Elsevier

Minerals Engineering

Volume 20, Issue 3, March 2007, Pages 290-302
Minerals Engineering

CFD validation for flyash particle classification in hydrocyclones

https://doi.org/10.1016/j.mineng.2006.10.009Get rights and content

Abstract

The investigation pertains to establishing a simulation methodology for understanding the flyash classification characteristics of a 76 and 50 mm diameter hydrocyclone where the work was carried out using commercially available CFD software. Comparative results on the simulated and experimental water throughput, split values are presented. Results indicted that there is a good match in water split between the experimental and simulated values with error values below 10% at different hydrocyclone designs. Further a discussion is made on the flow features at comparable ratio of cyclone diameter to spigot opening in the 76 and 50 mm designs. The vertical core region around the cyclone axis having static pressure equal to or below the atmospheric pressure is examined to be increasing in diameter from bottom of the spigot opening till the interface where vortex finder joins the main cylindrical cyclone body and remains more or less similar at the vortex finder outlet. The diameter of this zone at the spigot outlet is 0.6 and 5.4 mm at 3.2 and 9.4 mm spigot openings in case of 50 mm diameter hydrocyclone. The diameter of the core at spigot outlet is found to be around 9.2 and 11 mm at 15 and 20 mm for spigot openings in case of 76 mm diameter hydrocyclone.

Classification of flyash particulates is simulated through discrete phase modeling using particles injection technique and the simulated results are further validated with suitably performed experiments. With 50 mm diameter hydrocyclone, reasonable predictions are observed at 9.4 mm spigot opening. Considerable deviation in particle distribution points with this hydrocyclone is observed at narrowest spigot diameter of 3.2 mm. The simulated values of d50 in case of 50 mm diameter hydrocyclone are 8 and 10 μm at 9.4 and 3.2 mm diameter spigot openings. Better predictions are obtained with 76 mm diameter hydrocyclone at both 10 and 15 mm diameter spigot openings. Similarly, the simulated d50 values are 14 and 20 μm at 15 and 10 mm diameter hydrocyclones. Possible reasons for deviations in the results relating the spigot opening, solids concentration at the underflow and in turn role of slurry viscosity on the air core diameter are proposed.

Introduction

Hydrocyclone is a versatile unit operation for applications in liquid–liquid separation, liquid–solid separation, air–solid separation, and solid–solid separation. Due to several industrial advantages, like good separation efficiency, ease in operation, high throughput, less maintenance, less floor space requirement etc., the hydrocyclones are very popular in large number of industrial sectors like mineral processing, environmental, food processing, chemical engineering etc. The works related to understanding the principles began only in mid fifties (Kelsall, 1952) though the unit was originally invented in late 18th century. Kelsall’s studies on the axial, radial and tangential velocity profiles formed the basis for subsequent research on hydrocyclones. Due to the complex phenomenon involved in analyzing flow behavior coupled with non-availability of high-speed computational systems most of the research works till the recent past were focused on the empirical modeling (Lynch and Rao, 1975, Plitt, 1976). These models are material specific and their applications are limited to the boundary conditions within which the model suitability is evaluated. However, the advent of high speed computational systems, in last couple of decades made researchers focus performance simulations using Computational Fluid Dynamics (CFD) techniques (Pericleous and Rhodes, 1986, Hsieh and Rajamani, 1988, Monredon et al., 1992, Rajamani and Milin, 1992, Dyakowski and Williams, 1993, Dyakowski et al., 1994, Dyakowski et al., 1999, Malhotra et al., 1994, Hargreaves and Silvester, 1990, Devulapalli and Rajamani, 1996, Griffths and Boysan, 1996, Slack and Wraith, 1997, Slack and Boysan, 1998, Stovin and Saul, 1998, Suasnabar and Fletcher, 1999, Slack et al., 2000, Slack et al., 2003, Ma et al., 2000, Nowakowski et al., 2000, Nowakowski et al., 2004, Grady et al., 2002, Grady et al., 2003, Nowakowski and Dyakowski, 2003, Schuetz et al., 2004, Cullivan et al., 2003, Cullivan et al., 2004, Delgadillo and Rajamani, 2005, Narasimha et al., 2005, Narasimha et al., 2006, Udaya Bhaskar et al., in press). Gradual evolution has taken place from 2D axi-symmetric studies of hydrocyclones to 3D non-axisymmetric geometry simulation methodologies. Any numerical technique needs development of suitable methodology and extensive validation with the experimental data before its real applications in design and performance simulation. Among the reported studies, the validation was mostly carried out with water flow characteristics and only a few of them are related to solids separation behavior. The present study involves simulation of water flow behaviour at different design conditions of 50 and 76 mm diameter hydrocyclone and particle size distribution behaviour under both simulated and experimental conditions. Simulation was carried out using commercially available CFD software ‘Fluent 6.1.22’. The simulation results on cyclone cut size d50 values at different spigot openings are validated with experimental data generated in the laboratory treating flyash at different concentration of solids in the feed slurry.

Section snippets

Geometry

Experimental and CFD simulation works are carried out on 76 and 50 mm diameter standard hydrocyclone geometries. The geometrical description is discussed separately as follows.

Numerical simulation

A segregated, steady state, 3D double precision implicit solver was used for simulating the flow and turbulence inside the hydrocyclone. The pressure interpolation scheme adopted was PRESTO (Pressure staggered option), which is useful for predicting highly swirling flow characteristics prevailing inside the cyclone body (Fluent Europe Ltd., 2002). In order to reduce the effects of numerical diffusion, higher order discretization schemes are recommended for simulating cyclones. Accordingly, a

Experimental

The experimental setup consisted a slurry tank of 200 L capacity mounted on a stable platform. A centrifugal pump with 3-phase, 5.5 kw motor was connected to the slurry tank at the bottom. Feed slurry consisting of flyash material at different solids consistency was pumped into the cyclone body through the pipeline connected to the pump. The other end of the pipeline was connected to the inlet opening of hydrocyclone in study. The pressure drop inside the cyclone was maintained at required level

Validation results

The experimental and simulation results of inlet flow rates (water entering the cyclone) and water-split (% report of total water into the overflow) into overflow product of 76 and 50 mm diameter hydrocyclone are presented in Table 3.

It can be observed from the table that the actual and simulated water throughput values obtained in a 50 mm hydrocyclone are found matching over a range between 0.42 and 0.45 kg/s. Similarly, the experimental values of water flow split into the overflow product varied

Conclusions

  • A CFD simulation and validation study on 50 and 76 mm hydrocyclone designs has indicated that the water distribution values in terms of throughput and split closely match with the experimental results at all cyclone design conditions.

  • Diameter of the core portion associated with static pressure values below the atmospheric pressure is found to broaden from the spigot opening till the interface where the vortex finder joins the main cylindrical body and further the diameter remains more or less

References (35)

  • K.A. Pericleous et al.

    The hydrocyclone classifier—a numerical approach

    International Journal of Mineral Processing

    (1986)
  • S. Schuetz et al.

    Investigations on the flow and separation behavior of hydrocyclones using computational fluid dynamics

    International Journal of Mineral Processing

    (2004)
  • M.D. Slack et al.

    Advances in cyclone modelling using unstructured grids

    Transactions Institute Chemical Engineers

    (2000)
  • V.R. Stovin et al.

    A computational fluid dynamics (CFD) particle tracking approach to efficiency prediction

    Water Science and Technology

    (1998)
  • A.J. Delgadillo et al.

    A comparative study of three turbulence-closure models for the hydrocyclone problem

    International Journal of Mineral Processing

    (2005)
  • B. Devulapalli et al.

    A comprehensive CFD model for particle-size classification in industrial cyclones

  • T. Dyakowski et al.

    Simulation of non-Newtonian flow in a hydrocyclone

    Chemical Engineering Research and Design

    (1994)
  • Cited by (43)

    View all citing articles on Scopus
    View full text