Mesh size within the boundary layer
 

Hi everyone,

I have a question regarding the mesh size within the boundary layer. I am currently working on a high speed model (Mach 4 and up) with Reynold's numbers in the order of 1E8. I was taught that one should aim for a y+ value of 1, which gives me a cell height of the order 1E-7. When running this, the CFX solver makes round-off mistakes resulting in negative volume elements, so this is not possible with my current resources.

I have also encountered convergence issues (oscillation around a certain point) but I have convergence with a minimum cell height of 1E-3 (I could still investigate if this could decrease a bit, at 1E-4 there is no convergence anymore).

My boundary layer thickness has a minimum of about 0 and a maximum of 3.7 cm. Currently my interest is the pressure distribution over the body, however in a later stage the edge temperature and the temperature profile within the boundary layer are also of interest.

My question is: what would be an appropriate amount of cells within the boundary layer (or an appropriate cell height) in this case?

Some background about the model: It is a half-domain, 2D model of a high supersonic-hypersonic missile with a Von Karman Ogive nose cone and a simple, cylindrical body (no fins, wings, anything sticking out).