Resumo:
This work was developed in order to propose a welding GMAW optimise
mathematical model in a company who manufactures heavy truck and buses wheels in low
carbon steel. This optimisation was achieved by Design of Experiments application, using the
Response Surface Methodology technique. The experiment had 31 welding runs and it was
used standard tubeless wheels, which is the most popular model in internal and export market.
The input parameters (voltage, current, longitudinal angle and welding speed) were chosen by
literature review and process characteristics and constraints, as well as the quality bead
improvement added production cycle and wire rate decrease. The responses (rim and disc
penetration, convexity index, base angle and welding area) were determined based on the
weld bead internal specifications and economic aspects. The final mathematical models show,
in a general way, that all input parameters (isolated or combined) have influence in the
responses studied. Therefore it was possible to obtain the total process optimisation based on
technical and economics approaches.
