Early in the development of the Tesla Model S, studio designers and aerodynamicists decided to use a digital process in which CFD simulation was to be the principal engineering analysis tool while wind tunnel testing was to be used primarily for validation. Simulation made it possible to quickly analyze design iterations and optimize exterior surfaces such as the vehicle front, A-pillars, and other design features critical for aerodynamics.
“We used Exa’s PowerFLOW transient solver that directly resolves very large eddies to deliver superior accuracy compared to conventional CFD codes,” said Rob Palin, Lead Aerodynamicist at Tesla Motors. “This approach is especially beneficial in areas where flow physics are complex such as in the rear of a curved vehicle. The result is a full-size electric sedan with a drag coefficient of 0.24.” (Click here to view PowerFLOW animations of the side mirror and rear wake.)
The decision to deploy a simulation-driven design strategy put a huge premium on the ability to generate accurate simulation results, a particular challenge in automotive aerodynamic design because vehicles generate huge amounts of turbulence. Tesla began the design process using a conventional steady-state CFD code that uses the Reynolds Averaged Navier Stokes (RANS) approach to define the eddy viscosity on all scales of turbulent motion. “The results did not match our expectations, so we switched to PowerFLOW, which uses a fundamentally transient solver and sophisticated lattice-Boltzmann physics models to directly resolve anisotropic turbulence scales,” Palin said. “We had confidence based on validation and prior correlation with physical tests that PowerFLOW simulations were an accurate representation of the aerodynamics.”
Tesla used the traditional process of optimizing the overall shape of the vehicle followed by optimizing the individual details. Aerodynamicists worked closely with designers to provide feedback on alternative design concepts. Parametric studies were performed on key vehicle features to evaluate the effect of design choices on aerodynamic performance. Simulation was used to explore the design envelope to search for opportunities to achieve performance gains by iterating on the main design concept.
The front end of any vehicle is critical to aerodynamic performance because it sets up the flow around the entire vehicle. Tesla engineers worked with simulation results to manage the flow around the front end and sides to minimize the extent of the region of high pressure on the front profile while maximizing the suction on the outboard corners of the front fascia. Making the car rectangular in plan view acts to increase drag; however, the sharp corners generate suction that pulls the car forward in the same way that a wing lifts an aircraft. Exa consultants did a parametric study of changing the shape of the headlights and the area around the headlights with the goal of turning the air sharply so it adheres closely to the sides of the vehicle.
“The wheels generate up to 20% of the drag on the vehicle so airflow around the front wheels is critical,” Palin said. “One of the objectives with the Model S was to minimize the airflow around the front wheels and to line up the airflow so it hits the front wheels head on. We needed to avoid air hitting at an angle, as often the side of the tire acts like a bucket that catches the air, producing significant drag. This was an area where we made a huge improvement from the initial concept designs to the final design.”
The Model S has a smooth and tapering greenhouse with a shallow 66° windshield angle. Aerodynamicists used parametric analysis to optimize the curvature of the A and C pillars to minimize vortex formation. The side glass inset was minimized and the B pillar was eliminated entirely to further minimize drag. The front and rear headers were tweaked to minimize gaps and panel offsets. The rear Center High Mounted Stop Lamps (CHMSL) were integrated into the tailgate to minimize disruptions to the external airflow.
Tesla aerodynamicists also addressed many details of the exterior shape using PowerFLOW simulations. When the panoramic sunroof is opened, a small screen pops up. A parametric study was performed to optimize the height of the screen from a drag perspective and also to minimize sound pressure levels in the cabin. A booming low-frequency sound when a rear window is open is a common problem in aerodynamic design. In the case of the Model S, the flow was so clean around the sides of the vehicle that the problem moved to the front window. Aerodynamicists evaluated changes to the A pillar surface and side mirrors and came up with a new design that minimized the aeroacoustic noise.
Simulation results closely matched wind tunnel testing. The simulation results provided more information than physical testing, making it possible to diagnose and develop solutions to drag problems much faster than would otherwise be possible.
“The drag coefficient of early design concepts of the Model S was 0.32,” Palin said. “The major shape changes reduced the drag to 0.27, and the smaller changes provided further improvement to 0.24. These numbers were validated with wind tunnel testing using standard SAE International procedures, and extended using CFD modeling to add in factors that are not traditionally included, due to practical limitations in experimental testing, but do contribute to the on-road drag of the car.”
Brad Duncan, Director of Aerodynamics, Exa Corp., wrote this article for SAE Magazines.