The wear of carbon steel in corrosive environments has been observed under the potential controlled conditions, and the mechanism of wear and friction in corrosive environments is discussed. In the wear of metals in corrosive environments, the fracture of adhesion points, the fracture by thermal fatigue, the rubbing off of corroded layer and the damage by corrosion fatigue are observed in accordance with the variety of corrosive agents, the corrosion intensity, the type of friction, the rubbing velocity and the contact load. In the oxidizing neutral media, the lubrication due to the corrosion products plays an important role in the behavior of wear, the minimum wear rate being obtained under the mild corrosiveness in the type of face contact. In the reducing acid media, the lubrication of corrosion products is not found because of their dissolution. In the corrosive wear of metals, the wear rate is much affected by the dissolution rate of metals because of the great effect of corrosion fatigue. The cathodic potential sufficient to protect the static corrosion is not adequate to the members under friction in corrosive environments, owing to the increase of the surface activity of metals. However, the protection from corrosion is not always expected to decrease the wear rate as is mentioned above.

View full abstract

This paper contains the derivation of the basic differential equations which govern the elastic-plastic behaviours of rotationally symmetric shells. Geometrical nonlinearity is considered in the same degree as in the nonlinear theory of thin elastic shells derived by E. Reissner. Using the assumption of plane stress, the Euler-Bernoulli hypothesis, and the Prandtl-Reuss equations for stress-strain relations in the plastic region, the basic differential equations are expressed in terms of the increments of the variables. These differential equations are approximated in finite difference forms and they are solved by an elimination method. These solutions relating to the increments are integrated numerically, and the final states of shells are obtained. As an example of the numerical analysis a U-shaped bellows loaded by an axial force is discussed, and it is ascertained that our results coincide well with the solutions obtained by another method for the elastic problem.

View full abstract

Many papers on the behaviour of materials under hydrostatic pressure have been published; but it is still not sufficiently clear whether hydrostatic pressure has any effect on the elasto-plastic properties of metals. This may possibly be due to the difficulties in the measurement of the stress and strain under hydrostatic pressure, where various measuring errors are likely to be involved. In the present investigation on torsion, in order to eliminate the largest conceivable of these measuring errors, namely those due to friction and distortion of torsion grips, a new apparatus is introduced. In this apparatus, the specimen, with a torque bar connected in series, is twisted in a high pressure vessel and both the strains of the specimen and torque bar are measured by electric-resistance strain gages. The effects of hydrostatic pressures up to 4000 kg/cm² on the deformation of aluminum and a zinc alloy and on the lower yield point of low carbon steel were investigated, and the results are as follows: (1) With aluminum, in both annealed and extruded states, there is no effect on the shape of stress-strain curve. (2) With low carbon steel, there is no effect on the lower yield point. (3) With a zinc alloy, there is evidence of a little rise in the stress-strain curve.

View full abstract

In a horizontal reservoir, partially supported by saddles, great stresses arise near the supports, and the magnitude of these stresses is closely related with the form of the saddle supports. In this paper, we discussed the strength of such a reservoir and its relationship to the form of the saddle supports, utilizing the fundamental differential equation of cylindrical shells, and we pointed out the importance of the form of the saddle supports. To decrease the magnitude of the stresses arising near the supports, we further made researches on the optimum form of the saddle supports. From this study, we knew that the deformation of such a reservoir was very large, and so on.

View full abstract

In this paper an elastic analysis of the expansion joints used in pressure vessels and pipe lines under internal pressure is presented. Although these joints are widely used, few investigations on their strength have been published and at present designers are employing the rough formulas of M. W. Kellogg Co. for practical designing. Therefore in this investigation the problem is treated by using a shell theory and a calculating method is established by which the stresses and deformation of these joints can be evaluated. Numerical results of stress distributions and deformation for some dimensions of the expansion joints are shown, and these results are compared with experimental ones. The general solutions for the case of both axial load and internal pressure can be obtained by combining the solutions in this paper with ones in the previous paper.

View full abstract

Two types of flexural vibrations of an arbitrary ring whose cross section is symmetrical with respect to the principal axis in the axial direction are investigated. The frequency determinant is derived by means of the Ritz's method on the basis of the theory of a circular cylindrical shell, which considers the influence of rotatory inertia and transverse shear deformation in addition to the usual membrane and bending effects. Compared with the already-known and the experimental results, the theory is ascertained to be valid for the prediction of the natural frequencies of various rings whose dimensional parameters are comparable to those examined in the paper. It is shown by an experiment that the well-known theory of a circular arc bar is invalid for a general ring in the region of large length-to-diameter ratios, not only concerning the out-of-plane twist-bending vibration, which was examined as to the case of rectangular cross section in the authors' previous paper, but also concerning the in-plane bending vibration of a ring whose cross section is unsymmetrical with respect to the principal axis in the plane of the ring.

View full abstract

Major trends in the cone type dynamic speaker design are the realization of the uniresonance cones. On the contrary, the flat plane speaker has wide sound radiation area making its tone characteristics preferable to the general type speakers, however, the main defects of the flat speakers are their multiresonance of the vibrating membrane. After analysis of the flat membrane vibration problems, driving methods to excluded the multiresonance were found; that is, if the membrane were to be driven by the continuously distributed force proportionate to the fundamental vibration mode amplitude of the membrane, no multiresonance is proven to take place. While, within some restricted frequency range, the uniresonance is expected by the some discrete concentrated driving points. This driving method was applied to some cases taking account of various boundary conditions, and the method was applied to some cases taking account of various boundary conditions, and the experiments resulted in good cancellation of the multiresonance. This paper includes the improving methods of flat speakers distortion by the use of the velocity feedback compensation.

View full abstract

In this paper, the balancing of flexible rotors by using flexible bearing pedestals is discussed. The rotor systems having flexible bearing pedestals are analyzed and the relations between the natural frequency of a pedestal system and that of the whole system are shown in a nondimensional form. By using the mormal modes and the relation between the rigidities of the rotor and pedestals, a balancing technique is discussed. Then, it is concluded that in the practical balancing by using flexible bearing pedestals, it is desirable mainly to correct at the first and the second critical speeds for the static and the dynamic balancing of the rigid rotor and to correct at the third and more critical speeds for the balancing as the flexible rotor. For example, three speed three plane balancing is theoretically and experimentally tested.

View full abstract

A simple method is proposed to estimate the surface pressure distribution on a cascade airfoil in the retarded flow, is compared with theoretical and experimental results, and is discussed in detail. It is verified that the method is applicable to the design of axial-flow turbomachines considering the suction performance and efficiency. There are two assumptions in this procedure: The first; the flow velocity on the surface may be obtained by superposition of the velocity on an isolated airfoil and the additional velocity induced by the cascade effect. The second; the induced velocity has a tangential component which decreases linearly along the chord line. As an application, it is shown that a turbo-pump should be designed as follows, that is, a rather small flow coefficient when optimization of both suction performance and efficiency is desirable.

View full abstract

Results of the investigation about the translation of performance conducted in the manufacturers' shop using turbocompressors for any gas are presented in this paper. Estimating from the test results, it is reasonable to classify three ranges according to the converting accuracy, i.e. Range I where relatively correct conversion is possible, Range II where the conversion may be conducted within allowable errors and Range III where translation is difficult. However, even within the Range III, comparatively correct translation results may be obtained, if an appropriate empirical correcting factor is introduced. The perfect conversion may be obtain using a closed test loop with substitute gas or mixtures. But the compressor test loop is not always adopted by the manufacturer from an economical view point. At present, it is useful to prescribe about the test method as stated in ASME Power Test Code PTC 10-1965, namely there shall be a mutual agreement between all parties to the test on each of code items and agreements shall be recorded prior to the test.

View full abstract

The X-type pseudo-shock was investigated, which appeared at comparatively high Mach number, and was compared with the λ-type pseudo-shock reported in the previous paper. The similarity and difference between them were discussed. At the duct center, the shock loss by the leading shock wave is smaller for the X-type pseudo-shock than for the λ-type one, but near the wall and in the subsonic region, the flow conditions are similar. The over-all total pressure loss is affected by the mixing of the flow near the wall with the main stream and the so-called shock loss is rather small. the effect of the boundary layer and the wall friction force on the static pressure rise of the pseudo-shock is also made clear. The simple model for the mechanism of the pseudo-shock is presented in the appendix.

View full abstract

A method is presented for predicting the characteristics of practical nonisothermal-fin arrays from the results of the previous reports on isothermal ones. According to this method, the effects of the fin-geometry, thermal conductivity, and temperatures on the heat-transfer characteristics have been quantitatively clarified. A heat-transfer experiment on a fin array with the fin efficiency of about 80% and the hemispherical emissivity of 0.67 has verified the validity of the present method of prediction.

View full abstract

There has been no satisfactory method to obtain the distribution of radiant heat on water-cooled surfaces of a boiler furnace. This distribution may depend heavily upon the flow pattern of burning gas. From this point of view, the present paper relates to a method of defining the distribution of gas temperature and radiant heat by estimating the flow pattern in a furnace, leaving aside a number of unknown factors of combustion and thermal characters of burned gas. Result of the present calculation for a gas burning boiler whose distribution of radiant heat was measured by Mumford and Corey gives a good qualitative agreement with their measurement.

View full abstract

Distributions of flow rates and flow stabilities in parallel evaporator tube systems have been experimentally investigated. The evaporator is composed of three parallel copper tubes of 4mm inside diameter and 40 m long. Experiments were carried out by using refrigerant R-113 as the fluid in pressure range from atmospheric pressure to supercritical pressure, in order to simulate phenomena in steam generators. It has been shown that distributions of flow rates, their hysterisis and onset of excursion phenomena can be estimated from pressure drop versus flow rate curves for single tubes. Moreover, description is made of behaviors of thermal oscillative instability which are observed at relatively low flow rate. In addition to the experimental results, by a theoretical analysis of a simple model, stability criteria in parallel tube system have been described.

View full abstract

The effects of turbulence on the flame propagation in the propane-air mixture in a closed vessel were investigated by analyzing the high speed motion pictures of flame, the ion-currents due to combustion and the change of the gas compositions in combustion zone. In the combustion chamber, the uniform turbulent conditions were produced by moving suddenly a perforated plate which was driven by the spring and the electro-magnetic control system. The main results are as follows: (1) The combustion zone of turbulent flame may be composed of a lot of flame elements which are similar to laminar flame, and the apparent burning velocity is increased by enlargement of combustion zone due to turbulence. (2) during the early stage following ignition, the flame speed is affected by both factors of mixture strength and intensity of turbulence. with further development of the flame, however, the intensity of turbulence becomes a dominant factor determining thee flame speed.

View full abstract

The problem of deposit-induced surface ignition in internal combustion engines is studied analytically with perturbative method with respect to reaction term, to predict the quantitative criterion of ignition of the mixture. As the fuel has a higher concentration, a larger heat of reaction, a larger coefficient of reaction rate, or a smaller activation energy, the ignition resistance becomes greater. The large heat capacity of the mixture, the low temperature of inlet air, the low temperature at the surface of burning deposit and the high engine speed decreas the tendency towards the surface ignition. The combustion temperature of deposit is proportional to the square root of the ratio of the concentration diffusivity of oxygen to the thermal diffusivity of the material of deposit. In the case of turbulent motions of the mixture, the use of turbulent diffusivities could lead to a similar conclusion. As for the effect of engine speed, the heat associated with viscous motion of the mixture must be taken into consideration.

View full abstract

The experiments have been carried out in order to examine if it is possible to super -finish the whole convex surface smooth with newly developed MPR stones without changing the profile of the convex surface. MPR stones consist of abrasive particles, copper powders sustaining abrasive particles and synthetic rubber as a bonding material. The results are as follows: The convex surface can be finished smooth with an MPR stone No.1 without changing the profile of the convex surface if the height at the center of the convex surface is less than 13.8μ in case of 11 mm in width, and with an MPR stone No.2 if the height is less than 11.9μ. These values become larger as the surface irregularities of the abrasive stone become larger.

View full abstract

The bending stiffness of bolted joint has been investigated experimentally by using a simple bolted cantilever, in which the bending load is applied parallel to the joint surfaces. It is shown from the experimental results that the behaviour of joint stiffness depends largely on the pressure distribution on the joint surfaces and on the horizontal spring action between the joint surfaces. The basic experiments for the horizontal spring action and the slip displacement between the joint surfaces were carried out furthermore. By using the measured values of the horizontal spring action, the method for the theoretical analysis of joint stiffness has been investigated and a theoretical expression for the horizontal bending stiffness of bolted joint was given.

View full abstract