Winning on a Technicality

With the world championship fight closely poised into the second half of the season, the intensity of the technical battle between the teams is full-on. Because the technical regulations are so limiting, the biggest single factor driving developments has been the move to Bridgestone as a single tyre supplier. With most of the field previously on Michelins, it's had some profound effects on the how teams have adapted, and which technical directions they have pursued, in turn has affected the competitive state of play. Gary Anderson and Giorgio Piola report

The big challenge this year has been adapting to the Bridgestone tyre. Most teams felt that Ferrari would have a bit of an advantage because of its experience of this year's Bridgestone, which is different from the '06 tyre and comes from the same family as the '05 rubber.

The Michelin and the Bridgestone are very different in their construction. The Michelin had a very stiff tread and soft sidewall. The Bridgestone is nearer to crossply in its characteristics with stiffness all the way through it. The teams will have received information from Bridgestone on the tyre and compared it to the Michelin. But some teams got it right and some got it wrong.

Surprisingly, McLaren seemed to get it near enough right immediately, whereas Renault got it badly wrong.

Aero Implications

There are two tyre-related areas you need to get right: the aerodynamics and the weight distribution. From what we heard at the start of the year people were suffering big aerodynamic losses because of the different tyre profile.

The Michelin had a different straight-on profile anyway, but the difference becomes really big when you compare the profiles under load and this has serious implication on the aero. Move a tyre 20mm and you could easily lose five per cent of your downforce.

It's very difficult to simulate a loaded tyre profile in the windtunnel. You can change the tyre profile in the tunnel and change your aero quite dramatically as a result, but you still have to question if that is the profile you're actually getting on the track.

You sort it through correlation with time as you go from track to track, but it's not yet a full science because the simulation isn't really there yet.

Weight Distribution Implications

The tyre also works differently from a Michelin; it moves differently, mid-corner in particular. Weight distribution is crucial in getting the best from it, regardless of aero. And this is where most teams seem to have struggled. Plus or minus five per cent downforce can still be a good car, but of those teams that have lost their way with their designs, it's typically been about weight distribution rather than downforce.

With this year's tyre your car concept needs the weight forward compared to a Michelin. You then need to have the aero balance forward with it, and a car concept that doesn't allow you to bring the weight forward will make it even tougher to get the aero forward because the whole reasoning behind your package has been in a different direction.

Making the underfloor, bargeboards, front wing and chassis front work harder to make the front wing work harder is a tough task. Unless you're starting in a position where you can get most of that, you are fighting with it. Renault has shown that - it recognised the problem a while ago but is struggling to correct it because it started from the wrong place.

The Michelin rear concept was better for taking longitudinal acceleration. So when you went on the power the tyre had more capacity, so in order to fully use that capability you wanted the weight distribution rearwards.

The Bridgestone rear is not as good for longitudinal forces; its performance between longitudinal and lateral is more balanced. You therefore need a more forward distribution than the Michelin. This takes some of the load away from the rear tyre, making it feel as if it's bigger and as a consequence you abuse the front tyre a little bit.

With the Bridgestone you need to minimise the weight transfer in braking and acceleration. With the Michelin you didn't because of its longitudinal capability, so when you braked or accelerated it could handle the transfer because of its longitudinal performance. It all means the Bridgestone is more sensitive to weight distribution. The Michelin allowed a greater margin.

I'd imagine the weight distribution range of the cars to be between something like 46 per cent on the front at Hungary/Monaco and 48/49 per cent for high-speed, high-load tracks like Spa/ Silverstone.

t's difficult to get a bigger adjustment window than that because you don't have room to move the ballast. That's why the concept is crucial. In terms of ballast I'd be surprised if everyone wasn't in the 80kg ballpark. The bit you see of the 'tea tray' part of the floor at the front of the chassis is obviously a key place for ballast; you could carry as much as 35kg there.

With the Renault's latest front wing there's space for ballast (1). But you'd not get much more than 5kg in there. It's not an ideal place because it's outside the wheelbase and also higher up than the floor area, thereby raising the centre of gravity.

Being outside the wheelbase will make the car lazier in changing direction - though we're only talking small degrees here, with 5kg of ballast. Some think a bit of inertia is good, that it helps car the car to rotate, but I think it will cause more problems than it'll solve.

Two slots, three or two and a half?

We're seeing quite a few different concepts of front wing. Even forgetting the upper add-on top elements for a moment and looking only at the main elements, we see two-plane wings, three planes and even two and a half.

They're all about trying to control airflow separation - which you'll always get at some stage because there's a wing there with a fairly low point in the middle and that's always going to stall to some degree. Under braking or in a high-speed corner that wing will struggle to get enough flow to keep the airflow at the flap attached.

The flaps are very aggressively shaped and what you don't want is to have a catastrophic stall in the middle of the front wing. So the slot gaps are there to keep it attached, keeping the flaps fed with flow. It will reduce your peak downforce - which comes from the throat of the mainplane.

The whole wing has a low pressure underneath it and the velocity of air through that throat will cause it to separate; the last thing you want it to do is separate over the whole wing so the slots are positioned to keep those bits attached so the rest of the car doesn't suffer as dramatic a failure of airflow.

The Williams of a few years ago had a huge flap and a fairly small mainplane and that's the most catastrophic combination you can have because you fill that flap much sooner and then really struggle to get it reattached.

Obviously a two-plane wing has one slot gap, a three-plane has two. The Renault (2) has a centre section slot on its main plane, making it one and a half. The BMW also has one and a half but its additional slot is on the flap (3). CFD (computational fluid dynamics) will predict where the stall will happen and where it's best to put your slots to make it more controllable.

Upper elements

Looking at the upper elements over the top, only the outer 50-60 per cent of their section is used to create downforce. Those that do stretch across to the nose in the middle are neutral there - their shape is not aerofoil and doesn't therefore create downforce.

The McLaren has an unusual arrangement whereby the upper elements don't actually meet the nose but cross over the top of it (4). This is simply because the McLaren's nose drops down so sharply that this is the only way to have the elements in the right place.

Although they look different from anyone else's, they do much the same job. They give you some extra downforce in themselves, without disrupting the flow to the bargeboards (which is why their inner part is neutral).

On some cars the extra front downforce isn't really necessary but it's allowed them to back off the main flap section, making it less critical. Getting the front wing in its working window is where you'll get the most overall car downforce, as opposed to the most front peak downforce.

Keeping the flow attached feeds the rear downforce-producing devices better. If you put on more front flap angle you can take it out of its ideal working window. It disrupts the airflow further back. You do move the centre of pressure forward but the effect on the rear is bad so you're moving it forward only because you've reduced the rear downforce. It's easy to get tricked by that.

So these extra elements help keep the front wing in its working window, keep you from having to add too much extra flap angle and thereby damaging the rear. You're minimising the trouble caused to the airflow between the bargeboard area that feeds the floor and thereby the rear.

End plates

The most significant thing we've seen here is the move to wider endplates - as seen most notably on the McLaren (5) and Red Bull. Generically you're working the front wing very hard on that outboard bit to give you the front end you need, but then you put steering lock on and the front wheel crosses directly over the hardest-working bit of the wing. You don't want that mid-corner.

It becomes a vicious circle too because the cars tend to understeer in slow corners anyway, yet the slow corners are the ones that require more steering lock. So the more lock you put on, the more you hurt the front aero, the more understeer you get etc.

With a big endplate you minimise this effect although you're surrendering peak downforce because you're giving up some wing area. The trade-off of endplate width versus wingspan probably stops at 6-7cm wide. After that, the wing performance will drop away but the performance with lock on will still be with you. The McLaren has got around 15cm.

There is still a lot of divergence here between single, twin and zero keels. There is variation also on the height of the monocoque and this has very significant implications.

As ever, it's about compromise.

The chassis cross-section is defined by the FIA. It can be whatever height you want but it has to be symmetrical around the car's centre line. Anything else you add to it is aero blockage. The air that comes between the front tyres is what you can use. There are three things you can do with that: 1) brake cooling, 2) engine cooling, 3) downforce.

The more you have to use for brake cooling the less is available for the other. Any blockage you put between the front wheels is an area with less flow, so you want to reduce that to the minimum. Hence the appeal of twin and zero keels.

But the compromises are stiffness and suspension geometry. The initial windtunnel study makes twin keel look very good, but when you make it structural - make a package that would stand the loads - it disappears on you a bit. But in looking at the whole range from a traditional single keel to a zero keel we're talking single figures of kg of downforce at 150mph. So it's not make-or-break.

Toyota has taken the most extreme approach. Not only has it opted for zero keel (6, right, compared to '06 single keel), but it's made the chassis front higher than anyone else's, lifting the underside up high to maximise the airflow. But when you move that section of the chassis up by 40-50mm the centre of gravity goes up maybe five per cent and the geometry becomes flawed.

Watch it on the track and its front end is inconsistent to its driver input. When the driver has confidence in it it's pretty quick, but when he hasn't it's horrible. I believe that's coming from the compromises in the suspension geometry caused by having the lower wishbone as high as it is in order to have the zero keel.

With a grooved tyre, it's very different from a slick: you don't have a continuous load/grip relationship, each bit works on its own. With a suspension geometry like that the tyre will be rolling around more and you will get more contact patch movement across the tyre. Everyone has that problem, but the higher you have that wishbone the more you will get it.

Pushrod mounting

Ferrari was the only team with its pushrod mounted underneath the front wishbone (7, compared to R27 inset above), but since Magny-Cours Honda has adopted an almost identical layout. Having the pushrod mounted on the upright as opposed to the wishbone does more or less the same job for the suspension.

But it means you can change the characteristics of steering lock, change the weight transfer across the car on steering lock, which is good for low-speed corners and can reduce the understeer. It changes the feeling for the driver, so you have to be careful. But you can do a lot with it and it can actually give a feeling of less understeer even when it's actually the same.

McLaren seems to run its car more compliant than the Ferrari. It seems to ride the kerbs better whereas the Ferrari seems to have reasonable grip, and seems a very stiff, stable car.

The races where Ferrari wasn't good in qualifying it's often been okay in the race. The races where it's been good in qualifying, it's been dominant in the race.

This is a reflection of the fact that Ferrari doesn't get its front tyres working well enough over one lap in qualifying. And the McLaren does.

Furthermore, because Lewis Hamilton is able to run with quite a bit of oversteer it is able to run his car with quite a lot of front grip in the faster corners. That will certainly help. The McLaren uses the tyre quicker, heats it up quicker and gets access to its performance more instantly.

If you have a car that's very stiff it tends to be a car that wants to understeer. The worst thing you can do to a front tyre when you're trying to warm it up is slide.

You want to load up the carcass and make the tyre's structure warm. Whereas if you slide it all you're doing is heating up the surface and not getting any lateral load into the carcass where you need it.

The Ferrari tends to induce that sliding of a cold tyre, being a stiffer structure. The McLaren, being a bit more compliant, can get the weight transfer going a bit more and get that front end to bite.

Believe it or not, the Ferrari front-wheel 'hub-caps' (8), that remain stationary as the wheel turns, are theoretically part of the front brake ducts. This is to get around the regulations concerning the wheel rim having to be of a homogenous material.

These carbon bits are bolted onto magnesium rims and so are classed as an extension of the brake ducts, to which they are linked.

At the front, where you need a lot of brake cooling, you want your ducts as small as possible to reduce drag.

At the rear ,where you don't need all that much cooling, you want the ducts as big as possible so they can double up as downforce-generating devices.

This latest development ensures the front brakes are cooled in such a way that they keep airflow from spilling out undirected where it would otherwise increase drag. Instead it directs it out into the low-pressure area just behind the tyre contact patch.

By doing this it allows Ferrari to use a smaller duct intake because it's made it more efficient by directing it in under higher pressure and exiting with low pressure. It will reduce the drag of the car because the wheels account for around 35 per cent of a car's drag at no cost in downforce.

The big thing here is that it's become feasible to measure downforce effects of devices that aren't part of the car's sprung mass. It's only recently that windtunnels have had the bit of kit required to do this. CFD has helped too.

Front brake ducts three years ago lost you a lot more downforce than today. The development we see on the Ferrari is just the start.

There will be some very complex gizmos on the outside and flow devices. Everyone needs brake ducts but how far do you take it: when is a brake duct not a brake duct but an aerodynamic device? They'd be far better writing the regs to limit this pretty soon.

The further forward you put the bargeboard the more effect it has on the front wing. Conventional bargeboards in front of the rads give almost equal downforce distribution front/rear. The two tend to go in whatever direction they need to go to get the car balanced. Most cars have an element of both.

The leading edge at the front of the sidepod is all about inducing as high a velocity as possible to the flow, so you see lots of bits hanging off acting as flow conditioners to the underfloor, to allow the diffuser to do the maximum work on it, acting as it does like a multiplier. There's lots of different solutions (9) all trying to achieve the same thing.

Most people seem to have the shoulder wings on the top front corners of the sidepods. They pick up the wake of the front tyre and tidy it up.

McLaren has gone from a fairly conventional twisted rear wing (10) that reduced the load at the draggy outer ends and increased the load in the middle section, which is the downforce producing area, to a much more complex assembly (11) introduced in the last couple of races.

The rear wing runs at about the same overall efficiency as the car so it's an area where anything you get from it translates three for one. So you're not going to get any big advantages from it. You will find more downforce, but in the process you'll create more drag.

But you will get reasonably big downforce numbers, even if the drag goes with it. It won't necessarily make it a faster car, but it will help make better use of the tyres.

The front wing and the diffuser are to me the bits that are the real car. The McLaren has a very aggressive package of front wing assembly - narrow span, big endplates - to give good characteristics when on steering lock. So they have a good consistent package on the front.

This helps feed the diffuser. Look at McLaren's diffuser in concert with the front wing and you get a good idea of its aero philosophy: its interpretation of the regulations in terms of the height of the inner part of the diffuser is pretty clever. Some bits of the gearbox have become the diffuser, in essence.

McLaren has optimised it to a level in the concept of the car - the gearbox being quite a long lead item. Ferrari has a pretty good package in that area too, but perhaps not quite as good as the McLaren.

It's getting tough to find any more. It's been an area of extensive research in the past because it gives a great return of 10:1 in terms of downforce:drag.

Looking at some of the other teams, they have a very basic diffuser. The Honda has some bits that I'd question. It's the most basic car in the pitlane. Honda has gone past the stage of being far too adventurous, looking at the grey areas too much.

They've stopped doing that but haven't done anything on the real car either, have just left it sitting there. It looks like a three-year-old car now. It's a big team with a lot of money; I don't understand why it doesn't have more vision and more ideas.

There are a lot of different mechanisms out there. Minimum torque reduction would probably be a more accurate term than seamless shift.

Ferrari and Renault have introduced their systems this year without problem, but Red Bull has suffered big reliability problems with theirs. The Red Bull gearbox is packaged to the limit and it's probably suffering because of that.

People talk of finding 0.3sec per lap, but I have trouble believing that. The worst gearchange I've seen was 30 milliseconds and the difference between that and the best didn't find you anything like 0.3sec.

The area most teams have pushed hardest on is the rear bodywork. A lot of work has been done in getting the Coke bottle squeeze even tighter and reducing the gearbox size to enable this. The undercut of the sidepod feeds this; you need both working together.

Get a good undercut in the sidepod, narrowing the bodywork, and it helps get good attachment along the Coke bottle, which in turn will help produce dramatic downforce improvements. Also, you will actually get good downforce from the sidepods themselves because the velocity of airflow on that surface is 15-20deg to the ground. This has been the big area of development and some teams haven't done it as well as others.

Ferrari (12 - compare undercut sidepods to Toyota in illustration 6), McLaren, Renault and BMW have taken it a long way, while Honda and Toyota still haven't really bought into the philosophy. The numbers when you first look at it might not be big, but with this direction you find a succession of small improvements - added all together, it makes a big number.

The heat rejection figures will be lower than last year because of the 19,000rpm limit. An FIA pop-off valve that limits how much pressure you can have your cooling system running at has capped how small you can go on radiator size.

It used to be that you could pressurise it - the boiling point goes up with pressure so you could run the water as hot as 150C. The pop off valve runs at 3.5 bar which lets you run about 125C. We've seen McLaren consistently able to sit longer with its engine running. Partly that's electronics but basically it's the quantity of coolant in the system.

We want to hear from you!

Let us know what you would like to see from us in the future.

Take our survey

- The Autosport.com Team

Read and post comments