I am trying to perform a nonlinear buckling analysis on cylindrical shell in SolidWorks. I choose arc length method, fix the bottom and apply a load higher than critical load obtained from previous eigenvalue buckling study to the top of the shell. Moreover, I apply an additional trigger load to the whole cylinder in direction perpendicular to the cylinder's axis. The purpose of this force (with magnitude equal to 0,5% of the buckling load and amplitude constant throughout the simulation) is to provide an imperfection because from what I know it's not possible to import scaled mode shape from eigenvalue buckling to nonlinear buckling simulation in SolidWorks (or is it ?). And there has to be some kind of imperfection in the model as otherwise the response will be discontinuous beyond the bifurcation point. However when I run the analysis it seems that the solver can't get any further beond that bifurcation/critical load point. In some cases it gives error ("The solution may be at buckling or limit point ...") and sometimes it solves but when I plot the load factor vs axial displacement of all nodes (I haven't used Workflow sensitive sensors yet) it shows straight line up to load factor of about 0,8 which should be somewhere around the critical load, if I understand it correctly (0,8*applied load).
Do you know what can be done to make the solver go beyond critical load and converge giving correct post-buckling nonlinear load-displacement response ?
Thanks in advance for your help