Geometric Modifications and Their Impact On The Performance of The Vortex Tube

International Journal on Theoretical and Applied Research in Mechanical Engineering (IJTARME)
_______________________________________________________________________________________________
Geometric Modifications and Their Impact on the Performance of the

Vortex Tube
1
Arun Gopal Rajagopalan, 2Peter J Disimile
1,2
Department of Aerospace Engineering, University of Cincinnati

Email: [email protected], [email protected]
AbstractA vortex tube is a simple device that can

separate a single stream of compressed air into two streams,
one at a temperature higher than the inlet gas stream and
the other at a temperature lower than the inlet stream.
Vortex tubes are widely used in cooling applications where
compactness, safety, and low equipment cost are basic
factors. To better understand gas separation performance,
a 3D model of vortex tube was designed using Ansys
Workbench, a commercial CFD package. The turbulent
flow within the device was modeled using the k- model.
Since the aim of this study was to investigate geometric
modifications and their impact on the performance of the
vortex tube, several changes in the vortex tube geometry
was examined. Model validation was achieved using a base
vortex tube model and compared to published results. Then
the effect of the number of inlet nozzles, Length to tube
Diameter ratio (L/D), Cold Diameter to tube Diameter ratio
((Dc/D) and Cone valve inner diameter (Dco/D) variation on
the performance of the vortex tube were studied. This was
followed by examining the introduction of vortex chamber
and the effect of the non-dimensional chamber diameter
((Dch/D) on the performance of the vortex tube.
Experimental validation tests were also conducted on
vortex tube currently on the market and the CFD results
were compared.
Index TermsCFD Simulation, Parametric
Temperature separation, Vortex tube
separating a compressed air stream into hot and cold

streams with different mass fractions. This phenomenon
is referred to as energy (temperature) separation. A
vortex tube consists of one or more tangential inlet
nozzles, an axial tube with a vortex chamber, a cold gas
outlet and a hot gas outlet. Dust free dehumidified
compressed air in the range of 5 to 6 bars is injected into
the vortex chamber tangentially through the inlet nozzles.
As the air accelerates through the inlet nozzles, it exits
into a cylindrical section with a high angular velocity
which generates an intense swirling flow. The gas then
proceeds towards the hot outlet with the conical back
pressure control valve (Figure 1). However, not all the
flow exits through the control valve. A portion of the gas
reverses its direction and moves axially along the center
of the tube towards the cold outlet, against the oncoming
hot swirling flow.
Study,
I. INTRODUCTION
The vortex tube was invented by a student named G.J
Ranque in 1932. He was able to create two streams with
different temperatures, using a highly compressed inlet
pressure supply. In 1945, a German physicist Rudolf
Hilsch improved the vortex tube design of Ranque and he
published his work in a scientific paper [2]. This
publication of Hilsch was widely read and was very well
accepted by the scientific community and thereby the
energy separation phenomena in the vortex tube became a
topic of great interest. Even though, Hilsch fabricated a
less than optimized vortex tube geometry, he was able to
measure the temperature difference between the inlet
stream and the outlet streams; however he was not able to
explain the physical process behind this effect. This
device was later named the Ranque-Hilsch Vortex tube
acknowledging the work done by both G.J Ranque and
Hilsch.
Figure 1: Schematic diagram of a Vortex tube and the

phenomena of flow reversal [21]
As the reversed flow move towards the cold exit the
thermal energy possessed by the core flow is transferred
to the peripheral region of the flow. This results in the
peripheral flow temperature exceeding the inlet gas
temperature. The peripheral flow exits through the
annular space between the tube wall and the conical valve
at the hot outlet, whereas the core flow exits through the
cold outlet orifice. The relative mass flow rates of hot and
cold gases are controlled by the conical back pressure
valve at the hot outlet. This energy separation
phenomenon is known as Ranque effect.
Vortex tube is compact, lightweight and is free from any

moving part. It does not have any issues related to
maintenance as it uses neither electricity nor chemicals. It
is cheap compared to other cooling devices and it does
not require any refrigerant unlike other localized cooling
devices. It is wear resistant as it is made of stainless steel.
The vortex tube functions as a device capable of
_______________________________________________________________________________________________
ISSN (Print): 2319-3182, Volume -4, Issue-3, 2015
1

_______________________________________________________________________________________________
It is resistant to corrosion and oxidation as well. The hot
side temperature and thereby the total temperature
difference can be altered by changing the geometrical
settings (conical control valve setting) of the vortex tube.
Most importantly it gives instant cold air. But to produce
the instant cold air, it requires costly compressed air.
Another important factor to note is that the efficiency of
this separation mechanism is not very good. But the
disadvantages of using the vortex tube are far
overpowered by the advantages. Considering these facts
it is gradually becoming a very popular device in the
industries where localized cooling is required. It is used
in electric and electronic industry for cooling CCTV
cameras and electronic controls. It is also used for
machine part cooling, food item cooling and gas sample
cooling. It is used in the industries where quick cooling is
of ultimate importance; like setting hot melts and
soldered parts. Cooling heat seals and environmental
chambers are other applications of vortex tube. It is even
used in certain "cryogenic" applications like cooling lab
samples and circuit testing.
tube over the years. Aljuwayhel et al. [7] investigated the

energy separation mechanism using the fluid dynamics
model of the vortex tube. Scheper [8] used probes and
visualization techniques to carry out an experimental
study. They measured the total temperature gradient,
velocity and pressure (which were the basic flow
parameters measured) and concluded that the axial and
radial components of velocity were much smaller than the
tangential velocity. Takahama and Yokosawa [9]
examined the possibility of shortening the vortex tube
chamber length by introducing diverging section for the
vortex chamber. Eiamsa-ard and Promvonge [10]
conducted experimental studies to investigate the
phenomena of energy separation. They used the snail
entrance method to inject the compressed air into the
vortex chamber. They found that this method improved
the cold side temperature drop and thereby improved the
efficiency of the vortex tube as well. Although no single
process has been conclusively validated, the current work
looks at the impact on thermal performance of various
parameters.
Although several theories have been formulated to

explain the phenomena of temperature separation in the
vortex tube, none of them appears to be completely
validated. Hilsch [1] proposed that the transfer of
tangential shear work from inner to outer fluid layers is
the cause of energy separation. Ranque [2] proposed that
the work transfer caused by compression and expansion
effects is the main reasons for the temperature separation
in the tube. Harnett and Eckert [3] suggested that invoked
turbulent eddies are responsible for the temperature
separation mechanism in the vortex tube. Since turbulent
motion has a wide spectrum of eddy sizes, and large
eddies (associated with LOW frequency fluctuations) and
small eddies (associated with HIGH frequency
fluctuations) can coexist in the same volume of fluid,
they suggest that the wide spectrum of eddies are the
reason for the phenomena of temperature separation.
Ahlborn and Groves [4] proposed that the embedded
secondary circulation is the phenomena that cause the
energy separation. They measured the axial and
azimuthal velocity components to reach this conclusion.
Stephan et al. [5] proposed that the formation of Grtler
vortices on the inside wall of the vortex tube is the reason
that drives the fluid motion in the vortex tube. Grtler
vortices are stream-wise secondary flows that appear in a
boundary layer flow along a concave wall. If the
boundary layer thickness is comparable to the radius of
curvature, the centrifugal action creates a pressure
variation across the boundary layer. This leads to the
centrifugal instability (Grtler instability) of the
boundary layer and consequent formation of Grtler
vortices. Kurosaka [6] reported that acoustic streaming
effect is the reason for temperature separation. Acoustic
streaming is a steady flow in a fluid driven by the
absorption of high amplitude acoustic oscillations. It is
the less-known and opposite of sound generation
produced by a flow.
Table 1: Energy separation in a vortex tube: Theories
Numerous research activities were conducted on vortex
Name of the
Researcher(s)
Ranque [2]
Hilsch [1]
Energy
separation
Theories
Work transfer
caused
by
compression
and expansion
effects
Transfer
of
tangential shear
work from inner
to outer fluid
layers
Harnett
and
Eckert [3]
Invoked
turbulent eddies
Ahlborn and
Groves [4]
Embedded
secondary
circulation
Stephan et
al.[5]
Formation
of
Grtler vortices
on the inside
wall of the
vortex tube
Kurosaka [6]
Acoustic
streaming effect
Importance of the
model
Discovered
the
basic vortex tube.
Presented
a
modified
and
improved version of
Ranque's
vortex
tube
Presented a model
based on turbulent
rotating flow with
solid body rotation
Introduced
ananalytical model
which
was
experimentally
validated
Introduced the idea
of
stream-wise
secondary flows
Introduced the idea

of
temperature
separation
by
attenuated
sound
waves.
Table 2: Major Studies conducted on Vortex Tube

Name of the
Researcher(s)
Aljuwayhelet
al.[7]
Study
Conducted
Used the fluid
dynamics
model of the
vortex tube
Importance of the
study
The study showed that
the
work
transferseparates
the
cold flow region and the
_______________________________________________________________________________________________
2

_______________________________________________________________________________________________
Scheper [8]
Takahama and
Yokosawa [9]
Eiamsa-ard
and
Promvonge
[10]
Used probes
and
visualization
techniques
Shortened the
vortex
tube
chamber
length
Conducted
experimental
studies
hot flow region

Experimental
study
showed that the axial and
radial components of
velocity were much
smaller
than
the
tangential velocity
Introduced a diverging
section for the vortex
chamber. This improved
the thermal performance
Used snail entrance
method to inject the air
into the vortex chamber.
This improved the cold
side temperature drop
and
thereby
the
efficiency of the vortex
tube
Even though several studies were conducted, none of

them appears to be conclusive. The analysis of the vortex
tube using mathematical or analytical method was of
great difficulty considering the complexity of processes
present within the vortex tube. The flow is highly
irreversible, compressible and viscous. So the
mathematical modelling produced results that were
highly inconclusive. The aim of the current study was to
vary parameters that were never varied before and
determine the optimum values which can be used to
improve the performance of the vortex tube. The work
done in this study can be divided into two parts. The first
part is to conduct a parametric study on the base
computational model of the vortex tube, and second
validate the CFD model using experimental data obtained
on a small commercial vortex tube available.
II. MATHEMATICAL MODEL

Selecting the right geometry is very important when it
comes to the CFD analysis of a vortex tube. Based on the
literature survey numerous geometries have been
examined and unfortunately, majority of the CFD studies
were conducted on two dimensional models [17], [18],
[19]. It is stated that the two dimensional modeling with
axisymmetric approximation can provide a good
approximation of the 3D phenomena. However, the flow
behavior in the vortex tube is very complex and highly
turbulent. Since these complex 3D vortical effects are
absent in the 2D model, it was decided to forgo the
axisymmetric approximation and develop a 3D model.
Various 3D computational models were considered
before finalizing the current geometry. The 3D model
used by John Wills et al. [12] was used for the initial
validation study. Matching of the results enabled, the
vortex model by Wills et al. [12] to serve as the base
model for all the geometric modifications that were to be
implemented. The geometry used by R. S. Maurya et al.
[13] had same geometrical parameters, but they used 6
inlets instead of the 2 inlets used by John Wills et al. [12].
These published results [12] [13] gave us the luxury to
compare the results obtained after geometric
modifications of the base model, with them whenever
possible.
Table 3: Geometrical parameters of the vortex tube used
for validation study
Parameters
Vortex Tube Length
Number of Nozzles
Length to Diameter Ratio (L/D)
Inlet Area to Area of the vortex tube
Ratio (Ai/A)
Hot Outlet Area to Area of the vortex
tube Ratio (Aho/A)
Cold Outlet Area to Area of the vortex
tube Ratio (Ac/A)
Values
120 mm
2
10
0.07
0.3
0.34
A. Meshing Strategy
Meshing was done in Ansys Workbench v 14.5.0.
Literature survey [12] [13] was conducted to select the
best meshing strategy. For such a complex geometry, an
unstructured mesh was preferred, using mainly
tetrahedral volumes. Inflation layers were applied to the
boundary. The volume elements near the tangential inlet
portion of the domain are highly skewed, thus reducing
the quality of the mesh. To eliminate the errors due to
coarseness of grid, an analysis of different number of
cells was performed. Therefore a grid independence
study was performed where the criterion was based on the
variation of cold exit temperature and hot exit
temperature, as the number of cells was changed.
B. Assumptions
To aid the solution of this complex vortical flow field,
several
simplifying
assumptions
were
made.
Specifically, the working medium was assumed to be an
ideal gas with adiabatic tube walls. So there is no transfer
of heat between the system and the surroundings. No
slip conditions were used at the solid boundary and the
flow is considered to be highly turbulent. The flow is
steady and therefore the system properties are assumed
not to change with time. However the flow is
compressible and therefore the density of the fluid
changes significantly. Body forces were considered to be
negligible and the flow was considered to be three
dimensional and subsonic with uniform fluid properties
at the inlet.
C. Boundary Conditions
An appropriate boundary condition is a must for

obtaining a successful solution. But, the determination of
the appropriate boundary condition can be a very time
consuming process. A study of previous works [12] [13]
showed that good results were obtained when a pressure
or mass flow-rate boundary condition is applied at the
inlet of the vortex tube. Highly pressurized gas is injected
through the inlet nozzles which separate into two
different gas streams; one that exits through the hot side
and the other that exits through the cold side. So a
pressure boundary condition is considered ideal at the
inlets and at the exits. Therefore, a pressure boundary
condition is applied to the inlet region (inlet nozzle) of
_______________________________________________________________________________________________
3

_______________________________________________________________________________________________
the vortex tube with total pressure of 5bar (gauge) and
total temperature of 300 K. Even though the inlet
pressure was selected based on the previous studies
conducted; the current CFD and experimental validation
testing conducted also showed that the inlet pressure of 5
bars produced the highest total temperature difference.
From the literature survey conducted [12] [13] it was also
clear that relative pressure is imposed at the outflow
boundaries; in most of the cases. The earlier studies
conducted showed that it was ideal to use atmospheric
static pressure at the hot exit. Thus a pressure outlet
boundary condition was imposed to the cone valve just
upstream to the hot exit of vortex tube (hot outlet) with a
pressure of 1bar (gauge). Further, a pressure outlet
boundary condition to cold orifice of the vortex tube set a
pressure of 0bar (gauge). The operating condition was
absolute pressure of 1bar and a temperature of 300K. Due
to the fast energy separation process in the vortex tube the
wall was assumed to be adiabatic. Based on the previous
CFD studies conducted, it was determined that the
tangential speed was below Mach number 1, (sonic
speed) inside the vortex tube.
D. Modeling in Fluent
In the simulation of compressible turbulent flows, the
Navier-stokes equations (2) and (3), energy equations (4)
are solved by using a CFD package based on rectangular
finite volume method (Ansys Fluent). The solver selected
for this simulation was a 3D and steady state, density
based implicit solver. A second order discretization
technique was employed along with a k- model to
account for the turbulence.
Table 4: Thermo-physical properties of air
Properties (Working Medium Air)
Density
Values
Ideal
Equation
Specific Heat
1006.43 J/Kg-K
Thermal Conductivity
0.0242 w/m-k
Viscosity
1.7894e-5 kg/m-s
Gas
The ideal gas equation can be written as:

P
=RT
(1)
The mass balance, momentum balance and energy

balance equations are given as:

+ . u = 0
t
u
t
+ . uu = p + 2 u.
(2)
(3)
Cp
T
+ . uT
t
= k2 T
(4)
To tackle the phenomena of turbulence, the more

advanced turbulence models such as the Reynolds stress
equations as well as the random number generation k-
turbulence model were investigated. It was found that
these models could not be made to converge for this
simulation. Bramo et al. [14] showed that, because of
good agreement of numerical results with the
experimental the k- model with standard wall functions
can be selected to simulate the effect of turbulence inside
the vortex tube, computational domain.
Since the k- model was developed in the early 1970s its
advantages and disadvantages is well documented. It is
the simplest of the two equation model to implement. The
solution converges relatively easy and it provides
reasonable predictions for many physical flows. The k-
model is valid only for turbulent flows and it can be used
in highly turbulent applications. For the k- turbulence
model the equations (5) and (6), for turbulent kinetic
energy k and dissipation are solved by using Ansys
Fluent.
k
kui
t k
+
=
+ 2t Eij Eij
t
xi
xj k xj

() (ui )
t
+
=
t
xi
xj xj
2
+ C1 2t Eij Eij C2
k
k
(5)
(6)
III. EXPERIMENTAL VALIDATION

STUDY
The experiment was conducted using the Small Vortex
Tube Kit from Streamtek Corp. (Figure 2a). The vortex
tube had three generators with different cold diameters,
provided with it. The generator is an interchangeable,
stationary part in the commercial vortex tube that
regulates the volume of compressed air and also the
temperature ranges that one can produce using the tube
(Figure 2b). The generators are one of the most important
components in the vortex tube as they decide the diameter
of the cold side and also the number of nozzles. The
nozzles are modeled as slits in a generator. Air is supplied
to the inlet of the vortex tube from a pressurized chamber.
A pressure transducer is used to read the inlet static
pressure. Mass flowmeters provided at the inlet and the
hot outlet is used to measure the inlet mass flow rate and
the hot outlet flow rate. Thermocouples connected to a
multimeter provided the temperature measurements at the
cold and hot exits of the vortex tube.
_______________________________________________________________________________________________
4

_______________________________________________________________________________________________
a)
b)
Figure 2: a) Small Vortex tube kit by Streamtek and
b) one of the generators used
Tc = Ti Tc
(7)
Th = Ti Th
(8)
The Total Temperature Difference (T) is the

temperature difference between the cold side gas total
temperature (Tc) and the hot side gas total temperature
(Th).
IV. RESULTS AND DISCUSSIONS

A. Grid Independence Study
The mesh (cell) size was varied from 16,445 to 212,464,
based on changes in the total temperature difference.
While the cold side temperature approached an
asymptotic state for a mesh size of 123,598, the hot side
temperature approached the asymptotic state when the
cell number was increased to 176,728. Therefore it can be
clearly seen from the total temperature difference (Figure
4) plot that there is an advantage in increasing the number
of cells to 176,728 for the configurations studied.
50
Temperature Difference
(K)
45
40
35
Figure 3: Test stand setup for the experimental validation

study
The vortex tube had three generators provided with it and
based on the experimental study conducted it was
concluded that the generator with a cold diameter of 4
mm provided highest temperature difference. The
generator had 6 slits (rectangular inlets). The inlet air
with pressures ranging from 1.37 bar (20 psi) to 4.83 bar
(70 psi) was injected into the vortex tube inlet. The inlet
and the hot outlet mass flow rates were separately
measured using temperature corrected flowmeters
provided at the inlet and hot outlet. The corresponding
cold outlet and hot outlet temperatures were acquired by
using the thermocouples-multimeter combination. An
average of ten readings was taken for each of the inlet
pressure input, at the hot side and the cold side. The hot
side total temperature, cold side total temperature and the
corresponding total temperature difference for different
inlet pressures were noted.
30
25
20
0
50,000100,000150,000200,000250,000
Mesh Size
Figure 4: Thetotal temperature difference variation with
the number of cells
B. Validation Study
(a) Flow physics study
To understand the flow physics inside the vortex tube, the
tangential (swirl) velocity was plotted along the radius of
the tube. The plot obtained clearly shows that the
vortexsystem generated in the vortex tube is a
combination of both a forced vortex and free vortex. The
forced vortex is formed around the vortex tube center and
free vortex is formed near the vortex tube walls.
The Cold end total Temperature Difference is the

temperature difference (Tc) between the inlet gas total
temperature (Ti) and the cold side gas total temperature
(Tc), and the Hot end total Temperature Difference (Th)
is the temperature difference between the inlet gas total
temperature (Ti) and the hot side gas total temperature
(Th).
_______________________________________________________________________________________________
5

_______________________________________________________________________________________________
1.2
Normalized Tangential
Velocity (Vt / Vmax)
within 5% of the supply energy, for the base model, and

thereby satisfying the conservation of energy.
Table 5: The total temperature difference variation with

the number of cells
0.8
0.6
Flow
Properties
0.4
Current
CFD
study
0.2
0
0
0.5
Normalized
Radius1(r/R)
1.5
Figure 5: The tangential (swirl) velocity along the radius

(normalized) at X/L=0.5
The above plot provides an indication of the tangential
velocity variation along the radius for the base model, at
X/L = 0.5. The current study is compared with the work
of Dutta et al. [25]. The tangential velocity is normalized
by dividing the local tangential velocity at each radial
location by the maximum tangential velocity, in both the
cases. Both studies show the same trend and are within
5% of one another, at X/L = 0.5. For the current study, the
maximum tangential velocity approaches a Mach number
of 0.5 approximately at the given location (X/L=0.5).
a)
b)
Figure 6: Two expamples of vortices: a) Free vortex and

b) Forced vortex [23]
In figure 6 the velocity distributions that forms in the
vortex tube is presented. These two types of vortices are
referred to as; free vortex and forced vortex. In the free
vortex the peripheral velocity is inversely proportional to
the radius (Figure 6a) and in forced vortex the peripheral
velocity is proportional to the radius (Figure 6b). Thus
the swirling flow in the vortex tube can be characterized
as a Rankine vortex because of the presence of a forced
vortex in the central core, and a free vortex surrounding
it.
The next step in the vortex tube flow physics study is the
proof of conservation of energy. The law of conservation
of energy states that the total energy of an isolated
system remains a constant. The energy equation is given
by:
E = CP T
Mass
flow
rate
(kg/S)
Temperature
(K)
Energy
(J)
Location
Inlet
0.01897
300
5736.65
Cold Outlet
0.01024
270.516
2777.01
Hot outlet
0.00873
313.669
2728.92
(b) Base Model Validation Study

To prove confidence in the current base model used in the
present study, the computed CFD results were compared
with the results obtained by John Willset al. [12].
Matching of the results enabled, the current vortex tube
model to serve as the baseline model for all the geometric
modifications that were to be implemented. A 3D replica
of the model was designed according to the specifications
given by Willset al. [12]. However, the meshing tool and
the meshing method were different from the ones
provided in his paper. Comparison of the current total
temperature contours to those found in Wills et al.are
presented and found to be in very good agreement.
Specifically, the peripheral total hot temperature and the
core total cold temperature were computed along the
axial length and compared. In the current computation,
the temperature contour (Figure 8) indicates that
themaximum hot side total temperature is 314 K and the
minimum cold side total temperature is 270 K. When the
contours obtained for the present study was compared
with Wills et al. [12] contours in Figure 8; (hot side
temperature of 312 K and the cold side temperature of
271 K), the deviation was found to be less than5 %.
Figure 7: Total Temperature - current CFD study
(9)
Therefore, for energy conservation in the vortex tube,

i CP Ti = c CP Tc + h CP Th
Figure 8: Total Temperature - John Wills et al. [12] study

(10)
The energy calculation is given in table 5 and it shows

that the energy contained within the hot and cold flows is
(c) Experimental Validation Study

In this section, the current computational model was
modified to match the geometrical conditions of a
_______________________________________________________________________________________________
6

_______________________________________________________________________________________________
commercially purchased vortex tube which underwent
testing and is used for model validation. The geometrical
parameters of the commercially purchased vortex tube
are given in Table 4. Meshing was again performed using
Ansys workbench meshing tool and solving was
performed using Ansys Fluent. The meshing and the
solving technique used here was exactly same to the one
that was used for the base model, in the initial phase of
this study. A 3D and steady state, density based implicit
solver was used. Second order special discretization
technique was used in this analysis along with standard
k- model with standard wall functions.
The hot side and the cold side temperatures were
experimentally acquired and used as the validation data
for the CFD simulation. It can be noted that the
temperature difference increases linearly with the
increase in pressure for the experimental validation study.
The CFD results obtained also follows an increasing
trend. But the temperature difference plot for the CFD
study is not linear as in the case of the experimental
validation study. The lowest temperature difference
obtained in the CFD study was 26.15 K while the highest
temperature difference was 55.52 K.
Temperature
Difference (K)
60
50
40
30
20
10
0
Temp Diff Exp

Temp Diff CFD
Pressure (bar)
Figure 9: Total temperature difference variation with

Pressure
Table 6: Geometrical parameters of the small Vortex
Tube from Streamtek Corp. used for experimental
validation study
Parameters
Vortex Tube Diameter
Number of Nozzles
Length to Diameter Ratio (L/D)
Inlet Area to Area of the vortex tube Ratio
(Ai/A)
Hot Outlet Area to Area of the vortex tube
Ratio (Aho/A)
Cold Outlet Area to Area of the vortex tube
Ratio (Ac/A)
Values
6 mm
6
18
0.0277
0.1736
0.444
C. Parametric Study
In this section, a computational model was generated
with the exact dimensions provided by Willset al. [12]
and a parametric study was performed. The impact of
varying the geometrical parameters like vortex tube
length to diameter ratio (L/D), number of inlets, Chamber
Diameter to Diameter ratio (Dch/D), cone valve inner
diameter to vortex tube diameter ratio (Dco/D) and Cold

diameter to Diameter ratio (Dc/D) on the thermal
performance of the vortex tube were evaluated. In the
current study the inlet pressure was varied from 2 bars to
6 bars gauge. The results showed that the hot side
temperature increases and cold side temperature
decreases as the inlet pressure was increased. However,
the hot and cold side temperature curves reach an
asymptotic state as an inlet pressure of 5 bars gauge is
attained; thus clearly demonstrating inlet pressure above
5 bar gauge is not efficient. As a result, the inlet pressure
of 5 bar was maintained for all the remaining
computations.
(a) Length to Diameter Study
The length of the vortex tube was varied from 40mm to
240mm, while keeping all other geometrical parameters
constant. This equated to a non-dimensional length
variation (L/D) from 3.33 to 20. The total temperature
difference rises with increasing tube length and finally it
reaches an asymptotic state. The asymptotic state is
achieved when L/D ratio becomes 10. Increasing the L/D
ratio from 10 to 20 gives a total temperature difference
rise of just 0.8283 K. Since industries prefer the vortex
tube to be compact and light weight, the length of the
vortex tube should be kept as small as possible.
Therefore, increasing the L/D ratio more than 10 to
produce this slight improvement is not recommended.
Now, to enable a comparison to published work [13], the
effect of variation of L/D on different number of inlets
was also studied. The number of inlet nozzles was
increased from 2 (base model) to 6. Computations were
then conducted to determine the total temperature
difference for different L/D configurations; for the 6 inlet
model. The number of nozzles in the base model was then
increased to 4 and simulations were run to obtain the
Temperature difference for Length to Diameter ratio of
10. Results for the 2, 4 and the 6 inlet case were then
compared with the CFD results in the published work
[13].
Figure 10: Vortex Tube: Length to Diameter ratio study

It can be noted from the study conducted that, the 2 inlet
case reaches an asymptotic state after an L/D ratio of 10 is
achieved. The same trend can be observed in the current 6
inlets study as well. It can also be noted that the
temperature difference obtained in the current 2 inlet
study is lower than that of the current 6 inlet study, when
L/D is less than 10. But when L/D is greater than 10 the
temperature difference is almost the same for both 2 inlet
and 6 inlet arrangement. The total temperature difference
for different L/D ratios showed that the current 6 inlet
study compares very well with the 6 inlet study of
Mauryaet al. [13] with differences within 10 %. It can
also be observed that, for an L/D ratio of 10 the
_______________________________________________________________________________________________
7

_______________________________________________________________________________________________
temperature difference obtained for the 4 inlet case shows
no difference with the 2 and 6 inlet study.
Beyond an L/D ratio of 10, no significant improvement in
the temperature difference is observed. Maurya et al.
[13]showed that the L/D ratio should be 10 for
maximum energy separation. Their results suggested that
the formation of stagnation zone is responsible for this
phenomenon. For the vortex tube configuration studied,
the stagnation zone formation occurs after the L/D ratio
of 10 is achieved, which is in accordance with Mauryaet
al. result. In the current L/D variation study, it can be
noted that even for the 6 inlet configuration, the
stagnation zone formation occurred after an L/D ratio of
10 was achieved.
55
COP =
Desired Output
Required Output
(11)
Temperature Difference
(K)
c C p T i T c
COPR= RT
50
6 Inlets (Current CFD

study)
6 Inlets (R. S. Maurya
et al. [13] study)
2 inlets (Current CFD
study)
40
35
30
25
0.000 5.000 10.000 15.000 20.000 25.000

L/D Ratio
Figure 11: Total temperature difference curve for
different L/D ratios
(b) Number of Inlets :
The base model used for the current study had 2 inlets. To
study the effect of change due to the number of inlet
nozzles, we considered four nozzle arrangements: 1inlet,
2inlets, 4 inlets and 6 inlets. All the other geometrical
parameters were kept constant. The hot side temperature
increased considerably, as the number of inlets was
increase from one to two. The cold side temperature
dropped as the number of inlets was increased from 1 to
2. As the number of inlets is increased from 2 to 4, the
total temperature remained almost the same for both hot
and cold side.
320
310
300
Cold
Temperature
290
280
270
260
0
Number of Inlets
Figure 12: Total temperature variation with number of

Inlets
(12)
i ln P i P atm
Equation 12 is rearranged into equation 13 to produce a

few non-dimensional terms:
Cp
1T c /T i
(13)
COP =
c
ln P i P atm
The inlet mass flow rate held constant for all the cases
considered. Because of this, for a low number of inlets,
the resulting pressure loss is high. Thus the energy
remaining for vortex generation is not sufficiently strong
to support the energy separation. However, at high
number of nozzles, since more nozzles are available to
inject the compressed air, the air velocity exiting the
nozzle is reduced and also has a negative effect on the
temperature separation phenomena [16].
Thus for 2, 4, and 6 inlet case, a balance between the
pressure drop across the inlet nozzles and nozzle outlet
velocity must be obtained, so that the total temperature
remains almost constant for these three cases. Whereas
for the 1 inlet case the pressure loss is very high
compared to the other three cases and therefore the
advantage of higher nozzle outlet velocity is not able to
improve the performance of the vortex tube.
The total temperature difference increased considerably
as the number of inlets was increased from one to two.
But the increase in the temperature difference was only
1.4 K as the number of nozzles was increased from two to
four. An increase to 6 nozzles did not show any
noticeable change in the temperature difference.
Temperature
Difference (K)
45
Temperature (K)
This behavior can be attributed to the Coefficient of

Performance (COP). The COP of a vortex tube can be
defined in terms of heating (COPH) or cooling (COPR)
and it relates the compression energy required to produce
the desired temperature change. COPH is defined as the
Coefficient of Performance of heating and COPR is
defined as the Coefficient of Performance of
refrigeration. The value of COP increases as the number
of inlets was increased from 1 to 2. But this trend is not
consistent with the increase in number of inlets. It reaches
a maximum somewhere in between.
50
40
30
20
10
0
0
Number of Inlets

number of Inlets
_______________________________________________________________________________________________
8

_______________________________________________________________________________________________
(c) Chamber Diameter to Diameter ratio:

The base model did not have a separate vortex chamber.
It was a long tube with a cold outlet section of different
diameter and a hot outlet with a conical control valve.
However, in this portion of the study, a vortex chamber
was introduced. In the literature review conducted [22], it
was learned that the vortex chamber (swirl chamber) is
the region where the air injected through the inlet,
accelerates to achieve a high rate of rotation. Even though
the concept of a vortex chamber was already known, no
previous studies were conducted on the impact of
chamber diameter variations. Thus, the new geometrical
parameter to be considered here is Dch/D. To study the
impact of the introduction of the vortex chamber into the
base model, five cases were considered. The first case is
the default case with no chamber (Dch/D = 1) and the fifth
case had a Dch/D ratio of 1.66. The other three cases
considered were Dch/D ratio of 1.167, 1.33 and 1.5.
Temperature (K)
340
320
Cold
Temperature
300
280
260
0.800
1.300
1.800
Dch/D
Figure 15: Total temperature variation with Chamber
Diameter
Now, if we take a look at the total temperature difference;
it started rising till the Dch/Dvalue was 1.5 and then it
dropped to a really low value of 37.76 K. The increase in
the temperature difference can be attributed to increase in
the tangential velocity due to the chamber diameter
increase. The total temperature drop after a Dch/D ratio of
1.5 can be attributed to decrease in spin efficiency due to
increase in vortex diameter beyond an optimum value.
Temperature
Difference (K)
It is therefore concluded that using 4 or 6 nozzles in the

current vortex tube geometry appears to provide the same
temperature difference. In a similar arrangement the
vortex tube used for the experimental validation study
also, had six inlets and therefore the remaining
computational simulations utilized the 6 inlet
arrangement. So this study justifies the use of six inlets
for the computational simulation of the vortex tube model
used for the experiment.
56.5
51.5
46.5
41.5
36.5
0.000 0.500 1.000 1.500 2.000
Figure 14: Vortex tube diagram with the vortex chamber

added to the base geometry
The hot side temperature increased till Dch /D was 1.5 and
it started dropping after that (Figure 15). It can also be
noted that the cold side temperature increased with the
increase in the chamber diameter.
The introduction of a vortex chamber increases the spin
generated in the vortex tube. Higher spin causes the fluid
particles in the peripheral layer to move faster. This
results in an increased hot outlet temperature.
Simultaneously the air in the inner core transfers the
kinetic energy it possesses to the hot stream and moves
out through the cold orifice. Due to this, the air in the
inner core region cools down and thereby the air in the
peripheral region heats up further. Thus, as the diameter
of the vortex chamber increase, the hot side temperature
increases till a point where the vortex chamber diameter
geometry affected the vortex formation (Figure 15). But
the temperature drop in the cold side is not as significant
as the hot side. After a Dch/D value of 1.5, the vortex
efficiency is observed to drop and therefore the hot side
temperature also goes down.
Dch/D
Chamber Diameter
The vortex chamber cross section for two cases (Dch/D=
1.166and Dch/D= 2.33) were considered. The
vorticitycontour for Dch/Dof 1.166 is shown in Figure 17.
The decrease in total temperature difference after Dch/Dof
1.5 show that there is a decrease in swirl efficiency after
the chamber diameter of 18 mm (Dch/Dof 1.5) is
achieved. The decrease in swirling motion efficiency can
be clearly observed in the vorticity contour for Dch/D=
2.33 (Figure 18). This validates the two statements made
earlier. The first is that the increase in thevortex chamber
(swirl chamber) diameter increases the swirl generated in
the vortex tubetill a certain point where the vortex
geometry affects the temperature separation phenomena
and the second one is that the swirl efficiency goes down
as the chamber diameter was increased beyond 18
(Dch/D= 1.5).
_______________________________________________________________________________________________
9

_______________________________________________________________________________________________
0.5 and then it gradually increases. When the cone
diameter gradually approaches the value of the vortex
tube diameter, the hot side temperature reaches an
asymptotic state.
Figure 17: Vorticity contour at Chamber dia = 14mm

(Dch /D= 1.166)
Temperature (K)
320
300
Cold
Hot
280
260
0.000
0.500
Dco/D
1.000
Figure 20: Total temperature variation with Cone Valve

Inner Diameter
46
Temperature
Difference (K)
44
42
40
38
0.000 0.200 0.400 0.600 0.800 1.000
Dc/D
Figure 18: Vorticity contour at Chamber dia = 28mm

(Dch /D= 2.333)
(d) Cone Valve Inner Diameter to Diameter ratio:
In this section of the study, the inner diameter of the cone
valve was altered. Six cases were considered including
the default case to investigate the effect of change of the
inner cone diameter. All other geometrical parameters,
including the outer diameter of the cone valve, were kept
the same as in the base model. Thus, the geometrical
parameter to be considered here is the Dco/D (Figure 19).
The smallest Dco/D considered was 0.167 and in that case
the cone valve geometry was close to a cone. The biggest
Dco/D considered was 0.833 and in that case the cone
valve geometry was close to a cylinder. This will be the
first published study, to present this effect.
Figure 19: Vortex tube diagram with the representation of

cone valve inner diameter
The cold side temperature increases as the cone diameter
increases. It reaches the highest value of 274.1 K for a
Dco/D value of 0.5 and then it gradually decreases. When
the cone diameter gradually approaches the value of the
vortex tube diameter, the rise in the cold side temperature
(although very small) can be observed. The hot side
temperature decreases as the cone diameter increases. It
reaches the lowest value of 314.12 K for a Dco/D value of

Cone Valve Inner Diameter
The total temperature difference shows that at very small
Dco/D and at very high Dco/D (when cone diameter almost
approach the vortex tube diameter) the temperature
difference is low. But the lowest value of temperature
difference is found at a Dco/D value of 0.5. Thus the ideal
Dco/D was found to be 0.667.
(e) Cold Diameter Change
Cold end orifice diameter is another important
geometrical parameter that affects the thermal
performance of the vortex tube. Six orifice diameter cases
were considered including the default case of Dc/D of
0.583.
As the cold diameter increases, the hot side temperature
initially drops and then rises. This increasing trend is
stopped at Dc/D of 0.667, where another drop begins to
occur. Cold exit temperature initially drops till Dc/D is
0.667 and then there is a steep increase in the cold side
temperature as the cold orifice diameter increases. The
lowest cold side temperature was obtained at the default
7mm orifice diameter case (Dc/D = 0.583). The total
temperature difference therefore attains its highest value
at Dc/D of 0.667 and after that it drops. It reaches the
lowest value at cold diameter of 10mm (Dc/D = 0.833)
which is very close to the vortex tube diameter.
Mauryaet al. [13] suggests that the reduction in pressure
gradient causes energy separation as well as supply air
with cold air mixing, at the cold orifice. This reduction in
pressure gradient is the reason for rising cold temperature
_______________________________________________________________________________________________
10

_______________________________________________________________________________________________
according to them. The rise in cold temperaturebeyond
Dc/D ratio of 0.583 in the current study can therefore be
attributed to reduction in the pressure gradient. They also
suggest that the radial pressure gradient increase which
leads to more energy separation is the reason for the
falling cold temperature. In the current study a clear drop
in the cold temperature can be seen from Dc/D ratio of 0.5
to 0.583. Therefore, the current study agrees very well
with the Mauryaet al. study.
VI. REFERENCES
[1]
G. J.Ranque, Method and apparatus for obtaining

from a fluid under pressure two outputs of fluid at
different temperatures, US patent No.1:952,
pp.281, 1934.
[2]
R. Hilsch, The use of expansion of gases in a

centrifugal field as a cooling process, Rev.
Sci.Instrumentation,
vol.
18(2),
pp.
108113,1947.
[3]
Harnett, J. and E. Eckert, Experimental study of

the velocity and temperature distribution in a high
velocity vortex-type flows, Trans. ASME, 79:
751-758.( 1957)
[4]
Ahlborn, B. and J. Gordon, The vortex tube as a

classical thermodynamic refrigeration cycle, J.
Appl. Phys., 88: 3645-653.( 2000)
[5]
Stephan, K., S. Lin, M. Durst, F. Huang, D. Seher,

An investigation of energy separation in a vortex
tube, Int. J. Heat Mass Transfer, 26: 341-348.(
1983)
[6]
Kurosaka M., Acoustic streaming in swirling

flows. J. Fluid Mech., 124: 139-172.( 1982)
[7]
Aljuwayhel, N.F., G.F. Nellis and S.A. Klein,

Parametric and internal study of the vortex tube
using a CFD model, Int. J. Refrig., 28: 442-450.(
2005)
[8]
G.W.Scheper, The vortex tube; internal flow data

and a heat transfer theory, J ASRE Refrig. Eng.,
vol. 59, pp. 985989, 1951.
[9]
H. Takahama and H.Yokosawa, Energy

separation in vortex tubes with a divergent
chamber, Trans ASME J Heat Transfer, vol. 103,
pp. 196203, 1981.
[10]
S. Eiamsa-ard, and P.Promvonge, Numerical

investigation of the thermal separation in a
RanqueHilsch vortex tube, Int. J. Heat Mass
Transfer, vol. 50(56), pp. 821832, 2007.
[11]
C.U. Linderstrom-Lang, The three-dimensional

distributions of tangential velocity and
total-temperature in vortex tubes, J. Fluid Mech.,
vol. 45, pp.161187, 1971.
[12]
John Wills. N, Karthika. A. S, Numerical

Analysis of Flow Behavior and Energy Separation
in Vortex Tube, 10th National Conference on
Technological Trends (NCTT09) 6-7 Nov 2009
[13]
R. S. Maurya, Kunal Y. Bhavsar, "Energy and

Flow Separation in the Vortex Tube: A Numerical
Investigation,
International
Journal
on
Theoretical and Applied Research in Mechanical
Engineering
[14]
Bramo, A.R., Pourmahmoud, N, Computational

Fluid Dynamics Simulation of Length to Diameter
60
Temperature
Difference (K)
50
40
30
20
10
0
0.000 0.200 0.400 0.600 0.800 1.000
Dc/D

Cold Diameter
V. SUMMARY
Areference validation study was conducted to provide
confidence in the current model. This was achieved, by
comparing the contours obtained in the current study with
those of Wills et al. [12]. Since the deviation in results
were within 5%, the vortex model of Wills et al.was used
as the base model for all the geometric modifications that
were to be examined in the present study. A geometric
parametric study on the base computational model of the
vortex tube was done.
From the current parametric study, it was concluded that
the:
i.
Ideal chamber diameter to Diameter ratio (Dch/D)

for the vortex tube studied was 1.5.
ii.
The ideal cone diameter to Diameter ratio (Dco/D)

for the vortex tube studied was 0.667.
iii.
The ideal Length to Diameter ratio (L/D) for the

vortex tube studied was greater than 10.
iv.
Ideal inlet nozzle number for the vortex tube studied

was four.
v.
The ideal cold diameter to Diameter ratio (Dc/D) for

the vortex tube studied was 0.7.
An experimental validation study was conducted on the

small Vortex Tube Kit from Streamtek Corp. by
comparing the experimental results with the already
obtained CFD results. The deviation in cold and hot side
temperatures of the experimental study from the CFD
results was less than 10%, giving additional confidence to
the current CFD model.
_______________________________________________________________________________________________
11

_______________________________________________________________________________________________
Ratio Effect on the Energy Separation in a Vortex
Tube, Thermal Science, 15(2011), 3, pp. 833-848
[15]
[16]
Mohammad O. Hamdan, Basel Alsayyed,

EmadElnajjar Nozzle parameters affecting
vortex tube energy separation performance, Heat
Mass Transfer, DOI 10.1007/s00231-012-1099-2
Nader Pourmahmoud&Abdol Reza Bramo The
effect of l/d ratio on the temperature separation in
the counter flow vortex tube
[17]
JiLinhart, Mohamad Kalal, Richard Matas

Numerical Study Of Vortex Tube Properties
[18]
C. H. Marquesa, L. A. Isoldia,E. D. dos

Santosa,and L. A. O. Rochab, Constructal design
of a vortex tube for several inlet stagnation
pressures
[19]
[20]
http://download.autodesk.com/us/algor/userguide
s/mergedProjects/setting_up_the_analysis/Fluid_
Flow/analysis_Parameters/Steady_or_Unsteady_
Fluid_Flow_%28Turbulence_Options%29.htm
[21]
http://www.exair.com/enUS/Primary%20Navigat
ion/Products/Vortex%20Tubes%20and%20Spot
%20Cooling
[22]
http://en.wikipedia.org/wiki/Vortex_tube
[23]
http://www.jsme-fed.org/experiment-e/2011_2/0
03.html
[24]
Dr.Ing.Ramzi, Raphael IbraheemBarwari,Effect

of Changing Cone Valve Diameter on the
performance of Uni-Flow Vortex Tube
[25]
T.
Dutta,K.P.
Sinhamahapatra,
S.S.
Bandyopdhyay"Comparison
of
different
turbulence models in predicting the temperature
separation in a RanqueHilsch vortex tube"
N.Pourmahmoud, S.Akhesmeh, Numerical

Investigation of the Thermal Separation in a
Vortex Tube
_______________________________________________________________________________________________
12

Geometric Modifications and Their Impact On The Performance of The Vortex Tube

Uploaded by

Copyright:

Available Formats

Geometric Modifications and Their Impact On The Performance of The Vortex Tube

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Geometric Modifications and Their Impact On The Performance of The Vortex Tube

Uploaded by

Copyright:

Available Formats

International Journal on Theoretical and Applied Research in Mechanical Engineering (IJTARME)