Fatigue tests were carried out on mild steel plate specimens with various lengths of cracks. The relation between the alternating fatigue limit σ_w and the crack length was well expressed by following equation. R_p(c)w = αe(sec πσw/2σy - 1 ) = const. where, R_p(c)w is the cyclic plastic zone size at fatigue limit, αe is the equivalent crack length of an infinite plate, and σy is the yield strength of the material. The size of the persistent slip band zone on the fatigue limit is about one tenth of R_p(c)w and is nearly equal to the measured mean grain size of the material. R_p(c)w has close relations with the crack length corresponding to the fatigue limit of a plain specimen, the material constant which was already obtained in the criteria of fatigue crack initiation at notch roots, and the critical notch root radius at the branch point.

View full abstract

The effect of residual stress on the fatigue strength of machine members is investigated, with particular reference to an induction-hardened steel. In order to distinguish the effect of residual stress from the effect of changes in microstructure developed by hardening, the fatigue strengths of a solid specimen and a hollow specimen are compared. From the experimental results, the residual stress is found to have effect on the fatigue strength equivalent to that of the mean stress, if the relaxation of the residual stress is taken into account. The mechanism of relaxation of the residual stress under cyclic loadings is also investigated by measuring the stress-strain relation on the surface of an induction-hardened specimen. It is made clear that the relaxation due to plastic deformation in the core can occur, under some conditions determined by the hardness distribution, the residual distribution and the stress distribution due to the applies force.

View full abstract

The fatigue crack propagation by the cyclic stress below the fatigue limit in specimens with a pinhole is investigated after the crack initiation by the over-stress. The test results show that the critical condition for the start of crack propagation in a cracked specimen is indicated by σ³1 = 1.6 × 10³ ( (kg/mm²)³mm) as the relation between the crack length and the applied stress value. In some cases, a fatigue crack propagates under a cyclic stress below the fatigue limit and afterwards stops to propagates under a cyclic stress below the fatigue limit 1*__2 is greater in the case when the crack initiating stress is higher. When the cyclic stress over the fatigue limit is applies after a crack propagation has been stopped by the under-stress, the crack propagating rate is reduced in the first period of crack propagation.

View full abstract

The relations between the stress vector and the strain path were investigated for a complex loading to which the current plastic flow theory can not be applied. It was shown that by the use of the expression which related the stress vector to the strain path with a corner, the direction and the magnitude of the stress vector in loading along the strain path having various curvatures could be calculated by means of the Hereditary Integrals with respect to the length of strain path, and the trace of the strain vector in loading along the stress path with a corner could be determined by means of the graphic method. There was good agreement between the predicted results by the above mentioned method and the experimental results. The applicability of the current flow theory to non-proportional loading of brass was investigated on the basis of the calculated results of the stress vector for the strain path having various curvatures.

View full abstract

Thin-walled tubular specimens of annealed mild steel were prestressed plastically by tension or torsion, unloaded perfectly and reloaded proportionally by combined loadings of axial load and torque, as fundamental examples of investigating the plastic properties of the material used in received condition. The experimental results show that the shapes of equi-strain curves for the reloading on the deviatoric stress plane differ considerably from Mises' circle for smaller equivalent strains less than 0.5 %. This means that the isotropic and kineproperty of the material mentioned above. Then, a new expression of plastic stress-strain relation is proposed based on the general tensor equation between noncoaxial tensors. The plastic strain tensor for the reloading is expressed as a function of the tensors of the prestress and the reloading stress. The relation obtained approximates closely the experimental results.

View full abstract

Stationary solutions of motion of an accelerated unbalanced rotor with a linear oscillating system are introduced by an analysing method somewhat different from the methods used heretofor. By using the results, the stationary dynamic load torque generated by oscillations and the approximation expressions for calculating the stationary speed or rotation holding in the region far exceeding the critical speed, and for calculating the perturbation rates of the rotational speed are introduced. Furthermore, the transitional characteristics of oscillations and rotational speeds from the starting to the stationary states are calculated numerically by Runge-Kutta-Gill method. By comparison of the stationary characteristics obtained by the numerical calculation with the results analysed theoretically, it is shown that the theoretical solutions accord well with the results obtained by the numerical calculation. Furthermore, it is shown that the dynamic load torque is represented by a vector product of a centrifugal force vector and a displacement vector.

View full abstract

When the driving axles of an electric locomotive slip, self-excited vibration will occasionally occur in its trucks. To suppress such selfexcited vibration, detailed theoretical calculations of the vibrations during slippage of typical trucks were made and stability criteria were sought by simplifying the system. (1) Theoretical expressions were derived for the vibrations during slippage of typical trucks, that is, individual quill spring drive type trucks, and individual axle-hung motor type trucks, and numerical calculations regarding actual examples were made. The calculated and observed results generally agree with each other, revealing that the behavior of trucks can be predicted through theoretical calculations. (2) Stability criteria were sought for several problematical mode of vibration, that is, the torsional vibration of both individual and monomotor driving systems and the pitching of an axle-hung motor about the axle. The validity of the stability criteria was proved by comparing them with actual examples, and some instances of success in suppressing self-exited vibration were disclosed.

View full abstract

Making use of the mathematical model of a 2-axle wagon running on the track with irregularities described in the 1st report, the influences of the track irregularities and of the wagon parameters on the dynamics of the 2-axle wagon are investigated. An index is also proposed which might be employed as a measure of the magnitude of the influence of the track irregularities on the hunting motion of the 2-axle wagon.

View full abstract

This paper presents a numerical method of evaluating the density of probability distribution of a Bang-Bang control system excited by various types of pre-assigned signals in the presence of a Gaussian random noise. Namely, the Fokker-Plant-Kolomogorov equation for the probability density function may be solved as approximate solution of the given third boundary-value problem. For this purpose, the boundary conditions are applied along the switching surface in such a manner that the energy integral for the probability density function has a minimum value. Finally, the moments for the probability density function are calculated, and the convergency for the method of finite differences is examined.

View full abstract

Measurement of the pressure distribution on the two-dimensional rectangular cylinders are correlated with the characteristics of a smooth-wall turbulent boundary layer in which the cylinders are immersed. The contribution of the shear stress on the horizontal upper surface of the cylinders to the total drag is estimated to be a few per cent at most over the entire range of the width-to-height ratios tested. For flows with zero pressure gradient, correlations are obtained for the variation of pressure drag with the height of the cylinder h, which are analogous in form to the law of the wall for the boundary-layer velocity profile. The coefficients of the law of the wall for the drag are functions of simply the width-to-height ratio of the cylinder. This law of the wall is valid up to h/σ = 1.37, where σ is the thickness of the boundary layer at the location of the cylinders. The contribution of the base pressure to the pressure drag coefficient is found to be sensitive to the width-to-height ratio of the cylinders, while that of the pressure on the upstream face is almost constant over the entire range of the width-to-height ratios treated in this investigation.

View full abstract

As continuation of the foregoing paper, the present paper describes the flow patterns around the rectangular cylinders immersed in a turbulent boundary layer and the pressure distribution on the surface of the cylinders. The effect of three parameters, i.e. h/σ, u_τ/U_∞ and the width-to-height ratio b/h on the pressure distribution is examined in some detail, where b and h are respectively the width and height of the cylinder, σ is the boundary-layer thickness, u_τ is the shear velocity and U_∞ is the free- stream velocity outside the boundary layer. The main results will be summarized as follows : (a) Three sets of measurements of the mean flow and turbulence are presented for the width-to-height ratios of 2.0 and 4.0 with h/σ =1.13 and 0.554. (b) The mean velocity profile in a constant pressure half-jet theoretically obtained by Goertler. (c) The base pressure of the cylinders can be correlated with h in the form of a wall law. (d) The pressure distribution on the upstream face of the cylinders can be represented by a single curve which is irrespective of the width-to-height ratio and the boundary-layer characteristics.

View full abstract

Effects of " suction ring" and "separator", which are used to improve the off-design operations of axial-flow compressors, on the performance curves and stall conditions are studied experimentally. A suction ring has a large effect on the rotating stall characteristics. Application of suction ring with unsuitable geometrical dimensions results in undesirable rotating stall, and the decrease in discharge pressure becomes more remarkable than that of one with out suction ring. On the other hand, a separator has quite useful improving effects on the rotors in which rotating stall occurs. The most effective geometrical dimensions are shown in consideration of partial flow rate performance, stall condition and noise. In addition, the working mechanism of the separator is discussed.

View full abstract

Oscillation phenomena of an oil column in a conduit are analyzed. Both ends of the conduit are connected to a pulsating flow source (gear pump) and an elastically supported pilot valve. Results obtained are summarized as follows: (1) A self-excited oscillation is possible to occur even if the pump pulsation is ignored. (2) When the condition of self-excitation is fulfiled, an out standing oscillation arises in the system under the influence of the pump pulsation. (3) Both of harmonic and combination oscillations can occur depending on the operating condition of the system. (4) Analytical results based on a mathematical model of the system agree fairly well with experiments.

View full abstract

The statistical properties of the switching process of wall attachment amplifiers were studied in order to make the accuracy of the switching pressure clear. The switching process is classified into two modes : the initial switching and the random switching. The probability distribution functions of the switching delay were determined for both switching processes by considering the average deflection and the random fluctuation of the attached jet. A mathematical model was verified by experiments and the parameters in the model were determined. The relation or the input pressure vs. the switching probability for the input pulse with the finite duration is obtained by the mathematical model. A longer pulse yields a smaller pressure range of the uncertain switching. The theoretical estimation was verified by experiments.

View full abstract

For the purpose of describing spatial paths, two methods are suggested : one is realized by composing several planar mechanisms, and the other by using pairs with multiple degrees of freedom. In this paper, a spatial 5-link mechanism whose couplers are joined by a spherical pair is studied. The coupler motions of some triads composed of a fixed link, an arm, and a coupler are analyzed, and the relations between the shape of the triad and the position and moving direction of the coupler are derived. Then joining two couplers by a spherical pair, a mechanism which satisfies two precision positions, the directions of motion, and the path between them is synthesized. Moreover, it is experimentally proved that the mechanism has no change point and no intersection of links, and has high static and dynamic accuracy.

View full abstract

Spiral ropes are widely used as standing ropes in long bridges, suspension bridges or constructions with suspended ceilings. In these constructions, creep strain occurs inevitably. Also a relaxation of stress of PC strands ( a kind of spiral rope ) used in the prestressed concrete occurs. Therefore, to estimate the creep strain and the relaxation values of spiral ropes is one of the important problems in the design of bridges and prestressed concrete structures. The authors have carried out creep and relaxation tests of spiral ropes used as standing ropes and have calculated the creep strain and relaxation values by considering the creep strain of a single wire and by applying the analytical method, whose validity has been proven for the purpose of understanding the mechanical properties of ropes. Consequently, it has been made clear that the creep strain and the relaxation are influenced by the clearance between wire layers of the rope, and that they can be estimated by the analytical method presented in this paper.

View full abstract

The authors previously cleared from the experimental results which are derived from the four-ball test and the roller test that the condition or scoring could be anticipated by the surface temperature rise, and discussed how the thermal conductivity in the surface layer of material affects the surface temperature rise. In this report, the effects of the following factors on scoring are discussed ; the material and the hardness of gear, the viscosity of lubricant and the surface roughness of gear tooth. The results for scoring test are summarized as follows ; on both the normalized gears and the hardened gears, the progress of wear is especially remarkable in the zone of single pair of contact and the tendency of the increase of wear depth on the tooth surface well coincides with that of the weight loss due to wear. When the scoring failure of the hardened gear is compared with that of the normalized gear, the scar of the hardened gear tooth looks like a number of the micro fragments of the welding material gathered, and the size of the wear particles of the hardened gear is smaller than that of the normalized gear. As the thermal conductivity fall, the scoring limit decreases. Though the effect of the hardness on the scoring limit is larger than that of the thermal conductivity, the influence of the hardness of gear is inseparable from the scoring limit because the thermal conductivity and the hardness interact each other. Lastly, as the viscosity of lubricant grows higher, the scoring limit increases. In the lubricating condition of 140-Turbine oil, the index values of the flash surface temperature in AGMA Standard which were calculated by the measurement value of the surface roughness after running in, are within 161172°C on the carburized gear which consists of SCM22H material, and the mean value of the flash temperature is about 168°C.

View full abstract

The present test has been carried out for the purpose of investigating the change of stresses at the circumference of force-fitted surface caused by a load acting on tooth under running conditions. Momentary change of the stresses when a slip at the force-fitted surface between gear rim and hug occurs and when this slip is repeated three to four times, has been examined. Conclusions are that by a load that acts on tooth contact point, a partial decrease of internal pressure caused by force-fitting at the circumference of the inside diameter of gear rim occurs in the part corresponding to the meshing tooth and that the limit torque for the slip in a driving test of the force-fitting gear becomes much smaller than that in a slipping test of the force-fitting gear becomes much smaller than that in a slipping test under static condition.

View full abstract

The main objective of this paper is to discuss on the theoretical formula and practical equation for the fillet stress of the gear tooth in case when the hollow hub is force-fitting in the hollow hub, and also is to investigate the fillet stress in case when hollow hubs of various shapes are force-fitting. The authors measured the fillet stress of gear tooth by means of strain gauge and photo-elastic experiments for each force-fitted hub, and these experimental results were compared with calculated ones by the transform function and finite element method. Consequently the theoretical and practical formula could be introduced for the fillet stress of gear tooth with force-fitted hubs mentioned above, and it was found that calculated values agree fairly well with experimental results.

View full abstract

The principal object of this paper is to discuss how the fitting and extracting forces, slipping torque limit and the conditions of fitting surface vary with the repetition of force-fitting and extracting. Fitting and extracting forces of a cylinder with projections ( gear ), were compared with those of a cylinder without projecting, and the influence of size of the projections on fitting and extracting forces is examined, too. Moreover, the authors have introduced the most suitable conditions for force-fitting. From these experimental results, the authors came to the conclusion that, in the case of repetition of fitting and extracting processes, the increase of actual contract area of force-fitting surface and the decrease inner pressure are considered to be the most important factors influencing the change of fitting and extracting forces, and the influence of the former on the fitting, extracting forces and slipping torque limit is larger than that of the latter.

View full abstract