How to Build a Car: The Autobiography of the World’s Greatest Formula 1 Designer

by Adrian Newey

Thus, the aim of the chassis designer is to: One: ensure that the tyres are presented to the ground in an even and consistent manner through the braking, cornering and acceleration phases. Two: ensure the car is as light as possible. Three: ensure that the car generates as little drag as possible. Four: ensure that the car is generating as much downforce as possible in a balanced manner throughout the phases of the corner.

An aeroplane lifts because the contours of its wing cause air to flow at different speeds across the two sides, low pressure on the topside, high on the other, with the wing moving in the direction of the low pressure and giving us what we call ‘positive lift’ as a result. The wing on a racing car works the same way, but in reverse: ‘negative lift’, or ‘downforce’, pressing the car into the ground and hence allowing the tyres to generate more grip.

If you watch a heavy river bird such as a swan, it will often fly just above the water, with the tips of its wings on the edge of dipping in. In doing so, it harnesses two powerful effects: (1) If its wing tips just touch the water’s surface, the leak path is sealed, the low pressure on the suction surface is not compromised and the wing hence becomes much more efficient. (2) The downwash of air behind the wing (created by circulation) reacts against the river’s surface, creating a higher pressure underneath the wing – a phenomenon known as ‘ground effect’. Turn this upside down, so that you have a downforce-generating wing with its endplate rubbing on the ground, and suddenly you have a massively effective solution.

In those days, every motor racing team effectively had three engineering disciplines: the design and aerodynamics offices, and race engineering, though the race engineers would be doubling up with working in the design office during the week. Since then, the industry has mushroomed, and nobody crosses from one department to another. You’ll have, let’s say, 90 people in aerodynamics, another 70 in the design office, and perhaps 30 in race engineering and simulation, the latter being a relatively new area.

I’m known chiefly as an aerodynamicist, but that’s a product of the fact that aerodynamics is the biggest single performance differentiator. Therefore, I tend to spend most of my time looking at aerodynamics, with the mechanical layout a close second, in order to make sure the two complement one another in a package. In fact, my sole interest lies in improving the ability of the car to score points, and what helps me do that is my experience across the disciplines.

Tragically I was to learn how it felt the hard way. I’ve had one driver die in a car I’ve designed. Ayrton. That fact weighs heavily upon me, and while I’ve got many issues with the FIA and the way they have governed the sport over the years, I give them great credit for their contribution to improving safety in the sport.

At Red Bull I’ve introduced what I call the 24-hour rule, which is that we sit on an idea for a day or so, throw it around and talk about it, but don’t do anything concrete until it has been critiqued. Does it still stand up after 24 hours? If the answer’s no then we chuck it in the bin.

A Porsche 911 is a horrible car from a vehicle dynamics point of view. It’s all to do with the fact that the engine is hung out the back behind the rear axle. Owners who take early models on tracks often spin them. Once that heavy rear starts coming round, it’s like having a sack of coal in the boot: when it starts moving, it’s difficult to stop it. You have to make much bigger steering corrections.

The data showed that Ayrton held that position of 40 per cent throttle and low level of steering torque for half a second, then got very heavily on the brake. All we saw after that was extremely high brake pressure as he left the track. Again then, the sequence of events consistent with the data is that the rear stepped out, Ayrton reacted, doing his best to hold the slide by reducing to 40 per cent throttle and reducing steering torque before realising after half a second that he’d lost control, after which he jumped on the brakes. The initial stepping out of the car was nothing to do with any steering column failure. There had to be another explanation for that.

There is a photograph in Autosport (20 February 1997, page 6) that shows a piece of debris on the track, with Ayrton’s car about to pass over it. His right front and right rear tyres were completely destroyed in the accident, so it was impossible to examine them and say for certain, but a piece of debris that size could easily have caused a slow puncture. The puncture would have caused the bottoming we saw, and that in turn would have caused the rear to step out as it lost grip, since you’ve unloaded the tyres meaning that the weight of the car is now being taken on the skids, which have no lateral grip capability. Not only that, but with the car now flat on the deck the diffuser would be completely stalled, resulting in the rear losing most of its downforce.

There are two possibilities here. One is that the steering column failed at this point. The other is that as the car came off the back side of the hump pointing left, but with the front wheels still pointing straight ahead, the rear suddenly gripped and threw it sharply right. What we could see once we were allowed to inspect the steering column was that it did have a fatigue crack present, so it was going to fail sooner or later. It had fatigued roughly a third of the way around the circumference and the rest had snapped, either in the impact or from the pressure Ayrton exerted while trying to control the car after the rear stepped out. Where the steering column failed was where it had been locally reduced by 4mm in diameter.

Now, I am responsible for following that request of Ayrton’s to lower the steering wheel slightly to avoid him rubbing his knuckles on the inside of the chassis. I am responsible for giving the drawing office the instruction to lower it by 2mm, and when they came back to me to say that it would then interfere with the FIA cockpit template, I instructed them to reduce the steering column diameter locally by 4mm.