Abstract Numerous GMAW processes have appeared for one-pass welding of medium to high thickness. This concept originated in the 1980s with the T.I.M.E. technique and the 1990s and 2000s saw the emergence of Rapid Arc, Rapid Melt and Buried Arc techniques. These processes did not include any special control features, but rather parameter combinations to improve bead geometry. A new interpretation of the factors that are critical to achieving high performance GMAW (penetration) is presented herein. Tests with a voltage controlled power source with inductance control and the Rapid Arc process were conducted on bead-on-plate for high speed videography. Later, validation tests were conducted on butt joints, with the technique developed, with and without groove on AISI 1020, 3/8” thick steel plates. The power source may be of voltage control but has to provide a dynamic response able to sustain a metastable equilibrium of the molten pool applying the semi-keyhole technique. The technique developed was able to weld, in a single pass, 3/8” thick butt joint steel plates. The results showed that a power source, to achieve the GMAW buried arc condition, only needs to provide dynamic characteristics able to maintain the molten pool in a metastable equilibrium.

RESUMO A técnica de soldagem dupla camada utiliza método controlado de deposição, onde a relação entre os aportes térmicos da segunda para a primeira camada de solda (E2/E1) deve ser criteriosamente definida para promover o refino dos grãos da Zona Afetada pelo Calor (ZAC). Entretanto, a parametrização de processos de soldagem para a realização desta técnica tem sido de modo experimental, requerendo diversos ensaios, corpos de prova e consumíveis, demandando tempo e elevado custo. Uma possível solução para tal problemática é a análise computacional aplicada à soldagem, tornando o estudo desta de grande relevância. Este trabalho objetiva o estudo inicial do comportamento da ZAC em soldagem dupla camada utilizando o Método de Elementos Finitos (MEF). Experimentos práticos foram realizados para comparar e validar o modelo numérico. Dessa maneira, seis corpos de prova foram fabricados, três com camada simples e três com dupla camada. Foram utilizadas três energias de soldagem diferentes para os primeiros passes (0.571, 0.428 e 0.342 kJ/mm) e o dobro de tais energias para o segundo passe (taxa de energia de 1:2). Comparou-se as macrografias dos resultados experimentais e numéricos, tanto para a camada simples quanto para a dupla camada, bem como realizou-se a medição das dimensões da ZAC ao longo da seção transversal dos corpos de prova. Dessa maneira, notou-se que a taxa de energia de 1:2 é suficiente para atingir a zona de grãos grosseiros da primeira camada. Ainda, os resultados macrográficos e as medições das ZAC apresentaram ótima convergência entre as análises experimental e numérica, validando o modelo proposto.

ABSTRACT The double layer welding technique uses a controlled deposition method, where the relationship between the thermal inputs from the second layer to the first (E2/E1) must be carefully defined to promote grain refinement in the Heat Affected Zone (HAZ). However, the parameterization of the welding processes for this technique is usually experimental, requiring tests, samples and consumables, demanding time and high cost. A possible solution to this problem is a computational analysis applied to the welding process, making this study to be of great relevance. This work aims an initial study of HAZ behavior in double layer welding using the Finite Element Method (FEM). Experiments were performed to compare and validate the numerical model. In this way, six samples were made, three with a single layer and three with double layer. Three different welding energies were used for the first layer (0.571, 0.428 and 0.342 kJ/mm) and the double of those energies for the second layer (1:2 energy rate). Macrographs of the experimental and numerical results were compared for single and double layer conditions, and HAZ dimensions were measured along the sample cross section. Therefore, it was noted that 1:2 energy rate is sufficient to reach the coarse grain zone of the first layer. In addition, macrographic and HAZ measurements results showed excellent convergence between the experimental and numerical analysis, validating the proposed model.