See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/290788633

Numerical Analysis of a Vortex Tube: A Review

Article in Archives of Computational Methods in Engineering · January 2016

DOI: 10.1007/s11831-016-9166-3

CITATIONS READS

55 3,010

3 authors:

T.Karthikeya Sharma Amba Prasad Rao

National Institute of Technology, Andhra Pradesh National Institute of Technology Warangal
67 PUBLICATIONS 497 CITATIONS 91 PUBLICATIONS 1,243 CITATIONS

SEE PROFILE SEE PROFILE

Madhu Murthy K
National Institute of Technology Warangal
67 PUBLICATIONS 808 CITATIONS

SEE PROFILE

All content following this page was uploaded by T.Karthikeya Sharma on 08 March 2016.

The user has requested enhancement of the downloaded file.

Arch Computat Methods Eng
DOI 10.1007/s11831-016-9166-3

ORIGINAL PAPER

Numerical Analysis of a Vortex Tube: A Review

T. Karthikeya Sharma1 · G. Amba Prasad Rao1 · K. Madhu Murthy1

Received: 16 July 2015 / Accepted: 6 January 2016

© CIMNE, Barcelona, Spain 2016

Abstract Ranque–Hilsch vortex tube is a simple devise give a clear idea for the reader about the CFD models
with no moving parts which could generate cold and hot prediction capabilities.
air/gas streams simultaneously with compressed air/gas as
a working fluid. The energy and flow separation in a vortex
tube is highly depends on factors like nozzle shape, nozzle 1 Introduction
number, diameter and length of the vortex tube, inlet
pressure, control valve, diaphragm hole size and cold mass Ranque–Hilsch vortex tube (RHVT) simply called as Vortex
fraction. As the energy separation and flow patterns in a tube, is a device which generates simultaneously cold and
vortex tube are highly complex and were not explained hot gases using compressed gas or air as a source. The
successfully by any researcher, a computational study of vortex tube was invented by a French physics student
vortex tube flow and energy separation will give a better George Ranque by accident in 1931, while experimenting
understanding about the physics and mechanism involved. with a vortex-type pump delivering cold and hot air
Many researchers conducted computational fluid dynamic streams simultaneously from its ends [28]. Ranque started a
analysis of the vortex to have a deep insight about the small firm exploiting commercial potential of the device he
process of flow separation. In this paper computational developed having no moving parts. The vortex tube slipped
analysis of vortex by many researchers were presented in However, it soon failed and the vortex tube slipped into
along with the results obtained and suggestions to improve shade until 1945 when a German physicist Rudolph Hilsch,
the performance of the vortex tube. Researchers considered a scientific paper on the characteristics of the vortex tube.
Turbulence models which predict the performance pre- Since then vortex tube is considered as the “Ranque
cisely were discussed in the present paper. Researchers Vortex Tube”, the “Hilsch Tube”, the “Ranque–Hilsch
considered turbulence models like LES, k–ε, k–ω and RMS Tube”. Because of its compactness, easy maintenance and
to predict the energy separation in vortex tube. Some low cost vortex tube got acceptance by industrial applica-
researchers considered artificial neural networks (ANN) tions like spot cooling, cooling jackets, and chamber air
and Taguchi methods for their analysis. Comparison of the cooling etc. [42]. When the a compressed gas or air inducted
predictions with simulation results were also presented to into the vortex tube through a inlet nozzle mostly a tan-
gential nozzle a vortex flow is created inside the vortex
called as free vortex moving towards the other end, and
some portion of this leaves through the opening and other
got deflected by the control valve developing a forced vortex
moving at the central axis of the vortex tube and leaves the
vortex tube through opposite end. Heat exchange will occur
& T. Karthikeya Sharma between the two streams where the free vortex looses heat
[email protected]
and forced vortex gains heat. The two ends of the vortex
1
Department of Mechanical Engineering, NIT Warangal, tube are named as hot end and cold end depending on the
Warangal, AP 506004, India temperature of gas exiting through that end.

123
 T. Karthikeya Sharma et al.

1.1 Components of Vortex Tube Nozzle in a vortex tube in tangential in shape to allow
the inlet air to enter the vortex tube tangentially. The shape
A schematic diagram representing the vortex tube along of the nozzle plays major role in vortex creation thereby
with the various parts and the flow streams of free and the efficiency of the vortex tube. With increase in number
force vortices can be observed from the Fig. 1. of nozzles the vortex creation increases leading to
Vortex tube mainly consists of the following parts: improved performance of the vortex tube [16]. But there
observed to be a limit to the nozzle number because of the
1. Nozzle.
geometrical constraints of the vortex tube.
2. Diaphragms.
3. Control valve.
1.2 Working of Vortex Tube
4. Hot air side.
5. Cold air side.
Compressed air gas is inducted through the tangential
Vortex chamber is a portion where the diaphragm and nozzle and sectional view of vortex tube can be seen from
nozzle are place inside the vortex tube to facilitate the Fig. 1.3. Air expands and flows in spiral form because of
high pressure air stream to gain velocity and to form a the shape of the tangential nozzle. The vortex flow is called
free vortex [31]. The tangential nozzle plays a major role as free vortex flows close to the periphery of the vortex
in obtaining the spiral motion the air stream, the shape of tube towards hot side. Control valve at the end of the hot
the nozzle is designed in such a way that the air enters side deflects the free vortex creating a forced vortex
tangentially into the vortex tube and continuing a vortex diverted back towards the cold end close the central axis of
form [1]. the tube with a speed reaching 1,000,000 rpm.
Prof. Parulekar proposed the Hot air side cylindrical Sectional view of vortex tube can be seen from Fig. 2.
tube length should be three times the diameter of the vortex Control valve restricts the flow and helps in setting the cold
tube for the efficient operation of the vortex tube [36]. Hot mass fraction. By partly closing the control valve the pres-
streams are due to the free vortex moving close to the tube sure of the air near the valve is made more than the outside
periphery. The free vortex compressed because of the by, a reversed axial flow through the core of the hot side starts
expansion of the forced vortex to the walls gains heat and from high-pressure region. Energy transfer takes place
leaves through the hot end [22]. between the free and forced vortex streams because of the
Diaphragms or washer a small plate circular plate with a difference in their pressures and as the free vortex is com-
hole at the centre works as a throttle valve to reduce the pressed against the walls of the vortex tube because of the
temperatures of the cold streams further leaving towards expansion of the forced vortex. As pressure difference is
the cold end. Diaphragm is placed very next to the nozzle directly proportional to the temperature the free vortex
towards cold end. The diameter of the diaphragm hole absorbs heat and leaves the vortex tube through hot end and
affects the cold end temperatures to a considerable extent. forced vortex looses heat and leaves through cold end. The
Control valve is situated at the hot end. Control valve cold stream is escaped through the diaphragms hole reducing
is in conical shaped valve intended to deflect the free their temperatures further into the cold side, while hot stream
vortex to the other end. The conical angle of the control is passed through the opening of the control valve (Fig. 3).
valve plays a major role in splitting the air streams into
cold and hot streams [21]. It also controls the amount of 1.3 Types of Vortex Tube
air or gas to be deflected towards the hot and cold ends.
The cold fractions can be decided by controlling the The vortex tubes are classified as three types based on the
controlling valve opening and closing towards the hot cold exit position and flow patters they are shown below
end. proposed by Ranque:

Fig. 1 Vortex tube schematic.

(Courtesy: Science direct, K.
Dincer Paper, 2010)

123
Numerical Analysis of a Vortex Tube: A Review

Vortex Tube

Counter flow Uni- flow vortex Conical Vortex

vortex tube tube tube

Fig. 3 Schematic diagram of Counter flow vortex tube. (Courtesy:

jotse.com)
1.4 Counter Flow Vortex Tube

Counter flow vortex tube will have cold and hot flows on
its two opposite ends. The flow of forced and free vortices
will in opposite directions in this type of counter flow
vortex tube. In this type of flow hot air and cold air flow
in opposite directions [29]. Due to the heat exchange
taking place between opposite directions it is very effi-
cient, hot air and cold air leave the tube from opposite
ends of the inlet [49]. This type is most commonly used Fig. 4 Uni-flow vortex tube. (Courtesy: cielotech.com)
one (Fig. 4).

1.5 Uni-flow Vortex Tube

When the cold exhaust is placed at the same side of the hot
exhaust, it is named “uni flow”. In this type of flow hot air
and cold air flow in same direction. Less heat exchange
takes place between the same direction flowing air streams
so, it is less efficient. The hot and cold air comes out of the
vortex tube in one direction only [17].
From the experimental investigation it was found that
the performance of the uni flow system is worse than that
of the counter flow system. So, most of the time, the Fig. 5 Conical vortex tube. (Courtesy: Elsevier journal, Chengming
counter flow geometry was chosen (Fig. 5). Gaopaper, 2005)

Fig. 2 Sectional view of vortex

tube

123
 T. Karthikeya Sharma et al.

1.6 Conical Vortex Tube

Another type of geometry is the conical vortex tube (or

divergent vortex tube). In 1961, Paruleker designed a short
conical vortex tube. By varying the conical angle of the
vortex tube, he found that the parameter L/D can be as
small as 3. The conical type vortex tube can reduce the
need for lengthy vortex tubes to get the required drop in
temperature [30].

1.7 Advantages and Disadvantages of Vortex Tube

The vortex tube possesses several advantages as listed

below:
Fig. 6 Air suits of vortex tube (Courtesy: ExAir.Com)
● Since air is used as the working medium, the minor
leakages are insignificant.
● Its design is quite simple, and needs control of valves,
etc., for appropriate functioning.
● No moving parts are used, the life of the vortex tube is
expected to be infinite and maintenance is thus, almost
nil.
● It is light in weight, quite compact, easy to cool even in
a complicated space, just by introducing air to the point
under consideration since a stream of cool air comes out
of the vortex tube. Fig. 7 Simultaneous heating and cooling load maintenance by vortex
● Initial cost is low. tube. (Courtesy: Air Tx international.com)
● Overall no expert attendant is required.
● Low cost, maintenance free instant cold air.
economical way is to air-condition the operators working
● Adjustable temperature according to environment
near the hot places [33].
condition.
Presently, the largest single use is a cooling unit for
● At places where compressed air is readily available, it’s
protective clothing and helmets for such jobs as sand-
working expenses are low.
blasting, welding and handling toxic materials. These
● The main disadvantage is due to very poor C.O.P and
garments can be cooled simply by connecting a vortex tube
hence not suitable for large capacity refrigeration unit.
to an airline and attaching it to the suit (Fig. 7).
● Limited capacity is another disadvantage.
The air in the air suit is supplied to the mask and the top
half of the suit to maintain the normal body temperature.
1.8 Applications of Vortex Tube The air supply to the unit can be cooled or warmed by the
vortex tube. The supply of air can also be adjusted quickly
The use of vortex tube for small capacity applications is and easily. The safety and comfort provided by such a suit
always justified if the compressed air is readily available. can result in a very considerable saving in cost because the
The following few specific applications of vortex tube are operator has to work a long time under unpleasant condi-
described below (Fig. 6). tions. This is an ideal application of vortex tube, where
lightness, compactness and simplicity are of prime
1.8.1 Air Suits importance (Fig. 8).

A British company is manufacturing one piece air-cooled 1.8.2 Vortex Technology and Turbine Fuel Gas
suits. These suits are used by the operators entering vessels, Conditioning
tanks and pits where it is dangerous due to the concentra-
tion of toxic vapors, fumes or dust. It is commonly used by ● No line heaters
the workers working in the coal mines and foundries. It is ● No lost air
not always economical to condition the hot place like ● No maintenance
foundry where the heat load is considerably large. The only ● Environmentally friendly

123
Numerical Analysis of a Vortex Tube: A Review

dependent on cold storage facilities for storing a day’s

catch. Usually conventional cold storage rooms are
expensive for a local fisherman to afford. Conventional
refrigerators are expensive to buy for an average person.
Another important problem is transport of medicines,
specifically vaccines from one place to another. Several
vaccines require storage at low temperatures. Thus the
distance a medicine can be delivered via road is very
limited. Thus many remote areas do not receive vaccines at
local hospitals (Fig. 10).
We looked at a vortex tube based refrigeration system
for various needs described above. The goal of the exper-
iment was to study the fundamental limits of refrigeration
(temperature difference between cold and hot stream) that
Fig. 8 Vortex tube refrigerator and experimental setup. (Courtesy:
datacone.com)
can be achieved using vortex tubes in different fluidic
circuits, and measuring its cooling capacity. Compressed
air is employed in a vortex tube system which acts as the
energy source. Pressurized air is passed through a fluidic
structure which generates a vortex, separating the stream
into a hot and cold component. We experimented with
three different containers with various pressure settings.
Vortex tubes are easy to fabricate on a milling machine,
thus they can be locally built at the fables. All the com-
ponents are mechanical and thus once a design is finalized,
it can be very easily replicated. Being a purely mechanical
system, the wear and tear is very slow. Passive (mechanical
air pumps) or active (electrical air compressors) mecha-
nisms can be used to obtain compressed air (Fig. 11).

Fig. 9 Vortex tube Co2 separation unit. (Courtesy: P.A. Hilton Ltd)

Non-freeze pressure regulating unheated fuel air in the

Vortex Tube: the Vortex Tube in the course of its perfor-
mance discharges two low-pressure flows of different
temperatures: one very cold and one hot, thus creating the
cooling and the heating load. The availability of the cool-
ing load makes it possible to efficiently move it out of the
cold flow thus changing the level of the whole air internal
energy (Fig. 9).
● Removing the cooling load from the cold flow, this
results in its temperature increase.
● Combining the hot flow of the VT with the now warmed
cold flow in a way that the temperature of the combined
streams of the air meets its required value.

1.8.3 Vortex Tube Based Refrigeration

Refrigeration of food and medicine is an extensive prob- Fig. 10 Vortex tube cooling system for a drill bit. (Courtesy:
lem. Fishing communities living on the coast line are also arizonavortex.Com)

123
 T. Karthikeya Sharma et al.

Fig. 12 Vortex tube cooling system for a drill bit. (Courtesy:

nsltechnology.Com)

There is no doubt that the number of applications of

Fig. 11 Vortex tube spot cooling system. (Courtesy: exair.Com)
vortex tube is increasing and will probably continue to
increase. The availability of compressed air is an important
1.8.4 Vortex Tube for Carbon dioxide Separations factor in determining whether a vortex tube is in good
position.
Vortex tube can find its application in gasses separation
based on the density differences. A specific gas can be 1.8.6 Sub-zero Spot Cooling Using Compressed Air
separated from a combination of a two or more gasses
having control over the temperature drop at which a par- Vortex tubes our founding technology. More than 40 years
ticular gas can be liquefied [48]. Development and ago, we pioneered the application of the vortex tube prin-
demonstration of a unique gas–liquid contactor design for ciple to solve industrial cooling problems. Over the years,
the separation of carbon dioxide from natural gas and flue thousands of companies of all sizes have used vortex tubes
gas is being researched (Fig. 12). on their equipment to spot cool a part or process, thermal
The objective is to achieve significant improvements in test components, speed set solders or hot melt adhesives,
performance, operational flexibility and cost over conven- and so much more.
tional gas absorption technology for the separation of CO2 Vortex Tubes are an effective, low cost solution to a
from dilute CO2 -gas mixtures (\15 %). This work will wide variety of industrial spot and process cooling needs.
specifically address process and design optimization of a With no moving parts, a vortex tube spins compressed air
vortex tube CO2-liquid absorber. to separate the air into cold and hot air streams. While
French physicist Georges Ranque is credited with inventing
1.8.5 Vortex Cooling Systems the vortex tube in 1930, ITW Vortex was the first company
to develop and apply this phenomenon into practical and
The compact size of VCool vortex-cooling systems make them effective cooling solutions for industrial use. Vortex Tubes
ideally suited for small enclosures. Powered by compressed air have a very wide range of application for spot cooling on
(typically 100 psig, not to exceed 150 psig), VCool coolers machines, assembly lines and processes.
generate chilled air without refrigerants or moving parts. These
coolers provide exceptional reliability with minimal mainte- 1.8.7 Aviation
nance in even the most hot environmental conditions.
Another important vortex tube application is in cooling The cabins of high-speed gas turbine powered aero planes
enclosures. This use is commonly found in factories where are cooled with the use of bootstrap air-cycle. The vortex
high ambient temperature (near furnace) can be harmful to tube can also be used for the same purpose with less effi-
the tools and instruments, particularly electronic devices. ciency because more air would have to be bled-off from the
Many enclosure coolers are used on motor controls. In hot compressor at high pressure compared with bootstrap cycle
weather, motor controls were often trip-out even though the for the same cooling capacity. The use of vortex tube
motor is not overloaded. This nuisance tripping can rep- would result in overall reduction in weight, which is of
resent a major problem in a large plant on a hot day. prime importance. The use of vortex tube in military air-
Enclosure coolers solve this problem by air-conditioning craft may have a marked advantage over all other systems
the panel boxes. as a small cockpit is cooled.

123
Numerical Analysis of a Vortex Tube: A Review

1.8.8 Personal Air Conditioning below atmosphere develops rectangular model type box. A
sample, which is to be cooled, is kept in this box for a
Inside the PAC, a Vortex Tube spins the supplied com- specified time. This model is very useful in laboratories
pressed air, separating it into hot and cold airstreams. (See and research institutions.
Vortex Tubes for more detail) The cold air stream is
delivered to the vest via a ducting tube, while the hot air
exits out the side of the PAC unit. The low-pressure cold 2 Thermodynamic Analysis of Vortex Tube
air flows into the Diffuse-Air Vest’s perforated inner lin-
ing, which distributes the refrigerated air over the upper Vortex tube gets high pressure air from an air compressor
body. Unfolding the collars of the vest will reveal air holes through a tangential nozzle. Assume suffixes i, h, c stands
to cool the neck and face. (Model 220 uses hot and cold air for inlet to the nozzle, hot end and cold end, respectively
from both ends of the vortex tube to create a fully then the mass and energy conservation of control volume
adjustable heating or cooling option.) given by
Mass balance mi ¼ mc þ mh ð1Þ
1.8.9 Cutting Tools
Steady flow energy balance mi  hi ¼ mc  hc þ mh  hh
Many vortex tubes are used to cool machining operations ð2Þ
over small area. For example, many tubes are used to cool
Assuming the kinetic energies are negligible.
machining operations over a small area. For example,
The cold gas temperature difference or the temperature
many tubes are used to cool machining operations such as
drop of the cold air tube is defined as ΔTc = Ti − Tc.
drilling, milling, turning and reclaiming. As a rule those
The hot gas temperature difference or the temperature
materials which are difficult to machine are poor conduc-
raise of the hot air tube is defined as ΔTh = Th − Ti.
tors of heat. Therefore, machining operations heat does not
If the system is isentropic then the heat lost by the cold
readily flow away from the machining site and the tool
stream is equal to heat gained by hot stream.
overheats, causing excessive tool wear. Even a few degrees
of tool cooling by vortex tube can increase the tool life, m c ð Ti  Tc Þ ¼ m h ð Th  Ti Þ ¼ ð m i  m c Þ  ð Th  Ti Þ
improves surface finish and allows higher cutting speeds. 1
Ti  Tc ¼ ½ðmi /mc Þ  1  ðTh  Ti Þ ¼ ð  1Þ  ðTh  Ti Þ
Some materials are best cut without any lubrication. l
ð3Þ
1.8.10 Shrinks Fitting
where μ is the ratio of cold air to the air supplied, called as
Shrink fitting usually requires refrigeration for a short cold mass fraction.
period. Most factories have ring main air suppliers and From Eq. (3) we get.
vortex tubes could be connected at numerous points in the lðTi  Tc Þ ¼ ð1  lÞ  ðTh  Ti Þ
ring main where it is required. l½ðTi  Tc Þ þ ðTh  Ti Þ ¼ ðTh  Ti Þ
l ¼ ðTh  Ti Þ/½ðTi  Tc Þ þ ðTh  Ti Þ ð4Þ
1.8.11 Cooling of Gas Turbine Rotor Blades
¼ DTh /ðDTh þ DTc Þ
The research is going on from last 25 years to find out l ¼ mc /mi
effective and efficient method for cooling the gas turbine
If the process had undergone an isentropic expansion
rotor blades. The cooling of blades by passing the air
from inlet pressure Pi to atmospheric pressure Pa at the cold
through number of radial holes provided for the purpose is
end then the static temperature drop due to expansion is
successfully used in gas turbines used for aircraft and
given by
marine purposes. The cycle efficiency and specific output h i
both can be improved with the same quantity of air but at a DT0c ¼ Ti  T0c ¼ Ti 1  ðPa /Pi Þðc1Þ=c ð5Þ
lower temperature which can be made available with the
help of vortex tube. The compressed air can be bled from The temperature drop occurred in Vortex tube is DT0c .
the main compressor. The ratio of ΔTc to DT0c is called Relative Temperature
drop.
1.8.12 Laboratory Sample Cooling DTrel ¼ DTc /DT0c ð6Þ

Vortex Corporation USA in which cooled air from vortex The product of μ and ΔTrel represents the adiabatic
tube is circulated and the temperature inside is maintained efficiency of the Vortex tube because it is defined as

123
 T. Karthikeya Sharma et al.

actual cooling gained in vortex tube

gab ¼
Cooling possible with adiabatic exp ansion
    
gab ¼ mc DTc Cp / mi DT0c Cp ¼ lDTrel ¼ l  DTc /DT0c
 
gab ¼ ½DTh /ðDTh þ DTc Þ  DTc /DT0c
ð7Þ
The C.O.P of the vortex tube is defined as the ratio of
the cooling effect to the work input to the air compressor.
Cooling effect ¼ mc DTc Cp
 h i
Work input to air compressor ¼ Cp mi Ti ðPi /Pa Þðc1Þ=c 1 =gac

where ηac is the adiabatic efficiency of the compressor.

Coolingeffect
C:O:P ¼
Workinput
 "   # Fig. 13 Tangential velocity profiles calculated by using the k–ɛ
ðc1Þ=c
     turbulence model or the correlation from Keyes dimensionless tube
¼ mc DTc Cp  gac Cp mi Ti Pi Pa 1
length z/l0 = 0.98, z = 509.6 mm
( "  ðc1Þ=c #)
integrating the work done on the fluid by viscous force in
C:O:P ¼ lðDTc  gac Þ Ti P i P a 1
the numerical model. They incorporated Keyes correlation
to calculate turbulent viscosity instead of k–ε model as the
ð8Þ
k–ε model predictions of tangential velocity profiles were
Substituting the value of Ti from Eq. (5), in Eq. (8), we get not in good agreement with the measured values. Figure 13

flðDTc  gac Þg
C:O:P ¼  h ih i
ðc1Þ=c
DTc0
1  ðPa =Pi Þ ðPi =Pa Þðc1Þ=c 1
   nh i o
l  DTc DTc ga 0
ðPi Þðc1Þ=c ðPa Þðc1Þ=c  ðPa Þðc1Þ=c
ð9Þ
C:O:P ¼ nh i o
ðPi Þðc1Þ=c ðPa Þðc1Þ=c  ðPi Þðc1Þ=c
"   #
ðc1Þ=c
0
C:O:P ¼ l  ðDTc =DTc Þgac  Pa Pi

 
Substituting the value of l DTc =DT0c , from Eq. (7), we get shows the tangential velocity of flow obtained by experi-
h i mental by Bruun [11], CFX keys, CFX k–ε models. They
C:O:P ¼ gab  gac ðPa =Pi Þðc1Þ=c identified a circulating a secondary flow from axial and
radial flow diagrams which reviving energy from cold gas
It has been observed that the value of ηac is always con-
transmitted radially outwards to the hot gas. The expansion
siderably small, therefore the C.O.P of the vortex tube will
of axial flow causes a cooldown effect creating cold gas
also be very small of the order of 0.15–0.2 under normal
and temperature separation between cold and hot gasses.
operating conditions.
Behera et al. [6] conduct reported that the swirl, axial
and radial velocity components of both free and forced
3 Numerical Analysis of Vortex Tube vortex flow can be accurately determined by CFD analysis.
ed computational fluid dynamics (CFD) and experimental
Fröhlingsdorf and Unger [20] simulated the compressible studies on Ranque–Hilsch vortex tubes to optimize its
flow and energy separation using code system CFX. They parameters. They evaluated different nozzle numbers,
opined that energy transfer assessment can be done nozzle profiles along with swirl velocity, axial velocity and

123
Numerical Analysis of a Vortex Tube: A Review

Fig. 14 Temperature distribution in axial direction of vortex tube in

sections

Fig. 16 Heat and work transfer rate per unit length along the control
surface separating the hot and cold control volumes for the standard
k–e model

flow field into three regions, like hot flow region, cold flow
region and recirculation region inside the vortex tube. They
incorporated different turbulence models to identify the
effect of turbulence model prediction of vortex tube per-
formance and energy separation. They reported that
viscous shear acting on a rotating control surface produces
the torque responsible for the work transfer between the
cold flow and hot flow regions.
Fig. 15 Velocity magnitude of flow with six convergent nozzles
They opined that work transfer occurs from cold region to
radial velocity component as well as flow patterns includ- hot region whereas, the heat transfer occurs in reverse
ing secondary circulation flow of vortex tube. They direction from hot region to cold region tending to reduce the
validated their results with the experimental results. They temperature separation effect. They concluded that their
incorporated k–ɛ turbulence model of the Star-CD code. model successfully predicted the temperature separation
Their stated that the difference between cold and hot gas effect in vortex tube. The choice of turbulence model has
flows can be maximized increasing the length to diameter great effect on vortex tubes performance but does not change
ratio of vortex tube keeping the stagnation point within the the underlying physics mechanism. Figure 16 depicts the
tube and farthest from the nozzle inlet. Figure 14 shows the heat and work transfer rate per unit length along the control
temperature distribution in axial direction of vortex tube in surface separating the hot and cold control volumes for the
sections. They concluded that for a 12 mm diameter vortex standard k–e model Also, it was reported that the magnitude
tube—6 number nozzles, 7 mm cold end diameter, and of energy separation increases as the length of the vortex tube
6 mm cold end diameter for lowest possible cold gas increases to a critical length, later no variation in energy
temperature, length to diameter ratio of 25–35 are optimum separation was observed with increase in vortex tube length.
for best performance of the vortex tube yielding 361 K at Decrease in energy separation was increased was observed
hot end and 267 K at cold end at 60 % cold gas fraction. with increase in diameter of the vortex tube as the angular
Figure 15 shows the swirl patters of representing the velocity decreases with increase in vortex tube diameter.
velocity magnitudes of flow with six convergent nozzles. Skye et al. [47] conducted a comparison of exit temper-
Their CFD results revealed that performance of the vortex atures into and out of commercially available vortex tube by
tube will be degraded with secondary circulation flow. a CFD model with experimental results. They incorporated a
They obtained 0.59 COP as a heat engine and 0.83 as two dimensional axi-symmetric model that utilizing both
refrigerator. standard and renormalization k–ɛ turbulence models.
Aljuwayhel et al. [3] conducted CFD analysis on a They stated that the RNG k–ɛ model turbulence model
counter flow vortex tube using two-dimensional axi-sym- results of energy separation are not in good agreement with
metric CFD modelto study its energy separation measured results. Figure 17 presents the experimentally
mechanism and flow phenomena. They divided the vortex measured and predicted hot and cold power separation rate

123
 T. Karthikeya Sharma et al.

experimental data. Also, it was reported that diffusive

transport of mean kinetic energy has vital influence on
temperature separation, which occurs at the central axis
near the inlet region. They concluded that the static and
total temperatures are nearly same along the central line of
the tube. Figure 18 depicts the static temperature prediction
by (a) k–ɛ turbulence model (b) ASM for vortex tube of
Harnett and Eckert. Deviation was observed because of
convection and diffusion of total energy. It was reported
that along with diffusion of mean kinetic energy expansion
effect and stress production helps in increase in energy
separation far away from the inlet in the tube.
Farouk and Farouk [18] investigated numerically to
predict the flow field and the associated temperature sep-
aration within Ranque–Hilsch vortex tube. They employed
Fig. 17 Experimentally measured and predicted hot and cold power large eddy simulation (LES) technique to study a vortex
separation rate as a function of cold fraction tube with circumferential inlet with axial cold and cir-
cumferential hot stream outlets. They conducted
as a function of cold fraction. They stated that their model simulations for different cold mass fractions by changing
successfully predicted the maximum energy separation the hot end pressure. Figure 19 Depicts the stream trace for
operation point with respect to the cold fraction. They the inner core fluid flow in the entire vortex tube in three-
concluded that their model is capable of predicting the dimensional space. They reported that with increase in the
aspects of vortex tube accurately so it can be used for the cold mass fraction the hot exit temperature separation was
parametric studies on vortex tube. They opined that their increased and cold exit temperature was decreased. Also,
model as time saving cost effective tool in designing vortex maximum hot exit temperature separation was found for a
tubes. cold mass fraction of 0.78. Figure 20 shows the time
Eiamsa-ard and Promvonge [14] conducted numerical averaged total temperature contours for the vortex tube in
investigations to analyze the thermal separation in a the r–x plane (Case 1). Time average was performed
RHVT. They incorporated finite volume approach with k–ɛ between 0.5 and 1.0 s.
turbulence model and algebraic stress model (ASM) for the It was reported that the temperature separation predic-
computational analysis. They have used second order tions by LES modes were in good agreement with
upwind scheme to predict numerical diffusion and quick experimental results though it under predicted the cold exit
numerical schemes to compare with first order upwind and temperatures than k–ɛ turbulence model.
hybrid schemes. They stated that turbulent models are in Behera et al. [5] conducted three dimensional CFD
good agreement with measurements but ASM was analysis to investigate flow behavior and energy separation
observed to perform better than numerical and mechanism in RHVT using a commercial CFD tool STAR-

Fig. 18 Contours of static temperature predicted by (a) the k–e model and (b) the ASM for vortex tube of Hartnett and Eckert

123
Numerical Analysis of a Vortex Tube: A Review

Fig. 19 Stream trace for the

inner core fluid flow in the
entire vortex tube in three-
dimensional space(Case1)

Fig. 20 Time averaged total

temperature contours for the
vortex tube in the r–x plane
(Case 1). Time average was
performed between 0.5 and
1.0 s

Fig. 21 Radial profile of

angular velocities

CD. Their investigation concentrated on variation of fluid ● Flow in the vortex tube is largely governed by the
properties and flow parameters of fluid particles along the forced vortex regime except at inlet zone.
flow field. Parameters like stagnation temperature, static ● Large drop in static pressure was observed in the nozzle
temperature, static pressure, total pressure, axial, radial and till exiting the nozzle, there after no prominent drop in
swirl velocities were predicted along the length of the flow static pressure was observed till stagnation point.
field. They concluded the following points: ● The higher thermal efficiency of vortex tube at higher
hot gas fraction is due to substantial reduction in the
● Swirl velocity of hot end exiting fluid particles was 6 negative heat transfer.
times greater than axial velocity in the flow inlet zone.

123
 T. Karthikeya Sharma et al.

Fig. 22 Streamlines calculated by (A) the k–ε model and (B) the ASM

Fig. 23 Time averaged helium

mass fraction contours for the
vortex tube in the r–x plane
(case 1). Time averaging was
performed between 0.4 and
0.8 s

● Significant reduction in static total temperature after They concluded that both the turbulence models (k–ɛ
stagnation point to cold exit was observed due to energy and ASM) models yielded a recirculation zone near the
transfer to peripheral particles. inlet, but larger recirculation zone was predicted by ASM
● The viscous shear results in energy transfer between the than k–ɛ model. Also, they observed that the first order
fast moving inner core and slow moving outer core schemes like upwind and hybrid yielded better results than
because of the negative angular velocity gradient second order upwind scheme. Figure 22 shows the
radially outwards. Figure 21 shows the radial profile streamlines calculated by (A) the k–ε model and (B) the
of angular velocities. ASM.
Eiamsa-ard and Promvonge [15] analyzed the flow field Farouk et al. [19] incorporated LES model for predicting
and temperature separation in a uni-flow vortex tube. They gas flow species mass fraction and temperature fields in
studied the influence of turbulence modeling, grid density vortex tube. A vortex tube with circumferential inlet and
and numerical scheme in predicting the energy separation axial cold outlet and circumferential hot exit with nitrogen-
in the vortex tube. It was reported that the use of algebraic helium mixture was considered for the analysis to predict
Reynolds stress model (ASM) improved the accuracy of the parameters like temperature, pressure, species concen-
predictions than that of k–ɛ model. They reported that tration mass density and axial, radial and azimuthal
higher static temperature gradient was observed with at the components of velocity in the vortex tube. They stated that
circumferential region near the wall and separation effect very minimal gas separation was observed because of
was observed much in the core region near the nozzle inlet diffusion effects though large temperature separation was
in terms of total temperature. there. Also, separation mechanism was understood by

123
Numerical Analysis of a Vortex Tube: A Review

calculating the correlation between fluctuating components

of velocity, temperature and species mass fraction. Fig-
ure 23 shows the time averaged helium mass fraction
contours for the vortex tube in the r–x plane (case 1). Time
averaging was performed between 0.4 and 0.8 s. They
concluded that apart from 0.28 to 0.39 the cold exit tem-
perature was found to decrease with increase in cold mass
fraction but the hot exit temperature was observed to be
increasing. A maximum cold exit temperature separation of
38.7 K at a cold mass fraction of 0.39 and maximum hot
exit temperature separation of 41.8 K at 0369 cold mass
fraction was observed. A negligible gas separation of
0.8 9 10−4 − 1.0 9 10−4 was observed over the entire range
of old mass fraction. It was reported that an increase in gas
separation was observed with increase in cold mass frac-
tion as a result of increased separation effect of Fig. 25 Static temperature profiles in RANS and LES simulations at
temperature. different sections
Secchiaroli et al [43] investigated the flow inside a
commercially available vortex tube to simulate the turbu-
lent, compressible, high swirl flow was performed both by
RANS and LES techniques. They considered different
using a first order closure RNG k–ɛ and, a second order
RSM (Reynolds Stress differential Model) closure. They
compared the two dimensional RANS model results with
three dimensional LES computational grid. They consid-
ered Smagorinsky sub-grid scale model was used in LES
for their analysis using FLUENT 6.3.26. They observed a
split in the flow into two helical co-axial streams having
different thermal properties with hot stream near the
internal wall and cold one near the axis. Figure 24 shows
the three-dimensional visualization of streamlines patterns
with temperature map, obtained by LES with Smagorinsky
Fig. 26 The temperature gradient versus inlet mass flow rate and inlet
closure model. They observed a very close radial temper- pressures for N = 2
ature profile predictions between them far from the inlet by
RANS model, while the LES model predicts lower tem-
perature near the tube axis and different static temperature in RANS and LES simulations at different sections. They
radial profiles. Figure 25 depicts static temperature profiles concluded that the computational analysis can be used as
an effective tool to study the flow patterns inside the vortex
tube and the design parameters can be optimized for vari-
ous applications.
Uluer et al. [52] examined the effect of nozzle number
(N) and inlet pressure on the heating and cooling perfor-
mance of the counter flow type vortex tube using artificial
neural networks (ANN) designed in Pithiya software. They
considered inlet parameters such as the inlet pressure
(Pinlet), nozzle number (N), and cold mass fraction (μc) as
inlet parameters and temperature gradient between hot and
cold (ΔT) as the output parameter. They used Fermi
transfer function and back propagation learning algorithm
with Levenberg–Marquardt (LM) variant in the network
and validated its results with experimental results. Fig-
Fig. 24 Three-dimensional visualization of streamlines patterns with ure 26 depicts the the temperature gradient versus inlet
temperature map, obtained by LES with Smagorinsky closure model mass flow rate and inlet pressures for N = 2. They

123
 T. Karthikeya Sharma et al.

Fig. 27 The interaction plot for

S/N ratios

considered root mean square error (RMSE), mean absolute

percentage error (MAPE) and coefficient of determination
to determine temperature gradient between cold and hot
outputs. They reported that ANN was simple, fast in
deriving and calculating mathematical equations. They
concluded that ANN model can be successfully used for
predicting temperature gradient between the outputs. The
experimental data and the developed system analyses
showed that ANN reduces disadvantages such as time,
material and economical losses to a minimum.
Pinar et al. [37] discussed the exemplification of Tagu-
chi method in assessing maximum temperature gradient for
a counter flow RHVT performance. They considered
Taguchi L27 array for their analysis and conducted
experiments under different inlet pressure, nozzle number
and fluid type. They studied the effect of process param- Fig. 28 Radial distribution of turbulent viscosity at z/L = 0.5
eters on the performance of the vortex tube by Signal-to-
Noise ratio analysis (S/N), Analysis of variance (ANOVA) Dutta et al. [13] they incorporate an axi-symmetrical
and regression analysis to set optimal parameters. They CFD model to investigate the effect of different Reynolds
stated that inlet pressure (89.89 %) has dominant effect on Averaged Navier–Stokes (RANS) based turbulence models
total variation and it is followed by nozzle number in predicting the temperature separation and performance
(2.47 %) and inlet-pressure fluid type (1.78 %). Also, the in a RHVT. They considered standard k–ɛ model, RNG k–ɛ
effects of control factors and their two-way interactions on model, standard k–ω model, SST k–ω turbulence models
response were modeled via regression and correlation for their analysis. With an objective to find an optimum
analysis with R2-value of 94.1 %. Figure 27 exhibits the computational expense turbulence model in simulating the
interaction plot for S/N ratios. flow phenomena in a vortex tube they conduced numerical
They obtained an average error of 5.62 % between and experimental studies. They observed similar trends of
regression model results and experimental results, sup- temperature separation with all turbulence models, but the
porting application of Taguchi method for optimizing the magnitude of temperature separation models varied with
counter- flow RHVT performance. They concluded that turbulence model. Figure 28 represents the variation of
nozzle number 2 has given the maximum temperature turbulent viscosity at Z/L = 0.5 with different turbulent
gradient between hot and cold outlets. models. They stated that the static temperature and

123
Numerical Analysis of a Vortex Tube: A Review

Fig. 30 Comparison of the variation of total temperature at two

locations on the axis, x = 1 cm and x = 5 cm versus radial distance
(equal to y direction on the plane y=) for different number of nozzles
for the case m_ in = 13:1 g = s, ph = 156 kPa and pc = 103 kPa

Fig. 29 (Top) typical three-dimensional grid, (bottom) grids viewed

on the plane passing through the axis (plane y = 0)

pressure drops were observed to be smaller in radial

direction with standard k–ɛ model and observed to be
largest with RNG k–ɛ model.
They observed as radial expansion of air and turbulence
mixing resulting in a static temperature drop from the wall
to the axis of the vortex tube. The predictions of hot and
cold outlet temperature indicate that the standard k–3,
standard k–u and SST k–u models were reasonably good Fig. 31 Schematics of the air-standard cycle for the R–H tube
for practical design purposes. They concluded that more
complex turbulence analysis is required as the none of the
turbulence models considered couldn’t satisfactorily number of nozzles for the case min = 13:1 g = s,
resolved the turbulent swirling flow inside the vortex tube. ph = 156 kPa and pc = 103 kPa. They opined that practical
Shamsoddini and Nezhad [45] studied numerically the difficulties limit the number of nozzles in case of vortex
effect of nozzle number on the flow and cooling power of a tubes with small diameter.
vortex tube. They considered three dimensional numerical Simões-Moreira [46] incorporated a integral or control
fluid dynamic models for their analysis with number of volume approach to provide the conservation laws of mass
nozzles varying as 2, 3, 4, 6 and 8. They observed that and energy, as well as the constraint of the Second Law of
increase in number of nozzles increased the power of Thermodynamics. They envisaged a Thermodynamic air-
cooling significantly but the cold outlet temperature standard cycle for Ranque–Hilsh (R–H) or Vortex Tubes.
decreased moderately. Figure 29 shows the typical three Figure 31 shows the Schematics of the air-standard cycle
dimensional grid viewed through the axis (plane y = 0). for the R–H tube. Based on their analysis they stated that
They stated that an increase of 8.7 % in power of cooling the co-efficient of performance of Ranque–Hilsh vortex
was observed when the nozzle number was increased from tube doesn’t depend on the mode of operation (as a heat
2 to 8. pump or as a refrigerator). They reported that at a given
They reported that the extrapolation of numerical CFD operating pressure ratio isentropic COP delivers optima
results of cooling power showed 98 and 100 % matching values of cold and hot mass fractions. They evaluated their
with axi-symmetrical model results with the nozzle num- numerical analysis with experimental data. They concluded
bers 40 and 1000 respectively. Figure 30 shows that the Ranque–Hilsh vortex tubes are highly irreversible
comparison of the variation of total temperature at two devices. Figure 32 exhibits the Coefficient of performance,
locations on the axis, x = 1 cm and x = 5 cm versus radial COP, versus cold mass fraction, μH. Pressure ratio, RP = 5,
distance (equal to y direction on the plane y=) for different air atmospheric (k = 1.4). They obtained an optimum COP

123
 T. Karthikeya Sharma et al.

and Sigmoid transfer function for their back propagation

algorithm with variant in their neural network. They con-
ducted a statistical validation of their models results by
using RMSE, and relative absolute errors (RAE) and
coefficient of determination. They stated that the relation-
ship between the observed values and the MLR estimates a
high determination coefficient (R2) of 0.8924 and with a
RMSE value of 2.9165 but for the ANN model, R2 was
found as 0.9989 with a low root mean square error (RMSE
¼ 0.5016). Figure 33 depicts the observed and predicted
ΔT by ANN and MLR They concluded that ANN model is
simple to use, fast in calculating the different values and
formulation of different mathematical equation was easy
and the results are satisfactory.
Dutta et al. [12] employed a three dimensional CFD
Fig. 32 Coefficient of performance, COP, versus cold mass fraction,
μH. Pressure ratio, RP = 5, air atmospheric (k = 1.4) analysis to investigate the energy and species separation in
a vortex with compressed natural air as a working fuel at
normal atmospheric temperature and cryogenic tempera-
tures. They tried to predict the thermodynamic and
transport properties precisely inside the vortex tube using
NIST real gas model. They carried out their CFD analysis
using both real and Ideal gas laws. They compared the
results obtained by using both real and perfect gas laws
with experimental results at the normal atmospheric tem-
peratures. They reported that the separation of air into its
main streams (O2 and N2) was observed, though it is very
small. Figure 34 depicts the distribution of oxygen mass
fraction for Ti = 305 K. They concluded that the magni-
tude of separation of both species and energy was observed
to be small when compared with the out let temperatures
Fig. 33 Observed and predicted ΔT by ANN and MLR
predicted by real gas model at atmospheric temperatures
are closer to experimental results. Figure 35 shows the
for a cold mass fraction of 0.6. The actual COP was swirl velocity (m/s) distribution at the inlet region for Ti-
observed to be as low as 5 % of isentropic COP revealing = 305 K. (b) Swirl velocity (m/s) distribution along the
the need for further need for optimization of internal orifice vortex tube length for Ti = 305 K. They opined that in
and other parameters. cryogenic temperatures the swirl distribution was less when
Kocabas et al. [25] used ANN and multi linear regres- compared with the atmospheric temperatures. They opined
sion (MLR) to study the effect of cooling and heating that Soret diffusion was the main reason for species sepa-
performance of a vortex tube under varying nozzle number ration in the vortex tube based on the oxygen concentration
and inlet pressures (150–700 kPa). The designed their and lower species separation at cryogenic temperatures.
model to calculate the temperature gradient between cold Baghdad et al. [4] examined the vortex tube with an aim
and hot exists based on the parameters like number of to predict the energy separation and flow phenomena using
nozzles, inlet pressure, cold and hot mass fractions, mass a three dimensional CFD model considering 1/4th of the
flow rate. They considered Levenberg–Marquardt (LM) geometry model. They considered the k–ɛ, k–ɷ and SST k–

Fig. 34 Distribution of oxygen

mass fraction for Ti = 305 K

123
Numerical Analysis of a Vortex Tube: A Review

Fig. 35 a Swirl velocity (m/s)

distribution at the inlet region
for Ti = 305 K. b Swirl velocity
(m/s) distribution along the
vortex tube length for
Ti = 305 K

Fig. 37 Temperature distribution for GPL = 0.034

the streamlines colored by flow velocity for P = 380 kPa

and ξ = 0.3: (a) k–ɛ; (b) k–ɷ; (c) SST k–ɷ; (d) RSM. They
stated that except RSM model the other models over pre-
Fig. 36 Streamlines coloured by flow velocity for P = 380 kPa and
ξ = 0.3: a k–ɛ; b k–ɷ; c SST k–ɷ; d RSM dicted the temperature gradient results. They concluded
that despite the increased computational costs, second
ɷ two-equation models and the second moment closure moment closure seem to be more suitable for studying the
model (RSM) for the analysis with air as a working fuel flow in a vortex tube.
with varying inlet pressures from 200 to 380 kPa. They Pourmahmoud et al. [39] investigated the effect of
reported that all the considered four models are capable of helical nozzles on the refrigeration and energy separation
predicting the energy separation and flow phenomena of in the Ranquee–Hilsch Vortex Tube (RHVT) by using
vortex tube, but the advanced RSM predictions are in good CFD. They considered the radial gap between the helical
agreement with experimental results of temperature gra- nozzles from the vortex chamber (GPL) to be a significant
dient between the cold and hot outputs. Figure 36 shows design parameter affecting the performance of the vortex

123
 T. Karthikeya Sharma et al.

Fig. 40 Contours of total temperature inside a vortex tube with

divergent hot tube (β = 2)

hot inner peripheral region by the distribution of angular

velocity in radial direction. Figure 40 depicts the contours
of total temperature inside a vortex tube with divergent hot
Fig. 38 Hot exit temperature for vortex tube with 3 helical nozzles tube (β = 2). They concluded that the cooling performance
of vortex tube can be improved by using a hot divergent
tube. They obtained highest temperature drop at a diver-
tube. They incorporated k–ɛ model to simulate the turbu- gent angle of β = 2.
lence created by highly rotating complex flow field. Korkmaz et al. [26] investigated the effect of conical
They used helical nozzles having same profile but with valve angle (ø) and length to diameter ratio (L/D) of the
different GPL (0–0.169) for their analysis and nozzle vortex tube on its performance using ANN by using
numbers 1, 2 and 3. They stated that maximum cooling experimental data. They considered inlet pressure (Pi),
effect and swirl velocity was observed with GPL = 0.034. conical valve angle, L/D ratio and cold mass fractions as
Figure 37 shows the hot exit temperature for vortex tube input parameters and temperature gradient as the output
with 3 helical nozzles. They reported that keeping the parameter. They considered Levenberge–Marquardt learn-
GPL = 0.034 the variation of nozzle number showed sig- ing algorithm and multilayer feed forward model in the
nificant variation in temperature separation. Figure 38 ANN and the transfer function chosen was hyperbolic
temperature distribution for GPL = 0.034. They concluded tangent function for the analysis. They stated that increase
that no significant variation in coefficient of performance in temperature difference was observed with decrease in
was observed with variation in nozzle number. conical valve angle for the given configuration. Their
Pouraria and Zangooee [38] investigated the effect of analysis yielded best results at 30° conical angle. They
using divergent tube on the energy separation of vortex reported that decrease in L/D ratio increases the perfor-
tube and to find the optimum divergent angle for the best mance of the vortex tube except for L/D = 10. Increase in
performance of the vortex tube using k–ɛ turbulence model inlet pressure was observed to be increasing the perfor-
and experiments. They investigated the work and heat mance. Figure 41 details the total temperature versus cold
transfer under varying divergent angles and numerically mass fraction at different inlet pressures for h = 15 mm,
and experimentally. Figure 39 shows the schematic dia- L/D = 20 and conical angle (ø) = 30°. They concluded that
gram of the geometry and dimensions of vortex tube used ANN can be successfully used to predict the vortex tube
in this study. They stated that increase in cooling perfor- performance, and the best performance was observed at 30°
mance was observed with increase in divergent angle, but conical angle, L/D = 20, inlet pressure as 3 bar.
there exist a critical divergent angle after which the cooling Shamsoddini and Khorasani [44] conducted simple,
performance was observed to be decreasing. They stated thermodynamic analysis based on the conversation equa-
that the numerical results revealed that the swirl velocity is tions and second law of thermodynamics with some
much higher than axial and radial velocities. They auxiliary relations and interpolation tables provided by
observed transfer of shear work from the colder region to numerical simulations on the vortex tube. They stated that
the numerical simulation results are in good agreement
with the experimental results. Figure 42 shows the CFD
results of static (a) and total temperatures (b) for set #1 in
Table 4: T0in = 291.48 K, Pin = 256.6 kPa, Pc = 113 kPa,
Ph = 242 kPa. They obtained lower cold outlet temperature
in the lower cold mass fraction when compared with high
cold mass fractions. Also, they concluded that vortex tube
Fig. 39 Schematic diagram of the geometry and dimensions of vortex parameter optimization is more convenient with the pro-
tube used in this study posed approach and genetic algorithm.

123
Numerical Analysis of a Vortex Tube: A Review

Fig. 41 Total temperature versus cold mass fraction at different inlet pressures for h = 15 mm, L/D = 20 and conical angle (ø) = 30°

Figure 43 depicts the streamlines colored by flow velocity:

(a) Fixed thermophysical properties; (b) Variable thermo-
physical properties.
They obtained satisfactory results with their model
agreeing well with the experimental results considered for
validation. They reported that the variable thermophysicsl
properties have moderate effect on the numerical predic-
tions. They observed that an exergy efficiency of an actual
vortex tube ranging from 15 to 47 %. Figure 44 shows the
turbulent kinetic energy: (a).Fixed thermophysical proper-
Fig. 42 CFD results of static (a) and total temperatures (b) for set #1 ties; (b).Variable thermophysical properties. They
in Table 4: T0in = 291.48 K, Pin = 256.6 kPa, Pc = 113 kPa, concluded that regardless of the inlet pressure the exergy
Ph = 242 kPa efficiency curves showed the same trends, revealing the
need to focus more on the geometrical enhancements such
as convergent inlet channel to increase the flow velocity at
the inlet and to enhance the cooling effect and energy
separation.
Berber et al. (2013) [8] incorporated Rule-Based
Mamdani-Type Fuzzy (RBMTF) modeling technique and
experiments to study the cooling and heating performances
of a counter flow Ranquee–Hilsch vortex tubes (RHVT).
They observed increase in cold mass fraction due to
increase in mass flow rate of cold stream and also by
decrease in hot stream mass flow rate. At cold mass frac-
Fig. 43 Streamlines colored by flow velocity: a fixed thermophysical
tion = 0.9 the mass flow rate of cold stream was maximum
properties; b variable thermophysical properties
but the hot stream mass flow rate was minimum resulting in
maximum heating performance and minimum cooling
Ouadha et al. [35] conducted numerical study to inves- performance. They stated that an increase in heating per-
tigate the effect of fluid properties on energy separation, formance was observed with increase in fraction of cold
flow phenomena and the exergy efficiency using RSM flow. They reported that the level of fraction of cold flow
turbulence model and considering variable thermophysical significantly affects the heating performance of the RHVT.
properties. They stated that the three dimensional compu- Figure 45 shows the fuzzy membership functions for two
tations with constant and variable properties revealed that output variables: (a) ΔTh fuzzy sets graphics; (b) ΔTc fuzzy
constant thermo dynamical assumptions doesn’t signifi- sets graphics. They observed that maximum heating per-
cantly affecting the performance of the vortex tube. formance of the RHVT at 41.7 °C and maximum cooling

123
 T. Karthikeya Sharma et al.

Fig. 44 Turbulent kinetic

energy: a fixed thermophysical
properties; b variable
thermophysical properties

Fig. 47 Effect of the number of convergent nozzle on the cold

temperature difference
Fig. 45 Fuzzy membership functions for two output variables: a ΔTh
fuzzy sets graphics; b ΔTc fuzzy sets graphics
effect of convergence ratio, pressure inlet and number of
nozzle intake on the performance of vortex tube. They used
FLUENT 6.3 with k–ɛ turbulence model to solve flow
equations 3D-CFD model results were validated with
experimental results. Their analysis results revealed that
the convergence ratio (K) should be small and not more
than 1.9. Figure 46 shows the contours of total temperature
of the optimum model. They reported that 35.5 % reduction
in cold temperatures was obtained when the convergence
ratio was 1.9 and maximum cold temperature was obtained
when compared with the straight one. They observed that
the influence of nozzle inlets depends on the cold mass
fraction. They reported that sensitivity of cold drop tem-
perature was observed significantly improved with increase
Fig. 46 Contours of total temperature of the optimum model in convergent nozzles number and highest cold drop tem-
perature was obtained at lower cold mass fractions.
Figure 47 depicts the effect of the number of convergent
performance was observed at −31.1 °C. They concluded nozzle on the cold temperature difference They stated that
that RBMTF can be successfully used for the analysis of maximum cold temperature difference of 27.86 and
heating and cooling performances of counter-flow Ran- 26.52 K were obtained with convergence ratio of K = 2.22
quee–Hilsch vortex tubes with different geometric when the analysis was done with both experimentally and
constructions. numerically. An increase of 37.4 % in vortex tube perfor-
Rafiee and Rahimi [40] investigated experimentally and mance was observed with optimization. They concluded
by using three dimensional fluid dynamic analysis on the that computational analysis of optimizing vortex tube

123
Numerical Analysis of a Vortex Tube: A Review

parameters consumes less time and cost effective and CFD

analysis can be used as an effective tool for designing and
optimizing vortex tubes.
Bej and Sinhamahapatra [7] investigated numerically a
cascade type RHVT designed to have the hot exit of the
first vortex tube as the inlet to the second vortex tube to
have high heating value. They considered k–ɛ turbulence
model for their exergy analysis on the second stage vortex
tube for different cold fractions. They opined that the
exergy analysis of the hot cascade vortex tube helps in
predicting the available energy as pressure and tempera-
ture. They stated that the loss of energy was more when
heat loss due to irreversibility occurs at high temperatures,
but this was not observed at low temperatures. They
reported that the exergy losses at hot exit are more at Fig. 49 Total hot exergy in second stage RHVT as a function of cold
higher cold fractions. Figure 48 shows the comparison of fraction
exergy efficiency of a single RHVT. They opined that the
qualities of availability of energy at cold and hot exits were k–ɛ turbulence model for the analysis of flow and tem-
highly affected by pressure drop, mass flow rate, and exit perature separation in the vortex tube. They evaluated the
temperature. Figure 49 depicts the total hot exergy in total tangential and axial velocity of flow for both straight
second stage RHVT as a function of cold fraction. They and curved vortex tubes. They stated that tangential
concluded that combined effect of mass flow rate, pressure velocity profile changes significantly after the stagnation
drop and exit temperature showed that hot exit gas per- point. They observed higher heat transfer rates in the initial
forms better at low cold fraction where as cold exit gas length of the vortex tube before the stagnation point. They
more efficient at higher cold fractions. Also, they didn’t reported that straight vortex tube delivered higher COP and
observe any secondary circulation in the second stage of isentropic efficiency when compared with curved vortex
RHVT revealing its optimal design for exergy analysis. tube under varying cold mass fractions, while at 0.3 cold
They observed increased turbulence because of the radial mass fractions maximum efficiency was observed. Fig-
expansion of air form wall to axis. They stated that cascade ure 50 shows the contours of total, tangential and axial
system offers efficient energy utilization and produces velocities for (a) straight and (b) curved vortex tube with
higher temperature separation when compared with con- the inlet pressure of 2 bar and Cold mass fraction (Yc)-
ventional RHVT. = 0.5. They observed two regimes f forced and free vortex
Bovand et al. [9, 10] numerically investigated the effect in the vortex tube and increase in thickness of cold core
of curvature on RHVT refrigerators varying the curvature increased efficiency. They concluded that they obtained
angles from 0° and 110°. They verified their three dimen- maximum COP of the vortex tube at 0.6 cold mass fraction
sional model with experimental results. They utilized RNG as a refrigerator. Figure 51 depicts the variation of COP for
straight and curved vortex tube versus cold mass ratio.
Also, the efficiency of straight vortex tube is higher than
curved vortex tube. They observed vortices mixing the hot
and cold zones leading to efficiency reduction in curved
vortex tubes.
Kumar et al. [27] optimized the geometrical and oper-
ating parameters of vortex tube using Taguchi method.
They considered inlet air pressure, hot tube length and hot
tube internal diameter, orifice diameter and nozzle diame-
ter effect on the energy separation and flow patters inside
the vortex tube. They effect of above said parameters were
analyzed using Taguchi’s parameter design approach by
considering three levels each and L-27 orthogonal array for
experimentation. Figure 52 shows the main Effects Plot for
Means (Response vs. Factor levels) They concluded that
the parameters they considered have significant effect on
Fig. 48 Comparison of exergy efficiency of a single RHVT the performance of the vortex tube. They optimized the

123
 T. Karthikeya Sharma et al.

Fig. 50 Contours of total, tangential and axial velocities for (a) straight and (b) curved vortex tube with the inlet pressure of 2 bar and Cold mass
fraction (Yc) = 0.5

varied the angel of curvature from 0° to 150° and simulated

the flow field and temperature separation using RNG k–ɛ
turbulence model. They evaluated the tangential and axial
components of velocity and flow pattern including for
secondary circulation and validated their model with
experimental results available in the literature. They
observed split in the flow into a two thermally different
featured co-axial helical streams. They observed higher
temperature for the flows near the wall and lower tem-
peratures for the flows near the axis. Figure 53 shows the
total temperature maps with streamline patterns superim-
posed for (a) straight, (b) 30° curved and (c) 60° curved
vortex tube for P = 2 bar. They reported that decrease in
Fig. 51 Variation of COP for straight and curved vortex tube versus energy transfer was observed with curvature increase as the
cold mass ratio some negative values of tangential velocity in some upper
regions of vortex tube were observed. They concluded that
parameters for obtaining hot outlet temperature, the opti- decrease in swirl velocity and diffusion process in the
mized parameters are L-180 mm; Dt-14 mm; Dn-17 mm; vortex tube. Also, they concluded that straight and 150˚
Do-8 mm; and P-3 Kgf/cm2. curved vortex tubes have higher performance as a refrig-
Bovand et al. [9, 10] incorporated a three dimensional erator than the other curvature angles. Figure 54 exhibits
CFD using RNG k–ɛ turbulence model to investigate the the coefficient of performance versus cold mass fractions
effect of vortex tube curvature on its performance. They for different vortex tubes at P = 2 bar.

123
Numerical Analysis of a Vortex Tube: A Review

Fig. 52 Main effects plot for

means (response vs. factor
levels)

Fig. 53 Total temperature maps with streamline patterns superimposed for (a) straight, (b) 30° curved and (c) 60° curved vortex tube for
P = 2 bar

Rafiee and Sadeghiazad [41] conducted experimental fraction (α) of about 0.24. They observed increase in cold
and numerical simulations on a truncated cone throttle temperature difference with increase in cold mass fraction.
valve to study the geometrical parameters affect on the The maximum improvement in hot temperature difference
energy separation in a vortex tube. They varied cone length was observed at a cold mass fraction of 0.86. They reported
of the throttle valve (φ = 0 to 10 mm), inlet pressure and that increase in nozzle number increases the sensitivity of
nozzle number (N = 2, 3, 4, and 6). They conducted 3 (ΔT) of the vortex tube to variation of parameters and cold
dimensional numerical investigations using ANSYS- mass fractions (α). Figure 55 shows the contours of total
FLUENT 6.3.26 model utilizing k–ɛ turbulence model to temperature of the optimum model. They concluded that
solve the equations. They stated that the performance of the maximum temperature gradient between hot and cold
vortex tube can be improved by using truncated cone output was obtained with nozzle number N = 6 at a cold
throttle valve. They obtained higher temperature gradient mass fraction of (α) \0.5, while the minimum was
between the cold and hot output as (ΔT) = 20.21 K for a observed with N = 2 and the effect was reversed with cold
truncated cone length of (φ) = 6 mm at a cold mass mass fraction (α) [0.5. Figure 56 exhibits the numerical

123
 T. Karthikeya Sharma et al.

validation with experimental results for different model of

ø. They concluded that the comparison between the 3D
CFD results and the measured experimental data yielded
promising results relative to the model’s capability to
predict the energy separation.
Liu and Liu [32] investigated the effect of flow mech-
anism and energy separation inside a vortex tube. They
incorporated a three dimensional computational model
along with experiments to study the flow separation in a
vortex tube. They stated that CFD tool can be used as a
powerful tool to optimize the parameters of the vortex tube
and also to have a clear insight of the swirling behavior.
Figure 57 shows the total temperature distributions along
the vortex tube length. Closer agreements of prediction of
RSM and k–ɛ models with measured values were observed
Fig. 54 Coefficient of performance versus cold mass fractions for by them when compared with the other models. They
different vortex tubes at P = 2 bar suggested k–ɛ model to be best suitable for the numerical
analysis in terms of cost of computation. Figure 58 Depicts
the velocity patterns at the axial location z/L = 0.01. They
observed a significant of cold outlet diameter on the tem-
perature difference between cold and hot fluid outlets. They
concluded that the interactions of angular momentum
transaction and kinetic energy transfer were found to be
responsible for the temperature difference between cold
and hot fluids.
Mohammadi and Farhadi [34] conducted two dimen-
sional numerical experimental analysis of brass vortex tube
to study the performances of separation of a vortex tube for
a hydrocarbon mixture. They examined the effect of nozzle
number and cold mass fraction on gas species separation at
specific inlet pressure 236.37 kPa in a vortex tube with
LPG as hydrocarbon mixture and LPG with Nitrogen as
another mixture. Figure 59 shows the contour plot of pre-
dicted temperature for vortex tube at cold fraction 0.4.
They investigated the highly rotating field structure char-
Fig. 55 Contours of total temperature of the optimum model acteristics under different inlet pressures and different cold

Fig. 56 Numerical validation

with experimental results for
different model of ø

123
Numerical Analysis of a Vortex Tube: A Review

Fig. 57 Total temperature

distributions along the vortex
tube length

Fig. 58 Velocity patterns at the axial location z/L = 0.01

Fig. 59 Contour plot of

predicted temperature for vortex
tube at cold fraction 0.4

fractions. They calculated the numerical results of turbu- effect of inlet pressures (up to 850 kPa) on vortex tube
lent and compressible flows using k–ɛ turbulence model. separation performance. Figure 60 exhibits the CFD results
Finally the results are extrapolated for investigating the of cold and hot outlet molar recovery percent of LPG

123
 T. Karthikeya Sharma et al.

Fig. 62 Chart of isentropic energy efficiency coefficient of the vortex

Fig. 60 CFD results of cold and hot outlet molar recovery percent of tube versus cold mass flow fraction
LPG versus different cold fractions (inlet pressure = 236.37 kPa,
number of nozzles = 4)

Fig. 61 Computational domain of the vortex tube

Fig. 63 Axial velocity profile of different compound gases at axial

versus different cold fractions (inlet pressure = 236.37 kPa, distance x = 25.4 mm
number of nozzles = 4).
They observed nozzle number four giving highest sep-
aration performance for both mixtures considered (about They reported a drop in static and total temperatures of
65 % for C4+ in LPG as mixture of hydrocarbon com- the swirl core dropped and increase in outer swirl tem-
ponents and 79 % for LPG in LPG–N2 mixture). They peratures were observed with modified model when
reported that cold mass fraction 0.76 was observed as compared with the conventional model. They stated that
optimum cold mass fraction giving highest gas separation the swirl core static temperature was observed to be low
for both the mixtures (about 78 % for C4+ in LPG and when compared with outer part well agreeing with the
87 % for LPG in LPG–N2 mixture). experimental data was obtained by the modified model.
Khait et al. [24] conducted a three dimensional numer- They finally concluded that the computational power
ical model of a swirled compressible flow in a Ranquee– requirement of standard k–ɛ turbulence model was less
Hilsch vortex tube. Figure 61 shows the computational when compared with the other LES and RSM models.
domain of the vortex tube. They proposed some modifi- Thakare and Parekh [50] conducted CFD study to
cations to the energy conservation equation of Reynolds investigate the effect of thermo physical parameters like
Averaged Naviere Stokes (RANS) model. They incorpo- specific gas constant and thermal conductivity of different
rated additional heat transfer mechanism caused by gas gases, Prandtl number, and turbulence models on the per-
compressibility which is not applied in standard RANS formance of vortex tube. They considered eight different
model. They conducted their investigations with both gasses as working fluids for the analysis. They incorporated
standard and proposed energy conservation equations using different turbulent models, namely, one equation Spalart–
k–ɛ turbulence model. They stated that the application of Allmaras, two equations Standard k–ɛ and Standard k–ɷ
the modified energy conservation equation was found to model tested by Second Order Upwind (SOU) and Quad-
lead to re-distribution of temperatures along the computa- ratic Upstream Interpolation for Convective Kinetics
tional domain. Figure 62 exhibits the chart of isentropic (QUICK) scheme to understand the complex nature of
energy efficiency coefficient of the vortex tube versus cold highly compressible, turbulent and swirling flow within the
mass flow fraction. vortex tube. It was reported that all the models identified

123
Numerical Analysis of a Vortex Tube: A Review

turbulence models. They concluded that LES model pre-

dictions of simulated air flow were random and not precise,
like k–ε turbulence model which exhibited smooth hydro-
dynamic and thermodynamic parameters. The other models
show intermediate results between LES and k–ε turbulence
model. They stated that the radial velocity profiles of LES
predictions were higher than that of other models repre-
senting enhanced turbulent and heat and mass transfer.
Figure 65 represents the static pressure contour plots in the
cross sections of the vortex tube predicted using LES n.14
at different time: (a) t ¼ 0.011 s; (b) t ¼ 0.016 s. They
Fig. 64 Radial profile of axial velocity of air predicted by different concluded that for the simulation of Ranque–Hilsch dou-
turbulence models at various axial locations at x = 152.4 mm ble-circuit vortex tube LES model results are highly
accurate whereas, RSM-LRR and SAS- SST models are
the re-circulating secondary flow. Figure 63 Shows the good a good compromise between computational effi-
axial velocity profile of different Compound gases at axial ciency and accuracy.
distance x = 25.4 mm. They reported that increase in Kandil and Abdelghany [23] investigated computation-
energy separation was observed with increase in thermal ally the different effects on the performance of the
diffusivity and thermal conductivity of a gas, except for H2 Ranquee–Hilsch vortex tube. They considered three
gas. Also, it was reported that magnitude of energy sepa- dimensional axisymmetrical model using ANSYS FLU-
ration was decreased with specific gas constant value. They ENT CFD tool to investigate the effect of tube length to
concluded that energy separation was not influenced by diameter ratio (L/D) and the cold orifice size on the per-
Prandtl number. Figure 64 depicts the radial profile of axial formance the vortex tube. They stated that L/D ratio has
velocity of air predicted by different turbulence models at considerable effect on the performance of the vortex tube
various axial locations at x = 152.4 mm. The energy and it changes with change in cold mass fraction. They
separation predictions of Spalart–Allmaras model were reported that lowest cold orifice diameter to tube diameter
observed to be better when compared with Standard k–ɛ (dc/D) delivered the maximum cooling at lowest cold mass
and Standard k–ɷ models in the cold region. Finally, it was fractions (μc), while highest cold mass fraction and highest
reported that the mechanism of energy separation is inde- cold orifice diameter to tube diameter ratio delivered
pendent of turbulence model used. maximum heating. Figure 66 shows the velocity vectors at
Alekhin et al. [2] investigated to know the best suit- the cold outlet for dc/D = 0.544 at cold mass fraction of
able turbulence model to for predicting the aspects of a 0.3912. They observed secondary circulations when the
double-circuit RHVT. They considered k–ε, k–ε RNG, vortex tube was operated at lower cold orifice diameter to
SAS-SST, RSM-LRR and LES models for their analysis. tube diameter (dc/D) when operating at lower cold mass
They conducted a grid independent study for all the fractions and no prominent effect on the performance of the

Fig. 65 Static pressure contour plots in the cross sections of the vortex tube predicted using LES n. 14 at different time: a t ¼ 0.011 s; b t ¼
0.016 s

123
 T. Karthikeya Sharma et al.

Fig. 66 Velocity vectors at the cold outlet for dc/D = 0.544 at cold
Fig. 68 Variation of cold end temperature separation and with cold
mass fraction of 0.3912
mass fraction, predicted by different turbulence models

vortex tube was observed. They incorporated tube fins to

enhance the natural wall convection. Figure 67 shows the
total temperature contour of the RHVT with (a) adiabatic
wall and (b) circular fins.
Thakare and Parekh [51] examined the performance of
vortex tube and energy separation using different gasses at
varying cold mass fractions and using different turbulence
models. The turbulence models considered for the analysis
are one equation Spalart–Allmaras model along with
Standard k–ɷ model, Standard k–ε model and as well as
RSM (Reynolds Stress Model) and a comparison of the
results of all the turbulence models were made. They stated
that the magnitude of flow physics with different turbu-
lence models was different but the trends were observed to
be similar. They reported that RSM model under predicted
the magnitude of temperature separation where as Spalart– Fig. 69 Variation of cooling power separation with cold mass
Allmaras model over predicted the temperature separation. fraction for different gases

Fig. 67 Total temperature

contour of the RHVT with
(a) adiabatic wall and
(b) circular fins

123
Numerical Analysis of a Vortex Tube: A Review

Figure 68 depicts the variation of cold end temperature operating conditions and geometries, thereby providing a
separation and with cold mass fraction, predicted by dif- powerful tool that can be employed to optimize vortex tube
ferent turbulence models. They observed fluctuations in the design as well as assess its utility in the field of new
results predicted by Standard k–ɷ model where as con- applications and industries.
sistency was observed in the results predicted by Spalart–
Allmaras model and standard k–ε model over a range of
varying cold mass fractions. They concluded that maxi-
mum cold end and hot end temperature separation was References
observed for N2 whereas least for CO2 and closer results
1. Ahlborn B, Keller JU, Staudt R, Treitz G, Rebhan E (1994)
were obtained for N2, O2, and air. Figure 69 shows the Limits of temperature separation in a vortex tube. J Phys D Appl
variation of cooling power separation with cold mass Phys 27(3):480
fraction for different gases. Also, minimum cold end 2. Alekhin V, Bianco V, Khait A, Noskov A (2015) Numerical
temperatures were obtained with all the fuels when the cold investigation of a double-circuit Ranque–Hilsch vortex tube. Int J
Therm Sci 89:272–282
mass fraction was varied from 0.3 to 0.35, and maximum 3. Aljuwayhel NF, Nellis GF, Klein SA (2005) Parametric and
cooling power was observed for a cold mass fraction of internal study of the vortex tube using a CFD model. Int J Refrig
0.68. 28(3):442–450
4. Baghdad M, Ouadha A, Imine O, Addad Y (2011) Numerical
study of energy separation in a vortex tube with different RANS
models. Int J Therm Sci 50(12):2377–2385
4 Conclusion 5. Behera U, Paul PJ, Dinesh K, Jacob S (2008) Numerical inves-
tigations on flow behaviour and energy separation in Ranque–
Vortex tube is considered to be the simplest device having Hilsch vortex tube. Int J Heat Mass Transf 51(25):6077–6089
6. Behera U, Paul PJ, Kasthurirengan S, Karunanithi R, Ram SN,
applications in many industrial and laboratory applications. Dinesh K, Jacob S (2005) CFD analysis and experimental
The simplicity in usage, light and compactness, easy investigations towards optimizing the parameters of Ranque–
maintenance, and no moving parts made the vortex tube to Hilsch vortex tube. Int J Heat Mass Transf 48(10):1961–1973
adopt it in several applications where size and weight and 7. Bej N, Sinhamahapatra KP (2014) Exergy analysis of a hot
cascade type Ranque–Hilsch vortex tube using turbulence model.
simplicity matters. Simultaneous delivery of cold and hot Int J Refrig 45:13–24
air/gas streams makes vortex tube versatile device to use it 8. Berber A, Dincer K, Yılmaz Y, Ozen DN (2013) Rule-based
for both hot and cold air/gas needs. The mechanism of Mamdani-type fuzzy modeling of heating and cooling perfor-
energy separation is very complex in nature and involves mances of counter-flow Ranque–Hilsch vortex tubes with
different geometric construction for steel. Energy 51:297–304
high rates of turbulence. No accurate explanation was 9. Bovand M, Valipour MS, Dincer K, Tamayol A (2014) Numer-
available dealing with the energy separation in vortex tube. ical analysis of the curvature effects on Ranque–Hilsch vortex
To understand the concept of energy separation researchers tube refrigerators. Appl Therm Eng 65(1):176–183
conducted numerical and CFD analysis using different 10. Bovand M, Valipour MS, Eiamsa-ard S, Tamayol A (2014)
Numerical analysis for curved vortex tube optimization. Int
CFD tools having various turbulent models like LES, k–ɛ, Commun Heat Mass Transf 50:98–107
k–ɷ and RMS. Researchers reported that all the models 11. Bruun HH (1969) Experimental investigation of the energy sep-
predicted the results with same profile but variation in aration in vortex tubes. J Mech Eng Sci 11(6):567–582
magnitude. Many researchers observed that standard k–ɛ 12. Dutta T, Sinhamahapatra KP, Bandyopadhyay SS (2011)
Numerical investigation of gas species and energy separation in
model results are in good agreement with the experimental the Ranque–Hilsch vortex tube using real gas model. Int J Refrig
results. Some researchers incorporated ANN to predict the 34(8):2118–2128
optimum parameters of the vortex tube which will give 13. Dutta T, Sinhamahapatra KP, Bandyopdhyay SS (2010) Com-
maximum energy separation and cold and hot temperature parison of different turbulence models in predicting the
temperature separation in a Ranque–Hilsch vortex tube. Int J
difference. Researchers opine that parameters like nozzle Refrig 33(4):783–792
number, nozzle profile and diameter of the vortex tube, 14. Eiamsa-ard S, Promvonge P (2007) Numerical investigation of
length and cold mass fraction plays a major role in getting the thermal separation in a Ranque–Hilsch vortex tube. Int J Heat
the best performance from the vortex tube. Some Mass Transf 50(5):821–832
15. Eiamsa-ard S, Promvonge P (2008) Numerical simulation of flow
researchers tried with different shapes of vortex tubes like field and temperature separation in a vortex tube. Int Commun
conical, cascading type and curved vortex tubes and sug- Heat Mass Transf 35(8):937–947
gested that conical vortex tube can reduce the length of the 16. Entov VM, Kalashnikov VN, Raiskii YD (1967) Parameters
vortex tube, normal vortex tube is better than the curved which determine the vortex effect. Fluid Dyn 2(3):18–22
17. Escudier MP, Keller J (1985) Recirculation in swirling flow-a
vortex tube and cascading of the vortex tubes will improve manifestation of vortex breakdown. AIAA J 23(1):111–116
the exergy of the total system. Researchers concluded that 18. Farouk T, Farouk B (2007) Large eddy simulations of the flow
the successful use of CFD in order to develop a design tool field and temperature separation in the Ranque–Hilsch vortex
that can be utilized with confidence over a range of tube. Int J Heat Mass Transf 50(23):4724–4735

123
 T. Karthikeya Sharma et al.

19. Farouk T, Farouk B, Gutsol A (2009) Simulation of gas species 36. Parulekar BB (1961) The short vortex tube. J Refrig 4(4):74–80
and temperature separation in the counter-flow Ranque–Hilsch 37. Pinar AM, Uluer O, Kırmaci V (2009) Optimization of counter
vortex tube using the large eddy simulation technique. Int J Heat flow Ranque–Hilsch vortex tube performance using Taguchi
Mass Transf 52(13):3320–3333 method. Int J Refrig 32(6):1487–1494
20. Fröhlingsdorf W, Unger H (1999) Numerical investigations of the 38. Pouraria H, Zangooee MR (2012) Numerical investigation of
compressible flow and the energy separation in the Ranque– vortex tube refrigerator with a divergent hot tube. Energy Pro-
Hilsch vortex tube. Int J Heat Mass Transf 42(3):415–422 cedia 14:1554–1559
21. Hardebol, J. (1960). U.S. Patent No. 2,955,432. Washington, DC: 39. Pourmahmoud N, Hassanzadeh A, Moutaby O (2012) Numerical
U.S. Patent and Trademark Office analysis of the effect of helical nozzles gap on the cooling
22. Hartnett JP, Eckert ERG (1957) Experimental study of the capacity of Ranque–Hilsch vortex tube. Int J Refrig 35(5):1473–
velocity and temperature distribution in a high-velocity vortex- 1483
type flow. Trans ASME 79(4):751–758 40. Rafiee SE, Rahimi M (2013) Experimental study and three-di-
23. Kandil HA, Abdelghany ST (2015) Computational investigation mensional (3D) computational fluid dynamics (CFD) analysis on
of different effects on the performance of the Ranque–Hilsch the effect of the convergence ratio, pressure inlet and number of
vortex tube. Energy 84:207–218 nozzle intake on vortex tube performance-validation and CFD
24. Khait AV, Noskov AS, Lovtsov AV, Alekhin VN (2014) Semi- optimization. Energy 63:195–204
empirical turbulence model for numerical simulation of swirled 41. Rafiee SE, Sadeghiazad MM (2014) Three-dimensional and
compressible flows observed in Ranque–Hilsch vortex tube. Int J experimental investigation on the effect of cone length of throttle
Refrig 48:132–141 valve on thermal performance of a vortex tube using k–ɛ turbu-
25. Kocabas F, Korkmaz M, Sorgucu U, Donmez S (2010) Modeling lence model. Appl Therm Eng 66(1):65–74
of heating and cooling performance of counter flow type vortex 42. Saidi MH, Yazdi MA (1999) Exergy model of a vortex tube
tube by using artificial neural network. Int J Refrig 33(5):963– system with experimental results. Energy 24(7):625–632
972 43. Secchiaroli A, Ricci R, Montelpare S, D’Alessandro V (2009)
26. Korkmaz ME, Gümüşel L, Markal B (2012) Using artificial Numerical simulation of turbulent flow in a Ranque–Hilsch
neural network for predicting performance of the Ranque–Hilsch vortex tube. Int J Heat Mass Transf 52(23):5496–5511
vortex tube. Int J Refrig 35(6):1690–1696 44. Shamsoddini R, Khorasani AF (2012) A new approach to study
27. Kumar GS, Padmanabhan G, Sarma BD (2014) Optimizing the and optimize cooling performance of a Ranque–Hilsch vortex
temperature of hot outlet air of vortex tube using Taguchi tube. Int J Refrig 35(8):2339–2348
method. Procedia Eng 97:828–836 45. Shamsoddini R, Nezhad AH (2010) Numerical analysis of the
28. Kurosaka M (1982) Acoustic streaming in swirling flow and the effects of nozzles number on the flow and power of cooling of a
Ranque—Hilsch (vortex-tube) effect. J Fluid Mech 124:139–172 vortex tube. Int J Refrig 33(4):774–782
29. Lasheras JC, Choi H (1988) Three-dimensional instability of a 46. Simões-Moreira JR (2010) An air-standard cycle and a thermo-
plane free shear layer: an experimental study of the formation and dynamic perspective on operational limits of Ranque–Hilsh or
evolution of streamwise vortices. J Fluid Mech 189:53–86 vortex tubes. Int J Refrig 33(4):765–773
30. Leibovich S (1984) Vortex stability and breakdown-survey and 47. Skye HM, Nellis GF, Klein SA (2006) Comparison of CFD
extension. AIAA J 22(9):1192–1206 analysis to empirical data in a commercial vortex tube. Int J
31. Lewins J, Bejan A (1999) Vortex tube optimization theory. Refrig 29(1):71–80
Energy 24(11):931–943 48. Stephan K, Lin S, Durst M, Huang F, Seher D (1984) A similarity
32. Liu X, Liu Z (2014) Investigation of the energy separation effect relation for energy separation in a vortex tube. Int J Heat Mass
and flow mechanism inside a vortex tube. Appl Therm Eng 67 Transf 27(6):911–920
(1):494–506 49. Takahama H, Yokosawa H (1981) Energy separation in vortex
33. Mihal CP (1981) Effect of heat stress on physiological factors for tubes with a divergent chamber. J Heat Transfer 103(2):196–203
industrial workers performing routine work and wearing imper- 50. Thakare HR, Parekh AD (2014) CFD analysis of energy sepa-
meable vapor-barrier clothing. Am Ind Hyg Assoc J 42(2):97– ration of vortex tube employing different gases, turbulence
103 models and discretisation schemes. Int J Heat Mass Transf
34 Mohammadi S, Farhadi F (2014) Experimental and numerical 78:360–370
study of the gas–gas separation efficiency in a Ranque–Hilsch 51. Thakare HR, Parekh AD (2015) Computational analysis of
vortex tube. Sep Purif Technol 138:177–185 energy separation in counter: flow vortex tube. Energy 85:62–77
35. Ouadha A, Baghdad M, Addad Y (2013) Effects of variable 52. Uluer O, Kırmacı V, Ataş Ş (2009) Using the artificial neural
thermophysical properties on flow and energy separation in a network model for modeling the performance of the counter flow
vortex tube. Int J Refrig 36(8):2426–2437 vortex tube. Expert Syst Appl 36(10):12256–12263

123

View publication stats