Fluid Dynamics of the Continuous Casting of Steel

Project Lead and Collaborators: University of Leeds: Chris Jones, Onno Bokhove, Mike Fairweather. Materials Processing Institute: Alan Scholes, Chris Smith

Theme: Reacting Flows, Mixing and Safety

A collaboration between the Materials Processing Institute at Middlesbrough and the University of Leeds has been set up to study the fluid processes involved in steel manufacturing. With a group of three CDT students, Paul Allen, Caroline Shields and Fryderyk Wilczynski, we investigated the flow in the mould of a continuous caster. Molten steel is injected into the mould through a submerged entry nozzle (see left picture), and argon gas is added to prevent clogging and help remove impurities.

To get high quality steel, a steady flow pattern which doesn’t disturb the free top boundary in the mould, and reduces the turbulence, is best. This can be achieved by varying the nozzle angle of the entry of the steel into the mould. A 50% scaled water/air model was built, and the flow pattern was measured using PIV. The time-average flow of a steady case is shown in the right hand picture. CFD simulations using Fluent, with  Lagrangian tracking of the bubbles, were performed. These simulations can reproduce the observed flows in the water model, and can be run for parameters suitable for the steel/argon case. The results indicated that the turbulence level could be significantly reduced, hence improving the quality of the steel, if the nozzle angle was chosen to streamline the flow at entry into the mould as much as possible.

Building on the success of the first collaborative project, we are now investigating mixing in the molten steel ladle (see right figure above) with a group of CDT students, Andrew Clarke, Andrew Oates and Kristoffer Smedt. Various elements are added to the steel when it is in the ladle to improve quality. These need to be mixed in as rapidly as possible, and this is achieved by bubbling argon through porous plugs at the bottom of the ladle: see picture below. We are modelling the flow in the ladle induced by the bubbles, using a scaled down water/air model, measuring the flow using PIV. We are also doing CFD simulations. The objectives are first to understand how the flow pattern depends on the gas flow rate and size of the bubbles, and then to see how the flow is affected if the porous plug is put off centre. There is empirical evidence that this improves the mixing rate.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Materials Processing Institute: www.mpiuk.com

Leeds University Contact: cajones@maths.leeds.ac.uk