[sammyo]

Hi all,

I have been working on this problem for some time but have exhausted all possible solutions that I could think of!

Basically, I have a blunt trailing-edged NACA 0012 aerofoil (cut at 66% chord) and am running some simple initial 2D simulations on it. However, I am finding that the residuals all drop well to around 10e-4 but then rise again before settling at a value far too high to consider the solution solved.

When they begin to rise, if I view the scalar velocity scene as the scene is being solved it almost appears that there is vortex shedding off of the trailing edge! I know that it is not actually shedding as this is simply a frame-by-frame look at the solution trying to solve, but thats the only way that I can think to describe it!

My Re is far too high (3x10^6) for shedding to occur so the solution (I thought) should be steady.

I'm using low y+ treatment, with values of <1 across the entire aerofoil surface and around 40 prism layers to encompass the boundary layer.

The "shedding" problem occurs at all angles of attack right up to stall...

If anyone has any ideas on how this simulation could be stabilised it would be highly appreciated!

Thanks in advance

[ping]

most flow is unsteady in nature and by using rans etc we are trying to force a steady solution in many cases dampening the smaller eddies etc
your mesh might be good enough to be capturing that finer detail and so need to convert to unsteady to solve the model properly but averaging the solution you have could be fine