I believe the best way to settle down a model that does not converge is to reduce the nonlinear complexity so that it does converge. At least then you have results to review and you can evaluate the effect of your model simplification.
Use the tools in the analysis window to show which restraints are not converging. My guess is there is more than just friction, probably a +Y support or a gapped restraint that is involved. Make one of those linear and try again. When you review the results you can evaluate the significance of that simplification.
I suggest you do not look at occasional loads alone when you have friction in your model. For example, don't run WIND1 alone, instead use W+P1+T1+WIND1. Weight provides a normal load that will settle down friction.
Finally, if load case 6 is not converging, delete load cases after 5 to see results for the first 5 load cases. At least you can get started.
_________________________
Dave Diehl