Over the past few months, the aerodynamics team has been busy at work using tools at our disposal to better understand and analyze our vehicle in order to produce the best results possible. With a team comprised mostly of Windows-based computers, we’re extremely thankful to blueCAPE for their invaluable contribution of a Windows port of the open-source OpenFOAM software that allows us to simulate airflow across our car. The software port, called “blueCFD-Core”, has been an extremely useful and time-saving solution that has allowed us to use our own personal machines to learn the solver without the hassle of having to migrate operating systems.

An airflow simulation using the blueCFD-core OpenFOAM port

While OpenFOAM is able to solve simulations for our vehicle, it also includes an integrated meshing tool that allows us to produce accurate results within a reasonable time frame. It has allowed us to learn much of the basics for generating meshes, but does pose limitations towards customizing the discretization of cells and visualizing the mesh results. As a result, we use the dedicated meshing program Pointwise which allows us to set parameters such as structured or unstructured meshes, vary cell sizes for different regions, and create meshes with a lower cell count without sacrificing an unreasonable amount of accuracy. The program has been able to save us some simulation time as a result of its ability to reduce the overall cell count of our mesh, and is another tool that is extremely useful to have.

With a turnaround time of several days, we’re constantly looking for ways to improve our simulations and the time it takes to run them. As a result, we’ve turned to the Amazon AWS service to increase the computational power of our team in order to produce even better results. Where the average desktop computer contains anywhere from two to eight cores, our fastest AWS cluster contains 64 cores and a plethora of RAM that allows us to produce results much faster than we were able to before.
blueCFD-core simulation results showing pressure
With this arsenal of tools at our disposal, we have been able to generate post-processing results that give us better insight into the aerodynamic performance of our vehicle. We are able to identify areas where airflow conforms to the vehicle and areas where flow separation may be occurring. We can analyze different planes to observe how airflow changes around the vehicle and process that information in order to learn from this vehicle and make our next one even better. With the current vehicle having entered the production phase the aerodynamics team is getting a head start on the next one - using the extra time to run more in-depth simulations. We have been able to accomplish much over the past few months, and thanks to our sponsors who have contributed invaluable assistance to our mission, the skies above the team remain clear and sunny.