This content has been marked as final. Show 4 replies
Could you post some pictures of what you are trying to accomplish? What loads are you applying?
To get it to mesh you are going have to apply fine mesh control in the thread area. You could also try "automatic transition" under Options when you mesh.
I don't know if it would help but you might also want to look into bolted connectors under loads/restraints->connectors.
Also, modeling contact between threads usually requires nonlinear analysis. From COSMOS help:
A special class of nonlinear problems is concerned with the changing nature of the boundary conditions of the structures involved in the analysis during motion. This situation is encountered in the analysis of contact problems.
Pounding of structures, gear-tooth contacts, fitting problems, threaded connections, and impact bodies are several examples requiring the evaluation of the contact boundaries. The evaluation of contact boundaries (nodes, lines, or surfaces) can be achieved by using gap (contact) elements between nodes on the adjacent boundaries.
I have attached screenshots which will hopefully paint the picture more clearly. I have fully restrained the back of the boss because it is only the welds and threads I am concerned about. The load is a 1500 lbf load applied normal to the top plain applied at the end section of the shaft approximately 17 inches from the back of the boss.
I have tried to apply fine mesh control and I was successful but only the threaded areas appeared and I could not figure out how to apply fine mesh control locally. When I tried to apply fine mesh control over the entire model SW stated that I ran out of memory and then shut down (apparently 3 gigs of ram is not enough to handle that). I tried your automatic transition option to no avail. I have yet to look into bolted connectors, that is my next plan of action.
this is not good application for FEA. this can all be done using closed form solutions that will give excellent results and take less time.
stress in a cantilever s=M/C*I
give that this shaft is welded on the front and back, you can ignore the threads and simply look at the crossectional area of the weld joint under stress and to determine if it is strong enough.
I have already done those calculations on the weld joints. I was just experimenting with the COSMOS software and thought that it would be a complex problem to practice and learn on. I was also curious to see how the software results would compare to manual calculations.