The fatigue limits due to rotary bending for specially prepared mild steels containing silicate inclusions were obtained after the specimens were preworked by statical tensile deformations beyond the yield point. The results obtained are as follows; (1) Comparing with the results for annealed states, the fatigue limits decrease few percent for small strains, and increase for large strains of preworking. This tendency is similar to that obtained previously by Fukui and Sato for ordinary carbon steels, and the effect of the presence of inclusions can not be found. (2) Accordingly, in the preworked soft steels studied in this investigation, the presence of inclusions appears to have little influence on the fatigue strength. (3) 196-1 steel contains greater quantity of inclusions than the other steels. The fatigue limit of this steel, preworked to about 2% tensile strain by applying the tensile stress of 27.0kg/mm2, falls by 10% as compared with that for the annealed state, and then recover the same value for preworking of 40kg/mm2 stress and approximately 7% tensile strain. (4) Since the fatigue cracks initiate at inclusions, the fracture surface of a specimen which was subjected to a high stress level is very rough, and numerous micro-cracks are found on the outer smooth surface of many specimens near the fractured section.
Several fatigue tests were conducted on the low carbon steel at elevated temperatures. First of all, the fatigue strengths under rotating bending stress and the torsional fatigue stress were compared. Then the rotating bending fatigue tests were carried out with stepwise varying stress at 480°C, and the effect of the varying stress on the fatigue life was discussed. Finally, the torsional fatigue tests were carried out to see the effect of the variation of temperature on the fatigue strength. The mains conclusions from this study are as follows: (1) The fatigue strength of low carbon steel shows an approximately similar trend against the temperature ranging from room temperature to 500°C, irrespective of whether it is subjected to the rotating bending stress or to the torsional fatigue stress. But the temperature giving the peak of strength is somewhat different in each case. (2) When the low carbon steel is subjected to the stepwise varying stress at 480°C, Miner's rule is acceptable to hold as the first approximation. There is no significant difference in strengths obtained in the tests with increasing stress amplitude and with decreasing stress amplitude. (3) Miner's rule also holds in respect to temperature, when it varies cyclically between the two temperatures, in the range higher than 420°C, during the fatigue test. However, when the lower temperature approaches the temperature of blue brittleness, a large deviation from Miner's rule may occur, and a prolonged fatigue life will be obtained perhaps due to the strain aging.
It was found that the surface damage of annealed mild steel subjected to the cyclic straining in the plastic range is almost slip band cracks. A model is proposed to explain the fact that the slip band cracks are formed independently of the magnitude of stress if there are some large and strong obstacles in the microstructure, but only in the range of low stress if there are none such. With repeated cyclings, the screw parts of piled-up dislocations loops on the two neighboring parallel slip planes cross slip each other and annihilae themselves. As the result, an intrusion is formed, and the pile-up is made of edge dislocation loops enough to initiate a crack. Next, the intrusion is connected with the initiated crack, and hence a slip band crack is formed. Under high stress fatigue, no slip band crack is formed, if there is no large and strong obstacle, because the concentrated tensile stress is reduced by the pile-up of edge dislocations of opposite sign and the initiation of crack is thereby arrested.
The cumulative cycle ratio is usually larger than unity, when steels are tested under gradually increased stress amplitude, and is less than unity when tested under decreased stress amplitude. This fact seems to indicate that the strengthening effect, due to the work-hardening, during the test, prevails in the case of increased stress amplitude. What, then, is expected on the steel which was heavily work-hardened beforehand, when it is tested under the same stress conditions? With respect to this problem, the fatigue tests were carried out on the steel wires which have received different treatments, under the stepwise increasing and the stepwise decreasing stress amplitudes. The results on the heavily cold-drawn steel wire showed that the average of the cumulative cycle ratio is very close to unity under the increased stress amplitude, and is approximately 0.7 under the decreased stress amplitude. Therefore, it is clear that the heavily work-hardened material has no capacity for further work-hardening when it is subjected to the increased stress amplitude and the material is rather weakened or softened by the decreased stress amplitude. This inference was also confirmed by the examination of the hardness of specimens fractured under the respective stress conditions. On the other hand, the results on the steel wire, which was annealed at 400°C in vacuum after the same cold drawing, showed the increase of the cumulative cycle ratio more than two, in both cases of the increased and the decreased stress amplitudes. Therefore, it is inferred that the work-hardening can still take place for this annealed one. The increase of the cumulative cycle ratio in the case of the decreased stress amplitude is probably the result of the work-hardening, that particularly was occasioned by the large stresses applied at the initial stage. However, no more confirmation of the results was available from the examination of the hardness of the fractured specimen on this annealed steel wire.
The usual method of freezing photo-elastic stress patterns is to heat the specimens first and then cool them down very slowly without disturbing the stress patterns obtained at the high temperature. However, if a method which fixes stress patterns at room temperature could be developed, the method would be more useful. For instance, it could be applied to the study of the distribution of stresses in an object placed in a flow, such as a model in a wind tunnel. In this paper, an experiment which fixes stress patterns at room temperature during the curing process of resins was studied. The summary of the experimental results is as follows; (1) The fringe patterns fixed by this method show good coincidence with those obtained by ordinary photo-elastic experiments. (2) The fringe order is proportional to the principal stress difference in an elastic body. (3) The linear relation between the load and the fringe order is retained for a wide range of loads. (4) The fringes are not disturbed by cutting the specimen. Therefore, this method, as well as the heating method, can be used for the three-dimensional analysis of stresses.
A study of the machinability of leaded brass, with copper content ranging 52∼60% and additional 1∼4% lead ingredient, is reported in this paper, performed with respect to cutting resistance, surface roughness and chip formation among others, with particular reference to the part of copper content and lead content in affecting the machinability of the leaded brass piece which was used in the experiment as free cutting brass. The results are summarized as follows: With lead content as small as 1%, cutting resistance is smallest with brass of 58∼59%, copper content where phases α and β are co-existing, while cutting resistance is largest with brass of 53∼56% copper content where the phase is in transition stage from α+β to β. On the whole cutting resistance decreases with increase of lead content, and keeps up nearly constant value regardless of copper content. Cutting resistance decreases with the rise of side rake angle of tools and keeps up nearly constant value at the side rake angle 20°. With increase of lead content to 3∼4% cutting resistance becomes independent of the side rake angle. The roughness of the cut surface is fairly within the theoretical roughness of the feeding tool mark. Either copper content or lead content seems to have nothing to do with the surface roughness.
The present investigation was undertaken in order to describe the recovery characteristic of elongation of unplasticized polyvinyl chloride stretched at temperatures above its glass transition point (Tg). The experimental procedure was as follows; after the specimens have been stretched up to elongation ε1, at temperature T1(>Tg), they were cooled in cold water. They were then annealed at temperature T2(≤Tg) and the recovery of elongation was measured and its values are shown in Figs. 8 and 9. The data suggested that there was no particular recovery equilibrium in each curve but the strain might recover perfectly after sufficient duration time except at temperatures near the glass transition point or below. Approximate master recovery curves were drawn, as shown in Fig. 10, on the assumption that the method of reduced variables were valid to these experimental results. A simplified expression is also shown of the recovery characteristic derived from the results obtained at definite intervals.