CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   Rigid body (http://www.cfd-online.com/Forums/cfx/90356-rigid-body.html)

stefano rigosi July 7, 2011 13:21

Rigid body
 
Dear all,
I am dealing with floating body in the real sea state condition, which means irregular wave, high unsteadiness and wave-body interaction.
Anybody knows how CFX 13 computes the dynamic behaviour of a floating body??
Does it takes into account non-linear damping coefficient, the added mass of inertia and so on.
I would like to have some more information about this new topic, since in the Help it is not well explained.


I have got another easy question. Why when dealing with water free surface and two phases it is better to implement SST as turbulent model instead of k-e??

Thanks

ghorrocks July 7, 2011 19:02

CFX has a 6DOF solver. For details see the documentation.

What is the non-linear damping coefficient you refer to? The body is coupled to the fluid around it. There is no damping coefficients, the damping is modelled directly.

Mass - obviously, yes this is modelled.

Free surface and SST/k-e - I do not know where you got that one from. SST is generally a more widely applicable turbulence model than k-e, but that applies to single phase models as well.

stefano rigosi July 8, 2011 15:32

Dear Glenn,

Thank you for your quick answer.
Since I did the tutorial of a floating buoy in CFX13 I found that they suggest to use SST turbulence model, but they do not justify this choice.
I think, it is more robust for the free surface modelling, but I am not sure.

Moreover, I have some problem in my simulations. I used a timestep of 0,01 sec. and when I monitor the force and the torque acting on the rigid body I found very changeable values, such as +100 and -20 between two timesteps.
What can I do apart from reducing the timesteps, that leads to much longer simulations??

Thanks,

ghorrocks July 10, 2011 06:38

No, SST is recommended as the general purpose turbulence model. This recommendation is general, not specific to free surface models.

The rapidly oscillating forces suggest the 6DOF solver is not happy. Sometimes this can be helped with a smaller timestep, but there may be options in the solver to help damp oscillations. I don't know, I am not very familiar with the 6DOF solver.


All times are GMT -4. The time now is 13:48.