A new method for obtaining the stress intensity factor is proposed. The constant strain element is used to approximate the integrated value of the stress with singularity in the neighborhood of the crack tip. Many numerical examples are investigated by the present method, and the results are compared with those obtained by other previous methods. It is shown that usual division of elements which are similar to those used in ordinary two-dimensional stress concentration problems can yield sufficiently accurate values of the stress intensity factors. Further, the applicability of the present method to dynamic problems with or withoutcrack propagation is carefully examined. The results have also turned out to be fairly good.

View full abstract

The principal object of this paper are to show (i) a general method of solution for torsional body force problems of an elastic thick infinite plate, (ii) Green's functions for the torsional body force problems mentioned above and (iii) a numerical method of solution for elastic problems of a circular solid finite length cylinder twisted by a pair of axisymmetric rigid chucks. Green's functions shown in the paper are defined as solutions to the elastic problem of an infinite thick plate subjected to torsional body forces acting on a circle in the interior of the plate. The numerical method for solid cylinders is an application of Green's functions shown in the paper. That is, solutions to the solid cylinder are constructed by Green's functions for torsional body force problems of the thick plate. Numerical examples for the problem of the solid cylinder are given.

View full abstract

Free vibration of a simply-supported rectangular plate having a straight notch simulating a through-crack in the plate is analyzed to obtain its eigenvalues and the dynamic stress concentration in the front of the notch. That is, the plate is divided into two parts along the notch and then, Fredholm integral equations of the first kind are derived for the internal moment and shearing force, using Green functions satisfying the boundary conditions of each part and continuity conditions of deflection and deflection angle of the original plate. The integral equations are transformed into the algebraic equations by the numerical integration and subdomain method, to calculate the eigenvalues and the moment and shearing force distributions. They are numerically calculated with regard to the lower four modes and the effects of the aspect ratio of the plate and the length and location of the notch on them are discussed in detail.

View full abstract

A coiled pipe, a helically bent pipe and a wavily bent pipe are used in practical engineering for many purposes. In this report, the relationships between hydraulic losses and flow patterns are investigated in regard to wavily bent pipes and quasi-coiled pipes which are constructed with many 90° elbows or 90°bends with small curvature ratios, R/r_w=l, 3 and 6, respectively. The results of wavy pipes are compared with those of quasi-coiled pipes. The reasons why wavily bent pipes are adequate for application to the heat exchanger are explained.

View full abstract

Motion of a horizontal viscous liquid layer under the action of the gravity covered by an elastic thin plate subjected to gas stream is studied. Also, the nonlinear terms in the governing equations for the fluid and the elastic plate are taken into account. The wavelength of a deformed elastic plate is assumed to be very large compared with the depth of the viscous liquid layer. An evolutional equation governing the disturbance of the elastic plate is obtained to an accuracy of the fourth order of small parameters, the wave number and the amplitude. The equation is solved analytically and the neutral stability curves are obtained . It is found that there is an equilibrium state in the linearly unstable region for the wave number. The amplitude of the elastic plate in the equilibrium state is obtained as a function of Reynolds number of the liquid layer, the bending stiffness, Young's modulus, the in-plane force of the plate, and the velocity of the gas stream. It is not a function of the initial disturbance.

View full abstract

It has been reported by many other authors that the vortex breakdown phenomena may occur in several distinct types. The aim of this study is to obtain a reasonable and consistent explanation for the various flow patterns of breakdowns. In the preceding report, we have pointed out that the breakdown phenomena depend greatly upon the occurrence of internal waves within a swirling flow field, and also that the type of breakdown so far reported corresponds to an azimuthal mode of the wave disturbance. The preceding paper, however, treated only steady(time-independent) types of breakdowns. The present paper concerns unsteady(or periodicwith time) types of breakdowns. It is found that the time-dependency of breakdown has a close relation to the phase velocity of the wave disturbance.

View full abstract

This paper presents characteristics of fluidborne noise generated by an axial piston pump. In this analysis, a flow ripple due to compressibility of fluid is considered, in addition to one due to pumping mechanism. The ripple due to compressibility is caused by an unsteady internal leakage which flows backward intermittently, at the neighbourhood of bottom dead center, from the delivery chamber into the cylinder chamber. Since a steady and continuous leakage of piston pump is generally very little, only a capacitant type impedance due to compressibility of fluid in the delivery chamber is considered as pump source impedance. Experimental values of pressure ripple almost agree, up to about 10th harmonic, with calculated values based on foregoing mathematical models for flow ripple and source impedance. It is concluded that the flow ripple of piston pump is mostly due to fluid compressibility, and that the well-known effect of pumping mechanism on the flow ripple can almost be ignored in the case of a relatively high operating pressure.

View full abstract

To clarify the cause of cycle-to-cycle variations of combustion in a spark ignited engine, two types of combustion, conventional and fast burn (2-plug ignition), and two types of fuel, gasoline and propane, were tested and analyzed by measuring the cylinder pressure. An electronic computer was utilized for these analyses. From these analyses, it was established that cycle -to-cycle variations of combustion in a spark ignited engine were dependent on cycle-to-cycle variation of the air-fuel ratio of the mixture ; and that the shorter the heat release delay between ignition and start of heat release, the lower the cycle-to cycle variations of combustion.

View full abstract

The cycle-to-cycle fluctuation of combustion in spark-ignition engines operated with a homogeneous lean-mixture of propane and air was investigated. The main results are as follows : (l) The configuration and size of flames in the early stage of combustion fluctuate remarkably from cycle to cycle, even if the mixture composition is kept constant. (2) A reduction of combustion period suppresses the cycle-to-cycle fluctuation of combustion. (3) The influence of the combustion fluctuation in the early stage on the whole combustion fluctuation varies with the length of the main combustion period.

View full abstract

Heat transfer and thermal performance of the concentrating solar collector of horizontal coaxial cylinders for a solar thermal power source are studied theoretically and experimentally. The design parameters considered in the theory are the concentration ratio, the diameters of a semi-transparent glass tube and a selective absorber tube, the effective length and the optical property of an absorber tube, the storage temperature, the pressure in the annulus, and the inlet temperature and the flow rate of a heat transfer medium. The thermal performance of the absorber system is evaluated by the outlet temperature of the fluid, the collector efficiency, the exergy efficiency and the final conversion efficiency. The experimental results confirm the calculation method of heat transfer and thermal performance.

View full abstract

The analytical expression and representation of the dynamic characteristics of the frequency responses are discussed. All frequency responses are expressed as superposition of the dynamic responses of the systems with one degree of freedom. First, a method is introduced to represent the natural frequencies and the equivalent stiffness of the undamped responses calculated numerically by the reduced impedance method. Next, the usual method of curve fitting is improved to represent the natural frequencies, the equivalent stiffness and the equivalent damping ratios of the damped responses determined experimentally. The least square technique is adopted on the methods proposed in the present report in order to minimize the errors which appear in representation of the response curve with the dynamic characteristics. The present methods are applied to several model structures, and usefulness of these methods is proved as the result.

View full abstract

The coupler motions without any rotations of single-loop overconstrained mechanisms with degrees of mobility I have been analyzed considering the figures described by the open chains composing the mechanisms. The statical characteristics of overconstrained mechanisms and general mechanisms have been compared considering the joint clearances, and the errors of the output disp1acements of the above mechanisms have been derived by modeling the joint configurations.

View full abstract

In this paper, the theory of the part 1 for the two-bearing rotor system is extended to a multi bearing rotor system by using the transfer matrix method. The extended theory is applied to the three-bearing rotor system and it is experimentalily proved. Specially the effect of bearing pressure on the stability of the rotor system is experimented for the above rotor system and compared with the theoretical result. Finally it is concluded from the above discussions that the proposed theory is useful for diagnosing the unstable rotor system and for improving the bearings.

View full abstract

A strip-ironing type friction testing machine, which can measure simultaneously normal and frictional forces acting on the die during the process, has been made on an experimental basis to assess an ability of lubricants to resist the welding at the tool-work interface in deformation processes. It can estimate effects of the bulk plastic deformation of metal ; i.e., an increasing ratio of the frictional surface area η and a sliding distance h on the welding which are important factors for the welding phenomena of the deformation process. To measure anti-weldability by this testing machine, paraffinic mineral oils have been tested with a commercially pure aluminum strip. The following conclusions were made. (1) The frictional coefficient μ and the average frictional shear stress can be expressed by the ratio of the welding area to the total contact area. (2) The ability to resist the welding of lubricants can be assessed by the contour lines of μ = constant on the η-h diagram.

View full abstract

Using a two disc machine, conditions to give rise to scuffing or to prevent scuffing were investigated with 0.45% carbon steel annealed which has much ability to be work-hardened. Following three factors were considered to be most important associated with the occurrence of scuffing, that is, bulk surface temperature, severity of contact, and surface resistance to scuffing. Three different critical conditions of surface temperature related to scuffing should be taken account of. The first one is the transition temperature, the second one is about 180°C at which the composition of oxide films formed on the disc surfaces seems to be changed, and the third one is about 240°C at which black surface films begin to be formed. The severity of contact could be represented by dimensionless film thickness. The surface strength could be expressed by the surface hardness taking account of work-hardening during testing, which is always accompanied by the formation of oxide film. The existence of critical surface strength according to operating conditions, above which there is little chance of scuffing, is confirmed.

View full abstract

Poor surface finish commonly found on the gears used in the ashobbed condition often leads to a low surface durability. This paper presents detailed investigations about the effect and the formation of a built-up edge(BUE) which is one of the important factors influencing the surface roughness by performing hobbing experiments and simulated-hobbing tests. Moreover, the relation between the BUE and the cutting thickness which is dominated by the hobbing conditions is investigated. From the simulation tests, it is found that definite relationships exist among the size of the BUE formed on the cutting edge, the size of the cutting thickness and the surface roughness. Based on the results, it is clearly confirmed that these relationships are applicable to hobbing also.

View full abstract

Involute gears have been widely used because of their many advantageous properties. However, involute gears have a drawback that their tooth contact strength is weak, because the convex tooth surfaces contact each other in meshing. In the present study, the authors designed an involute-cycloid composite tooth profile gear having higher bending and contact strengths than those of an involute gear, and confirmed experimentally the superiority of this gear.

View full abstract

In this report, tufftrided steel rollers having four kinds of hardened depths are tested under two kinds of relative radii of curvature using sliding rolling contact fatigue testing machines, and effects of tufftrided depth and relative radius of curvature on surface durability and surface failure of these rollers are experimentally made clear. These experimental results are discussed in terms of amplitude of ratio of stress to hardness, and it is confirmed from the calculation of this amplitude that the dominant mode in the surface failures of the tufftrided rollers is spalling and the surface durability tends to increase as the tufftrided depth increases and as the relative radius of curvature decreases.

View full abstract