Cast iron is widely used as the material for various machine parts, and the fracturing of the members made of cast iron, presumably due to thermal fatigue, is often experienced in practice. In spite of its importance as an engineering problem, however, the study of thermal fatigue of cast iron is very limited. In the present paper, the experimental results of thermal fatigue on spheroidal graphite cast iron are presented. Comparison has been made of the thermal fatigue tests made under uniaxial thermal stress and those made under multiaxial thermal stresses, using hollow cylindrical specimens and solid cylindrical specimens, and discussion has been made on the relationship between the strength of material under both types of thermal cycling. The conclusions obtained are follows: (1) In the case of thermal fatigue of brittle material like cast iron, the number of cycles to fracture is close to that to crack initiation. (2) Thermal cracks are made near the spheroidal graphites, and propagated so as to stitch them. (3) The strength of the material subjected to multiaxial thermal stress cycling can be predicted from the fracture life of uniaxial thermal fatigue under the same constraint of thermal strain by employing von Mises' equivalent total strain on the surface of the specimen.
In order to investigate the general view of tensile properties which are the basic mechanical properties for various rigid PVC plates, ten kinds of plates were tested according to the ASTM Tensile Testing Method (D638). The results obtained are summarized as follows: (1) The rigid PVC plates are divided into two groups from the standpoint of the tensile properties. The first group has lower tensile strength and larger elongation, and the second one has high tensile strength and smaller elongation. (2) The value of tensile strength (upper yield strength) σ1 of the test plates exists between 5.1 and 7.9kg/mm2, and the value of lower yield strength σ2 is almost constant, i.e. 4.2kg/mm2. The value of tensile strength at break σB is roughly identical with σ2. The value of modulus of elasticity E exists between 320 and 390kg/mm2, and a correlation between E and σ1 can be found. (3) There is a reciprocal-correlation between elongation δ and σ1. The nominal absorbed energy to tensile break W is approximately proportional to the increment of distance between grips ΔD. The value of mean stress σm, which is the divided value of W by ΔD, is almost constant and is approximately equal to σ2. (4) The percentage of ultimate elongation of rigid PVC plates is about 150∼160%, and the true tensile strength at break corresponding to the ultimate elongation is about 12∼13kg/mm2. (5) The behaviors of the necking phenomenon at its arising and those at the breaking differ also between these two groups.
The most general fracture criterion of cement mortar subjected to triaxial compression is experimentally obtained. The criterion is expressed in the principal stress space as a convex surface with space diagonal of its axis. The surface expands almost isotropically with an increase of hydrostatic pressure. The right sections of the fracture surface have a slightly bulged shape as compared with the equilateral triangle. The fracture criterion obtained by the authors can be approximated by Mohr's envelope of parabola. It is also found that the fracture criterion is independent of loading paths within the experimental range.
The most accurate method to determine the fatigue limits is to conduct the conventional fatigue tests actually, but it is very laborious, taking many hours and requiring many specimens. Then, various methods are proposed for estimating fatigue limits rapidly, one of which is temperature measurement of fatigue specimens under gradually increasing loads. Recently one of the present authors has devised a new accurate temperature measuring method, using the ceramic condenser, the capacitance of which changes with temperature according to the Curie-Weiss law. By this method, it is easy to measure the temperature of rotating specimen accurately without any contact part like slip ring, by searching for the resonant frequency of L-C circuit consisting of coil and ceramic condenser attached to the specimen. This method is readily applicable to the rotating bending fatigue testing machines of various types widely used in Japan. In this investigation, the behaviors of rising temperature in the rotating bending specimens of Ono type fatigue testing machine under gradually increasing loads were measured and compared with the fatigue limits which were determined by conventional fatigue tests. The most specimens were machined from S20C, S45C, SNC-2 and 14S-T6 as received bars except some that were of S20C which were fully annealed in vacuum after machining. The behavior of rising temperature due to increasing stress amplitude were expressed typically in three parts, the gradually increasing part, the linearly increasing part and the rapidly increasing part. And for all materials, the fatigue limits were very close to the stress amplitude, at which the temperature began to deviate from the linear part to the rapid part, with maximum error 2.8 percent (non-conservative side).
In this study, the stress concentration factors of a slotted curved beam subjected to uniform bending are reported. The experiment was carried out by the photoelastic method. Epoxy resin plates with thickness of 6mm were used as specimens, and a uniform bending moment was applied. The width of the curved beam and the length of the slot were maintained as constant. The radius of the curvature was varied in four steps (including the case of a straight beam) and the radius of semi-circle at both ends of the slot, and also the eccentricity of the slot from the center line of the beam were changed in three ways, respectively. The stress concentrations along the upper and lower edges, through the minimum cross section and around the slot were examined, and the stress concentration factors at four main points were measured for several cases of the test conditions mentioned above. Moreover, the ratio of stress of the curved beam to that of the straight beam at each corresponding point was also investigated. From the experimental results, the following conclusion has been made. (1) The s.c.f. at the upper or lower edge of the beam decreases or increases, respectively, with the increase of curvature. But the s.c.f. at the upper or lower end of the slot shows an inverse tendency to that at the edge. As to the effect of eccentricity, the approach of the slot to each edge raises the s.c.f. at the edge. This trend is identical for the s.c.f. at the upper or lower end of the slot, when the slot approaches the upper or lower edge, respectively. The maximum values of s.c.f. obtained in this experiment are 1.4 and 1.8, respectively, for the lower edge of the beam and for the upper end of the slot. (2) The ratio of stress of the curved beam to that of the straight beam is more than unity at the lower edge of the beam as well as at the upper end of the slot in all cases. This fact confines our attention to these points. (3) In order to secure that none of the stresses at several main points of the slotted curved beam become specially large, it is desirable to locate the slot on the centre line of the beam, avoiding the eccentricity of the slot.
The floor of rolling-stocks will be in the most severe condition when it has to support concentrated compressive load such as the weight of a woman carrying heavy luggage and standing on a high heel. Particularely is it the case with the floor of rigid plastic foam sandwich, the most delicate material with acoustic and thermal insulation effect, but not tough enough against the concentrated compressive load. The formulas given below give the relations between the rigidity of the face plate, the compressive modulus of elasticity of the core and deflection of the face, when the rigid plastic foam sandwich floor supports such concentrated load. w=P/8πD(a2-b2/2+b2logeb/a) (A) Ec=64Dtc/(a2-b2)2 (B) in which D=EFtF3/12(1-μ2) w: deflection of the face P: concentrated compressive load a: radius of loaded circular area of the core b: radius of bottom surface of a high heel Ec: compressive modulus of elasticity of the core tc: thickness of the core EF: flexural modulus of elasticity of the face tF: thickness of the face μ: Poisson's ratio of the face D: flexural rigidity from (B), we find a2=b2+8√D·tc/Ec (C) From fomulas (A) and (C), we obtain the maximum deflection of sandwich or the deflection of the face w. But w/tc must be less than the proportional limit strain of the core. The calculated value of deflection w was found to be 1∼1.5 times as high as the experimental value.