Engineering topics
Publications and source records attributed to Kline, S. J..
Two-part report reviews knowledge of coherent structures in turbulent boundary layers. Part I describes processes and status of cooperative project to summarize data from research on boundary-layer turbulence. Part II presents results of study of numerically simulated flat-plate canonical turbulent boundary layer.
A review is presented of the approach, motivation and early results of a reevaluation of the knowledge collected by the research community during thirty years of research on the structure of turbulent boundary layers. Four distinctions or criteria concerning the need to improve the practices in this field are discussed: data versus inferences, possible versus actual versus significant events, one structure versus many, and one or two flows versus the totality of data. Attention given to the known quasi-coherent structures show that three factors significantly reduce the information available from laboratory data as a basis for forming a complete model of the quasi-coherent structure in the turbulent boundary layer: (1) the necessity to ensemble average probe output with resulting loss of phase information, (2) the inability to see more than one or two of the various types of structures simultaneously, and (3) the inability to see the spatial relations between the various quasi-coherent structures resulting from (1) and (2).
Explore the source record for details and available documents.
Results of a cooperative effort to review the structure of the turbulent boundary layer are summarized. The principal concepts relevant to the study are briefly reviewed, with particular attention given to vorticity lines and vortices, vortex elements and vortex structures, and average models and realizations. The discussion covers the preliminary ordering of the importance of structure elements, the strength of vortex elements, and time evolution of structures. Finally, some areas of further research are suggested.
Preliminary results of a comprehensive study of the structural aspects of a numerically simulated number turbulent boundary layer are presented. A direct Navier-Stokes simulation of a flat-plate, zero pressure gradient boundary layer at Re0 = 670 was used. Most of the known nonrandom, coherent features of turbulent boundary layers are confirmed in the simulation, and several new aspects of their spatial character are reported. The spatial relationships between many of the various structures are described, forming the basis for a more complete kinematical picture of boundary layer physics than has been previously known. In particular, the importance of vortex structures of various forms to the generation of Reynolds shear stress is investigated.
Coherent structures are discussed as a sequence of events (identifiable motions) in the flow which convert significant amounts of mechanical energies of the mean flow stream, into turbulent fluctuations. The use of structure information in modeling is also discussed.
Computational techniques for simulating turbulent flows were explored, together with the results of experimental investigations. Particular attention was devoted to the possibility of defining a universal closure model, applicable for all turbulence situations; however, conclusions were drawn that zonal models, describing localized structures, were the most promising techniques to date. The taxonomy of turbulent flows was summarized, as were algebraic, differential, integral, and partial differential methods for numerical depiction of turbulent flows. Numerous comparisons of theoretically predicted and experimentally obtained data for wall pressure distributions, velocity profiles, turbulent kinetic energy profiles, Reynolds shear stress profiles, and flows around transonic airfoils were presented. Simplifying techniques for reducing the necessary computational time for modeling complex flowfields were surveyed, together with the industrial requirements and applications of computational fluid dynamics techniques.
The paper studies incompressible flow over a backward-facing step in order to investigate the flow characteristics in the separated shear layer, the reattachment zone, and the redeveloping boundary layer after reattachment. It is shown that turbulent intensities and shear stress reach maxima in the reattachment zone, followed by rapid decay near the surface after reattachment. In addition, it is found that downstream of reattachment, the flow returns very slowly to the structure of an ordinary turbulent boundary layer.
A procedure is described for computation of incompressible, steady, two-dimensional flows in fully stalled diffusers with plenum exit. The procedure is successful in predicting pressure distributions and patterns to the accuracy of the data. The procedure employs a zonal model; this maintains close connections between the modeling and the physics thereby providing insight into critical aspects of modeling separated flows. The procedure presented is also convenient for computing unstalled flows in passages with turbulent boundary layers for either direct or indirect design problems. Computing times are well within engineering feasibility. The concepts developed can be extended to other classes of separated flows; some of these extensions have already been completed and are referenced.
A classification according to the level of detail of description the computational method provides is probably most useful. Flow calculations can be classified into five categories: (1) correlations; (2) zonal method; (3) time-averaged equations; (4) large-eddy simulation; and (5) Navier-Stokes solution. There are methods that fall into more than one category, and there are sub-divisions of each category. A discussion of the advantages and disadvantages of each of these five categories is given.
Experiments in free turbulent flows are recommended with the primary concern placed on classical flows in order to augment understanding and for model building. Five classes of experiments dealing with classical free turbulent flows are outlined and proposed as being of particular significance for the near future. These classes include the following: (1) Experiments clarifying the effect of density variation owing to use of different gases, with and without the additional effect of density variation due to high Mach number or other effects; (2) experiments clarifying the role and importance of various parameters which determine the behavior of the near field as well as the condictions under which any of these parameters can be neglected; (3) experiments determining the cumulative effect of initial conditions in terms of distance to fully established flow; (4) experiments for cases where two layers of distinctly different initial turbulence structure flow side by side at the same mean speed; and (5) experiment using contemporary experimental techniques to study structure in free turbulent shear flows in order to compliment and support contemporary work on boundary layers.
Heat-transfer rates, velocity profiles, and temperature profiles for the turbulent incompressible flow of air over a flat plate with a constant surface temperature have been measured at Reynolds numbers up to 3.5 x lO(exp 6). The turbulent heat-transfer measurements agree well with the von Karman analogy, and the velocity profiles agree with the data of previous investigators. The temperature profiles are similar to the velocity profiles, both being adequately described by power formulas.
Heat-transfer rates and temperature profiles for the turbulent incompressible flow of air over a flat plate with a stepwise temperature distribution (unheated starting length) were measured for a variety of step positions at Reynolds numbers up to 3.5 x 10(exp 6). Comparison of the data with existing heat-transfer analyses indicates that an improved analysis is needed. An integral analysis is made that agrees very well with the data and allows a simple correction for the unheated starting length. In addition, a differential analysis is made that allows prediction of the temperature profiles from the velocity profiles, and good agreement with experimental profiles is obtained.
Superposition techniques are used to calculate the rate of heat transfer from a flat plate to a turbulent incompressible boundary layer for several cases of variable surface temperature. The predictions of a number of these calculations are compared with experimental heat-transfer rates, and good agreement is obtained. A simple computing procedure for determining the heat-transfer rates from surfaces with arbitrary wall-temperature distributions is presented and illustrated by two examples. The inverse problem of determining the temperature distribution from an arbitrarily prescribed heat flux is also treated, both experimentally and analytically.
Superposition techniques are used to calculate the rate of heat transfer from a flat plate to a turbulent incompressible boundary layer for several cases of variable surface temperature. The predictions of a number of these calculations are compared with experimental heat- transfer rates, and good agreement is obtained. A simple computing procedure for determining the heat-transfer rates from surfaces with arbitrary wall-temperature distributions is presented and illustrated by two examples. The inverse problem of determining the temperature distribution from an arbitrarily prescribed heat flux is also treated, both experimentally and analytically.
The effect of the location of transition on the heat transfer to the turbulent incompressible boundary layer is analyzed. The analysis indicates that considerably higher heat-transfer rates may occur for some distance downstream if the transition is very late. The results of a limited experimental investigation are in substantial agreement with the results of the analysis. If the extent of the transition region is known, the analysis also allows adequate prediction of heat-transfer coefficients within the transition region. The nature of this analysis is such that it should predict local shear coefficients in the transition region equally well.