The X-ray method of stress measurement is particularly important in the determination of residual stress since it is the only accurate method for measuring residual stress non-destructively and is very useful for determining local stress distribution. With a great progress in the equipment and technique for X-ray stress measurement, its accuracy has been much improved recently. For example, there has been almost no problem for measuring the residual stresses induced by heat treatment. As for the residual stress developed by welding, however, the possibility of applying the X-ray method is not certainly confirmed because of lack of actual data. In the present study, errors in stress measurements were examined for three specimens which were welded by large or small heat input. Especially, the dependence of the errors on the parameters representing the shape of X-ray diffraction profiles was studied. The main results are summarized as follows: (1) The residual stress induced in weld metal by the welding of small heat input can be measured with the error less than ±4kg/mm2 if the limit of confidence is set at 95 percent. The error is much more for the residual stress induced by the welding of large heat input. (2) The residual stress developed in heat affected zones can be measured with the errors less than ±5kg/mm2 irrespective of the level of heat input. (3) A good correlation is observed between the errors in stress measurements and the standard deviations of half-value breadth of X-ray diffraction profiles. (4) The loading of additional stress affects the errors of stress measurements.
A series of fatigue tests were conducted, from a statistical viewpoint to obtain foundamental data on the scattering characteristics of fatigue strength, on 3 kinds of commercial plain carbon steels of 0.45% C type and 2 kinds of 0.25% C type. The analysis of the results revealed that the scatter in fatigue strength of smooth specimens may generally be fitted to the normal distribution, and it is possible to obtain P-S-N curves by shifting laterally on S-N diagram the plot of median fatigue strengths estimated by Probit analysis from the data obtained at different numbers of cycles. The scatter in fatigue strength of smooth specimens made of the same steel subjected to a constant condition is affected by the difference in the sample rods from which the specimens were made. For the specimens subjected to different heat treatment conditions, the scatter tends to increase as the fatigue strength becomes higher. The variation coefficient of fatigue strength at 107 cycles is estimated to be approximately 2.5% for all the cases studied, except for the case of 0.45% C steel tempered at 350°C to achieve high brittleness, where it is 4.9%. For notched specimens of the same steel, on the other hand, the variation coefficients are 6.3% and 5.3% for the specimens tempered at 600°C and 350°C, respectively. From the fact that the notch causes a large increase in the variation coefficient for ductile materials but not so much for brittle materials, the process of crack propagation is thought to have great influence on the scattering phenomenon of fatigue strength.
The standard method for X-ray stress measurement was recently established by the Committee on Mechanical Behaviour of Material, Japan. Thus, the stress determination by means of X-rays will be more widely used in the future at laboratories or factories on the basis of this standard, and the automatic measurement will be required as the numbers of measurement increase. In“the standard method”, the diffraction profile is recorded as the intensity-angle curve by means of a counter. The peak position of the profile, 2θ, is graphically determined by the half-value breadth method, and the stress σ is then calculated from the gradient of 2θ-sin2ψ diagram. In this paper presented is a half-value breadth method by digital computation to determine the peak position of profile, which is suitable for the automatic measurement of X-ray stress. Several examples are also shown. In this method, the diffraction intensities are converted to the digital values at equal intervals in the diffraction angle. The background line of profile is calculated from the data of both ends of the profile on the basis of correlation coefficient of linear regression. The half-value breadth line is then drawn parallel to the background line. The values at the intersections between the half-value breadth line and the profile are calculated approximately by the simultaneous equations, and the midpoint between two intersections is defined as the peak position. The measurements of stress were performed by the standard method and this computation method on several carbon steel specimens. The measured values of peak position of profiles by these two methods almost agreed each other, and the difference of stresses by these methods was as small as under 1kg/mm2.
The residual stress in a large steel casting (1000×1000×400mm) was measured by means of the photographic X-ray method developed in our laboratory. Since the grain size of the steel casting is large, the obtained diffraction lines on the films were spotty and the spots scattered in the radial direction. The peak positions of the diffraction lines was determined by averaging the distance between each diffracted spot and line of the standard substance arithmetically. The mean value of the residual stress for the specimen annealed at 940°C for 16 hours was 7.6kg/mm2 in tension on the surface of the feeder head side and 10.5kg/mm2 in tension on the surface of the bottom side. The specimen was then cut 15mm in depth from the surface of the feeder head side. The mean residual stress on the newly appeared surface was 3.2kg/mm2 in compression and it was 7.1kg/mm2 in tension on the opposite side. When the specimen was annealed at 650°C for 8 hours, the residual stresses on both surface became almost zero. The distribution of the residual stress in the sample were discussed.
The effects of hydrostatic pressure up to 3000kg/cm2 on the mechanical properties of polyvinyl chloride, polycarbonate, nylon 6 and epoxy resin were investigated. The yield strength increased linealy with increasing pressure for all the polymers investigated. The elastic modulus was little affected by pressure except epoxy resin. With increasing pressure, the elongation decreased for polyvinyl chloride, polycarbonate and nylon 6, while it increased for epoxy resin. The strain rate effect on the tensile properties was also investigated for polycarbonate and nylon 6. A linear relationship was found to exist between the yield strength and logarithmic strain rate both under a high pressure of 2000kg/cm2 and atmospheric pressure. It was considered that there may be a possibility of predicting the pressure dependence of the mechanical properties of polymers from the temperature dependence of those through the pressure dependence of their glass transition temperature at least for polyvinyl chloride, polycarbonate and nylon 6. Further study is needed for epoxy resin.
In the previous report, the various factors that affect the endurance limit of induction hardened steel were examined. It was found that the microstrain and particle size are more influential than the residual stress on the endurance limit of both induction hardened smooth and notched specimens. The present paper describes an experimental study on induction hardened smooth specimens tempered at various temperatures to find out main factors affecting the tension and compression fatigue strength and changes of the integral width and structure during fatigue. The main results are as follows: (1) The compression and tension endurance limit decreases with rising tempering temperature. But the influence of tempering temperature on the compression and tension endurance limit is smaller than on the rotating bending and the plane bending. (2) A linear relation exists between integral width, or micro-strain of particle size obtained from profile analysis and endurance limit. (3) The integral width and microstructure of hardened material such as induction hardened steel change during fatigue, and the fatigue damage exists in a wide area.
This paper is concerned with the fatigue fracture of the clad plate between mild steel and high strength steel, which is considered to be one of the simple examples of the laminated inhomogeneous metals such as the induction hardening specimens, the carburized hardening specimens, the preliminary working specimens and other surface treated specimens. The results obtained are as follows: Although the static tensile strength of the clad plate can be expressed by the law of mixture, its fatigue strength can not. It is expected that the proper composite ratio exists in order to obtain the maximum fatigue limit of the clad plate which has high strength steel layers on the surfaces of the mild steel.
Studies on fatigue strength under varying stress conditions are very important in the fields of fatigue research and machine design. So, in the past many researchers have conducted a great deal of fatigue tests under varying stress. The majority of these studies under varying stress have been done from the viewpoint of fatigue crack initiation and failure, and there were only few studies on the fatigue slip band initiation. In this paper the effect of cumulative cycle ratio on fatigue slip band initiation was discussed. Fatigue tests were done on the specimens made from S10C and S35C (JIS standard). The experimental results are summarized as follows: (1) Σ(n/N)≥1 for low-high stress sequence. (2) Σ(n/N)≤1 for high-low and two level multiple repeated stress sequences.
The transition from flat to shear mode fracture occurs during the crack propagation in polymeric materials under repeated tensile loading. The present study has been made under the condition fulfilling Irwin's criterion for fracture mode transition, to clarify how the transition behavior of an edge-slotted polycarbonate plate is affected by temperature, strain rate and stress amplitude. It has been found that the ratio of the size of plastic zone at the transition point to the plate thickness increases with increasing the temperature or the ratio of stress amplitude to yield stress, but decreases with increasing the strain rate. This size ratio is roughly equal to one half at the static fracture under a given strain rate, but it is less than one half at the fatigue fracture. However, its value calculated with consideration for the local yield stress at the tip of fatigue crack approaches to one half and becomes in good agreement with the ratio of the measured plastic zone size to the plate thickness.
It is important to determine the distribution of plastic strain at the tip of a crack, in order to clarify the mechanisms of fracture. In this paper, a method is presented for measuring local plastic strain by the X-ray microbeam diffraction technique. The total misorientation β, which is calculated from the tangential breadth of the spotty diffraction image in the stationary photography, is closely correlated to the equivalent plastic strain εp. This correlation is independent of the deformation modes; tension or torsion. The total misorientation can not be measured for a heavily deformed material because the diffraction image becomes a continuous ring. The radial breadth of a diffraction ring is adopted as a measure of the change in diffraction image. Measurement of this quantity is acculately performed by using a specially designed X-ray micro-camera with the film- and specimen-swinging devices. The half-value breadth B of a diffraction intensity profile, representing the radial breadth of a diffraction ring, is also uniquely correlated to the equivalent plastic strain εp independently of the deformation modes. The correlation between εp and β or B obtained above offers us a method for plastic strain measurement in a localized region by the X-ray microbeam diffraction technique. The technique based on the measurement of the half-value breadth may particularly suit for the investigation of the fracture process at the tip of a notch or a crack, because of its ability for measuring the large plastic strain up to about 3.6. The utility of this technique is confirmed by applying it to the measurement of the plastic strain distribution at the neck in a stretched specimen.
In order to clarify the mechanism of fatigue crack propagation, the plastically deformed zones at the crack tips were measured by means of two microbeam X-ray diffraction techniques: The transmission Laue method and the back reflection Debye-Scherrer method. The material used was annealed, extremely pure aluminum of 99.999%, and the specimens were subjected to cyclic tension and compression loads. The distribution of maximum shear stress, τmax, around the crack tip was estimated by Sneddon's formula. It was found that the plastically deformed zones appear in accord with the τmax distribution; that is, the minimum deformed region exists at the front of the crack tip and the maximum one appears in the direction 40°∼50° off from the crack propagating direction. I was also clarified that the total misorientation of the grain near the crack tip inside of the specimen is slightly larger than that on the surface.
Measurements and observations on the closures of both propagating fatigue cracks and socalled non-propagating fatigue cracks initiated from notch roots in three kinds of steel during rotating bending tests were made successively at various stress levels in one stress cycle by using the plastic replica method. The results for the propagating cracks having about the same propagation rate (0.05μ/cycle) show that fatigue cracks can be closed at the crack tips for up to three fourths in the annealed 0.13%C steel, two thirds in the annealed 0.54%C steel and one half in the heat treated 0.54%C steel, of the respective full stress ranges. Therefore, the stress up to these levels does not cause further propagation of the cracks. On the other hand, the result for the non-propagating cracks in the annealed 0.13%C steel shows that the crack tip is kept closed even at the maximum stress level. Therefore, the fatigue damage is hardly accumulated by the repetitions of the stress which has produced the non-propagating cracks. The plastic deformation remaining in the wake of crack extension is considered to be the factor controlling the crack closure mentioned above.