Evaluation of Partial Flow Dilution Methodology for Light Duty Particulate Mass Measurement

Two different implementations of Partial Flow Dilution (PFD) methodology designed for gravimetric particulate matter (PM) sampling are evaluated for applicability to light-duty chassis emissions testing. Filter PM measurements were collected and compared to constant volume sampler (CVS) full dilution tunnel PM filter measurements and other real-time PM measurement technologies, using gasoline vehicles generating a range of 0.1 to 10.0 mg/mile PM. Exhaust samples were collected for each phase of the Federal Test Procedure (FTP-75) with a fourth filter sample collected for the US06 supplemental cycle. Both PFDs satisfactorily met proportionality criteria for conventional combustion engines, but some improvements are needed for hybrid electric vehicles (HEVs). The PM mass collected scaled linearly with the CVS tunnel samples, with slopes of 1.03 and 0.74 for the two PFDs. For each sampling method, the filter-based PM mass variability is about ± 0.5 mg/mi; thus, the uncertainty in these slopes is about ± 0.2. Real-time capability was also tested by comparing simultaneous PM measurements with two Dekati Mass Monitors (DMM) sampling from PFD and CVS tunnel, with significantly different results from the two PFD instruments. Additional real-time measurements produced particle number counts and soot measurements that generally trended well with gravimetric results. The soot results particularly served to draw out possible artifact deposition conditions that were sensitive to PM output levels and exhaust temperatures.