Thirty days at the "Brickyard"

Senior Editor

The 88 ^th running of the Indianapolis 500 this Memorial Day weekend will likely draw 400,000 race fans. Millions more will tune in at home. But when the checkered flag drops on the final lap, all attention will focus on one person: the winner.

"Racing has one clear and concrete goal — winning," says Team Manager Steve Dickson of Columbus, Ohio-based Team Rahal. And for Indy Racing League teams the Indy 500 is the race to win. The front-runner will collect a cool $1.4 million. The purse for finishing dead last bests winning money at lesser races so qualifying at all is a big deal. Doing well at the " Brickyard" also attracts those all-important sponsor dollars. Fielding an IRL car costs teams about 8 to $9 million annually and sponsors bankroll most of it. The Indy 500 alone consumes nearly $3 million, substantially more than any other single race. Why so?

An average race weekend lasts three or days days. Teams "dialing in" race cars for Indy spend an entire month at the track including 17 days of practice and qualifying. A practice session lasts about an hour and there can be several each day. Time between sessions has teams pouring over results of the previous run and readying cars for the next. The combination of 12-hr-plus workdays and 15-min lunches can wear on team members. But in racing, track time is precious and the more the better.

"Indy is where we most refine the car," says Team Rahal Race Engineer, Todd Bolen. Bolen is responsible for setup, the precise combination of suspension, gearing, and aerodynamic downforce that lets cars go fast and handle properly. "We go through 35 sets of tires optimizing the car into an eversmaller envelope," adds Bolen. "It's easy to get lost and fall off the optimal setup." Complicating matters, there are two types of setups, qualifying and racing. In qualifying the goal is to go as fast as possible for four laps. Minimizing drag is a key factor. A racing setup, on the other hand, may trade some flat-out speed for improved handling.

Weather also plays a role in setup, especially at high-speed oval tracks such as the Indianapolis Motor Speedway. Elevated temperatures, for example, equate to less drag, downforce, and subsequently higher speeds. The 2.5-mile Indy oval has wide, 9°-banked independent turns (flatter than at most tracks) which tend to limit corner speeds, while long, 5/8-mile straightaways routinely push speeds above 220 mph. The challenging combination requires an Indy-only single-element rear wing that trims and balances the car as it travels through the gently banked corners yet minimizes drag. Tracks with higher banking angles such as Homestead-Miami Speedway use a two-element wing, while short tracks such as Phoenix International Raceway favor a more "draggy" three-element wing. Teams monitor an on-track weather station and make last-minute adjustments to gearing and wing surfaces if necessary.

The ultimate test of a good racing setup is one that lets a driver go fast and finish a race with minimal fatigue. "If the driver is not comfortable throwing the car into turns at well over 200 mph you have to make it so without compromising speed potential," says Chief Mechanic Ricardo Nault. "Drivers routinely push cars to 80 or 90% of their potential and keep the rest in reserve for qualifying and passing."

About 75% of setup is empirical; knowing what works well at a particular track comes from years of experience. Breakingnew ground typically involves simulation and testing, as well as input from drivers, engine and chassis suppliers, and team members who have been around "forever." The IRL prohibits private testing but allows Open Test days of which there are four before Indy, two each at Homestead-Miami Speedway and Phoenix International Speedway.

These days the distinction among simulation, testing, and racing has blurred because track data continually refines simulation software. And better simulation, in turn, improves car setup. Here, ever-more powerful computers and data-acquisition systems have melded the art and science of going fast.

SPEEDIER COMPUTERS = FASTER RACE CARS
For comparison, in 1991 Indy cars typically used Toshiba 1200 data-acquisition systems with 10 or 12 channels. Simulation software was primitive at best. Today three onboard computers monitor and log 65 or more data channels, then beam the (encoded) information to trackside laptops where it is decoded. Onboard memory has doubled over the last couple of years, boosting data-logging rates and resolution. So much information streams in from its race cars that Team Rahal dedicates eight team members to data acquisition and analysis alone. Jim Foley, assistant race engineer, is one of them.

"Chassis contain 45 sensors for measuring pitch, yaw, and downforce, including potentiometers, accelerometers, strain gages, and laser-distance devices for ride height," Foley explains. "Another 20 sensors monitor engine parameters including rpm, air-box temperature and pressure, oil temperature and pressure, fuel pressure, and water temperature." The IRL mandates teams can only receive data and bans traction control and other active systems.

The sensor data feeds to visualization and simulation software. Also fed into simulations are results of windtunneland dynamometer tests as well as previous track data spanning 10 years. Engineers tweak software parameters including drag and drivetrain efficiency to make simulations match actual ride heights, top speed, minimum corner speed, and lap times to within tenths of a second.

Obviously a big part of what sensors measure is the direct result of driver input. "If the driver can move it there is a sensor attached to it," says Foley. Professional race drivers such as Team Rahal's Kenny Brack basically know by feel how a car is performing, though certain nuances are below human perception. For example, rear tires spinning about 10% more revs than the front signal loss of traction and the need for a tire change. Foley must keep close tabs on tire wear because a flat tire would likely cost them the race. The special Firestone tires have a tread depth of just 3/32 in. and last about 100 miles.

Other sensor data simply relieve drivers of certain details so they can focus on the task of racing. One such system lets engineers monitor remaining fuel within 1% by totaling the number of squirts made by the engine's calibrated fuel injectors. IRL cars hold 35 gallons of methanol fuel and their thirsty, 700-hp V8 engines get less than 2 miles/gallon so fuel burn is a key metric for planning pit stops.

MEANWHILE, BACK AT THE SHOP...
Most of the action is at trackside though efforts there keep team members back at the shop busy as well. For example, "At Long Beach recently, race engineers predicted from track data and simulation that a ratio change to a steering arm could improve lap times," explains Team Rahal Design Engineer Bryant Holzinger. "They e-mailed the suspension coordinates and we took it from there."

Using Pro Engineer CAD software Holzinger first checked that a part matching those coordinates wouldn't interfere with existing hardware then optimized the part with FEA. Machinists converted the geometry to G-code and finally into a total of 10 aluminum-alloy parts, four for each race car and two spares. The parts arrived for a Friday practice. Total time: five days. The change helped both cars qualify in pole position.

At the Indy 500, lowering drag is a main goal. And the typical way to do this is by altering the aerodynamic surfaces that provide downforce. Again, race engineers relay the basic data to Holzinger who turns it into CAD renderings. After a pass through nonisotropic FEA, design drawings go to NC machines that cut out mold bucks. The bucks provide the shape for the vacuum-cured carbonfiber parts. "We can deliver a finished, painted carbon-fiber wing section in three days," says Holzinger.

While such engineering gymnastics go with the territory, Holzinger and crew have done much of the hard work of preparing the cars months before Indy. This year they have their work cut out for them. In an effort to curb speeds and contain costs, the IRL (starting with the Indy 500) will slash engine displacement from 3.5 to 3.0 liters, challenging teams to overcome the deficit. Three companies supply engines to IRL teams: Chevrolet, Toyota, and Honda. Team Rahal partners with Honda which leases the motors and provides a full-time technician who doubles as a Rahal team member. It's a win-win because both Rahal and Honda benefit from ongoing R&D.

Probably an even bigger challenge for Team Rahal came from the decision last November to switch chassis suppliers from the more established Dallara Automobili to relative upstart Panoz G-Force. One reason for switching: Far fewer cars on the circuit — six G-Force cars versus 28 Dallara — lets G-Force engineers respond faster to customer input.

"Dallara listens to 12 drivers and G-Force, to three or four," Bolen explains. "Our testing program at Homestead in February, for instance, prompted big and aggressive changes at G-Force that got us pole position." Not to mention, G-Force is currently on a roll. It was a Panoz G-Force car driven by Marlboro/Team Penske's Gil de Ferran that won last year's Indy 500.

Both Dallara and Panoz G-Force cars cost about $0.5 million/copy and are considered race-ready right out of the box, though for most teams it is merely a starting point. To the untrained eye both chassis look remarkably similar, but "we basically had to start over with the simulation and 'best guess' drag and downforce numbers," says Foley. "We'll probably redesign 20 to 50% of the components to make the cars more competitive and reliable," adds Nault. "All components are mileaged out. A half shaft lasts 2,000 miles, for example. We do nondestructive testing before each race on critical components such as suspension wishbones. Parts get tossed when twothirds of usable life is up, even if they test out OK."

Other design changes help shorten time needed for pit stops. Though details are secret, Nault says that a tweak in geometry of the fuel inlet boosts flow rate, shaving time off refueling. And the built-in air jacks that raise cars for tire changes were redesigned so they actuate faster. Every little bit helps considering pit crews can change four tires in about 5.2 sec. Pit crews work out and practice on a pit-stop simulator (a retired race car) at the Columbus shop. And just like pro sports teams they review videotapes of their work searching for clues to improve.

LOOKING FOR AN EDGE
Take a walk through the paddock at Indy and other races and you'll see race engineers scoping out the competition, especially the league leaders. "Go-fast" secrets are tough to keep under wraps. And when team members switch employers they take valuable information with them. The result is "Everyone moves toward a kind of setup over the course of several years," observes Bolen. "Some amount of turnover and cross-pollination is good for the sport because it keeps one team from dominating as Ferrari has Formula One."

Steve Dickson has a similar take on the subject: "Everything in racing is an evolution. Lessons learned from the first day on at the Indy 500 this year will be catalogued for next. We're always looking to the next race using information from previous races." In that regard Team Rahal will have plenty of material to study. They recently announced plans to run three cars at Indy instead of the normal two. And there are 15 other races in 2004, though probably none more nerve-racking for teams than the Indy 500. Jim Foley disagrees. "I approach it the same as any other race. Special preparations? I pay the bills and buy my wife flowers before I go because for the next month I'm married to racing."