Author's personal copy

Powder Technology 191 (2009) 349–357

Contents lists available at ScienceDirect

Powder Technology
j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / p o w t e c

CFD analysis of flow field in square cyclones

Arman Raoufi, Mehrzad Shams ⁎, Homayoon Kanani
Department of Mechanical Engineering, K.N. Toosi University of Technology, Pardis St., Mollasadra St., Vanak Sq., Tehran, Iran

a r t i c l e i n f o a b s t r a c t

Article history: In this study, computational fluid dynamic method is used to predict and evaluate the flow field inside a
Received 19 August 2008 square cyclone. The flow field is calculated using 3D Reynolds-averaged Naveir–Stokes equations. The
Received in revised form 30 September 2008 Reynolds stress transport model (RSTM) is used to simulate the Reynolds stresses. The Eulerian–Lagrangian
Accepted 6 November 2008
computational procedure is implemented to predict particle trajectory in the cyclone. The Newton's second
Available online 27 November 2008
law is used to study the particle trajectory with modeling the drag and gravity forces acting on the particles.
Keywords:
The velocity fluctuations are simulated using the discrete random walk (DRW). Two square cyclones which
CFD have different geometries are studied. The cyclones are simulated at different flow rates. The details of the
Square cyclone flow field are studied in the cyclones and the effect of varying the flow rates is observed. Tangential velocity
Two phase is investigated in different sections inside the square cyclone. Contour of pressure and turbulence intensity is
Flow field shown for different inlet velocities inside the cyclones. It is observed that different geometries, also different
inlet velocities, could affect on the pressure drop. The collection efficiency and the flow patterns obtained
numerically are compared with the experimental data and good agreement is observed.
© 2008 Elsevier B.V. All rights reserved.

1. Introduction separation mechanism of a square cyclone at high inlet particle

concentration. They proposed an instantaneous separation model
Due to relative simplicity to fabricate, low cost to operate, and well based on experimental observation and measurement. Junfu et al. [6]
adaptability to extremely harsh conditions, cyclones have become one investigated the square cyclone at 75 t/h CFB boiler and presented a
of the most important gas–solid separators which are preferably model to study the cyclones. Effects of different parameters including
utilized in both engineering and process operations. Cyclone is a key length, diameter of vortex finder, and inlet velocity on performance of
part for the circulating fluidized bed (CFB) boiler which has great square cyclones were studied by Qiang et al. [11]. Su and Mao [15] used
effects on the combustion efficiency, the circulation rate and the a three-dimensional particle dynamic analyzer (3D-PDA) to under-
desulfurization efficiency by the circulation of the solid particles in the stand the nature and characteristics of the suspension flow in the
furnace. The performance of a cyclone is critical to the boiler's safe and square cyclone separator and found out the factors affecting particle
economic operation. The arrangement and structure of the cyclone motion. The turbulent flow field inside a square cyclone was
have influence on the overall arrangement of a boiler. experimentally investigated by Su [16] to study the mechanism of
Conventional cyclone which has circular cross section was the particle separation and provide guidance for the optimization of its
commonly used cyclone for the CFB boiler. With the development of structure. Junfu et al. [7] evaluated the performance of advanced
large CFB boilers, the huge body of the conventional cyclone became a water-cooled square cyclone with curved inlet. The results were
major shortcoming because of the thick refractory wall that needs a compared with other cyclones through fly ash analysis and showed
long period to start the boiler. An alternative way to overcome these that the overall performance of the square cyclone in such capacity is
problems is the use of square cyclone. A square cyclone has more compatible with the conventional cyclones.
advantages over the conventional cyclone including convenient Laser Doppler anemometry (LDA) and hot-wire anemometry are
construction, easier membrane wall arrangement, shorter start–stop frequently employed to study the flow pattern in the cyclones
time and at the same time easy integration with the boiler [17]. experimentally. Recently, research efforts by computational fluid
A considerable number of experimental investigations have been dynamics are frequently carried out for the resolution of flow field
performed on the square cyclones. Wang et al. [17] studied the and dust particle behavior with different degrees of numerical and
modeling accuracy in order to assist in the time consuming experi-
mental works. In conjunction with the complex flow structure,
numerical simulation is momentarily not able to completely sub-
⁎ Corresponding author. Tel.: +98 912 359 6236; fax: +98 21 88877273. stitute experiments but can reduce, to a certain degree, experimental
E-mail address: [email protected] (M. Shams). costs for design and optimization.

0032-5910/$ – see front matter © 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.powtec.2008.11.007
 Author's personal copy

350 A. Raoufi et al. / Powder Technology 191 (2009) 349–357

CFD has a great potential to predict the flow field characteristics the cyclones are shown. Tangential velocity profiles and velocity
and particle trajectories inside the cyclone as well as the pressure vectors in different sections are investigated. The numerical results
drop [4]. The complicated swirling turbulent flow in a cyclone places are compared with experimental data and good agreement is
great demand on the numerical techniques and the turbulence observed.
models employed in the CFD codes. CFD was widely used to
investigate flow field inside conventional cyclones. It has been 2. Flow simulation
shown in the previous works on the prediction of the cyclone
performance under different operating temperatures and inlet For an incompressible fluid flow, the equations of continuity and
velocities (Gimbun et al., 2005). Raoufi et al. [13] used computatio- balance of momentum for the mean motion are given as
nal fluid dynamics to simulate and optimize vortex finder of
conventional cyclones. Although many numerical works have been
Au i
conducted on the conventional cyclones, there is a little numerical =0 ð1Þ
Axi
study about square ones. As numerical investigations of square
cyclones could have an important role to better understanding of
the flow parameters, this study is intended to obtain detailed Au i Au i 1 Ap A2 u i A
flow information by CFD simulation within square cyclones. Two + uj =− +m − Rij ð2Þ
At Axj ρ Axi Axj Axj Axj
cases with different geometries are selected to study. Case 1 is the
square cyclone studied by Wang et al. [17] with upward outlet.
Also, a square cyclone investigated by Su and Mao [15] is where ūi is the mean velocity, xi is the position, p̄ is the mean
simulated numerically, as case 2. The square cyclones are modeled pressure, ρ is the constant gas density, v is the kinematic viscosity, and
at different flow rates and flow fields are evaluated inside these Rij = u i ′j is the Reynolds stress tensor. Here, ui′ = ui − ūi is the ith fluid
¯¯¯¯¯
′u
cyclones. Contours of pressure and turbulence intensity within fluctuation velocity component.

Fig. 1. The geometry of cyclone for (a) case 1 and (b) case 2, unit: mm.
 Author's personal copy

A. Raoufi et al. / Powder Technology 191 (2009) 349–357 351

Fig. 2. Generated mesh of the cyclones for (a) case 1 and (b) case 2.

The RSTM provides differential transport equations for evaluation are random functions of space and time. In this study, a discrete
of the turbulence stress components i.e., random walk (DRW) model is used for evaluating the instantaneous

   velocity fluctuations. The values of u′, v′ and w′ that prevail during the
A A A vt A Au j Au lifetime of the turbulent eddy, Te, are sampled by assuming that they
Rij + u k Rij = Rij − Rik + Rjk i ð3Þ
At Axk Axk σ Axk
k Axk Axk obey a Gaussian probability distribution. In this model the instanta-
    neous velocity in the ith direction is given as
e 2 2 2
−C1 Rij − δij k −C2 Pij − δij P − δij e
k 3 3 3 qffiffiffiffiffiffiffiffiffiffi
i = f uV
uV i uV i: ð6Þ
where the turbulence production terms are defined as
In Eq. (6), ζ is aq
zero-mean,
ffiffiffiffiffiffiffiffiffi unit-variance, normally distributed,
Au j Au i 1
Pij = −Rik −R ; P = Pij ð4Þ random number, uV uV is the local root mean-square (RMS)
Axk jk Axk 2 i i

with P being the fluctuation kinetic energy production. vt is the turbulent

(eddy) viscosity; and σk =1, C1 =1.8, C2 = 0.6 are empirical constants [8].
The transport equation for the turbulence dissipation rate, ε, is given as
 
Ae Ae A  mt  Ae e Au e2
+ uj = m+ e −C e1 Rij i −C e2 : ð5Þ
At Axj Axj σ Axj k Axj k

In Eq. (5), k = 12 uiVuV

i is the fluctuation kinetic energy, and ε is the
turbulence dissipation rate. The values of constants are σε = 1.3,
Cε1 = 1.44 and Cε2 = 1.92.
The dispersion of small particles is strongly affected by the
instantaneous fluctuation of fluid velocity. The turbulence fluctuations

Table 1
False time steps used for the simulation

Parameters False time step

Pressure 0.2
u (x-velocity) 0.4
v (y-velocity) 0.4
w (z-velocity) 0.4
k (turbulent kinetic energy) 0.5
ε (turbulent dissipation rate) 0.5
Reynolds stresses 0.5 Fig. 3. Comparison of numerical, mathematical and experimental [17] result for
collection efficiency in the case 1.
 Author's personal copy

352 A. Raoufi et al. / Powder Technology 191 (2009) 349–357

fluctuation velocity in the ith direction; and the summation conven- Eulerian–Eulerian and the Eulerian–Lagrangian approaches. The
tion on i is suspended. former is based on the concept of interpenetrating continua, for
The characteristic lifetime of the eddy is defined as a constant which all the phases are treated as continuous media with properties
given by analogous to those of a fluid. The Eulerian–Lagrangian approach
adopts a continuum description for the liquid phase and tracks the
Te = 2TL ð7Þ discrete phases using Lagrangian particle trajectory analysis.
where TL is the eddy turnover time given as Te = 0.3(k / ε) in the RSTM. In present study, one way coupling method is used to solve of two-
The other option allows for a log-normal random variation of eddy phase flow and the Eulerian–Lagrangian approach is implemented for
lifetime that is given by simulation of second discrete phase (particles). In this model, the air is
the continuous phase and the particles are treated as the dispersed
Te = −TL logðr Þ ð8Þ discrete phase. The volume-averaged and steady state Navier–Stokes
equation is solved for the gas phase. The particle motions are
where r is a uniform random number between 0 and 1. The particle is simulated by the Lagrangian trajectory analysis procedure. Forces
assumed to interact with the fluid fluctuation field, which stays fixed acting on the dispersed phases include drag and gravity. The discrete-
over the eddy lifetime. When the eddy lifetime is reached, a new value phase equations are solved with the Runge–Kutta for particles.
of the instantaneous velocity is obtained by introducing a new value of To calculate the trajectories of particles in the flow, the discrete
ζ in Eq. (6). phase model (DPM) was used to track individual particles through the
There are two main approaches to modeling multiphase flows continuum fluid. The particle loading in a cyclone separator is
that account for the interactions between the phases. These are the typically small, and therefore, it can be safely assumed that the

Fig. 4. Contour of static pressure (Pa) for two cyclones at different inlet velocities. Case 1 (a) v = 22, (b) v = 30 and case 2 (c) v = 20, (d), v = 28.32, unit: m/s.
 Author's personal copy

A. Raoufi et al. / Powder Technology 191 (2009) 349–357 353

presence of the particles does not affect the flow field (one-way order to make sure that the solution is not grid dependent. The finite
coupling). volume methods have been used to discretize the partial differential
The equation of motion of small particles, including the effects of equation. The SIMPLE method is used for pressure–velocity coupling
nonlinear drag and gravitational forces, is given by and the second-order upwind scheme is implemented to interpolate
the variables on the surface of the control volumes. Turbulence
duPi 3vCD Rep
fluctuations are simulated using Reynolds Stress Transport Model
= ui −uPi + gi ð9Þ
dt 4d2 S (RSTM). The computation is continued until the solution converged
with a total relative error of less than 0.0005. The Lagrangian method
dxi is used for tracking of particles in the simulation. For investigation of
= uPi : ð10Þ
dt particles trajectory inside the cyclones, seven and nine millions of
particles are released at the inlet of cases 1 and 2, respectively. The
Here, uPi is the velocity of the particle and xi is its position, d is the choice of the time step influences the convergence behavior: if taken
particle diameter, S is the ratio of particle density to fluid density, and too large the simulations diverge, if taken too small the computation
gi is the acceleration of gravity. The buoyancy, virtual mass and Basset times go up. For steady-state problems, the time steps are false time
forces are negligible because of the small fluid-to-particle density steps. False time steps used for simulation of flow field are brought
ratio. The first term on the right-hand side (RHS) of Eq. (10) is the drag in Table 1. Time step of 0.2 ms is used for calculation of particle
force due to the relative slip between the particle and the fluid. The trajectories in the numerical model.
drag force is, generally, the dominating force. According to [5], the It is difficult to understand the cyclone separation behavior
drag coefficient, CD, is given as without some information about the flow field. CFD is a very useful
tool to obtain details of the flow inside a cyclone. Fig. 3 shows the CFD
24
CD = for ReP b1 ð11Þ
ReP


24 1 2=3
CD = 1 + ReP for 1 bReP b400 ð12Þ
ReP 6

where, ReP is the particle Reynolds number defined as

djuj −uPj j
ReP = : ð13Þ
m

The particle equation of motion requires the instantaneous

turbulent fluid velocity values at particle locations. The mean liquid
velocity was evaluated by the use of the Reynolds stress transport
turbulence model (RSTM) and the fluctuation velocity components
were calculated from Eq. (6). The drag coefficient for spherical
particles is calculated by using the correlations developed by Morsi
and Alexander (1972). The ordinary differential equation (Eq. (10)) is
integrated along the trajectory of an individual particle. Collection
efficiency statistics are obtained by releasing a specified number of
mono dispersed particles at the inlet of the cyclone and by monitoring
the number escaping through the underflow. Collisions between
particles and the walls of the cyclone were assumed to be perfectly
elastic (coefficient of restitution is equal to 1) [13]. Also, particle–
particle collision is negligible.

3. Results

The simulations are performed for two cyclone types with different
geometries which have been studied experimentally by [17,15]. Fig. 1
shows the geometry and dimension of the studied cases. As seen in
Fig. 1 (a) the cyclone of case 1 has an upward outlet that also plays the
role of vortex finder. The particles size range is 0–2 mm of a material
whose density and average diameter are respectively 0.205 mm and
2550 kg/m3 [17]. The simulations are performed at inlet velocities of
22, 26, 30 and 34 m/s.
As indicated in Fig. 1 (b), the exhaust gas went through downward
exit in case 2. The particle used was glass beads of mean diameter of
30–40 μm and density of 2400 kg/m3 [15]. This cyclone is studied at
inlet velocities of 20, 25.3 and 28.32 m/s.
For analyzing the flow in the cyclones, the numbers of 186,298 and
258,976 hexahedral cells are generated for cases 1 and 2, respectively.
Fig. 2 shows the details of the computational grid for the cyclones. The
hexahedral computational grids are generated by dividing the whole
cyclone geometry into a number of blocks. A “velocity inlet” boundary
condition is used at the cyclone inlet. A fully developed boundary Fig. 5. Pressure drop in the cyclones at different inlet velocities. (a) Static pressure drop
condition is used at the outlet. Grid refinement tests are conducted in and (b) Dynamic pressure drop.
 Author's personal copy

354 A. Raoufi et al. / Powder Technology 191 (2009) 349–357

prediction of collection efficiency on the square cyclone in the case 1. empirical models have been obtained for the pressure drop in the
The CFD predicts the effect of the inlet velocity on the cyclone conventional cyclones [14,1,3,2]. Generally cyclone pressure drop is
collection efficiency with an acceptable deviation in comparison with proportional to the inlet velocity and can be written as
the experimental data [17]. Comparison between the CFD simulation
result and mathematical model of Wang et al. [17] shows that the ρg m2i
ΔP = α ð14Þ
numerical simulation has better agreement with experimental data 2
than the mathematical model. For square cyclone in case 2, Wang et al.
where ρg is gas density and vi is inlet velocity. In the Shepherd and
[17] reported the collection efficiency only for velocity inlet of
Lapple [14] model, α is obtained by assuming static pressure drop
28.32 m/s equal to 81%. At this inlet velocity, a value of 79% is
given as
obtained by numerical simulation.
The contours of static pressure are shown in Fig. 4 for cases 1 and 2 ab
at different inlet velocities. As shown and anticipated in the figure, by α = 16 ð15Þ
D2e
increasing the inlet velocity in the cyclone, the pressure drop is
increased. For both cases the low pressure zone in the centre of the where a and b are dimensions of the inlet section. Also De is vortex
cyclone is seen to be similar to the conventional cyclones. One of the finder diameter. Using dimensions of two square cyclones in this
most important parameters in the investigation of a cyclone study to obtain pressure drop of corresponding conventional ones,
performance is the pressure drop in the cyclone. The static and ΔPcase1 = 4.5v2i and ΔPcase2 = 3.13v2i are obtained. Comparison between
dynamic pressure drop between inlet and outlet is shown as a pressure drop of the square cyclones presented in Fig. 5 with
function of inlet velocity for two cases in Fig. 5 (a) and (b). Many the derived equations for the conventional cyclones shows that

Fig. 6. Contour of turbulence intensity for two cyclones at different inlet velocities. Case 1 (a) v = 22, (b) v = 30 and case 2 (c) v = 20, (d), v = 28.32, unit: m/s.
 Author's personal copy

A. Raoufi et al. / Powder Technology 191 (2009) 349–357 355

pressure drop in the square cyclones is less than comparable con-

ventional ones.
Fig. 6 shows the contour of turbulent intensity for different inlet
velocities. As shown in the figure the local turbulent intensity has a
maximum value at the outlet of the cyclones same as the conventional
ones. Also by increasing the mass flow rate in both cases, the turbulent
intensity increases.
The velocity field in a cyclone has three components: tangential,
axial and radial. Since the flow is strongly swirling, the tangential
velocity component is more important than the axial and radial
components. The radial component is the smallest one and can be
neglected for basic calculations [10]. The tangential velocity profile for
conventional cyclones can be seen as two regions with an outer region
of quasi-free vortex flow surrounding an inner region of quasi-forced
vortex flow. Wang et al. [17] mentioned that the tangential velocity
distribution inside a square cyclone basically agrees with the
rotational flow that consists of inner forced and outer free vortexes.
But, they didn't obtain the tangential velocity profile inside the square
cyclone and the effect of inlet velocity on it. In this study, the
tangential velocity is investigated in the different sections of square

Fig. 8. Tangential velocity vectors for the cyclone with downward outlet at different
inlet velocities: (a) v = 20 and (b) v = 28.32, unit: m/s.

Fig. 7. Tangential velocity vectors for the cyclone with upward outlet at different inlet
velocities: (a) v = 22 and (b) v = 30, unit: m/s. Fig. 9. Y-velocity distribution at Section 2-3 of case 2.
 Author's personal copy

356 A. Raoufi et al. / Powder Technology 191 (2009) 349–357

distribution in the inner region is rather similar at different sections.

In the outer region, due to the steep drop in velocity magnitude near
the wall region, the distribution is completely different. The tangential
velocity distribution only varies slightly with axial positions. As
seen in the figures, by increasing the distance from the bottom, the
tangential velocity magnitude decreases. The subfigures of Figs. 7 and
8 represented by (a) and (b) show the results of tangential velocity for
different inlet velocities. Varying the inlet velocity has a poor effect on
pattern of the tangential velocity. Only, by increasing of the inlet
velocity, the magnitude of the tangential velocity increases.
In addition to tangential velocity, the axial velocity is important for
the particle collection in the cyclone. Fig. 9 shows the y-velocity
distribution at the Section 2-3 for case 2. According to the
experimental data, the y-velocity in near wall region is downward
and has a greater amount in the right side. In the center of the cyclone,
upward y-velocity is observed in the numerical result that has a good
agreement with the experimental data in this region.
Velocity vectors are shown at the middle plane inside the cyclones
in Fig. 10. Flow enters the cyclone and moves toward the outlet that
upward in case 1 and downward in case 2. Generally, in the square
cyclones, flow pattern is similar to the conventional cyclones.
According to the available experimental data, the corner vortex
increases the collection efficiency [12,9]. Fig. 11 shows the velocity
vectors at a y-plane in the cyclones. Vortexes are obviously shown at
the corners inside the cyclone according to the experimental work
[15]. Velocity vectors at the y-plane located 30 and 120 mm from top
of the cyclone of case 2 are shown in the Fig. 12 (a) and (b),
respectively. In both planes the vectors have a configuration similar to
the experimental work. Vortexes presented at the corners of the
cyclone in the experimental data are clearly observed in these figures.
Fig. 10. Velocity vectors at the middle plane inside the cyclones. (a) case 1, v = 26 m/s and Also the center vortex in the Fig. 12 (b) is not absolutely at the center of
(b) case 2, v = 25.3 m/s. the cyclone and it is located a little nearer to the wall at the z = 0.12 m.
According to the experimental work, the pattern of particle velocity
vectors is same to the gas velocity vectors.
cyclone by CFD. Velocity profiles are numerically obtained along
the radial orientation at three axial stations in the cyclone. The 4. Conclusions
Sections 1-1, 1-2 and 1-3 are at 280, 320 and 350 mm from top of
cyclone 1 and the Sections 2-1, 2-2 and 2-3 are at 100, 110 and 120 mm A numerical approach is employed to study flow in square cyclones
from top of cyclone 2. Figs. 7 and 8 present the radial profiles of with different outlets. The presented numerical results are compared
tangential velocity at three different axial stations for cases 1 and 2, with the experimental data and good agreement is obtained. The flow
respectively. The tangential velocity profiles at these three sections field parameters at the different flow rates are numerically investi-
clearly show that the swirling flow inside the square cyclone consists gated. Pressure drop for two cyclones at different flow rates is
of two parts: an outer free vortex and an inner forced vortex in the obtained and compared. The radial profiles of tangential velocity at
centre same as conventional cyclones. The tangential velocity three axial stations in the cyclones are investigated. Effect of flow rate

Fig. 11. Velocity vectors inside the cyclones: (a) for case 1 at v = 22 m/s and (b) case 2 at v = 25.3 m/s.
 Author's personal copy

A. Raoufi et al. / Powder Technology 191 (2009) 349–357 357

3. Comparison between pressure drop of the square cyclones

obtained in this study with the empirical equations given in the
literature for the conventional cyclones, shows that pressure drop
in the square cyclones is less than the corresponding conventional
ones.
4. The tangential velocity profiles at different sections show that the
swirling flow inside the square cyclone consists of two parts: an
outer free vortex and an inner forced vortex in the centre same as
conventional cyclones.
5. In the outer region, due to the steep drop in velocity magnitude
near the wall region, the distribution is completely different. The
tangential velocity distribution varies slightly with axial positions.
6. Simulation results show that vortexes are generated at the corners
of the square cyclone.

References

[1] J. Casal, J.M. Martinez, A better way to calculate cyclone pressure drop, Chemical
Engineer (1983) 90–99.
[2] A.K. Coker, Understand cyclone design, Chemical Engineering Progress 28 (1993)
51–55.
[3] Dirgo, J., 1988, Relationships between cyclone dimensions and performance,
Doctoral Thesis, Havarad University, USA.
[4] W.D. Griffiths, F. Boysan, Computational fluid dynamics (CFD) and empirical
modeling of the performance of a number of cyclone samplers, Journal of Aerosol
Science 27 (1996) 281–304.
[5] W.C. Hinds, Aerosol Technology, Properties Behavior and Measurement of
Airborne Particles, John Wiley and Sons, New York, 1982.
[6] L. Junfu, Z. Jiansheng, Y. Guangxi, L. Qing, L. Xiaoma, L. Jiling, The progress of the
water cooled separator CFB boiler in China, Proceedings of the 15th International
Conference on Fluidized Bed Combustion, Savannah, Georgia Paper No. FBC99-
0038, 1999.
[7] L. Junfu, Z. Jiansheng, Z. Hai, L. Qing, Y. Guangxi, Performance evaluation of a 220t/h
CFB boiler with water-cooled square cyclones, Fuel Processing Technology 88
(2007) 129–135.
[8] B.E. Launder, G.J. Reece, W. Rodi, Progress in the development of a Reynolds stress
turbulent closure, Journal of Fluid Mechanics 68 (1975) 537–538.
[9] Liang, S., 1997, Gas–solid separation mechanism at high temperature and it's
experimental study, Ph.D. Thesis, Zhejiang University, China.Morsi, S.A., A.J.
Alexander, 1972, An investigation of particle trajectories in two phase flow
systems, J. Fluid Mech., 55,193–208.
[10] W. Peng, A.C. Hoffmann, P.J.A.J. Boot, A. Udding, H.W.A. Dries, A. Ekker, J. Kater,
Flow pattern in reverse-flow centrifugal separators, Powder Technology 127
(2002) 212–222.
[11] G. Qiang, Y. Hairui, L. Junfu, L. Qing, W. Yu, Y. Guangxi, K. Makino, S. Watababe, M.
Nakamura, Experimental Investigation on the Performance and Characteristic
Parameter of the Square Separator, the 8th China Japan Fluidization, Gifu, Japan,
2003.
[12] K. Qiu, J. Yan, X. Li, et al., Experimental study and structure optimization of a
uniflow square-shaped cyclone separator, Journal of Engineering for Thermal
Energy and Power 14 (3) (1999) 193–199.
[13] A. Raoufi, M. Shams, M. Farzaneh, R. Ebrahimi, Numerical simulation and
optimization of fluid flow in cyclone vortex finder, Chemical Engineering and
Fig. 12. Velocity vectors inside the cyclones: for case 2 at the y-plane located. (a) 30 mm Processing 47 (2008) 128–137.
from top of the cyclone at v = 25.3 m/s and (b) 120 mm from top of the cyclone [14] C.B. Shepherd, C.E. Lapple, Air Pollution Control: A Design Approach, Second ed.,
at v = 28.32 m/s. Woveland Press Inc., 1939, pp. 127–139.
[15] Y. Su, Y. Mao, Experimental study on the gas–solid suspension flow in a square
cyclone separator, Chemical Engineering Journal 121 (2006) 51–58.
on the shape of velocity profile is studied. Based on the presented [16] Y. Su, The turbulent characteristics of the gas–solid suspension in a square cyclone
separator, Chemical Engineering Science 61 (2006) 1395–1400.
results, the following conclusions may be drawn:
[17] S. Wang, M. Fang, Z. Luo, X. Li, M. Ni, K. Cen, Instantaneous separation model of a
square cyclone, Powder Technology 102 (1999) 65–70.
1. Obtained results show that CFD is a good tool for the study of the
flow in square cyclones and can be used for optimizing it.
2. The low pressure zone in the centre of the square cyclones is similar
to the conventional cyclones.