[Yan77]

Dear everyone,

i calculated a moving piston inside the cylinder with a small gap using Overset Meshes of Star-CCM+ (compressible flow). I compared the results with theoretic estimations. Without a gap (applying the zero-gap-interface) the pressure and temperature rise is correct. However as soon as a gap allows backflow unrealistic high velocities occur in the gap (>Ma=1) and the pressure and temperature rise is much too small. To get a closer look on the flow in the gap I tried calculating the couette flow with Overset Meshes between two cylinders and two cubics (one fixed one moving). The shear stress is not transmitted (not constant) between the walls and the velocity profile is far away from the (linear) analytical solution. It seems like there is a wrong wall treatment/friction. Has someone experiences with this problem?

Thanks a lot