I have not actually used Simulation to test the stress in a threaded connection partially for the exact reasons you are listing. You would need No Penetration contacts between the two touching faces which should not be difficult to set up but will definitely add considerably to solve time. Without seeing what your model looks like it is hard to say how difficult it would be to mesh but you would need to be careful to try and get a nice, uniform mesh, especially at the root of the thread. I can't really comment too much on how plausible your analysis is without seeing it. It is possible to do this analysis but I cannot say how accurate your results would be. You would definitely want to do some hand calculations to try and verify your Simualtion results.
Ryan has it spot-on. It is really difficult, even impractical (esp. on a PC) to do full 3D helical threads. I attached a white paper (one page) we did at GoEngineer on the subject. You can conduct a simplified analysis using the 2D Axisymmetric capability of SW Simulation (shown in the white paper), that can get you some answers.fairly quickly. However you'll still need to insert mesh controls and lots of contacts. I hope this helps a little. -Tony
ThreadFEAExampleWPSS0711.pdf 454.7 KB
Hi Ryan and Anthony,
Thanks a lot for your input and help. I was really hoping to get different answer to my question but we can do only what we can do. I will keep trying with the options you gave me and see how it goes. Thanks a lot again.