Modifying Under-Car Airflow, Part 1

This article was first published in 2005.

Read any web forum or newsgroup big on automotive modification and you’re sure to find at least occasionally some discussion of car undertrays. Someone will ask about directing the way the airflow moves under the car (usually in order to reduce lift) and a whole heap of ‘experts’ will spring to their keyboards. “Ah,” they’ll say, “what you need is an undertray complete with diffuser.” They’ll then point to race car aero textbooks like Competition Downforce or Race Car Aerodynamics. Trouble is, when the original poster actually reads those (very good) books, they’ll find next to nothing that is relevant to road cars.

Don’t agree?

Well, take just the fact that the race cars have very stiff suspension, so ground clearance remains far more constant than in a road car. Or the fact that the rake (the angle of the underside of the vehicle to the road) changes very little in a race car but is all over the place in a road car – especially if passengers are carried. A final point - aero drag is of nearly no consequence in a race car (pretty well all of ‘em have terrible drag coefficients) but in a road car, it is very important.

The end result of all that is that the person with a genuine interest in improving the underside aero of their car ends up without a lot to go on. In fact: just theories and practice which are usually near-irrelevant.

Obviously, with this sort of prelude, we’re hoping to achieve something different. We’ll will concentrate strictly on road cars, and the majority of technical resources that we’ll use are sourced from road car literature. The modifications that we’ll do are also on road cars. However, a word of warning. Without completely redesigning the underside of your car – and that includes potentially re-routing the exhaust, changing the suspension stiffness and developing custom all-enveloping underside panels – you are very unlikely to generate a large amount of downforce. It’s just not possible to achieve that with only tweaks.

However, you are likely to be able to either reduce lift or decrease drag. We choose to reduce drag sufficiently to give a real and measurable decrease in open road fuel consumption without noticeably affecting cross-wind or high-speed stability. The fact that we did it on a car that is already highly aero-optimised from the factory shows what is possible.

But let’s start at the beginning.

Underflow Aero

In the old days it was pretty well a given – the lower the front spoiler, the better. This idea was predicated on the fact that the underside of a car is rough – there are suspension and exhaust and engine bits all hanging down into the airstream. This roughness caused drag – so better to have the air flow over the smooth exterior parts of the car than battle its way along the underside.

Despite the fact that a lower front spoiler increased the frontal area of the car (and total drag is worked out by the frontal area [A] multiplied by the drag coefficient [Cd]), the increase in area was more than compensated for by the decrease in drag.

However, cars are changing. The exterior panels are now being shaped to reduce drag – and cars of the last 10 years or so are achieving this very well. As a result of this excellent exterior body design, the attention of designers is turning to the hidden underside. (Something, it must be said, that companies like Porsche, Volkswagen and Citroen having known about for more than 50 years!) While underside aero refinement is still absolutely primitive (in production cars, underside design is at least 30 years behind exterior aero design), progress is being made.

Examples of undercar flow design include in-fill panels to shield off cavities, deflectors located ahead of wheels (here highlighted on the Honda S2000), and small undertrays.

But in these cars with relatively slippery undersides, a low front spoiler can actually cause an increase in drag – the underside is smooth enough that the increase in frontal area is greater than the decrease in drag resulting from more air passing over the car than under it. So a low front spoiler is no longer necessarily good for decreasing drag – it depends on how good the existing under-car aero is. And for lift? Well, that depends a lot on where the spoiler is located – something that we’ll come back to in a moment.

So things are getting more complex.

A seminal paper was recently written by Volvo’s Dr Simone Sebben (SAE paper 2004-01-1307). It covers the computer modelling of undercar airflow, concentrating on the changes in lift and drag caused by variations in the size and positioning of deflectors located ahead of the front wheels. (We’ll come back to these deflectors in more detail later.)

Some of the data presented in the paper concerns the various component drags that make up the total drag of the car. (The car is unnamed but it is likely a soon-to-be-released Volvo, ie a current, aerodynamically designed car.)

As a percentage of the total drag coefficient of 0.29, the various elements that make up this drag are:

Parts	Percentage of Total Drag
Cooling package (including radiator, intercooler, oil cooler, etc)	33.4%
Exterior	31.7%
Front wheels	13.1%
Rear wheels	6.9%
Floor	6.9%
Rear Axle	3.1%
Engine	3.1%
Front Suspension	1.4%
Exhaust	0.7%

This is a fascinating breakdown. It can be seen that the cooling flows for the radiators create a huge amount of drag – at about one-third of the total, the greatest single component. The exterior (that’s all the exterior!) creates a little less than one-third of the total, and the remaining 35 per cent is caused by the underside of the car.

So what drag reductions are possible on the underside of a typical modern car?

As this graph shows (click on it to enlarge it), the greatest contributors to underside drag on the Volvo are the front wheels, followed by the rear wheels and floor. You might be puzzled by the amount of drag caused by the front wheels. However, another brilliant paper (yes, still fantastic despite its age) is SAE 900317, published in 1990. It’s on the aero development of the Opel Calibra – a car with a (still stunning) Cd of 0.26.

In the case of the Calibra, it was found that air reaching the front wheels was being deflected out the sides of the car, so increasing the size of the wake (the wake is the area of disturbed air dragged along behind the car – the larger the wake, the bigger the drag). It helps to remember that the wheels are not static – instead they’re rotating rapidly, with a huge swirl of air pulled around with them. When the air flow caused by the forward movement of the car reaches this swirling air, fireworks can happen.  

In the case of the Calibra, the centre section of the front spoiler was lifted to allow more (yes more!) air to flow under the middle of the car, so making the total flow more parallel with the car’s long axis. This reduced the amount of air being deflected outwards by the front wheels, reducing the size of the wake and so drag.

So where’s lift in this discussion? Well, compared with underside flows, lift is rather an exterior surface thing. As this diagram for the Mazda RX7shows, there are plenty of curved surfaces on the surface of a car that generate lift (indicated here by the yellow). However, it’s still important to realise that the underside of a car does have a bearing on lift – effectively, if there is a pressure build-up on any body surface parallel to the road, lift is being generated. In other words, a front spoiler positioned back from the leading edge of the car will have a horizontal surface positioned ahead of it (eg the front portion of an undertray) on which pressure can build up, causing lift. This effect was clearly measured in 'Real World Spoiler Development'.

Conclusion

So where does all that leave us?

• In a modern aero car, a low front spoiler doesn’t guarantee a reduction in drag.
• About one-third of all drag in current cars results from underside flows.
• Most of that undercar drag is created by the front wheels, followed by the rear wheels and floor.
• Lift is caused mostly by the exterior surfaces, although any positive pressure on flat undercar surfaces will also generate lift.
• A front spoiler positioned back from the leading edge of the bumper is likely to cause lift.

Using that information, next week we’ll modify and test a car.

Cooling Performance There is a strong relationship between the aerodynamic flows under a car and the cooling performance of radiators, condensers, oil coolers and intercoolers. To gain greatest airflow through these coolers, there must be a high pressure differential across them – the air pressure in front must be greater than the air pressure behind, so that flow will occur through the heat exchanger. Since in nearly all cases the exit air path is under the car, the undercar pressure must be kept low if cooling system flow is to be good. AutoSpeed’s ‘Undertrays, Spoilers & Bonnet Vents’ 3-part series covers the modification of a car to achieve a major improvement in flow through the heat exchangers, concentrating on the intercooler. A large proportion of the gain was achieved with a new front undertray. The series starts at Undertrays, Spoilers & Bonnet Vents, Part 1.

Other stories – AutoSpeed has covered a lot of aerodynamic topics, including DIY tuft testing, real world front spoiler testing, the aero development of specific cars, patented pneumatic aerofoils, wind tunnels, and other topics. Do a site search to find them!

Interested in do-it-yourself car aerodynamics? You're sure then to be interested in the Amateur Car Aerodynamics Sourcebook, available now.