Dealing with porpoising has been a challenge for several Formula 1 teams in the new ground-effect era. As a result, the engineering company ENGYS decided to explore the problem and developed useful techniques to simulate and visualize the phenomenon.
The problem was at its worst in 2022, with multiple drivers complaining about the issue before FIA rule changes largely eliminated the problem for 2023. Notably, many teams failed to foresee the extent of the issue before on-track testing, having not picked up the gravity of the problem in design simulations or wind tunnel testing. As highlighted by ENGYS in a blog post, that’s not a huge surprise, given the difficulty of simulating the phenomenon using Computational Fluid Dynamics (CFD) software.
Porpoising involves severe vertical oscillations as an F1 car’s underbody gets too close to the ground. When this happens, the underfloor aerodynamics can stall, leading to a sudden reduction of downforce. This leads to the car rising in height, at least until the aero kicks back in and pulls it down again. When this happens rapidly, it leads to the uncomfortable porpoising phenomenon.
Porpoising is difficult to simulate due to a variety of factors. One is that typically, CFD simulations and wind tunnel work both fail to accurately simulate the action of the car as a whole system, including the actions of the car’s tires and suspension. As explained by McLaren’s James Key last year, the porpoising phenomenon is not solely down to aerodynamic action, but by the interaction of the forces with the vehicle’s suspension, too. If that isn’t accounted for, the problem won’t reveal itself.
As for CFD simulations, they face a greater problem. They simulate flow by creating a “mesh” of distinct volumes around the car. The flow parameters are calculated for each mesh cell, and the aerodynamic behavior can thus be simulated.
The problem with doing this for porpoising is that it involves the car’s underbody becoming ever closer to the track surface. This creates difficulties as the CFD simulation mesh must become ever denser and smaller to predict the flow in this vanishingly small space. As the mesh becomes finer and finer, it becomes more expensive to simulate, computationally speaking, particularly regarding F1’s restrictions on simulation time.
ENGYS developed a technique to get around this problem. If you’re not an expert in CFD, it won’t make a lot of sense, but it involves simulating the situation as if the ground is moving relative to the car. The CFD software then snaps a mesh to the track surface underneath the car, which changes as the car’s ride height changes.
Even with these optimizations, and only simulating half a car, the simulation still took 7.5 hours to model 0.5 seconds of porpoising behavior on a 128-core computer. F1 places significant restrictions on the amount of CFD simulation that teams are allowed to use, so any efficiency gains in this area are highly useful.
If you’re studying at university and want to get into motorsport, diving into this area could serve you very well. Understanding how simulations can best be made to represent the real world is key to success, and it has fast become one of the biggest parts of top-tier motorsport.
Got a tip? Let the author know: firstname.lastname@example.org