As mined ores are becoming finer grained, the use of stirred milling is increasing rapidly within the minerals industry. Due to the large power requirements for such fine grinding, optimal grinding efficiency is extremely important. There are a large number of parameters that can effect grinding, and the correct combination is required to maximise energy efficiency. Three parameters, namely stirrer speed, media size and media density, can be combined to describe the stress intensity (Sib) of the grinding beads (Blecher et al, 1996).
Db= diameter of grinding beads (m)
Pb = density of grinding beads (kg/m3)
p = density of fluid (kglm3)
vd =tip speed of stirrer (m/s)
Stress intensity effectively describes the energy available for collisions between media particles. If the energy in these collisions is too low, any ore particles trapped between the media will not be broken. Conversely, if the energy is too high, the ore will be broken, but energy wasted. The most efficient stress intensity occurs when the energy of the grinding beads is just enough to break the ore particles. Consequently, for any specific energy input to the mill, an optimum stress intensity exists which will give the smallest product size. Stress intensity is very useful for finding the best conditions for a mill to work with maximum efficiency.
All previous work on stress intensity was done on horizontal mills with disc type stirrers. The aim of this investigation to test the hypothesis:
Stress intensity, as calculated by Blecher et al, can be applied to mills with different stirrer designs and mill orientations.
To test this hypothesis, a vertical mill with a pin-type stirrer has been used. Grinding tests were carried out at the JKMRC, with power measurements and product sizing taken at set time intervals. The stress intensity was altered by using different types of grinding media. Results were then plotted on a stress intensity vs. product size graph, and compared to data available in the literature. The results where indeed similar, with a distinct drop in efficiency as the stress intensity moved away from the optimal value.
Further testwork would be required to extend the trend either side of this optimum stress intensity value, and to confirm these results. Tests at different stirrer speeds are also required, as this is a parameter in the stress intensity formula.
Investigation into the media wear was also carried out, as this is an important economic factor in stirred mill operation. Different wear rates were found for different grinding media and wear kinetics were found to follow those in the literature.
A fault tree was constructed to identify errors that might occur during both testwork and subsequent analysis. It helped explain why some unexpected results occurred. This fault tree would be invaluable for any further testwork, as errors came from some very unexpected sources.