This is an attempt at interpretting arc welding as a kind of wetting phenomenon, i.e., a mechanical phenomenon which is governed by three kinds of interfacial tension taking part in it. Fundamental and general behaviour of a liquid drop confined within a solid surface was discussed first, and the findings were applied to explain the welding phenomena. The results were as follows. (1) "Mother Drop", the molten weld metal phase, should be considered as a single liquid drop, surrounded by an interface between molten metal and slag or gas phase, stressed by interfacial tension. (2) Mother Drop is confined within a concave "Surface of Penetration", which is composed of the melting surface of mother plate and the solidifying surface of a weld bead."Confinement" means that the contact line of a liquid drop can not leave the intersecting line of two surfaces until the liquid surface takes the equilibrium contact angle against the adjoining surface. (3) Internal pressure of a liquid drop kept within a concave conical surface upon a horizontal plate is minimum when its contact line reaches the boundary line and its contact angle is equilibrium value. Mother drop on the surface of penetration seems to be in a similar condition. (4) Generally there are two types of drop, "Convex" and "Flat", in a single V-groove. If the inclination of groove face is too small as compared with the contact angle of the drop, there emerges flat to convex transition with an increasing volume of the drop. Mother drop of normal weld bead must be a flat type.
In this report, the problems of generation processes of welding residual stresses were studied with experimental techniques using the special resistance wire strain gauges usable up to 400°C, directly attached on the samples. The apparent strains were continuously recorded an automatic recorder. The inherent strains were calculated by subtracting rectified values ascribed to the temperatures characteristic of strain gauges and the thermal expansion of the samples from apparent strains. The experimental results are summarized as follows. As to the generation processes of stress parallel to weld lines in the parts adjacent to weld lines, compressive stress first grows with the approach of the welding heat source, and falls off swiftly when the heat source is kept away, tensile stress growing rapidly instead. This tendency of stress changes when welding becomes violent with the increase of an input heat. The behaviour of stresses during the welding and cooling at the points farther away from the welding lines is quite opposite to that in the neighbourhood of the welding lines. Close to the time when weld starts or finishes, the weld is discontinuous, and the generation processes of stresses perpendicular to the weld lines are remarkably different from those at the time when weld reaches its stationary state. In general, although the generation processes of stresses are not essentially influenced by preheating all over the specimens, the curves of stress behaviour become more or less slow and the residual stresses slacken. The local preheating, however, of the groove tend to cause a great change in stress and enlarge the residual stresses.
In the present report, the Author studied experimentally the influences of peening on the processes of relaxation and the processes of generation of restraint stresses using a restraint frame apparatus. The same experiments were also made to examine the influences of peening on residual stresses. The main conclusions obtained from the above results are as follows : (1) Although the decrease of welding stresses by peening is remarkable at early stages, it becomes gradually saturated because peened specimens workharden. In other words the relation between restraint stress and log. of peening time is linear. (2) Also a linear relation exists between log. of relaxation rate of stresses and peening pressure. (3) There are ranges of a certain optimum temperature in the influence of hot peening on suppression and relaxation of stresses, that is 200-300°C. (4) The effect of peening on base metals is about the same in its extent as that on deposit metal, but peening on a heat affected zone is not so effective as on the others. (5) The tip areas of peening tools have scarcely any influence on the stress relief. (6) Although the growth of residual stress caused by the approach of the heat source can not be suppressed by peening, stresses that grow as the heat source gets kept away can be considerably suppressed by giving a suitable hot peening.
The welding conditions have complicated influence on the porosity in aluminium alloy welds. The authors reported in their first report that the heat input in welds influences considerably the gas content in welds and further in their second report that several different inspection methods presently available of welding porosity have respective merits and demerits. In this report (the third report) an attempt is made to simplify and analyse the welding conditions in the case of experimental welding which influence the porosity in aluminium alloy welds. The apparatus prepared for the purpose is schematically shown in Fig. 1. The following results were obtained : 1) The increase of heat input in welds decreases the percentage voids of porosity in spot deposit (Fig. 6) ; the same result was also obtained in practical welding. Since the increase of heat input decreases the cooling rate of molten pool, it is clear that the cooling rate has remarkable influence on the welding porosity. 2) Heat input of welds, however, is a complicated factor for welding porosity. It is therefore considered necessary to divide the factor into two minor ones, i.e., the cooling rate and the temperature of molten pool, and then investigate both, As the result of temperature measurement in molten pool during welding, it was found that the temperature and the porosity in weld deposits simultaneously increased linearly with the welding current.
The reverse bend crack test-in some cases, slit type specimen crack test-was applied to several casting aluminum alloys of Japanese specifications which were the objects of fish-bone type crack test in the previous paper. The results are shown in Fig. 3 and Fig. 5-8 as the function of the chemical compositions of alloys, and are compared with similar results obtained by other authors. Some notes on the crystal behaviours of weld metals are given.
Since a new interest is being aroused in alumiuum rolling stocks in recent years the domestic rolling stock makers have started a serious study of the Sigma Spot Welding for aluminum alloys as a part of thier shop practice researches with considern the attention paid to its practical application. The author has investigated various welding factors (weld current, arc voltage, arc time, argon flow, etc.), weld joint strengths, X-ray inspection data, weld section contours, etc. in respect to 2 mm +4mm plate combination of an aluminum alloy (52S-O) with a view to applying this welding process to sections where a spot welding is not practicable at joints between external panels and carbody framework of aluminum rolling stocks, in future, and has selected an optimum welding condition for this type of welding work. He has, also, touched on the peripheral cracks around nuggets, which have developed on the test pieces in this experiment.