Hi San, i'm a bit lost.
Can you give us the bigger picture?
can you show us the whole assembly, describe the setup and then make a list of the simulations you want to run and how you're thinking of applying the loads and boundary conditions? listing out the assumptions you know exist and are ok with would be helpful.
then list where your concerns are per study.
i don't think you're outside the scope of the software that you have, but i need to better understand what you're trying to do before we give an answer about the validity of the setup..etc.
Sorry for the confussion.
Ive attached 2 more pictures to help me describe what Im trying to achieve.
This is a wheel steel disc bolted to a hub. I then will apply a remote load to the hub in an attempt to simulate cornering forces, which I hope then can be used in a fatigue study to varify that the disc design can pass a 300,000 cycles with full stress reversal.
At this point, the fatigue suggested the bolt holes would just simply fail at circ 3000 cycles!
if you were going to do a fatigue test by hand, what stress would you use?
have you read up on stress concentrations and singularities?
i might suggest setting up this analysis with bolts that you've modeled. don't owrry about the preload for now and then compare those results to an analysis with bolts with no preload.
where i'm going with this is that i think you're using a stress concentration/singularity that is artififically developed by the bolt connectors to make a fatigue decision. you may need to read away from those locations to better understand what is going on.
Jared is right, there are usually concentrations wherever rigid restraints OR connectors are applied. In this example, if you are using the Rigid vs. Direct Transfer in the Remote Load settings, then you are in fact over-rigidizing the bolt holes, which will lead to high stress. In reality the fasteners are elastic.
I agree with jared to add physical bolts, since their interaction with the disc is most likely going to result in complex contact stress that can't be completely idealized. Another option is to run with the "direct transfer" option which would bracket your analysis the other way, assuming the fasteners add no rigidity to the system.