Catarina Ericsson

Abstract

In the beginning of this project the focus was set on exploring how different flowability in a falling position could affect the occurrence of Lack Of Fusion type defects (LOF) in MIG/MAG-welding. The general idea was that if a relative measure of the flowability of a molten metal could be established then this could perhaps indicate whether the gravity force would promote the molten metal to flow away and to produce LOF.

Bead on plate trials were therefore performed in a horizontal and in a falling position for different base-/filler metal combinations. Thorough evaluations of different geometrical properties of the weld bead such as bead to plate angles, cross section areas etc. were then made. Large differences were noticed between the samples welded in a falling position compared to these welded in a horizontal position. Those welded in a falling position penetrated less and they possessed a smaller cross section area. In the falling position a difference in behaviour between the different base-/filler metal combinations could also be noticed and a certain ranking could be established. This difference in flowability depends on the type of filler metal used and most likely on the influence of surface active elements such as silicium, manganese and sulphur. Under these circumstances one could assume that those base-/filler metal combinations that flowed easily would promote more LOF when welding was performed in a falling position. To confirm this assumption, welding trials in a joint were to be performed using the same base-/filler metal combinations as in the bead on plate trials.

Before that though a complementary study was made to determine whether the LOF found in project [3] were caused by too low or too high deposition efficiency that is to say if they were “cold or warm". This study revealed that these defects in fact had been surrounded by molten metal on both sides, which indicated that the planar defects thought to contain gas, instead were filled with slag that most likely originated from the surface of the weld bead. This somewhat changed the focus of this project. For this reason the defects investigated in this study are mentioned as “LOF/slag" and not just “LOF" throughout this report..

The final welding trials performed in a joint confirmed that there in fact were slag inclusions present in the weld. Depending on base-/filler metal combination used the percentage amount of these slag inclusions varied. This variation might be due to that certain combinations also showed less slag on the surface than others and that the slag on the surface seemed to be placed differently for different combinations. This surface-slag forming aspect in the falling position is thought to be of greater importance than the flowability with regard to the formation of these non-surface planar defects.