The criterion determining the fatigue crack propagation path near or across the interface of hardness discontinuity in material inhomogeneous with respect to hardness was experimentally investigated. The Δδθ maximum criterion was used as a reference. Single-edge-cracked panels of S45C hard steel subjected to uniform pulsating axial stresses of R=0.1 were used as test specimens, which had in each a partial slant band-like or circular area hardened by induction-heat-treatment; the fatigue cracks were intended to propagate perpendicular to the specimen axis, if the Δδθ maximum criterion was valid. It was observed that the fatigue cracks propagated in all cases straight and perpendicularly to the specimen axis, crossing the interface of the hardness discontinuity, indicating that the Δδθ maximum criterion was valid in this case.
The purpose of the present study is to create a better understanding of the fatigue behaviour in 6063 aluminium alloy. Rotating bending fatigue tests have been conducted using smooth specimens in laboratory air at ambient temperature. Crack initiation site and growth path were found to depend strongly on stress level. Particularly, the stress level dependence of crack initiation site largely changed at a certain stress level, above which the crack initiation was predominantly intergranular, while transgranular below that stress level. Such crack initiation behaviour was significantly related to fatigue life and a morphological change of the S-N curve was seen. Additional fatigue tests were performed using the materials with different grain sizes, and the grain size dependence of fracture mechanisms and associated S-N characteristics are further discussed.
The mechanism for fatigue failure in extremely high cycle fatigue in the regime of N>107 is studied on a bearing steel, JIS SUJ2. Special focus was given to the fracture morphology in the vicinity of fracture origin (subsurface nonmetallic inclusion) of a heat treated bearing steel (Specimen QT). The particular morphology looks dark during optical microscopic observation. Specimens with short fatigue life of the order of Nf=105 did not have such a dark area, ODA (optically dark area). To investigate the influence of the hydrogen trapped by nonmetallic inclusions on fatigue properties, specimens heat treated in a vacuum followed by quenching and tempering (Specimen VQ) were prepared. Specimens VQ contained 0.07ppm hydrogen as compared to 0.80ppm hydrogen for conventional Specimens QT. Specimens VQ had a slightly smaller ODA than Specimens QT. Hydrogen was detected by a Secondary Ion Mass Spectrometer around the inclusion at fracture origin of Specimens QT and Specimens VQ. Thus, it can be concluded that the formation of ODA is closely related to hydrogen trapped by nonmetallic inclusions. Estimations of fatigue limit by the √area parameter model based on the original size of inclusions for fatigue limit defined for 107 cycles are -10% unconservative. Considering the size of ODA into fatigue limit estimation, the √area parameter model can predict the mechanical fatigue threshold for small cracks without influence of hydrogen. The mechanism of duplex S-N curve is also discussed.
The main purpose of this study is to investigate the mechanism of transition from the surface fracture to the interior fracture in Ti-6Al-4V alloy. During the investigation, we paid attention to the relation between the fracture mode and fatigue life distribution. We analyzed fatigue life distribution using a competing risk model, that regards surface fracture and interior fracture as independent fracture modes. As a result, the distribution of the surface fracture shows a tendency similar to that of nonferrous metals, and the interior fracture distribution is similar to that of low and medium carbon steel that shows a saturated failure probability. In this case, the latter (saturated failure probability) was caused by arrest of interior crack propagations. The fatigue life distribution shows a peculiarity when we do not consider fracture modes. However, there is no peculiar aspect when using the competing risk model to analyze fatigue life distribution individually. It shows that the competing risk model, which consists of two fracture modes, describes well the fatigue characteristics of Ti-6Al-4V alloy.
To produce the precise stamping dies for ultra fine LSI lead frame, fatigue strength properties of WC-Co alloy plates of 0.1mm in thickness with different WC particle size, machined by wire electric discharge (WED) and/or by subsequent abrasive flow (AF), were investigated. The WED machining significantly reduced the fatigue strength of the ultra thin alloy plates from about a half to a quarter that of ground specimens. Coarsening of WC particles and surface removal of around 1-2μm by AF machining, however, enhanced the fatigue strength. This was principally caused by that they lowered tensile residual stress in machined surface layer. It was suggested that suppression of initial micro-crack propagation due to coarse WC particles and elimination of surface defects due to the AF machining also improved the fatigue strength although their contributions were very smaller than that of the residual stress.
The fatigue strength of lap welded joints of thin steel plate for automobile use was improved by using low transformation temperature welding material which induced the compressive residual stress around welded part. The transformation of welding material from austenite to martensite begins around 200°C and finishes around room temperature. The expansion of weld metal due to the transformation was constrained by the surrounding base metal, and the compressive residual stress is induced in the as welded condition. The mean stress effect due to this compressive residual stress improved the fatigue strength.
The effects of prestraining on high cycle fatigue strength of newly developed low alloy TRIP steels with different matrix structure and different retained austenite characteristics were investigated for the automotive applications. The prestraining to 10% in tension increased fatigue limit of the TRIP steels, especially in steel with polygonal ferrite matrix. It was considered that the polygonal ferrite matrix in the steel brought high fatigue limit mainly due to TRIP of the retained austenite and high compressive internal stress in the matrix, resulting from hard second phase on prestraining. On the other hand, the TRIP steel with bainitic ferrite lath matrix exhibited a small increment in fatigue limit after prestraining. This was supposed to be mainly owing to strain-induced martensite content increased by prestraining, with small effects of TRIP of retained austenite and compressive internal stress in the matrix. In addition, a very interesting finding was obtained that the internal stress is the most effective parameter among some parameters to increase the fatigue strength in the low alloy TRIP steels.
Mode III fatigue crack is usually observed at the root of circumferentially notched round bar. As the crack grows in the depth direction of specimen, the growth rate of crack decreases. In this respect, crack mode transition from mode III to mode I is understood responsible. However, quantitative evaluation on crack mode transition has been hardly done. Therefore, values of stress intensity factor KI under mode III fatigue crack growth tests on a low carbon steel were calculated with using principal stress, as crack mode transition began. Then KI values were compared with crack growth rate curve of mode I. As the result, it was observed that the condition of crack mode transition was given by the crack growth rate curve of mode I.
Fatigue characteristics are of fundamental importance when choosing a structural materials, especially when deciding upon the materials for reactor structure where long-term safety is primary importance. In this field, recently, the propagation behavior of mode II and mode III fatigue cracks has attracted special interest. The mode II and mode III fatigue testings involve greater difficulties compared with the mode I testing and the testing procedure has not been established. SS400 carbon steel was fatigue tested by a torsion fatigue testing machine. In the torsion test, fatigue cracks can propagate in all modes. To identify their behavior, surfaces of the specimen were observed by CCD camera and the behavior of surface cracks were recorded in a time-lapse video recorder which can record up to for 40 days. To reveal the morphology of the mode II and mode III crack propagation, some of test interrupted specimens were sliced and observed by a two dimension measuring microscope. To investigate the crack morphology three dimension crack demonstrate computer program was developed. In the torsion test of plain surface specimen the mode II crack initiates first, but it transits to mode I in the early stage. However, according to the 3D view a large mode III crack propagates into inside in this stage. Persistent slip band like patterns were observed on the surface and the cross section of the fatigue cracked specimen and all mode III cracks propagate in the band. Some of them grew across the center of the specimen. The da/dN-ΔK diagram of the surface mode II fatigue crack propagation was compared with that of mode I by CT type specimen of the same material. mode II propagation rate agreed with mode I propagation rate where the stress intensity factor range was converted as ΔKI=2ΔKII. It means that to design the structure by mode I crack propagation data alone may be acceptable under above conversion. To establish this relationship and to elucidate the behavior of mode III fatigue cracking, more investigations are required.
Flaking type failure in rolling contact fatigue is usually attributed to the fatigue induced by cyclic subsurface shearing stress caused by contact loading. We suppose that this type of subsurface crack growth is essentially a mode II fatigue crack growth under the condition in which tensile mode growth is suppressed due to the compression stress field arising from the contact stress. Although the newly developed apparatus is, as the former apparatus, based on the principle that the static KI and the compressive stress parallel to the pre-crack are superimposed to mode II loading, we employ a new mechanism for mode II load application. In the new apparatus, a direct loading system is employed instead of the four-points-shear loading system used in the former apparatus. A new device for the application of compressive stress parallel to the pre-crack is also developed. Due to these alterations, mode II cyclic loading tests for hard steels have become possible at arbitrary stress ratios, including fully reversed loading (R=-1) which is the case of rolling contact fatigue. The tests were made on bearing steel SUJ2 and 0.75% carbon steel. The hardness of these two steels was around HRC60. The test results showed that stable mode II fatigue crack growth occurred in the region of ΔKII=3-10MPa√m, and that ΔKII-threshold values for mode II fatigue crack growth of these two steels were estimated to be around 3MPa√m. The apparent effect of load ratio was not observed in the data of R=0 and R=-1.
We have studied the effect of tensile axis on strain induced martensitic formation in SUS304 steel. Single crystal specimens having various tensile axes were deformed by the Instron type testing machine at 293K. After deformation, quantitative analysis of α' martensite formed in specimens were carried out by using a newly developed apparatus. This apparatus estimates an amount of α' martensite with measuring the magnetic force between a specimen and a permanent magnet, of which force was calibrated by standard samples. The formation of α' martensite appeared to be classified into 3 groups. (1) In case of ‹001› specimen, no martensite was found without necking down part. (2) In ‹111› specimen, relatively small amount of martensite formation was found. (3) In specimen with the other tensile axes, large amount of martensite was formed. This classification agrees with previously reported hydrogen induced fracture
In order to clarify the correlation of the fracture mechanism and the behavior of the residual stress in the ceramic/metal joint, the residual stress was measured by X-ray diffraction method during applied bending stress. Sintered silicon nitride (Si3N4) and stainless steel (SUS304) were used as specimen materials. The stress analysis by finite element method (FEM) was carried out, and was compared with X-ray measurement results. The obtained results were as follows: (1) The measured stress was larger with the applied stress. (2) As the applied stress was larger, intensity of the stress singularity, K was larger and λ which shows the order of the stress singularity was smaller and then became constant. (3) The behavior of K and λ agreed in the experimental results and the analytical results qualitatively well. (4) At the low-loaded region, there was a difference for the behavior of K and λ depending on the shape and dimension of specimen. But, when the applied stress was larger, K and λ become constant respectively. (5) To observe the change of stress singularity at the interface, it is possible to estimate the stress of fracture initiation.
A maraging steel has low fatigue strength for its high static strength. In the present study, the influences of shot peening and reversion austenite on fatigue strength of a maraging steel were investigated to achieve the improvement of fatigue strength of the steel. By shot peening, fatigue strength was markedly increased and fracture mode changed in order of surface, combined type of surface and interior and interior fractures with decrease of the stress level. Fatigue strength in the surface fracture region was still increased by double shot peening which improved the surface integrity i.e. surface roughness, hardness and compressive residual stress. On the other hand, fatigue strength in the interior fracture region was increased by the formation of reversion austenite where as it was not affected by the shot peening condition.
Recently, it has been necessary to improve the packaging density of printed wiring board (PWB), because the demand for size reduction of the electric devices has begun to increase. Therefore, the pitches between through holes in electric circuits have become fine for the high package density. On the other hand, various forms of damage occur around the holes after drilling. Among them, internal damage has attracted attention in order to prevent the reduction of the insulation resistance between the holes after through-hole plating, and the blow holes after soldering. Therefore, this study researches the damage factors around the drilled hole wall in small diameter drilling of a printed wiring board in order to prevent the ion migration between the holes. First, the internal damage around a drilled hole is examined. Next, the dynamic force (thrust force) during drilling is measured in order to examine the factors causing the internal damage around the drilled hole. As a result, it is shown that the outer corner of drill is one of the most important part in the small diameter drilling for the damage around the drilled hole.
The tensile test was conducted on the FRPA with the crack initiated by the fatigue test, including fiber content in weight of 50% made by direct injection molding method. The relation between the fractal dimensions obtained from the AE amplitude distributions, the damage (white) zone at the crack tip, the microfractographs analysis and the strain energy were examined during the process of its tensile test to fracture. The fractal dimensions obtained from the AE amplitude distributions of fatigued specimens became lower when the stress amplitude and the number of cycles became higher. The microfractographs analysis showed that the specimens of the higher fractal dimensions had fracture surfaces of larger roughness, but the ones of the lower fractal dimensions had fracture surfaces of smaller roughness. It was found on the fatigued specimens of the region II where AE was detected constantly that the fractal dimension obtained by normalizing the experimental values of fatigued specimens with the experimental one of non-fatigued specimen decreased with a decrease in the normalized strain energy.
Matrix hybrid laminates are defined to be materials fabricated by the combination of two or more kinds of matrices. The wide ranges of stiffness and damping property may be produced by changing various designing parameters, the resin ratio and/or the stacking sequence. In this study, two kinds of resins are used as matrices. One is a vinyl ester resin with high stiffness and the other is unsaturated polyester resin with low stiffness. As a reinforcement, plain woven glass cloth is used. Matrix hybrid laminates with various flexible resin ratios and stacking sequences are fabricated. The static and dynamic bending properties are investigated by the three-point bending and the vibration tests, respectively. In addition, antisymmetric matrix hybrid laminates are also investigated by the above tests. For the above all experimental results, the influences of designing parameters on both static and dynamic bending properties are discussed in detail. It is confirmed that matrix hybrid laminates can have the wide ranges of stiffness and damping properties by changing the flexible resin ratio and the stacking sequence.
Accoustic emission (AE) due to microcracking process in the proximity of a macrocrack is theoretically studied. On the basis of some approximations, the AEwave from a crack-opening event is directly related to crack-opening volume as a function of time. It is assumed that two-dimensional cracks open spontaneously under a tensile stress applied at infinity, and that the crack volumes grow to the values of the static solution by a constant rate.The sources are approximated as points located at the center of the cracks. The two-dimensional Green's function for an unbounded medium is utilized in the analysis. It is shown that the interaction between a macrocrack and a microcrack generates additional crack-opening volume for both cracks, whose signals tend to be much larger than those that would emanate from nucleating a solitary microcrack, and the waveforms due to other models.