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 heating and movement of the upper atmosphere at ionospheric levels in response to electric currents are discussed. Joule dissipation, generation of winds, and pressure gradients are significant factors in the energetics of the ionospheric electric currents flowing during magnetic storms and also of the Sq current system.
The linear hydrodynamic stability of a thin-liquid layer flowing along the inside wall of a vertical tube rotating about its axis in the presence of a core-gas flow is examined. The stability problem is formulated under the conditions that the liquid film is thin, the density and viscosity ratios of gas to liquid are small and the relative (axial) pressure gradient in the gas is of the same order as gravity. The resulting eigenvalue problem is first solved by a perturbation method appropriate to axisymmetric long-wave disturbances. The damped nature (to within the thin-film and other approximations made) of the nonaxisymmetric and short-wave disturbances is noted. In view of the limitations on a truncated perturbation solution when the disturbance wavenumber is not small, an initial value method using digital computer is presented. Stability characteristics of neutral, growing, and damped modes are presented showing the influences of rotation, surface tension, and the core-gas flow. Energy balance in a neutral mode is also illustrated.
Extensive numerical calculations have been carried out to determine the structure of the primitive solar nebula. The resulting structure shows a dominance of the radial pressure gradient near the center, and a dominance of the centrifugal forces throughout most of the outer region. Thermal convection and Eddington-Sweet circulation currents can be expected to cause a rapid dissipation of the disk, requiring only a few centuries. It is therefore postulated that the planets were formed in an exceedingly hot initial state due to their rapid rates of accumulation. Beyond one or two astronomical units from the center, the original interstellar grains should not be completely destroyed during the formation of the disk, and these will form the nuclei for subsequent chemical condensations and chemical accumulation. Inside this distance, nucleation of condensed materials must occur within the nebular gases.
Results of an analytical and experimental study of boundary layer flow over an aerodynamic surface rejecting heat to a cool environment. This occurs following reentry of a Space Shuttle vehicle. Analytical studies revealed that a surface to freestream temperature ratio, greater than unity tended to destabilize the boundary layer, hastening transition and separation. Therefore, heat transfer accentuated the effect of an adverse pressure gradient. Wind tunnel tests of a 0012-64 NACA airfoil showed that the stall angle was significantly reduced while drag tended to increase for freestream temperature ratios up to 2.2.
The nozzle wall boundary layer of an M-50 helium tunnel was investigated with pitot pressure, total temperature, skin friction, and wall heat transfer measurements at five stations and hot wire measurements at two stations. The results indicated that the boundary layer was turbulent with a thick viscous sublayer. Pressure gradients were observed across the boundary layer; the effect of these gradients on the equations of motion are discussed. The direct skin friction measurements were higher than expected from empirical predictions; the Reynolds analogy factors however were lower than expected. Hot wire measurements indicated mass flow fluctuations as large as 80 percent of the local mean mass flow at the edge of the viscous sublayer with a maximum value relative to the edge mass flow of about 15 percent at gamma about equal to 0.8.
The response of a rectangular panel under the excitation of a turbulent boundary layer with a zero longitudinal mean pressure gradient in a subsonic flow was studied in detail. The method of normal mode was used together with the technique of spectral analysis. Both simply supported and clamped edge conditions of a panel were considered, and the displacement power spectral density of the panel response was computed. The results for the clamped edge panel compare favorably with existing data. Charts of structural acceptance, which provide a framework for estimating the response of other rectangular panels, are presented, and the physical significances of structural acceptances discussed.
The governing equations for an unsteady turbulent boundary layer on a swept infinite cylinder, composed of a continuity equation, a pair of momentum equations and a pair of turbulent energy equations which include upstream history efforts, are solved numerically. An explicit finite difference analog to the partial differential equations is formulated and developed into a computer program. Calculations were made for a variety of unsteady flows in both two and three dimensions but primarily for two dimensional flow fields in order to first understand some of the fundamental physical aspects of unsteady turbulent boundary layers. Oscillating free stream flows without pressure gradient, oscillating retarded free stream flows and monotonically time-varying flows have all been studied for a wide frequency range. It was found that to the lowest frequency considered, the lower frequency bound being determined by economic considerations (machine time), there were significant unsteady effects on the turbulent boundary layer.
A method is presented for calculating the aerodynamic heating and shear stresses at the wall for tangent ogive noses that are slender enough to maintain an attached nose shock through that portion of flight during which heat transfer from the boundary layer to the wall is significant. The lower entropy of the attached nose shock combined with the inclusion of the streamwise pressure gradient yields a reasonable estimate of the actual flow conditions. Both laminar and turbulent boundary layers are examined and an approximation of the effects of (up to) moderate angles-of-attack is included in the analysis. The analytical method has been programmed in FORTRAN 4 for an IBM 360/91 computer.
Prandtl's basic mixing length model was used to compute 22 test cases on free turbulent shear flows. The calculations employed appropriate algebraic length scale equations and single values of mixing length constant for planar and axisymmetric flows, respectively. Good agreement with data was obtained except for flows, such as supersonic free shear layers, where large sustained sensitivity changes occur. The inability to predict the more gradual mixing in these flows is tentatively ascribed to the presence of a significant turbulence-induced transverse static pressure gradient which is neglected in conventional solution procedures. Some type of an equation for length scale development was found to be necessary for successful computation of highly nonsimilar flow regions such as jet or wake development from thick wall flows.
Calculation of free shear flows was carried out to investigate the usefulness of several concepts which were previously successfully applied to wall flows. The method belongs to the class of differential approaches. The turbulence is taken into account by the introduction of one additional partial differential equation, the transport equation for the turbulent shear stress. The structure of turbulence is modeled after Bradshaw et al. This model was used successfully in boundary layers and its applicability to other flows is demonstrated. The work reported differs substantially from that of an earlier attempt to use this approach for calculation of free flows. The most important difference is that the region around the center line is treated by invoking the interaction hypothesis (concerning the structure of turbulence in the regions separated by the velocity extrema). The compressibility effects on shear layer spreading at low and moderate Mach numbers were investigated. In the absence of detailed experiments in free flows, the evidence from boundary layers that at low Mach numbers the structure of turbulence is unaffected by the compressibility was relied on. The present model was tested over a range of self-preserving and developing flows including pressure gradients using identical empirical input. The dependence of the structure of turbulence on the spreading rate of the shear layer was established.
In order to determine the convective heat-transfer distribution for the nose region of the space shuttle entry configurations, a three-dimensional flow-field is described which may include extensive regions of separated flow. Because of the complexity of the flow field for the nose region, experimental data are needed to define the relation between the nose geometry and the resultant flow field. According to theoretical solutions of the three-dimensional boundary layer, the boundary layer separates from the leeward generator of a blunted cone at an alpha equal to the cone half-angle. Separation results from the transverse pressure gradient, i.e., the velocity derivative due to crossflow. The boundary layer limiting streamlines converge toward the singular point of sep aration. The separated region is bounded by an ordinary line of separation.
Turbulent shear-stress, eddy-viscosity, mixing-length, heat-flux and Prandtl number distributions across a hypersonic turbulent boundary layer have been determined from the 'time-averaged' conservation equations using experimental mean profile data obtained at several streamwise locations in a fully developed turbulent boundary layer with negligible pressure gradient. The eddy-viscosity, mixing-length and Prandtl number results show general agreement with previous incompressible and adiabatic compressible correlations. However, when the turbulent Prandtl number is defined using total enthalpy as opposed to static enthalpy no clear correlation of the results in the outer portion of the boundary layer could be obtained.
Sinusoidal instability of aircraft trailing vortices was induced by differentially oscillating inboard and outboard flaps on a model wing in a ship towing tank. Measured amplification rates qualitatively agree with theoretical predictions. Axial velocities in the vortex core were directed towards the wing and measured to be approximately 25% of the tow speed. Vortex breakdown was observed to occur along trailers undergoing sinusoidal instability near but ahead (towards the wing) of positions of maximum trailer separation. It is shown that axisymmetric pressure gradients are imposed along the vortex core by the other sinusoidally deformed trailer and are responsible for the observed changes in core diameter. A theoretical model predicts the short-time behavior of the vortex core and shows that the response of the vortex depends crucially on the axial velocity in the vortex.
Investigation of the nonsimilar flows which in the course of their development in two considered branches of the Falkner-Skan equation show pressure gradient parameter variations that span the range of parameter values, or 'window,' in which no similar solutions have been shown to exist by Libby and Liu (1967). The investigation results obtained are shown to support Chen and Libby's (1968) considerations on the spatial stability of some solutions of the Falkner-Skan equation.
A method is presented for calculating the aerodynamic heating and shear stresses at the wall for tangent ogive noses that are slender enough to maintain an attached nose shock through that portion of flight during which heat transfer from the boundary layer to the wall is significant. The lower entropy of the attached nose shock combined with the inclusion of the streamwise pressure gradient yields a reasonable estimate of the actual flow conditions. Both laminar and turbulent boundary layers are examined and an approximation of the effects of (up to) moderate angles-of-attack is included in the analysis. The analytical method has been programmed in FORTRAN IV for an IBM 360/91 computer.
The governing equations for an unsteady turbulent boundary layer on a swept infinite cylinder, composed of a continuity equation, a pair of momentum equations and a pair of rate equations for the shear stress, based on the turbulent kinetic-energy equation, are solved numerically. These rate equations for the shear-stress express a balance between the convection, production, dissipation and diffusion of the turbulent shear stress components. Due to the physical model of turbulence employed, this system of equations is hyperbolic. An explicit, second-order accurate, conditionally stable difference scheme for this system, together with appropriate boundary and initial conditions, is formulated and developed into a computer program. Calculations for oscillating free stream flows with no pressure gradient show significant unsteady effects on the turbulent boundary layer.
The paper presents the results of some calculations aimed at studying the characteristics of turbulent boundary layers in unsteady flow. The calculations were performed using the method of Singleton and Nash, which is based on the turbulent kinetic energy equation. Emphasis is placed on the approach to zero wall stress, and the results show the extent to which onset of this condition can be delayed by the effects of time dependence. The delay is not associated merely with alleviation of the pressure gradients, and quasi-steady methods cannot readily be used to calculate this type of flow.
Turbulent boundary-layer flows of nonreacting gases are predicted for both internal (nozzle) and external flows. Effects of favorable pressure gradients on two eddy viscosity models were studied in rocket and hypervelocity wind tunnel flows. Nozzle flows of equilibrium air with stagnation temperatures up to 10,000 K were computed. Predictions of equilibrium nitrogen flows through hypervelocity nozzles were compared with experimental data. A slender spherically blunted cone was studied at 70,000 ft altitude and 19,000 ft/sec in the earth's atmosphere. Comparisons with available experimental data showed good agreement. A computer program was developed and fully documented during this investigation for use by interested individuals.