Acta hydrotechnica 37/67 (2024), Ljubljana                                                                                       Open Access Journal 
ISSN 1581-0267                       Odprtodostopna revija 
137 
UDK/UDC: 544.272:551.51(078.7) Prejeto/Received: 26.09.2024 
Izvirni znanstveni članek – Original scientific paper Sprejeto/Accepted: 06.01.2025 
DOI: 10.15292/acta.hydro.2024.08 Objavljeno na spletu/Published online: 03.03.2025 
NUMERICAL STUDY OF HIGHLY EFFICIENT CENTRIFUGAL 
CYCLONES  
NUMERIČNA ŠTUDIJA VISOKOUČINKOVITIH CENTRIFUGALNIH 
CIKLONOV 
Murodil Madaliev
1
, Zokhidjon Abdulkhaev
1,*
, Yunusali Khusanov
1
, Saxiba Mirzababayeva
1
, 
Zebuniso Abobakirova
1
 
1
 Fergana Polytechnic Institute, Fergana, Uzbekistan 
Abstract 
Centrifugal cyclones have been developed for centuries to treat process streams, but their efficiency in 
removing fine dust remains below 80%. The widespread use of cyclones in various industries is the result of 
their simplicity in design and their reliable operation. However, the process occurring inside a cyclone is a 
complex scientific problem that still remains unsolved within the framework of aerohydromechanics, as 
evidenced by the variety of cyclone designs. The current level of cyclone cleaning's efficiency for process 
streams does not meet sanitary standards and significantly affects the level of environmental pollution. This 
paper compares various configurations of centrifugal cyclones, including cyclones without screws, with a 
screw of a uniform pitch, and with a screw of a variable pitch to regulate the twist of the flow. Numerical 
simulations were performed using the Comsol Multiphysics 5.6 software package using the SST turbulence 
model. The obtained numerical data show that the efficiency of a cyclone with a variable screw pitch is 
significantly higher than that of cyclones without screws and with a screw of a uniform pitch. 
Keywords: cyclone, mathematical modeling, turbulence models, Reynolds-averaged Navier–Stokes 
equations. 
Izvleček 
Centrifugalni cikloni se že stoletja razvijajo za obdelavo procesnih tokov, vendar njihova učinkovitost pri 
odstranjevanju drobnega prahu ostaja pod 80 %. Široka uporaba ciklonov v različnih panogah je posledica 
njihove preproste zasnove in zanesljivega delovanja. Proces znotraj ciklonov je zapleten znanstveni problem 
in v okviru aerohidromehanike ostaja nerešen, kar dokazuje raznolikost modelov. Trenutna raven učinkovitosti 
ciklonskega čiščenja procesnih tokov ne ustreza sanitarnim standardom in bistveno vpliva na onesnaženost 
okolja. Članek primerja različne konfiguracije centrifugalnih ciklonov, vključno s cikloni brez vijakov, z 
vijakom enakomernega koraka in z vijakom spremenljivega koraka, ki uravnava zasuk toka. Numerične 
simulacije so bile izvedene z uporabo programskega paketa Comsol Multiphysics 5.6 z uporabo modela 
 
*
 Stik / Correspondence: zokhidjon@ferpi.uz   
© Madaliev M. et al.; This is an open-access article distributed under the terms of the Creative Commons Attribution – NonCommercial 
– ShareAlike 4.0 Licence. 
© Madaliev M. et al.; Vsebina tega članka se sme uporabljati v skladu s pogoji licence Creative Commons Priznanje avtorstva – 
Nekomercialno – Deljenje pod enakimi pogoji 4.0.  

Madaliev et al.: Numerical study of highly efficient centrifugal cyclones –Numerična študija visokoučinkovitih 
centrifugalnih ciklonov 
Acta hydrotechnica 37/67 (2024), 137–151, Ljubljana 
138 
turbulence SST. Dobljeni numerični podatki kažejo, da je izkoristek ciklona z vijakom spremenljivega koraka 
bistveno večji kot pri ciklonih brez vijakov in cikloni z vijakom enakomernega koraka. 
Ključne besede: ciklon, matematično modeliranje, turbulenčni modeli, Reynoldsovo povprečene Navier-
Stokesove enačbe. 
 
1. Introduction 
Gas cyclones are devices that use the vortex effect 
to separate the dispersed phase (dust) from the 
continuous phase (gas). These devices are widely 
used for collecting solid particles from gas streams 
and have been used in various industries for many 
years. Cyclones are popular due to their simple 
design, low operating costs, lack of moving parts, 
high reliability, and ability to operate in harsh 
environments. When a gas cyclone is on, vortex 
centrifugal forces act on every point inside the 
device, which leads to the separation of the 
dispersed phase, being heavier, and its movement 
towards the walls of the cyclone. These particles are 
then driven downwards by axial velocity towards 
the collector, where they are gathered. The purified 
air is then vortexed upward again and leaves the 
cyclone through the outlet pipe. Although the 
operating principle behind a gas cyclone seems very 
simple, the physics governing the liquid flow inside 
the device makes for a complex and not yet fully 
understood area. 
The wide variety of applications of cyclone 
separators has led many researchers to study the 
effect of cyclone size on performance and flow 
field. 
Studies conducted by Hoffmann et al. (2001), Xiang 
and Lee (2005), Zhu and Lee (1999), Avci and 
Karagoz (2003), Chuah et al. (2006), and Surmen et 
al. (2011) provide valuable data on the influence of 
various parameters on the operation of gas cyclones. 
In particular, research data indicate the importance 
of the length and height of the cyclone, as well as 
the geometry of the conical tip, on its effectiveness. 
Although the works of Hoffman and Xiang present 
different results on the effect of cyclone length, the 
studies of Zhu et al. highlight the importance of 
cyclone pitch on separation efficiency. These 
inconsistencies may be explained by differences in 
methodology and experimental conditions. The 
mathematical model developed by Avsiev et al. 
provides a new tool for predicting the efficiency of 
cyclones and agrees well with experimental data. In 
addition, the work of Chuah et al. confirms that 
CFD computer modelling can accurately reproduce 
the performance of gas cyclones, allowing for more 
accurate predictions. The research of Surmen et al. 
(2011), based on a theoretical approach, is also 
important because it predicts the cutoff size of a 
cyclone based on its geometry and fluid properties. 
Overall, these works make significant contributions 
to the understanding of the physics and optimization 
of gas cyclones, which has important implications 
for their application in various industrial processes. 
Studies by Lee et al. (2006) and Hsu et al. (2014) 
add to our knowledge of the influence of various 
parameters on the performance of gas cyclones. In 
particular, they emphasize the importance of the 
length of the cone and cylinder, as well as other 
geometric parameters. The work of Lee et al. 
(2006), which doubled the length of the cyclone 
cone compared to the Stairmand model, revealed 
instability of the flow field and incomplete 
development of flow in the cylinder region as a 
result of the abrupt transition to the conical segment. 
These results highlight the importance of an optimal 
cylinder-to-cone-length ratio for efficient cyclone 
operation. Experiments by Hsu et al. (2014), 
covering different cyclone geometries, including 
cylinder and cone lengths, revealed a decrease in 
pressure drop and an increase in aerosol penetration 
with increasing these parameters. However, detailed 
data on flow field variables were not presented, 
which may limit the understanding of the 
mechanisms underlying these changes. Overall, 
these studies highlight the importance of 
considering various geometric parameters when 
designing and optimizing gas cyclones, as well as 
the need for further research to better understand the 
physical processes occurring within these devices. 
Indeed, there is limited literature examining the 
effect of cyclone length on cyclone performance. 

Madaliev et al.: Numerical study of highly efficient centrifugal cyclones –Numerična študija visokoučinkovitih 
centrifugalnih ciklonov 
Acta hydrotechnica 37/67 (2024), 137–151, Ljubljana 
 
139 
Most published work on cyclone length is based on 
comparing the increase in cylinder or cone length 
using a standard cyclone model. However, these 
results are often not applicable to each other due to 
differences in cyclone models and experimental 
conditions. Comparing the effects of geometric 
changes in specific cyclone models (Kuznetsov et 
al., 2009) operating under similar conditions is 
extremely important for understanding the effect of 
cyclone length on its performance. This approach 
will eliminate the influence of factors not related to 
geometric changes and more accurately determine 
the influence of the cyclone length on its operation 
(Tarasova, 2010). Thus, future research in this area 
should pay particular attention to comparing 
geometric changes in specific cyclone models under 
similar operating conditions to provide more 
accurate and comparable results. 
To eliminate the shortcomings inherent in classical 
centrifugal cyclones, article (Zlochevskiy and 
Mukhopad, 2015) proposes to use a new type of 
cyclone for cleaning process air. The highly 
efficient centrifugal cyclone is equipped with a 
conical screw insert housed in a perforated, 
truncated cone. This creates a confined space in the 
form of a helical channel with a decreasing cross-
sectional area. In this case, the exhaust pipe limits 
the interaction between the main flow, containing 
dust particles, and the counterflow, containing 
purified air. 
As part of this work, the effect of a flow swirling 
screw on the cyclone’s efficiency was investigated. 
This new approach can improve the efficiency of 
cleaning air from dust and other pollutants, which 
can lead to improved working conditions in various 
industrial processes. However, to fully understand 
the benefits and evaluate the effectiveness of this 
new type of cyclone, additional research and 
experimental testing of its operation under various 
operating conditions is required. 
 
2. Physical formulation of the problem 
The proposed device operates on the following 
principle (see Fig. 1a): a mixture of dust and air 
enters the device along a tangent line from above 
and begins to move downward, rotating between the 
conical and cylindrical parts. After this, the cone 
moves into a second, larger cone, where the 
particles begin to move toward the cone’s wall. A 
vortex is formed in the center of the cyclone, 
capable of carrying small particles into the bunker. 
To prevent this effect, a reverse cone is installed at 
the bottom of the device. To more effectively 
capture small particles, it is necessary to increase 
the centrifugal speed inside the cyclone. For this 
purpose, a screw is installed inside the cyclone, as 
shown in Fig. 1b. The pitch of the auger in Fig. 1b 
is fixed. However, Fig. 1c shows a modification 
where the screw’s pitch is gradually reduced, which 
leads to a further increase in the rotation speed of 
the particles inside the cyclone.  
Details of the studied cyclones are given in Table 1, 
corresponding to Fig. 2. 
 
2.1 Mathematical formulation of the problem  
The present study uses numerical simulation to 
predict the flow field variables, pressure drop, and 
collection efficiency. The Navier–Stokes equations 
are discretized using the finite element method 
(FEM). The choice of a particular turbulence model 
depends on the nature of the flow, the available 
hardware resources, and its ability to capture the 
complete physics of fluid flow. To model Reynolds 
stresses in RANS equations, Comsol Multiphysics 
5.6 (COMSOL Inc. Retrieved, 2015) provides 
various closure models; the SST model was used in 
this work. SST models are very robust turbulence 
models suitable for most engineering and industrial 
flows. 
To study the flow dynamics, consider a system of 
hydrodynamic equations. In this case, we will make 
the assumption that the volume density of particles 
in the main flow zone is negligible compared to the 
gas flow density. Indeed, in real cyclones, the dust 
density does not exceed 5 g/m
3
, the air density is 
1.29 kg/m
3
, and the particle density is 2700 kg/m
3
. 
However, the density of dust near the wall due to 
accumulation can greatly exceed this indicator. But 
this happens in a rather narrow layer, whose size can 
be neglected in comparison with the characteristic 
parameters of the cyclone. 

Madaliev et al.: Numerical study of highly efficient centrifugal cyclones –Numerična študija visokoučinkovitih 
centrifugalnih ciklonov 
Acta hydrotechnica 37/67 (2024), 137–151, Ljubljana 
140 
  
 
а) b) c) 
Figure. 1: Highly efficient centrifugal cyclones: a) without a screw cyclone, b) a screw cyclone with a uniform 
screw pitch, c) a screw cyclone with a variable screw pitch. 
Slika 1: Visokoučinkoviti centrifugalni cikloni: a) brez vijaka, b) z vijakom enakomernega koraka, c) z 
vijakom spremenljivega koraka. 
 
 
 
Figure 2: Details of the studied cyclones. 
Slika 2: Podrobnosti preučevanih ciklonov. 
Table 1: Dimensions of high-efficiency cyclones. 
Preglednica 1: Mere ciklonov z visokim izkoristkom. 
 
To numerically study the problem posed, a system 
of Reynolds-averaged Navier–Stokes equations is 
used (Malikov and Madaliev, 2020; Son and 
Murodil, 2020). The system of equations does not 
take into account the forces due to the effects of 
turbulent migration, namely Seffman, Magnus (lift), 
and Coriolis, because they are significantly less than 
the centrifugal force. Thus, for mathematically 
modelling the processes of transfer of dust particles 
and aerosols in dust collectors, it is sufficient to take 
into account centrifugal, gravitational, and Stokes 
forces. 
Dimen
sions 
Dimensionle
ss ratio 
(Size)/D 1-
cyclone 
Dimension
less ratio 
(Size)/D 2-
cyclone 
Dimension
less ratio 
(Size)/D 3-
cyclone 
D D D D 
De 0.4D 0.4D 0.4D 
C 0.3D 0.3D 0.3D 
H 2.4D 2.4D 2.4D 
h 1.7D 1.7D 1.7D 
a 0.25D 0.25D 0.25D 
b 0.3D 0.3D 0.3D 
e 0.92D 0.92D 0.92D 
f 0.2D 0.2D 0.2D 
d 0.25D 0.25D 0.25D 
Ds 1.3D 1.3D 1.3D 
g - 0.3D 
0.3D≈0.15
D 
dn 0.3D 0.3D 0.3D 

Madaliev et al.: Numerical study of highly efficient centrifugal cyclones –Numerična študija visokoučinkovitih 
centrifugalnih ciklonov 
Acta hydrotechnica 37/67 (2024), 137–151, Ljubljana 
 
141 
( )
2
0,
,
,
.
eff
p
mp
eff T
p
t
k
t

  

  
= 


= −  +  +



=−


=+

u
du
ug
d
du
uu
d
              (1) 
here u – is the air flow velocity vector; 
p
u – similar 
velocity vectors for the nth fraction of dust;  –  
gas density; ,
t
 –  molecular and turbulent 
viscosities, p – hydrostatic pressure, g – 
acceleration of gravity. 
The interaction coefficient between phases was 
determined through the Stokes parameter: 
   
2
18
.
m
m
k



=                                        (2) 
here 

 represents the density of the dust particles' 
material, and 
m
 denotes the 'effective' diameter of 
the particles. 
 
2.2 Turbulence models  
Menter's shear stress transport (SST) model is a 
combination of the k−ε and k−ω models. In this 
scheme, k−ω is used for the wall layer, while k−ε is 
used for the outer region. This combination allows 
the flow features both near the wall and in the outer 
regions to be taken into account, providing more 
accurate modeling of turbulent flows. The SST 
model is currently very popular and is widely 
included in many CFD programs. Its efficiency and 
accuracy make it the preferred choice for modeling 
shear stress transfer in a variety of applications. It is 
important to note that the SST model provides 
effective tools for modeling turbulent flows and has 
been successfully used in various hydrodynamic 
calculation packages. Its ability to account for flow 
patterns both within the wall layer and in external 
regions makes it a valuable tool for engineers and 
researchers involved in analyzing and modeling 
complex turbulent flows. 
 
 
*
2 2
1
( ) ,
( ) 2(1 ) .
kt
t
t
k
v v k P k
t
v v P F k
tv


  
  
    


=  +  + −




=  +  + − + −  


d
d
d
d
(3) 
Here k is the specific turbulent kinetic energy (m
2
 
s
−2
) and ω is the specific rate of turbulent dissipation 
(s
−1
). Additional data are presented in studies by 
Menter (1994), Smirnov and Menter (2009), and 
Spalart and Rumsey (2007). These works contain 
details and extended information about other 
meanings.  
 
2.3 Solution method 
COMSOL Multiphysics 5.6 provides several 
effective solvers for problems of various physical 
phenomena (Agullo et al., 2013; Kuzmin et al., 
2012). When choosing a suitable solver, it is 
important to consider the type of physics being 
modeled, the complexity of the problem, the 
required accuracy, and the available computational 
resources. To solve the equations of a two-phase 
turbulent model, the Segregated method with the 
PARDISO direct solver algorithm was used in this 
work. This method allows the interaction of various 
physical processes in the system to be taken into 
account by solving the systems of linear equations 
arising at each step of the iterative method. To 
accelerate convergence, the Newton iterative 
method with a damping factor of 0.1 was used. The 
iterative process was used for nonlinear equations, 
where, at each iteration, the linear systems were 
solved using the PARDISO algorithm. The 
calculations were continued for up to 250 iterations 
with a specified tolerance factor of 1 and a residual 
factor of 1000. The tolerance factor (1) specifies the 
accuracy of the solution for each iteration, and the 
residual factor (1000) indicates the maximum 
permissible level of discrepancy between the 
calculated and exact values. These parameters are 
necessary to control the convergence of the iterative 
method to ensure a balance between the 
computational load and the accuracy of the result. 
These parameters were tuned to the required 
accuracy and specificity of the problem, ensuring an 
efficient and reliable solution to the turbulent model 
equations within the framework of the software 
used. The computational experiment was carried out 

Madaliev et al.: Numerical study of highly efficient centrifugal cyclones –Numerična študija visokoučinkovitih 
centrifugalnih ciklonov 
Acta hydrotechnica 37/67 (2024), 137–151, Ljubljana 
 
142 
at flow rate 
0
18 / U m s = . Particle sizes are taken 
in the range from 0.2 to 10 microns. Solid particles 
are injected using the normal surface condition at a 
speed equal to the speed of the fluid entering the 
cyclone. Particles reaching the cyclone’s walls 
collide with it and are reflected back. For the 
condition of low dust load, it is quite reasonable to 
assume that the collision of discrete phase particles 
with the walls of the cyclone is ideally elastic, 
wherefore the recovery coefficients in both the 
tangential and normal directions of the wall are 
taken as unity. The total density of the solid phase 
at the inlet was equal and distributed uniformly over 
the cross section. Fig. 3 shows a grid diagram for 
calculating cyclones. Table 2 provides complete 
information about the computational normal grid. 
To ensure convergence to the accurate solution, a 
mesh independence study was conducted using the 
grids presented in Fig. 3. Specifically, we compared 
the axial and tangential velocity profiles obtained 
from the coarse, normal, and fine grids. The results 
showed negligible differences between the normal 
and fine grids, confirming that the normal grid 
provided a solution close to the converged and 
accurate result. 
2.4 Checking the model’s adequacy  
Since the process within a centrifugal cyclone 
involves complex turbulent dynamics, testing the 
model’s adequacy before applying it to describe the 
process in a new cyclone is an important step. The 
Stairmand cyclone, for which experimental data is 
available, is considered in this context. This allows 
you to compare the results of numerical simulations 
with real measurements and evaluate the model’s 
accuracy. To obtain complete information about the 
initial conditions and design of the Stairmand 
cyclone, it is recommended to refer to the relevant 
sources indicated in the articles (Brar et al., 2015; 
Hoekstra, 2000 and Wei et al., 2017), where an 
experimental study of this device was carried out. 
Fig. 4 shows the average axial and tangential 
velocities at z=217.5 mm from the cyclone roof with 
experimental values (Hoekstra, 2000).  
As can be seen from Fig. 4, the result obtained from 
the SST model is in good agreement with the 
experimental results. Therefore, this model can be 
used for our cyclone. 
 
2.5. Calculation results and their discussion 
Fig. 5 shows contour plots of the average axial 
velocity of cyclones in the midplane. 
 
   
Figure 3: Normal grids for calculating cyclones. 
Slika 3: Normalne mreže za izračun ciklonov. 

Madaliev et al.: Numerical study of highly efficient centrifugal cyclones –Numerična študija visokoučinkovitih 
centrifugalnih ciklonov 
Acta hydrotechnica 37/67 (2024), 137–151, Ljubljana 
143 
 
 
а) b) 
Figure 4: The average axial a) and tangential b) velocities at z=217.5 mm from the cyclone roof are 
represented by experimental values (Hoekstra, 2000). 
Slika 4: Povprečne aksialne a) in tangencialne b) hitrosti pri z = 217,5 mm od strehe ciklona so predstavljene 
z eksperimentalnimi vrednostmi (Hoekstra, 2000). 
Table 2: Provides complete information about the computational normal grid. 
Preglednica 2: Celostne informacije o računski normalni mreži. 
 
without a screw 
cyclone 
a screw cyclone with a 
uniform screw pitch 
a screw cyclone with a 
variable screw pitch 
Mesh vertices 82873 522 786 535 099 
Tetrahedra 214 572 1 309 184 1 364 717 
Pyramids 3546 43 880 56 624 
Prisms 86 436 537 442 524 384 
Triangles 13 552 105 106 119 532 
Quads 318 390 348 
Edge elements 860 7063 7853 
Vertex element 26 96 101 
Number of elements 304 554 1 890 506 1 945 725 
Minimum element quality 0.01873 0.006998 0.005481 
Average element quality 0.6853 0.6758 0.6627 
Element volume ratio 7.451E-6 1.668E-5 7.667E-6 
Mesh volume mm
3
 6.642E7 6.481E7 6.431E7 
 

Madaliev et al.: Numerical study of highly efficient centrifugal cyclones –Numerična študija visokoučinkovitih 
centrifugalnih ciklonov 
Acta hydrotechnica 37/67 (2024), 137–151, Ljubljana 
 
144 
   
а) b) с) 
Figure 5: Isolines of the velocity field in cyclones [m/s]: a) without a screw cyclone, b) a screw cyclone with 
a uniform screw pitch, c) a screw cyclone with a variable screw pitch. 
Slika 5: Izolinije hitrostnega polja v ciklonih [m/s]: a) brez vijaka, b) z vijakom enakomernega koraka, c) z 
vijakom spremenljivega koraka. 
 
   
а) b) с) 
Figure 6: Isolines of the tangential velocity field in cyclones [m/s]: a) without a screw cyclone, b) a screw 
cyclone with a uniform screw pitch, c) a screw cyclone with a variable screw pitch. 
Slika 6: Izolinije tangencialnega hitrostnega polja v ciklonih [m/s]: a) brez vijaka, b) z vijakom 
enakomernega koraka, c) z vijakom spremenljivega koraka. 
 

Madaliev et al.: Numerical study of highly efficient centrifugal cyclones –Numerična študija visokoučinkovitih 
centrifugalnih ciklonov 
Acta hydrotechnica 37/67 (2024), 137–151, Ljubljana 
 
145 
 
  
а) b) с) 
Figure 7: Isolines of the radial velocity field in cyclones [m/s]: a) without a screw cyclone, b) a screw 
cyclone with a uniform screw pitch, c) a screw cyclone with a variable screw pitch. 
Slika 7: Izolinije radialnega hitrostnega polja v ciklonih [m/s]: a) brez vijaka, b) z vijakom enakomernega 
koraka, c) z vijakom spremenljivega koraka. 
 
  
 
а) b) с) 
Figure 8: Isolines of the pressure field in cyclones [Pa]: a) without a screw cyclone, b) a screw cyclone 
with a uniform screw pitch, c) a screw cyclone with a variable screw pitch. 
Slika 8: Izolinije tlačnega polja v ciklonih [Pa]: a) brez vijaka, b) z vijakom enakomernega koraka, c) z 
vijakom spremenljivega koraka. 
 
 

Madaliev et al.: Numerical study of highly efficient centrifugal cyclones –Numerična študija visokoučinkovitih 
centrifugalnih ciklonov 
Acta hydrotechnica 37/67 (2024), 137–151, Ljubljana 
 
146 
  
а) b) 
 
Figure 9: Average axial speed at z=190, 490, and 
690 mm from the cyclone roof: a) without a screw 
cyclone, b) a screw cyclone with a uniform screw 
pitch, c) a screw cyclone with a variable screw 
pitch. 
Slika 9: Povprečna aksialna hitrost pri z = 190, 490 
in 690 mm od strehe ciklona: a) brez vijaka, b) z 
vijakom enakomernega koraka, c) z vijakom 
spremenljivega koraka. 
с) 
 
 
  
а) б) 
 
Figure 10: Average tangential speed at z=190, 490, 
and 690 mm from the cyclone roof: a) without a 
screw cyclone, b) a screw cyclone with a uniform 
screw pitch, c) a screw cyclone with a variable 
screw pitch. 
Slika 10: Povprečna tangencialna hitrost pri 
z = 190, 490 in 690 mm od strehe ciklona: a) brez 
vijaka, b) z vijakom enakomernega koraka, c) z 
vijakom spremenljivega koraka. 
с) 

Madaliev et al.: Numerical study of highly efficient centrifugal cyclones –Numerična študija visokoučinkovitih 
centrifugalnih ciklonov 
Acta hydrotechnica 37/67 (2024), 137–151, Ljubljana 
 
147 
 
 
а) b) 
Figure 11: Average axial a) and tangential b) velocities at z=720 mm from the cyclone roof. 1 – without a 
screw cyclone, 2 – a screw cyclone with a uniform screw pitch, 3 – a screw cyclone with a variable screw 
pitch. 
Slika 11: Povprečne aksialne a) in tangencialne b) hitrosti pri z = 720 mm od strehe ciklona: 1 – brez 
vijaka, 2 – z vijakom enakomernega koraka, 3 – z vijakom spremenljivega koraka. 
 

Madaliev et al.: Numerical study of highly efficient centrifugal cyclones –Numerična študija visokoučinkovitih 
centrifugalnih ciklonov 
Acta hydrotechnica 37/67 (2024), 137–151, Ljubljana 
 
148 
 
   
 
 
Without screw cyclone 
Ciklon brez vijaka 
 
   
 
 
Screw cyclone with uniform screw pitch 
Ciklon z vijakom enakomernega koraka 
    
  
Screw cyclone with variable screw pitch 
Ciklon z vijakom spremenljivega koraka 
0.2 s 0.4 s 0.6 s 0.8 s 1 s 1.2 s 
Figure 12: Particle movements at different times. Here the colors represent the particle’s velocity at the 
indicated moment. 
Slika 12: Gibanje delcev ob različnih časih. Barve predstavljajo hitrost delca v določenem trenutku. 
 
 

Madaliev et al.: Numerical study of highly efficient centrifugal cyclones –Numerična študija visokoučinkovitih 
centrifugalnih ciklonov 
Acta hydrotechnica 37/67 (2024), 137–151, Ljubljana 
 
149 
 
Figure 13: Percentage of lagging particles in cyclones at velocity 
0
18 / U m s = . 1 – without a screw 
cyclone, 2 – a screw cyclone with a uniform screw pitch, 3 – a screw cyclone with a variable screw pitch. 
Slika 13: Odstotek zaostalih delcev v ciklonih pri hitrosti 
0
18 / U m s = . 1 – brez vijaka, 2 – z vijakom 
enakomernega koraka, 3 – z vijakom spremenljivega koraka. 
Fig. 6 shows isolines and tangential velocities in 
cyclones. Fig. 7 shows isolines and radial velocities 
in cyclones. Figures 5–7 clearly show that, when 
using a screw (screw cyclone with uniform screw 
pitch), both the tangential and transverse speeds of 
the cyclone in Figure 6b increase by 40%. Similarly, 
when using a screw (screw cyclone with variable 
screw pitch) in Figure 6c, the tangential speeds 
increase by 60%. This leads to an increase in 
centrifugal force. Fig. 8 shows isolines and pressure 
in cyclones. Fig. 8 shows that the pressure drop 
without a screw is 722 Pa, 1457 Pa when using a 
screw cyclone with a uniform screw pitch, and 3115 
Pa when using a screw cyclone with a variable 
screw pitch. Fig. 9 shows the average axial velocity 
at z=190, 490, and 690 mm from the cyclone roof. 
Fig. 10 shows the average tangential speed at z=190, 
490, and 690 mm from the cyclone roof. Fig. 10 also 
shows that the tangential speed increases after 
inserting the screw. Fig. 11 shows the average axial 
and tangential velocities at z=720 mm from the 
cyclone roof. Fig. 12 illustrates particle movement 
over time in the upper part of the cyclones, with 
colors indicating their sizes, based on the inlet 
velocity 
0
18 / U m s = . Fig. 13 illustrates the 
percentage of outgoing dust particles for different 
size fractions at the inlet air velocity 
0
18 / U m s =
. Fig. 13 demonstrates that the efficiency of a screw 
cyclone (with a uniform screw pitch) for particles 
with a diameter of 0.1 μm is 27%, for particles with 
a diameter of 0.5 μm – 30%, and for particles with 
a diameter of 1 μm – 13%. This type of cyclone is 
better at collecting zinc particles than cyclones 
without an auger. On the other hand, the efficiency 
of a screw cyclone (with variable screw pitch) for 
particles with a diameter of 0.1 microns is 39%, for 
particles with a diameter of 0.5 microns 40%, and 
for particles with a diameter of 1 microns 23%. This 
cyclone option also captures zinc particles better 
than a cyclone without an auger. Conclusion: When 
using a screw cyclone (with variable screw pitch), 
particle retention is 15% more effective than with 
other types of cyclones. Thus, the use of a screw 
cyclone is economically justified in those 
applications where high separation efficiency is 
critical. However, if lower operating costs are 
required, two- or multi-stage screwless cyclones 
may be a more suitable option. A more detailed 
economic analysis taking into account the specific 
application is necessary to make a decision. 
 
 

Madaliev et al.: Numerical study of highly efficient centrifugal cyclones –Numerična študija visokoučinkovitih 
centrifugalnih ciklonov 
Acta hydrotechnica 37/67 (2024), 137–151, Ljubljana 
 
150 
2.6 Conclusions 
This work used the Comsol Multiphysics 5.6 
software package to simulate the aerodynamic 
processes of two-phase flow inside cyclones. In this 
case, the SST turbulence model was used. This 
model should be used to calculate internal flows in 
systems with curved boundaries, for example, when 
modeling separator cyclones. Based on the 
presented data, we can conclude that the use of a 
screw cyclone, especially with a variable screw 
pitch, significantly increases the efficiency of 
collecting zinc particles. Compared to traditional 
cyclones, auger cyclones show significant 
improvement in the retention of particles ranging 
from 0.1 µm to 1 µm in diameter. This indicates the 
potential of auger cyclones to improve the filtration 
and purification processes of gas streams, especially 
with regard to the removal of zinc particles. 
 
References 
Agullo, E., Giraud, L., Guermouche, A., Haidar, A., 
Roman, J. (2013). Parallel algebraic domain 
decomposition solver for the solution of augmented 
systems. Adv. Eng. Softw. 60, 23–30. 
https://doi.org/10.1016/j.advengsoft.2012.07.004.  
Avci, A., Karagoz, I. (2003). Effects of flow and 
geometrical parameters on the collection efficiency in 
cyclone separators. J. Aerosol Sci. 34(7), 937–955. 
https://doi.org/10.1016/S0021-8502(03)00054-5.  
Brar, L. S., Sharma, R. P., Dwivedi, R. (2015). Effect of 
vortex finder diameter on flow field and collection 
efficiency of cyclone separators. Part. Sci. Technol. 
33(1), 34–40. 
https://doi.org/10.1080/02726351.2014.933144.  
Chuah, T. G., Gimbun, J., Choong, T. S. Y. (2006). A 
CFD study of the effect of cone dimensions on sampling 
aerocyclones performance and hydrodynamics. Powder 
Technol. 162(2), 126–132. 
https://doi.org/10.1016/j.powtec.2005.12.010.  
“COMSOL Modeling Software”. COMSOL.com. 
Comsol, Inc. Retrieved [WWW Document] (2015). 
Hoekstra, A. J. (2000). Gas flow field and collection 
efficiency of cyclone separators. TU Delft, Ph. D. Thesis, 
Delft Univ. Technol. 
Hoffmann, A. C., De Groot, M., Peng, W., Dries, H. W. 
A., Kater, J. (2001). Advantages and risks in increasing 
cyclone separator length. AIChE J. 47(11), 2452–2460. 
https://doi.org/10.1002/aic.690471109. 
Hsu, C.-W., Huang, S.-H., Lin, C.-W., Hsiao, T.-C., Lin, 
W.-Y., Chen, C.-C. (2014). An experimental study on 
performance improvement of the stairmand cyclone 
design. Aerosol Air Qual. Res. 14(3), 1003–1016. 
https://doi.org/10.4209/aaqr.2013.04.0129.    
Kuzmin, D., Löhner, R., Turek, S. (2012). Flux-corrected 
transport: principles, algorithms, and applications. 
Springer Science & Business Media. 
Lee, J. W., Yang, H. J., Lee, D. Y. (2006). Effect of the 
cylinder shape of a long-coned cyclone on the stable 
flow-field establishment. Powder Technol. 165(1), 30–
38. https://doi.org/10.1016/j.powtec.2006.03.011.  
Malikov, Z. M., Madaliev, M. E. (2020). Numerical 
Simulation of Two-Phase Flow in a Centrifugal 
Separator. Fluid Dyn. 55(8), 1012–1028. 
https://doi.org/10.1134/S0015462820080066. 
Menter, F. R. (1994). Two-equation eddy-viscosity 
turbulence models for engineering applications. AIAA J. 
32(8), 1598–1605. https://doi.org/10.2514/3.12149.  
Smirnov, P. E., Menter, F. R. (2009). Sensitization of the 
SST turbulence model to rotation and curvature by 
applying the Spalart–Shur correction term. J. 
Turbomach. 131(4): 041010. 
https://doi.org/10.1115/1.3070573. 
Son, E., Murodil, M. (2020). Numerical calculation of an 
air centrifugal separator based on the SARC turbulence 
model. J. Appl. Comput. Mech. 6(Special Issue), 1133–
1140. https://doi.org/10.22055/jacm.2020.31423.1871. 
Spalart, P. R., Rumsey, C. L. (2007). Effective inflow 
conditions for turbulence models in aerodynamic 
calculations. AIAA J. 45(10), 2544–2553. 
https://doi.org/10.2514/1.29373. 
Surmen, A., Avci, A., Karamangil, M. I. (2011). 
Prediction of the maximum-efficiency cyclone length for 
a cyclone with a tangential entry. Powder Technol. 
207(1–3), 1–8. 
https://doi.org/10.1016/j.powtec.2010.10.002. 
Wei, J., Zhang, H., Wang, Y., Wen, Z., Yao, B., Dong, J. 
(2017). The gas-solid flow characteristics of cyclones. 
Powder Technol. 308, 178–192. 
https://doi.org/10.1016/j.powtec.2016.11.044.  
Xiang, R. B., Lee, K. W. (2005). Numerical study of flow 
field in cyclones of different height. Chem. Eng. Process. 
Process Intensif. 44(8), 877–883. 
https://doi.org/10.1016/j.cep.2004.09.006. 
Zhu, Y., Lee, K. W. (1999). Experimental study on small 

Madaliev et al.: Numerical study of highly efficient centrifugal cyclones –Numerična študija visokoučinkovitih 
centrifugalnih ciklonov 
Acta hydrotechnica 37/67 (2024), 137–151, Ljubljana 
 
151 
cyclones operating at high flowrates. J. Aerosol Sci. 
30(10), 1303–1315. https://doi.org/10.1016/S0021-
8502(99)00024-5.  
Zlochevsky, V. L., Mukhopad, K. A. (2015). Analysis of 
air flow formation in a cyclone. South Siberian Scientific 
Bulletin (4), 5–13. 
Kuznetsov, S. I., Mikhailik, V. D., Rusanov, S. A. 
(2009). Modeling the operation of a highly efficient 
cyclone-rotary dust collector. Bulletin of Kherson 
National Technical University (3), 36. 
Tarasova, L. A. (2010). Improving the technological 
efficiency of vortex-type devices in gas cleaning systems 
(Doctoral dissertation, Moscow State University of 
Environmental Engineering).