circular motion of the whole surface

[Sagar]

Dear all,

I am solving a 3D thick Cook's membrane problem. The left surface is fixed and I would like to apply a circular load on the right surface so that, the whole surface rotate along a center point while keeping the face straight. I have tried using spin command but it will twist the surface along the center point. I also did it using proportional loading of type 1 along with CSURF command using tangential and traction commands but the motion is not exactly circular.

[Prof. R.L. Taylor]

Have you tried to link the dof's on the surface you want to remain planar and then twist it?

[Sagar]

Dear Prof.

I have tried using LINK command as well. The problem is that, the whole surface twist along the center. I want the whole surface to rotate along the center. This means the center of the surface will also rotate along the specified center while keeping all  the edges perpendicular all the time. The surface rotates along the y axis while the xz plane of the surface is fixed. I have attached the input file.

[Prof. R.L. Taylor]

Can you describe the motion you want the end to follow?  When you say twist around the center, do you mean center of the cross-section?  Can you prescribe boundary conditions on the center to keep it where you want?

[Sagar]

Dear Prof.

I have attached the description of the motion of the surface. I want the end surface to rotate around a point in circular motion while keeping the xz plane of the moving surface that is fixed to the surface to remain perpendicular to the rotating axis all the time.

[Prof. R.L. Taylor]

If all the end nodes are linked to have the same "y" displacement, I would think this should happen.

[Prof. R.L. Taylor]

The attach file results in all the y displacements being the same.

[Sagar]

Dear Prof.

Thank you for quick responses.

Still, this is not what I want.  I have uploaded a picture to make it more clear how the center point of the end surface rotate around the circular path while keeping its face completely perpendicular to the rotating axis all the time.  I have also attached input file where I tried to do it using proportional loading but it somehow rotates the surface in circular path but it is not very smooth.

[Prof. R.L. Taylor]

You should be able to do this with two proportional loads that are something like

u = c_1 + u_o * sin(at)
w = c_2 + u_o * sin(at + pi/2)

also link the y direction still to keep the surface planar.

The nodes on the surface need all the u & w dof's fixed. 

[Sagar]

Thank you prof. it is working fine now.

[Prof. S. Govindjee]

For those that are interest in SPIN, there is an example on the wiki page: http://feap.berkeley.edu/wiki/index.php?title=Spin_boundary