Fatigue crack propagation tests were conducted on butt-bonded two layered composite plates consisting of a low carbon steel and a ferritic or austenitic stainless steel under pulsating tension varying from zero to maximum, to study the effects of residual thermal stress and the difference in mechanical properties on the crack propagation and crack opening behaviors. It was observed that the propagation behavior of fatigue crack in the composite plate was rather complicated when the crack growth rate was plotted against K_max. And it was found that ΔK_eff was a much better parameter for controlling the crack propagation in the composite plates, though not completely sufficient. In discussions it was shown that the crack opening behavior in the composite plate was well estimated by considering an initial distribution of the residual stresses and the crack opening behavior of the monolithic plate, when the crack propagated in the field of compressive residual stress. But this was not the case, when the crack was in the field of tensile residual stress.

View full abstract

Corrosion fatigue crack growth rate of an austenitic stainless steel (SUS304) was measured in Cl^- containing water. Acceleration mechanisms of crack growth in corrosive environment were discussed, and an acceleration of da/dN in SUS3O4 was compared with that in a high-tension steel (HT55). The first step acceleration of da/dN in water environment from that in dry air was due to the reverse slip suppressing effect induced by water molecule adsorption. The second step acceleration at a low frequency (f=0.1Hz) was due to hydrogen embrittlement of plasticity-induced martensite formed at crack tips. In 3% NaCl solution at f=0.1Hz, the acceleration by stress-assisted dissolution was also observed. The crack opening ratio (U) was independent of Cl^- concentration as well as of cycle frequencies. The acceleration of da/dN in SUS304 amounted to 9.2 times that in HT55, and a water environment sensitive crack growth behavior under cyclic stresses was clear. However, there existed a threshold ΔK to commencing to accelerate da/dN from that in air, and the acceleration in HT55 exceeded that in SUS304 at f=5Hz near ΔK_th.

View full abstract

Rotating bending fatigue tests were performed for forged and cast steels of which specimen sizes are in the range from 8 to 40mm in diameter. The cast steel including originally many defects exhibits a decrease of about 27% in the fatigue limit with an increase in the size from 8 to 40mm, while such a decrease is about 13% in the forged steel of comparatively homogeneous substance. The measurement of the strain during tests manifests that a plastic strain of about 80×10^-6 is produced at the fatigue limit in the forged steel, independently of specimen size, and on the other hand, the cast steel yields no macroscopic plastic strain at the fatigue limit of this material. From these results, it is suggested that in the former the size dependent plastic strain, which is caused by the difference in the stress gradient produced in the cross section of the specimen, and in the latter the scatter in the fatigue strength induced by the defects, are the dominant factors controlling the size effect on the fatigue limit of steels.

View full abstract

A contact stress problem of an infinite plate with a rigid disk inserted in a circular hole is solved numerically. It is assumed that diameters of the disk and of the hole are identical in the unstressed state and that an in-plane load is applied to the disk. The method of stress analysis is a numerical one using recurrently the general form of a stress function expressed in the polar coordinates. Distributions of stresses and displacements around the hole are calculated and shown in some illustrated figures. The presented results are compared with those obtained analytically and experimentally by other authors. The problem of a large circular plate fixed at the edge with a concentric circular hole where a rigid disk is inserted is also solved in order to check the condition of support of the infinite plate.

View full abstract

In this paper, the authors analyze exactly the stress state of an elastic layer indented by an axisymmetric concave rigid punch when the indented load is small and the contact region is an annulus. Numerical examples are presented for parabolic and conical concave punches. For each case, we show the distributions of surface displacements and contact stresses and clarify the relationships among the parameters of the indented load, the indented depth, the punch face confuguration and the layer's thickness.

View full abstract

A dynamic substructuring method is presented for the efficient calculation of the Finite Element Method. The method is mainly composed of the Runge-Kutta-Gill method, the lumped mass matrix, and the substructuring stiffness matrices. To verify the validity of the method, it is applied to the longitudinal elastoimpact analysis of two-dimensional elastic rods having a uniform cross section and a variable one in collision with a rigid wall. The calculated values of the substructuring models agree precisely with those of the whole rod models. The use of this method makes it effective to decrease computer memory capacity and computing time. If the method is combined with the condensation method, the impact behavior of two-dimensional and/or three-dimensional structures with complicated shapes can be calculated more efficiently using smaller computer memory capacity and shorter computing time.

View full abstract

An experimental investigation was conducted to clarify the validity and the limitations of the analytic predictions of an unsteady pressure on a blade under periodic gusts. The unsteady pressure distributions around both cambered and symmetrical blades with angle of attack were measured. The results were compared with the theoretical ones considering the effects of angle of attack and camber. An attempt was made to predict the unsteady pressure due to an effect of blade-thickness theoretically. The theoretical predictions taking account of this effect agreed well with the experimental results.

View full abstract

Inviscid flow in an annular diffuser composed of a cylindrical casing and a conical hub was analyzed theoretically, and effects of the boundary layer thickness on the flow pattern and the performance of the diffuser were examined experimentally. When the inlet flow has a whirl component, theoretical results show that the axial velocity comes to increase near the hub side with a spread of the free vortex flow region independently on the inlet whirl condition. The agreements between the theory and the experiment are good except for effects of the boundary layer. On the other hand, the flow has tendency to run toward the hub side with an increasing displacement thickness of the boundary layer on the casing wall even if the flow has not a whirl component. As these phenomena are effective to prevent the separation when the convergence angle of the hub is rather large, the performance of the diffuser is improved by increasing the thickness on the casing wall and generating a mild whirl flow in the inlet cross section.

View full abstract

Pulsating flow through an orifice in a circular pipe is investigated by flow visualization method. Aluminium powder is suspended in a fluid. The movements of many particles are traced by taking a photograph exposed for a certain period of time. Various flow patterns are presented as the functions of Reynolds number, Strouhal number and time. Numerical solutions of Navier-Stokes equation are obtained and compared with the experimental results. It is confirmed that the calculated streamline patterns well coincide with the experimental ones. The behavior of wall shear stress is also investigated both numerically and experimentally considering the interaction between the shear stress and the flow pattern. It is made clear that the wall shear stress is very large closely behind the orifice, and that it is due to the motion of eddies (vortices).

View full abstract

In this experiment, in connection with the cavitation microjet, the behavior of a model microjet, produced by a water capsule, is photographed by a high-speed camera and its impact pressure is measured by a piezo-type pressure transducer. When the model microjet, whose diameter is 1.3 to 1.7 mm and impact velocity is 85 to 115 m/s, impacts directly against the solid wall, an impact pressure of 10 to 15 MPa is measured. The impact pressure, however, decreases very much, when a water layer exists on the solid wall and its thickness is greater than 1.5 mm. The measured time of the maximum impact pressure, detected on the solid wall, coinsides with our previous experimental results, obtained by the photoelastic method. Moreover it is shown that most of the test pieces of pure aluminum are damaged by the model microjet, whose diameter is 1.3 to 1.7 mm and velocity is 85 to 120 m/s, and that the damage pattern is clearly influenced by the shape of the microjet front.

View full abstract

This study reports on an oscillatory phenomenon of the vortex in a cyclone separator shaped vessel. In the vessel the primary flow field is such that the axial flow reverses and the average flow rate is zero in any section of the vessel. Under some conditions the vortex turns and attaches itself to the cylindrical wall of the vessel instead of being attached to the center of the end plate. The point of attachment rotates in a horizontal plane, and periodic fluctuations are produced in the flow. The unique behaviour of the vortex mentioned above is examined by detailed experiments. It is found that the configuration of the precessing vortex core is almost a plane curve, and that the frequency of the precession depends directly on the local strength of the primary swirling flow near the attachment position. A simple mathematical model is proposed to explain the qualitative properties of the phenomenon.

View full abstract

In order to detect the concentration of some material in a field, use is made of the light absorption method. The most importamt problem to be solved is the deviation from Lambert-Beer's law in the relation between photomultiplier output E^^∼_p and concentration value Γ^^∼ which appears under use of a broad band light source. An analysis is made of the deviation from Lambert-Beer's law, in consideration of various spectral characteristics of each optical element in the measuring system. A matching method is shown to make E^^∼_p - Γ^^∼ relation follow a simple power law over a broad range of the values of the concentration.

View full abstract

This paper concerns a piston pump without valves which consists of a piping system with a T-junction, two water tanks connected with both ends of the main pipe and a piston installed in the lateral pipe. Pump effect occurring in the system is examined theoretically and experimentally. The study is confined to the case that the time average flow in the main pipe is zero. The pump effect is a phenomenon that part of the energy corresponding to the time averages of kinetic energy in both of the main pipe is stored in one side of the tanks. Influences of system parameters on the pumping effect were clarified. The results obtained from a mathematical model proposed in this paper were found to be in good agreement with the experimental data.

View full abstract

The objective of investigation is to get a clear idea of the nature of wind penetration effects in clothing. The preceding study of wind penetration through fabric is extended, in this paper, to determine the heat transfer of fabric system. With the aid of heat transfer equation of porous media, the thermal insulation value of fabric layer can be described as a function of such arguments as α^*, some pertinent properties of fabric and wind velocity, where α^* is the characteristic number of contact layer referring to the manner of contact between fabric and cylinder. Quantitative agreement' between the theory and experimental data is considerably good. Then it is seen that the dynamic pressure of wind sets up penetration, and thereby augments heat transfer process. The insulation value decreases with an increasing wind velocity and the behaviour of the decreasing depends largely on α^*. Therefore α^* is an essential factor for assessing the value of a sample fabric in wind.

View full abstract

The maximum work which can be taken out of a system is entirely determined by a few parameters, and does not directly depend on the detail in the system. One of the parameters is negentropy in the system, which represents non-equilibrium in the system, and the others are concerned with the final equilibrium state of the system. Available energy (or exergy) is considered to be a limit of the maximum work which can be taken out of a system. Negentropy can be stored in a system in various forms, such as the form of the difference of densities, the form of mechanical energy, etc. Since negentropies in these forms can be converted equivalently from one to another, the quantity corresponding to one form, which can be obtained from the system in another form is easily calculated by equating the expressions for negentropies in these forms.

View full abstract

Experimental investigation was made on circumferential variations of the characteristics of a liquid film formed on the inner surface of a horizontal tube of 26.0 mm I.D. in which air and water flowed concurrently. Superficial velocities of the both phases ranged 10∼50 m/s for air and 0.006∼0.40 m/s for water. The flow pattern concerned was mainly an annular flow. Annular flow is subdivided into two regions characterized by liquid slug like frothy liquid lump (A_F-region) and disturbance wave (A_D-region). Classification can be made by examining the shape of the cumulative wave frequency distribution, plotted on normal probability paper against the radial distance from tube inner wall. A new method is proposed for separating base film from disturbance wave, which is processed on the wave frequency distribution chart. Experimental data are presented on principal parameters of liquid film such as thickness of base film, disturbance wave-height, -frequency and -velocity. A consideration is also made on the liquid transportation in disturbance wave region.

View full abstract

Uneven distribution of combustion gas in the cylinder occurs in some open chamber type diesel engines. This phenomenon is generated by an interference of fuel spray on the combustion chamber wall. So the concentration distribution of combustion gas can be used as a measure of interference. In this report, the mean relative burning rate is examined for possibility to express the combustion performance of fuel spray quantitatively. Then, the influence of spray impingement on the combustion performance is examined about the correlation between the concentration distribution of combustion gas and the mean relative burning rate. As a result, the following are revealed. An increase in the mean velocity of fuel spray causes the speed of mixture formation and combustion to increase through whole period, and the impingement of fuel spray suppresses the mixture formation in the injection period and promotes the combustion in the period after the end of injection.

View full abstract

Accurate switching times of manipulated vartables for a bang-bang control of a manipulator with two-degrees of freedom are easily calculated by approximating the velocity of a DC servomotor. Such bang-bang control is compared with a near-minimum-time control whose number of switching times of manipulated variables is two and the dynamic characteristics of the manipulator controlled by those methods are analyzed theoretically and also investigated experimentally.

View full abstract

The two-phase flow state that originates with the transient self-evaporation of hot water on the upstream side of a water level regulating valve, decreases the flow rate capacity of the valve, and accordingly causes the water level in equipment to rise. The purposee of our research was to establish an advanced control technique to prevent an excessive rise in the water level in equipment. To do this, we proposed a feedforward water level control technique using dynamic analyses with a computer model and confirm the applicability of this technique in experiments with a model test apparatus. One feature of the new technique is that valve openings are decided according to feedforward signals that indicate the degree of self-evaporation that has occurred, and then, valve openings are corrected by feedback signals that indicate a deviation in the water level in equipment. Test results of the new feedforward control system showed that rises in the water level could be reduced to thirty percent of those seen when conventional feedback control systems are used.

View full abstract

Since the glass-fiber-reinforced plastics (GFRP) is non-homogeneous, the cutting behavior of this material differs from that of a homogeneous material such as steel. It is, therefore, difficult to analogize the tool wear mechanism in GFRP cutting from the data obtained in the cutting of homogeneous materials. In this study, a face turning test was performed on GFRP cylindrical pipes which contained uni-direction fibers (right-hand-winding or left-hand-winding) and others. The cutting resistance and surface roughness were measured to analyze the machinability and the tool wear mechanism in GFRP cutting. Thus, the relation between the fiber orientation and tool wear becomes clear ; e.g., there is a significant difference in the wear rate between right-hand-wound material and left-hand-wound one.

View full abstract

The influence of geometrical contact conditions at the edge of a tool-material interface on the frictional shearing stress in massive forming was investigated. The contact conditions were classified into three types : a) the die-surface extends beyond the edge of the workpiece, Edge-E_I ; b) the workpiece-surface extends beyond the edge of the die, Edge-E_II ; and c) the die and the workpiece surface continuously contact into each other at the edge, Edge-E_III. Based on slip-line or upper-bound analysis of the deformation, characteristics of the deformation and the frictional slip near the edges were shown. A frictional testing apparatus for the plane strain forging which measures the normal and frictional forces operating on a referring frictional plane between edges was made. By using this apparatus, aluminum blanks were forged at slow speed : 1 mm/min. It was shown that the frictional shearing stress was critically affected by the contact conditions.

View full abstract

In the proposed precision die forging utilizing divided flow^(1)∼(3), the filling up of materials into die corner can be made more effective by suppressing an increase of the centripetal flow rate and by promoting an increase of the centrifugal one. In this research, the following is confirmed. When both the step height of a die and the size of the relief-hole diameter are chosen suitably, the working pressure required for a complete filling up becomes lower than that in the process utilizing the relief-hole principle without decreasing the product height. This suggests that a suitable restriction of the centripetal flow will be effective to improve the tool-life and save the material. The two step forging method proposed in the third report ⁽³⁾ is found to be a preferable one in the case of this flow restriction.

View full abstract