Extending the application of JKFBC for gravity induced stirred mills feed ore characterisation

摘要

Although stirred mills have been used widely in the mineral processing concentrators for fine grinding purposes, till date a standard ore characterisation test is not available for this technology. Advanced process modelling requires ore hardness index and ore specific breakage functions for accurate predictive capability. The Julius Kruttschnitt Mineral Research Centre (JKMRC) has pioneered the initiative to develop a practical ore characterisation test for the fine particles (within stirred mill regime). The JKFBC (JK fine-particle breakage characteriser) was used to develop an index that represents the ore hardness of the stirred mill feed. The index is called Stirred mill index (SMi). This index measures the particles hardness and generate a generic tn-family curve (to establish energy base breakage function). Four types of materials were chosen for the test work i.e. Au-Cu ore, limestone, Fe ore and Au ore. The JKFBC test works were carried out by varying the applied vertical load, mill rotational speed and number of revolutions to achieve various levels of energies. The energy regimes chosen in this test work were within the typical operational conditions of the industrial scale gravity induced stirred mills. The test was modified from the original Shi's test methodology by reducing the thickness of the particles bed in the mill and specific energy consumption. The volume of the material in the mill is kept constant to avoid the particle density effect. A new tu, function that incorporates the size effect was used to describe the breakage in the JKFBC. Ore hardness of the fine particles (between 1180 and 53 mu m) was quantified through SMi. A preliminary to-t(10) family curves was developed that responds based on the hardness of the samples 7 SMi and energy. Further work in extending this procedure with other types Of ores are essential for a more robust model parameters. This preliminary development has enabled a more predictive stirred mill process model development that incorporates the ore characteristics in the selection function. (C) 2016 Published by Elsevier Ltd.