problem with rotating mesh
 

Hi all,

I face a problem with rotating mesh as follows: In a 2D transient fluid domain, there is a ellipse wall or oval wall. This wall can rotate about is axis with sinusoidal function, let's say theta = A*sin(w*t). I write the functions of motion in x and y direction to apply to the moving mesh, but I could not get the result. The error always appears due to the mesh distortion when I run the problem.

My functions in CEL are: XDIS = r*cos(theta)-r and YDIS = r*sin(theta). Are these functions correct? Please help me.

Re: problem with rotating mesh
 

The CFX mesh motion definitions and options are rather counter intuitive.

There are two possibilities: Either your motion definition is incorrect or you are being too aggressive in the motion demanded per timestep.

Reduce the amplitude of the oscilation to check that your formulation is correct.

Re: problem with rotating mesh
 

If you want to calculate moving distance of a node between two timesteps, XDIS should be r*cos(A*sin(omega*(t+dt)))-r*cos(A*sin(omega*t)), while YDIS be r*sin(A*sin(omega*(t+dt)))-r*sin(A*sin(omega*t)).

Re: problem with rotating mesh
 

Hi, Thanks for your response, I follow your suggestions but my problem still can not be solved. I think my input function is not correct. First, I try to simulate a simple problem in which a circle wall rotates harmonically about its axis with a sinusoidal function alpha=A*sin(omega*t). Can you suggest any moving mesh function for that? Thank you

Re: problem with rotating mesh
 

Does the center of circle wall move? What's the rotating axis of it, x, y, or z?

Re: problem with rotating mesh
 

The circle wall is in xy-plane. It doesn't have any translation movement, but rotates harmonically about its center axis (z-axis). Thank you in advance.