It was only a few years ago that some of Formula 1's brightest minds were predicting a future where wind tunnels were no longer needed. Think of the wasted expense powering a giant fan in a massive, empty room filled with nothing but a 60% scale model of a car, and all the time spent poring through the data readouts.
It seemed that technological progress was reaching a cross-over point, one where computer simulations could do all of that and accurately – and effectively – predict flow patterns around a car.
And here we are, the year 2024 of the common era, and two F1 teams are in the process of either building or finalising brand-spanking-new wind tunnel designs. Aston Martin is in the commissioning phase of its new in-house facility, while Red Bull is aiming to have a new wind tunnel in operation by 2026. McLaren, for its part, had finished its own new build in 2023.
The anticipation for both incoming wind tunnels among the teams will be stratospheric. These are often held up as a significant influence on performance, one that can change the fortunes of a team and stall any hints of a decline. But, as ever, it's not quite as simple as that – even after construction, there are months of work ahead before the wind tunnel is even remotely useable.
In those months, there will be significant challenges in getting the wind tunnels ship-shape. Aston Martin is already delving into those, while Red Bull will have to go through the entire rigmarole in about 12 months' time when it moves out of its Bedford tunnel – which dates to the Cold War as part of an aeronautical research site.
These are the challenges we don't hear about; the narrative always seems to lean towards the following phraseology: "Team X has a new wind tunnel, it'll work off the bat and the car will be competitive". Let's set the record straight and explain how to set up a wind tunnel, and detail the work that Aston Martin and Red Bull have ahead.
To do that, we've asked former F1 aerodynamicist Jean-Claude Migeot – formerly of Tyrrell and Ferrari, and the architect of F1's high-nose era – to assist. Migeot has been heavily involved in the development of multiple wind tunnels over his post-F1 career, including the Aerolab and Fondtech facilities in Italy, so he's the perfect person to help us untangle the key details.
"First of all, I don't think you get benefits straight away," Migeot begins. "You have to build confidence. A wind tunnel is not reality, it's a laboratory. So, you have to trust it behaves like a laboratory, which means if you're ever going to repeat something, it's that the results are going to repeat perfectly.
"That's the minimum required, and then [you build] confidence between the real thing and your simulation in the wind tunnel, and it takes some time to build that confidence. But it depends on the background of the people using it."
In this, he cites the example of McLaren switching to its new wind tunnel – ending its reliance on the Toyota facilities in Cologne – and the perceived impact that this had on the Woking team. This coincided with the team's upward mobility in 2023, as upgrades took the team from one floundering in the mid-pack to one fighting at the sharp end of the grid. Building the wind tunnel is only the first step.
Once the building has been completed, with all monitoring equipment installed, the tunnel then needs to go through what's known as 'commissioning', and that's before the work can even start with a wind tunnel calibration model. It's a lengthy process, and one that Aston Martin is currently mired within as it hopes to get its tunnel working in time for the 2025 season.
"Step zero is commissioning," Migeot explains. "This is where you have to check what has been sold to you is effectively achieved. You know, the air distribution, the turbulence, the temperature control, which is hugely important.
"You have to run the tunnel in all conditions and first check your builder has done a good job, then run all the electronics for the control [units]. It's a huge work for a group of people and is taking at least one month. And then if that's okay - because if that's not okay, you are fixing something which is not hours, but days and weeks of changes - then you start to work with the model."
Explaining commissioning and correlation
Formula 1's current ruleset is now into its third year, and the increased competitiveness of the 2024 field has demonstrated the level of convergence that is starting to be achieved with steady regulations. Teams are having to dig deeper to find lucrative opportunities to improve, but this is leading a number of teams to encounter correlation issues between the wind tunnel and their on-track findings.
For example, a new floor tested in the wind tunnel might churn out data that suggests it would be an upgrade worthy of entering the production phase for a future race. Once ready, the team tries it over a race weekend – but it either fails to offer the promised downforce gains or, more alarmingly, creates instability in the car.
There are a few notable examples of this happening over 2024. Ferrari went through a mid-season phase of floor iterations that progressively worsened high-speed bouncing through corners, while Aston Martin's Imola upgrade made the car much harder for the drivers to handle. RB and Mercedes have both rolled back on floor upgrades, and Red Bull's updates mid-season took a long time to untangle after introducing a variety of balance issues.
The catch-all explanation is that it's a correlation problem and, at surface level, this is correct. Some of that is down to how a wind tunnel is calibrated, and perhaps there are losses somewhere in the tunnel that make it hard to achieve the right figure. Take Renault's 2007 decline, for example; the then-reigning champion team appeared to suffer with its switch from Michelin to Bridgestone tyres due to the change in tyre profile.
Migeot explains that a tunnel doesn't have to be perfectly calibrated to what is experienced on track, but the difference between the tunnel and real-world readings must be consistent.
"The tunnel will not give you the truth. It will give you something which, in the best scenario, is consistent with reality. If you have a quite fixed difference between reality and the tunnel, that's fine; it can't be identical. If you're confident that, when you improve in the tunnel, you know it's going to improve on the track, you're saving a huge amount of time.
"I hear this doesn't work very much like that for many, the upgrades are brought to the track and then they return to the old configuration quite often these days. So I think a lot of teams are in correlation problems."
So, can you ever achieve perfect correlation? "It's endless, it's a quest you never finish," Migeot states. "You're in a happy situation when you have confidence, but the situation can change quickly, and it would be radically changed by these rules of two years ago with a completely new floor.
"There's many imperfections, like in the rubber tyres (on the model), maybe you don't have the right tyre profile under yaw, there are many compromises and there's maybe a hundred differences which you cannot have in the tunnel. You try your best and, if these things are kept to a very small influence, you're okay.
"Then you are in the case of a fixed error, and you can make your development and improvements. But if suddenly one of these things becomes influential on the results, and you are again in correlation work. For example, a change of tyre dimension is not very often these days, but even the rubber state between new and used can make a difference on the track.
"All these things you have to collect, and it takes time. So it's an endless subject. I think you have people on that all the time, measuring what's happened on the track. If it's only in some occasions you don't correlate, it doesn't stop you. But when you see a car which is on the front row on one race, and on the third or fourth row on the next race, then you have a problem."
