When Ti-6Al-4V alloy was solution-treated and aged (STA) in air, the oxygen contaminated layer which consisted of an embrittle layer and a hardened layer was formed below the TiO2 scale. The influence of the oxygen contaminated layer was studied on the mechanical properties (static tensile properties and fatigue strength). Except the 0.2% proof stress and tensile strength, the static tensile properties and fatigue strength were degraded by the embrittle layer. On the other hand, the hardened layer hardly influenced on the mechanical properties. Furthermore, the influences of mechanical polishing and electropolish were discussed on the fatigue strength of annealed and STA materials. As a result, it was found that the mechanical polishing was effective to the improvement of the fatigue strength of annealed material.
Authors have made a fatigue strength data-base FRINT in the late 1970's and after that developed KAFASAS for the purposes of a common data use and an easy data management. Several fatigue strength data-bases have been developed since then and used in the fields of fatigue strength design and research but most of them impose difficulties on users to modify data structure and renew data. On the other hand, a remarkable progress has been made in hard and soft-ware of computers. Low cost, high speed and powerful personal computers, as well as multifunction data-base soft ware are now available for designer and researcher. In this paper the construction of fatigue strength data is reviewed, and a personal fatigue strength data-base allowing any user to construct and modify the data-base and renew the data is shown. This data-base which contains about 1200 S-N curves on metallic material operates in the environment of any personal computer using the data-base soft ware “Microsoft Excel”.
Characteristics of crack initiation and propagation in rotating bending fatigue tests were investigated on Ti-6Al-4V alloy. Successive observations were made by the plastic replica method to determine the physical meaning of fatigue limit σw, and crack propagation behavior was studied using a small crack growth law. The main results obtained are summarized as follows. (1) When σ is over σw, microcracks appear at α phase. (2) The fatigue life is determined by the crack propagation period except for the stress near the fatigue limit. (3) When σ is equal to σw, no microcracks are observed on the specimen after 107 cycling of stress under a metallurgical microscope. Therefore, the fatigue limit is determined by the limiting condition for the initiation of microcrack. (4) The relation dl/dN∝σnl holds when σ/σ0.2>0.6, and the relation dl/dN∝ΔKm holds when σ/σ0.2<0.5.
Fatigue tests were carried out on 2024-T3 aluminum alloy sheet material under 11 kinds of flight simulation loadings and its crack propagation behavior was observed. In this experiment, flight simulation loadings were made to simplify a complex stress history so that they consisted of the ground to air cycle of large stress range and gust cycles of relatively small stress range. The results obtained were as follows: (1) When the maximum stress of the ground to air cycle was equal to the maximum stress of gust cycles, the crack propagation rate under flight simulation loading was faster than that estimated by the linear damage law on the basis of crack propagation of constant stress amplitudes. (2) When the maximum stress of gust cycles was 40MPa or 80MPa lower than that of the ground to air cycle, the crack propagation rate in a short crack length was almost the same as the estimated value. With increasing crack length, the crack propagation clearly showed the retardation behavior. As the crack length increased further and approached to the critical value of final fracture, the crack propagation rate gradually recovered to the estimated values. (3) When the maximum stress of gust cycles was 120MPa or 160MPa lower than that of the ground to air cycle, the crack propagation rate under the flight simulation was nearly equal to the estimated value and the retardation behavior could not be observed in the whole range of crack propagation. (4) The retardation behavior of crack propagation was not observed just after changing the stress of the ground to air cycle to the gust stress, but the crack propagation clearly showed the retardation behavior with increasing the number of gust cycles in the same flight.
In order to investigate the physical basis of scatter in fatigue life, rotating bending fatigue tests of smooth specimens were carried out using two steels with different microstructures. 16 or 18 specimens were fatigued at each stress range, and the initiation and propagation behaviors of a crack which led to the fracture were examined for all the specimens. The fatigue data exhibited a large scatter in fatigue life Nf at a stress near the fatigue limit. The scatter in Nf is controlled by the behavior of a microcrack, namely it is controlled by the scatter in N0.3 for an annealed steel and N0.1 for a heat-treated steel, where N0.3 and N0.1 are the number of cycles to attain a crack length of 0.3 and 0.1mm, respectively. On the other hand, after the crack became 0.3 or 0.1mm, the propagation life exhibited small scatter, and its scatter was hardly affected by the stress range. To estimate the scatter characteristics quantitatively, the statistical treatment of fatigue data was performed by assuming a Weibull distribution.
The purpose of this research was to study the fatigue behavior and the test method for S2 glass fiber reinforced plastic under an interlaminar shear fracture mode. Fatigue tests were carried out under the short beam shear (SBS) condition of three point bending. The main results obtained are as follows. (1) Interlaminar shear fatigue fracture was realized for SBS specimens with the span ratio of L/t=4 and 3.5. (2) Fatigue crack propagation was detected during interlaminar shear tests by using scanning acoustic tomography. (3) In interlaminar shear fatigue tests under the condition of varying stress amplitude, the cumulative fatigue damage calculated by Miner's law was approximately unity when the scattering of fatigue data was taken into consideration.
Dynamic behavior of cathode ray tube (CRT) under the ball impact test condition was analyzed by computer simulation. Experiments were also carried out by using an instrumented drop-weight impact testing machine to investigate the impact load and stresses of CRT, and the phenomena of impact fracture of CRT were discussed. The following results were derived. (1) Dynamic behavior of CRT under several impact load conditions changing the impact speed and the hitting point was clarified by using the instrumented testing machine. (2) The results of computer simulation with the impact analysis program DYNA3D showed good agreement with the experimental dynamic behavior of CRT. (3) Static load due to vacuum in the tube or tightening force by the tension band had no effct on the dynamic characteristics of CRT before breaking. (4) It was explained quantitatively that hitting the corner of panel face at the impact test was more severe to the dynamic strength of CRT than the hitting the center of panel face. (5) The macroscopic impact fracture mode of CRT depended on the tensile stress at the edge of panel face and a criterion of the implosion was proposed.
Morphology and growth of α plates which appear during annealing in the temperature range from 150 to 300°C were investigated by using repeated electropolishing technique. The results obtained are summarized as follows. All the α plates were triangles with a tip angle of 70 degree, but the opposite side of the tip, or growing front, showed various shapes depending upon individual α plate. The α plates became thicker near the tip, and the cross section of that portion had almost the same thickness. On the other hand, the α plates became thinner near the opposite side of the tip, and the thickness decreased with approaching to the both edges. The values of width and length of α plates were almost the same after the same annealing time below 225°C. On the other hand, above 250°C, the growth rate of the longitudinal direction (length) was faster than that of the transverse direction (width). The growth rate of the horizontal direction of α plates was obviously slower than those of the longitudinal and transverse directions at any annealing temperature. The activation energy for growth of α plates was about 120-156kJ/mol. This value is of the same order as those for diffusion of copper and zink atoms in Cu-Zn alloys.
The effect of Mg addition to the matrix of almina powder dispersed Al-4%Cu composite materials on the aging behavior has been investigated by means of microvickers hardness measurement and transmission electron microscopy. Age-hardening appeared in the composite materials with Mg addition when aged at 150°C, whereas the hardness decreased monotonically at 200°C aging. The aging behavior of the composite material with 0.6%Mg addition was the same as that of the ternary Al-4%Cu-0.6%Mg Alloy. In contrast, the aging process in the composite materials with 1.6 and 4%Mg addition differed remarkably from that of base alloy. That is, GP zone and θ' were observed instead of GPB zone and S' precipitates. Spinel was scarcely observed in the 0.6%Mg composite, and the amount of spinel near almina particle increased with an increase in the amount of Mg addition.
This paper is concerned with the effect of stress ratio on the fatigue strength of butt-welded joints made by two kinds of steels (0.18C steel and 780MPa high strength steel). The effect of mean stress on the fatigue strength of welded joints depends on the type of steel and the value of a notch root radius at the weld toe. The fatigue strength of a welded joint with a small notch root radius at the weld toe is remarkably influenced by the mean stress. In this case, the fatigue limit of the welded joint increases when the positive mean stress changes into the negative. The fatigue strength of a welded joint with a large notch root radius at the weld toe is a little affected by the mean stress. The benefitial effect of TIG treatment on the fatigue strength is more remarkable at the stress ratio R=0 than at R=-1 and also more remarkable in 780MPa high strength steel than in 0.18C steel. These phenomena are closely related to the value of ρ0 which is the value of ρ at the branch point and the material constant depending on both the type of steel and the stress ratio.
Fatigue tests of 0° unidirectional and ±45° cross-ply composites of epoxy resin reinforced by two types of aramid fibers (HM50, Teijin and Kevlar 49, Dupont) were carried out in 71°C hot water. An emphasis was placed on the influences of hot water on the fatigue strength and fatigue mechanism, and the following was made clear. In case of ±45° cross-ply composites, the fatigue strength of both HM50 and Kevlar 49 composites decreased in 71°C hot water from that in 30°C water. It is considered that the fatigue strength of the resin and the interface between fiber and resin became weaker by the exposure to hot water. Both fiber/resin interfacial debonding and interlaminar delamination occurred more distinctly in 71°C hot water than those in 30°C water. In case of 0° unidirectionally reinforced composites, the fatigue strength of both composites became stronger in 71°C hot water than that in air and almost similar to that in 30°C water. It is considered that the resin absorbing water facilitated the local re-alignment of fibers under atensile stress. Kevlar 49 fibers splitted more easily than HM50 composites.
Fracture process of creep fatigue in a Type 304 stainless steel at 923K can be divided into two types; “crack initiation type” and “crack growth type”. The former is characterized by multiple small cracks initiating one after another along grain boundaries on the specimen surface and coalescing each other at the late stage of failure life, when the material is subjected to relatively large strain ranges in slow tension-fast compression (so called c-p type) strain waveforms. This failure is mostly governed by crack initiation. The latter is distinctive of the initiation and growth of little coalescence, which is revealed under small strain ranges and/or in slow-slow (c-c type) or slow very slow (c-s type) strain waveforms. Therefore, the failure life is controlled by the growth rate of small cracks in this case. In this study, a numerical simulation was done based on a model of two kinds of driving forces for crack initiation and crack growth with random fracture resistance of grain boundaries, in order to distinguish between the two types of fracture process and to estimate the failure life in relation to the cracking behavior.
In order to make clear the effect of tangential component of impact force on the particle impact damage of ceramics, a silicon carbide or steel particle was impacted obliquely on a silicon carbide specimen at various impact angles. For low impact velocity, partial ring cracks were initiated on the surface of specimen. With an increase in impact velocity, they became completely closed ring cracks, and an alignment of the non-concentric ring cracks along the impact direction was observed on the surface of specimen at the impact site. The oblique particle impact produced an cone crack inclined with respect to the normal of the surface. Therefore, the residual strength of specimen evaluated by the bending test after particle impact was dependent on the impact direction. The residual strength of transversely impacted specimen was smaller than the case of longitudinally impacted one in the high impact velocity range. An explanation for the tangential component of impact on the inner damage could be given by using the static sliding contact model from the viewpoint of fracture mechanics. However, the extent of surface damage and the decrease in residual strength were generally depended upon the normal component of impact velocity, regardless of impact angle of particle.
Low-cycle fatigue tests were carried out on a reactor pressure vessel steel at 290°C in order to investigate an evaluation method for fatigue damage and fatigue straining histories. The initiation behavior of surface microcracks was examined, and the following results were obtained. (1) The initiation of surface cracks begins from the early stage of fatigue life and the crack density (number of cracks per unit surface area) continues to increase throughout the fatigue life. (2) The increasing behavior of the crack density can be correlated with the Coffin-Manson type parameter ΔεP·N0·6 irrespective of an applied strain range (Δεt). Therefore, fatigue straining histories represented by ΔεP·N0·6 can be estimated from the measurement of surface crack density. (3) The grain boundary is one of the preferential sites of crack initiation, and micro-cracks which just initiated have a tendency to propagate along the grain boundary. The electrochemical etching was performed at a midplane of the fatigue damaged specimens where no geometrical damage could be expected. Etching properties of fatigue damaged specimens were also studied and were found to change with damage accumulation. The results obtained can be summarized as follows. (4) Many random-shape pits appeared only on the surface of the fatigue damaged material by the electrochemical etching. These random-shape pits initiated mainly at the triple point of grain boundaries and grew preferentially along the grain boundaries. These pits can be considered as a result from degradation of corrosion resistance of the demaged material caused by microstructural changes at or near grain boundaries during fatigue. (5) The area fraction of corroded region on the electrochemically etched surface shows a single correlation with ΔεP·N0·6, and quantitative evaluation of fatigue damage accumulation can be made nondestructively by the electrochemical etching.
The effect of slit depth on the impact value of austempered spheroidal graphite cast iron was studied. The principal results and conclusions are as follows: (1) The impact value could not be improved, though the tensile strength was raised by austempering treatment. (2) The impact value decreased as the slit depth and the angle of the fall of a pendulum increased. (3) Dimple rupture, which consists of both large dimples originating from graphites and small ones occurring from inclusions, was mainly observed on the impact fracture surface in ferrite structure. Also the cleavage facets were observed on the surface with an increase in slit depth. (4) Quasi-cleavage pattern was observed on the surface of austempered or upper bainitic structure. The number of graphite particles on the fracture surface was less than that on the ferrite one.
In general, a plastic sheet displays vibration-damping by converting vibration energy into thermal energy. In this study, the vibration-damping for copolymers including butyl acrylate was evaluated. Copolymer of butyl acrylate (B) grafted onto acrylonitrile (A)-styrene (S) copolymers and methyl methacrylate (M)-butyl acrylate (B) random copolymer were used to prepare eight kinds of specimens having different butyl acrylate contents. A cantilever method was adopted for obtaining frequency response spectra of the specimens. The loss factor of laminated steel plates was also obtained. The loss factor and shear modulus were calculated for copolymers from that of the laminated steel plates. AS-B graft copolymers seemed to give a maximum loss factor at temperatures above 120°C. Therefore the loss factor at room temperature is not sufficiently large. The loss factor for the M-B copolymer having butyl acrylate of 54 mol% was 2.0 in the frequency range from 1Hz to 1kHz at 40°C. Therefore it is expected that M-B copolymer is very useful for vibration-damping at around room temperature.
The residual stresses of gradation coating components induced by a fabrication process were analyzed by comparison with two-layer composites prepared by a direct bonding. Firstly, the gradation coating components of stabilized zirconia/nickel based alloy composites were chosen for a finite element method analysis of residual stress distributions. It was verified that the residual stresses at the interface produced by a uniform heating process decreased by use of the gradation techniques. Especially, stress singularities at the edge of the interface of the direct bonding composites disappeared in case of the gradation coating components. However, the residual stresses at the surface of the gradation coating could not be decreased using the gradation techniques. Secondly, the effects of the gradation geometry and material constants, such as Young's modulus and thermal expansion coefficients on the residual stresses of the gradation coating components were investigated by the thermoelastic analysis using a finite element method. As a result, it was found that the residual stresses could be effectively analyzed by application of the dimensionless parameter, which is σ(1-μ2)/[E2(α1-α2)ΔT](σ; residual stress, μ1=μ2; Poisson's ratio, E2; Young's modulus of coating film 2, (α1-α2); difference of thermal expansion coefficients between substrate 1 and coating film 2, ΔT; temperature difference) in case of the gradation coating components. The analytical results indicated that the dimensionless residual stresses decreased with increasing the coating thickness ratio, t/T (t; coating thickness, T; substrate thickness) and Young's modulus ratio, E2/E1.