The dynamic behaviour of a large‐scale 250‐mm hydrocyclone: A CFD study

Abstract

The fluctuating feed solid contents usually cause instationarities in industrial hydrocyclone separation. This demands a controlling mechanism for stabilizing the separation effect. Understanding the transitional/dynamic behaviour is necessary to design such control mechanisms. In this context, the present paper is aimed to understand the dynamic behaviour of a large-scale 250-mm hydrocyclone in terms of air core development and solid concentration distributions using the modified mixture model coupled with large eddy simulation turbulence model. Numerically predicted two-phase flow field data is validated against laser Doppler velocimetry data at two axial positions. A close agreement is found between the modified mixture model predicted particle classification curve and experimental data. Longer residence time is observed with near cut size particle compared to the and fine size particles. Further, the analysis of the equilibrium radius of different solid particles inside the cyclone in the presence of turbulence is compared with the classical Kelsall experiments and observed that the classification of particles influenced by both the turbulent dispersion and the centrifugal force. Finally, the effect of feed solids concentration (10-50 wt%) on the hydrocyclone performance and particle segregation is also undertaken in this study.

Comprehensive Dense Slurry CFD Model for Performance Evaluation of Industrial Hydrocyclones

“…Thereby, a reduction in the tangential velocities can be expected. 24 As the standard ASM model-predicted mixture viscosities are similar to the water (Figure 6), the existence of high tangential velocities throughout the hydrocyclone lead to fine cut size. Similar to the experiments, a close match in the coarser size particle behavior is noted with the modified ASM model predictions.…”

“…Detailed information about this rheological model can be found in Narasimha et al 46 The modified ASM model with the aforementioned additional corrections accurately predicted the hydrocyclone and dense medium cyclone performance up to intermediate solids content. 10,17,24,47 3. NUMERICAL MODELING Figure 1 displays the grids of different cyclone geometries used for numerical studies.…”

“…8,16,17,21,27−29 The VOF model is best suited to model two-phase flows with a sharp interface between an air core and a continuous liquid phase. 2,8,24,28 A lot of studies 8,16,21,30,31 have reported Lagrangian approach for modeling the particle phase in hydrocyclone. But the discrete phase model (DPM)/ Lagrangian particle tracking (LPT) is not suitable when the feed solids content is greater than 10% by weight, because it tracks an individual particle and neglects the effect of particle concentration on the flow medium and the inability to account the particle−particle interaction.…”

“…The Eulerian and Lagrangian approaches are two types of models employed for multiphase studies in hydrocyclones. The use of the full Eulerian–Eulerian (EE) approach for performance predictions is restricted because of its high computational cost. , The simplified Eulerian approaches such as the volume of fluid (VOF) and algebraic slip mixture (ASM) model are widely adopted for simulating two-phase and multiphase flows in hydrocyclones. ,,,− This is because of solving less number of transport equations compared to full EE model. Therefore, it requires less computation time and numerical efficiency.…”

“…Therefore, it requires less computation time and numerical efficiency. Frequently, VOF coupled with RSM or LES turbulence models are selected to predict the accurate two-phase flow field and air core in hydrocyclones of various sizes under a wide range of operating conditions. ,,,,− The VOF model is best suited to model two-phase flows with a sharp interface between an air core and a continuous liquid phase. ,,, A lot of studies ,,,, have reported Lagrangian approach for modeling the particle phase in hydrocyclone. But the discrete phase model (DPM)/Lagrangian particle tracking (LPT) is not suitable when the feed solids content is greater than 10% by weight, because it tracks an individual particle and neglects the effect of particle concentration on the flow medium and the inability to account the particle–particle interaction .…”