The creative spark of a true free spirit

From March to Red Bull via Williams and McLaren, a sprinkling of ideas that set Newey apart from his contemporaries

March 881 vee-shaped chassis

The Leyton House March 881 was by far the most aerodynamically advanced car on the 1988 grid, and a key part of that was how Newey took the driver’s feet – wider at the toes than at the heels - on the pedals as his cue for the chassis shape, to give significantly less frontal area than the conventional rectangle-section tub. “That came on one of those transatlantic flights. I was thinking about how to treat the underside of the chassis as an extension of the front wing rather than as something that happened to be there and the wing got bolted onto it. That seemed a bit basic so I tried to create a sympathetic and minimal structure behind the front wing - hence raising the underside of the chassis to, roughly speaking, put the other side of the chassis on a streamline off the wing in side view. Then you have to come down at some point to get down to the reference plane at the back edge of the front wheel. That meant having to depart from the natural streamline so I wanted to get that part as narrow as possible. You couldn’t go any narrower at the balls of the feet but you could go narrower at the heels. That’s where the vee keel came from.”

March 881 nose/wing underside

With a combined nose/wing, downforce was being created across the whole width rather than just from the two separate wings bolted onto a central nose. As an extension of that, undercutting the nose on the underside lowered the air pressure there, the air rushed to fill the vacuum, pulling the flow over the wing faster. “I didn’t think to go as far as Jean-Claude Migeot later did with the Tyrrell [019] when [by lifting the driver slightly] he undercut the whole keel [rather than just the nose], but aerodynamically I think the March was better than anything out there at the time. F1 cars at that time had become very unsophisticated aerodynamically. It was actually more sophisticated in Indycar where I was working at the time because everyone had pretty much the same power unit whereas F1 in the turbo era people had been finding more and more power which meant that drag wasn’t that important, so the cheap and easy way was just to put ever greater wings on the things. They tended to be somewhat inelegant as a result. Indycar heightened my realisation that aero was the key.”

March 881 sculpted endplates

Before the 881, front wing endplates were flat across their height and width. Which meant that as the wheels were steered it changed the gap between the endplate and the tyre, giving an inconsistent airflow. Newey sculpted the shape of the plates to keep that gap consistent as the wheels were turned. “I guess I’m unusual in being a race engineer, an aerodynamicist and a mechanical designer. In those days I don’t think anyone was bothering to steer the wheels in the wind tunnel. If you’re an aerodynamicist spending all day in the tunnel you might forget that steering actually happened! Whereas I’d be at the track with Mario, watching him come in and out of the pits, working with the driver in trying to understand what the car was doing around the lap. The train of thought that led to the sculpted plates came from talking to Mario, one of the greatest drivers ever, and understanding what he needed out of the car.”

Williams FW18 cockpit proportions

New regulations were introduced for 1996 that defined a minimum cockpit height for enhanced head protection.

Newey’s FW18 appeared to contravene the rule, but a careful reading of the regulations proved otherwise.

“The rules appeared to be asking for these very high ugly [and aerodynamically disruptive] side headrests, but when you read through it carefully it specified the height of the side of the chassis and it specified a minimum area for the side of the headrest but it didn’t anywhere say that the side headrest top edge had to match the profile of the chassis.

“An easy goal!

“I prefer not to know the intent of a regulation. Because that can lead you to pre-judge. I’d prefer just to read the regulation in black and white without knowing what it’s trying to achieve. That’s one of the reasons I resigned from the F1 Technical Working Group. The others being that it was brain-numbingly dull and a bit depressing in that my fellow technical directors were like turkeys voting for Christmas and wanting ever-more restrictions!”

Williams FW18 Modified airbox

A newly elongated section between the back of the cockpit and the airbox inlet, introduced at Hockenheim in ’96, cured the long-running problem of airflow separation starving the front two inlet trumpets of the engine.

“On holiday during a flight from Barbados to St Barts, I looked out of the window and saw the shape of the engine intake just below the propeller and it triggered a thought. If you look at most turbo-props you have the hub of the propeller, then a bit of a gap down, then the engine intake. Typically that intake will be stand-alone and the bit in between the top of the intake and the underside of the propeller hub will be a nice rounded shape.

“So turn that upside down and it suddenly became the solution to something I’d been battling with for best part of a year of how to stop the air off the back of the driver’s helmet and onto the top of the headrest separating and spoiling the flow to the trumpets.”

Williams FW18 undercut diffuser

The FW18 had a massive aerodynamic advantage over the rest of the 1996 field by dint of a diffuser that was undercut so as to go up over the exhausts rather than beneath them, giving a hugely enhanced diffuser exit area and flow volume.

“This was another one of those reading the rules carefully to see if there was a way of doing what you wanted rather than what they wanted. I realised that although the reference plane specified a width and the step plane had to be 50mm above that at each side, at no point did it say the step plane had to extend to the rear axle, like the reference plane did. It meant you could consider the step plane ending at the edge of the rear wheel. It meant there was this grey area where you had freedom to cut in above the reference plane to give a much bigger diffuser exit. The diffusers were very exit-limited and by undercutting it we got a huge advantage by drawing so much more mass flow through it.

“We actually introduced it on the FW17B at Estoril late in the ’95 season. I thought that was the worst possible time to introduce it because everyone would see it and copy it for ’96. But somehow they didn’t.”

Williams FW14 front wing foot plate extension

Newey’s first design for Williams, the March-like FW14 of 1991, featured an extension to its endplates that prevented the dirty airflow from the tyre squirt (the air gets squirted either side of the tyre at the contact patch as it has nowhere else to go) polluting the flow to the underbody.

“This one wasn’t very health and safety. We were using the Southampton University wind tunnel for the critical surfaces. It had an 11 foot by 8 foot working section and we ran a 40 per cent model. On each side of the rolling road there was a platform, about 10 inches wide, which if you were careful you could just about walk along. I was trying to understand the flow physics of the car because in those days we didn’t have CFD or PIV and all the things we now have. So it was flo-viz and bits of wool. Flo-viz can only teach you on-surface. The only way to understand the off-surface flow characteristic was to go with a wand of wool. That’s exactly what I did. I got the tunnel technician to slow the tunnel down to a speed I could just about not fall over and break my ankle and I walked along the side with my wand of wool and observed all this very dirty air just around the inboard side of the front contact patch area. That was wheel squish, which hadn’t been named at the time because we hadn’t properly understood it.

“I had a eureka moment in the shower shortly afterwards. We needed a physical barrier there. So I then had a look through the rules and realised there was a loophole between the centreline of the front wheel and the back edge of the front wheel. It was unrestricted in height, didn’t have to be 25mm above the reference plane (which the front wing itself had to be).”

McLaren MP4-13 chassis shape

The 1998 McLaren was an amalgam of Newey’s own ideas during his gardening leave from Williams and research already conducted by McLaren on how best to configure a car for the radically different regulations that demanded a reduction in width from 2-metres to 1.8 and the introduction of grooved tyres.

“I didn’t have any data to back up my initial thoughts – because I was working from my bedroom – but I felt that lowering the centre of gravity would become crucial because the narrow track would bring more weight transfer and the grooved tyres would be less consistent as the grip co-efficient would fall away more as loads went up. So you’d need to minimise weight transfer as much as possible. I started working with Mario Illien at Ilmor in secret, to get the crankshaft height down because that not only lowers the whole engine but also the gearbox internals. A low seating position obviously became important because if you could get the driver’s upper body lower, then you could lower the surrounding structure, the roll hoop etc. Then visibility becomes a problem. Another regulation change that year was go to a box chassis where the regs called for a certain width and bulk of chassis at front bulkhead and a certain width and bulk at dash bulkhead and a linear taper in between. Although the drawing [in the regs] showed a rectangular box, it actually said depth. It didn’t say it had to be a rectangle.

“One of the guys working at McLaren when I started there had spotted that and had proposed effectively making a very shallow chassis in height then putting a couple of fins up the side to legalise it. But I felt that wouldn’t be legal because it said depth. In the middle you wouldn’t have had the depth. But potentially what you could do was create something not unlike Batman with the two triangular ears sticking up at the top and the bottom chamfered. So at any given point across the width it complied with the depth - but wasn’t a simple rectangle.

“So there seemed to be a loophole. You could get straight ahead visibility by dropping the chassis down in the middle and still be legal. So then we started working with David Coulthard. It was him who said, ‘My vision is binary; I’m either looking straight ahead for the braking area or I’m looking diagonally across at the apex.’ So we could put these fins in the bits in between without really bothering the driver. That’s what led to MP4-13 chassis which was a good step.”

Red Bull RB5 pull-rod rear suspension

The key innovation of this 2009 car – which would be the basis of the title winners of the next four years – was the thorough repackaging of the rear end in response to radically different aero regulations that limited the diffuser height.

“Because we now had a full-length diffuser of very limited height, to get it working as well as possible you needed to create a very low pressure at the diffuser exit. Feeding really good quality flow to the beam wing would become vital because the harder you can get that to work, not only is it creating its own downforce but it acts as an extractor for the diffuser. So you work your way to upflow and ask, ‘What’s the dirty flow that’s impeding the beam wing from working as efficiently as possible?’ One very obvious thing was the push-rod and the rocker assembly around it. So convert to a pull-rod and it allows you to raise the rocker out of the way.

“The push-rod had become a convention for good reason. On the previous generation of cars the diffuser started much earlier – at the front edge of the rear wheel rather than the centre line of the rear axle – making it difficult to package a pull-rod because it would have been going straight through the diffuser. But now we had a change in the regs.

“When something changes, what else does that allow you to alter rather than just saying. ‘Okay, I’ll just adopt the change?’ It wasn’t a new technology, just adapting an existing one in response. Motor racing as an industry is very good at that. It’s too small to do genuine research – we’re not going to create new materials or write start-from-scratch CFD codes. But we are prolific users of technology and that’s what makes the industry so interesting to corporate suppliers we now have as tech partners. We can take that product and work it much harder than other industries. A good example is the driver-in-loop simulators we use, which take a lot of software from the gaming industry.

“Sometimes coming up with a wacky idea is quite easy. But then it’s a case of being critical of them. I apply the 24-hour rule. Scribble something down then leave it and if after 24 hours it still seems like a good idea, pursue it further. Even a team the size of Red Bull can’t pursue every idea that someone proposes. We have to choose the ones most likely. I do see that with some of my colleagues – at both Red Bull and previous teams. They will come up with an idea and start to pursue it and it will start to become obvious it’s a turkey and isn’t going to fly. But you’ll see some will belligerently keep pushing it because it’s their idea. That’s where the disciplined engineering mind comes in – you have to keep critiquing it and saying, do I think this can work or is it time to ditch it and move onto the next one?’ ”

McLaren MP4-20 front suspension

Rules that lifted the front wing 50mm for 2005 potentially put the wake off the back of it smack into the lower wishbone, destroying the flow further back. The convention with the raised noses of the time was for the lower wishbone to be mounted to a keel beneath the nose. Newey thought it didn’t need to be that way.

“The tip vortexes were clattering into the wishbone and making a mess. So the obvious thing to do was raise the wishbone. So how do you do that and get the required structural strength? The drawing office and stress guys came up with it. Having raised the wishbone at the outboard end you might as well join the wishbone straight into the bottom corner of the chassis, and not have the flexible bit at the bottom. It sounds blindingly simple now and the surprise almost was no other team did it.

“This thought came to me while on holiday and faxes were flying back and forward from the hotel. I wasn’t very popular on holiday...”