The uncertainty of yield load and ultimate load measurements in high-speed impact tension test of metallic materials is discussed. For obtaining the yield land within the error of 5%, the impact of up to 5m/s velecity is to be used. However, since the lower yield load appears distinctly flat on the load-time curve with impact velocities of up to 20m/s, such tests may not be meaningless if generous allowance is made for the error. As for the ultimate load, it drops significantly when the impact velocity exceeds 60m/s due to the effect of stress wave propagation. The critical velocities obtained from the elongation drop and the change of strain distribution of fractured test pieces are compared with the respective values calculated by von Karman's method. The effect of temperature on the deformation rate of tensile properties is described for mild steel, pure copper and pure magnesium with some remarks.
A high speed tensile testing machine of a constant speed type, in which high speed tension is given to the test piece through a striking jaw from a projectile ejected by explosion pressure of explosive, is constructed. In the present state, the speed of the projectile reaches up to 210m/s, and the actual tensile speed of specimens up to 120m/s. With this machine, high speed uniaxial tensile tests of materials, such as 2S-0 aluminium, 2024C-0 super duralumin, SPC-1 mild steel, 18-8 stainless steel, and ST-60 titanium, are carried out to obtain the relation between percent elongation and tensile speed in a range up to 120m/s. High speed elongation is larger than the static one for aluminium and its alloy, but smaller for mild steel and titanium. For 18-8 stainless steel the elongation seems to show little change with tensile speed. These experimental results seem to suggest that the high speed elongation is larger than or nearly equal to, the static elongation for metals of face centered cubic lattice, but smaller for metals of hexagonal close packed lattice, and that the behaviour of metals of body centered cubic lattice lies in between. The generality of this supposition should be studied further. Drastic decreasing of percentage elongation at velocities beyond the critical impact velocities calculated by Karman's theory on plastic wave propagation is not clearly found in the range of the experiment. This seems to show the necessity of considering the factors such as the strain rate effect in stress-strain relation, etc. which are ignored in Karman's theory. A theoretical analysis of plastic wave propagation in a specimen of finite length is carried out, and the difference in analytical results on specimens of finite and infinite lengths is clarified.
In high-speed impact test of metals with a rotating disk type high-speed tensile testing machine, the stress-time record obtained by the use of a load cell attached to the specimen in series does not always tell how the specimen itself undergoes. This will become serious at high impact velocities, for the irregularity of strain distribution in such a system becomes conspicuous with in-creasing impact velocity. Therefore, to make clear the relation of stress in the specimen to that in the load cell, a round bar made in two sections of different diameters as a model of the system comprising the specimen and the load cell is pictured, and the longitudinal stress wave propagation in it is discussed in detail. Important problems first encountered with this model are the reflection and transmission of stress wave involving plastic wave at the discontinuity. Since these problems for elastic wave have hitherto been dealt with, calculations are made for plastic wave with some interesting results, and the conducted test and how the stress wave propagates are outlined. Furthermore, the case of elastic wave propagation in a bar of continuously varying cross-section is also dealt with by the use of hypergeometric function and Bessel function. This way of solving can be applied to cases of elastic wave propagation in various forms of bar of varying cross-section, for hypergeometric function involves three arbitrary parameters.
In September, 1961, and September, 1962, large scale explosion experiments with 150 kg to 3 t of explosive were carried out in Hokkaido and near Lake Biwa, and data were obtained on blast pressure, damage on houses, attenuation of shock wave by banks etc. which are supplemented by high speed photographs taken by two Fastax cameras at a rate of approximately 4, 000 fps. Together with detailed explanation of how the high speed photographs were taken, information concerning the following items are given. (1) Generation and growth of fire ball. (2) Shock wave perceived over the background or by Wilson cloud chamber effect. (3) Transformation of flame into smoke. (4) Disintegration of smoke. (5) Collapse of structure.
Epitaxial silicon layers are usually expected to have high purity, but when some carbon com-pound is present in the reaction tube, the layer is found contaminated by the formation of Si-C bonding beside Si-Si bonding. In the infrared reflection spectrum of Si-C contaminated epitaxial layers, the residual line of silicon carbide (cubic-β) is observed at 12.6μ and etch patterns of screw dislocation are seen on etched surface of the contaminated layer. SiC is also formed on the back surface layer of the substrate crystal by direct chemical reac-tion of silicon crystal with carbon at 1200-1300°C. Electron diffraction pattern shows that the contaminated epitaxial layer has zinc-blende structure and the back surface layer has multi-crystalline structure. The SiC contamination can be avoided by removal of carbon block from the reaction tube.
Servomechanism is applied to an electron trajectory plotter which is provided with an electro-lytic tank. The principle of measurement is that, in an electrostatic field, the radius of curvature of electron trajectory at the point where the electron is found is proportional to the ratio of poten-tial V to potential gradient εnV and a. are detected by a pair of probes inserted in the electrolytic tank. εn is measured by a differential amplifier to avoid the effect of phase-shift of the probe voltages, while V is measured by a cathode follower amplifier. The two are connected to a poten-tiometer after being precisely adjusted in phase shift and wave form. The position of the poten-tiometer contact is controlled by a servo-motor, which control at the same time, the direction of the motion of trolley for converting the ratio 2V/εn to the radius of curvature. The position and rotation of the pair of probes and the motion of the trolley on which a stylus is fixed are in coincidence by gantry mechanism. The experimental results are in good agreement with the outcome of mathematical calculation. Detail of the electronic circuit, the mechanism, and the measured examples are discussed.
This paper describes how the boundary lubrication actually affects the relative tangential micro-displacement between two bodies in contact, the displacement being caused by a tangential force weaker than the force of maximum friction. The micro-displacement is measured with a differential transformer and the tangential force with an electromagnet, its current determining the magnitude of the force. The curve of tangential force vs. tangential displacement and that of tangential force vs. contact resistance are recorded on an electromagnetic oscillograph. The tangential displacement caused by a tangential force weaker than to give rise to sliding becomes larger for a wide range of the force when the contact is lubricated. This confirms the obvious effect of lubricants on the process of plastic deformation before the sliding sets in. The coefficient of maximum friction and the tangential displacement just before the beginning of sliding are measured under boundary-lubricated condition at the temperature ranging from 25°C to about 180°C. As the temperature rises, the displacement becomes smaller and the observed values of the coefficient become random. With unlubricated steel specimens, irregular stick-slip occurs even during the micro-displace-ments prior to the sliding. The magnitude of this stick-slip is 0.2μ_??_1μ which is considered to be of the order of the area of individual real contacts dotted over the apparent area of contact. When the tangential force is increased, the contact resistance decreases and remains very nearly so even the former is decreased again. The increase of real contact area following the increase of tangential force seems therefore to be due chiefly to plastic deformation in normal direction at the real contacts.