On Thursday in Silverstone, England, ahead of this weekend’s British Grand Prix, Formula 1 revealed next year’s car to the public. 2022 will see the biggest shake-up to the sport’s technical regulations since the introduction of the turbocharged hybrid powertrains in 2014. There has been a fundamental change in the way the car creates its aerodynamic downforce, with the goal being to make it easier for F1 cars to race each other closely. Ars spoke to Rob Smedley, director of data systems at F1, to find out why and how the new car came to be.
What’s the problem?
The cars that will race each other at Silverstone this weekend use the air to generate grip through a combination of the front wing and rear diffuser. And they make an awful lot of downforce, which is part of the reason F1 lap times have reached historic lows. The problem is what happens to the air after it has passed over an F1 car’s body—it becomes a massive wake of disturbed air. A wing running in turbulent air won’t work nearly as efficiently as a wing running in clean air, and that means it’s very hard for one car to follow another closely enough to try and overtake—something that F1 fans have told the sport they want to see more often.
“As the  car moves in, let’s say a second behind, it’s losing around 25 percent of its downforce,” Smedley said. “As it moves in to about half a second—a closing distance and getting to the point where they could start to have this wheel to wheel interaction—at that point, it loses 40 percent of its downforce. So the loss is immense.”
An F1 car’s performance has always been a mix of aerodynamic performance, tire grip, and engine power, but in the 25+ years that I’ve been a fan of the sport, aerodynamics has been the most important of those three elements. “We’ve never really thought about the importance of the wake effect—the air flow coming off the car on the car behind—we’ve just accepted that’s what happens in Formula One,” Smedley told me. That is, until now. (Or next year, to be accurate.)
How are they fixing it?
In recent years, F1 teams have made massive investments in server farms as the sport cracked down on unrestricted wind tunnel testing (one of its repeated attempts to control costs, something now done much more effectively with a cost cap). But that decision came from the teams—the sport itself does not have those kinds of in-house resources to rely upon. Instead, it turned to Amazon.
“Partnering with AWS, we were able to basically redesign a Formula One car using the computational fluid dynamics (CFD) simulation. And it was important that we were designing the car with the following car,” Smedley explained.
Designing a two-car simulation is significantly more complex than a single car simulation, and in fact, the process would have been perilously slow had each team tried to do it themselves. “The teams are very limited by what’s called the aerodynamic test restriction, which is part of the technical regulations, so they’re allowed to run a certain amount of computing power, which equates to roughly 190 cores of compute power when they do their CFD simulations,” Smedley said.
Going with EC2 gave Smedley’s team a lot of flexibility and access to much more computing power, as much as all 10 teams could muster together. “We started off with around about 1,100 cores, and then we got to about 2,500 cores. At one point we were spinning up upwards of 7,000 cores, which is just mind blowing,” he said.
That meant his team could solve the problems quickly. Smedley’s team was able to run two-car simulations in about six hours, compared to the four days it would take using a team’s CFD infrastructure. “Which meant developing the new car in an agile way, and that really was the game changer,” he added. “We knew what effectively the concept of the car was, but we could then prove that in a scientific way.”
How an F1 car will make downforce in 2022
As you can see in the photos, the 2022 car is a much cleaner design than the cars that are racing currently. There are many fewer aerodynamic appendages sprouting up here and there (as that’s something that will not be allowed under next year’s technical regulations). There is a less complicated front wing, control devices to mitigate some of the turbulent air coming off the front wheels, and a shaped underfloor that generates most of the car’s downforce via ground effect, something F1 last saw in the early 1980s.
Redirecting the wake of disturbed air from an F1 appears to be an important change. “Because of the change in pressure differential across the top and bottom of the car, you get a very, very strong upwashing effect,” Smedley said. “So the wake actually lifts up rather than out-washing and creating this this massive void of really turbulent, non-Newtonian flow behind the car.”
Sorry, the wings stay for now
I asked Smedley whether F1 had considered anything radical, like ditching the wings altogether in favor of mechanical grip. Many fans believe that kind of tweak would result in much better racing. After all, decreasing drag is more important in terms of road car relevance than adding downforce at triple-digit speeds, and Peugeot has even opted for a wingless design for its next hybrid Le Mans racer, the 9X8.
“If you went away from it—an aerodynamic formula—you’d struggle to claw back that performance and probably end up with significantly slower lap times, and these would not be the fastest race cars on Earth. So I think that it’s important that we didn’t take away from that, but I also think it’s important that we had a really good eye on efficiency,” Smedley replied. “We’ve reduced the ability to create a lot of downforce—and therefore a lot of drag—with the furniture around the cockpit; you see these very intricate bargeboards and turning vanes and vortex generating solutions on all of the [current] cars.”
As a result, next year’s cars will have a much higher lift-to-drag ratio than the current race cars (which are between 7:1 and 9:1, according to Smedley).
Listing image by Formula 1