Back to Search Start Over

Representative domain size for the simulation of coalescence filtration in nonwoven and foam media.

Authors :
Abishek, S.
King, A.J.C.
Schuler, J.
Kasper, G.
Schmid, H.-j.
Mullins, B.J.
Source :
Separation & Purification Technology. Dec2018, Vol. 207, p344-352. 9p.
Publication Year :
2018

Abstract

Graphical abstract Highlights • Pore scale simulations of air-oil flow through fibrous and foam filters are performed. • Influence of domain size on the CFD predictions are characterized for the two filters. • Two phase flow is more sensitive to mesh/domain size compared to single phase flow. • CFD predictions are validated with measured saturation data for different velocities. • Statistical uncertainties can be reduced to under 10% with sizes between 50 and 100 d. Abstract Pore-scale filtration simulations require high spatio-temporal resolutions and significant computational effort, hence, keeping the domain size to a minimum is desirable. Previous studies have considered domains based on Brinkman length, or are limited by computing power, and little information is available for conditions involving high fluid saturation – typical of steady state mist filtration. In this study, simulations are performed to characterize the effect of domain size on pressure drop, residual saturation, liquid film thickness and interfacial area concentration, using virtual nonwoven and foam filters with similar micro-structural properties. Further, experiments using micro-CT are performed to validate the present computational simulations. It is found that two phase flow through filters are more sensitive to local geometric variations or mesh resolution in the porous media than single phase flow. Statistical uncertainties in the steady state quantities of less than ± 10% can be expected to cope with the increase in computing power required for practical mesh sizes. A computational domain size of about 50–100 × d (where d is the strut or fibre diameter) was found to be required for CFD for the operating conditions considered. [ABSTRACT FROM AUTHOR]

Details

Language :
English
ISSN :
13835866
Volume :
207
Database :
Academic Search Index
Journal :
Separation & Purification Technology
Publication Type :
Academic Journal
Accession number :
131543261
Full Text :
https://doi.org/10.1016/j.seppur.2018.06.051