup.

Description: Microgravity processes must rely on mechanisms other than bouyancy to move bubbles or droplets from one region to another in a continuous liquid phase. One suggested method is thermocapillary migration in which a temperature gradient is applied to the continuous phase. When a fluid particle contacts this gradient, one pole of the particle becomes warmer than the opposing pole. The interfacial tension between the drop or bubble phase and the continuous phase usually decreases with temperature. Thus the cooler pole is of higher interfacial tension than the warmer pole, and the interface is tugged in the direction of the cooler end. This thermocapillary or thermally induced Marangoni surface stress causes a fluid streaming in the continuous phase from which develops a viscous shear traction and pressure gradient which together propel the particle in the direction of the warmer fluid. In this paper, we provide a theoretical basis for remobilizing surfactant retarded fluid particle interfaces in an effort to make viable the use of thermocapillary migrations for the management of bubbles and drops in microgravity,

Description: This is a review of our experimental and theoretical studies relating to equilibrium and stability of liquid drops, typically of low viscosity, levitated in air by a sound field. The major emphasis here is on the physical principles and understanding behind the stability of levitated drops. A comparison with experimental data is also given, along with some fascinating pictures from high-speed photography. One of the aspects we shall deal with is how a drop can suddenly burst in an intense sound field; a phenomenon which can find applications in atomization technology. Also, we are currently investigating the phenomenon of suppression of coalescence between drops levitated in intense acoustic fields.

Description: Oscillations of supported liquid drops are the subject of wide scientific interest, with applications in areas as diverse as liquid-liquid extraction, synthesis of ceramic powders, growing of pure crystals in low gravity, and measurement of dynamic surface tension. In this research, axisymmetric forced oscillations of arbitrary amplitude of viscous liquid drops of fixed volume which are pendant from or sessile on a rod with a fixed or moving contact line and surrounded by an inviscid ambient gas are induced by moving the rod in the vertical direction sinusiodally in time. In this paper, a preliminary report is made on the computational analysis of the oscillations of supported drops that have 'clean' interfaces and whose contact lines remain fixed throughout their motions. The relative importance of forcing to damping can be increased by either increasing the amplitude of rod motion A or Reynolds number Re. It is shown that as the ratio of forcing to damping rises, for drops starting from an initial rest state a sharp increase in deformation can occur when they are forced to oscillate in the vicinity of their resonance frequencies, indicating the incipience of hysteresis. However, it is also shown that the existence of a second stable limit cycle and the occurrence of hysteresis can be observed if the drop is subjected to a so-called frequency sweep, where the forcing frequency is first increased and then decreased over a suitable range. Because the change in drop deformation response is abrupt in the vicinity of the forcing frequencies where hysteresis occurs, it should be possible to exploit the phenomenon to accurately measure the viscosity and surface tension of the drop liquid.

Description: In this work, we study the combined thermocapillary and natural convective flow generated by a bubble on a heated solid surface. The interaction between gas and vapor bubbles with the surrounding fluid is of interest for both space and ground-based processing. On earth, the volumetric forces are dominant, especially, in apparatuses with large volume to surface ratio. But in the reduced gravity environment of orbiting spacecraft, surface forces become more important and the effects of Marangoni convection are easily unmasked. In order to delineate the roles of the various interacting phenomena, a combined numerical-experimental approach is adopted. The temperature field is visualized using Mach-Zehnder interferometry and the flow field is observed by a laser sheet flow visualization technique. A finite element numerical model is developed which solves the two-dimensional momentum and energy equations and includes the effects of bubble surface deformation. Steady state temperature and velocity fields predicted by the finite element model are in excellent qualitative agreement with the experimental results. A parametric study of the interaction between Marangoni and natural convective flows including conditions pertinent to microgravity space experiments is presented. Numerical simulations clearly indicate that there is a considerable difference between 1-g and low-g temperature and flow fields induced by the bubble.

Description: Marangoni-Benard convection in evaporating liquid thin layers has been investigated through flow visualization and temperature profile measurement. Twelve liquids, namely ethyl alcohol, methanol, chloroform, acetone, cyclohexane, benzine, methylene chloride, carbon tetrachloride, ethyl acetate, n-pentane, silicone oil (0.65 cSt.), and freon-113, were tested and convection patterns in thin layers of these samples were observed. Comparison among these tested samples shows that some liquids are sensitive to surface contamination from aluminum powder but some are not. The latter is excellent to be used for the investigation of surface-tension driven convection through visualization using the tracer. Two sample liquids, alcohol and freon-113 were particularly selected for systematic study. It was found that the wavelength of Benard cells would not change with thickness of the layer when it evaporates at room temperature. Special attention was focused on cases in which a liquid layer was cooled from below, and some interesting results were obtained. Convection patterns were recorded during the evaporation process and the patterns at certain time frame were compared. Benard cells were observed in thin layers with a nonlinear temperature profile and even with a zero or positive temperature gradient. Wavelength of the cells was found to increase as the evaporation progressed.

Description: Ground-based experiments together with analytical studies are presently being conducted for levitated drops. Both acoustic and electrostatic techniques are being employed to achieve levitation of drops in a gaseous environment. The scientific effort is principally on the thermal and the fluid phenomena associated with the local heating of levitated drops, both at 1-g and at low-g. In particular, the thermocapillary flow associated with local spot heating is being studied. Fairly stable acoustic levitation of drops has been achieved with some exceptions when random rotational motion of the drop persists. The flow visualization has been carried out by light scattering from smoke particles for the exterior flow and fluorescent tracer particles in the drop. The results indicate a lack of axial symmetry in the internal flow even though the apparatus and the heating are symmetric. The theoretical studies for the past year have included fundamental analyses of acoustically levitated spherical drops. The flow associated with a particle near the velocity antinode is being investigated by the singular perturbation technique. As a first step towards understanding the effect of the particle displacement from the antinode, the flow field about the node has been calculated for the first time. The effect of the acoustic field on the interior of a liquid drop has also been investigated. The results predict that the internal flow field is very weak.

Description: The deformation of the fluid column by an action of a low-frequency vibration is considered. It is shown that behavior of the free fluid surface depends on the frequency of applied vibration and its amplitude. In the area of very low frequencies when fluid has time to comment on travel of bounding solid walls limiting column, the harmonical oscillations of free surface with given frequency are observed. With increase of vibration frequency the steady-state relief on free fluid surface is formed. If the amplitude of vibration is very small and the frequency corresponding to the first peak in the vibration spectrum on the Mir orbital station, the deformation of free surface tends to zero. Fluid flow induced thermocapillary effect on deformed free surface is more unstable as in the case of smooth cylindrical surface. It was shown that width of heating zone affects very essentially the flow pattern and transition to oscillatory regime of thermocapillary convection.

Description: Theoretical and experimental investigation of the stability of nonaxisymmetric and nonaxisymmetric bridges contained between equal and unequal radii disks as a function of Bond and Weber number with emphasis on the transition from unstable axisymmetric to stable nonaxisymmetric shapes, are conducted. Numerical analysis of the stability of nonaxisymmetric bridges between unequal disks for various orientations of the gravity vector is performed. Experimental and theoretical investigation of large (nonaxisymmetric) oscillations and breaking of liquid bridges are also conducted.

Description: Surfactant-induced Marangoni effects strongly alter the stresses exerted along fluid particle interfaces. In low gravity processes, these stresses can dictate the system behavior. The dependence of Marangoni effects on surfactant physical chemistry is not understood, severely impacting our ability to predict and control fluid particle flows. A droplet in an extensional flow allows the controlled study of stretching and deforming interfaces. The deformations of the drop allow both Marangoni stresses, which resist tangential shear, and Marangoni elasticities, which resist surface dilatation, to develop. This flow presents an ideal model system for studying these effects. Prior surfactant-related work in this flow considered a linear dependence of the surface tension on the surface concentration, valid only at dilute surface concentrations, or a non-linear framework at concentrations sufficiently dilute that the linear approximation was valid. The linear framework becomes inadequate for several reasons. The finite dimensions of surfactant molecules must be taken into account with a model that includes surfaces saturation. Nonideal interactions between adsorbed surfactant molecules alter the partitioning of surfactant between the bulk and the interface, the dynamics of surfactant adsorptive/desorptive exchange, and the sensitivity of the surface tension to adsorbed surfactant. For example, cohesion between hydrocarbon chains favors strong adsorption. Cohesion also slows the rate of desorption from interfaces, and decreases the sensitivity of the surface tension to adsorbed surfactant. Strong cohesive interactions result in first order surface phase changes with a plateau in the surface tension vs surface concentration. Within this surface concentration range, the surface tension is decoupled from surface concentration gradients. We are engaged in the study of the role of surfactant physical chemistry in determining the Marangoni stresses on a drop in an extensional flow in a numerical and experimental program. Using surfactants whose dynamics and equilibrium behavior have been characterized in our laboratory, drop deformation will be studied in ground-based experiment. In an accompanying numerical study, predictive drop deformations will be determined based on the isotherm and equation of state determined in our laboratory. This work will improve our abilities to predict and control all fluid particle flows.

Description: The object of this work, started in March of 1995, is to approach the problem of determining the transport conditions (and effects of residual acceleration) during the plane-front directional solidification of a tin-bismuth alloy under low gravity conditions. The work involves using a combination of 2- and 3-D numerical models, scaling analyses, 1-D models and the results of ground-based and low-gravity experiments. The experiments conducted in the MEPHISTO furnace facility during the USMP-3 spaceflight which took place earlier this year (22 Feb. - 6 Mar. 1996). This experiment represents an unprecedented opportunity to make a quantitative correlation between residual accelerations and the response of an actual experimental solidification system

Description: Turbulent gas flows laden with small, dense particles are encountered in a wide number of important applications in both industrial settings and aerodynamics applications. Particle interactions with the underlying turbulent flow are exceedingly complex and, consequently, difficult to accurately model. The difficulty arises primarily due to the fact that response of a particle to the local environment is dictated by turbulence properties in the reference frame moving with the particle (particle-Lagrangian). The particle-Lagrangian reference frame is in turn dependent upon the particle relaxation time (time constant) as well as gravitational drift. The combination of inertial and gravitational effects in this frame complicates our ability to accurately predict particle-laden flows since measurements in the particle-Lagrangian reference frame are difficult to obtain. Therefore, in this work we will examine separately the effects of inertia and gravitational drift on particle dispersion through a combination of physical and numerical experiments. In this study, particle-Lagrangian measurements will be obtained in physical experiments using stereo image velocimetry. Gravitational drift will be varied in the variable-g environments of the NASA DC-9 and in the zero-g environment at the drop tower at NASA-Lewis. Direct numerical simulations will be used to corroborate the measurements from the variable-g experiments. We expect that this work will generate new insight into the underlying physics of particle dispersion and will, in turn, lead to more accurate models of particle transport in turbulent flows.

Description: This is a new project which started in May 1996. The main objective of the experimental/numerical study is to improve the understanding of the physics of two-way coupling between the dispersed phase and turbulence in a prototypical turbulent shear flow - homogeneous shear, laden with small liquid droplets (in gas) or gaseous bubbles (in liquid). The method of direct numerical simulation (DNS) is used to solve the full three-dimensional, time-dependent Navier-Stokes equations including the terms describing the two-way coupling between the dispersed phase and the carrier flow. The results include the temporal evolution of the three-dimensional energy and dissipation spectra and the rate of energy transfer across the energy spectrum to understand the fundamental physics of turbulence modulation, especially the effects of varying the magnitude of gravitational acceleration. The mean-square displacement and diffusivity of the droplets (or bubbles) of a given size and the preferential accumulation of droplets in low vorticity regions and bubbles in high vorticity regions will be examined in detail for different magnitudes of gravitational acceleration. These numerical results which will be compared with their corresponding measured data will provide a data base from which a subgrid-scale (SGS) model can be developed and validated for use in large-eddy simulation (LES) of particle-laden shear flows. Two parallel sets of experiments will be conducted: bubbles in an immiscible liquid and droplets in air. In both experiments homogeneous shear will be imposed on the turbulent carrier flow. The instantaneous velocities of the fluid and polydispersed-size particles (droplets or bubbles) will be measured simultaneously using a two-component Phase-Doppler Particle Analyzer (PDPA). Also, the velocity statistics and energy spectra for the carrier flow will be measured.

Description: The effects of imposed nonlinear oscillatory shear upon the onset of Marangoni-Bernard convection, as predicted by linear theory, in a layer of liquid with a deformable free surface were reported upon by Or and Kelly for small amplitude oscillations. Depending on the operating conditions, either stabilization or destabilization might occur. The aim of the current paper is to report the results for finite amplitude imposed oscillations so that the actual amount of stabilization or destabilization can be determined for prescribed operating conditions.

Description: A computational study of thermally-driven convection in multilayered fluid structures will be performed to examine the effect of interactions among deformable fluid-fluid interfaces on the structure of time-dependent flow in these systems. Multilayered fluid structures in two models configurations will be considered: the differentially heated rectangular cavity with a free surface, and the encapsulated cylindrical liquid bridge. An extension of a numerical method developed as part of our recent NASA Fluid Physics grant will be used to account for finite deformations of fluid-fluid interfaces.

Description: Capillary Pumped Loops (CPL's) and Loop Heat Pipes (LHP's) are finding increased acceptance on upcoming NASA spacecraft missions, as well as military and commercial applications, The transition from a research and development tool to an "off the shelf' system is underway. The state of the art of two phase systems (TPS) is reviewed and applications on various NASA missions are discussed, with particular emphasis on new technology developments. Upcoming research areas and flight experiments are also addressed, along with recommendations for future activities.

Description: This paper details the flight configuration and pre-flight performance test results of the fifth generation cryogenic capillary pumped loop (CCPL-5). This device will fly on STS-95 in October 1998 as part of the CRYOTSU Flight Experiment. This flight represents the first in-space demonstration of a CCPL, a miniaturized two-phase fluid circulator for thermally linking cryogenic cooling sources to remote cryogenic components. CCPL-5 utilizes N2 as the working fluid and has a practical operating range of 75-110 K. Test results indicate that CCPL-5, which weighs about 200 grams, can transport over 10 W of cooling a distance of 0.25 m (or more) with less than a 5 K temperature drop.

Description: Capillary Pumped Loops (CPL's) and Loop Heat Pipes (LHP's) are finding increased acceptance on upcoming NASA spacecraft missions, as well as military and commercial applications. The transition from a research and development tool to an "off the shelf' system is underway. The state of the art of two phase systems (TPS) is reviewed and applications on various NASA missions are discussed, with particular emphasis on new technology developments. Upcoming research areas and flight experiments are also addressed, along with recommendations for future activities.