There has been adoped in our industries the Unionmelt Walding Process, as a submerged arc Walding, which has two special phenomena and problems to be solved. One is porosity on bead and the other is sulphur cracks grown from Sulphur Bands. These two defectal, we consider, occur mainly, by hydrogen gas, which generates from the dissociation of a bit of moisture contained in flux, caused by arc. 1) When the absorbed hydrogen in deposited metal is more than its solubility limit of solid metal at melting point and its speed of solidification is fairly faster, porosity on bead is formed. 2) When deposited metal comes to have hydrogen in super saturation, Sulphur cracks are easily to start from sulphur band which involved sulphide of low melting point used to be observed in rimmed steel plates. 3) Intergranular crack in welding bead of killed steel plates is as like as one in ice flower structure.
If you take the oscillographs of the arc current or voltage during the welding whith the automatic or hand welder, you will find the ripple in the arc voltage though the arc current is constant. Ripple occurs as such in the case of D. C. source, while it is superposed to the commercial frequency with A. C. source. Up to the present, the above phenomenon has been simply reduced to the spattering of the metal, the influence of the electric source, or the effect of the automatic welder. It seems, however, that there are problems which are inexplicable by the above explanations. For example, the period of the ripple is under 1/100 sec., while the response time is as long as 17/100 sec. The Authors attribnted it to the self-regulation of the arc, and according to their theory, the. spattering of the metal should be included in this self-regulation. Under the reasonable assumption, they have tried the mathematical analysis, and recognized that it satisfactorily coincides with the experimental results.
Welded Kahn tear test piece was used to evaluate the effect of submerged arc welding on the notch sensitivity of structural mild steel plates. The saw cut was located alongside the deposited bead to determine the fracture transition temperature of heataffected zone. Welded plates generally showed higher transition temperature than unwelded plates, ranging from 6 to 20°C in 13 mm thick arid 5 to 15°C in 23 mm thick ones of rimmed steels, from 0 to 12°C in killed steels, 23 mm thick. These adverse effects of submerged arc welbing were almost the same as manual arc welding. But, generally speaking, as these effects of welding were comparatively slight, it could be concluded that the notch sensitivity of mild steel plates was but little increased by submerged arc welding so far as the fracture transition was concerned.
The welding residual stresses in both directions, parallel to and perpendicular to the welded edge, were measured by the X-ray diffraction method at six points over the surface of mild steel rectangular plates, which had a weld bead placed along one edge; as shown in Fig. 1. As it was found that the longitudinal stresses in the zone from the point. 3to 5. were in. compression, the zone was heated to 110° to 370°C. Progressively from one end to another by a gas flame. The changes. in residual stresses due to the local heating are shown in Fig. 3 to Fig. 15. The stress near the weld was reduced to a small value, but the one around the heated zone turned to be in large tension. This fact seems to be due to improperly slow heating and cooling, and a more appropriate measure will be introduced in the next report together with the clarification bf the mechanism for the relieving of welding residual stress by local heating.