Variation of dislocation density during the fatigue process of 0.1% aluminum-killed carbon steel was explored by the aid of transmission electron microscopy technique. The fatigue tests were conducted under a constant axial stress amplitude with zero mean stress at a frequency of 30Hz. The results showed that the dislocation density of about 108/cm2 under the annealed state increased with cycles and reached a value of 5×109/cm2 at ten percent of the total life. This trend of growth well corresponded to that of plastic strain amplitude. By the analyses of multiplication and annihilation processes of dislocations, the change of dislocation density ρ with the number of stress cycles n was expressed by ρ=α/β1/1+ke-αn, k=α/βρ0-1, where α and β are the constants, and ρ0 is an initial density before fatigue tests. It was demonstrated that this equation was applicable not only to the present low carbon steel, but also to pure iron, pure aluminum and prestrained S15C steel.
The present paper deals with a problem of propagation of longitudinal plastic waves in a semi-infinite bar subjected to a constant impact velocity, according to Malvern's strain-rate dependent theory. Especially, we shall discuss whether or not Malvern's theory can account for the plateau of uniform plastic strain adjacent to the impact end of the bar, which is predicted by Karman's theory neglecting a strain-rate effect and is observed experimentally. Malvern could not explain the strain plateau in his own calculations. This is considered so far to be an only weak point in the strain-rate dependent theory, though it can account for the extensive phenomena of high-rate deformation and plastic wave propagation in a bar. In this study, the following conclusions were obtained. (1) The existence of the strain plateau can be also predicted by Malvern's theory. (2) Its appearance requires a relaxation time which is governed by the strain-rate dependence of material and the impact velocity at the end of a bar. (3) As the strain-rate dependence of material and the impact velocity increase, the relaxation time needed for the appearance of the strain plateau increases.
The problem of a cylindrical flat-ended projectile impinging on a flat rigid anvil has been analyzed by many investigators in order to assess the dynamic stress-strain relation at high strain rate. As the impact speed used in such an experiment can be higher than that of other tests such as Split Hopkinson Bar Method, temperature rise caused by plastic work cannot be neglected. Thus, in this paper, the simulation of elasto-plastic wave propagated in a finite bar, which impinged on a flat rigid anvil, has been made with a computer by taking temperature rise into account. The analysis in this paper was based on the following three items. (1) A constitutive equation used in the analysis is rate-dependent. (2) Temperature rise has influence upon the static stress-strain relation without delay time. (3) The static stress is a linear function of absolute temperature. The problem was formulated and solved numerically by the method of characteristics with a supplemental rectanguler net. The numerical results show that temperature rise has influence upon the deformation resistance of specimen and unloading wave from free end. Consequently, the specimen length, impact stress, material constant K defined plastic strain rate and initial temperature have considerable influence upon stress, strain and time of contact.
The effect of creep deformation on microstructural change in 12% Cr heat resisting steel was investigated by comparing the gauge and head regions of creep ruptured specimens. Such microstructural changes as recovery, precipitation rate and coarseness of precipitates were promoted by creep deformation. When the precipitation behaviors of the gauge and head regions with the same hardness were compared, the precipitation behavior of the gauge region was smaller in amount, smaller in mean diameter of precipitates and shorter in interparticle spacing than that of the head region. The promoting ratio of recovery increased with increasing the strain and with lowering the temperature. This result supports the thought that increase in diffusion coefficient by creep deformation promotes microstructual change.
In the previous paper the fracture criterion of alumina ceramics subjected to triaxial stresses had been examined by carrying out bending tests under various hydrostatic pressures. By considering that in bending the effect of surface crack on fracture would be remarkable and that of non-uniform stress distribution produced in the material was probable, it was concluded that the specimen should be tested under a uniform stress state as possible. In this paper, the material was fractured by applying internal pressure to a thin-walled tubular specimen with open ends under hydrostatic pressure. The fracture criterion was reexamined by comparing the present experimental results with the results obtained in the previous paper. (1) The fracture stress plotted in the principal stress space showed a similar tendency to the theoretical results for the fracture criterion obtained by taking account of small cracks in the material, in which the material constants had been decided so as to satisfy the experimental results in bending tests under hydrostatic pressures. Thus, the validity of the assumptions in deriving the theoretical results was confirmed. (2) The experimentally obtained fracture stress was somewhat lower than the theoretical results. This fact may indicate that the theoretical results include the effect of surface crack or non-uniform stress distribution. (3) In the fracture criterion, the respective material constants of πKIc2/4a and KIc/KIIc were corrected, where KIc and KIIc are the respective fracture toughness values for mode I and mode II, and a is the radius of a penny-shaped crack. The corrected fracture criterion agreed quantitatively with that determined from the experimental results.
Fatigue tests have been made on a SUS 316 stainless steel at 550°C under stress patterns of intermittent strain wave form which is composed of cycling of a large pulsating strain and superposed small strain cycling. A rest time was 5sec. and it was introduced at zero strain of a large strain cycle. The method of life prediction for the strain cycling was examined on the basis of a linear cumulative damage concept. A linear rule based on the fatigue life curve under a constant strain amplitude gave a reasonable prediction in the case that the superposed small strain was a larger fraction of the basic large pulsating strain, although it gave a poor prediction in the case that the superposed small strain was a small fraction of the latter. By using Manson's 10per cent rule, a reasonable life was predicted regardless the fraction of the superposed small strain to the basic large pulsating strain.
The purpose of this research was to study the fatigue strength and mechanical properties of anti-tufftrided steel. The non-tufftrided, tufftrided and anti-tufftrided specimens of SS 41 and SM 50A were used and their hardness, tensile strength and fatigue limit were measured. The relation between the tensile load on spot welded cross joints and the tufftriding time was investigated and the fracture surfaces were observed by scanning electron micrographs in order to study the effects of inorganic coating of the anti-tufftrided specimens. The results obtained are summarized as follows. The differences in microstructures, hardness, tensile strength and compressive residual stress were obviously recognized between the tufftrided and anti-tufftrided specimens. But the fatigue limits of both specimens were approximately the same. It is note-worthy that the anti-tufftrided mechanical parts maintains the mechanical properties similar to those of the non-tufftrided one and retains the fatigue strength equal to that of the tufftrided one. Against tensile load on spot welded cross joints, the tufftrided specimen markedly reduced its strength compared with the non-tufftrided specimen, but this strength reduction was not so much on the anti-tufftrided specimen. The effect of anti-tufftriding coating was confirmed also by the results of fractographical observation.
The coaxing effect in plane bending fatigue was investigated on the plain specimens of annealed 0.5%C steel under zero and positive mean stresses. The main results obtained are as follows: (1) The mean stress has hardly any effect on crack initiation, but crack propagation is remarkably affected by the value of mean stress. (2) The Fatigue limit, in the case where the mean stress is positive, is controlled by the limiting condition for propagation of a micro-crack. This is closely related to the coaxing effect. (3) The Coaxing effect of a plain specimen, in the case where a non-propagating micro-crack formed under the first stress level is short, is more remarkable than that in the case where the micro-crack is long. (4) The coaxing effect of a plain specimen is bigger in the case of positive mean stress than in the case of zero mean stress.
In order to understand the effects of various factors on fatigue life of concrete under varying repeated load, two-stage and three-stage stress fatigue tests were performed. As the factors, the order of repeated loads, the number of cycles in initial stress level (S1) n1, and the following stress level S2, S3 were selected. Since the values of fatigue life of concrete are known to scatter widely, the fatigue test results were evaluated statistically. The experimental results show the followings: (1) The sum of cycles ratio for the same test condition follows a logarithmic normal distribution. (2) The estimation of fatigue life with Miner's rule predicts on the unsafety side when the initial stress level is higher than the following stress level. (3) The Miner's rule, however, may be applicable for the prediction of fatigue life of concrete under varying repeated load when the order of repeated load is uncertain.
The effects of stress ratio and microstructure on the fatigue crack propagation near the threshold in a high-strength low alloy steel were analysed based on the crack closure measurement and the microscopic observation of fracture surfaces. The following results were obtained. (1) The effects of microstructure and stress ratio on the relation between the crack propagation rate and the stress intensity range were large in region A near the threshold. The threshold stress intensity range increased linearly with the square root of the mean free path. When compared at the same crack propagation rate, the relation between the stress intensity range and the stress ratio was dividable into two regions, i.e., Kmax controlled region with low R values, and ΔK controlled with high R values. (2) Region A was dividable into regions A1 and A2 in the relation between the rate and the effective stress intensity range. In regions A2 and B, the rate was expressed in a unique power function of the effective range without respect to the microstructure and the stress ratio. In region A1, very close to the threshold, the rate depended on the microstructure but not on the stress ratio. Thus, the effect of stress ratio on the crack growth rate can be ascribed to its effect on the crack closure. (3) The transition from region A to B can be caused by either the transition in crack closure behavior or the change in crack growth mechanism. The latter condition becomes predominant when the reverse plastic zone size is equal to the mean free path of dislocation motion. The transition in crack closure is dependent on the ratio of the reversed plastic zone size to the prior-austenite grain size as well as on the material itself.
The effect of stress frequency on fatigue crack propagation in 99.5% pure titanium which has remarkable strain rate dependency in the plastic region was studied. Fatigue crack propagation tests were carried out under three kinds of stress frequencies (0.02, 0.2, 20Hz). FEM elasto/visco-plastic analysis of fatigue crack propagation was performed, and the comparison between the dependency of crack propagation rate on stess frequency and the strain behavior at the crack tip calculated by the analysis was made. The results obtained in this study are summarized as follows: (1) It was found from the experiments that the crack propagation rate was approximately in inverse proportion to fn(n>0), where f is the stress frequency. (2) The parameter closely related to the fatigue crack growth rate is the strain range Δεy or visco-plastic strain range Δεyvp at the crack tip. (3) The effect of stress frequency on fatigue crack propagation rate may be explained by the variation of Δεy or Δεyvp based on the strain rate dependency of material.
The mechanochemical reactions of polyvinyl chloride (PVC) were investigated at room temperature by using a self-made high pressure apparatus capable of simultaneous shear deformation. Static pressures applied were 5, 10, 15kbar for hard PVC resin and less than 5kbar for soft PVC resin. The shearing curves for hard PVC resin showed the decrease of shearing stress at the initial stage of shearing because of the occurrence of the major mechanochemical reaction at this stage. The mechanochemical reaction rate and the limiting molecular weight were evaluated by fitting the rate equation to the experimental data on molecular weight of hard PVC resin. All shearing curves for soft PVC resin did not show any definite yield point. Although the addition of plasticizer affected the shearing behavior, the shear-strength under pressure was not appreciably affected by adding over 50wt-% to pure PVC. No general conclusion about the effect of plasticizer content on the molecular weight was obtained.