Further developments in the modelling and scale up of industrial hydrocyclones

Abstract

Based on dimensional reasoning, equations have been developed for predicting the performance characteristics of industrial hydrocyclone classifiers.As a result of the present study, it is now possible to predict, without any experimental work, the relative changes in the performance characteristics of industrial hydrocyclones, namely:a. the throughput, Q,b. the classification size,' d50c',c. the recovery of water to underflow, Rf, andd. the volumetric recovery of pulp to underflow, Rv.caused by changes in the following commonly adjusted variables:(i) vortex finder diameter,(ii) spigot diameter,(iii) the feed pressure, and(iv) the solids concentration in the feed pulp.The influence of inlet diameter, cone angle and cyclone length on the behaviour of a hydrocyclone have also been quantified, albeit approximately.The influence of the characteristics of the feed solids on the behaviour of hydrocyclones has not been evaluated. As a result, data from at least one classification test is required to determine the material dependent constants in the performance equations.The reduced efficiency curve is shown to be independent of the cyclone geometry, the inlet and outlet dimensions and the size distribution of the feed solids.From a study of classification behaviour of a 120 cm cyclone it is concluded that cyclones with nominal diameters as high as 120 cm can be used as efficient classifiers.