Prediction and optimization of heat input on mild steel weldments

Excessive weld penetration is a weld defect that affects weld quality and the service life of Engineering Structures, such defects can be eliminated by the optimum selection of welding process parameters. This study is carried out to optimize the heat input, An experimental matrix was developed with the design expert software, which resulted in a composite design. Mild steel plate of 10mm thickness was cut to size measuring 80mm in length and 40mm in width, the samples were grinded, polished and welded with the Tungsten Inert Gas. Thereafter the heat input was determined for the weld samples, the experimental results was used as the data. The Response Surface Methodology, was employed to optimize the responses from input parameters which includes current, voltage and gas flow rate. The experimental results shows that the minimum value for current is 120, voltage 20, gas flow rate 12 and the maximum value are 170 Amp, 25 volt, 14Lit/mm. The RSM model selected the quadratic model as the suitable model for the heat input, deposition rate, weld penetration size because it has a p – value < 0.05. The models developed possessed good noise to signal ratio of 9.47 for heat input. The optimal numerical solution was obtained and the result shows that a current of 157.67 will combine with 24.44 volt and 12.34 Lit/min to produce a weld having 0.603 heat input.