Product Support

Who is Bill Mitchell?

Fundamentals of Data Acquisition - part 1

[ Home ] [ Up ] [ Fundamentals - part 1 ] [ Fundamentals - part 2 ] [ Fundamentals - part 3 ] [ Fundamentals - part 4 ] [ Fundamentals - part 5 ] [ Data Acquisition Procedures ] [ Force-Based Roll Centers ]

Introduction to Data Acquisition

Computer data acquisition is a common tool of the modern racing team. It is an electronic system carried in the race car which records data acquired from sensors mounted about the car. The sensors may measure engine performance, chassis movement, aerodynamic data and driver activity. Each sensor may be recorded once a second (water temperature) or as often as 1000 times a second for damper movement. The data gives engineers, technicians and drivers information to improve their performance. With track time getting more and more expensive, and more restricted, it is crucial for the winning team to get as much information as possible from every minute the car is on the track.

When the car pits and the crew swarms over the car, one crew member (usually known as the data geek) inserts a cable connected to a laptop computer into the car and downloads the information to the computer for analysis. Telemetry is a version of data acquisition which transmits the data via radio from the car to the pits. The transmission may be continuous during the lap or may be a burst transmission when the car is near the pits. This has the advantage of immediate knowledge regarding engine problems which may prevent an expensive engine failure. But telemetry is expensive and sometimes gives less than complete coverage of the track. More complex data is usually stored and transmitted by cable.

The basic engine sensors include rpm, oil pressure and water temperature. More complex systems may measure air flows, combustion efficiency, and even engine torque. Cars with electronic engine management systems can usually record much of the information used for engine control. Indeed, electronic engine management systems and data acquisition systems have much in common. They certainly should use the same sensors. Some data systems even provide a video dash showing the driver some of the relevant information. These displays can even be programmed to show detailed information when a problem is brewing.

Racing is all about speed. Measuring vehicle speed is the most important chassis variable, and not as easy to measure as you might think. After all, every street car has a speedometer. Vehicle speed is usually measured as wheel speed, but the tire may not be going the same speed as the vehicle. The rear wheels (of a rear-wheel drive car) are often going faster than the vehicle (wheel spin) and all wheels sometimes stop while the vehicle keeps going (brake lock). Indeed, one of the toughest problems is measuring the speed on a four-wheel-drive rally car on gravel. You may have four different wheel speeds, none of which may correspond to the actual speed of the car.

Generating accurate vehicle speed is crucial to understanding how the car is behaving and in generating distance data. Two different laps should matched by distance in order to compare the speed of the two laps at the same point. Most data acquisition systems have a beacon system which identifies one point on the track, usually the Start/Finish line. This allows the data system to break the data into individual laps.

Additional chassis sensors include suspension travel, which is usually measured with a linear potentiometer. This measures how much the spring and/or shock is extended or compressed. Shock velocity can be measured directly or by differentiating the position data. More sophisticated chassis measurements can include ride height, which is measured by bouncing a laser beam off the ground. Lasers are expensive and complex. The difference between measuring suspension travel and ride height is tire squish. It is easy to determine where the chassis is relative to the wheel, but measuring where the wheel is relative to the ground is difficult. A rubber doughnut rotating 30-50 times a second while absorbing varying vertical, lateral and horizontal forces is a very complex device. Most of the experts in this subject work in Akron. 

The wheels measure ground speed but wind gusts change the forces acting on a race car. And modern formula cars are often closer to low-flying aircraft than automobiles. Air pressure in or near the car can be measured by pitot tubes. Cowl pressure is often measured when Winston Cup teams test at Daytona. But this year the rules were changed to simplify the air induction.

But the most crucial sensors are linked to the driver. These include throttle pedal and brake pedal position, steering wheel position. , and gear choice. This information is crucial to comparing driver performance and is a great advantage in developing young drivers. Seat time is so expensive and it is very difficult to detect driver differences when their lap times are close. Data acquisition allows driver to grow to their potential with less seat time than otherwise. 

Every driver with a data acquisition system will make some comment like "Now, I won't have any secrets." And there are many stories of drivers being surprised by what the data shows. But good drivers utilize the data to improve their performance. They may not be interested in parts of the track where they are faster than their teammate, but they are very interested in parts of the track where they are slower. Rick Mears was one of the best Indy Car drivers ever, and there are reports he would take a computer back to the hotel to study his performance.

Costs:
Data acquisition systems cost anywhere from $500 to $35,000 or more. A Competition Data System starts at about $5,000 and may run to $8,000 with additional sensors. PI has a wide variety of systems and it quite easy to spend $15,000-$35,000. There are less expensive systems coming onto the market and even a one-channel system can provide useful information. Like anything else electronic, the prices are coming down while the capabilities grow.

Rules:

Sanctioning bodies have rules regarding data acquisition systems. They are intended to contain costs by eliminating the escalation of expense while maintaining cost-effective systems. Of course, each sanctioning body has their own definition of cost-effective. NASCAR forbids any computer data acquisition systems during race weekends. But most Winston Cup teams use data for private tests and the truck teams may even use data during "test days", which may be part of a race weekend. But all data acquisition systems, sensors and wires must be removed before the official practice, qualifying and racing.

SCCA limits data acquisition for it's premier series, Trans-Am, but there are no limits in club racing. The Trans-Am rules have a $5,000 claiming rule for the data hardware and permit only ten specific sensors. This is a fairly effective rule and no systems have ever been claimed. The $5,000 claiming rule makes the Competition Data System almost a spec system.

CART recently adopted a list of permitted sensors. Teams may use one wheel speed sensor, damper position, tire pressures, lateral G, brake pressure and any engine-related function as well as driver controls (steering wheel position, gear, anti-roll bar controls and weight jackers). They may not measure or record longitudinal G, wheel loads or ride heights.

Formula One has no limits and many cars have separate systems for chassis (the team) and engine (the engine manufacturer). With the "marriages" between team and engine manufacturer now often reduced to "one year stands" the two do not always share data. The larger teams have their own proprietory systems while PI has the back portion of the grid.

Formula One and CART also have data acquisition systems designed for accident investigation. These are usually supplied by the sanctioning body specifically to measure the impact forces experienced during a crash. These systems are adapted from the automotive industry and have clever algorithms to record data at a very high sampling rate during a crash and retain that information. The crash systems measure very brief extreme forces (up to 100 g) but without the precision of a 5 g sensor which is more common on a chassis system. Impact sensors are cheap enough to be placed in computer boxes to determine if they are dropped during shipping.

Benefits:
Track time is expensive and getting more expensive. Test time is expensive and often restricted. Teams must get as as possible much from each moment of track time. This means recording data for later analysis.

Link to Tech Reports

Link to part two Fundamentals - part 2