[jordan]

I am currently involved in a research project investigating the effects of repairs on nozzle guide vanes in gas turbine engines. We are performing experiments in our transonic cascade and also trying to simulate the flow with CFX. The cascade is a linear cascade of blades so I am trying to build a two dimensional, periodic hex mesh with ICEM. The flow is transonic and we have schlieren images that demonstrate the presence of a strong shock in the flow just after the trailing edge, this shock of course interacts with the vane above where it strikes the suction surface at about mid chord.

As this is my first serious attempt at meshing such a domain, I am hoping to receive some feedback regarding my meshing approach and perhaps some pointers for improving my mesh. I have managed to get good results with the blocking shown; however, I am forced to use a mesh with a very fine density in regions that I know I don't need it because of the periodicity. Does anybody know a way around this? I also need resolution normal to the blade surface in the region near the shock, but I don't need it through the entire domain. Increasing the resolution in this direction results in a much larger number of nodes than I need.

The images I have attached show my blocking and the geometry. The inlet and outlet are both parallel to the pitch line of the translationally periodic geometry. The airfoil geometry shown is simply representative, I am not able to share the actual geometry of the vane.

I look forward to any feedback that anyone might have. Thanks.