Wind tunnels: Lifting the lid on F1′s biggest secrets

In the secretive world of Formula 1 design, wind tunnels may be literally the biggest of all secrets.

These huge, purpose-built buildings are hotbeds of innovation and absolutely crucial to the success of a car.

And, because of that, wind tunnels are the ‘Area 51s’ of Formula 1, large in scale but shrouded in understandable secrecy (the photo is of a General Motors wind tunnel… it’s tricky to get into an F1 equivalent with a camera).

With the regulation changes to F1 for 2017 making it a sport that will rely much more heavily on aerodynamic prowess, wind tunnels are going to be even more important.

But, in terms of innovation, wind tunnels can help teams build winning cars only if the tunnels themselves are innovative.

So good are the wind tunnels used by F1 teams that some of the technology has been adopted by the aerospace industry, to help improve understanding of how aircraft behave, particularly during takeoff and landing.

So just how big a deal are these big wind tunnels? Here are a few areas in which the scale and technology involved show that F1 wind tunnels are a very big deal indeed.

A 23-tonne engine. Yes, 23 tonnes

The fans in F1 wind tunnels are huge and the motor driving them has to be equally huge.

For example, Sauber’s fans are powered by a 3MW (3,000,000 Watts) motor, and can push out 1.2million litres of air per second at up to 80m/sec – if you want to mix up your measurement units, that’s more than 1.5tonnes of air travelling at almost 180mph.

The motor alone weighs 23tonnes, 19tonnes of which are copper winding. That’s equivalent to about 150 modern F1 engines (or 8-10 African forest elephants, pictured).

A machine this size will shake, no matter how well it is built. To minimise the effects of this, the motor is planted on huge concrete blocks.

Then, under a layer of springs, there are even bigger concrete blocks.

Finally, to make sure any vibrations which do get through are isolated, this whole structure sits on foundations that are independent of the rest of the building in which it is housed.

This is important – when you’re examining the effects of tiny aerodynamic tweaks on a scale model in a wind tunnel, the vibrations from such a big motor could destroy the accuracy of your measurements.

This is no ordinary tunnel…

The tunnel that leads from the fan to the working area gets constantly wider. That slows the airflow down and reduces friction between the air and the tunnel walls (aerodynamic forces increase with the square of speed, so a small reduction in airspeed can result in a substantial reduction in friction).

But don’t worry, the air will soon be speeded up again. First of all, though, it has to be guided around a couple of corners by vanes.

Corners? Yes, wind tunnels are generally closed systems – the wind that passes over a model F1 car is travelling at up to 50m/s (the maximum speed allowed) and it would be incredibly wasteful just to let that wash out through a window.

So this wind is recycled – it is channelled back to the main fan, which now only has to accelerate it a little, rather than moving it from a standing start.

That’s why the whole tunnel is rectangular or oval in shape.

There is also a ‘settling chamber’ where the air expands quickly, reducing turbulence, and mesh grids to further cut turbulence.

And there’s a huge radiator – the spinning fan and the friction of air on the tunnel walls creates a lot of heat, and so the water-cooled radiator can easily be 9m x 9m in size, drawing away heat without causing too much of a blockage to the airflow.

Finally, just before the tunnel reaches the working area, it contracts in size, speeding the air back up just as the water from a hosepipe speeds up when you squeeze the end closed a little.

The rolling road… somewhat more innovative than a conveyor belt

Inside the working area of a wind tunnel, it’s important to be able to replicate how the ground behaves in the real world, not just the car.

You can’t just stick the car on rollers, like you’d find in your local garage, because that wouldn’t replicate the effects of the car travelling over a track at speed.

When F1 teams first started using crude wind tunnels, the rolling road was simply a big belt, like a scaled-up version of the belt you dump your groceries on at the supermarket checkout.

Not surprisingly, this had its limitations. Not least of which was that the belt was sucked up by the aerodynamic effects of the car, causing all sorts of problems – especially when the belt was sucked up high enough to make contact with the test vehicle.

Being an innovative lot, the wind tunnel experts installed vacuums to suck the belts back down. This had the unfortunate effect of increasing friction on the belts and they wore out far more quickly.

Then a US company, called MTS, introduced F1 to a genuinely innovative solution.

They used a steel belt supported on air bearings – the ‘conveyor belt’ literally floated on air.

The genius of this was that, the harder the car pushed down on the belt, the more air resistance pushed back.

So efficient has this system become that the weight of a scale model (50-60 per cent of life size) and several tonnes of aerodynamic pressure can be tested at high speeds without stressing the test rig.

This is particularly important when it comes to examining how turbulence around rotating wheels affects aerodynamics on the rest of the car. Because F1 is becoming an ‘aero formula’ in 2017, innovation in this area may well be the difference between success and failure.

About the smoke we use to check airflow…

Those photos you see of boffins examining smoke streaming over an F1 car in a wind tunnel… they were probably taken for PR purposes and little else.

Apart from anything else, think about how much effort F1 teams put into tweaking the tiny ‘turning vanes’ on a front wing, as well as the myriad other minuscule aerodynamic tweaks that we hear about before every race.

Sticking a smoking wand in front of these sensitive aero parts would disrupt the airflow sufficiently to make any measurements useless.

And, anyway, what could the smoke tell us that a bank of hundreds of electronic measurements couldn’t?

If you were able to watch an F1 wind tunnel session (they are rather secretive) you’d see a model that was 50 or 60 per cent as big as a real F1 car, kept in position by a column that (usually) rises from the midpoint of the vehicle.

You might be able to see what are known as ‘six-component load cells’, which measure the overall forces acting on a car. Aerodynamicists want to know exactly what the downforce, lateral forces, drag figures are, and how they vary from front to rear.

These ‘load cells’ also allow them to analyse what happens as a car pitches and yaws.

Similar but simpler load cells will be found attached to the wings and axles.

Under the ‘rolling road’, four contact patches will measure the forces acting down through the wheels. This is a particularly difficult set of measurements to take, under a surface that is floating on air and travelling at high speed – a lot of experimentation has gone into these measurements to make sure that instruments are calibrated precisely, despite the challenges.

Anything that can move will be measured – wheels, suspension, flexible car body parts – and there will be hundreds of sensors checking air movement and pressure.

Each round of wind tunnel innovation brings new levels of accuracy, and can bring vast quantities of new data.

Often, the challenge for teams is to be able to make best use of the huge amount of information pouring out of such sessions – interpretation of data is as important as gathering of data.

Not so fast… there are rules about testing

Wind tunnels are expensive to build and run, and rules have been put in place to limit how much time (and therefore money) teams can spend in such facilities.

They are not allowed to use any models bigger than 60 per cent of life-size, and can test only at speeds up to 50m/s (112mph).

The rules around testing are complicated – one ‘unit’ of testing is regarded as being one hour in a wind tunnel with the wind on (‘on’ means greater than 5m/s).

One unit of testing is also regarded as being one terraflop of CFD (Computational Fluid Dynamics … aerodynamics done on a computer) power, and teams have to supply the FIA with details of their computers and, indeed, the chipsets powering these computers.

Confused? Teams are allowed to ‘spend’ a limited number of both these units – in 2015, it was set at a total of 25 per week. There are more restrictions, far more, but you get the idea.

What it all boils down to is this: if you want to innovate in F1 aerodynamics, you have a limited time in which to test your ideas, no matter how big your budget is.

And, with aerodynamics potentially becoming more significant than engines in the sport, that means you’ve less time in which to create a winning package.

In the 2017 season, that means the races between aero specialists may well be more exciting than some of the on-track races – it’s such a pity that they’ll be conducted in secret.