Small diameter (1＇＇ - 2＇＇) hydrocyclones are used in different applications, in industrial minerals processing and in desliming ahead of flotation. The study of hydrocyclone performance, in terms of both fundamental and operational variables, has been undertaken by several authors. Empirical models are used for predicting the hydrocyclone performance in terms of capacity, cut size (d50c), water split and partition curves. However, these models have been developed for larger hydrocyclones, while for smaller cyclones little modelling has been done. Only a few previous studies exist, which propose either new model equations or modifications to the existing ones. This paper reports experimental research on small hydrocyclones. The suitability of existing models has been checked and their accuracy compared. The experimental results show, in some aspects, coherency with what is already known. In other aspects, the results are in disagreement with previous findings. Accordingly, new model equations are proposed, which improve the predictive capability for small hydrocyclone performance.