In this paper, a method for determination of the depth and location of a transverse surface crack in a beam is proposed. The study, which is treated as an inverse problem, was performed using a error-back propagation neural network approach. In the analysis, the crack is simulated by an equivalent rotational spring, connecting the two segments of the beam. Analysis of this approximate model leads to a frequency equation which relates the natural frequencies to the crack location and depth. The equation is then applied to the inverse problem-identification of the crack location and depth from frequency measurement results. First three or four natural frequencies, obtained by solving the non-linear frequency equation, are adopted as training data for the neural network. After the training stage has been completed, the crack parameters are estimated using the output of the trained network. The simulation results indicate that the proposed method is valid and practical.

View full abstract

The bonded dissimilar plates are often used for chemical apparatus and pressure vessels. The failures of those apparatus and vessels can lead to serious accidents. In the present paper, an explosion clad plate composed of copper and mild steel is dealt with as a typical example of the bonded dissimilar plate. Tensile tests are carried out by using rectangular plate specimens extracted from the clad plate. Each specimen has an artificial through-the-thickness edge crack close and perpendicular to the explosive interface. The effects of the material inhomogeneity, the residual stress and the hardened zone on the deformation near a crack are examined by the experiment and the finite element analysis. The lower-strength material ahead of a crack and the tensile residual stress near the bonded interface increase the crack opening displacement. The higher-strength material ahead of a crack, the compressive residual stress and the hardened zone near the bonded interface decrease the crack opening displacement.

View full abstract

A method is proposed for simplified nondestructive evaluation of multiple cracks by means of measurement of magnetic field produced in the air by d. c. current flow in a material. First approximate equation which relates the change in magnetic flux density due to a 3-D surface crack to the crack length and crack depth is derived based on the results of the numerical analysis. The equation is the basis of developing a method for simplified nondestructive evaluation of 3-D multiple cracks. Next the equation for multiple cracks is expressed as the sum of the magnetic flux density for a plate without crack, increment of the magnetic flux density due to the respective cracks which is calculated by assuming that the plate contains one of these cracks and a term representing the interaction between these cracks. Finally the inverse problem to evaluate the crack depths is analyzed by comparing the magnetic flux density calculated in this way with the measurement. It is shown that multiple cracks are sized accurately by the present method.

View full abstract

Reversed plane bending fatigue tests were conducted on SiC whisker reinforced aluminum composite specimens with shallow surface notch. Initation and growth behavior of small surface fatigue cracks were investigated by means of a plastic replica technique. Fatigue strength of composite was higher than that of unreinforced alloy because SiC whiskers restricted cyclic slip deformation of matrix. which resulted in the delay of fatigue crack initiation. the initiation of small fatigue crack in composite was found characterized by the interface cracking between matrix and SiC whiskers, where the lump of SiC whiskers was composed in a manufacturing process. The many small cracks initiated as the lumps of SiC whiskers distributed around specimen surface irrespective of the stress revel, while several cracks grew simultaneously in the tests with high stress amplitude and one main crack propagated in those with low stress one. Microscopically, the coalescence of main crack and micro cracks was observed frequently. The difference of small fatigue crack growth rate between composite and unreinforced alloy was small, which implied SiC whisker was not resistant to crack growth but to crack initiation. The crack growth rate of small cracks was found faster than that of long cracks.

View full abstract

An elastic-plastic finite element analysis is presented for the axisymmetric problem of frictionless indentation of an elastic sphere into a half-space with a ceramic coating. The analysis is performed for a variety of combinations of material properties, i. e. elastic modulus, strain-hardening exponent and yield stress, for both the coating and the substrate. Results for stresses on the surface and residual stresses after unloading are presented. The effect of elastic moduli, strain-hardening exponents and yield stresses of the coating and substrate on the stresses is examined in detail. Finally, the ring crack initiation on the ceramic coating surface is discussed, emphasizing especially the critical load for the ring crack initiation.

View full abstract

Mode I fracture behavior of rubber toughened PMMA (RTPMMA) was studied over five decades of displacement rates (10^-4-13 m/s) using a high-speed tensile testing machine. Two kinds of RTPMMA, extruded RTPMMA containing 20 wt.% rubber particles (EM20) and pressed RTPMMA containing 40 wt.% rubber particles (PM40), exhibited different rate dependences of the maximum stress intensity factor K_1max and the maximum energy release rate G_1max. K_1max and G_1max of EM20 decreased rapidly at high rates (1-11 m/s). This suggests that the toughening effect due to rubber blending was dramatically reduced at the high rates. On the other hand, PM40 maintained high K_1max and G_1max values at all rates tested in this study. Thus, PM40 maintained the toughening effect even at high rates up to 13 m/s. For comparison, unmodified PMMA was also examined, and the results showed that K_1maxof neat PMMA increased with the increase in displacement rates and G_1max remained constant as the rate increased. Discussion is given on the rate-dependent K_1max and G_1max values, taking (K_1max)²/G_1maxinto account. The fracture surface morphology of the specimens is also discussed.

View full abstract

Rotating bending fatigue tests were carried out using specimens of aluminum-oxide-dispersion-strengthened copper (ODSC) alloy in order to investigate the crack initiation and its growth behavior. The ODSC alloy has excellent properties not only in the static strength but also in the fatigue strength. Most of the fatigue life is occupied by the growth of a small crack and the crack growth rate can be evaluated using the small crack growth law. Therefore, the superior fatigue strength is caused by the increase in the crack growth resistance due to the dispersed aluminum oxide.

View full abstract

Torsional fatigue tests and creep tests were carried out on a 60Sn-40Pb solder. In both tests three temperatures (303∼333K) were used, and in the fatigue test five cycling frequencies (0.001∼5.0 Hz) were used. An evaluation method for the solder fatigue life under relatively high temperature or low cycling frequency was proposed from the test results, noting that the characteristics of fatigue under the above mentioned conditions were similar to the creep characteristics. Furthermore, this evaluation method was applied to the estimation of the crack initiation life of through-hole-type joints. By comparing the estimated values with the experimental ones, it was clarified that this method was valid for the assessment of the crack initiation life of solder joints.

View full abstract

In this paper, out-of-phase thermo-mecahnical fatigue tests are performed on the aluminum alloy that is used for engines of automobiles. The tests are conducted with three constant total strain amplitudes, three periods of cycle of loading and three temperature renges. These test conditions are chosen taking account of actual use of the alloy in engines. First, the detailed observation on the inelastic deformation due to thermo-mechanical loading is done, and we show the effect of strain amplitude, period of cycle of loading and temperature range on the deformation of the alloy. Second, the method to estimate number of cycle to fatigue failure is discussed. As a result, the relationship between the plastic work density per unit time, which is based on the concept of the plastic strain energy density, and number of cycle to fatigue failure is expressed by a formula irrespective of the test condition. Then, it is found that the plastic work density per unit time has an advantage to estimate number of cycle to fatigue failure due to the thermo-mechanical loading.

View full abstract

Large deflection occurring in flexible materials is of both analytical and technological interest. This report deals with the postbuckling behavior of a slender rod. The spring-hinged slender rod is subjected to compressive forces at its clamped ends. With contact between the two different parts of the rod, several analytical formulae are derived in terms of elliptic integrals for arc length, deflection, distance between the two ends, curvature, bending moment and bending stress. Moreover, an experiment is carried out to confirm the applicability of the proposed theory. The experimental results agree well with the theoretical ones.

View full abstract

In this study, a theoretical treatment of an elastic behavior for a nonhomogeneous medium is developed. As one nonhomogeneous medium, a semi-infinite body subjected to the action of a rigid punch on its surface is considered. It is assumed for the nonhomogeneous material property of the semi-infinite body that the shear modulus of elasticity G varies with the variable of the axial coordinate z by an arbitrary power product form. Making use of a fundamental equation system for such a nonhomogeneous medium, an axisymmetric problem for such a singular stress field is developed theoretically. Numerical calculations are carried out for several cases taking into account the variations of the nonhomogeneous parameter, and the numerical results for displacements, stresses and the stress intensity factor at the edge of the rigid punch are shown graphically. Thereafter, the influences of the nonhomogeneous material property on the elastic behaviors such as displacements, stresses and the stress intensity factor are examined.

View full abstract

In the present paper, we summarize the basic equations for a cylindrically anisotropic elastic medium of two-dimensional in-plane problems and out-of-plane shear problems, and show the fundamental closed-form solutions of displacement and stress components by using stress functions. The results contain some new multivalued solutions for in-plane problems and can be widely ultilized for the fields of solid mechanics, strength of materials and theory of structures. Further, we show general solutions for a circular hollow ring with given stress boundary conditions of Fourier trigonometric series.

View full abstract

This paper presents an analysis of general temperature distributions in two confocal elliptic boundaries of an isotropic elastic medium. The methodology uses the complex stress functions Φ and ⩛. The complex function Φ is found by solving the remaining boundary condition. Φ is proposed in the form of a Laurent series with coefficients determined using matrix representations. Using the analytic continuation through one of the boundaries. ⩛ can be written in terms of the function Φ. Several numerical results are given by graphical representation.

View full abstract

The main break down modes of constant velocity joints are the fatigue break at the valley portion of the bellows and the wear break caused by friction in the bellows. We measure the stress which ar the causes of break down by the photoelastic method. Maximum stress is located the surface of the valley portion of the bellows, and it's location agree that of the fatigue break. We analyzed the stress, strain and the contact pressure in the bellows by finite element method, and proved that the FEM results is accurate by comparison between experimental results and FEM results. The Ogden model which is obtained from the data of the stress and strain through the test of the strip biaxial extension is effective in the nonlinear problem of the rubber.

View full abstract

The stress in a human temporomandibular joint (TMJ) during clenching was analyzed using a 2-dimensional model of a TMJ. The finite element method was employed and the contact between soft tissue and bone including articular cartilage was taken into consideration. The stress perpendicular to the surface in cortical bone is compressive on the anterior sides of the condyle and the glenoid fossa. The maximum and minimum principal stresses in the articular cartilage are nearly equal and the equivalent stress is relatively small in the region between the parietal and anterior directions. The equivalent stress is distributed in the articular disk when the condyle is displaced anteriorly in the TMJ. On the other hand, concentration of equivalent stress is observed in the disk when the condyle is displaced posteriorly.

View full abstract

The diagonally split connecting rod is widely used for medium-speed diesel engines for the purpose of reducing its width to easily perform assembling and disassembling. Thus, shearing forces are occurred along the interface between the shank and the cap of connecting rod. Consequently, so called serrations, which have similar shapes of screw threads, are machined at the mating interfaces of shank and cap to support the shearing force. It is sometimes reported that failures are observed there because of stress concentrations, fatigue or fretting around the serration. In this paper, the mechanical behaviors of the big end are analyzed by FEM as a plain strain contact problem, where selective reduced integration (SRI) scheme is introduced for integrating element matrices of bolt model for better accuracy, and the effects of both ignition and inertia forces are discussed.

View full abstract

The authors developed a new type of horn named "simple round bar horn"which is based on a different magnification principle from a conventional one. The horn is constructed by connecting simple round bars of different specific acoustic impedance, which gives uniform cross section along its axis. However, two different materials must be joined firmly to endure any vibration stress for practical applications. In this paper, we have attemped to use friction-welded joints consisting of SUS 304 stainless steel and A6061 aluminum alloy as the simple round bar horn. The characteristics of the input-output displacement amplitude, of the elevated temperature and of the frequency are investigated experimentally on the friction-welded simple round bar horn and on the conventional horns of step and exponential type, consisting of SUS 304 and A6061 respectively. Furthermore, ultrasonic fatigue tests have been performed on the friction-welded simple round bar horns and A6061 monometallic bars. The results shows promising possibilities for this new type of horn.

View full abstract

The need for plastic composites increases with the increasing use of electronic instruments such as robots, personal computers and cellular phones come into wider use. Carbon fiber-reinforced thermoplastic, CFRTP, is one plastic composite used for electronic instruments. It is favored as a casing for multimedia instruments because it has good deformation and recycling capabilities. It is important to study the shielding effectiveness, SE, of CFRTP against electromagnetic radiation. In the present paper, the shielding effectiveness was experimentally investigated in an electronic shielding room for the CFRTP of which the resin materials were varied in PC, PP, PEI, PMMA and PA. The effects of compressive and impact damage on the shielding effectiveness were studied when the carbon fiber orientation and the space between fiber bundles were changed. The shielding effectiveness strongly depended on the orientation of the carbon fiber. It also decreased when the space between the fiber bundles was increased but was scarcely influenced by the resin materials. When the CFRTP formed a compressive trace on the plate surface by compressive load, the shielding effectiveness increased when the compressive load was increased, and thus depended on the shape of compressive trace.

View full abstract

The stress of joints combining the bonding force of adhesives with the clamping force of bolts is analyzed using axisymmetrical theory of elasticity and FEM, when an internal pressure is applied to the combination joint in which two pipe flanges are clamped together by bolts and nuts with an initial claming force after being joined by anaerobic sealant. The anaerobic sealant is applied at the interfaces partially instead of formed gaskets in pipe flange connections. In the numerical calculations, the effects of the bonding area and position and Youngs' modulus of sealants (epoxy adhesive) on the contact stress at the interfaces are clarified. In addition, using the interface stress distribution, the strength of the combination joint is predicted. Experiments were performed and the analytical results are consistent with the experimental results concerning the variation in axial bolt force and the strength of combination joints. It is found that the joint strength changes as the bolt pitch circle diameter, the bonding position and area, the initial clamping force and the adhesive material properties change. Furthermore, it is seen that the sealing performance of such combination joints in which the interfaces are bonded partially is improved over that of pipe flange connections with metallic gaskets.

View full abstract

For finite element analysis of hyperelasticity such as rubber material, the mixed method and the selective integration method have been used widely. These methods can optimize volumetric constraint and degree of freedom of stiffness matrix. In the mixed method, the mean stress and the deviatoric strain are separeted and formulated. In the selective integration method, the reduced and full integration rule is applied to the terms including the mean stress and the other. Laminated rubber bearings consisting of thin rubber films and steel plates are used as seismic isolation device to reduce horizontal earthquake response of structure. Recently, laminated rubber bearings using thick rubber films is considered to use as seismic isolation device to reduce both horizontal and vertical response. In this paper, formulation of a selective integration method in rate form is made and results of three dimensional finite element analysis of rubber bearings are shown along with the comparison with experimental results.

View full abstract

This paper presents r-and hr-adaptive boundary element methods. The error analysis is done by the sample point error estimation scheme. Then, the error indicator is defined by the computational error. Mesh refinement is carried out so that the error indicator distributes uniformly on whole boundary. The present method is applied to the two-dimensional potential problem in order to confirm its validity.

View full abstract

A new formulation is presented in this paper for the boundary element analysis of inhomogeneous-nonlinear potential-type problems. The formulation aims to transform the domain integral relevant to the inhomogeneous-nonlinear term to the corresponding boundary integral, different from the well-populated schemes for the purpose, such as the Dual Reciprocity and Multiple Reciprocity Methods. The inhomogeneous-nonlinear term is first approximated with a polynomial in terms of the space coordinates with unknown coefficients. Integral equations on the selected points (computing points) on the boundary as well as inside domain are employed to determine the abovementioned unknow coefficients using the least square method. Numbers of computing points affect the accuracy of the result, which is discussed by some numerical examples in two-dimensional space.

View full abstract

Microscopic sintering behavior in compacts formed by pressing ceramic granules is examined to develop a constitutive model for analysis of macroscopic deformation. Spray-dried alumina granules are compacted by CIPing and fired at various temperatures. The internal structure composed of fractured and unfractured granules, and large crack-like cavities caused by their inhomogeneous shrinkage during sintering are observed. The macroscopic sintering rate becomes slow because of the formation of the large cavities. A basic constitutive equation for sintering of ceramic powder compacts is proposed, with the grain boundary diffusion and the grain growth taken into account. The relationship between the change in the microstructure and the shrinkage rate of the powder compacts is modelled using the basic constitutive equation. The constitutive model is applied to the finite element analysis to predict the shape change of the compacts during sintering. The calculated results show good agreement with the experimental results.

View full abstract

The asymptotic homogenization method is applied to composites whose microstructural configurations reveal irregularity. Although the method usually requires periodic boundary conditions for microstructures, i. e., volume elements (RVE), actual composite materials do not have geometrical periodicity when arbitrary regions are taken as RVEs. After reviewing the theory of the mathematical homogenization method, we show that the homogenization for periodic media be appled to arbitrary statistically homogeneous media. Using several sizes of unit cells in numerical analyses, the convergence trends of macroscopic and microscopic variables are considered, and the theoretical results are justified. Also, we try to show how to determine the RVE as a unit cell for a particular type of composites. For this purpose, the systematic modeling technique utilizing digitized images is extensively utilized to properly model the irregular material configurations for the homogenization analyses and to take into account the actual microstructural geometry in evaluating the micro and macroscopic variables. Furthermore, we briefly discuss the image-based evaluation of the results in the linear and nonlinear homogenization.

View full abstract

A systematic modeling technique of utilizing digitized images for micromechanical analysis in the asymptotic homogenization method is presented. By use of digital images, irregular material configurations are accurately modeled for the material pre/post-processing and therefore the specific microstructural geometry is taken into account in evaluation of the micro and macroscopic variables. Each picture element (pixel) in the images is identified with a regular and fixed-size finite element, and a three-dimensional (3 D) unit cell model can be generated by stacking of two-dimensional (2 D) images into a 3 D image consisting of volume picture elements (voxels). After the concrete modeling procedure is presented, its applicability to the analysis and advantages from a computational point of view are discussed. With a very primitive unit cell model that is generated using digital images, the numerical accuracy in homogenization analyses is investigated and some practical aspects of its usage are identified. Furthermore, several applications are presented in order to show the rigorousness and the flexibility of this modeling technique in terms of the pixel (or voxel) information and image manipulations.

View full abstract

By performing the thermomechanical loadings with tension and heating-cooling for a TiNi shape-memory alloy wire, the two-way deformation properties due to the R-phase transformation were investigated. The results are summarized as follows. The two-way strain due to the R-phase transformation occurs based on internal stress which is induced by cyclic deformation with the martensitic transformation. If temperature is high and strain is large for cyclic SME and PE with the martensitic transformation, large two-way strain appears. The two-way strain of 0.6% appears in several cycles for PE in the completion region of the martensitic transformation. The two-way strain changes little under cyclic motion.

View full abstract

In order to describe the thermomechanical properties in shape memory polymer of polyurethane series, a thermomechanical constitute model is developed by modifying a standard linear viscoelastic model. The model involves a slip element due to internal friction and takes account of thermal expansion. In order to describe the variation in mechanical properties due to glass transition. coefficients in the model are expressed by a single exponential function of temperature. Several kinds of thermomechanical tests are carried out. The proposed theory expresses well the thermomechanical properties of the material, such as shape fixity. shape recovery and recovery stress. The proposed model is useful for design of shape-memory polymer elements, in which the amount of recovery deformation, the tightening force and the working start and completion temperature are specified.

View full abstract

The condition necessary for the multiplicity in deformation rate was give by Hill. The condition is Δε_uΔs_u=0, where s_u denotes the nominal stress rate, and Δ shows the difference between any two multiple solutions. However, many proposed criteria of plastic instability in published literature did not take into consideration this necessary condition. They were based on phenomenological aspects, for example, simultaneous stationarity of loads, that is, s_i=0, which corresponds to the special case of the necessary condition. We discovered that the necessary condition can make the equivalent strain rate difference Δε_g indefinite at the critical deformation. The sufficient condition for the multiplicity is deduced from the requirement which makes the equivalent strain rate ε_g indefinite. This is given by ε_us_u=0. In the plane stress state, the sequential onset of diffuse necking (by Swift) and localized necking (by Hill) is observed, and they have been thought to be phenomena based on plastic instability. Detailed examination of their occurrences reveals that the former satisfies the instability condition but the latter does not. Therefore, in localized necking, multiplicity does not arise, and localized necking is merely the plastic deformation confined into a thin layer with a rigid region on its both sides.

View full abstract

A fundamental study was carried out for applying the method of the X-ray stress measurement to a composite material consisted of high chromium steel and titanium nitride manufactured by the powder metallurgy. The materials used in this study has been developed for the valve seat insert, which is a part of diesel engines in automobiles, because of its high wear-resistance and heatresistance. The influence of the volume fraction of titanium nitride on the composite X-ray elastic constants and each phase stresses was investigated. X-ray diffraction data obtained from both phases were compared to the Eshelby/Mori-Tanaka model, which was developed by Lin and Mura et al based on Eshelby approach and Mori-Tanaka theorem. It was found that the relation between the composite X-ray elastic constants and the volume fraction of the second phase agrees well with the theoretical estimation.

View full abstract

Cyclic plastic deformation subsequent to pre-strain of SUS 304 stainless steel is experimentally investigated at low temperature in liquid nitrogen under torsional, tensile and compressive loading conditions. Thin-walled tubular specimens given pre-strain were subjected to cyclic loading under stress-controlled conditions. The pre-strain is conducted at 77 K, and subsequent cyclic lading under constant strain amplitude is given at 77 K. The cyclic stress-strain curves are saturated by increasing number of cyclic. At small number of cyclic, cyclic plastic deformation depends on the pre-strain direction. The pre-strain perpendicular to cyclic loading directions results in large hardening at the same accumulated plastic strain value comparing with the pre-strains of the same direction as cyclic loading directions. The directional effect of pre-strain on cyclic loading with increasing number of cycles becomes small.

View full abstract

The objective of the present study is to exploit the feasibility of piezoelectric polymer film sensors to evaluate in-plane strain distributions of structural components. In contrast to ordinary use of piezoelectric films as strain and strain-rate sensors detecting electric charges and currents through electrodes, the proposed strategy consists of directly measuring the distribution of electric potentials induced in the piezoelectric film mounted on the surface of a structural component in loading. The stress and strain distributions are determined from the measured potentials taking into account the piezoelectric constitutive law of film materials. As an illustrative example, thin films of polarized polyvinylidene fluoride (PVDF) are mounted on a holed elastic plate subjected to in-plane loading, and the induced potential distributions are measured on the film surface by an electrostatic voltmeter of non-contact type. It is demonstrated that the determined strain distributions are in fair conformity with those analytically predicted.

View full abstract

In recent years, steam injection wells, which are enhanced oil recovery wells, have been developed. Since well casings installed in steam injection well are subjected to excessive tensile loading during oil production, problems such as casing collapse, connection fracture and connection jump out (pull out) can occur during service life. However, the joint strengths for fracture and jump out have been calculated for cases of axial loading alone. Therefore, no formulas for estimation of the strengths under combined external pressure and axial tensile loading are available. We conducted jump out tests under conditions of biaxial loading with high axial stress close to the yield strength of the casing material and derived an empirical formula for estimation of the jump out pressure by regression analysis. This formula can be used for accurate estimation of jump out pressures under conditions of biaxial loading, since jump out pressures calculated using this formula agree well with ones determined experimentally.

View full abstract

Design problems of a beam with a maximum eigenfrequency of an arbitrary prescribed order are described. Two types of optimization problems are formulated and solved. The first is tansverse shape optimization and the second optimization of the mixture rate distribution. In the second problem, it is assumed that the beam is made from two materials and has a uniform transverse shape. The characteristic equation with respect to the eigenfrequency is derived from the well-known Hamilton's principle and is discretized by the finite element method. For solving the optimization problems, the constructive algorithm proposed by the authors is used. The optimal transverse shape and the optimal mixture rate distribution along the beam axis are given as a piecewise linear function.

View full abstract

In this paper, a new technique for optimum rib layout in plate and shell structures is presented using the genetic algorithm. In the technique, the design model is discretized into bending plate elements. Element thicknesses such as t_plate and t_rlb for plate and rib elements correspond to GA genes 0 and 1, respectively. The chromosomes of GA are arranged in a matrix and have block crossover process between the blocks of 3×3 genes. To avoid rib element concentration, this block is subjected to several constraints. According to the strain energy density of elements, a local rule is formulated. Using this local rule, manipulation among genes can progress and convergence is rapid while the reliability of the GA is simultaneously improved. In this technique, the rib elements can be placed in optional positions. According to the connection of rib elements, optimal stiffener can be obtained without any special assumptions.

View full abstract

The estimation of surface defect shape has been demonstrated by photoacoustic (PA) microscope using specimen with an artificially made surface defect. The shape of the defect introduced for measurement is arc with a number of aspect ratio. Furthermore, in order to establish PA signal dependence on defect depth, a simulated defect the depth of which is linearly increasing is fabricated on a metal plate by mechanical processing. The results show that imaging of defects by PA microscopy enable precise determination of both location of edges and configuration of the defect.

View full abstract