You can easily check if tie constraints work properly by running a frequency analysis. If anything is disconnected, it will be visible in low frequency modes.
My first mode starts at 50.65 Hz. I will not see any rigid body motion as my side plates are properly connected to the bottom frame. I still have the doubt that something is wrong as I see the maximum deformation in the plate where I suspect that tied constrain is not working there properly
In CalculiX, one surface must not be a master surface (red) in a tied constraint more than once. You have constraints part-81_to_part-49 and part-81_to_part-41, which violate this principle. You should merge such constraints into one (select both, right click, merge).
I couldn’t check the file yesterday but right, that’s also why there’s a bunch of warning messages such as:
*WARNING in gentiedmpc:
DOF 1 of node 168558 is not active;
no tied constraint is generated
*WARNING in gentiedmpc:
DOF 2 of node 168558 is not active;
no tied constraint is generated
*WARNING in gentiedmpc:
DOF 3 of node 168558 is not active;
no tied constraint is generated