Modeliing a thin-walled cylinder

[biomedman]

I am trying to model a cylinder (1mm thick, 16mm dia, 5mm height) of stainless steel. The cylinder has cuts made in its wall with the intention of making it flexible in pure compression. What kind of elements would be good such an analysis?

Regards,
Ravi

 

[pjhype]

It depends on the analysis you want to do. If you are interested in only the compliance of the cylinder, the standard 8-node isoparametric shell or plate element should be adequate.

If you are worried about the stresses in the vicinity of the cuts, you might want to use solid (brick) elements.

If the compression direction is parallel to the axis of the cylinder and the cuts in the wall are designed to make the cylinder compliant along this axis, you may want to consider performing a buckling analysis if the remaining "beams" are loaded in pure compression.

Alternately, if you need compliance along the axial direction, you might want to design the cuts so that the remaining beams are loaded in shear or bending.

pj

 

[corus]

For a thin walled analysis the 8 noded shell elements should be used. If you require a detailed analysis of the region around the cuts, for say stress relief holes, then use a sub-model on that region. If you believe that the stress distribution through the thickness of the tube will not be linear then use brick elements. I would stick to shell elements for the sub-model though for such relatively thin plate. Using brick elements throughout or trying to model details in the whole model is too expensive for such a model and sub-modelling is the better option.