There have been a number of researchers that have investigated comminution and classification, however not a lot of this has been focused on the cement industry. This thesis utilised Whiten’s Perfect Mixing Ball Mill Model to model the different materials in cement manufacture. In addition the Tromp curve was used to assess the efficiency of separation.
Samples were taken at Sunstate Cement Ltd to obtain data to use in the models. The samples of the feed material, limestone, gypsum and two types of clinker, Adelaide Brighton Cement (ABC) and Taiheiyo were prepared to a size of approximately 1000ìm. These samples were then crushed using the drop weight tester at the JKMRC. The particle size distribution (PSD) was found for all streams to and from the separator. The breakage test and PSD data were then used in the ball mill model.
The breakage tests revealed all four materials have basically the same breakage characteristics. The two clinkers (ABC and Taiheiyo) and the limestone, have very similar breakage characteristics. These three materials all have the same gradient, for percent passing versus size, which suggests they all break in a similar manner. Gypsum is the only material that differs. The gypsum has a slightly different curve, however follows the basic shape of the other materials relatively closely, showing a steep gradient for large sizes and more gradual for the smaller sizes.
It was discovered from the model that the gypsum is significantly harder than the other materials and thus requires a greater amount of energy to be crushed. It was also discovered that the limestone was considerably softer than the other materials and is crushed to a smaller size fraction for the same energy applied.
Determining the relative hardness of each material is quite significant in the overall process as this affects the amount of energy required for the desired size reduction. This could also effect the amount of time the material needs to be in the mill and as a result could increase or decrease the milling rate depending on the hardness of the material.
This thesis shows how analysis of clinker grinding can be undertaken, but clearly there is potential for additional work. A number of suggestions are made in this section for further work and analysis to achieve the ultimate goal of a more efficient cement processing plant.
It is suggested that a greater amount of data be obtained for the feed materials. This should include; more breakage tests conducted over a larger range of particle sizes, drop heights and energies. Further work should also be completed over the separator, as this is a major issue in relation to the total efficiency of the process.
It is believed that the data contained within this thesis has the potential to contribute to the development of a more efficient cement milling process.