Premium
Resin flow through fiber beds during composite manufacturing processes. Part I: Review of newtonian flow through fiber beds
Author(s) -
Skartsis L.,
Kardos J. L.,
Khomami B.
Publication year - 1992
Publication title -
polymer engineering and science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.503
H-Index - 111
eISSN - 1548-2634
pISSN - 0032-3888
DOI - 10.1002/pen.760320402
Subject(s) - porosity , materials science , fiber , newtonian fluid , permeability (electromagnetism) , composite number , composite material , mechanics , porous medium , flow (mathematics) , physics , chemistry , membrane , biochemistry
A critical part of any master model used to simulate or control a composite material manufacturing process is the description of resin flow through the fiber bed. We present here a review of both theoretical and experimental studies of fluid flow through porous media, including fiber beds. For the practical porosity range of interest in continuous fiber composites processing (0.3< ϵ < 0.6), the permeability cannot be accurately described using the Blake‐Kozeny‐Carman equation, even though the flow is Newtonian at very low Reynold's number. For aligned fiber situations, the Kozeny constant, k , deviates radically from theory, depends on bed nonuniformities, and is only constant over very narrow porosity ranges. Thus, one cannot experimentally determine k at high porosities and use this value to describe low porosity situations. Theoretical attempts, based on perfectly spaced and aligned arrays of cylinders, adequately describe the transverse permeability of ideal fiber beds in the high porosity range, but do not succeed at porosities below 0.6. For axial flow through aligned fiber beds, the theory yields permeabilities much lower than are experimentally observed throughout the entire porosity range. For randomly arranged fibers, random cylinder theory also predicts permeabilities that are significantly lower than are measured.