"If I use the "COMPONENTS FEATURE" to add the 10000N in all directions will this result in an overall 30000N load."
I don't think so. Forces are vectors, so three principle vectors with magnitude of 10000 yield an equivalent of ~17320 magnitude vector, not a 30000 magnitude vector. (I'm surprised no one commented on this. BTW, unconstrained sketches have no effect on the results of the FEA as long as your solid is the desired size and shape.)
I can't remember, but does IV11 have a "bearing load" option? I believe this is probably the best choice (in combination w/ the fixed constraint you already have in place). See attached for a simulation of a couple of simple cases. Although the bearing load won't allow for axial loading of the hole, so another force will need to be added to the side face. (I believe the force is applied over the whole face and is equivalent to placing the load at the centroid)
You will need to define different simulations for different loading conditions. Unfortunately, IV only allows you to define one simulation at a time and doesn't allow you to save different loading conditions. So you will spend alot of time redefining different loading conditions and running the simulation, especially if your iterating a design. (maybe this has changed in IV2010, I certainly hope so). Like others have said, the FEA analysis is a static analysis. You can either use Dynamic Simulation to determine the loads at each time period, or estimate worst case loading and use that force.
Just from looking at the part, I believe the worst case scenario would be a bending about the long dimension of the bottom face. It will probably be the weld in tension that fails before ripping of the material. For this, I would get the Machinery Handbook and look at cantilever beams for analysis methods. Also look at the weld analysis. If you look at the reports I attached, it also suggests the same.
Remember, that while you are using IV Professional, it is using FEA Light, so be critical of your results and compare with other solution methods. Don't just assume they are correct.
