Optimisation of Operational Parameters of a Spiral Classifier Using Design of Experiment (DOE)

Abstract Classifying Mn Ore to improve upon the grade and the properties has become a crucial activity for the Mn industry since it increases the market value of the ore. Ghana Manganese Company (GMC) has renovated their oxide washing plant by integrating it with a spiral classifier to make a batch system operating process. Particle size of <3.35mm (Mn grade of 43-47%) obtained from the primary section of the plant served as feed to the classifier and with the plant condition (i.e. at 60 min washing time, 30 rev/min speed of spirals and feed tonnage of 6 t), Mn grade of 49% was achieved. This paper focused on the optimisation of some selected operational parameters of the classifier to obtain a Mn grade >50% using Design of Experiment (DOE). Series of test works were designed using the DOE for the classifier using the constraints of washing time (30-90 min), speed of spirals (20-40 rev/min) and feed tonnage (6-9 t). The outcome of the test work after simulation showed that all the selected parameters had a great influence on Mn grade. The spiral speed and feed tonnage correlated negatively to the Mn grade with washing time correlating positively. Operating the spiral classifier at a feed rate, spiral speed and washing time of 6 t, 25 rev/min, and 30 min, respectively, yielded Mn grade of 53%. A Confirmatory test using the established conditions gave a Mn grade of 53%, which is a 4% increment in the previous Mn grade which was 47%. The outcome of the studies is the new established operational conditions which is adhered to by the plant, producing a manganese concentrate grade ranging between 52-54%.   Keywords: Design of Experiment (DOE), Spiral Classifier, Grade, Manganese