Showing results for 
Search instead for 
Do you mean 
Reply

SOL601-106 ATS with contact yields in large strains and very distorted element

[ Edited ]

Hello experts,

 

I have a problem with an analysis with large displacements / strains in solution 601-106.

- isotropic material behaviour

- linear heaxeder elements

- symmetrical constraints for quarter modeling

- surface to surface contact

 

In my model I press a kind of stamp (solid) onto a solid plate with high pressure a I want to get it deformed in a very strong way with buckling and so on.

Everything works fine until I increase my pressure.

I 've got a distorted element in contact zone with large strains. The element penetrate itself with a strain of nearly 32 % compression. Otherwise the next element in vertical direction above contact zone shows a strain of positive tension in vertical direction.

Only the utmost element layer is compressed, the next is not. I expected that compression would procceed through element layers.

 

If I analyze other incrementation steps earlier, I could see that the effect is growing but not from the early beginning.

 

I've got the following error message:

>>>>>>>>>>>>

Element distorted, Jacobian DET= -1.138862E-01  GROUP=        2  EL=    200830
       MESH TOO DISTORTED (JACOBIAN DETERMINANT NOT POSITIVE)
 NUMBER OF ELEMENTS WITH NEGATIVE JACOBIAN DETERMINANT ...=         1
<<<<<<<<<<<<<<<<<<<<<< 

 

And that is true.

Here some pictures:

 

1. Overlay plot of deformation during load incrementation (deformation scaling 1.00 absolute)

2. nonlinear strain component zz of compression zone of stamp (source region) at the end of aborted analysis:

 

Now my question:

Is this a limitation of solution 601 or may be an error in my model?

I' ve found a statement in help-documentation that acceptable maximum strains in solution 601 are nearby 30 %.

 

Is this limitation induced by such deformation effects or vice versa?

It looks for me as there is something like hourglass - deformation.

Does anybody know the internal relationships of NX and can help?

Could I change parameters for contact or NXSTRAT to improve the solution?

 

Any suggestion is welcome.

Best regards,

Michael