MOTOR-VEHICLE EMISSIONS VARIABILITY

Test-to-test variability has been observed by many current testing met hods, including the Federal Test Procedure, the IM240 dynamometer test , the idle test common to many Inspection and Maintenance programs, an d on-road remote sensing. The variability is attributable to the vehic le, not to the testing procedure. Because the vehicles are the dominan t source of variability, the only way such vehicles can be reliably id entified is through the use of multiple tests. The emissions variabili ty increases with increasing average emissions, and it appears to be p revalent among the few newer technology vehicles with defective, but u ntampered, closed-loop emissions control systems (1981 and newer model s). In one fleet the variable emitters constitute 2.2% to 4.8% of the vehicles and contribute 8.5% to 22% Of the total carbon monoxide emiss ions. Scheduled I/M programs that fail to ensure repair of these vehic les allow a significant portion of vehicles with excess emissions to e scape reduction measures.