Turbulent Shear Layers In Supersonic Flow

A good understanding of turbulent compressible flows is essential to the design and operation of high-speed vehicles. Such flows occur, for example, in the external flow over the surfaces of supersonic aircraft, and in the internal flow through the engines. Our ability to predict the aerodynamic lift, drag, propulsion and maneuverability of high-speed vehicles is crucially dependent on our knowledge of turbulent shear layers, and our understanding of their behavior in the presence of shock waves and regions of changing pressure. Turbulent Shear Layers in Supersonic Flow provides a comprehensive introduction to the field, and helps provide a basis for future work in this area. Wherever possible we use the available experimental work, and the results from numerical simulations to illustrate and develop a physical understanding of turbulent compressible flows.

It was on a proposal of the late Professor Maurice Roy, member of the French Academy of Sciences, that in 1982, the General Assembly of the International Union of Theoretical and Applied Mechanics decided to sponsor a symposium on Turbulent Shear-Layer/Shock-Wave Interactions. This sympo sium might be arranged in Paris -or in its immediate vicinity-during the year 1985. Upon request of Professor Robert Legendre, member of the French Academy of Sciences, the organization of the symposium might be provided by the Office National d'Etudes et de Recherches Aerospatiales (ONERA). The request was very favorably received by Monsieur l'Ingenieur General Andre Auriol, then General Director of ONERA. The subject of interactions between shock-waves and turbulent dissipative layers is of considerable importance for many practical devices and has a wide range of engineering applications. Such phenomena occur almost inevitably in any transonic or supersonic flow and the subject has given rise to an important research effort since the advent of high speed fluid mechanics, more than forty years ago. However, with the coming of age of modern computers and the development of new sophisticated measurement techniques, considerable progress has been made in the field over the past fifteen years. The aim of the symposium was to provide an updated status of the research effort devoted to shear layer/shock-wave interactions and to present the most significant results obtained recently.

Experiments have been conducted in the GALCIT Supersonic Shear Layer Facility to investigate some aspects of high Reynolds number, turbulent, shear layer flows in both incompressible and compressible flow regimes. Experiments designed to address several issues were performed; effects of inflow conditions, freestream conditions (supersonic/subsonic flow), and compressibility, on both large scale dynamics and small scale mixing, are described. Chemically reacting and non reacting flows were investigated, the former relying on the (H2+NO)/F2 chemical system, in the fast-kinetic regime, to infer the structure and amount of molecular scale mixing through use of "flip" experiments. Both inflow conditions and compressibility are found to have significant effects on the flow. In particular, inflow conditions are "remembered" for long distances downstream, a sensitivity similar to that observed in low-dimensionality, nonlinear (chaotic) systems. A previously proposed mode selection rule for turbulent structure convection speeds, based on the presence of a lab frame subsonic freestream, was demonstrated to be incorrect. Compressibility, when decoupled from other parameters, e.g., Reynolds number, velocity and density ratios, reduces large scale entrainment and growth, but slightly enhances mixing, and changes the molecularly mixed fluid structure. This reduction in shear layer growth rate is examined and a new parameter that interprets compressibility as an energy exchange mechanism is proposed. The parameter reconciles and collapses experimentally observed growth rates.

A study of the time-averaged mean flow and the turbulence in a Mach 5 free turbulent shear layer has been performed. When the experimental data were reduced with the assumption of constant static pressure through the shear layer, the mean-velocity profile in similarity coordinates was in good agreement with the low speed velocity profile. The intensities of the velocity fluctuations were about a factor of 2 lower than the low speed measurements but with the maximum velocity fluctuations occurring in the same regions of the supersonic and low speed shear layers. A large density fluctuation was observed in the outer part of the shear layer near the boundary of the shear layer and the potential core.