Error analysis for bond work index determinations part 1: Accuracy and reproducibility

Abstract

An error analysis of the Bond rod mill work index is performed. A first source of error is the representativity of the sample used for the test, estimated using Gy's sampling theory. A second source is the test itself. Test errors likely to originate from screening are estimated using the ASTM specifications, and from grinding using experimental repeats. Accuracy ranged between 0.20 and 0.33 kWh/st, and reproducibility was estimated at 0.13 kWh/st. The main sources of error were found to be screen mean opening size (for accuracy) and grinding reproducibility (for both accuracy and reproducibility). Sampling errors were found to be insignificant.