This paper is concerned with the analytical and experimental studies of a cylindrical shell cut by a spiral groove, the so-called flexible shaft. The analysis of the flexible shaft under a uniform bending is carried out by using the theory of a cylindrical shell derived by one of the authors. The relations between the angle of the spiral groove, thickness of shell, radius of cylinder and width of band element, and the flexibility, displacement and stresses of the flexible shaft are made clear. The discussion on the flexibility of the shaft with or without groove suggests how sensibly the groove affects the flexibility of the shaft. Some experimental data are presented for the displacement and strain, and the comparison between them assures the validity of theoretical analysis.

View full abstract

The translation rule in the so-called kinematic hardening model with respect to elasto-plastic constitutive relations of metal is investigated. The representative translation rules of Prager's and Ziegler's type and of the stress increment type and of the stress increment type which was proposed by the author in the past are pointed out to have some limits in their applicabilities with respect to an accuracy of analysis. A new translation rule of an integral type with respect to the pre-loading paths is proposed. The newly-proposed translation rule was applied to the ratcheting deformation analysis with a yield function of Mises' type. There is a good agreement between the calculated results and the experimental ones. It is confirmed that by using the new translation rule, some complex deformation analysis like the ratcheting deformation analysis can be simply and precisely performed.

View full abstract

This paper contains a three-dimensional solution for the stresses in a semi-infinite body having a hemi-oblate spheroidal pit under all-around tension parallel to the free surface. In the analysis, Boussinesq's two harmonic stress functions approach is used. The solution is presented by three sets of oblate spheroidal harmonics. These are chosen so as to satisfy the boundary conditions on the free surface and at infinity automatically. The remaining boundary conditions on the surface of the pit are satisfied with the aid of half-range expansions of Legendre functions. Numerical calculations are carried out for four shape ratios of the oblate spheroidal pit.

View full abstract

The authors start with an analytical study on the propagation of plane harmonic waves in an infinitely extended, cubi-anisotropic, elastic, and micropolar medium. It is possible to have six fundamental waves. Of these, a longitudinal uncoupled displacement wave and two coupled transverse waves of displacement and microrotation which become uncoupled with become uncoupled displacement waves at the limit of long wavelength are compared with the acoustical waves in the classical theory. The micropolar elastic constants of diamond, as well as its local moment of inertia, are estimated under the requirement that the theoretical dispersion curves for these waves fit the experimental data. Numerical computations are carried out and the results are made into a table.

View full abstract

A finite element method was developed for analysing steady two-dimensional inviscid subsonic flows through an arbitrary shaped cascade on an arbitrary revolutional surface with a varying channel height. The flows were assumed to be homentropic and homorothalpic. In the method Poisson's equation for stream function is calculated repeatedly by minimization of the functional using linear triangular elements. In order to in crease the accuracy and save the computer time, the treatments of the periodic condition and the kutta condition were improved reasonably and the calculation of velocity was performed by the least square method. As numerical examples, flows through a compressor rotor cascade on a plane were computed and the streamlines, the Mach number contours, pressure contours, and the pressure distributions on the blade surface were obtained. These results agreed well with the results of experiment and the method of streamlines.

View full abstract

An experimental investigation is carried out on a flow around a circular cylinder vertically mounted on a smooth plane wall along which a turbulent boundary layer is fully developed. Pressure drag of the cylinder is correlated with the height h, the diameter d of the cylinder and the characteristics of the boundary layer. The pressure drag coefficient C_D is found to be expressed in the form C_D=a(h/δ)^b, where δ is the thickness of the boundary layer and the values of a and b are dependent on h/δ. when h/δ&lsin; 1. 1, a and b are linear functions of d/δ, while they are constant in the range h/δ> 1. 1.

View full abstract

An experimental investigation has been carried out to study acoustic interferences in vortex-shedding from a circular cylinder in a crossflow subjected to the acoustical vibrations directed along the axis vortex-shedding along the cylinder axis. The results indicate that the effects of the acoustical vibrations directed along the axis of the cylinder are similar to the effects of the cylinder on the vortex-shedding from the cylinder, and both vibrations increase the spanwise correlation of the cylinder wake. The acoustic frequencies which produce powerful effects on vortex-shedding correspond to the frequencies of laminar-turbulent transition wave in a separated shear layer, and there are critical sound pressure levels in these acoustical vibrations.

View full abstract

The hydrodynamic stability of a liquid layer down the inner and outer surfaces of a vertical circular cylinder rotating about its axis is investigated taking exactly the curvature of a cylinder into account. In the case of long waves, the stability characteristics of this flow is clarified from the neutral stability curves and the phase velocity. The rotation of a cylinder stabilizes the flow of a liquid layer down the inner wall of a cylinder, while the curvature of a cylinder destabilizes it. For the flow down the inner wall of a cylinder, there exists a critical Reynolds number corresponding to an angular velocity of a cylinder. In the case of small Reynolds numbers, the neutral stability curves do not depend on the rotation of a cylinder but depend on its curvature.

View full abstract

Approximate equations for a pulsating turbulent flow in a circular tube containing a slightly compressible fluid are numerically calculated by the method of characteristics and finite difference. Axial distributions of pressures and velocities along the pipe axis and a cross-sectional distribution of velocities are computed by using a time-dependent friction velocity or time-independent one. Pressure and velocity distributions calculated at various frequencies are shown schematically, and those calculated with a time-dependent friction velocity agree well with experimental ones.

View full abstract

Friction factors in a pulsating turbulent pipe flow are estimated from the flow pattern diagrams shown in a previous paper and the effect of frequency and time-averaged Reynolds number Re+a on the frictional losses is verified with experimental data. The instantaneous friction factor λ_u, ta may be generally expressed as functions of ω'/Re³/4_ta, where ω' is a dimensionless frequency. Here, λ/u(t) is almost equal to the quasi-steady friction factor λ_g in the quasi-steady region, but in the intermediate and the inertia dominant regions, λ_u, ta increases with an increase in both ω'/Re³/4_ta and velocity amplitude ratio A₁. In addition, more adequate values indicating the limits between the above mentioned three regions are proposed.

View full abstract

By introducing four characteristic parameters defined for a pulsating turbulent pipe flow, the pattern diagrams for a pulsating laminar pipe flow are shown. The flow patterns are classified into three types, i. e., quasi-steady, intermediate, and inertia dominant ones with respect to the dimensionless frequency ω' and the limits between these three regions are determined. An alternative analytical representation for wall shear stress τ_w including acceleration term and its approximation are proposed. Various approximate representations for τ_w reported previously or proposed here are compared with the analytical one by means of the four characteristic parameters and their applicability is examined.

View full abstract

A study has been made of the attachment of a radial jet flow discharged from an inclined nozzle on an adjacent offset disc plate. Mean velocity, turbulent velocity and static pressure distributions were measured. The effects of the angle of inclined nozzle to the disc plate, and also of the distance of nozzle center from the offset plate on flow pattern were investigated. Length and average negative pressure in the region of recirculating flow were examined. The results were compared with those obtained for a two-dimensional attaching jet flow to an inclined flat plate. Changes of maximum jet center velocity and jet half width of the flows before and after an attaching point along the flow were measured, and the results were compared with those obtained for a radial free jet flow or a radial wall jet flow, respectively.

View full abstract

A pulsating flow of liquid metal in a circular pipe under a transverse magnetic field is analysed numerically and experimentally. A numerical analysis is carried out under the assumpution of a parallel flow to the axis of pipe and the effects of Hartmann, Womersley, and magnetic Prandtl numbers on the velocity, the induced magnetic field, and others are made clear. In an experiment, the sectional mean velocity, peripheral mean shearing stress at the wall, and their phase-lags from the wave of pressure gradient are measured and compared with the results of numerical analysis.

View full abstract

The behaviour of the Maxwell fluid in the inlet region of a channel composed of flat plates is revealed by solving the governing equations numerically. Two cases are studied, one with the boundary condition of a uniform flow far upstream of the entrance and the other with the condition of a uniform flow at the entrance. It is shown that a negative vorticity and a distinct flow pattern appear near the leading edge of the plate caused by a steep gradient of a normal stress for the former case. The numerical results also indicate that the profiles of kinematic variables in cross sections have inflection points and extreme values, that the region where the internal energy of fluid elements decreases exists near the leading edge and that the vorticity distribution along the wall gas extreme values.

View full abstract

The distribution of rotational speed fluctuation rates on the performance map is investigated by a model test. The random fluctuation is small in the range of high speed ratios 0.90∼1.0, and lower speed ratios under the small charging quantity of the working fluid. The domain of large random fluctuation exists in the intermediate range. The periodic fluctuations occur in a narrow region where the output torque rises with an increase of speed ratio. Flow-visualizations with stroboscope, oil-film method and tuft method are performed to study the relation between the internal flow and the rotational speed fluctuation. A vortex perpendicular to the pressure side surface of turbine blade is observed, and the flow from the vortex is found to be unstable when the random fluctuation is large. Furthermore, the state of free surface and air-bubble flow is examined.

View full abstract

This paper presents the experimental results and analyses on the vibratory characteristics of the hydraulic oscillator which utilizes the self-excited vibration phenomenon occurring in valves such as a spool valve or a poppet valve. Experiments on the two improved forms of the hydraulic oscillator - an oscillator with saturation restoring torque and one with reinforced hydraulic restoring torque - were carried out by systematically varying the factors affecting the vibratory characteristics. The vibratory characteristics having the frequencies in the range of 14 to 85 Hz and the angle amplitudes in the range of 10 to 34 degrees were obtained in this experiment. By comparing the theoretical vibratory characteristics presented in previous report with the experimental ones, it became clear that they almost agreed in the relation of the frequency to the angle amplitude, and available data for the design of the hydraulic oscillator were obtained.

View full abstract

This paper presents a numerical analysis of transient turbulent heat transfer from a flat plate with a constant heat capacity. A flat plate which was set parallel to a uniform flow was initially at the same temperature as the flow, and the plate was heated stepwise with time. The responses of turbulent heat transfer on the plate were calculated by the Laplace transform and the numerical inversion of the Laplace transform. The variations of a dimensionless wall temperature and a Spalding function were obtained. The relation between the numerical results and the one dimensional heat conduction solution or the quasi-static solution was also obtained

View full abstract

The effects of free-stream turbulence on the average heat and mass transfer coefficients around spheres were experimentally investigated. The Reynolds numbers ranged from 3.5×10³ to 1.4×10⁵ and the turbulence intensity was up to 8.5%. The average mass transfer rates measured by means of naphthalene sublimation technique were found to increase in direct proportion to the turbulence intensity, though the rate of increase decreased in the flow above the critical Reynolds number. The results of heat transfer were compared with those of mass transfer experments in order to study the effects of free-stream turbulence on the analogy between heat and mass transfer.

View full abstract

The effects of free-stream turbulence on the local mass transfer coefficients of a single sphere situated in air flow were studied with naphthalene spheres. The flow conditions covered Reynolds numbers between 3.5×10³ and 1.4×10⁵ and the turbulence intensities from below 0.3% to 8.5%. The influence of freestream turbulence was found to be much more remarkable on the mass transfer from the rear hemisphere, the mechanism of which was investigated by means of flow visualization, measurements of velocity fluctuations and wall-pressure distributions around the spheres. The appearance of a peak in the local Sherwood number distribution at the rear hemisphere with the increase of Reynolds number and of turbulence intensity was attributed to the intermittent reattachment of a separated free shear flow.

View full abstract

A numerical method for the two-dimensional freezing problem around a horizontal cylinder involving a maximum density point was presented. This problem offers a complex one involving the natural convection flow around the freezing front and the heat conduction inside it, and no exact analysis of it has been performed so far. The present numerical method utilizes the so-called boundary fixing method in which both the boundaries of body and freezing front are immobilized and it is applicable to general multi-dimensional moving boundary problems. Example calculations are made for steady freezing front shape, temperature distribution in solid phase, and also isotherms, streamlines, and Nusselt number distribution around the cylinder

View full abstract

Soot and oxygen concentrations, and temperature distribution in kerosene spray flames were measured to study the relation between soot formation and characteristics of spray flames. The distributions of soot and oxygen concentrations have an inverse tendency to each other. The maximum soot concentration and the minimum oxygen concentration are observed at the same location, which is the inside of the maximum temperature zone in the flame. As the ignition location in the spray comes closer to an atomizer, the maximum value of soot concentration on spray axis becomes lower. When the Sauter mean diameter of droplets in the spray is kept constant, the maximum soot concentration increases and its axial location comes closer to the atomizer, as the issuing momentum flux of a spray decreases.

View full abstract

When the flame propagates over liquid fuel at sub-flash temperature, heating the surface of the flame are required, and the surface-tension-driven flow, which is caused by the temperature distribution at the surface and flow, which is caused by the temperature distribution at the surface and controls the heat transfer at the tip of spreading flame, plays an important role. The characteristics of the surface-tension-driven flow are classified in to three groups (laminar, transient and turbulent). The relations between the flame spreading rate and the initial temperature are also classified into three groups according to the characteristics of the flow in the neighbourhood of the tip of a spreading flame. The flame spreading rates at sub-flash temperatures are calculated by using only one experimental value of flame spreading rate at the flash temperature. The numerical results coincide with the experimental results.

View full abstract

An experimental study has been conducted to explore the effects of the potential, size, and velocity of a cylindrical, electrostatic probe on the ion current fluctuation recorded when the probe traverses a premixed, two-dimensional nozzle burner flame parallel to the probe axis. The maximum ion current was found to be proportional to the probe length Lp and to increase with increasing probe diameter dp, probe velocity Vp, or absolute value |Ψp| of the probe potential Ψp. In the ranges of Lp from 1 to 14 mm and Vp from 1 to 22 mm/s, the estimated half-value thickness δh of the high-ion-density region near the flame decreased and approached about 0.4 mm as |Ψp| and dp were decreased. When the probe was operated under practical conditions, e.g., |Ψp|&fim; 40 V and dp &fim; 0.4 mm, the value of δh would be between 0.4 and 0.8 mm.

View full abstract

In order to discuss the influence of turbulence on turbulent burning velocity and to compare the experimental results of many workers with each other quantitatively, it is necessary to represent a turbulent field using the parameters with general definition. Some arbitrariness was unavoidable in the definition of turbulence parameters especially in the study of turbulent combustion in a closed bomb, in which the turbulence was measured usually with a hot-wire anemometer. A procedure for the analysis of a turbulent field of any turbulence intensity by processing the DC and AC indication of a squared anemometer output signal of the effective cooling velocity, is described and the results of the turbulence parameters in a closed bomb, such as intensity, temporal scale and spatial scale, in comparison with those by conventional method, are shown in this paper.

View full abstract

We have already reported about subharmonic oscillations of an unsymmetrical shaft with nonlinear spring characteristics, and made clear that unique vibration phenomena which were not observed in a shaft with circular cross section appeared. Similar phenomena appear in summed-and differential harmonic oscillations of the same system. We show analytically in this report that the following types of resonance curves are obtained depending on the degrees of nonlinearity and unbalance. They are a stable resonance curve of hard spring type, a stable resonance curve of soft spring type, an unstable region where unstable vibration appears, and the case where no summed-and-differential harmonic oscillation occurs. We made experiments with an unsymmetrical shaft supported by single-row deep groove ball bearings. The above-mentioned resonance curves were obtained in experiments.

View full abstract

Subharmonic and summed-and-differential harmonic oscillations of a nonlinear rotating shaft system where an unsymmetrical rotor is mounted on a shaft with circular cross section are treated. It becomes clear theoretically that the resonance curves of subharmonic oscillations of order 1/2 and summed-and-differential harmonic oscillations are classified in to four types. Therefore, similarly to those of an unsymmetrical shaft system reported previously, stable oscillation, unstable oscillation, or no oscillation appears. In an experimental apparatus in which single-row deep groove ball bearings were used, we observed the same types of the resonance curves as the theoretical results by changing the assembly of the apparatus and the size and location of the existing unbalance.

View full abstract

Observing a structural damping caused by the method of fitting the rotor into the rotating shaft, this paper has studied experimentally the structural damping, the estimated values of the internal damping and stabilities of many rotors. From these experimental results, it has been found that the fitting methods bear relations to the stabilities and one view has been obtained. And then, this paper has made clear theoretically and experimentally that the self-excited vibration of this kind can be prevented by supporting the rotating shaft flexibly.

View full abstract

This paper deals with the error analysis of the vibration of rotor/bearing systems. If the statistical properties of errors (i. e., mean values and standard deviations) are known, the mean value and standard deviation of unbalance response and critical speed may be obtained and the probability of instability may be calculated. It is concluded that the critical speed is not greatly affected by the errors, but the real part of the eigenvalue which is very important for instability is sensitively affected. The amplitude of the vibration for mass and stiffness errors varies with high sensitivity near the critical speed.

View full abstract

Coupled oscillations of liquid and a structure of a circular cylindrical tank are studied for small motions. The liquid is assumed as a perfect fluid, the side wall of the tank is treated as an elastic circular cylindrical shell, and the tank bottom is assumed rigid. The method of separation of variables is used to obtain series from solutions for the axisymmetric case. Velocity potential is introduced to express liquid motions in series expansions of an orthogonal system. Frequency equations are obtained from a kinematical condition along the side wall and numerically solved to illustrate the effect of coupling, showing that bat lower eigenfrequencies the motions are mainly due to liquid surface sloshing, and that at higher eigenfrequencies emerge so-called bulging modes consisting mainly of sinusoidal motions of the side wall.

View full abstract

This paper deals with the nonlinear vibration problem concerning mechanical equipment-piping system in nuclear power plants and others. A simplified dynamical model of these systems consists of an elastic beam with one end fixed and the other end supported by a piecewise-linear spring with clearance and hysteretic damping. Analytical methods of exact and approximate solutions are introduced for the model as a continuous system with nonlinear boundary conditions, and some numerical examples are shown. Finally some numerical results obtained from exact solutions are compared with those from approximate solutions.

View full abstract

The vibrations of circular plates with variable thickness are studied by the method of Mindlin's thick plate theory. From the Lagrangian of vibration of the circular plate with thickness varied along radius, the equations of vibration and the boundary conditions are deduced. The vibration problems of circular plates with two or more nodal diameters are solved in the case that the thickness is varied exponentially, the effects of some parameters on frequencies are discussed and some frequencies are compared with those from the classical theory.

View full abstract

The latest magnetic disk files require lightly loaded air-lubricated slider bearings operating with film thicknesses less than 0.5μm to achieve high recording density. This paper reports results on flying characteristics obtained from experiments with such air bearings, which are influenced considerably by slip-flow effects. Experiments are carried out in the range of 30 to 70% load reductions due to slip-flow. White light interferometric techniques for measuring the spacing between a rotating glass disk and the slider are improved. For load reductions less than 60%, experimental data agree well with the numerical solution of a modified Reynolds equation which takes into account the velocity-slip boundary conditions. It is verified that the modified Reynolds equation can be used to predict the static behavior of slider bearings in designing magnetic flying heads. An effective calculation technique for the Reynolds equation by the finite element method is presented.

View full abstract

Distributions of flash temperature T_f of on tooth flanks of spur and helical gears were calculated. The concentration of T_f at the beginning point of contact on leading side edge of tooth and at the ending point of contact on trailing side edge of tooth was found on helical gears, and the maximum T_f appeared on helical gears of small overlap ratio. By computer simulation for progress of scoring and by tooth flank observation of gears scored on test bench, the fashion of scoring progress on tooth and it proceeds in longitudinal direction on tooth flank of approaching contact, and then it occurs at the ending point of contact on trailing side edge of tooth and it proceeds on tooth flank of recess-contact in counter longitudinal direction.

View full abstract

The modular construction capable of manufacturing different kinds of machine tools using common modules (real modular construction) is recently of importance, not only to rationalize the design and manufacturing processes of machine tools, but also to evolve the computer aided manufacturing system. To realize the real modular construction system, a design methodology of it has to be developed, considering the extreme necessity of establishing the description method of machine tool structures. In this report, therefore, a method of the structural description, in which the concepts of the group technology and the flow of force in machine tool structures are used, has been proposed, also investigating its propriety through many practical applications.

View full abstract

To establish a design methodology for the real modular construction of machine tools, a method of structural description, by which the machine tool can be represented with the structural pattern, was proposed in the first report. As the next step, in this report, the evaluation method for the similarity of structural pattern has been investigated. Choosing the necessary informations from the structural pattern, the similarity in terms of `rate of commonness' has been computed for various kinds of machine tools, and the computed results of evaluation have also been compared with designer's intuition, which has been used in the designing procedure of machine tools to give a concept of deciding the similar structure.

View full abstract