The remaining strength of ceramics after repeated thermal shocking was analyzed theoretically, and the probability of crack occurrence was predicted. In this analysis, the Weibull statistical theory of fracture was applied to repeated thermal shocking of ceramics assuming that thermal shock fracture of ceramics was due to crack growth from one of initial flaws. For the verification of this proposed analysis, tests on alumina, silicon carbide and silicon nitride samples with various diameters were conducted under various repeated thermal shock conditions of water quenching. As a result, analytical results were found to be in good agreement with results of repeated thermal shock tests. Thus, the probability of crack occurrence can be evaluated from the normalized fracture strength using effective volume and effective stress hold time.

抄録全体を表示

In the present study, the sampling inspection condition was analyzed for a method of ensuring the ring crack initiation strength of ceramic rolling bearing balls on the basis of the 2-parameter Weibull distribution. It was clarified how the allowable lower limits of sample size is estimated not only in the case in which parameters of both shape and cale are unknown but also in the case in which only the shape parameter is known. The former case is superior to the latter case as regards the smaple size necessary for a sampling inspection test, but the latter case is advantageous for the practical tests, considering that 8 data can be measured on a ball in the case of the sampling inspection test for ring crack initiation strength. Ceramics, Reliability, Statistical Treatment, Weibull Distribution, Sampling Inspection, Consumer's Risk, Producer's Risk, Confidence Level, Ceramic Bearing Ball, Ring Crack

抄録全体を表示

For structures with slow-growing cracks, a suitable maintenance programme can be performed to avoid the occurrence of safety and/or functional problems. An economical maintenance programme is desired and the maintenance economy can be characterized by the so-called economic life. The occurrence of widespread damage in a structure which is uneconomical to repair and, if not repaired, could cause functional problems, was used to define qualitatively the economic life of the structure. In this paper, by using a quantitative economic life criterion, and taking into account the stochastic crack growth, a general formula for determining the economic life is derived, and a numerical example is given. The effects of economic repair limit, the detectability and the initial fatigue quality on the economic life are discussed.

抄録全体を表示

Nuclear power plants have the greatest potential for use in space ships, underwater vehicles, and so on. It is very important to develop new radiation shielding materials which have excellent mechanical strength and heat-resistant characteristics. Titanium boride ceramic (TiB₂) is an excellent neutron absorber and has been investigated as a radiation shielding material. This report presents an evaluation of thermal shock resistance by the laser irradiation technique for three kinds of titanium boride ceramics with 3, 8, and 15% porosity. Their temperature and thermal stress distributions are analyzed by the finite-element method, and the relationship between the spot diameter of the laser beam and the maximum tensile thermal stress is derived for various power densities of the laser beams. Critical fracture curves are obtained from these relationships, which can specify a critical power density for a given laser beam spot diameter. The irradiation experiments are performed on titanium boride ceramics using a C0₂ laser. Finally, theoretical results are compared with experimental ones.

抄録全体を表示

The strength evaluation method based on Weibull's statistical theory is often used for estimating the stlength or the reliability of actual ceramic components. However, the validity or the reliability of the method itself has not yet been investigated fully. Comparison is made here between results calculated by the method and experimental results, which are those of bending of prismatic bars, tension and torsion of round bars, bending of notched bars, out-of-plane bending of circular plates with a shoulder fillet and combined bending and torsion of rotor blade root models. It is shown that the method can predict the strength of ceramic components of various sizes and shapes fairly well. However, large portion of the reliability of the method is dependent on that of the value of Weibull modulus m. It is also revealed that there are still several points that remain to be overcome for the method to be used more reliably.

抄録全体を表示

The design concept based on reliability is effective for advanced materials, because advanced materials such as fiber-reinforced composites have variable mechanical properties. In addition, as the failure mode in the low cycle fatigue region differs from that in the high cycle fatigue region, it is difficult to predict the fatigue life of advanced materials. So far, we have developed a design margin decision system based on reliability by evaluating fatigue life. In this study, the system is improved for application to advanced materials and a method to divide whole sets of data into the above two regions and to calculate the design margin in each region is presented. From numerical examples for CF/PEEK laminates and epoxy resin, it is revealed that a more reasonable S-N curve and design margin are obtained by the proposed system.

抄録全体を表示

Pinhole defects in two different TiN films were evaluated by conducting tests employing various electrochemical methods. TiN-coated stainless steels prepared by plasma CVD are best evaluated by electrochemical measurements under the coupling condition between specimen and counter electrode of Pt, because there is little spalling-off of the coated film in this evaluation method. On the other hand, TiN coated steels prepared by dynamic ion mixing can be examined using potential sweeping electrochemical measurements, such as the critical passivation current density (CPCD) method and polarization curve method, because there is sufficient adhesive strength between the substrate and the coated film. Then, the CPCD method was applied for evaluation of the defects in plasma-sprayed ceramic coating whose thickness is more than 200 μm. The corrosion damage of plasma-sprayed ceramic coating was brought about by the anodic dissolution of the undercoated NiCrAlY layer at the boundary between the zirconia top coat and the undercoated layer. Therefore, the ratio of the ratio of the undercoated layer's damage, that is the defect ratio of top-coated layer, due to corrosive solution penetrating the inner layer through the top coat layer can be evaluated through the CPCD method.

抄録全体を表示

In-situ observation using a "SEM servo" of fatigue crack initiation and growth behaviors of Hastelloy-X at 973K in vacuum was carried out. The main results obtained were as follows. (1) A strong correlation between fatigue crack growth (FCG) rate da/dN and stress intensity factor range ΔK=K_max-K_min in the lower rate region from 2×10^-11 to 4×10^-8 m/cycle was found, which suggested a relationship in the form of da/dN=C(ΔK)ⁿ, where n was 7. 2. Furthermore, the threshold for FCG, ΔK_th, could not be recognized. (2) In-situ observation showed that the fatigue crack propagated within the grains as a transgranular crack and also combined with microcracks in grains. (3) The S-N diagram for an ultralong life of 10⁸ cycles was obtained, which showed the relation S=-32 log N_f+348. Also, by in-situ observation, the relation between S and crack initiation life, N_i, was found to be S=-32 log N_i+328. (4) By means of a reliability engineering approach to distributed microcrack growth, it was found that a two-parameter Weibull distribution fitted the experimental data of crack length well, which suggested a mixed mode with two values of shape parameter m_t(=m_g+m_c) : m_g for growth and m_c for combination. Therefore, it could be concluded that the fatigue life was shortened due to the combination of microcracks.

抄録全体を表示

The three-point bending test was conducted on ground specimens (30×8×4, mm) of a commercial WC-Co alloy having four different surface roughnesses with R_max of 3.0, 0.3, 0.1 and 0.03 μm. It was found that the bending fracture strength and its scatter were almost the same in all specimens irrespective of surface roughness, except for exceptionally low strengths obtained in a few specimens with R_max of 0.03 μm. This negligible effect of grinding on the strength was explained by assuming that the two effects of surface roughness and compressive residual stress measured in this study cancelled each other. In addition, it was proposed that the grinding process generally "healed" relatively large surface defects, and this "healing effect" resulted in the increased strength. The exceptionally low strength of a few specimens of R_max 0.03 μm is principally caused by lack of the healing effect.

抄録全体を表示

Oxide-dispersion-strengthened Ni-NiO composites were prepared by the powder metallurgy technique. The nickel oxides were added at the 1∼15wt.% level. Control materials were also prepared without NiO added. The distribution of the NiO phase and grain size of nickel matrix were observed metallographically in materials sintered at 1473∼1673K for 10.8 ks. The Vickers micro hardness of as-sintered materials was measured in order to investigate the correlation between the microstructure and the room-temperature mechanical properties. The Vickers micro hardness of the samples sintered over 1573K increased almost linearly with NiO content. That of samples sintered below 1573K was smaller than that of samples sintered over 1573K. From the viewpoint of hardness, it was desirable to sinter samples at a temperature over 1573K. NiO particle growth was found to be governed by the Ostwald ripening law. The Vickers micro hardness was proportional to the root of the reciprocal grain size.

抄録全体を表示

The effect of silicon carbide (SiC) on small fatigue crack initiation and propagation behavior of A6061-T6 metal matrix composites reinforced with 20 vol. % or 30 vol. % SiC particles of various diameters was investigated under axial cyclic loading at the stress ratio R of 0.1. The results obtained are summarized as follows. (1) The fatigue cracks of these composites initiate preferentially at the sharp end of a SiC particle in the matrix aluminum material, because the stress near the sharp end of a particle appears to be much higher than the average stress. (2) Small cracks exhibited much higher growth rates just after crack initiation, followed by a decrease in growth rate to a minimum value, which corresponded to that of a Stage I crack, and a monotonic increase with increasing ΔK. (3) The highest fatigue strength of the 20 vol. % SiC_p (mean diameter is 7 μm) composite was due to the excellent small crack growth resistance of this material.

抄録全体を表示

ABS resin is widely used in the fabrication of general-purpose plastics. This material has a structure in which butadiene rubber particles are dispersed like minute islets within an acrylonitrile-styrene copolymer matrix, and the phenomenon whereby butadiene rubber particles are deformed in the molding process is wellknown. This paper reports the strong relationship between the deformation of the butadiene rubber particles dispersed near the root region of the pole and the error of the pole perpendicularity when the plain-plate-shaped specimen of ABS-resin on which the pole stands vertically, is made by the injection molding process.

抄録全体を表示

The effects of the fiber volume fraction of CFRTP (Carbon Fiber Reinforced Thermoplastic polyimide) on creep behavior were investigated. The bending creep tests for resin and CFRTPs where fiber volume fractions were 3.7, 7.6 and 15.6% were carried out at constant temperatures from 190 to 240°C. For all materials, master curves of creep compliance could be drawn and the reciprocation law of time and temperature of Arrhenius type was applied. With shift not only on the time axis but also on the compliance axis, these 4 master curves were found to correspond to each other. Hence it was concluded that master curves of CFRTPs mainly depended on the viscoelastic property of resin. The effects of the fiber volume fraction on the shift factor on the compliance axis when adjusting master curves of CFRTPs to the master curve of resin is related to the law of mixture.

抄録全体を表示

The effect of packaging stress on device characteristics was determined experimentally. The sensitivity to stress of the electronic characteristics of MOSFETs was measured by applying uni axial stress to the transistor. The sensitivity to stress was shown to depend on the transistor conduction type and the current flow direction. The packaging stress effect on a simple inverter amplifier using MOSFETs was measured by changing the packaging materials. The residual stress distribution was also measured using stress sensing gauges embedded in LSI chips. The distribution of the amplifier gain change rate agreed well with the residual stress distribution in the LSI chip. The amplifier gain change rate was calculated based on the experimental results of the sensitivity to stress of the electronic characteristics of MOSFETs and the residual stress distribution. This predicted amplifier gain change rate agreed well with the measured data.

抄録全体を表示

A thermal stress problem of a coated half-space due to a temperature discontinuity on the interface is solved exactly using the two-dimensional thermoelastic theory. Stress singularities at the point of discontinuity of temperature are obtained in closed form. The results show that the normal stresses have jumps of the inverse tangent type and the shear stress has singularity of the logarithmic type. From the comparison with the singularities of the corresponding problem of a point heat source at the interface, the important result is obtained that the types of singularities for the two problems are different but the intensities are the same. In other words, the types of singularities depend on the type of loading but their intensities depend only on the combinations of materials and do not depend on the type of thermal loading.

抄録全体を表示

A new X-ray method was proposed to measure the triaxial residual stress in Al thin films sputtered on (100)-oriented silicon single-crystal wafers. The film had [111] fiber texture perpendicular to the film. The linear relation between strain and sin² ψ obtained from Al 222 and 311 diffractions was used to determine the triaxial residual stress in Al films. The residual stress in Al films was equibiaxial tension, and the magnitude decreased with increasing film thickness. The residual stress determined by the curvature method was nearly equal to that determined by X-ray analysis. The loading stress in Al-coated film was found to be biaxial even under uniaxial tension of Si substrate. The stress value measured by the X-ray method agreed with the prediction based on the mismatch of Poisson's ratio between coating and substrate.

抄録全体を表示

In this paper we describe an evaluation of multistage honeycomb panels, the buffer characteristics of which consist of the impact load, the deformation in the direction of the axis, the crush pattern and absorbed energy. The test specimens are the uniformtype laminated honeycomb panels of uniform cross-sectional area and the pyramidtype of variable cross-sectional area. Furthermore, the uniformtype are of the two kinds of aluminum alloy, namely artificially aged (T6) and naturally aged (T4). The various processes of the time history to the impact load are classified into three regions. These processes that depend on the deformation velocity are repeated the number of the crushed panels. When the absorbed energies are compared between the different aging processes, the panels of T6 excel in capacity of the absorbed energy and those of T4 excel in efficiency because the deformation for T4 increases more than that for T6. Furthermore, it is evident that the absorbed energy characteristics of the pyramidtype exceed those of the uniformtype in both efficiency and capacity.

抄録全体を表示

The stress distributions in single-lap adhesive joints with circular hole defects in adhesives are analyzed using a two-dimensional theory of elasticity when the joints are subjected to tensile shear loads. In the numerical calculations, the effects of the size, location, number of holes and the ratio of Young's modulus for an adhesive to that for adherends on the principal stress distributions at the interfaces and around the circular holes are examined. As a result, it is found that the stress concentration around holes increases and the stress singularity at the edges of the interfaces decreases as the hole size increases. The stress distributions in joints are analyzed by the finiteelement method. Photoelastic experiments were also performed. Fairly good agreement is seen among the analytical results, the F.E.M results and the experimental results.

抄録全体を表示

The stress distributions and the deformations in single-lap adhesive joints consisting of similar adherends are analyzed using a two-dimensional theory of elasticity when the joints are subjected to tensile shear loads. In numerical calculations, the effects of the ratios of Young's modulus and the thickness ratios between an adhesive and adherends on the stress distributions at the interfaces are examined. As the results, it is seen that the stress singularity caused at the edges of the interfaces increases with a decrease of Young's modulus of adherends. In addition, the stress distributions in joints are analyzed by the finite-element method. A fairly good agreement is seen between the analytical result and the result obtained by the finite-element method.

抄録全体を表示

Plastic-strain-controlled low cycle fatigue tests were carried out using cylindrical specimens of two kinds of spheroidal graphite cast iron (ferritic ; FDI and pearlitic ; PDI). The relationship between plastic strain range and fatigue life was investigated in the two cast irons, and the propagation of small cracks was measured through successive observations by the plastic replica method. The fatigue life satisfies the relation of Manson-Coffin's type and the fatigue lives of the two cast irons are almost equal. That is, the difference in fracture ductility does not reflect the fatigue life. The behavior of a main crack is the factor controlling fatigue fracture and most part of fatigue life is consumed by the growth of a small crack. Crack growth rate dl/dN of the main crack is determined uniquely by σ_aⁿl in these materials.

抄録全体を表示

In the present paper, in order to elucidate micromechanically the crack length dependence of threshold stress for no crack growth in steel, we propose a main crack-slip band interaction model which consists of a main crack, a dislocation free zone, and a slip band. Analyzing the above model by the method based on the continuously distributed theory of dislocations, we obtain the distribution functions and the stress intensity factors K both at the tip of the main crack and the slip band in a closed form. Using these K values, we can elucidate the effective range of the above interaction model and the crack shielding effect by a slip band. Furthermore, the analytical expression for the crack length dependence of threshold stress for no crack growth, which involves the effect of the crack length and grain size, can be obtained successfully by using the present model.

抄録全体を表示

The mechanical properties of Ti-15V-3Cr-3Sn-3Al alloy depend considerably on heat treatment conditions, that is, solution treatment temperature and age hardening method. Thus, rotating bending fatigue tests have been performed to investigate the effect of heat treatment on the fatigue strength of this alloy in the high cycle range. The fatigue crack initiates at the β single phase in β matrix, because the strength of β phase is lower than that of a phase and the number of slip systems of β is higher than that of a phase. Therefore, the fatigue strength increases with increasing the (α+β) microduplex structure in β matrix under the conventional heat treatment. From the above results, the optimum heat treatment condition for this alloy was determined to be as follows : heating at 800°C, which is slightly higher than β transition temperature, for 20 min and quenching in water as a solution treatment, with subsequent heating at 490°C for 8hrs and air cooling as an aging treatment.

抄録全体を表示

This paper is concerned with a transient thermal stress problem of a rectangular plate under nonuniform heat supply and the fracture behavior (stress intensity factor) of the plate with a crack under the axial compressive stress state. Assuming that the crack is sufficiently small and is closed for a compressive stress field, then the temperature field is not disturbed by the crack. Theoretical analyses for the temperature and the stress field are developed on the basis of the two-dimensional plane-stress problem using a heat conduction equation and Airy's stress function method, and the stress intensity factor is analyzed for the biaxial compressive stress field. As an analytical model, we consider the mechanical boundary condition of prescribed displacement, and estimate the stress intensity factor of the crack tip using parameters of the crack configuration such as the location, direction, length and coefficient of friction. These numerical results are shown in graphical form.

抄録全体を表示

Mesh dependence of local approach for creep crack growth analysis of a perforated plate without initial stress singularity was investigated. According to the results of calculations using various finite, element meshes, it was observed that the effects of the element types and mesh patterns were insignificant, and the creep crack growth rates showed significant dependence on the size of crack tip elements. The essential cause of this mesh dependence was attributed to the different stress concentrations in front of the growing crack (or the completely damaged zone) for different mesh sizes. The effects of elastic-damage coupling on the mesh dependence were also examined. Finally, a simple stress limitation method based on ideal plastic deformation was proposed to regularize the mesh dependence of crack growth analysis, and was shown to be an effective method for this purpose.

抄録全体を表示

Tensile creep and torsional creep rupture tests of a commercial austenitic heat-resistant steel, JIS SUH 38, were conducted at 973 K, using two types of specimens : one with a weak rectilinear grain boundary (WQ) and the other with a strong zigzag grain boundary (FA). It was found that the tensile creep rupture strength of FA specimens was considerably higher than that of WQ specimens, and that all specimens tested showed brittle grain boundary fracture. On the other hand, in torsional creep rupture tests, WQ specimens had comparatively higher strength than FA specimens at a higher stress level, while at lower stress levels, WQ specimens became weaker than FA specimens. In the present test, WQ specimens displayed ductile transgranular fracture at higher stresses, but with decreasing stress, they showed brittle intergranular fracture. In all FA specimens, however, the fracture modes were relatively ductile transgranular irrespective of the stress level. These results were explained by the difference in mean normal stress between tensile creep and torsional creep rupture tests, as well as the difference in grain boundary strength between WQ and FA specimens.

抄録全体を表示

An application limit of linear notch mechanics in elastic-plastic solids is proposed, based on theoretical and FEM elastic-plastic analyses ; this limit corresponds to the condition of small-scale yielding in linear fracture mechanics. Moreover, tensile tests of notched plates in practical materials, which are advanced thermoplastics and their reinforced materials by means of short carbon fibers, are carried out. Then, the proposed equation for the application limit of LNM in real materials is examined from the experimental fracture data.

抄録全体を表示

This paper deals with numerical solutions of singular integral equations in interaction problems of diamond-shaped holes with angular corners under various loading conditions. The problems are formulated as a system of singular integral equations on the basis of the body force method. In order to analyse the problems accurately, the unknown functions of the body force densities are expressed as a linear combination of two types of fundamental density functions and power series, where the fundamental density functions are chosen to express the symmetric stress singularity of 1/γ^1-λ1 and the skew-symmetric stress singularity of 1/γ^1-λ2. The stress intensity factors (SIFs) of the angular corners are systematically calculated for various shapes, spacings and numbers of the diamondshaped holes subjected to uniaxial tension, biaxial tension and in-plane shear in comparison with the SIFs of ordinary cracks. The present method is found to give rapidly converging numerical results for the interaction problem. For all types of diamond-shaped holes, the maximum SIF is shown to be linearly related to the reciprocal of the number of diamond-shaped holes.

抄録全体を表示

Stress intensity factors around three cracks, which are situated perpendicular to the stress-free surfaces of an infinite elastic strip, have been determined. Two cracks which are small and of equal length, are placed ahead of a large crack so as to allow for geometrical symmetry. To solve the problem, two solutions are combined. One of them is a solution for a center crack in an infinite strip and the other is that for two collinear cracks in an infinite strip. The Schmidt method is used to satisfy the boundary conditions on the crack surfaces, while the conditions at the plane surfaces are satisfied with use of the Fourier transform method. Stress intensity factors are calculated numerically for some of these crack configurations. Stress Intensity Factor, Three Cracks, Strip, Schmidt Method, Integral Equation, Fracture Mechanics

抄録全体を表示

A simplified identification method by the boundary element method is proposed to estimate the residual stress distribution due to butt-welds of thin plates. In general, the residual stress distribution is produced by the shrinkage source in the plate after welding. Therefore, if the value of the shrinkage source is known, the residual stress distribution can be obtained from the shrinkage source by using a numerical analysis method. In this paper, the shrinkage source is replaced by the body forces of the boundary element method, and the body forces are identified by a few measured stress that can be obtained by, for example, X-ray method. The validity of the proposed method is demonstrated by numerical simulation. By this simulation, the applicability and effectiveness of the proposed method is confirmed from examples without error. Also, the same numerical simulation with error is conducted. From the result, if the number of measured points was more than ten, a better agreement between the assumed and the identified values could be obtained.

抄録全体を表示

This paper concerns a construction of adaptive boundary elements for the two-dimensional problems with mixed boundary conditions such as heat transfer condition between heated body surface and surrounding medium. The scheme is an extension of the method proposed by the present authors, called sample point error analysis and related error indicator for the h-version mesh refinement. Examples in steady heat conduction and linear elastostatics are considered.

抄録全体を表示

An incremental boundary element method for finite deflection analysis of von Karman-type thin elastic plates is presented. Since the governing equations involve the coupled in-plane and out-of-plane deformations as the nonlinear terms, the boundary integral equations are formulated in terms of the incremental quantities by using the fundamental solutions for linear parts of the differential operators. In order to implement the higher-order interpolation function, the boundary integral equations are regularized by using the subtracting and adding-back technique. The boundary integral equations have unknown displacements also in the domain integral terms. Therefore, the boundary-domain element method is applied to solve the discretized boundary integral equations. Some numerical results for sample problems are presented and obtained results are compared with the available analytical solutions to show the effectiveness of the proposed method.

抄録全体を表示

The analysis of a two-dimensional elastic medium containing an elliptical elastic ring with free or fixed inner boundary under uniform antiplane stress at infinity is performed. This analysis is based on the complex variable method using a conformal mapping technique, and assumes the sliding boundary condition between the elastic ring and the matrix. Using this solution, we show the effects of thickness and elastic properties of the elastic ring on distributions of stress and displacement.

抄録全体を表示

Sensitivity analysis is important in inverse analysis and optimization. In such problems, an iterative procedure is employed to find the minimum value of the cost function, which requires accurate and efficient computation of sensitivities with respect to unknowns or design parameters. We consider an isotropic, homogeneous, linear elastic body which is subjected to time-harmonic excitations. Differentiating the regularized boundary integral equations for these elastodynamic problems, we obtain the relationships between the displacements, tractions and design sensitivities. The accuracy and usefulness of the proposed method are demonstrated through sensitivity analysis of some examples in the two-dimensional elastodynamic problem. Finally, the method is applied to optimal shape design for reduction of stress concentration.

抄録全体を表示