Herein is contained details of a comprehensive finite element survey and experimental investigation into the buckling and postbuckling characteristics of thin laminated square Carbon-Epoxy panels with various cutout geometries, subjected to uniaxial compression. The plate edges are considered to be fully fixed with constant edge displacement loading. The panels were quasi isotropic in nature with a stacking sequence of (0/90/±45)2,. Square, circular and elliptical centrally located cutouts were considered with cutout dimension/panel widths ranging from 0.1 to 0.7 in increments of 0.1. Eccentrically located circular and square cutouts were considered for cutout dimension/panel width ratios ranging from 0.1 to 0.4 with vertical and horizontal eccentricity varying from 0 to 20% of the panels width. Multiple circular cutouts with cutout dimension/panel width ranging from 0 to 0.3 with separation distance/panel width ratios ranging from 0.2 to 0.65. A finite element eigenvalue analysis was adopted to determine the critical buckling loads and buckle mode shapes for the panels. The postbuckling response of the panels were investigated by adopting a non-linear finite element analysis approach using an Incremental Newton-Raphson Iterative solution scheme. A limited experimental test programme was undertaken to act as verification to the finite element solutions. A purpose built buckling rig was designed and manufactured for the purposes of the tests. It has been confirmed that the critical buckling loads for centrally located circular and square cutouts initially reduces as the cutout size increases. After reaching a minimum value it thereafter increases with large cutout sizes, the exact changeover point being dependant upon the shape of the cutout. The orientation of ellipse major axis significantly affects the critical buckling load of a panel. A horizontally aligned ellipse exhibits similar behaviour as that to a circular or square cutout. However when the major axis is rotated relative to the horizontal axis its buckling capacity reduces monotonically till it has a buckling load less than that for an unperforated panel when vertical aligned. It has been shown when a circular cutout is eccentrically placed in a panel, for small cutout sizes the buckling load reduces with horizontal eccentricity while a small increase is experienced for vertical eccentricity. Multiple circular cutouts significantly reduce the buckling capacity of the panel for all cutout sizes and separation distances. Initial geometric imperfection in the panel does not affect the critical buckling load significantly. The postbuckled response of such panels are also insensitive to the magnitude of imperfection. Panels with circular, square and elliptical cutouts exhibit substantial postbuckled strength. The post buckling response of such panels are insensitive to cutout geometry shape.

No description available.

This research explores the use of carbon/epoxy and fiberglass/epoxy fiber-reinforced polymer (FRP) composite rebar manufactured on a three-dimensional braiding machine for use as reinforcement in concrete beams under four-point bending loads. Multiple tows of prepreg composite fibers were pulled to form a unidirectional core. The core was consolidated with spirally wound Kevlar fibers which were designed to also act as ribs to increase pullout strength. The rebar was cured at 121â—¦C (250â—¦F) in an inline oven while keeping tension on the fibers. Five configurations of reinforcing bars were used in this study as reinforcement in concrete beam specimens: carbon/epoxy rebar and fiberglass/epoxy rebar were manufactured on the three-dimensional braiding machine and cured in an inline oven while still under tension immediately after production; carbon/epoxy rebar was manufactured by IsoTruss industries on the three-dimensional braiding machine and was rolled and stored before curing; fiberglass/epoxy rebar was purchased from American Fiberglass; conventional No. 4 steel rebar was also purchased. All bars were embedded in 152 cm (60 in) long, 11 cm (4.5 in) wide, and 15 cm (6.0 in) tall concrete beams. Beams were tested under four-point bending loads after which three 30 cm (12 in) specimens were taken from the ends of each configuration to be tested under axial compression loads in order to investigate the effects of the concrete voids on the concrete strength. Concrete beams reinforced with BYU glass/epoxy rebar manufactured on the three-dimensional braiding machine exhibited 5% greater compression bending stress and 11% greater tension bending stress than concrete beams reinforced with industry manufactured glass/epoxy rebar. Concrete beams reinforced with BYU carbon/epoxy rebar manufactured on the three-dimensional braiding machine exhibited 18% lower compression bending stress and 64% lower tension bending stress than concrete beams reinforced with industry manufactured carbon/epoxy rebar. BYU glass/epoxy rebar has a 3% greater stiffness and 1% greater displacement than industry manufactured glass/epoxy rebar and BYU carbon/epoxy rebar has a 40% greater bending stiffness and 19% lower displacement than industry carbon/epoxy rebar. BYU carbon/epoxy rebar has 49% lower compression bending stress, 1% lower tension bending stress, 28% lower displacement, and a 68% greater bending stiffness than BYU glass/epoxy rebar. BYU glass/epoxy rebar has 38% greater compression bending stress, 30% lower tension bending stress, 26% greater center displacement, and a 105% lower bending stiffness than conventional steel. BYU carbon/epoxy rebar has 8% lower compression bending stress, 31% lower tension bending stress, and 22% lower bending stiffness than steel. The deflections of steel reinforced concrete and BYU carbon/epoxy reinforced concrete are comparable with steel rebar displaying a 1% greater center displacement than BYU carbon/epoxy rebar.

No description available.

Cette étude présente deux objectifs : industriellement elle contribue à la certification d’unaéronef léger de quatre places (MCR 4S) suivant la norme CS 23 régie par l’EASA(European Aviation Safety Agency) et scientifiquement, elle cherche à mieux appréhenderle lien entre la variabilité des matériaux et du comportement des structures. L’étude devariabilité montre une grande dispersion des caractéristiques mécaniques telle qu’il nousest impossible d’optimiser le coefficient de sécurité. A partir de ce constat, une étude plusapprofondie est effectuée et a permis de déterminer les origines de la variabilité à savoir, lanon homogénéité de l’épaisseur et donc de la teneur volumique de fibre dus au processusde fabrication utilisé (moulage au contact).A partir d’une instrumentation par jauges de déformation sur des zones ciblées de lavoilure d’un aéronef prototype, un essai en vol est effectué dans le but d’extraire un spectred’utilisation en cours de service et ce dans les différentes configurations de vol (roulage,décollage, montée, vol en palier, descente, virage et atterrissage). Ce spectre est utilisé parla suite dans un essai de fatigue sur l’aile afin de comprendre les mécanismesd’endommagement dans ce cas de figure. Conjointement, un modèle numérique de lavoilure validé par des essais statiques sur la voilure a permis de mieux comprendre lecomportement de la voilure grâce à une cartographie détaillée des contraintes. / This study has two objectives: industrially it contributes to the certification of a four-seaterlight aircraft (MCR 4S) following the CS 23 standard governed by EASA (EuropeanAviation Safety Agency) and scientifically, it searches to better understand the linkbetween the variability of materials and structural behavior. The study of variability showsa wide dispersion of mechanical characteristics such that is impossible to optimize thesafety factor. From this, further study is conducted and identified the origins of variabilityi.e., non homogeneity of the thickness and therefore the fiber volume content due to themanufacturing process used (contact molding).From instrumentation by strain gages on targeted areas of the wing, a flight test isperformed in order to extract a spectrum of use in service on different configurations flight(taxiing, takeoff, climb, level flight, descent, turn and landing). This spectrum issubsequently used in a fatigue test on the wing in order to understand the damagemechanisms in this case. Together, a numerical model of the wing validated by static testson the wing helped to better understand the behavior of the wing through a detailed stressdistribution.

Various configurations of unidirectional carbon/epoxy composite rods were impacted radially, inspected using micro-CT scanning equipment, and tested in axial compression to measure the residual strength after impact. This data was used to correlate the relationship between impact energy, residual strength, and the peak crack area and total crack volume along the length of the specimens. These specimens represent local members of open three-dimensional composite lattice structures (e.g., based on isogrid or IsoTruss® geometries) that are continuously fabricated using advanced three-dimensional braiding techniques. The specimens were radially impacted with 2.5 J (1.9 ft-lbs), 5.0 J (3.7 ft-lbs), 7.5 J (5.6 ft-lbs), 10 J (7.4 ft-lb), 15 J (11 ft-lbs), and 20 J (15 ft-lbs) of energy, and compared to undamaged control specimens. The unidirectional core specimens were 8 mm (5/16") in diameter and were consolidated with various sleeve configurations and materials. Sleeves differed in types (bi-directional braided sleeves or unidirectional spiral wraps), nominal sleeve coverage of the core fibers (full or half), and sleeve material (Nomex Thread or Dunstone Hi-Shrink Tape). The unsupported length of the specimens used in this research was 50.8 mm (2") to ensure a strength-controlled compression failure rather than a failure due to buckling. After impact, the specimens were scanned using a micro-CT scanner at resolutions of 50 and 35 microns and subsequently tested in axial compression. The micro-CT scan images were analyzed to measure the crack areas along the specimen. From this analysis, the peak crack area and total crack volume along the length of the specimen was calculated. Similar to past research, as the impact energy increases, the residual compression-strength-after-impact decreases. As the impact energy increases, specimens with shrink tape sleeves had the largest increase in peak crack area and overall crack volume while specimens with full spiral sleeves had the lowest increase in peak crack area and overall crack volume. A bimodal increase is evident in the peak crack area and total crack volume over the length of the specimen where specimens impacted at 15 J (11 ft-lbs) showed the highest peak crack area across all sleeve types. There is a slight correlation between the increase in peak crack area and overall crack volume and the decrease in residual compression strength after impact. Shrink Tape, while yielding a higher quality specimen with greater compression strength prior to impact, did not protect the specimens against damage due to impact as well as other sleeve types. This was shown by the large decrease in residual compression strength after impact and increase in peak crack area and overall crack volume as the impact energy increased.

Ce travail prend naissance dans un vaste programme européen d’allègement des structures aéronautiques, mis en application par un projet de recherche (PRC-Composites) et donnant suite à des développements de procédés et matériaux nouveaux. Cette thèse s’inscrit donc dans ce projet global et vise à mettre en évidence, analyser, comprendre puis modéliser les mécanismes d’absorption d’eau dans un composite à matrice organique (carbone/époxy) tissé 2D mis en oeuvre par RTM et à évaluer l’influence de l’humidité sur son comportement mécanique.Une méthodologique consistant à étudier l’effet du vieillissement sur les composants du matériau : résine pure, interface/interphase résine/toron, composite a été développée.Pour la résine pure (RTM6), un couplage complexe entre absorption d’eau et thermo-oxydation de la matrice à 70°C a été démontré, découplé, puis quantifié expérimentalement en s’appuyant sur un modèle numérique diffuso-mécanique couplé. Une étude basée sur l’emploi d’éprouvettes modèles « mono-toron », conçues à cet effet, a montré que – à l’échelle microscopique – la présence d’interfaces/interphases résine/toron n’a pas d’impact significatif sur la cinétique de diffusion dans la matrice. Le comportement du composite (RTM6/AS7) en milieu humide/oxydant a été exploré en s’appuyant sur les notions établies sur les composants. La diffusion d’eau a été modélisée à travers une représentation explicite de la microstructure et en incluant le modèle établi sur la résine pure. Le comportement du composite vieilli a été évalué via des essais de traction uni-axiale dans les 3 directions du plan (0°, 90° et 45°). Ces essais montrent l’intérêt d’une pré-étude sur les constituants. / The initiative of this work is based on weight saving European program of aeronautic structures. It takes part of PRC-Composites project and aim to develop processes and new materials. This thesis belong with this whole project and wants to highlight, to analyze, to understand, then to modelize water absorption phenomena in an organic composite (carbon/epoxy) 2D woven, manufactured by RTM process. The final objective is to evaluate the effect of moisture on the mechanical behavior.The method consists in studying the effect of water ageing on components: neat resin, interface/interphase resin/tow, and composite.For the neat resin (RTM6), a complicated coupling between water sorption and thermo-oxidation of the matrix at 70°C has been demonstrated, uncoupled and then experimentally quantified using a coupled diffuso-mechanical model. A study based on the use of “mono-tow” samples, specially manufactured for this work, has shown – on the microscopic scale – the presence of interface/interphase resin/tow has no significant effect on the diffusion kinetic through the matrix. In moist/oxidant environment, the composite behavior (RTM6/AS7) has been explored relaying on established ideas related to components. Water diffusion has been modelized through an explicit representation of the microstructure and by including neat resin established model. The aged composite behavior has been evaluated with uni-axial tensile tests in 3 directions (0°, 90° and 45°). These experiments show the interest of the component pre-study.

Cette thèse s’inscrit dans le cadre d’un vaste programme visant à établir un premier retour d’expérience sur des structures aéronautiques en fin de vie. L’objectif des travaux présentés ici est donc de caractériser le vieillissement et les propriétés résiduelles de pièces provenant d’avions après démantèlement, et donc après service. Plus précisément, deux matériaux de nature différente sont considérés : un alliage d’aluminium 2024-T351, constitutif d’une voilure d’A320 ; et un composite carbone/époxy T300/914, prélevé sur les voilures d’un Falcon X et d’un ATR.Pour les voilures composites, les travaux ont porté sur les effets de l’eau des stratifiés. L’évolution de la température de transition vitreuse en DMA a été étudiée en fonction du taux d’humidité présente dans le stratifié. Les résultats d’essais de sorption set de désorption ont été confrontés à différents modèles de diffusion. Les propriétés résiduelles ont été évaluée au travers de divers essais mécaniques. Il ressort de cette étude un très bon comportement du composite après service.Les travaux concernant la voilure métallique se focalisent sur les propriétés résiduelles en fatigue de l’alliage de voilure. Les résultats montrent qu’un durcissement structural apparaît en service, et qu’un léger abattement de la durée de vie en fatigue est observable. Cependant, le comportement à la fissuration est inchangé en comparaison avec un matériau « neuf ».Au final, ce premier retour d’expérience est positif. Il peut maintenant permettre aux avionneurs de vérifier les règles utilisées lors de la conception ou d’optimiser certains dimensionnements, mais aussi de justifier des extensions de durée de vie des avions. / The thesis is part of a larger program aimed at establishing a first feedback on structural health of aeronautical structures at the end of life. The aim of the work presented here is to characterize the residual properties after aging of parts from aircraft after teardown, and therefore after service. Specifically, two different types of materials are considered: an aluminum alloy 2024 T351, constituting the underside of an A320 wing, and a composite carbon/epoxy T300/914, taken from the wing of a Falcon X and the wing of an ATR.Concerning the composite wings, the study focused on the effects of water on laminated composites. The evolution of the glass transition temperature by DMA has been studied as function of moisture present in the composite. The results of sorption and desorption tests were confronted to different diffusion models. Residual properties were evaluated through various mechanical tests. It is clear from this study a very good behavior of the composite after service.The work on the metal wing is focused on the residual fatigue properties of these alloys. The results show that hardening occurs in service, and a slight reduction of the fatigue life is observed, the number of cycles to failure ranging between 104 and 106. However, the fatigue crack growth resistance is unchanged in comparison with a “virgin” material.Finally, this initial feedback is positive: It can now enable manufacturers to check the design rules or to optimize the design, but also to justify aircraft life extensions.

The present study considers the dependence of mechanical properties in composite laminates on the fibre architecture. The objective is to characterise the mechanical properties of composite plates while varying the fibre distribution but keeping the constituent materials unchanged. Image analysis and fractal dimension have been used to quantify fibre distribution and resin-rich volumes (RRV) and to correlate these with the mechanical properties of the fibre-reinforced composites. The formation, shape and size of RRV in composites with different fabric architectures is discussed. The majority of studies in literatures show a negative effect of the RRV on the mechanical behaviour of composite materials. RRV arise primarily as a result of (a) the clustering of fibres as bundles in textiles, (b) the stacking sequence, and/ or stacking process, (c) the resin properties and flow characteristics, (d) the heating rate as this directly affects viscosity and (e) the consolidation pressure. Woven glass and carbon/epoxy fabric composites were manufactured either by the infusion or the resin transfer moulding (RTM) process. The fractal dimension (D) has been employed to explore the correlation between fabric architecture and mechanical properties (in glass or/ carbon fibre reinforced composites with different weave styles and fibre volume fraction). The fractal dimension was determined using optical microscopy images and ImageJ with FracLac software, and the D has been correlated with the flexural modulus, ultimate flexural strength (UFS), interlaminar shear strength (ILSS) and the fatigue properties of the woven carbon/epoxy fabric composites. The present study also considers the dependence of fatigue properties in composite laminates on static properties and fibre architecture. Four-point flexural fatigue test was conducted under load control, at sinusoidal frequency of 10 Hz with amplitude control. Using a stress ratio (R=σmin/σmax) of 0.1 for the tension side and 10 for the compression side, specimens were subjected to maximum fatigue stresses of 95% to 82.5% step 2.5% of the ultimate flexural strength (UFS). The fatigue data were correlated with the static properties and the fibre distribution, in order to obtain a useful general description of the laminate behaviour under flexural fatigue load. The analysis of variance (ANOVA) technique was applied to the results obtained to identify statistically the significance of the correlations. Composite strength and ILSS show a clear dependence on the fibre distribution quantified using D. For the carbon fabric architectures considered in this study, the fatigue properties of composite laminates have significant correlations with the fibre distribution and the static properties of the laminates. The loss of 5-6 % in the flexural modulus of composite laminates indicates an increasing risk of failure of the composite laminates under fatigue loads. The endurance limits, based on either the static properties or the fibre distribution, were inversely proportional to the strength for all laminates.

This research demonstrated the fabrication, flexural testing, and analysis of nominally 5 ft (1.5 m) 6-bay and 10 ft (3 m) 12-bay carbon/epoxy IsoBeam™ structures. The rectangular cross-section was 5 in (12.7 cm) wide by 10 in (25.4 cm) high. The IsoBeam structure is a composite lattice structure that is a geometric derivative of the IsoTruss® structure. Modifying the geometry to yield a rectangular cross-section provides additional applications for these beams as structural elements in buildings, aircraft, vehicles, and other structures. The diameters of the constituent members of the IsoBeam, namely the longitudinal and diagonal members, were sized such that the IsoBeam could hold the design load of a 10K1 steel joist 550 plf (818.5 kg/m). Three IsoBeam structures were manufactured: two 5 ft (1.5 m) long and one 10 ft (3 m) long. The IsoBeam structures were manufactured with carbon/epoxy composite tows comprised of T700SC-12K-50C carbon fibers and UF3369-100 pre-impregnated (pre-preg) epoxy resin. The pre-preg tows were positioned on a modified pin-mandrel under tension using a combination of hand and machine filament winding in an interwoven pattern to create the complex geometry of the IsoBeam structure. Each member was circumferentially wrapped with 1 in (2.5 cm) wide strips of Dunstone Hi-Shrink Tape (polyester) to consolidate the tows during the manufacturer’s recommended curing process. Microscopic measurements after testing established that these careful manufacturing techniques produced high-quality specimens with an average void ratio of 0.72% and an average fiber volume fraction of 69.5%. The average compression stiffness and strength were 18.7 ksi (129 GPa) and 115.1 ksi (793 MPa), respectively.Each IsoBeam was loaded in four-point bending to failure, with other tests performed in the linear-elastic range to study load path behavior of the IsoBeam. Strain, deflection, and load data were collected to provide a detailed understanding of the behavior of individual members under load and their corresponding stresses. The 6-bay IsoBeam structures experienced failure at 8055 lbs (35.8 kN) and 11224 lbs (49.9 kN), in compression initiated by buckling. The longer 12-bay IsoBeam structure failed in a similar manner at 8035 lbs (35.7 kN) but also exhibited delamination, due to insufficient interweaving.Experimental results were compared to the predicted strength of the IsoBeam based on a linear finite element model (created using SAP 2000) and hand calculations. Validation of the design through the comparison of experimental and predicted values gave insight on design techniques and overall understanding of the performance of the IsoBeam in bending, with excellent correlation in the linear range. The assumption that longitunals are primarily responsible for bending strength and diagonals primarily carry shear was validated, indicating a strong correlation between manufacturing quality and performance of IsoBeam structures.