In repeated friction, the wear rate varies with increasing the sliding length, i.e. it is mostly very high at the beginning, but later decreases to a lower constant value. Mechanism of these transient phenomena of wear is discussed in this paper. Special wear machine, with which tests can be carried out in virgin and repeated friction, is used for rubbing of steel and copper on steel in dry and lubricated states. Effects of surface roughness and wear dust accumulated on rubbing surface are examined in several methods. Wear processes of undeformed and deformed surface layers by frictional force are compared. Then, it is concluded that the transient phenomena of wear are mainly affected by accumulation of wear dust on rubbing surface, not by formation of deformed and hard surface layer, or variation of surface roughness.

View full abstract

In this paper, the undetermined multiplier method in the calculus of variations under constrained conditions is applied to the problems of transverse vibrations (free and forced), statical deflections, bucklings and others of rectangular plates, some edges of which are clamped and the others are supported. The numerical examples of various kinds are solved in order to explain the process of this method.

View full abstract

We can recognize that there are two steps in the yield process of unplasticized P. V. C.. The first step of yielding is a process, during which the stress-strain curve reaches its maximum point and drops to the lower yield point ; we may call it the "initial yielding". The original local contraction can be seen in a part of the specimen. When such local contraction proceeded to a limiting value corresponding to the strain ε₁, a new yielding or contraction of the cross section is caused in the neighbouring part of yielded ; such yielding or the second step of yield may be called the "progressive yielding". The above two steps of yielding may be considered as analogous to the process of the upper and lower yield of the mild steel. Such phenomena of yielding are remarkably affected by the stretching speed ν_st. We established a relation between the stretching Speed ν_st and the yield strain ε₁, the velocity of propagation of plastic strain wave c, i.e. ν_st=c·ε₁.

View full abstract

Employing 0.14 per cent carbon steel, temperature and stress dependences of minimum creep rate and steady state creep rate in secondarv stage of creep are studied with particular reference to the characteristics of primary stage creep. The results are presented and discussed in terms of the empirical expression of creep behaviour of the material. The strain rate in secondary stage of creep is directly related with that of primary stage creep at arbitrary time. It facilitates the prediction of the characteristics of long term creep from the data of short term creep tests. Overall relation between strain rate, temperature and stress in secondary stage of creep is presented in the new "Master creep curve", which is established by taking applied stress on ordinate and new parameter including the properties of primary stage creep on abscissa.

View full abstract

In this paper, creep characteristics of mild steel which are to be determined by long term creep tests are interrelated with that of short term creep tests. It leads to the prediction of long term creep limit from the data of short term creep tests together with that of single long term creep test datum. The prediction is demonstrated by examples. Temperature dependence of creep characteristics of the material is also discussed, and a proposition is made on the estimation of creep limit at arbitrary temperatures from the data of creep tests conducted at a reference temperature.

View full abstract

The transient behaviour of creep strain, which appeared on partial removal of applied stress at the stage of secondary creep, was studied. The strain rate immediately after partial reduction in stress was found smaller than the steady creep rate, at which the material was expected to creep under the reduced stress, and then gradually increased and reached the expected value. It was also found that this behaviour was well interpreted on the basis of the theory of creep deformation, which had been proposed by the present authors.

View full abstract

Fatigue tests were made under the condition of constant deflection with a rotating type machine at temperatures of 450°C, 300°C and at room temperature by applying stress cycles in the range from 170 to 3000 cpm. The results were arranged on the basis of strain amplitude and discussed by referring to the results presented by Forrest and Tapsell. The conclusions obtained may be summarized as follows : 1. At a temperature of 450°C, frequency of repeated cycles has the following influence on fatigue life, (a) under a definite stress amplitude, the number of cycles to fracture is less at lower frequency and time to fracture has a definite value irrespective of frequency. (b) under a definite inelastic strain amplitude, (in other words, when the inelastic deformation in one cycle has a definite magnitude) the number of cycles to fracture does not vary with frequency. 2. At a temperature of 300°C, higher frequency results in decreasing the number of cycles to fracture in the range of amplitude near the fatigue limit. This perhaps results from anchoring of dislocations by solute atoms of carbon or nitrogen. 3. It seems that the general characteristics of fatigue at elevated temperature do not appear until the temperature is raised over the range of blue brittleness. Maximum of fatigue strength appears at temperature 300°C.

View full abstract

Concerning the mechanism of plastic deformation of metals at high temperatures, a theory is presented, in which the effect of stress and temperature history is embodied. After Becker and Orowan, the concept of internal yield stress is introduced in the present theory. The internal yield stress is defined as the critical applied stress at which flow can occur without help from thermal agitations, In the theory it is assumed that, in course of deformation, internal yield stress varies as the result of recovery as well as that of work hardening, and that rate of strain is determined by the present values of internal yield stress, applied stress and temperature. Therefore, the current value of internal yield stress, and hence the rate of strain, would be affected by the stress and temperature history experienced before. Treatments on application of the theory are shown to the results of creep test, tension test and relaxation test.

View full abstract

One of the authors presented a fracture criterion of metals in a general form previously, explaining the fractures under polyaxial stresses and the notch brittleness of ductile metals. To ascertain the validity and applicability of this criterion to brittle metals, we carried out fracture tests on thin-walled-tubes of gray cast irons under various ratios of combined tension and torsion, measuring all strain components independently. In this paper, only the deformation behaviours before fracture are reported. First, the distinguishing characteristics under fundamental tension and compression, which have not been seen in ductile metals, are explained by taking account of the notch effect due to heterogeneous graphites in perlitic matrix. Then, a number of specularities in elastic and plastic behaviours under combined stresses can be well explained analytically. On the basis of the above mechanism, the fracture will be discussed in a following paper by applying the fracture criterion.

View full abstract

Fracture tests on thin-walled-tubes of gray cast irons were carried out under tension and torsion combined. From the latest paper concerning the deformation behaviors, the Fisher's hypothesis was approved as reasonable. In the present paper, only the fracture behaviors are discussed by taking this notch effect into account and applying the fracture criterion at the tips of graphite flakes. The fracture stresses decrease with increasing the amount of torsion in comparison with those predicted by the maximum tensile stress theory. These experimental evidences can be accounted for by considering the additional stress in tensile direction at the tip of notch and the weakening effect of material strength due to polyaxiality. In order to compare with other proposed criteria, the fracture results under other kinds of polyaxial stresses by other reseach workers are cited and are all well explained consistently with our concept.

View full abstract

In order to obtain the correct data in plastic domain under various kinds of loads, it is necessary that the material is isotropic in the virgin state. In this paper, the author examined the relations among pure shear, tension and compression in plastic domain with a kind of brass that was considered to be isotropic in the virgin state. The experimental results are shown in Fig. 4. In this figure, the stresses are taken in the ordinate and the strains correspond to the amounts of total slip are taken in the abscissa. This results evidently show that the relation between tension and compression coincide well and pure shear is higher by about 14% than tension or compression in all plastic domains.

View full abstract

The authors examined, from the standpoint of rapid heating, the influence of induction hardening on the fatigue limit of circular specimens of induction-hardened, plain carbon steel containing 0·41% C. The results were as follows : [table] From the above data it has been revealed that : 1) Adequate heating time gives maximum fatigue limit. The fatigue limit 80 kg/mm² is the highest value ever attained with the plain carbon steel containg 0·41% C. 2) The adequate heating time nearly coincide with the critical time deciding whether ferrite disappear or not. 3) There are proportional relations between fatigue limit and surface hardness. Such relations have been recognized at lower hardness, also when electric furnace quenched, but have not invariably been acknowledged in the case of ordinary quenching. 4) Some commercial criteria of optimum heating time were induced in connection with the diffusion of carbon in pearlite structure.

View full abstract

A testing equipment and measurements are reported of the fatigue of metals under supersonic frequency. This equipment has an advantage over conventional fatigue machines that stress alternations up to 10⁸ reversals can be made within an hour or so. At the resonant condition, metal samples are subjected to alternating stresses above the fatigue limit by means of a magnetostriction transducer through an exponential brass horn. By this method the fatigue properties of various metals were investigated. To avoid temperature rise of the samples, water cooling was necessary. Under insufficient cooling, it was not possible to obtain large motions and high strains, pressumably due to the increase of internal damping in the samples. Even marks in the most strained part of the samples such as might be made by a very light scratch with a knife edge, had a considerable effect on the fatigue limit.

View full abstract

We have shown a general view of the extant theories on the orthogonal machining in steady state, especially paying attention to the existing limits of the complete solution. It is convenient for estimating these limits to investigate the existing range of the cutting resistance P/Tk related to the shear angle φ applying Hill's theorem to the rigid regions. We investigated the generalized case of Lee and Shaffer's solution with built-up edge and it is proved that the chip should be rigid contrary to Lee and Shaffer. We attempted some experiments using the lead specimens and the obtained relations between φ and λ-α coincided with the theory, but the values of P/Tk were scattered and slightly larger than the theoretical range as a whole. It seemed that φ was affected by the initial condition.

View full abstract

In this report, the effect of inclination of the measuring tube of the tester on the Shore hardness number was investigated. The experiment was carried out with three Shore scleroscopes of D-type and many specimens different in hardness or in material. It need scarcely be mentioned that the general requests for testing were satisfied in this experiment. Especially operating conditions of the tester were carefully maintained constant throughout this experiment without being effected by the speed of the operating handle and the force with which the specimen was pressed. As a result, the effect of inclination of the measuring tube on the hardness number was made clear. That is, on the whole, the behaviour of this effect is separated into three sections. Namely, 1st is a horizontal line section where the hardness number has no change, 2nd is a inclined line section where the hardness number dropes at constant rate, and 3rd is a horizontal linear section where the hardness numbers are scattered over a certain range. It is shown that the hardness of the specimen, and the structure of the tester, dimensions and finishing grade of its parts have some influence upon this effect. And, after all, it is represented that the effect of inclination of the measuring tube is so strong that it is necessary to attend to hold the measuring tube vertical in such cases as testing with the measuring tube hold by hand or by the swing arm holder.

View full abstract

Usually steel balls are used as indenters in Brinell hardness test and also in Rockwell hardness test. In Japan, tungsten carbide balls have been recently made in such size that we could use them as indenters in ordinary tests. In the present research, these balls were used to measure the hardness of hard metals and their permanent deformation was observed to determine the maximum hardness number which we could measure with such balls without their permanent deformation, and also the difference between hardness values of the same metals was measured with balls of different materials to determine the new conversion table of hardness. This table was compared with the already issued one such as ASTM table, and then the disagreement between them was discussed.

View full abstract

For the purpose of researching the strength and the rigidity etc. of box type girders used for overhead travelling cranes, full-sized model experiments were conducted. Two model girders were used. The top deck and two web plates of one model (girder 2) were reinforced by longitudinal stiffners, while those of the other (girder 1) were not. Load was applied on the top of the inner web plate at the center of the span of the each girder, which was supported at its ends. Results are as follows : (1) The stress distributions on the plate surfaces were generally irregular, while deflections, buckling loads, failure loads of girders and shapes of plates after failures were coincident with their calculated values. (2) The effects of longitudinal stiffners on the strength of girder 2 appeared clearly. (3) The torsion produced by the eccentric load had not remarkably bad effect upon the box type girders used for the present experiment.

View full abstract

Summery ; This paper reports on the test results, their data and also circuit elements of an Analog-Digital converter, constructed in our laboratory. This converter is a device which converts the d-c voltage to the binary code signal. The pulse counting method is applied and the converting process is as follows. (1) A sawtooth voltage is generated in the device. (2) The d-c voltage is compared with the sawtooth voltage. (3) The elapsed time between the starting point of the sawtooth and the point where it coincides with the d-c voltage is measured by chronometric pulses. (4) These pulses are digitized by the binary counter. This converter is of an all electronic type and we obtained rather satisfactory results.

View full abstract

The purpose of this paper is to determine the equivalent gains of non-linear devices to an amplitude modulated signal in the presence of random noise, and to evaluate how these equivalent gains depend on such important variables as ; (a) the amplitude of signal, (b) various non-linearities, and (c) variance of random noise. The investigation of these problems is of considerable interest in the study of non-linear control system. From an analytical point of view, the problem lies in an evaluation of the correlation function of the output of a non-linear device to the input. Then we develop an equivalent linearization technique, using these results.

View full abstract

The low-temperature quenching is greatly effective on improvement of the fatigue strength of steel, and the effect is more remarkable in notched specimen than in plain one. The reason for this was discussed on the basis of the strain-age-hardening of the ferrite, which took place during repeated loading. In spite of their low strength and hardness, the fatigue strength of low-temperature quenched steel is higher than that of tempered ones, because the low-temperature quenched steel has a higher age-hardening power. It may be concluded that the age-hardening power of the ferrite is one of the most important factors to control the fatigue failure as well as slip resistance of material does, and in notched specimen, the former is rather dominant factor than the latter. And it may be emphasised that the sharp notches in machine parts are not so harmful if the age-hardening power of the material is high enough.

View full abstract