Modeling transverse micro flow in dry fiber placement preforms

Abstract

Dry fiber placement has a large potential for manufacturing preforms for primary-load components at minimum scrap rate and fiber crimp. Yet, challenging impregnation behavior due to low permeability of these preforms during liquid composite molding imposes a need for further research to optimize preform structure for higher permeability. For full understanding of flow behavior within these preforms, flow has to be considered on micro scale (in between single fibers), on meso scale (in between single rovings or strands), and on macro scale (on scale of parts to be manufactured). While macro and meso scale can be measured in experiments or derived from filling times in real processes, micro scale is usually not easily assessable and accessible for standard textile materials. Analytical approaches are limited to regular fiber arrangements (square and hexagonal) that are strongly differing from real arrangements. The present work deals with application of a numerical solver to random fiber arrangements to determine micro permeability transverse to the fiber orientation, for later use in meso- and macro-scaled models. As a premise for reliable calculation, guidelines for boundary conditions as well as size and resolution of the representative volume element are elaborated in the course of this work. Calculated out-of-plane micro permeabilities are subsequently compared to real experiments and show good accordance. The influence of binder particles on micro permeability has not yet been conclusively clarified.

A combined experimental–numerical approach for permeability characterization of engineering textiles

“…Based on the guidelines for microscale modeling developed by Rimmel et al, [ 37 ] the models were generated with randomly positioned fibers, with a resolution of 1/20 of the fiber diameter of 10 μm. The model size was 512 × 512 × 32 voxels, which corresponds to between 460 and 540 fibers at a FVCs between 55 and 65%.…”

“…An error bound of 3% was used as a termination criterion, that is, an estimation of the final value from the asymptotic convergence of the calculated values is made and if the currently calculated value deviates from this by less than 3% the calculation is stopped. Boundary conditions (BC) were defined according to Rimmel et al [ 37 ] in order to generate realistic flow at relatively low computational times: Symmetrical BC tangential to the flow direction and periodic BC perpendicular to it, with an added 10 voxel inlet and outlet area free of fibers, which allows the fluid to disperse before it flows into the fiber structure. These inlet and outlet zones are necessary for the flow simulations transverse to the fiber direction, where the fiber‐free zones are necessary in order not to close‐off any flow channels, which would lead to an underestimation of the permeability.…”

“…•Alternatively, micro permeability can be determined via numerical simulations based on computer‐generated microscale models using suitable assumptions. [ 34–37 ] Usage of such values as roving permeability in dual‐scale simulation models were carried out by Dittmann et al [ 38,39 ] and Chen et al [ 40 ]…”

The influence of fabric architecture on impregnation behavior and void formation: Artificial neural network and statistical‐based analysis

“…This creates effective combinations of impregnation behaviors. [25][26][27][28] All these factors influence the impregnation flow and consequently the formation, morphology, and location of voids. [25,27,29] A complete analysis and classification are critical tasks to ensure highmechanical behavior since voids are connected to poor mechanical performance and the initiation of fracture mechanisms (crack initiation and propagation, delamination, tensile and shear strength, among others).…”

Micro-Scale Permeability Characterization of Carbon Fiber Composites Using Micrograph Volume Elements

“…For the case of high fiber volume fraction, their values lie between the results using square and hexagonal arrangements. Rimmel and May (2020) concluded that a minimum of 10 simulations is necessary to capture the Gaussian distribution of the permeabilities for one fiber volume fraction. They also analyzed the influence of different boundary condition types but without ensuring periodicity at the edges of the micro-scale elements, which makes a comparison to periodic models difficult.…”