The piezoelectric ceramics with different dimensions are embedded into the composite laminates. The impedance of the piezoelectric are measured by a LCR meter in the autoclave processing. The impedance of the piezoelectric ceramics at non-resonant frequencies is adopted as the cure index. The piezoelectric ceramics have strong temperature dependency. The temperature dependency is corrected by using a dummy piezoelectric ceramics. A dielectric sensor is also embedded in the composite laminate as a reference sensor for the degree of cure. The change in corrected impedance at non-resonant frequency shows good correspondence with change in the log ion viscosity which is measured by the dielectric cure monitoring sensor. At the cooling stage, the changes in the impedance of the piezoelectric ceramics depend on the dimension of the piezoelectric. A finite element method is used for analyzing the deformations of the piezoelectric ceramics in the composite laminate at the cooling stage. The changes of the impedance of the piezoelectric ceramics are dominated by the deformation of polarizing direction.

View full abstract

The application results of transmission/reflection-type hybrid extrinsic Fabry-Perot interferometric (TRHEFPI) optical fiber sensors to structural health monitoring were reported. Conventional interferometric optical fiber sensors have a disadvantage of directional ambiguity generally. In a TRHEFPI optical fiber sensor, the reflection and transmission sensor signals can be obtained simultaneously. The reflection signal contains the information of only measurement quantity. On the other hand, the transmission signal contains the information of both measurement quantity and directions. The interferometric fringes of the linear combination between the transmission and reflection signals have phase shifts from those of the reflection signal. Due to the phase shifts, which show the lead or lag behavior according to measurement directions, the clear information of measurement directions can be obtained. To control the phase shift magnitude, the coefficient in the linear combination signal must be designed adequately. The precise measurement quantity can be obtained from the reflection sensor signal with high visibility. For strain measurement, the sensor was attached on a surface of an aluminum-alloy tension specimen. In this research, the repeated sinusoidal strain of the fatigue test was measured using TRHEFPI sensors. The health monitoring of the smart structures was possible with the fatigue monitoring.

View full abstract

Stimulated Brillouin scattering in optical fibers can be used to measure strain or temperature distribution along the fibers. To operate BOTDA system with high spatial resolution, short-duration pulses must be used, resulting in reduced signal powers causing a decrease of the dynamic range. In this paper, a new technique for enhancement of spatial resolution of BOTDA using a double-pulse was proposed. Experimental results showed that the ability to resolve two adjacent events could be enhanced by using a double-pulse as a pump lightwave without decreases in the dynamic range.

View full abstract

Thermo-mechanical fatigue (TMF) tests of a unidirectionally reinforced SP700/SCS-6 composite were carried out, where the matrix alloy, SP700 is a new generation high strength titanium alloy. The damage evolution during the TMF was explored, through the investigations of the following articles, compared with that in the monolithic matrix alloy: (i) mechanical properties of both the SP700/SCS-6 composite and the matrix alloy at elevated temperatures; (ii) TMF and isothermal low cycle fatigue (LCF) failure lives of the composite and the matrix alloy, (iii) fiber push-out tests at elevated temperatures to represent the fiber/matrix inter facial strength; (iv) observation and the characterization of the interface by means of a transmission electron microscope (TEM) and an energy dispersive X-ray spectrometer (EDS); (v) crack initiation and the propagation during the thermal cycle test under no external loading; and (vi) changes of the interface during the TMF, LCF and thermal cycles. These systematic investigations showed that the lives and the failure mechanisms of the TMF should be clearly distinguished from those of the LCF, whereas there were a few similarities between them.

View full abstract

This paper describes an applicability of passive electric potential CT (computed tomography) method which does not require electric current application for damage detection. In this method, piezoelectric film is glued on the surface of structures as a sensor. Electric potential values on the piezoelectric film change due to the strain distribution on the structure, when the structure is subjected to external load. The strain distribution of the cracked body induces a characteristic electric potential distribution on the piezoelectric film. Therefore passively observed electric potential values on piezoelectric film can be used for the defect identification. An inverse method based on the least residual method was applied to the defect identification from the electric potential distribution. In this inverse method, square sum of residuals is evaluated between the measured electric potential distributions and those computed by using the finite element method. This method may be applied to develop an intelligent structure with a function of self-monitoring of flaws and defects. The electric potential on piezoelectric film was measured by the contact type and the non-contact type methods. The measured electric potential distribution was used to identify a crack by using the present inverse method. It was found that the location, size and depth of the crack can be quantitatively identified by the passive electric potential CT method.

View full abstract

This paper presents modal-based structural damage detection. Specifically, we focus on localized flexibility properties that can be deduced from the experimentally determined global flexibility matrix. We present the underlying theory that can be viewed a generalized flexibility formulation in three different generalized coordinates, namely, localized or substructural displacement-basis, elemental deformation-basis and element strain-basis. Then, the present methods are applied to a CFRP space frame structure having interior damage and the numerical and experimental results show that the elemental strain-basis method is quite useful for detecting the inside damage of the CFRP space frame elements.

View full abstract

This paper addresses a sensitive strain monitoring composite as compared with carbon fiber reinforced polymer composites (CFRP). Carbon fibers in CFRP can be used as sensors to detect fiber fracture and strain of composites using electrical resistance change. However, the sensitivity is not sufficient because of the electrical conduction between neighboring fibers. In this study, surface oxidized SiC fibers were applied to prevent electrical conduction due to contacts between neighboring fibers. Model specimens were fabricated by embedding as-received or oxidized SiC fibers into epoxy resin, and electrical changes were measured under monotonic and loading-unloading tensile tests. As a result, electrical resistance of oxidized SiC fiber specimens increased remarkably with increasing strain as compared to as-received SiC fiber specimens. Cumulative fracture probabilities of oxidized fibers embedded in the model specimens were predicted from measured electrical resistance change, and the strength distributions were analyzed based on Weibull statistics. The difference of electrical resistance behavior about two kinds of specimens with different effective gage length was also indicated.

View full abstract

Structural health monitoring is a noticeable technology for advanced composite structures and civil structures. The present paper proposes a new diagnostic tool for the structural health monitoring that employs a statistical diagnosis of self-learning method. Most of the structural health monitoring systems adopt parametric method based on modeling or non-parametric method such as artificial neural networks. The former method requires modeling of each structure and latter method requires data for the training. And these modeling and data for the training demand much cost. The new statistic diagnosis method does not require the complicated modeling and a learning data of damaged structure for the artificial neural networks. The present study deals monitoring of delamination of composite beam using change of strain judged by the statistical tools such as Response Surface and F-Statistics. Response surfaces among the measured strain data of surface of composite beam are produced at intact state and monitoring state, and the difference of the each response surface is statistically tested using F-test. As a result, the new method successfully diagnoses the damage without using modeling and a learning data of damaged structure.

View full abstract

The use of thin actuator plates bonded to surfaces of FRP laminates to suppress the growth of delamination cracks is investigated. A composite beam analysis was performed to estimate the effect of strains induced by actuators on the strain energy release rate of delamination cracks. When compared at the same value of actuation strain, the apparent crack growth resistance, G_<app>/G_C , is larger for smaller FRP thickness and for smaller FRP fracture toughness, G_C , under both mode I and II loadings. Mode I delamination suppression tests were conducted by using thin SMA actuators. The value of the apparent crack growth resistance, G_<Iapp>, under actuation strains of about 0.001 was 2.7 times larger than fracture toughness of FRP, G_<IC>, and was also about 1.6 times as large as that just before actuation.

View full abstract

Shape memory polymer as one of smart materials has received much attention. It is characterized by the recoverability of shape memory effect, but its mechanical property such as the strength isn't strong enough. In this study, for the applications in the field of industry, the carbon fabric reinforced shape memory polymer was developed. The bending recoverability was investigated and compared between the SMP simple substance and the development composite material. Both of the materials were loaded and then unloaded repeatedly beyond Tg, and the bending recoverability was characterized by the bending angle and the influence of weight of specimen was also examined.

View full abstract

This study proposes an intelligent composite material structure that suppresses Mode II delamination crack in CFRP laminates. Pre-strained shape memory alloy TiNi sheets were attached on the surface of CFRP specimens in order to investigate the effect of the recovery strain of the SMA actuators on the delamination crack propagation behavior under static loading. Mode II interlaminar fracture toughness tests were conducted on the two types of specimens: One is CFRP laminates with a compressively pre-strained SMA actuator on the compressive side of the specimen. The other has an additional SMA actuator with tensional pre-strain on the tensional side of the specimen. The results show that the fracture toughness increased due to the attachment of the SMA actuators even in the range away from the actuators for the both types of the specimens. In the specimen with the SMA actuator on one side, the fracture toughness increased further by the application of heat to the actuator. This is attributed to the effect of the recovery strain to suppress the mode II interlaminar crack. On the other hand, specimens with the SMA actuators on the both sides didn't exhibit much improvement in fracture toughness.

View full abstract

In this paper, experimental and theoretical researches on the viscoplasticity of lead and lead free solder alloys are conducted. First, experiments such as tension-compression loading with several stress amplitudes and strain rates, creep tests at several stress levels, and stress relaxation tests at several strain levels are conducted using specimens made of 63Sn/37Pb as the lead solder alloy and Sn/3Ag/0.5Cu as the lead-free solder alloy at several temperatures. The test results show that the lead-free solder alloy has smaller strain rate effects, creep strain and stress relaxation than those of the lead solder alloy. The viscoplastic deformations of both lead and lead-free solder alloys are simulated by a viscoplastic constitutive model. The simulations show that the constitutive model can successfully apply to the pure tension, the cyclic loading, the creep deformation and the stress relaxation of both the lead and the lead-free solder alloys. FEM analysis of the bend of the substrate is also conducted by MSC.Marc, in which the constitutive model for viscoplasticity is incorporated. As a result, it is found that the bend of substrate using the lead-free solder joints has much larger bend than that using the lead solder joints.

View full abstract

This paper describes the tensile, low cycle fatigue, creep and creep-fatigue standard testing methods for solders recommended by the solder strength subcommittee in JSMS. These methods recommended material preparation, specimen geometry, testing method and data acquisition. Benchmark experiments were carried out based on the standards and they demonstrated that the methods are available for stable data generation. Tensile and low cycle fatigue database for Sn-37Pb and Sn-3.5Ag were published as the solder database.

View full abstract

In this paper the authors have investigated the thermal fatigue reliability of lead-free solder joints. They have focused their attention to the formation of the intermetallic compound and its effect on the initiation and propagation behaviors of fatigue cracks. An isothermal fatigue test method was used in this study to improve the efficiency of fatigue study, and several different lead-free solder alloys, Sn-Ag-Cu, Sn-Ag-Cu-Bi, Sn-Cu and Sn-Zn-Bi were investigated. There are two kinds of fracture mode in lead-free solder joints, one is solder fatigue mode, and the other is an interface fatigue mode. Based upon the experimental results, it was found that not only is the mode transition of the fatigue crack affected by the properties of the intermetallic layer but also is affected by the tension strength of the solder material. If the tension strength is lower than a critical value, the fatigue cracks in the solder joints appear within the solder domain, that is the solder fatigue mode, and their fatigue life can be assessed by Manson-Coffin's law of the bulk solder material. On the other hand, if the strength stress is higher than that value, the interface between solder and Cu pad breaks much earlier than the solder fatigue life, that is the interface mode.

View full abstract

A convenient method for predicting mechanical properties of solder materials was proposed. Indentation method was used to obtain mechanical properties including elastic, plastic and creep deformations. Elastic-plastic FEM correlated Young's modulus and yield stress with unloading portion of load-displacement curve and Vickers hardness, respectively. Tensile creep properties were lead by measured indentation creep properties, where the creep properties were evaluated in terms of the Norton's law. The agreement of the prediction was confirmed for Sn-37Pb (mass %) eutectic solder in the range between room temperature and 125℃. The above prediction methods were established based on elastic-plastic-creep analysis of FEM for an indenter with rotational symmetry, whereas the Vickers indenter is a three-dimensional configuration of quadrangular pyramid. The agreement of the prediction methods was confirmed by three-dimensional analysis of FEM. Using the proposed method, mechanical properties of lead-free solders such as Sn-3.5Ag and Sn-3Ag-0.5Cu-0〜3Bi alloy systems were determined for the temperature between -20℃ and 160℃. The predicted results revealed that mechanical properties of Sn-3.5Ag and Sn-3Ag-0.5Cu alloys were similar, and that Young's modulus, yield stress and creep properties of Sn-3Ag-0.5Cu-0〜3Bi lead-free solders were strongly influenced by Bi content.

View full abstract

In this paper, a viscoplastic constitutive model was proposed considering both the determination method of the material constants and the implementation method for a general purpose FEM program. The model was incorporated into FEM code "ANSYS" and its applicability to the lead-free solder alloys was verified by the simulations of the viscoplastic deformation. As a result, the proposed model was successfully incorporated into "ANSYS", and then it could accurately describe the creep and cyclic loading of three lead-free solder alloys using the material constants determined from only the pure tensile tests.

View full abstract

In this paper, the authors have investigated mechanical fatigue strength of lead-free solder joints. The use of Sn-Zn-Bi and Sn-Zn solder is increasing because of the advantage of low cost and low melting point. Therefore, it is important to ensure the fatigue strength of Sn-Zn-Bi and Sn-Zn solder joints. In this study, the mechanical fatigue strength of Sn-Zn-Bi and Sn-Zn solder joints has been studied, by using the CSP(Chip Size Package) specimen. In order to investigate the influence of plating treatment on the fatigue strength of the solder joint, the specimens with Ni/Au or Cu plating treatment on Cu-pad were prepared. And these specimens were aged by holding 200, 500, 1000 hours at 85, 125, 150℃, respectively to investigate the influence of the growth of the intermetallic compounds on the fatigue strength. Through a series of isothermal mechanical shear fatigue tests and FEM(Finite Element Method) analysis, it has been found that if the Sn-Zn-Bi and Sn-Zn solder with Ni/Au plating treatment were aged at the temperature above 125℃, the fatigue strength decreased remarkably in comparison with the specimens with Cu plating. By plating with the Ni/Au on the substrate side, the interface in the substrate side was strengthened. But, the interface in the chip side is weakened. It causes the interface fatigue crack between intermetallic compound layer and solder at the chip side, and the decrease of fatigue strength.

View full abstract

Isothermal axial strain controlled creep-fatigue tests of a eutectic alloy Sn-37Pb were carried out at room temperature. Four kinds of triangular strain waveform, so called fast-fast, fast-slow, slow-fast and slow-slow, were used. First, constant amplitude fully reversed creep-fatigue tests were done in order to obtain the creep-fatigue properties of the material as the partitioned inelastic strain range versus life relationships, that is, Δε_<ij>-N_<ij> relationships (ij=pp, pc, cp and cc). The obtained Δε_<ij>-N_<ij> relationships were compared with the literature data. Second, two-step variable amplitude fully reversed creep-fatigue tests, so called low-high and high-low, were conducted under fast-fast and slow-fast straining conditions. The results were analyzed and discussed based on the creep-fatigue damage rule proposed by the author and his colleague in the previous work for other high temperature materials. Finally, the effects of ratcheting strain on creep-fatigue life were studied by conducting creep-fatigue tests under the strain waveforms accompanying a small tensile ratcheting strain.

View full abstract

The thermomechanical behavior of soldered joints have received much attention, due to the importance of thermal fatigue and creep as causes of degradation and failure. Previous work in this area mainly involved optical methods, diffraction, mechanical testing and modeling. The experimental techniques were commonly applied before and after a certain number of cycles of temperature and/or stress variation, so the observations were limited to effects that were irreversible. However, an effect may be partly reversible. In order to observe the reversible part of a phenomenon, observation must be conducted in real time using a nondestructive technique. Measurement of the DC electrical resistivity was found to be effective for monitoring in real time the effects of stress and thermal cycling on a soldered joint. The solder was eutectic tin-lead. The adjoining materials were copper. The resistivity was that of the overall sandwich. Both reversible and irreversible effects were observed.

View full abstract

Fatigue tests have been conducted in 3%NaCl solution using stainless steels in order to understand the effect of volume fraction of ferritic(α)/austenitic(γ) phase on crack initiation and growth behavior. Materials with five different volume fractions of α/γ phase were prepared. Fatigue cracks were initiated predominantly from slip bands within J grains regardless of volume fraction of α/γ phase. In all the materials, fatigue strengths were slightly reduced in 3%NaCl solution compared with in laboratory air. In very small crack region following crack initiation, crack growth was enhanced in 3%NaCl solution. Environmental effect was also seen in high ΔK region for long cracks, where crack growth rates werefaster in 3%NaCl solution than in laboratory air.

View full abstract

Damage initiation and growth under quasi-static and cyclic loading of cross-ply and multidirection laminates are presented. Damage states such as initiation of matrix cracking, onset of delamination and fiber fracture are identified. A linear relationship is found between the maximum applied cyclic stress and logarithm of number of cycles for each damage mode. The effect of transverse cracks on the stiffness reduction has been quantified. Again a linear relationship between the normalized stiffness and crack density is shown to be a good representation of the test data, irrespective of the applied maximum cyclic load. Under cyclic loading matrix cracks initiated at a lower stress compared to the quasi-static loading. Also higher crack densities were observed under cyclic loading than that of quasistatic one. A major difference between the two types of loading is in the initiation and growth of delamination.

View full abstract

High temperature fatigue (R=0) damage and deformation behavior of SUS304 steel thermally sprayed with Al_2O_3/NiCr coating were investigated using an electronic speckle pattern interferometry(ESPI) method. Surface cracks and delamination occurred after 1x10^5 cycles test when σ_<max> was 202MPa at 873K. The lengths and number of cracks or delamination largely decreased when σ_<max> or temperature decreased to 115MPa or 573K, respectively. Strain values measured with the ESPI method along cracks were much larger than those in other areas due to crack opening under the tensile load. The positions of strain concentration zones on strain distribution figures measured by ESPI method corresponded well to those of cracks on sprayed coatings. Strain values decreased largely where local delamination occurred.

View full abstract

The fatigue strength and the fracture mechanisms of surface treated specimens were investigated using five types of specimens - annealed, nitrided, surface-removed, shot peened and multi treated specimens with ion-nitriding and shot peening. In the case of surface-removed specimens, 0.1mm of surface layers were removed after nitriding. The shot peening was carried out after or before nitriding. The fatigue limit of the nitrided specimens was higher than that of the annealed specimens, but was almost the same as that of the surface-removed specimens. This is related to the hardness distribution of the specimens. In the case of the nitrided specimens, the estimated strength deduced from the hardness is higher than the measured fatigue limit from surface to a depth of 0.1mm. This shows that fatigue limit and strength of the nitrided specimens are determined by the internal properties of the materials, and thus the fatigue strength of the nitrided specimens are not decreased even if the surface layer is removed. It is also found out that the fatigue strength of nitriding specimen could be improved by the shot peening. The fatigue strength of the nitrided low carbon steel was related to the hardness distributions.

View full abstract

In the present work, crack initiation in biaxial fatigue was first modeled as the slip band formation in an individual grain constituting a polycrystalline metal. In modeled grains, normal directions of slip planes and slip directions on respective planes were independently given at random. The stress states in individual grains were determined in the following two ways. In the first model, it was assumed that every grain is subjected to the same stress state as that in the bulk material. In the second model, stress variations were supposed to exist in distinct grains, and the stresses in each grain were also randomly given. As a crack initiation criterion, a slip-band crack was presumed to be initiated along the given slip direction on the specified slip plane when the resolved shear stress calculated in the slip direction exceeded a critical shear stress. The crack initiation life was evaluated using a dislocation pile-up model, in which the calculated resolved shear stress was incorporated. Simulations on crack initiation were carried out for three values of biaxiality; i.e. axial, combined axial-torsional and torsional modes. Space-time information on initiated cracks was able to be obtained by simulation using the aforementioned model. Simulated directions of initiated cracks under biaxial modes were compared with experimental results which had been observed in fatigue tests using tubular specimens of pure copper under axial, combined axial-torsional and torsional loading modes. As a whole trend, the estimation based on the proposed model showed a good agreement with the experimental observation.

View full abstract

Microstructural effects on near-threshold fatigue crack growth behavior in a Ni-base superalloy have been studied. TEM and SEM observations have been performed on corresponding microstructures near crack tip and fracture surfaces. The fatigue crack growth resistance increases with increasing grain size, i.e. the threshold value increases and the fatigue crack growth rate (FCGR) decreases; Correspondingly the fracture surface changes from "fatigue striations" into remarkable "facets". However, increasing γ' particle size at aging-peak condition leads to decrease of the fatigue crack growth resistance, i.e., the threshold value decreases and the FCGR increases; In this case the fracture surface shows "fatigue striations" feature. The high resistance to fatigue crack growth is closely related to the planar slip mode of dislocations in the plastic zone near crack tip, which gives rise to lower accumulation of damage and enhances crack closure resulted from facet-feature cracking. The optimum microstructure for best fatigue crack resistance is suggested as coarse grains with fine γ' particles just before aging-peak is reached.

View full abstract

Both the effect of compressive mean stress on fatigue properties of non-rolled specimen and the effect of compressive residual stress on fatigue properties of plastic deformed specimens by roller-working were investigated. Pulsating fatigue test have been performed using the specimens with or without surface deformation. The main results obtained in this study are as follows; (1) According to increase of the compressive mean stress from 0, -61.25, -122.5 and -183.75 MPa, the fatigue limits of these specimens also increase to 255, 288, 388, and 428 MPa. (2) The fatigue limit of rolled specimen D2 is 500 MPa, which is higher than that of non-rolled specimen ND0 by 196 %. The above difference would be caused by compressive residual stress and work-hardening due to roller working. (3) The rolled specimen shows two steps of fracture mode, i.e., shear fracture mode and tensile fracture mode. (4) According to the above results, the compressive mean stress and compressive residual stress delay crack initiation and suppress remarkably the transition from crack initiation to propagation.

View full abstract

As one of the effective and low cost surface modification, the authors have been studying and clarified roller working to be able to improve the fatigue properties of bar shape components. In this study, fatigue test is performed to research the influence of roller working on fatigue properties and determine optimum value by roller working using eutectoid steel. Six kinds of specimens (the deformation value are 0, 0.05, 0.1, 0.15, 0.25 and 0.3mm, respectively) are used for this study by keeping the same final shape and dimensions. According to the results of fatigue test, the compressive residual stress generated in the roller working is one of the main factors for the improvement of fatigue strength, and the fatigue limit does not necessarily increase with the increase of deformation values. There exists an optimum value by roller working for improving the fatigue strength. The reasons why the fatigue limits change with the variation in plastic deformation values and why there exists an optimum deformation value are examined by analyzing residual stress, the variation in micro structure and Vickers hardness distribution.

View full abstract

For the maintenance of machine structures from fatigue damage, it is important to evaluate the fatigue damage before fatigue crack initiation and growth. Therefore, in this study, we propose a new method to evaluate microstructure changes during fatigue process using ultrasonic spectroscopy. The dislocation behavior, i.e. density of movable dislocation and eigen-frequency of dislocation strings, are closely related to fatigue damage of materials. We can estimate the density of movable dislocation under fatigue process by measuring ultrasonic surface wave velocities with three different propagating frequencies via Granato-Lucke's dislocation string theory. This proposal using the method of presuming movable dislocation density is examined whether it is one index of evaluating fatigue damage in engineering or not. Therefore, we experimented in two cases, i.e. the first experiment is a pulsating tension-compression fatigue test, and the second is a rolling contact fatigue test. In the first case, the correlation of the movable dislocation density with fatigue damages concerning both of aluminum alloy and chromium molybdenum alloy steels was examined. The results obtained are as follows. The ultrasonic surface wave velocities with three different propagating frequencies (5, 10 and 15 MHz) decreased due to increasing of number of fatigue cycles in the both cases of aluminum alloy and chromium-molybdenum alloy steel. The density of movable dislocation in the both cases estimated by Granato-Lucke's model increased in the fatigue damage progress. In the second case, we estimated the density of movable dislocation under the rolling contact fatigue process of chromium molybdenum forged steel by measuring ultrasonic surface wave velocities with three different propagating frequencies via Granato-Lucke's dislocation string theory. As a result, the ultrasonic surface wave velocities with three different propagating frequencies (5,10 and 14 MHz) decreased due to increasing of the rolling number of rolling contact fatigue and the density of movable dislocation estimated by Granato-Lucke's model increased in the fatigue damage progress. The comparison of changes of the density of movable dislocation with changes of half value breadth of X-ray diffraction line profile analysis under the rolling contact fatigue process suggests promising availability of the proposed method.

View full abstract

The biggest increase in materials used for light-weighting cars and trucks are neither aluminium, magnesium nor some other alternative metals. Instead, it is the high strength steel sheets, such as Extra-low carbon steels, that have shown the most growth in the light-vehicle market. The continuing emphasis on vehicle mass reduction has driven researchers to investigate the use of thin, high-strength steels in automobiles. Reducing weight by using thinner high strength steel sheets allows car companies to meet reduced fuel requirements without compromising the size and safety of cars as well as the buyers' affordability. Extra-low carbons steels, commonly known as Interstitial Free steels (IF) have been widely used for automobile plates. IF-steels have also been widely used in many kinds of products by press deformation particularly of stamped components in automobile application because of their excellent deep-draw ability and low deformation resistance.

View full abstract

A hypothesis of fatigue plastic adaptation is proposed and modeled as one idea that is available to combine microscopic and macroscopic approaches to fatigue plasticity. The hypothesis expresses that, in a surface layer, at a notch root and at a fatigue crack tip, elastic deformation for the maximum stress is transformed into localized inhomogeneous plastic deformation inherent in fatigue. Based on the hypothesis, an equivalent stress ratio is formulated as a parameter for correspondence between cyclic stress conditions of notched and unnotched specimens and is applied to some published experimental data including different cyclic loading types of torsion, bending and their combination.

View full abstract

Rotating bending fatigue tests on shot-peening (SP) treated AC4CH aluminum alloy were carried out in order to investigate the effect of SP treatment on the high cycle fatigue properties. As comparing fatigue property of the SP-treated material with fatigue properties of non-peened material, hot isostatic pressure (HIP) treated material and semi-liquid (SL) die casting material, the following conclusions were obtained. (1) Although the fatigue property of SP-treated material is most excellent in all materials, they are a few differences as compared with HIP-treated material and SL material. (2) The fatigue life property of AC4CH alloys is significantly affected by fatigue crack initiation behavior. The reason why the SP-treated material has longer fatigue life than that of other material is that it has no cast defects near the surface by the effect of SP treatment. Moreover, the fine granular microstructure near the surface caused by SP treatment affects fatigue life since this fine granular structure prevents generation of the slip in crystal grain, which causes the surface fatigue crack initiation.

View full abstract

The surface of metallic materials can be modified by introducing compressive residual stress thereby inhibiting crack initiation and/or growth. We have developed a new method, cavitation shotless peening (CSP), for improving the fatigue life of metallic materials. In this paper, beneficial residual stress were introduced by CSP. CSP is a new method of surface enhancement technique without using shots which differs from shot peening. CSP makes use of a high speed water jet with cavitation, herein referred to as cavitating jet, whose intensity and occurring region can be controlled by parameters such as upstream pressure and nozzle size. The cavitation impacts induced by the collapse of the cavitation bubbles produce compressive residual stress and work-harden the materials. The residual stress was measured using the X-ray diffraction method. To determine the improvement of fatigue strength of Duralumin (JIS A2017) using CSP, the specimen was tested using the rotating bending fatigue test while the surface analysis was performed by surface profile meter. Hardness changing with exposure times was determined using Microvickers. Comparisons of mechanical properties of non-peened and CSP specimens have been discussed. Experimental results showed that specimen treated by CSP in air had an increase of 65MPa (54%) in comparison to the non-peened specimen at fatigue limit.

View full abstract

Rotating bending fatigue tests were carried out and typical factors to govern the fatigue life were experimentally examined in the long life region. Fracture mode in the long life region was an interior inclusion induced fracture with a fish-eye. At the center of the fish-eye, a non-metallic inclusion was necessarily observed, and a fine granular area (FGA) was also necessarily observed in the vicinity around the inclusion on the fracture surface. It was found that the fatigue lives were related to following three factors; (1) the stress amplitude at the inclusion, (2) the inclusion size, and (3) the FGA size. These factors can be connected to the parameters in the formula of Paris' law. Based on this aspect, the fatigue life of this steel in the above fracture mode was formulated from a viewpoint of fracture mechanics. Thus, it was found that the period forming the FGA occupied more than 90% of the total fatigue life. It was another finding that the effective factor to expand the fatigue life was smallness of the inclusion size.

View full abstract

The effects of frictional wear characteristics and microstructural factors such as volume fraction and diameter of primary α phase on fretting fatigue strength of Ti-4.5Al-3V-2Mo-2Fe conducted with various annealing treatments were investigated in this study. Fretting fatigue life of Ti-4.5Al-3V-2Mo-2Fe is hardly affected by the volume fraction and diameter of primary α phase in the low cycle fretting fatigue region. The fretting fatigue limit of Ti-4.5Al-3V-2Mo-2Fe tends to increase with decreasing the volume fraction and average diameter of primary α phase. However, when volume fraction and average diameter of primary α phase are below certain values, the fretting fatigue limit is lowered by coarsening of acicular α phase, which precipitates in β phase region. The weight loss of Ti-4.5Al-3V-2Mo -2Fe tends to increase with decreasing the hardness. The fretting fatigue life of Ti-4.5Al-3V-2Mo-2Fe tends to decrease with increasing the estimated weight loss of the slip region and increasing the depth of the crack growth zone caused by the contact pressure.

View full abstract

The present study investigates plasticity-induced martensitic transformation around a part-through crack in an austenitic stainless steel SUS304 fatigued at room temperature in air. Distributions of the volume fraction of α' martensitic phase around a semi-elliptical fatigue crack were measured with ferrite scope. The results were compared with the distributions of vertical magnetic flux density B_z above and below the crack measured by in a specimen magnetized by a strong electromagnet. It was revealed that the B_z distributions reflected the distributions of α' volume fraction ξ_α' around a fatigue crack: i.e., the distance between the points where B_z reached the maximum and the minimum values B_<zmax> and B_<zmin> had linear correlations with surface crack length 2a. The B_<zmax> and the B_<zmin> values also showed linear relations with maximum stress intensity factors at the surface tips K_<amax> and at the depth position K_<bmax>. These results imply that not only 2a but also K_<amax> and K_<bmax> values can be estimated in an electromagnetic non-destructive way.

View full abstract

To ensure long life reliability of the machine parts made of thin sheet metals, ultrasonic plain bending fatigue test method for thin sheet metals was developed. One end of the sheet specimen is fixed at a loop point of the bar oscillating at a frequency of 20kHz and the other end is set free. As a result, when a frequency of the first characteristic bending mode vibration of the specimen is also 20kHz, a resonance occurs in the specimen due to the forced vibration at the fixed end. This is a similar situation to the plain bending fatigue and the maximum cyclic stresses of tension-compression result near at the fixed end of the specimen. To calculate stress, deflections both at the free end and at the fix end of the specimen oscillating at a frequency of 20kHz have to be measured. A laser displacement sensor and a gap sensor based on variation of the capacitance were used to measure these deflections. Thus, long life fatigue strengths of thin sheet metals can be obtained in a very short period of time. Stainless steel, phosphorous bronze, commercially pure aluminum and maraging steel were used as the test materials and fatigue properties of those sheet metals of around 0.2mm thick were obtained up to the giga-cycle regime. From the results, it is confirmed that this method can be used practically to obtain very high cycle fatigue properties of sheet metals in a very short time.

View full abstract

Focusing on the singular stress field near the contact edge, the evaluation of the fretting fatigue strength can be intrinsically carried out. Two kinds of singular stress field near the contact edge are possible to appear; one corresponds to the slipping deformation of the edge, and the other corresponds to the sticking deformation. The fretting fatigue tests with these two singular stress fields were designed and carried out. From cross-section of observation of the specimens at stress levels lower than the fatigue limit, fretting fatigue cracks were observed when the cyclic stress was above a certain level. This fact means that the specimen may be fractured at stresses lower than fatigue limit in the very high cycle regime. Therefore, the crack initiation limit, which corresponds to infinite fatigue life, is valuable to be investigated. Through the numerical analysis of the stress intensity coefficient of the singular stress fields, such limit corresponding to different singular stress field were quantitatively evaluated.

View full abstract

Using a fatigue testing method by which fatigue cracks can be initiated and propagate in a film adhered to cover an elliptical through-hole in a base plate subjected to push-pull cyclic loads, annealed rolled pure copper films of 100μm and 30μm thickness were fatigued under constant stress amplitudes with a stress ratio of R = 0. The effects of microstructures such as crystal orientation due to rolling on fatigue crack propagation behavior were studied using two types of specimens in which the rolling direction was parallel and perpendicular to the loading direction. The crystallographic characteristic of rolling textures in the annealed film was analyzed using "electron back-scatter diffraction (EBSD)" system. As a result, the fatigue crack propagated slower toward the perpendicular direction to the rolling direction than toward the parallel direction in the film with the thickness of 30μm. This is probable that the number of slip systems in the film loaded parallel to the rolling direction was found to be less than in the film loaded perpendicular to the rolling direction from a result of the slip analysis of the rolling texture in copper film materials. The effects of the anisotropy of the crystal orientation seemed to be greater in the thin film with a weak interaction between grains through the thickness than in the thick one.

View full abstract

As model specimens of surface film-bonded materials with or without resin interlayer, pure copper and commercial grade iron films were bonded to the surface of steel base plates by epoxy resin bonding or by diffusion bonding. The film thickness was 100 and 50μm for the copper and 100μm for the iron, respectively. In the fatigue testing results using these specimens, the fatigue crack initiation life is smaller on the film bonded with epoxy resin than on the film bonded by diffusion. This tendency is particularly strong for the iron film. On the other hand, the epoxy bonding layer protects the inner base plate against fatigue damage, thus increasing the fatigue crack propagation life of the film-bonded plates with epoxy resin. In this connection, both the compressive residual stress on the iron film and the smaller tensile residual stress on the thinner copper film increases the fatigue crack propagation life significantly as compared with the larger tensile residual stress on the thicker copper film. Finally, the effect of the epoxy bonding layer on the fatigue crack propagation rate of the film is discussed in terms of the measured crack opening displacement range at 250μm behind from the crack tip, Δφ_<250>.

View full abstract

The requirement of the high processing, high strength and light weight steel sheet recently rises in automobile industries based on various consideration such as safety, cost reduction, save energy, or even environmental protection. In the present paper, fatigue properties of extra-low carbon steels with different Sn (tin) contents, ranging from 0.002 to 0.22% mass, were investigated. The results show that the fatigue strength by 1×10^7 cycles and hardness number increase with increasing Sn content. In addition, the grain size of materials decreases with increasing Sn content. Fatigue cracks are initiated from the surface of specimens and propagate mainly in an intergranular mode for all of the extra-low carbon steels.

View full abstract

In order to clarify the influence factor on high-cycle fatigue of high strength steel SUJ2, the rotating bending fatigue tests have been carried out up to 10^8 cycles at room temperature. According to the experimental results, in the constant stress condition, the fatigue crack is initiated from a specimen's surface. On the other hand, in the case of two-step stress amplitude condition the fatigue crack is initiated in the specimen's surface parts. Furthermore, when the fatigue damage under the two-step stress condition is evaluated by a modified Miner's low, it will be better way to evaluate the fatigue life with S-N curve by surface failure and that by internal one respectively.

View full abstract

As far as the surface defects on the rail head, there are two kinds of defects called on "head check" and "dark spots" (shelling defects) in the Shinkansen's rail. In addition "White Phase" exists in the surface layer of rail head. The contents discussed in this paper are Contact Rolling Fatigue of Rails in the high-speed rail tracks. The authors have focused on "White Phase". It is clear that the initiation of "contact rolling defects" relates to Hertzian stress, Lubrication, tangential force etc. The main results obtained in the present study are as follows: (1) The occurrence of White Phase is attributed to alternate between water lubrication condition and dry condition. (2) The tangential force affect cracks initiation and propagation materially. (3) White Phase occurs in the origination of shelling early.

View full abstract

The object of this study is to investigate the effect of roller-working on the torsional fatigue strength of machine structural steel. The material used in this test was carbon steel (JIS standards S25C). Three types of specimens were plastically deformed from plain specimen to notched one and their deformation values were 0mm, 0.5mm and 1.0mm, respectively. The main results obtained in this study are as follows: (1) When the plastic deformation value is 1.0mm, the torsional fatigue strength of roller-worked specimen increases more than twice of that of non-roller-worked specimens. (2) When observing the transverse section of roller-worked specimens, a compressed and elongated structure is formed at the notch bottom of the specimen. (3) The hardness number of notched specimen at the notch bottom is higher than that of the non-roller-worked one. (4) According to surface observation of roller-worked notch specimen, there exist many non-propagating micro-cracks just before the specimen fail. (5) Using FEM analysis, the residual stress at the notch bottom of specimen with roller-working is estimated.

View full abstract

A study of the fatigue crack growth process in a low carbon steel has been carried out over a wide range of crack growth rates in air and vacuum at R=0.05 using the techniques of stereo-microscopy and scanning-transmission electron microscopy(STEM). Examination of the profiles of striations indicated that they are of the stair-case type, as originally proposed by Schivje. In this steel the striations were not resolvable below a growth rate of 10^<-4> mm/cycle. In the crack growth rate range from 10^<-4>mm/cycle down to 10^<-8>mm/cycle a dislocation cell structure developed whose dimension in the direction of fatigue crack growth was approximately 1.5 x 10^<-4> mm/cycle, independent of the ΔK level. The constancy of this dimension has been interpreted to indicate that the fatigue crack growth process at low crack growth rates is discontinuous rather than cycle-by-cycle in nature. However the evidence gathered in this investigation supports the view that the fatigue crack growth process occurs on a cycle-by-cycle basis even at low fatigue crack growth rates. As is known, striations are difficult to find on the fracture surfaces of specimens tested in vacuum. The use of the STEM technique provides evidence that a striation is created in vacuum on unloading, but that due to the greater extent of crack tip blunting in vacuum as compared to in air, in the next loading cycle the previously created striation is swept into the flanks of the blunted crack and largely eliminated. For comparison purposes an aluminum alloy, 2519-T87, was also tested.

View full abstract

Uniaxial tension fatigue tests were carried out in air and high vacuum environments using pieces made of Ti-6Al-4V alloy with several surface finishings. Effects of high vacuum environment on high cycle fatigue were investigated based on results of fatigue tests and the data arranged by initial stress intensity factor range (ΔK_<ini>). Considering results of fatigue tests that interior-originating fractures occurred near yield stress in high vacuum environment, it is necessary to pay attention to the interior-originating fracture in case of using this material in vacuum environment like space. Fatigue lives in high vacuum strongly depended on cyclic stress compared with those in air as stress decreased. As for data arranged by ΔK_<ini>, the dependence of the threshold stress intensity factor range (ΔK_<th>) on crack size in high vacuum is expected to be stronger than that in air.

View full abstract

Fatigue tests were carried out on a 6061 Al alloy under the ultrasonic fatigue test (test frequency 20kHz) and the conventional tension compression fatigue test (50-80Hz). In the ultrasonic fatigue test, the temperature of the specimen rises due to internal friction. In order to suppress the temperature rise in the specimen, the specimens under the ultrasonic fatigue test were cooled by air blow. Small artificial holes were introduced onto the specimen surface to investigate the crack initiation and growth behavior. In the crack initiation period, the crack growth rate at lower stress amplitude under the ultrasonic fatigue test and the conventional tension compression fatigue test decreased with crack propagation. The crack growth rate started to increase at a critical crack length. The critical crack length at lower stress amplitude under the ultrasonic fatigue test was longer than that under the conventional tension compression fatigue test. This might be the reason for that the fatigue lives at lower stress amplitude under the ultrasonic fatigue test were longer than those under the conventional tension compression fatigue test. A few small cracks were observed in the vicinity of the main crack tip of the specimen under the ultrasonic fatigue test. However, such small cracks were not observed on the specimen under the conventional fatigue test. The small cracks shield the tip of the main crack under the ultrasonic fatigue test and influence the crack growth rate.

View full abstract

Fretting fatigue tests in ultra high cycle region up to 10^9 cycles were performed to investigate whether the fatigue limit which means infinite life really exists in fretting fatigue. S-N curve had knee point around 2×10^7 cycles and it was verified that clear fatigue limit was observed up to10^9 cycles. An electric potential drop technique was developed to detect the crack under contact pad. The real time measurement of crack depth during fretting fatigue showed that initiated crack ceased to grow at about 2×10^7 cycles and the crack became a non-propagating crack at least up to 10^9 cycles. These results indicated that the fatigue limit can exist also in fretting fatigue. The infinite endurance can be achieved by the mechanism to become a non-propagating crack.

View full abstract

Fatigue failure in the very high cycle (gigacycle) regime plays an important role in many technical fields. Appropriate testing techniques have been developed and improved in recent years. Results of investigations on different materials like high strength steels, aluminium alloys, magnesium cast alloys and titanium alloys are reviewed in the present paper. Life time as well as fatigue crack growth measurements including endurance limit and threshold crack growth behaviour are reported.

View full abstract