This paper considers a cylindrical crack in a transversely isotropic elastic body on the macroscopic level. which models a crack-like defect on an interface between fiber and matrix in the fiber reinforced composite materials. Numerical calculations are carried out for E.glass/epoxy and graphite/epoxy composites when they are subjected to a constant internal pressure or shear traction linearly distributing in the axial direction on the crack surfaces. Stress intensity factors are evaluated for different values of the crack-length and the elastic constants.

View full abstract

In numerical calculations of the Hankel Transform to solve axisymmetric elastic problems of semi-infinite bodies or infinite thick plates, Simpson's rule is usually adopted. But it is difficult to obtain accurate results by Simpson's rule because the Bessel function in the kernel of the Hankel transform oscillates. In this paper, f(x), a function to be transformed, xf(x) is approximated piecewise by the quadratic and the result of the Hankel transform is expressed by Lommel's functions. In this way accurate Hankel transforms are obtained. Two examples are illustrated to show the availability of this method.

View full abstract

The states of stress distribution in mitred bend subjected to in-plane uniform bending moment have been studied systematically by means of photoelastic stress freezing method. The relations between the stress concentration factor of fiber stress σ_l and of hoop stress σ_θ near the bent art and the angle of mitred bend are thoroughly investigated. The effects of fillet radius of the bent-part and wall thickness on the stress concentration factors are also discussed.

View full abstract

Fracture under biaxial tension is one of important problems in the engineering design. Although there has been a great deal of reasearch made on this problem, it is difficult to say which procedure is better to solve the problem. With an aid of the photoelasticity method. successfully the authors determined plastic stress concentration factors concerning a circular hole under equal biaxial tension. And these were in good agreement with theoretical values.

View full abstract

The growth and collapse of cavitation bubbles are observed in distilled water using a hydro shock tube which alternately generates low The cavitation bubbles pressure and high pressure in the test section. are fragmentized into numerous small bubbles (so called nuclei) after their collapse. The number of remaining nuclei increases rapidly as the gas content increases in the water. The nuclei population varies In the water during the growth and collapse of cavitation bubbles, and finally it becomes comparable with the one measured in some cavitation tunnels. The cluster of cavitation bubbles, grown from the remaining nuclei, behaves like a single bubble in the initial stage, but at the middle stage it separates into small bubbles, and they collapse faster.

View full abstract

The knowledge of the behavior of a cavitation bubble in a narrow space, such as a cylinder liner, is still very scanty. In this study, the behavior of a spark-induced bubble between parallel walls is photographed by a high-speed camera, and the generated impulsive pressure waves are detected by piezo-type transducers. The experiment is carried out by changing the distance between the walls and the location of the bubble formation, and it is made clear that when the non-dimensional wall distance is small, the collapsing bubble becomes "gourd" shaped and at the final stage of the collapse, the bubble is broken down into two small bubbles, and their sizes depend on the initial bubble location. Moreover, such behavior of the bubble is simulated by the numerical method of Plesset-Chapman.

View full abstract

An aerodynamic approach to an optimum compressor cascade in a two-dimensional incompressible flow is described on the basis of two inverse problems ; namely, an inverse boundary layer problem in order to produce the optimum surface velocity distribution and an inverse cascade problem in order to obtain the blade section which realizes a prescribed velocity distribution. A new method for the former problem is developed in this paper; namely, the momentum equation is integrated for a prescribed shape factor distribution in order to get an unseparated turbulent flow approaching the equilibrium state. An inverse cascade design method is introduced by combining the thus obtained optimum surface velocity distribution with the inverse cascade method proposed in the previous report. Highly loaded compressor cascade designed by this method was tested in a low speed cascade wind-tunnel. Reasonable agreement between the theory and the experiments is obtained and the maximum lift-drag ratio is about 1.4 times as high as that of corresponding NACA 65 series cascade at design operation.

View full abstract

It is important to estimate the leakage flow rate through a rotor clearance of positive-displacement type fluid machine and flowmeter. In order to :make clear the flow through the top clearance between the lobe rotor and the easing wall not only in laminar regime but also in turbulent regime, experiments are carried out by using a model passage composed of a rotating cylinder and a stationary straight wall. In the laminar regime, the leakage flow rate through the top clearance is estimated theoretically. When the cylinder is at rest, the transition from laminar to turbulent flows occurs at Reynolds number Re=150 based on the minimum clearance and the mean velocity in it. The effect of the rotating direction and speed on the critical Reynolds number is also discussed. In the turbulent regime, the empirical equation for the leakage flow rate is obtained.

View full abstract

Vibration of a pseudo-shock in a duct often causes noise, vibration and breakdown of the duct. In the present study the wall static pressure fluctuation was measured along the duct and statistically analyzed to make clear the mechanism of the pseudo-shock vibration. The result showed that the vibration had two main frequencies, high and low ones. The former was the natural frequency of the air column in the divergent passage attached downstream of the duct, and the pressure fluctuation propagated in the upstream direction. The latter was the natural frequency of the air column in the duct downstream of the pseudo-shock and this disturbance propagated in the downstream direction. Each vibration was maintained by the interaction of themselves.

View full abstract

In the 1st report pressure fluctuation at each point in the duct was measured and analyzed. In the present study pressure fluctuations at two points along the duct were simultaneously measured and correlated statistically in the forms of correlation factors and coherences. Thereby the propagation of the pressure fluctuation was investigated. The variation of a pseudo-shock had two main frequencies, high and low ones. The former, the vibration of the air column in the divergent passage propagated upstream as a pressure wave, while the latter propagated covectionally with the main flow.

View full abstract

Oscillatory air flows in ducts of rectangular cross-section with aspect ratios 1, 5 and 10 are investigated experimentally. On the basis of the inspections of hot-wire signals, flow types can be classified into laminar, transitional and turbulent regimes, and these regimes are well delimitated by using two parameters, Re_os=D_h|u_m, os, 1|/ν and √(ω')=(D_h/2) √(ω/ν), where D_h : hydraulic diameter, |u_m, os, 1| : cross-sectional mean velocity amplitude, ν : kinematic viscosity, ω : angular frequency. For fully developed steady turbulent flows in a square and a two-dimensional duct, the approximate solutions for the axial mean velocity profiles are derived on the assumption of eddy viscosity distributions. In the phase when turbulent bursts occur, instantaneous velocity profiles of measured oscillatory duet flows are well described by the above-mentioned solutions.

View full abstract

Self-flashing two-phase flow caused erosion damage to the piping just downstream from the constricted parts such as valves, orifices, etc. of the hot water piping. The cause and allowable value of erosion, caused at the diffuser behind valve, were experimentally analyzed using a half-cut diffuser in order to make the observation of inner surface easy. As the result, the intensity of the erosion was affected by the two-phase flow pattern and the divergent angle of the diffuser. The erosion was mainly caused by an eddy flow generated at the diffuser. Allowable values of erosion factors, such as averaged velocity, averaged kinetic pressure and cavitation coefficient, depended on the divergent angle of the diffuser.

View full abstract

Experimental and theoretical studies were made to gain a deeper understanding of the radiative properties of nonequilibrium argon plasma flows in a circular tube. The self-absorption effects were taken into account as rigorously as possible. Experimentally, the radial profiles of the population densities of argon atoms at the excited 4s, 4p, 5p, and 5d levels were obtained from the lateral distributions of the absolute intensities of ArI spectral lines originating from these levels. On the other hand, theoretical profiles of the population densities for the same levels were calculated based on the optically thick model for collisional and radiative processes proposed by Bates et al. and experimentally measured atom temperature, electron temperature. electron density and gas pressure. Comparison of the experimental and theoretical results showed a reasonably good agreement and the importance of the self-absorption effects.

View full abstract

Simple approximate equations were obtained with high accuracy for the periodic steady outlet water temperature amplitude and the efficiency of flat plate solar collectors which were subjected to a comparatively short period of incident solar energy fluctuation. These equations were based on Taylor's expansion, concerning eleven non-dimensional parameters which composed the basic equations shown in the first report, around a typical flat plate solar collector specifications. Furthermore, considering the effects of each parameter, those approximate equations were simplified. In an ordinary case, the above two quantities could be obtained easily by summing about ten terms, respectively, with sufficient accuracies.

View full abstract

A phenomenological treatment is given for steady magnetic fluid flows in annular pipes in the presence of uniform external magnetic fields. It is found that the apparent viscosity of a dilute colloidal suspension of ferromagnetic particles is a marked function of both the magnitude and direction of an externally applied magnetic field. The experimental fact shown by McTague(1969) that the additional viscosity is greater by a factor of 2 for the field parallel to the flow than for the field perpendicular to the flow is theoretically explained. Analytical expressions are derived for the distributions of velocities and angular velocities of the magnetic fluid in annular pipes under the influence of the external field. The induced magnetization calculated here to the first order is skewsymmetric with respect to the plane through the center line of the pipe and perpendicular to the applied magnetic field.

View full abstract

A fully elliptic numerical scheme based on the pressure correction procedure has been successfully employed for the prediction of a three-dimensional developing turbulent flow in a square duct. The calculations have been performed using a two-equation turbulence model coupled to algebraic stress model for the secondary flow of the second kind developed by Launder and Ying. In order to reveal full features of this particular stress model. the calculation results are compared extensively with available experimental data with strong emphasis on the local structures of turbulence responsible for the secondary flow generation.

View full abstract

High-speed type cross flow turbine was proposed by the present authors as one of the most desirable water turbine to utilize low head water energy resources and the results of the theoretical analysis of its performance characteristics were presented in the previous paper. This paper compliments the authors' proposition by the experiments that confirms the validity of the theory and by the extension of the theory examining the effects of dynamic stall and of draft-tube. The measurement of the rotor performance was carried out in air stream in a straight duct. Although the fragile structure of the rotor did not allow to conduct the measurement at sufficiently high Reynolds number, the obtained data were enough to support the results of the theory. The influence of dynamic stall of the rotor blades which inevitably takes place at designing tip speed ratio was reasonably taken into account. Furthermore, the effect of draft-tube which is indispensable for low head water turbine was also clarified.

View full abstract

This paper describes a flow past a circular cylinder whose one end is set on a flat plate and the other free end is immersed in a shear flow. The surface pressure distribution on the cylinder was measured and drag coefficients were determined by integration of the result. Experimental results for a shear flow were compared with those for a uniform stream. The main results are as follows : (1) The effect of the free end is limited to within 3d from the top of the cylinder. (2) The local drag coefficient divided by local dynamic pressure increases from the neighborhood of the top to the bottom of the cylinder as the length to diameter ratio and the velocity gradient increase. (3) In the wake of the cylinder, the velocity of the center portion of the wake, which is accelerated by the downwash, is larger than the local mean velocity.

View full abstract

The vortex theory of screw propellers is modified and applied to the high-speed horizontal-axis wind turbines. The turbine blade can be replaced by a bound vortex system, and when the energy loss per unit time is a minimum, the trailing vortices move backwards with a constant velocity and build a helical vortex sheet. A velocity potential function which represents the vortex system is obtained. The circulation distribution along the blade is calculated iteratively until the hydrodynamic pitch of helical vortices converges. This method is applied to two examples, one with a sectional shape of modified Clark-Y type and the other with a sectional shape of NACA 65-series type. Torque, power and resistance are computed for a turbine which has two to six blades, respectively. Numerical values agree well with the sectional shape of experimental ones and it is concluded that the vortex theory is applicable to high-speed horizontal-axis wind turbines.

View full abstract

The static and dynamic properties of a pseudo-shock in a radial supersonic flow with a swirl between parallel plates were experimentally investigated. The length and pressure recovery of the pseudo-shock were smaller than those in a straight channel. Pressure recovery decreased with a growth of the swirl. Pressure fluctuation at two different points along the radius was statistically correlated, so that the mechanism of propagation of a pressure fluctuation could be analyzed. There were two main fluctuations with high and low frequencies, each of which propagated in opposite direction. The former was excited by an oscillation of the air column in the exhaust chamber and propagated upstream as a pressure wave, while the latter was due to an oscillation of the first shock and was convected downstream.

View full abstract

Stable hydraulic pressure pulse series can be generated by an appropriate valve operation at a pipe end. The pressure of the pulse series is amplified by a booster piston, consequently the pressure pulse series with a pressure higher than 400 MPa is obtained for a frequency of 15.2 Hz. By a pressure intensifier applying the amplified pressure pulse series, a continuous oil flow with an amplified pressure higher than 300 MPa is obtained for a source pressure of 6.6 MPa, Using this pressure pulse series generator as a lateral pressure cutting device of a ductile material, an aluminum column of 20 mm in diameter with a notch is cut after 580 cycles of the repeating pressure of 204 MPa. Hydraulic pressure pulse series generator discussed in this paper is practical enough to be used as a pressure intensifier and a lateral pressure cutting device.

View full abstract

Small design changes are usually repeated at design phase of structures, and it appears to be of practical significance to develop effective numerical methods for analysis of free vibration problems of structures with small perturbations. This paper presents a numerical method for perturbation analysis of matrix eigenvalue problems, where the eigenpairs necessary to be considered are assumed to be non-degenerate in unperturbed states. The proposed method is based on an iteration scheme combined with the use of generalized inverses of singular matrices. The scheme is designed to be easily implemented on large-scale finite element programs. To see effectiveness and fundamental properties of the proposed method, some numerical results are given for lateral vibrations of cantilever beams and circular plates.

View full abstract

The vibration of a diesel generator package is analyzed by the substructure synthesis method. namely the building block approach. The diesel generator package is divided into three substructures, namely a generator, a common plate and a diesel engine. The transfer function of each substructure is measured by a vibration test under the free condition. The transfer functions of all substructures are combined to make an equation of motion for the total structure. This equation is solved, and natural modes. natural frequencies and dynamic responses are obtained. The connecting parts between three substructures which are composed of bolts and supporting metals are considered springs of one degree of freedom. The dynamic characteristics of rubber isolators placed between the common plate and the base are obtained experimentally, and are introduced into the analysis. The analyses are performed for both a model and an actual machine, and the results are compared with experimental ones.

View full abstract

This paper analyzes the steady state response of a single-degree-of-freedom system with clearance and triangular hysteresis loop characteristics which are described by a symmetrical piecewise-linear model. Analytical methods of exact and approximate solutions are introduced for the system, and the resonance curves are obtained for the amplitude versus the frequency ratios with respect to several fixed exciting ratios and system characteristics by those methods utilizing a digital computer. Some numerical results obtained from exact solutions are compared with those from approximate solutions, and these theoretical results are confirmed by analog computer solutions. It is concluded that the approximate solutions are more practical than the exact solutions for a system with low ratios of nonlinear parameters.

View full abstract

This study is concerned with a newly developed facility for controlling noise propagation and thereby producing a wide region of reduced sound power, which is expected to be effective to alleviate railroad noise, highway noise, machine noise etc.. The mechanism of noise reduction by this device is destructive sound wave interference between refractive sound waves propagating through multiple passages of different lengths eqipped with the device and direct propagation sound waves. In this paper, the structure and principle of the facility are described and the technical data for practical application are collected.

View full abstract

A passive magnetic radial bearing with an eight-pole stator is statically stable. But its inherent damping characteristics are not sufficient to prevent dynamic instability. The dynamic stability of the magnetic bearing is commonly attained by introducing the friction of a mechanical viscous fluid. Though the static performance in this type of magnitic bearing has been theoretically analyzed by some investigators. its dynamic stability has not been fully determined. This study presents an analysis of the dynamic characteristics of a passive magnetic radial bearing with an eight-pole stator. In the analysis the perturbation method has been used to obtain a transfer function for the magnetic bearing system. Thus the dynamically as well as statically stable region has been calculated. The analytical results are confirmed by experimental observation.

View full abstract

This paper deals with distributions of the contact stresses when a T-flange is fastened with tap bolts to a base which is large enough compared with the T-flange. In an analysis, by replacing a T-flange and a base with two finite strips respectively, the distribution of contact stresses is analyzed as a two body contact problem using two dimensional theory of elasticity. In order to know the distribution of contact stresses near a bolt hole, the contact stress between two hollow cylinders is analyzed using three dimensional theory of elasticity. In addition. the force ratio, that is the ratio of an increment of a bolt axial force to a load when the load is applied, is analyzed. Moreover, taking a bending moment caused in a bolt into consideration the maximum stress is also analyzed. For verification of analyses, experiments are carried out and it is shown that analytical results are in a fairly good agreement with experimental ones.

View full abstract

Equations for designing Novikov gear pairs with 3 - 5 pinion teeth for high gear ratios were obtained and they were numerically analysed using an electronic computer. The main calculated results were shown in the for:n of figures. For testing of both the tooth profiles and the running performance, Novikov gear pairs with Z₁=3, Z₂=27, a helix angle of about 22° and a center distance of 84.32 mm were made using a hobbing machine and two kinds of fly tools. The proposed contact markings were obtained at the designed center distance. When the trial-made gear pairs were tested at a pinion speed of 2000 rpm, a power transmission efficiency of about 99% was obtained at tangential loads in the range of 1500 to 3000 N. This efficiency is higher by about 4% than that of gear pairs with involute profiles with much the same size as that of the Novikov gears.

View full abstract

In the eariler report, it was investigated basically how the surface roughness of a steel cylinder influenced the abrasion of a nylon cylinder. In this report, steel gears were ground to several degrees of roughness by O° teeth ground method of MAAG. And it has been investigated through the experiment how these roughnesses of teeth of steel gears influence the abrasion of nylon gears. These results were compared with the abrasion of a nylon gear, which meshed with a hobbed steel gear. In the intermeshing with a power transmission nylon gear, it has been found that a steel gear whose teeth have been ground, is useful.

View full abstract

The externally-pressurized gas journal bearings with circular slot restrictors are investigated theoretically and experimentally to determine their dynamic properties. The theoretical analyses are presented for a single-row admission bearing and a double-row admission one. respectively, by using Reynolds equations in which inertial forces of a gas film are considered. The accuracy of the theoretical method is consequently shown that the present theoretical method can accurately predict the dynamic properties of a gas journal bearing with a circular slot restrictor.

View full abstract

A system program of curve-fitting for general purpose is worked out to analyze the modal parameters from a transfer function obtained by a vibration test and to reconstruct the frequency curve of the transfer function. Both the frequency domain and the time domain curve-fittings are possible under the assumption of three kinds of dampings. namely pro port i proportional viscous. gene general viscous and hysteretic dampings. A transfer function of a mass-spring system of three degrees of freedom is fitted by six methods using this program, and calculated results of modal parameters are compared with rigorous solutions. Next. a transfer function of a steel plate obtained by the vibration test is fitted using this program. and the result is compared with the experimental one.

View full abstract

Generation method of appropriate pulse trains which should be utilized as alternative excitation inputs for the structural response analysis in terms of various types of random excitation is presented. Criteria for getting optimum pulse trains are established based on the evaluation of the structural response spectra. Fundamental consideration are briefly made concerning the comparison of the response properties through random motions with those through pulse trains taking a single-degree-of-freedom systems as a structure model. Results obtained fairly well correspond to the experimental results of a shaking test. Shape and number of the pulse trains can be optimally determined by the calculation of the difference of the response spectra among both input cases within the periods including main natural periods of structures. Optimization can be successfully carried out by taking artificially simulated narrow-band excitations and nonstationary broad-band excitations like seismic motions as inputs for numerical examples.

View full abstract