الفهرس 
Chapter One: IntroductionOnly 14 pages are availabe for public view
 |  | from | 145 |  |  |
| | | from | 145 | | |
Abstract
Cyclone separators are considered one the most important devices that are used for the purpose of separating solid particles from a fluid. The importance has came from its simple design, as there is no need to connect it with a source of power and lacking of moving parts. A cyclone separator essentially consists of three parts: the intake cylinder (Barrel), the cone and the immersion tube (cyclone hopper). Many attempts have been done to understand and predict cyclone performance in terms of pressure loss and collection efficiency (cut-off size). In this thesis, a new cyclone separator design without barrel part has been studied numerically and experimentally to investigate the gas flow field and predict its performance. The new cyclone design was modified by introducing the tangential inlet of the standard Stairmand cyclone into the cone part directly. The design modification performed by increasing the cone length gradually using two ways, one with fixing the cone diameter (design (A) cyclones) and the other with fixing the cone angle (design (B) cyclones).
For the numerical solution, a computational mesh for the standard Stairmand cyclone was generated using ICEM 14.0 code software. The Reynolds stress model (RSM) and the discrete phase model (DPM) were solved by the commercial CFD code ”ANSYS-Fluent V. 14.0” to predict the turbulent flow field and the dispersed phase inside the cyclone. The model was validated by comparing the numerical results to published experimental data. The velocities predicted by the RSM and the pressure DROP were found in good agreement with the experimental data. Also, a grid independence test was performed to ensure that the model provides an acceptable level of overall grid independency and provides reasonable computational time.
For the gas flow field, the results show that the pressure DROP and the value of tangential and axial velocity components increase with increasing the cone length with fixed cone angle, design (B) cyclones. The pressure DROP increases about 25% greater than the standard design, while the maximum value of tangential velocity increases about 2.26 times of the inlet velocity. It was found also that increasing the cone length with the same cone diameter increases the back flow and the lip leakage below the vortex finder, design (A) cyclones. The back flow and the lip leakage phenomena were observed to be reduced by increasing cone length with fixed cone angle. The effect of contraction ratio (D_x⁄D), which is defined as the ratio between the vortex finder diameter and the cone diameter, on the turbulent kinetic energy and the velocity components was also investigated. As the contraction ratio decreases the maximum value of the turbulent kinetic energy and the tangential and axial velocity components increase. The pressure DROP was fitted into a correlation, represented by a dimensionless form using Euler number (Eu), as a function of dimensionless quantities Reynolds number and contraction ratio. The new equation was found to be:
Eu=0.9×〖Re〗^(0.163)/((D_x⁄D)^(0.5)×(l_c⁄D)^(0.5) )
The dispersed phase, results show that the collection efficiency of design (B) cyclones is significantly increased as a result of increasing the tangential velocity due to decrease of contraction ratio, meanwhile the pressure DROP increases. The critical Stokes number, corresponds to the cut-off size, was calculated for the tested cyclones at different Reynolds numbers. The calculated data was used to derive a new correlation for critical Stokes number in terms of Reynolds number and contraction ratio to help predicting the cyclone cut-off size. The new equation was found to be:
〖Stk〗_50=0.003×((D_x⁄D)^(0.5) (l_c⁄D)^(0.5))/〖Re〗^(0.12)
The derived dimensionless equation for Euler number was verified experimentally with three different types of industrial cyclone without barrel part working in Upper Egypt Flour Mills Co., Sohag, Egypt. The calculated pressure DROP from the new equation were in good agreement with the measured pressure DROP for the industrial cyclones. This indicates that the new derived equations in this thesis are valid for calculating the performance of cyclones without barrel part in the industrial applications.