Fatigue behavior of Si₃N₄/Si₃N₄ and Al₂O₃/Al₂O₃ butt joints at elevated temperatures was investigated in air environment. The Si₃N₄/Si₃N₄ joint with titanium, silver and copper foils as interlayer materials was bonded at a temperature of 1 273 K for 3.6 ks at a pressure of 2 MPa in vacuum environment, whereas the Al₂O₃/Al₂O₃ joint with an aluminum foil was fabricated at 1073K for 1.8 ks at a pressure of 24 MPa. Three-point bending tests were conducted under both static and cyclic loading. Waveforms applied in the cyclic fatigue test were sinusoidal and triangular, and the frequency for cyclic load was usually 1 Hz. The fatigue tests were conducted at temperatures of 773-1 038 K. The main results obtained are as follows. (1) At the same temperature, cyclic lifetime is much longer than static fatigue lifetime at the same maximum stress. This behavior is observed to be much more marked in the case of loading a triangular waveform. Hence the cyclic lifetime depends on waveform but not on frequency of loading. Thus, the fatigue process of the joint at high temperatures is found to be time dependent. (2) The same value of apparent activation energy for the process is obtained in static and cyclic fatigue. Then the strength degradation observed in fatigue seems to be controlled by the same thermally activated process in both cases of static and cyclic fatigue. (3) Fatigue behavior of the joint depends on the material properties of the interlayer and the state of strain restriction in the joining layer regardless of the adherent material. (4) Very good agreement is obtained between the observed and calculated fatigue lifetimes.

View full abstract

Metallurgical factors controlling the fatigue properties of gas-nitrided titanium have been studied through some critical experiments concerning the fatigue strength of nitrided pure titanium whose grain size had been changed from 100 μm to 2800 μm through annealing prior to nitriding. The results are summarized as follows. (1) There exists a Hall-Petch-type relationship between the fatigue strength of nitrided pure titanium and its grain size. (2) The fatigue strength of pure titanium is improved by low-temperature nitriding (620°C) where the grain growth can be suppressed. (3) The fatigue strength of nitrided pure titanium decreases with increasing the thickness of compound layer at the specimen surface.

View full abstract

We carried out high-cycle internal pressure fatigue tests using a large number of thin-wall cylinders of austempered ductile iron (ADI). As a result, we observed a distinctive feature often seen in high-strength steel, that the SN diagram inclined again after about 10⁷ cycles, in spite of once revealing a horizontal part at 10⁶- 10⁷ cycles, and that an obvious fatigue limit could not been observed. Corresponding to this characteristic SN diagram, Weibull plots of the fatigue data were dramatically changed between high-and low-pressure levels. The thorough SEM observation of the fracture surfaces clarified that the fracture origins changed from casting defects to relatively small graphite particles between high-and low-pressure levels. This means that different fracture mechanisms exist between high-and low-pressure levels. These two mechanisms competed in the high-cycle region and brought about the distinctive fatigue behavior of ADI.

View full abstract

Closed-form expressions of the stress intensity factor are so useful in fracture mechanics analysis that a number of expressions have been obtained for various kinds of cracked specimens. However, most of actual structures are statically indeterminate, and stress intensity factors for such statically indeterminate cracked structures cannot be estimated only in terms of the above expressions. Such crack problems in statically indeterminate structures were systematically discussed by Okamura et al., and stress intensity factors for statically indeterminate cracked structures were shown to be estimated in terms of the crack compliance, i. e., the variation of compliance due to the presence of cracks. In this paper, an estimation scheme of closed-form expressions for crack compliance is presented. First, asymptotic properties of the crack compliance are obtained for single edge notched (SEN) specimens under both tension and bending. Then, closed-form expressions are estimated based on the asymptotic properties. The obtained expressions are shown to be quite accurate and useful for not only engineering discussion but also theoretical study.

View full abstract

Molecular dynamic simulation for ductile crack propagation of α-iron is carried out to study the effects of thermal control by using a scaling technique and Nose's methods as well as the direct simulation method. The equation of motion and the energy conservation law are formulated for the displacement loading condition, where a periodical boundary condition is applied in the thickness direction and the displacement rate is prescribed on the atoms at the boundary. Atomic rearrangement, temperature distribution and energy exchange are investigated for mode I loading to compare the three simulation methods. Effects of loading rate are also studied to obtain reasonable simulation results in crack propagation analysis.

View full abstract

In this paper, the axisymmetric torsion problem of two compounded infinite cylinder with an outer annular crack is considered on the basis of the theory of 3-dimensional elasticity. This problem reduced to the solution of an infinite system of simultaneous equations. The coefficient matrix is given by the product of three matrices involving the inverse matrix. Numerical results are illustlated for the distributions of displacement and shear stress, and for the variations of the stress intensity factor K_III various ratios of inner material's shear modulus to outer material and various ratio of inner radius to outer radius of the cylinder.

View full abstract

A life prediction method for creep crack initiation is proposed based on time dependent fracture mechanics. The material tested was Co-based superalloy. Crack initiation was investigated through microscopic observation under creep. That is, crack tip opening displacement, microcavity and its density around the notch root were clarified. The observed crack initiation behavior can be explained successfully by a model for crack extension caused by cavity coalescence. If local failure by cavity coalescence is controlled by the accumulation of critical strain over microstructural length, crack initiation lifetime is derived from the equation using time dependent fracture mechanics. In addition, creep crack initiation lifetime for a notched specimen was investigated. The same type life prediction could be derived using the Barenblatt-Dugdale model.

View full abstract

The body force method is one of the most powerful methods to analyze bent or branched cracks. In particular, a more accurate solution of the stress intensity factor can be obtained by applying the resultant force boundary condition (resultant force method). However, the crack displacement of bent or branced cracks cannot be obtained continuously by the resultant force method, and it is necessary to divide a crack into very small elements to achieve high accuracy. In this paper, we propose a new resultant force method to obtain continuous crack displacement of bent or branched cracks, which dramatically improves the accuracy of the solution. The effectiveness of this new method is illustrated by solving some typical crack problems. Moreover, simulation of crack propagation through grain boundaries is performed.

View full abstract

Load-controlled fatigue crack extension tests with a single peakoverload were carried out in copper. The localized strains in the vicinity of a crack tip were measured under cyclic loadings by means of the grid method, where special attention was given to the strain behavior during one load cycle before and after a single peakoverload. The application of a single peakoverload greatly opened the crack and intensely developed both the longitudinal and shear strains in the direction of θ≃±60°C ahead of the crack tip, where θ is the angle inclined with respect to the general direction of crack extension. Such strains were virtually retained and the crack did not grow during some loading cycles subsequent to the peakoverload application. The variation of the two strains and the crack tip opening displacement during half a load cycle were fairly small at the retardation period of crack extension, compared with those in the stable crack growth stage. When the crack extends beyond the region strained intensely by the peakoverload, the crack began growing again in a stable manner. At that time, the amounts of the strains around the crack tip and crack tip opening displacement returned to almost the same as those in the crack extension process before overload application.

View full abstract

In the process of studying the mechanism of the Bauschinger effect, we tested the fracture ductility of compressive prestrained copper. Then we compared it with that of materials prestrained in different conditions. As a result, two types of work hardening were considered. One is inherent work hardening caused by reversed strain, and the other is work softening caused by the reversibility of slip bands. We found that the degree of fracture ductility is mainly due to the latter. This work softening phenomenon is derived from the disappearance of dislocation.

View full abstract

Strength and fracture toughness tests were carried out on HIP-sintered SiAlON at elevated temperatures. Above 1200°C the strength decreased with increasing temperature. In the temperature range from 1400°C to 1600°C, the strength increased with temperature. On the other hand, the fracture toughness monotonously increased with temperature up to 1400°C. Above 1400°C the fracture toughness gradually decreased with increasing temperature. A stable crack growth region was observed both on the strength specimens and fracture toughness specimens, tested beyond 1200°C. From the detailed observation of the stable crack growth region, β-SiAlON particles were found to be oriented along the tensile direction. Increment of strength and fracture toughness beyond 1400°C may be due to the stable crack growth. The difference in temperature dependence between the strength and fracture toughness may be caused by high- temperature slow crack growth.

View full abstract

A boundary element method formulation is developed for forced steady-state bending vibration of thin elastic plates with viscous damping. An approximate fundamental solution for the biharmonic operator is used for the derivation of the integral representation. The integral equations for the deflection and the rotation are regularized up to an integrable order and then discretized by means of the boundary-domain element method. In addition to discretization of the boundary, the inner domain is discretized into domain elements. The final set of discretized equations consists of both the unknown nodal values on the boundary and the nodal displacements in the inner domain. A viscous damping effect is taken into consideration in the formulation. Numerical analysis is carried out for several examples, whereby computational aspects of this method are investigated in detail and its usefulness is demonstrated.

View full abstract

A particular formulation of the three-dimensional finite-element method is established for the elastic-plastic analysis of a triangular pyramidal indentation. The triangular pyramidal indentation on 0. 46% carbon steel and 70 : 30 brass is analyzed by the proposed method using the stress-strain curve of the tensile test. The calculated results are in good agreement with the experimental results. It is shown that the loading portion of the calculated load-displacement curves can be correlated with Vickers hardness, and the unloading portion, with Young's modulus. The extension of the plastic zone induced by the indentation made by an indenter is investigated using the analysis.

View full abstract

This paper is concerned with elastodynamic inverse analysis in which unknown defects, cracks and/or boundary conditions should be identified from displacement responses on the boundary of an elastic body. In the previous paper, the authors proposed the inverse analysis method for identification of unknown parameters using the extended Kalman filter algorithm and the boundary element analysis for steady-state elastodynamics. In this study, computational aspects of the inverse analysis method are investigated with respect to the influence of errors in the measured data and to arrangement of the measuring points. Furthermore, computational efficiency and accuracy are compared between the inverse analyses using the extended Kalman filter algorithm and the optimization method. Numerical simulation is carried out for some example problems to identify an unknown defect and unknown boundary values as a parameter identification problem.

View full abstract

In turbine design, because the joint between the turbine blade and the rotor bears the largest stress, it is important to study the stress at this position. The fir-tree type is one of the most efficient joint configurations for blade fastenings. Many papers have reported on the fir- tree turbine blade, but these mostly dealt with experiments using two-dimensional analysis. We therefore experimented with three-dimensional stress analysis in frozen photoelasticity. In this examination, the study was made of characteristic occurrences under centrifugal force produced by rotation simulating an actual turbine. The analysis of the fir-tree joint was performed for two tests. One was for straight entry and the other was for curved entry. Photographs were obtained for the photoelastic fringe, stress distribution, and stress concentration factors. The results were as follows radially, in both the straight and curved entry types, the highest stress concentration occurred at the tip serration. Axially, in the curved entry type, the highest stress concentration point was the axial center, and in the straight entry type, the distribution was uniform.

View full abstract

In this paper, an approximate solution of the elastic contact problem between film-coated bodies is proposed. This solution is analytically derived by applying the Hertzian contact theory and by assuming a stress field in the film-coated body. As a numerical example, the elastic contact problem between a film-coated cylinder and an elastic body, which is composed of elastic film and a semi-infinite elastic body, is treated. From the results, effectiveness of the present solution is examined by comparson with the corresponding solutions by the finite element method and Hertzian contact theory.

View full abstract

A new type of finite element method is presented for analyzing dynamic behavior of viscoelastic materials. In this method, the finite element formulation can be accomplished in simple form based on the correspondence principle between elasticity and viscoelasticity. Namely, the finite element equations for a viscoelastic problem are found by applying the Laplace transform to those for the corresponding elastic problem and then replacing elastic constants by viscoelastic functions. Laplace transforms of the nodal displacements can be obtained by solving the matrix equation described in the Laplace plane (complex s domain). Laplace inverse transformation is subsequently performed with the FFT (fast Fourier transform) algorithm to obtain solutions in the time domain. The present method is successfully applied to analyze some typical problems, and the numerical results show the adequacy of the method.

View full abstract

This paper is concerned with a new boundary element approach based on the time-stepping scheme for two-dimensional thermoelasticity in unsteady heat conduction. Coupling between the temperature and displacement fields is considered in the context of a quasi-static transient state. The time derivative in the differential equations is approximated by the time-stepping scheme. The reduced differential equations are transformed into a set of boundary integral equations using the exact fundamental solutions, which can be derived by the Hormander method. The standard boundary element method is applied to obtain numerical solutions at each time step. Computation begins in the initial state, and advances in turn by a time-marching procedure until the final state is realized. The mathematical formulation is presented in detail. The proposed solution procedure is applied to an example problem to show its usefulness.

View full abstract

A simple method for measuring contact pressure at a tool-workpiece interface is presented. The strain of the tool near the contact surface is measured using a pipe-type strain gage, and is converted into the pressure. The measured pressure distributions in plastic compression of rectangular blocks almost agreed with those estimated from the strain distribution. The pressure was distributed uniformly over the contact surface in the well-lubricated specimens regardless of the height-to-width ratio of the specimen, H₀/L₀ Friction hills were observed in the unlubricated specimens with H₀/L₀=0. 25 and 0. 5. A concave pressure distribution was obtained for the specimen with H₀/L₀=1, where the pressure increased from the center to the periphery of the contact surface. The present method can be applied to measure die pressure in practical metalworking processes.

View full abstract

In the displacement and strain measurement using moire interferometry, the out-of-plane rotation of a specimen caused by bending, for example, will strongly affect the measured values. When a specimen is deformed with an out-of-plane rotation angle θ_y about the y-axis, the normal strainε_x in the x-direction is given byε_x = (1/f)(ΔN_x/Δx)+(θ²_y/2), where f is the frequency of the reference grating and N_x is the fringe number of the fringe pattern. A photoresist master grating of 900 lines/mm was replicated on the surface of a cantilever beam specimen using silicon rubber. The measured displacement u and strainε_x in the x-direction deviated greatly from the theoretical values, but the values corrected for the rotation angle θ_y agreed well with the theoretical values.

View full abstract

It is considered that the strain rate dependence of the flow stress level of inelastic deformation of metal materials is caused by both viscoplasticity and strain ageing. In this study, the effect of strain rate on flow stress change, including temporal softening/hardening, which is observed after a sudden change in strain rate, is experimentally investigated. Each contribution of these two dominant factors to flow stress change is assessed quantitatively and is formulated for both SCM435 Cr-Mo steel and SUS316 stainless steel. There is a very good agreement in stress-strain responses to various changes in strain rate, between the calculated results based on the obtained expressions, and the experimental results.

View full abstract

Detailed observation of inhomogeneous deformation of polycrystalline metals and roughening of free surface under cyclic strain is important for basic research on the study of mechanisms of microscopic deformation, as well as on that of fatigue damage. Two-dimensional surface roughness and three-dimensional microscopic shapes of free surfaces of annealed copper and iron specimens were measured during cyclic plastic deformation with four-point bending. The relationship between the change in the three-dimensional surface shape and the behaviour of respective grains was observed with a mapping technique. It was found that the surface roughness changes with the cycle during the half-cycle of deformation, and the maximum and the minimum values of the surface roughness in a cycle increased gradually with the number of cycles. It was observed from the three-dimensional analysis that surface irregularities such as mountains and valleys caused by tensile and compressive strain are reversed in their shapes with the reversal of strain. The irregularity of surface with cyclic plastic deformation seems to be caused by preferential deformation in some specific areas or grains.

View full abstract

This paper describes a constitutive equation of dynamic plasticity using the internal time theory based on the concept of overstress of the yield surface. The proposed model employing the power law and exponential law is regarded as an extension of the static plasticity theory incorporating the combined isotropic/kinematic hardening. This paper shows a method of determining material constants in dynamic uniaxial loading for isotropic, kinematic and combined hardening models. The present model can accurately predict the experimental results of aluminum and mild steel under monotonic and cyclic loading over a wide range of strain rate conditions.

View full abstract

A numerical method using a finite element approach has been developed to analyze thermal deformation of short-fiber reinforced polymer products. In this paper, we propose a method for calculation of thermal deformation with an anisotropic stiffness matrix caused by fiber orientation ; the method takes an incremental approach, using the arc length and the modified Newton-Raphson method. Calculated results for a square plate and one-support beam were comparable with the analytical solution. In an other example, the calculated result obtained by the present method for a box-shaped polymer product is shown.

View full abstract

For shock drawing of thin plates using high-speed dry-ice projectiles, it is very important to understand the dynamical deformation process of the dry-ice projectiles and plates. In the present report, high-speed photographs are taken to visualize the direct impact sequence of circular cylindrical dry-ice projectiles on a rigid wall, and the hydrodynamical pressure generated at the impact surface is measured simultaneously. Also, the shock drawing of thin plates is simulated by means of the computer code AUTODYN-2D for the same configurations as in the experiments. In the simulations, paraffin projectiles are successfully used instead of dry-ice projectiles, although the effect of sublimation of the dry-ice on the impact phenomena cannot be taken into account.

View full abstract

In the preceding two papers, the shock drawing of thin circular plates using high-speed dry-ice projectiles was investigated experimentally and by numerical simulation. The austenite stainless steel SUS304 used in the experiments is partially transformed into the martensite structure by machine processing, and ferrite is deposited according to the degree of strength of the machine processing. In this study, the distribution of ferrite content of the shock-drawn plates is measured and microscopic observations are made in orde to estimate the shock load in the drawing process.

View full abstract

The density of sintered materials becomes inhomogeneous partially because of the friction between the powder and the die during compaction. In this paper, we evaluate the inhomogeneity of the sintered materials by means of measurement of the ultrasonic wave velocity. It is shown that the ultrasonic wave velocity depends linearly on the density when the density variation is small. The relation between the density of the sintered copper and the ultrasonic wave velocity is determined by experiment, and the density distribution of the sintered materials is evaluated. The velocity and the attenuation also depend on the frequency of the wave and the powder diameter ; hence the powder diameter can be evaluated from the dependency of the velocity or of the attenuation on the frequency.

View full abstract

In order to utilize magnetoelastic interaction effect in ferromagnetic materials for measuring stress, it is preferable that measured value affected by the interaction is obtained independently of the detecting system. When a sheet is magnetized, magnetic flux density near the sheet is affected by the interaction, the value of which can be measured independently of the detecting system. A detecting system using Gauss meters is proposed to measure the relationship between B_b/B_a and uniaxial stress J, where B_a and B_b are the magnetic flux densities near the back surface of the sheet and near the sheet surface, respectively. By conducting experiments which use various values ofσ, magnetizing frequencies and low magnetic fields, it is shown that the system can detect the relationship sensitively, and that frequencies more sensitive to the relationship exist. Also, a method for evaluating the value of stress in the sheet independently of the detecting system is shown based on the results obtained.

View full abstract

Functionally gradient material (FGM) is one of the typical nonhomogeneous materials. In this paper, an analytical solution based on a kind of integral transform developed by Vodicka for the composite regions is presented for the transient heat conduction problem in a nonhomogeneous plate with arbitrary nonhomogeneous thermophysical properties. The material properties of SiC/Al FGM with the uncertainty of distinction between the matrix phase and the filler phase in an intermediate composite and the uncertain change in the microstructure of the filler phase are estimated based on Mamdani's method of fuzzy inference. The associated transient thermal stresses under the various thermal boundary conditions are analysed with the use of the closed- form solution derived by one of the authors for a nonhomogeneous plate with arbitrary variations in mechanical properties through the thickness, and the composition suitable for the reduction of thermal stress is discussed.

View full abstract

An inverse analysis using the optimization technique with a boundary element method is presented for quantitatively estimating residual stresses caused by volume changes in a semi-elliptical region in a structural component. In the present paper, this inverse problem is reduced to the problem of minimizing the optimization function defined as the sum of the square of differences between experimental values and numerical solutions, where experimental data include deformation at the free surface of the block with rectangular cross section subjected to surface treatment, and numerical values are calculated by using the boundary element method. In order to investigate the influence of measurement errors included in the given additional information on identification results, numerical simulation is carried out for the case where the given displacement information includes probabilistic errors with the normal distribution. Furthermore, the distributions of residual stresses measured by the X-ray diffraction technique for partially quenched S45C steel are compared with the results calculated by inverse analysis, and discussed as to the validity of the identification procedure.

View full abstract

This paper deals with the singular stress field near the interface edge point in a bonded strip. The stress field around the singular point is described in terms of one constant, K_σ. The parameter K_σis calculated by using the body force method. In the numerical analysis, the singularity of the stress field is characterized by introducing a proper basic density function of the body forces. The use of the basic density function enables one to obtain accurate solutions.

View full abstract

In this paper, a method for study of the logarithmic singularity in two bonded wedges is proposed. By mean of the proposed method, the general expression of the stress field with logarithmic singularity is derived for a half-plane bonded to a wedge and an interface crack. It is found that surface traction is necessary to produce the logarithmic singular stress field. For the interface crack, the logarithmic singularity occurs only when the applied uniform shear stresses are not equal (t₁≠t₂).

View full abstract

The method of estimating residual stresses in a film and a substrate has been proposed through use of the relationship between inherent strains and inherent stresses which is a kind of an inverse problem. The well-known method for determining the residual stress was the use of Stoney's equation estimating the average residual stress in the film from a deflection of the substrate. The proposed method focuses on the inherent strain generating the inherent stress such as a residual stress, then the strain is estimated from the relationship between the thickness of film and the deflection. The new method does not require the assumption of the constant stress in the film, and the stresses in the film and the substrate are simultaneously calculated from the inherent strain.

View full abstract

In recent years, composite materials and metal-ceramic joints have been used widely as industrial materials. Failure in these materials frequently occurs at the vicinity of bonded edges. It is therefore important to clarify the stress and displacement fields around the bonded edges. In this study, the authors first proposed the use of a triangle element for accurately analyzing a stress field with a singular line occurring along the section of interface and free surfaces in a three-dimensional bonded structure. The fundamental solution, including a boundary condition for the bonded interface (Rongved' solution) was next derived, and the stress distribution around the bonded edges was clarified by means of the three-dimensional boundary element method using this solution. It was found that the order of the stress singularity of three-dimensional dissimilar materials is between that of plane strain and plane stress, and that as an internal point approaches the free surface along the bonded edge, the order of stress singularity approaches that of the plane stress. The order of stress singularity was investigated in the spherical coordinate system with the origin at the bonded edge. The order is constant in all directions, and is larger than those for plane strain and for plane stress. When the apex angle of the bonded edge is changed to an angle incurring no stress singularity in plane strain or plane stress, the stress singularity vanishes in three-dimensional dissimilar materials.

View full abstract

The plane problem for a three-phase bonded structure formed from three isotropic homogeneous rectangular wedges is analyzed using the theory of elasticity. The stress field near the apex is represented by the type Kr^P-1, where p depends on three shear modulus and on three Poisson's ratios, and K indicates the intensity of the singular stress field. Emphasis is placed on investigating a singular stress field at the vicinity of the intersection of the interfaces in the three-phase bonded structure and a free surface subjected to normal loading. Numerical results of K for different combinations of materials are indicated for the case where wedge angles ψ₁, ψ₂, ψ₃ are ψ₁ =ψ ₂= ψ₃ =π/2. The stress behavior in the three-phase materials for the limit of r→0 is studied in detail on the basis of the numerical results.

View full abstract

Experiments on exfoliation temperature at the interface of a bonded joint have been made to estimate the joint strength of dissimilar materials and the effect of specimen size on the strength. The joint was made of a resin cured on steel. The exfoliation temperature was measure by a thermocouple and visual observation. It is shown from the experimental results that the exfoliation temperature becomes constant when the width of resin is greater than 8 times the thickness of the resin and when the thickness of the steel is greater than twice that of the resin. The energy release rate g-value of virtual crack length, which exists within a singular field in the vicinity of the corner due to a combination of dissimilar materials and the wedge angle, is a useful parameter for predicting the exfoliation temperature.

View full abstract

We investigate the main-boundary behavior of coupled sliding and migration induced by thermal activation, using a molecular dynamics simulation. The two [001] (310) Σ=5 tilt grain boundaries are simulated in aluminum, where Morse potential and periodic boundary conditions are used. The simulations show that when the temperature is low, or when the distance between the two boundaries is long, the boundary structure is likely to be preserved, and the sliding and migration of one grain boundary tends to be accompanied by sliding and migration of the other grain boundary. When the temperature is high, or when the distance is short, these tendencies are weakened. Boundary sliding and migration is unlikely in crystals where the boundary structure is disordered, because dislocation generation and mobility are difficult in such crystals.

View full abstract

Shape memory properties of Ti-Ni-Cu alloy under a mild hydrogen absorbing condition have been studied. Decreases in mechanical properties such as tensile strength and total elongation occurred because of hydrogen absorption. Critical value of hydrogen content for the decrease in mechanical properties was higher in comparison to Ti-Ni alloy. Decreases in shape recovery stress and recovery rate due to hydrogen absorption during the cyclic shape memory test were not as great as that of Ti-Ni alloy. These results were considered to be caused by relatively small plastic deformation and dislocations introduced during the cyclic test, resulting from small absorption content of hydrogen due to the presence of some protective surface layer. In conclusion, Ti-Ni-Cu alloy is suitable for applications in hydrogen environment, in comparison to Ti-Ni alloy.

View full abstract

Mechanical components such as steam turbine blades may experience heavy damage when subjected to repeated impingement of liquid droplets. In order to evaluate the erosion damage quantitatively, a high-speed erosion test rig utilizing water droplets was used to study the erosion resistance of 12Cr steel, cobalt base alloy and titanium alloy. From the experimental results, it reflects that erosion does not proceed at a constant rate. The characteristic erosion-time curve consists four stages : the incubation period, the acceleration period, the deceleration period, and the steady-state region. In this paper, a universal curve fitting approach is proposed in an attempt to define this erosion process. With optimal values of the five main parameters, it may be possible to more closely correlate the experimental data with the proposed model. Furthermore, the relation-ships among these parameters, impact velocity, droplet diameter and material hardness are discussed, and a prediction method of erosion resistance is presented.

View full abstract

The structural sensitivity analysis for elasto-plastic problems is more difficult than that for the other nonlinear problems because of the path dependency of sensitivity. In this paper, the sensitivity analysis method that the authors have developed is applied to a realistic example of an elastic-plastic large displacement problem of a grid spring which supports fuel rods in a PWR. The sensitivity of the reaction force and maximum Mises stress with respect to the thickness of the spring is analyzed. The sensitivities obtained by the present method are compared with those by the finite difference method to show the good agreement. The stochastic finite element analysis is also exemplified based on the sensitivities and the experimental data of the random thickness of the spring.

View full abstract

In injection molds and compression molds, slender cooling channels or heating pipes are used to control the temperature over the cavity surfaces of the molds. The optimum design of layout and size of such slender channels is important for high-quality production and reduction of the forming cycle time of products. In this paper, a boundary element approach for design sensitivity analysis of cooling or heating of such molds is presented. The boundary integral equation for steady-state heat conduction problems implemented with a special type of pipe elements is first differentiated with respect to an arbitrary design parameter, and a boundary integral equation relating the temperatures and heat fluxes to their sensitivity coefficients is derived. The effectiveness of the system developed based on the above formulation is illustrated through some numerical test examples.

View full abstract

A digital measurement system for the time-of-flight of ultrasonic longitudinal and transverse waves is constructed in terms of a conventional pulser/receiver, a transducer and a high-speed digital wave memory for material characterization and acoustoelastic stress measurement of solid materials with weak dispersion. Three methods, zero-crossing, cross-correlation and phase spectral methods, are compared from the viewpoint of precision for the time-of-flight measurement. The measurements of aluminum (10 mm in thickness) and acrylic (5 mm) plates have shown that the precision (2σ) of the time-of-flight is estimated to be less than 0.01 ns by the cross-correlation methods, which corresponds to the relative velocity variation of 2×10^-6 for the longitudinal velocity of the aluminum plate. The cross-correlation method is found to have the highest stability.

View full abstract

A three-dimensional finite element computer program for thermal stress analysis was developed for the cracking of trigonal class 3 m oxide bulk single crystals such as lithium niobate (LN) and lithium tantalate (LT) during Czochralski growth. A tensor transformation technique was use to obtain the elastic constant matrix and the thermal expansion coefficient vector corresponding to an arbitrary pulling direction. The anisotropy of the elastic constants and the thermal expansion coefficients as well as their temperature dependence was considered in the program. Using this program, we analyzed thermal stress in LN bulk single crystals with three types of temperature profiles or crystal-melt interface shapes. The analyses were carried out for the pulling directions of both the a-axis [100] and the c-axis [001]. The results show that thermal stress is very sensitive to the crystal-melt interface shape, and that thermal stress is lower in the c-axis pulling than in the a-axis pulling. The relation between thermal stress and the cracking and quality of a bulk single crystal is also discussed.

View full abstract

In the numerical treatment of the optimum design of a both-ends-clamped beam which is subjected to a concentrated load at a position other than the center of the span, the computation process is not stable. In this study, I analytically solve the optimization problem of both-ends-clamped beams, based on the inverse variational principle and consider the cause of the instability of the computation. The shapes obtained are classified into three types depending on the position of the load. It is found that the solution of the problem in the true sense of the word exists in a limited case of the position of the load. Moreover, the behavior of beams which have cantilever-like shapes is compared with the solutions and it is shown that beams with cantilever-like shape have comparatively small strain energies in a numerical sense.

View full abstract

A multiblade fan is generally used as a sturdy and easy-to-use fan in many industrial fields. In conventional design, the structure of a fan blade is analyzed using Wiesmann's equation in which only bending deformation under centrifugal force is considered with the beam model. However, it is often reported that fatigue failures occur in fan blades designed using this equation. In this paper, the stress distributions on the blade were measured and the problem of Wiesmann's equation was clarified by comparison with experimental results. On the basis of experimental results, an original algorithm of deformation occurring in blades was proposed, and structural analysis of the fan blade was carried out by means of the finite element method (FEM) under the proposed algorithm. From these investigations, the results obtained are as follows. The body force due to centrifugal force generates torsional moment in addition to bending moment because positions of the center of gravity and the shear center don't coinside caused by the open section of the blade. This phenomenon causes stress concentration at both ends of the blade, and the stress caused by the torsional moment is dominant. It is suitable in design that the stress caused by trorsional moment be taken into consideration, as well as stress caused by bending moment, from the viewpoint of the fatigue strength.

View full abstract

This paper deals with structural optimization in view of topology. The numerical examples are concerned with the partition of a certain area into equal parts of prescribed number with the shortest boundary line. The problem is to find the nodes and boundaries which satisfy these conditions. The boundary cycles are utilized to represent the topology of the area partition, and the genetic algorithm is used in the process to find the optimal boundary cycles in order to minimize the total length of the boundaries, which is given as a function of the nodal coordinates and curvatures of the boundary lines. The examples verify the validity of the proposed method.

View full abstract

Cylindrical shells with multi-step variable thickness are often encountered in the design of storage tanks for oil, LNG and so on. The sidewalls of the tanks increase in thickness from the top to the bottom, and are subjected to external pressure caused by wind load or insulation weight. The standard design procedure for the sidewall is based upon an assumption of a constant-thickness shell as represented by the weighted-average thickness of the sidewalls. This paper presents an analysis of the buckling strength of the thin cylindrical shells with two-step variable thickness under external pressure. The objective is to investigate the applicability of the standard design procedure. Using the finite-element method and taking geometrical nonlinearity in the prebuckling deformation into consideration, the problem is solved parametrically. Results show that the standard design procedure underestimates the buckling pressure on the shell for the cases in which the thinner portion of the sidewall is longer than the thicker portion.

View full abstract

The hydrodynamic effect of shear flow on red blood cell deformation was investigated in vitro with a rheoscope. Human red blood cells were suspended in a dextran water solution and were sheared between two counterrotating transparent discs. Deformation of red blood cells in the shear field was observed with the aid of an optical microscope, a charge coupled device camera and a video tape recorder. Red blood cell deformation was evaluated from a deformation index, which was calculated from the ellipsoidal shape. Results show that the deformation index increases with increase of shear stress of < 6 Pa.

View full abstract

To investigate the wear mechanism of the ultra-high-molecular-weight polyethylene (UHMWPE) for an artificial knee joint, the pin-on-disk-type wear test and the numerical analysis of contact mechanics of the UHMWPE are performed. In this analysis, to calculate the cyclic deformation of the UHMWPE, the constitutive equation for cyclic plasticity is used. In the initial stage of the deformation, the UHMWPE is deformed only plastically, but beyond this stage, the wear of the UHMWPE progresses rapidly with weight loss. In this initial stage, the shear stress and the equivalent plastic strain are accumulated by cyclic loading at the contact surface of the UHMWPE and it is expected that the microcracks that create wear particles will be generated at the surface.

View full abstract