DEVELOPMENT OF INDUSTRIAL BENCHMARK FINITE ELEMENT ANALYSIS MODEL TO STUDY ENERGY EFFICIENT ELECTRICAL CONNECTIONS FOR PRIMARY ALUMINIUM SMELTERS D. Molenaar 1 , K. Ding 2 and A. Kapoor 2 1 CSIRO. Melbourne, Australia. 2 Swinburne University of Technology. Melbourne Australia Keywords: contact resistance, contact pressure, stub to carbon voltage drop, cast iron, anode carbon, anode assembly Abstract Process improvements of 5 MW per plant (50,000 t CO 2 e pa for coal based electricity) are possible through optimisation of the complex cast iron to carbon contacts within aluminium smelter anode and cathode assemblies. Finite element analysis is considered the tool of choice within industry for assessing potential improvements; however there are limitations with existing models regarding handling of contact resistance and carbon stress state. A study has been undertaken using thermo- electrical-mechanical finite element analysis of the cast iron to carbon contact for an anode assembly. The contact pressure and electrical resistance and its dependence on temperature have been derived from data available in the public domain. This paper presents development of the benchmark model including results. The benchmark model will be used as the reference point for the development of more advanced models in ongoing studies to assist primary aluminium smelters achieve these substantial savings in energy efficiency and reduced greenhouse gas emissions. Introduction Aluminium smelters operate at currents in the range of 100-400 kA DC. Most reduction cells operate at approximately 4.5 volts DC each and, as the theoretical reduction of alumina requires about 1.8 volts DC, it is clear that there are significant energy losses in the process. Of the excess 2.7 volts, power losses associated with electrical conductors and connections represent approximately 0.2-0.4 volts and a significant portion of these losses are contained within the complex cast iron to carbon contacts within aluminium smelter anode and cathode assemblies [1]. Figure 1 shows the components of a typical anode assembly used within the aluminium smelting process comprising (1) aluminium rod, (2) aluminium-titanium-steel transition joint or clad, (3) steel crossbar or yoke, (4) steel stubs, (5) cast iron thimbles and (6) carbon anode. The anode assembly shown in Figure 1 has been specifically designed for this study. Of particular interest in this paper is the electrical connection system of the steel stub to cast iron thimble to anode carbon, referred to as the stub to carbon (STC) connection. There are generally three methods employed by industry to study potential energy efficiency savings that may be present in the anode assembly STC connection. The first has historically been to undertake extensive in-plant trials of suitably instrumented anode assemblies set into actual reduction cells and monitored over the entire period of operation [2]. The operating environment within a reduction cell offers many technical challenges to overcome in order to ensure that a measurement system is sufficiently robust and will generate valid data throughout the measurement campaign. In-plant trials are logistically demanding, expensive and labour intensive. Figure 1. Components of a typical anode assembly They require many months preparation and will generally mandate a very high number of sample repeats to form a statistically valid result due to the inherently high levels of variation within and between anode assemblies. The second method is the off-line experimental laboratory. Major issues with this approach are; (a) significantly reduced current density making it very difficult to detect small changes in contact resistance, (b) heating and protection of the anode carbon from combustion during testing and (c) it is common to employ smaller portions of an actual anode in the test rig resulting in non-representative overall geometry which will cause significant alterations to the constriction of current through the portion of carbon being tested. Also, in the latter case the smaller sized carbon may not withstand the stress generated from differential thermal expansion and will likely cause the carbon to crack, invalidating the test data. It is not practical or efficient to test entire anode assemblies in an off- line laboratory. The third approach employed is finite element analysis (FEA) and this is considered the tool of choice within industry for assessing these potential improvements [3-6]; however there are numerous limitations with existing models, the two most important ones being the handling of contact resistance and carbon stress state. The purpose of this paper is to describe in detail the development of a benchmark finite element model using a proper analysis procedure to achieve a thermo-electrical-mechanical analysis of the cast iron to carbon contact for an anode assembly. This benchmark model will be used as the reference point for the development of more advanced models in ongoing studies towards assisting primary aluminium smelters achieve substantial savings in energy efficiency and reduced greenhouse gas emissions. 985 Light Metals 2011 Edited by: Stephen J. Lindsay TMS (The Minerals, Metals & Materials Society), 2011