A constitutive model to describe time-dependent inelastic behavior of unidirectionally fiber-reinforced metal matrix composites is developed from phenomenological and continuum mechanics points of view. The fibrous composite is treated as a homogeneous medium which hardens with inelastic deformation. The inherent anisotropy is assumed to be transversely isotropic. The constitutive modeling is based on the well-established thermodynamic formalism for internal state variable theories, where the thermodynamic potentials are defined using a transversely isotropic tensor of the fourth order. First, a kinematic hardening model is derived in the invariant form. In this model, the evolution of the internal state variable is prescribed by the Bailey-Orowan type so that both the transient and steady-state creep behavior can be described. Then, an isotropic hardening model is formulated by assuming a particular representation of the kinematic hardening variable. It is found that the isotropic hardening model can yield the power law relation for steady state responses.

View full abstract

CFGFRP (carbon fiber-glass fiber-reinforced plastics) composites are materials with a self-diagnostic function for detecting latent damage. Three type of carbon fiber bundles with different ultimate elongations and one type of glass fiber bundle with a large ultimate elongation were tested as conductive fibers and insulating fibers, respectively. Electrical resistance increased with increasing strain, and a tremendous change was seen at the transition point where the carbon fiber bundles fractured. Data obtained by the acoustic emission method were measured simultaneously, and these changes showed results similar to changes in the electrical resistance. Permanent, residual strain and residual electrical resistance were observed even after the removal of load, and these changes were dependent on the maximum strain applied in the past. Latent damage of CFGFRP composites can be detected by a method based on the measurement of their changes in residual electrical resistance after unloading. The method based on measuring changes in the electrical resistance of the CFGFRP composites during and after loading is simple in terms of technique and manageable in terms of cost for detecting latent damage and preventing fatal fracture.

View full abstract

In order to study the effects of dissolved hydrogen atoms and hydrogen damage on elongation of mild steel in relation to the susceptibility of iron and steel to hydrogen embrittlement, tensile tests were carried out immediately after cathodic hydrogen charging and after degassing of hydrogen at room temperature. Because cathodic hydrogen charging caused internal damage, such as microcracks (blisters) and plastic deformation owing to hydrogen precipitation, the influence of the dissolved hydrogen on the elongation was distinguished from that of the hydrogen damage. The dissolved hydrogen and hydrogen damage were found to cause reversible and irreversible reductions of elongation, respectively.

View full abstract

Rotating bending fatigue tests at room temperature were carried out on a ferritic spheroidal cast iron (FDI) and a ferritic/pearlitic spheroidal cast iron (FPDI). Two series of fatigue specimens were machined out from a lower and an upper positions of the cast Y-blocks, which were called "lower specimen" and "upper specimen", respectively. Using these specimens, the fatigue strength and fatigue life distributions were investigated. The major results obtained were as follows. (1) The fatigue strength of lower specimens was higher than that of upper ones for FDI and FPDI. (2) In both materials, the average graphite size of the lower specimens was smaller than that of the upper one. (3) The average graphite size had a main effect on fatigue strength and fatigue life distributions.

View full abstract

Low-cycle fatigue crack initiation process of Austenitic stainless steel SUS316 have been studied at 700°C in air and vacuum with an attention to the role of local oxidation on grain boundary sliding and transgranular slip. In vacuum, local cumulative displacements at grain boundary due to sliding saturates in an early stage of fatigue life, while local cumulative displacements due to transgranular slip develops proportionally to the number of strain cycles and results in slip bands crack initiation. In air, local oxidation attack on the newly originated surface due to grain boundary sliding can be a primary reason for the intergranular crack initiation. Intergranular oxidized cracking is induced by wedge shaped oxide with micro-cavities in the process of grain boundary sliding, while transgranular slip bands simply cause small steps on the surface and this oxidized transgranular slip steps do not develop into transgranular cracking in air.

View full abstract

Low cycle fatigue tests and fatigue crack growth rate tests were conducted at ambient temperature on fine and coarse-grained IN 718 of which grain size numbers were ASTM No. 10 and ASTM No. 5, respectively. In the low cycle fatigue tests on fine-grained material at maximum stress of 1180 MPa and R=0, fatigue lives of ASTM standard cylindrical specimens ranged from 50000 to 70000 cycles, and every fatigue crack was initiated from the defect on the surface of specimen, i.e., nonmetallic inclusion or carbide. The Fatigue crack growth rate test showed that small surface cracks in the coarse-grained material were propagated considerably faster than those in the fine-grained material. If the small surface cracks started to be propagated from the defects immediately after the start of cyclic loading, the difference in fatigue life between fine-grained material and coarse-grained material resulted from the difference in the small surface crack growth rate.

View full abstract

A method of detecting fatigue crack initiation in some weak section of a machine element is examined using a holed specimen under random load conditions. Waveforms of two strain functions, that is h1 = ε_y1 + λε_y2 and h2 = ε_y1 - λε_y2, are successively observed. ε_y1 and ε_y2 are the strains in the vicinity of the hole and λ is the strain range ratio Δε_y1/Δε_y2. The function h1 can be used to substitute for the load parameter and the waveform of h2 changes its shape after the crack initiation due to the crack closure behavior. Therefore, we can detect the fatigue crack initiation at the hole by comparison of these two waveforms. Also, these waveforms are analyzed with FFT (Fast Fourier Transformation) to allow for a more effective detection of crack initiation. Power spectrum density of the waveform h2 varies immediately when the crack length measured from hole edge reaches about 0.5 mm. The figures showing the coherence of two waveforms h1 and h2 clearly vary at the same length 0.5 mm. Thus, the crack initiation of its length of 0.5 mm can be easily detected with the analysis of the strain waveform.

View full abstract

The fracture behavior and crack propagation process of carbon-fiber-reinforced thermosoftened plastic (CFRTP) materials are examined under SEM in situ, and acoustic emission signals are also measured simultaneously. As a result, in the model materials fiber breakages do give rise to AE signals. In CFRTP materials, the relationships between fracture factors and AE parameters can be made clearer since the fractography technique is combined with the method of acoustic emission in real time. It is found that the crack suspended after a rapid propagation, and this phenomenon recurred although the tensile load in the experiment was increased monotonously. Corresponding to this phenomenon of crack propagation/suspense, the characteristics of AE events and power spectrum varied sensitively. Furthermore, it was shown that the degree of damage of materials during the fracture process can be evaluated by using AE cumulative energy and analysis of power spectra.

View full abstract

In the previous report, it was found that low-cycle fatigue strengths were accurately estimated by a multivariate analysis on basic material factors such as chemical composition and heat treatment temperature under limited conditions which were classified by the materials used, the heat treatment conditions being the same as those in the case of high-cycle fatigue properties. In this report, a procedure for the prediction of S-N curves was examined using the same analytical procedure developed previously. It was concluded that the procedure developed in this report is quite suitable for the prediction of S-N curves.

View full abstract

An investigation was made concerning the influence of surface removing on the microfracture strength of the surface layer of structural ceramics as evaluated by sphere indentation tests, where the microfracture strength was defined as the maximum tensile stress at ring crack initiation. Indentation tests were carried out on polished Si₃N₄ plates using a Si₃N₄ spherical indenter. Conventional bending tests were also conducted. As in unpolished specimens, the scatter of the microfracture strength was found to be much smaller than that of the bending fracture strength. Thus, surface removing does not alter our previous finding that the microfracture strength can not be related to the bending fracture strength based on the concept of effective area. An idea based on fracture mechanics was suggested to relate the two kinds of strength.

View full abstract

Two specified loads, P_b and P_c, of short-fiber-reinforced ABS composites with various fiber contents are determined by the stress intensity factor method. The loads of P_b and P_C determined by this method are in good agreement with loads determined by the AE energy gradient method. The damage at the notch tip is observed through a monitor microscope at or near the loads of P_b and P_C. The fiber breakage estimated from AE frequency analysis at the loads is considered to be the dominant factor of the damage mode. It was found that the loads of P_b and P_c are important parameters to clarify the fracture mechanisms of the composites.

View full abstract

Fracture toughness tests have been made in a transition temperature region at stress ratio R=-1.0 using 1TCT specimens which were cut out from SM41C carbon steel. Considerations of a plane strain fracture toughness and a plane strain condition at a brittle fracture were performed. The main results obtained were as follows. (1) The plane strain fracture toughness in the transition temperature region was simply obtained from a test which was carried out at R=-1.0 using a few small specimens. (2) A stress condition occurring in front of a crack at a brittle fracture satisfied the plane strain condition. (3) da/dN from the fatigue test data at stress ratio R=-1.0 can be extrapolated by ΔJ even in the transition temperature region. (4) The plane strain fracture toughness can be obtained from the minimum value of the fracture toughness test data using 20 three-point bend specimens in the transition temperature region.

View full abstract

To assess the structural integrity of high-pressure autofrettaged vessels, systematic sensitivity analysis was conducted for fatigue crack growth life N_f from an initial defect. The factors with the greatest influence on N_f are the autofrettage radius (i.e., compressive residual stresses) and initial defect size. Each contribution was clarified quantitatively. Fracture toughness K_IC of the material and occurrence of partial crack closure due to compressive residual stress have no marked effect on N_f. However, increase in K_IC increases the critical defect size which promotes integrity through high detectability of the defect during in-service inspection. N_f for a semielliptical crack (aspect ratio = 1/6) is about 60%∼70% that for a straight crack. The latter assumption is relatively useful considering its simplicity. The concept of "leak before break" does not hold for this particular type of vessel. However, as N_f is very large compared to the assumed size of the allowable defect described in ASME Boiler and Pressure Vessel Code, Sec. XI, structural integrity is satisfactorily assured.

View full abstract

Currenty there are a number of general purpose computerized symbolic manipulation systems in use such as REDUCE, Mathematica and MACSYMA. These systems can significantly reduce the tedium of analytic calculation and increase its reliability. However, in general, the speed of symbolic computation is considerably slower than that of numerical computations. Therefore, more efficient methods for performing the symbolic operations to reduce the CPU time required are needed. In this paper, in the transient analysis of a stress wave, we show the relationship between the number of rays and the CPU time required in two analytical methods, Maclorin expansion and inverse matrix expansion, numerically and reveal that the method of inverse matrix expansion is more desirable

View full abstract

Stress analysis of transistor structures is performed considering internal stress of thin fims. Internal stress of newly employed films such as amorphous-silicon and tungsten-silicide films are measured by detecting surface curvature change of the film-covered substrates, as a function of temperature. Internal stress of both films changes upon annealing due to phase transition, and reaches about 1000 MPa. The predicted stress of transistor structures without considering the internal stress of the films differs markedly from the measured result obtained using microscopic Raman spectroscopy. On the other hand, the predicted stress considering the internal stress of the films agrees very well with the measured data. Stress design is performed on the actual transistor structure considering annealing temperature dependence on internal stress of the amorphous-silicon thin film to eliminate dislocation generation. It is confirmed that stress design is effective in improving device reliability.

View full abstract

A method of elastic analysis for SCF (single-wall corrugated fiberboard) under uniform bending in the moment axis perpendicular to the machine direction is studied. The following results are obtained. (1) The deformation of SCM (semichemical corrugating medium) is produced from the fixed moment at the position where SCM and KL (kraft liner) are joined. (2) In the SCF, the bending stress increases markedly with the increase of the distance from the center of the thickness. The maximum value σ_{s max} of the absolute stress for SCM is at the inner surface of the joint positions. The ratio [(the maximum stress of KL)/σ_{s max}] is about 30.

View full abstract

In the case of conventional analysis of beams, lateral components (along the width) of the beam are neglected, but this assumption is not always suitable for the case of real beams. In this paper we propose general higher-order equations for one-dimensional static and dynamic beam theories including the effects of lateral components as well as transverse components of general anisotropic beams by expanding the displacements of double infinite-power series with respect to the transverse and lateral coordinates of the beam. Some typical examples of the general higher-order equations given by the truncation of the series are compared with the previously proposed beam theories, and the interrelations among those theories are examined. Futhermore, static and dynamic numerical examples are shown for the cases of infinite beams subjected to sinusoidal loadings and frequency spectrum of simply supported beams.

View full abstract

In this paper, static and dynamic numerical examples are presented to examine the accuracy characteristics of the theories which were formulated based on the most generalized successive higher-order approximation method in the 1st report. Simultaneously, typical successive approximation theories generalizing several previously proposed kinds of consistent theories are examined ; these were deduced by introducing some specializing assumption into the formulation method in the 1st report. From these studies, the accuracy characteristics and applicable limits of successive approximation type plate theories will be shown

View full abstract

Tensile tests with loading and unloading under various strain rates ε were performed for TiNi shape memory alloy wires at various constant temperatures. The influence of strain rate on the stress-strain-temperature relationship was investigated. The main results were summarized as follows. (1) For ε<10%/min, the martensitic transformation stress and its reverse transformation stress did not depend on ε. These stresses were almost constant in the transformation regions. (2) For ε≥10%/min, the martensitic transformation stress increased and its reverse transformation stress decreased with an increase in ε. These stresses fluctuated in the transformation regions. (3) The influence of ε on the transformation stress due to R-phase transformation was slight. (4) The recoverable strain energy density depended slightly on ε but increased significantly in proportion to temperature. The dissipated strain energy density depended slightly on temperature but increased in proportion to ε for ε≥1%/min. (5) The influence of ε on deformation properties is important for evaluation of mechanical properties of shape memory alloy elements.

View full abstract

In the bifurcation analysis of the multilayered cylindrical pressure vessel, it has been assumed that the adjacent layer deforms continuously. The condition necessary for the solution of the incremental boundary value problem was also the continuous deformation. However, the conventional analysis may not be suitable when the slip between the layers of the multilayered cylindrical pressure vessel is taken into consideration. In this study, we propose the bifurcation analysis method of the multilayered cylindrical pressure vessel including the slip between the layers. For the simplicity and the lower-limit solution, we assumed a frictionless slip between the layers. With this assumption, the conventional analysis method can be easily extended. However, the plastic instability of multilayered cylindrical pressure vessel obtained by the proposed bifurcation analysis showed a significant difference from that by the conventional method.

View full abstract

Up to now, behaviors of link chains in chain hoist subjected to impact loads have not been clarified yet. In this paper, an elastoplastic analysis is carried out using the finite-element method (DYNA3D) in order to clarify the stress distribution and to predict the tensile strength of the link chains subjected to impact loads. The tensile strength is evaluated by von Mises' criterion. The effects of the dimensions of the link chains on the stress distributions and the tensile strength are shown. For verification, experiments were performed on strains and tensile strength of link chains when impact loads are applied to the chains. The analytical results are in fairly good agreement with the experimental results. As a result, it is shown that the strength can be evaluated by F. E. M. analysis. In addition, it is clarified that the impact tensile strength of link chains can be improved by optimizing the pitch, the width, the shape of the cross section and the compressed depth.

View full abstract

Many automatic mesh generation methods for the finite-element method (FEM) have been reported. However, for the case with complicated heat generation, a large amount of data depending on the position must be added to mesh data. The another examples, which need a large amount of data depending on the position, are functionally gradient material and biomechanics. In these cases, it is difficult to prepare the distributed data. This paper shows that these problems can be solved by using an improved multiple-reciprocity boundary element method. In this method, contour lines of distribution are used and these distributions are assumed to approximately satisfy the Poisson equation.

View full abstract

The discretization of a whole domain is a prerequisite for FEM. Since manual mesh generation is a tedious process, the mesh generation scheme has been studied in recent years. In iterative analysis, the geometry of the domain is drastically changed during iterations. It is known that mesh regeneration eliminates mesh distortion caused by geometry change, which exerts unfavorable effects on the convergence of the whole analysis. If error control of the analysis is additionally considered, such a process should be adaptive ; adaptive remeshing is required for a reliable solution. There exist numerous mesh generators on the market ; however, most of them are based on an interactive procedure with a user, thus they cannot be used for the adaptive remeshing process. In this paper, a sophisticated mesh generation scheme for the adaptive remeshing process in 2D FEM is discussed in detail. After comparison with other schemes, it can be concluded that our scheme is superior to the others in terms of flexibility and time complexity.

View full abstract

Using the constitutive equation for cyclic plasticity, the effects of thickness and fixation of ultrahigh-molecular-weight polyethylene (UHMWPE) plate and of traction caused by the contact of the femoral and tibial components on wear behavior are evaluated with the aim of improving the artificial knee joint. As a result, it is clarified that the thickness and the fixation of the UHMWPE plate affect the distribution of stress and strain in UHMWPE. To prevent the wear of UHMWPE and to extend the life of the artificial knee joint, the thickness of the UHMWPE plate should be at least 6 mm and the plate should be completely fixed all over the attachment surface. The coefficient of friction also affects the deformation of UHMWPE, especially beneath the surface, and the wear of UHMWPE progresses rapidly when the coefficient of friction increases, as expected.

View full abstract

In this paper we consider optimal distribution of a gradient material at the circular cross section of a cantilever. The design variables are the parameters which express the distribution of the material density or the elastic constant, and Young's modulus is assumed to be linear in the material density. Two problems are analyzed : (1) maximization of the bending stiffness with constant weight, and (2) minimization of the maximum stress at the cross section with constant bending stiffness. In a general case, the above problems reduce to generalized eigenvalue problems which are formulated and solved by symbolic manipulation and numerical calculation. The optimal distribution indicates that for (1), the density is concentrated at the circular surface such as of a circular cylinder, and for (2), Young's modulus decreases with respect to the radius. The obtained results are also applicable to the optimal design of composite materials where the distribution of the material density is not continuous.

View full abstract

In fatigue strength design, there are many influence factors and estimating methods, and they influence each other. For the development of an expert system, the management of knowledge is necessary. In the present paper, a new procedure for knowledge management is proposed, and is applied to management for estimating fatigue strengthes of notched bars.

View full abstract

The development of the shape optimization design technique of elastoplastic bodies can realize severe structural design such as limited material design, and is useful for the preliminary study of elastoplastic impact design. In this study, an exact and direct sensitivity analysis technique for the elastoplastic material governed by the bilinear constitutive law is developed, based on the implicit differentiation method. The sensitivity analysis technique suggested here is applied to determine optimum shapes of minimum weight under the elastoplastic stress constraints.

View full abstract

In our previous study, as a result of the air-blast erosion tests at room temperature, the wear of materials by impact of powder could be estimated from the impinging conditions and the properties of materials and particles. In this study, some alloy steels were tested using the high-temperature erosion test equipment, and the effects of the wear factors on the wear rates were investigated. At a relatively low temperature (300°C), using the predicting method reported in our previous study, the wear rates calculated from the hardness of the materials and the particles at room temperature were close to the experimental results. At a higher temperature (700°C), the wear rates of the materials were increased because the materials were more ductile and softer. The wear rates calculated from the hardness at high temperature, using the same predicting method, were relatively close to the experimental results when the hardness of the materials at high temperature, data which are not obtained easily, were estimated on the assumption that the hardness is proportional to the tensile strength.

View full abstract