The purpose of this paper is to make clear the reason why the fatigue limit and the coaxing effect exist in steel, using 0.13%C steel and copper specimens. The main results obtained are as follows. In 0.13%C steel specimens ; (1) The incipient fatigue cracks, revealed by light electro-polishing (1∼2μ) after 10⁷ cycles under the stress repetitions of the fatigue limit of the specimen, exist in the ferrite not only at the surface but also in the inside and the tip parts of those cracks become 24∼34% harder than in the virgin state. (2) This incipient fatigue crack appearing by the stress of the fatigue limit is the starting point of fracture, when the specimen fails by a stress greater than the fatigue limit. In copper specimens ; (3) The tip parts of the incipient fatigue crack in copper, appearing after 10⁷ cycles of the stress which corresponds to the fatigue strength for 2×10⁷ cycles, do not become so hard than the ones in 0.13%C steel. These results show that the existence of the fatigue limit in steel is due to the strengthening of the tip parts of the non-propagating micro-crack.

View full abstract

In this report the hypothesis of non-propagating fatigue crack proporsed in a previous report was applied to the case of an infinite plate with an elliptical hole notch which was subjected to push-pull fatigue stress at zero mean stress. Following results were obtained by the calculation. The non-propagating fatigue crack curve obtained by the application of the hypothesis well explained not only the relation between non-propagating crack length and given alternating stress but also the crack propagation fatigue limit. Further, it was shown by the hypothesis that there was the critical crack size as the constant by which the crack initiation fatigue limit in the low stress concentration factor region could be decided.

View full abstract

The fatigue crack propagation behaviors were investigated on HT80 steel plates under net section stress control in which stress range and minimum stresses were widely varied. The fatigue crack growth rates were proportional to the effective crack length 2a_e which was an actual crack length 2a modified by the finite width effect of the specimen. When minimum stress δ_min was in compression side, the growth rate depended on maximum stress δ_max and in tension side, it depended on stress range Δδ_n. As the result, the effects of crack length, stress range and minimum stress on the fatigue crack propagation can unifyingly be shown such that the crack growth rate is proportional to the cyclic plastic zone size at a crack tip[numerical formula]. Where, δ_y is yield strength of material and δ_n is δ_max when δ_min≤0 and is Δδ_n When δ_min≥0

View full abstract

The biaxial low-cycle fatigue properties of anisotropic rolled steel plates were investigated. Biaxial low-cycle fatigue tests were conducted by utilizing the principle of anticlastic bending of flat rhombic plates. Crack initiation and propagation were examined with special reference to angles between the applied principal stress axes and the anisotropy axes of the material studied. The biaxial fatigue life of anisotropic specimens cut out in a given direction satisfied a linear relation on a log-log scale between an equivalent strain range and the number of cycles to failure. The direction of crack initiation under the uniaxial and biaxial stress conditions was coincident with that of the maximum shear stress. The fatigue life under the uniaxial stress condition was shorter than that determined under the biaxial stress condition for a given value of the equivalent strain range. The discrepancies were interpreted in terms of both normal stress component acting on the crack surface and the modes of crack propagation.

View full abstract

In this paper, a method is treated which predicts the low-cycle fatigue failure of bellows subjected to completely-reversed loading, with use of the solution procedure employing the finite difference technique proposed previously by the authors for the large deflection problems of elastic-plastic shells of revolution. Namely, a procedure is discussed which estimates the low-cycle fatigue life, applying the plastic strain range under cyclic loading assessed by the elastic-plastic analysis to the Manson-Coffin equation. A U-shaped bellows subjected to deflection-controlled cyclic loading is taken as an example for the fatigue life analysis, and comparison is made with the corresponding experimental results to ascertain the appropriateness of the present method. Moreover, the fatigue lives predicted by other methods are compared with those by the present method, and fairly good agreement is recognized between the fatigue lives by the modified Langer equation which takes account of the strain concentration factor obtained from the elastic-plastic analysis and the experimental ones.

View full abstract

This is a fundamental analysis of periodic solutions of the free vibration of a two degree-of-freedom nonlinear spring-mass system. After transforming the differential equation under consideration into a canonical form, the relation between the initial values which yield periodic solutions and the period has been established. It has been proved that, by the perturbation method, the periodic solution and the period can actually be determined uniquely as powers of a small parameter. The condition that two masses vibrate in phase with each other has been ascertained. The calculation for a case with nonlinear terms of a cubic expression has been executed yielding the result that there holds the same relation between the amplitude and the period as in the case of a single degree-of-freedom nonlinear spring-mass system, if all of the three springs are either hard springs or soft ones.

View full abstract

When single-row deep groove ball bearings with so-called small "angular clearance" are used in a shaft rotating with an angular velocity ω, there appear rotating anisotropies both of shaft stiffness and of unsymmetrical nonlinearity in the spring characteristics of the shaft. These anisotropies are induced by irregularities in rotating parts of the shaft system and also rotate with an angular velocity ω. These spring characteristics result in that, in the neighborhood of the major critical speed, an unstable region where unstable vibrations of frequency ω build up takes place or disappears, and furthermore resonance curves of forced vibrations caused by the unbalance of the rotor vary with the change of the angular position of the unbalance. In this paper, these particular phenomena of lateral vibrations of the rotating shaft are treated both analytically and experimentally.

View full abstract

This paper presents an approach to finding optimum parameters of vibration isolators for systems subjected to stationary random force and/or foundation motion excitations. The vibration isolators to be determined consist of linear passive elements. Performance indices for the systems are expressed as nonlinear functions of the parameters of the vibration isolators. The optimum seeking method in nonlinear programming is applied to the determination of the optimum values of the parameters. The approach proposed in this paper is more versatile than the other approaches such as graphical technique using the trade-off curve. The procedure is illustrated on single- and two-degree-of-freedom systems with random excitations.

View full abstract

Transitional region of pipe flows of dilute polymer solutions is investigated. Experiment is made with HEC water solutions and the following results are obtained. Anomaly of the resistance coefficient is very small in the flow region before transition, though it is comparatively large for PEO solutions as reported in the previous paper. In this flow region a kind of fluctuations of static pressure is recognized and once the fluctuations occur, the appearance of issuing jets shows some anomalous winding effects. An approximate analysis is made on the anomaly of the resistance coefficient in the flow region before transition. The results show that the resistance coefficient becomes larger than that of Poiseuille flow and that if Reynolds numbers and the fluctuations are constant, the rate of increase of the resistance coefficient enlarges as Weissenberg number increases. Comparison between analytical results and experimental data shows an approximate agreement.

View full abstract

In this paper, in order to improve the measuring accuracy of this kind of flow meter, it is experimentally pointed out that the commonly used discharge coefficients of float-area-type flow meters do not always have constant values over a wide range of characteristic Reynolds number even in the region of Re larger than critical characteristic Reynolds number Rec. It is shown that only in the case of the floats having special shapes or geometrical dimensions the discharge coefficients of this kind of flow meters have constant values independing on the non-dimensional position of float. When employing such float, semi-empirical formula to predict the flow rate characteristics is obtained, and the limits of applicability of the formula are experimentally explained.

View full abstract

This paper presents a fundamental study on a bistable fluid amplifier load oscillator. To make preliminary clarification of the effect of the amplifier on the oscillating system, the flow characteristics of the amplifier is discussed in terms of solving steady two-dimensional Navier-Stokes equations by computer. And experiments are also carried out. The results obtained are summarized as follows : (1) Wall-attachment phenomenon of the jet which is essential to the operation of a bistable fluid amplifier may appear in laminar flow region. (2) To construct a mathematical model of the fluidic load oscillator, it would be reasonable to represent the amplifier's characteristics by the plots of the pressure difference versus the flow difference between both amplifier branches. (3) Using thus represented characteristics, it is easily explained that a bistable fluid amplifier can cause self-sustained oscillations in the system.

View full abstract

This paper presents the experimental results on bistable fluid amplifier characteristics and checks its representation which was introduced in the previous report. Then a mathematical model for a fluidic load oscillator is presented. And the analytical results derived from the model are compared with the experimental data. The results obtained are summarized as follows : (1) Experimental results show that the representation of the bistable fluid amplifier's characteristics given in the previous report is favorable. (2) The mathematical model well explains the features of the sustained oscillations occurring in the fluidic load oscillator. (3) The model also well explains the relation between the sustained oscillations in the system and the natural oscillations. Furthermore, the experimental data of the fluidic load oscillation are given for a wide Reynolds number range.

View full abstract

In analysing the heat transfer in vertical falling films it has been customary to apply the boundary layer theory in the single-phase turbulent flow. In this paper the limit of the applicability of this method is examined by reviewing papers on this method, and it is estimated that the Reynolds number must not be smaller than 800. Experiments on heat transfer were carried out with liquids of Prandtl numbers between 300 and 5500. By analysing the experimental data it is found that the equivalent eddy thermal diffusivity ε_H/υ is equal to 3 × 10^-5(y⁺)³ for the stable wavy flow, while the order of the first term of the Taylor expansion of the eddy viscosity for the single-phase boundary layer is larger by one ; in other words ε/υ is equal to (const.) x (y⁺)⁴. This difference indicates that, if the Reynolds number is small, the mechanism of the heat transfer in vertical falling liquid films differs largely from that in the flow Which has no free surface

View full abstract

Pollutants are transported by air stream and diffused by eddy flow. The turbulent diffusion equation describing these phenomena is common to the case of diffusion of automobile exhaust and the case of smoke diffusion from stacks. The Gaussian plume model is available for the case of primary pollution where no chemical reaction is included. The model is easily applied when the parameters of the model can be determined by the meteorological conditions. However the diffusion equation should be solved numerically if the air flow is not uniform or the secondary pollution is included. In this case adequate estimation of diffusion coefficients is rather difficult. The characteristics of air pollution caused by automobiles are introduced in comparison with those of smoke from stacks. The concentration of carbon monoxide in a street canyon at various instants of traffic signals cycle is described. The three dimensional solutions of the diffusion equation in a region of a busy cross section of roads are compared with field survey. CO concentration in San Jose city is also worked out and the results are compared with those of Stanford Research Institute. Finally the method of analysis of secondary pollution including the formation of photochemical smog is discussed. Diurnal variation as well as the isopleths of pollutants in Tokyo urban district is worked out.

View full abstract

In such cases as conveying bodies by use of motions of coupler links, a kinematic synthesis has to be made considering the coupler directions. In this paper, dimensional analysis and synthesis of a six-link Watt mechanism are carried out. The parameters of the mechanism are split into the ones which determine the shape of the coupler curve or the direction of the coupler and the ones which determine the position and magnitude of the curve. Moreover they are qualitatively classified into three kinds. By use of these parameters, an optimum synthesis of a Watt mechanism is carried out such as to satisfy every planar path with two directions of a body, without any trial and error.

View full abstract

The present investigation relates to a rotation-reciprocation mechanism and more particularly to a perfectly balanced, vibrationless, geared device or machine part, in which linear reciprocating motion is converted to uniform speed circular rotating motion or vice versa. In this report, fundamental principles both of generation of linear reciprocating motion by combining two uniform speed circular rotating motions and of perfect balance are theoretically analyzed. And theoretical equations on equivalent moment of inertia about crankshaft are obtained in order to ascertain that the vibration of this mechanism is not produced by polar moment of inertia. Fundamental forms of these mechanisms have been classified by constitutions and discussed. From this classification, crankshaft rotary motion system and crankshaft planetary motion system are obtained, and as an example, perfectly balanced theory of an internal geared device of crankshaft rotary motion system for the sake of simple constitutions is described.

View full abstract

This report relates to an eccentric geared device of crankshaft planetary motion system designed to satisfy two theories in the 1st report and a dynamic balancing machine with a vibrating table designed for this study. The construction of its device and constructional comparison between this device and known slider crank device, theoretical analysis on its motion and perfect balance, are described. From these results, it is made clear that the inertia force of reciprocating mass is perfectly balanced, and a moment due to its inertia force does not occur in this device. Further, vibration analysis on the vibrating table with a testing device in the dynamic balancing machine and its characteristics are described.

View full abstract

The authors have already synthesized the noncircular planetary gear mechanisms with nonuniform motion by using the noncircular gears for the planetary gear mechanisms. In this paper, the dynamic characteristics of those mechanisms are made clear by analysis and experiment. First, the torque fluctuation occurring in the driving axle of the mechanism in operating the mechanism is analysed and its equation is induced. Next, the mechanism is modeled to a vibration system and the equation of motion is induced and the dynamic displacement curve is calculated. Moreover, considering the mechanism as an equivalent moment of inertia, the equation of motion is induced and the value of the torque fluctuation is calculated. And then, by comparing and discussing the theoretical results With experimental results, the basis of design of the mechanisms is made clear.

View full abstract

The coefficient of friction between vulcanized natural rubbers or polyurethane elastomers and metals is measured under the pressure up to 900 kg/cm². The effects of several factors on the coefficient of friction are examined and an experimental formula is derived from the experiment in which threaded metal bars are used as a model of metal surfaces. The pressure, the mechanical property of rubber and the geometry of metal surface are considered in the above formula. Moreover, the validity of applying the formula obtained by using screws to the surfaces of rolled sheets or drawn bars with microscopic asperities is discussed. The coefficient of friction decreases with an increasing applied pressure, but approaches a definite value in the range of high pressures. The definite values for the case of using soapsuds and powdered rosin are about 1/5 and 5 times the value for no lubricant respectively.

View full abstract