ATLAS F1   Volume 6, Issue 52
		Comprehensive Traction Control in F1
		by William Shoebotham, U.S.A.

The anticipated re-introduction of Traction Control into Formula One sparked a heated debate among fans and professionals alike. However, none seem to pay attention to the full potential of traction control, either in terms of how difficult it can be to implement, or how much it can accomplish. William Shoebotham, an automotive engineer, attempts to fill that gap The technical directors of most Formula One teams recently recommended legalizing traction control due to concerns about gray areas in the rules and outright cheating. Following that, the sport's governing body, the FIA, said it would adopt the recommendation, in return for a few safety concessions from the teams. There remains a question of exactly when it will be adopted, but it appears certain to happen in the near future. It's natural to assume that traction control will be simple to implement and will reduce the gap between the front and back of the grid. However, traction control opens an entire world of active car control that may increase the performance difference between "haves" and "have-nots." Traction control isn't limited to preventing wheel spin during acceleration. It allows traction at the rear wheels to be actively controlled by the drivetrain at anytime. This control can be used to influence any aspect of a car's performance, including braking and cornering. Control of rear wheel traction may not be an engineer's first choice to influence braking or cornering, but it certainly suffices. This comprehensive traction control requires a complex algorithm. Teams will program traction control computers with the car's desired behavior for any situation. Sensors on the car will tell the traction control computer what the car's behavior actually is. The computer will then use mathematical models of the drivetrain and vehicle dynamics to determine what drivetrain actions should be taken to make actual car behavior most closely match desired car behavior. Closed loop feedback can be used to fine tune any part of this algorithm while the car is being driven. The mathematical drivetrain model is needed to predict the throttle, ignition, and fuel required to produce a desired torque at the rear wheels. This model would include engine torque for any RPM, engine temperature, air pressure, etc. Furthermore, this torque can be negative (braking torque). The model would also include things like engine rotational inertia, gear ratios, and driveshaft torsional stiffness. The mathematical vehicle dynamics model is needed to predict the rear wheel torque required to produce a desired behavior on the track. This model would include things like aerodynamic forces, masses, moments of inertia, tire information, and suspension positions. The vehicle dynamics model would be improved with track mapping to including bumps, hills, corners, and other influences on car behavior. The models help traction control get from current behavior to desired behavior, but what is the desired behavior? During braking the car will effectively have active rear brake control. If there is too little rear brake force then the drivetrain can put negative torque into the rear wheels to compliment the brakes. If there is too much rear brake force then the drivetrain can put positive torque into the rear wheels to partially counteract the brakes. The driver will have quicker braking and less concern of locking the rear end and spinning going into a turn. In a steady state corner, if a driver responds to oversteer by lifting, then lift-throttle oversteer can make the situation worse. Traction control can use the vehicle dynamics model to predict how quickly the throttle can be rolled off without making the oversteer worse. Traction control can then override the driver until it determines that the car would be stable again under driver control. Traction control can also sense or predict oversteer on its own and gently roll off the throttle without any input from the driver. Traction control isn't even simple during acceleration. A tire turning at ground speed creates no forward force, so the tire must slip slightly to accelerate the car. The acceleration increases with slip until it peaks (typically at 20% slip for street tires) and then decreases with additional slip beyond this. By contrast, tire wear doesn't reach a similar peak and always increases with more slip. A good driver without traction control can keep acceleration near peak, but generally errs on the side of less slip and less wear. Well executed traction control can keep acceleration closer to ideal than any driver but, contrary to intuition, traction control's greater slip can also increase wear. During qualifying a team might choose to maximize grip at the expense of wear and then change the tradeoff for the race. So far only pure braking, cornering, and acceleration have been considered. However, an F1 car will often be in complex combinations of these situations. What does the traction control system do to maximize performance in these instances? How should traction control influence the car so the driver still has confidence the car will response like he wants? Think traction control is still simple? Maximizing the potential of traction control will be the work of entire departments of engineers. This complexity will be a long-term challenge for Ferrari and McLaren. At the other end of the pitlane Minardi is fighting to survive, and any notion of additional engineers to work on traction control is quixotic. Crude systems that merely prevent excessive wheelspin during acceleration are easy to make, so the entire grid will have various forms of traction control once it's legalized. They will likely get what they pay for. However, all teams want traction control at some point in the future despite its burdens. This is a strong indication of how frustrated the teams have become with the potential cheating that legalized traction control seeks to avoid.

William Shoebotham	© 2000 Kaizar.Com, Incorporated.
Send comments to: william@altair.com	Terms & Conditions
William Shoebotham has a Batchelors of Science degree in Mechanical Engineering from The University of New Mexico. He created suspension kinematics software for an Albuquerque company called Auto-Ware, and is currently employed as an automotive engineer for Altair in the Detroit area.