2016 | OriginalPaper | Chapter
Rodding in Hall-Héroult Cells: An Fea Model that Predicts Room Temperature Mechanical Properties and Cracking Tendency of Thimbles
Authors : D. R. Gunasegaram, D. Molenaar
Published in: Light Metals 2014
Publisher: Springer International Publishing
Activate our intelligent search to find suitable subject content or patents.
Select sections of text to find matching patents with Artificial Intelligence. powered by
Select sections of text to find additional relevant content using AI-assisted search. powered by
The quality and extent of the contact between the thimble and the anode comprising the anode assembly of a Hall-Héroult cell are influenced by the mechanical properties of the thimble. The contact is established when the thimble differentially expands with increasing temperature during the cell start-up phase and touches the anode surface. The size and shape of the contact area and the magnitude of the interfacial pressure are subsequently modified as the thimble deforms with further increases in temperature. Crucially, this deformation mechanism is complicated by the fact that the thimble properties vary from location to location based on processing history unique to each location. It is therefore necessary to account for such variations if realistic predictions are to be made for electrical and thermal flux profiles across the critical thimble-anode interface. In the present work, a fully coupled transient thermal-mechanical model is developed for thimble solidification using the finite element code Abaqus. This continuum scale model predicts the local mechanical properties of the slightly hypereutectic gray iron casting at room temperature by recreating the phase fractions based on local non-equilibrium cooling rates. These properties may be modified for elevated temperatures using experimentally obtained relationships available in the literature. The model also predicts the cracking tendency of the solidifying thimble based on calculated equivalent plastic strain profiles. The computation time for this model is relatively short.