Pulsating tension fatigue tests were made of Hi-Strand and of JIS PC-strand, and the following conclusions have been obtained: (1) Hi-Strand and JIS PC-strand have almost same fatigue strength at N=106 in stress amplitude, and as the former has larger ultimate tensile strength and larger yield strength, the fatigue strength diagram (in σa-σm representation) of Hi-Strand is extended along the abscissa (σm), the values of the ordinate (σa) remaining unchanged. (2) But at N=107, it is expected that Hi-Strand has smaller allowable stress amplitude for fatigue, so if the fatigue behavior of PC-strand is brought into question, the materials which have very high tensile strength are not preferable.
In the previous report, the fatigue behaviour of 1×7 PC-strand was investigated. In this paper a report is made of the fatigue tests that were made on the wire which the said strands composed. The fatigue failure of the test specimen at chucked-off end impedes the experiment to find the fatigue strength of the specimen. To meet this difficulty the following method was employed. Several long specimen pieces were used, and at every failure at the chuck rechucking was performed on the shortened specimen. The fatigue test was continued by subjecting the specimens to repetition of failure and rechucking, and replacing one specimen with another when shortest limit was reached. The following conclusions have been obtained: (1) The notch factor β for the strength at chuck is estimated at between about 1.5 and 2.0. (2) The notch factor β for the strength of the strand compared to the wire is estimated at 2.0∼3.0. These values of β are calculated between allowable stress amplitudes at the same mean stress.
In order to account for the embrittled phenomenon in steels charged with hydrogen, several considerations have so far been proposed, such for instance as the rift network theory, the void theory, the formation of cottrell atmosphere, and the interaction of hydrogen in the lattice work and in the stress field. The authors have tried to make an X-ray study of the mechanism of steel getting charged with hydrogen by performing comparative experiments on the polycrystalline material and on the single crystal material, both electrolytically charged with hydrogen. The former has been subjected to analysis of the X-ray diffraction line profiles, and the latter to analysis of its lattice distortion. The results indicate that thus introduced hydrogen gives rise to plastic deformation in the crystal lattice and there is no recovery phenomenon observable after having been subjected to room temperature aging. By annealing treatment of the material at higher temperature, however, the crystal have been recovered in a similar fashion to mechanically deformed crystal. It is believed that the recovery in this case is not due to the diffusion process of hydrogen. In the case of single crystal, the micro-focus X-ray source (30∼20 micron) is employed to take Laue pictures which showed the heavily deformed spots on the films. This is believed to be due to the distortion in the crystal lattice as well as to the existence of void caused by being charged with hydrogen.
The X-ray method of stress measurement is taken as a unique method of non-destructive measurement of local surface stress. In the past few years, the experimental technique and theory of the stress measurement by X-rays have been remarkably improved, and the recent studies in this field have achieved the sufficiently accurate measurement of applied or residual stress for homogeneous metallic materials such as iron, steel, light metal, alloy and so on. However, the method of X-ray stress measurement is not established for inhomogeneous metallic materials such as cast iron, and there remain yet a few problems upon which there is much controversy. One of the most important problems is to find out the suitable elastic constant for the practical X-ray stress measurement of cast iron. In the previous papers, the authors discussed the effects of graphite on the deformation behavior of spheroidal and flake graphite cast iron from the lattice strain measured by X-rays. It was also suggested that the X-ray elastic constant obtained at after 3 to 10 stress cycles should be used for X-ray stress measurement of cast iron, and that its constant obtained by X-rays is different from the value obtained by the mechanical method. Besides, there is a wide difference in these values obtained by X-rays in cast iron from those obtained in wrought iron or steel. In the present investigation, further experiments were made of the application of X-ray stress measurement to cast iron. The elastic constant of spheroidal and flake graphite cast iron was measured by the mechanical method and by X-rays in detail. The X-ray measurement of applied stress and residual stress induced by grinding was carried out using the obtained X-ray elastic constant. The following conclusions were mainly obtained for these materials; (1) The macroscopic elastic modulus (EM) decreased with the increase of stress. For flake graphite cast iron it decreased having a straight line relation with stress. On the other hand, the elastic modulus pasted through two-stages for spheroidal graphite cast iron. It showed gradual decrease in a linear relation to stress to the stress of 25kg/mm2 and, after that, the rate of its decrease became larger until fracture occurred. (2) The macroscopic elastic modulus (EM) of flake graphite cast iron was markedly different from that of spheroidal graphite cast iron. It is believed that this wide difference is due to the effect of difference in the shape of the graphite on each deformation behavior. (3) On the other hand, the X-ray measurement yielded almost the same elastic constant for the two materials. It is considered that the value of the X-ray elastic constant of cast iron is controlled not so much by the shape of graphite as by the volume ratio of graphite to matrix. Therefore, it may be said in conclusion that the elastic constant measured by X-rays should be used in the X-ray stress measurement of cast iron. (4) If the X-ray elastic constant was used for stress calculation, the applied stress could be measured by the X-rays having a good correlation with mechanically applied stress. (5) A good correlation between the value measured by the X-ray method and by the mechanical method was found for the residual stress induced by grinding.
A method of calculating residual stress distribution in a hollow cylinder during successive removal from its surface residual stress obtained by its X-ray measurement on its outer or inner layer is reported in the present paper, and comparison is made of the result reached by this method with that reached by the conventional method of mechanical measurement. It is reported also how it is possible by adaptation of Schaal's method to measure tangential stress on the inner surface of cylinder directly without cutting off the cylinder in order to set it in longitudinal direction. This procedure may be applied practically to the stress measurement of the specimens of geometrical size where it is difficult to apply normal X-ray method.
The changes in residual stress occurring in hardened steels as a result of rolling contact operation between the cylinder and the ball are described. The relationship of contact load, the number of the loading cycles and the hardness of the specimen (HRC63-54) to the extent of residual stress alteration have been examined. The threshold stress exists, below which no marked residual stress change occurs. For HRC 62, 58 specimen, the threshold stress is about Pmax=350kg/mm2, and for HRC54 specimen, it is about Pmax=250kg/mm2. With the rise of contact stress, the peak value of compressive residual stress due to the rolling contact becomes larger, and the location of the peak becomes deeper. The peak depth corresponds to the depth of maximum shear stress. When Pmax=600kg/mm2, the residual compressive stress in the HRC58, 54 specimen reaches to the saturation within the loading cycle of 10 or so, and the lower the hardness of the specimen is, the lower the saturation value is. In case of the rolling contact between the cylinder and the ball of the same hardness (HRC58, 54), the residual stress changes due to the operation are similar in both specimens. On the contrary, the contact between the HRC62 cylinder and the HRC63 ball produces a different residual stress distribution in each specimen. The reason for this producing unsimilar distributions is not clear, but it is thought to be due to the difference in hardness or in residual stress distribution of specimens before the operation.
Many of recent papers on fatigue crack propagation have pointed out the importance of applied stress as factor determining crack propagation rate. However, the relation between plastic deformation ahead of crack tip and propagation rate has been observed by few investigators. In this study, 0.01% C steel specimens were initially cold-rolled to 15 reduction percent and the slip bands zone ahead of carck tip during fatigue was investigated by means of an optical microscope. The mechanism of fatigue crack propagation in this material is discussed on the basis of the interrelation found among the three parameters; the propagation rate, the slip bands zone size ahead of crack tip and the applied stress amplitude. Discussion was also made about the dormant or delayed period seen in the crack propagation. The results are summarized as follows: (1) The fatigue crack propagation process of 15% cold-rolled specimens has three stages, that is, the rapid propagation stage (Stage 1), the constant growth rate stage of long period (Stage 2) and the final stage of faster growth yielding complete fracture (Stage 3). (2) The crack propagation rate Rc in Stage 2 has no relation with the stress concentration factor of notched specimens and is expressed in terms of nominal stress amplitude σ as follows, Rc=5.3×10-12σ5.6 (3) The relation between Rc and the slip bands zone size ξ0 ahead of crack tip is established as Rc=2.1×10-4ξ01.8 Therefore, Rc is approximately proportional to plastic energy ahead of crack tip. (4) The dormant or delayed period in crack growth is caused in the forms as (i) grain boundary blocking, (ii) crack branching, (iii) grain boundary propagation and (iv) unfavorable predeformation for crack propagation. The crack propagation blocked by grain boundary is found more often under lower stress amplitude, while the higher stress amplitude fatigue shows crack propagation delayed by crack branching.
The rates of fatigue crack propagation in circumferentially cracked 0.35, 0.47 and 0.81% carbon steel specimens were examined in rotating bending. The stress intensity factor, K, of the circumferentially cracked specimen in bending was analysed using Neuber's results. These experimental and analytical approaches to the problem of the fatigue crack propagation suggest the following results. (1) The rate of fatigue crack propagation in the circumferentially cracked specimen under rotating bending may be expressed as a function of K, as is shown in the equation, dl/dN=CK4, that has been proposed by Paris et al. (2) The fatigue crack propagation characteristics are found to be not widely different for any of the three materials.
The increasing use of liquid metal coolants for compact heat exchanger demands the basic knowledge of the fatigue properties of solid metals when brought in contact with liquid metals. In the present paper, the fatigue test of notched specimens made of low carbon steel was carried out in molten tin, lead and zinc at elevated temperature with the fatigue testing machine of the rotating loading and fixed-specimen type. The results obtained are summarized as follows; (1) The fatigue strength of a notched specimen in molten tin is lower in the higher stress with a smaller number of cycle range at the test temperature of 250∼550°C than that in air, but higher by 13∼34% in the lower stress with a larger number of cycle fatigue. (2) The fatigue strength of a notched specimen in molten lead compared with that in air is considerably reduced at every test temperature over the whole range of number of stress cycles. (3) In molten zinc the fatigue strength of a notched specimen is not so extremely reduced as in the case of an unnotched specimen in the lower stress with a larger number of cycle range because of mechanical strengthening by intermetallic compounds at the root of the notch. As mentioned above, the fatigue strength in molten metal depends upon the kind of molten metal, the shape of the specimen, test temperature, the number of stress cycles and so on. In order to arrange the effects of environment on the fatigue strength, fatigue-environmental factor, fenv, was newly introduced, which is the ratio of fatigue strength in a liquid metal environment to that in air. Consequently, it is considered that the factor is adequate in estimating the fatigue strength in various environments qualitatively and quantitatively.
Annealed mild steel specimens have been subjected to rotation under the bending stress of 21∼31kg/mm2 with the speed of stress repetition at 3∼3000rpm at room temperature to examine its dynamic strain amplitudes and creep strains. A wire strain gage and an ocillogram have been employed by means of a slip ring to record the process. (1) Fatigue process, which is expressed by the dynamic strain amplitude, can be divided into three stages: first stage of softening, second stage of hardening, and third stage of apparent softening. (2) When the speed of stress repetition is slow and the stress amplitude is high, the amplitude of dynamic strain increases to a larger value, and the effect of stress amplitude on the specimens is larger than that of the speed of stress repetition. From the experimental results, it has been made clear that fatigue life is in inverse proportion to the peak of dynamic strain amplitude, i. e. the maximum value of softening process. (3) With the exception of the third stage, the dynamic strain amplitude on fatigue process is proportional to the transient creep strain during fatigue process.
This investigation has been conducted to make clear the effect of strain rate on the internal structure and the mechanical properties of rapidly stretched copper and copper alloy. Tough pitch copper, 7/3 brass and free cutting brass specimens have been used in this study. The main results obtained are as follows; (1) The strain rate sensitivity of mechanical properties is recognized in annealed and prestretched specimens of tough pitch copper, 7/3 brass and free cutting brass. Their yield strength, tensile strength and elongation increase with increment in their strain rate. (2) It is proved that the work of hardening has been more effectively made in the rapidly stretched tough pitch copper than in the statically stretched material. (3) Concerning 7/3 brass and free cutting brass, there is little difference in the effective work of hardening it between the case in the rapidly stretched specimen and that in the statically stretched one.
The present study has been made experimentally and theoretically to inquire into the characteristics of Shore hardness in the measurement of the hardness of plastics by comparing it with that of metals. The effect of the measuring conditions, such as the radius of the hammer edge, the weight and the falling height of the hammer, on the value of Shore hardness has also been experimentally investigated. As far as the principle is concerned the value of Shore hardness means the rate of elasticity in the strain. The rate of elasticity in plastics is generally much greater than in metals, as its modulus in plastics is far smaller than in metals, so that Shore hardness value of plastics becomes generally higher than that of metals, although the Vickers hardness value of plastics, which is one of the representative indentation hardness value, is generally much lower than that of metals. The severer impact stress is given by the hammer of Shore hardness tester, that is, the smaller is the radius of the hammer edge, and the greater the weight and height of the hammer, the lower will the Shore hardness value of plastics become.