As a part of series investigation on the low-cycle fatigue properties under the multiaxial stress condition, low-cycle fatigue tests were conducted under axial and reversed-torsion strain cycling conditions. Two kinds of carbon steel, a Ni-Cr-Mo steel, a commercially pure aluminum, a 7075-T6 aluminum alloy, and cold-worked and annealed coppers were tested in this investigation. The equivalent plastic strain range based on the octahedral shear strain theory was used in comparing these axial and reversed-torsional results. In general fairly good agreement was observed between the two cycling results, although some discrepancies existed in the case of the carbon steels and the annealed copper investigated. Some amount of anisotropy was found in fracture ductility between the axial direction and the diametral direction of specimens. This might affect more the reversed-torsion results. For more precise correlation of the two cycling test results, therefore, the influence of the anisotropy as well as that of the crack propagation characteristics should be taken into consideration.

View full abstract

To investigate stress distributions in a plate under the tension through rivets, we consider a semi-infinite plate with equal spaced identical holes which are in a row parallel to the boundary of the plate and equally loaded at their edges. Applying the theory of two-dimensional elasticity and deriving a series of periodic stress functions containing unfixed coefficients which have poles at the centers of the holes and satisfy the condition of free boundary at the straight boundary, we solve the problem by determining the coefficients in such manner that each hole is under given boundary conditions on the edge. Using these solutions, we calculate stress distributions in the plate for several cases.

View full abstract

The problem of the stress concentration in a thin cylindrical shell with one or two circular holes, which is subjected to axial tension or internal pressure, is treated. A numerical method is proposed, which uses the Fourier expansion in the circumferential direction and the difference approximation in the radial direction. The results here obtained for the case of one hole are ascertained to be consistent with those previously published by other researchers. For the case of two holes, the boundary conditions around the holds are satisfied by iterated calculations, and it is found that the tendency of the stress concentration by the axial tension is similar to that for an infinite plate with two holes, although the values of the stress concentration factors are higher than those for plates.

View full abstract

Stress and strain distributions in polycrystalline aggregates during uniaxial elastic deformation were investigated adopting the finite element method. The effect of grain shape on the deformation mode was studied. The distribution of the constraint ratio between grains defined previously was calculated for each grain shape. It was found that the strain is continuous across the grain boundary parallel to the stress axis, while the stress is continuous across the grain boundary perpendicular or inclined to the stress axis. The deformation of island grains was also discussed.

View full abstract

The elasto-plastic stress analysis of a straight prismatic bar bend under the shearing force, is very difficult even for a simple cross section. Such problems have been investigated by an elementarily analytical method, provided that all stress components apart from bending stress are zero. The analytical method for this problem is formulated in this, both for a complete elastic material, and for a compressible isotropic work-hardening material obeying a non-linear stress strain law. Then, the fundamental equation for an elastic bending is reduced to Laplace's equation and on the other hand, one for an elasto-plastic bending, represented as a system of nonlinear, second-order, partial differential equations, can be linearized by adopting new parameters in the system of the stress space. Provided that the Ramberg-Osgood's law is employed as a nonlinear stress strain relation, the linearized governing equation can be reduced to a hypergeometric differential equation.

View full abstract

An approximate analysis is made of the effects of wall-conductance and external-circuit condition on the velocity profiles of steady magnetohydrodynamic flow of an incompressible, viscous, and electrically-conducting fluid between two-parallel rotating disks in an axial magnetic field. Perturbation method is employed to take the inertia effects into account. By numerical examples, the effect of these parameters on the velocity profiles and hence the torque of rotating disks are clearly shown. (i) The circumferential velocity profile is directly affected by the wall-conductance, but the effects on the radial and axial- velocities are second order. (ii) The effect of total current I* on the circumferential velocity is large only for the insulating-wall. (iii) Total torque of the rotating disks changes largely with the wall-conductance.

View full abstract

A gas flow in and behind tube banks with tuber axes normal to the gas flow is highly turbulent containing numerous vortices of different sizes and intensities. Each vortex in this complicated wake is the result of a periodic generation, and the periodicity sometimes induces serious vibration problems in steam generators and heat exchangers. The structure of the flow may be treated ass a complex synthesized result of Karman vortex flow. This paper is the first report of a series of studies carried out form the above mentioned viewpoint, and presents the experimental results about a Karman vortex flow from two tubes at various spacings. The vortex formation region was visualized by means of the Schlieren method and was recorded by a 16mm movie camera. The vortex shedding frequency was systematically studied. The frequency for two tubes is entirely different from that for a single tube. It was found that the size of the vortex formation domain has a dominant effect on the frequency.

View full abstract

One method to measure the instantaneous flame temperature within the cylinder of a gasoline engine by means of the absolute radiation method is proposed. The radiation energy of D-line emitted from a trace of sodium, which is added to fuel, is measured by an optical-electronic system. Then, the measured energy is converted into the equivalent black body temperature. This temperature may be considered as the flame temperature which is averaged through an optical path in the luminous flame zone. In the process of conversion, the half-width of D-line is determined experimentally and the number of Na atoms radiating in the optical path is evaluated theoretically. As the optical system, a quartz window, 4 limiting diaphragms, an interference filter and a fiber-scope are arranged in order. And a light beam having 1 mm in diameter is introduced to a photo-multiplier by means of the fiberscope. By this technique, the flame temperature in CFR engine is measured under various operating conditions.

View full abstract

The temperature measured in the 1st Report is the one which is averaged through an optical path in the luminous flame zone. In the first step of this paper, the physical meaning of this average temperature is discussed. For this purpose, the flame is divided into many layers and the radiation from the whole flame is considered as the result composed of radiation and absorption of each layer. In the second step, based upon the above consideration, the temperature distribution in a flame is measured. To divide the flame into many layers, and to measure the radiation effect of each layer, a light-intercepter is inserted into the flame and displaced successively across the flame. By this technique, a change of the flame temperature distribution versus crank angle is measured on a CFR engine. As one of the results, Hopkinson's effect is ascertained. Namely, at the instant when the flame spreads over the combustion chamber, it is found that the temperature of gas around the park plug is about 300°C higher than that of gas near the cylinder wall.

View full abstract

We studied the effects of double injection by a single nozzle on fuel consumption and noise level of a single cylinder 4-stroke direct injection diesel engine. It was observed that, within a certain range of fuel quantity of pilot injection, the combustion noise was reduced considerably. This range of the fuel quantity depends on the timing of the pilot injection. In view of the fuel consumption, it was found desirable to inject pilot fuel at the end of exhaust stroke of the previous cycle or just before the main injection. Furthermore, a thermodynamic analysis was made to explain qualitatively the mechanism of the effects of pilot injection.

View full abstract

The distribution of the slopes of projections of surface asperity is necessary knowledge to analyze the deformation of contacting rough surfaces. In this report, the distribution of the slopes of projections is analyzed theoretically for ground surfaces of carbon steel and cast iron on the supposition that the asperity of surface is a wedge. The slope of projections is smoother than the original when a surface is measured by the profilometer with a stylus. Assuming that the profilograph is hyperbolic, the distribution of the slopes can be analyzed theoretically. The results coincided with experimental data. This distribution defined in this report varies depending on the form parameter, but the variance of the slope has no relation with this parameter. In this report, the slope is calculated from the discrete data. As a result, the variance of the slope could be calculated for any sampling space and then the authors showed the method to determine the variance of the slope from that of surface roughness height.

View full abstract

This report deals with the calculation method to determine the highest point of ground surfaces (the extreme value of surface roughness) which is important to analyze the elastic and plastic deformation of contacting rough surfaces. On ground surfaces, a probability density function of surface asperity has always been assumed to be a normal distribution. For ground surfaces, however, the Weibull distribution gives a better approximation. In this report, using the Weibull distribution of which the form parameter is m=4.0, the extreme value of ground surfaces can be calculated statistically. With instruments having an optical flat and a precision steel ball, we could measure the extreme value of 0.2∼0.5μ accuracy. The experimental results reveal that the extreme value of surface asperity coincides with the value calculated by the Weibull distribution which is smaller than by a normal distribution. And further we explain that the standard deviation of surface asperity measured along the grinding direction can be calculated from that measured along the direction perpendicular to the grinding one.

View full abstract

The elastic and plastic deformation of plane joint lowers the static stiffness of machine tool structure. Its deformation is considered to be that of contact regions between a rough surface and a smooth one. In this report, the elastic and plastic deformation of contact regions between a directional surface asperity and a rigid flat one was theoretically analyzed. For this analysis, the following assumptions were made. (1) The projections of surface asperity are wedges and the probability density function of topography is a normal distribution. (2) The contact of rough surfaces can be replaced by that between a rough surface and a rigid flat surface and the stress acting on contacting points is equivalent to the plastic flow pressure. As a result, the depth of elastic deformation of asperity is found to be about twice the mean distance of contacting points and the computation formula coincides quantitatively with experimental data. The deformation depends on the contact surface coefficient K₁ which is newly defined in this report. It is concluded that the relation between computed values of contact stress and elastic deformation is that of a straight line on a double logarithmic chart.

View full abstract

With an increasing capacity of large marine engines, the axial vibration of a large ship propeller shaft becomes a new problem. In order to solve this, investigation of the dynamic characteristics of thrust bearings is necessary. The authors theoretically analysed the variation of the oil film thickness and the pressure distribution of hydrodynamic thrust bearings under periodical load and verified the results experimentally. In the theoretical calculation, two different ;methods were used. One of them is a strict method, in which the Reynolds equation is solved at every small time interval (nonlinear solution), and the other is an approximate method, using spring and damping coefficients of the oil film (linear solution). The results are: (1) The linear solution method can be applied up to a considerably large amplitude of load. (2) Phase delay of the oil film thickness against the load is large, which has been verified both analytically and experimentally.

View full abstract