Dynamic air classification occupies a central position in the grinding circuits of cement and several other industrial minerals. An important recent application of this device is in the reduction of the proportion of fine (< 75 µm) material contained in fine aggregate for construction that is produced by crushing. The present paper analyzes the influence of a number of variables on the performance of a pilot-scale Sturtevant classifier, aiming at providing data for mathematical modeling. It is demonstrated that the performance of the classifier is significantly influenced by feed moisture content, rotor frequency and the position of the rejection elements, being just marginally influenced by particle shape and the feed rate, within the range of the studied conditions.
Dry classification; performance evaluation; efficiency
