On the issue of selecting and optimizing parameters of continuous laser welding of cast iron

OBRABOTKAMETALLOV Vol. 23 No. 3 2021 technology Fig. 1. Microstructure of sample No. 5: 1 – source material; 2 – hardening area; 3 – penetration area; 4 – welding defect graphite. Due to the high crystallization rate, releasing excess phases was suppressed; as a result, the entire melt passed into a finely dispersed mixture of austenite and cementite (quasi-eutectic ledeburite). An accurate analytical description of the hydrodynamics processes, heat and mass transfer in the weld pool, affecting the weld geometry is not an easy task in view of its complexity. Therefore, to optimize the welding parameters, it is convenient to make simplified mathematical models. For this aim, a regression analysis of the obtained data was carried out, the results of which are presented in table. 3 Regression models explain a significant proportion of the dependent variable dispersion. Regression coefficients, like the model itself, are statistically significant. Regression prerequisites for each model are met. All this testifies to the close linear relationship between the weld geometry and process parameters. High laser power and at the same time high welding speed provide the melt-pool smallest width and obtain “a knife fusion” penetration (Fig. 2). In this way, narrow, deep weld seams can be obtained. Due to the low heat capacity, graphite inclusions are heated to a significantly higher temperature than the base metal. The metal vaporisation and the graphite sublimation cause excessive pressure in the weld channel, which leads to the melt displacement from the melt-pool, as a result of which discontinuities of significant size are formed in this area, mainly in the weld root part. In this case, the bead convexity considerably increases. Reducing the laser speed at a high laser power increases the weld pool width and decreases the penetration. As a result, a wide bead with a small convexity is formed. The volumes of discontinuities in the melting zone are significantly reduced (Fig. 1). However, decreasing the penetration ratio to the weld width increases the level of tensile stresses arising from the cast iron shrinkage in the weld pool. Reducing the Ta b l e 3 Regression analysis results Statistics Dependent variables Width, mm Penetration, mm Convexity, mm Hardening area, mm Laser power, kW 1.127*** (Std. Error 0.112) 2.889*** (Std. Error 0.141) 0.798** (Std. Error 0.188) 0.083*** (Std. Error 0.017) Welding speed, mm/s –0.023*** (Std. Error 0.002) 0.021*** (Std. Error 0.002) 0.015*** (Std. Error 0.003) –0.001** (Std. Error 0.001) Intercept 1.170*** (Std. Error 0.142) –1.979*** (Std. Error 0.180) –1.034** (Std. Error 0.278) 0.079** (Std. Error 0.022) Observations 20 20 10 20 R 2 0.966 0.961 0.794 0.765 Adjusted R 2 0.962 0.956 0.735 0.737 Residual Std. Error 0.115 (df = 17) 0.145 (df = 17) 0.127 (df = 7) 0.018 (df = 17) F-statistic 240.182*** (df = 2; 17) 209.358*** (df = 2; 17) 13.484*** (df = 2; 7) 27.64*** (df = 2; 17) Note: * p < 0.1; ** p < 0.05; *** p < 0.01.