Filtration efficiency and pressure drop modelling of particulate filters with rear plug damage

Abstract

A model suitable for wall-flow particulate filters with partial rear plug damage is developed and experimentally validated in this work. A ceramic filter with 16% of the rear plugs mechanically removed is tested at steady-state conditions on the engine bench and transient driving cycle conditions on the chassis dynamometer. After decanning of the monolith, destructive analysis is conducted to identify deposit loading variations and scanning electron microscopy is used to study the deposit structures in the channels. It is shown that channels without rear plugs develop distinct deposit structures in the entry zone. Hence, a local pressure loss coefficient is applied to model the effect of entrance flow constrictions, taking also into account deposit restructuring phenomena at higher flow rates. In addition, a deep-bed filtration submodel is used to capture the effect of non-uniform wall velocities on deposit accumulation in the wall. The modified model is first fitted to the engine bench data and then validated in a wider range of conditions using the driving cycle tests. With the exception of prolonged steady-state loading conditions, good pressure drop and filtration efficiency predictions are obtained throughout the tests in conjunction with correct deposit property profiles. Notably, the cold-start worldwide harmonized light vehicles test cycle shows that the current European on-board diagnosis threshold limit for particulate mass is too relaxed to trigger a malfunction indication for moderate filter faults. In conclusion, the model can be applied in damaged particulate filter studies for the assessment of leaked particulate mass, the specification of more effective legislation limits and the development of rigorous on-board diagnosis systems and algorithms.