A rolled aluminum alloy plate has a remarkable texture. This texture affects the mechanical properties. In this study, rotating bending fatigue tests were carried out on plain specimens cut out from a rolled plate of an age-hardened aluminum alloy, 7NO1-T4, in the rolling direction (material L) and its longtransverse direction (material T) in order to investigate the influence of the texture on the fatigue properties. Slight anisotropy of the crack growth rate is recognized. This is caused by the difference in the crack propagation mechanism, that is, the fracture mode is a tensile-type in material T and a shear-type in material L.

View full abstract

Using the CCT specimen made by an aluminum plate, the effect of anisotropy on the fatigue crack propagation path was investigated, and also the crack propagation path in semicircular notched plate was estimated. The obtained results were as follows. A marked effect of anisotropy appeared in the specimen whose rolling direction was inclined 15 or 30 degrees from the loading direction. Namely, in this case, all the cracks in plate specimens propagate along the rolling direction. However, in the other case, the cracks propagate along the Mode I direction. These results are considered to be based on the effect of Mode II crack propagation. In order to predict the crack propagation path under complicated stress conditions with FEM, the authors assume a condition in which the crack propagates along the rolling direction in the case of G_II/G_I≥2.0 and G_I >G_{I th}. If the above conditions are not satisfied, the crack propagates along the Mode I or maximum energy release rate direction. This assumption is proved by the crack propagation path of several types of semicircular notched plate specimens.

View full abstract

In this investigation, singular fields of stresses and heat fluxes at the tip of a crack in a nonhomogeneous body, or functionally gradient material #(FGM), are studied under both elastic and plastic conditions. It is found that the crack tip fields have the same forms as those in the homogeneous material, provided that the material properties of FGM are continuous. The heat fluxes have a square-root singularity at the crack tip, and the elastic crack tip stress field is characterized by the K-field of linear fracture mechanics and the plastic singular field is the HRR field for a power-law material. The relationship between stress intensity factors and the path-independent integral J proposed by Aoki, Kishimoto, Matsumoto, Tachihara and Sakata is discussed.

View full abstract

Abstract-In this paper, we consider the thermal-mechanical stress problem of cracked G-10CR glass-epoxy laminates with temperature dependent properties. The layered composite is made of one cracked layer bonded between two other layers of different physical properties, and it is suddenly cooled at the surfaces. The ends of the crack are situated at equal distances away from the interfaces which are directed normal to the crack plane. The composite material in generalized plane strain is assumed. Fourier transforms are used to reduce the problem to the solution of a pair of dual integral equations. The solution of the dual integral equations is expressed in terms of a Fredholm integral equation of the second kind. Numerical results on the transient stress intensity factor are obtained and are presented in a graphical form.

View full abstract

The stress distribution along grain boundaries of polycrystalline materials ahead of a cavity under Coble creep (grain boundary creep) condition is numerically analyzed in order to elucidate the effect of the grain boundary network on cavity growth. The stress gradient at the cavity tip, which is proportional to the changing rate in the cavity volume, points out that the grain boundary triple points interfere with the cavity growth, though the cavity grows faster beyond them. The deceleration effect due to the triple points diminishes as the cavity becomes larger. The transition from "crack-like cavity" to "crack" is also discussed on the basis of the stress distribution. "Crack-like cavity" and "crack" are distinguished by the fact that the latter possesses stress singularity in the vicinity of the tip while the former does not. The calculation reveals that the transition under the Coble creep condition does not take place before the cavity length exceeds 15 grain boundaries. The transgranular deformation due to dislocation creep drastically accelerates the transition.

View full abstract

Fatigue crack growth tests were carried out in order to give a unified explanation of the formal contradiction between two fatigue crack growth laws in small and large cracks, dl/dN∝σⁿl and dl/dN∝ΔK^m. Cracks were monitored using the plastic replica method. Crack opening displacement ranges (ΔCOD 's) were measured from the positive metal replicas made from the plastic replicas. In small and large cracks having the same growth rate, the relationships between ΔCOD 's near the crack tips and the distance from the crack tip were almost the same. Therefore, ΔCOD near the crack tip can be considered as the main factor controlling the crack growth rate. Based on this, the two fatigue crack growth laws can be explained consistently.

View full abstract

Intergranular cracking due to oxidation under low-cycle fatigue loading at 700°C has been studied with a comparison of results obtained from two different types of specimens, with and without oxidation. Morphology of the surface oxidation of the specimen under low-cycle fatigueloading is quite different from that of the specimen without fatigue loading. Selective oxidation along the grain boundary was observed on the surface of fatigue-loaded specimens, while on the other hand, the surface of specimens without fatigue loading was oxidized uniformly all over the area. Intergranular cracking occurs under the combined effects of preferential oxidation along the grain boundary and grain boundary sliding in high-temperature low-cycle fatigue in air.

View full abstract

This paper deals with the effect of nodule count on the low-cycle fatigue life of ferritic ductile cast iron. Stress amplitude-controlled fatigue tests are conducted on ferritic ductile cast iron with different nodule count, i. e. high nodule count (HNC), and low nodule count (LNC) with same chemical composition and heat treatment. The difference of nodule count yields difference of fatigue mechanism in HNC and LNC. The fatigue life of both materials satisfies the Manson-Coffin relationship and shows an opposite trend depending on the fatigue region. The Fractgraphic observation reveals that the crack initiation site of LNC is originated at spheroidal graphite, while a coarse graphite controls the crack initiation in HNC. The increase of graphite size in LNC results in enhancing the internal notch effect of graphite.

View full abstract

The initiation and growth of intralaminar fatigue cracks in notched unidirectional carbon/epoxy laminates, Toray T800H/#3631, were investigated under tension-tension cyclic loading with constant stress amplitudes. The initiation of fatigue cracks at the notch root was caused by shear stress concentration. The fatigue life for crack initiation was a unique function of the maximum shear stress divided by the critical value of the maximum shear stress for static tensile fracture. Fatigue cracks propagated under mixed mode, mode I and mode II. When the crack length was short, the crack propagation rate was determined by the mode I stress intensity range, ΔK_I. As the crack length increased, the mode II stress intensity range, ΔK_II, became a dominant crack-driving force. Although the ΔK_I value increased with crack length, the propagation rate monotonically decreased and the crack finally stopped. Such deceleration and non propagation of cracks were caused by the increase of interlocking and friction between fracture surfaces. The condition of the non propagation of fatigue cracks was determined by a constant value of ΔK_I for short cracks in the ΔK_I-dominant region. 0n the other hand, for long cracks in the ΔK_II-dominant region, the threshold value of ΔK_II increased with increasing crack length.

View full abstract

Double cantilever beam (DCB) specimens were used to study mode I interlaminar fracture of unidirectional carbon fiber/epoxy composites. An improved analytical model was introduced to analyze the crack opening displacement (COD) of the DCB specimen. Theoretical values of COD were compared with those measured by moire interferometry to evaluate an effective interlaminar tensile modulus and strain in the vicinity of a crack tip. The results showed that the modulus plays an important role in increasing the values of COD, and that there is a correlation between the modulus and the strain.

View full abstract

Matrix cracking is the most important damage mechanism of FRP structures because matrix cracking causes the stiffness degradation of structures and becomes the staring point of delaminations. In this study, therefore, with specimens of stacking sequence [0/90₄]_s tension tests were conducted and in-situ observations of the matrix cracking in the 90° lamina were conducted with a replica method. As a result, the distributions of the matrix crack position were experimentally proven to be fractal distributions of D=0.45 and is constant at any loading levels. By using these fractal distributions, the stiffness degradation was evaluated based on an imaginary crack density induced by the Renormalization Group Theory. Moreover, the distribution of an initial defect size was experimentally proven to be a fractal distribution of D=1.6. Based on these results, the fractal distribution of the initial defects was modeled and applied successfully to predict variations of crack density with increasing applied load by using a damage mechanics.

View full abstract

To evalute fracture toughness of C/C composites, four-point bending tests (SEN) are performed. The C/C composites are carbonized from the CFRPs, which are fabricated by means of two different methods (a hot-press method and a vacuum-bagging one). It is found that each fabrication method leads to a different distribution of bending strength. The vacuum-bagging method is preferable to obtain uniform bending strength properties. Crack growth resistance curves are obtained based on the compliance matching method. These curves have an inclination of convergence at a nearly equal constant value regardless of fabrication method. Existence of the stress shielding mechanism in the C/C composites is suggested by the R curves. From a fractographic observation, crack propagates three-dimensionally, and then pullouts of the fiber bundle occur at a large scale in the fiber-bridging mechanism.

View full abstract

It is well known that the fracture strength of ceramics with small defects is lower than the strength estimated by linear fracture mechanics, based on the strength of a large defect. This phenomenon can be explained by the increase in the crack resistance with crack propagation. This R-curve behavior is caused mainly by grain bridging in the wake of the crack. Recently the grain-bridging stresses for alumina and silicon nitride have been measured experimentally. In this study, the fracture strength is estimated by numerical simulation, based on grain-bridging stresses. In the simulation, the effective stress intensity factor at the crack tip, K_tip, and the effective fracture toughness at the crack tip, (K_tip)_c, are used for the conditions of stable and unstable crack propagation. It is found that the fracture toughness for long cracks, K_Ic, depends on specimen geometries, and the maximum values of K_c approach (K_tip)_c as the initial crack length approaches the width of the specimen. Furthermore a convenient method to evaluate (K_tip)_c experimentally is discussed.

View full abstract

The misalignment of the flat anvil of a testing machine with the end of a circular rod specimen causes local crushing due to the stress concentration and bending moment, which must be avoided in compression tests. When very hard material, such as fine ceramics, is tested using the compression split Hopkinson bar, the hard specimen may be indented into the incident and transmission bars. In this paper, the insertion of a thin metal plate between the bar-specimen interface in the conventional compression split Hopkinson bar system was examined to prevent indentation. The stress pulse in the incident and transmission bars is measured using the thin steel plate (thickness 0.4 and 1 mm), copper (1 mm) and aluminium (1 mm), and the pulse profile is investigated. The configuration of the wave observed using the thin steel plate (0.4 mm) is the sharpest, and is similar to that without the plate. The rise of pulses using the steel plate (1 mm) is ramplike. With use of a copper plate and aluminium, the two-step profile appears. The stress-strain relationship evaluated from these stress pulses was compared with that obtained in the static test. The dynamic strength of ceramics is less than about half the static strength.

View full abstract

Stress distributions in double lap adhesive joints are analyzed using an elastoplastic finite-element method in order to predict the joint strength. The effects of the yield stress and Young's modulus of the adherends and the lap length on the stress distributions in the joints are demonstrated. In addition, the joint strength is estimated taking into account the peel fracture produced at the interface, by von Mises's criterion. Experiments were also performed. The analytical results are in fairly good agreement with the experimental results concerning the joint strength. It is seen that the joint strength increases as the adherend rigidity increases. In addition, it is shown that the joint strength of double-lap joint is twice as large as that of a single-lap joint when the yield stress of the adherends is increased.

View full abstract

Periodic wear is often seen on the circumferential surface of a rotating disk made of polymeric material when it is loaded with contact at a point, such as the rubber tire of vehicles. This type of wear is markedly dependent not only on environmental and loading conditions, but also on the mechanical properties of the material. Many factors such as heat generation due to viscoelastic response and retardation of creep recovery of the material, and vibration of the system are considered to be responsible for this problem. It is pointed out that the first step in determining the mechanism of periodic wear may be established from the viscoelastic response of the material.

View full abstract

In order to investigate the elastic and plastic bending deformation of a sandwiched beam with honeycomb core, three-point bending tests were carried out statically. Tensile tests were also performed to obtain the mechanical properties of surface plate material. It was found from bending tests that the applied load began to decrease just after the occurrence of plastic buckling of the honeycomb cell underneath the loading point. Welded joints never failed even after buckling of a honeycomb cell. By extending the elementary bending theory into the plastic region, elastoplastic load-deflection curves were obtained. The effect of shear force on honeycomb core was also taken into account. The load-deflection curve obtained by the analysis showed reasonable agreement with that obtained from experiments.

View full abstract

A finite-element polycrystal model is proposed to predict plastic deformations of polycrystal metals. Each element in FEM formulation is assumed to be a crystal of cubic shape, in which plastic strains are assumed to be uniform. For the determination of time-independent slips, a new numerical scheme, the "successive integration method", is proposed on the basis of Schmid's law. The numerical code of the present model is quite simple and can be applied to any arbitrary loading path. A stress-strain curve in simple tension is calculated for a non-workhardening FCC metal and compared with the other theories.

View full abstract

A rational constitutive model for initially orthotropic but subsequently skewed-anisotropic materials is proposed in this paper. The variation of anisotropic axes is discussed, and the evolution of the axes is then formulated. It is shown that the yield function proposed is identical to Hill's quadratic form in a very special case of orthotropy. Applying the conventional laws for both workhardening and material flow to the anisotropic yield function, an anisotropic constitutive model is completed. A finite element equation is obtained next, based on the up-dated Lagrangian type formulation. Three-dimensional finite element analyses are carried out to predict the necking of anisotropic sheet metals. The direction of the localization is strongly affected by the initial configuration of the anisotropic axes.

View full abstract

Inverse analysis for estimating the geometry or the shape and the elastic properties of a structure is generally carried out by using additional information such as experimental data of displacements or strain and velocity of the structure the subjected to any mechanical loading or constraint. Such inverse analysis has some problems as to the uniqueness of the solution and the stability of the inverse procedure. The model for the inverse analysis in the present paper is a square plate with several circular defects subjected to a bidirectional load. The position, the size and the number of internal defects are estimated by using the genetic algorithm with additional data of displacement in the outer boundary. A new procedure for estimating the number of defects is presented in the paper and is shown to also be useful in the inverse analysis using data with several errors.

View full abstract

This paper presents a method of optimum design for transient heat conduction fields using inverse analysis with the boundary element method and the Kalman filter algorithm. It is expected that the inverse analysis method can be applied to optimization problems of determining unknown parameters by minimizing cost function. The optimization problem under consideration is such that the boundary conditions on part of the boundary are to be determined so that requirements of responses on the remainder of the boundary are satisfied. A few sample problems of optimization are computed by means of the inverse analysis method to demonstrate its potential usefulness.

View full abstract

The θ projection concept, recently developed to predict not only creep curves even in tertiary creep but also long-term creep properties, required constant-stress creep data. The authors designed a constant-stress creep machine, only changing the conventional loading lever of the constrant load creep machine into the profiled cam lever. The purpose of this study is to examine whether there is relationship of parameters between constant-stress and constant load creep curves derived from the θ projection concept. Both creep tests were conducted using duralumin at 250°C. Six kinds of cutoff strains were proposed to systematically obtain four parameters, e. g., values of standard error or dimensionless standard error. The creep curves can be described precisely within the standard error of 0.02%. However, poor relationships of parameters between both creep tests were recognized because creep strain at rupture was small in the test material.

View full abstract

Thermal residual stress distributions on the interface and in the vicinity of the intersections of the surfaces and the interface of dissimilar materials are calculated by the boundary element method. Thermoelastic constant terms are calculated using Airy's stress function. The thermal residual stresses, when the thermoelastic constant terms are subtracted, show free-edge stress singularity. The value of the order of thermal residual stress singularity and the stress distributions agree well with the theoretical ones calculated using Airy's stress function. The thermal stress singularity disappears for certain ranges of wedge angles of the pair of materials, as predicted.

View full abstract

This study deals with the thermal stress in a surface mount joint under uniform temperature change, where an upper element is mounted on a substrate and is adhesively bonded/soldered near both ends. The analytical approach is developed mainly in the case of the plane strain state and general solutions for the thermal stress and the strain distributions in the joint are derived using the two-dimensional theory of elasticity. It is shown by numerical calculations that the thermal stress is singular at both edges of the interfaces between the upper element and the adhesive, and between the substrate material and the adhesive. Maximum principal thermal stresses are examined in several types of joints in order to predict a fracture initiating point in a surface mount joint. In the experiments, the thermal stress distribution in an adhesive layer is measured by photoelasticity. From the comparisons between the analytical results and the experimental ones obtained by photoelasticity, fairly good agreement is shown.

View full abstract

The θ projection concept, which gives one of creep constitutive equations, can predict various creep curves over a wide range of strain, including the tertiary stage. However, parameter estimation procedures for solving the linearized maximum-likelihood function derived from the original nonlinear equation, involve some numerical difficulties e. g., nonconvergence. In order to improve the accuracy of parameter estimation, to eliminate the numerical difficulties and to be free from approximations by using the original equation as the objective function, the authors have developed two procedures for parameter estimation employing nonlinear programming. One is to estimate material parameters individually for each creep curve. The other is to estimate the coefficients formulating the material parameter as a simple polynomial function of stress and temperature to allow interpolation and/or extrapolation of the material parameter. According to estimation from the strain data referring to the literatures, these procedures can successfully predict the creep curves obtained for some steels.

View full abstract

This paper deals with the algorithm of structural analysis by the ε-method for a two-dimensional multistoried and multispan rigid frame with circular members loaded in arbitrary directions of the frame's plane, in which the bases of the columns are all fixed. The different member's revolution angles and the axial forces are taken into consideration and the digital computing program based on this method is developed. At the end of the paper, the results from one example are given.

View full abstract

This paper describes an investigation on the tensile and fatigue fracture behavior of an SiC whisker reinforced 6061 aluminum alloy fabricated by a squeeze casting process under combined tension/torsion loading at room and elevated temperatures. The tests were conducted under a load-controlled condition keeping a constant value of the combined stress ratio, α=τ/σ. Except for the elongation at failure, mechanical properties of the composite including fatigue strength were superior to those of an unreinforced 6061 alloy, not only under uniaxial loading, but also under combined tension/torsion loading. The static strength of the composite and the unreinforced matrix material showed good agreement with the Tsai-Hill failure criterion. Fatigue strength under α≤1 was determined by maximum principal stress, whereas the strength under α>1 was determined by maximum equivalent stress. Fracture surfaces were closely examined with a scanning electron microscope, and the fracture mechanisms under combined tension/torsion loading were discussed.

View full abstract

Stainless steel, titanium alloys and aluminum alloys are metallic materials widely used in the marine environment, by reason of their excellent anticorrosion character. The characteristics of these materials depend strongly on the passivity film over their surface, which is a very thin oxide film with intensively chemical stability. Therefore, if a repeated load accompanied by wear such as fretting works on these materials, the thin film may be successively destroyed and the surface of metal is exposed directly to the environment. The question arises whether or not the anticorrosion quality could be maintained. This study is a basic experimental investigation to shed light on the question. In this paper, the author reports the fretting wear of aluminum and its alloys in some surroundings. The chief measuring items in fretting processes are wearing volume, variation of friction and dropping down of electrode potential of materials. According to the experiment, some interesting results are obtained, which are characteristic of aluminum and its alloys.

View full abstract

This paper deals with the adaptive procedure for structural analysis using the boundary element method for two-dimensional problems with linear elements. The algorithm in this paper is simple and fast for computations of error estimations and the mesh refinement. We used the h-refinement and hr-refinement. We showed the experimental rule for the mesh refinement. The numerical results for some problems show the validity of the adaptive procedure in this paper.

View full abstract

On August 28, 1992, a terrible explosion occurred at a cold evaporator for liquefied nitrogen at the Ishikari plant of Suzuki Sogo Shokuhin Co. Disruptive failure of pressure vessels generated high-speed projectiles. The maximum extent of the zone affected by these projectiles was about a 350 m radius. The cause was pressure increase due to locking of original valves of safety valves. The cause and process of disruptive failure of pressure vessels were analyzed.

View full abstract

When a flexible magnetic medium is driven, it is in contact with the magnetic head. To evaluate the tribological behavior of fiexible media, it is necessary to measure the mechanical properties. However, it is very difficult to measure the mechanical properties, since the magnetic layer is very thin and is attached to the substrate. First, this paper proposes a convenient method for separately determining Young's moduli of two thin films of a magnetic medium. Second, the particular stress distribution inside a magnetic film (bonded two thin films) is analyzed. The internal stress in the medium is not uniform when the medium is subjected to tension. The analysis shows that the difference in Poisson's ratio in two bonded thin films causes flexure of the medium under tension as well as shearing stress at the interface between the magnetic layer and the substrate. Since the shearing stress may cause fatigue damage of the tape edge, detailed FEM analysis on its distribution is carried out.

View full abstract

Optical birefringence is measured by a high-frequency modulation method using a photoelastic modulator and polarized laser. The modified synthesis method was developed and applied for the measurement of stress distributions of a glass plate. The distributions of the differences of principal stresses and their directions were obtained directly by this method. The stress distributions of a pulled rectangular glass plate with a small hole at its center were measured. The results of stress distributions agreed with the analytical results. It was confirmed experimentally that the spatial resolution of stress measurement had the same size as the diameter of laser light. The stress at many points could be obtained quickly by the synthesis method and the scanning stress distribution measurement was realized.

View full abstract