In order to investigate the characteristics of fracture stress curve of steel by notched bar tension specimen, we must clarify first the range of notch shapes which would cause an initial crack not at the contour, but at the center of notched section. Authors carried out extensive experiments and obtained some results, which are summarized as follows:
(1) Initiation of center crack of fibrous type could be observed after sectioning longitudinally the specimens which were strained and unloaded just before their fracture by means of very careful and minute adjustment of load, using a rigid and small tension testing machine of screw type (capacity, 5tons) (cf. Photos. 1-4).
(2) It was proved that “rim effect” could be recognized conspicuously on the flow stress curve of the specimen in which the center crack was initiated. By observing this rim effect, the range of notch shapes that would produce center crack was determined in detail as shown in Fig. 3.
(3) It was also proved that no contour crack was found in the case where center crack would initiate and that this phenomenon might also be observed in the reverse way. Moreover, we observed that contour crack grew slowly in comparatively a long period and was quite stable, and that, on the contrary, center crack was very unstable and accordingly the specimen broke just after its initiation. The fracture mode was shear at the contour in almost all specimens and fibrous at the center.
(4) It was presumed that micro metallurgical defects of steel might influence strongly on the initiation of ductile crack.

View full abstract

In the present report, we describe the results of the fatigue strength of the spring steel (Sup 3) under the stress waves which are composed of superposition of the two stress waves, with the mean stress 40kg/mm².
The primary waves are about 70cpm and the secondary ones about 730cpm. The former has a large amplitude and the latter a small one (the ratio: 0-0.8). The specimens which have been treated in ordinary heat-treated conditions are tested in the full size.
We have summarized the results of this experiment as follows:
1) Designing the laminated spring, we should pay attention to the frequency of the fatigue test.
2) When the secondary waves are superposed on the primary ones in such an extent as described in this report, we can estimate the fatigue strength as the following, that is, a) the stress amplitude is of the primary stress waves superposed on the secondary ones, b) the cycle of the repetition is that of the primary waves.

View full abstract

This paper presents the results of compressive fatigue and creep tests on piano wire coiled spring, setting-treated at elevated temperature conditions. The specimen is treated for 4min at a temperature of 250°C under the compressive load of 40kg, for this condition gives the material the highest values of proportional limit and spring constant.
The results obtained are as follows;
(1) The fatigue limit of setting-treated spring is about 14% higher than the conventional bluing-treated one.
(2) The value of deformation limit is nearly constant for the value of τ_a/τ_m from 0.6 to 1.0, while it becomes higher as the value of τ_a/τ_m becomes lower.

View full abstract

Some torsional fatigue tests were carried out to obtain the fundamental data of the fatigue notch effect and the fatigue size effect on the large steel shaft with shoulder fillet.
Two kinds of materials were used for the tests. They were 0.37% carbon forged steel and chromium molybdenum forged steel, and their chemical compositions and mechanical properties are shown in Table 1 and Table 2. The working sections of the specimens are 50mm in diameter for the large shafts, and 7mm for the conventional small specimens, the forms and dimensions are shown in Fig. 1 and Table 3.
As the tests results, the following aspects were obtained.
(1) Good agreement was observed between the measured stress concentration factor α' (by means of electric resistance wire strain gauge) at fillet radii on large shafts and α presented by Peterson excepting a case of 1.5mm in fillet radius.
(2) The fatigue limit strengths at 10⁷ cycle are shown in Table 4. On the size effects, the reduction of the fatigue strength of 50∅ straight shaft compared with that of 7∅ was 13.5% for carbon steel, and 22% for chromium molybdenum steel. In case of the specimen with shoulder fillet, the reduction was 17% for carbon steel, and 13% for chromium molybdenum steel when α was about 1.5.
(3) The reduction between the fatigue notch factor β₂ (based on 7∅ straight shafts) and the stress concentration factor α is shown in Fig. 3. From this figure it seems that β₂ is considerably smaller than α in case of carbon steel.
(4) From the diagram of η (the fatigue notch sensitivity factor) plotted against notch radius as shown in Fig. 5, it seems that η of annealed carbon steel falls considerably lower than the average curve presented by Peterson.

View full abstract

Rotating bending and reversed flexure tests on mild steel coated with paints were carried out. Paints used were vinyl and oily ship bottom paints and zinc-rich paint (ZRP). The effectiveness of the impressed current cathodic protection on the painted surface was also examined. In the rotating bending, the fatigue limit was obtained for the specimens coated with vinyl or oily paint, but in the reversed flexure tests, the S-N curves of the painted specimens were similar to those of the bare metal in sea water and no definite fatigue limits were obtained. The difference in the S-N diagrams of the two test methods was probably due to the large deformation at the side edges of the specimen in the reversed flexure test. The S-N diagram of the ZRP coated specimens at relatively high stresses nearly coincided with those of the bare metal in the air, but at the stresses below the fatigue limit in the air, specimens fractured at a nearly equal number of reversals of stress. This might be due to the consumption of the protective ability of ZRP, judging from the change of the electrode potential of the specimens. The fracture of the specimen was protected by repeated painting. The application of the impressed current cathodic protection togather with the oily paint was very effective and the current densities necessary to protect corrosion were reduced to about one-twentith or-thirtith of those of the bare specimens cathodically protected.

View full abstract

In this paper we have discussed on the relation between the stress and Bauschinger strain, which means the plastic strain caused by a reversed load in a transient-softening region.
The authors consider that Bauschinger strain can be caused by the random distribution of barriers against the movement of dislocations. The expression for the stress-strain relation in this region is established by the following assumptions: (1) The distribution of strength of barriers is normal, and (2) the effect of increase or decrease of the numher of dislocations on the strain is neglegible. According to this “random-barriers theory”, the Bauschinger effect and some incidental phenomena can be explained considerably well. Moreover, the stress-strain relation derived by the theory agrees well with the experiments on low carbon steels. The authors conclude that Bauschinger strain is mainly caused by the random distribution of barriers especially in materials containing many obstacles such as internal stresses of various scales, segregates, inclusions, pile-up of dislocations and so on.

View full abstract

In this paper, the torsional strength and the flexual rigidity of flexible shaft are measured, and then the effect of the constructed wire diameter and the number of coiled wire in the outer layer are obtained. As the results of the test, two types of faliure, that of breaking and that of buckling have been indicated. The breaking failure is caused basing on the strength of constructed wire and the number of wire.
The following are the results obtained;
(1) The torsional strength of flexible shaft depends on the wire of the outer layer.
(2) The breaking torsional moment T_max are obtained as follows; T_max=πd_m²σ_Bn_m(D-d_m)/8k in which d_m the diameter of outer wire, n_m the number of wire and k the experimental coefficient.
(3) The torsional buckling is produced in the case of long length of shaft.
(4) The critical torsional buckling moment T_cr is obtained as follows; T_cr=k₁B/l in which B the flexual rigidity of shaft.
(5) In the case T_cr>T_max, the flexible shaft is broken, and when T_cr<T_max, the shaft is buckled.

View full abstract

In this investigation we tried to analyze the mechanism of evolution of abrasion of wire drawing die made of tungsten carbide for the purpose of finding the efficient lubricant or high-durable die for the abrasion.
An apparatus for testing the abrasion of die was constructed and the electron microscope was used in order to observe the abrasion processes of die. By using the Nishihara's metal abrasion tester, the rotating abrasion was also tested to compare with the abrasion of die in wire drawing. In regard to these experiments thousands of electron photo-micrographs on the abrasion processes were taken and the other mechanical properties of the drawn wire were investigated.
The following conclusions were obtained from the above experiments as to the mechanism of evolution of abrasion of die.
1) The abrasion of the die is caused mainly by the particles of tungsten carbide leaving from die surface.
2) This takes place at the time when a part of the wire adhering on die surface due to contact pressure leaves from the surface. The particles are taken off easily from the small part where is compressed repeatedly without the abrasion of the wire.
By this study only a few points were solved as to the mechanism of evolution of abrasion of die. We hope this study will contribute to further studies on lubricant or die.

View full abstract

The author has inquired experimentally into the effect of the temperature ranging between 20°C and 200°C upon the Rockwell hardness and the thermal expansion behavior of several virgin plasticses after casting, such as phenolic, urea, melamine, polyester, epoxy, nylon 6, polypropylene, polytrifluoroethylene and polytetrafluoroethylene.
Generally, the hardness of plastics goes down gradually as temperature rises and drops suddenly at a special temperature region, which nearly corresponds with the deflexion point in thermal linear expansion-temperature curve. The hardness is somewhat recovered again when the temperature surpasses this discontinuous point.
The thermoplastic plasticses have generally lower values and more sensitive temperature dependence in Rockwell hardness number as compared with the thermosetting plasticses.
The thermosetting plasticses, which contain the filler having more durability against high temperature than the resin itself, show lower temperature dependence for hardness, and a little amount of drop in hardness number at the discontinuous point as compared with the plastics having no filler.

View full abstract

In the presence of oxygen, the radicals produced by irradiation in high density polyethylene become trapped for a long time and can not immediately crosslink but react on oxygen to form carbonyl groups. If quick-heating was applied to the polymer followed by irradiation, the trapped radicals are set free to move and to form crosslinks. This reaction is accompanied by the decrease in crystallinity. While, the per cent gel corresponding to the degree of crosslinking is less in the case when polyethylene is irradiated in higher crystalline state than when it is done in lower crystalline one for the same irradiation dose. It may be expected therefore that several heat-treatments during irradiation would improve the effect of irradiation.
The authors reported previously that the effect of heat-treatments on tearing properties of irradiated and/or rolled polyethylene. Now, it is the purpose of this paper to study and show the same effect on tensile properties of irradiated or rolled and then irradiated polyethylene using an instron tensile tester. Test temperature was limited to 30°C and cross-head speed was made 100%/min.
The results obtained from the experiments are as follows: The yield point becomes higher generally when the heat-treatments are performed ten times than when they are given once (All the specimens are heat-treated once at least following the irradiation to eliminate the time effect). Especially, it is worth nothing that the irradiation dose, at which the yield point ratio takes the maximum value, becomes smaller gradually with the repetition of heat-treatments for the specimen rolled and cut in 0° direction to the direction of rolling and then irradiated.

View full abstract

Of great importance in the design of earth structures, such as earth dams, river levees and subgrade or base course of roads, is to know the change of density and strength of compacted soils with their moisture contents. The paper concerns this problem on the basis of pore pressure exerted during shear of soils, through a series of triaxial compression test accompanied by the pore pressure measurement.
Respecting the behavior of pore pressure during shear of compacted unsaturated soils, a lot of experimental data have been obtained which are clearly proved to be quite different from those of undisturbed saturated soils.

View full abstract

When the forms of the stress wave change or the order of the stress-amplitudes and stress-frequencies change, the fatigue strength becomes very complicated in comparison with that at the time of the normal sine wave.
In order to clarify these problems, we devised for trial a random type fatigue testing machine. This machine is composed of two parts, that is, the ordinary Schenck-type machine (used the double excentric mechanism and drived by induction-motor) and a newly designed random loading apparatus.
This random loading apparatus is composed of the following five parts:
(i) Program oscillator is composed of ink-writing oscillogram to be entered in in special ink (it can lead the electric-current) and the tracing apparatus of oscillogram (on-off type).
(ii) Thervo-amplifier
(iii) Oil-pressure source
(iv) Oil-pressure mechanism (Servo-valve and ram-cylinder)
(v) Load-checker and load-indicator.
We can carry on the fatigue test only by the Schenck-type machine, or only by the random-type loading apparatus, and also we can practise the same test superposing the two apparatus on one another.
We tested some fatigue tests as follows:
(1) Torsional fatigue test of C=0.34% steel (d=15mm plain specimen)
(i) Ordinary fatigue test of 1450rpm by Schenk-type only.
(ii) Ordinary fatigue test of 30cpm by the random-type only.
(iii) Secondary sine wave by motor (1450rpm) superposed on the first sine wave of 12cpm by the random-type.
(iv) Secondary sine wave by motor (1450rpm) superposed on the first rectangular wave by the random-type.
(v) Secondary sine wave by motor (1450rpm) superposed on the first random wave by the random-type.
(2) Plane-bending fatigue test of the induction-hardened V-notched specimen (low carbon steel)
(i) Ordinary fatigue test of 1450rpm by Schenk-type only.
(ii) Ordinary fatigue test of 60cpm by the random-type only.
(iii) At first fatigue test of 1450rpm and secondly fatigue test of 60cpm.
(iv) At first fatigue test of 60cpm and secondly fatigue test of 1450rpm.

View full abstract