Your Flexible Friend

Wings and other bendy things.

Don't you wish sometimes things would just stay put? Like that bookshelf that strains alarmingly at the mounting brackets or the car keys that aren't in the place you definitely left them last night. Well, engineers often feel the same.

We spend enormous amounts of energy fixing parts to F1 cars so that they won't fall off or move out of position. No one likes components becoming detached from cars mid-race; it turns something normally innocent into a 200mph missile. Also, most parts on the car are fit to a very high precision. Suspension parts, for example, are built on custom designed jigs and use laser metrology systems to fit them to ensure they result in the wheels being just where they were intended to be.

With this obsession with parts being in the right place, why then would F1 teams want to allow some of them to move out of position? Why are there regulations which define limits of bodywork flexibility and why do these in particular get so much attention?

The answers open a window into one of the most intriguing aspects of competitive sport - the desire to extract as much performance from every conceivable aspect of the sport as possible.

Wheels - definitely preferable for these to stay attached to the car

Much of the deformation or flexibility of car components as it drives at speed is a giant headache for the regulators of F1, the FiA. The problem is that nothing physical is infinitely stiff so the components on the car will respond to the sometimes significant forces to which they are exposed by moving - bending or shifting around their fixings. The extent to which they do that can be controlled through design but it is inevitable that to make parts deform less, they will have to be heavier and designers are obsessive about keeping weight down.

In the majority of cases, we do not want parts to move relative to the rest of the car as this puts our carefully designed geometry out of whack. With the suspension, the geometry of the members which support and control the position of the wheels can be predicted by modelling its kinematics through the normal ranges of motion. Great care is taken to keep the wheels, and therefore the tyres, in as optimal a position as possible around the lap. But under the huge lateral and accelerative forces generated, the suspension legs and supporting upright will bend out of position, affecting in turn the position of the wheel.

In rare cases this can be used to actually improve the behaviour of the car, for example to align the wheels differently under braking to increase stablity of the car. Normally however the wheels don't move to a 'better' postion, they go somwhere you don't want them to. Luckily, a lot of this behaviour can be predicted, firstly through simulation and then by what I would describe as 'wanging it around' on some kind of rig. Then there is some judgment to be made to balance the requirements of low weight with the disadvantages of the wheels moving out of position.

The first F1 car I had any involvement with was the Jaguar Racing R3. This famously suffered from an issue with the rear suspension which resulted in Eddie Irvine describing it as "like driving a shopping trolley". Unfortunately (for us) the new, complex suspension was significantly less stiff than expected from the structural calculations; as a result the rear wheels tended to go where they felt like going, rather than where the designers intended. With approaching 1000bhp going through them, you can see where this may quickly become a problem. Drivers in the media sometimes call this behaviour "lively" - they normally mean "actually quite dangerous - fix it immediately".

There is, however, one area where allowing movement of parts, whether through bending or deforming their whole shape, can be rather beneficial - aerodynamic bodywork.

Move your body(work)

Regulations specify the posiiton of all of the bodywork surfaces on the car which are 'wetted' by the air as the car moves. This allows something of a level playing field and ensures all the cars look vaguely similar - or sometimes very similar, but that is for another time...

There are also prescriptions in many areas for the amount any of the bodywork can deflect, along with some detailed tests to ensure these requirements are met. We also have an overriding principle which forms part of the regulations - bodywork should not be allowed to move relative to the rest of the car. However, as we know, nothing is infinitely stiff and so it would seem every car will contravene this rule as soon as any aerodynamic load is applied.

What every engineer understands this to mean in reality is that the bodywork should not move too much under normal loading conditions. Of course, that is highly subjective and a bit unsatisfactory so we rely on the FiA to police this in a reasonable way. In the past this has been known as the 'finger test' - if I come along to your floor edge and can bend it down 50mm with a press of one finger, you are going to be in trouble. Often it is genuinely difficult to get bodywork to be stiff, particularly when it is hanging a long way out from its associated mounting point. However, F1 engineers are highly skilled and capable of making lightweight carbon fibre parts almost unbelievably stiff. Therefore, if something moves more than a millimetre or two under a hard press of a finger, it has been designed to do so.

Why, you might ask, would this even become a conversation? Surely a car is best staying in roughly the shape you designed it rather than wobbling about all over the place?

There are two main things (there are some others) that aerodynamicists would like to achieve by deflecting parts under load:

1. Get even more aerodynamic load from parts being closer to the ground

2. Lose aero load from parts bending - usually lying flatter

Front wings, for example, augment their downforce-generating powers with ground effect - something which gets stronger the closer the wing gets to the ground (up to a point). Therefore, anything which allows the wing to drop at speed will give more front downforce. The faster the car goes, the more downforce that wing will create.

A current style front wing - it’s the wing. At the front.

Let's say we wanted to do the opposite though - imagine that we had a surfeit of front downforce at speed and wanted to actually lose some as the car speed increases. We can achieve that by stopping the wing from dropping down towards the ground as much as possbile and then allowing the rearward wing elements to flatten themselves off - what we call 'backing off'. Current F1 regulations allow for a four-element front wing where the rearward two pieces can rotate along a portion of their span. If these rearward two wing elements are allowed to rotate nose-up under increased aerodynamic loading, the result will be to reduce the amount of overall downforce from the front wing compared to them staying in their designed position.

As another example, consider the rear wing which tends to produce a significant amount of downforce but in a relatively inefficient way. That is, it generates a lot of drag along with the downforce. Could we imagine a way of getting the maximium downforce out of the wing at low speeds when cornering speed is paramount and then reducing the attendant drag at top speed on the straights? Well, yes by allowing the wing to 'back off' - rotate nose-up - either through rotation of the rearward element or from the whole assembly moving.

This could be done by clever design of the wing attachment points, permitting some movement of the endplates which bookend the wing and form part of the mounting, or even allowing the whole assembly including the mountings to move in a controlled way.

Indeed, this is done by every F1 team and is perfectly legal... up to a point.

The reasoning is this: we know there is a tension between making parts structurally sound and reasonably stiff, and making them light. We will therefore establish an acceptable limit for deflection of these main parts such that some delection is allowed but no more than that. There are several of these limits and they form part of the technical regulations.

What do F1 teams do with these regulations? They turn the regulated limits into targets. Suddenly what was a practical 'go no further' number is now something for the designers to aim at.

Among the many quandries the FiA find themselves having to navigate, the problem of how much deflection is too much is one of the worst. Testing of parts for excess movement can only realistically be done statically. That is, they can only hang weights from things or press them with hydraulic rams whilst the car is sitting on the ground. Also, it is notoriously hard to replicate the kind of loads a car sees as a result of air flowing over it. Prodding the car in one place is just not the same thing.

Often, the result is that regulations are reactive rather than proactive. A common example is with front wing deflection which can be seen quite clearly by the TV cameras cars have to run on the front of the chassis. As an aerodynamicist, there is a sinking feeling associated with seeing onboard footage of your front wing moving significantly as the car speeds up. If it moves more than your competititors' ones do, you know you are in for a conversation with an FiA representative some time soon.

Normally, with every new set of regulations there is a period of time when the deflection limits are set but the teams have yet to found ways of reaching the limits. Inevitably they will find those limits in the end though and perhaps even find clever ways of passing the relevant tests but gaining more of an advantage when the car is on circuit. This is when eagled eyed fans and engineers from rival teams will pore through TV footage and shout loudly that they see an outlier.

We all love rules

The reason this gets so much attention is that the potential gains are huge. Difficult car balance characteristics can be helped greatly with some smart aerodynamic changes with varying speed. A car that loses more drag as it goes faster will be more competitive than another that holds onto it. Conversely, ignoring the deflection of your car's bodywork will result in something on track that doesn't resemble your simulations, particularly at high speed. There may be any number of small issues that creep in when parts of the car move relative to your original designed position.

From a team perspective, they are always on the lookout for other teams crossing the invisible line from inevitable and acceptable deflection to a clear breach of the principle of no bodywork moving relative to the rest of the car. If they find an example on another car they will complain bitterly and loudly. If someone accuses them in return they will point to having passed all of the required static tests and say that their car is perfectly legal. And so it is, year after year.

It is difficult to see a resolution to this in the near term. Regulators are the most motivated to find a workable solution whcih can be consistently applied no matter the regulations. It would seem that this could only realistically involve on-car measurements whilst it is in motion. Modern measurement techniques certainly allow this, and have been used recently to determine the motion of F1 teams' front wings before imposing more stringent static tests. It would seem inevitable that this kind of policing will become more prevalent in the future.

Another view is this: cars need to be safe - that is the number one priority - but if bodywork or any other component can be made to deflect without becoming damaged or detached, that should be fully permitted. Imagine a series where there were no limits on aerodynamic deflections and the cars were constantly changing as they went around the track. It would be up to teams to find the limits of acceptability and they would have to take full responsiblity for keeping the car intact.

The truth is that this would result in such an increase in overall performance that the cars would quickly become dangerous to drive. Such is the power of 'bendy bits'.