Navier—stokes computations of hypersonic flows

Abstract

The current interest in hypersonic flows is leading to significant effort both to develop CFD methods and also to provide experimental data for their evaluation. In our research we attempt to integrateCFD and experiments as closely as possible so much so that most of our experimental model designs are based upon preliminary flow field computations in order to identify likely regions of importance and distribute instrumentation as efficiently as possible. The experiments must also have the CFD requirements clearly in mind. In particular we consider it important to separate evaluation on the numerics (essentially the algorithm) from modelling of the physics (which includes the uncertainties of turbulence modelling) to this end our experiments include laminar studies, for both attached and separated flows, for which the physical equations are known exactly, as well as turbulent flow studies. This paper concentrates mainly on our CFD efforts and presents details of a high resolution solver for viscous flows together with their predictions for a range of problems which are the subject of our current and planned experiments.

Heat transfer measurement of turbulent spots in a hypersonic blunt-body boundary layer

“…The primary reference quantities required are laminar so that the following CFD comments, unless otherwise stated, relate specifically to the laminar simulation. The numerical procedure has been described elsewhere (Hillier, Kirk & Soltani 1995;Jackson, Hillier & Soltani 2001;Hillier et al 2003), and is formulated here as a second-order accurate 'convection-diffusion-split' axisymmetric Navier-Stokes code. The fluxes for the convective, or Euler part, are solved using the explicit generalized Riemann problem of Ben-Artzi & Falcovitz (1984) whilst the diffusive, or viscous, fluxes are evaluated by an implicit centred-differencing procedure.…”

Experimental and computational study of unsteady hypersonic cavity flows

“…Certainly 3D turbulent simulations would have been too ambitious for the range of cases that we wished to survey here. Most details of the method are available elsewhere (24,28) , so that only the main points are repeated here. It uses 'convection-diffusion' splitting, solving the convective (i.e.…”

Application of axisymmetric analog to unstructured grid for aeroheating prediction of hypersonic vehicles

“…Therefore they cannot predict the peak heating which occurs in the vicinity of boundary layer reattachment. In the case of flow separation, full Navier-Stokes equations should be solved around the vehicle (Hillier and Soltani, 1995;Mehta, 2000;Mulas et al, 2002;Soltani et al, 1993).…”