It is a basic problem in the deformation of polycrystalline aggregates composed of anisotropic crystals, whether the stress or the strain is constant in the polycrystals. In order to treat the problem analytically, the idea of constraint ratio was introduced as a measure of deformation behaviour of each grain. Basic relations between stress and strain were deduced in terms of the constraint ratio. On the basis of the relations acquired, energetic consideration was made on the deformation characteristics of the polycrystals. It was found that, if the mutual restriction between the grains is absent, the constant stress condition is energetically stable. In real polycrystals, however, the strain is to be continuous through the grain boundary during deformation. It was concluded that, the mode of deformation changes gradually from constant stress to constant strain in proportion to the increase in strained region around the grain boundary.

View full abstract

Fatigue tests of S 20 C electropolished specimens were conducted with a rotating bending testing machine under double repeated stress in two stress levels. The results are as follows. (1) The values of accumulated cycle ratio based on the initiation of crack (length of 0.1 mm) are between 0.9 and 1.1 and independent of the loading order of stress amplitudes. This may imply that the fatigue damage before the initiation of crack is accumulated in a linear way and mainly at the same places. (2) In the process of crack propagation, the values of accumulated cycle ratio based on the fracture of specimen depend on the loading order of stress amplitudes. This may be explained from the fact that the crack lengths at the same cycle ratio under different stress amplitudes do not coincide. (3) The repetition of the second stress increases the density of slip traces mainly in the grains whose slip traces appear under the repetition of the first stress. (4) French's damage curve is close to the crack (length of 0.1 mm) initiation curve.

View full abstract

In the present paper, the construction and features of a fretting fatigue testing machine designed by the authors are explained, and the experimental results obtained by using the testing machine are stated concerning mainly the relative slip which is considered to be one of the most significant factors affecting the fretting fatigue strength. The following are known from this investigation. In fretting fatigue, many shallow micro-cracks which seem not to propagate to a complete fracture are formed and develop somewhat obliquely from the surface of specimen because of the effect of frictional force due to fretting. Fatigue strength is reduced significantly by the effect of fretting, and the strength reduction factors, concerning both nucleation of fatigue cracks and fatigue fracture, are greatly influenced by the magnitudes of relative slips. The fatigue strength is most significantly reduced when the relative slips are about 0.015∼0.020 mm, and reduced to less than 1/5 of its strength without fretting, and if the relative slips could be kept less than 0.005 mm, a marked improvement would be expected.

View full abstract

In order to clarify how the number of strain cycles to failure is influenced by temperature and frequency of straining in low cycle fatigue, it is necessary to observe the process of the fatigue failure. To obtain the knowledge of the process, crack appearances existing on the failed specimens were compared with each other. Fatigue tests were carried out in a push-pull testing machine on the AISI 321 stainless steel under various temperatures and frequencies of straining conditions. According to the results, the crack appearance does not depend on the temperature and frequency of straining conditions but on the plastic strain range. If fatigue tests are carried out in the same plastic strain range but at different temperatures and frequencies of straining conditions, similar distributions of crack density from the outer surface to the interior of specimens are always observed in the failed specimens. Thus, it is concluded that in low cycle fatigue the behaviour of fatigue cracks is governed primarily by the plastic strain range and it is deduced that the rate of fatigue process depends on temperature, strain rate, and plastic strain energy per cycle.

View full abstract

An analysis of wave propagation in an infinite elastic beam on an elastic foundation is presented, considering the effect of the mass of a foundation. The foundation is represented as a base consisting of closely spaced, independent bars or linear springs. The frequency equation is derived and the spectrum of frequencies for real, imaginary, and complex wave numbers is investigated. The phase velocity and group velocity versus wave number are also calculated for various ratios of the mass of a foundation to that of a beam. Results from the present method of solution are compared with those obtained under the assumption that a beam is supported by a massless elastic foundation whose local restoring force per unit length depends only on the local deflection.

View full abstract

When the power supply to the pump is suddenly cut off, the pressure drop and rapid pressure rise are caused by a pressure surge. These phenomena are well known as a waterhammer. Some works on this subject have been published in recent years. In this paper, in the first place the quasi-linear system of hyperbolic partial differential equations is transformed into a problem of the characteristic equation and eigen values, next a new graphical method based on the characteristics is presented and the effect of hydraulic friction subsequent to the water column separation is discussed. As an illustration of the method, calculation examples of mathematical models of Isogo and Chita Pumping Plants associated with the cooling-water supply for thermal power stations are shown.

View full abstract

In this paper, the mechanism of back flow to the suction side of mixed-flow pump impeller was studied by measuring the pressure distribution and the flow direction on the outside of impeller blades. Furthermore, the effect of setting angle of impeller blades on the back flow mechanism was investigated. The obtained results are mainly as follows : (1) The pump performance was made clear. (2) Pressure distributions and flow patterns on the outside of impeller blades were made clear. (3) The difference of flow on the outsides of axial-flow and mixed-flow impellers was made clear. (4) Two different reasons for the beginning of the back flow were discovered with the mixed-flow impeller.

View full abstract

On the flow properties of stable slurries of several polymer particles, experiments were carried out, using a Couette-type rotational viscometer and two capillary viscometers, in the range of concentration : 0∼45 (%) and rate of shear : 5∼7×10³ (sec^-1), changing the size of particles, the viscosity of suspending liquids and the temperature. The results obtained are as follows. (1) The flow properties of slurries are not independent of the kind of viscometer (rotational or capillary) and the apparent viscosity by the capillary viscometer is lower than one by the rotational viscometer, at the same rate of shear. (2) P.M.M.A.-slurry is a Newtonian fluid in the range of concentration from 0 to 20 (%). But, P.E.-and P.V.C.-slurries become non-Newtonian fluids at lower concentrations. (3) Oliver's constant of P.M.M.A.-slurry is 2.60 and its value agrees with the results of experiments by others.

View full abstract

In 1st report, we reported the discrepancy between the rotational and the capillary viscometers concerning the flow properties of stable slurries. In this paper, this discrepant phenomenon is analyzed from the standpoint that it is due to the concentration distribution of particles in slurries in the circular tube flow and discussed on the experimental results. In the analysis, it is assumed that slurries are stable and composed of rigid spherical particles and Newtonian suspending liquid. It is shown that the concentration distribution is affected by the slurry concentration, the ratio of particle radius to capillary one, the ratio of length to diameter of capillary and Reyonlds number. It is shown too that the additional loss due to the concentration distribution is negative. Next, on the viscometry using the capillary viscometer of non-spherical particle slurries, a practical approach is taken.

View full abstract

One-dimensional unsteady heat conduction problems can usually and easily be solved numerically by use of an analog computer. In this paper, the following problems about linear one-dimensional transient heat conduction are revealed. (1) Computer circuits for boundary conditions of no flux across the surface, prescribed heat flux across the surface, linear heat transfer at the surface, non-linear heat transfer, contact with a well-stirred fluid or perfect conductor, and the surface of separation of two media of different conductivities are given in addition to the well-known circuit for that of prescribed surface temperature. (2) The accuracy of numerical solutions using the analog computer is shown to be governed by the degree of approximation in the ordinary differential equation, which is converted from the original equation of heat conduction. (3) [numerical formula]=0.4∼2.6 is shown to be of sufficient condition to solve the above problems by analog computers, when the required accuracy is within 2∼3% of the maximum temperature change.

View full abstract

Convective heat flux from a nonisothermal surface can be calculated by employing the Spalding function as the turbulent heat transfer coefficient and by the method of superimpositions of temperatures. And it gives a better accuracy for the calculation than the other heat transfer coefficients. However, the result is not always in good agreement with the measured one. An experiment is carried out in this paper to examine whether the discrepancy originates in wall temperature approximation for the superimposition or in the calculation method itself. The experimental surface is isothermal including many stepwise discontinuities on the wall temperature, so that any temperature approximation is unnecessary for the calculation. An idea to normalize the data from different wall temperature distributions is introduced. The experimental result shows that the method of superimposition with the Spalding function is proper.

View full abstract

The rate of sublimation of ice was investigated by experiment. The pressure range was from 10 to 10^-3 mmHg. The following three processes were shown to determine the rate in the mass transfer of sublimation of ice. (1) In the pressure range (P>1 mmHg), the rate of mass transfer is governed by the diffusion velocity of vapor in vapor phase, and the maximum rate is reached when the vapor supersaturates and recondensates on the ice surface. (2) Under 10^-2 mmHg, the velocity of molecular flow, known in the kinetic theory of gases, governs the rate of mass transfer. (3) In the pressure range 1 to 10^-2 mmHg, the rate is governed by the velocity of phase transition from ice to vapor on the ice surface. In this range, the ice surface becomes very rough. Maximum rate is reached when the surface area of ice increases to the maximum. (4) For finding the correlation between heat and mass transfer from these experimental results, use of the following relation is recommended. α/h=_ρC_p(α/D)(P^^-_g/P) α : coefficient of heat transfer, h : coefficient of mass transfer.

View full abstract

A method of measuring Seebeck coefficient, electric conductivity, thermal diffusivity and thermal conductivity of semiconductors with a single apparatus was investigated. The feature of this method, improved from Harman's method is to measure the above properties separately and easily with a single apparatus. P-type and N-type semiconductor-rods are soldered and sandwiched between copper blocks. Thermal diffusivity and thermal conductivity were measured by heating the sample periodically by the Peltier effect. The principle of the above method was derived from the solution of the equation of heat conduction in semiconductor specimens. Seebeck coefficient and electric conductivity were measured from the steady state temperature and electric potential distribution in semiconductorrods. The dimensions of specimen and the measuring conditions for the measurement with an accuracy of 2% were discussed. Test results for Bi₂Te₃ by this method were shown to agree with these of Goldsmid, Harman and Nii.

View full abstract

As to the bubble growth in saturated pool boiling under low heat flux conditions, a theoretical equation was derived by introducing the contribution of superheated liquid around the bubble as well as that of evaporation of thin liquid film between bubble and heating surface. The theoretical prediction showed good agreement with experimental results obtained under atmospheric pressure. It was found that the contribution of surrounding liquid predominates at first and later the effect of thin film evaporation gradually happens to appear, as far as the heat flux is considerably low.

View full abstract

Local void fractions averaged with time were measured above and adjacent to a horizontal heating surface in saturated pool boiling of water at atmospheric pressure. Measurements were performed by the"Probe Method"previously developed by authors. Main results are as follows : (1) Characteristic changes of void fraction distribution occur below a height of 4 mm. (2) An explanation about relationship of mechanism among three boiling regions can be derived. (3) In nucleate boiling, the distribution curves versus height have a peak near the heating surface. However, there is a liquid-rich layer under the peak, and the layer is considered as the region which governs heat transfer. (4) Void fractions over 90% were measured at a height of 0.3∼1.0 mm near the condition of the burn-out point. (5) In transition boiling, liquid is in contact with the heating surface. (6) In our experiment on film boiling, the minimum film thickness is very little.

View full abstract

A new equation of state for compressed water was formulated in the ranges of temperatures from 0 to 350°C and of pressures from nearly zero to 1000 bar based on the International Skeleton Tables (1963). This new equation expresses the specific volume as the function of pressure and temperature, and is suitable for computer calculation. Calculated values of the specific volume and the enthalpy agree well with the values of the International Skeleton Table (1963). This new equation was transformed into the nondimensional canonical form (the equation of Gibbs function). The equations for various properties such as enthalpy, entropy and specific heat were derived from this new equation of state.

View full abstract

This paper deals with the low frequency type of oscillations in a simple combustion chamber where the fuel used is propane. The phenomenon was found to be closely connected with the progressive flame necking from the results of flame photographs by a high speed camera and the frequencies of oscillations were able to be caloulated as a Helmholtz resonator. From both results mentioned above and measurements of pressure drop along the combustion chamber, it was found that the oscillations were caused by the negative gradient of the characteristic curve which represents the relation between the pressure drop and the secondary air velocity in the vicinity of the flame necking. Some theoretical analyses were made and a differential equation of Van der Pol type was obtained. The analytical values of the frequencies well agreed with the experimental results and an example of oscillation patterns on phase-plane was in good agreement with that of oscillograms.

View full abstract

This study attempts to produce multifuel capability for a standard diesel engine with a few simple modifications, and attempts to obtain a correlation existing between the performance characteristics of the engine and the combustion behaviour on a variety of fuels. In the first paper, two methods of reducing the ignition lag were attempted fairly successfully : 1. Reduction of the heat transfer by promoting the insulating effect of the pre-chamber with a sleeve to raise its mean wall temperature. 2. Employment of higher compression ratio up to 19.05. As a result, a possibility of operating a diesel engine on automotive gasoline was assured.

View full abstract

The idea that the ignition lag of the main spray can be reduced by a preceding injection of pilot spray has been investigated several times. This means was found effective to realize multifuel operation of a conventional diesel engine. In order to realize pilot injection, many ideas have been conceived, but in this study two independent injection pumps and one common nozzle were employed. First, an investigation was made to determine whether injection characteristics of the proposed injection system were suitable for practical engine operation or not, and desirable results could be obtained. Secondly, we attempted at the operation of a pre-chamber diesel engine on low cetane fuel, and excellent results could be obtained with pilot injection if automotive gasoline was used. And the reduction in the ignition lag of the main fuel was caused mainly by the pre-reaction of the pilot fuel.

View full abstract

Up to now, equations used for the design of cavity type mufflers have generally been TL. But, since TL is not for design purposes, much inconvenience results when it is used as a practical equation for muffler design. For this reason, the author has explained in this paper the theoretical equations for muffler design which he developed, analyzed characteristics in overall muffler construction, and shown the experimental example for each muffler.

View full abstract

It takes more time to design link mechanisms than to do cam and Geneva and other mechanisms because high-order equations must be solved by trial and error methods and complicated data treated. In this paper, these problems are solved by the analog computer. The analog computers have been used as the simulators in investigations of link mechanisms but never been used to synthesize mechanisms. In this paper, the analog computer is used for three-points' synthesis when planar four-bar linkages are used as function-generators. Furthermore, a new simulator is shown to be applicable for examining precisely the conditions of motion of the synthesized mechanisms.

View full abstract

One of the important problems to be solved on the friction and lubrication in the cold striprolling is to know how much amount of lubricant is carried in the interface of the roll and strip. In order to facilitate its estimation, the authors proposed a new parameter ; t_d=η(U₀+U₁)/(αp₁) where η is viscosity of lubricant, U₀ peripheral speed of rolls, U₁ entry speed of strip, α angle of bite and p₁ yield stress of strip in plane strain. Experimental results on aluminium sheets show that oil quantity dragged in the roll bite increases with the value of t_ψ and the larger the t_d the smaller the calculated coefficient of friction. By applying the t_d on the results of further investigation, some other factors affecting the lubrication are revealed. They are strip thickness, reduction, roll diameter, previous heat treatment of strip, intensity of roll pressure and others.

View full abstract

It is well known that the surface condition of deformed metal is affected by lubrication. In this paper, experimental and theoretical analyses are presented for surface condition of the specimens which are initially polished to flat surfaces and compressed by flat tools with liquid lubricants. Simple models are suggested for roughening of surface under free deformation and flattening of the surface by tool. With these models and some assumptions, the relationship between compressive strain and the fraction of the true contact area is derived as a function of roughening rate of the free surface and the thickness of the initially trapped lubricant film. The experimental result agrees well with the theoretical result. A relationship among contact area, C.L.A. roughness and lubricant film thickness is also derived from the surface model.

View full abstract

Experimental studies were made on the effects of back pressure in the extrusion process of some brittle metals, in connection with their fracture characteristics under high pressure. Extrusion apparatus with solid back pressure was newly developed for this purpose. The results show that : (1) the fracture strain in compression test increases almost linearly with the surrounding pressure ; (2) a certain critical back pressure is present to prevent cracking in the extruded products of the brittle metals ; (3) the V-type crack is produced in the core of the magnesium billet, while in Al-Cu-Si alloy the fir-tree type crack also appears on the surface ; (4) the extrusion temperature as well as the back pressure has effect on the cracking of Mg-Al alloy ; (5) the prevention of cracking in tube extrusion is more difficult than in rod extrusion ; (6) the critical back pressure in high speed extrusion of Al-Cu-Si alloy is much lower than that in low speed extrusion ; and (7) the sintered copper powder is extruded without cracking at high speed by the application of appropriate back pressure.

View full abstract

Because of various non-uniformities of grinding wheel and high rotational speed of wheel spindle, forced vibrations are considered to be liable to occur during grinding operation. Influence of the unbalance and the eccentricity of grinding wheel on the amplitude of workpiece waves is studied theoretically, and results are verified through experiments using a surface grinding machine. In the theoretical analysis the coefficient of grinding force, that is the grinding stiffness, is introduced and a significant effect of the grinding stiffness on the workpiece geometry is pointed out. Influence of forced vibration on the surface roughness of workpiece is also investigated.

View full abstract