A Methodology to Estimate the Mass of Particulate Matter Retained in a Catalyzed Particulate Filter as Applied to Active Regeneration and On-Board Diagnostics to Detect Filter Failures

Steady-state loading characterization experiments were conducted on a catalyzed diesel particulate filter (CPF) in a Johnson Matthey CCRT® (catalyzed continuously regenerating trap) system. The experiments were performed using a 10.8 L 2002 Cummins ISM heavy-duty diesel engine. Experiments were conducted at 20, 60 and 75% of full engine load (1120 Nm) and rated speed (2100 rpm) to measure the pressure drop, transient filtration efficiency, particulate mass balance, and gaseous emissions. During the research the CPF cracked with a characteristic ring-off failure. Data and analysis on the cracked filter were compared to the data on the same un-cracked filter.

Recent literature using experimental data and the MTU 1-D 2-Layer CPF model has shown that the mass of particulate matter (PM) in the wall can oxidize under temperatures above 350 C and significantly affect the wall and in turn the total pressure drop. This affect would result in a poor correlation between pressure-drop across the CPF and the mass of PM in the CPF.

An algebraic equation relating PM mass to total pressure drop was formulated from previous theory along with an approach of using an average exhaust temperature into the CPF to estimate the mass in the wall and in turn the permeability and pressure drop of the wall. The results for the estimated CPF mass retained for the un-cracked filter obtained from this methodology developed were within ± 2% from those obtained using the MTU 1-D 2-Layer CPF Model.

This method also proved to be a useful tool to detect filter failure. The pressure drop for the cracked filter was lower by 1 to 8 kPa than that for the un-cracked filter during loading characterization experiments. The pressure drop for the cracked filter peaked during the initial 5-10 minutes of the experiment (due to temperature) and then became nearly constant. The peak in the un-cracked filter was at about 30 minutes of the experiment due to temperature and particulate mass loading. The particulate filtration efficiency was between 50 and 90% for a cracked filter compared to 99% for the un-cracked filter.