Large-size welded chains made of low carbon structure steel are sometimes quenched and tempered after manufactured for the purpose of increasing in their strength. This low carbon steel contains C 0.12-0.18% and Mn 0.8-1.0%. It is often observed that if the tension test is made on the quenched low carbon steel without tempering, the values of its tensile strength scatter to a great extent. As a cause of this wide scattering is presumed that this sort of steel is much influenced by the slight differences of heating or cooling conditions in the heat treatment.
This experiment designated to investigate the scattering states of tensile strength and to pursue its causes, is made in two stages from the practical point of view. The test pieces used in the first stage, 120 in all, are made of the C 0.17%, Mn 0.85% low carbon steel of one and the same charge; and those used in the second, 920 in all, of the same sort of steel of the other 92 charges, -e.g. ten pieces out of each charge, -to some of which boron is added. In the second stage, the experiment is made in the same way, but with less accuracy for the purpose of merely getting more general conclusions.
Although the values of tensile strength scatter in a fairly wider range, they can be put on a normal curve. An example is given for the non-tempered test pieces as shown in Fig. 1.
For quenching, the following methods were applied: the agitated, the normal, the gradual, and the oil quenching. Investigation was carried out. On the test pieces quenched in those methods to know what kind of influences the cooling rate has on their mean tensile strength, how much the size of their diameters effect the mean tensile strength and the scattering states of tensile strength. Non-tempered test pieces as well as those of different hardenability were used for this purpose. The effect of tempering on these test pieces were also examined.
Among the reasons of the scattering of tensile strength, the difference of the size of test pieces and that of tempering conditions are taken into consideration, but the most influencial one is the dynamical behavior of cooling water for hardening the test pieces.
We come to the conclusion: in the case of quenching such a low carborn steel as is used in this experiment, the scatterng of tensile strength is inevitable.

View full abstract

This paper presents the experimental investigation of the proper grading of aggregates for the proportioning of concrete which should be selected to produce economical concrete of the required strength and workability.
For this purpose, the authors examined the effects of various gradings, especially gap grading, of aggregates on the bulk density, compressive strength and workability of concrete. The aggregates used are river gravel and sand, and the particle shape may practically be considered as constant.
The experimental results are as follows;
(a) The proper gap grading of aggregate gives the weighty bulk density and the concentrated grading gives the minimum value.
(b) Two fineness modulus of aggregates having the same property give the same bulk density.
(c) Relationship between the thickness of cement paste film and consistency of concrete, is approximately linear. Consistency of concrete is decreased as the cement paste film is reduced, if other factors are equal.
(d) Compressive strength of concrete which is made of the proper gap grading of aggregate is higher than that of concrete made of the continuous grading of aggregate.
(e) When cement content and consistency of conrete are constant, the compressive strength of concrete is decreased, as the void ratio or specific surface of aggregates is increased.
Herewith, the authors propose to take the specific surface and the voids of aggergates as parameters for designing the proportion of concrete.

View full abstract

The authors previously noticed on the depression in case of measuring hardness, that the Vickers depression had some piling-up even in diagonal direction, which is believed to have little piling-up. The Vickers hardness is measured by the surface area of depression caloulated from the diagonal length of square depression, so if the degree of piling-up differs, the influence of measuring error in the hardness number will be different.
In this paper, the change of the piling-up caused by the heat-treatment of materials are examined as a help for obtaining the inquiring data of the measuring error of Vickers hardness.
Five sorts of the carbon steel and two sorts of the brass are heat-treated and depressed by Vickers indentator, and the surface of depression are traced by the Ogoshi Surface Roughness Tester (contact needle type). Some of the results are summarized as follows: (1) The degree of piling-up is much affected by the heat-treatment in both materials. (2) The change of the degree agrees with the change of Vickers hardness in case of carbon steels, but does not always agrees with the hardness in that of brass. (3) The degree of piling-up varies by 200%, in case of carbon steel, and by 100%, in that of brass according to the rise of tempering temperature, if the quenching value is taken as standard. These degrees have a great influence on the hardness number. (4) As the ratio of the major diagonal of Knoop depression to the minor diagonal is in the lever relation to the degree of piling-up. One rising, the other falling, so we may easily obtain the general trend of the piling-up.

View full abstract

The authors treat in this paper the experiment on specimens with double notches. Double notches are deviled into two kinds. In the first, the notches consist of two circumferential grooves, one (secondary notch) of the two being situated at the root of the other (primary notch). In the second, the double notches consist of a groove (primary notch) and a drilled hole (secondary notch), the latter being situated at the root of the former.
As to the first kind mentioned above, the authors reported previously on the case where secondary notch is sharp but very shallow. In the present test the authors treated the case in which secondary notch was comparatively deep.
Fatigue tests were made on specimens having such notches.
From the results of experiments, the following conclusions are reached.
The notch factor of a double notch β_{k1, 2} is equal to the product of β_k1 and β_k2, where β_k1 and β_k2 are the notch factors corresponding to the primary and secondary notches respectively. But as reported previously, when the sharpness of two notches becomes closer to each other, β_{k1, 2} becomes less than the product of β_k1 and β_k2.
Thus designers have to pay their attention to the fact that the fatigue strength of the materials with double notches will be essentially less than that which would be expected from each notch.

View full abstract

Ultrasonic attenuation in solid is very sensitive to conditions under which the material is placed. Therefore the attenuation may be considered as a measuring factor of the properties of solid under the given condition. In the present paper, the result of measurement of the ultrasonic attenuation in normalized steel of 0.15, 0.4 and 0.55°C at low temperatures down to -180°C is reported and some conclusions are derived as to possible relation of the properties with the attenuation. The result shows that the attenuation decreases with the lowering of temperature and at around -60° to -80°C, the attenuation drops discontinuously by an appreciable amount, corresponding to the drop of Charpy's impact value of similar steels.
Extension of Maxwell-type model to continuous body leads to the wave equation
∂²u/∂t₂=a²∂²u/∂x²+c∂³u/∂x²∂t, u=u(x, t)
where c is the coefficient of internal friction.
The solution of this equation with the boundary conditions
u(∞, t)=0
u_x(0, t)=Acoswt
is given in the form of
u(x, t)=A₀e^-αxsin(wt-ηx)
where, A₀=A√a²+c²ω² and as the first approximation, α is proportional to w² and c. The quadratic change of α with w agrees with the result of experiment within a certain range of frequency but, with specimens of lower carbon content, additional attenuation appears to superpose, resulting a maximum attenuation at a frequency. Since α decreases with the temperature fall, the coefficient of viscous friction c in steels decreases with the lowering of temperature, contrary to the viscous friction of fluid which generally increases with the temperature fall. This seems to indicate that the drop of absorbed energy of a shock at low temperatures cannot be explained by the decrease of internal viscous friction of Maxwell model.

View full abstract