Membrane wrinkling simulation in ABAQUS

[rcsingh723]

Hello all,

I've been trying to perform an FEA simulation of a membrane wrinkling in tension in ABAQUS 6.7. It's a 100 mm x 100 mm square Kapton (.013 mm thick, 2600 MPa secant modulus, v = .34) membrane meshed with S4R5 elements with an element size of about 1.33 mm. It's in the xy-plane, and the boundary conditions are applied as follows:

Initial - left side of membrane fixed along x-axis, z-axis, and right side fixed along y-axis, z-axis

Step1 (static, general analysis)- keep boundary conditions from initial, add 1 mm displacement load to top right corner of membrane in the diagonal direction

Step2 (Lanczos buckling)- keep boundary conditions from Step1, add 1 mm displacement load to same corner.

I know I'm not doing this right, because despite the fact that the static analysis step runs fine, the buckling analysis step does not. It says that the model is either preloaded above the bifurcation load or has rigid body modes. I've attached my most recent CAE-generated .INP file for reference.

Please help if you can. I need to get the proper mode shapes so I can apply the imperfections and get some good results by the end of the week.

Thanks,

Raymond

 

[Nileo2005]

Whats the .dat file error complaining about?

 

[Sleight]

You can also apply random imperfections to the nodes. You can write a program to generate random imperfections based on the thickness of the membrane and apply it to the nodes. Therefore, the results won't have any bias towards the applied imperfections from the mode shapes.

A better way would be to apply random out-of-plane loads to the membrane to initiate the wrinkling. Then, once the wrinkling starts you can remove the out-of-plane loads.

Check out papers by Tessler and Sleight (me) for random imperfections and Leifer for random out-of-plane loads. These papers provide results for your problem.