The X-ray elastic constant of sintered alumina with powder purities of 99%, 96% and 92% was determined for two diffraction planes, (2. 1. 10) and (1 4 6), by using Fe-Kα and Cu-Kα radiations. The effects of purity and diffraction plane on the X-ray elastic constant were studied. The results obtained were summarized as follows: (1) The diffraction from the (2. 1. 10) plane yielded higher accuracy in the X-ray stress measurement of sintered alumina than that from the (1 4 6) plane, because of its high intensity, isolation of the diffraction profile, and low magnitude of the stress constant. (2) The X-ray value of the elastic constant, E/(1+ν)(E=Young's modulus, ν=Poisson's ratio), as well as the mechanical value, became smaller with decreasing purity or bulk density. The ratio of the X-ray elastic constant to the mechanical value did not vary with bulk density for the both diffraction planes. (3) A simple method to estimate the X-ray elastic constant of sintered alumina with arbitrary bulk density was proposed on the basis of the Reuss-model analysis of alumina polycrystals.
Rolling contact fatigue tests were performed on the cylindrical test pieces taken out of a fresh rail, changing the conditions of contact stress, slip rate and velocity. The behavior of plastic deformation was analyzed in terms of the texture development under the contact surface of the test pieces. The results indicated that two kinds of textures developed in the surface layer of the test pieces; one at the topmost surface, especially remarkable in the testing condition without slip, and the other in the layer a half milimeter deep, being caused by Hertz's maximum shearing stress. A slight effect of the velocity was detected such that the former layer got thinner with an increasing velocity. Transmission electron microscopic observations revealed that the texture was formed due to the crystal slip between the pearlite lamellae and its development corresponded to the development of a cell structure between them. The results were also discussed as compared with those of full-size used rails.
Stress corrosion cracking tests were conducted by using the compact tension (CT) specimens of 200°C tempered AISI 4340 steel in a 3.5% NaCl solution environment under various electrode potentials. The distribution of the residual stress beneath the fracture surface was measured with the X-ray diffraction technique. The fracture surface and the crack morphology were also observed by scanning electron microscopy. The Effect of electrode potential on the growth kinetics of stress corrosion cracking was discussed on the bases of the results of X-ray observation and scanning microfractography. The results obtained are summarized as follows: (1) The residual stress measured on the fracture surface was tension under cathodic potential. It decreased with increasing potential, and changed to compression under anodic potential. (2) The relation between the plastic zone depth ωy and the stress intensity factor K can be approximated by the following fracture mechanics equation: ωy=α(K/σY)2 where σY is the yield strength. The α value was 0.084 under the free corrosion condition and 0.053 under the cathodic polarization condition.
To investigate the influence of mean stress on the tensile and compressive fatigue strength of the heat-treated high-speed steel, fatigue tests were conducted on the test pieces of 9mm and 6mm in diameter which had been heat-treated at two levels of temperature. The fatigue strength at a finite life was evaluated at three different levels of mean stress by using the stair case method for small sample size (JSME Standard S02-1981 14S-N testing method). Furthermore, to clarify the fracture mechanism, fracture toughness tests were conducted on the test pieces of 1/2 compact tension specimens. The following results were obtained. (1) It was found that the fatigue life for 50% survival was strongly affected by mean stress. The fatigue life obtained was much shorter than the estimated value by the modified Goodman law but close to that by the Soderberg diagram. The effect of mean stress was stronger in the harder material. (2) In the tensile and compressive fatigue tests, “Fish eye” was found on the fracture surface. By assuming that “Fish eye” was an initial crack, the value of fracture toughness KIC was estimated from the radius of Fish eye, which showed 15-25 MPa√m for the specimen 1 (HRC60) and 22-30MPa√m for the specimen 2 (HRC56). (3) From the fracture toughness tests on the test pieces of 1/2 compact tension specimens, it was found that the values of KIC for the specimens 1 and 2 were 20.9-21.2MPa√m and 23.2-26.9MPa√m, respectively. These values were in good agreement with the mean value of fracture toughness KIC estimated by the tensile and compressive fatigue tests.
To investigate the influence of mean stress on the tensile and compressive fatigue strength of the heat-treated high-speed steel, fatigue tests were conducted on the test pieces of 9mm and 6mm in diameter which had been heat-treated at two levels of temperature, and the correlation between the X-ray parameters obtained from the X-ray measurements and the fracture mechanics parameters was examined. An X-ray microbeam stress measurement equipment was used to measure the residual strees distribution and the integral breadth distribution on the fracture surface. The results obtained are summerized as follows. (1) In the region of Fish eye, or the fatigue fracture surface, the residual stress on the surface was found to change from compressive residual stress to tensile residual stress and reached to a maximum at a certain value of the maximum stress intensity factor Kmax. The residual stress on the brittle fracture surface was about constant at Kmax. (2) The integral breadth on the fatigue fracture surface increased with Kmax, while the integral breadth on the brittle fracture surface was about constant irrespective of Kmax. (3) In the region of Kmax below 20MPa√m, the residual stress was compression. The reason of this behavior was discussed in terms of the correlation between the residual stress σr1 controlled by Kmax and the residual stress -σr2 controlled by ΔK. It was found that the effect of -σr2 was greater than σr1 in this region.
This paper summarized a study of X-ray stress measurement of stainless steel SUS410 brazed joints with two types of palladium containing filler metals, BPd-7 (5Pd-95Ag) and BPd-8 (18Pd-82Cu). The brazing was performed in a resistance heated vacuum furnace. The joint clearance of brazed joints ranged from 0.30 to 0.35mm. The brazing surfaces were polished by four types of emery papers, #80, #240, #400, and #1500, to clarify the effect of brazing surface roughness on the residual stress after brazing. The stress measurement of brazed joints was made in the longitudinal and transversal directions of the specimen by using the X-ray stress measurement technique (sin-squarepsi technique). The residual stresses σxr and σyr were compressive at the brazed joints with BPd-7 and BPd-8, and the residual stresses of the brazed joint with BPd-8 were greater than those of the brazed joint with BPd-7. In the case of brazed joint with BPd-8, the rough brazing surface produced greater residual stresses than the smooth brazing surface.
A magnetic sensor acting under high biasing field was developed to measure stress in steel nondestructively, based on the linear dependence of reversible permeability on stress, and it was applied to measure uniaxial tensile stress. It was found that the product of reversible permeability and measured area changed linearly with stress. This result was explained by the magnetostriction theory under high field amplitude. In order to extend the applicability of this method to the measurement of plane stress, the relationship between the measured value of the product and the stress in the transverse direction was examined. Fundamentally, a linear relation was found between them, but some problems, such as hysteresis due to loading and the boundary effect on the test piece, remained to be solved.
In order to investigate the effect of residual stress on the fatigue strength of nodular cast iron in use hardened by high frequency induction, rotating bending fatigue tests were made on the specimens drilled along the central axis to change the surface residual stress of the induction hardened FCD45 and FCD70. Some irregular shape graphite exists in nodular cast iron in use. In this examination, the irregular shape graphite was the initiation point of fatigue fracture. When the applied stress (cyclic stress and residual stress) was under the yield stress of the matrix, the relation between the residual stress and the endurance limit was shown by the straight line of gradient -1.0. When the applied stress was over the yield stress of the matrix, the gradient of the relation line became low. But in this case, the residual compressive stress became smaller by cyclic loading. Accordingly, when the final residual stress was used in the relation between the residual stress and the endurance limit, the gradient of the relation line was also -1.0. This decreased value of the residual stress corresponded about 1:1 to the difference between the yield stress and the applied stress.
Fatigue strength and fatigue crack growth behaviors were studied for two types of dual phase steels with duplex ferrite and martensite structure: One was an ordinary dual phase steel processed to replace pearlite by martensite and another was a fine grained one obtained with the method developed by N. Matsumura and M. Tokizane. Major attention was paid to the fatigue properties of the fine grained dual phase steel. The results of this study revealed the fatigue characteristics as follows: (1) The fatigue strength of the dual phase steel, especially that of the fine grained one was excellent. (2) The crack propagation life to failure from the initiation of small crack with the length of 0.1mm was longer for the fine grained dual phase steel than the ordinary one, indicating that the former material has a higher resistance against the crack growth of short cracks. (3) The crack growth rate of long cracks for both materials was higher than those of the conventional structural carbon and alloy steels in the relatively low stress intensity region.
Fretting fatigue tests of high strength stainless steel 440C were carried out to investigate the basic properties of fretting fatigue and the effects of contact material and variable loading on fretting fatigue strength. The fatigue life and the fatigue limit of 440C steel decreased significantly under the fretting fatigue condition. A soft contact material was effective to improve the fretting fatigue strength. The difference in fatigue life among contact materials will be caused by the difference in the concentration of frictional force in the edge region of contact surface: The softer the contact material, the lower the concentration of frictional force. Therefore, the softer contact material induces the lower acceleration of crack growth rate, and gives the higher fatigue limit and the longer fatigue life. The modified Miner's rule seemed to hold in the case of fretting fatigue under a program loading simulating the actual load of bearing for airplane.
Fretting fatigue tests were carried out using S20C steel to investigate the effects of relative slip amplitude and contact pressure between the specimen and the contact pad on fretting fatigue strength. The fretting fatigue life was reduced with an increase in relative slip amplitude, and it came to a constant value beyond a certain value of relative slip amplitude (12-13μm). With an increase in contact pressure, the fretting fatigue life was reduced. The reason why the fretting fatigue life is reduced with increasing relative slip amplitude and contact pressure may be as follows. An increase in relative slip amplitude or contact pressure produces an increase in frictional force which acts on the contact surface of specimen. Since the growth rate of fretting fatigue crack is influenced strongly by the frictional force, the increase in frictional force produces the acceleration of the crack growth rate and consequently gives a lower fatigue life. Since the frictional force comes to a constant value beyond the certain relative slip amplitude, a constant fatigue life is observed in this region of relative slip amplitude. This speculation was confirmed by the prediction of fretting fatigue life based on the elastic-plastic fracture mechanics, where the frictional force was taken into account.
The growth behavior of small fatigue cracks was investigated in a low carbon steel under axial loading at the stress ratios R of -1 (tension-compression) and 0 (pulsating-tension). Crack closure was measured to evaluate the effects of stress ratio and stress level on the growth behavior of small cracks. Except for the accelerated growth behavior at the higher stress levels close to the yield stress σys of the material, i.e. at σmax=0.84σys for R=-1 and σmax=1.12σys for R=0, at R=-1 small cracks grew faster than large cracks below a certain crack length, but at R=0 the crack growth rate for small cracks was coincident with that for large cracks in the whole region of crack length investigated. The critical crack length 2cc above which the growth behavior of small cracks was similar to that of large cracks depended on stress ratio, being 1-2mm at R=-1 and less than 0.7mm at R=0. The 2cc value at R=-1 agreed with that obtained under rotating bending (R=-1). The crack growth rate for small cracks was well correlated with that for large cracks by the effective stress intensity range ΔKeff; thus ΔKeff was found to be a controlling parameter for small crack growth including the growth behavior at the higher stress levels.
This paper deals with the effect of reinforcement shape factor on the stress intensity factor (KI) at the tip of a nonpropagating crack at the toe by using a Boundary element method. As micro-cracks exist in the welded joint, the fatigue strength of welded joint is determined by the propagating condition of nonpropagating cracks. The following conclusions were derived from the numerical results of KI and the investigation of the shape of reinforcement made by shielded metal arc welding and CO2 gas welding; (1) The micro-crack does not affect KI; (2) The toe radius considerably influences KI; (3) The reinforcement angle affects KI; (4) The height and width of reinforcement do not affect KI.
The influences of bending and torsional loading modes on cyclic SCC crack initiation behavior were investigated in a high-strength steel, SCM435, sensitive to hydrogen embrittlement type SCC. Under bending loads, an intergranular cyclic SCC crack by tensile stress is initiated at the bottom of a corrosion pit, and the strength of cyclic SCC is smaller than that of fatigue in laboratory air. Under torsional loads in laboratory air and in a 3.5% NaCl solution at higher stress levels of τmax≥780MPa, specimens are fractured by transgranular cracking normal to the longitudinal direction, with no influence of environment on strength. At τmax<780MPa, however, cyclic SCC strength is smaller than fatigue one. At 500MPa≤τmax<780MPa, a logitudinal crack by shear stress, which is considered to be caused by anodic dissolution, is initiated at the bottom of a corrosion pit, and then an intergranular cyclic SCC crack by hydrogen embrittlement at an angle of 45 degrees against the longitudinal direction (45-degree crack), which is dominated by a principal stress, is initiated at the tip of the longitudinal crack. At τmax<500MPa, however, an intergranular 45-degree crack by hydrogen embrittlement is immediately initiated at the bottom of a corrosion pit. In the case of bending and torsional loads at τmax<500MPa, the mechanical condition for crack initiation is determined by the KIFSCC value, which is obtained by assuming corrosion pits as sharp cracks. In a similar way, a longitudinal crack is initiated when a KIII value exceeds the KIIIFSCC; a 45-degree crack is initiated when a Mode I stress intensity factor of a longitudinal crack exceeds the KIFSCC value.
The constant tensile speed tests were carried out in liquid zinc and in air on the precracked specimens of alloyed tool steel (JIS: SKD6) quenched and tempered at 923K, in order to investigate the crack propagation and crack branching behaviors in a liquid zinc embrittlement. The unstable fracture occurred at smaller load in liquid zinc than in air in all the testing temperatures. Grain boundary fracture occurred in liquid zinc, and fine zinc particles were observed on the fracture surface, which suggested that the adsorption of zinc vapor to a crack tip was the reason for such an unstable crack propagation. The crack propagation velocity in liquid zinc took a minimum at a certain temperature. The crack branching was often observed during the crack propagation in liquid zinc. Well-developed crack branching occurred when the temperature was low and the crack propagation velocity was high.
Rail-shear tests were carried out on the glass cloth/epoxy laminates, and then the mode II fracture toughness value, KIIin, was obtained as the pre-crack length, 2a/W, was altered from 0.2 to 0.45. In addition, the spread of damage was obtained under simulation by the finite element method and then compared with the experimental results. The following results were obtained; (1) In this test method, the recommendable pre-crack length existed in order to obtain the fracture toughness value. (2) The material used in this study was assumed to be a homogeneous anisotropic material of a tetragonal system, and then the finite width correction factor of the specimen geometry was calculated. Consequently, KIIin became almost constant in the range of 2a/W=0.25-0.45. (3) The spread of damage was simulated by the finite element method, and it was found that the analytical result agreed with the initial failure process obtained in the experiment. In addition, the JII value was calculated from the path integral, and discussed.