Hi, my problem here seems simple. But it's good enough to keep me awake at nite. I'm trying to simulate a falling ball (can be any material) and analyze its terminal velocity. I'm confused here, should I write two UDFs to specify the drag and grid motion (for dynamic mesh)? Thanks.

You do not have to do that, you already have a 6 degree-of-freedom solver imbedeed into Fluent 6.2. All you have to do is activate "dynamic mesh", read and compile a udf for 6DOF (on which you will find all information needed in documentation) to specify mass and inertia of the ball and set motion type for every zone in your model.

Here is a very precious suggestion for you: encapsulate your ball in a bigger sphere and for the resulting volume assign rigid motion; inside use boundary-layer meshing combined with some prisms layers to fill the volume (you cannot fill the whole volume with B-L mesh, so leave just enough space to insert one or two more layers); the exterior mesh it with tetrahedrals (no other way) and this will be the deforming and remeshing zone.

This approach is very nice because of two reasons:

- you keep a high quality mesh near the ball

- you will not have so many problems with negative volumes because the first layer of deforming elements has bigger cells being away from the ball's wall.

And that's all you need to do!

Best whishes, Razvan

I'll give it a try. Thx.

By the way, I am using Fluent 6.1.22. Will there be a problem?

That's bad: you don't have 6DOF implemented in 6.1.22. If you have access to Fluent Users Centre, you might find a 6DOF UDF to download and use instead. Or ask for it on the forum, there must be someone who has something like that.

I can't help you more than this. Sorry.

Best wishes, Razvan

Thx.