Numerical investigation of transverse sonic injection in a non-reacting supersonic combustor

Abstract

Efficient combustion and heat release in scramjet flows depend on effective mixing of the fuel in supersonic streams. Usually, transverse sonic injection in-stages are employed as one of the suitable means for efficient supersonic combustor design. Numerical simulations are carried out to study the mixing characteristics of staged sonic air injections in supersonic stream (M ¼ 2.07) behind a backward-facing step in scramjet combustor by solving threedimensional Navier -Stokes equations along with K -1 turbulence model with a commercial CFD software CFX-TASCFlow. Computed results of the jet penetration and spreading show very good agreement with the experimental values and the results of other computations. A good overall match has been obtained between the experimental values and the computation for various flow profiles at various axial locations in the combustor. However, the values differ in the near-field region at the injection plane. The assumed uniformity of the flow-field properties at the injection orifice and/or the inadequacy of the turbulence model considered in this study is conjectured to be the cause of the difference.

Numerical Exploration of Staged Transverse Injection into Confined Supersonic

“…Two sets of calculations were carried out for with and without transverse injection and qualitative behaviour of the flow has been analysed and the quantitative comparisons has been made for various flow profiles at different axial stations by comparing with experimental results 10 and other numerical calculations 20,21 .…”

“…Experimental condition of staged sonic injection behind a backward-facing step in rectangular duct has been explored numerically by Chakraborty 19 , et al using a Cartesian-based three-dimensional Navier Stokes solver alongwith k-e turbulence model and obtained reasonable agreement with the experimental value of injectant penetration and various flow profiles at various axial locations of the combustor. The same experimental conditions were simulated by Manna and Chakraborty 20,21 with a 3D Reynolds Averaged Navier Stokes (RANS) solver alongwith k-e turbulence model using a commercial solver and obtained very good comparisons of injectant penetration and spreading with the experimental results. Although, predicted flow profiles at various axial locations match with the experimental results, the values differ in the nearfield region.…”

N umerical Investigation of Hydrogen-fuelled Scramjet Combustor with Cavity Flame Holder

“…The details of the governing equations, thermodynamics and the discretization schemes are given in the following subsections. To find out the accuracy and the range of applications, the software has been validated for various reacting and non-reacting flows pertaining to the scramjet combustor including transverse sonic injection in a supersonic flow 12 , transverse H 2 injection in constant area duct 13 , H 2 injection from struts 14,15 and pylon injectors 16 . All these validation exercises have revealed that although the computed pressures overpredict the experimental values in the injection zone, the computational and experimental values of the flow parameters match fairly well in the divergent portion of the combustor where the major portion of thrust is produced.…”

Mixing Flow Characteristics for a Transverse Sonic Jet Injecting into a Supersonic Crossflow

“…Due to the formation of the shock-wave, and the consequent strong adverse pressure gradient, the incoming turbulent boundary layer will separate, leading to a large recirculation region. So far, majority simulations of flush-wall jet injection into crossflow use the Reynolds-averaged Navier-Stokes (RANS) turbulence modeling approaches, including predictions that use a variety of turbulence models by Uenishi et al [4], Tam et al [5], Palekare et al [6], and Manna and Chakraborty [7], among others. However, it is widely accepted that for three-dimensional complex flows, turbulence models are generally incapable to capture 'real' flow physics, and thus RANS simulations can merely provide reasonable 'mean' quantities, but not the dynamic process of unsteady flow field, the latter is in fact very important and crucial in evaluating the mixing flow such as JISCF.…”