SEARCH · Search NASA
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 Direct Simulation Monte Carlo (DSMC) method is applied to a 10 km/s flow at 0.1 torr. The assumptions used to model plasmas and electronic excitation with DSMC are investigated and new models for these processes are proposed. The results from the new modeling techniques are compared with results obtained using the previous models and the significant differences are discussed.
New experimental data are used to show that the direct simulation Monte Carlo (DSMC) method provides an accurate description of a low-density flow in the nozzle and the near-field expansion of a small rocket for two slightly different experimental configurations. These results verify the DSMC method in an expansion flow for the first time. From a number of different gas/surface interaction models, it is found that fully diffuse reflection gives the best agreement with experiment.
Results of a numerical study using the direct simulation Monte Carlo (DSMC) method are presented for hypersonic rarefied flow over an aeroassisted space transfer vehicle (ASTV). The emphasis of the study is the characterization of the near wake region which includes the ASTV payload. The study covered the transitional flow regime from near continuum to free molecular. Calculations show that the character of the near wake is significantly affected by the presence of the payload. Flow separation occurs when an afterbody is present throughout the transitional flow regime. In contrast, when no afterbody is present, no separation is observed until the flow approaches continuum.
Pitot pressures and flow angles are measured in the plume of a nozzle flowing nitrogen and exhausting to a vacuum. Total pressures are measured with Pitot tubes sized for specific regions of the plume and flow angles measured with a conical probe. The measurement area for total pressure extends 480 mm (16 exit diameters) downstream of the nozzle exit plane and radially to 60 mm (1.9 exit diameters) off the plume axis. The measurement area for flow angle extends to 160 mm (5 exit diameters) downstream and radially to 60 mm. The measurements are compared to results from a numerical simulation of the flow that is based on kinetic theory and uses the direct-simulation Monte Carlo (DSMC) method. Comparisons of computed results from the DSMC method with measurements of flow angle display good agreement in the far-field of the plume and improve with increasing distance from the exit plane. Pitot pressures computed from the DSMC method are in reasonably good agreement with experimental results over the entire measurement area.
The recontamination of a spacecraft surface directly exposed to the ram flux is investigated. It is assumed that the surface has been contaminated by a rocket plume contact and that, subsequent to the exposure, the contaminated surface is placed in the ram flux. A BGK-simulation was compared to both a Direct Simulation Monte Carlo (DSMC) and a Full Flow Monte Carlo (FFMC) using sensitivity analysis. Results indicate that the BGK-based simulation underpredicted the return flux contamination as compared to both the DSMC and FFMC techniques. Results from the full flow model suggest that the BGK model is inadequate to represent this physical problem.
A higher-order viscous shock-layer (HVSL) method has been developed and is used to obtain physically consistent results under varying degrees of low-density conditions for perfect-gas and nonequilibrium flows past long slender bodies. This method is implemented with higher-order body and shock slip conditions. Detailed comparisons with the direct simulation Monte Carlo (DSMC) method and Navier-Stokes (NS) calculations establish HVSL as a computationally efficient code for the low-density calculations. Further, it is shown that the deficiencies in the standard viscous shock-layer (VSL) method when predicting low-density flows lie in the governing equations and not with the slip boundary conditions as considered by earlier researchers.
The study assesses the effects of the upstream domain size and grid resolution upon flow properties and body aerodynamics computed for rarefied flows over cold blunt bodies with a direct simulation Monte Carlo (DSMC) particle method. Empirical correlations are suggested for aerodynamic coefficients for two-dimensional flows past a perpendicular flat plate. Free-stream parameters which were varied in the study include the Mach number, Knudsen number, surface temperature, and intermolecular potential. Insufficient grid resolution leads to overprediction of aerodynamic heating and forces in the DSMC method. Solution accuracy correlates well with the Reynolds number defined at the wall temperature and the stagnation mean free path relative to the cell dimension. Insufficient upstream domain size in the DSMC method leads to overprediction of heating and drag. Errors in aerodynamic coefficients correlate well with the distance ahead of the body where flow temperature reaches half of its peak value. Simulation of a hard-sphere gas is more sensitive to grid resolution, while simulation of a Maxwell gas is more sensitive to upstream domain size.
The Direct Simulation Monte Carlo (DSMC) method is applied to a radiating, hypersonic, axisymmetric flow over a blunt body in the near continuum regime. The ability of the method to predict the flowfield radiation and the radiative heating is investigated for flow over the Project Fire II configuration at 11.36 kilometers per second at an altitude of 76.42 kilometers. Two methods that differ in the manner in which they treat ionization and estimate electronic excitation are employed. The calculated results are presented and compared with both experimental data and solutions where radiation effects were not included. Differences in the results are discussed. Both methods ignore self absorption and, as a result, overpredict measured radiative heating.
Aerodynamic loads upon the Magellan spacecraft during aerobraking through the atmosphere of Venus are computed at off-design attitudes with a direct simulation Monte Carlo (DSMC) particle method. Simulated rarefied flows at nominal altitudes near 140 km and an entry speed of 8.6 km/s were compared to simulated and analytic free molecular results. Aerodynamic moments, forces, and heating for rarefied entry at all attitudes were 7-10 percent below free molecular results. All moments acted to restore the vehicle to its nominal zero-pitch, zero-yaw attitude. Suggested canting of the solar panels is an innovative configuration to assess gas-surface interaction during aerobraking. The resulting roll torques about the central body-axis as predicted in rarefied flow simulations were nearly twice that predicted for free molecular flow, although differences became less distinct for thermal accommodation coefficients well below unity. Roll torques increased dramatically with reduced accommodation coefficients employed in the simulation. In the DSMC code, periodic free-molecule boundary conditions and a coarse computational grid and body resolution served to minimize the simulation size and cost while retaining solution validity.
Comparisons between particle and continuum simulations of hypersonic near-continuum flows are presented. The particle approach employs the direct simulation Monte Carlo (DSMC) method, and the continuum approach solves the appropriate equations of fluid flow. Both simulations have thermochemistry models for air implemented including ionization. A new axisymmetric DSMC code that is efficiently vectorized is developed for this study. In this DSMC code, particular attention is paid to matching the relaxation rates employed in the continuum approach. This investigation represents a continuum of a previous study that considered thermochemical relaxation in one-dimensional shock waves of nitrogen. Comparison of the particle and continuum methods is first made for an axisymmetric blunt-body flow of air at 7 km/s. Very good agreement is obtained for the two solutions. The two techniques also compare well for a one-dimensional shock wave in air at 10 km/s. In both applications, the results are found to be sensitive to various aspects of the chemistry model employed.
The direct simulation Monte Carlo (DSMC) method is the established technique for the simulation of rarefied gas flows. In some flows of engineering interest, such as occur for aero-braking spacecraft in the upper atmosphere, DSMC can become prohibitively expensive in CPU time because some regions of the flow, particularly on the windward side of blunt bodies, become collision dominated. As an alternative to using a hybrid DSMC and continuum gas solver (Euler or Navier-Stokes solver) this work is aimed at making the particle simulation method efficient in the high density regions of the flow. A high density, infinite collision rate limit of DSMC, the Equilibrium Particle Simulation method (EPSM) was proposed some 15 years ago. EPSM is developed here for the flow of a gas consisting of many different species of molecules and is shown to be computationally efficient (compared to DSMC) for high collision rate flows. It thus offers great potential as part of a hybrid DSMC/EPSM code which could handle flows in the transition regime between rarefied gas flows and fully continuum flows. As a first step towards this goal a pure EPSM code is described. The next step of combining DSMC and EPSM is not attempted here but should be straightforward. EPSM and DSMC are applied to Taylor-Couette flow with Kn = 0.02 and 0.0133 and S(omega) = 3). Toroidal vortices develop for both methods but some differences are found, as might be expected for the given flow conditions. EPSM appears to be less sensitive to the sequence of random numbers used in the simulation than is DSMC and may also be more dissipative. The question of the origin and the magnitude of the dissipation in EPSM is addressed. It is suggested that this analysis is also relevant to DSMC when the usual accuracy requirements on the cell size and decoupling time step are relaxed in the interests of computational efficiency.
Four tasks were completed in this period and results were published in AIAA papers. First, a Boltzmann-2D code, was developed and applied to compute MSFC-A2 nozzle/plume flow field. It solved the two-dimensional Boltzmann-BGK equation using the Finite Difference Discrete Ordinate (FDDO) numerical technique. The code was validated by experimental data for one-dimensional shock structure predictions, paper 95-2056. Successful results for nozzle/plume flow simulation using the developed Boltzmann-2D code were presented at the 1995 AIAA Aerospace Science Conference, paper 95-0627. Second, a computer code solving two-dimensional Burnett equations was developed and applied to low-density nozzle flow field calculation. Results were also published at the 1994 AIAA Thermophysics Conference, paper 94-2055. Third, the developed two-dimensional Burnett code was extended to compute axisymmetric flow field inside MSFC-A2 nozzle, paper 95-2008. The computed nozzle exit conditions are used as input data for Direct Simulation Monte Carlo (DSMC) plume calculation. Fourth, a DSMC code was modified to compute the exhausted plume near the nozzle exit and in the backflow region.
Results are provided from a viscous shock layer (VSL) analysis of the reentry flowfield around the forebody of the Japanese Orbital Reentry Experiment (OREX) vehicle. This vehicle is a 50 deg. spherically blunted cone with a nose radius of 1.35 m and a base diameter of 3.4 m. Calculations are done for the OREX trajectory from 105 to 48.4 km altitude range. A 7-species chemical model is found adequate for the flowfield analysis. However, for altitudes greater than 84 km, the low density effects (such as thermal nonequilibrium and slip) are to be implemented for good agreement between the predictions and flight inferred heat-transfer rate data. Further, at altitudes lower than 84 km, a finite surface recombination probability is to be employed in place of a non-catalytic surface for better comparison between the calculations and data. VSL results are also compared with the direct simulation Monte Carlo (DSMC) predictions at high altitudes (greater than 80 km) and the electron number density data for three altitudes in the OREX trajectory. Overall, there is a good comparison between the flight data and calculated results. With the ongoing refinements in data extraction procedures, the OREX data should prove valuable for validating theoretical models employed in flowfield codes for calculation of reacting-gas flowfields.
Recent results of experimental and computational studies concerning hypersonic flows about blunted cones including their near wake are reviewed. Attention is focused on conditions where rarefaction effects are present, particularly in the wake. The experiments have been performed for a common model configuration (70 deg spherically-blunted cone) in five hypersonic facilities that encompass a significant range of rarefaction and nonequilibrium effects. Computational studies using direct simulation Monte Carlo (DSMC) and Navier-Stokes solvers have been applied to selected experiments performed in each of the facilities. In addition, computations have been made for typical flight conditions in both Earth and Mars atmospheres, hence more energetic flows than produced in the ground-based tests. Also, comparisons of DSMC calculations and forebody measurements made for the Japanese Orbital Reentry Experiment (OREX) vehicle (a 50 deg spherically-blunted cone) are presented to bridge the spectrum of ground to flight conditions.
There is a growing practical and theoretical interest in developing accurate macroscopic modelling for flows arising in chemical or physical vapor transport (VT) crystal growth experiments, including those conducted in reduced gravity environments. Rosner was the first person to point out that previously neglected rarefield gas dynamics phenomena (Stefan and bouyancy-driven flows) become rather important sources of convection. In particular, the combination of rarefaction and strong gradients of temperature (and/or concentration) tangential to the side-walls of the ampoule induces convective flows known as thermal (and concentration) 'creep' respectively. His order-of-magnitude estimates revealed that thermal creep effects can be non-negligible even at normal gravitational levels. On the macroscopic level, the bulk fluid mechanics can be adequately described by the familiar macroscopic equations as long as the boundary conditions are modified to account for the integrated effect of kinetic boundary layers adjacent to solid boundaries. Motivated by the growing importance of these phenomena, we have embarked on a series of computational studies to elucidate these fundamental creep-induced effects for a rarefied gas in simple, two-dimensional confined geometries. However, unlike previous related studies, we resort to a microscopic description of the gas, mathematically expressed by the Boltzmann integro-differential equation. We employ the direct simulation Monte Carlo (DSMC) method of Bird, the theoretical foundations and several practical applications. In the case of thermally induced flows, the no-time counter method of Bird is used, as implemented for a hard-sphere gas. The scheme has been also extended to account for realistic molecular interaction models, an extension necessary if the diffusion physics underlying concentration creep are to be captured.
The objective of this study was to modify an existing parallel particle code based on the direct simulation Monte Carlo (DSMC) method to include a Navier-Stokes (NS) calculation so that a hybrid solution could be developed. In carrying out this work, it was determined that the following five issues had to be addressed before extensive program development of a three dimensional capability was pursued: (1) find a set of one-sided kinetic fluxes that are fully compatible with the DSMC method, (2) develop a finite volume scheme to make use of these one-sided kinetic fluxes, (3) make use of the one-sided kinetic fluxes together with DSMC type boundary conditions at a material surface so that velocity slip and temperature slip arise naturally for near-continuum conditions, (4) find a suitable sampling scheme so that the values of the one-sided fluxes predicted by the NS solution at an interface between the two domains can be converted into the correct distribution of particles to be introduced into the DSMC domain, (5) carry out a suitable number of tests to confirm that the developed concepts are valid, individually and in concert for a hybrid scheme.
Current Connection Machine (CM) Fortran codes developed for the CM-2 and the CM-5 represent an important class of parallel applications. Several users have employed CM Fortran codes in production mode on the CM-2 and the CM-5 for the last five to six years, constituting a heavy investment in terms of cost and time. With Thinking Machines Corporation's decision to withdraw from the hardware business and with the decommissioning of many CM-2 and CM-5 machines, the best way to protect the substantial investment in CM Fortran codes is to port the codes to High Performance Fortran (HPF) on highly parallel systems. HPF is very similar to CM Fortran and thus represents a natural transition. Conversion issues involved in porting CM Fortran codes on the CM-5 to HPF are presented. In particular, the differences between data distribution directives and the CM Fortran Utility Routines Library, as well as the equivalent functionality in the HPF Library are discussed. Several CM Fortran codes (Cannon algorithm for matrix-matrix multiplication, Linear solver Ax=b, 1-D convolution for 2-D datasets, Laplace's Equation solver, and Direct Simulation Monte Carlo (DSMC) codes have been ported to Subset HPF on the IBM SP2 and the SGI Power Challenge. Speedup ratios versus number of processors for the Linear solver and DSMC code are presented.
Two different approaches, the direct simulation Monte Carlo (DSMC) method based on molecular gasdynamics, and a finite-volume approximation of the Navier-Stokes equations, which are based on continuum gasdynamics, are employed in the analysis of a low-density gas flow in a small converging-diverging nozzle. The fluid experiences various kinds of flow regimes including continuum, slip, transition, and free-molecular. Results from the two numerical methods are compared with Rothe's experimental data, in which density and rotational temperature variations along the centerline and at various locations inside a low-density nozzle were measured by the electron-beam fluorescence technique. The continuum approach showed good agreement with the experimental data as far as density is concerned. The results from the DSMC method showed good agreement with the experimental data, both in the density and the rotational temperature. It is also shown that the simulation parameters, such as the gas/surface interaction model, the energy exchange model between rotational and translational modes, and the viscosity-temperature exponent, have substantial effects on the results of the DSMC method.