FEA and Crash Analysis: Questions Answered by Skill-Lync Experts

When a material is plastically deformed, unloaded and then reloaded past its previous maximum stress, its resistance to plastic flow has increased. Its yield point is higher than in the preceding cycle. This is strain hardening. The two main models are isotropic and kinematic hardening.

Isotropic hardening

• The yield surface is symmetric about the stress axes, so yield strength in tension and compression is the same.
• It does not account for the Bauschinger effect.

Kinematic hardening

• In real materials, an increase in tensile yield strength usually comes at the expense of compressive yield strength. This is the Bauschinger effect: plastic deformation in one direction affects the response in the other.
• Kinematic hardening accounts for it. The yield surface keeps the same shape and size but translates in stress space.

Bilinear vs multilinear When you define a material in ANSYS you sometimes need to provide a hardening curve. If the hardening curve has two linear regions, it is bilinear. If it has several linear segments, it is multilinear.

• Adjust to Touch: the program ignores any initial gap or penetration between the contact surfaces and creates an initial stress-free state. The surfaces are just touching. Initial gaps are closed; if they were not, the bodies would fly off during the simulation. It cannot be used for concentric cylinders with a gap.
• Add Offset, Ramped Effects: an offset is an extra user-entered distance. It brings the two bodies into initial contact, with the load values ramped over the first load step. You enter the offset value manually. A positive value brings the bodies closer, a negative value moves them apart.
• Add Offset, No Ramping: works like the ramped option, but the loading is not ramped.

Adjust to Touch assumes both bodies are in contact with no gap. When the model has a gap that you want the analysis to consider rather than ignore, you do not use Adjust to Touch.

Which option is best depends on the type of application.

Ideally you would add gravity in all cases: frontal impact, side impact and roof crush.

• In the frontal and side impacts, the vehicle is dynamic and we use a reduced vehicle model, so gravity has no significant effect on the result. That is why it is neglected.
• In roof crush the vehicle is static, so gravity does have an effect, and it is added.

You can include gravity in the other cases too, but you must then constrain the vehicle so it cannot move in the negative Z direction. That is the only change in model building. You can use a rigid wall, or arrest the Z translation DOF, as is done on the shock towers in the project.

Adding gravity helps show how the vehicle behaves in a crash, as in real life.