Steel delivery to the continuous casting mould is a major determinant of product quality. However, although much modelling has been done to optimise steel flow patterns, evidence from measurements within the nozzle (SES or SEN) during casting is difficult to obtain. Reliable data would enhance understanding of real flows, permitting process improvements. The aim of this project, involving Corus RD & T, Sidenor I + D, MEFOS and BFI with associate partner Saarstahl, was to develop a number of potential techniques for determining the flow conditions in the casting nozzle. At Corus, an electromagnetic probe was developed, placed around the SEN. The recorded signal was related to known flow types generated in pilot plant trials, using transparent quartz casting tubes. This provided a reference for plant trials at the Corus Scunthorpe works, showing deviations from 'ideal' flow (e.g. stopper flush events). Sidenor I + D focused on stopper/nozzle pressure measurement at the Basauri works. First, this required development of a gas-tight stopper. Results clearly showed influence of pressure on nozzle flows and gave indications related to accretions at the stopper. BFI developed an ultrasonic sensor applied to the SEN, addressing significant materials and coupling issues. Progressing from laboratory development to production trials at Saarstahl, variation of the signal over time was related to changes in casting condition. Comprehensive assessment of inclusions was provided by Saarstahl in support of cleanness aspects. MEFOS investigated vibration of the SEN using laser detection. Although some correlation to nozzle flows was noted in trials at SSAB, interpretation proved difficult. MEFOS obtained better results from SEN thermal measurements, which were able to indicate biased or non-filled flows. Sidenor also used thermal measurements together with instrumented mould tests, with similar conclusions. Overall, several of these techniques appear viable. Valuable results have been obtained, indicating that real steel flows, at times, differ significantly from the ideal condition.
