ADVEA supervised Aydin Kulekci in his final year of Automotive Engineering on the topic of reverse engineering and re-designing a steering knuckle from a 2012 Toyota Camry. The purpose of the project was to teach Aydin topological optimisation and common linear elastic Finite Element Method solutions.
The 2012 Camry was used simply because the geometry was readily available. This model has been developed by The National Crash Analysis Center (NCAC) of The George Washington University under a contract with the FHWA and NHTSA of the US DOT. With this model we were able to achieve the goal of working with a commercial level assembly with complex geometry and connectivity which is a good challenge for a final year engineering student.
Aydin was required to manage as many original constraints as possible, and considered packaging, manufacturing, material and stiffness constraints for the topological design. The first thing required was to translate the model from the existing LS-Dyna model into a Simulia Abaqus model using Altair HyperMesh. His analysis background to date was pure Abaqus implicit/explicit analysis using Abaqus/CAE for pre-/post-, and so he was keen to investigate the Abaqus Topology and Optimisation Module (ATOM) – so that was the agreed direction.
The steering knuckle was successfully remodeled with C3D10 second-order tetrahedral elements with thin S3 linear triangular shell elements on the surface for more accurate surface stress prediction. The original material model was matched for the linear elastic terms (Young’s Modulus, Poisson’s ratio and density were used).
A normal modes of vibration analysis was completed as an initial measure of stiffness. Frequencies and mode shapes were recorded to be compared. A subsequent set of linear static stress/displacement analyses were completed using 1N and 1N.mm forces and moments at all attachment points. The translational displacement magnitudes were used to determine initial stiffness constraints for the component. The boundary condition for these cases was fixed at the wheel-hub connection.
The topological analysis was used to identify areas where material could be removed with as little sacrifice to stiffness as possible. The areas which were identified are shown the below figure, highlighted orange.
A comprehensive re-analysis was performed for all normal modes and stiffness conditions that were initially investigated. It was found that maximum von Mises stress and the fundamental natural frequency could be maintained with under 1% alteration, but if stiffness in one particular loading path was able to be traded off by around 3% then a total weight reduction of 5% would be possible.
Clearly not knowing any more about the part than we were able to determine from the unit loads, the project cannot go further than that – however for an educational exercise for a final year student it was an interesting process and allowed the student to gain an insight into engineering methods which are used for virtual design in OEMs today.