The applicability of Hill's anisotropic yield condition was studied experimentally for the anisotropic state given by uniaxial prestrain. The change of Hill's anisotropic constants with the increase of prestrain was investigated by the test of the small tensile specimens cut from the initially prestrained brass plate in various orientations to the initial prestraining direction. The subsequent yield surfaces after uniaxial prestraining and the necking angles of the small specimens were calculated by Hill's anisotropic constants and compared with the experimental results. Hill's original expression of the anisotropic yield condition became a simpler form for the plate with the anisotropy given by the uniaxial prestrain. The calculated yield surfaces by this formula did not coincide well with the experimental values when the value of the subsequent yield stress in the initial prestraining direction was determined as that of the yield point in the reloading stress-strain diagram. But the coincidence between them became a little better when the applied stress value for the prestrain was adopted for the subsequent yield stress. Hill's formula of the necking angle was found to be impertinent in instances where the amount of the prestrain was smaller than that of the strain at fracture, unless the change of the anisotropy due to the deformation was taken into consideration.
We performed the fatigue tests of single spot-welded steel SS41P under completely reversed direct stress at room temperature. Two modes of failures, that is, the through-the-thickness failure and shear failure at the interface of sheets were observed through these tests. The former occupied 84% of total failures and had higher frequency of occurrence at low stress level than at high stress level. We postulate that through-the-thickness failure will occur when a maximum principal stress reaches the limit value which depends on a number of stress cycles. In order to estimate this stress, the following stress components were taken into consideration, the shearing stress at the interface of sheets and normal stress, which are in the direction of loading, and the normal stress in the thickness direction. In addition to the original residual stress, the stress in the thickness direction includes the stresses which are developed by bending due to the attachment of the specimen on testing machine and the eccentric loading of external force during the fatigue test. In a S-N diagram, in which the nominal stress is adopted as ordinate, a considerable scattering of plotted data is shown and it is difficult to recognize the effects of such factors as diameter of electrode and spot-welding conditions. In a modified S-N diagram, however, in which the maximum principal stress is taken as ordinate, its scattering of data is much reduced. In this diagram the S-N curve appears to depend primarily on a diameter of faying area rather than on spot-welding conditions. In comparison with the S-N curve of the mother material, we obtained the strength reduction factor β=2.73 for the number of stress cycles until the through-the-thickness failure Nf=5×106 is reached, and we consider that the main cause of the low fatigue strength in the spot-welded joint may be the high stress concentration at the origin of the crack.
In view of the fact that the residual stress, affecting the fatigue phenomena of materials, varies with repeated strain, it is not considered to be altogether proper to evaluate fatigue strength solely on the value of the residual stress in the virgin materials, but it is necessary to study the process of changes in the residual stress under cyclic loading, which is also an important process in making nondestructive prediction of fatigue life. In this papers reported the experimental study made on this problem about cast iron, particularly malleable cast iron. The following conclusions have been obtained: (1) If there is residual stress in the virgin materials, the changes in the stress are in three stages. (2) If there is no residual stress in the virgin materials, the changes in the stress are in two stages. (3) These changes correspond to the fatigue behaviour of the materials. (4) These relationships will be utilized in making nondestructive prediction of fatigue life of cast iron.
The double-repeated rotating bending fatigue tests of 18-8 austenitic steel were carried out at room temperature, 400°C, 500°C and 600°C. The main conclusions arrived at are as follows. (1) Remarkable surface temperature rise is measured during the fatigue tests of solution-treated 18-8 austenitic steel. Its behavior has the same tendency as that of the deflection of the specimens during the fatigue tests. (2) Generally, cumulative cycle ratio Σ(n/N) is larger than one when the secondary stress level σ2 is higher than the primary stress level σ1. However, the effect of stress sequence on Σ(n/N) is not observed at 600°C, nor is it observed at any temperature when the cycle ratio of the primary stress level n1/N1 is 0.25. (3) The relation between Σ(n/N) and the surface temperature and deflection during the double-repeated fatigue tests are as follows. When σ1<σ2, the surface temperature and deflection under the secondary stress level σ2 are smaller than those under the constant stress σ2 of the same level. When σ1>σ2, the former is larger than the latter. When the stress sequence have not an effect on Σ(n/N), the former is much the same as the latter.
In the previous paper7), the author proposed a new measuring method for tracing fatigue behavior of plastics. Here in the present experiment this method has been put into practice to study the fatigue behavior of the seven plastic materials, in addition to the three in the previous experiment, such as rise in temperature of the specimens and changes of their dynamic deflection. As the result the fatigue behavior can be classified into several types. Besides, an attempt has been made to relate the types of fatigue behavior to fatigue processes, with the aid of other experiments. It is difficult to classify exactly the materials by types of fatigue processes. However it is observed that in the cases of epoxide, polycarbonate, polystylene and polyester, the rise of temperature is not remarkable and there occur some changes in quality of the materials (e.g. elastic modulus changes). In the cases of polyamide, polytetrafluoroethylene and polymethyl methacrylate, the rise of temperature is remarkable and the materials become softer. In the cases of polycarbonate and polyvinyl chloride, the specimens have been observed as changing their shape by slow development of crack.
The preparation of graphite has been worked by pyrolysis of propane gas on a directly heated substrate in the gas pressure range of 5∼200mmHg at the deposition temperature of 2025°C and gas flow rate of 640cm3/min. The formation process and structural features (microstructure, density, and crystal structure) of the pyrolytic graphite were examined on several samples deposited under various gas pressure conditions. The effect of the gas pressure on them is appreciable. At the lower pressures of 5∼10mmHg and the higher pressures above 40mmHg, the formation rate and the structural features vary continuously with the increasing gas pressure. In the pressure range of 20∼30 mmHg, however, the rate of formation drops, and sootiness features its structure. These results with the pressure change are discussed in relation to the convection of the vapor phase.
Measurement was made of the electron emitted from the oxide-coated cathodes in the directly heated rectifier during both the normal and the accelerated life test. The accelerated life test was made by applying cathode voltage 10% higher than the normal rating. The emitted current was measured by two methods, the measurement of the component dc of the current put out Io and the measurement of the dc saturated in the emission Is. It seems that the most effective and shortest life test in the evaluation of the cathodes in rectifier on the quality level is the measurement of Is at accelerated life test. It was found in the above mentioned life test that increase of oxide-coating was effective to prevent decay in the emission of electron. Addition of nickel powder into oxide coating was effective to prevent rectifier from sparking between anode and cathode and gave no harmful influence on electron emission from cathode.
The model experiments of direct extrusion through the wedge-shaped dies have been executed. The material of the specimen is acetyloid, and four kinds of dies have been adopted, whose dimensions are: die half angle 15°, 10°and 0° and reduction ratio 1/4 and 1/2. The square grid lattice (0.5mm interval) are traced on the specimen and the photo-plastic isochromatic patterns and grid paterns are taken on the films. The fringe-order of the point on the central line of the vicinity of the die entrance always remains in zero fringe-order and the point is suppressed by the hydrostatic pressure. The distribution along the central line of the fringe-order and of the difference of the principal strains of the die (15°, 1/2, symmetry) has been measured. The latter vanishes on the above mentioned point. The method of visioplasticity has been applied. By the measurement of the displacement vector of each material point on the grid lines, the strain-rate pattern has been obtained. According to the Lévy-Mises plastic equation the pattern of the directions of the principal shearing stress has been calculated. On the other-hand the slip-line field has been obtained in the case of the friction angle 7°. The two patterns are comparatively in good agreement. The analysis of the stress distribution is yet to be made.