Large-Eddy Simulation of Flame-Turbulence Interactions in a Shear Coaxial Injector

Abstract

The turbulent combustion inside the main chamber of a liquid rocket engine is difficult to numerically model because of the wide range of scales involved, the complex thermodynamics at elevated pressure, and the influence of heat transfer. Here, a single-element gaseous-hydrogen-gaseous-oxygen shear-coaxial injector is studied. This case involves complex physical processes that are simulated using a large-eddy simulation approach. In particular, modeling of the nonpremixed turbulent combustion and its impact on the wall heat flux is addressed. This large-eddy simulation of the compressible multispecies Navier-Stokes equations is solved using a hybrid central-upwind scheme with a thermally perfect formulation. To keep the computational cost reasonable, priority was given to the near-field resolution over the near-wall resolution. The modeled flame anchoring and dynamics provide a satisfactory estimate of the wall heat flux. The unsteady and three-dimensional features of the flow are discussed in length, and the implications of the unique features of this shear-coaxial GH 2 -GO 2 injector for turbulent combustion modeling are analyzed.Nomenclature C , C = model coefficients for the subgrid closure D = mass diffusivity, m 2 s 1 e T = total energy (internal kinetic) per unit mass, J kg 1 H sgs = subgrid enthalpy flux, J m 2 h k = partial massic enthalpy of species k, J kg 1 J k = mass diffusion flux, kg m 2 k sgs = turbulent kinetic energy, m 2 s 2 N S = number of species considered P sgs = production of turbulent kinetic energy, kg m 1 s 3 p = pressure, pa Q IK = heat flux in the Irving-Kirkwood form, J m 2 Q sgs k = subgrid enthalpy flux due to diffusion of species k, J m 2 R = gas constant, J kg 1 K 1 T = temperature, k t = time, s u = velocity, m s 1 V k = diffusion velocity of species k, m s 1 x i = Cartesian coordinate, m Y k = mass fraction of species k = grid filter size, m sgs = subgrid dissipation of turbulent kinetic energy, kg m 1 s 3 sgs k = subgrid species diffusive flux, kg m 2 = thermal conductivity, W m 1 K 1 = dynamic viscosity, Pa s = density, kg m 3 sgs = subgrid heat flux, J m 2 ij = stress tensor Subscripts i, j = Cartesian direction k = species k property m = mixture property Superscripts x = spatially filtered quantitỹ x = Favre-filtered quantity x sgs = subgrid quantity

Stochastic modelling of transverse wave instability in a liquid-propellant rocket engine

“…At this time, large-eddy simulations exist for simple configurations. There are several single injector studies (Oefelein & Yang 1998;Oefelein 2006;Tucker et al 2007;Yang, Cuoco & Oschwald 2007;Tucker et al 2008;Masquelet et al 2009;Masquelet & Menon 2010;Schmitt et al 2010Schmitt et al , 2011Guézennec, Masquelet & Menon 2012), but systematic large-eddy simulation of a multi-injector system is still to be demonstrated.…”

Studies of shock/turbulent shear layer interaction using Large-Eddy Simulation

“…The performance of the implementation on parallel clusters depends on the domain decomposition, and the amount of switching between numerical schemes inside a given domain. The implementation of both the upwind and central schemes independently have been found to scale very well up to 1024 processors, on multiple architectures [38].…”

Large-Eddy simulation of a reactive shear coaxial injector configuration

“…Assuming no interaction between chemistry and turbulence leads to the laminar chemistry approach, which has been used in the past on supercritical conditions with success. 13 This approach requires proper temporal integration of the reaction rates. However, the time scale characteristic of the system of sti↵ di↵erential equations is much smaller than the fluid mechanics time-step.…”