Ball mill media optimization through functional performance modeling

W. Conger, J.F. DuPont, R.E. McIvor, T.P. WeldumW. Conger, J.F. DuPont, R.E. McIvor, T.P. Weldum

Mining Engineering, November 2018, pp. 28-37

The objective of the study conducted by the authors is to determine what media size(s) addition will maximize any given plant ball mill’s grinding efficiency. The functional performance parameters “mill grinding rate through the size of interest,” and “cumulative mill grinding rates” from both plant and small-scale tests are applied to this task. A plant media sizing methodology, and industrial case studies, are provided.

This work is ongoing. However, a number of key observations and conclusions can be made at this time.

  • Size by size cumulative energy specific grinding rates can be used to characterize mill grinding performance and changes in performance, as a function of media sizing. They can be used for both plant and test mill data. The cumulative grinding rate through the size of interest is the identical parameter as the ball mill grinding rate calculated from the functional performance equation. Its maximization is the technical optimization criterion for ball sizing studies.
  • Plant ball mill preferential retention of coarser particles, ignored by most population balance modeling, confounds interpretation of plant grinding rates. However, when the calculated apparent cumulative grinding rates increase from the smallest to the largest particle sizes, the potential for increasing efficiency with a media size change is greatest.
  • A suitably sized, torque-metered test mill can be used to experiment with the currently used plant ball charge and alternative ball charges. Three types of tests are being conducted: batch tests on plant ball mill feed; locked-cycle tests on plant circuit feed; and batch “grind through” tests on plant circuit feed to circuit product size.
  • Comparative batch tests on ball mill feed with the current and an alternative ball charge may show increased grinding rate through the size of interest, but lower rates at coarser particle sizes. Locked cycle tests are then needed to determine whether a net increase in grinding efficiency will result.
  • “Grind through” tests on a circuit feed to circuit product sizing (P80) that have displayed increased efficiency with a media size change in the test mill have shown similar results in closed-circuit grinding in the plant.
  • The optimum media sizing depends on the fineness of the circuit product sizing (P80), as well as the circuit feed size.
  • Multiple size charging can sometimes outperform charging any single size.
  • Different plant media sizing can result in the same grinding rate (efficiency) on the same ore at a specific particle size, even though they perform differently at other particle sizes. This explains why changing media sizing may produce no effect on circuit performance.
  • Numerous torque mill tests, in addition to the growing plant data base compiled to date, show that media charging near the size that provides highest ball mill grinding efficiency provides a degree of robustness with changing ore grindability.
  • It is rare that the media is incapable of breaking the largest particles as they become tougher. Rather, they do so at a lower rate. (References are available from the authors.)


Conger, R.E. McIvor and T.P. Weldum, members SME, are director of application engineering, ME Elecmetal; chief metallurgist, Metcom Technologies Inc and senior metallurgical engineer, Cleveland Cliffs, Inc. and J.F. Dupont, is mill superintendent, Detour Gold Inc., email rob@metcomtech.


Copyright 2012 by the Society for Mining, Metallurgy and Exploration, Inc. All rights reserved. MINING ENGINEERING magazine (ISSN 0026–5187) print and online is published monthly by the Society for Mining, Metallurgy, and Exploration, Inc., at 12999 E. Adam Aircraft Circle, Englewood, CO, 80112 USA. The paper can be accessed by subscribers here.