[fated-oracle]

Hi!

I'm a student, quite new to Ansys CFX. I'm supposed to simulate a wedge dropping onto the calm water surface. It is a transient 2-D simulation (hence I used symmetry planes), with translation in the y-axis only. The time domain is limited to 0.75s in 0.002 time steps, as the impact is expected after about 0.010s from the start of the simulation and the impact speeds are believed to be high. In order to simulate the wedge falling onto the surface I used the beta feature rigid body in Ansys CFX Pre v12 and subjected it to both the gravity force and an initial downward velocity. I used an internal subdomain to allow the mesh deformation around the body as it falls. However, after only 0.01s from the start of the simulation an error occurs. It has to do with the mesh deformation and some cells becoming too small, as it says that too skew a mesh cell has been formed next to the lower wall boundary.
Initially I even set a subdomain that would move with the rigid body, but this would move along the y-axis (the direction of the motion) and so I dropped the idea.
What's worse is I think the equations I used to set up the simulation for the two fluids must be wrong, as I get a mixture of water and air although I used the bouyant model as well.
I will include the ccl file I used. I know I haven't been too clear and I'm sorry, but if anyone's willing to help I have some video files of my simulations as well. They have been mostly based on the buoy tutorial.

[ghorrocks]

I would strongly recommend you upgrade to V13. It has major enhancements to the rigid body solvers and they are no longer beta featers. Frankly you are wasting your time doing this sort of work in V12.

And I would make sure your free surface model is working properly before you add the rigid body component. Do not start off with everything activated, then you have no idea where the problem is.

[fated-oracle]

Thank you very much for your reply. Unluckily, I cannot upgrade my version of Ansys, as it is provided by my university. I'll make do with it. I've now changed the mesh a bit, using a single domain and larger inflation margins, and corrected my step functions. Now everything's working fine as I can see from the keyframes simulations. However, I need to validate my data. I was wondering if there a way to monitor the pressure at one point only, which however moves with the deforming mesh (hence no fixed coordinates can be specified). In this way I can manage to produce a graph of pressure against time at a specific location.

[belgacem]

Hi!
Friends I have the same problem. I'm supposed to simulate wave generated by sliding mass. I have considered a wedge falling on inclined slope. It is a transient 2-D simulation (hence I used symmetry planes), with translation in the x and y-axis. In order to simulate the water wave generated by the falling wedge under the gravity, I used the beta feature rigid body in Ansys CFX Pre v12 .

[ghorrocks]

What is your problem?

Again, thsi sort of stuff is far easier in the current version (now V14). Things like immersed solids and rigid body solvers make this sort of stuff much more accurate, and simpler to set up.

[belgacem]

hi friend
thank you for replying to my question . i am having some difficulties in the generation of mesh of this case. i don't have the correct version of ANSYS V14 . could you give a tutorial of this kind of stuff. precisely,i want an exemple of meshing using software Pointwise and how i can create the subdomain and boundary condition. thanks

[ghorrocks]

The tutorials provided with CFX show this sort of stuff. I do not have time to write tutorials for you.

[belgacem]

But i have the time to you say Thanks. I apperciate your help