By the progress in resin, mould and moulding technique, it is now possible to make various products of plastics. Mainly, these products are made of thermoplastics or fibre reinforced thermoplastics (FRTP) and are moulded by injection. In the injection moulded parts, the orientation of short fibres is determined by the flow of molten resin and therefore the mechanical properties are highly dependent on the flow condition. From this reason, it is important to make clear the flow condition. In this paper, the moulding process is investigated microscopically in order to solve the unstable flow problem. In experiment, the investigation on the flow front shape of cylindrical products in short shot moulding was carried out to observe the unstable flow. The result showed the asymmetrical flow front shape resulting from the unstable flow behavior. The flow analysis was carried out, by applying the eigenvalue analysis. The fibre orientation analysis was also performed, by using the results of this flow analysis. These numerical analyses were in good agreement with the experimental results. It is considered that this flow analysis using eigenvalue analysis is effective to express the unstable flow.
This paper deals with the optimum design of fibrous laminated composite plates. The plates are subjected to any concentrated and/or distributed loads and they are optimized under the constraint that the strain energy or the maximum deflection of the plates is minimized for the given thickness. The design variables are the fiber orientation angles of the small sections of the plates. Thus, the optimum fiber orientation becomes curvilinear. The deflection analysis is performed by using the Rayleigh-Ritz method and the complexity in developing the analytical formulation for the sectional plate is reduced by using the symbolic calculation system, Mathematica. The optimizer has a SUMT-type algorism. The calculated results show that the laminated plate with the sectionally-optimized fiber orientation angles yield a considerable improvement in the strain energy and maximum deflection.
Generally woven cloth laminates have a typical right-angled anisotropy. Their thermal and mechanical properties are also dependent upon temperature at cryogenic atmosphare. Therefore, it is necessary for cryogenic use to understand the anisotropy of FRP. In this report, the thermal contract and bending property of woven cloth laminates have been studied by changing the off-axis angle in order to analyze the anisotropy of FRP for cryogenic use. The angle-ply FRP, made by piling up the woven cloth lamina at a fixed angle, and hybrid FRP have also been investigated so as to reduce the anisotropy of the thermal contraction and bending property. The following results are obtained. (1) The thermal contraction of cross ply FRP has a peak value when the off-axis angle becomes 45°, that is the maximum mismatch angle made by the principal axis and fiber direction of the specimen. Especially AFRP shows a notable anisotropy. (2) The 45° angle ply FRP and hybrid FRP are effective to reduce the anisotropy of thermal contraction and bending property at low temperatures.
A terminal of Kevlar rope with a metal socket was loaded statically and repeatedly to assess the effect of high water pressure. The experimental results show that the water pressure increased both the static and fatigue strengths of the terminal. An analytical simulation of this terminal was also performed using the finite element method, where contact pressure, friction and slip process were taken into account. The detailed stress distribution and concentration were estimated along the interface between the rope and tapered socket. It was found that there are two stress concentrations on the Kevlar rod surface. One is at a concave being made in the forming process and another is at a contact point between the rod and socket. When the water pressure is low enough to keep the rod-slip out of the socket within a small quantity, two stress concentrations yield a single peak near the socket edge. High water pressure decreases the stress concentration factor at a contact point, which explains the present experimental result.
A personal computer program by Finite Element Method (FEM) has been developed in order to analyze the fracture mechanism of composite materials with nonlinear properties. The mechanical behavior of a structure with local failures or damages may be shown by a decrease in Young's modulus and Poisson's ratio of the broken elements in FEM. Therefore, Young's modules and Poisson's ratio for fracture elements were assumed to be 0.01(MPa) and 0.0, respectively. As the examples, the mechanical behaviors of a hybrid composite and an angle-ply laminate under tensile load and of a concrete beam with FRP sheet under bending load were analyzed by the developed computer program. The diagrams showing the relation between the applied load and the displacement up to failure were obtained. These results have revealed that the developed program is very useful for the fracture simulation of structures made of composites.
In this paper, the fracture mechanism and AE behavior of a short carbon fiber reinforced nylon 66 composite with a center crack through the thickness during monotonous tensile deformation are studied. It is shown that most of AE signals occur in the region of stable crack propagation and there is an obvious increase in AE activity at the initial stable cracking. The dimple patterns can be observed in SEM for all of materials and the AE characteristics of them are clarified. It is also shown that a proportional relation between J integral and COD holds well for the composites, but the proportional coefficient reduces as the fiber content or the ratio of crack length to specimen width increases. Using AE energy aspect ratio, which is defined as the ratio of the cumulative AE energy to the one at the initial stable crack, an exponent relationship between AE energy aspect ratio and J integral can be obtained experimentally. The value of the exponent m in the above relation is roughly the same during stable crack propagating even if the fiber content or the ratio of crack length to specimen width varies.
In order to improve the reliability and safety of composite materials, it is necessary to have non-destractive testing (NDT) available to detect defects in materials. Although conventional ultrasonic methods have, in principle, been used for NDT, they are inefficient and difficult to inspect large structures, such as ship hulls, because of the high attenuation of ultrasonic waves. In these cases, low frequency waves have an advantage over ultrasonic waves. Therefore, by using low frequency Lamb waves generated by tapping as a means of NDT in composite laminates, the following examinations were made; (1) predicting the thickness of plates, (2) detecting the delamination, (3) predicting the fiber contents in the composite laminates, and (4) measuring the wave velocities depending on propagating directions. (1) and (2) were carried out by using antisymmetric A0 mode waves. The results showed that it was possible to predict the thickness and to detect the delamination based on the relationship between the phase velocities of Lamb waves, which has dispersion characteristics, and the product of frequency and plate thickness. (3) was examined by the resonant method of symmetric S0 mode waves, and (4) was done by S0 waves. The results showed that the fiber content could be predicted from the stationary waves and the wave velocity propagating at an angle to the fiber direction could be calculated from the Young's modulus.
Optical fibers were integrated into graphite-epoxy composites in their manufacturing process. Two types of embedded fiber optic sensors were tested. One was an embedded fiber optic strain sensor, and the other was an embedded sensor for detection of cracking, utilizing optical fiber breaks. First, in order to evaluate the bonding property between optical fiber and epoxy resin, the optical fibers were integrated into epoxy resins, and the tensile test was conducted. As a result, by removing an acrylate coating of the optical fiber with H2SO4, the optical fiber showed good adhesion to the surrounding epoxy resin. The ultimate strain of the uncoated optical fiber was about 1.4 to 1.5%. Second, the embedded fiber optic strain sensor was evaluated. The fiber optic strain sensor consisted of Michelson interferometer using a single-mode coupler. The strains in graphite-epoxy composite laminates were measured by the embedded fiber optic strain sensor, and their strain sensitivities agreed well with the theoretical values. Third, the crack detection experiment using the embedded optical fiber was conducted. When the crack was induced by the tensile load, the embedded optical fiber was broken and the light through the optical fiber went out immediately. Such a relation between cracking and optical fiber break was confirmed by tension test with ultrasonic C-scan images. The cracking induced by the impact load in the composite laminates was also monitored. Internal cracking was induced at the early stage of impact load. As the results of these evaluations, it was proved that the embedded fiber optic sensors were effective for the measurement of strain and the detection of cracking in graphite-epoxy composite laminates.
The hybridization of glass fiber with aramid or carbon fiber in FRP laminates offer a promising method to exhibit the excellent performance of aramid and carbon fibers while restricting an increase in material cost. In this paper, the static tensile, compressive and flexural properties of glass/aramid and glass/carbon hybrid FRP laminates have been evaluated systematically from the viewpoint of hybridization merit of FRP. As a results, it has been made clear that the hybridization of glass and aramid fibers is a very useful method to overcome the weak point of aramid fiber which has a low compressive strength.
Effect of surface modification by ion implantation on fatigue behaviors was investigated on two sorts of 13Cr martensitic stainless steels with high tensile strength of about 1900MPa, under an out of plane bending load condition. Nitrogen ions were implanted into the surface of thin plate specimen of 1mm thick with several energy levels from 0.8 to 1.2MeV and with doses in the range of 1013-1016ions/cm2. Effect of post heat-treatment on the fatigue strength was also investigated for the specimens ion-implanted with several different conditions. Results of a series of fatigue tests revealed that the surface modification by ion implantation raised fatigue resistance of the materials by choosing appropriate implantation conditions. Furthermore, a stable increase in fatigue strength could be attained when the specimen was annealed after ion implantation. Observations of initiation and propagation of surface fatigue cracks indicated that the thin surface layer modified by ion implantation had an arresting effect against crack initiation though it gave no effect on the propagation of large crack. Characteristics of the modified surface layer were evaluated from several viewpoints; penetration depth of nitrogen, hardness and residual stress by X-ray stress measurement method. The results revealed the following features: 1) The surface layer thickness modified by ion implantation was less than 1μm in the case as implanted but was extended to 1-2μm by the post heat-treatment, 2) the hardness at the surface layer showed a trend to rise with an increase in annealing temperature, and 3) the high compressive residual stress was measured at the implantation dose of 1016ions/cm2.
Creep fatigue tests were conducted on smooth specimens of a Type 304 stainless steel under a tensile stress hold (cp-type) condition at 1073K, and pre-creep damage was introduced in the form of small intergranular cracks distributed uniformly through the thickness. Macrocrack propagation tests in the high temperature fatigue were then carried out with the pre-damaged specimens. The results obtained are summarized as follows: (1) The pre-damage accelerated the successive macrocrack propagation in cp-type fatigue at 923K for an equal value of creep J-integral range, ΔJc, which governs the crack propagation rate of virgin material in time dependent fatigue. (2) Although the crack propagation rate of pre-damaged material in cp-type fatigue at 1073K was faster than that at 923K for ah equal ΔJc, it coincided with that of virgin one. The multiple small cracks were found on the specimens of virgin material as well as pre-damaged one near the macrocrack, which implied that the small cracks were generated not only in prior creep fatigue but also during the propagation test. Thus, the acceleration in cp-type is caused by the small cracks. (3) The crack propagation rate of pre-damaged material in cycle dependent (pp-type) fatigue was nearly 10 times faster than that of virgin one. The observation of crack morphology revealed that the accelerated fatigue crack propagation was caused by the coalescence with intergranular small cracks.
Creep-fatigue crack growth behaviours of a Ti-6242 alloy, a low carbon steel, a Type 304 stainless steel and Hastelloy X under reversed loading patterns (C-P, C-C, P-P and saw-tooth of slow-fast type) were investigated in air and vacuum environment in the light of fracture mechanics and fractography. The crack growth rate in each of the materials tested was successfully correlated in terms of the cyclic J integral range ΔJ irrespective of loading patterns. In the region of low growth rate, the crack growth rates of all the materials were about the same for the same value of ΔJ. In the region of high growth rate, on the other hand, the growth rates were somewhat different depending on the creep ductility of material, those with lower ductility giving higher growth rate for the same ΔJ value. Significant difference was not found between the crack growth rates in air and vacuum, and it was consistent with little difference observed between the fracture surface morphologies in these environments. It was confirmed by creep void observation on the cross section of specimens that the difference in fracture morphology between C-C type and C-P type loading was due to the recovery of grain boundary sliding during compression hold in the C-C type loading. An experimental evidence was obtained to suggest that the better correlation in crack growth rate in terms of the separated J integrals-fatigue and creep J integrals, which was reported on a low carbon steel, was possibly related to the specimen geometry rather than the creep-fatigue interaction.
The unidirectional CFRP has a weak point in interlaminar strength and therefore, delamination resistance has been evaluated on many types of unidirectional composites and the evaluation method is now being established. In these studies, however, unidirectional reinforced CFRP, in which the stacking sequence is n, was used. Thus, the measured delamination resistance is not applicable to evaluate the delamination resistance of actual composite structures made from multidirectionally reinforced CFRP. Moreover, the delamination micromechanism of the multidirectionally reinforced CFRP has not been understood yet. Therefore, the material design to improve the delamination resistance of the actual composite structures has not been conducted. In this study, the objective was to examine the delamination resistance curves of multidirectional reinforced CF/Epoxy composites and to investigate problems on the testing method. Additionally, the micromechanism was modeled based on the observations by a scanning acoustic microscope and a scanning electron microscope. As the results, it was shown that the delamination crack propagating between 90° laminae has the highest resistance, and the crack deflection and branching are most effective for increasing the delamination resistance.
Thermoplastic composites have several advantages over thermoset composites in productivity and properties. However, thermoplastic composites have not been used so widely. Several impregnation techniques have been specially developed for thermoplastic composites to solve the difficulty in impregnation of their matrix resins. In this paper, we propose a new impregnating system using a commingled yarn of spun yarn, in order to facilitate the impregnation process. This yarn consists of carbon fibers and Nylon 6 fibers. We fabricated unidirectional composites using spun yarn. We performed a longitudinal bending tests and examined the influence of processing conditions on the bending properties. It was observed that the resin-rich regions between fiber bundles in composites were formed due to the original twist structure of fiber bundle. Impregnating behavior of resin was affected by processing conditions and progressed with two steps, that is, the resin firstly impregnated inside of fiber bundles and next impregnated between fiber bundles. Longitudinal bending strength was saturated at a rather lower pressure and in a shorter period of compression.
In order to clarify the mechanical behavior of living trees, bending tests were carried out. Two sites in the Miura peninsula where wind blows second strongestly in Kanagawa Prefecture, one site at the tip of the peninsula and the other in a valley about 300m inland, were chosen for the study. The mechanical properties (σm, εm and Um: stress, strain and strain energy at maximum load) were determined by three point loading bend test. The species in bending were Enoki (Celtis sinensis var. japonica), Mochinoki (Ilex integra), Hisakaki (Eurya japonica), Tobera (Pittosporum tobira), Masaki (Euonymus japonicus) and Yabunikkei (Cinnamomum japonicum). The experimental results obtained can be summarized as follows: (1) σm were not independent of the part or branch size of the tree. (2) εm decreased with an increase in sample size. (3) σm was higher in stronger wind than in weaker wind. (4) Um was higher in stronger wind than in weaker wind. (5) The ratio of the strain energy per volume Um/V to σm·εm/24 did not depend on wind conditions, so σm·εm/24 could be a mechanical index for the particular species. (6) Um/V was in proportion to εm.
The main purpose of this study is to make clear the variation of strain induced by the excavation of a shaft with a diameter of 6m and a depth of 150m. Variation of strain around the shaft was measured using 8-element borehole guages embeded before excavation, and a numerical analysis was conducted for comparison with the results of in situ strain measurement. The following conclusions were obtained: (1) Strain variation around shaft is induced by the extraction force acting on the wall of the opening. (2) Variation of strain in sedimentary rock with wide to moderately spaced joints can be simulated by the FEM analysis using an elastic model.
In order to express impact strength of materials in stress unit over a wide range of impact velocity, a new testing method is proposed. The experimental apparatus used in this study was composed of a gas gun and a loading apparatus for impact tension or impact bending test. Fracture strength was evaluated based on the strain history measured by the strain gage cemented on a specimen. Commercial JIS. SS41 carbon steel specimens were used to show the usefulness of this new testing apparatus. In order to carry out impact tension or impact bending test, a plate or conventional Charpy specimen conforming to JIS. 5 was used, respectively. A strain gage was cemented on the central portion or notch root of each type specimen. Then, impact bending strength was expressed by the nominal stress taking the stress concentration by the notch into account. Thus, the experimental results were expressed in stress unit instead of energy unit. By using this apparatus, low temperature tests can be carried out under more strictly controlled temperature conditions in comparison with a pendulam type tester such as Charpy impact tester. The impact tensile strength Im, t and the impact bending strength Im, b evaluated by this method increased with an increase in impact velocity. At low temperature, Im, t was higher than that of room temperature. These impact strengths, Im, t and Im, b, took the values of about 2 GPa at the impact velocity of 80m/s. For the lower impact velocity range, the impact bending strength was slightly lower than the impact tensile strength at 77K.