My Goal is to combine the benefits of BCONTACT based contact formulation to efficiently model Shell-to-Shell contact with the benefits of linear gaps, modelling a fastener that can only take tension. If it is not possible, how to deal with 1D Elements that should only take tension in MSC.Nastran?
This should not work, since a SUPORT entry in a contact analysis results in the following error:
*** USER FATAL MESSAGE 9008 (SUBDMAP PHASE0)
A SUPORT BULK DATA ENTRY IS PRESENT.
USER INFORMATION: THIS ENTRY IS NOT ALLOWED IN LINEAR CONTACT ANALYSIS.
And Linear Gaps require the use of SUPORT entries.
More than likely, you will need to use nonlinear to get the contact and nonlinear effects for the bar (using nonlinear material).
That being said, the old tried and true method of running it once with the 1d element connected and the other body contact on, then checking the loads in the 1d element to determine if it is in tension or compression, then removing it if it is in tension, then running again to see what happens... you can 'iterate' manually. If you use a spring (CBUSH) then you can leave the element in place, but modify the stiffness to be 'stiff' or 'soft' as you iterate.
I'm probably missing something, so maybe someone else will add some comments.