It has already been indicated, as a general trend in impact fatigue, that the impact fatigue strength is lower than the ordinary fatigue strength, and that the crack growth rate under impact fatigue is higher than that under non-impact fatigue, excepting a few materials with relatively high static strength. The first aim of this study is to clarify the impact fatigue behaviors of high-strength alloy steels with the ultimate tensile strength of about 1800 MPa, and the second is to evaluate the crack growth behavior under impact fatigue load in connection with the dynamic stress-strain response. For these purposes, a series of impact fatigue tests were first carried out on martensitic stainless steel, and then single impact tests were performed to obtain dynamic stress-strain relation-ships of the martensitic stainless steel and the other three steels whose crack growth behavior in impact fatigue were already obtained by the authors. The S-N property of the martensitic stainless steels is similar to the other steels but the fatigue crack growth rate under impact fatigue load is equivalent to that under ordinary fatigue load. An interesting finding is that the crack growth behavior of each material under impact fatigue load is well correlated with the dynamic stress-strain response of that material.

View full abstract

This paper describes the low-cycle fatigue life for three kinds of cast iron. Strain-controlled low-cycle fatigue tests were carried out for flake graphite, compacted vermicular graphite and spheroidal graphite cast iron at 723 K in air. The fatigue life of spheroidal graphite cast iron was the largest while that of flake graphite cast iron was the smallest. Compacted vermicular graphite cast iron showed intermediate fatigue strength. The modified universal slope method was developed and it could correlate the fatigue life of the three types of cast iron uniquely within a factor of two scatter bands.

View full abstract

Soda-lime glass specimens with an indented induced flaw at the center were fractured under a constant load and a cyclic load applied by the four-point bending of plates, concentric-ring loading of plates and diametral compression of disks, respectively. All glass specimens showed a susceptibility to static and cyclic fatigue failure, and their lifetimes increased with decreasing maximum applied stress. Equivalent time-to-failure obtained from the cyclic data, using the assumption that there is no enhanced effect of cycling on the rate of subcritical crack growth, were compared with the static time-to-failure data measured under polyaxial stress states. The influence of cyclic loading on fatigue life was not significant in cases with uniaxial tension, but the cyclic loading effect was obviously seen in both cases of tension-compression and equibiaxial tension.

View full abstract

Effects of grain-boundary reaction (GBR) precipitates on the improvement of fatigue-crack initiation and propagation resistances were investigated in an austenitic 21Cr-4Ni-9Mn (21-4N)steel at room temperature. Grain boundaries were considerably serrated by the GBR precipitates. Fatigue cracks were initiated along slip lines and then propagated within the grains. The short cracks, whose length was one or two grain diameters, were deflected at the GBR nodules and propagated around the GBR nodules. Long cracks, which were longer than about 1.0mm, propagated initially within grains and then along grain boundaries. When the cracks propagated along grain boundaries, the crack propagation rates were lower along the serrated grain boundaries than along the straight ones. The geometrical effects such as the crack deflection and the crack closure, which were induced at the deflected portions of the crack paths, were important in improving the short and long crack propagation resistances.

View full abstract

A static fatigue test was carried out with a steel adhesive joint in distilled water at 18∼50°C, and the static fatigue damage of the adhesive was analyzed on the basis of the theory of diffusion. The main conclusions obtained are as follows : The part of the adhesive containing more water than certain critical concentration, c_r, can be supposed to be a crack. The process of the static fatigue of adhesive joints in water is considered to be a thermally activated process controlled by the stress-accelerated diffusion of water. The static fatigue life, t_f, is expressed as follows;[numericai fornula] where D^^∼ is the effective diffusion constant, ψ and c^* material constants, σ_f* the initial strength, and h the height of the section of specimen. The effect of side diffusion on the fatigue life is not so significant.

View full abstract

Statistical fatigue tests were conducted on structural carbon steel, S 45 C, in room air and in 3 % NaCl solution using three cantilever-type rotating bending fatigue testing machines which were specially manufactured for the purpose of the present study. Fatigue life distribution was examined at three stress levels in both environments, and twenty specimens were allocated to each stress level. In room air, it was found that fatigue life distributions followed the three-parameter Weibull distribution and were closely related to fracture morphology. In 3 %NaCl solution, they also followed the three-parameter Weibull distribution, but the shape parameter and the scatter in fatigue life were smaller than those in room air. It is suggested that the decrease in the scatter of fatigue life may be attributed to the small number of cycles to crack initiation in 3 %NaCl solution.

View full abstract

Fatigue crack propagation can be represented by a stochastic process such as a Markov process, which accounts for the distribution of crack propagation fatigue life. However, since we normally are able to obtain only limited information about the actual stochastic properties of the fatigue crack propagation process, it is necessary to represent the process by an appropriate model. This paper presents a new stochastic model which treats the material's resistance against fatigue crack growth as a spatial stochastic process evolving along the path of the crack. To obtain the stochastic properties of the fatigue crack propagation process, constant ΔK tests were conducted for several ΔK cases, and the parameters of the model were calculated by a spectral analysis using the maximum entropy method (MEM), which accounts for the statistical correlation that has been observed between adjacent fatigue crack growth rates. Using a proposed random crack propagation model, Monte-Carlo simulations were also performed and satisfactory agreement with the results of experiments were obtained.

View full abstract

It is widely recognized that the fatigue crack propagation is fundamentally a random process which can be predicted only in terms of probability. The primary source of statistical variation of fatigue crack propagation is material inhomogenity. To explain its effects, the authors proposed in the previous paper a new stochastic model which treats the material's resistance against fatigue crack growth as a spatial stochastic process along the path of the crack. In this paper, using a proposed random propagation model, Monte-Carlo simulations are performed for the constant load amplitude and the influence of the parameter variation on the results is investigated. It is shown that the statistical correlation of resistance at different points along the crack path has great influence on the statistical properties of the distribution of the crack propagation fatigue life. The results are also compared with well-known experimental data sets and satisfactory agreements are obtained.

View full abstract

The authors have already derived the equation by which crack opening stress distributions and the resulting stress intensity factors (SIF) can be calculated from the crack opening displacements along a crack, based on the complex-valued formulas on the stress and displacement of a two-dimensional crack. In this study, this equation is used to evaluate the SIF of a three-dimensional crack. For elliptical internal cracks, the SIF at the principal axis could be evaluated theoretically under a plane strain condition from the crack opening displacement distributed in the vertical plane including the principal axis. For surface part-through cracks, the SIF evaluated under a plane stress condition from the crack opening measured along a crack on the specimen surface were in good agreement with the SIF determined from Newman-Raju's equation necessary for both the applied stress and the surface crack shape.

View full abstract

The crack nucleation mechanism in an alumite substrate for thin film magnetic film disks is investigated by using the scratch technique. The alumite substrate consists of a hard-brittle anodized alumite layer of μm-order thickness and a soft-ductile Al-Mg alloy base. In the scratching of the alumite substrate even with a relatively blunt indenter, the so-called median crack is observed at a low load prior to the tensile bow crack. This crack is a typica nucleation process when bulk hard-brittle materials are scratched by very a sharp indenter such as a Vickers indenter. It is also observed that two types of median cracks exist. One is nucleated at the elastic/plastic boundary just below the indenter and is propagated downward. Another is nucleated at the alumite/Al-Mg alloy boundary and is propagated upward. These crack nucleation mechanisms are investigated by stress analysis using FEM as well.

View full abstract

The crack propagation velocity in tensile fracture is determined by comparison of the unloading wave observed by the strain gage method with the theoretical analysis based on postulating the crack growth models which are backed by the fractographic analysis using the photograph taken by the scanning electron microscope (SEM). The effect of elongation, contraction of notch bottom and load rate just before rupture on the crack propagation velocity is investigated using the notched round bar specimens of annealed JIS-SS41, S45C and quenched SKS93 steel. The fracture surface of SS41 and S45C is a dimple and SKS93 a cleavage. The deformation of SS41 is largest and in turn S45C, SKS93. It is shown that the crack speed slightly decreases as elongation or contraction of notch bottom increases. Particuraly, the decrease of crack speed of quenched SKS93 is a very little. The reduction of crack speed attributes to softening of notch bottom.

View full abstract

In an idealized 2-dimensional crack, the deformation constraint is kept constant along the crack front line. Thus it is expected that the fracture mode is also kept constant along the crack front. However, the deformation constraint and fracture mode change generally along the crack front line in a 3-dimensional crack, and this deformation mode change cannot be dealt with as far as conventional parameters such as K or J are applied. The CED (Crack Energy Density) was proposed as a parameter which has no restriction on the constitutive equation and is expected to be applicable to almost all kinds of crack problems. This paper purports to demonstrate that the deformation mode change above can be evaluated quantitatively when the CED is applied.

View full abstract

In linear fracture mechanics, stress intensity factors controlling the fracture behavior of materials are calculated from stress distributions acting on the crack surfaces by the aid of superposition even for remote applied stresses. An equation was derived by which such stress distributions can be assessed from each mode-discontinous displacement along a crack with mode I and II components, based on the complex-valued formulas on the stress and displacement of a two-dimensional crack. The validity of the derived equation was theoretically checked by a closed-form solution of a slant crack under biaxial loading. For a slant crack 45 deg to the tensile loading axis in a rectangular specimen, the stress distributions and stress intensity factors K_I and K_II, which were evaluated from the discontinous displacements measured along a crack, were in good agreement with those values calculated from the applied stresses.

View full abstract

This paper describes an application of statically indeterminate fracture mechanics, SIFM, to evaluating the structural integrity of axially cracked internally pressurized and spinning cylinders. The following points were made clear : The J-integral obtained from SIFM agrees well with the conventional fully plastic solution given by Kumar et al. in the case of the internally pressurized cylinder ; SIFM can give the J-integral value equivalent to the J-integral keeping path independence in the case of the spinning cylinder : Based on these results, a quantitative study was carried out for the United Kingdom Atomic Energy Authority, UKAEA, test : here, a structural integrity test was performed under pressurized thermal shock (PTS) by using the spinning cylinder.

View full abstract

The notched strength of unidirectional carbon/epoxy laminate, Toray T800H/#3631, was evaluated under tensile loading using center-notched specimens having various fiber orientations relative to the loading axis. On the basis of acoustic emission (AE) monitoring and fractography, failure mechanisms can be classified into two modes: tensile failure and shear failure. For the case of tensile failure, no AE signal was detected before the split initiation; the fracture surface was covered with epoxy resin. On the other hand, for the case of shear failure, a gradual increase in total AE energy was observed before the split initiation; the fracture surface was characterized by many hackles and fiber/matrix interfacial debonding. The measured strength for the split initiation from notches showed good agreement with the predictions from the modified average stress model combined with the maximum stress criterion proposed by the authors. No interaction was observed between tensile failure and shear failure for split initiation.

View full abstract

A new theory of propagation of elastic waves in porous materials was developed for the pore characterization. The idea is based on the arrival probability of a direct ray, reflecting ray and creeping ray. The theory gives a good account of the nature of elastic waves propagated in porous solids not only quantitatively but also quantitatively. The authors showed a rather good correlation between pore characteristics (porosity and pore size) and ultrasonic characteristics (waveform and arrival time), and proposed a nondestructive technique based on the presented theory for evaluating the porosity and pore size.

View full abstract

In the impact 3-point bending test, it had been shown that a loss of contact between specimen and anvils occurred in the early stage of impact, and after that the specimen was subjected to the reaction force from the anvils. As a loss of contact is induced by the inertia effects, the dynamic fracture behavior in the test depends on the dimensions of specimen. In this paper, the dynamic fracture behavior of the specimen made of epoxy resin is investigated by changing the specimen length. In the impact 3-point bending test, the stress intensity factor at the crack tip is analyzed by the strain gage technique and the dynamic photoelastic technique using high speed camera. The loss of contact between the specimen and the anvils are also examined. The experimental results are compared with the theoretical values by using the simple method evaluating the dynamic stress intensity factor and dynamic finite element method. The influence of overhang on the fracture toughness in this test are presented.

View full abstract

The buckling strength was evaluated for a nonsymmetrical reinforced cylindrical shell with a branch joint subjected to external pressure. The property on stresses of the shell was analyzed by FEM around the joint where the branch/main bore ratio is close to one. Large deformation theory based on FEM was applied to the structure around the joint against the buckling in the following two cases, a) circumferencial stress and axial stress are applied on the shell, and b) initial irregularity exists in shape in addition to the above stresses. This evaluation method was verified by the model experiments simulating the branch joint. The above-mentioned technique was proven to be valid in practical use.

View full abstract

Impulsive responses of a circular cylinder filled with fluid and subjected to uniform load are analyzed. The coupling effects on the responses are investigated using analytical solutions and approximate solutions of mass-spring models. Impulsive responses of the cylinder are divided into three groups with a nondimensional number β=KR(1-ν²)/Eh where E, ν, h and R are Young's modulus, Poisson's ratio, thickness and radius of the cylinder, respectively, and K is the bulk modulus of fluid. For β of the order of approximately one, the coupling effect on the responses appears remarkably. When β<< 1, the fluid motion hardly influences the behavior of the cylinder. When β>>1, the behavior of the cylinder is dependent on the fluid motion.

View full abstract

The thermal shock behavior of cermets and cemented carbides was studied in detail by using unnotched smooth bar specimens. Thermal shock was given to the specimens by plunging them into a water bath at 20°C. After the thermal shock experiments, bending strength, micro vickers hardness and fracture toughness of the specimens were investigated. As a result, two different types of thermal shock behavior were observed between cermets and cemented carbides. These differences indicate that microcracks occur more easily in cemented carbides than cermets during thermal shock experiments.

View full abstract

The effect of shock pressure on the stress-strain behavior of four plain carbon steels with 0.06, 0.17, 0.44 and 0.85 wt % C was investigated. The shock pressure was generated in water by an explosive. The shock pressure effect was compared with the effect of pressurizing or temper rolling on the stress-strain behavior. Susceptibility to twinning caused by the shock loading was also examined in the steels with different carbon content. Results obtained in this investigation are summarized as follows. The yield stress of three steels except for 0.85 wt % C steel began to decrease at the shock pressure of 300 MPa and continued to decrease to the minimum with increasing pressure. However, the yield stress of 0.85 wt % C steel was never lowered by the shock pressure. For the steel with larger carbon content, the amount of decrease in the yield stress became smaller and the shock pressure which minimized the stress became higher. Yield ratios of the steels containing 0.06, 0, 17 and 0.44 wt % C, which had the original ratios of 0.65, 0.65 and 0.53 respectively, were reduced to 0.38∼0.40 by the shock loading, On the other hand, the ratio 0.37 of 0.85 wt % C steel was not lowered. The shock pressure (P_s)_t at which the mechanical twins began to be formed in the specimen increased linearly from 720MPa for 0.06 wt % C steel to 3100 MPa for 0.85 wt % C steel with increasing carbon content. The pressure (P_s)_t did not always depend on the yield stress value. Twin density defined as the ratio of number of crystal grains with twins to total number of grains was smaller in the steel with larger carbon content and the density monotonously increased with increasing shock pressure. The decrease of the yield stress by the shock loading was considerably larger than the decrease in the cases of pressurizing and temper rolling.

View full abstract

In this paper, the theory is developed for the thermoelasto/viscoplastic response of a transversely isotropic thick-walled spherical shell suddenly subjected to a uniform temperature rise over its cross section. The dynamic response is studied by applying the theory of rays, which give rise to the exact solution of the transient response. Numerical results indicate the variations in the dynamic thermoelasto /viscoplastic stresses in a transversely isotropic spherical shell with time and show the effect of viscoplastic properties explicitly.

View full abstract

A stochastic thermal stress problem in a nonhomogeneous hollow circular plate is studied. The circular plate is subjected to unaxisymmetric random heating with respect to time on the inner boundary surface and has nonhomogeneities of thermal conductivity and Young's modulus expressed in forms of different power laws of the radial coordinate, the coefficient of linear thermal expansion given as an arbitrary function of the radial coordinate and constant Poisson's ratio. In this analysis the autocorrelation function and the power spectral density are obtained as statistics of temperature and thermal stress which are stochastic processes depending on time. The analysis is done by applying the solution of the thermal stress problem in a nonhomogeneous hollow circular plate which has been derived by one of the present authors. Numerical calculations of the statistics of tempera-ture and thermal stress are carried out for the case in which the random temperature on the inner boundary surface is a stationary Gaussian process which has Markov property.

View full abstract

This paper is concerned with a transient thermal stresses and thermal deformations of the axisymmetric problem of a semi-infinite solid cylinder on consideration of a moving boundary. Assuming that an end surface of the cylinder is heated with arbitrary relative heat transfer and moves with a constant velocity, the temperature distribution of the cylinder is analyzed using with a moving coordinate system. Thereafter, the associated thermal stress distributions and thermal displacements are determined with the aid of the thermoelastic potential function method and Love's displacement function method. As an illustration, numerical calculations are carried out for several velocity values of the moving boundary and the relative heat transfer coefficient of the heated end surface. The influence of the velocity and the heat transfer on the temperature, the thermal stresses and the thermal deformations are examined precisely.

View full abstract

A three-dimensional finite element proqram is developed for calculating thermal stress in bulk single crystals during Czochralski growth. Elastic anisotropy is taken into account in this program. Thermal stress analyses of a GaAs bulk single crystal are performed in the cases of [001] and [111] pulling directions using its temperature distribution obtained from a heat conduction analysis and its material properties. The stress component and the parameter representing dislocation density are compared between the anisotropic analysis taking account of elastic anisotropy and the isotropic analysis using the Young's modulus and the Poisson's ratio in the {111} plane. Significant differences are found in their values and distribution patterns between both analyses.

View full abstract

Residual stress seems to be a main factor controlling the fracture of dissimilar materials such as metal/ceramic bonded materials. In this study, we develop a simple method to analyze the residual stress caused by the thermal mismatch of dissimilar materials, using the boundary element method. The present results present a stress singularity at the edge of the interface and therefore it can be supposed that a fracture occurs at the edge of the interface into either material. This fracture behavior is evaluated by σ θ_max criterion with the use of the obtained numerical results on the stress distribution around the edge point and discussed in comparison with the experimental results.

View full abstract

In this paper, the development of a 2-dimensional elastic contact stress analysis procedure using a hybrid method of FEM and BEM is described. FEM is valid for slender bodies and BEM is valid for massive bodies. Therefore, the hybrid method of FEM and BEM is useful to solve the contact problems of a magnetic disk drive, which is composed of thin films, substrates, and head sliders. In order to improve the accuracy of the analysis, the equilibrium condition of the BEM region is considered when the equivalent stiffness matrix is formed. Furthermore, the penalty function method is applied to relax the contact conditions. The results of example analysis of the contact problem between a magnetic disk and a slider show the validity and accuracy of this procedure.

View full abstract

This paper is concerned with a contact problem between an elastic half-space and a rigid elliptical punch subjected to translation parallel to the surface of the half-space. In such a problem, the normal contact stress within the contact region is much smaller than the tangential stress and is usually assumed to be zero for simplicity of the analysis. In this study, we treat the problem as a mixed boundary value problem in which the surface displacements are specified inside the contact region and the surface stresses are zero outside. Therefore, the normal contact stress is not zero within the contact region. The problem is formulated in the form of simultaneous integral equations with kernels involving Bessel functions. It is reduced to the inhomogeneous Hilbert problem with infinite unknown functions by using Abel transformation and Plemelj formulae. A general solution for the contact stress and the surface displacement is given. The distributions of stress and displacement are shown for the case of a flat-bottom punch and compared with those of the result when the normal contact stress is assumed to be zero.

View full abstract

Three-dimensional elastic contact problems are analyzed by the boundary element method. The boundary line of the contact area is calculated accurately by the method called "dynamic mesh". For such calculation, linear triangular elements are used. These elements are integrated analytically to estimate the singular integrals accurately. Inner stresses of contact bodies are calculated by numerical differentiation of displacements. The conclusion of this report is compared with the results of other methods which have already been reported in public. The effectiveness of "dynamic mesh" is certified by practical examples.

View full abstract

In the previous reports, the authors proposed the concept of a higher-stage micropolar continuum, and then formulated their kinematics and mechanical balance laws. In the present paper, strain tensors characterizing deformation of micropolar materials of stage-2 are newly defined on the basis of the kinematics presented in the 1st report. Some equations are derived, which express relations between material strain tensors and spactial ones, and between the strain rate tensors and the kinematical quantities (e.g., deformation rate tensor, angular velocity vector, etc.) defined in the lst report. While, new displacement vectors and tensors which are peculiar to micropolar materials of stage-2 are defined, and the relations between strains and displacements are obtained. Furthermore, linearizing the strain-displacement relations, strain tensors defined here are discussed geometrically. The strain tensors are suitable for expressing the constitutive equation of bicouple stress which is the generalized bimoment in the theory of thin walls.

View full abstract

Constitutive equations for higher-stage micropolar materials must be discussed ther-modynamically as a preliminary to the construction of concrete constitutive equations. In the present paper, averaging the lst and 2nd laws of thermodynamics is the microscopic volume elements up to stage-2, the energy balance equation and the entropy inequality for the micropolar materials of stage-2 are formulated on the basis of the results of the previous reports. Thermodynamic relation-ships between the generalized forces (stress, couple stress, bicouple stress, heat flux density) and thermodynamic functions (free energy and dissipation function) which govern the constitutive equations are derived using the Clausius-Duhem inequality obtained by coupling the energy balance equation and entropy inequality. In this way, the thermodynamic limitations which the constitutive equations must satisfy are clarified, and the response functions and their arguments which characterize the constitutive equations are determined.

View full abstract

The residual stress distribution of a ceramic-metal joint specimen was determined by both the two-and three-dimensional thermoelastoplastic stress analyses using the finite-element method (FEM) and the X-ray method. The specimen was prepared by brazing silicon-nitride and carbon-steel rectangular plates. The highest tensile stress σ_x perpendicular to the interface appeared at the corner of the silicon nitride adjacent to the interface. The maximum compressive stress σ_y parallel to the interface occurred at the center of the interface of the silicon nitride. In the steel plate, the stress distribution was reversed. The measured stress distribution was similar to that calculated from the three-dimensional FEM, but it differed remarkably from the result by the two-dimensional FEM especially at the center of the specimen. This is because the sharp stress concentration occurs and the stress distributes three-dimensionally at the interface.

View full abstract

In the previous paper, we have proposed the numerical analysis method for estimating the elastic modulus of braided composite. This analytical model is based on the faithful expression of fiber the orientation state. In this report, the analysis method to estimate the tensile strength and to simulate the fracture process of a braided composite is extended. The braided flat bar is composed of three local parts are which have different fiber weave geometres. The stress states of each part are calculated by nonlinear analysis. The tensile strength and fracture process of the braided flat bar are estimated from the numerical results of local parts of flat bar. The results predicted by numerical analysis are compared with the experimental results. The numerical results of the tensile strengths agree well with experimental results. Moreover, in the predictions of the fracture process, an agreement is found between the numerical and experimental results. Also, numerical analysis of the braided composite flat bar cutting edges is carried out. These results suggest that the continuity of fibers at the side edges is important.

View full abstract

Bearing in mind the boundary element analysis in composite materials, this study described a method that used a data base paying attention to the similarity of kernel functions. It is found that the coefficient matrices of a region can be used to compose those of another region if the shape is similar even if the numbers of divisions differ. This indicates that the coefficient matrices can be composed without a numerical integration scheme so that a drastic reduction in cputime is expected. Using the results, a new analysis system which utilizes the data base has been constructed. The advantages of this system are verified through a few examples.

View full abstract

In this study, interfacial mechanical properties of glass/epoxy composites were evaluated for some kinds of silane treatments on glass fiber with the aid of an embedded single filament tensile test that is generally employed for measurement of interfacial shear strength. However, this technique has some problems such as difficulty of specimen manufacturing and determination of critical state. A new evaluation method of the interface is proposed in this paper to overcome these problems. In this method, the fiber fracture was monitored by a video camera and image processor during the tensile test for the embedded single-filament coupon. The first fracture of the embedded filament in the fiber fracture process was used to evaluate interfacial properties. The applied stress of the embedded filament at the first fracture, σ_f1b, is calculated on the basis of the elastic model and compared with the strength of filaments s_f. The fact that s_f is smaller than σ_f1b means the inefficiency of the tensile load transfer due to the interfacial mechanical properties. The ratio of s_f to σ_f1b is named 'interfacial transmissibility'. The effect of the chemical structure of a silane coupling agent on interfacial transmissibility was investigated.

View full abstract

Supersaturated Al-Si and Al-Ge solid solutions were prepared at a high pressure of 5.4 GPa. Superconducting transition temperature T_C and low-temperature specific heat of the obtained samples were measured. It was confirmed that the dependence of Tc on the electron concentration e/a was much greater than that of other nontransition alloys. With increasing e/a the electronic-specific-heat coefficient was approximately constant within experimental errors, but in the Debye temperature, a remarkable decreasing trend was observed when compared with noble and simple alloy superconductors. The above results could explain the dependence of T_C on e/a qualitatively, using free electron approximation.

View full abstract

This paper discusses the use of a small punch (SP) testing method to determine the elastic modulus and fracture strength of the material. Numerical analyses using the FEM were carried out to calibrate a SP specimen. Three loading modes were analysed to examine the punch tip effect on the deformation and fracture behaviour. Several ceramics and ceramics/metal composites were tested using the SP method, and their mechanical properties were determined using the results of FEM analyses. It is shown that the SP method provides a reliable means for measuring the elastic modulus. Furthermore, it is demonstrated that the fracture strength determined by the SP method shows a good correlation with the bending strength obtained from the three-point bend specimen which is commonly used for the determination of the fracture strength of ceramics.

View full abstract

A new boundary element method for dynamic viscoelastic problems is suggested to improve the accuracy of numerical results at the vicinity of boundaries. The singular terms in the fundamental solutions are associated with static solutions and are separated from the dynamic solutions. The strong singularities in the fundamental solutions are weakened by the partial integration of the integral equations for internal points. The yielded equations are numerically solved using displacements and tangential differences of displacements on the boundaries in the Laplace transformed domain. The validity of this method is confirmed by the traditional analysis. The numerical results are found to exhibit good accuracy of the stresses at the vicinity of the boundaries. It is shown that the numerical accuracies with this method are improved, compared with the results with the standard method requesting more numbers of boundary elements.

View full abstract

As a basic study for the establishment of an accuracy estimation method in the finite element method, this report deals with the problems of accuracy estimation in the analyses of in-plane stress and transverse bending of thin, curved plates. From numerical experiments for uniform mesh divisions, the following relation was deduced, ε∝(h/a)^k, k&ge;1, where ε is the error of the value by the finite element method relative to the exact value and h/a is the dimensionless mesh diameter. From this relation, an accuracy estimation method, which was based on the recursive simulation method, was presented. A computer program using this accuracy estimation method was developed and applied to eight problems of various shapes. The usefulness of this accuracy estimation method was illustrated by these application results.

View full abstract

An analysis method to evaluate the sensitivity of structural dynamic response is developed for nonlinear systems in the context of an implicit time integration scheme. The formulation originates in the perturbation method and retains its advantage. Namely, the decomposed effective stiffness matrix after the iteration for each time increment is efficiently used to compute the sensitivity of any parameter in the system. Because it allows the nonlinearity including load path dependency, the proposed method can be applied to a wide range of nonlinear finite element dynamic analyses. An elastic-plastic truss structure is solved to examine the validity of the method, and satisfactory results are obtained.

View full abstract

A method is proposed for the shift synthesis of structural responses which satisfy inequality constraint conditions. The first-order form is taken to express the inequality constraint conditions with respect to design variables and slack variables introduced to cope with the inequality. The design variables are obtained by solving the inequality constraint equations by means of the generalized inverse matrix. The particular solution or the complimentary solution added to it enables us to find the design variables in the feasible domain without any iterative procedure. The validity of the proposed method is verified by the numerical results regarding the eigenvalue and mode of beam vibration based on the finite-element sensitivity analysis.

View full abstract

Acoustic emission (AE) from two types of columnar grained ice specimens, plain and sharply notched, subjected to bending load at -5°C was examined. Two different levels of thresholds, 25 mV and 125 mV, were set for the "average signal" of AE, and the signals which exceeded these levels were counted. The count of the AE signal from the sharply notched specimen was less than that from the plain specimen. The Kaiser effect on AE was observed in the reloading stage and it vanished when the pause time was over after about three minutes. The pop-in of the sharply notched specimen can be detected by AE. The ratio of σ_AES and σ_f, which express the calculated stress at the start of AE and the broken stress respectively, was nearly proportional to the value of σ_AES.

View full abstract

The unloaded photoplasticity technique can be employed in a large variety of applications such as plane stress, plane strain, axisymmetric and general three-dimensional problems. Depending on the type of application, different sets of results may be obtaind : for example distributions of plastic strain differences, principal strain directions (isoclinics), slip-line directions and elastoplastic boundaries. The compression of a short cylindrical strut is the most typical and essential process in plastic forming. It is universally significant to study this kind of deformation, and to apply the obtained information to complex deformation analysis. This paper presents how an experimental method of photoplastic strain freezing, which was proposed by the authors, was used to obtain the three-dimensional strain distribution for compressed cylinders. The results were intended to present a very good accuracy of photoplastic strain freezing solutions to compare overall displacements calculated by integration of strains determined by photoplasticity with displacements measured by a micrometer directly over the models or slices.

View full abstract

Roughening of the free surface of products is one of the most impportant problems in metal forming processes. In the present paper, spectrum analysis was carried out for the surface wave data of iron specimens during compression and tension. The maximum entropy method is adopted in the spectrum analysis using the personal computer. The effects of the applied strain and the grain diameter on the wavelength of the surface roughness were examined. The influence of the cut-off value in the surface roughness measurement was also studied. The results show that the primary wavelength of the surface roughness is independent of the strain, though the surface roughness and its standard deviation increase with the applied strain. A model is proposed to explain the experimental data, in which small waves of the surface roughness disappear during compression.

View full abstract

Internal friction of carbon steels at about 17kHz is studied by using a sensitive electromagnetic amplitude sensor attached to an ultrasonic fatigue testing machine under the strain amplitude within 10^-3. The influence of applied magnetic field and the effect of heat treatment on the internal friction of carbon steels are as follows. The peak of the internal friction Q^-1 found in carbon steels at the strain amplitude of about 3×10^-4 is decreased with increasing applied magnetic field. The dependence of resonance frequency on the strain amplitude decreases with increasing of the applied magnetic field. The peak of Q^-1 is explained by the magnetic property by the motion of the magnetic domain wall.

View full abstract

In this paper, dynamic analyses of normal walking of men are performed by measuring the foot pressure distributions. The measuring system is a sensing system of piezoelectric ceramics that was previously proposed by the authors. The sensor part of this system has 16×32 measuring points and the measuring area is 174mm×352mm, sufficient to measure foot pressure distributions of men. A microcomputer system is also used to analyze the numerous measured datums, and contour lines of foot pressure distribution and its related information are drawn on the display. By using this system, the dynamic change in foot pressure of standing and walking typical men are measured under the conditions of open or closed eyes, load or noload, and various walking speeds. From the results, the dynamic behavior of normal walking of men is discussed.

View full abstract

The effect of steady axial load on internal remodeling of a hollow cylindrical bone tissue, which is a model of diaphysis, is investigated. The coupling between the change of the volume fraction and the stress distribution is considered. The internal remodeling theory developed by Cowin and Van Buskirk is employed, and the displacement field is represented, including the second-order terms. The geverning equations are solved numerically, and the changes of the volume fraction and stress distribution due to internal remodeling are clarified.

View full abstract

A minimum compliance design technique of plate thickness variance for general shell structures assembled by linear-or curved-shell elements is proposed. In this technique, the shell structures are discretized into isoparametric finite shell elements, and nodal point thicknesses are taken as design variables. Thickness distributions on subregions of each element are interpolated by quadratic shape function using the nodal values. Therefore, one can obtain optimal design in which the thickness varys continuously over the whole region of the shell structure. A mathematical programming method is employed to solve the optimization problems using design sensitivity analysis. As a check on effectiveness, the proposed method is applied to a classical cantilever beam problem and a plate clamped at two corners. Furthermore, the problem of maximizing the torsional rigidity of the motorcycle body, which consists of shell elements, is solved.

View full abstract

It has come to be an essential procedure for us to perform the optimal shape design of root/groove structure of turbine blades in order to improve its structural reliability. However, various complexities, such as multiobjective characteristics, nonlinear formulation and complex shapes have made it difficult to solve this type of problem. In this paper, an effective method of solving this type of problem is proposed by combining sensitivity analysis based on finite element method with linear programming. In the present procedure, alternative iterations of design sensitivity analysis and linear programming analysis are performed to obtain a convergent solution. The additional constraints are also used at the stress-concentration points in order to reduce all peak stresses.

View full abstract