A method for performing uniaxial thermal fatigue test is described. This method utilizes a testing machine with the closed loop servo system. The advantage of this method is that the strain- and stress-ranges and the ratio of constrained strain to thermal strain, that is, constraint ratio, can be set to any condition. Particular emphasis is placed on the following description:
(1) the relationship between shapes of heating coil and temperature distributions,
(2) the method of measuring strain,
(3) the system controlling constraint ratio by using the closed loop servo system,
(4) the difference between the stress-strain relation during thermal fatigue test obtained by the present method and that by the Coffin type testing method, and
(5) the presentation and discussion of experimental data on 0.16% carbon steel.

抄録全体を表示

The nonlinear-defect model, which had been previously proposed to analyze fracture and residual strain theoretically, was applied to the real stress (σ)-strain (ε) curve of steel. The results obtained are as follows
(1) When it is assumed that the cross-sectional area decreases and the stress-ccncentration appears during the final fracture resulting from propagation of a visual crack, a σ-ε curve becomes similar to that of a nonlinear-defect.
(2) The three regions of strain in a σ-ε curve were analyzed and discussed. The equivalent defects in the initial region R_i should be different from those in the final region R_f, and at the character of equivalent defects in the transient region R_t should change from that in R_i to that in R_f. The σ-ε curve can be given as
ε{1-β_(ε)ε}=A_(ε)σ
where β and A are parameters characteristic of each region.
(3) The above expression represents quantitatively the actual σ-ε curve of steel.

抄録全体を表示

As one of the series of studies on the influence of hydrostatic pressure on the mechanical behavior of polycrystalline metals, the transient behavior of plastic flow stress of prestrained polycrystalline aluminum at atmospheric pressure immediately after hydrostatic pressure soaking of 2000kg/cm² for 1 hour was discussed experimentally in the present paper. From the present study, the following conclusions have been obtained:
(1) At reloading after the pressure soaking, it was observed that the yield stress of the metal at reloading decreases transiently in comparison with the flow stress of the metal in the tension test at atmospheric pressure but it returns asymptotically to the original flow curve as the deformation advances. It was also observed that the pressure soaking has no influence on ductility of fracture of polycrystalline aluminum.
(2) It was observed that the yield stress at reloading after pressure soaking decreases remarkably with increasing the prestrain formed by pulling at atmospheric pressure. This becomes more and more evident for the test specimen with finer grains.
(3) As a structural change caused by pressure soaking of the material, the transient increase in number of moving dislocations in the metal resulting from decrease of frictional stress and locking force in the material can be considered. It is suggested that the yield stress at reloading after pressure soaking decreases because of the accelerative effect of moving dislocations.
(4) As for the strain-rate dependence of the influence of pressure soaking, it was obtained that the decrease of the yield stress under high strain-rate at reloading after pressure soaking is smaller than that under low strain-rate. This result was explained in terms of the increase of frictional stress caused by the strain-rate dependency of the material.

抄録全体を表示

This paper presents the dynamic stress-strain properties of styrofoams. Two ballistic pendulums were used in this experiment. The specimens, which had been bonded on the end surface of the pendulum being impacted, were loaded by the impact pendulum. The acceleration-time curves induced in the impacted pendulum were measured. The dynamic stress and strain were calculated by integrating these acceleration-time curves numerically. The effects of the strain rates and the number of impact loadings on the dynamic properties of styrofoams were examined and the accuracy of the test result was also discussed. The results obtained are as follows:
(1) The dynamic stress-strain properties of foamed plastics can be obtained with the accuracy of about 3 percent by this experimental method.
(2) The strength and stiffness of styrofoams increase with increasing the strain rate up to about 20/sec, beyond which they start to decrease.
(3) The stiffness of styrofoams is remarkably reduced by the first application of impact loading but does not change any more by its further successive applications.
(4) At the strain rate of about 35∼40/sec, styrofoams disintegrate in an explosive manner. Therefore, the styrofoams seem to have the critical deformation rate in the vicinity of this strain rate.

抄録全体を表示

The viscosity of glasses in the systems As-Se and As-Se-X (X=I, S) was measured by the beam-bending method in the transition region ranging approximately from the transition temperature to the deformation temperature. The plots of the logarithm of the viscosity versus the reciprocal of the absolute temperature gave essentially straight lines for all samples. The apparent activation energy for the viscous flow of glass was calculated from the linear relation shown in Figs. 1, 2 and 3 by the equation ΔE=R d(logη)/d(1/T), where R is the gas constant. The results are shown in Figs. 5 and 6. Also the delayed elasticity becomes evident in the transition region for the glasses in the systems As-Se and As-Se-S, as shown in Figs. 8 and 9.

抄録全体を表示

This investigation has been undertaken to determine the influence of microstructural factors (prior austenite grain boundaries, lenticular martensite leaves, non-metallic inclusions, etc.) on the characteristics of low-cycle fatigue cracks longer than 10μ (initiation period and sources, growing rate, propagation paths, etc.) in Fe-25% Ni-0.17% C martensitic steel, by conducting optical microscopic observations of the surface of smooth specimens. The fatigue tests were carried out at 20°C under a constant total-strain range of ±1.2% by alternating bending. The results obtained are summarized as follows:
A number of microcracks originate mostly from the prior austenite grain boundaries (γ_B), boundaries of martensite leaves (α'_B) and non-metallic inclusions, and some of these microcracks spread out by means of growing and joining. The growing rate of cracks along α'_B is almost equal to that along γ_B, but the propagating manners of these cracks are different each other. That is, a microcrack occurred in γ_B tends to grow into larger size than that occurred in α'_B which often remains as a non-propagating. As for the propagation path, the path along γ_B is slightly longer than that along α'_B, and the total length of these two paths accounts for above 80% of the whole crack length.

抄録全体を表示

One of the purposes of the present study was to investigate the effect of the program period on the fatigue life under programmed load. The load programs of two-stress repeated-block type were adopted in the experiment, in which the lower stress was set at a level below the endurance limit. The ratio of the number of cycles at the lower stress level to that at the higher stress level was fixed, and only the program period was varied in a wide range. Another purpose of the study was to investigate the possibility of fatigue failure by only the stresses below the ordinary endurance limit. Similar load programs were used in this experiment, but both of the stress levels were set at suitable values below the endurance limit. Carbon steel S50C was used and the tests were conducted under cantilever type rotating bending stress.
The following conclusions were obtained from the present investigation.
(1) The fatigue life under the program load composed of stresses above and below the endurance limit was influenced remarkably by the program period and the fatigue life showed a rapid decrease with the decrease of program period. Moreover, the effect of the program period became larger as the lower stress level became closer to the endurance limit.
(2) Occurrences of the fatigue failure caused by the stresses below the endurance limit were observed when the applied load was altered rather rapidly.

抄録全体を表示

In this paper, the authors discuss the failure process of rock which would result in the extremely wide fluctuation of fatigue lives observed in their fatigue tests under pulsating compressive or tensile stress.
Under the assumption that the process of fatigue failure is a stochastic process, it is indicated that the probability of survival of rock specimen is expressed as a linear combination of two exponential functions. This means that the failure process of rock is neither serial nor cumulative, but consists of two parallel Poisson processes of first order.
Other probable explanations of test results are assumed and discussed. The first of such assumptions is that each of rock specimen has a different rate constant of failure. This is discussed but denied from the theoretical point of view.
The second one is that the failure process is serial and cumulative. In order to verify whether the failure process of rock is serial or not, the fatigue test by the multiple-repeated tensile stress was carried out, using the same rock samples as in the previous fatigue tests. The test result shows that the failure process of rock is not a serial process.
Together with other test results reported in the previous papers, it is concluded that the process of fatigue failure of rock consists of two parallel Poisson processes of first order.

抄録全体を表示

The ASTM D 495-61 method has been regarded as a representative test for the high voltage-low current arc resistance of insulating materials. In this method, however, the electrodes are placed directly on the sample, so that the heat delivered to the fire spot at the tip of advancing track is of the orders of magnitude greater than that on the surface. Accordingly, the failure by the heat delivered to the fire spot appears to be emphasized in comparison with the failure by the heat delivered to the surface from the body side of arc.
From these view points, it is considered that, when the electrodes are detached from the surface, the formation of fire spot can be avoided and consequently the failure caused by only the body of arc can be taken out and examined under various conditions excluding the influence of electrode position and its state in order to evaluate the arc resistance. In fact, under such conditions, the formation of tracking is not observed and the burning part spreads out widely on the surface.
The experimental method used in this study was similar to the ASTM D 495-61 method, but the electrodes were placed horizontally about 1∼2mm apart from the surface of the sample as shown in Fig. 1. The elapsed time of failure were selected to be 10, 20, 30sec and so on. After allowing 30 minutes of cooling, the surface resistance was measured in order to detect the failure. Various types of thermosetting resins were used as the test samples.
The following results were obtained from these experiments:
(1) At more than 10∼40 seconds of failure time, the relation between the failure time and the ratio of the insulating resistance after failure to that before failure begins to deteriorate. From this relationship the characteristics of failure can be derived.
(2) For the evaluation of arc resistance, the measurement of insulating resistance after the failure is superior than that of breakdown voltage.
(3) The time dependency of insulating resistance during failure, which can be followed by the measuring circuit arranged between the electrodes (Fig. 1), shows that the process of failure is extremely complicated.
(4) The present method is found to be useful as a practical testing method from the comparison between the present results and the results obtained by various arc resistance tests.

抄録全体を表示