International Journal of Engineering Research & Technology (IJERT)

ISSN: 2278-0181
Vol. 3 Issue 8, August - 2014

Design and Analysis of Cyclone Seperator

Bharath Raj Reddy Dere1

G. Mahesh Babu2,

Mechanical department
Chaitanya Bharathi Institute Of Technology
Hyderabad,India.

Mechanical department
Chaitanya Bharathi Institute Of Technology
Hyderabad,India

A . Divya Sree3,

S. Rajiv Rao4

Mechanical department
Chaitanya Bharathi Institute Of Technology
Hyderabad,India.

ABSTRACT: The gas-solids cyclone separator is industrial

equipment that has been widely used. Due to its industrial
relevance, a large number of computational studies have been
reported in the literature aimed at understanding and
predicting the performance of cyclones in terms of pressure
and velocity variation. One of the approaches is to simulate
the gas-particle flow field in a cyclone by computational fluid
dynamics (CFD).
The cyclone performance parameters are governed
by many operational parameters (e.g., the gas flow rate and
temperature) and geometrical parameters. This study focuses
only on the effect of the geometrical parameters on the flow
field pattern and performance of the tangential inlet cyclone
separators using, CFD approach. The objective of this study is
three-fold. First, to study the optimized stairmands design by
understanding the pressure and velocity variations. Second, to
study the performance of a cyclone separator by varying its
geometrical parameters. Third, to study the performance of a
cyclone separator by varying its flow temperatures. Finally to
study the performance of a cyclone separator with collector
and compare to that without collector. Four geometrical
factors have significant effects on the cyclone performance
viz., the inlet width, the inlet height and the cyclone total
height. There are strong interactions between the effect of
inlet dimensions and the vortex finder diameter on the cyclone
performance. In this study, inlet geometry, cone- tip diameter,
cone height is taken for analysis.
The modeling of the cyclonic flow by
computational fluid dynamics (CFD) simulation has been
reported. The effect of the cone tip-diameter, cone height and
inlet geometry in the flow field and performance of cyclone
separators was investigated because of the discrepancies and
uncertainties in the literature about its influence. The flow
field pattern has been simulated and analyzed with the aid of
velocity components and static pressure contour plots. The
CFD model was used to predict the pressure and velocity
variations of cyclone geometries based on Stairmands high
efficiency design.

geometrical ratios. By considering this geometric ratios

the modeling of the cyclone done in solid works.

IJE
RT

Mechanical department
Chaitanya Bharathi Institute Of Technology
Hyderabad,India.
Email

1.

Figure 1.1.1: cyclone geometry in solid works

TABLE 1: Cyclone geometry used in this simulation

(stairmands

optimized design)
Geometry

a/D

b /D

Stairmands
High
Efficiency

0.5

0.2

D
x/D
0.5

S /D

h /D

0.5

1.5

H
/D
4

B /D
0.375

1.2. MODELLING IN SOLID WORKS

To design the cyclone the diameter (D) is
considered as 20 mm.
Step 1: Draw the sketch according to the stairmands ratios.
Step 2: specify the dimensions.
Step 3: Revolve the sketch 360 degrees and give thickness
as 0.1mm.
Step 4: to get the inlet, draw a rectangle on the part which
is developed by revolving.

MODELLING

Step 5: extrude the rectangle along the Z-axis (L=D).

1.1 THE STAIRMAND OPTIMIZED DESIGN
Stairmands conducted so many experiments on
the cyclone separator and finally developed the optimized

Step 6: draw a rectangle on the extruded plan to get the

hallow inlet. Select the rectangle and give extrude cut
throughout the extruded rectangle.
Step 7: save the geometry.

IJERTV3IS080015

www.ijert.org
(This work is licensed under a Creative Commons Attribution 4.0 International License.)

12

International Journal of Engineering Research & Technology (IJERT)

ISSN: 2278-0181
Vol. 3 Issue 8, August - 2014

2.

GOVERNING EQUATIONS

2.1 CONTINUITY EQUATION

The continuity equation describes the conservation of
mass.
Consider a differential control volume
Let be the density of fluid, u be the x-component, m
be the mass of fluid element
a= (density)
a=

[m = /x( dxdydz)]
b=u (velocity)

( / x)dx

b=u+ (u / x)dx

Fig 1.2.1 shows the dimensions of the cyclone separator

fig.4.1.1 fluid particle differential control volume

IJE
RT

Fig.1.2.2 cyclone separator design.

Rate of change of mass differential volume

= /t( dx dy dz)
M ass efflux in x - direction = /x( u) dxdydz
M ass efflux in x - direction = /y(v) dx dy dz
M ass efflux in x - direction = /z(w) dx dy dz
Since, Rate of accumulati on + M ass efflux = 0
/t(dx dy dz) + /x( u) dx dy dz +
/y(v) dx dy dz + /z(w) dx dy dz = 0
/t + /x( u) + /y(v) + /z(w) = 0
[ /t + . ( V) = 0]

Fig.1.2.3 Half sectional view of cyclone.

IJERTV3IS080015

www.ijert.org
(This work is licensed under a Creative Commons Attribution 4.0 International License.)

13

International Journal of Engineering Research & Technology (IJERT)

ISSN: 2278-0181
Vol. 3 Issue 8, August - 2014

2.2 ENERGY EQUATION

The energy equation can be derived by applyin
g I - law of thermodynamics to a moving fluid element.
Rate of change of energy inside the fluid element =
Net flux of heat into fluid element +
rate of work done on fluid element due to body and
surface forces
dE/dt = dQ/dt + dW/dt
]
[E = Q + W
E = Total energy of fluid per unit mass
= e + 1/2 (u^2 + v^2)
Rate of change total energy of fluid element per
unit mass = D/Dt [e + 1/2 (u^2 + v^2)]
Rate of change total energy of fluid element
= D/Dt [e + 1/2 (u^2 + v^2)] ?dxdy
Considerin g heat transfer across surface due
temperature gradient (conduction)

2.3 Significance of k- equation

K-epsilon (k-) turbulence model is the most
common model
used
in Computational
Fluid
Dynamics (CFD) to simulate turbulent conditions. It is a
two equation model which gives a general description
of turbulence by means of two transport equations (PDEs).
The original impetus for the K-epsilon model was to
improve the mixing-length model, as well as to find an
alternative to algebraically prescribing turbulent length
scales in moderate to high complexity flows. The first
transported variable determines the energy in
the turbulence and is called turbulent kinetic energy (k).
The second transported variable is the turbulent dissipation
( ) which determines the rate of dissipation of the
turbulent kinetic energy.
3 .CFD ANALYSIS
3.1
STAIRMANDS
OPTIMIZED
DESIGN
ANALYSIS: Open Ansys work bench and Select fluent
analysis system for the current analysis.

/x(u xx + vxy) + /y(vyy + uyx)] dxdy

]
[E = Q + W

IJE
RT

Q = K[ 2 T/x 2 + 2 T/?y 2 ] dxdy

Rate of work done = Force x velocit y
Total work done = (work done by body forces) +
(work done by surface forces)
) = [ufx + vfy +
Total rate of work done (W

Fig.3.1.1. fluent project schematic

3.1.1 CYCLONE GEOMETRY

Import the cyclone design from the solid works.
open the design modeler. Click on generate the imported
geometry appears. Select the part body in the tree outline
.select the body click on the screen. Change the solid body
into the fluid body. Close the design modeler and save the
project.

[De/Dt + 1/2 D/Dt (u^2 + v^2)]dxdy =

K[ 2 T/x 2 + 2 T/y 2 ] dxdy + [ufx + vfy +
/x(u xx + vxy) + /y(vyy + uyx)]dxdy

De/Dt == K[ 2 T/x 2 + 2 T/y 2 ] + u/x xx +

v/y yy + v/x xy + u/y yx
We have xx = u/x
yy = v/y
xy = yx = ( u/x + v/y)
C P DT/Dt = K[ 2 T/x 2 + 2 T/y 2 ]
C P [

T T
u
+ v T/y] = K[ 2 T/x 2 + 2 T/y 2 ]
t x

IJERTV3IS080015

Fig.3.1.2: solid cyclone geometry for the simulation.

3.1.2 MESH
Open mesh.>create named sections
1. Select the inlet face.name it as velocity inlet
2. Select the outlet face and name it as pressure outlet.
3. Select the rest of the faces and name them as wall.
Select mesh in tree outline. In mesh details default
conditions are set to be CFD and FLUENT solver as shown
in the fig 5.1.3. Give high smoothing condition and fine
relevance. And change the transition slow to fast to reduce
the no. of elements. Select mesh and click generate mesh to
obtain mesh. The generated mesh contains 74128

www.ijert.org
(This work is licensed under a Creative Commons Attribution 4.0 International License.)

14

International Journal of Engineering Research & Technology (IJERT)

ISSN: 2278-0181
Vol. 3 Issue 8, August - 2014

tetrahedron elements and 14577 nodes. Close the mesh and

update the mesh in project schematic.

STEP 1: General > check mesh (To verify the mesh is

correct or not) Enable pressure based type, absolute
velocity formulation and transient time steps.

Fig.3.1.7: general conditions

STEP 2: In models select the realizable k-epsilon (2eqn)

model and enable the enhanced wall treatment because the
design contains many walls.

Fig3.1.4: mesh top view

IJE
RT

Fig.3.1.3: mesh details and no of elements

Fig.3.1.8: defining the models

STEP 3: Boundary condition

3.1. Velocity inlet> z-velocity =20 m/s
Turbulence: specification method> k- model

Fig 3.1.5: mesh front view

3.1.3 SETUP
Double click on the fluent set up to set the
simulation conditions. The software automatically
recognizes the 3d dimension. The display mesh after
reading, embed graphics windows and work bench color
scheme must be enabled. Enable the double precision and
serial processing options. Then click ok to open the fluent.

Fig.3.1.9: inlet velocity boundary conditions.

3.2. Outlet
Turbulence: specification method> k- model
Backflow Turbulence kinetic energy= 5 m2/s2
Backflow Turbulence dissipation rate=10 m2/s3

Fig. 3.1.6: fluent launcher

IJERTV3IS080015

www.ijert.org
(This work is licensed under a Creative Commons Attribution 4.0 International License.)

15

International Journal of Engineering Research & Technology (IJERT)

ISSN: 2278-0181
Vol. 3 Issue 8, August - 2014

Fig.3.1.10: outlet boundary conditions.

3.3. Wall
Wall motion> stationary wall
Shear condition> no slip
STEP 4: Solution methods

Fig.3.1.13: set up for the calculation

3.1.4 SOLUTION
Residuals

Fig.3.1.14 residual graph

IJE
RT

Fig.3.1.11: details of solution methods

The solution is converged at the 355th iteration.

The vector fig 5.15 shows that the path followed by the
fluid inside the cyclone. The flow follows swirl flow
conditions as explained in the principle of the cyclone
separator. The left side bar shows various velocity ranges.
The color obtained in the vectors show the variation of
velocity at the different sections. The values of the velocity
can be studied from the left side scale. To understand the
complete flow inside the cyclone the contours at the z=0 is
computed. The velocity and pressure contours are shown in
fig 5.1.16. which gives complete flow variation in the
cyclone separator.
Vectors

STEP 5: Initialization: Select standard initialization and

compute from inlet velocity

Fig.3.1.15: vectors of velocity
Fig.3.1.12: initialization conditions

STEP 6: RUN> Check case>close

Time step size(s) =1; Number of time steps =50; Max.
Iterations / time step = 20 > calculate

IJERTV3IS080015

Pressure- velocity contours

The pressure in the cyclone separator increases
from the center to wall. The maximum and minimum static
pressures are 4.122e+002 (pa) and -6.44e+001 (pa)
respectively. The velocity first increases then decreases
from center to towards wall. The maximum and minimum
velocity magnitudes are 23.62 m/s and 0 m/s respectively.

www.ijert.org
(This work is licensed under a Creative Commons Attribution 4.0 International License.)

16

International Journal of Engineering Research & Technology (IJERT)

ISSN: 2278-0181
Vol. 3 Issue 8, August - 2014

Fig.3.1.16: the pressure and velocity contours at the section

Z=0 .

Pressure-velocity charts
The following charts shows the pressure and
velocity variation along the y-axis at different sections. The
graphs are plotted with y-axis as pressure/velocity and xaxis as radial distance along the x-axis.
Table 2: The sections
SERIES 1
AT Y=0
SERIES 2

AT Y=0.01

SERIES 3

AT Y=0.02

SERIES 4

AT Y= -0.01

SERIES 5

AT Y= -0.02

3.2 TEMPERATURE ANALYSIS

This analysis involves various flow studies at
various temperatures. The stairmands design is used for
the simulation in fluent. The same set up is used for the
temperature analysis as the stairmands design analysis.
Additional to that energy equation is activated to start the
temperature analysis. In the velocity inlet boundary
conditions the temperature of the inlet flow is added. This
study involves in the simulation of the cyclone at 4
different temperatures. The variations in the pressure and
velocity are noted and compared. The effect of the
temperature is justified.
The analysis is done at the temperatures
290,300,310 & 320 (k).
3.2.1 SOLUTION
Table 3: pressure and velocity readings at different temperatures
Temp (k)
290
300
310
320
Maxpressure (pa)

533.5

535.35

522.15

522.6

Mini pressure(pa)
Maxvelocity (m/s)

-185
25.25

-192.7
25.22

-194.7
25.12

194.7
25.12

Minivelocity (m/s)

Graphs: The graphs are drawn by taking y-axis as

pressure/velocity and the x-axis as radial distance along X.

IJE
RT

Graphs: The graphs are drawn by taking y-axis as

pressure/velocity and the x-axis as radial distance along X.

Graph 3.2.1 The variation of pressure velocity in cyclone and at section

Z=0 .

`Pressure contours

fig.3.1.17: the pressure and velocity graphs at different section along the
y-axis.

3.1.5 RESULTS AND DISCUSSIONS

The stairmands optimized design results are
shown in above graphs. To study the variations in pressure
and velocity 5 different sections are created along the yaxis as shown in the table 5.1.1. The variation in the
pressure and velocity inside the cyclone is shown in the fig
5.1.16 & fig 5.1.17. The solution got converged at the 15 th
time step, total time is 15s.
The pressure inside the cyclone along the radial
distance (x-axis) increases from centre to the ends. The
pressure variations are also shown as a graph in fig5.1.17.
The shape of the graph is U which shows the fall down of
pressure at the centre of the cyclone (x=0).
The velocity in the cyclone along the radial
direction first increases and then decreases from centre to
the wall. The shape of the graph is inverted W which
shows the fall down of velocity at the centre and at the
wall. The high velocity obtains in the middle portion of the
centre and the wall.

Fig 3.2.1 variation of pressure along the radial distance (x-axis) at z=0.

velocity contours

IJERTV3IS080015

www.ijert.org
(This work is licensed under a Creative Commons Attribution 4.0 International License.)

17

International Journal of Engineering Research & Technology (IJERT)

ISSN: 2278-0181
Vol. 3 Issue 8, August - 2014

Fig 3.3.2 mesh of the cyclone with collector

3.3.2 SET UP IS SAME AS THE STARMANDS

DESIGN ANALYSIS
3.3.3 SOLUTION
The solution of the simulation of the cyclone with
collector is compared with the cyclone without collector.
The variations in pressure and velocity are noted and
compared & the cyclone performance is justified.
Table 4: The pressure and velocity readings.
Fig 3.2.2 variation of velocity along the radial distance (x-axis) at z=0.

Without collector
Max pressure (pa)
Mini pressure (pa)
Maxvelocity (m/s)
Minivelocit (m/s)

418.5
-66.28
23.65
0

With
collector
403.4
-55.11
23.5
0

Graphs: The graphs are drawn by taking the y-axis as

pressure/velocity and x-axis as radial distance along X

IJE
RT

3.2.2 RESULTS AND DISCUSSIONS

The results are concluded that the cone height has
significant effect on the performance of the cyclone. The
pressure in the cyclone varies along the X-axis as shown in
the contours. The pressure first decreases and then
increases. The minimum pressure occurs at the mid section
(x=0). The graph shows the variation in the pressure along
the radial direction. The curve is in U shape explains the
decrease and increase of pressure. The velocity in the
cyclone first increase from the centre and then decreases at
the wall. The curve will be in M shape or reversed W
shape. The velocity is high at the middle portion of the
center and the wall.
The variation in the flow temperature slightly
varies the pressure for every 20k. The velocity of the flow
doesnt vary with the temperature. So we can say that the
temperature cannot affect the performance of the cyclone
because of the slight variations we can neglect the effect of
temperature.
3.3 THE ANALYSIS OF STAIRMANDS DESIGN
WITH COLLECTOR
The collector is the attachment to the cyclone
where the particles are collected. Collector is of any shape
(ex: cube, cylindrical). It locates at the end of the cone tip
and it prevents the re-entertainment of particles. In this
analysis a cube shaped collector is provided at the bottom
(with dimensions 15mm*15mm*15mm). The same set up
is used to simulate the cyclone with collector.
3.3.1CYCLONE GEOMETRY and MESH WITH
COLLECTOR

Graph 3.3.1: variation of pressure and velocity.

Pressure contours

Fig 3.3.3 variation of pressure along the radial distance (x-axis) at z=0.

Fig 3.3.1 geometry of cyclone with collector

IJERTV3IS080015

www.ijert.org
(This work is licensed under a Creative Commons Attribution 4.0 International License.)

18

International Journal of Engineering Research & Technology (IJERT)

ISSN: 2278-0181
Vol. 3 Issue 8, August - 2014

velocity contours

Pressure contours

Fig 3.3.4 variation of velocity along the radial distance (x-axis) at z=0.

Fig 3.4.1 variation of pressure along the radial distance (x-axis) at z=0.

Velocity contours

IJE
RT

3.3.5 RESULTS AND DISCUSSIONS

The results are concluded that the cone height has
significant effect on the performance of the cyclone. The
pressure in the cyclone varies along the X-axis as shown in
the contours. The pressure first decreases and then
increases. The minimum pressure occurs at the mid section
(x=0). The graph shows the variation in the pressure along
the radial direction. The curve is in U shape explains the
decrease and increase of pressure. The velocity in the
cyclone first increase from the centre and then decreases at
the wall. The curve will be in M shape or reversed W
shape. The velocity is high at the middle portion of the
center and the wall.The study shows that the collector
doesnt have much effect on the cyclone performance.
There are slight variations in the pressure and velocity
which can be neglected.
3.4 INLET GEOMETRY
The inlet geometry is one of most important
parameter which affects the cyclone performance. This
study involves with various inlet geometries. By varying
the height and width of inlet the inlet geometry varies. The
analysis of 3 different cyclones with different inlet
geometry is simulated in the fluent. By comparing all the 3
cyclones th ne better design is concluded. The same set up
is used for the simulation as stairmands design analysis.
3.4.1 SOLUTION
Table 5: pressure and velocity readings
a=12,
a=10 ,
b=6
b=5
Max pressure (pa)
513.5
415.7
Mini pressure (pa)
-102.9
-76.28
Maxvelocity (m/s)
Minivelocity (m/s)

24.25
0

23.84
0

Fig 3.4.2 variation of velocity along the radial distance (x-axis) at z=0.
a=8, b=4
334.8
-46.68
22.89
0

Graphs: The graphs are drawn by taking the y-axis as

pressure/velocity and x-axis as radial distance along X

Graphs 3.4.1 the pressure and velocity plots

IJERTV3IS080015

3.4.2 RESULTS AND DISCUSSIONS

The results are concluded that the cone height has
significant effect on the performance of the cyclone. The
pressure in the cyclone varies along the X-axis as shown in
the contours. The pressure first decreases and then
increases. The minimum pressure occurs at the mid section
(x=0). The graph shows the variation in the pressure along
the radial direction. The curve is in U shape explains the
decrease and increase of pressure. The velocity in the
cyclone first increase from the centre and then decreases at
the wall. The curve will be in M shape or reversed W
shape. The velocity is high at the middle portion of the
center and the wall.
The inlet geometry is the most important
geometrical parameter of the cyclone design. By varying
the inlet dimensions there is a huge variation in the
pressure and velocity. The maximum pressure in the
cyclone falls drastically by the decrease in the inlet

www.ijert.org
(This work is licensed under a Creative Commons Attribution 4.0 International License.)

19

International Journal of Engineering Research & Technology (IJERT)

ISSN: 2278-0181
Vol. 3 Issue 8, August - 2014

dimensions. For every 2mm decrease in the inlet height and

1mm decrease in width gives 20% decrease in the pressure.
By the decrease of this inlet dimensions the pressure drop
also decreases and we can say that the one with minimum
pressure drop is works more efficiently. So the inlet
dimension shows a large effect on the performance of the
cyclone. The velocity also decreases by the increase in the
inlet geometry, if velocity decreases the collection
efficiency decreases so the inlet dimensions must be high.
So the cyclone with the inlet height 10mm and inlet width
5mm is better design.
3.

REFERENCES
[1]

[2]

Khairy Elsayed 2011, PhD thesis on Analysis and

Optimization of Cyclone Separators Geometry using RANS
and LES Methodologies. Pages from 20-160.
John Anderson 2011, A Text Book on Computational Fluid
Dynamics, vol. 1

CONCLUSIONS

IJE
RT

After studying the existing literature and

performing analysis on certain cyclone parameters, the
following conclusion can be drawn:
The separation mechanism inside cyclone
separators is not well understood yet, and needs
more investigations.
Nearly all published articles have no systematic
and complete study for the effect of geometrical
parameters on the ow eld and performance.
In all operating conditions and cyclone types the
FLUENT CFD was found to be much closer to the
experimental measurement.
This project is done taking into consideration
single parameter at a time, the results may vary if
multiple parameters are considered at a time.
Some parameters have less interest compared with
others like the effect of inlet conditions and
cyclone height.
Ex: 1. A cone is not an essential part for
cyclone operation, whereas it serves the
practical purpose of delivering collect
particles to the central discharge point.
Ex: 2.Comparison between cyclone with
and without dustbin was done and
observed that a negligible effect of the
dustbin on the performance

IJERTV3IS080015

www.ijert.org
(This work is licensed under a Creative Commons Attribution 4.0 International License.)

20