I am trying to model the thermal stress of the part shown below. I first made a thermal model to approximate the thermal profile. Since the walls are very thin they were meshed as a shell. I then made a static model using the imported thermal results as the external load. The results are not what I would expect. There are a lot of splotches, but the mesh is pretty fine and I don't see the shear stress between the walls and end plate which I would expect. Is this approach valid for modeling the shear stress of the thin walls, or is there another way I could predict the thermal stress.