Has anyone had any hands-on experience with the ALE option in Flotran? After reading the Ansys documentation, it is not clear to me if I must specify both displacements and velocity on the moving boundary, or just displacements.

I set up a somewhat simple 2/D test problem which involves a body with slow, transient rectilinear motion in an "infinite" medium. If I've calculated everything correctly, the flow should be laminar. I applied both the displacement and velocity boundary conditions to the nodes on the surface of the moving body. While the solution appears to converge and looks somewhat "reasonable", the mesh does become highly distorted. In fact, it looks like the elements immediately adjacent to the moving body are being stretched/compressed, while the rest of the mesh looks essentially unchanged.

Any thoughts or suggestions?

Thanks.

Mark

In my opinion, you couldn't fix both displacement and velocity on the same boundary because deplacement is derivated from velocity.

Well, they obviously have to be consistent. I think that the displacement-time history is just for moving the mesh, while the velocity is the actual BC for the flow field solution. My question was that it is not at all clear what FLOTRAN actually wants you to input for the moving boundary; displacement, velocity, or both?

Mark