Bad Buckling Modes with Soft Material

[headrush]

I have been trying to run a non-linear analysis on the compression of a rotationally symmetric part made of a very soft elastomer. The modulus is a mere 300 psi and I have tried poission's ratio's of .48 and .49. The elastomer is bonded to some nylon (I used the material library for these properties) at each end and I created a cavity for the nylon parts in the elastomer, so they should fit perfectly.


I keep gettingbuckling modes which do not make sense to me. It is not a collapse of the structure or even part of the structure, but one or two elements which go spastic. Has anyone else experienced similar problems? Since I have many buckling modes occuring below my applied load I can not run a non-linear analysis.


The first 10 modes all eminate from the same place and are very close in value. It appears to be a meshing error. Maybe some nodes didn't merge? Is it possible for Mechnanica to not merge nodes? Ifthese werereal modes I think I should be seeing multiple modes fromvarious points around the perimeter.


I have modeled 1/4 of the part (90 degrees) and these modesoccur at or near 45 degrees. The buckling occurs near constraints at one end that set all DOF to zero (I'm using solid elements so the rotational DOF are meaningless). Could Mechanical be getting confused as the cartesian coordinate system wraps around the semicircle?Also, the bucklingoccurs near a material interface.


Any help will be appreciated.

 

[AHA-D]

All I can say is that I saw elastomer simulations at work at different occasions and every time it was specialised software. The first occasion wasa research centre near Paris that did the bulk of rubber products for French car industry. That was end nineties and hey had an airconditionedsupercomputer working on the subject. Second time was at a plastics conference on a "normal" computer, 4-5 years ago, with less complex products. Don't remember the software unfortunately.


The thing i got from the information given is that it needs a lot of input of material parameters to get it right, and that it is difficult because the model practically needs to be rebuilt after each iteration to take in account the deformations. Load points moving, parts getting in touch with eachother, ... Maybe that's what's causing the problem : a small stress or model problem that "explodes" because it gets blown up by the calculation.


Sorry that I can't give any real help.

 

[napoleonm]

Could you post some pictures to see what you have and what you got?.


Regards

 

[headrush]

A picture is not always worth 1000 words. I can't show you the entire structure but I am hoping thisdemonstrates the problem. Hopefully it is clear that there is no structural instability shown. The spastic element is located at a material interface. The other part is hidden.





View attachment 3015