The Fisker Karma plug-in hybrid, which launched in late 2011 in the U.S., utilizes a subframe with a cage that supports the fuel tank and traction motors. This unique design increases the challenge of ensuring the integrity of the fuel tank in a rear end crash as required for Federal Motor Vehicle Safety Standard (FMVSS) 301 certification.
FMVSS 301 requires that the rear of the vehicle be impacted with a deformable moving barrier at a speed of 80 km/h (50 mph). The standard limits fuel spillage due to these required impact tests to 28 g (1 oz) by weight during the time from the start of the impact until motion of the vehicle has stopped and to a total of 142 g (5 oz) by weight in the 5-min period after the stop. In a conventional vehicle, the fuel tank is positioned underneath the rear seat and is secured to the body by a pair of straps wrapped around the bottom of the tank. The trunk provides a substantial crumple zone that simplifies FMVSS 301 engineering.
The Karma’s unique subframe consists of two longitudinal members made of aluminum castings and a series of extruded cross-members that form a cage that houses the motors and the fuel tank. The traction motors are fastened on the subframe with cast aluminum brackets. They are positioned on the same axial centerline, and a transmission/differential case is mounted between them. The fuel tank is made of upper and lower steel shells and mounted inside the subframe above the rear motor. The depth of the tank requires a multiple step forming process. Straps wrap around the top of the tank to hold them to the cage while a pair of support brackets on the subframe cradles each side of the tank. The subframe is soft mounted with four rubber isolators to the body. The vehicle’s innovative design creates the need to manage the impact energy in a rear crash with a smaller than normal crumple zone.
Fisker used ESI Group’s Virtual Performance Solution extensively in the design of the rear structure and other areas of the vehicle. The Virtual Performance Solution includes a validated barrier model that substantially reduces the amount of time needed to simulate FMVSS 301 certification testing. It also offers a broad and fully validated library that covers current regulatory barriers for frontal and side impact tests: frontal offset deformable barrier (ODB, R94), NHTSA (FMVSS 214/301), European (Advanced, R95), and IIHS mobile deformable barriers.
The vehicle designers provided the simulation team with a CATIA model of the concept design of the vehicle. They built a relatively coarse model of the entire vehicle and a fine model that included the subframe, traction motors, and fuel tank. Virtual Performance Solution was used to mesh the models, using solid elements for castings and shell elements for extrusions. They applied boundary conditions such as tension in the straps holding the fuel tank to the model.
Once the vehicle model was completed, the FMVSS 301 test could be simulated by selecting the deformable barrier model from the library and crashing it into the vehicle at a speed of 80 km/h (50 mph). In addition to ensuring the integrity of the fuel tank, analysts were careful to protect the battery pack located in a structural tunnel running down the car’s centerline.
Engineers also applied a forcing function to the model designed to replicate conditions that the vehicle would see in driving under extreme conditions such as driving over rough roads, hard braking, and high speed. These simulations were designed to ensure that the motor and the fuel tank would not interfere with each other while the vehicle is being driven.
Fisker analysts began running crash test simulations shortly before the company began testing the mule or initial prototype. The physical tests were used to validate the simulation, and the simulation was used to diagnose the test results and evaluate potential solutions for issues that were found in the tests. In the initial design, some movement of the tank was seen in both the simulation and the physical tests. Analysts modified the design to improve the retention of the fuel tank and re-ran the simulation to compare the results to the baseline. They considered different options such as material, gauge, and section geometry of the crossmembers and the tension of the fuel tank straps in more than 100 simulation runs. Through an iterative process of changing the design and evaluating the effect on the crash performance, analysts were able to progressively improve the design.
In this way, the analysts developed an improved design for the rear structure that meets all of the FMVSS 301 requirements while also providing excellent performance in ordinary and extreme driving conditions. The improved rear structure design was used in the CP prototype. Physical testing of the CP prototyped showed that it performed very well and matched the simulation results. The design remained unchanged during the product validation phase, the final prototypes that were used to certify the ability of the vehicle to meet FMVSS regulations.
Fisker is currently developing its second vehicle—the Atlantic—again basing crash safety design on Virtual Performance Solution simulation.
Chun-Heng Chou, Computer Aided Engineering Supervisor, Fisker Automotive, and Peter Üllrich, VPS Product Manager, ESI Group, wrote this article for AEI.