CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM (http://www.cfd-online.com/Forums/openfoam/)
-   -   Fixed mesh method for rigid body motion (http://www.cfd-online.com/Forums/openfoam/60715-fixed-mesh-method-rigid-body-motion.html)

philippose January 10, 2009 15:50

Hello, A Good Evening to ev
 
Hello,

A Good Evening to everyone :-)! Hope its not overly cold! Havent seen such temperatures in a long time now!

Anyway, I was looking through the net for methods of treating rigid body motion for CFD, and came across some interesting papers at the website of the CFD software "Flow-3D", by Flow Science (www.flow3d.com)

They call it "General Moving Objects (GMO)" and the "FAVOR" method (which seems to be trademarked by them).

In this approach, a fixed mesh is used, in which the rigid bodies are defined via a combination of Volume Fractions and Area Fractions (which looks very similar to the VOF method) in order to simulate 6-dof motion coupled with fluid flow.

Motion then becomes a variation of the volume and area fraction fields (as a function of time), rather than a physical deformation / modification of the mesh itself.

Has anyone used a similar approach in OpenFOAM, and what would the main disadvantages of this method be? At the face of it, it seems to be "the" solution to large scale motion, without having to worry about mesh quality...remeshing...etc...etc...

It would be great if someone could shed some light on this concept...

As a passing question.... any idea why Flow-3D has not made it bigger? Its been around a while, and their approaches seem to be very novel and interesting....!

Have a great weekend!

Philippose

nico765 January 12, 2009 05:57

Hello, I'd say that
 
Hello,



I'd say that Flow-3D is great at single/multiple body (large) motion. It's probably less accurate at getting good values of resistance (see
Validation of the CFD code Flow-3D for the free surface flow around the ships).
I've tried to raise the question of the moving mesh approach (vs a deforming mesh) in this thread (about a 2D wedge). This approach can and has been successfully used on other cases. CD-Adapco, in their last versions of Star-CCM, have implemented what they called DFBI (dynamic fluid body interaction). Mesh can be body-fitted (could force calculation) and handles well large motions.

In OpenFoam, the only missing parts (that I can see) are the handling of boundary conditions when handling motion. (MRF, 6DOF and VOF solvers are already available)

Regards,


All times are GMT -4. The time now is 17:22.