A real-time prediction model of electrode extension for GMAW

This paper presents the development of an electrode extension model for the gas metal arc welding process based on the process voltage. The full dynamic model for the electrode extension is derived by combining a dynamic resistivity model with the voltage model. The electrode extension model was found to be represented mathematically by a nonlinear, time-varying, second-order ordinary differential equation. This model can be used in through-the-arc sensing and arc length control systems. To experimentally verify the model, the process dynamics were excited by a continuous sinusoidal variation of arc current. Using a constant current power source with the electrode positive, sinusoidal perturbations of variable amplitude were superimposed on the current to allow direct measurement of changes in electrode extension, arc length, and total voltage. A high-speed video system was used to capture the experimental electrode extension dynamics. The model was verified by comparing the frequency response of the model to the frequency response of the real process. Agreement between the simulations and the experimental results was found to be very good. The accuracy of this model was found to be approximately /spl plusmn/0.6 mm, which is considered to be suitable for process control applications.