Browse Publications Technical Papers 2004-01-3019

Prediction of the Efficiency of an Automotive Oil Separator: Comparison of Numerical Simulations with Experiments 2004-01-3019

RENAULT has developed a numerical method for predicting the efficiency of two types of oil separator. Numerical simulation is performed with the CFD package FLUENT6. Blow-by flow is considered as an aerosol mixture and is simulated as a continuous phase (air flow) carrying a discrete phase (oil droplets). The separator is meshed using a uniform 1mm tetrahedral mesh. The inlet volume flow is assumed steady. A standard k-epsilon model is used for flow calculation. Then droplets are introduced and their trajectories are computed. The wall boundary condition is straightforward: a particle touching a wall is assumed trapped and removed from the flow.
The experimental set-up for efficiency measurement includes an oil generator which produces an air/oil aerosol mixture with a user-fixed volume flow, oil mass-flow and oil droplets diameter distribution. The generator is connected to a prototype oil separator. It is equipped with length and position-adjustable plates. A spectrometer bridges the inlet and the outlet. It can measure the concentration of both incoming and leaving aerosol flow in respect with droplets diameter, hence efficiency.
Some preliminary results have been obtained on a 3 chicanes configuration. Since there is an average 3% error on the measurement, the separator is found to separate nearly no droplets of diameter lower than 2.5 micrometers. Then the efficiency increases up to 30% for 4 micrometers droplets. The numerical simulations do not reproduce this evolution. The statement can be made with the second separator. Efficiency is overpredicted by numerical simulations.


Subscribers can view annotate, and download all of SAE's content. Learn More »


Members save up to 40% off list price.
Login to see discount.
Special Offer: With TechSelect, you decide what SAE Technical Papers you need, when you need them, and how much you want to pay.