Study on the Evaluation of Weld Soundness Based on Arc Stability and Bead Quality using the Statistical Analysis of Arc Power and Arc Dynamic Resistance

Abstract

The goal of this study was to evaluate weld soundness based on arc stability and bead quality using a statistical analysis of the arc power and the arc dynamic resistance. To achieve this, 50 passes of welding experiments were carried out. The welding current and voltage for every pass were obtained at a sampling rate of 5,000 Hz. The weld beads were classified into 4 quality levels by visual inspection and X-ray tests. The measured arc current and voltage were converted to a dimensionless value that was independent of the value itself, and then mapped onto a non-dimensional current-voltage plane. The welding data in the non-dimensional current-voltage plane were grouped into 50 areas (A0 ~ A49) surrounded by the lines of constant arc power and arc dynamic resistance, which changed every ±2% of mean value. The number of welding data points (N0 ~ N49) in each group were distributed in a quasi-lognormal distribution form. When this distribution skewed more leftward, the data in the center became denser and the arc became more stable. The statistical analysis of the entire welding data distribution of 50 passes showed that the cumulative probability=0.8 at the percentile α = 0.22 was an indication of weld soundness. The cumulative probability for a good bead was higher than 0.8, and was lower for a poor bead. This tendency remained consistent in a piecewise analysis of 125 sections of each bead. The sections with irregular bead or pores showed a much lower cumulative probability than 0.8, while sound bead sections had higher than 0.8, that is, a higher arc stability. The piecewise analysis makes it possible to evaluate weld quality in real time during welding Key words: Welding current and voltage, Arc stability, Arc power, Arc dynamic resistance, Weld quality, Statistical analysis