Solidworks deformation before yield point

Question

I'm running a solidworks drop test simulation and I'm having some trouble making sense of the results. The idea is basically a hollow shell made of aluminum that will impact the ground at 5 m/s. It is carrying some weight inside which I was unsure how to model so I just created a custom material and said that it has the exact same properties as AL 6061 but has a higher density which makes the mass come out to the real world mass of the shell+internal mass. This alone might be causing some issues but I don't know how else to say that the shell is carrying a mass (I thought of perhaps increasing gravity, but that seems even more roundabout).

My main concern has to do with the displacement, below you can see that the heavy shell has impacted the ground and stresses are propagating upward in a wave, the scale on the side shows that the max stress was 273 MPa, and that the yield is 275 MPa.

The shell should therefore not have any permanent deformations, however when I show the "deformed result", this is what it looks like:

Solidworks shows this large displacement (Which I made sure is on a true 1:1 scale and not exagerated). It corresponds to 3.5mm, but if you look closely the displacement will actually be double this. I noticed this by superimposing the undeformed model in the white silouhette. As shown, the whole part was moved a certain amount,downward, and the nose was dented upward, solidworks is subtracting the difference but in reality the nose is twice as far away from it's original position. Basically, besides the curent 3.5mm, you also have to account for moving the whole model down so that the bottom noses correspond to one another, meaning the displacement will be almost a centimeter.

My confusion stems from the fact that this all looks like far too much displacement for the deformation to be plastic, but as I mentioned before, we have not exceeded the yield point, so it should theoretically all spring back into place after the impact. Should I trust that since the YP stress wasn't reached the shell will come out intact?

Answer 1

Looks like "snap through" buckling of a shell to me.

When you say "propagating like a wave", are you saying that this is modeled as a transient hyperbolic analysis?

Are you modeling it as a large strain, large displacement problem with the appropriate strain measure and stress conjugate? I'd assume Green Lagrange strain.

You describe the body as a shell; your model looks like an axi-symmetric model might be appropriate. The number of elements through the thickness is key here. You need at least two to pick up bending behavior properly; more elements or higher order elements will be better. Bending usually means a linear strain distribution through the thickness. If you don't have enough linear elements, or aren't using higher order elements, you won't pick it up correctly.

You can check by assessing convergence: refine your mesh in that area, reduce the time step, and run it again.

You are close to yield, so perhaps your model isn't refined enough in that area to pick it up. Modify the mesh in the area of the hinge and re-run.