otaviobarodrigues
Student
Hello, everyone.
I'm modeling the contact between two concentric tubes in ABAQUS. The inner tube is elastic, while the outer one is rigid. I used PIPE31 and ITT31 elements. The inner tube is under buckling.
First, I obtain the buckling modes to introduce the imperfections in the structure. Second, I add the material imperfections in the post-buckling simulation and create a step to apply a line load due to self-weight. In another step, I add a concentrated load to generate buckling.
I understood that it is not possible to apply the imperfections only in the last step. Thus, in the loading simulation due to the self-weight, the solution does not converge. Warnings such as:
"The nlgeom flag was not active in the previous general (non-perturbation) step. The activation of nlgeom in the current step may cause convergence problems if the displacements at the end of the previous general step are not small. Discontinuities in displacement response histories should also be expected."
Due to this warning, I set NLGEOM=ON and the problem still does not converge.
I will add a .inp files to better visualize the problem.
I'm modeling the contact between two concentric tubes in ABAQUS. The inner tube is elastic, while the outer one is rigid. I used PIPE31 and ITT31 elements. The inner tube is under buckling.
First, I obtain the buckling modes to introduce the imperfections in the structure. Second, I add the material imperfections in the post-buckling simulation and create a step to apply a line load due to self-weight. In another step, I add a concentrated load to generate buckling.
I understood that it is not possible to apply the imperfections only in the last step. Thus, in the loading simulation due to the self-weight, the solution does not converge. Warnings such as:
"The nlgeom flag was not active in the previous general (non-perturbation) step. The activation of nlgeom in the current step may cause convergence problems if the displacements at the end of the previous general step are not small. Discontinuities in displacement response histories should also be expected."
Due to this warning, I set NLGEOM=ON and the problem still does not converge.
I will add a .inp files to better visualize the problem.
