A method for the simulation of carburized-quenching process is presented in this paper, which takes into account the coupling effect of temperature, stress and metallic structure. Here, emphasis is placed on the role of transformation plasticity. The constitutive equation of the transformation plastic strain was determined experimentally assuming that the strain rate is proportional to the applied stress for both pearlitic and martensitic transformations. The carburized-quenching process of a steel gear of 1.8%Ni-0.6%Cr-0.25%Mo with 0.2%C (SNCM420HK) was analyzed by use of the finite element method, and the calculated results of carbon content, temperature, metallic structure and stress were compared with the experimental data. The distribution of the residual stress revealed that the transformation plasticity should be introduced in the simulation.
In this study, in order to control artificially the inhomogeneity of the passive film of austenitic stainless steel SUS 304, a catholic reduction treatment was conducted in boiling 42% magnesium chloride aqueous solution at 416±0.5K. To see how these inhomogeneities change depending upon catholic reduction treatment, the morphology of the corroded surface of specimens were observed. In this way, the effect of catholic reduction treatment on the stress corrosion cracking susceptibility of test material was investigated. At the same time, the change of the character of passive film with catholic reduction treatment, which is deeply associated with the susceptibility of this material to the stress corrosion cracking, was studied through examination of physical properties. Namely, the hardness of its surface layer was determined and its structural change was inspected by the electron spectroscopy for chemical analysis (XPS). Then, the stress corrosion cracking tests were performed in a wide stress range from near the threshold stress to 0.2% proof stress, where no plastic deformation occurred macroscopically either in the untreated specimen or in the specimens cathodic reduction treated to various degrees in the same environment. As the results, it was shown that the difference of the current density used for the catholic reduction treatment affected the inhomogenity of the passive film. In addition, the change in stress corrosion cracking susceptibility was recongnized depending upon catholic reduction treatment conditions. Thus, it is clear that this fact can be explained well by the local corrosion susceptibility and the resistance to deformation of passive film.
The information obtained from fractured surface is quite relevant to failure analysis. In the present study, it was investigated whether the Exo-electron emission phenomena can be successfully applied to failure analysis. The Exo-electron (EE) emission from the fracture surface of a toughness test specimen was observed by using a scanning Exo-electron emission microscope. The difference in EE emission behavior between fatigue fracture and ductile fracture surfaces as well as the time decay of EE emission intensity were investigated. A significant difference in EE emission intensity was observed between fatigue fracture and ductile fracture surfaces. A good correlation was found between the J-integral range at fatigue cracking and the EE emission intensity of the fatigue fracture surface. The characteristics of decay of EE emission intensity was given as I(t)=I0exp(k1t)[A+Bexp(-k2t)] where I0 is the intensity of EE emission at t=0, k1, k2, A and B are constants. There is a good correlation between ΔJ and k1. ΔJ can be estimated by determining the k1 value from the decay curve.
In this study, statistical fatigue tests in rotating bending were carried out by using round bar specimens of alumina ceramics, and the fatigue life distribution was examined at five stress levels by assigning 20 specimens at each level. Distribution characteristics of static bending strength were also examined, and the relationship between static and fatigue strengths was discussed as P-S-N characteristics in a wide range of fatigue life. Furthermore, SEM observations of the fracture surfaces were made to investigate the effect of material defects on the fatigue life. The main results obtained are summarized as follows; (1) The distribution characteristics of static bending strength in alumina ceramics were well approximated by three-parameter Weibull distribution including a location parameter. (2) The fatigue life of this material was also governed by three-parameter Weibull distribution, and its quantitative P-S-N characteristics were obtained in a wide range of fatigue life, 1≤N≤108. (3) The fatigue life tends to increase with decrease of defect size, but it has a large scatter even if the defect size is same. This trend is more marked in the case of small defects.
This paper deals with a statistical data pooling method for two different fatigue life data and its application to multiple samples from the data base. The proposed data pooling method is based on the statistical tests of significant difference between regression curves fitted to S-N data. Those treated in this paper are: (a) for inclined parts, (1) test of linearity, (2) test of equality of residual squares and (3) test of equality of regressive coefficients, and (b) for horizontal parts, (4) test of equality of mean and (5) test of equality of variances. As an extension of the method, the algorithm of pooling multiple samples is also developed. The proposed method is applied to the fatigue life data base of metallic materials compiled by the Society of Materials Science, Japan, as an illustrative example.
There exist many well-known uncertainty factors associated with the fatigue crack propagation process, among which two major uncertainty factors, one due to material inhomogeneity and the other due to the randomness of applied loads, undoutedly play an inevitably important role. Uncertainties caused by these two factors must be clarified not only individually but also jointly, since the cross-effect of both factors may arise under random loading and this is often the case in engineering reality. In this respect, the present paper deals with the case where crack propagation resistance and stress amplitude of applied load are both stochastic processes and, in addition, each correlation decays exponentially with the increase of time difference. Towards this case has first been constructed a theoretical stochastic model with the aid of a Markov approximation method. Then the aforementioned cross-effect has been evaluated and discussed in detail, and the distribution characteristics of the model have been investigated by utilizing numerical computation techniques. Finally, the wide applicability of the proposed model has been emphasized by exemplifying a useful application of this model towards the inspection period determination problem which is often encountered in the practical reliability-based design of machines and/or structures.
The effect of the variation of stress frequency on fatigue crack propagation in pure titanium plate, which has remarkable strain rate dependence in the plastic region and notable orthotropy, was studied. Fatigue crack propagation tests were carried out under four stress frequencies: two constant stress frequencies (20Hz, 0.02Hz) and two variable stress frequencies changed step-wise (20→0.02Hz, 0.02→20Hz). The test specimens were cut out from the rolling direction and transverse to the rolling direction of the plate. An anisotropic elasto/visco-plastic analysis of the fatigue crack propagation was performed by the use of the orthotropic elasto/visco-plastic overlay model, and a comparison between the change of crack propagation rate due to variation of stress frequency and the visco-plastic strain behavior at the crack tip calculated by the analysis was made. The results obtained are summarized as follows: (1) It was found from the experiment that the crack propagation rate changed characteristically depending on cutting out direction of the specimen because of the anisotropy of titanium plate and on the variation of stress frequency. (2) A parameter closely related to the fatigue crack propagation rate is the visco-plastic strain range at the crack tip. (3) The effect of the variation of stress frequency on the fatigue crack propagation rate may be explained by the variation of visco-plastic strain range at the crack tip based on the strain rate dependence and the anisotropy of the material.
In order to improve the fracture toughness of an epoxy resin, five kinds of glycidyl compounds were examined as a modifier of the resin. Butyl, phenyl and allyl glycidyl ethers, and glycidyl methacrylate were found to be effective to lower the viscosity of the resin. However, the fracture toughness of the resin was not improved by the 10phr modification of these monoglycidyl compounds. Polypropylene glycol diglycidyl ether (PPG) was effective both for the reduction of viscosity of the liquid resin and for the promotion of fracture toughness of the cured resin. The fracture toughness of the epoxy resin was doubled by the modification with more than 30wt% of PPG. The heat treatment of the PPG modified resin resulted in the improvement of fracture toughness, at a small sacrifice of the deflection temperature under load. In the case of PPG modification, it was found that the smaller the crosslinking density is, the higher the fracture toughness is.
The present study was carried out to give a physical interpretation to the absorbed energy in Charpy impact test for specimens with mechanical heterogeneity. Standard Charpy impact tests were conducted on several kinds of structural steel with weld bead made by electron beam welding. The results obtained were as follows: (1) It became clear that a large fraction of the absorbed energy in Charpy impact test was associated with the plastic-strain energy consumed to deform a specimen. (2) The absorbed energy of a specimen with mechanical heterogeneity in Charpy impact test was affected not only by the mechanical properties of material under the notch root but also by the mechanical heterogeneity of weld bead in the vicinity of notch. (3) The relation between the absorbed energy and test temperature was determined quantitatively as a function of mechanical properties and constraint distance.
Mechanical tensile properties of hybrid composites built up from many thin Al (aluminum-IN30) sheets and unidirectional CFRP prepleg layers (fiber volume contents 60%) were investigated. The thickness of Al sheets and of CFRP layers was 0.1mm and 0.12mm each. Their total number was 20. They were laid up alternately and symmetrically. The results obtained are summarized as follows: (1) The estimated values of elastic moduli agreed well with the experimental ones. (2) The anisotropic residual stress after curing was calculated and its correctness was checked by using the stress-strain diagram. (3) Dependence of yielding strength on prestrain was verified theoretically and experimentally. (4) The average value of longitudinal strength was about 35% greater than that expected by the rule of mixture. Also the average transverse ultimate strain was about 65% greater than that of monolithic CFRP. The reason for these results was assumed to be that the Al sheets which divide the CFRP into many thin layers tend to arrest the crack propagation, and consequently the failure mechanism of CFRP layers due to crack propagation changes to the cumulative weakening one.
Many trials of fine ceramics machining are being conducted because of their increased importance in industry. One of them is the machining utilizing electrochemical discharge phenomena, which is tested as wire EDM, combined energy machining (combining chemical reaction, electrodischarge phenomena and mechanical action) and so on. However, the effect of the electrochemical discharge phenomena on fine ceramics has not been known yet. In this paper, the influence of electrochemical discharge phenomena on the surface of fine ceramics was studied. Through this investigation, it was found that the NaOH electrolyte caused alkaline erosion significantly under the condition of existing electrochemical discharge. The affected layer of ceramics decreased its surface hardness, and the thickness of the layer was from 1 to 5μm in the condition of this experiment. In the case of more severe electrochemical discharge conditions, the ceramic material was removed by its energy, and the removal rate increased with increasing alkaline concentration. Ceramics with low thermal conductivity such as ZrO2, Al2O3 or Pyrex glass caused cracks due to thermal stress by severe electrochemical discharge.
An optimization method of the proof test loading factor is described. The decision of an optimum load factor under the consideration of prior test results was made by using the Bayesian analysis. Moreover, the reliability utility function which shows the utility value of reliability was newly introduced, so that the present optimization method became similar to the Bayesian decision making procedure defined as the maximization of expected utility. The initial conditions used were the prior distribution A0(μ) of mean value μ, the coefficient of variation COV of the strength distribution which was assumed to be the normal distribution, and the degree of belief Rcoeff used to evaluate the reliability by the Bayesian analysis. As the result of the numerical calculation, it was concluded that the effective estimation of the reliability could be performed based on the test results using the optimum loading factor obtained by the present method.
Hardness of carbon fiber reinforced plastics (CFRP) and Al-2 and 4wt% Cu alloys containing columnar grains was measured with a wedge-shaped indenter, and was compared with their static tensile properties. Hardness measurement was conducted with a Rockwell hardness tester. The hardness and mean pressure of CFRP decreased with the angle θ between indentation normal and fiber direction. The orientation dependences of the hardness and the mean pressure agreed with those of Young's modulus and the fracture strength of CFRP. However, no linear relation existed between the mean pressure and the strength. The hardness and the mean pressure of Al-Cu alloys took a minimum when the angle θ was about π/4 rad, and the orientation dependence of the mean pressure disagreed with those of the yield stress and the ultimate tensile strength. The relation between the mean pressure and the yield stress was described by the equation: σy=(P/3)(0.0017)n, where n is the strain hardening coefficient.