CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   Optimal mesh for moving boundary problem (http://www.cfd-online.com/Forums/cfx/23843-optimal-mesh-moving-boundary-problem.html)

Luk_Fiz April 11, 2007 13:59

Optimal mesh for moving boundary problem
 
Hello all, I simulate very simple arrangement consisted on box (one side inflow, opposite outflow). Through the centerline of the box, perpendiculary to the flow direction there is a cylinder that slowly puts into the box and stops the flow. The problem is: I meshed the arrangement with tetras and notices negative volume elements with certain position of the cylinder no matter about timesteps and other. I want to discuss-are there any limitations of CFX 10 in mesh characterizations for moving walls? If so how to mesh more complicated arrangement f.e. cylinder of the IC engine with the nozzles and valves??

A second: is that true that in CFX 10 other than "wall" BC's cannot be moving walls?

With best regards, Luk

Glenn Horrocks April 12, 2007 02:48

Re: Optimal mesh for moving boundary problem
 
Hi,

The moving mesh in CFX can stretch meshes reliably but squashing a mesh up can lead to negative volume elements. Try to start the simulation in the squashed up position if possible.

Glenn Horrocks

Luk April 12, 2007 03:04

Re: Optimal mesh for moving boundary problem
 
I guess that the best possible solution is to use hexas? But what with complicated geometry and inflation layer? I'll try to start steady state from the moment in which error appears.

Luk

Glenn Horrocks April 12, 2007 18:59

Re: Optimal mesh for moving boundary problem
 
Hi,

Hexas work better as high aspect ratio elements so that can help but you can still get negative volume elements with hexas so it will not stop the problem.

There are also some controls where you can weight the smoothing function and this can help, but no guarantees.

Glenn

Luk April 13, 2007 02:30

Re: Optimal mesh for moving boundary problem
 
You mean mesh stiffnes?

Thanks for help. Luk

Glenn Horrocks April 14, 2007 06:21

Re: Optimal mesh for moving boundary problem
 
Hi,

Yes, that's correct. I have not found it to be very useful but it might help.

Glenn Horrocks


All times are GMT -4. The time now is 01:38.