Steady state thickener modelling from the compressive yield stress and hindered settling function

Abstract

An algorithm has been developed to predict steady state thickener operation from fundamental material properties, properly accounting for compression of the suspension network structure within the sediment bed. The material properties include the compressive yield stress, Py(ϕ), and the hindered settling function, R(ϕ). Py(ϕ) reflects the suspension network strength as a function of solids volume fraction ϕ, while R(ϕ) is inversely related to the permeability. The required inputs to the model include Py(ϕ) and R(ϕ) curve fits, thickener diameter as a function of height, solids density, liquid density and feed solids volume fraction. The model output is either solids throughput or solids flux as a function of underflow solids concentration, for a range of suspension bed heights. As a bonus, the solids residence time in the suspension bed can also be determined. The algorithm involves prediction of the solids throughput versus underflow solids concentration in two parts; free settling (clarification) and compression within the suspension bed (thickening). The free settling prediction utilises an adaption of the simple Coe and Clevenger method, while prediction of compression in the bed is achieved through integration of a differential equation developed from the fundamental dewatering theory of Buscall and White. The limiting steady state solids flux is the minimum of the two predicted values for each underflow solids concentration and bed height. In just minutes, this algorithm can produce tabulated and graphical results providing useful insights into the inter-relationship between solids throughput, bed height and underflow solids concentration. For steady state thickener operation, the outputs reveal three general modes of stable operation; permeability limited at high solids fluxes, compressibility and permeability dependant at intermediate solids fluxes and compressibility limited at very low solids fluxes. Knowledge of the conditions under which each of these modes is applicable enables process operators to understand the effect of variations in process conditions and assists in process optimisation.