Low magnification electron fractographs of the cup and cone fatigue fracture showed parallel markings which are said to be characteristics of the stage I fatigue crack. However, high magnification fractographs of the etch pitted fracture surfaces revealed {100} fracture surface-<110> crack propagation direction and striations as well as {111} surface-<110> direction. This suggests that the opening and closing mode mechanism of fatigue crack propagation is operating as well as the shear mode mechanism in spite of the same fracture surface appearance as the stage I crack. A correlation of the fracture with the texture of material is also suggested.

抄録全体を表示

The effect of prestrain at room temperature on the relationship between non-ductile transition temperature and tensile speed has been examined on a low carbon steel.
The results obtained are summarized as follows.
(1) The transition temperature of the prestrained specimen rises with increasing prestrain at room temperature.
(2) The relation between the transition temperature T_B and the tensile speed V is represented by the following equation,
1/T_B=a-blogV
where a is dependent on the amount of prestrain and b is a constant.
(3) There are little deformation twins in the prestrained specimen tested in the vicinity of its non-ductile transition temperature. The micro-cracks remain slightly in the prestrained specimen tested at a low tensile speed but they disappear at a high tensile speed.
(4) A number of results seem to support the dislocation piled-up mechanism of the cleavage fracture.

抄録全体を表示

The diametral compression testing method, in which a simple penny shaped disk is loaded along a diameter until it splits, is convenient for determining the fracture strength of brittle materials. The maximum tensile stress at the center of disk σ_H is derived accurately by considering the Hertzian contact width as follows;
σ_H={1-1.15(b/R)²+0.22(b/R)³}σ_P; σ_P=P/πRt
where R, t and b are the radius, thickness and contact half width of the disk specimen, respectively. σ_P is the uniform tensile stress on the loading axis when the disk is compressed by the point force P. Furthermore, the correction in the diametral compressive strength for the effects of stress distribution and specimen volume is considered according to the Weibull's statistical theory, in order to apply this method to practical uses.

抄録全体を表示

A new approach to the diametral compressive tests using circular anvils is proposed on the basis of the analytical study given in the preceding paper. In this approach, the collapse at the contact edges can be avioded. The experimental results obtained by this method for several kinds of graphite and Italian Ondagata light marble are compared with those of the uniaxial tensile strength, and the discrepancy is discussed for a wide range of brittle materials.

抄録全体を表示

The diametral compression testing method proposed in the previous paper for simple measurement of tensile strength is needed to be modified from the view point of fracture mechanics when it is applied for biaxial stress problems. In this paper, a new macroscopic fracture criterion for brittle materials in biaxial stress states is discussed by considering their strength in the diametral compressive testing where initial cracking arises due to shearing stress in the mode II fracture and then propagation to final fracture occurs due to tensile stress. According to this fracture criterion, tensile strength can be derived from the uniaxial compressive strength and the diametral compressive strength. It is ascertained experimentally that the estimated values agree very well with the usual uniaxial tensile strength in a wide range of brittle materials such as graphite and marble.

抄録全体を表示

The exisiting data reported by several investigators for three kinds of structural steels and two kinds of Al-alloys were analyzed by a statistical model based on the Weibull distributions of mixed type. The main conclusions are as follows.
(1) In the case of steel, the following mixed distribution is well fitted to the fatigue life distributions at stress levels around the fatigue limit σ_w.
F(N)=p₁F₁(N)+p₂F₂(N)+p₃F₃(N), p₁+p₂+p₃=1,
where F₁(N) is a distribution which appears mainly at stress levels well above σ_w, F₂(N) is that appearing around σ_w, and F₃(N) is the distribution around and below σ_w. There are fixed relationships between the parameters of F₁(N), F₂(N) and F₃(N) and the stress, as well as between the probabilities of occurrence of F₁(N), F₂(N) and F₃(N), p₁, p₂ and p₃, and the stress. Namely, the shape parameters m₁ of F₁(N), m₂ of F₂(N) and m₃ of F₃(N) are virtually constant, and the scale and location parameters a₁, γ₁ of F₁(N), a₂, γ₂ of F₂(N) and a₃, γ₃ of F₃(N) show approximately linear relationships on log a-σ and log γ-σ or log γ-log σ diagrams. Moreover, the probabilities of occurrence p₁ and p₃ change almost linearly with σ on a normal probability paper, while p₂ shows a maximum at certain σ.
(2) In the case of Al-alloy, the following mixed Weibull distribution is well fitted to the fatigue life distributions at stress levels around σ_w, the fatigue strength at about 10⁷.
F(N)=p₁F₁(N)+p₂F₂(N), p₁+p₂=1,
where F₁(N) and F₂(N) are single distributions which appear at stress levels above and around σ_w, respectively. There are fixed relationships between the parameters of F₁(N) and F₂(N) and the stress, as well as between the probability of occurrence of F₁(N) and F₂(N), p₁ and p₂ and the stress. Namely, the shape parameters m₁ of F₁(N) and m₂ of F₂(N) are almost constant independent of σ, while the scale and location parameters a₁, a₂ and γ₁, γ₂ of F₁(N), F₂(N) show approximately linear relationships on log a-σ and log γ-σ or log γ-log σ diagrams. Moreover, the probabilities of occurrence p₁ and p₂ change almost linearly with σ on the normal probability paper.

抄録全体を表示

Propagation of part-through cracks (semi-elliptical surface cracks) by alternating bending stresses was studied with carbon steels of different strength level (S38C-N, S50C-QT). In the high stress intensity range, the propagation rate of part-through cracks could be correlated with that of through-thickness cracks in terms of the effective stress intensity based on the crack closure, irrespective of stress ratio and materials. In the low stress intensity range near threshold, on the other hand, no correlation could be found between the two, because the propagation rate-effective stress intensity relation in this range varied depending on several factors such as the geometry of crack, stress ratio, material property, etc.. These results seem to raise a question on the validity of using the threshold stress intensity as a design criterion.
The crack opening ratios were essentially the same for the two materials of different strength levels. This does not agree with the result reported in the literature, and the discrepancy was attributed to the difference of crack closure on the surface and in the interior of the specimen.

抄録全体を表示

The tensile creep-rupture properties of commercial pure copper were studied in vacuum of 760torr, 660torr, 470torr, 0.3torr and 9×10^-3torr at 320°C. The following conclusions were obtained.
(1) The creep-rupture times in both vacuums of 0.3torr and of 9×10^-3torr were five and two times larger than that in air, respectively. Therefore, the existence of the effect of vacuum environment was confirmed.
(2) With the decrease from 9×10^-3torr to 0.3torr, the coherent oxide film Cu₂O was observed on the surface of crept specimen. Besides, the exfoliating film of CuO was observed preferentially with the decrease from 0.3torr to 760torr.
(3) The creep test was interrupted at the time when the total strain reached 7 percent under σ_n=6.0kg/mm² in both 9×10^-3torr and 0.3torr. It was considered that the remarkable incease of the steady state creep rate in vacuum of 9×10^-3torr resulted from the increase of dislocation density or the associated lattice strain on the basis of the measurement of micro strain by the X-ray diffraction technique.

抄録全体を表示

Polarization measurement was conducted on iron, nickel, and austenitic stainless steels in hot caustic soda solutions containing ClO₃^- to investigate the effect of ClO₃^- on corrosion behaviors of those metals and alloys. Chlorate ions decompose into ClO^-, and this chemical reaction is stimulated by the addition of some metallic species and/or ions such as Mo (III) and Fe (II). Cathodic reduction of ClO^- to Cl^- is easy but ClO₃^- is electrochemically stable, so that cathodic reduction of ClO₃^- to ClO^- and/or Cl^- is difficult to occur directly. However, since ClO₃^- converts into ClO^- and Cl^- by consecutive reactions, ClO₃^- is detrimental to the passive film on metals and alloys in a caustic soda environment.

抄録全体を表示

Up to this time, only a little studies have been undertaken on the effect of sample size on arc resistance of electrical insulating materials. They are usually done with an assembly of electrodes in contact with the testing insulator but are scarcely done with an assembly of electrodes detached from the testing insulator.
In this study, arc resistance tests were carried out with the latter assembly under higher applied voltage and in a longer electrode gap length, that is, in a larger sample size. Furthermore, the results were compared with those of the smaller sample size in order to discuss the effect of sample size.
The experimental method used was similar to the ASTM method, but the plate electrodes were placed horizontally about 1∼3mm apart from the surface of the sample. The electrode gap length was 5∼10mm and the applied voltage was max. 50kV (commercial frequency 60Hz). To prevent the variation of arc current due to the change in electrode shape, electrode gap length, applied voltage, etc., the same arc current as that in the ASTM method was maintained by adjusting the current controlling resistance of the primary circuit of testing circuit. The arc intermittence time was also the same as in the ASTM method.
Generally, during the arc deterioration in these electrode assemblies, no arc failure point was observable. Consequently, the total time required for the deterioration path to progress and stop at the neighbourhood of the under part of tip of electrodes was regarded as the arc resistance time.
Furthermore, the experiments on a larger sample to examine the effect of sample size were done not only for the case that one electrode was placed on the testing insulator and the other was detached from the testing insulator, but also for the case that the air gap between the electrode and the testing insulator was changed.
The change in arc deterioration time of the sample with the variation of electrode gap length shows a V type curve for the both cases of the equal air gap and the different air gap, but the style of the curve differs a little depending upon the kind of samples. This result seems to indicate that the effect of sample size exists on arc resistance of electrical insulating materials.

抄録全体を表示