SEARCH · Engineering Papers
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
The basic numerical techniques used for the direct simulation of transition to turbulence are described. Applications are presented for the computation of finite amplitude, two-dimensional steady states, the simulation of the secondary instability of channels and boundary-layers, and the simulation of the breakdown of detached shear layers.
The primary objective of our work was to evaluate and test several wake management schemes for the reduction of turbomachinery fan noise. Throughout the course of this work we relied on several tools. These include 1) Two-dimensional steady boundary-layer and wake analyses using MISES (a thin-shear layer Navier-Stokes code), 2) Two-dimensional unsteady wake-stator interaction simulations using UNSFLO, 3) Three-dimensional, steady Navier-Stokes rotor simulations using NEWT, 4) Internal blade passage design using quasi-one-dimensional passage flow models developed at MIT, 5) Acoustic modeling using LINSUB, 6) Acoustic modeling using VO72, 7) Experiments in a low-speed cascade wind-tunnel, and 8) ADP fan rig tests in the MIT Blowdown Compressor.
This report describes the work accomplished during a first-year investigation of base drag reduction for launch vehicles. Interest in this work has arisen due to the large base areas associated with the current generation of launch vehicles. For example, the X-33 and Venture Star configurations all have large base areas because of the aero-spike engines they use. Because of the large base-to-wetted area ratios of these vehicles, the majority of the vehicle drag is due to base drag. Based on previous research, there appears to be a means for reducing the base drag on such vehicles. A clear relationship between viscous fore-body drag and base drag was demonstrated. For subsonic flow conditions, an increase in fore-body drag causes a decrease in base drag. This base-drag reduction is a result of boundary layer effects at the base of the vehicle. The shear layer that develops from the boundary layer separating at the back of the vehicle is the conduit through which momentum is transferred from the high energy free-stream flow to the low-energy fluid in the base area. One way to think of the base drag is that it is the momentum required to accelerate this low-energy fluid. The boundary layer that develops on the fore body acts as an "insulator" between the external flow and the low-energy air at the base. As the viscous fore-body drag is increased, the boundary layer thickness at the aft end of the fore body increases, thereby reducing the rate at which momentum is transferred to the base area. As a result of the lower momentum transfer (or reduced pumping), the base pressure coefficient rises resulting in a reduction of base drag.
A series of Direct Numerical Simulations of a temporally evolving shear layer subject to both harmonic (2D) and streamwise (3D) forcing were performed. The interaction and coupling of these various 2D and 3D modes is shown to significantly alter the development of the flow. The scale of the 3D modes is quite important to the coupling process with greatly enhanced mixing and product formation resulting from 3D modes that are rapidly amplified by the spanwise vorticity. In general, the longer wavelength 3D modes are found to be highly efficient at increasing the momentum transport while the shorter wavelengths increase mass transport.
A study of emerging turbulent scales entropy production is conducted for a supercritical shear layer as a precursor to the eventual modeling of Subgrid Scales (from a turbulent state) leading to Large Eddy Simulations.
The work is directed toward understanding and modeling compressibility effects in turbulent flows. The primary objective is to investigate how compressibility influences the basic turbulence processes such as turbulence production and dissipation, length scale modification, spectral energy transfer, etc. The second objective is to develop parameterization schemes and models to incorporate compressibility into one-point closure models and into the subgrid scale models for LES techniques. The ultimate purpose of this research is to develop compressible turbulence models which are capable of handling the hypersonic regime. Specifically, the emphasis will be on the model capability of predicting: (1) heat and momentum transfer in hypersonic boundary layers; and (2) mixing and growth of high Mach number shear layers, jet and wakes.
Among the new significant aspects of the present work are: (1) the treatment of the far-field boundary; (2) the use of C-grid topology, with the branch-cut singularity treated analytically; (3) evaluation of the effect of the envelope of prevailing initial states, and finally; (4) the ability to employ streakline/pathline 'visualization' to probe the unsteady features prevailing in vortex-dominated flows. The far-field boundary is placed at infinity, using appropriate grid stretching. This contributes to the accuracy of the solutions, but raised a number of important issues which needed to be resolved; this includes determining the equivalent time-dependent circulation for the pitching airfoil. A secondary counter-clockwise vortex erupts from within the boundary layer and immediately pinches off the energetic leading-edge shear layer which then, through hydrodynamic instability, rolls up into the dynamic stall vortex. The streakline/pathline visualization serves to provide information for insight into the physics of the unsteady separated flow.
The purpose of this paper is to gain a better understanding of rotor wake evolution in hover and axial flow by deriving an analytical solution for the time dependent behavior of vortex filament circulation and core size. This solution is applicable only for vortex filaments in the rotor far-wake. A primarily inviscid vortex/shear layer interaction (where the slipstream boundary is modeled as a shear layer) has been identified in this analytical treatment. This vortex/shear layer interaction results in decreasing, vortex filament circulation and core size with time. The inviscid vortex/shear layer interaction is shown, in a first-order treatment, to be of greater magnitude than viscous diffusion effects. The rate of contraction, and ultimate collapse, of the vortex filament core is found to be directly proportional to the rotor inflow velocity. This new insight into vortex filament decay promises to help reconcile several disparate observations made in the literature and will, hopefully, promote new advances in theoretical modeling of rotor wakes.
This article is part of a series of studies culminating from the multi-objective Flow Boiling and Condensation Experiment (FBCE) onboard the International Space Station, which utilized the Flow Boiling Module (FBM) for experiments between February and July 2022. This study investigates microgravity flow boiling of n-Perfluorohexane with liquid-vapor mixture (two-phase) inlet conditions to the FBM with either one or two opposite walls heated. The FBM’s channel has a rectangular cross-sectional area of 5.0 × 2.5 mm 2 and a heated length of 114.6 mm. Key parameters of interest include mass velocity (180 – 2400 kg/m 2 s), inlet quality (-0.01 – 0.87), inlet pressure (120 – 200 kPa), and heat flux (1.8 W/cm 2 to critical heat flux), and a large database is amassed. The flow is visualized via a high-speed video camera and photographs are recorded at each heating increment to assess the periodic flow patterns within the channel and the near-wall interfacial behavior. Flow patterns are complex and mainly characterized by high- and low-density fronts alternately traversing the channel to yield high- and low-density-dominant periods of boiling. At all operating conditions, high-density fronts are faster during high-density-dominant periods. At low inlet qualities, the flow is annular near the channel inlet with a central vapor core surrounded by an annular liquid layer. Each high-density front having a high liquid fraction leaves a thin liquid layer sheared onto the heated walls. Boiling occurs within the liquid layer and a vapor layer is formed next to the heated wall. Inlet quality and mass velocity most dictate the overall flow patterns followed by heating configuration, and to a much lesser extent, heat flux and inlet pressure. Heat transfer characteristics are assessed via averaged boiling curves, streamwise profiles of wall temperature and heat transfer coefficient, and parametric curves of local and averaged heat transfer coefficients. Inlet pressure has an insignificant effect on heat transfer. At similar operating conditions, both the heating configurations yield similar trends and values of heat transfer coefficient and critical heat flux (CHF, slightly higher for single-sided) even though double-sided heating adds twice the heat to the fluid and doubly raises local quality. The heat fluxes required for both onset of nucleate boiling degradation and CHF are larger at high mass velocities and low inlet qualities. For a fixed inlet quality, high mass velocities yield higher average heat transfer coefficients at both lower and higher heat fluxes, while the nucleate boiling regime at intermediate heat fluxes is unaffected. For a fixed mass velocity, higher inlet qualities yield higher and lower average heat transfer coefficients at lower and higher heat fluxes, respectively.
Report discusses issues involved in computation of growth of instabilities and vorticity in mixing (shear) layer between two parallel flows having different speeds and separated initially by flat surface. Study focusses upon degree to which these growing flow structures hasten or retard their own further growth, and upon related question of differences between flows computed according to mathematical models of S-layer and T-layer mixing. Transformation method used offers suitable tool for study of such flows.
The evolution of a small perturbation introduced periodically into a boundary layer with a moderately strong adverse pressure gradient (APG) was investigated experimentally using the hot-wire method. The magnitude of the disturbance is found to decay with the streamwise distance in the APG before undergoing rapid growth and triggering transition in a very repetitive manner. The evidence suggests that the laminar layer separates and that the transition mechanism is an inviscid shear-layer type of instability. Contours of spanwise vorticity in the plane of the disturbance show the formation of roll-ups in the initial stages. Subsequent interaction with the wall leads to the formation of large-scale vortex loops which merge into a single vortex loop further downstream.
The presented paper shows the effect of local Mach number on the turbulent disturbance spreading angle relative to the wall and on lateral disturbance spreading angles. Almost all the disturbances angles relative to the wall were determined from investigations where hot-wire contours or hot-film surveys of a 'laminar' boundary layer were obtained. Lateral disturbance spreading angles were obtained from investigations of various conditions including turbulent bursts, reported observations of transverse contamination, and observed transitional flow. It is noted that the disturbance spreading angle relative to the wall seems to remain essentially invariant with Mach number, while the lateral spreading angle decreases sharply with increasing Mach number up to about 6. The good agreement between lateral disturbance spreading angle data and data for the variation of turbulent jet spreading angle with Mach number implies that in the lateral dimension, turbulence in a boundary layer may develop essentially free of wall constraints (similar to a free shear layer).
Finite difference methods were applied to solve the parabolic Navier-Stokes equations for the flow over a finite width plate at 0 deg and 10 deg angles of attack. The methods were developed on the basis of the operator factorization concept resulting in the split of a three dimensional equation into successive two dimensional equations. Backward and centered implicit factorization schemes, were used and their results were compared. Available numerical solutions and experimental data obtained at low Reynolds number conditions were also used for comparison. The backward implicit method provides a more successful solution, which ranges from the merged layer to the strong interaction regimes. Detailed structures were revealed of the shear layer around and behind the side edge.
Calculations were made for the gaseous H2/O2 thruster for the Space Station by using the RPLUS code which employs an implicit finite volume, LUSSOR scheme to solve the Navier-Stokes equations and the species equations. The combustion processes of hydrogen and oxygen are modeled by a 9 species and 18 step reaction mechanism and the turbulence is simulated by the Baldwin-Lomax turbulence model for the thruster wall boundary layer and the modified Prandtl's mixing length model for the reacting shear layer. Results are presented for different mixture ratios and fuel film cooling percents and compared with the experimental data. The calculated performance predictions for the thruster agree well with the experimental data and the results demonstrate that the RPLUS code can be used for design and analysis of thrusters and rockets.
Modem low-pressure turbines, in general, utilize highly loaded airfoils in an effort to improve efficiency and to lower the number of airfoils needed. Typically, the airfoil boundary layers are turbulent and fully attached at takeoff conditions, whereas a substantial fraction of the boundary layers on the airfoils may be transitional at cruise conditions due to the change of density with altitude. The strong adverse pressure gradients on the suction side of these airfoils can lead to boundary-layer separation at the latter low Reynolds number conditions. Large separation bubbles, particularly those which fail to reattach, cause a significant degradation of engine efficiency. A component efficiency drop of the order 2% may occur between takeoff and cruise conditions for large commercial transport engines and could be as large as 7% for smaller engines at higher altitude. An efficient means of of separation elimination/reduction is, therefore, crucial to improved turbine design. Because the large change in the Reynolds number from takeoff to cruise leads to a distinct change in the airfoil flow physics, a separation control strategy intended for cruise conditions will need to be carefully constructed so as to incur minimum impact/penalty at takeoff. A complicating factor, but also a potential advantage in the quest for an efficient strategy, is the intricate interplay between separation and transition for the situation at hand. Volino gives a comprehensive discussion of several recent studies on transition and separation under low-pressure-turbine conditions, among them one in the present facility. Transition may begin before or after separation, depending on the Reynolds number and other flow conditions. If the transition occurs early in the boundary layer then separation may be reduced or completely eliminated. Transition in the shear layer of a separation bubble can lead to rapid reattachment. This suggests using control mechanisms to trigger and enhance early transition. Gad-el-Hak provides a review of various techniques for flow control in general and Volino discusses recent studies on separation control under low-pressure-turbine conditions utilizing passive as well as active devices. As pointed out by Volino, passive devices optimized for separation control at low Reynolds numbers tend to increase losses at high Reynolds numbers, Active devices have the attractive feature that they can be utilized only in operational regimes where they are needed and when turned off would not affect the flow. The focus in the present paper is an experimental Separation is induced on a flat plate installed in a closed-circuit wind tunnel by a shaped insert on the opposite wall. The flow conditions represent flow over the suction surface of a modem low-pressure-turbine airfoil ('Pak-B'). The Reynolds number, based on wetted plate length and nominal exit velocity, is varied from 50,000 to 300,000, covering cruise to takeoff conditions. Low (0.2%) and high (2.5%) Gee-stream turbulence intensities are set using passive grids. A spanwise-oriented phased-plasma-array actuator, fabricated on a printed circuit board, is surface- flush-mounted upstream of the separation point and can provide forcing in a wide frequency range. Static surface pressure measurements and hot-wire anemometry of the base and controlled flows are performed and indicate that the glow-discharge plasma actuator is an effective device for separation control. of active separation control using glow discharge plasma actuators.
Research has been underway for a number of years, both in the United States and abroad, to develop advanced aerospace propulsion systems for use late in this century and beyond. One program is now underway at the NASA Langley Research Research has been underway for a number of years, both in the United States and abroad, to develop advanced aerospace propulsion systems for use late in this century and beyond. One program is now underway at the NASA Langley Research Center to develop a hydrogen-fueled supersonic combustion ramjet (scramjet) that is capable of propelling a vehicle at hypersonic speeds in the atmosphere. A part of that research has been directed toward the optimization of the scramjet combustor and, in particular, the efficiency of fuel-air mixing and reaction taking place in the engine. In the high-speed vehicle configurations currently being considered, achieving a high combustor efficiency becomes particularly difficult. With increasing combustor Mach number, the degree of fuel-air mixing that can be achieved through natural convective and diffusive processes is reduced leading to an overall decrease in combustion efficiency and thrust. Compressible shear layers and jets provide a good model for studying the physical processes occurring in high speed mixing and combustion in a scramjet. Mixing layers are characterized by large scale eddies that form due to the high shear that is present between the fuel and the air streams. These eddies entrain the fuel and air into the mixing region between the fluids leading to increased surface area and locally steep concentration gradients. Molecular diffusion then occurs across the strained interfaces. In an early study of high-speed mixing, Brown and Roshko show that the spreading rate of a supersonic mixing layer decreases with increasing Mach number, exhibiting a factor of three decrease in spread rate as compared with an incompressible mixing layer with the same density ratio. They conclude that the reduced spread rate is primarily due to compressibility. Papamoschou and Roshko also observe that the spreading rate of compressible mixing layers is significantly reduced over that of incompressible layers. To characterize the structure of the flow quantitatively, they define a convective Mach number. The reduction in mixing layer spreading rate is shown in these experiments to correlate well with increasing convective Mach number.
Thermal convection in horizontal fluid layer heated from below, with zero shear boundaries above and below, noting critical Rayleigh number measured value
A modification of Prandtl's mixing-length model is presented which takes into account the effects of compressibility on turbulence for high speed flows. A parameter is introduced into the turbulent transport formula which acts like an effective turbulent Schmidt number for mixtures of gases or a turbulent Prandtl number for a homogeneous gas. Results presented for such cases as high Mach number turbulent boundary layer flows over a flat surface, tangential slot injection problems, and shock/turbulent shear-layer and boundary-layer interactions agree well with experimental data.