How do I represent a bolt/pin inside of a block?

[mojojojo69]

I've got a problem where I need to analyze using NX NASTRAN/FEMAP (I didn't get to choose which program). I need to model a bolt going through a 3d mesh of blocks that are tied to some surfaces (See pic). The bolts are tightened into the blocks w/ threads but the connecting arm is free to move. What I've tried, and have not gotten to work, is filling in the holes in the 3d mesh. Then applying a beam w/ the properties of the bolt through the solid. Then I merge the nodes of the beam that exit the solid. The arm is modeled as a beam attached to the center nodes of the bolt beam representations. Each solid is connected to the surfaces by merging at least 4 nodes in 2 directions. I try to run this in NX Nastran and built it in Femap. I get errors that don't seem to make sense (Fatal error 3001). Can anybody tell me what to do or is there a better way of modeling this. All I really want is the stiffness of the bolt in the overall analysis and to be able to extract the shear data out of the bolts.

 

[mikeneal]

I cannot recall what exactly error 3001 is and away from FEMAP so cant look it up. General thoughts, can you trouble shoot it by doing a modal analysis (use bailout if you must) and seeing how it behaves? If it wont run add some secondary "extra" constraints if you have to in order to get it to run without errors. As far as the modeling technique I have sometimes modeled a solid rod of the root or pitch diameter and glued it to the fixed side (block) and done a contact surface to the movable part in order to replicate the bearing /shear interaction. As a rule I personally like to use contact / glued surfaces opposed to merging nodes across interfaces but that's just my personal preference.