Get instant digital access.
Subscribe Now!

Wing It: Simple Talk on the Complex Subject of Aerodynamics

story by steve stafford

Steve Stafford is an aerodynamic engineer with Toyota Racing Development, where his current beat covers the NASCAR Nationwide Series. He has done aero work at nearly all levels of motorsport, from autocross and club racing to Indy and the Rolex 24 At Daytona.

On a fall day in 2001, amid the greater concerns of society, I was also worried about drag. What can I say? I’m an aero engineer.

I happened to be in Germany at the time, and I had a specific concern: How were my buddies and I going to get my VW rental car past the 130 mph barrier on the Autobahn between Dresden and Lausitz? That ribbon of unlimited-speed highway allowed us to reach terminal velocity for several minutes before we had to slow for a speed-limited section.

The first trip along this stretch of highway had shown that the rental car was capable of 126 mph in stock trim. Now the work could begin, although there were two restrictions: a limited number of tools plus the requirement that the car not be permanently altered.

We came up with a plan and implemented one step each time we crossed this section of highway. Pulling in the side-view mirrors, removing the radio antennae, and shutting down all engine accessories helped us reach 128.5 mph.

Thinking back, we might have nudged 130 if we removed the luggage rack crossbars. The German beer consumed during the planning sessions must have made us overlook that idea.

Methodical Approach

Our approach to increasing the VW’s aerodynamic efficiency was focused on drag by way of two methods. The first was to reduce the drag by pulling in the mirrors and removing the radio antennae—taking off that luggage rack would have helped here, too. The second method was to overcome the existing drag by reducing the engine accessory loads and freeing up more power. Other aerodynamic goals can include increasing downforce and generating more stability, usually longitudinal.

No matter what the particular goals, any aerodynamic testing can be tackled using the same approach. It doesn’t take a huge budget or access to wind tunnels to make aerodynamic gains. You simply need a basic understanding of aerodynamic principles, a detailed plan of attack, and a method to test the changes you make.

Don’t forget, however, that most aero changes will affect more than one aspect of a car. Increasing downforce nearly always results in more drag, for example.

As aerodynamicist of your project, you first need to get a better idea of your starting point. This technical analysis requires some math, but we can get through most of the ideas without crunching any numbers. Here are the important things to know:

• Most production cars generate downforce at the front axle and negative downforce (also known as lift) at the rear.
• Drag—that’s the air resistance—and downforce increase as speed increases.
• The downforce and drag will change with suspension movement, including both roll and compression.
• Tires typically generate lift when they rotate.

Seeing Air

Before we can discuss aero improvements, we need to visualize what’s happening as a car cuts through the air. The fact that air is usually colorless makes this difficult.

Most people only picture a car’s side view when thinking about aerodynamics, ignoring the fact that a vehicle is really a 3D object. Additionally, the air moving over the surface is probably not going to flow smoothly everywhere.

The speed of the air bypassing the car can range from the car’s road speed to a complete halt. To further complicate matters, in some areas the air can actually flow against the car’s direction.

So, how can you find out how your car is interacting with the air? The simplest method is to apply a flow visualization fluid and see how it streaks when the car is at speed. The trick is using a liquid that won’t dry too quickly and won’t run too easily. It also has to be thin enough to streak.

The recipe that I use is safe, easy to make, and can be mixed for cars of any color. The ingredients are water-based powdered paint—you can find it at any arts and crafts store—plus 3-In-One Oil and a little bit of rubbing alcohol. This mix provides good results, but it’s important to achieve the right consistency: not too thick, not too runny.

The testing process is easy as well. First, use a small paintbrush or a syringe to apply the solution to the car’s body. I often make a line of dots ahead of the area I want to study. Then I simply go for a drive and decipher the information upon my return.

You want to look at the length and direction of the streaks. You can even snap some photos to save these “measurements” for later.

Here’s what’s going on: The longer the line, the faster the air is moving over that area of the car. The direction of the streaks will be pretty easy to decipher.

You should use this information to correctly line up ducts and also find the high- and low-pressure areas on the car. (Long streaks signal low pressures, while short streaks or no streaking at all signal high pressure.) This process can also be used to test new equipment. Do the paint test before and after installing a piece of hardware to see what’s going on.

Downforce: Traction From Air

A car encounters two aerodynamic forces at speed. Let’s start with downforce: It’s simply the pressure directed toward the ground, and it helps increase traction between the car and the road.

Since most production cars generate almost no downforce at the rear, the balance of downforce between the front and rear first needs to be considered. This balance is the percentage of front downforce over the total downforce. For the typical production car, the balance is nearly 100 percent, meaning the front tires receive almost all of the downforce.

Increasing the rear downforce balances the chassis. It also adds to the total downforce. If there’s a disadvantage, it’s that increasing downforce also increases drag. We’ll get to drag in a bit.

Downforce can be added to the rear in a number of ways. Some methods are very visible, like adding wings and spoilers to the back of the car. Others are more subtle, like increasing the suction under the car or upping the rear ride height.

Most of us will simply add a rear wing or spoiler. It’s an effective move, and the amount of downforce can be adjusted by changing the wing angle or spoiler size. Since these pieces are placed at the top of the car, they’re also less influenced by ride height variations and body roll.

Floors, diffusers and rear ride height are very sensitive to roll and ground clearance, especially at the front of the car. However, the easier options are usually not as efficient in drag.

Front downforce isn’t too hard to generate, since most cars already have some. However, more can usually be added easily. Lowering the front will generally add downforce; so will adding a valance or chin spoiler. The latest “cool” aero touch is to use a splitter.

All of these methods are working from the same premise: Lower the point where the air splits over the top and bottom of the car. This point, known as the stagnation point, continues to move downward as cars evolve—just look at the shift toward lower radiator intakes and sloped hoods.

Fitting a flat pan or diffuser under the front of the car can also increase front downforce. This method smooths out the under-car airflow and also generates downforce as more low pressure is formed between the car and the ground.

Sealing the hood seams can also increase front downforce, as this prevents the pressure in the engine bay from leaking to the topside of the car. (In some cases, however, the leakage of the underhood pressure can be used to reduce the front downforce and help move the balance rearward.)

A last-resort method for increasing front downforce is to add dive planes: really short, steep wings that stick out from either side of the front bumper. They do make downforce, but they’re notoriously draggy.

Balancing the Front and Rear

Before blindly adding rear downforce, the balance between both ends must be considered. The typical production car is set up to provide understeer at the limit of handling, so decreasing the front downforce or adding more rear downforce makes the car understeer more as speed increases. At the extreme, you can wind up with a car that oversteers at low speeds and understeers at high speeds.

This balance problem is not easily solved, and even the top teams spend much time and effort on making things as close to constant as they can. Production cars make this task even tougher thanks to their big ranges of suspension movement, both in heave and roll.

The large range of suspension travel in a production-based car generally causes big swings in chassis balance, which a driver is more likely to feel than pure downforce. I have seen drivers complain about balance shifts as small as 0.2 percent without noticing the addition of nearly 100 pounds of downforce.

Unfortunately, adjusting one end of the car usually impacts the other. For example, adding a rear wing can do three things: greatly increase the rear downforce, somewhat increase the overall downforce, or slightly increase the front downforce. (The last one on the list is caused by the extra rear flow drawing more air under the front of the car.)

Balance shift also needs to be considered. Balance shift also needs to be considered. There is a point—and it’s different for every car—when the aero load on the front suspension causes the nose to drop to a point where the airflow under the car becomes insufficient. As a result, the front loses huge amounts of downforce and pops up. It’s the start of a vicious cycle.

Fortunately, this porpoising can be fixed. One option is to restrict the front end from diving so far. Another is to slightly raise the spring heights or bodywork.

How do you figure out the right aero balance? Lap times and driver feedback are going to be your most important barometers.

Fighting Drag

Drag is another aerodynamic force, and this one is easier to feel. Hold your hand out of the window while driving, and you can get an idea of the forces working against the car and its engine. The drag is a function of the car’s shape, speed and frontal area.

The speed is pretty easy to figure out: look at the speedometer. The frontal area is also straightforward and can be approximated pretty closely with a few simple measurements. The shape of the car is the hard one to work out.

Let’s first tackle something easy. We could reduce the frontal area by about the equivalent of two open hands if we removed the side-view mirrors. As demonstrated by our experience in Germany, this does help.

However, in the grand scheme of things, it’s pretty hard to significantly reduce the frontal area. The mirrors on a typical car only account for about half a percent of it. In the terminal speed scenario, this is significant. For most purposes, though, the change is not enough to worry about. Besides, if you remove the mirrors, you won’t be able to see who’s behind you.

The shape of a car can be changed to help reduce the total drag. Every car leaves a wake when it travels. The wake is equal to all of the changes experienced by the air that the car has encountered.

One way to reduce the wake is to extend the car’s tail to a pointed tip. Theoretically, this allows the air to rejoin as the car passes through. It would be as if the car had never been there. This is impossible—there will always be some wake—but it can be approached as a goal.

A more realistic solution is to reduce the sharp edges and abrupt surface changes. Antennae and other short, stubby pieces create drag. Find them and remove them. Don’t forget to look under the car, too.

While the side view is often pictured during aerodynamic analysis, the top view is just as important. The air flowing around a car splits from side to side as well as top to bottom. The car’s corners can be big drag contributors. Fortunately, radiusing the fender edges and A-pillars to about 2 or 3 inches can result in significant drag reductions.

For real-world examples, look at recent production cars. Vehicles from the ’90s all started to look like jellybeans in order to deliver reduced drag. The engineers smoothed the corners and reduced seams by placing the glass flush with the body.

Current cars are finally getting away from the jellybean shape thanks to development time in the wind tunnel. Rear bumper lines have been raised and deck lids have been lowered to reduce the size of the wake. Lower hood lines that slope up have also reduced some drag from the front of the car.

There is another huge drag contributor that’s harder to reduce but still needs to be addressed: the cooling system. On most cars, the air encountered by the front end cools the engine by way of a radiator. After passing through the radiator, this air is unceremoniously dumped into the engine bay and allowed to go wherever it wants.

A better—although more difficult—solution is to duct the radiator exhaust air to a low-pressure area. This helps keep the flow moving past the car. Ducting past the engine is challenging at best, but it does pay dividends. (Just remember to vent some cool air from the front of the car to the engine bay to help keep the rest of the bits from overheating.)

Testing a Hypothesis

Testing? Yes, actual testing will reveal whether all of these aerodynamic changes are helping, hurting or just distracting us. Top-tier groups have access to wind tunnels and powerful computer systems, but they still test their parts at the track.

A top-speed test run at the Autobahn isn’t a likely option, but fortunately tests for drag and downforce can be accomplished easily and inexpensively. Lap time comparisons will also show whether a modification is worthwhile.

To perform a speed test for downforce, you first need to find the right stretch of road: reasonably straight, level, long, and suitable for driving at highway speeds. The conditions need to be reasonably controlled, so testing in traffic or during weather events isn’t your best bet.

Next comes the basic information gathering. The first step is to come up with a way to mark the suspension travel. Here’s a simple solution: Snug a zip tie around each shock absorber shaft. Then, with the car at static ride height, slide the zip tie along the shaft so it’s touching the body of the damper.

You’ll also want to determine the wheel rate for each corner. Basically, this indicates how much force is required to move the suspension one inch. Spring rates and suspension geometry will yield this figure, and you need to be as accurate as possible.

To start the actual testing, very gently and slowly get the car up to speed. Once at speed, coast down to a stop. Now measure how far the zip tie moved. This test will reveal how much downforce is applied for the given configuration.

This method may not be the most accurate one out there, but it is a reasonable—and free—way to test changes. It works as long as there’s sufficient downforce to generate a fraction of an inch of suspension movement. For an example, if the wheel rate is 200 lbs./in., then the smallest feasible amount of downforce that can be determined is about 50 pounds. When spread across the axle, that’s 1/8 inch of travel at both wheels. Don’t forget that the dampers will show less travel than the wheels experience.

A low-cost drag test is both simpler and more complicated: There’s no need to measure anything other than time, but the final output can’t be turned into a useful number very easily. The test involves seeing how long it takes to coast from one speed to another. The longer the time, the lower the drag. Again, this isn’t something to try during rush hour.

The best way to perform this test is to first accelerate past your starting speed. Next, let off the gas. Start the clock when you decelerate past the first speed, then stop it as you pass the second. Even if you’ve been trying to develop downforce, it would be a good idea to test the drag using this method as a reference.

Remember, every test should be done with the conditions as close to constant as possible. If the conditions change too much, the data may be unreliable. The methods can be used with data systems as well, perhaps with more accurate results. In fact, some data systems even produce some of the same types of values.

If you have a system that can handle these kinds of calculations, then use it as a tool for relative comparison between aero configurations. You can effectively use a horsepower output as a correlation to drag.

As with any aerodynamic testing, the higher the test speeds, the better the resolution of the output: A 10-percent increase in test speed increases the forces experienced by 21 percent.

Math as Proof

Now that the testing basics are covered, we need to discuss at least a little bit of the math. Sorry. The big one to remember is that the aero forces increase with the square of the speed. What does that mean to you? If you drive through a corner at 60 mph on the first lap and at 70 mph on the second lap, the aero forces will increase by 36 percent. That fact applies to the drag as well as the downforce.

Here’s the math that backs up that statement:

wing and splitter aero

We have a little more math that you should know. The frontal area for production cars can be approximated by multiplying three figures: the car’s height, its body width, and the number 0.81. Since the aero forces are directly linked to the frontal area, the forces on a 9/10-scale car would be 81 percent of the full-scale car. This is part of why folding in those side-view mirrors can shave off some time on the Autobahn. It’s also part of the theory behind a few of Smokey Yunick’s more infamous creations.

All Cars Experience Aerodynamic Effects

Every car is affected by aero forces, whether it’s a Miata at a local autocross or a Ferrari at Le Mans. Fortunately, there are almost always ways to use these forces to your advantage. Remember, the top teams spend huge amounts of money and time maximizing every last detail.

For those of us playing a notch or two below Formula 1, gains can still be made—and for little money, too. All it takes is some basic knowledge and a willingness to test and try things.

Aero Analysis

Steve Stafford uses his aerodynamics expertise to evaluate various cars built for a range of competition formats. Which ones get his aero seal of approval?

Subaru Impreza WRX autocrosser

The rear wing appears to be in the working range—not too much angle—and the splitter isn’t crowded by the bumper. The downside—even at slow speeds—is the drag caused by the flaps in front of the rear wheels and open windows.

Ford Mustang road racer

The high-mounted rear wing is up in the clean air and looks like it’s in the correct operating range. It’s good that the front overhang has been removed, but the splitter is too thin and is visibly deflecting. The brake ducts could also be moved closer to the radiator inlet for better pressure and cooling.

Chevy Corvette autocrosser

Steve says he likes the big end plates on the rear wing. The wing’s high placement also allows it to receive clean air. Unfortunately, there are some problems: No additional front downforce has been added to balance the chassis; the windshield wipers parked in the up position create extra drag; and the exposed hood pins disrupt the airflow. Most importantly, the rear wing elements look reversed and may have too much angle. Generally, the largest element of a wing is the closest to horizontal; the smaller ones follow.

Honda Civic time trial car

The rear wing is mounted high enough to work in clean air, and both elements look like they’re in the working range. The dive planes are big enough to do real work instead of just adding drag. The smaller mirrors help reduce drag. The tire-to-body gaps are small, which is usually good for drag reduction. Unfortunately, the splitter is deflecting and is overhung by the dive planes, reducing the efficiency of both.

Scion tC time trial car

Both wings are mounted high enough to reach clean air, and the windows are up and flush for drag reduction. The thick splitter does not seem prone to deflecting. If there is one potential problem, it’s that the wake from the front wing may reduce the efficiency of the rear wing.

A Modified autocrosser

The massive multi-element wings with endplates are great for low-speed operation. The diffuser and side skirts help maintain the downforce on what is probably a flat floor. The lack of bodywork may reduce the effect of the diffuser, however, as air isn’t smoothly directed toward the back of the car. The exhaust could also be influencing the diffuser, which may reduce rear downforce off the throttle.

Pikes Peak Suzuki

This car is developed for maximum low-speed downforce at the cost of drag. The massive splitter is kept off the ground by endplates that feature dive planes. The engine and cooling intakes may be squared off, but they will definitely move the air to where it needs to go. The rockers are extended to increase the floor area and downforce. There are louvers over the wheels to vent high pressure into low-pressure areas. The rear wing is mounted high enough to reach clean air, and the multi-element design keeps the air attached to extreme angles for maximum downforce.

Join Free Join our community to easily find more articles.


View comments on the GRM forums
sesto elemento
sesto elemento SuperDork
6/22/17 11:52 a.m.

More of this please!

WAKman New Reader
6/22/17 8:18 p.m.

Great article! I especially liked the photo analyses of various cars.

David S. Wallens
David S. Wallens Editorial Director
6/22/17 8:31 p.m.

Glad to hear that you're enjoying these cool articles from past issues.

Fitzauto Dork
6/22/17 10:37 p.m.

Awesome article! Gonna go,mix up some test paint now.

David S. Wallens
David S. Wallens Editorial Director
6/26/17 8:34 p.m.
Fitzauto wrote: Awesome article! Gonna go,mix up some test paint now.

So, any data yet?

stafford1500 HalfDork
6/27/17 1:49 p.m.

In reply to Fitzauto:

If you run into any issues with your chosen flow viz fluid let me know and I will try to help troubleshoot it.

Grassroots Motorsports Magazine

Subscribe Today

Also get your instant access to the digital edition of Grassroots Motorsports Magazine!

Learn More