How Should Bending Stress Be Calculated in a Spinal Disc Endplate Design?

[cabellos6]

http://img341.imageshack.us/my.php?image=endplateballsocketij7.jpg"

I am designing the appropriate size of an upper endplate for a spinal disc and was wondering what would be the best way to calculate bending stress in this design. The image is shown in the link above. It is a ball and socket arrangement - would it be suitable to simplify this to a single knife edge when considering the stresses in the upper endplate?

 

[Pyrrhus]

Your FBD is an unstable structure. Is the plate bolted to the disc?

 

[cabellos6]

The two upper and lower endplates with ball and socket arrangment form the 'disc'. I need to do some calculations to gain an idea for an appropriate thickness of endplate. I am beginning to think beam bending is not appropriate in this case then as the top endplate is not fixed (e.g it can't be modeled as a cantilever beam). The endplate is able to provide side lateral bending, extension/flexion and lateral rotation about the central axis of the ball.

Does it seem this is not a beam bending case? More of a direct 'compressive' stress case Stress = Load/Area ?