Within an investigation focused on effect of casting conditions on steel quality in an industrial beam blanks mould, different nozzle geometries were tested with Computational Fluid Dynamics modelling. The innovative nature of the work consisted in feeding with only one nozzle, whereas two nozzles in the flange-tip zone are commonly used. This configuration has the advantage of a simplified mono-slide gate casting layout, but single nozzle feeding can bring about risks of too high steel velocity in the mould, harmful for shell integrity and meniscus stability.Having in mind the mentioned constraints, different geometry solutions were checked and, to assess the solutions found, suitable indices were defined, related to flow conditions able to prevent slag entrapment at the meniscus, and hot-spotting at walls, harmful for the solid shell integrity.The modelling work gave general indications on undesired flow features and guidelines to improve reference conditions, involving number of holes and holes angle, size and shape. For the caster mould and the operating conditions under concern, a solution was found satisfying the indices, and expected to fulfil the quality requests. It consisted of a nozzle with a 50 mm diameter throat, a 50 mm× 60 mm elliptical lateral port inclined 25° downwards and a 20 mm-diameter bottom hole. A water model check with such a nozzle prototype validated the model supporting the solution identified to be used on plant.KEY WORDS: beam blanks; flow field; nozzle; cracks; shell remelting; slag entrapment. ForewordIn continuous casting of beam blanks, the occurrence of longitudinal surface cracks plays a relevant role in affecting as-cast quality. Not so many studies on surface crack occurrence in beam blanks are shown in literature. Lee and others, 1) within the as-cast beam blank shape (Fig. 1) identified the web and fillet region as mostly prone to crack formation, also noticing that shell formation is somewhat retarded at the flange-center region, due to the steel stream impingement from the feeding nozzle. The conclusions were gained based on a suitably defined crack susceptibility coefficient. Furthermore, Seok and Yoon 2) investigated the effect of steel composition on longitudinal crack formation. The most relevant results were the identification of: a) a mostly sensitive crack susceptibility range of carbon contents (0.12-0.13 wt.%); b) the effect of casting speed on crack formation (a linear relationship). Then, Hibbeler et al., 3) performed a thorough 3D thermo-mechanical analysis of beam blank steel solidification without accounting for liquid metal flow. A coupled thermal-mechanical model of steel casting was applied to accurately simulate casting of steel beam blanks, validated with plant measurements. An efficient local-global numerical procedure is given to integrate a realistic elasticvisco-plastic phase-dependent constitutive model implemented into the commercial package ABAQUS, and insights are provided into the mechanisms of shell failure triggered first by a thinner she...

No abstract