The data used in this paper were obtained from chassis testing on five heavy-duty trucks from the E-55/59 program [8]-[10], which was jointly sponsored by the Coordinating Research Council (CRC), CARB, USEPA, Department of Energy (DOE), Office of Freedom CAR and Vehicle Technologies, National Renewable Energy
Laboratory (NREL), South Coast Air Quality Management District and Engine Manufacturers Association. Data
were obtained from the chassis dynamometer testing at the West Virginia University Transportable Heavy-Duty
Vehicle Emissions Testing Laboratories (TRANS-LAB). A comprehensive explanation of the experimental
procedures can be found in prior papers [11]-[13]. A brief description of the experimental set up is as follows.
The dynamometer was a platform with flywheels, power absorbers and rollers. The vehicle was mounted on a
test bed with the drive wheels on rollers. The rear wheels were allowed to rotate freely on the rollers. The power
was absorbed from the vehicle wheel hubs by the power absorbers mounted on either side of the chassis bed,
simulating the load on the vehicle. The power absorbers simulated real-world driving conditions by accounting
for the aerodynamic and the frictional load. The flywheels were connected to the vehicle hubs and the vehicle
load was established using a coast down procedure on the dynamometer. The torque produced by the vehicle
was translated to the sensors through shafts and gear boxes. Sum of the readings of the sensors on either side
should be equivalent to the axle torque. The vehicle was driven to follow the speed-time trace of the desired
drive cycle. The target speed was provided on the computer screen to the driver while the test was running and
the vehicle was driven to meet that speed which simulates the drive cycle used. The emissions were measured
with exhaust gas analyzers and a data acquisition system. The losses associated with the tire-roller interaction
have been discussed elsewhere [14].