Hello cfd online users!

I was trying to simulate flow around a rectangular wind rotor blade. I was using k-epsilon turbulence model with enhanced wall treatment. But though the solutions are converged, but the residual for moment coefficients are very high, and that too, negative.

Will someone please help me!

Thanks. -- Agni.

Have you checked under what conditions a k-e model performs satisfactorily ? To my knowledge, K-e model can have issues when you encounter adverse pressure gradient and also flow transition. I would expect that for your configuration, there would be flow separation and hence there might be a problem with k-e model.

Second thing, you never mentioned about the numerical method being employed. That can also cause some errors.

I was using simple pressure-velocity coupling with ist order upwind discretization scheme for momentum, turbulent k.e. and sp. dissipation rate.

Can I use K-omega model, since it is good for flows involving separation?

Thanks for your suggestions!

First order methods are highly dissipative. You might want to go to second order central or upwind schemes. K-omega model is fine for flows involving separation. Try to change to second order and check your results.

THANKS A LOT!