Design of a Novel Crash Can

Tesla

The results of this projects resulted in a journal paper publication:

Kenyon, Daneesha, Yi Shu, Xingchen Fan, Sekhar Reddy, Guang Dong, and Adrian J. Lew. "Parametric design of multi-cell thin-walled structures for improved crashworthiness with stable progressive buckling mode." Thin-Walled Structures 131 (2018): 76-87. (link)

Goal:

To optimize the design of a five-cell crash can, shown in the figure above, so as to maximize the amount of energy per unit mass absorbed by the structure upon a frontal collision.

Problem:

The performance of crash cans crucially relies on the onset of progressive buckling when impacted with a high-enough impulse, since deformations of this type absorb a large amount of energy as plastic deformation. However, the crash can can also deform by buckling in the first mode, as any column would do, or switch to buckling in the first mode after a short length of progressive buckling. The problem is to define the geometry of the crash can so that it reliably deforms through progressive buckling upon impact. In particular, the team was tasked to design an optimal five-cell crash can and compare its performance with optimized four- and nine-cell crash cans.

What did the team do?

The team decided to parametrically explore design alternatives following two parallel and complementary paths. On one side, they investigated rough performance measures through a simplified model, known as "Super Folding Element Theory," which allowed them to identify promising regions in the design space. On the other hand, detailed LS-Dyna simulations of the entire geometry of the crash can were performed for a variety of impact and boundary conditions. By performing these two studies for a range of potential designs for four-cell, nine-cell, and five-cell structures, they arrived to stability or buckling-mode diagrams like the one shown in the figure above. Among these designs, the one with the best Specific Energy Absorption, or energy absorbed per unit mass, was selected, leading to the design shown in the top figure. Both this design as well as the optimal nine-cell design were built and tested at Tesla, with some of the results shown across the above figures.

Because of the promising initial results, both the team and the Tesla mentors decided to continue the project for an additional quarter, leading to the publication cited and linked above.