Modeling and Experimentation of GDI-Sized Particulate Filtration and Pressure-Drop Behavior in Uncoated Commercial DPF Substrates

2019-01-0052

01/15/2019

Event
International Powertrains, Fuels & Lubricants Meeting
Authors Abstract
Content
Gasoline Direct Injection (GDI) is known to produce lower concentrations of smaller particulate matter (PM) compared to diesel combustion [1]. The lower concentration results in the absence of soot-cake formation on the filter channel wall and therefore filtration behavior deviates from the expected diesel particulate filter (DPF) performance. Therefore, studies of cake-less filtration regimes for smaller sized particulates is of interest for GDI PM mitigation. This work investigates the filtration efficiency of laboratory-generated particulates, representative of GDI-sized PM, in uncoated, commercial DPF cordierite substrates of varying porosities. Size-dependent particulate concentrations were measured using a Scanning Mobility Particle Sizer (SMPS), both upstream and downstream of the filters. By comparing these measured concentrations, the particle size-dependent filtration efficiency of filter samples was calculated. To predict filtration efficiency for these non-loaded particulate traps, the Opris and Johnson flow field model was updated to include sedimentation and thermophoretic terms and with soot-cake related filtration approximations removed. Experimental results showed excellent agreement with model predictions. Our study demonstrated that current DPFs are insufficient for deployment on GDI vehicles due to their low filtration efficiency for GDI-sized particles. GPFs (gasoline particulate filters) are essential and the newly developed filtration model can serve well to facilitate their design.
Meta TagsDetails
DOI
https://doi.org/10.4271/2019-01-0052
Pages
9
Citation
Sheppard, J., Yang, P., and Strzelec, A., "Modeling and Experimentation of GDI-Sized Particulate Filtration and Pressure-Drop Behavior in Uncoated Commercial DPF Substrates," SAE Technical Paper 2019-01-0052, 2019, https://doi.org/10.4271/2019-01-0052.
Additional Details
Publisher
Published
Jan 15, 2019
Product Code
2019-01-0052
Content Type
Technical Paper
Language
English