12th International Conference on Zinc & Zinc Alloy Coated Steel Sheet - GALVATECH 2021

Accurate and stable snout temperature can improve the stability of Zn pot temperature and bath chemistry, avoid dross related defects, reduce Zn dust formation, and achieve consistent galvannealing performance. In order to make sure the reliability/accuracy of the snout pyrometer (multiwavelength) for various steel grades (e.g. DP600, DP780 and DP980 etc.) and to monitor the deterioration of the pyrometer due to Zn dust buildup or other causes, predictive models were developed for #5CGL, ArcelorMittal Dofasco. Two approaches (a regression method based on rapid cool wedge-pyrometer readings, and a Zn bath thermal balance method based on Zn pot thermocouples readings) were developed and compared. During stable periods, with the regression method, the prediction bias (vs. snout pyrometer) is -0.1 ~ +0.8 0C, the prediction accuracy (1σ) is ≤ ±2.1 0C. With the thermal balance method, the prediction bias is -0.5 ~ +0.4 0C, the prediction accuracy is ≤ ±2.9 0C. The deviation between two models is ≤ ±2.4 0C. The snout pyrometer generally provides a reliable measurement, including for DP grades. However, large error can occur occasionally (~2% of coils), with measurements lower than predictions by 10~40 0C, which correspond to sudden changes in surface emissivity (related mostly to sudden line speed slowdown and/or product transition). The models can be applied to monitor the performance of the snout pyrometer, Zn bath thermocouples and Zn pot heating inductors.