The authors proposed previously the method for predicting the fatigue limit of smoothed specimens made of tufftrided steel from the hardened depth h and the distance from surface to center of specimen a. It was expected that the linear relation should hold between the improvement ratio of fatigue limit σw·t/σw·nt and h/(a-h) according to the prediction method. In case of the tufftrided steel supersaturated with nitrogen in the matrix, the linear relation was observed between these two quantities as expected and the validity of the method was clarified. On the other hand, it has been already reported that the fatigue limit of tufftrided steel decreases with the precipitation of nitrides of Fe4N. In this paper, the applicability of this prediction method is investigated on the tufftrided steel which has the nitrides of Fe4N. The results obtained are summarized as follows: (1) The fatigue limit of tufftrided steel decreases with the increase of nitride, and the decrease stops when the precipitation of nitride is perfectly completed. Its fatigue limit is still larger than that of non-tufftrided steel. (2) The linear relation is also observed between σw·t/σw·nt and h/(a-h) on the tufftrided steel which contains nitrides. Consequently, it is concluded that the prediction method proposed by the authors can be applied regardless of the condition of nitrogen. However, the linear relation equations differ depending upon the condition of nitrogen.
In this paper, the method of jet-electropolishing is described, by which few-microns thick foil specimens containing a considerably sharp crack can be prepared from bulk fatigued-specimens of iron, and the substructures in the vicinity of the cracks observed through a 2000kV electron microscope are shown. They are; (1) a fine cell structure formed in the areas near the crack sides, and (2) dislocation tangles of extremely high density developed around the crack tips, where the cell structure is indistinguishable. In the present study, many cracks were observed crossing cells, but few propagated along cell-boundaries. Neither microcracks nor voids were observed ahead of the tips of the cracks inside bulk specimens. The results obtained here suggest that, as a crack passes, the tangled dislocations ahead of the crack tip disappear or are rearranged to form the boundaries of cells, and, therefore, the propagation of fatigue cracks (Stage II) occurs in a work-hardened area, accompanied by plastic relaxation.
In the previous paper, the authors revealed that fatigue-induced substructures such as tangled dislocations of high density were responsible for the formation of nonpropagating fatigue cracks in sharp notched-specimens of iron. In stainless steel, such substructures are expected to be developed extensively in the course of fatigue tests. Nevertheless, it is still in dispute whether nonpropagating cracks are formed in this material. To clarify this point and to get more information on nonpropagating crack formation, fatigue tests were made on sharp notched-specimens of stainless steel. Furthermore, the changes in hardness and dislocation structures with an increasing number of stress cycles were investigated, and these were compared with those in the case of iron, respectively. Also, the martensitic transformation due to cyclic strain was examined. It was concluded that, in stainless steel, fatigue-induced substructures were developed at the root of the notch in the very early stage of fatigue tests, so that the stress required for the crack initiation was higher than those in iron and steel. This seemed to be the reason why nonpropagating fatigue cracks were difficult to be formed in this material.
The microstructural factors affecting the yield strength of Mn-Cr-B spring steel ausformed at a comparatively low temperature in the stable phase of austenite and tempered at temperatures of 400°-600°C were investigated by Conrad's method. The higher yield strength was obtained by this ausforming, but the ductility dropped slightly. The yield strength of the ausformed specimens increased almost linearly with the degree of deformation. The results of the two-step replica microscopy indicated that relatively large carbides were located preferentially at the martensite boundaries in the conventionally heat-treated specimens, but in the ausformed steel smaller carbides precipitated in a much more uniform and random morphology. Therefore, the mean-free path of carbide was smaller in the ausformed steel than in the conventionally processed specimen. It has been shown by the structural analysis that the major increase of yield strength by ausforming is attributable to the increase of dislocation density and the contribution of carbide dispersion is relatively small.
An investigation has been made on the influence of ausforming of stable austenite at a comparatively low temperature on the fatigue crack growth behaviour of Mn-Cr-B spring steel under alternate bending stress. The following results have been obtained. (1) The fatigue strength of the notched specimen is improved by this ausforming, and about 20-35% improvement in the endurance limit has been realized within 56% reduction in thickness. It seems that ausforming has no influence on the fatigue notch sensitivity. (2) Ausforming is an effective treatment for lowering the crack growth rate in the range of stable propagation from the early stage to the middle stage. In the range of unstable propagation, however, the crack growth rate of the ausformed specimen is higher than that of the conventionally heat-treated one, and in this case, the power m in the Paris' formula increases with the reduction ratio in ausforming process, while the coefficient c drops gradually. (3) The fractography reveals that the fracture surface of the conventionally treated sample consists of both the intergranular and the transgranular fracture, while the ausformed specimen gives only the transgranular fracture. The lowering of the crack growth rate by ausforming is related to the disappearance of the intergranular fracture, and this improvement of the fracture mode may be attributed to the precipitation tendency of carbide.
As a method to predict the fatigue lives of metallic materials when mean strains are given, the fatigue fracture criterion based on varying plastic strains with mean strains taken into consideration has been varified experimentally by carrying out the low-cycle fatigue tests under tension-compression. This time, the authors carried out the low-cycle torsional fatigue tests under completely reversible stresses programmed between the stresses more than the yield point and less than the elastic limit, by using 0.84% C piano wires. As the test results, comparatively large mean strains appeared in the specimens in the opposite direction to that of load applied at first. Based on the test results, the authors investigated experimentally the applicability of the above-mentioned fatigue fracture criterion which was derived in the tension-compression fatigue tests. Furthermore, in the torsional programmed fatigue tests with intermittent high level stresses the influence of high level stresses on the fatigue lives at low level stresses were also examined. And the applicability of the above fracture criterion has also been discussed on the basis of the experimentally determined correlation between the equivalent stress and the equivalent plastic strain amplitude. As the main conclusion, for estimation of fatigue lives by using the fatigue fracture criterion based on varying plastic strains, it is more accurate to calculate the fatigue lives by taking the mean strains into consideration rather than by ignoring them.
An acrylate beam of 10×10×55mm with a notch was subjected to lateral impact at supports 40mm apart in distance by using a steel hammer of 0.615kg in weight sliding on an air bearing runway. The impact velocity ranged from 0.9 to 1.4mm/ms. The notch sizes were 1, 2, and 5mm in depth with each 1mm bottom radius. The displacement of the hammer was measured as a function of time by the optical interferometry. From the result, the time dependence of load on the specimen was estimated. The theory of contact was successfully utilized to calculate the magnitude of the load. Also the stress in the specimen was obtained by means of the photoelastic method using a multi-giant-pulse laser as a light source. The time to the initiation of crack in the specimen, measured from the instance of the first contact of the hammer, ranged from 500 to 900μs depending upon the notch sizes. On the other hand, the duration from crack initiation to fracture was almost constant and was estimated about 50μs irrespectively of the notch dimensions. As was known before, the impact value calculated on the basis of the kinematic energy of the hammer varied as the notch size changed. However, nearly the same number of isochromatic fringes was observed at the notch bottom for all specimens at the moment of crack appearance. By using a photoelastic sensitivity-0.0527mm/kg, the stress was calculated to be in the range of 8 to 9kg/mm2.
The inhomogeneity and the residual stress are known to be important factors in determining the fatigue strength of laminated inhomogeneous metals such as induction hardened materials, carburized hardened materials, superficially worked materials and other surface treated materials. This paper is concerned with the inhomogeneity and the residual stress after some heat treatments of a clad plate composed of mild steel and high strength steel, which can be regarded as an example of the above mentioned materials. The results obtained are as follows: The microstructure, the hardness and the residual stress are discontinuous at the boundary between two materials of the clad plate. The distributions of the hardness and the residual stress were explained quantitatively in terms of carbon diffusion at the boundary in the clad plate.
In recent years, with the high faculty and the large capacity of electrical machines, the size of insulating materials used in those machines has become larger and larger, and the high arc resistance has been demanded on the electrical machines with arcing, especially. Since various spatial configurations of these constitution materials are being used, the effect of spatial configurations of samples on arc resistance of electrical insulating materials is a quite important factor for insulation design. In this paper, the arc resistance tests of large size sample were carried out under the different conditions from the previous arc resistance test, and the effect of spatial configurations of samples was studied. As the test procedure, the max. applied voltage was 50kV, the arc current was 10mA and the electrode gap length was 6∼50mm. Other test conditions were the same as those of the ASTM method. As the test samples, various plastics were used. The variations of arc resistances were tested by changing the angle of inclination. As the result, the effect of spatial configurations of samples on arc resistance was noticed when the gap length was comparatively large, and the arc resistance decreased when the samples were at the inclined configuration from the horizontal one. The arc resistance decreased only 61∼65% of the value at the horizontal configuration at a certain electrode gap length. In conclusion, the arc resistance is affected significantly by the spatial configurations of samples as well as their size, such as gap length.