No convergence in the second step

[sps210]

Hello,

I have a certain static loading on an object in the first step, and the abaqus direct solver converges in a single iteration. I have another step in which I propogate the same boundary conditions and the loads, but the solver does not converge. I am not sure why.

I am using the second step to model surface contact between a solid model and a surface, but I disabled the interaction to troubleshoot when I discovered that even having the exact same static load and boundary conditions do not work properly.

Is there anything I am missing about the second step?

thanks!

 

[rstupplebeen]

More details, images or your model is needed to really help.

Rob Stupplebeen

https://sites.google.com/site/robertkstupplebeen/

 

[sps210]

Here is the image file of my model.

http://files.engineering.com/getfil...2-0972-4be3-87fb-8d57825f4124&file=abaqus.jpg

The pic on the left is the undeformed solid object. We apply cylidrically inward force (step 1) on the "legs," and we deform the object as shown in the top-right image. This object is inside a almost-cylidrical solid shell shown in the bottom-right image. The undeformed object protrudes out of the cylinder. When I let go off the forces (step 2), I want the legs to just touch the cylinder. Instead, the solver never converges even when I disable surface-to-surface interaction step. The solver doesn't converge for step 2 if I simply propogate the load and the boundary conditions from the initial step to step 1 and step 2.

I have the following boundary conditions propogated from the initial step:
a. three non-collinear point on the solid object are held stationary.
b. three non-collinear point on the cylindrical shell are held stationary.

The solid object is made of a superelastic material and the cylindrical shell is made of any metal.

 

[rstupplebeen]

Try ramping down the force applied in step 2. I believe the discontinuity in load is causing the solution to fail. I hope this helps.

Rob Stupplebeen

https://sites.google.com/site/robertkstupplebeen/

 

[sps210]

Even when I propogate the same load (and boundary conditions) in step two (and deactivate the interaction module), the solver does not converge. I use a direct solver.