Prediction of process input interactions of Floatex Density Separator performance for separating medium density particles C. Raghu Kumar a, ⁎, Srijith Mohanan a , Sunil Kumar Tripathy a , Y. Ramamurthy a , T. Venugopalan a , Nikkam Suresh b a Tata Steel Limited, Jamshedpur, Jharkhand, India b Indian School of Mines, Dhanbad, Jharkhand, India abstract article info Article history: Received 31 March 2010 Received in revised form 20 May 2011 Accepted 22 May 2011 Available online 27 May 2011 Keywords: Hindered settling Floatex Density Separator Classification Galvin model The Floatex Density Separator (FDS) is a classifier-concentrator. In the concentrator fluidization coupled with hindered settling leads to separation in a teeter bed column. Due to the apparent density of the suspension, the viscosity of the bed offers more resistance and particles settle in terms of the drag force. The separation principle of the FDS is complex, hence it was decided to study separation based on size alone by considering interactions between different process parameters to predict the performance of FDS. The effects of four important operating parameters—set point, teeter water flow rate, feed rate and pulp density of feed—on the performance of the FDS have been studied through simulation. From the theoretical values, interaction between various process inputs has been studied. A detailed understanding has been established about particle behavior under different feed input conditions for pure silica (S.G. 2.65). The predicted values are validated against the experimental results. It was found that variations in TW flow rate and feed rate are predicted with considerable degree of accuracy. © 2011 Elsevier B.V. All rights reserved. 1. Introduction Floatex Density Separators (FDS) are relatively new, affecting separation based on both size and density of the solids. A FDS is a classifier-concentrator in which fluidization coupled with hindered settling (resulting in a pseudo liquid) results in effective separation in a teeter column. Feed slurry of definite solid concentration is fed from top through a feedwell and teeter water (TW) is introduced from the bottom uniformly throughout its cross section. The heavier particles settle through the dense medium, whereas the lighter particles are hydraulically transported to the overflow. Separation in FDS is quite complex, since besides the apparent density of the suspension, the viscosity of the bed offers more resistance to particles from settling due to the drag force. Unlike in dense media separation, where the media density alone dictates the separation, in FDS both the pseudo liquid density and upward liquid (teeter) velocity have substantial influence on the separation. Among all operating parameters, the set point and the teeter water flow rate are the most important operating parameters which govern the nature of the bed. A schematic diagram of FDS is shown in Fig. 1. It essentially consists of a vertical square tank with a conical bottom. Feed slurry is introduced through a feed well and TW is introduced uniformly from the bottom of the square tank through a bank of perforated tubes placed in a plane adjacent to each other. A pressure sensor is fitted at the bottom of the square tank. An underflow pinch valve, automatically operated using a PID controller, provides a means to dewater the underflow product. Richardson and Zaki (1954) have proposed an empirical slip velocity model to describe the particle movement in sedimentation and liquid fluidization processes where slip velocity was defined as a function of void fraction and particle terminal settling velocity. Lockett and Al-Habbooby (1974) have described hindered settling of two-species suspension. However it was claimed by Galvin et al. (1999a) that this correlation was not valid for species of varying densities. They proposed an empirical slip velocity equation consid- ering the terminal settling velocity and the density difference. As per their claim and through work of other workers in the related area (Galvin et al., 1999b, Sarkar et al., 2008, Sarkar and Das, 2010), the model is applicable to particles of varying densities in homogeneous suspensions. Though, it is well established that in a teeter bed separator both the density and size play significant role (Galvin et al., 1999b, Sarkar et al., 2008), the effect of process parameters on the behavior of teeter bed separator has not been carried out in detail. Since the separation principle of an FDS is quite complex, in the present study it was decided to study the size separation effect with the help of the interaction effects between different process variables. All the work has been done using a single density material (using silica of S.G. 2.65) of size range b 1 mm. International Journal of Mineral Processing 100 (2011) 136–141 ⁎ Corresponding author. E-mail address: chittapragada_rk@yahoo.com (C.R. Kumar). 0301-7516/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.minpro.2011.05.011 Contents lists available at ScienceDirect International Journal of Mineral Processing journal homepage: www.elsevier.com/locate/ijminpro