Fatigue reliability of ferritic ductile iron FDI hardened by Micropulse system was evaluated with an attention to meso-scopic fatigue mechanism. Especially, the influence of the martensite phase around graphite nodules on the fatigue crack initiation and propagation was examined. The main results were as follows ; (1) By Micropulse system, the fatugue strength of FDI was improved, and the fatigue limit of the hardened material could be increased by 30%. (2) The fatigue crack initiation in hardened FDI was induced by slip in the ferritic matrix, because debonding between graphite and matrix was prevented from the martensite phase around graphite nodules. Thus, it could be interpreted that the crack initiation stress would be increased by improved mesostructure in the surface. (3) The fatigue crack propagation in the hardened layer was restricted by the presence of the martensite phase and compressive residual stress, and the crack propagation behavior was zigzag path in the ferritic matrix.

View full abstract

The fatigue property of Al-Si powder metallurgical alloy at the elevated temperature was investigated with a special view of mesoscopic fracture analysis. Especially, the influence of difference of manufacturing process (extrusion and forging) on the fatigue property was confirmed experimentally. The main results obtained are as follows : (1) The fatigue property of Al-Si powder metallurgical alloy at the elevated temperature was mainly dominated by difference of manufacturing process. Namely, the fatigue strength of forging material was lower than that of extrusion material at the elevated temperature. (2) The lowering of fatigue strength in forging material was attributed to the breaking of a Al-Si-Fe intermetallic compound and not the debonding on the interface between Al matrix and Si particle. A intermetallic compound was shattered in the extrusion manufacturing process. In contrast, a intermetallic compound was not shattered in the forging manufacturing process. These things led to conclusions (1).

View full abstract

SEM servo analysis using in-situ observation on a fatigue mechanism at elevated temperature was carried out. The main results were as follows. (1) An Al-Si powdered alloy exhibits superior fatigue property at elevated temperature. Namely, the fatigue strength ratio, that is the ratio of the fatigue strength to the tensile strength, at 200°C is higher than that at room temperature. The fatigue limit of an Al-Si powdered alloy at 200°C and room temperature isn't recognized. (2) The crack initiation life of an Al-Si powdered alloy represented the majority of the total fatigue life. The fatigue crack was initiated by the debonding on the interface between the Al matrix and Si particle. (3) The transition condition from fatigue to brittle fracture was determined by Griffith criterion. The load energy was spent to surface energy due to the debonding when mesocracks grew to critical crack length of about 200 μm following the final brittle failure.

View full abstract

In the case of anisotropic pure aluminum thin plate specimens, a small inclination between the loading direction and the rolling direction affects the crack propagation path, and the fatigue crack propagates in the rolling direction under appropriate conditions. In this report, this phenomenon is investigated for both constant and cyclic varying load conditions. The results are as follows. (1) Under constant loading conditions, the fatigue crack propagates in the rolling direction when the stress intensity factor range settles at 5.0∼7.0MPa√(m). In the case of varying load conditions, the same phenomenon can also be explained using the equivalent stress intensity factor range. (2) Fatigue crack propagation in the rolling direction under cyclic varying load conditions is determined by the damage ratio calculated using the linear cumulative damage law. The lower limit of the damage ratio at which the crack propagates in the rolling direction is very small and is only about 10^-3 for this specimen.

View full abstract

Three-dimensional surface fatigue crack propagation experiments were conducted for PMMA with residual stresses to examine the life prediction method proposed previously by the present authors. The residual stresses were introduced by adhering three plates together when the central plate was loaded in tension or in compression. These three-layered specimens had tensile residual stresses in the center layer and compressive ones in the other layers, or vice versa. In addition, direct optical observations of the crack closure behavior were conducted. This direct observation was made possible based on the fact that sodium light penetrated straight through fracture surfaces when they were in contact with each other, while it was reflected on the fracture surfaces when they were separated. It was found that the life prediction method worked fairly well for predicting crack propagation behavior in PMMA with residual Stresses.

View full abstract

Rotating bending fatigue tests were carried out for Inconel 718 at room temperature and elevated temperatures of 300°C, 500°C and 600°C to investigate the influence of temperature on the fatigue crack growth resistance. The crack growth rate can be expressed by the small crack growth law, dl/dN=C₁σⁿ_al=C₂(σ_a/σ_B)ⁿl, at all the temperatures, where σ_B, σ_a and l are the tensile strength, the stress amplitude and the crack length, and C₁, C₂ and n are constants. Therefore, the resistance to crack growth was evaluated using the small crack growth law. The fatigue crack growth resistance decreased with the increase of temperature.

View full abstract

A fatigue crack growth test on Al-Zn-Mg-Cu alloys with various grain sizes was carried out using CCT specimens (T-direction) to investigate the relationship between fatigue crack growth resistance and the grain size of the alloy. The main results are as follows : (1) The effect of grain size was observed relationship between fatigue crack growth rate, d (2a) / dN, and stress intensity factor range, ΔK. The fatigue crack growth resistance in these alloys depended on the average grain size, ds. (2) The fatigue fracture pattern changed at d_s=24.1μm. In d_s⪈24.1μm, the intergranular fracture decreased with increasing d_s. On the other hand, the intergranular fracture in d_s>24.1μm increasing d_s, and the de-touched cracks and branching along the grain boundary were observed with increasing d_s. Therefore, the transition behavior of the fatigue crack route was closely related to the fatigue crack growth resistance in this alloy. (3) The relationship between grain size and plastic zone size corresponding to the applied stress level at the crack tip is one of the most important factors in fatigue crack growth resistance of Al-Zn-Mg-Cu system alloys.

View full abstract

The objective of the present paper is to evaluate the effect of a contact edge profile on stress distribution and fretting fatigue crack initiation in press-fitted axles. The press-fitted axles were fatigue tested under rotating bending conditions. The profiles of the axle and boss contact edge, the crack initiation site and the fretting corrosion region were investigated after fatigue testing. Stress states of the press-fitted axles were calculated by the finite element method (FEM), taking into consideration the contact edge profile. It was found that the maximum axial stress increases and its location shifts from the contact edge to the inside with increasing contact edge rounding. Moreover, crack initiation sites well correspond to the maximum stress location calculated by FEM considering the contact edge profile.

View full abstract

The energy-release rate (ERR) due to transverse matrix cracking in a 90°layer of FRP crossply laminate is theoretically calculated. The ERR of the whole laminate is derived for two cases of displacement-controlled and dead-load conditions. The exact expression of ERR is derived by a precise examination of load-displacement characteristics taking account of residual displacements due to relaxation of thermal residual stresses. For displacement-controlled tensile testing, the energy released by cracking is equivalent to the change in the strain energy of the laminate ; therefore, the ERR of the whole laminate is divided into two parts, i.e., those of the 90°layer and the 0°layer. The ERR of each layer is theoretically calculated for use in parts II and III in order to formulate the evolution law of cracking damage.

View full abstract

In part I, the energy-release rates (ERRs) of the whole laminate and of the 90°and 0°layers were theoretically obtained. Part II discusses an experimental determination of the ERRs. Real-time measurements of cracking stress and positions of all cracks are conducted via making use of GFRP crossply laminates in which the transverse matrix cracks are visible to the naked eye. Then the ERR can be calculated using the theoretical results obtained in part I and the experimental data of cracking stress and positions of cracks. It is found that the ERR of the whole laminate is not constant but increases with increasing crack density ; therefore, the fundamental assumption made by Nairn is not appropriate. It is also found that the ERR of the 90°layer is constant during damage evolution, which will serve as a material parameter yielding an evolution law of the cracking damage in part III. The characteristics of the energy-release rate of the 0°layer are also discussed.

View full abstract

In parts I and II of our study, the energy-release rate of the 90°layer has been theoretically calculated and experimentally shown to be constant during cracking damage evolution. Part III establishes the damage evolution law utilizing the experimental fact that the energy-release rate of the 90°layer is constant during damage evolution. Taking account of the statistical nature of cracking position, it is possible to obtain an analytical expression of the relation between the laminate stress and the crack density, i.e., the damage evolution law. The theoretical damage evolution law shows excellent agreement with experimental results of the present study. The experimental results obtained by other researchers can also be explained by the proposed evolution law.

View full abstract

Stress analysis was conducted for a traversely reinforced composite plate with an interface crack at the center under tension in plane strain condition by the boundary element method (BEM). The results obtained are summarized as follows ; (1) When the normalized crack length, a=a/d, (a : crack length, d : thickness of matrix layer) is much smaller than unity, the magnitude of stress intensity factors (SIFs), K₁, K₂, and K_i=(K²₁+K²₂)^1/2, are close to the SIFs for an interface crack in a dissimilar infinite body. The crack is termed as the small crack for composite materials. (2) As a becomes larger, the normalized SIF, K_i decreases and shows the minimum near a=1, where K_i is the SIF, K_i, divided by the SIF for a homogeneous orthotropic body composed of a Mode I crack, K₁h. It increases and converges on a constant value as a increases. (3) The energy release rate calculated from the convergent K_i coincides with that of the crack in a homogeneous orthotropic body where the elastic constants are determined by a mixture of those in the matrix and fibers. The crack is termed as the large crack for composite materials. (4) Since the region of oscillated stress distribution is confined in the vicinity of the crack tip, the stress intensity is represented by K_i, of which values are nearly equal to the SIF of the large crack in a homogeneous orthotropic body.

View full abstract

In this paper, a singular integral equation method is applied to calculate the variation of the stress intensity factor along the front of a 3-D oblique semi-elliptical surface crack. The stress field induced by the body force doublet in a semi-infinite body is used as the fundamental solution. Then, the problem is formulated as a system of integral equations with singularities of the form γ^-3. In the numerical calculation, the unknown body force doublets are approximated by the product of fundamental density functions and polynomials. The results show that the present method yields smooth variations of stress intensity factors along the crack front accurately for various geometrical conditions.

View full abstract

In this paper, the axisymmetric torsion problem of compound infinite hollow cylinder with an external annular crack is considered on the basis of the three-dimensional theory of elasticity. The cylinder is composed of two layers of different materials. The problem is reduced to solution of an infinite system of simultaneous equations. Distributions of displacement and shear stress and, in addition, the stress intensity factor KIII are computed for various ratios of inner to outer material shear modulus and for various ratios of inner to outer radius of the cylinder.

View full abstract

Rate dependence of mode I interlaminar fracture toughness of interlayer-toughened carbonfiber/epoxy laminates (T800H / 3900-2, Toray) was studied over a very wide range of loading rate from quasi-static to impact. A SHPB (Split Hopkinson Pressure Bar) system using a ramped incident stress wave was employed to estimate the accurate fracture toughness at high loading rates. Finite element results proved the validity of a simple estimation formula, which is based on static beam theory, for impact fracture toughness. Experimental results showed that the fracture toughness is strongly rate-sensitive. The fracture toughness decreased stepwise interposing a transition region where the fracture toughness dropped rapidly with increasing loading rate. The decrease of fracture toughness in the transition region was larger in the initial stage of crack growth than in the propagation stage. The crack path was inside the interlayer region in the initial stage, but it shifted to near the interface between the interlayer and carbon-fiber/epoxy lamina in the propagation stage. Rough fracture surfaces including microcracks of resin were observed at low loading rates, but relatively smooth fracture surfaces were observed at high loading rates. The transition of fracture characteristics approximately corresponded to the transition of fracture toughness.

View full abstract

In this study, based on the fracture surface observations using a SEM (Scanning Electron Microscope), a numerical unit model and 3-dimensional non-uniform models for FEM (Finite Element Method) analyses are proposed using Gurson's model as a constitutive equation. The effect of the non-uniformity, the reinforcement volume fraction and the aspect ratio are considered for the quantitative evaluation of the stress-strain relationship of SiC particle-reinforced aluminum alloy. The results show that the dimple fracture process of the matrix aluminum alloy is well simulated after a large amount of plastic deformation. The non-uniformity of the SiC reinforcement locations and the aspect ratio have a strong effect on the local and global damage behavior and the stressstrain relations. It is shown that the fracture strain greatly increases when the aspect ratio of SiC particles is nearly 1.0, and the SiC particles are distributed uniformly. To improve both tensile strength and fracture strain, it is better to increase the SiC particle volume fraction rather than to increase the SiC particle aspect ratio. The numerical model is improved by considering the nonuniformity effect on the tensile stress and fracture strain.

View full abstract

Recently auto-mobile airbag systems are usually used. This paper describes a study to analyse experimentally an effect of airbag as a shock absorber for impact in car crash. Experimental systems are a compact laboratorial type and a practical type for high energy. The former type uses an airgun to make impact force and the latter type falling striker. The fracture behaviors of windshield glass and airbag deformation are recorded by the high speed video-camera. As the results, it is obvious that an initial pressure of airbag and an initial crack in laminated glass have an important effect for deformation and penetration fracture behaviors of windshield glass.

View full abstract

The criterion for cleavage fracture in nickel crystals was elucidated by the atomic analysis in order to determine the evaluation method of strength of micro-materials. The analysis was based on a molecular dynamics simulation using the EAM (embedded atom method) potential. First, a simulation was conducted on a nickel single crystal with several size of the simulation cell under tension along the [001] direction where the transverse deformation was constraint. As the number of atoms in a cluster, n, increases, the magnitude of strain at fracture decreases monotonically. In n=4, the cleavage fracture takes place at the tensile strain, ε=0.263, which is smaller than the critical strain of lattice instability, ε=0.304, calculated on the basis of the Born's analysis. Since the displacement of each lattice have fluctuation, the maximum local strain reaches the critical strain of lattice instability when the global strain is 0.263. In order words, the freedom of mobility brings about the reduction of fracture strain. In 32≤n≤256, the cleavage does not take place due to the transformation which is out of scope in this study. In n≥500, the cleavage initiates from the local lattice where the shear deformation on (111) planes meet when the local strain exceeds the lattice instability. A tensile simulation was also carried out for a nickel with Σ5 twist grain boudaries under traverse constraint. Cleavage fracture at a grain bouudary is observed at ε=0.130 with shear deformation in a grain. It is revealed from the detail observation that the local strain at grain boundaries are much higher than that of grains and the cleavage fracture at the grain boundary takes place when it reaches the lattice instability. Thus, the criterion for cleavage fracture is quantitatively estimated by applying the lattice instability of Born's analysis to the local strain.

View full abstract

Fatigue tests were carried out under two kinds of stress ratios in constant amplitude loading of Ti-6Al-4V alloy after heat treatment of solution at 1223K and aging at 813 K (STA). Retardation behavior of fatigue crack propagation was observed. It is considered that this retardation is caused a stress relieving at front of crack tip by growth of secondary cracks. X-ray fractography, which is a technique to estimate the stress conditions and to be used for analysing a mechanisms when a fracture of structures occurs, is applied on fractured surface. Distributions of residual stress and halfvalue breadth by X-ray technique on the fractured surface along the crack propagation direction as a function of a stress intensity factor. It was found that the residual stress on the fatigue fractured surface had a peak value when maximum stress intensity factor was nearly equal to 25MPa√(m). Strain induced transformation was observed on the fatigue fractured surface from X-ray diffraction pattern. The half-value breadth on fractured surfaces increased with increasing maximum stress intensity factors. Depth of monotonous plastic zone size was estimated from a change of the distribution of the residual stress in direction of depth near fractured surface. Depth of monotonous plastic zone size ω_y was expressed by following equation ; ω_y=a(K_max/σ_y)², where the value of a was 0.17. By using X-ray fractography, maximum stress intensity factor K_max in fatigue crack propagation of Ti-6Al-4V alloy can be predicted.

View full abstract

A topographical observation system with a scanning laser microscope has been newly developed. This system is consisted of a scanning laser microscope, a personal computer having video-capture equipment and an automatic X-Y stage. We applied this system to a 3-dimensional fractography of TiAl creep-cracked specimen and constructed a 3-dimensional image covered a wide area on fracture surface. Furthermore, we tried to simulate a fracture process by using topographical data obtained through this system. Since creep crack initiation and creep voids nucleation around the creep crack were clearly reproduced, the reliability of this simulation system was confirmed.

View full abstract

Fatigue tests were carried out on notched and shouldered specimens of a cast stainless steel under rotating bending and reversed torsion. It is shown that the fatigue strength of a notched or a shouldered specimen having an arbitrary size under the two loading conditions can be predicted systematically based on the concept of the linear notch mechanics (LNM). The main results are summarized as follows : (1) In the case of rotating bending, it was found that the fatigue strength of notched and shouldered specimens having an arbitrary size can be estimated on the basis of the LNM. (2) In the case of reversed torsion, the value of τ_w1 (fatigue limit based on crack initiation) can be estimated successfully using the LNM. But the value of τ_w2 cannot be estimated using the LNM, because the final fracture of the specimen is not controlled by the stress fields of the minimum section alone. (3) The fatigue limit based on crack initiation under the two loading conditions can be estimated systematically using the LNM.

View full abstract

This paper is concerned with a compressive experiment on carbon steel thin walled square pipes with a square hole in a short column range, supported by spherical seats. Collapse of these carbon steel square pipes occurred as the local buckling of the flat wall with a square hole. It was found that the buckling stress decreased as the opening ratio increased. Consideration has been given to the formula and the diagram wich are the combination of the analysis based on the energy method and the experimental result. The measurement results show that the formula and the diagram are in good design data, and are in good agreement with those obtained experimentally for a short column range.

View full abstract

Very few studies on anisotropic elliptical or circular ring have been done because of some difficulties caused by anisotropy. The present paper shows the succussive analytical method by constraint release technique and gives the distributions of stress and the displacement by using Lekhnitskii's complex variable method. The boundary stresses of the elliptic ring are given by arbitrary distributed loads such as hydraulic pressure, trigonometric series and so on.

View full abstract

Intrinsic mechanical stresses exist in a general interface because of interface atomic structure changes relative to the bulk. This paper presents fundamental solutions using the interrelationships between the surface stresses and the bulk stress deduced in the previous paper. The interfacial stresses have a close relationship with the interfacial energy and the geometry (mean curvature) of an interface. In the present paper, fundamental solutions for a prismatic dislocation loop with radius a in a transversely isotropic material are derived using the three-dimensional theory of elasticity. Furthermore, solutions for a two-phase transversely isotropic material are deduced, and potential functions for arbitrary shape loop are derived from the solution. The difference in the stress distribution between cases in which the interface energy is considered those in which it is not was investigated using the solution. All of the stress components which consider the interface energy are larger than those which do. Finally, we discussed the relationship of a dislocation loop with a dislocation produced by locally changing free energy.

View full abstract

Based on analyses of stress wave propagation, cavitation erosion was studied on coating materials such as welding deposit metals and platings. The MDPR (mean depth of penetration rate) increased, decreased or kept constant as a function of the coating thickness depending on the combination of the acoustic impedances of the coating and its substrate. A critical thickness T_c was observed, above which the MDPR kept constant for all coatings. The critical thickness T_c and the MDPR at a given thickness was predicted in terms of physical parameters of the material such as the modulus of longitudinal elasticity and the density of both coating and its substrate. In case of a very thin coating, the time to entire removal of coating layer was greatly influenced by the combination of the acoustic impedances as well as by the erosion resistance of the materials. Thus, it was concluded that the erosion life of the coatings can be predicted from analyses of the stress wave propagation.

View full abstract

A new optimization method, which will globally optimize even a multimodal function, is proposed. Non-linear programming is an excellent way to optimize a convex function, but this is not the case for a multimodal function. A numerical differentiation scheme with variable increment sizes was applied to a BFGS method, which is a popular non linear programming method, in order to overcome the above-mentioned deficiency. A multimodal test function was designed to examine global convergence. In the optimization process, the increments, which were targe in the beginning, were reduced as the process approached the global minimum. It was found that the process works well even for a multimodal function when the initial size of an increment size is properly selected.

View full abstract

Sintered ceramics often have different properties at the near surface from in the middle. With surface waves of different frequency or wavelength, we can estimate the depth-dependent subsurface properties, because the surface wave propagates in surface layer of about one wavelength. In this paper, the leaky surface wave velocities of Si₃N₄ ceramics with various porosity have been measured by a scanning acoustic microscope at 195 and 440 MHz and by time-of-flight measurement with a PVDF line focused sensor at 36 MHz. The velocities measured at 440 MHz are much higher than those at 36MHz, while the latter are very close to the Rayleigh wave velocities calculated with bulk velocities measured at 5 MHz. This shows elastic constant near surface is higher than that in the middle.

View full abstract

A method of measuring leaky Rayleigh wave velocity precisely has been developed with a line-focused PVDF ultrasonic transducer and a digital signal processing. A focused PVDF ultrasonic transducer is constructed by cementing PVDF film on a concave surface. The measurement system is composed of an ultrasonic pulser/receiver, a high-speed digital wave memory and a personal computer. The relative precision of the measurement for the leaky Rayleigh wave velocity of fused silica, the standard deviation divided by the mean velocity, is 10^-4. PVDF ultrasonic transducers have broad band width, therefore, leaky creeping wave and surface SV wave are clearly identified and these wave velocities have been measured simultaneously.

View full abstract

We had developed the computer simulation program in order to estimate the spread of damages by fire and explosion in chemical plant with many tanks. In this paper, a probabilistic estimation approach of the risk based on the developed simulation is described. The risk can be determined with two factors, which are the relative values for equipment in plant and, the occurrence of probability of fire. The probability of fire is calculated with the developed simulation and Monte Carlo method considering variance of heat radiation from fire to tanks due to the inclination of the shape of tank fire by the effect of various velocity and direction of wind. As a numerical example, the estimation approach was applied for a practical plant model with several tank units and pipeline, and the risk in process with time can be evaluated quantitatively. These results indicate that the proposed method is very useful for the evaluations of the safety and reliability of chemical plants.

View full abstract

The effects of mechanical stimuli on fibroblasts were investigated, using the collagen gel culture method. Specimens of thin collagen gel membrane involving cells were supported by a stainless steel wire mesh and subjected to static and dynamic uniaxial tensile/compressional loads in an incubator. The cell alignment, the shrinkage of the collagen gel matrix, and the mechanical properties of the specimens were examined. We found that both of the static and dynamic uniaxial tensile loads significantly strengthened the collagen gel matrix. From morphological evidence, dynamic stimuli were proven to be more effective in matrix strengthening than static stimuli. However, since the specimens were not strengthened uniformly and portions where compressional stresses were induced remained weak, the macroscopic tensile test failed to show the differences between static and dynamic stimuli on mechanical properties of the specimens.

View full abstract