Noise as a signal for on-line estimation and monitoring of welding process

In this paper noise generated during the gas-metal arc welding process was studied. Theoretical and experimental analyses of the acoustic signals have shown that there are two main noise-generating mechanisms. The first mechanism generating characteristic sound impulses is arc extinction and ignition; the second noise-generating mechanism is the arc itself acting as an ionization sound source and producing mainly high frequency "turbulence" noise of a low level. The much higher level of impulse noise dominates in the total sound pressure level and is used for assessing and monitoring of the welding process and for prediction of welding process stability and quality. A new algorithm based on the measured welding current was established for the calculation of noise during the welding process. The algorithm was verified on different noise generating mechanisms and for different welding parameters. The comparisons have shown that the calculated values are in good agreement with the measured values of noise.