Cancel
Showing results for 
Search instead for 
Did you mean: 

Constraints with RBE2 PROBLEMS NX NASTRAN

Experimenter
Experimenter

Hello guys,

 

I have some problems with my Nastran simulations. I want to lock the bearing seats of my assembly. So i used a 1D Connection to connect the bearing surface with an rbe2 element to a point in the middle. Now i locked dof 1 and 3 on that point. I applied some loads but the bearing seat can still move. What am I doing wrong?

 

Thanks for your helpassembly1_assyfem1_sim1.pngassembly1_assyfem1_sim2.pngassembly1_assyfem2.png

7 REPLIES

Re: Constraints with RBE2 PROBLEMS NX NASTRAN

Legend
Legend

I am not sure but it seems to be that the component is rotating about "y" axis?

Re: Constraints with RBE2 PROBLEMS NX NASTRAN

Experimenter
Experimenter

that DOF is locked by the flange on the left with the four holes in it

Re: Constraints with RBE2 PROBLEMS NX NASTRAN

Siemens Phenom Siemens Phenom
Siemens Phenom

I would agree with @jon_morga. Fixing X and Z at the centerline grid will prevent translation, but not rotation. The deformations and the contour distribution sure look like a rotation to me.

 

Even if the rotation is fixed by the left component, there is bending in the legs that connect the middle disk to the right disk, sothe right-most piece is rotating with respect to the fixed flange.

Re: Constraints with RBE2 PROBLEMS NX NASTRAN

Legend
Legend

You can perfom a modal analysis and see the mode shapes with natural frequencies close to 0. This indicates solid rigid movement. 

Re: Constraints with RBE2 PROBLEMS NX NASTRAN

Experimenter
Experimenter

Thanks for your help! I didn't read you answers correctly. You are right with your opinion about rotation. It should be able to rotate there. I was concerned about the expanding.

Re: Constraints with RBE2 PROBLEMS NX NASTRAN

Experimenter
Experimenter
I will try that. Thanks for your help!

Re: Constraints with RBE2 PROBLEMS NX NASTRAN

Siemens Phenom Siemens Phenom
Siemens Phenom

The "expansion" is a different issue. This is due to the fact that you are running a linear analysis. This means that the model (nodal coordinates, load directions, etc.) is not updated with nodal displacements. This is a good assumption for small displacement analysis.

 

If you truly have large displacements and need the model to update with displacements, you need to run a geometric nonlinear (large displacement) analysis.