Modeling and Sensing for Penetration Control of the SAW Process in the Presence of Welding Perturbations

Abstract

A study was conducted to investigate the feasibility of using a point infrared sensor to monitor and control the penetration during the Submerged Arc welding (SAW) process. The SAW process, using small tractor welding units, is extensively used in the ship building industry because of the high deposition rates that may be achieved. Typical problems encountered by these welding units include seam gap variations, plate offset, changes in thickness, tracking misalignment, plate inclinations and heat sinks such as stiffeners, chill blocks and temporary buckling restraints. Each of these perturbations causes a change in the temperature distribution that may be identified using infrared sensors. A simple numerical heat transfer model using the Finite Element Analysis (FEA) package ALGOR was constructed to predict the heat flow patterns and thermal distributions that occur in the presence of step changes in thickness and heat sink perturbations. Bead-on-plate welding trials were then done on A-36 carbon steel and monitored using an infrared sensor to compare predictions with experiment. Good agreement was observed when predictions were compared with experiments. The measured changes in infrared signal corresponded well with the predicted results both in terms of time of occurrence and relative magnitude of signal. The results of the experiments suggest that infrared sensors may be used to identify and control process perturbations that occur during the SAW process. Future experiments will utilize the signals from IR sensors to control welding current and provide automatic control to obtain constant penetration during the SAW process.