Increased interest in ramjet propulsion systems with higher performance requirements and tighter constraints on system size and weight has lead to the need for improved techniques for analyzing and designing such systems A computer program has been developed to analyze the turbulent reacting flow field in a ramjet dump combustor configuration The program solves the axisymmetric Navier-Stokes and species equations throughout the engine diffuser and combustor providing a unified analysis of the complete engine flow field, including flow separation, fuel-air mixing, and preliminary results with chemical reaction Details of the program development are given, along with a comparison of program results with data from a dump combustor simulation experiment, to allow assessment of the flow field modeling that is employed

Numerical study of a ramjet dump combustor flowfield

A two-component laser Doppler anemometer (LDA) was used to measure mean and turbulent velocities in an axisymmetric dump combustor. Data were acquired for the axial and tangential components and provide a comparison between combusting and isothermal flows. Inlet profiles were carefully documented to provide a suitable boundary condition for future comparison to predictive models. The results show significant differences between the reacting and nonreacting flows. One large-scale observable effect was the difference between the recirculation regions for each case. Combustion decreased the length of the region by approximately 50%, while increasing the maximum velocities. This made for a more compact, but stronger, recirculation region. The reduction in recirculation zone size is recommended as a benchmark criterion in modeling studies. The presence of reaction caused higher turbulent fluctuations near the expansion, and lower fluctuations in downstream regions. The turbulence was observed to be nonisotropic in both the reacting and nonreacting cases.

Velocity characteristics of reacting and nonreacting flows in a dump combustor

A ducted, subsonic, hydrogen/air turbulent jet mixing flowfield was experimentally investigated, both with and without combustion. The geometric configuration approximated a sudden expansion or "dump" combustor with a central air jet surrounded by a low-velocity hydrogen stream at an overall equivalence ratio of 0.12. The ratio of the duct to inner nozzle diameter was 2.5. Radial distributions of mean axial and radial velocity, axial and radial turbulent intensity, velocity cross correlation, gas composition, static temperature, and total pressure, as well as axial distributions of wall static pressure, were obtained for axial stations 0-6 duct diameters from the combustor entrance. The data indicate that mixing is slower in the chemically reactive flowfield than in the nonreactive flowfield, and that the presence of combustion has a significant effect on the size and location of the recirculation zone within the mixing duct.

Measurements of reactive recirculating jet mixing in a combustor

Numerical study of reactive ramjet dump combustor flowfields with a pressure based CFD method

The interactions between turbulence and combustion are numerically examined. The effects of the combustion on the turbulence are studied by use of density-weighted (Favre) and density-unweighted (Reynolds) formulisms. The effects of the turbulence on the combustion are studied by use of the presumed probability-density-function (pdf) method, and three different forms of pdf: double β function, ß function and clipped Gausian function are examined. Comparison results show that the choice of ensemble-averaging methods has greater influence on model simulation than the choice of presumed pdf forms. Nomenclature ß beta function Γ diffusion coefficient a,b parameters in ß function « delta function Α,Β constants in clipped Gaussian e dissipation rate of turbulent kinetic function energy Cj^Cgj turbulence-combustion constants η maximum probability D diameter μ viscosity f mixture fraction ξ variance of distribution F/A fuel/air ratio Ρ density g variance of mixture fraction σ turbulent constant correlated with G production term diffusion coefficient h enthalpy Φ dependent variable Η Heaviside step function Lc combustor length Superscripts k turbulent kinetic energy Ρ pressure density-unweighted mean Ρ probability density function — density-weighted mean r radial coordinate Λ pseudo value S source term u axial velocity component Subscripts V radial velocity component X axial coordinate c combustor α constant in double β function g variance of mixture fraction 'To whom all correspondence should be addressed.

Examination of Interactions Between Turbulence and Combustion in Diffusion Flame

: An experimental investigation of ducted, two-stream, subsonic, turbulent jet mixing with recirculation was conducted A primary jet of air at a mass flow rate of 0580 1bm/sec and bulk velocity of 335 ft/sec was surrounded by an outer hydrogen stream at a mass flow rate of 0002 1bm/sec and bulk velocity of 3 ft/sec (overall equivalence ratio of 012) The ratio of the duct to inner nozzle diameter was 25 Radial distributions of mean axial and radial velocity, axial and radial turbulence intensity, velocity cross correlation, gas composition, static temperature, and total pressure, as well as axial distribution of wall static pressure, are presented for axial stations from zero to six duct diameters from the nozzle exit plane, both with and without chemical reactions The maximum turbulent intensities which occurred in the center of the mixing layer and within the recirculation eddy were very high, having values in excess of 20 percent of the jet exit velocity The data clearly indicate that the mixing is slower in the chemically reactive flow field than in the nonreactive flow field and that the presence of chemical reactions had a significant effect on the size and location of the recirculation zone within the mixing duct

G. D. Smith

Papers

An Experimental Investigation of Reactive, Turbulent, Recirculating Jet Mixing