This work is concerned with the determination of transient thermal stresses in a long circular cylinder inserted into a short rigid heated ring. It is assumed that the heating condition is asymmetrical with respect to θ. The problem is formulated in terms of a integral equation using the Fourier series, Fourier transform and Neumann series. The radial, hoop and axial stresses have singularities at the end of a short rigid ring on a cylindrical surface. The factors are defined to evaluate the singularity of stresses, and the relation between these factors is derived.

View full abstract

Structural ceramics are the materials most expected to realize improvements in the efficiency and reliability of heat enigines. This report presents a transient thermal stress analysis of circular ceramic rings, the inner and outer boundaries of which are heated by a hot environment through a given heat transfer. Taking four kinds of ceramics, silicon nitride, silicon carbide, zirconia and alumina, it focuses on how the material charcteristics of those ceramics influences on the thermal stresses which arise in the circular ceramic rings. Since the temperature distribution depends on the thermal conducutivity and stress components, a quantity "αET" (product of thermal expansion coefficient, Youngs modulus and temperature), the computed results show that the maximum values of thermal stress in the zirconia and alumina rings are considerably higher than the ones in the silicon nitride and silicon carbide rings.

View full abstract

The microscopic and macroscopic residual stresses in metals deformed plastically under multi-axial stress are studied with the models of polycrystals. In the analysis, the Taylor model is adopted for the plastic deformation and the uniform stress and uniform strain models are used for the unloading process after the plastic deformation. The state of microscopic residual stress in a grain changes with the orientation of the crystal and it is much affected by the condition of yield stress in the plastic deformation. The microscopic residual stress is different from the macroscopic one. In the case where a compressive residual stress is observed macroscopically, there exists a tensile residual stress in a particular grain. The residual lattice strain is not linearly related to sin² φ in X-ray stress measurement. The residual lattice strains measured by X-rays agree with the results analyzed by the Taylor model and the uniform stress model.

View full abstract

The quasi-static loading of a curved strip compressed by a flat, rigid plate is considered, with particular reference to large deformations and the ensuing buckling behavior. Experiments were performed on curved strips of constant width, but of different thickness. The strips were initially deformed to a fixed radius of curvature and stress was relieved before pinning the ends. The deformation characteristics have been analyzed using an incremental finite element technique. Particular attention has been paid to modelling the situation when a node contacts the plate and the condition for separation of the strip from the plate. The predicted loads and deformation modes agreed well with experimental results from tests on steel and aluminum specimens. The experimental and theoretical procedures are pertinent to the study of dent resistance of sheet metal stampings, particularly for automotive panels.

View full abstract

The infinitesimal free longitudinal or torsional vibrations of a thin straight rod, and the infinitesimal free transverse vibrations of a taut string are all governed by a Sturm-Liouville differential equation, and their discrete approximations by a second order difference scheme. The inverse procedure is presented for constructing the mass and stiffness parameters of the discrete model from a knowledge of the natural frequencies and the static displacement of the system.

View full abstract

Inverse analysis using the boundary element method is considered theoretically very useful for the qantitative estimation of complicated stresses in a solid material induced by machining, heating and so on. However, a more refind fundamental consideration is still needed for a reliable evaluation of the stresses. In order to find a more practical estimation method, the authors have carried out several numerical simulations by using rectangular plate models with residual stresses induced by spot heating in the plate. They have also made an improved identification program of the boundary element method. Through typical calculations using this program, they found that unknown residual stress distribution in the plate was identified with enough accuracy from input data of the boundary displacements and tractions.

View full abstract

The applicability of a local approach to the analysis of creep crack growth in a plate is discussed from the view point of continuum damage mechanics. The anisotropic creep damage theory developed by the present authors is implemented into the finite element analysis based on two-dimensional simplex elements and the creep damage process leading to the initiation and propagation of creep cracks in a thin plate with an inner crack under proportional and non-proportional loadings is elucidated. The dependence of the rupture times and the crack extension pattern on the mesh discretization are discussed in detail.

View full abstract

The characteristics of small fatigue crack growth for a high strength steel SNCM 439 were investigated. The tests were conducted under constant stress range or constant stress intensity range conditions at a cyclic frequency of 1 Hz in room air and a 1 % NaCl solution. Crack growth rates of small cracks were accelerated markedly in the 1 % NaCl solution, as compared to those of long cracks under the same conditions. The crack growth rates under constant stress intensity range in the 1 % NaCl solution decreased with increasing crack length. This result indicates that the chemical driving force decreases as the crack grows. Under static SCC, the small crack grew at a stress intensity below K_{I SCC} of the long crack. It is suggested that the SCC behavior of the small crack is responsible for the chemically short crack.

View full abstract

Rotating bending fatigue tests of a low carbon steel were carried out on plain specimens and specimens with a small blind hole at room temperature and at elevated temperatures of 200°C, 375°C and 450°C. The crack growth rate is proportional to σⁿ_al(σ_a : stress amplitude, I : crack length, n : constant) at each elevated temperature similarly as at room temperature. The values of constant 'n' are nearly equal to 8.2 at room temperature, 200°C and 375°C, and 5.6 at 450°C. Moreover, the cracked surfaces in the former temperatures are of the wavy type and those in the latter temperature are of the planar type.

View full abstract

Software for the automatic extraction of intergranular failures from a mixed mode of inter-granular and transgranular ones is developed based upon computer image processing and pattern recongnition techniques. Without human assistance, calculations of intergranular fracture area fractions can be done quickly and accurately by computing co-occurrence matrices. Using this technique, the influences of stress cycle frequency on dynamic SCC crack growth, and those of stress wave form on cyclic SCC crack growth are investigated in a high-strength steel, 300M (σ_B=1310 MPa), sentitive to hydrogen embrittlement type SCC. At higher stress cycle frequencies of f≥30 Hz, threshold values, K_DSCC, are smaller than the static SCC threshold, K_ISCC. However, at lower stress cycle frequencies of f≤5 Hz, the repassivation of oxide films becomes dominant, resulting in an increase in K_DSCC up to K_ISCC. Threshold values for cyclic SCC crack growth are independent of stress wave form, including superimposed waves. At intermediate stress intensity factor ranges, however, crack growth rates increase with increasing stress rise time. In the case of a superimposed wave, a negative pulse wave with a superposition of small vibratory stresses, crack growth rates are much accelerated by promoted dissolution at the crack tips.

View full abstract

Fatigue crack propagation behavior in a tensile residual stress field has been experimentally examined, using plate specimens with residual stress distributed in a profile of thermal stress type. The fatigue crack growth law, using both range ΔK and maximum K_max of the stress intensity factor, has been obtained from fatigue tests under various tensile mean stresses in specimens without residual stress, and thereby changes of fatigue crack propagation rate due to tensile residual stress have been evaluated in terms of the value of K_max which can take into account initial residual stresses distributed along a crack, at relatively lower stress amplitudes. At higher stress amplitudes, however, the residual stress distribution in front of the crack tip has resulted also in changing the fatigue crack propagation rate in accordance with changes of plastic deformation at the crack tip, which has been discussed from a fatigue crack analysis based on the Dugdale model in which initial residual stresses can be taken into account in opening a fictitious crack governed by the surrounding elastic field.

View full abstract

First, the effects of nonmetallic inclusions on the fatique strength of metals are reviewed and the influential factors are revealed. Next, it is emphasized that the effects of nonmetallic inclusions must be analyzed from the viewpoint of small defects or small cracks, because the threshold condition at the fatigue limit is not the condition for crack initiation but the condition for the non-propagation of a crack emanating from defects or inclusions. Finally. from this point of view, the equation for the prediction of the threshold stress intensity factor range ΔK_th and fatigue limit σw for defects and small cracks was applied to predict those for inclusions contained in high strength steels. It is demonstrated that the square root of the projected area of inclusions and the H_v of the matrix are the crucial parameters to predict the fatigue limit of metals containing inclusions. The predictions by the proposed equations were in very good agreement with the experimental results obtained from the fracture surface showing fish-eye patterns. The reasons why the fatigue limit σw of high strength steels does not increase linearly with increasing hardness, and why the scatters of fatigue limit are so large are made clear.

View full abstract

The author has reported the effect of laser welding on the prevention of the fatigue fracture of cracked shafts of a low carbon steel (JIS S 15 CK) in the previous paper. This treatment was found to be very effective for the low carbon steel. In this investigation, the laser welding was applied to specimens of a high carbon steel (JIS SK 3). After specimens were welded at the cracked portion using a CO₂ laser, static tensile strength and rotary bending fatigue strength were measured. Since the welded region became extremely hard and brittle by laser welding, both strengths of welded specimens were lower than those of non-welded virgin specimens. However, the strengths become higher than those of virgin specimens after being subjected to tempering. It was found that when tempered at 600°C, the rotary bending fatigue strength of specimens with a crack smaller than 12 mm was higher than that of the base metal, and when tempered at 700°C, the static tensile strength of specimens with a crack smaller than 12 mm was higher than that of the base metal. Therefore, it is concluded that laser welding is effective in preventing the fracture of high carbon steel with fatigue cracks.

View full abstract

The dynamic stress intensity factors are presented for the fast running crack of the double-cantilever beam type specimens with various heights machined from the model material PMMA. The dynamic stress intensity factors decrease at the beginning of the crack propagation phase, remain almost unchanged in the next stage, and then increase in all specimens. After this period the behavior of the stress intensity factors varies with the specimen geometry. The values of the stress intensity factors at three characteristic points in the stress intensity factor-crack extension curve are discussed with relation to the crack arrest.

View full abstract

Deformation of the craze during fatigue cycling was measured continuously at the fatigue crack tip in polymethylmethacrylate by an optical interference method. Based on the line-zone craze model proposed previously, the measured craze contour was analyzed. With the craze stress, being discretized along the craze length, simultaneous non-linear equations were solved numerically. The craze stress concentrates gradually near the crack tip with applied load, which may cause fibril breakage there, and hence crack propagation. At an almost unloaded condition, about half of the craze length was subjected to compression, but the craze-crack boundary was still clearly distinguished. Any value of the elastic modulus of the craze fibril gives similar craze stress, but only the appropriate value, 100 MPa, resulted in the invariant original craze thickness with the load. The geometry of craze contours may also be used to evaluate the stress state around the crack tip without referring to the applied load. More than 30 % of the applied stress intensity was estimated.

View full abstract

Crack branching in epoxy (Araldite D) was studied by the method of caustics in combination with high-speed photography. Conditions for crack branching were examined using values such as crack velocity a^^·_b and crack extension resistance R^*_b at the onset of crack branching. No characteristic critical values could be defined for R^*_b or a^^·_b alone. Instead, it is found that the product R^*_b· a^^·_b, i. e., dissipated energy per unit crack front and unit time, becomes constant when the crack branching occurs, and that the faster the crack, the larger the number of successful branches.

View full abstract

Crack extension resistance in fiber-reinforced composites is considered from the view point of crack surface interactions, which may be induced by some micromechanisms such as fiber bridging behind the crack tip. Using a concept of interactive stress, equivalent cohesive stress is evaluated as a function of the relative displacement between upper and lower crack surfaces. With some experimental results of crack extension parallel to fibers in a unidirectional continuous fiber-reinforced epoxy composite, we conclude that this inherent relationship between cohesive stress and relative crack surface displacement completely determines the crack extension behaviour of the composite. We also compare opening mode crack extension with in-plane-shear mode crack extension, and find that in-plane-shear mode cohesive stress is much larger than that of opening mode. Furthermore, in-plane-shear mode cohesive stress is shown to decline sharply under large relative crack surface displacement, while opening mode cohesive stress tends to remain a constant value. This difference of cohesive stresses accounts for the characteristic behaviour of crack extension in each mode.

View full abstract

The strength evaluation of single lap joints between glass cloth laminated plates (GFRP) and metals (carbon steel, copper, brass, aluminum alloy) bonded with epoxy resin was investigated both analytically and experimentally. The stress and strain distributions under tensile shear loads of the joints were analyzed by applying the finite element method. The strength of the joints was predicted by applying the strength laws of GFRP, metals, the adhesive layer and their interfaces to the calculated stress distributions. The calculated strain distributions in the joints were in good agreement with the experimental observations. The predicted strengths of the joints were close to the experimental results of the initial cracking load. On the basis of the results obtained, the diagrams for the relation between the joint strength and the adherend stiffness were obtained.

View full abstract

This paper is concerned with the complete adhesion problem when an elastic space is subjected to moment by an elliptical rigid punch. By using the Abel transform, we reduce the problem to solve the Hilbert problem with infinite unknown functions. A general method for solving the problem is given. Numerical results are shown for distributions of contact stresses and surface displacements in several cases of the punch's aspect ratio.

View full abstract

Slow strain rate tests (SSRT) have been carried out on high strength steel HT80 in air and the synthetic sea water at the free corrosion potential Ecorr, and cathodic protection potential, Specimens tested were subjected to fractographic examination to examine SCC crack on a fracture surface. The objectives of this study were to investigate the suitable cathodic protection condition and its dependence upon the flow condition of sea water. Furthermore, emphasis was placed on finding the most suitable evaluation parameters for SCC susceptibility from SSRT characteristics, considering the effect of size. From this study, fracture mechanisms seems to be changed with shifting potential to the cathodic direction, and maximum crack length was increased at more basic potentials. It was found that the most suitable parameters to evaluate SCC susceptibility are reduction in area Ra and the maximum crack length Δa.

View full abstract

A method for estimating the Weibull Parameters of reinforcing-fiber embedded in a matrix has been proposed. In general, the weakest fibers fracture first in FRM, and the strength is variable as shown by tensile-testing several times. That is to say, the strength is a random variable based on the order statistics, and the Weibull parameters can be estimated by applying a single Weibull distribution to it. B₄C coated boron fiber reinforced Al composites produced by hot-pressing were employed as the test materials. The fracture of the weakest fibers was identified by the AE method. The result showed that the shape parameter obtained from the present method gave a smaller value compared to that obtained from the conventional extracted-fiber test, and the scale parameter gave a larger value. Using these Weibull parameters, the strength distribution curves of the composites, reported by Harlow-Phoenix, were predicted. The curves by the present parameters were in good agreement with the experimental data of the composites, as compared with the those by the conventional test parameters. Additionally, the efficiency of the present method was discussed from the Weibull parameters converted by the moment method, on the condition that the fiber strength was given by the bi-modal Weibull distribution.

View full abstract

Pulsed holographic microscopy is applied to take an instantaneous microscopic photograph of the neighborhood of a crack tip propagating in a PMMA plate specimen at a speed of several hundred meters per second. From the photograph, crack opening displacements (COD) are measured in the vicinity of the crack tip, and the dynamic fracture toughness K_ID is calculated through the formula of COD in the singular stress field of a propagating crack. Two types of optical systems are used for holographic recording of cracks. The values of K_ID obtained through one optical system are in agreement with those obtained through the other. One of the holographic optical systems can simultaneously record the caustic of a fast propagating crack at a moment when the crack tip is recorded as a hologram. The value of K_ID caluculated from COD measurement is roughly in agreement with that from the caustic method.

View full abstract

Recently the authors have derived the path-independent J' integral, which gives the energy release rate for a dynamically propagating crack. The authors have also shown the exsistence of many path independent integrals which do not give the energy release rate. Using the asymptotic solutions for elastodynamic crack propagation, the relations between the various path-independent integrals and the instantaneous stress intensity factors can be determined. Howevor, those relations may depend on the shape of the near-field path which shrinks to the crack-tip. In this paper, the shape independence of the J' integral was numerically verified. An analytical proof for the shape independence of the crack-axis component J'₁ was also given. The present paper also shows the finite limits of apparent non-integrable singularities in the J' integral.

View full abstract

The virtual work principle of rate type is discussed. The incremental elastic-plastic finite element formulation (EP-FEM-F) for the analyses of large deformations is presented with respect to the constitutive equation proposed by the author in the previous reports (MG c. e.) . It becomes non-linear in the terms of the nodal velocities. A linearization of MG c.e. is given and substituted in the EP-FEM -F. A new computer program is constructed for the analyses of the two-dimensional and axi-symmetrical forming processes. By the use of the program the upsetting of cylindrical bars, the lateral compression of the rectangular bars into cylindrical shape and the forging of the cylindrical bars into spheres are analyzed for J2-F and MG c.e.

View full abstract

This paper deals with the numerical analysis of large elastic-plastic deformation of shells in contact with each other. The penalty function method is used in order to treat accurately the contact condition, involving a finite amount of sliding. Axisymmetric large deformations of contacting spherical shells are analysed, and it is shown that elastic unloading resulting from the sliding affects definitely the crushing of one shell due to another. The possibility of non-axisymmetric bifurcation of deformation is examined on the basis of J2-deformation thoery.

View full abstract

In the present paper, the finite element method is applied to the creep buckling of a partial spherical shell subject to a uniform external pressure. In the analysis, plastic strain is included in addition to elastic and creep strains. Not only the axisymmetric mode but also the bifurcation mode are considered in the creep buckling analysis. Three types of plastic constitutive equation that is, the flow theory the modified flow theory, and the deformation theory are used to predict the critical time of creep buckling. It is found from the analysis that the critical times of the modified flow theory and the deformation theory are smaller than that of the flow theory.

View full abstract

The formability of two sheet steels, SSPDX (0.8 mm thickness) and SPMY (0.9 mm thickness), are examined by means of a plane-strain tensile test and an in-plane stretching test. The parameter m of Hill's non-polynomial yield function is calculated by applying Wagoner's method. The assumption m=2 is recognized for both steels because the difference between effective stress-strain curves of the uniaxial and plane-strain test can almost be neglected. The experimental data of the in-plane stretching test are in good agreement with those predicted by modified M-K analysis. Because of less surface roughness growth, the limit strain for SSPDX is greater than that for SPMY. The fracture mechanism of the in-plane stretching specimen observed is almost identical to those for isotropic fcc metals reported. However, that of the plane-strain tensile specimen is quite different. The crack is initiated in the center of the specimen by shear deformation and then propagated in the manner of a mode I crack. This difference is well explained in terms of anisotropy and strain distribution.

View full abstract

In the injection molding process, when the melted resin solidified by the cooling effect from the die, non-equivalence of the cooling history caused the Sink-mark growth in the resin, near boss structures. In this study, relations between Sink-Mark depth in the boss structure and some molding parameters are experimentally evaluated, and especially the effect of the forming pin is considered. Also, cooling simulation by thermal analysis is done. In considering the P-V-T characteristics and cooling history of the resin, the experimental results are quantitatively explained.

View full abstract

Instrumented Charpy tests and static J_IC tests were conducted on fatigue-cracked and V-notched specimens of A508 Cl.3 steel at 100°C· A proposed key-curve method to compare the P-S curves of the two types of specimens and a partially loaded instrumented Charpy test using a newly designed stopper proved to be applicable not only at the transition temperature, but also at the ductile fracture temperature region, regardless of testing speed. Also, a simple formula to evaluate the Ernst's Modified J-integral was derived, and showed to have a sufficient accuracy well beyond the J-control led crack growth regime.

View full abstract

Many studies have been done on the acoustoelastic technique to measure residual stresses. However, the influence of the principal direction of the material anisotropy is not taken into account for the measurement. In this study, the influence of the material anisotropy on the acoustoelastic effect are evaluated. The residual stresses of the welded plate. of carbon steel are measured. The results obtained are as follows. (1) The influence of the material anisotropy an the acoustoelastic effect is significant far a welded plate of carbon steel. In this case, first, the polarization directions and the relative difference in velocity should be measured. Then, the principal directions of stress and the difference in principal stresses should be calculated using analytical results by Iwashimizu. (2) The sign of the relative difference in velocity in the weld metal is opposite (positive) to that in the base metal. The absolute value of the relative difference in velocity in the weld metal is larger than that in the base metal.

View full abstract

A systematic theory to describe the anisotropic damage states of materials and a consistent definition of effective stress tensors are developed. By introducing a fictitious undamaged configuration mechanically equivalent to the real damaged configuration, the classical creep damage theory is extended to the general three-dimensional states of material damage. It is shown that the damage state can be described in terms of a symmetric second rank tensor. The physical implications, mathematical restrictions and the limitations of this damage tensor, as well as the effects of finite deformation on the damage state are discussed in some detail. The notion of the fictitious undamaged configuration is then applied to the definition of effective stresses. Finally, the extension of the effective stresses incorporating the effects of crack closure is discussed. The resulting effective stress tensor is employed to analyze the stress-path dependence of the elastic behavior of a cracked material.

View full abstract

The shape of a structure must be determined optimally from the viewpoint of safety and economy. Especially if we have uncertainty concerning the loads or strengths, the conventional treatment which is represented by the safety factor can not satisfy the requirement. This report proposes a method for the determination of a structural shape for the case where the structure is subjected to static loads whose magnitudes vary according to the normal density function and excited by random loads which cause dynamical response, and where the local sizes, material properties and strengths are also assumed to be random variables which are governed by normal distributions. The object of the design is to minimize the volume of the structure; and as design constraints, the probabilities of the first excursion failure and fatigue failure are taken into consideration so that they don't exceed a certain value. In the numerical treatment, an effective procedure is used in order to obtain pseudo-optimum shapes.

View full abstract

A new general-purpose computer program for optimum structural design synthesis is developed. Its principal components are, in general terms, optimizer and analyzer routines. The analyzer routine provides considerable flexibility and efficiency in the computational treatment of structural design problems. Special processor interfacing of the analyzer to the optimizer is composed in the system. By applying this programming system, an optimum design problem on the framed structure of the motor cycle to find the framed structure that will maximize the torsional rigidity subject to the constant volume, is studied.

View full abstract

A programmable automated X-ray diffractometer has been developed, making use of a micro-computer and a commercial X-ray goniometer, in order to use it for materials evaluation by various new X-ray diffraction techniques. A scalar-timer and stepper moter controllers were newly designed using several highly efficient large-scale integrated devices {LSI} . The use of these devices could simplify greatly the electric circuits of the units, and could improve their reliability and maintenance. The viability of this system was confirmed by using an annealed mild steel and an aluminum oxide specimen. Accurate diffraction line profiles could be obtained by using the fixed time method. The diffraction line broadening of the mild steel caused by stretching and residual stress of the aluminum oxide were measured by the Gaussian curve method. They could be determined by using K_α1 line profiles for materials whose K_α1, and K_α2 lines are separated

View full abstract

Cavitation damage was studied by both the measurements of cavitation bubble collapse pressures (impact loads) using our developed method with piezoelectric ceramics and the cavitation erosion tests of various metals in a magnetostrictive vibratory facility. Although the size of impact loads produced by bubble collapses distributes over a wide range, a threshold value for impact load which contributes to fatigue failure on the eroded surface exists and is considered to correspond to fatigue limit. Thus, this threshold value moves to the high impact load for harder materials in Al, Cu and steel but scarecely increases with the tensile strength for steels. Comparing the distributions of impact load with the assumed S-N curves of fatigue, Miner's law is realized for the incubation period and the reciprocal of volume loss rate during the stationary stage regardless of the cavitation conditions and the materials.

View full abstract