Abstract

The paper describes in some detail the difficulties encountered when conducting simulations intended to be realistic, and hence useful from a predictive viewpoint, of localised impact events on composite shells.Experimental results of drop tower tests were available on this occasion, consisting of an impactor with semi-spherical end being dropped from various heights onto a square shell made of composite material.The exhaustive simulations performed included different constitutive behaviours of the material, damage models, element types, and simulation strategies (three-dimensional and axially symmetric).The results show that, as the damage level increases, the problem becomes very complex because it requires capturing well the onset of interlaminar failures, as well as reproducing accurately the transition from a bending to a membrane resisting mechanism, which is what happens as those failures accumulate. The best results were obtained using very refined meshes, with four levels of elements per layer, and with cohesive elements between adjacent layers.Apart from the difficulties posed by the numerical simulation, it is worth pointing out that the study of these problems also tends to suffer because the characterisation of the parameters and properties governing the material behaviour is often less precise and reliable than desirable.

Abstract

A numerical approach has been developed for modelling the behaviour of a carbon fibre woven laminate under low-level energy impact. The model results have been compared with dart drop tests under sub and supercritical impact energies in fibre carbon textile laminates of different thickness.The models were developed using Ls-Dyna explicit finite element code. The validation approach was based in increasingly complexity approach. First models were developed and compared with subcritical impact energy test. Afterwards, supercritical impact energy has been applied in the laminates and delamination failure has been considered. In order to model delamination damage, cohesive law has been applied using Ls-Dyna cohesive and contact elements. The validation of the model has been carried out by comparing force and displacements.

Abstract

Carbon/epoxy pre-impregnated has become one of the most demanded materials for some aeronautics structures. Boeing 787 and Airbus A350 are pioneers in incorporating carbon fiber-reinforced composite materials in primary structures as wings and airframe. Tailoring the laminate ply sequence involves a difficult challenge owing to the rigorous design and operational criteria. Given that composite materials present a high sensitivity to impact loads, it has been indispensable to examine the influence of different factors in the failure mechanisms after an impact event. In this work, the effect of impact energy, stacking sequence and thickness on the impact damage resistance of CFRP composites was investigated. Damage response parameters as peak force, absorbed energy and delamination threshold were obtained from the impact curves. Then, a C-Scan imagen showed the damage extension using the non-destructive inspection Phased Array.

Abstract

The aim of the present study is the analytical determination of the energy release rate during the DCB test of angle-ply laminates. With this purpose, an analytic approach based on the complementary energy of a laminated beam including hygrothermal effects is proposed. These effects lead to initial strains that indeed, are impeded by the configuration of the DCB test. The consequent forces and moments are determined by means of Engesser-Castigliano theorem. Once these loads are known, an analytical expression for the energy release rate based on the complementary energy is obtained. Symmetrical and anti-symmetrical angle-ply laminates with 16 plies have been prepared and used for the tests. The configuration of the cracked arms is the same in both laminates, however, the delamination occurs between plies with different orientation. Experimental results of the energy release rate obtained by means of the area method agree with those determined by the proposed approach.

Abstract

Crack propagation performance in the ENFR (End Notched Flexure test with roller) test configuration is analyzed. The test configuration is based on the ENF test with a roller introduced between the two surfaces of the crack in order to get mixed-mode. Based on the reliability of the linear criterion, an equivalent energy release rate is proposed taking into account the interaction between the two modes and II. The stability of the crack propagation is defined by means of the derivative of the proposed equivalent energy release rate. Experimental tests with specimens of F593/T300 carbon/epoxy unidirectional composite have been carried out to assess the proposed equivalent energy release rate and to evaluate the crack stability condition. Finally, some recommendations for performing tests in the ENFR configuration under optimum conditions are summarized.

Abstract

Interlaminar failure tests are usually carried out in specimens where crack arms are symmetric. Asymmetric Double Cantiliver Beam (ADCB) is an exception where both crack arms have different thickness in order to get mixed mode I/II. The asymmetry in ADCB specimens could be applied also to te case of End Notched Flexure Tests, obtaining also mixed mode I/II. The decomposition proposed by Williams is not valid in the case of ADCB tests. On the other hand, articles related to asymmetric ENF tests including the mode decompositiion have not been found. The present study deals with the analsys of the mode decomposition in the case of specimens with asymmetric arms. The asymmetry may be related to the different thickness of the crack arms, to different materials or a combination of both factors. The results obtained will be compared with the results of mode decompsition that have been published in the case of ADCB tests.

Abstract

The aim of this work is the development of a new test methodology for the characterization of delamination of composites under mode III of fracture and dynamic stress (fatigue). The phenomenon of delamination is one of the main failure mechanisms in composite materials configured by stacking several layers. The mode of fracture that has been studied is mode III to fatigue, with the difficulty that its simulation implies. The test device employed, which allows subjecting to the laminates to this loading mode, has already been previously tested in static tests. Characterization tests were performed on two different materials, both with different matrices and carbon fibers. An analytical model developed for the case of static fracture has been used and a statistical analysis of the results obtained has been carried out. A fractography of the fracture surface was also performed in order to determine the characteristics of this type of fracture in a dynamic regime.

Abstract

In a same way to polymeric materials, some composite materials as it is the case of green composites, are characterized by a viscoelastic behaviour which plays a key role in their mechanical properties with time. The use of some engineering materials could be restricted as their mechanical properties could be highly affected by temperature or time. For this reason, this work is focused on the modeling of the viscoelastic behaviour of this biomaterial composite with a biobased polymer matrix from poly(lactic acid) and a lignocellulosic reinforcement/filler coming from hazelnut shell with different plasticizer content. In particular, a biobased plasticizer, namely epoxidized linseed oil (ELO) has been employed. A first approach to the viscoelastic behaviour of these composites is summarized in this work. The Maxwell model considers a series combination of the elastic reponse (represented by a spring) and the viscous contribution (represented by a dashpot). The Kelvin-Voigt model considers the same physical elements (spring and dashpot) but coupled in parallel. As these two basic visceolastic models show important restrictions regarding the real viscoelastic behaviour, this work also considers a series combination of the Maxwell model and the Kelvin-Voigt model, known as Burgers model. This works focuses on determining the different parameters of the Burgers model to fit experimental data from PLA-hazelnut shell flour biocomposites with varying plasticizer (ELO) content.

Abstract

Composite materials manufactured by means of the superposition of layers and some specific morphologies, are susceptible to losing part of their resistant capacity due to the presence of delaminations between their  layers. The objective of this work is to evaluate experimentally the generation and propagation of these interlaminar cracks in a carbon-epoxy composite subjected to dynamic stress under mixed fracture mode I/II, using for this purpose two different test methodologies, on the one hand the standardized MMB (Mixed-Mode Bending) and on the other hand the ADCB (Asymmetric Double Cantilever Beam), with the intention of exploring the viability of the ADCB test as a simpler alternative of execution, especially in fatigue tests, to the MMB test. For this purpose and after a previous static characterization of the material in which the critical values of the energy release rate were determined in both test methods, the energy release rate levels to be applied in the fatigue tests were defined for two GII/Gc mixed mode grades of 0.2 and 0.4(0,33 ADCB)   and a ratio of stress levels to fatigue of R= Gmin/Gmax=0.1.DG-N fatigue initiation curves were obtained from these experimental data. The most relevant result of the work is that, in general, the MMB mixed mode provides different fatigue limits depending on the degree of mixed mode, whereas in the ADCB type test the fatigue limits tend to have the same value.

Abstract

Mode decomposition in interlaminar fracture tests initially designed for pure modes I and II as Double Cantilever Beam (DCB) and End Notcehd Flexure (ENF) is analyzed. In the initial proposal of those tests crack arms have geometric and material symmetry and the crack advance occurs through the middle plane of the laminate. In the current study, the possibilty of crack arms of different materials or with different thickness is included. Two mode decomposition criterions are considered after the analysis of a slice near the crack tip. Those criterions are based on the stress and displacement state in the non-cracked zone near the crack tip. Two points that before the crack advance were the same point are named homologous. When the cohesive zone is developing, the stresses in homologous points are the same due to the principle of action and reaction. Nevertheless, relative displacement can be normal, related to mode I, tangential, related to mode II, or a combiantion of both, related to mixed mode I-II. In the present work the relative tangential displacement related to mode III is not considered. The contributions to the energy release rate corresponding to bending and shear in each mode are analyzed. The results obtained are applied to asymmetric DCB and ENF tests.