Original Article

Proc IMechE Part C:

J Mechanical Engineering Science
Effects of helical fins with semi-circular 0(0) 1–12
! IMechE 2020

in cross-section on the performance Article reuse guidelines:

sagepub.com/journals-permissions

characteristics of novel finned-cyclone DOI: 10.1177/0954406220916485

journals.sagepub.com/home/pic

separators

Mahesh Dasar and Ranjit S Patil

Abstract
Separation of particles (solid) from hot gases were carried out with the help of cyclone separators in most of the power
generation industries (fluidized bed boilers) and process industries. Present research focuses on improving the perform-
ances of such devices by modifying the geometry. Conventional 2D2D Lapple type cyclone separator geometry was
modified by attaching semi-circular cross-section helical fins along the height of barrel wall to make it as finned-cyclone
separators. Semi-circular helical fins attached on the barrel wall varies (six-sets) with size (5 mm, 7.5 mm, and 10 mm) and
pitch (30 mm and 50 mm). 3D computational fluid dynamics simulations were carried out using ANSYS 15.0 tool and
validation were performed with experimental data available in the literature for conventional 2D2D Lapple type cyclone
separator before proceeding towards simulations of semi-circular finned-cyclone separators. Improved performance was
noticed from all selected six cyclone separators with semi-circular fins (cswf) when compared to conventional cyclone
separators while studying the fluid mechanics characteristics, which directly influences on collection-efficiency of cyclone
separators. Among selected cyclone separators, the cswf with size of fin 7.5 mm and fin-pitch 50 mm was noted to
be performing improved collection efficiency for the particles size <3 mm. Around 5%–11% improved collection effi-
ciency when compared to conventional cyclone separators were observed in the study by fixing the semi-circular fins
along the height of barrel wall without disturbing the main role of cyclone separator, which is separating the very fine
particles, which would otherwise causes pollution.

Keywords
Computational fluid-dynamics, collection-efficiency, semi-circular helical fin

Date received: 7 July 2019; accepted: 7 March 2020

Background these parameters prompts to increase in pressure drop

Separation of particles (solid) from hot gases were and decreased separation efficiency. Zhang et al.11
carried out with the help of cyclone separators in have focused on new multi-cyclone dual-stage separ-
most of the power generation industries (fluidized ators (DSCS) to boost flow distribution. DSCS gives
bed boilers) and process-industries. This is a signifi- greater collection efficiency for a given inlet speed
cant function for the assortment of fine particles so as than other selected cyclone separators. Mariani
to diminish environmental pollution. Fluid mechanics et al.12 recorded improvement in separation efficiency
attributes such as axial-velocity, tangential-velocity by adjusting the angle of conicity and the duration of
(TV), pressure-drop, and so on recorded in detail the vortex finder. Wasilewski and Duda13 recorded 57
through the quantitative examination, which impacts separate geometric configurations of cyclone separ-
the cyclone separator collection efficiency.1–5 ators. Few researchers14,15 discovered the effects of
Literatures’ claim that simulations on two phase
flow to investigate flow behavior on the separation
efficiency.6–8 Avci and Karagoz9 build up a numerical Department of Mechanical Engineering, BITS Pilani, India
model for foreseeing the separating efficiency. Further
Corresponding author:
Zhang et al.10 recorded results on the output of the
Ranjit S Patil, Department of Mechanical Engineering, BITS Pilani, K K
hexahedral cyclone separator, which varied param- Birla Goa Campus, NH-17(B), Air Port Road, Zuarinagar, Goa 403726,
eters, for example, external body round corner diam- India.
eter and twist angle. It was found that the ascent in Email: [email protected]
2 Proc IMechE Part C: J Mechanical Engineering Science 0(0)

incorporating the fins on the heat transfer features. by modifying inside surface of the barrel of the trad-
Nag and Gupta14 detailed different parameters, itional cyclone separator with semi-circular fins.
which impacts the heat transfer in case of suspended
fins in the drum. Bodo15 mentioned external fins
Numerical method
mounted in the gas-outlet area to increase the heat
transfer rate in cyclone separators. Two phase flow simulations were performed using
As observed, the completely modifying drum of Ansys 15 research code in case of semi-circular
device by fixing semi-circular helical fins so as to finned-cyclone separators. Several researchers1,2,4,5,16,17
enhance the performance of cyclone separators has have claimed that, for cyclone separator simulation,
not yet been reported. the Reynolds stress model (RSM) is most appropriate
The present investigation in this manner centers turbulence model over k-p turbulence model. Recent
around computational research on cyclone separators studies have suggested RSM to model turbulence

Figure 1. (a) 2D2D Lapple cyclone separator with dimensions mentioned in mm. (b) Helical semi-circular fin-geometry. (c) Helical
cswf with size of fin as 5 mm, 7.5 mm, and 10 mm (left side to right) and pitch-size 30 mm. (d) The cswf with size of fin as 5 mm, 7.5 mm,
and 10 mm (left side to right) and pitch-size 50 mm.
Dasar and Patil 3

precisely without error since the simulation of the cyc- particles’ flow simulation and separation efficiency
lone separator deals with heavy swirling.18 carried out using Eulerian-Lagrangian approach.23
Many industries use RSM for all the turbulent flow Equation of Particle motion given as
simulations.19 RSM is the strongest turbulence model 
for complicated 3D models with such as heavy curva- dupi  gi  p  
ture of the streamline, spinning motion of rotational ¼ FD ui  upi þ
dt p ð4Þ
flow.20
3D conservation equation as defined by Safikhani
et al.21 and Hesham and El-Batsh22 for the constant
dxpi
isothermal and incompressible flow ¼ upi ð5Þ
dt

@u i where FD ui  upi is the drag force per unit particle
¼0 ð1Þ
@xj mass
  
@ ðu i Þ @P @ @u i @u j @ij FD ¼
18 CD Rep
ð6Þ
uj ¼ þ  þ þ ð2Þ p d2p 24
@xj @xi @xj @xj @xi @xj

where, p dp
u  up

u i and u j Mean velocity in i and j directions, Rep ¼ ð7Þ


respectively.
xj is the position,

P is the mean pressure, Table 1. Geometrical parameters of cyclone separator.

 is the constant gas density,
D H/D h/D De/D B/D S/D a/D b/D
ij is the Reynolds stress tensor given by
ij ¼ u0i u0j . 200 mm 4.0 2.0 0.50 0.25 0.625 0.25 0.50
The RSM transport equation21,22 can be written as

@  
uk uk u0i u0j ¼ Dij þ Pij þ ij  "ij ð3Þ
@xk

Dij is diffusive transport term

!
@ t @u0i u0j
Dij ¼  ð3aÞ
@xk k @xk

Pij is generation of stress term

 
@uj @ui
Pij ¼  u0i u0k þ u0j u0k ð3bÞ
@xk @xk

ij is pressure-strain correlation term

 0 
@ui @u0j
ij ¼ P þ ð3cÞ
@xj @xi

"ij is dissipative term

@u0i @u0j
"ij ¼ 2 ð3dÞ
@xk @xk

Discrete phase model

Figure 2. (a) Cut-section of mesh generated on cswf. (b)
Two-phase flow model is based on discrete phase Cut-section of mesh generated on helical semi-circular finned-
mechanism that is Eulerian-Lagrangian method. cyclone separator. (c) Cut-section of mesh generated on helical
Particle tracking method was carried-out with isolated semi-circular finned-cyclone separator.
4 Proc IMechE Part C: J Mechanical Engineering Science 0(0)

Figure 3. Mesh independence study of conventional cyclone separator.

Figure 4. Validating the computational data with experimental data of Wang et al.5

Drag coefficient CD is a function of relative

Modeling of semi-circular finned-cyclone separators
Reynolds numbers, which is in the form of
Lapple24 proposed a 2D2D type cyclone separators,
CD ¼ 24=Rep For 0 5 Rep 41 which are highly accomplishable and desirable among
  all other types of cyclone separators available.25
CD ¼ 24=Rep 1 þ 0:15Re0:687
p For 0 5 Rep 41000
Figure 1(a) shows the 2D2D Lapple type normal cyc-
ð8Þ lone separator and dimensions used as in Table 1 in
the present research. As observed from Figure 1(b)
Dasar and Patil 5

Table 2. Mesh used and Yþ value measured in the computational study.

Chosen
Pitch Fin size number Identification name Yþ value
Fin shape size (mm) (mm) of nodes henceforth measured

Semi-circular fin 30 5 591 K Set-1 33.86

7.5 606 K Set-2 32.92
10 612 K Set-3 32.06
50 5 566 K Set-4 25.82
7.5 575 K Set-5 23.62
10 576 K Set-6 30.37

Figure 5. (a) Axial-velocity measured for cswf with fin-size 5 mm and pitch-size 30 mm compared with conventional cyclone sep-
arator at three different locations. (b) Axial-velocity measured for cswf with fin-size 7.5 mm and pitch-size 30 mm compared with
conventional cyclone separator at three different locations. (c) Axial-velocity measured for cswf with fin-size 10 mm and pitch-size
30 mm compared with conventional cyclone separator at three different locations. (d) Axial-velocity measured for cswf separator with
fin-size 5 mm and pitch-size 50 mm compared with conventional cyclone separator at three different locations. (e) Axial-velocity
measured for cswf with fin-size 7.5 mm and pitch-size 50 mm compared with conventional cyclone separator at three different
locations. (f) Axial-velocity measured cswf with fin-size 10 mm and pitch-size 50 mm compared with conventional cyclone separator at
three different locations. (g) Axial-velocity comparison between all six sets of cswf.
6 Proc IMechE Part C: J Mechanical Engineering Science 0(0)

the solid base dark in color on which the hollow semi- scalable-wall function using the standard constants.
circular fin (light colored) is fitted on inner surface of Inlet boundary condition was chosen as velocity-
barrel. Helical cyclone separators with semi-circular inlet with constant velocity-magnitude of 20 m/s
fins (cswf) with fin size of 5 mm, 7.5 mm, and 10 mm through which solid-gas mixture enters (Figure 1(a))
with pitch 30 mm and 50 mm are as shown on Figure cyclone separator and turbulence intensity 3% with
1(c) and (d). hydraulic diameter as 0.06 m utilized.5,25 Gas-outlet
(Figure 1(a)) was chosen as pressure-outlet boundary
condition turbulent intensity 3% and 0.1 m hydraulic
Mesh generation diameter. Solid-outlet and cyclone body (Figure 1(a))
Hexahedral unstructured mesh was used at the center was given as wall boundary condition with no-slip
while three-layer prismatic cells were used near wall. conditions. With 0.6 explicit-relaxation factor for
Tetrahedral meshes were used to link hexahedral the pressure, 0.08 for momentum and 40 flow-courant
and prismatic meshes as shown in Figure 2(a) to (c). number the solution was controlled.
For six sets of cswf, mesh quality above 0.2 was Many researchers conducted steady5,26,27 and
reached and nearly 80% mesh quality ranged from unsteady state5 simulations by varying outflow
0.9 to 1. length and observed slight variations in the flow-
field behavior with unsteady state solver.
Computational parameters
Computation research work was carried out using
Grid dependency and validation
RSM turbulence model with convergence criteria of Figure 3 shows the grid independency study, which
105 and near-wall treatment were supervised with was carried out on the conventional cyclone

Figure 5. Continued.
Dasar and Patil 7

Figure 6. Axial pressure drop (P).

Figure 7. (a) TV measured for cswf with 5 mm and 30 mm fin and pitch size, respectively. (b) TV measured for cswf with 7.5 mm and
30 mm fin and pitch size, respectively. (c) TV measured for cswf with 10 mm and 30 mm fin and pitch size, respectively. (d) TV
measured for cswf with 5 mm and 50 mm fin and pitch size, respectively. (e) TV measured for cswf with 7.5 mm and 50 mm fin and
pitch size, respectively. (f) TV measured for cswf with 10 mm and 50 mm fin and pitch size, respectively. (g) TV comparisons between
all six sets of cswf.
8 Proc IMechE Part C: J Mechanical Engineering Science 0(0)

separator. As presented in Figure 4, results obtained separator and comparison was performed between
using 617 K nodes were validated using experimental cswf and normal cyclone separator. Axial-velocity
data of Wang et al.5 profiles obtained in the present study are qualitatively
cognate to Wang et al.5,28–32 for both fined and con-
Grid dependency study of cyclone separators with ventional cyclone separators.
As noted from Figure 5(a) to (f), the axial-velocity
semicircular fins
measured for cswf at a location 700 mm from the
Mesh independency study was carried out for all six- bottom of cswf and near the wall (around 20 mm
sets of semi-circular fin-based cyclone separators with from the wall) the axial-velocity is noted to be in nega-
the same procedure used for conventional cyclone tive value, that signifies that, this velocity in the down
separator. Table 2 provides further details. direction, which helps particles to move in downward
direction along the wall towards the collection point.
The velocity was noted in the range of 6–7 m/s mea-
Results and discussion sured at 80 mm from the center of the cyclone, which
was slightly higher when compared to conventional
Axial-velocity
cyclone separator. The axial-velocity decreases due to
The axial-velocity has a strong influence on the friction until the end of the barrel section and increases
separation efficiency. Figure 5(a) to (f) show the in the conical section due to sloping edges.
axial-velocity profiles measured at the three different Also, as shown in Figure 5(g) the axial-velocity
locations along the height of cswf and normal cyclone measured for cswf at a location 500 mm from the
separator. The axial-velocity was measured at bottom of all six cswf were compared to check the
the same three locations for conventional cyclone effect of fin-size (5 mm, 7.5 mm, and 10 mm) and

Figure 7. Continued.
Dasar and Patil 9

fin-pitch (30 mm and 50 mm). It was noted that the higher when compared with remaining five-sets of
fin-size and fin-pitch variations on the axial-velocity semi-circular finned-cyclone separators, which further
profile was not significant. However, axial-velocity in experienced as separation efficiency for the same is
cswf with pitch size 50 mm were observed to slightly higher compared to other five-sets of semi-circular
higher when compared with cswf with pitch size finned-cyclone separators.
30 mm near the all (dense solid particle area), which
further experienced as higher separation efficiency for
Separation efficiency
cswf with pitch size 50 mm when compared to cswf
with 30 mm pitch sizes. Figure 8(a) and (b) shows the comparative separation
efficiency cswf with pitch size of 30 mm and 50 mm
with normal cyclone separator.
Axial pressure-drop As observed in Figure 5(a) to (f), due to
Figure 6 shows the comparisons of axial pressure improved TV in case of cswf, it gives better separation
drop (P) measured from all six types of cswf and of particles hence efficiency than normal cyclone
normal cyclone separator. Difference of total pressure separator. Which results in forcing more particles
measured at start (208 mm) and end (493 mm) of (Figure 9(c) and (d)) towards wall in case cswf than
helical fin is noted as axial pressure drop. The same normal cyclone separator (Figure 9(a) and (b)), add-
location was used to find the pressure difference itionally axial velocity forces these particles towards
for conventional cyclone separator. Due to helical bottom outlet.
fin, increase in axial pressure drop is observed Also due to helical pathway created by helical fin
in case of cswf when compared to conventional cyc- through which particles travel along the wall towards
lone separator. This deficit can however be compen- solid outlet, additionally which creates friction, which
sated for by increasing the separation efficiency helps more particles to separate from gas/air.
(section ‘‘Separation efficiency’’). It is also observed
that P was decreased while the separation efficiency
(section ‘‘Separation efficiency’’) was increased with
increase in pitch of the fin.

Tangential-velocity
The mixture flow (air þ solid) inside the device will
be controlled by TV, which creates a centrifugal
force that defines the particle separation in effect.33
Figure 7(a) to (f) shows the TV profiles measured at
the three different locations (300 mm, 500 mm, and
700 mm) along the height of device. With intention
to see the effect of helical semi-circular fins attached,
the TV was measured at the same three locations for
conventional cyclone separator and comparison was
performed between cswf and normal separator. The
patterns in TV profiles were observed qualitatively the
same as reported in the literature.34–40
The TV for semi-circular finned-cyclone separator
was observed between the region 40 mm to 80 mm
from core of cyclone separators is little higher when
compared with conventional cyclone separator at the
measured location 700 mm (from the bottom outlet),
which is the higher particles concentration area and
higher TV at that area helps to improved separation
efficiency for semi-circular finned-cyclone separators
when compared to conventional cyclone separator.
As noted from Figure 7(a) to (f) TV profiles, the
TV decreasing as the flow moves down along the
height of barrel wall and gradually becomes 0 near
the wall due friction.
Also, as shown in Figure 7(g), it was noted that the Figure 8. (a) Separation efficiency comparison between cswf
TV was not much affected due to variations in sizes of with 30 mm fin-pitch and conventional cyclone separator. (b)
fin or pitch. However, TV for cswf with fin-size Separation efficiency comparison between cswf with 50 mm fin-
7.5 mm and pitch size 50 mm was noted to slightly pitch and conventional cyclone separator.
10 Proc IMechE Part C: J Mechanical Engineering Science 0(0)

Figure 9. (a) and (b) Front view and top view of particle tracking of conventional cyclone separator at barrel wall (left) and cone
(right) location. (c) and (d) Front view and top view of particle tracking of semi-circular finned-cyclone separator at barrel wall (left)
and cone (right) location.

Figure 10. Flow reversal points for six sets of cswf and unfinned separator.

Also, as observed from Figure 10, central air jet

reversal takes place close to the bottom of cswf with
Conclusions
size of fin as 7.5 mm with fin pitch 50 mm, which . Improvement in the collection efficiency was
results in increased particles’ separation. observed for fin-based cyclone separator compared
Dasar and Patil 11

to conventional cyclone separator because of 8. Zhou LX and Soo SL. Gas-solid flow and collection of
improvement in the TV. solids in a cyclone separator. Powder Technol 1990; 63:
. Separation efficiency of semi-circular finned-cyclone 45–53.
separator of fin-size 7.5 mm with pitch-size 50 mm 9. Avci A and Karagoz I. Effects of flow and geo-
metrical parameters on the collection efficiency in
was observed to be highest among the selected other
cyclone separators. J Aerosol Sci 2003; 34: 937–955.
five sets of semi-circular finned-cyclone-separators.
10. Zhang T, Liu C, Guo K, et al. Analysis of flow field in
Also, it was giving around 11% more efficiency than optimal cyclone separators with hexagonal structure
conventional cyclone separator. using mathematical models and computational fluid
. With increase in fin size the axial pressure-drop has dynamics simulation. Ind Eng Chem Res 2016; 55:
increased, however this increment in the axial pres- 351365.
sure drop was compensated by increment noted in 11. Zhang T, Guo K, Liu C, et al. Experimental and
the collection efficiency. Considerable deviation in numerical investigations of a dual-stage cyclone separ-
both axial and TV and separating efficiency was ator. Chem Eng Technol 2018; 41: 606–617.
observed by changing the fins sizes. 12. Mariani F, Risi F and Grimaldi C N. Separation effi-
. Helical semi-circular fin creates helical pathway ciency and heat exchange optimization in a cyclone. Sep
Purif Technol 2017; 179: 393–402.
through which particles travel along the wall towards
13. Wasilewski M and Duda J. Multicriteria opti-
solid outlet, additionally this pathway creates the
misation of first-stage cyclones in the clinker
friction with the particles, which prevents the par- burning system by means of numerical modelling and
ticles’ movement towards the central air flow experimental research. Powder Technol 2016; 289:
moving towards upper gas outlet, which was resulted 143–158.
as the improvement in the collection efficiency. 14. Nag PK and Gupta AVSSKS. Fin heat transfer studies
. Comparatively large numbers of very small par- in a cyclone separator of a circulating fluidized bed.
ticles below 3 mm were separated using proposed Heat Transfer Eng 1999; 20: 28–34.
novel fin based cyclone separators than normal 15. Bodo K. Cyclone separator vortex finder with exterior
non-finned cyclone separator. heat fins. Patent 780525, 304031, Canadian Patent File,
1977.
16. Elsayed K and Lacor C. The effect of cyclone inlet
Declaration of Conflicting Interests dimensions on the flow pattern and performance.
The author(s) declared no potential conflicts of interest with Appl Math Model 2011; 35: 1952–1968.
respect to the research, authorship, and/or publication of 17. Hanjalic K and Launder BE. A Reynolds stress model
this article. of turbulence and its application to thin shear flows.
J Fluid Mech 1972; 52: 609–638.
18. Parekh J and Rzehak R. Euler–Euler multiphase
Funding CFD-simulation with full Reynolds stress model and
The author(s) received no financial support for the research, anisotropic bubble-induced turbulence. Int J
authorship, and/or publication of this article. Multiphase Flow 2018; 99: 231–245.
19. Slack MD, Prasad RO, Bakker A, et al. Advances in
cyclone modelling using unstructured grids. Trans
ORCID iD IChemE 2000; 78: 1098–1104.
Ranjit S Patil https://orcid.org/0000-0003-4935-593X 20. Modeling Turbulent Flow. Introductory fluent training,
www.southampton.ac.uk/nwb/lectures/. . ./Turbulence_
References Notes_Fluent-v6.3.06.pdf (2006, accessed 27 July 2017).
1. Azadi M, Azadi M and Ali M. A CFD study of the effect 21. Safikhani H, Akhavan-Behabadi MA, Shams M, et al.
of cyclone size on its performance parameters. J Hazard Numerical simulation of flow field in three types of
Mater 2010; 182: 835–841. standard cyclone separators. Adv Powder Technol
2. Hoekstra AJ, Derksen JJ and Van Den Akker HEA. An 2010; 21: 435–442.
experimental and numerical study of turbulent swirling 22. Hesham M and El-Batsh. Improving cyclone perform-
flow in gas cyclones. Chem Eng Sci 1999; 54: 2055–2065. ance by proper selection of the exit pipe. Appl Math
3. Shi L and Bayless DJ. Comparison of boundary condi- Model 2013; 37: 5286–5303.
tions for predicting the collection efficiency of cyclones. 23. Zhao B, Su Y and Zhang J. Simulation of gas flow
Powder Technol 2007; 173: 29–37. pattern and separation efficiency in cyclone with con-
4. Wan G, Sun G, Xue X, et al. Solids concentration simu- ventional single and spiral double inlet configuration.
lation of different size particles in a cyclone separator. Chem Eng Res Des 2006; 84: 1158–1165.
Powder Technol 2008; 183: 94–104. 24. Lapple CE. Processes use many collector types. Chem
5. Wang B, Xu DL, Chu KW, et al. Numerical study of Eng 1951; 58: 144–151.
gas–solid flow in a cyclone separator. Appl Math Model 25. Mothilal T, Pitchandi K, Velukumar V, et al. CFD and
2006; 30: 1326–1342. statistical approach for optimization of operating par-
6. Dietz PW. Collection efficiency of cyclone separator. ameters in a tangential cyclone heat exchanger. J Appl
AIchE J 1981; 27: 882–892. Fluid Mech 2018; 11: 459–466.
7. Trefz M and Muschelknautz E. Extended cyclone theory 26. Udaya Bhaskar K, Rama Murthy Y, Ravi Raju M,
for gas flows with high solid concentrations. Chem Eng et al. CFD simulation and experimental validation stu-
Technol 1993; 16: 153–160. dies on hydrocyclone. Miner Eng 2007; 20: 60–71.
12 Proc IMechE Part C: J Mechanical Engineering Science 0(0)

27. Rama Murthy Y and Udaya Bhaskar K. Parametric 39. Zhou H, Hu Z, Zang Q, et al. Numerical study on gas-
CFD studies on hydrocyclone. Powder Technol 2012; solid flow characteristics of ultra-light particles in a cyc-
230: 36–47. lone separator. Powder Technol 2019; 3441: 784–796.
28. M’Bouana NLP, Lia D, Yangjia T, et al. Numerical 40. Zhang J, Zha Z, Che P, et al. Influences of inlet height
simulation of gas and particle flow in cyclone separ- and velocity on main performances in the cyclone sep-
ators. AIP Conf Proc 2013; 1547: 545–554. arator. Particul Sci Technol 2018; 37: 669–676.
29. Farzad P, Seyyed HH, Goodarz A, et al. Impacts of the
vortex finder eccentricity on the flow pattern and per-
formance of a gas cyclone. Sep Purif Technol 2017; 187:
1–13. Appendix
30. Dzmitry M, Anders GA and Tord SL. Effects of the
inlet angle on the flow pattern and pressure drop of a
Notation
cyclone with helical-roof inlet. Chem Eng Res Des 2015; a cyclone inlet width (mm)
102: 307–321. b cyclone inlet height (mm)
31. Sakura GB and Andrew YTL. CFD simulation of cyc- B cyclone solid-outlet diameter (mm)
lone separators to reduce air pollution. Powder Technol CD drag coefficient (–)
2015; 286: 488–506.
D cyclone diameter (mm)
32. Utikar R, Darmawan N, Tade M, et al. Computational
De cyclone gas-outlet diameter (mm)
fluid dynamics. Australia: Griffith University, 2010,
p.420. Dij  diffusive transport term
33. Cortes C and Gil A. Modeling the gas and particle flow FD ui  upi drag force per unit particle mass
inside cyclone separators. J Prog Energy Combust Sci gi acceleration of gravity (m/s2)
2007; 33: 409–452. h cyclone barrel wall height (mm)
34. Faqi Z, Guogang S, Yuming Z, et al. Experimental and H cyclone height (mm)
CFD study on the effects of surface roughness on cyc- P mean pressure (Pa)
lone Performance. Sep Purif Technol 2018; 193: Pij generation of stress term
175–183. Rep particle relative Reynolds number (–)
35. Kumar V and Jha K. Numerical investigations of the S cyclone vertex finder height (mm)
cone-shaped vortex finders on the performance of cyc-
u i mean velocity in i-direction (m/s)
lone separators. J Mech Sci Technol 2018; 32:
u j mean velocity in j-direction (m/s)
5293–5303.
36. Bunyawanichakul P, Kirkpatrick MP, Sargison JE, xj position (m)
et al. Numerical and experimental studies of the flow "ij dissipative term
field in a cyclone dryer. J Fluids Eng 2006; 128:  gas density (kg/m3)
1240–1250.
p particle density (kg/m3)
37. Akiyama O and Kato C. Numerical investigations of
unsteady flows and particle behavior in a cyclone sep-
ij pressure-strain correlation term
arator. J Fluids Eng 2017; 139: 913021–9130211.  viscosity of the gas (Pa s)
38. Jafari PH, Misiulia D, Hellstrom JGI, et al. Modeling t the eddy viscosity (Pa s)
of particle-laden cold flow in a cyclone gasifier. J Fluids ij Reynolds stresses (Pa)
Eng 2019; 141: 0213021–02130213.