CFX-Post results of Moving Mesh.
 

Hello everyone.

I have been reading the tutorials for moving mesh on these two links :http://www.edr.se/blogg/blogg/ansys_...cfx_re_meshing
http://www.edr.no/blogg/ansys_blogge...nch_cfx_remesh

Now I finally want to use the workbench approach since I want to solve a problem which has a rotating and translating mesh. However, for now, I tried to implement a simple translation motion combing the techniques used from both tutorials.

In my project, I have a created a box domain which encloses a sphere. This sphere moves in a certain direction across the fluid. The problem I'm facing is that in CFD Post, I have the entire domain moving, instead of just the sphere.
Here is the animation for the result produced using CFD-Post.https://dl.dropboxusercontent.com/u/.../test3_003.wmv

I have no idea what is causing this and I've tried to search for a solution for over a week, on my own as well as on these forums.

I've attached a few images to show my setup. Kindly help me through this.

https://dl.dropboxusercontent.com/u/...ectForMesh.jpg
https://dl.dropboxusercontent.com/u/...ignModeler.jpg
https://dl.dropboxusercontent.com/u/...FX/WB_Mesh.jpg
https://dl.dropboxusercontent.com/u/...re_Project.jpg
https://dl.dropboxusercontent.com/u/...nsientFlow.jpg
https://dl.dropboxusercontent.com/u/...re_Ball_BC.jpg
https://dl.dropboxusercontent.com/u/...penings_BC.jpg
https://dl.dropboxusercontent.com/u/...deWalls_BC.jpg
https://dl.dropboxusercontent.com/u/...xpressions.jpghttps://dl.dropboxusercontent.com/u/...hingConfig.jpg

You have the outer boundaries set as "unspecified" motion. You need to define them as zero motion.

I've tried that already with Outer Boundaries set as 'Stationary' and the 'Free Slip' and 'No Slip' settings but each time I've got an error stating that an element with negative element volume has been detected and the solver execution terminates, inspite of the remeshing procedure taking place according to the .out file. I've tried decreasing the timesteps from 0.01s to 0.001s as well, but still get the same error.

If you want the outer boundaries to stay still then you have to set them to stationary (obviously). If you leave them as unspecified then they can move around - which is exactly what is happening.

The error of the negative volume element needs to be fixed properly, not ignored. It means you need to include some remeshing steps, improve the mesh smoothing algorithm or use another approach for the body.

Can you model the sphere by an immersed solid? This does not suffer from the mesh quality and folding issues. Much easier if it is applicable.

I'm sorry to retake this thread, but does anyone have the tutorials for the moving mesh? I tried to click on them and they are not longer available...

If anyone has it, can you send me those tutorials to josed@sadpe.es?

Thank you in advance