[lost.identity]

Hi,

I've been stuck on this problem for a while. I'm carrying out turbulent flame calculations, therefore I use the compressible versions of the turbulent kinetic energy equation (see Wilcox book on CFD modelling for example).

The problem is that the flame is thin, which leads to large changes in flow properties across a small length. Hence, this leads a large pressure gradient across the flame. The pressure work term in the turbulent K.E equation is given by



There are further modelling involved for but I want to focus on the pressure gradient term. When I have this large pressure gradient, the turbulent kinetic energy goes negative (unphysical) and my calculations diverge. Also this usually happens during the first iteration. I'm wondering if there are any remedies for such behaviour?

Making the gradient less steeper solves the problem, but my calculations need to have this large gradient. I've tried increasing the number of grid points considerably, which didn't help.

Additional information, I'm using the pressure-correction method of Patankar as the pressure solver.

Many thanks