In the usual moire method and the multiplication method, the sensitivity to deformation is not sufficient at present. Then, how to increase the sensitivity is studied. In the measurement of a usual sheet metal specimen, moire fringes with large mis-match is photographed on a film at first. Such fringes are equivalent to grating lines on the specimen, as thy displace corresponding to strain, and they can be amplified by applying the multiplication method to that film. When strain is expressed by two beam interference fringes, one of the beams is tilted by a small amount, then fine fringes are obtained. They are equivalent to moire fringes with large mis-match and the multiplication method can be applied. Applications in various conditions have been shown. Consequently, concerning a sheet metal, measurement of elastic-plastic strain (ε⋍ 0.001) becomes possible and in the special condition that the surface of a specimen has been finished mirror-like, elastic strain (ε⋍ 0.001) can be measured.

View full abstract

The present work treats a free-flexural vibration problem of a stepped elliptical plate consisting of confocal elliptical parts. Based on the classical theory of thin plates, the exact analysis is developed in the elliptical cylinder coordinate system. The procedure leading to the frequency equation of the normal mode of symmetrical vibration about both axes is given for a stepped elliptical plate consisting of two confocal elliptical parts. Natural frequencies of the first two normal vibrations are calculated numerically and compared with experimental results by a resonance method.

View full abstract

Several papers have been published concerning the vibration of a rotating shaft carrying a rotor at the time of passing through the critical speed. But most of them treat the rotor as a concentrated mass. In this paper, the equations of motion of the rotor when the shaft speed varies arbitrarily are introduced using a rotating coordinate system and taking into account the gyroscopic effect of the rotor. The equations are solved by Runge-Kutta-Gill's numerical method in various cases where the rotor and the bending rigidity of the shaft are symmetrical or asymmetrical and the shaft passes through the critical speed with a constant or a variable angular acceleration. The exact solutions expressed with Fresnel's integrals in the case when the rotor is a concentrated mass are compared with the results of calculation by the numerical method in order to check the accuracy of this method. In addition, the results of simulation by an analogue computer are shown.

View full abstract

While a number of studies on impact dampers having a free additional mass have been published, impact dampers having a spring-supported additional mass have been dealt with by only a few investigators. In this study, we devised a model of an impact damper having a spring-supported additional mass and made theoretical and experimental analyses of it, thus obtaining data useful for practical planning application. The obtained results are as follows : (1) Generally, an impact damper of one-sided impact design is superior in damping effect to one of two-sided impact design. (2) If the static distance between the impact damper and the stopper is chosen properly and the ratio of the natural frequency of the impact damper to that of the damped mass is set at a value of 0.6 - 0.8, a damping effect is achieved over a wide range of excitation frequencies. (3) The calculated and experimental results are in fairly close agreement with each other.

View full abstract

In high-head and large-sized radial flow turbomachinery some devices, such as balance holes or balance disks, are usually used to balance the axial thrust. But these devices sometimes result in an unbalanced axial force owing to wear or aging of wearing rings. This report puts forward a new device which is quite effective to control and balance the axial thrust with ease and convenience. Even if the above mentioned unbalanced force is caused, it is possible to balance it by this new device even during the operation of a machine. In the former report, it was suggested that it was possible to balance the axial force by controlling the angular momentum of the flow between an impeller shroud and a casing wall. this device uses the phenomenon of angular-momentum defect caused by the wake flow behind several short cylindrical pegs attached onto the casing wall in the rotating flow. The effects of cylindrical pegs on the thrust control are studied theoretically and experimentally.

View full abstract

Swirling flows in cylindrical ducts were investigated by many researchers to meet practical requirements, but the detailed nature of the flows has not yet been clarified. This paper describes the experimental results of swirling flow obtained in stationary cylindrical pipes. Except the inlet region of the pipes, velocity distributions of the swirling flow are governed by a single dimensionless parameter Ω defined by (angular momentum flux)/(axial momentum flux), irrespective of the inlet conditions of flows and wall roughness. This swirling flow shows three kinds of flow patterns in its course of decay; (1) forced-free vortex type, (2) transitional type, (3) forced vortex type. The strength of the swirl Ω decreases exponentially along the pipe axis and its decrement varies with change of the flow patterns. Shear stresses for these flow were calculated, and also the total amount of flow energy was discussed in connection with wall pressure measured along the pipe axis.

View full abstract

As part of a serial study of the effect of a damping screen on the velocity distribution of pipe flow, the case of a flow with S-shape velocity distribution is analysed theoretically. The analyses are performed for the case of a single isotropic damping screen and for three other combinations of (1) an isotropic damping screen and an isotropic lattice, (2) the screen and an anular lattice and (3) the screen and a radial lattice. Theoretical results show that, for a single isotropic screen, the final damping ratio is the same as in the case of an axi-symmetric axial velocity distribution, but the flow uniformalizing process is more gradual than the axi-symmetric case. It is also made clear, for cases of combined damping devices, that the damping process vanishes downstream of an isotropic lattice, and a secondary flow is generated downstream of an annular or radial lattice.

View full abstract

Although the Navier-Stokes equations have been solved directly by using the finite-difference method, the calculation requires in general method for time-dependent two-dimensional laminar flows is proposed with the aim of reducing the time spent and increasing the accuracy of the solutions is investigated theoretically from the truncation errors, and the stability criteria are derived from the inherited errors. A comparison of such accuracy and also some numerical examples of the present and the other available methods show that the second approximation of the present method is the most suitable.

View full abstract

In this report, the numerical method to solve the problem of time-dependent two-dimensional laminar flows which was proposed in Part 1 is modified from some practical points of view. First, the previous method using a square grid is extend to use a nonuniform rectangular grid with the aim of saving the computer time. Next, an iterative method is developed in order to solve the problems of prescribing the flow rate through the duct for which explanation is given on some cases. Finally, some numerical examples are demonstrated. In these example are briefly discussed the instability of the solutions at high Reynolds number and its location, the dependence on interpolation formula and on mesh size.

View full abstract

For laminar incompressible viscous flow past a semi-infinite flat plate with an injection or suction velocity v₀=C₀/√(2Re), the numerical solutions of the Navier-Stokes equations are obtained by the alternating direction implicit method. It is shown that there exists a reverse flow region near the leading edge of a flat plate with injection so that the dividing streamline is detached from the leading edge. The value of the surface vorticity function at zero Reynolds number limit is fount to become smaller than that of the high Reynolds number limit for large suction rates in contrast to both cases of flows without mass transfer and with injection. It is found at finite Reynolds number above 20 that the skin-friction value increasingly deviates from the boundary-layer solution with increase of the injection rate C₀, and that it remains finite even at the critical 'blow off' value of C₀.

View full abstract

In the present report, a single phase alternating flow hydraulic system the output end of which is connected to a load with mass, spring and friction is treated theoretically. The general equations for the transmission power and the efficiency of the system are derived, and then the load conditions are examined which give the maximum power and efficiency. It is made clear that the maximum give the maximum power and efficiency. It is made clear that the maximum efficiency is not always equal to that of the impedance matching. From a viewpoint of obtaining higher efficiency for the system with constant load, the relations between the efficiency and the conduit conditions are also discussed.

View full abstract

In order to obtain fundamental knowledge of performance and optimum dimensions of a disk friction pump, an analysis is made of the flow between two infinitely large corotating disks with throughflow supplied at the inner periphery of finite radius. The flow field is assumed to be either everywhere laminar or everywhere turbulent. The laminar case is treated by dividing the field into three regions; the region in the vicinity of the entrance periphery to which Gortler series is applied, the region far from the center axis for which analytical perturbation solution is used and the mid-region connecting the above two extreme regions for which finite difference computation is carried out. The solution gives a complete scheme of the flow field with satisfactory indications for the engineering scheme of the flow field with satisfactory indications for the engineering purpose. Meanwhile, the turbulent case is treated by means of momentum equations with elaborated curvilinear coordinate axes, which leads to a physically acceptable flow pattern.

View full abstract

In this paper, first, experimental investigations were carried out about the starting torque, the friction torque in low speed range and the leakage flow. For these experiments, a test rig utilizing a worm gear as the load of a test motor was used. Second, using a pressure loading type gear motor, in which the pushing force of the side plate was adjustable, the effect of he pushing force on the friction torque was investigated. Finally, the relationship between the variation of the starting torque and that of the leakage flow according to the rotational position of the drive shaft was investigated.

View full abstract

An attempt was made to simulate the forced flow boiling in flow characteristics with air-water two-phase fluids, into which air was injected through the porous wall of a round tube. The test section of porous wall is 1450 mm in length and 26 mm in inner diameter. This report treats the spatial (i.e., radial and axial) distributions in both void fraction and bubble frequency for the bubble flow regime under atmospheric pressure and room temperature. The experiments suggest the existence of three regions differentiated in the radial distribution of bubbles. The 1st. one is the bubble sublayer formed adjacent to inner surface. The 2nd. one is bubble transition layer characterized by rather high concentration of bubbles. The 3rd. one is the core outside the bubble transition layer. Results and discussion are presented on the thickness, void fraction and bubble frequency in two bubble layers. Development of these flow parameters along the tube is also discussed.

View full abstract

An experimental study has been made for ascertaining the relation between the combustion phenomena of a propagating flame and the ion current from the flame to an electrostatic probe or ionization gap. A schlieren system was simultaneously used together with the electrostatic probe to study a flame propagating through a stoichiometric propane-air mixture in a closed combustion tube in the initial pressure range from 100 to 600 mm Hg. It was found that the flame front might be disturbed by the turbulence of the unburned gas flow. The measured distance between the locations of schlieren front and maximum ion current lay in the range from 2 to 6 mm. When the flame front was disturbed, the ion current fluctuations were observed. These ion current of the flame front and were inferred to be closely to the flame front configuration.

View full abstract

An important point in designing a bolted joint is to know the increment of an axial force produced in a bolt, when a load is applied on an assembly. For this purpose, it is necessary to know the spring constant K_c for a connected part and K_t for the bolt with a nut. In the calculation of K_t, up to now, the deformations of bolt head and nut have not been taken into consideration. A purpose of this paper is to analyze the value of K_t, taking into account these deformation, and to rationalize the design of the bolted joint. Assuming that the bolt head is a finite cylinder and the nut is a finite hollow cylinder, these deformations are analyzed by using the three-dimensional theory of elasticity. Experiments on these deformations are carried out, and the theoretical value is compared with the experimental one. They show a fairly good agreement with each other. Taking into account these deformations, for example, the value of K_t for M24 (clamping length 50 mm) is calculated. It is found that this value is 20% smaller than that of K_t by the conventional method.

View full abstract