Press-fitting is widely used for railroad axles, crank shafts, etc.. A lot of investigations on fatigue strength of press-fitted specimens have been so far carried out. However, many of them were done under constant stress amplitude conditions and only few investigations under varying stress amplitude have been reported. In the present study, rotating bending stress tests were carried out on press-fitted specimens under varying stress amplitude conditions which were composed of two-level multiple-step stresses, ; one stress amplitude was higher than fatigue limit (the over stress) and the other was lower than that (the under stress). The specimens were made of medium carbon steel (S35C), and the diameters of the specimens were 10mm and 50mm. The 10mm diameter specimens were in the as received condition and the 50mm diameter specimens were in four heat treatment conditions; as received, tufftrided, induction hardened, and tufftrided after induction hardened. The results obtained are summarized as follows; (1) Under the varying stress amplitude, the under stress gave a fatigue damage to the specimens and the cumulative damage calculated by Miner's rule was very much less than unity. (2) Haibach's modification of S-N diagram was more suitable than the modified Miner's rule for calculating the damage of the under stress in the range of these experiments.
The present investigation is concerned with the influence of solute element atoms on the recrystallization temperature of cold-rolled silver. The materials containing 0.03∼1.8 at% each of Be, Mg, Pd, Cu, Au, Cd, Al, In, Ge, Sn, Pb, Sb, or Bi were cold-rolled to 90% reduction of area and annealed for 10 minutes at various temperatures. The recrystallization temperature was obtained by hardness measurement. At any rate, all solute elements raised the recrystallization temperature of silver with increasing their content. The most pronounced influence on the recrystallization temperature was found to be brought in by the addition of the elements which have a much larger absolute linear size factor, excess valence and smaller maximum solid solubility. The magnitude of the increase in recrystallization temperature seems to be related to the binding energy of solute atoms with vacancy, dislocation and grain boundary.
It is well known that residual deformation is retained in hot dip zinc galvanized steel structural members. This deformation is mainly caused by thermal deformation which occurs when the members are dipped into molten zinc. Therefore, in order to consider the countermeasure for the occurence of such residual deformation, it is very important to investigate the thermal deformation and thermal stress caused by dipping. However there are very few reports on these kinds of transient deformation and stress. In this study, the cross-sectional deformation and the thermal strain of a steel pipe (φ660.4× 18, L=3500mm) were measured on the specimens dipped into 460°C molten zinc from 20°C by using displacement transducers and capsule type high-temperature strain gages, respectively. They were also analyzed by using the finite element method (FEM) and the results were compared with those of the experiments. The main results obtained are as follows: (1) The measured variations of deformation and thermal stress for a steel pipe dipped into molten zinc coincided well the results analyzed by using FEM in sequence. (2) The total deformation and the thermal stress of the steel pipe varied with the dipping speed: the thermal deformation as well as the thermal stress decreased with increasing dipping speed.
The yield behavior of polycarbonate and polyvinyl chloride subjected to bending was investigated by uniform bending tests using beams with several kinds of cross section. Particular attention was given to the effect of the macroscopic stress distribution in the specimen on the yield behavior. It has been shown that by observation in tension and compression tests, the yield behavior of rigid plastics can be deviled into two distinct processes; the initiation process of yield accompanying load drop and the progressive process of yield under constant load. This suggests that there are two kinds of yield stresses. Thus, in order to get a general interpretation of yield of plastics, the two kinds of yield stresses should be taken into consideration. The yield behavior in bending has been analysed in terms of two kinds of yield stresses for rigid plastics, and the relation between the bending moment and strain on the surface of beam has been determined quantitatively. The calculated results were confirmed by the experiments.
The fatigue crack growth rate was investigated on the basis of a dislocation multiplication model for damage accumulation on which the fatigue life up to crack initiation was evaluated. A two-dimensional version of a stochastic model proposed by Tsurui and one of the authors (A. I.) was used in order to obtain the expression of the rate of fatigue crack growth, and a distribution function of the strength of material elements was derived by the use of the dislocation multiplication model. The results show that the fatigue crack growth rate is proportional to some power of the range of the stress intensity factor. The theoretical results were compared with the experimental data.
In this paper, a statistical method to evaluate the fatigue data has been proposed by considering a two dimensional distribution on the life-strength plane (S-N plane). This evaluation method is based on the assumption that the data of S-N curve can be treated as the members of the set, which has a two dimensional normal distribution around the peak point in the coordinates of life and strength axis, and the variance of this two dimensional normal distribution can be given from the life and strength distributions of the original S-N data. A contour line of this two dimensional distribution is of elliptical shape, and represents the probability range. Furthermore, the contour line having the same probability of S-N curve can be obtained from the envelope curve of the contour lines having the same probability of subgroups which have also the two dimensional normal distribution. The region enclosed with this curve expresses the probability of fracture occurrence. The fatigue limit is defined as the tangent line of this contour line, which is paralled to the life axis. As an example, the S-N data obtained by Nishijima et al. were analyzed. It was confirmed that the contour lines of the same probability obtained by the above mentioned method were in good agreement to that of the sample S-N data.
The influence of corrosive environments on the fatigue crack growth rate of SNCM8 was investigated. The material used was tempered at 200°C to give a yield strength level of 150kgf/mm2·Tests were made at stress intensity levels below KISCC in room air, distilled water and 3.5% NaCl solution. Since the formula of stress intensity was unknown, the data were analyzed by using the stress intensity of an infinite cracked plate with uniform stress at infinity. Moreover, in order to consider the effect of hydrogen on this material, hydrogen-precharged specimens were tested in the same environments. The results obtained are as follows: (1) The fatigue crack growth rate in corrosive environments was 3-5 times faster than that in room air. However, it is considered that the growth rate in room air is accelerated by water vapor in air. The growth rate at lower stress intensity in 3.5% NaCl was a little faster than that in distilled water. (2) Independent of the stress level during precracking and corrosive fatigue stressing, the growth rate was represented well by the formula above-mentioned. (3) Precharged hydrogen accelerated the fatigue crack growth. This effect was remarkable for precracked specimens and in room air. (4) Fracture surfaces have both striation-like pattern and intergranular facet pattern in the above environments. In corrosive environments intergranular facet pattern increased. These results suggest the occurrence of hydrogen embrittlement.
A C-shaped specimen is suitable for testing thick-walled cylinders. Fatigue crack growth tests were conducted by using the through cracked specimens with different ratios of outside to inside radius. It was shown that the fatigue crack propagation rate, da/dN, was related uniquely to the stress intensity factor range, ΔK, by following expression, da/dN=C(ΔK)m This relationship is consisted with that obtained by using the compact specimens fabricated from the same orientation. From the fatigue crack growth tests using the surface cracked specimen, it was shown that the crack depth could be measured accurately by the End-on ultrasonic method, and the aspect ratio variation chracteristics with increasing crack depth was clarified. Moreover, it was found that the fatigue crack propagation rates, da/dN and db/dN, in the specimen width and through thickness directions were related to the presumed stress intensity factor ranges, ΔKa and ΔKb, respectively.
In this study, the proposed elastic-plastic fracture toughness test procedure based on Top-on and End-on ultrasonic methods was applied to JIc tests for the C-shaped specimen, and it was pointed out that the load and load point displacement at the onset of stable crack growth during fracture toughness loading can be detected from a single specimen by the End-on ultrasonic method. JIc, measurements obtained for the C-shaped specimen were in excellent agreement with that obtained for the compact specimen. The JIc value evaluated according to the proposed JIc test procedure was also in agreement with that determined according to the SZ method, but was not consistent with that determined according to the R-curve method.
The glass-forming region in the B2O3-Na2O system was tried to interpret on the basis of an assumption concerning the glass-forming ability of a substance that higher viscosity melts will make a glass when quenched at an appropriately rapid rate, while lower viscosity melts will not. Ten grams of B2O3-Na2O melts were water-quenched at a rate of 7°C/sec, and two glass-forming regions were determined. The upper glass region was 100∼61 mole per cent B2O3 and the lower glass region was 32∼28 mole per cent B2O3. Viscosity was measured by a counter-balanced viscometer at elevated temperatures above the liquidus temperature, and the liquidus viscosity log ηl was obtained for the compositions of 100∼25 mole per cent B2O3. The liquidus viscosity curve plotted as a function of B2O3 content for the B2O3-Na2O system exhibited two maxima and two minima in viscosity. It was found that the glass regions, 100∼61 and 32∼28 mole per cent B2O3, corresponded quite closely with the regions where the liquidus viscosity was higher than log ηl (poise)=0∼0.1. The single-bond strength BM-O and the Rawson's parameter BM-O/Tl for glass formation were calculated and were compared with the glass-forming regions of this system. These parameters were found unsatisfactory to explain the glass-forming ability of the lower glass region. But the liquidus viscosity described rather quantitatively the existence of the separate glass regions in the B2O3-Na2O system.
In the present paper, an experimental method of testing multicycle fatigue of ceramics was investigated. When the inner surface of a ring is heated by a constant heat flux, the resulting thermal stress can be easily estimated. By using this merit, the relation between the thermal stress and the fracture time of some ceramics was measured. A nearly linear relation was obtained when the data were plotted on a log-log graph paper. The technical meaning of the slope of the line was also considered.