Verification of Accelerated PM Loading for DPF Qualification Studies

High gas prices combined with demand for improved fuel economy have prompted increased interest in diesel engine applications for both light-duty and heavy-duty vehicles. The development of aftertreatment systems for these vehicles requires significant investments of capital and time. A reliable and robust qualification testing procedure will allow for more rapid development with lower associated costs. Qualification testing for DPFs has its basis in methods similar to DOCs but also incorporates a PM loading method and regeneration testing of loaded samples. This paper examines the effects of accelerated loading using a PM generator and compares PM generator loaded DPFs to engine dynamometer loaded samples. DPFs were evaluated based on pressure drop and regeneration performance for samples loaded slowly and for samples loaded under accelerated conditions. A regeneration reactor was designed and built to help evaluate the DPFs loaded using the PM generator and an engine dynamometer.

Data demonstrates that the procedures used for accelerated PM loading are reasonable. Normalized regeneration results of samples loaded for 45, 60 and 90 minutes using the PM generator show less then 5% difference. CO and CO₂ light-off results obtained during regeneration tests were within 5% for the three different loading times. The slopes of the loading curve between PM generator (35 min.) and engine dynamometer (180 min.) were 0.28 and 0.29, respectively when samples were loaded to the same mass. We believe the main difference is the chemical composition of the PM for the dynamometer vs. PM generator. Regeneration results show a lower temperature light-off for PM generator loaded samples when compared to engine dynamometer loaded samples. Dynamometer loaded samples have a slightly faster regeneration response, but the qualitative features track with overall loading and PM oxidation characteristics. It was determined that the method was not only accurate but was also repeatable for multiple samples.