5th July 2017 by Richard Yen
Before we continue our look back at the 2016 Enlighten Award entries, be sure to check out this year’s nominations which were revealed to the world last week featuring Aston Martin, Honda, GM, Faurecia, Volvo, Constellium, Meridian and many more.
Now that’s done, let’s review American Axle’s 2016 entry which concerned a Power Take Off Unit Housing (PTU) for the 2016 Chevrolet Equinox vehicle. The PTUs are mechanical gearboxes that are attached to the transmissions to transmit torque from the engine to the rear axle. With pressures to lightweight vehicles being felt by suppliers as much as the car brand owners, American Axle wanted to investigate ways of removing mass and material from the PTU while still meeting the high performance standards expected of their products.
To achieve this goal, American Axle utilized topology optimization techniques to find areas where material could be removed. The process works by using specialist optimization software to define areas of the initial design that is ‘designable’ (i.e. the software is allowed to make geometry changes) and ‘non-designable’ (i.e. areas where the software is not allowed to modify such as fixing points). The known loads on the components are entered into the design optimization software and any relevant design or manufacturing constraints are added, such as ensuring that the resulting design does not contain any holes in the components’ structure.
The optimization technology (in this case OptiStruct from Altair) is then able to use this data to suggest a new, material efficient design which removes material while still meeting design targets.
The study resulted in a 17% weight reduction compared to the baseline component, but what really set this entry apart was the link the company made with a reduction in part rejection as a result of the optimization process. With margins often very tight for automotive suppliers, this nomination showed that the benefits that design optimization studies can deliver go beyond weight savings alone.