Abstract

Charge motion is of prime importance in the efficiency of comminution in tumbling mills. Since direct observation of charge shape and its motion in industrial mills are not possible, a combination of analytical and physical studies was used to determine charge trajectory. Software packages, which predict charge motion such as the GMT (Grinding Media Trajectory) only consider the outermost layer of charge (single ball) and ignore the charge and the interactions of grinding elements. In this research, the charge trajectory measured in a laboratory mill (model mill) with the diameter of 100cm and length of 21.5cm with the transparent end was compared with that of the GMT. Three types of polyurethane rings were accurately machined to scale down the liners arrangements at two industrial mills. To explore various charge shapes and trajectories, the model mills were operated at 55%, 70% and 85% of critical speed for five levels of mill filling (10%, 15%, 20%, 25% and 30% by volume). The special design of the model mill which enabled gradual increase of the mill length so as to minimize the impact of the end-wall effect. The experiments indicated that the ends wall effect was negligible for the model mill with the length of 10.8cm and beyond. The proposed relationships to correct the trajectory and charge shape obtained by the GMT were validated by using the new liner of the Gol-E-Gohar iron ore company AG mill. The average relative error of prediction was found to be 1%. The results indicated that when the lifter face angle increased from 7° to 30°, the distance between the charge impact point and the toe decreased from 40.1° to 11.2° for 30% filling. This meant increasing the probability of charge impacting the toe not the liner which favoured more efficient comminution practice. After converting AG mills to SAG mills on the account of liner profile change, 31% increase in throughput (from 419 to 548t/h) in addition to 4% decrease in the product size (from 516 to 496μm) were realized which was a significant contribution to the plant performance improvement.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call