African Fusion November 2020

IIW Paper: Laser buffering for Duplex welds

plate, but leave the area A1 (Figure 4) free of material. Since the laser beamweldingwas done without any additional filler material there is a undercut. To overcome this problem, LMD-tack lines were used. For this, instead of tack- ing the plates at three single points the tacking was done with LMD-lines on the upper and the lower part. On the one hand, tacking is necessary to prevent distortion, so the effect of additional LMD-tack lines is relatively low, and on the other hand, these tack tracks provide the extra mate- rial to fill the gap ends, thus enhancing the performance of the laser weld. Figure 5 shows a cross section of the weld seam, with only a slight relapse on the root side visible. To evaluate themicrostructural compo- sition of the different zones, an electron backscatter diffraction (EBSD) analysiswas done. The region of interest ROI 1 in Figure 6 shows a balancedmicrostructurewith an average austenite (blue) content of 55%. In comparison, the welds done without any cladding display an austenite content of about 10 %, so the welds were almost completely ferritic. For ROI 2 in Figure 7, a narrow heat affected zone (HAZ) with a predominantly austenitic microstructure was detected. Where the coating is not completely molten in the HAZ (ROI 3) between the laser beamwelding and the coating, is the only area where a notable in a change in microstructure froma balanced austenite- ferrite ratio in the weld seam and the residual coating can be seen. This region transformed to austenite as can also be seen in ROI 4. The HAZ of the coating is visible in ROI 4, were a narrow red ferritic layer is evident. To characterise the mechanical proper- ties of thewelds, Charpy impact testingwas executed. The impact toughness and the fractured surfaces are shown in Figure 8. Coated as well as uncoated laser-welded joints were tested. The fractured surface of the welds without any cladding showed

Figure 4: Missing material in Area A1.

Figure 5: Cross section of weld seam with LMD-tacking.

Figure 6: Regions of interest of weld seam (left side); EBSD of ROI 1: blue- austenite, red- ferrite (right side).

width were tested as well, but the results were too wavy for the laser welding pro- cess, where a zero gapbetween thewelding partners is preferred, which leads to irregu- larities in the upper and the lower part of the weld seams. Even with the more even cladding there was an underfill on both sides of the weld seam. A more detailed examination of the coatings showed that the first and the last line do not form a 90° angle with the base

of the powder mixture in comparison to the unmixed duplex powder, the optical analysis of themicrostructure shows a bal- anced ratio between austenite and ferrite. The first layer is more ferritic, most likely because the base material is cold and warms up with an increasing number of layers. The cross section shows strong overlapping of the single LMD-tracks and relative evenness of the coating. Coatings with a normal stepover of 30% of the line

Figure 7: EBSD of ROI 2 (left side); ROI 3 (middle); ROI 4 (right side).

16

November 2020

AFRICAN FUSION