Efficiency and grinding media filling level

Advances in materials, instrumentation, monitoring techniques, process knowledge and an ever increasing demand for efficiency and conservation have opened the way to research in areas of comminution that have been overlooked or abandoned time and time again. There are a multitude of papers and books in existence concerning the “art” of grinding. Over the last 50 years, the specific energy required to reduce a given mineral or material to a specific fineness has been cut in half and we know there may still be possibilities to reduce this even further. About 60 years ago, Fred C. Bond, George Krouse, and a few others involved in comminution research conducted some preliminary studies on low filling level of grinding media in laboratory ball mills. Initial indications were that there was a very good possibility for increased efficiency if ball mills were run with drastically reduced filling levels. Unfortunately there was only a verbal presentation and discussion on the results; no written report was made and both men have since passed away. Mr. Longhurst was fortunate to have been associated with them and his initial introduction to grinding was through them, which is how he came to know of these tests. This presentation was given at a time when a 1000 horsepower ball mill was considered a “giant” and the goal was to get as much energy from the lowest mill volume requirement. For this reason, funded research for possible efficiency gains from low ball charge filling levels was not allowed by the corporation they worked for. To put maximum energy between the trunnion bearings in a short distance required a high filling level in the mill, making the actual grinding efficiency a secondary concern. These decisions made a great deal of sense then as slide shoe supported mills were not in use. As demand for mills with increased energy grew, the bending stresses on the mill shell between trunnion bearings also grew. Mill shell thicknesses inc- eased and the head and discharge walls of trunnion supported mills became massive. The question as to whether or not energy was being effectively transmitted to the material being ground as the layers of balls increased in the grinding chamber remained unanswered. In other words, “Was there a change in efficiency relative to the filling degree of ball mills?” With the advent of shell supported mills and larger diameters, the bending stresses on mill shells were alleviated to a great degree. Slide shoe supported mills eased those concerns and made possible the lowering of “normal” filling levels of 40% to 45% to a new “normal” of 28% to 35%. Ten to fifteen years ago two events took place which seemed to support the results of the preliminary investigations of low level filling and efficiency. In the first case, one of the drives failed on a dual drive ball mill operating with a High Pressure Grinding Roll (HPGR)⁽¹⁾. Since delivery time for parts to repair the damaged drive was lengthy, a decision was made to reduce the charge in the ball mill from approximate 32% filling level to a point low enough to allow the mill to be operated with the single remaining drive. Surprisingly the result was a minimal loss in production, with a significant gain in efficiency. In the second case several years later, an opportunity arose which allowed commissioning of a new mill with a low ball charge level grinding granulated blast furnace slag to produce Grade 120 Slag cement (~5750 cm²/g). The results of this were briefly presented at the 2000 IEEE in a paper titled “SEPARATE SLAG GRINDING IN DIFFERENT MILLING SYSTEMS”⁽²⁾. Since that time, little has been done to investigate the possible benefits of low filling level of ball mills on a commercial scale. The time, expense, and possible loss in production meant it was not feasible to perform such a study on a commercial mill until recently