Numerical And Experimental Study Of Hybrid Laser-GMA Welding Of Low Alloy Steels

Hybrid laser arc welding combined laser and arc together, which have significantly different natures as heat sources, to obtain better welding qualities and higher welding productivities compared with autogenous laser welding or arc welding alone. Current researches about hybrid laser arc welding shows that various welding variables have significant influences on weld pool profiles and weld geometries. The cooling rates of the welds will also be affected and further lead to different weld metal microstructures and mechanical properties. Although a large amount of experimental work have been done for hybrid laser arc welding, the temperature distribution and fluid flow within the weld pool cannot be demonstrated exhaustively without the aid of numerical simulation. Consequently, three-dimensional modeling is critical for improved understanding of the physical process during hybrid laser arc welding.Currently little work has been done for comprehensive studies of the weld pool profiles, weld geometries and weld metal microstructures of hybrid laser-GMA(Gas Metal Arc)welding. In this thesis the effects of heat input and laser arc spacing of full penetration hybrid laser-GMA welding of low alloy steel on weld pool profiles, weld geometries, weld metal cooling rates, weld metal microstructures and weld metal mechanical properties are investigated both numerically and experimentally. The numerical model has been constructed based on the understanding of laser keyhole welding and GMA welding processes. Various physical processes such as the absorption of laser and arc energy by the workpiece, the formation of keyhole, the vaporization of the metal within the keyhole, the turbulent fluid flow due to Marangoni force and electromagnetic force within the liquid weld pool and corresponding heat transfer are considered in the model.The effects of welding speed and laser arc separation distance on temperature distribution and fluid flow with in the weld pool have been studied first based on the three-dimensional heat transfer and fluid flow model. The weld pool dimensions increased with the decrease of welding speed from 40 mm/s to 30 mm/s and 20 mm/s at laser arc separation distance of 1mm. The weld pool dimensions also increased with the increase of laser arc separation distance from 1 mm to 3 mm and 5 mm at welding speed of 40 mm/s, although the welding heat input was kept constant. The fluid flow within the weld pool was turbulent flow because the Reynolds Number of the fluid was significantly higher than the critical value for turbulence. The turbulence extent was higher at the circulation center of fluid flow loops where the velocities of the fluid flow are also relatively higher. In order to validate the numerical model the calculated results were compared with corresponding experimental results and it showed the results agreed well for all the cases.Experimental researches were conducted in order to study the effects of heat input and spacing between the heat sources on weld microstructures. The welds were sectioned, polished, etched and photographed to reveal the weld fusion zone profile and microstructures. The volume fractions of selected microconstituents in the weld metal are determined by using the point counting method following the International Institute of Welding(IIW) guidelines. Numerical simulations were applied to further investigate the influences of the welding variables besides the study of the experimental results. The microstructure numerical model is based on thermodynamics and phase transformation kinetics, with output of TTT(Time Temperature Transformation) and CCT(Continuous Cooling Transformation) diagrams together with corresponding cooling curves of the welds.The cooling curves only intersected with the upper curve of the CCT diagrams when the welding speeds were 30 mm/s and 20 mm/s at laser arc separation distance of 1mm and with welding speed of 40 mm/s and laser arc separation distance of 5 mm. The weld microstructures contained grain boundary ferrite, Widmanstatten ferrite and acicular ferrite. The volume fraction of grain boundary ferrite increased with the decrease of welding speed or the increase of the heat input; the volume fractions of Widmanstatten and acicular ferrites decreased with the increase of welding heat input. The cooling curves intersected with both the upper and lower curves of the CCT diagrams when the welding speeds was 40 mm/s at laser arc separation distance of 1mm. Consequently the weld microstructures contained martensite besides grain boundary, Widmanstatten and acicular ferrites. The phase transformation model was validated with the comparison of the experimental and calculated volume fractions of the microstructures.