A cyclic fatigue mechanism based upon cyclic degradation of the bridging effect due to cyclic load has been verified through extensive experimental results obtained by the authors. A fretting wear model for cyclic fatigue has been developed and compared with the experimental results. As the results, the model was proven to be excellent. Cyclic degradation of the bridging effect ΔK_s is found to be as follows: ΔK_s=1.54{(1-R)-0.08}log[10^-4/(da/dt)_eq]. R is stress ratio and (da/dt)_eq is equivalent crack propagation rate. This means that ΔK_s is proportional to the amplitude of cyclic load and to log N, where N is the number of stress cycles. This cyclic fatigue mechanism can successfully explain the main cyclic fatigue behaviour. Cycle-dependent crack propagation at low K_Imax results because the wear process is the rate controling process at bridging sites.

View full abstract

In this study, we investigated the dynamic mechanical properties of polycarbonate materials, noncrystal polymeric materials in the low-cycle-fatigue region. The fundamental differences between metal and polycarbonate materials are in their stress-strain hysteresis loops. We payed our attention in the plastic energy, which is a common parameter of metal and polycarbonate materials, by measuring the area of the stress-strain hysteresis loop under constant strain control. As a result, we found that the expected fatigue failure takes place when the cumulative energy observed reaches a certain energy value Wf.

View full abstract

Polymers are used in various industrial fields for machine parts and structural members. Because of advances in industry, the usage conditions of polymers have undergone various changes, and this has intensified the problems of fatigue and environmental fractures. In this research, an experiment on low cycle fatigue fracture of polycarbonate was performed. Local strain in the vicinity of the notch root and crack tip was measured in real time by means of the fine grid method. The relationships between local strain, crack initiation from the notch and crack propagation have been studied. Finally a method for more precise life estimation is suggested in this paper.

View full abstract

In order to estimate and improve the fatigue reliability of spheroidal graphite iron (SGI) with high tension and toughness, the influence of the matrix structure on the fatigue limit was examined through fractography, with a special focus on austempered ductile iron (ADI) with the bainitic ferrite-retained austenite dual structure. Under the fatigue test of tension-tension loading, R=0.05, it was found that ADI is superior to the SGI with all-pearlite matrix in terms of the fatigue limit and also the ratio of fatigue limit/tensile strength. The fractography showed the fatigue crack initiated at the microporosity in both of the above mentioned SGIs with the matrix structure of high tension. Therefore, it could be understood that bainitic ferrite-retained austenite dual structures had lower notch sensitivity to the microporosity of casting defects than the full pearlite matrix.

View full abstract

A Monte Carlo simulation of corrosion fatigue life distribution has been conducted on low-alloy steel, SNCM439. In a simulation, it was assumed that fatigue life consisted of the growths of corrosion pits and fatigue cracks. The parameters used in the simulation were obtained from fatigue tests at 250 MPa. The simulated results were in good agreement with the experimental results at 250 MPa, but not with those at 100 and 400 MPa. The cause of poor agreement at 100 and 400 MPa was attributed to the stress level dependence of the growth characteristics of corrosion pits and fatigue cracks. The same simulation was applied to the results at 100 MPa using the parameters obtained from the fatigue tests at this stress level. Consequently, the experimental distribution was accurately represented by the simulation. Therefore, it was concluded that the distributions of corrosion fatigue life could be predicted by a Monte Carlo simulation taking into account the scatter in the growth processes of corrosion pits and fatigue cracks.

View full abstract

A Study has been made to clarify the effect of gas-nitriding on the corrosion resistance and the fatigue properties of pure titanium in a hydrochloric environment. The results are summarized as follows : ( 1 ) most of the fatigue life of nitrided pure titanium is spent in the period of the destruction of the TiN layer associated with pit formation due to hydrochloric acid attack ; ( 2 ) the fatigue strength of annealed pure titanium is greatly reduced in a hydrochloric environment because a fatigue crack initiation along the grain boundary is promoted by hydrochloric acid attack.

View full abstract

We carried out internal pressure tests on a large number of thin wall cylinders of austempered ductile iron (ADI), and investigated the influence of the casting defects on fatigue life distributions. As a result, it was observed that the fatigue life distributions at low-pressure levels were divided into two groups : a short-life group below 1×10⁶ cycles and a long-life group fractured at near 1×10⁷ cycles. After the SEM observation of the fracture surface around the casting defects, the short-life group showed significant roughness, and the long-life group tended to be flat ; it was supposed that the fracture mechanisms of the two were different. Specifically, in the short-life group, a proportional relationship was observed between the initial stress intensity factor of the casting defect and the fatigue life on the double log scale. It is apparent that crack propagation of the short-life group was caused under the Stage 2 process in which the Paris law was effective.

View full abstract

Fatigue crack growth is a major factor that must be considered in the design and life prediction of fatigue-critical structures, yet recent studies suggest that much of fatigue crack growth data exhibit a large amount of statistical variability and deterministic models do not adequately describe the crack propagation behavior. Thus, the need to use probabilistic methods to predict fatigue crack growth becomes evident. In this paper, focus is centered on the formulation of a stochastic fatigue crack growth model. With the use of the diffusion Markov process theory, a partial differential equation called the Fokker-Planck equation which satisfies the probability density of the random time to reach a given crack size is introduced. Then, the equation is numerically solved by the finite differential method, and the statistical properties of crack growth fatigue life are evaluated.

View full abstract

In order to estimate the fracture toughness of a ductile bimaterial with an interface crack under mixed mode loading, an elastic-plastic finite element calculation based on finite strain theory is carried out under the assumption of small scale yielding and plane strain condition. The modified Gurson's constitutive equation and the element vanish technique introduced by Tvergaard et al. are used to simulate the crack growth due to ductile fracture. The results shows that the values of the fracture toughness of bimaterial are not only much lower than those of homogeneous material, but also vary greatly with mixed mode ratio and take the minimum value at a certain critical ratio.

View full abstract

When the tensile stress is applied to the material, the activation energy for the generation of submicrocrack decreases and the submicrocracks are thermally generated at the crack tip. The crack propagates very slowly, when the stress intensity factor of the crack is smaller than the fracture toughness of the material. Based on the postulate that the crack velocity in this regime is determined by the behavior of the submicrocrack generated at the tip of the crack, we have analyzed the experiment performed by Regel and Leksovskii in 1970. Assuming the density distribution function n(r, t) of the submicrocrack at the crack tip, we find that the stress and temperature dependence of the crack velocity agress qualitatively with their results.

View full abstract

The effect of interfacial peeling of nonmetallic inclusion on fatigue crack initiation was studied by applying a finite element method (FEM) contact problem. In this study, the interfacial stress between an inclusion and matrix was evaluated by the bond element method. It was determined that the stress concentration near the inclusion-matrix interface increased as a result of interfacial peeling. In addition, the local stress amplitude was found to increase and the fatigue process was shown to be promoted by the interfacial peeling in the fatigue incubation period.

View full abstract

The growth behavior of intralaminar fatigue cracks in a unidirectional carbon/epoxy laminate, Toray T300/#2500, was studied under mixed-mode loading with a ratio of mode II stress intensity factor (SIF) to mode I SIF equal to 2.50. Under a constant value of the applied SIF range, ΔK_Iap, the crack propagation rate decreased with the crack extension because of crack-tip shielding by crack bridging. For each value of ΔK_Iap, the relationship between the crack propagation rate, da/dN, and the crack extension, Δa, was expressed by the following equation: log(da/dN)=log V₀-Δa log V₁, where V₀ and V₁ are functions of the applied SIF value. The value of V₀ which is the crack propagation rate without crack bridging was approximated by a power law function of the crack-tip value of the SIF range, ΔK_Itip. On the basis of the power law relationship, a method for evaluating the shielding SIF range, ΔK_Ibridge, was proposed. ΔK_Ibridge monotonically increased with the crack extension. Futhermore, ΔK_Ibridge for a small ΔK_Iap value was higher than that for a large ΔK_Iap value when compared to the same amount of crack extension.

View full abstract

This paper presents an application of fracture mechanics to the evaluation of strength of an interface crack between dissimilar materials. Special emphasis is given to mixed mode fracture criteria. The stress intensity factor is one of the basic fracture mechanics parameters, but there exist few applications of the stress intensity factor to fracture criteria of an interface crack. In this study, a concept of the stress intensity factor for an interface crack is discussed in detail, and various types of specimens are tested experimentally for investigating the mixed mode fracture toughness of an interface crack. The specimens are acrylic resin-epoxy resin and aluminum-epoxy resin joints. The mixed mode stress intensity factors for various specimens are analyzed by using BEM and M_I-integral method, an extended version of the J-integral. Two types of the mixed mode fracture criteria of an interface crack are derived. One is characterized by the total energy release rate G for acrylic resih-epoxy resin joints, and the other is characterized by the stress intensity factors K_I and K_II for aluminum-epoxy resin joints.

View full abstract

In the present study, crack extension behavior and fracture toughness of ductile cast iron are examined by three-point bending tests with various detection methods of crack initation under static and dynamic loading conditions. Loading on specimens is interrupted at various displacement points, and fracture surfaces of the specimens are observed by scanning electron microscopy (SEM). Each method is effective in detecting the strict crack initiation at the middle of specimen thickness under static and dynamic loading conditions. J (mid) is proposed to define the fracture toughness of ductile cast iron. Crack tip opening displacement(CTOD) under the dynamic loading condition is smaller according to the results of calculation and SEM observation. It is considered that crack initiation under the dynamic loading condition is accelerated since the plastic deformation cannot follow up deflection of the specimen.

View full abstract

The microstructures of stainless steel weld metals which had absorbed nitrogen were analyzed quantitatively by means of a computer-aided image analyzer. Then, the relationships between quantitative microstructural factors and the low-temperature fracture behaviors were investigated. Generally, the δ ferrite which is included in the stainless steel weld metal causes cleavage fracture at low temperatures and brings about a decrease in the fracture toughness property. There is much potential for improving the toughness by controlling the amount and shape of δ ferrite. The amount of δ ferrite decreased and its shape became more globular with increasing nitrogen content in weld metal. Consequently, cleavage fractures rarely occurred at low temperature, and the fracture toughness values were improved. The δ ferrite amount (A) and the shape factor ( S ) were combined into the factor of A·S. The A·S factor has a linear relationship with fracture toughness values.

View full abstract

This paper describes an analytical method for evaluating the elastic properties of three-dimensional fiber reinforced composite materials on the basis of a micromechanical analysis of a unit cell structure. The unit cell is a small repeating geometrical unit of fiber bundles. A special feature of this study is to treat the unit cell as a rigid frame structure constructed of fiber-beam and matrix-beam elements. This analytical structure model can faithfully represent actual fiber composition. For two types of carbon/epoxy three-dimensional composite test pieces cut out of I-beam structure, the tensile elastic moduli are computed by finite element method. Comparisons between analytical and experimental results are also made. This analytical method shows good agreement with the experimental data.

View full abstract

This paper presents a unified analysis of isotropic out-of-plane shear problems containing an elliptic hole or an elliptic rigid inclusion, and shows the closed-form analytical solutions. The applied forces considered in this paper are longitudinal shear stress at infinity, point force, screw dislocation, dipole force, and dipole dislocation at an arbitrary point. The analysis is based on the complex variable method using the conformal mapping technique. By making use of the results, the stress intensity factors (or singularity coefficients) of a crack (or rigid line inclusion) are given. Stress distribution and displacement field are shown by many graphical representations as several numerical examples. The previous results published by several authors can be included as particular cases of our solutions.

View full abstract

A boundary element method is presented for analyzing the thermoelastic contact problem. In this problem, temperature and stress fields are coupled since the thermal distortion changes the size of the contact zone and then affects the heat flow. The displacement, traction, temperature and temperature gradient are unknown in the contact zone. By the use of residual vectors of discretized field equations obtained by the boundary element method, the objective function is defined. The contact conditions are also introduced to the objective function. The thermal resistance of the contact surface is considered as a function of the contact pressure. The solution is obtained through minimizing the objective function by means of the conjugate gradient method. In order to demonstrate the practical applicability of the method, several numerical examples are presented.

View full abstract

The indirect fictitious-boundary integral method is applied to unsteady elastodynamic problems. At first, the impulsive stresses are analyzed for a spherical cavity in infinite medium subjected to impulsive pressure at the surface boundary, and these results are compared in detail with the analytical solution and the other one. Next, the longitudinal impact problem of a circular bar with a semi circular groove is treated, and the dynamic stress concentration factors and the stress distributions are accurately analyzed. Finaly, the availability of this method to unsteady elastodynamic problems is shown.

View full abstract

The use of bellows expansion joints is an effective measure for rationalizing various piping systems in industries. Essentially, bellows should be designed so as not to be subjected to torsional loads. Usually, the bellows is protected so as not to be twisted by the hardware. However, it is a fact that there are some sources of torsional deformation, e. g., the clearance of the bearing structure and the deformation of the hardware. Therefore, it will be useful to elucidate the allowable torsional displacement of bellows from the point of view of improvement of reliability in design and fabrication. In this paper the stress due to torsion and the torsional buckling behavior of bellows are investigated through both theoretical and numerical analyses. The numerical solutions agree with the theoretical results very well in a wide range of dimensions and the number of convolutions. Based on these results, the framework of the design rule related to the torsion of bellows is proposed.

View full abstract

It is known that intrinsic mechanical stresses exists in a general free surface or interface, because of surface atomic structure changes relative to the bulk. This paper presents the inter relationships between surface stress, free surface and volume stress on the basis of membrane model. The equations for the force balance taking into account the surface stresses is generally derived by means of surface geometry. Furthermore, a boundary integral equation is newly developed by using the force balance equations. In this deduction, the relevant dimensionless parameter called elastic capillary number which is a function of surface energy, elastic modulus and the mean curvature of curved surface is introduced.

View full abstract

Decreasing characteristics of both stress and stress gradient with propagation distance at a 2-dimensional linear viscoelasticity wavefront are derived by using our 3-dimensional theoretical equation for particle velocity discontinuities. By finite-element method code DYNA3D, stress at a noncurvature dilatation wavefront of linear viscoelasticity is shown to decrease exponentially. This result is in good accordance with our theory. By dynamic photoelasticity experiment, stress gradients of urethane rubber plates at 3 types of wavefronts are shown to decrease exponentially at a noncurvature wavefront and are shown to be a decreasing function of (1/√(R)) exp (α²₁/(2α³₀ξ)) at a curvature wavefront. These experiment results are in good accordance with our theory.

View full abstract

The stress singularity for the tip of a sharp symmetrical or stepped circumferential notch of cylindrical bars under tension and bending is numerically analyzed by the body force method. The parameter of equivalent stress intensity factor K_e, instead of stress intensity factor K_I for a crack, is defined considering the singular stress distribution. The value of K_e for the fatigue limit of sharply notched cylindrical specimens of S45C, SS41 and SCM440 steels are evalutated. It is shown that the threshold range of K_e, i. e., ΔK_{e. th} for fatigue limits becomes a material constant and accordingly ΔK_{e. th} can be used as the parameter to predict fatigue strengh for a sharp notch.

View full abstract

In this paper, the singular stress fields near a corner of a fiber are discussed based on the expressions of the stress fields obtained recently by the authors. The stress fields are expressed as a sum of two terms : a symmetric term and a screw-symmetric term. The intensities of singular stress field are calculated by the method developed in the first report. K_{I, λ1} and K_{II, λ2} are shown for various elastic constants and aspect ratios of fiber. The singular stresses at the interfaces are discussed on the basic of the results of calculation.

View full abstract

The stress distribution introduced in joined dissimilar plates under tension was calculated from two-dimensional (2-D : the plane stress and plane strain states) and three-dimensional (3-D) elastic stress analyses using the finite-element method. A silicon nitride plate joined to copper plate was used. In the interfacial region of the composite specimen, a high stress concentration was found to occur, distributed three-dimensionally. Particularly, at the center of the specimen, the stresses, σ_x, in the direction of the applied stress, calculated from the 2-D and 3-D stress analyses, distributed quite differently near the interface. However, at a further distance from the interface, they agreed more closely. The highest stress, σ_x, appeared at the corners of the silicon nitride plate adjacent to the interface. Near the interface in the cross section of the specimen, the stresses calculated from the 3-D stress analysis varied rapidly, in contradiction to the assumption made in the 2-D analysis.

View full abstract

It is important to obtain information on the effect of the interfacial slip, which is induced by the thermal activation process in different material bonding processes to execute the interfacial fracture analysis. In this paper, the effect of the interfacial slip in different material bonding processes on the stress singularity at the free edge was studied by finite element method (FEM) with bond elements. The results are as follows. Another stress singularity field was produced at the slip tip, although the interfacial slip released the stress singularity at the free edge. The stress field was found to resemble the crack problem, so that the singular stress field was evaluated by the stress intensity factor.

View full abstract

X-Ray stress measurement was applied to TiC and SiC monolithic films and TiC/SiC composite films coated by CVD process on graphite substrates. Textures of films were examined by the X-ray diffraction method. TiC coating had preferential growth of the (2 2 0) plane. SiC coating had preferential growth of the (1 1 1) plane. TiC/SiC composite coating showed a strong texture of the (2 2 0) plane perpendicular to the substrate surface. The diffractions from TiC (3 3 1) and SiC (3 3 1) planes by Fe-Kα characteristic X-rays are suited for X-ray residual stress measurement in thin films. In TiC/SiC composite coating, X-ray diffraction profiles from the TiC phase and SiC phase overlap. These line profiles were separated into TiC and SiC line profiles by the modified DFP method. Phase stresses in the TiC phase and SiC phase in TiC/SiC composite coating showed a triaxial stress state. The measured residual stress of the TiC phase is tensile and that of the SiC phase is compressive. Although the surface residual stress in SiC monolithic coating with thicknesses below 30μm was compressive, the residual stress was tensile for thick coating, and increased with increasing film thickness.

View full abstract

In this paper, numerical solutions of singular integral equations of the body force method in notch problems are discussed. The stress field due to a point force in a body is used as a fundamental solution. Then, the problems are formulated as a system of singular integral equations with Cauchy-type singularities. The unknown functions of the body force densities which satisfy the boundary conditions are approximated by means of the products of the fundamental density functions and polynomials. The accuracy of the present analysis is verified by comparison with the results obtained by the previous method where the unknown functions are approximated by the products of the fundamental density functions and stepped functions. The present method is found to give accurate stress distribution along the notch boundary with short CPU time.

View full abstract

In the conventional boundary element method, all the boundaries including those of inclusions must be discretized. Therefore, the size of the coefficient matrix of the resulting system of algebraic equations becomes quite large for a potential field with many small inclusions. In this paper, the potential distribution in a circular inclusion is approximated with a simple Fourier series in the angular direction and a second-order polynomial in the radial direction. In the present formulation, discretization of the boundaries of the inclusions is not needed, hence the total number of degrees of freedom is quite reduced in comparison with the conventional method that actually discretizes the inclusion boundary. The effectiveness of the present formulation is illustrated in several numerical examples.

View full abstract

A new algorithm for branch-switching at bifurcation points in the nonlinear finite elenent method is developed. The procedure is simple and neither eigenvalue analysis nor an alternative cpu-extensive task is necessary for the calculation of buckling modes at bifurcation points. The basic concept is that a displacement corrector vector in the Newton-Raphson iteration can be used in place of an eigenvalue mode which is usually used for branch-switching, due to the high similarity of the governing equations at bifurcation points. Two numerical examples, buckling of a square plate under compressive loading and 'diamond buckling' of a cylinder, are demonstrated to show the validity of the algorithm.

View full abstract

Since the quality of FEM analysis directly depends on the quality of meshes, various mesh adaptation schemes have been studied. There are two stages in adaptive mesh refinement: to derive error measure and to control meshes according to error estimation. The former has been well researched. However, the importance of the latter aspect has not been adequately considered. Even if the error measure is well estimated, the total performance of mesh adaptation might be poor with poor mesh control. We provided an effective remeshing scheme in the previous paper, although the desirable mesh size was not well derived there. In this paper, the explicit relation between error measure and mesh size is proposed. After evaluation on several examples, it is confirmed that the proposed error measure-mesh size relation is acceptable. The limitation of mesh adaptation is also discussed in this paper.

View full abstract

The paper is concerned with the inverse problem in which unknown boundary values in transient heat conduction should be estimated by inverse analysis using the boundary element method. It is assumed that the measured data of time-dependent variables at some measuring points on the boundary can be used for inverse analysis. Unknown boundary values are approximated by using a spline function in space and time. The inverse problem is reduced to an optimum problem in which a set of parameters describing variations in unknown boundary values should be estimated by using the standard optimization technique. Numerical simulation is carried out for several sample problems to confirm the usefulness of the proposed method.

View full abstract

We develop a new method to calculate impact forces of a thin target and of a projectile at normal perforation of a cone-nose rigid projectile by means of a 3-dimensional finite-element method (FEM) code DYNA 3 D. We use a new fracture condition (strain fracture condition that depends on pressure) and a small hole target model at perforation calculation. Impact forces calculated by the FEM code are compared with impact forces of 2-type experiments ; target force in a forward ballistic experiment (Yoshida) and projectile force in a reverse ballistic experiment (Arajs). Target forces by calculation have a high frequency mode and have good maximum values compared with Yoshida's experimental forces. Futhermore, projectile forces determined by calculation are in good agreement with Arajs's experimental forces.

View full abstract

The analysis using a neural network was carried out to determine the acoustic emission (AE) source waveform from the artificial AE signal detected by a piezoelectric AE sensor. We assume that each sampling point of the waveform corresponds to one unit of a layer in the network. It was found that the two-layer network assuming convolution form would be effective to determine source waveforms. Since linearity of the input/output relationship of the system would be limited, two-layer networks might not be available to deal with a general problem in AE source analysis. Hence, the network structure was extended to three layers, in which was used the back-propagation algorithm for learning. It was demonstrated that if appropriate waveform data were provided for learning, analogous source waveforms could be exactly determined. It was also revealed that the number of intermediate layer units might be of greater necessity than the number of learning patterns, but the number of units exerted little influence on the calculations.

View full abstract

A new nondestructive inspection technique using infrared thermography was proposed, in which the thermal image of the surface temperature on a heated sample was used to identify flaws and defects. Joule effect heating by an electric current was employed to heat the sample instantaneously. Both numerical and experimental studies were conducted on the resolution and the availability in the detection of the through-thickness and surface cracks embedded in steel plates. The results showed that a singular concentration was observed at the crack tips in the surface temperature field in the transient stage of heat conduction, and the cracks were found to be sensitively detected from such a singular temperature field in the early transient stage. This technique was also applied to the inspection of the delamination defect in carbon-fiber reinforced plastics (CFRP), again using Joule effect heating. The subsurface defect was found to be identified from the localized low-temperature region appearing on the sample surface. It was found that the proposed technique is applicable to the damage inspection in CFRP as well as the flaw inspection in metallic materials.

View full abstract