This paper reports the result of experiments made of the effect of grinding, lapping and barrel-finishing operations on the fatigue strength of quenched and tempered 13Cr stainless steel used for compressor valve. The reversed plane-bending fatigue tests were carried out by means of Schenk-type fatigue testing machine of 10kgm at about 1500rpm. The results obtained can be summarized as follows: (1) In longitudinal grinding the fatigue limit of roughly ground specimens was almost the same as finely ground ones, but in transverse grinding the latter was about 10% higher than the former. (2) In rough grinding the fatigue limit of transversely ground specimens was contrary to general expectations, nearly equal to that of longitudinally ground ones, but in fine grinding the former was about 20% higher than the latter. (3) The lapped specimens, when barrel-finished, get their surface roughness reduced by half, with but little increase in fatigue limit. (4) In the surface finishing operations under the conditions used in this experiment it was necessary that the residual stress both in the surface layer and the surface roughness should be taken into consideration as having influence on the fatigue strength.
For the purpose of performing the fatigue test of plastic materials under strictly controlled condition, the following four accessory test equipments were devised: (1) Reversed bending fixture (2) Reversed torsion fixture (3) Constant static strain maintaining device (4) Constant mean load maintaining device In consequence of the several fatigue tests performed of plastic laminates reinforced with fiberglass it has been clarified that the functioning of these accessory test devices are satisfactorily efficient.
The authors have studied on the problem of the relative effect of work-hardening and residual stress in the machine-worked surface layer on the endurance limit of 18-8 stainless steel by the rotary bending fatigue tests which consisted of three series as follows: (A) Series: Tests of the specimens with machine-worked layer and negligible residual stress. (B) Series: Tests of the specimens with thermal residual stress and without machine-worked layer. (C) Series: Tests of the specimens with machine-worked layer and thermal residual stress, corresponding to the synthesis of the two Series (A) and (B). The following conclusions may be drawn from this investigation. The endurance limit depends on three factors, such as surface hardness, the depth of the machine-worked layer and its residual stress. Comparing with the endurance limit of the standard specimen with no machine-worked layer (accurately with a slightly polished layer by 0/4 emery paper) and no residual stress, increasing ratio of the endurance limit is simply given as the sum of increasing ratio resulting from the work hardening and that from the residual stress. The increasing ratio by work-hardening is obtained by multiplying that depending on increase of surface hardness by the amendment coefficient which is a function of the depth of the machine-worked layer.
Hot corrosions of marine engine steels have often been experienced in operations with fuel oil C. Fuel oil C has approximately 40∼400ppm vanadium, 50ppm Na and 3∼3.5wt% sulphur. The controlling factors against its corrosive effect are oil ash deposits containing V2O5, Na2SO4, SO3 at elevated temperature 500 to 700°C, pulsating combustion pressure, and so on. It is the aim of the present paper to describe an aluminized process as protective to hot corrosions of heat resistant steels on marine engine service. The results obtained from this study were as follows. (1) None of the alloys used in the typical marine engine and investigated in the present experiment has satisfactory resistance to vanadium attack corrosions, although austenitic alloys-14 Cr-14 Ni-2W and territic alloy-2 Si-11 Cr-1 Mo, have been found as having comparatively high resistance. In general the resistance may be the function of the base composition, i. e., the Ni; Cr; Fe. (2) At service temperature, 600°C, the alloy's resistance to vanadium attack as well as to the scaling loss and such mechnical properties as fatigue life, shockwearing, etc., has been remarkably improved by the aluminized process. (3) However, it is less effective if micro cracks are initiated in the intermetallic compound formed on the surface of base material during the pulsating heats. It is likely that the intermetallic compound layer is stable against crack-initiations during the heating process if proper base materials, such as austenitic alloys 14 Cr-14 Ni-2 W, are selected, whose thermal expansion is approximate to that of the surface layer.
A method to evaluate the recrystallization of tungsten wire was studied by measuring its tensile strength at room temperature after heat treatment. The tungsten wire was heated by electric current in the atmosphere of dissociated ammonia gas. We found a suitable method to determine the necessary electric current in order to heat the tungsten wire up to the desired temperature. The results are summarized as follows: (1) The necessary electric current to heat the tungsten wire up to the desired temperature in the range between 1200°C and 2000°C will be estimated approximately using the following formula; i=71.2MG·V/T·l (2) The tensile strength of the tungsten wire decreased rapidly by short time heat treatment, this means that recrystallization takes place for the most part in a short time. (3) The tungsten wire with low recrystallization temperature has low tensile strength at room temperature, and shows exaggerated grain growth after heat treatment. (4) There is close correlation between the tensile strength before the heat treatment and the tensile strength after the heat treatment.
A study was made by photoelastic analysis of determination of uniaxial stress and stress concentration of epoxy cast model with Dumb-Bell type subjected to shrinkage. The model was cast at the temperature of 140°C and cooled at temperature reduction rate of 5°C pre hour. The value of the stress at freezing temperature was (1.19±0.17)×10-2kg/mm2. Stress concentration factors for R/d obtained from these experiments are compared with the results calculated from Heywood's projection formula. In the experiments the observing errors of fringes are less than 0.1 fringe order. The result is applicable to derermining the shrinkage stress distribution and stress concentration in complex epoxy cast model.
In this report, a convenient method for the estimation of program fatigue life is proposed. The concepts of basic load program, program S-N curve and equivalent load program are introduced. Under some conditions, it is predictable that (1) in log-log diagram, the program S-N curve is parallel to the conventional S-N curve, (2) program S-N curve can be estimated from one program test, in other words, it can be estimated if one point on a program S-N curve is determined experimentally, (3) according to Miner's low, the fatigue life of one type of specimen under a load program coinsides with the fatigue life of another type of specimen under the equivalent load program, and, therefore, the program S-N curve of one type of specimen can be estimated from only one test of another type of specimen under the equivalent load program. To make sure the appropriateness of these predictions, three types of specimen were tested under some series of load programs and equivalent load programs, and it has been observed that the predictions are valid with in the range of the test.
In order to clarify the relations between the progressive stress and fatigue strength of steel wires, a fatigue testing machine was devised and constructed, utilizing the buckling which was produced when one end of the wire was rigidly fixed and the other was hinged without friction. This machine has several advantages. It has but small resistance at the hinged end, and it is fit to perform the test under progressive stress by application of a simple device. The fatigue tests were carried out under the logarithmically increased stress and it has been found that the rupture stress is approximately expressed by the following formula: logσR=logσw+K(σR/NR)1/4 where σR: rupture stress, σw: fatigue limit, K: constant and NR: the number of cycles to failure.
The mechanical behavior of rock mass under load is influenced by the nature of discontinuous geological planes rather than that of rock material itself. Model tests were made with idealized models with discontinuous planes orientated vertically and horizontally. The results obtained are as follows: stress distribution shows concentration along the vertical central axis of the loading plate. The state of fractures is different from that of the homogeneous and isotropic elastic solid. Cracks are initiated either by partial tension or partial shear, and then they are propagated toward lower parts of the foundation. Hence it is necessary to examine sufficiently the nature of discontinuous planes involved, in order to understand the characteristics of such discontinuous rock foundation.
Poisson's ratio and Young's modulus are two important material constants for isotropic elastic materials, but there are usually very difficult problems regarding Poisson's ratio since high accuracy is required for its measurement. And also, for orthotropic elastic materials it is required that these material constants will be determined in line with the respective principal structure direction. In this paper, we describe the methods of measuring Young's moduli and Poisson's ratios for isotropic elastic material and orthotropic elastic materials using biaxial tensile tester. For isotropic elastic materials the Young's modulus and Poisson's ratio can be measured by strip biaxial testing, where the stretch ratio is variable to the axis of X1 and the stretch ratio to the axis of X2 is kept in unity. And the material constants are given by the following equations. E=1/ε1σ1(σ12-σ22) ν=σ2/σ1 where E is Young's modulus, ν is Poisson's ratio, σ1, σ2 are stress to the axis of X1, X2 and ε1 is strain to the axis of X1. For orthotropic elastic materials these material constants are measured by three operations, that is, the strip biaxial tensile testings where the extension direction is selected to 0°(Lateral), 90°(Transverse) and 45°from one of structure principal axis. And the material constants are given by the following equations. EL=1/εL(σL-σTσL'/σT') ET=1/εT(σT'-σL'σT/σL) νLT=σL'/σT'ET/EL and GLT=σθ1/εθ1-EL(1+2νTL)+ET/4(1-νLTνTL) where EL, ET are Young's moduli of lateral and transverse directions respectively, νLT is Poisson's ratio, GLT is shearing modulus, σL, σT are stresses in lateral and transverse axis and εL, εT are the strains. And the prime means the case of transverse direction tensile, and suffix θ1 indicates the case of 45°direction tensile.
Concerning arc resistance of electric insulating materials, development of a reasonable and simple method of testing was recommended for consideration. Taking the deterioration in the low voltage high current arc to be due to the deterioration of its energy source, which was thought to be the low voltage high current arc itself, the bunsen burner was used for the thermal energy source, considering that the equivalent energy source was not to be necessarily confined to electric thermal energy source, but the use of simple energy source such as chemical flame might as efficiently substitute. The test insulator was held horizontally at a constant distance from the nozzle of the burner, and was subjected to deterioration. Then the temperature of the part of the flame which was in contact with the insulator was fixed at 1000°C and at the distance of 50mm. from the nozzle in this condition. The length of the flame was kept at about 70mm. constantly. And then the degree of deterioration was examined by measuring the post-test surface resistance. The two surface resistances, pre-test and post-test, were compared, and the ratio of the lowering determined the character of the deterioration and of the arc resistance. The above mentioned subject was also studied from the point of arc resistance test by low voltage high current. The comparative studies were carried out smoothly. The measurement agreed fairly with the results of various other arc resistance tests by low voltage high current. The present test has thus proved a practical method.