In this paper, the effects of forms of stress waves on the fatigue strength are investigated, using three low carbon steels. The forms of stress waves adopted are sinusoidal and nonsinusoidal ones.
The fatigue tests were made in plane bending and torsion under completely and partly reversed stresses. The testing machines used were the Upton-Lewis plane bending fatigue testing machine and the Upton-Lewis torsion fatigue testing machine. The eccentric mechanisms of these machines were modified so that they might produce expected forms of stress waves. The driving speed was 100 to 200cpm.
The test results were: (1) the fatigue lives at the cyclic nonsinusoidal waves, shown in Fig. 1, seem to be about the same as those at the cyclic sinusoidal waves, (2) the fatigue lives at the cyclic twin waves, shown in Fig. 2, are different from those at the cyclic sinusoidal waves, that is, the secondary wave in the twin wave has an effect on the fatigue life depending on the stress magnitude, and (3) the changes of the stress amplitude in the process of fatigue were similar for the all forms of cyclic stress waves under the constant strain amplitude.

View full abstract

The present study was made to examine high temperature strength of high Cr-Ni austenitic steels. Creep rupture tests and short time tensile tests at 650, 800 and 1000°C were conducted for AISI type 309 (25Cr-12Ni), 310 (25Cr-20Ni) and 330 (15Cr-35Ni) specimens to determine the effect of alloying elements, solution treatment and micro-structure on high temperature strength.
The results obtained are summarized as follows:
(1) Differences of tensile properties and creep rupture strengths between those of three grades of austenitic steels tested are not clear. But grade 309 possesses the highest rupture strength for 1000h at 800°C, while grade 310 has the lowest strength, and the latter shows the highest elongation at any temperature.
(2) As to the effect of alloying elements, C and N, which are interstitial atoms, have a remarkable effect for the improvement of high temperature strength, while substitutional atoms such as Cr, Si and Ni are only slightly effective for creep rupture strength.
(3) Because of the transformations of ferrite into sigma phase during test, its very effect is not clear at 650 and 800°C tests, while detrimental effect is observed at 1000°C where the transformation does not occur. Elongation at rupture of specimens containing ferrite at 1000°C is higher than that of fully austenitic specimen.
(4) Specimens containing 0.16% or more of C show a clear grain growth when solution treatment temperature is increased from 1000 to 1300°C while low C specimens do not show further grain growth over 1100°C. On the tensile properties at room temperature, the higher the solution temperature is, the lower is the strength and the higher ductility. But both of strength and ductility in short time tensile test at 650 and 800°C decrease with the increase of treatment temperature and remarkable embrittlement appears at 800°C.
Creep rupture strengths of higher C steels are improved as solution treatment temperature raises up to 1200°C. Lower C specimens do not show definite improvement of creep rupture strength, while elongation decreases remarkably. In this respect, creep rupture strengths are influenced more by behavior of C and N than by grain size.
(5) Type of fracture is transgranular at room temperature test and intergranular at 650°C and above, independent of steel grades and heat treatment temperatures.

View full abstract

The literature about the fatigue of a spiral spring can scarcely be found as compared with that of the coiled helical spring, and the report of Mr. Kawada and Mr. Watanabe is the only one in our country. In compliance with a request of a maker of spring balancer in this district, the fatigue test of spiral spring was carried out, and some data were obtained.
For this purpose, a testing machine with a shifting rod was used in our test, and the number of repetition in loading up to the fructure as well as the drop due to fatigue seen until the fructure takes place were carefully observed.
From the result of the test, it has been found that in case of a repetition number of 5000 times, the spring dropped 2-5% in torque owing to the fatigue, and that a further drop could not be seen after that. It was attempted to make an experimental formula for the relation between the load drop and the number of repetition. The formula can only be applied to a range up to 5000 time repetition.

View full abstract

Rotary bending fatigue tests were conducted at 500°C, using annealed 0.54% C-steel, to investigate the effects on fatigue lives by rest in the midst of a run and also changing of stress level in two steps. The fatigue lives are prolonged by twenty hours' rest at 500°C at various cycle ratios R. Especially, when the rest is given at R=70%, the fatigue life is made longer as much as 37% than the case where the fatigue damage is considered to be fully recovered by annealing effect during the rest, the specimen being equalized to the virgin material. This may be due to the partial recovery of fatigue damage and aging effect during the rest period at high temperature. The results of varying stress tests do not obey the Miner's criterion, the life ratio, or the so-called cumulative cycle ratio, L being larger or smaller than unity at different cycle ratio R. However, the damage ratio D for tests in which lower stress is applied first is always less than the one where higher stress is applied first. The relation of cycle ratios vs. life ratios and damage ratios show similar figure for both the rest test and varying stress test, and the difference of the damage ratios between each test corresponds to the strengthened rate during the rest period.

View full abstract

The variations of the deformation and the impact load are electrically measured on the specimens which are subjected to the repeated impact bending energies of 20, 30, 40 and 50kg-cm at the testing temperatures of 20, 0, -18 and -40°C. Then, if plotted in the load versus deformation diagram, deviations from the linear elastic lines take place, and the so-called hysteresis curves are obtained. The area of this closed curves represents the hysteresis energy which is the work to be given from the impact bending to material.
The results obtained show that the relations between the hysteresis energy and the number of repetitions to failure are expressed by the same straight line for 20, 0 and -18°C, while the relations for -40°C not expressed by the above line, if plotted each in logarithmic scale.

View full abstract

Plate specimens of mild steel were stretched so that the nominal stress would have a value equal to the yield point of the material. The result of the fatigue test in reversed bending with these specimens showed a slight improvement in fatigue strength over that of not stretched specimens. The fading of surface residual stress due to alternating stressing was investigated at the two stress amplitudes, one equal to the fatigue limit, and the other slightly above it. The residual stress was found to vanish after 10⁷ stress cycles or at fracture. The linear law of fading of surface residual stress was confirmed to hold for this case also.
Another specimens were shot peened on both surfaces by a centrifugal type machine. The fatigue test with these specimens resulted in a considerable improvement in fatigue strength. The fading of surface residual stress due to cyclic stressing was examined at four stress amplitudes, below and above the fatigue limit.
The analytical calculations carried out on the annealed steels were applied to the above two cases, and the contributions of residual stress to the fatigue limit and life were made clear. Especially, for the shot peened specimens, the ideal S-N curve which involves the sole contribution of residual stress was established. In the case of plastic tension, the role of residual stress in improving the fatigue strength is rather secondary, but the residual stress plays a leading part in the shot peening treatment. Taking the fading of residual stress due to alternating stressing into account, a convenient prediction of fatigue limit for the plastically cold worked specimens is possible from the fatigue limit of the original specimens. This procedure can be extended to the case of pulsating bending in a similar way.

View full abstract

A heat-resistant semi silicon carbide sagger body could be produced by using black silicon carbide. Experiments were made to examine the effects of kinds of aggregate, combining clays and other additives on thermal shock resistance and oxidation of silicon carbide.
The results obtained are as follows:
(1) On thermal shock resistance, “Kibushi” clay of high plasticity both as an aggregate and as combining clay gives good effects. Burned “Kibushi” clay used as an aggregate gives better effects at S.K. 14 than S.K. 9, but fused “Kibushi” clay such as corhart is not effective as an aggregate.
(2) Effects of other additives on oxidation of black silicon carbide by short-time heating are investigated through D.T.A., weight increase of black silicon carbide after burning and X-ray analysis. Copper carbonate and manganese fluoride increase the oxidation of silicon carbide, and borax which shows lower melting point prevents the oxidation. These results can be explained by the partial pressure of oxygen and their eutectic points with silica.
(3) Borax as an additive gives better effects not only on the above physical properties but also on preventing the oxidation of silicon carbide.

View full abstract