"ball joint" sounds like pinned (in three directions)
"sliding" sounds like pinned in one or two directions (does the point slide along a line or free to translate on a plane).
you'll need to consider rigid body motion to correctly restrain your model.
The rear wheel of bicycle is sliding in the x-axis only. So, i should let x-axis free to move (TX uncheck), and check the y and z translational, as well as the 3 rotational?
From the ball-joint boundary conditions, you said it should constraint the 3 translational motion, and uncheck the rotational motion because it is free to rotate in the 3 axes(x, y and z). There's no translational motion.
But sliding boundary conditions, the rear wheel can only move in the x-direction, so shouldn't i should uncheck x-axis, and constraint all the others?
sliding doesn't require moment fixity. in particular in your case you're looking at the rear axle of a bike, clearly one axis has no moment stiffness.
ask around. it is better to constraint the model either minimally (reacting the 6 degrees of rigid body motion) or as realistically as possible (which in my mind says avoid rotational fixity as much as possible ... what is truly "fixed" in reality ?). adding redundant constraints to your model will at least mess with the loadpaths, at worse mess with all your results.
if you want try it either way and see the effect. which answer is right (or righter) ?
Dear Steve,
The following picture can help you to understand how to setup the FE problem correctly using FEMAP & NX NASTRAN:
Prescribe different beam cross section properties to every member of the frame and play with them to reach the best design that fits the prescribed requirements of both weight & stress -- good luck!.
Best regards,
Blas.
PD
Here you are a real life model of a mountain-bike meshed with FEMAP & NX NASTRAN at
