Fatigue crack initiates in the slip band and exists also in it near fatigue limit; many slip bands are apt to appear in the direction of the maximum shearing stress; crack propagates by the normal tensile stress; the maximum shearing stress on a plane at fatigue limit is reduced by the effect of the normal stress on the same plane. From these results of the experiment, a new criterion is proposed, which coincides with the Gough's empirical formula for the brittle materials under combined stress. As the plane of the maximum shearing stress is varied by the various combination of torsion and bending, the isotropic material should be used in the combined stress experiment. In this paper, the results of experiments on three isotropic and one anisotropic materials are discussed and compared with the criteria proposed up to the present.

View full abstract

Measurements and observations on the closure of fatigue crack initiated from notch roots in an annealed and a quenched-and-tempered 0.54% carbon steel in tension-compression fatigue tests were made successively while increasing the stress level, in order to investigate the influence of mean stress on fatigue crack propagation and crack closure phenomenon. The following are the results obtained: (1) The crack closure phenomenon in tension-compression fatigue can be recognized more clearly in the annealed steel than in the tempered steel. (2) The crack propagation rate dl/dN can be expressed as [numerical formula] where Δσ_eff means the effective stress range under which the crack is fully open and l is the crack length. The exponent m is about 1.6 for the annealed steel and about 3 for the tempered steel. Mean stress affects Δσ_eff alone, therefore, the constant C is not related to the mean stress.

View full abstract

Plastic strain is mainly due to the crystallographic slip in grains. The slip quantity Δλ_εs reversed in individual slip bands is more important than the macroscopically observed plastic strain Δε_p as the criterion of a crack initiation along a slip band. In the present work, it was investigated on the basis of microscopic observation how the macroscopic plastic strain was distributed in the individual slip bands. Manson-Coffin type relation was expressed by using the slip quantity Δλ_εs reversed in the individual slip band. The material used was an anti-corrosion aluminium alloy.

View full abstract

High velocity tensile tests of JIS Type 304 stainless steel have been made at room temperature, 400°C and 600°C. Tests were done with an explosive impact tensile tester. The purpose of this work was to determine the effect of test temperature and strain rate on the mechanical properties of SUS 304 stainless steel. Various quantities of explosive have been used as a parameter to cover various ranges of strain rates. The dependencies of strain rate upon 0.2% yield strength, ultimate strength, elongation and deformation energy were shown.

View full abstract

In this paper the large plastic deformations of thin elliptical and rectangular metal diaphragms being edge-clamped and bulged by hydrostatic pressure are calculated, using the incremental finite element method which was formulated by the authors. The predictions are obtained in terms of the bulge pressure - central height relation, the bulge pressure - polar thickness strain relation, strain distributions on the major and minor aces of diaphragm, and strain path and stress path at the pole of diaphragm. In general, theoretical predictions are reasonably in agreement with the experimental results of aluminium sheet.

View full abstract

In reducing the total mass of structures, it is sometimes necessary to pay such dynamical considerations as avoiding resonant phenomena between the structures and external systems involving the sources of vibration. From this point of view, a kind of problems named "type-1" have been proposed in literatures in this field: i.e. the problem of finding the optimum shapes of structures maximizing their fundamental natural frequencies for prescribed mass of them. In the first report, a general method has already been presented for optimum designs of structures in which natural frequencies have been involved in objective functions and conditions of constraints. Thus in this reports, some demonstrative applications of this method are shown in the type-1-problems of vibrating structural elements of 1-section-freedom^<****%gt; and 2-section-freedom, which contain extensional, shearing and torsional vibration problems of elastic rods and flexural vibration problems of beams. And numerical results obtained are discussed in detail from many points of view.

View full abstract

In the first report, there has been presented a general method of optimum designs of structures in which natural frequencies are involved in objective functions and/or conditions of constraints. In the second report, the method was demonstrated in the type-1-problems, which is one of basic problems of frequency maximization. Concrete objects were taken from structural elements having 1- and 2-section-freedom^**** and numerical results obtained there were discussed in detail from many points of view. In this report, another basic type-w-problem (i.e. the problem of finding the optimum shapes of structures which give minimum total mass under specified fundamental frequencies) are treated numerically by our same method, taking example of vibrating rods and cantilevers with a point mass. Moreover, reciprocal relations between type-1-problem and type-2-problem are also clarified analytically in discrete systems and continuous ones, and the relations are verifyed numerically by making use of the results in rods and cantilevers.

View full abstract

In this report, results of an experimental investigation which was performed to eliminate the streamwise vortices arising in the nominally two-dimensional turbulent boundary layer, using the procedures which made the main stream uniform completely by replacing bad screens with appropriate ones and cutting those off, are presented. The results are compared with the previous authors' values and the effects of the open area ratio of the screen are confirmed. The differences between the flow fields in which the streamwise vortices exist and the completely two-dimensional flow fields are discussed and the feature of velocity defect region is clarified.

View full abstract

Three-dimensional turbulent boundary layer developing along a long streamwise bar of a rectangular cross section (60 mm × 20 mm) placed on a flat plate was investigated experimentally. Two different wall configurations are possible by placing the longer sides of a rectangle perpendicular or parallel to the plate surface. This class of turbulent boundary layers is made complicated by secondary currents which inevitably arise near the streamwise corners. Experimental results obtained for both wall shapes are described mainly concerning isovels, lines of constant turbulence intensity, secondary currents measurements by vortex-meter, distributions of wall shear stress and total drags. Considerations are made on the velocity profiles in a plane of flow symmetry where secondary currents flow toward the wall, and a self-preserving distribution in the outer layer is proposed.

View full abstract

The present work is concerned with an experimental investigation of the wake flow behind a two-dimensional wedge. The experiment was performed in the wind tunnel at the Reynolds number R_e = 9.0×10³. The pressure distribution on the wedge surface, the vortex-shedding frequency behind the wedge and the distributions of velocities and static pressures in the wake flow were measured. From the surface pressure the drag of wedge was obtained. The velocity distribution and the width of the wake in the similar profile region were compared with the theoretical results obtained by Schlichting and Reichardt, and discussed in comparison with the results for a circular cylinder.

View full abstract

This paper presents an experimental investigation on the flow past a cone placed on a float plate and its wake. The vertex angle of the cone was varied 60°, 90°, 120° and 150°. The surface-pressure distributions on the cone and the flat plate, the velocity and the static pressure in the wake were measured, and the separation and the vortex formation were observed by visual means. The main results are as follows : (1) There occur necklace-like vortices around the cone of vertex angle 60°120°, and horse-shoe vortices generated behind the cone are shed downstream periodically. (2) The lift acting on the cone and the drag decrease with an increase in the vertex angle. (3) There exists a region of negative wake, which becomes maximum at a downstream distance of about twelve times the height of cone.

View full abstract

Two kinds of fast acting valves have been developed to replace diaphragm-breaking which is usually employed in the conventional shock tubes. One valve, which is called a type-H valve, opens in the direction of the shock tube axis and has an advantage that the moved distance required for full opening is very short. Another valve, called a type-V, moves vertical to the shock tube axis. The motion and the opening time of these valves were analyzed. The experimental results agree well with the calculated values, especially at a high actuating pressure of the valves. The type-H valve gives stronger shock waves, therefore it is more suitable for use in the shock tubes than the type-V valve. But the type-V valve has an advantage that the compression chamber has the same cross section as that of the expansion chamber, which may be desirable to the Ludwieg tubes.

View full abstract

Considering the Mechanisms to be constructed with several kinds of basic open chains, the relative displacements of links in each basic open chain and the connecting conditions of the open chains have been generally described by the transformation functions. The transformation functions enable the systematic displacement analysis of planarmultilink mechanisms without the derivation of displacement equations. In the analysis, all of the possible motions can be separately evaluated in correspondence with the conjugate positions of mechanisms. The method has been applied also to the error analysis of mechanisms.

View full abstract

In this paper the techniques are presented which allow one to solve rapidly the displacement equation of the kinematic linkage. Especially the following two techniques make a significant contribution to the development of the method. (1) The new method of formulation of the problem saves one from the cumbersome manipulation of the equations, which is usually required to translate the statement about a given linkage into an analytical one. (2) A set of displacement equations is solved by the Newton-Raphson method, which is remarkable in convergence speed but is apt to be affected by the choice of the starting value. Owing to the closer estimation by the parabolic extrapolation through the last three solutions for the preceding three input crank angles, the convergence rate is about twice that of the usually chosen starting value of the solution corresponding to the preceding input crank angle.

View full abstract

In designing a bolted joint, it is important to find the ratio of the increment of an axial force produced in a bolt to a load applied on an assembly, that is, the force ratio. For the force ratio, up to today, the formula proposed by Thum has been used. Recently it has been found that the value of the force ratio varies with the position of the load application on the assembly, and the method to compensate Thum's formula was proposed by Junker. But this compensation is based on experience and the theoretical basis is not made clear. In this paper, the concept of the tensile spring constant K_pt for the clamped part is introduced when a load is applied on the outer circumference of clamped parts and the force ratio is calculated by using K_pt. The value of K_pt is analysed by the three dimensional theory of elasticity. In order to verify this theory, an experiment is carried out and the theoretical results are compared with the experimental ones. They show a fairly good agreement. In the case where a part of the interface of clamped parts causes a separation with an increase of the load, theoretical and experimental results show a fairly good agreement, too.

View full abstract

This paper studies an approach of two reference planes between two rough surfaces in contact, i.e. an approach due to the elastic and plastic deformations of the contact asperities. Assuming that in the first loading the shape of the contact asperities is pyramidal and in unloading it is pyramidal with a spherical tip, the approach between two ground surfaces in contact resulting from the changes of the surfaces roughness and the relative position of the machined lays are analyzed and discussed. The approaches between two ground surfaces of low carbon steel (S15CK) were measured in loading and unloading. After various loads were applied and removed, the radius of curvature of the contact asperities deformed plastically and the peak heights distribution were also measured. The predictions for the approach, the distribution of the radii of curvature and that of the peak heights agreed with the experimental results.

View full abstract

Pitting fatigue limits obtained from many past investigations were in the range p_max = (0.2=0.35) HB and they were less than the lowest shakedown limit (p_max = 0.4 HB; past theoretical pitting limit) when test rollers made of steel were rotated under rolling and sliding conditions. In the case of pure rolling, it was recently demonstrated by the present authors that pitting fatigue limits became greater ( e.g., p_max = 0.7 HB) when a testing machine made by the authors was used with precise test specimens. However, in the case of combined rolling and sliding contact, an attempt to increase the pitting limits of rollers was unsuccessful, as expected by many investigators. After unfruitful experiments for about one year, the authors succeeded in appreciably increasing pitting limits using another new testing machine with an improved performance. Not even a single pit occurred at specific slidings 1.5 to 29% at a Hertzian stress of 123 kg/mm² (p_max = 0.68 HB) when 180 HB steel rollers ground to a surface roughness of 1 μm R_max were used in equal hardness combinations. Moreover, no pitting occurred on a 180 HB roller up to 10⁷ rotations at a Hertzian stress of 137 kg/mm² (p_max =0.76 HB) when the soft roller was combined with a superfinished hard (420 HB) roller and rotated at a specific sliding of about 20%.

View full abstract

Film thickness and pressure distributions in foil bearings are analysed in connection with magnetic tape memory devices. The finite element technique is used. As examples, analyses are done in cases of a circular head and a double apex head. The results are given in graphs. The region of constant film thickness and the pressure spike in the exit region, both of which are typical features of elasto-hydrodynamic lubrication, are clearly seen in them. Experiments are also carried out in case of a circular head. Close agreements between calculated and experimental results are obtained.

View full abstract

Two-dimensional cutting theories of soils (dry quartz sand and plastic loam) are made. Such quantities of the cutting as the magnitude and directions of the cutting force, its application point, and the direction of shear plane are determined from geometrical cutting conditions, soil failure conditions and frictional conditions between soil and a cutting blade. Experimental results agree approximately with the theoretical results, and it is confirmed both experimentally and theoretically that the cutting force is reduced effectively by the application of electro-osmosis to the low-frictional cutting blade.

View full abstract