The stresses in cemented joints, specially lap joints, were studied photoelastically. The photoelastic mcdels were made of epoxy resin and epoxy rubber or epoxy resin and metal, namely, each material had its own elastic constant. The shapes of the adhesive joints are a rectangular type, with a fillet, with convex or concave semi-circular ends. According to the results of experiments, the fillet is more effective for reducing the high stresses in lap joints. The other result has revealed that the joint with concave semi-circular ends is best, one with rectangular ends is second best and one with convex semi-circular ends is worst of all so far as the stresses in lap joints are concerned.

View full abstract

The object of the work described was to obtain information about the stresses due to tension in an I-beam with a circular hole at the web, which were studied photoelastically. Test pieces were made of epoxy resin with adhesive method. The stress concentration occurred on the hole boundary at the minimum section. The values of stress concentration of the I-beam were calculated from the values of the stress concentration with a circular hole of a beam having the equal sectional area without a flange.

View full abstract

When there is a crack on the straight edge of a plate under tension, to prevent further development of the crack, a circular hole is drilled at the end of the crack. Considering the above case, the author has solved the problem of a semi-infinite plate with a circular hole connected to the straight edge by a straight and narrow slit subjected to uniform tension parallel to the straight edge. He has obtained stress distributions on the plate and the factors of stress concentration at the bottom of the circular hole.

View full abstract

Here is described an analytical method on the steady-state creep of a rotating disk with variable thickness and nonuniform temperature distribution. The numerical computation by this method is easy. Creep tests of rotating solid disks of forged 18-8 Mo stainless steel were carried out in a hot spin tester prepared by the author and his associates. At the same time, specimens were cut out from the various parts of the disks and put to creep tests and other material tests. The analytical results were compared with experimental ones. Mises criterion has shown better agreement with the test data.

View full abstract

Canti-lever type fatigue tests were carried out on press-fitted and related specimens of 50 mm dia. Results obtained were as follows : (1) The hardness of a collar had little effect on the fatigue limit of a press-fitted specimen. (2) If the collar was press-fitted by superposing over the fillet of shouldered shaft, the fatigue limit was lowered remarkably as compared with that in the case of press-fitting alone. (3) The fatigue limit of the press-fitted specimen was reduced even under very light press-fit pressures, and the press-fit pressure had little effect on the fatigue limit. (4) When the collar was press-fitted closely to the shoulder with a very sharp fillet, the fatigue limit was lowest but not lowered as compared with that of a shouldered specimen with a very sharp fillet.

View full abstract

One of the important distortions of the gear teeth due to quenching is the one produced on the inner diameter, which must be fitted to the shaft properly. Distortion of the inner diameter of the gear, caused by induction hardening with simultaneous quenching of all teeth using a circular inductor surrounding the gear, was investigated. In such case of hardening, we can deal with a disk instead of a gear to study the tendency of the distortion. The stress and the distortion due to thermal change only were considered theoretically with the disk and the effects of quenching transformation were also discussed briefly. From these considerations, the mechanisms of the inner diameter distortion have been explained and it has been made possible to choose the method of induction hardening effective to reduce the inner diameter distortion.

View full abstract

Young's modulus of polyethylene, γ-ray irradiated, was measured by a vibration method and the effects of the radiation dose, the atmosphere of radiation, the thickness of samples and so forth on cross-linking were studied. Main results are as follows : (1) the degree of crosslinking of samples irradiated in vacuum is remarkably is remarkably larger than that irradiated in air, (2) the thinner the samples, the lower the degree of cross-linking produced in them and (3) for thin samples irradiated in air, the proportional relationship between the cross-linking and the radiation dose is not observed, and irradiation has complicated, rather harmful, effects on the mechanical properties such as yield stress and wear resistance. In Appendix, theoretical approach is shown for a composite oscillator which consists of the sample and the reed.

View full abstract

In this paper, the state of vibration of a circular cylindrical shell which is rotating about its axis with a constant angular velocity is discussed utilizing the fundamental differential equation introduced by the present author. The same problem has researched by G.H. Bryan⁽¹⁾ ; the utilization of his results, however, is very limited because of his assumption of inextensionality. In this paper, no such limitation is imposed and several problems which cannot be solved by Bryan's method are exhibited. The relations between the angular velocities of the nodes and those of the shells in various wave patterns are obtained, and in the particular case where the shell vibrates as a beam, the relation between the critical speed and the natural frequency is also discussed.

View full abstract

In the previous report⁽¹⁾, the strength of a horizontal reservoir supported at both ends and completely filled with a liquid was discussed, utilizing the fundamental differential equation introduced by Dr. K. Mizoguchi, and on comparing the results with those obtained by Timoshenko's method, it was pointed out that the latter's method was inadequate. In this paper, the strength of the same reservoir, being partially filled with a liquid, is discussed on utilization of the same methcd as before. In this case, the behavior of the deformation of the shell is very complicated ; e.g. the deflection is larger at a lower head than at a higher head of the liquid, which shows that the use of the approximate method such as that of simplified equation or the energy method is very dangerous.

View full abstract

The strict analysis of the strength and deformation of a thin circular cylindrical shell under a varying radial line load distributed around the circumference at a section has not been developed. In this paper, as an example of the problem, the strength and deformation of a drum supported at both ends under a line load in the middle through a rope is discussed, utilizing the fundamental differential equation introduced by Dr. K. Mizoguchi, and an experiment is performed for checking this analysis. In this case, the behavior of the deformation of the shell is very complicated as in the case of the previous report⁽¹⁾ ; e.g. the deformation of the shape of the cross section is so large that the uppermost and lowest side deflect upwards, quite opposed to the deflection obtained on the beam theory, which shows that the use of the approximate method such as that of simplified equation or the energy method is very dangerous.

View full abstract

Measurements of the torque consumed by an impeller rotating in a closed vessel were taken under various geometrical configurations by means of a torsion balance technique. The experimental results give the following conclusions : (1) When the boss ratio has a value between zero and 0.6, the resisting coefficient takes a constant value approximately. (2) The resisting coefficient takes the maximum value for a certain number of blades of the impeller and varies in proportion to (sinθ)^1/2 (θ≐̸0) where θ is the attack angle of blades. (3) The relation between the resisting coefficient C_f and Reynolds number R₆ in a smooth vessel is given by C_f ∽ R₆^-1/6. (4) When the width of the impeller is large, the resisting coefficient of the impeller is identical with that of a disc with ribs on both sides of the surfaces. (5) The effect of the number of blades disappears when the impeller has a shroud and the angle of blade is larger than 45 degrees.

View full abstract

The effects of runner surface roughness on the performance of a Francis turbine have been studied experimentally. The performance data were obtained for seven various type roughnesses from 0.007 to 1.0 mm in protuberant height. Preliminarily, the frictional resistance of enclosed rotating disks with respective roughnesses was measured for laminar and turbulent flows, in order to relate these different roughnesses to hydraulic frictional resistance. Through these disk-friction tests, the roughness of these surfaces could be expressed by the equivalent Nikuradse sand-grain diameter. The following conclusions may be drawn from this experimental study. (1) The turbine efficiency dercreases in proportion to the frictional resistance of its equivalent sand roughness. (2) With an increasing roughness, the maximum efficiency point shifts to the lower speed and the lower discharge. (3) Concerning the admissible roughness, the Schlichiting's formula is in good agreement with this experiment.

View full abstract

The first air curtain apparatus in Japan was installed several years ago. Since then, it has been used in places extensively with many practical experiences on its utility and performance, and a further wide range of application is promised. However, little has been studied concerning the operating principle and the performance of the air curtain, particularly, the effects of the outside wind and the outlet opening on the floor are fully taken into account. In this respect, comprehensive tests and analysis have been made so as to find how the air curtain is deflected and how much outdoor air blows into the room under various conditions of geometrical configuration of air curtain and outside wind velocity. As a result, general instructions for designing the air curtain have been made available, which are reported herein.

View full abstract

The natural convection of an electrically conducting fluid on a vertical plate in the presence of a transverse magnetic field is analysed for the magnetic field depending on the λth power of the vertical distance. The effect of the magnetic field on heat transfer may be uniformly described by the magnetic parameter, Z=Z₀(x/L)^2λ+1/2 where Z₀=2M²/G_rL^1/2. The local heat transfer coefficient is reduced as the magnetic parameter Z is increased, for example, 14% at Z=1, 45% at Z=5 for Prandtl number P_r∼0.01, while 11% at Z=1, 36% at Z=5 for P_r∼1. Also, if the total amount or the mean value of the magnetic field intensity over x=0∼L is equal, the mean effectiveness on the magnetic field becomes remarkable as the value λ is changed from a negative value to a positive one.

View full abstract

This paper deals with an analysis to determine the heat transfer characteristics of thermal entrance region for the established laminar flow of a heat generating pseudo-plastic fluid through a circular tube with wall heat transfer. The conditions of wall heat transfer treated in this paper are for both cases of uniform wall heat flux and uniform wall temperature. In this analysis, the numerical calculations were done by means of a digital computer KDC-1. The results for no internal heat generation with uniform wall temperature which were analyzed as a special case agree very well with those by Bird et al⁽²⁾. And the heat transfer characteristics in the case of uniform wall heat flux approach those of a fully developed region which were calculated by U. Grigull⁽⁵⁾.

View full abstract

This paper considers convective heat transfer to a flow in an annular gap with an axial flow between a rotating inner cylinder and a stationary outer cylinder. The experimental results indicate that the heat flow from the surface of the inner rotating cylinder is the sum of the following two kinds of the heat flow : one by turbulent vortices induced by rotation instead of Taylor vortices and the other by the turbulent axial flow. For turbulent vortices we have the empirical equation for the overall heat transfer coefficient when without an axial flow : [numerical formula] Moreover for the turbulent axial flow we have the empirical equation on the heat transfer coefficient [numerical formula] Here it is shown how this correlation of summing up the two heat flows is applicable to the wide regime of such mixed flow.

View full abstract

Generally it is possible to start the boiling at the surface temperature of a heated wall somewhat exceeding the saturated temperature of the liquid, and the heat flux of initial boiling varies depending on the fluid temperature, liquid velocity, system pressure, surface condition of the heated wall, and so on. In this paper, an analytical consideration was tried to make clear the conditions of incipient subcooled-boiling with forced convection, and the equation to estimate the heat flux under that condition was determined. By comparing with the authors' data at atmospheric pressure or the experimental data of previous investigators at higher pressure, it is known that the analytical results agree very well with experimental ones. The relation to estimate the conditions of incipient boiling is as follows ; [numerical formula] where, A : heat equivalent of work, D_e : hydraulic diameter, P_r : Prandtle number, q : heat flux at incipient boiling condition, γ : latent heat, T_s : saturated temperature, ΔT_sub : subcooling, u : velocity, v : specific volume, λ : thermal conductivity, ν : kinematic coefficient of viscosity, σ : surface tension. Suffix ; g : vapour, l : liquid, 0 : value of bulk temperature.

View full abstract

E. Zielinski discovered experimentally a comparatively easy method to reproduce film boiling. According to his method, if a wire which is making nucleate boiling with heating current adjusted to a rather low heat flux is pulled out of water for a short time, the surface of the wire dries up and its temperature rises because of the contact of the wire and air, even when heat flux is kept constant. If the wire is dipped in water later on, for the second time, it heats red-hot in a few seconds, followed by the appearance of film boiling. Film boiling is to be reproduced with comparative easiness by breaking temporarily the contact of heating surface with water. The authors who paid attention to Zielinski's method followed it with the result in which nucleate boiling and film boiling are made to coexist stably on the same heating surface by heating them with a given heat flux. This is considered to be a kind of transition boiling, and the present paper deals with the boiling phenomena at the coexistence of nucleate and film regions in saturated water, subcooled water and sodium oleate solution as well as their relation with boiling characteristic curve. In this region the coefficient of heat transfer decreases with the increase of temperature difference, while heat flux increases with the increase of temperature difference.

View full abstract

Through an experiment of nucleate boiling under free convection in subcooled water, we came to the following conclusions : (1) The effective stirring length of bubbles in surface boiling is reasonably estimated to be one-third of their condensing height which is determined identically with the dgree of subcooling of bulk water. (2) Correlation of heat transfer in surface boiling using Δθ_s instead of Δθ corresponds to the replacing of H_e by H_es, where Δθ_s : temperature difference between heating surface and saturated liquid, Δθ : temperature difference between heating surface and bulk liquid, H_e : effective stirring length of bubbles in saturated boiling, H_es : effective stirring length of bubbles in surface boiling. From these conclusions, we derived an extension of author's correlating equation of nucleate boiling heat transfer, and confirmed that it fairly agreed with the experimental results in surface boiling.

View full abstract

In this paper, the ignition of pulverized coal by radiation is dealt with first. Three kinds of heat are taken account of : heat radiated from high temperature bodies, such as flame and furnace wall, to the pulverized coal particle ; heat transferred from the particle to the surrounding air, and heat caused by chemical reactions on the particle's surface. Thus we show how the particle starts to ignite and what effects various factors such as ash in coal, primary air/coal ratio, size of coal, and temperature of heat source would have on the ignition. Next, we show the correlation between the unburned fraction of fixed carbon in coal and the furnace combustion rate, modifying the Hottel's method, and its correction curves according to the size of particles and excess air ratio.

View full abstract

The piston temperatures of two 4-stroke agricultural engines and a 2-stroke motor cycle engine were measured accurately by means of the thermocouple method, in which the thermocouple-wires were connected with the moving piston and a fixed point without any joint on the way. For this purpose we used the following two devices : (i) The thermocouple-wires were put on a spring which was fixed to the piston skirt at one end and to the crank case wall at the other end. (ii) A conduit containing the thermocouple-wires was movable with the piston through the combustion chamber. With the former device, the measurement was carried out at the speed up to 4000 rpm for a 4-stroke engine of stroke 65 mm, and with the latter, at the speed up to 8000 rpm for a 2-stroke engine of 40 mm stroke.

View full abstract

Power curves in hot strip mill both in roughing and finishing train can be expressed as the following equation. E=E₀ (λ^m-1) where E is horsepower-hour per net ton, λ is elongation rate and E₀ and m are power curve constants. From the power curve formula, we can deduce the equation of roll separating force, and the equation of delivery thickness in tandem rolling. If the roll speed of the last stand is given, the roll speed of each stand is set by the following equation, V_i=V₆·h₆/h_i And we can set the screw-down position by the following equation. S_i=h_i-P_i/M+C_i' The validity of these equations is verified by the experiments in Tobata No. 2 Hot Strip Mill, and we can apply these equations to computer control in hot strip mill.

View full abstract

The rolling of thermoplastic sheet at the plastic flow temperature seems to be different from the rolling of metal because the elastic after-effect of thermoplastic is very remarkable. However, according to the experimental result of compression of the sheet, indentation by roll into the sheet and rolling of the sheet of rigid polyvinyl chloride, the rolling theory which is utilized in metal working can be also applied to the rolling of thermoplastic sheet if the elastic after-effect is evaluated. This paper shows that the necessary data on the elastic after-effect can be obtained from the tension test.

View full abstract

As a basic experiment on gear scoring, the surface temperatures of gear teeth are measured. So far, we can find the experiment for measuring the mean surface temperatures of gear teeth. But it is significant in considering a countermeasure of scoring that from another point of view we know the change of the instantaneous temperature with the change of the meshing position. In our experiment the thermocouple consisted of a constantan pinion and a steel wheel, and the thermoelectricity produced in it was observed by oscilloscope. As a result we reached the following conclusion : (1) On the relation between the meshing position and the surface temperature of gear teeth, the experimental value is nearly equal to the value calculated theoretically. (2) As the surface temperature of gear teeth is largely influenced by the load distribution, it is most desirable for designing the gear that we try to change the profile of gear teeth so that the surface temperature may be minimum, and to avoid the local contact of gears. (3) The surface temperature of gear teeth rises in proportion to the circumferential speed of gears in the range of lower speeds, but in the range of higher speeds its change becomes small. Speaking of the influence of loads, the temperature changes in proportion to 1/2 the power of the load, but in the range of higher speeds the change becomes small.

View full abstract

In this report, first, the temperature rise of oil film is analysed. In this case, the heat transmission through the surface to the rollers together with heat generation by internal friction loss in the oil film is taken into consideration. Next, the effects of the elastic deformation of rolling surfaces are analysed, and approximate calculating formulas for the maximum pressure on the rolling surface and the minimum film thickness are investigated. Further, the measured film thickness is compared with the value given by the above theory. It is shown that the measured values are involved in the region where the effect of elastic deformation is negligible ; and, when the rolling velocity is high, the measured values are smaller than the values obtained by the theory assuming constant viscosity, but, when the rolling velocity is low, the relation is reversed. This is explained reasonably by the above theory.

View full abstract

The author has already reported that the lubricating conditions of the gears remarkably affect the endurance limit of the pitting. To explain these phenomena theoretically, the author has analyzed the hydrodynamic lubricating film of the gears, taking into account the elastic deformations of the rolling surfaces and the film temperature. In the present report, however, it is shown that the experimental results on the pitting are reasonably explained with the aid of the above theoretical analysis. That is, the thicker the theoretical film is, the more of the cut-marks of grinding of the gear surfaces remain after the running. Further, if the maximum film pressure calculated on the basis of the above theory is taken instead of Hertz stress, the Wohler curve becomes only one curve under various lubricating conditions. These results show that the pitting must be analyzed with the aid of the hydrodynamic film lubrication theory.

View full abstract

In externally pressurized gas-lubricated circurlar thrust bearings, the occurrence of depression in the pressure profile is observed under some working conditions. Following the previous paper on the theoretical investigation of pressure depression, this paper investigates theoretically the pressure distribution in externally pressurized circular thrust gas bearings with a circular recess by assuming the generation of a normal shock wave in the bearing recess. Good quantitative agreement is found between the analytical results and available experimental data except the case of small bearing clearance height or deep recess. In order to prevent the pressure depression just after the gas supply hole, it is effective to provide a round or tapered chamfer at the entrance edge of the hole to the bearing clearance or to make the recess tapered doubly. Through these alternations, the shape of pressure distribution becomes flatter so that the load capacity can increase remarkably.

View full abstract

There were many studies on electrohydraulic forming in which the relationship between the deformation of metallic meterial and the stored electrical energy was investigated. It is necessary for us, however, to explain the correlation between the deformation resistance for each of these metals and the pressure pulse in order to analyze the deformation characteristic of metals. The experiments were carried out by means of pressure gage (BaTiO₈) on deformation pressure and pressure pulse which were directly related to the deformation, and further, the shock wave energy was calculated. The efficiency of these apparatus was discussed. The clearance between the electrodes creating pressure was short-circuited with several kinds of fine wire, and the influences of the diameters and lengths of wires were presented. Also in the case when there existed water only between the electrodes, the influences of electrode diameters and clearances were presented. It is found from experimental results that cases where the clearance between the electrodes was short-circuited with a fine wire are superior to the case where water was the only pressure source between the electrodes, and the aluminum wire which had a low boiling point metal was effective. The rise time of pressure pulse is about 5μsec and its duration is about 15μsec. Such shapes of pressure pulse do not change up to 1000 joule charged energy. Shorter wires are better than longer wires as pressure sources, while the diameter of wires creating the largest pressure is limited by characteristic of apparatus and the charged energy.

View full abstract