In this paper, an effective and convenient method for estimating low cycle fatigue life based on the smail crack growth law was proposed, and the factors controlling fatigue life were clarified. Also, plastic-strain-controlled low cycle fatigue tests were carried out on pearlitic spheroidal graphite cast lron (PDI) and its matrix material (0.85% carbon steel ; JIS SK5). The effect of detects on fatigue life and the factors controlling fatigue life were examined from a microscopic point of view. The main results obtained are as follows : (1) Low cycle fatigue life is controlled by three factors : crack initiation length, crack propagation limit length and relative crack growth rate. (2) Fracture of PDI starts from a microshrinkage that exists on the neighboring surface in the case of a cylindrical specimen ; thus, the crack initiation length of PDI is longer than that of SK5. (3) Cracks caused by defects shorten the crack propagation limit length. (4) The relative crack growth rate is accelerated by the increase in crack opening displacement range at a crack tip.

抄録全体を表示

Low-cycle fatigue tests were performed at -140°C using commercial-grade pure iron having an average ferrite grain size of 400 μm. The effect of fracture mechanism on fatigue life properties was investigated, especially on the dispersion of fatigue lives in the high plastic strain range (Δεp) where fracture occurs in an internal fracture mode. Fracture surfaces were observed using a scanning electron microscope. The dispersion of fatigue lives is due mainly to the difference of internal fracting mode. Specimens fractured by intersection of a deformation twin and a grain boundary (T-B type) show longer fatigue lives than those fractured by intersection of deformation twins (T-T type) or a deformation twin and an inclusion (T-I type). The effect of grain size at the internal fracture origination site was also discussed.

抄録全体を表示

It is know that the fretting fatigue limit is affected by various factors such as contact pressure, relative slip, contact length, specimen size and so on. In this study, the reason why these factors affect the fretting fatigue was investigated based on the stress distribution near the contact edge. A measuring technique of local stress distribution near the contact edge was developed using a small stress concentration gage. It was shown that stress concentration occurred at the contact edge by the fretting process depending on these factors. It was found that the peak stress amplitude at the contact edge was the unified factor which governs the fretting fatigue limit.

抄録全体を表示

Faligue creep transition in Hastelloy-X at elevated temperatures under tensile-tensile cyclic leading was studied by in-situ observation employing a SEM servo. The main results were as follows (1) Fatigue limit was not found at any temperature, since oxide-induced crack closure did not occur in vacuum. (2) In-situ observation showed that surface crack length increased at 700°C and crack opening displacement increased at 800°C. Thus, fatigue crack initiated and propagated at 700°C, while, creep deformation accumulated at 800°C. (3) The fatigue life of specimens tested at 800°C ere decreased under cyclic loading frequency from 30 Hz to 0.3 Hz. under these conditions, the fatigue at 800°C was dominated by time dependent creep rupture. (4) Fatigue-creep transition occurred from 700°C to 800°C. The mechanism of fracture at 700°C was fatigue and that at 800°C was creep deformation

抄録全体を表示

A plane bending fatigue test of Al-Si sintered alloy was carried out employing a compact testing machine. Crack initiation and growth behavior were investigated in comparison with those of casting alloy from the mesoscopic point of view. The main results were as follows. (1) The fatigue lint of sintered alloy could not be found. The fatigue strength of sintered alloy was higher that that of casting alloy. (2) Fatigue fracture of sintered alloy were controlled by brittleness. Crack initiation life accounted for the greater part of fatigue life. (3) Crack propagation characteristics of sintered alloy were determined controlled by ΔK. Crack propagation rate was higher than that of casting alloy since the fracture toughness of Al-Si sintered alloy is higher than that of HIP-treated casting alloy. (4) the crack propagation rate of Al-Si sintered alloy was very high and fracture mode rapidly changed into monotonic brittle facture. It is suggested that in engineering design emplcying Al-Si sintered alloy, consideration of brittle fracture is important.

抄録全体を表示

The fatigue reliability of recycled advanced fiber-reinforced thermoplastic resin (FRTP) was examined, with particular attention to the effects of mesoscopic factors on the fatigue crack propagation. The main results obtained were as follows. (1) The recycled PPS materials were inferior in fatigue reliability to recycled PA·66 materials. The'm'values of Paris's law for recycled PPS materials were considerably higher than that of PA·66. FCP properties of recycled PPS materials were unstable. (2) From fractography using SEM, it was found in recycled PPS materials that the reinforced fibers were separated from the brittle resin matrix. Therefore it was understood that the effect of the ductility of resin matrix on the fatigue crack propagation for recycled FRTP was considerable.

抄録全体を表示

The estimation of fatigue reliability at elevated temperatured for advanced carbon fiber reinforced PA46 was examined. Especially meso crack forming stage developed on notch bottom and its effect on fatigue reliability at elevated temperature. The main results are as follows, (1) The fatigue strenght, life of crack initiation and life of crack propagation for V-notch specimens were almost the same at elevated temperature. (2) From fractography using SEM, it was recognized that the matrix was elongated and the cementing of the interface between fiber and matrix was poor. So it was thought that the crack propagation resistance at elevated temperature was not influence by the kinds of fiber. (3) It was considered that the difference of crack tip heat change accompanying fatigue damage and crack propagation was controlled under the linear fracture mechanics.

抄録全体を表示

In order to investigate cyclic deformation and fatigue of a TiNi shape-memory alloy wire, rotating-bending fatigue tests were carried out at various strain amplitudes, temperatures and rotational speeds in air and in water. The results were summarized as follows. (1) If strain amplitude was found in the rhombohedral-phase transformation region, the fatigue life was longer than 10⁷ cycles. The deformation properties did not change under cyclic deformation. (2) If strain amplitude was found in the martensitic transformation region, the fatigue life was shorter than 10⁵ cycles. The higher the temperature, the shorter the fatigue life. (3) The rotational speed did not affect the fatigue life in water. The higher the rotational speed, the shorter the fatigue life in air.

抄録全体を表示

Creep-fatlgue tests were conducted using 316FR stainless steel at 650°C in air under a variety of strain time programs including symmetric continuous, unsymmetric continuous and strain hold cycles. By comparison of the results with test data obtained in vacuum, it was found that the life reduction which occurred for a slow tension-fast compression strain wave and the tensile strain holding wave was smaller in air than in vacuum, and that the symmetric continuous cycle gave the largest life reduction from the data obtained in vacuum. Following the creep-fatigue tests, fracture surfaces were investigated by SEM. When time/rate-dependent life reduction occurred, the fracture mode in vacuum was intergranular, but in air, striations were observed and integranular fracture was rarely observed.

抄録全体を表示

Futiguc limit and crack propagation behavior of TRIP aided dual-phase (TDP) steel associated with the TRIP of retained austenite were investigated under the stress ration R of 0.1 and at three testing temperatures 20, 100 and 200°C. For comparison, a conventional ferrite-marlensite dual phese (DP) sted was also tested. The fatigue limit of TDP steel was considerably higher than that of DP steel. Furthermore, the fatigue limit decreased with increasing testing temperature, differing from the case of DP steel. These results were ascribed to the solution strengthening of the ferrite matrix due to the large amount of added silicon and the "block effect"against crack initiation due to the strain-induced transformation of the retained austenite. In a low ΔK region the TDP steel crack propagation rate da/dN was higher than that of DP steel, and was accompanied by a smaller threshold value ΔK_m. On the other hand, in a high ΔK region, da/dN was somewhat lower than that of DP stecl. Such crack propagation behavior was mainly explained on the basis of the crack closure effect induced by fracture surface roughness and fretting oxide films.

抄録全体を表示

Molecular dynamics is applicable for only a small region of simulation. To simulate a large region. it is necessary to combine molecular dynamics with continuum mechanics. Therefore, we propose a new model in which molecular dynamics is combined with micromechanics. We apply a molecular dynanics model to the crack tip region and apply a micromechanics model to the surrounding region. Problems exist at the boundary. In this study, we solve the boundary problems and devolop a method for combining molecular dynamics with micromechanics. We make possible the simulation of the process of crack propagation and dislocation increase. In order to examine the vaildity of this model, we use α-iron in simulation. If the proposed model is valid, stress and displacement must vary continuously across the boundary between the molecular dynamics region and the micromechanics region. Our results show that the two parameters vary continuously across the boundary.

抄録全体を表示

In one of the fields of fracture mechanics, the determination of the stress intensity factors of a crack is of great importance, when a cracked body is affected under a heating load. The results heated from the same side of the crack were reported in our previous paper. In this paper, the additional results are reported on the stress intensity factors of specimens heated abruptly at a constant temperature from the opposite side of the crack. Both caustic and photoelastic methods were employed in our experiments, and the obtained values of the stress intensity factors were compared with those obtained in our previous experiments and those reported by others in the similar experiments.

抄録全体を表示

This paper is concerned with the mixed boundary value problems of an infinite plate with a square hole into which a smooth rigid square plug with rounded corners is simply inserted. Since the infinite plate is subjected to a uniaxial tension at infinity, gaps are produced along the boundary between the inserted plug and the hole. The contact pressure between the inserted plug and the hole is expressed in a convergent series, so that the stress and displacement along the boundary can be numerically analyzed by the point matching method. Using the numerical results for the shape of a square hole with rounded corners, the influence of the interference is shown with the stress around a square hole.

抄録全体を表示

This peper deals with a theoretical treatment of elastic behavior for a medium with Kassir's nonhomogeneous material property, which is defined by the relation [numerical formula], i.e., shear modulus of elasticity G varies with the dimensionless axial coordinate z by the power product form, arbitracily. The fundamental differential equation for such nonhomogeneous medium was previously proposed by Kassir, and is given by a second-order partial differential equation. However, it is found that this basic equation is not sufficient in general for solving boundary value problems. Making use of the fundamental equation system for such a nonhomogeneous medium, which was previously proppsed by us, axisymmetric problems for a semi-infinite body subjected to an arbitrarily shaped distributed load and a concentrated load on its boundary surface are developed theoretically. Numerical calculations are carried out for several cases taking into account the variation of a nonhomogeneous parameter m. and the numerical results for displacements and stress components are shown in graphical form.

抄録全体を表示

Functionally graded materials (FGMs) have been developed for case as ultrahigh-heat-resistant materials in aircraft, space and nuclear engineering fields. The composition of FGMs varies, but most FGM are particle-reinforced FGMs. In the study of FGMs, the material properties due to macroscopic combine-low have been adopted. However in order to obtain more correct thermal stresses in the FGMs, it is necessary to analyze thermal stresses by use of the microscopic combinelow. Using the constitutive equations for the particle-reinforced composites, the stress-strain relations of particle-reinforced composites taking into account the effect of the manufacturing process, and the stress distribution in an FGM plate subjected to heating are discussed. The number of damaged particles and the possibility of crack generation taking into account the manufacturing process are also discussed.

抄録全体を表示

This paper deals with the strength characteristics of bolted joints subjected to torsional loads. The joints are composed of two hollow cylinders which are clamped by a bolt and a nut or a threaded tap bolt. The stress distributions at the contact surfaces and the variation in torsional loads occuring in the bolts are analyzed using an axisymmetrical theory of elasticity. A method for cstimating the joint strength for slipping is proposed. Analytical results concerning the variation of the torsional loads in the bolt and the joint strength are in fairly good agreement with the experimental ones. It is found that the joint strength for slipping increases as the initial clamping force increases. It is also found that the strength of joint using a bolt and a nut is greater than that using a threaded tap bolt. The joint strength when a torsional load is applied in the same direction as the screw fastening direction is greater than that in the inverse direction.

抄録全体を表示

In the present paper, analytical treatments of dynamic behavior of beam mode of bimodulus materials under moving load are shown by using the transfer matrix method to determine the natural frequencies and mode shapes of the beam and application of the mode superposition technique. We adopt analytical theories such as classical, Timoshenko's and Levinson's for the computations of dynamle behavlor of a simple beam fixed at both ends, in which the latter two theories take into consideration the shear deformation and rotatory inertia of the beam. The accuracy, dynamic responses and dynamic magnification factors for the beam of bimodulus materials are illustrated by several numerical results obtained using the three theories.

抄録全体を表示

Honeycomb structure is used as the shock absorbent to protect precision machines during transport. The purpose of this study is to establish the relation between the energy absorption ability and the honeycomb parameters such as cell size, cell wall thickness and material properties. In this work, as the first step of study, the authors carried out experiments on the collapse of honeycomb structure under quasi-static condition, and clarified the energy absorbing mechanism which is due to progressive buckling. Based on the observed buckling mode, the authors established a finite element model which can represent the debonding failure between the bonded cell walls. The analysis results calculated using this model agreed well with the experimental results for mean collapsing stress and buckling mode.

抄録全体を表示

The validity of the concept of linear notch mechanics for evaluating the damage near the notch root of notched FRP plates is investigated experimentally. An experimental method is presented for examining the effect of notch geometry on the damage near the notch root of FRP plates in tension and bending. Two kinds of FRP plate were used : a woven roving glass fabric/epoxy laminate and a short-glass-fiber reinforced polycarbonate. To evaluate the damage near the notch root, we measured the luminance distributions by means of the luminance measuring technique using a CCD camera. The results show that the growth of the damaged zone near the notch root was governed predominantly by both the notch-root radius and the maximum elastic stress at the notch root, while it was independent of notch depth and type of loading. Using linear notch mechanics, the experimental results can be clearly explained.

抄録全体を表示

This paper concerns the dynamic response of partially delaminated composite laminates impacted by a rigid ball. Nine-node isoparametric elements including the influence of shear deformation were used in the finite element analysis. In order to evaluate the contact force between the impactor and the laminate, the Hertzian contact law was adopted. The Newmark algorithm was employed to solve the motion of the impactor. To deal with the dynamic contact problem between two delamlnated layers effectively, a Lagrange multiplier technique incorporated with a special time integration algorithm was presented to overcome the deficiency of the classical Lagrange multiplier method. The numerical results give much information of impact phenomena of delaminated composite laminates.

抄録全体を表示

The purpose of this study is to contribute to the development and the safety design of plasma facing components for fusion reactor devices. We evaluated the thermal shock resistance and the thermal shock fracture toughness of a bonding material which was jointed a carbon-fiber-reinforced carbon composite (C/C composite) to oxygen-free copper. We also examined the microstructures of the bonding layers using a scanning electron microscope before and after thermal shock tests. The bonding material did not fracture during thermal shock tests. However, thermal cracks and delamination cracks were observed in the bonding layers.

抄録全体を表示

The present paper deals with a deformation control of a laminated composite plate with piezoelectric actuators under static loading. Deflection of the plate is controlled using actuators bonded symmetrically to the plate surfaces. Static response of the plate with distributed actuators is calculated by a two-dimensional finite element analysis incorporating the effect of actutors. In order to minimize the deflection of the plate at the specified locations, the optimal voltage applied to the actuators is obtained by the least squares method. The effect of the geometry and the number of actuators on the deformation control is discussed through several numerical examples for cantilevered laminated plates.

抄録全体を表示

In this paper, a woven structure consisting of fiber bundles is defined as a mesostructure and we evaluate the strength of woven fabric composite materials mesoscopically by the homogenization method and anisotropic damage mechanics. A numerical simulation system for virtual strength test has been developed to predict the initial mesoscopic fracture, the fractured location in both macro and mesostructure and the fracture mode. The mesoscopic fracture mode is shown schematically, which greatly aids in the understanding of the mesomechanics of woven fabric composite materials. This system is effective for material and structure design and proves the necessity of considering the interaction between macro-and mesostructures. Furthermore, the formulation and a numerical example for mesoscopic fracture propagation considering the fracture mode are shown.

抄録全体を表示

Micro-Vickers hardness measurement is used to establish a method for determining the composition in a functionally graded material (FGM). Materials used in the present study are Al-SiC, Al Shirasu and Al-Al₃Ni systems, which are manufacured by a centrifugal method. The hardness of aluminum matrix markedly increases with an increase of volume fraction of particles in the case of Al-SiC FGM. However the tendency is not soclear for Al-Shirasu and Al-Al₃Ni FGMs. Plastic strain energy due to thermal misfit between matrix and distributed particles is calculated based on the elastoplastic analysis by applying a shell theory after Lee et. al. The hardness is correlated theoretically with composition gradient through the mean plastic strain energy that generates higher disiocation density. It is concluded that the evaluation of the composition gradient of metal-based FGM can be achieved from hardness measurement.

抄録全体を表示

Numerical stress analysis of resin cracking occurring in LSI packages during temperature cycling tests was carried out using a thermoelastic finite element method for nonlinear contact problems. Through the use of the linear fracture mechanics approach, the analysis provides a fracture parameter and indicates the direction in which resin cracking occurs. Temperature cycling tests using actual LSI packages were also performed to verify the validity of the analyses. It was found that the test results agreed well with the analytical results and that leadframe materials and the delamination occurring between dissimilar materials determine quantitatively the tendency of resin cracking. The resin cracking mechanism was also studied and countermeasures against it were proposed.

抄録全体を表示

Benchmark tests of the strength of adhesive joints were performed by twelve members (institutes), tocusing on the difference in strength between several different types of joints. A series of experiments on four types of joints on (1) butt joints, (2) single-lap joints, (3) double-lap joints and (4) double cantilever beams (DCB). was carried out using two types of epoxy adhesives (adhesive A, ductile ; adhesive B, brittle) and steel adherends. The adhesive joint strengths are strongly affected by the mechanical properties of adhesives and adherends. The adhesive strengths of the butt and double lap joints of adhesive B are found to be higher than those of adhesive A, whereas the single-lap joint and DCB of adhesive A are stronger that those of adhesive B. The difference in adhesive strength between these joints can be explained by considering two different types of fracture criteria : (1) the fracture criterion by crack initiation and (2) the fracture criterlon by crack propagation. The results obtained suggest that we should make an appropriate choice of adhesive test to evaluate the strength of actual adhesive joints.

抄録全体を表示

The purpose of this study is to evaluate the effects of burn-off at 500°C by air oxidation on strengths and fracture toughnesses of a carbon-fiber-reinforced carbon composite and three finegrained isotropic graphite materials that are used as main materials of plasma facing components for fusion reactor devices and core structural materials of gas-cooled nuclear reactors. The Young's modul, the bending strengths, the Rockwell hardnesses, the mode I fracture toughnesses and the load and depth by a manufactured hardness tester were measured and the microstructures of the fracture surfaces were examined. The degradations due to burn-off of these properties of CX-2002 U composite were the smallest and those of IG-430 U graphite were less marked than those of other graphite materials. After oxidation of the C/C composite, preferential removal of the boundary layer between carbon fiber and pyrolytic carbon matrix and carbon fiber were observed. After oxidation of graphite materials, the size and the number of pores were increased and the fracture surfaces were rough due to oxidation of graphite grain boundaries.

抄録全体を表示

In order to investigate the effects of temperature and atmosphere on the strength of C/C composites, the elevated-temperature strengths of C/C composites were measured in vacuum and in alr. The strongth in vacuum increases with increasing temperature. The exposure of C/C composites to alr at elevated temperatures leads to a decrease of the room temperature strength due to oxidation. The residual strength after elevated temperature oxidation shows good correlation with weight loss or decrease of rigidity.

抄録全体を表示

In this paper, we propose a new algorithm for parallel computing of combination of BEM and FEM. Each analysis is performed independently by introducing domain decomposition of an entire domain into the boundary element and finite element subdomains. An iterative scheme for parallel computing is the Schwarz method which was previously adopted to the domain decomposition parallel scheme for the BEM analysis. Cluster parallel workstations are employed for efficient analysis We examine the convergence and the accuracy of solutions and show some numerical examples.

抄録全体を表示

Based on the assumption that a solid or a liquid which consists of a vast number of atoms can be regarded as a collection of atom clusters which themselves can be regarded as particles. a renormallzation procedure has been applied to the cluster system in oder to produce a new reduced system having the same scale as the atomic one. This new system consists of particles of atomic size but having the same mass as the original clusters. Then it is shown that molecular dynamics simulation carried out on this system using the same interatomic potential as that of the original atcmlc system reproduces results corresponding to those of the original huge atomic system with regard to phenomena of a scale larger than that of clusters. We then discuss the damping effect which should be introduced in the cluster system. An example of a numerical algorithm for the variable scale molecular dynamics developed based on the above discussion is also presented.

抄録全体を表示

We establish a general theory of maximum stiffness design of elastic beams subject to a given volume. Determination of the maximum stiffness of a beam, with the height taken as an independent variable can be treated as the Bolza problem. The beam is taken to be symmetrical about its center line. The theory thus obtained can easily be extended to take account of variable width or section area of the beam, by defining a control vector u. An illustrative application is presented in cases of cantilever and simple beam designs subjected to a restriction of the lower bound under an arbitrary external load.

抄録全体を表示

Complex surface defects simulated on a metal specimen are detected using the photoacoustic (PA) method and signals corresponding to defects are reconstructed to form PA images. The depths of two defects and the distance between them range from 0.5 to 1.5 mm and 0.5 to 2.0 mm, respectively. Both measurements were in increments of 0.5 mm. The PA image obtained clearly shows the locations of the edges of two defects. It is found that complex defects composed of two sections are detected separately as individual single defects.

抄録全体を表示