A sampling procedure validated

Extensive work was done previously on the wear of grinding media and on the comminution processes operative in ball milling based on the determination of the number-size and mass-size distributions of samples of balls taken from ball mills by a specially designed device. Proof that a sample is as nearly representative as possible of the whole mass of material from which it is drawn is usually extremely difficult. This paper reports a special case in which a direct comparison was made between the number-size and mass-size distributions obtained from a sample drawn from a mill and those from the whole ball charge emptied from the mill. Exposure of the internal defects of the grinding balls caused them to develop irregular shapes after wear had reduced their diameters by about 50 per cent. As a result of the marked change in the morphology of the grinding components, an anomaly was observed between the proportion of small components collected by the sampler and that determined in the total charge. Except for this anomaly, a marked similarity was found in the data relating to the larger components in the sample and the total charge, indicating that the sample obtained by the procedure outlined is representative of the mass of the grinding charge in a mill.