On Electric Butt Welding

Abstract

In this paper the electric butt welding of soft steel bars is treated and the conditions in order to obtain good results as well as the advices for practical operators are concluded from numerous experiments and the result obtained from them. In the first place oscillograms are taken to show the amplitudes of the welding current and the voltage across the butt joint and also that between the clamping block and the bar. From these we can ascertain that the voltage across the joint is comparatively small and its initial value is usually less than one volt and even this decreases pretty rapidly owing to the gradual melting of the joint surfaces. On the contrary the voltage between the block and the bar is not so small remains almost constant throughout the welding time and moreover appears on bath sides of the joint. The value of this voltage depends largely on the condition of the surface of the bar, which is more or less covered with rust. These voltages corresponding to a very rusty surface and a polished one are shown on the oscillograms. The rusty survace makes the power loss between the block and the bar large and renders the welding time long and irregular. The voltage across the joint is comparatively small and decreases rapidly as before mentioned, therefore the heat generated at the joint is not so large as might be expected; the head quantily necessary for raising the temperature and melting the part adjacent to the joint is rather produced in the metalle'c part itself owing to the increased resistivity of the metal, thus raising the temperature with acceleration. From this we can conclude that the metal with a smaller resistivity and temperature coefficient is more difficult to weld; that is, for example, copper bars are more difficult to weld than steel bars. In the next place, bars of different diameters were welded with various powers and the corresponding welding times were measured and these relations between the power and time for each diameter are shown by curves, which resemble hyperbolas but gradually deviate from them upwards for smaller power and larger time. This fact shows that the welding with the smaller power requires larger energy and time. The tensile test was made on these welded bars and the result shows that the welds made with smaller power, that is to say, with larger time are smaller in their tensile strength, although this tendency is less remarkable for bars of the larger diameters. In either case we must make the welding time as small as possible. The effect of the spring pressure which is (applied) to the bar axially was tested. Bars were welded under various pressures, but with a constant power, and the result of this test is that the welding time is almost independent of the pressure and remains constant, but the tensile strength generally increases with the pressure. The joint is liable to the oxydation and this the more remarkable the longer the welding time and the less the spring pressure. In order to be able to investigate the physical properties of the welds the knowledge of the maximum temperature along the bar and also the manner of the variation of the temperature is necessary and these are calculated from differential equations containing as many factors as possible, such as the heat transmission from the surface, the temperature coefficient of the resistivity and the heated generated at the joint etc., and these results are shown by curves.