Two stress steps multiple tests including understress are carried out using smooth specimens of two kinds of carbon steel, S10C and S45C, under rotating bending and reversed torsion. The effects of understressing on crack initiation and fracture are examined and discussed. Cumulative cycle ratio for crack initiation and fracture calculated by Miner's rule decreases with decreasing number of cycles in one block, n₀. Understressing influences more strongly the crack propagation behaviour than the crack initiation and results in an increase in the crack propagation rate under overstress for smaller n₀. Acceleration and deceleration phenomena are recognized in crack propagation behaviour during overstressing just after the stress changes. The former occurs in the case of rotating bending, of S45C and of smaller n₀, and the latter in the case of reversed torsion, of S10C and of larger n₀.

View full abstract

Effects of the strength and microstructure of the coarse second phase on tensile ductile fracture processes were studied. The materials used were 0.16% carbon steel composed of ferrite (matrix) and martensite or pearlite colony (coarse second phase). With an increase in martensite strength which was adjusted by tempering temperature, void formation mode was changed from a fracture of the colony to a decohesion at the interface, and fracture strain (ε_f) was decreased. This decrease of ε_f was due to a decrease of void initiation strain (ε_f). In comparison with martensite colony's case, number of voids produced during deformation and ε_f vs ε_i relation in ferrite-pearlite material were quite different. The reason seemed to be in the anistropic fracture strength of pearlite colony.

View full abstract

Most of the previous theories about brittle fracture, based on the weakest link theory, ignore the effects of shear stresses. In this paper, on the contrary, more general multiaxial distribution functions are formulated taking into consideration both the normal and shear stresses and assuming straight surface cracks and penny-shaped cracks. Using new multiaxial distribution functions, the brittle fracture loci under bi-axial stress state are analysed.

View full abstract

In this paper, the transient responses of a cantilever circular cylindrical shell subjected to impulsive loads on its free edge are analyzed on the basis of an improved dynamic shell theory, namely, Mirsky-Herrmann's theory, considering the influences of the rotatory inertia and the transverse shear deformation by the use of Laplace transformation. The numerical results are compared with those previously obtained by the author via Flugge's classical dynamic theory. It is shown that the results predicted by the classical and the improved dynamic shell theories are in good agreement about the displacement components and the in-plane stress resultants, but are considerably different from each other about the out-of-plane stress resultants, namely, the bending moments and the transverse shearing forces.

View full abstract

In the present paper, we consider the problem of thermal stresses in an infinite plate with an infinite row of circular holes which are filled with elastic inclusions of another material. The steady state of temperature is induced by the infinite row of pairs of heat sources (positive and negative sources) periodically placed on either side of the infinite row of inclusions. The analysis is developed by the Airy's stress function in the generalized plane stress. In the construction of temperature and stress functions, periodical harmonic functions are utilized as the fundamental stress function. The successive approximation is adopted for the numerical determination of the unknown coefficients involved in the solution. Moreover, numerical calculations about the stress distributions along main parts of the infinite plate and the inclusions are worked out in some detail, in order to clarify the effects of an infinite row of circular inclusions.

View full abstract

Deformation behaviour of softened celluloid is observed by applying combined loadings of tension after torsion along orthogonal bi-linear stress trajectories of a thin-walled tubular specimen. The experiment is performed under two conditions : creep test and continuous loading test. From the experimental decreases just after the corner of stress trajectory, and this trend decreases with the development of a deformation after the corner. The trend appears more remarkably for higher level of pre-stress. In order to formulate a stress-strain relation from the experimental results, the relation having a principal term of the third order in the non-linear viscoelastic theory is extended to application for the above-mentioned trajectories. The influence function contained in the relation is determined by using experimental results. The stress-strain relation obtained is confirmed to approximate precisely the experimental results.

View full abstract

Experimental investigations on the characteristics of the flow around two circular cylinders in a cross flow were carried out. Reynolds number was varied in the range of subcritical values, 8.7×10³ <__= Re <__= 5.2×10⁴ and the distance between the axes of the cylinders in the spacings 1.03 <__= L_d <__= 5.0, where d is the diameter of the cylinder. Up to the spacings of L/d=3.5, where the quasi-stationary vortices are formed between the cylinders, the changes of the flow structure around the two cylinders were observed at the spacings of L/d=1.1, 1.6, 2.3 and 3.1. Particularly, the dependency of the Reynolds number was recognized in the range of 1.1 <__= L/d <__= 2.0.

View full abstract

Experimental and theoretical investigations are carried out on the secondary flow in the downstream of a rotor row in a single stage axial-flow fan with a stator row ahead of the fan wheel. A five hole pitot tube was placed in the downstream of the fan wheel, and the relative flow field was measured for several hub ratios by traversing the tube over one blade spacing. An analysis is made of an incompressible, inviscid, but rotational flow through the stator row and the successive rotor row. The relative streamwise component of vorticity is obtained analytically in the downstream of the rotor row, and is used to calculate the secondary flow velocities. Discussions are made on the stream function of the secondary flow, the air outlet angle from the rotor row, the distribution of loss coefficients, the swirling flow rate of the secondary flow, and its kinetic energy. Hub ratio effects on these items are examined.

View full abstract

The flow pattern in a rotary flow dust collector was changed on a large scale by the flow rates of the primary flow Q₁ and the secondary flow Q₂. Considering the flow characteristics of a turbulent rotational flow, one of the authors newly derived theoretical formulae for the cut-sizes which were determined by another conception in comparison with the derivative means of the cut-size of the ordinary types of cyclones. In these results, the authors confirmed that the relationship between the experimentally determined cut-size Xc₅₀ and the theoretically calculated cut-size Xc showed good coincidence in the range of flow rates Q₂/Q₁ = 1.0∼1.5.

View full abstract

The turbulent near wakes starting from the turbulent boundary layers on both sides of a flat plate and subjected to pressure gradients have been investigated. The experiments have been conducted in zero, adverse, favourable and mixed pressure gradients, and the results show that the total pressure along a streamline of the near wake in a pressure gradient changes at a rate nearly equal to that along the same streamline in zero pressure gradient, which suggests that the transverse gradients of Reynolds stress are little affected by pressure gradients in the turbulent near wakes. On the basis of the experimental results a model to predict the development of a near wake in a pressure gradient has been presented. The calculated results agree well with the experimental data except for a strong adverse pressure gradient case.

View full abstract

The governing equations for a steady fully developed laminar flow in a helically coiled pipe are derived and the solutions are obtained for small values of a^*/R^*, where a^* is the radius of the pipe cross-section and R^* is the radius of the circular cylinder on which the pipe center-line is helically coiled. The torsion of the pipe center-line causes asymmetry of the flow pattern in a pipe cross-section, an example which appears in the inclination of the flow pattern at large values of the parameter D₁. Here, D₁ = (C^*a^*3/ v^*2ρ^*)(2a^*cos²β/R^*)^1/2, where C^* is a costant pressure drop along the pipe center-line, v^* is the kinematic viscosity, ρ^* is the density of the fluid and β is a constant angle which the normal to the helix makes with the generating line of the cylinder. It is concluded that the resistance formula for a toroidally curved pipe is also applicable to a coiled pipe, if the curvature of the coiled pipe is used instead of that of the toroidally curved one.

View full abstract

From the experimental results of a two-dimensional free jet flow, it was found that the flow depends on the initial turbulence intensity at the nozzle exit and this dependency diminishes as turbulence intensity becomes large. The relations between the empirical constants contained in the eddy kinematic viscosity and the initial turbulence intensity were obtained comparing the experimental values of a velocity decay on the jet center line with its approximate calculations. These relations were used for the approximate calculations of a two-dimensional reattachment jet flow. The calculations of the reattachment flow having different turbulence intensities agree well with the experimental results in a wide range of offset ratios, except so small offset ratios.

View full abstract

We test the performance of a wave power absorber of a rocking body of Salters Duck of 37cm in diameter and 150cm in width combined with a hydro-static power conversion mechanism consisting of a hydraulic cylinder, check valves and an accumulator, clarify the effect of both the shape of the body and the load characteristics on the efficiency, and propose a control method of hydraulic power conversion mechanism for efficient storage and absorption of wave energy.

View full abstract

A transient hot-wire method to measure the thermal conductivity of fluid was applied to liquids under high pressure. One of the advantages of the method is that it can eliminate errors due to convection, as examined in the previous report by the present authors. Furthermore, simple structure of the test section is suitable for a high pressure instrument. Measurements were carried out with the aid of a new apparatus for two fluorinated hydrocarbons, R11 and R113, and toluene in the range of temperatures 212 K -353 K and pressures up to 54 MPa. Comparisons with existing data show that the only one set of data available for R113 under high pressure are larger than those of the present study.

View full abstract

An experimental study was made to investigate the heat transfer and flow around the third and fourth ones of four cylinders of in-line in a cross flow of air. The in-line pitch ratio was in the range 1.15 <__= c/d <__= 5.0, where c is the center to center distance and d the cylinder diameter, and in the Reynolds number 12000 < Re_d < 50000. The heat transfer characteristics of the third and fourth cylinders are found to exhibit a strong dependency upon the separation point of their upstream cylinders.

View full abstract

A self-excited oscillation in a pipe containing heat sources has been studied both experimentally and theoretically. Various parameters, total pipe length, relative location of heat sources, configuration of heat sources and so on have been clarified through quantitative experiments concerning the onset of an oscillation with three kinds of gases, air, helium, and carbon dioxide. New type of critical curve with pressure and temperature difference between the heat sources is presented in this paper. Thermodynamic state near heat sources has been visualized with an M-Z interferometer, which clearly shows the existence of heat cycles near them for the first time.

View full abstract

This paper deals with the influences of swirl and turbulence of a gas on the burning velocity in a cylinder using a single-compression test equipment, in which the flow conditions can be varied by changing the flow-in direction of the scavenging gas with adjustable vanes. The influences of the spark location and the rotational speed of the equipment on the burning velocity were also studied by means of both pressure measurements and direct photography of the flame propagation. The gas motion during the compression stroke was measured with a hot-wire anemometer and the result of analysis showed that in the case of a turbulent flow without swirl, the process of a change in kinetic energy of the gas could be expressed as follows. (ε/ν³)^1/4L =1.91(u^^-L/ν)^0.700 where ε denotes the rate of a change in kinetic energy, ν the kinematic viscosity, u^^- the mean velocity of response by a hot-wire, and L the height of the combustion chamber. The process of a change in kinetic energy was also discussed in the case of a turbulent flow with swirl.

View full abstract

The nonstationary vibration of a rotating shaft which passes through two critical speeds successively under conditions of uniform acceleration is theoretically analyzed by both an exact solution and approximate expressions for the vibration, neglecting the damping force. The analytical results show that the nonstationary vibration is characterized by one parameter containing both the critical speed ratio and the acceleration rate, and that the vibration is directly affected by the unbalance distribution in the case of fast acceleration. Furthermore, formulae for evaluating the maximum amplitudes and the rotational speeds of the nonstationary response are derived, and the evaluated values are compared with those obtained from the exact solution.

View full abstract

This paper presents closed form narrow bearing approximate solutions of the incompressible Reynolds equation applicable to estimating the static and dynamic behaviors of sub-micron slider bearings used in a current disk storage unit. General trends of the static and dynamic characteristics of slider bearings are discussed from these narrow bearing solutions, considering a slip flow effect. By comparing these analytical solutions with numerical results calculated from a generalized compressible Reynolds equation, the valid regions of these solutions are clarified in connection with the slider width to length ratio and compressibility parameters. As a result, it is shown that the film fore can be estimated quantitatively for a sub-micron slider bearing whose width to length ratio and bearing number are smaller than 0.1 and 10 respectively.

View full abstract

This paper is concerned with the effects of a turbulent flow in hydrostatic thrust bearings. Applying an equation representing the "low of wall" to the three-dimensional turbulent shear flow, the local velocity distributions between a rotating disk and a fixed wall are calculated as a function of Reynolds number, local pressure gradient and the ratio of the film thickness to the radius. Using above results, the pressure distribution and the load capacity of hydrostatic thrust bearings with a circular recess are presented. The theoretical results agree well with the experimental data.

View full abstract

Using an inductance type vibrometer for shaft, the displacement of bevel gears is measured on a rectangular-coordinate, and the results are discussed. The results are as follows : The displacements of three directions perpendicularly intersecting each other are influenced by the bending stiffness of gear shaft and the stiffness of bearings. The displacements of three directions are influenced by the natural frequencies for torsional vibration of a gearing system and the natural frequency for bending vibration of a gear shaft. The vibration of high frequencies of 2fz, 3fz, 4fz etc. appears in the axial displacement of bevel gears. The vibration of low frequency is considerable influenced by the stiffness of bearing materials.

View full abstract

In the previous reports, in order to obtain high surface durability of the H_B &ap; 340 (hardened and tempered) gears, the authors proposed that the initial tooth surface should be finished as smooth as possible. In this report, surface durability of gears finished by various methods (skiving --- finish-hobbing with carbide tipped hob, honing with screw-type hone, lapping and grinding) are compared with that of hobbed gears by using a power circulating-type gear testing machine. From these results, it is found that the smoother the surface, i) the less the number of pits and the smaller their size, ii) the smaller the rate of increase of pits with repeated contact stress, and iii) the higher the load-carrying capacity. Therefore, the influence of surface roughness on the surface durability is much more clarified, and the effect of various surface finishes can be grasped quantitatively.

View full abstract

Equation of motion for free vibrations of a pair of gears is derived in a general form. A vibration system of two-degrees of freedom which represents the fundamental characteristics of bevel gears is constructed mathematically. It is found from the analysis of the system that the fundamental characteristics of vibrations which distinguish bevel gears from spur and helical gears are caused by a change in the direction of the tooth-normal in tooth-meshing. the influences of parameters such as the contact ratio, stiffness of gear-carrying shafts and damping ratios on the band width of unstable regions of the vibration are also shown.

View full abstract

The authors calculated the deflection of various spur gear teeth by using a two-dimensional elastic theory and the mapping function, and showed a simplified formula for the tooth deflection, which was deduced from the results of calculation based on the elastic theory. The deformation by Hertzian contact stress was also calculated in the present study. When this deformation was compared with the values obtained from Weber's and Lundberg's formulas, it was found that their formulas involve in themselves a problem which arises in the calculation of tooth deflection. Furthermore, in the present paper, the authors showed an approximate expression for the deformation by Hertzian contact stress.

View full abstract