You probably want to revisit your constraints to verify them. You could add contact definitions, but that's likely overkill and will increase your simulation time. Keep in mind that the displayed results are exaggerated, your bottom results aren't as different as you think: top graph has...
If you use the "OP=NEW" option when defining loads and boundary conditions for each steps, it should reset rather than carry over previous definitions.
That'd be my guess, but I generally don't spend much time in the GUI except for results visualization/post-processing. You can verify in the .inp after writing it from the GUI though.
Upgrading the older odb to the current version with the "abaqus upgrade" command (Section 3.2.18 of the 6.14 user's guide manual) should allow you to then use your script.
@GregLocock is correct. Assuming all your input units are consistent, I'd guess the option to scale the modes based on mass (rather than displacement) was selected. Your generalized masses should all be unity in that case.
Why? You could allocate an array of the appropriate size and store all the values in said array. I'm not sure what you're asking for is even possible in Fortran or similar compiled languages (see http://stackoverflow.com/questions/9536346/fortran-dynamic-variable-names) like it would be in an...
For the shell thickness, you can view it as a contour plot if you output the STH variable. It won't render the thickness geometrically, but you can at least get an idea of it from the contour.
Sounds like OP works for Company A providing services to Company B and is asking whether it's a conflict of interest to accept money from Company B under his small business, Company C. Sounds like at minimum a perceived conflict of interest. While OP has not signed an NDA or anything with...
