A simple theory is developed to explain the effect of hydrostatic pressure on the torsional strength of brittle solids which contain numerous flaws uniformly. The effect of hydrostatic pressure on the stress intensity factor for an internal crack is estimated, and the statistical treatment of brittle strength, developed by Weibull and extended by Oh et al., is used. The mean values of the torsional strength and fracture angle for solid- and hollow-cylinders under hydrostatic pressure are calculated. The torsional and tensile strength thus predicted are compared each other. The torsional strength and fracture angle of glass cylinders, which are supposed to contain only surface flaws, are also considered.
By applying a method of analyzing the plastic and quasi-elastic strain energies on the actual fatigue tests of carbon steel, the authors previously made clear the effect of temperature change in the atmosphere around specimens on the change in strain energy of one kind of carbon steel and obtained the ralationships between these energies and the structural changes appearing gradually on the surface of specimen. Furthermore, the dependence of the features of fractographs on the temperature conditions was also investigated. In the present experiment, the following four cases were adopted as the temperature conditions of atmosphere around the specimen: (1) the constant temperature at +20±2°C up to rupture of the specimen, (2) the constant temperature at -20±2°C, (3) changing the temperature from +20°C to -20°C after a certain number of cycles and (4) changing the temperature in the reverse way of (3). The main results obtained are as follows: (1) The circumferential temperature change brought in a certain stage on the fatigue process has important effects upon several fatigue phenomena including the strain energy changes and the failure repetitions of a specimen. (2) The change in the sum of quasi-elastic strain energy with the progress of fatigue can be expressed generally in Σqq=1(U)q=A·qk up to rupture of specimen, where A is the initial value defined by the ordinary elastic energy calculation, q is the number of stress repetitions, k is a constant depending upon the property of test material, the method of experiment and the circumferential temperature condition. Using this expression and the statistical values for the fatigue life (N) and Σqq=1(U)q obtained by the test under the constant temperature condition (+20±2°C), the fatigue lives under other temperature conditions can be inferred approximately. (3) If the sum of the plastic strain energy, Σqq=1 (Ep)q, is plotted logarithmically, two nodal points C and D appear. The point C appears to have relation with the occurrence of a slip band, and the energy quantity at this point was a constant about 10kg-m, independent of the stress amplitudes and of the temperature conditions. Therefore, it was found that the critical value of the energy accumulation in a specimen was related probably to the origination of the slip band. On the other hand, the point D appears to have connection with the appearance of intrusions, and the critical energy is considered to have relation to their appearance. (4) Fractographs collected in the vicinity of the central portion of the fracture surfaces were studied and it was found that their features were dependent upon the temperature conditions after change.
In order to clarify the effect of coaxing in 18 Cr-8 Ni austenitic stainless steel at room temperature, progressive stressing fatigue tests were carried out and the fatigue-hardened layers produced by coaxing or non-coaxing were examined by means of microscopy, microhardness test and X-ray diffraction technique. The results obtained are summarized as follows: (1) At room temperature the effect of coaxing is observed remarkably in 18-8 stainless steel, and under the progressive stressing conditions of stress increment of 0.85kg/mm2 and cycle increment of 1×105, 1×106 and 1×107 the rate of increase in fatigue strength are 10.8, 14.2 and 20.5per cent, respectively. (2) The strain-induced martensite transformation takes place during the fatigue deformation process at room temperature and the amount of martensite in the fatigue-hardened layer of specimen under coaxing is much larger than that of non-coaxing specimen. (3) The strengthening of 18-8 stainless steel caused by the effect of coaxing at room temperature is attributed to the strain-induced martensite transformation as well as the strain aging of the martensite and austenite during fatigue process, and the strengthening due to the former occurs mostly before the fatigue crack initiates.
It is well-known that the polymer impregnated concrete (PIC) has many excellent characteristics such as high compressive strength of over 1500kg/cm2, large elongation capacity, negligibly small water permeability and no-shrinkage. And it has also been reported by BNL and USBR that PIC shows good resistance to chemical attack. However, the chemical stability of PIC has been evaluated only form its weight loss, but scarecely from the loss of its mechanical strength. In this paper, the losses in weight and strength of a plain mortar, a mortar (PIC) impregnated with poly-methyl-methacrylate polymerized by thermal-catalyzer with BPO=5%, and a mortar impregnated with polypropylene fiber reinforced polymer were examined in order to obtain their resistance to chemical attack by 10%-HCl, 10%-MgSO4, 10%-CH3COOH and pure C6H6. The results showed that the weight loss of PIC (18.6%) was less than that of the plain mortar (24%), when attacked by 10%-HCl for 28 days. The losses in compressive strength of PIC and plain mortar after attacked by HCl were 67% and 97%, respectively; and after attacked by CH3COOH the losses were 26% and 57%, respectively. So, it can be said from the viewpoint of compressive strength that the chemical resistance of PIC is 23 times that of the plain mortar against HCl and 3.6 times against CH3COOH.
In this paper proposed are a new method of separating Kα-doublet in the diffraction curve and a new method of correcting back ground in X-ray stress measurement. The separation method of Kα-doublet is a computational one using Fourier transformation, and the correction method is also a computational one based on the assumption that the back ground is composed of two lines and one elliptical curve. The whole process of calculation can be carried out by a digital computer. The favorable results of the correction of back ground and the separation of Kα-doublet were obtained by the above mentioned methods in all cases experimented in this study.
In order to describe the elastic constants of the orthotropic plate having the unknown principal direction under a plane stress condition, we have to determine 4 more independent elastic constants in addition to that in the principal direction. If we try to evaluate these values by an ordinary tensile test, we have no alternative but to carry out the tensile tests in the directions other than the principal direction (the tensile tests of off-angle). However, there are always some difficulties in the off-angle tests. The present report describes a method of determining the principal direction and elastic constants of an orthotropic plate by carrying out the compressive tests on the ring specimens cut out from the plate. When an orthotropic ring is compressed in θ direction against its principal direction, the deflection δc of the ring will be approximately described as follows: δc=f(El, E', Et, sin2θ, cos2θ), 1/E'=1/2(1/Glt-2νlt/El) Here, El, Et, νlt and Glt are its Young's modulus, Poisson's ratio and shear modulus in the principal direction, respectively. Since the relation of δc-θ assumes the center of symmetry at θ=0°and 90°, we can determine the principal direction by finding out the symmetric point of the δc-θ' curve obtained by the compressive tests in several θ' directions against its reference axis. After the principal direction has been determined, compressive tests are carried out in the directions of 0°and 90°on the open ring samples made from the ring by cutting off one part. Then, the values of elastic constants are obtained from the values of δc and deflection δh of the open ring by the following procedure. First, the constants c1, c2 and c3 are calculated by the next equations. δ'c or h=2I/PR3δc or h c1=1/π[δ'h(0°)-δ'h(90°)] c2=15π/8[2δ'c(45°)-δ'c(0°)-δ'c(90°)]-10c12/62π/[δ'h(0°)+δ'h(90°)] c3=[δ'h(0°)+δ'h(90°)]/2π-c2/4 Here, P is the load, R is the mean radius of the ring and I is the moment of inertia of area. The value of elastic constants are obtained by the following equations. El=1/c3-c1 Et=1/c3+c1 E'=1/c3+c2 In order to separate νlt and Glt from the evaluated value of E', we have to introduce other tests. As far as we deal with usual 2-dimensional stress problem, however, we have only the values of El, Et and E' to be solved, being relieved of the trouble of separating them. The present method enables us to determine the principal direction and the values of necessary elastic constants by cutting one ring out from a small portion of a given original plate and testing it.