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.
Similarity laws for jet noise, model line emitter, near field pressure covariances and spectra, phase coherence of near pressure fields, and nonlinear coupling in turbulence
The detailed turbulence structure of a weakly ionized, continuum plasma is analyzed. This analysis, when extended to high Reynolds number turbulent boundary layers, shows that the ratio of the electron to ion saturation currents decreases by an order of magnitude as compared to the laminar case.
The simplifed statistical theory developed previously is employed to analyze the equilibrium and near-equilibrium combustion of initially unmixed reactants. It is found that the flame zone in the limit of large Damkohler numbers is very thick and is of the order of the local integral scale of turbulence. This is in contrast to the existing phenomenological theories which predict the infinitesimally thin flame sheet, in the same limit, as it is with the laminar diffusion flame. Qualitative agreements with the available experimental results are shown. It is found that singularities exist at the edges of the flame which are removed as Damkohler number is reduced. Also, it is found that the heat transfer may take place against the local mean temperature gradient in certain regions within the flame.
Studies are described in which the effort was essentially directed to classes of problems where the phenomenologically interpreted effective transport coefficients could be absorbed by, and subsequently extracted from (by comparison with experimental data), appropriate coordinate transformations. The transformed system of differential equations could then be solved without further specifications or assumptions by numerical integration procedures. An attempt was made to delineate different regimes for which specific eddy viscosity models could be formulated. In particular, this would account for the carryover of turbulence from attached boundary layers, the transitory adjustment, and the asymptotic behavior of initially disturbed mixing regions. Such models were subsequently used in seeking solutions for the prescribed two-dimensional test cases, yielding a better insight into overall aspects of the exchange mechanisms.
Work during the period of this report has been in three areas: (1) pressure transducer error analysis, (2) fluctuating velocity and pressure measurements in the NASA Lewis 6-inch diameter quiet jet facility, and (3) measurement analysis. A theory was developed and experimentally verified to quantify the pressure transducer velocity interference error. The theory and supporting experimental evidence show that the errors are a function of the velocity field's turbulent structure. It is shown that near the mixing layer center the errors are negligible. Turbulent velocity and pressure measurements were made in the NASA Lewis quiet jet facility. Some preliminary results are included.
The turbulence structure of weakly ionized, continuum plasma is analyzed with emphasis on the turbulent fluctuation and the transport properties of the ionized species. The mean gradients of the velocity and ionized-species concentrations are considered to be uniform. It is assumed that the turbulence Reynolds number is small and that the turbulence field is homogeneous in order to close the dynamic equations of turbulence. It was found that the ratio of the Debye shielding length to the local integral scale of turbulence must be sufficiently small in order for the detailed quasi-neutrality to prevail. The condition of the detailed quasi-neutrality is necessary, in addition to the equal mean concentrations of electrons and ions, in order for the turbulence properties involving the electrons to be the same as those involving the ions.
A closed-form solution for the sound radiation from multipole sources imbedded in an infinite cylindrical jet with an arbitrary velocity profile is obtained. It is valid in the limit where the wavelength is large compared with the jet radius. Simple formulae for the acoustic pressure field due to convected point sources are also obtained. The results show (in a simple way) how the mean flow affects the radiation pattern from the sources. For convected lateral quadrupoles it causes the exponent of the Doppler factor multiplying the far-field pressure signal to be increased from the value of 3 used by Lighthill to 5.
An uncoupled time asymptotic alternating direction implicit method for solving the Navier-Stokes equations was tested on two laminar parallel mixing flows. A constant total temperature was assumed in order to eliminate the need to solve the full energy equation; consequently, static temperature was evaluated by using algebraic relationship. For the mixing of two supersonic streams at a Reynolds number of 1,000, convergent solutions were obtained for a time step 5 times the maximum allowable size for an explicit method. The solution diverged for a time step 10 times the explicit limit. Improved convergence was obtained when upwind differencing was used for convective terms. Larger time steps were not possible with either upwind differencing or the diagonally dominant scheme. Artificial viscosity was added to the continuity equation in order to eliminate divergence for the mixing of a subsonic stream with a supersonic stream at a Reynolds number of 1,000.
A previous analysis of the acoustic radiation from multipole sources is extended to include additional components of the dipole and quadrupole sources. It is found that, unlike the components of the sources considered in the previous paper, the exponent of the Doppler factor now depends on the location of the sources within the jet.
The principle of local similarity, which has been used to model the two-dimensional boundary layers in the oceanic upper mantle, permits calculation of the temperature, velocity, and stress fields with essentially analytic techniques. Finite difference numerical methods are hard pressed to resolve the detail required by the large variation of viscosity between the lithosphere and the asthenosphere. In this paper the local similarity approximation has been justified by quantitatively evaluating the effect of nonsimilarity due to viscous heating, nonlinear temperature- and pressure-dependent rheology, buoyancy, adiabatic cooling, etc. Nonsimilar effects produce only small modifications of the locally similar boundary layers; important geophysical observables such as surface heat flux and ocean floor topography are given to better than 10 percent by the locally similar solution. A posteriori evaluations of the terms neglected in the boundary layer simplification of the complete equations have been conducted on the locally similar temperature and velocity profiles close to the spreading ridge. The boundary layer models are valid to depths of 100 km at 3 m.y. and 10 km at 0.3 m.y.
The interactions between large-scale structure and fine-grained turbulence in a spatially developing mixing layer are investigated. Both a conditioned average and a time average are employed to separate the nonrandom and random fluctuations in the kinetic energy treatment of the interactions. Shape assumptions applying to the interaction integrals are adopted to obtain a solution to the closure problem. Energy-transfer mechanisms among the mean flow, the fine-grained turbulence and the large-scale structure are discussed. The magnitude of the large-scale structure and its streamwise lifetime both prove to be important in accounting for fine-grained turbulence enhancement.
Stereoscopic photography was used to record the three-dimensional motions of fluid elements along a flat-plate turbulent boundary layer flow. The outer region of the boundary layer was dominated by the formation and convection of transverse vortices, formed at the high-speed front between low- and high-speed fluid elements. Vortical motions observed in the wall region consisted of (1) motions of single particles around part of a circle, (2) streamwise vortices, and (3) transverse vortices.
The full time-dependent two dimensional Navier-Stokes equations are solved using spectral methods to achieve high spatial accuracy and high order time-splitting methods together with conformal mapping methods to allow simulation of flow over steep waves, including general turbulent Reynolds stresses. The programs are used to study the properties of laminar and turbulent flow over a flat plate and in various wavy geometries. Results are compared with both available theory and experiment. For laminar flow, good agreement is achieved. For turbulent flow, the agreement is less satisfactory and may be explained by the simplified turbulent modelling use in the calculations.
The results of a numerical computation of the interactions between the horizontally periodic monochromatic component of a large-scale coherent structure and the fine-grained turbulence in a mixing layer are presented. In the numerical calculations, the appropriate dependent variable is one which comprises both the mean and the large-scale coherent structure. The dynamical equations obtained for such a total coherent structure quantity are identical to the unsteady equations for the mean quantities in the Reynolds sense, except that the fine-grained turbulent stresses are interpreted as being conditionally averaged. It is shown how three dimensional fine-grained turbulence can be produced indirectly from the two dimensional large-scale structure via the isotropizing process of the approximated pressure-strain correlation.
The conceptual foundation underlying Large Eddy Simulation (LES) is summarized, and the numerical methods developed for simulation of the time-developing turbulent mixing layer and turbulent plane Poiseuille flow are discussed. Computational results show that the average Reynolds stress profile nearly attains the equilibrium shape which balances the downstream mean pressure gradient in the regions away from the walls. In the vicinity of the walls, viscous stresses are shown to be significant; together with the Reynolds stresses, these stresses balance the mean pressure gradient. It is stressed that the subgrid scale contribution to the total Reynolds stress is significant only in the vicinity of the walls. The continued development of LES is urged.
Developments in three dimensional, time dependent numerical simulation of turbulent flows bounded by a wall are reviewed. Both direct and large eddy simulation techniques are considered within the same computational framework. The computational spatial grid requirements as dictated by the known structure of turbulent boundary layers are presented. The numerical methods currently in use are reviewed and some of the features of these algorithms, including spatial differencing and accuracy, time advancement, and data management are discussed. A selection of the results of the recent calculations of turbulent channel flow, including the effects of system rotation and transpiration on the flow are included.
The two-dimensional steady and time dependent properties of plane Poiseuille and plane Couette flows are analyzed using iterative techniques and full numerical simulation of the Navier-Stokes equations. It is shown that the finite-amplitude two-dimensional states investigated are strongly unstable to very small three-dimensional perturbations. It is also shown, through full numerical simulation, that this explosive secondary instability can explain the subcritical transitions that occur in real flows. Finally, it is shown that the three-dimensional instability can be analyzed by a linear stability analysis of a two-dimensional flow consisting of the basic parallel flow and a steady (or quasi-steady) finite-amplitude two-dimensional cellular motion.
Numerical simulations of wakes of axisymmetric bodies and of turbulent mixing layers are reported. The flows were assumed to be statistically homogeneous in the mean flow direction, in concert with experimental data and the self-similarity theorem. The nonlinear Navier-Stokes equations were solved by a pseudo-spectral numerical method using a 32 x 32 x 33 point grid and an algorithm for fast Fourier transforms and inverse transforms. Leapfrog time differencing was employed on nonlinear terms and time differencing on viscous terms. Towed wakes and wakes behind a self-propelled body were simulated, showing that the towed wakes exhibited a proper temporal behavior after an initial period of adjustment, including the development of a kurtosis near the wake edge, which is experimentally verifiable. The mixing-layer simulation displayed the laboratory demonstrated presence of large scale features such as vortex cores, while the lateral coherence was weak.