STAR CCM+ Adjoint convergence
 

I want to use adjoint on an external aero case with about 75mio cells, coupled k-w and Reynolds of 10^5. My costfunction is drag. As the residual plot show there isn't any convergence or even any real change after the first iteration running the adjoint.

Primal sultion is well converged (apart from SDR which is a little high).

I already swithed to flexible GMRES and changed courant but that doesn't help. Does sb. have an ideas or can give me some pointers?


Attachment 84770