From a mesoscopic viewpoint, the effects of SiC particles on fatigue crack initiation and growth behavior in an A6061-T6 metal matrix composite (MMC) were investigated with a topographical approach employing a laser-type profile measurement microscope. The results are summarized as follows. Fatigue crack of MMC grew and stopped sporadically and crack propagation rate decreased with a decrease in the number of particles. Fatigue crack of MMC originated from the aluminum matrix and the crack growth was a shear mode. Thus, the fractured surface showed crystallographic facets. The number of facets on the fractured surface decreased with increasing number of particles, since SiC particles in the aluminum matrix were resistant to shear mode crack growth. Crack growth changed from shear mode to brittle-like monotonic fracture mode when SiC particles cracked.

抄録全体を表示

Since Bourdon tubes are small in size and complicated in geometry, an ordinary fatigue design procedure is difficult to be applied, in which the design fatigue curve of the component material and the stress analysis of the component based on the operation mode are necessary. Therefore, the entire coil configuration was used to evaluate the integrity against fatigue of the Bourdon tube. The fatigue resistance property of the Bourdon tube was directly obtained by applying cyclic internal pressure (P_a), which simulates the internal pressure fluctuation during the normal operation of the Bourdon tubes in BWR power plants. As a consequence, a reasonable fatigue design margin was demonstrated for the proposed exchange period of 4 years, of the Bourdon tubes in BWR power plants by extrapolating P_a-N_f (fatigue life) curve according to the Modified Miner's and Haibach's methods, and by the evaluation procedures recommended in the ASME Code, Sec.III, Appendix II.

抄録全体を表示

A rotating-bending fatigue testing machine was newly developed for the evaluation of the fatigue strength of a shape memory alloy wire. The rotating-bending fatigue tests were successfully performed for the TiNi shape memory alloy wire at various temperatures in water. The results were summarized as follows. If the maximum strain amplitude was in the regions of the rhombohedral-phase (R-phase) transformation, the fatigue life was longer than 10⁷ cycles. If the maximum strain amplitude was in the region of the R-phase transformation completion and the martensitic transformation, the fatigue life became shorter as the strain amplitude increased. If the fatigue tests were performed in air, the temperature of the material rose and the transformation stress increased under cyclic deformation, yielding short fatigue life. The higher the ambient temperature, the shorter the fatigue life of the wire. The oxide film formed on the surface of the Wire did not affect the low cycle fatigue.

抄録全体を表示

Structural components operating at high temperatures in thermal and nuclear power plants are subjected to interaction of thermal fatigue and creep which results in creep-fatigue damage. In evaluating the life of those components, it is important to comprehend detailed microscopic damage evolution under creep-fatigue condition. We had observed continuously the growth and coalescence of cavity and microcrack on specimen surface during the test. In this study, creep and creep-fatigue tests with tensile hold-time were conducted on SUS304 stainless steel, with large grain size producted by a particular heat treatment, by using a high-temperature fatigue machine combined with a scanning electron microscope. Microstrain distribution in the grains on the specimen surface during the test was measured by analyzing the distortion of microgrids. The grid pattern with a pitch of about 10 μm was induced by a diamond stylus on the specimen. As result, it was found that the grain deformed uniformly in a direction of stress axis under creep condition. The deformation progressed with inhomogeneous microstrain distribution under creep-fatigue condition. Initiation of microcrack under creep-fatigue condition was mainly related to strain concentration in the grain near grain boundaries. The microcrack under creep condition propagated after reaching a critical crack opening displacement. The crack opening displacement under creep-fatigue condition was smaller than that under the creep condition.

抄録全体を表示

The contour J-intergral values were calculated using the three/dimensional elastic-plastic finite element method for semielliptical surface cracks originating at notch roots under tensile loading. Then these were compared with the J-integral values calculated using the conventional equation that we proposed in earlier papers, which we refer to as J hereafter. The difference between the two was quite small at the deepest point A in the semielliptical surface crack, but was fairly large at the edge of a crack on the specimen surface, C. The convensional equation was modified by considering stress/strain gradients developed in the cross section at the notch root. The J-integral value, J'calculated using the proposed equation was in quite good agreement with the contour J-integral value at both points, A and C in a semielliptical surface crack. The ratio J/J'decreased with increase in crack depth, depending upon the aspect ratio of the surface crack, magnitude of the applied stress and notch root radius. The parameter R_w which was determined from F. E. M. calculation was proposed so that J/J'could be expressed by a unique curve as a functien of R_w over quite wide ranges of notch root radius, surface-crack shape, and magnitude of stress. The limitation for applying the simple J-integral estimation proposed earlier was determined using this curve. The crack growth rates obtained from fatigue test of notched specimen were successfully plotted against J-integral range calculated using the modified conventional equation.

抄録全体を表示

Two-dimensional finite element analysis has been carried out on compact tension and shear specimens under mixed-mode loading. The complex stress-intensity factor K associated with an elastic interface crack is discussed via three different approaches : virtual crack extension method, modified crack-closure integral and displacement extrapolation. A comparison of stress-intensity factors K_I and K_II obtained by these three methods is carried out. The accuracy of these methods is discussed for specimens comprised of various dissimilar materials under various kinds of mixed-mode loading. The accuracy of the virtual crack extension method was revealed by comparison with the J-integral. The modified crack-closure integral varies markedly depending on the mesh length at the crack tip. However, the stress-intensity factor can be estimated with high accuracy using an appropriate mesh length.

抄録全体を表示

A finite, element method for calculating the effect of lubricating of trapped in cracks was presented. The specimen with a crack was divided by the elastic elements and the lubricating oil was divided by the viscous elements. The behavior of the lubricating oil under compressive stress was calculated The lubricating oil is trapped in the crack and moves to the crack tip. The crack opens near the tip and hence the stress intensity factor increases. These results agree with the results calculated by the method presented in the former paper. In the case of a three-dimensional through crack, the lubricating oil is not trapped completely and flows from the sides of the specimen.

抄録全体を表示

Moire interferometry was utilized to study fracture behavior of polymer specimens under mode I loading. Displacement fields around a crack tip were experimentally determined for two kind specimens of brittle and ductile polymers. The results were compared with the displacements given by linear fracture mechanics or HRR fields to study their applicability to the polymers. Examinations were also made on the effect of material properties and the influence of load applied to the specimens.

抄録全体を表示

The fracture behavior and interfacial properties of single fiber composites (SFC) containing a single optical fiber are examined by acoustic emission and a polarized microscope in a monotonous tensile and bending test. As a result, in both tests, initial and main AE events occur in an earlier loading region, and near the peak loading point, respectively. Fiber fragment length can be expressed by Weibull distribution. However, the differences in these two tests are clearly seen in AE activity patterns, amplitude distributions, power spectra, and fiber fragment distributions. the shape parameter of a Weibull distribution in the bending test is 5.7, which is about twice as large as that in the tensile test. This means a smaller dispersion of fiber fragment length in the bending test than in the tensile one. A transition of failure modes due to tensile, compressive, or shear stresses or a combination of these is found by changing the fiber position along the thickness direction of a specimen in the bending test. Variance of AE characteristics and fracture behavior at a fiber breakage can also be observed for different fiber positions.

抄録全体を表示

Polycrystalline metals usually have internal structures, thus there are cases where it is difficult to apply Fick's laws to diffusion. In the present paper, the authors introduce mesodomains, which are regarded as the grains in metals, to the conventional diffusion theory. State variables are varied in mesodomains and are divided into the average value and fluctuating value. Consequently, the mass balance is expressed by not only the mass concentration but also the mesoscopic average mass concentration considering nonlocality. A new entropy flux is introduced to entropy inequality and a new mesoscopic variable is added to helmholtz free energy. Furthermore, the constitutive equations of Cauchy's stress and of diffusion flux are influenced by mesoscopic average mass concentration which is one of the behaviors of materials in grains.

抄録全体を表示

The purpose of the present study is to develop a procedure for calculating mechanical Material constants of simple crystalline metal by the molecular dynamic method (MD). Suitable parameters for total number of particles, incremental time and numbers in numerical integration of motion equations were investigated in MD simulations. Simulations for NVT ensemble MD in which a model simple crystal of α-Fe was stretched in simple or triaxial stress were carried out using Johnson or Morse potential. Mechanical material constants, that is, Young's modulus and Poisson's ratio for α-Fe, were calculated using the results obtained by the MD and statistical thermodynamics. We proposed a modified Johnson potential such that mechanical material constants calculated by the MD coincide with experimentally determined ones.

抄録全体を表示

The mechanical properties of low-carbon martensitic steels were studied with the aim of their application to structural materials. The seven kinds of steels containing different carbon contents up to 0.3 mass percent were examined. The carbon content dependences of hardness, tensile properties and impact toughness in as-quenched martensite are shown. The low-carbon martensite is plastically deformable, i. e., more than 60% cold rolling is easily achieved in steels containing less than 0.2 mass percent carbon and the strength is explained reasonably from the viewpoint of dislocation density. The 0.1 mass percent carbon as-quenched martensite is revealed to show comparable impact value to those of quenched and tempered steels at strength level of more than 1 GPa. Several points relevant to practical use, including, residual stress and strain rate effect, are discussed.

抄録全体を表示

Uniaxial tension tests under various environmental temperatures are carried out for 5 kinds of polymers : high-density polyethylene (HDPE), poly (vinylidene fluoride)(PVdF), polyoxymethylene (POM), polyamideimide (PAI) and polyether-etherketone (PEEK). As temperature increases, yield stress decreases and yield strain except for the case of PAI increases in all of the matelials tested. HDPE, PVdF and POM are strongly influenced by environmental temperature, but PAI and PEEK are weakly influenced. The authors previously proposed a spherical model based on intermolecular force (Lennard-Jones type) for understanding the influence of hydrostatic pressure, environmental temperature and strain rate on elastic modulus. In this paper, this model is extended to explain the yield behavior and to propose an experimental equation of yield stress and strain. We found that this equation is useful for describing yield stress under various environmental temperatures.

抄録全体を表示

Basic properties of creep and stress relaxation in a film of shape memory polymers of polyurethane series above and below the glass transition temperature T_g were investigated experimentally. The results are summarized as follows. Modulus of elasticity, yield stress and 100% modulus are large below T_g but small above T_g. The relationships between normalized stress at each temperature and creep strain almost coincide without depending on temperature. Creep strain above T_g recovers after unloading but creep strain below T_g does not recover. Irrecoverable creep strain which remains below T_g recovers at temperatures in the vicinity of T_g during the heating process. A half of initial stress decreases under constant strain after a certain duration of time for each temperature and strain. The ratio of relaxed stress to initial stress increases with an increase in strain below T_g. The deformation properties of the materials with different T_g are almost same except for difference in T_g.

抄録全体を表示

Very few studies on an anisotropic elliptical or circular disk have been carried out because of difficulties caused by the anisotropy. The present paper shows the analytical solutions and gives the distributions not only of the stress but also of the displacements obtained by using Lekhnitskii's complex variable method. The disk is assumed to be affected by the arbitrarily distributed loads such as concentrated force and hydraulic pressure. Some numerical results are shown by graphical representation.

抄録全体を表示

The general condition for the logarithmic stress singularities obtained by the author in a previous study is applied to a semi-infinite plate consisting of two edge-bonded wedges. When the eigenvalue is a multiple root, logarithmic stress singularities the may occur. The results of analysis show that the logarithmic singularities in the stress field may be of order (log γ)², log γ or γ^λ-1 log γ, depending on the combination of materials and the vertex angle of the wedges.

抄録全体を表示

A general solution for an elastic medium with an arbitrary-shaped elastic inclusion is shown for the case of out-of-plane loading at infinity. The analysis is based on the complex variable method using a conformal mapping technique. Using the exact elastic solution, the stress and displacement at the boundary of the elastic inclusion with the cusp are discussed. The stress concentration is given at the cusp for various ratios of Young's moduli of the inclusion to that of the elastic medium. Several numerical examples are shown by graphical representation for various arbitrary-shaped elastic inclusion. Our numerical results agreed with those of other investigators for these cases.

抄録全体を表示

Elastoplastic constitutive equations which take into account yield vertex effects are important in the study of localization instabilities of plastic deformations. However, they have never been discussed thermomechanically. In the present paper, a method of deriving the above equations is proposed which is based on the second law of thermodynamics and the principle of maximal entropy production rate. Elastic strain as a strain measure which is conjugate to objective stress rate is separated from total strain so that the Clausius-Duhem inequality, in which the Gibbs function is introduced as an elastic potential, can be always satisfied. The strain rate and stress rate are expressed by the same objective rate in the rate form of the elastic constitutive equation obtained. The plastic constitutive equation is derived using the principle of maximal dissipation rate. Since this equation is regarded as a flow rule in which the complementary dissipation function assumes the role of a plastic potential, it is indicated that the yield vertex can exist on the dissipation surface. Furthermore, the spin used in the objective stress rate is selected by taking into account not only conventional requirements but also thermomechanical ones.

抄録全体を表示

In this study, the one-dimensional transient thermal stress problem of a nonhomogeneous hollow circular cylinder with arbitrarily distributed and continuously varied material properties, such as functionally graded materials, is evaluated theoretically. Using the analytical procedure of a laminated hollow circular cylinder model, the analytical temperature solution for the cylinder is derived approximately. Furthermore, making use of Airy's stress function method, the thermal stress components are formulated under the mechanical condition of being traction-free. As a numerical example, a hollow circular cylinder composed of zirconium oxide (ZrO₂) and titanium alloy (Ti-6Al-4V) is considered. To optimize (i. e., minimize) the thermal stress distribution, numerical calculations are carried out, and the optimum material composition is determined taking into account the effect of the temperature dependence of the material properties. Furthermore, taking into account The variation in the thickness of the cylinder, the temperature rise of the surrounding medium, and the relative heat transfer coefficients on the inner and outer surfaces, the optimum values of the material composition are obtained. Numerical data for the calculations are shown graphically.

抄録全体を表示

Effective creep compliance of carbon-fiber-reinforced composites is evaluated for applications of composites at elevated temperatures. A homogenization theory with two-scale asymptotic expansion in the Laplace domain is used to solve viscoelastic problems of composites. Effective constitutive equations and microscopic disturbed displacements are derived from the homogenization theory. A hexagonal array of fibers is employed for the microstructure of the composite and a hexagonal unit cell is placed in the microscopic field to represent a boundary value problem. In numerical calculations, a carbon-fiber-reinforced composite with thermoplastic matrix is considered at the glass transition temperature of the matrix. The matrix is viscoelastic and is represented using the generalized Maxwell model at the glass transition temperature, and fibers are considered as a transversely isotropic elastic medium. The effective creep compliance of the composite is determined by numerically solving a set of equations for macro-and microscopic fields.

抄録全体を表示

Bamboo is a composite material reinforced by fibers called bundle sheaths. It also has functionally graded structures on macro-and micro-scales. Bamboo has a cylindrical structure with many nodes, and its diameter and thickness vary with height from the ground, which is called the macroscopic graded structure. The fibers are densely distributed in the outer region, which is called the microscopic graded structure. This report presents an analysis of the large-deflective behavior of bamboo based on large-deflection beam theory. The analysis takes gravity and wind load into consideration The beam is modeled using the measured data of the sample bamboo. The maximum surface stress of the bamboo culm keeps approximately a constant even if wind load increases. Unique and smart structural behavior of bamboo deflection is described with respect to the graded structures.

抄録全体を表示

This paper examines the stress distributions in friction welded joints of dissimilar materials of copper and stainless steel under cyclic tensile stress. The thermoelastic stress analysis method is used to measure the stress distribution. This technique for stress measurement is based on detection of the temperature changes by thermoelastic effect. The stress distributions in the bonded joints of dissimilar materials on an elastic modulus are also calculated from the elastic stress analysis using the finite element method. The experimental results indicate that this technique is very useful for stress analysis on friction welded joints of dissimilar materials. Because the stress distribution map is quickly and directly obtained. The tendency of stress distribution is similar to the results obtained by elastic stress analysis. Particularly, near the interface for plate specimen, the experimental resut agrees with the result of three dimensional elastic stress analysis. The stress distribution in copper near the interface on the surface for round bar specimen differs from the result of elastic stress analysis due to the effect of change in mechanical properties and microstructures produced by friction welding. The results of elastic stress analysis considering anisotropic area near the interface agree with the experimental values.

抄録全体を表示

When a piece of nickel foil is attached on the surface of a specimen subjected to repeated loads at various temperatures in the range of room temperature to 350°C, the elastic surface stress is measured by making microscopic observations of slip bands in the foil, caused by repeated shearing strains. An image-processing system using a personal computer is applied for an automatic measurement of the density of slip bands in the attached foil. In order to apply the technique to the foil attached to a large specimen, the replica method using a plastic film is successfully applied to replicate the slip bands in the attached foil. The density of the replicated slip bands is also measured by the image-processing system and is found to be in good agreement with the density of the original slip bands. The peak stresses in grooved shafts under bending are measured with nickel foil gages and the image-processing system, and it is found that the results are satisfactorily accurate.

抄録全体を表示

A generalized phase-shift method is developed for automatic measurement of isochromatic fringe orders in the entire field of a model. This method uses two sets of light intensities obtained from the rotation of polaroids in dark-field and light-field plane polariscopes at each point. In order to investigate the accuracy of the measurement, light intensity curves in time series with different noise and different numbers of data which constitute the curves are used in computer simulation. The accuracy of the fringe orders obtained by the method is hardly affected by noise when the number of data exceeds l00. This method is applied to a circular disk subjected to a concentrated load. As a result, the fringe orders obtained by the method are found to be in good agreement with the theory.

抄録全体を表示

Three-dimensional large-rotation analysis of a beam by the finite-element method is possible using a beam element with the special orthogonal group SO(3). It is known that the stress rate used in the constitutive equation governs the tangent stiffness. Therefore, in this paper, the relationship between the stress rate and the tangent stiffness will be shown. The stress rates used are Truesdell stress rate, Jaumann stress rate, Neo-Green stress rate and Ishihara stress rate. It will also be shown that all tangent stiffnesses have the second variation of velocity gradient tensor because of S0(3), and the tangent stiffness of Ishihara stress rate has only material stiffness and geometric stiffness of rigid rotation. In the upcoming paper, tangent stiffnesses of those stress rates of a beam element will be shown.

抄録全体を表示

Finite-element analysis of a 3-dimensional large-rotation problem of a beam has been made possible by means of the nonlinear beam model developed by Simo and Vu-Quoc. Their beam model is regarded as geometrically exact, and is described by a configuration manifold which involves the rotation group. As a result, the tangent stiffness matrix becomes non-symmetric away from equilibrium. However, it have been shown in Ishihara that the tangent stiffness of semitangential rotation became symmetric, and by analyzing of fundamental problem, it was proven that symmetric tangent stiffness was valid. The beam model was assumed to have large rotation and small strain. Next, it was shown in Ishihara that the geometric stiffness of large strain in terms of the variation of the deformation gradient tensor before rigid-body rotation. However, that formulation was a total lagrangian formulation parametrized by the coordinate of the centerline of the reforence configuration. It was also shown in Ishihara that the large-strain beam model of an updated lagrangian formulation pararetrized by the coordinate of the centerline of the current configuration. Here, it will be shown that numerical examples of large-rotation and largestrain analysis of a beam.

抄録全体を表示

We have developed an error estimation technique for tetrahedral quadratic elements in a 3-dimensional structural analysis system. The technique is based on estimating the traction jump at element boundaries, where the exact traction force is continuous. The global error and element wise errors are defined in the same way as in the Z-Z (Zienkiewicz and Zhu) method except that average traction and traction jumps are weighted by the volume of each element. We also included an adaptive procedure which uses the error estimator in this analysis system for generating optimized meshes. We test this system on a simple problem, an infinite body with a spherical cavity under uniform tension, to examine the accuracy of this error estimator. Another example, a centrifuge rotor subjected to centrifugal force is presented to validate the practical use of this adaptive procedure.

抄録全体を表示

Link chains used in chain hoists are subjected to combined loads of torsion, tension, impact and bending moments. In this paper, fatigue tests for link chains concerning combined loads are performed. Moreover, a procedure for predicting S-N curves is proposed using the three-dimensional elastoplastic finite-element analysis and Coffin-Manson's law. Fatigue tests were conducted for link chains in order to examine fatigue strength of l0⁶ cycles. From the results of the fatigue tests, it is found that crack initiates from the border between the straight part and the curved part of the inner surface of link chains. It is found that the procedure suggested in this paper is quite suitable for the prediction of S-N curves. As a result, fundamental method of predicting fatigue strength of link chain is obtained.

抄録全体を表示

Three-dimensional elastoplastic analysis is carried out using the finite-element method (DYNA3D) in order to evaluate the stress distribution and to optimize the shape of link chains subjected to combined loads of impact tension and bending moment. The yielded region is evaluated using von Mises'criterion. The effects of chains geometry and contact angles of the link chains on the stress distributions are shown. Experiments were performed to evaluate the strain in link chains subjected to combined loads. Fairly good agreement is seen between the numerical and experimental results. Thus, it is clarified that von Mises'equivalent stress in link chains which pass through a sheave can be improved by optimizing the chain width. Von Mises' equivalent stress in chains when both contact angles θ₁ and θ₂ are 60 degrees, is about 8% greater than that when the contact angles θ₁ and θ₂ are zero degrees.

抄録全体を表示

This paper presents an inverse analysis method for the identification of unknown boundary condition in elastoplastic state, using a finite element method (FEM) on a small memory capacity machine such as a microcomputer and a personal computer. This inverse analysis method is composed of following two calculation processes: an identification process which obtains a solution in elastic state, and a direct analysis in elastoplastic state based on the identified results from. For efficient identification, uses a penalty function method and an attack technique based on both accelerative moving lower and upper bounds. A zooming identification method is available for improvement of the identification accuracy. adopts a triple-tandem-type substructuring method as an efficient calculation method. Therefore, we establish an efficient inverse analysis method for the identification of unknown boundary condition in elastoplastic state. Finally, the validity of the inverse analysis method is confirmed from the numerical solution of a three-dimensional finite element model, as an example of V-shaped grooved structure with reinforced members under tension, executed on a small machine.

抄録全体を表示

Green's functions are derived for axisymmetric surface force problems of an elastic half-space with transverse isotropy. Green's functions are defined as solutions to the problem of a transversely isotropic elastic half-space subjected axisymmetric surface forces acting on a circle. The surface forces acting on a circle are classified as (i) a radial force, (ii) a torsional force and (iii) an axial force in cylindrical coordinates. Green's functions for an isotropic elastic half-space were derived in a previous paper by applying Love's stress function. In this paper we apply Lekhnitsky's stress function in order to obtain Green's functions for an elastic half space with transverse isotropy. As an example of application of these Green's functions, we consider a problem of a transversely isotropic elastic half-space twisted by a flat annular rigid stamp.

抄録全体を表示

A new method for optimizing the cathodic protection of a ship in operation is proposed. First, from electric potential data measured using several sensors located on the wall of the ship, the polarization curve of the painted ship wall is estimated by solving an inverse problem using the boundary element method. Then, using the polarization curve, the optimum current to be supplied to each electrode (several electrodes are located on the ship wall) is determined in such a way that the electric potential of every part of the ship wall becomes lower than a critical value using minimum the necessary electric power. In case when the electric potentials at the sensors change abruptly due to paint damage, the location of the damage is estimated from the potential change by solving another inverse problem, and then the cathodic protection is optimized. An example problem is solved to demonstrate the applicability of the proposed method.

抄録全体を表示

To introduce a numerical simulation technique for the design sites of industrial products, a "CAD/CAE seamless" system from modeling to computation is desired. A meshless or element-free method is expected to be an effective procedure for realizing such a system, because time-consuming mesh generation processes can be omitted. The element-free Galerkin method (EFGM) proposed by Belytschko et al. is one practical meshless method. This study describes an application of EFGM to solve two-dimensional potential problems. A new methodology of enforcing the essential boundary condition in the EFGM formulation is proposed. Numerical examples are shown and the solutions by EFGM are examined by comparing with exact solutions or those by the conventional FEM.

抄録全体を表示

Because future advances in LSI (large-scale integrated circuit) technology will require further reduction of the size, the multilayered structure is being intensively investigated. The formation process of interconnections between the layers is one of the key issues in order to accomplish the structure. Reflow, whereby the via-hole is filled with the conductor metal by surface diffusion, is a promising method though its process has not yet been clarified in detail. In this study, the reflow process is simulated numerically for a two-dimensional model (filling into a groove) and an axisymmetrical one (filling into a circular cylinder) by solving the differential equation of surface diffusion by a finite dellevence method. The results show that perfect filling cannot be obtained for the groove or the cylinder when the aspect ratio, depth/half width or depth/radius, is larger than 3. Moreover, the time for the filling depends strongly on the aspect ratio.

抄録全体を表示