The X-ray microbeam diffraction technique is a powerful method for observation of substructures in metallic materials, because of its ability to measure the microstructural changes in a localized region quantitatively and non-destructively. In recent years, the microstructure in the vicinity of fatigue crack tip has been investigated by using this technique and the mechanism of crack propagation has been discussed based on the results obtained. However, the correlation of the X-ray microbeam diffraction patterns and the microstructures observed by the other methods has not been made sufficiently clear. In the present study, the X-ray microbeam diffraction patterns from deformed specimens were compared with the substructures photographed by means of transmission electron microscopy. The material used was 0.12% carbon steel, and the low cycle fatigue test, the high cycle fatigue test and the impact test were performed to form substructures in the specimens. A good correlation was obtained qualitatively between the X-ray microbeam patterns and substructures in both the low cycle fatigued and impacted specimens. It was possible to distinguish the mode of substructure through X-ray microbeam diffraction patterns. A disagreement existed, however, between the values of subgrain size obtained by different analytical methods. It is considered that getting agreement between these values is difficult in the present stage, because of the assumptions on analyses. But the subgrain size should be a good parameter representing the plastic deformation, if an analytical method suitable for the mode of diffraction pattern is adopted.
In order to evaluate the fatigue strength of blades of an axial flow compressor, we have done the strain measurements in service condition, the fatigue tests on the materials of blade under several conditions (under varying stress amplitude and superimposed stress, in corrosion, etc.) and the fatigue life prediction. In this paper, the results of strain measurement of 1st and 14th stages of rotor blades and prediction of life are explained. Also the problems in evaluating the fatigue strength of the blade are treated. The main results are as follows; (1) Stress-frequency curves depend very much on the cycle counting method and the rate of resonance of blade. (2) Fatigue life is determined by low level stress having many number of cycles rather than the maximum stress level. (3) The S-N curve estimated can be used to evaluate the service life of blades.
The exploitation of geothermal energy as one of new energy sources has been highlighted, and now it is mainly utilized for generating electric power. However, it has become a problem that the fatigue strength of stainless steels used for the turbine blades and other parts decreases due to corrosion by H2S contained in geothermal steam. In order to solve this problem, the corrosion fatigue test on 13 Cr-Mo, 12 Cr-Mo-V, 3 Cr-Mo, Cr-Mo-V and 3.7 Ni-Cr-Mo steels was carried out in geothermal steam at the Onikobe Geothermal Power Station. The main results obtained are summarized as follows: (1) The fatigue strength of steels in geothermal steam was much lower than that in air. The rate of decrease in fatigue strength was smallest for 13 Cr-Mo steel, and was largest for 3.7 Ni-Cr-Mo steel among five steels. (2) 13 Cr-Mo steel was corroded very little even if being exposed in geothermal steam for a very long period. (3) A method of estimating fatigue strength at 109∼1012 number of cycles in geothermal steam from the corrosion fatigue strength of the specimen and the fatigue strength of the corroded specimen was established by separating the cycle-dependent effect of fatigue damage from the time dependent effect of corrosion damage. For this purpose the corrosion coefficient Kc=1+B·t0n1·Nfn2, where t0 and Nf are corrosion time and number of cycles was defined. For 13Cr-Mo steel, B=4.33×10-3, n1=0.11, n2=0.22 were obtained.
There are very few reports about X-ray stress measurement of austenitic stainless steel and consequently, there exist many problems which must be clarified. In the present study, the X-ray stress measurement of seven kinds of specimens (grain size number 7.2∼7.0) obtained by various treatments from cold-rolled 18 Cr-8 Ni austenitic stainless steel plate were carried out by following heat the standard method used for ferritic steel, and their X-ray elastic constants, constant K and accuracy of measurement were determined by using different diffraction planes with Cr Kα, Cr Kβ the radiations and two methods of X-ray incidence. The results are summarized as follows: (1) The error in stress measurement of ten incident angles by the fixed ψ method with γ(311) diffraction of Cr Kβ radiation is from ±2.17 to ±9.91kg/mm2, or the average of ±5.85kg/mm2, if the limit of confidence is set at 95per cent, and the X-ray elastic constants and constant K are, (1+ν/E)X-ray=6.83±1.30×10-5mm2/kg (ν/E)X-ray=1.60±0.54×10-5mm2/kg KX-ray=-36.14±7.03kg/mm2/deg. respectively. The mean error by the usual measurement of four incident angles is ±11.60kg/mm2. (2) The mean errors of measurement by the fixed ψ0 method with Cr Kβ γ(311) diffraction and the fixed ψ method with Cr Kα γ(220) diffraction are ±9.33kg/mm2 and ±24.11kg/mm2 respectively. The accuracy of the measurement by the latter is insufficient for practical usage. (3) In the measurement using γ(311) diffraction of Cr Kβ radiation, it is very necessary that the peak intensity of diffracted X-ray is increased in order to minimize the error, and that the recorder is equipped with differential curcuit for bringing down the back-ground because its intensity tends to become high. (4) It seems that chromium carbides precipitated in grain boundaries or a trace of martensite in austenite do not cause the error in stress measurement to increase.
Deep drawing is one of the most important plastic working processes and it is being used to manufacture many products. However, the deep drawing has very complex and severe modes of deformation, such as ironing, bending and unbending. In the case of austenite stainless steel, on which deep drawing is used most extensively, there is a well known property of stress induced transformation that makes the stress state more complex. The residual stress analysis is needed from the view point of preventing cracks caused by residual stress. So, the measurement of residual stress in austenite matrix phase and in transformed alpha' phase has been carried out by using X-ray method. After the X-ray examination, the deep drawened cup was cracked by cathodic charge of hydrogen without any mechanical disturbance to obtain the relation between residual stress and cracking.
Thermal stress behavior of W/Cu composites was studied by the X-ray stress measurement method. With the X-ray technique, the thermal stress of the composites due to the thermal expansion coefficient mismatch of component materials has been measured in the form of residual stress at room temperature. This paper reports the thermal stress behavior during heating and cooling of the composites, which was measured by using a new apparatus consisting of an X-ray diffractometer and a vacuum furnace. Generally, the thermal stress behavior of composites is given by the following equation. σm=KΔαΔT where σm is stress in matrix, K is a constant derived from elastic constants and volume fraction of constituents, Δα is a difference in thermal expansion between the fiber and the matrix and ΔT is a temperature difference. That is, σm is a linear function of temperature. During this study, it has been found that the thermal stress behavior of W/Cu composites during heating and cooling does not show a linear relation with respect to temperature, but a hysteresis loop. The loop can be related to the matrix stress-strain behavior at high temperatures.
Pure aluminum deformed in tension was studied by the transmission Laue method. The asterism of diffraction spots was expressed in terms of its magnitude and diffracting direction as follows: Let the magnitude Δθa be the dispersion of diffration angle, and φ the angle between the load the diffracting directions on a film. The locus of vector (Δθa, φ) can be approximated by one or two circles in many cases. These circles represent the directional asterism distributions and the longest vector corresponds to the asterism maximum. The experimental results obtained were as follows: for coarse grained polycrystalline aluminum, the maximum value of asterism is approximately proportional to the elongation up to a several percent. For both bi-crystal and polycrystal specimens, the asterism maximum vector in a grain oriented in the easy slip direction is large and has the direction along the load direction at the mid point of grain, but it becomes small near the grain boundary. On the other hand, the asterism maximum vector in a grain oriented in the hard slip direction is small at the mid-point of grain, but it becomes large near the grain boundary, with its direction varying differently.
The analysis of X-ray diffraction profile can provide us with much useful information on crystalline structure, and so it has been used for the studies on metal behavior during plastic deformation, fatigue process, recovery process and so on. In the first part of this paper the amount of deformation, ∈, and the flow stress, σ, given to the specimens of plain carbon steels (0.05% to 0.50% carbon) are quantitatively related to the integral breadth, B, by using the yield stress (σY), the uniform elongation (∈U), the integral breadth corresponding to the uniform elongation (BU), and the integral breadth of each annealed specimen (B0). The X-ray diffraction technique is believed useful to study the anisotropic deformation behaviors of pre-deformed polycrystalline metals, since some microscopic parameters involved in such phenomena can be detected by means of this method. The relationship between the anisotropic deformation properties induced in a pre-stretched steel sheet and the directional dependence of the breadth of X-ray diffraction lines is described in the latter part of this paper. The experimental results show that the diffraction profiles obtained with respect to the macroscopically active slip planes are relatively narrower than those with respect to the other planes. This phenomenon is interpreted due to the so-called latent hardening on non-active slip planes, since the directional dependence of their breadth is considered to represent the anisotropy of crystalline structures.
Single-edge-notched specimens of 3% Si-Fe polycrystals were stretched under pin-point loading. The feature of plastic defomation near the notch tip was observed by the etch pit method, and the distributions of strains in the loading direction (y-direction) and in the thickness direction (z-direction) were measured by a talysurf and a microscope. The characteristics of strain distribution in the near-tip field were clarified and the fracture criterion was discussed based on the observed strain distributions. Main results are summarized as follows: (1) The plastic deformation near the notch tip was confirmed to concentrate on two sets of mutually perpendicular planes inclining 45°to the loading axis and the specimen surface. This mode of deformation results in the necking of the surface and the corresponding crack tip opening displacement. (2) The slopes of strain distribution curves in a log-log plot of the z-directional strain vs. distance ahead of the crack tip were -0.13 and -1.0 within and beyond the region of half thickness distance from the tip, respectively. The intensity of strain distribution was found to be proportional to the COD value. For the y-directional strain, the slopes were -0.71 and the strain intensity was proportional to the J integral. (3) The empirical relation between the J integral and the gross displacement has the same tendency with the prediction of the linear elastic solution in small scale yielding and with that of the rigid plastic solution after general yielding. (4) The critical value of the z-directional strain at the notch tip at the time of unstable fracture was found to be independent of the notch tip radius, but decreased with increasing crack length In fractographic observation of fracture surfaces, the larger extension of stable crack growth was detected for the smaller initial notch length.
Mechanism of fatigue fracture has been widely studied by many investigators. It has not been clarified, however, whether surface features or sub-structures are essential to fatigue crack initiation. In this investigation, as the first step to challenge this problem, the crystal orientation dependence of fatigue damage is studied on 99.99 % pure aluminium single crystals by X-ray micro-beam technique. The results obtained are as follows: (1) Fatigue damage is accelerated in the crystal orientation which has equally large Schmid factor both on Primary and Critical slip planes, but it is restrained as Schmid factor on Conjugate slip plane increases. (2) The formation of extrusion and intrusion is restrained near  direction which has large Schmid factor of cross slip system. It is suggested that the persistent slip bands are coplanar. (3) The configuration of fatigue damage depends on the crystal orientation. (4) Secondary slip system plays an important part in the formation of fatigue damage. It is remarkable in the Stage II of crack propagation.
To clarify the fatigue crack propagation process, thin foils of annealed and 90% rolled, aluminum and aluminum-7% magnesium alloy are fatigued and their dislocation structures near the fatigue cracks are observed by a transmission electron microscope at 2000Kv. Within about 1∼2μ from the fatigue crack, fine subgrains less than 1μφ are observed. The size of these subgrains in the annealed specimen is similar to that formed in the rolled specimen of the same material, but the average size of these subgrains formed in aluminum specimen is larger than that in aluminum-7% magnesium alloy specimen. In the region more than 2μ away from the fatigue crack, the different structures of subgrains are obviously observed in the annealed and in the rolled specimens, and this region might be responsible for the difference in the mode of fatigue crack propagation of these specimens.
The fatigue test has been carried out on coarse grained aluminum at the conditions of ambient pressure and 6×10-5 Torr, in order to investigate the effect of atmospheric-corrosion on the fatigue strength of aluminum. Furthermore, to understand the difference of fracture modes at these two environmental conditions, the external and fracture surfaces were observed by means of optical and scanning electron microscopies. The results were discussed from the view point of grain size dependence. The main results obtained were as follows: (1) The fatigue strength of aluminum in vacuum was higher than that in air. The fatigue resistance of a specimen decreased with increasing the grain size. However, the effect of grain size was not so much when the applied stress level was low. (2) The fatigue test on single crystal showed that the effect of atmospheric-corrosion on the fatigue process is different for each crystal. The fatigue behavior of a single crystal in vacuum is considered to be purely dependent upon the crystallographic matters. (3) Many cracks of trans-granular type are observed in the specimens tested in vacuum, while those of inter-granular type exist in the specimens tested in air. Also, there are striations on the fracture surface of the specimens tested in air, but not in vacuum.
It has been pointed out that the fatigue strength of steel in air is improved by tufftriding treatment. The present authors have clarified in the previous study that such improvement results from (1) the appearance of compressive residual stress, (2) the rise of hardness at the surface, (3) the strengthening of matrix by supersaturated nitrogen and (4) the increase of hardness on the fatigue process. On the other hand, a serious view is recently being taken on the corrosion fatigue fractures of machine and construction, as high strength steel is developed and utilized. However, the corrosion fatigue behavior of tufftrided steel is hardly clear, since it is influenced by many factors. Therefore, the authors carried out the static dipping corrosion test and the corrosion fatigue test of tufftrided steel. The specimen used in this investigation was a 0.15% low carbon steel, and pure water and 3% NaCl solution were used as corrosive environments. As the results, the following conclusions have been obtained: (1) The corrosion resistance of tufftrided steel is good in pure water and it results from protection to corrosion by a compound layer formed on the surface. Even when cyclic stress is applied on it, the bonding between the compound layer and the diffusion layer is very strong. So, the diffusion layer is protected from corrosion in its fatigue process until the breakdown of compound layer takes place. Therefore, the diffusion layer contributes to the fatigue in pure water as well as that in air, and the fatigue strength of steel is improved by tufftriding treatment in pure water. (2) On the other hand, the corrosion of tufftrided steel is severe in 3% NaCl solution because of formation of etch pit, which originates from the localized damage of compound layer. Such formation of etch pit is accelerated by interaction of corrosion and stress cycles. A crack initiates at the root of etch pit due to cyclic stress, and the crack propagates to become a fracture when the solution infiltrates into the tip of the crack. Therefore, the fatigue strength of tufftridedsteel decreases radically in 3% NaCl solution with the increase of stress cycles.
An X-ray study was conducted to investigate (I) the size effect of notched specimens on fatigue, (II) the surface area effect on fatigue and (III) a probability problem. The cyclically induced micro-yielding of the specimen surface, detected by X-ray, is the controlling factor of fatigue damage in the above three fatigue problems. This micro-yielding value is found to be interpretable in the same way as the plastic strain amplitude of low cycle fatigue in terms of the number of cycle to failure. Hence, X-ray observation was confirmed to be a very useful method in detecting fatigue damage in machine parts, and in picking out the parts that might suffer damage under service load.