Applied Thermal Engineering 30 (2010) 2505e2511

Contents lists available at ScienceDirect

Applied Thermal Engineering

journal homepage: www.elsevier.com/locate/apthermeng

A new vortex generator geometry for a counter-flow RanqueeHilsch vortex tube

Orhan Aydın*, Burak Markal, Mete Avcı
Karadeniz Technical University, 61080 Trabzon, Turkey

a r t i c l e i n f o a b s t r a c t

Article history: In this study, a new geometry is introduced for the cold end side (i.e. where the swirl flow is introduced
Received 30 March 2010 into the tube), which is called ‘helical swirl flow generator’. Effect of the helical length of the swirl flow
Accepted 24 June 2010 generator on the performance of the vortex tube are investigated for varying values of other geometrical
parameters as a function of the cold mass fraction, yc. Finally, it’s disclosed that the effect of the helical
length on the performance changes critically according to the value of L/D.
Keywords:
Ó 2010 Elsevier Ltd. All rights reserved.
RanqueeHilsch vortex tube
Vortex tube
Energy separation
Helical swirl flow generator

1. Introduction detailed explanation about this phenomena belongs to Hilsch [6].

He claimed that the reason is the internal friction. The most
The vortex tube (also called RanqueeHilsch vortex tube) is common theory is the Fulton’s [7] one. In this theory, fluid enters
a simple mechanical device which splits a compressed gas stream the vortex tube tangentially from nozzles and is compressed
into cold and hot streams without any chemical reactions or straight to the tube walls by centrifugal forces. In this situation,
external energy supply [1], whose schematic is shown in Fig. 1. there exists pressure difference between center and tube walls.
Vortex tubes have advantages compared to other refrigerating Therefore, fluid particles move to the center radially. Because of the
or heating devices, being simple, having no moving parts, using no conservation of the angular momentum, velocity of particle in
electricity or chemicals and having long operation time. They need center becomes higher than the one near wall. Fluid particles near
only compressed gas to operate. Their critical disadvantage is their the center try to accelerate the outer ones. As a result, mechanical
low thermal efficiency. energy is transferred to the outer regions. Fluid temperature near
In the existing literature, there are many theoretical and the tube wall increases due to friction by transferring mechanical
experimental studies on vortex tubes. Experimental studies have energy from center. Contrast to this, inner flow become colder.
been generally focusing on increasing the performance of vortex Kurosaka [8] explained energy separation with acoustic streaming,
tubes. Different geometrical and thermo-physical parameters have while Stephan et al. [9] did it in view of Görtler vortex.
been tested to obtain an optimum vortex tube design. In theoretical There are also some studies on the performance enhancement
studies, aim is concentrated at determining velocity, pressure and of the vortex tubes. Wu et al. [4] designed a new nozzle to increase
temperature distribution. Recently, commercial CFD packages have efficiency. Dinçer et al. [10] studied effects of a mobile plug located
been used to obtain internal flow pattern. An excellent review of at the hot outlet. Xue and Arjomandi [11] examined effect of the
the studies existing in the literature can be found in a recent study vortex angle on efficiency. Nimbalkar and Muller [1] investigated
by Aydin and Baki [2] and a more recent study by Eiamsa and effects of various cold end geometries. Behera et al. [12] conducted
Promvenge [3]. experimental and CFD studies towards optimization of the Ran-
The mystery topic for the vortex tubes is the energy separation queeHilsch vortex tubes. Dincer et al. [13] used the analysis of the
effect. Although the vortex tubes have been known for decades, the artificial neural networks in modeling of some geometrical effects
mechanism producing temperature separation phenomenon as on the performance of a counter-flow RanqueeHilsch vortex tube.
a gas or vapor passes through a vortex tube hasn’t been fully Pinar et al. [14] used the Taguchi method in the performance
understood yet [4]. According to Ranque [5], the reason for energy analysis of a counter-flow RanqueeHilsch vortex tube. There are
separation is the adiabatic expansion and compression. The first also some numerical attempts to analyze performance of Ran-
queeHilsch vortex tube [15e21].
* Corresponding author. Tel.: þ90 462 377 29 74; fax: þ90 462 325 55 26. In a recent study [22], we developed a new geometry for the
E-mail address: oaydin@ktu.edu.tr (O. Aydın). cold end side of a counter-flow vortex tube. As an extension of that

1359-4311/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.applthermaleng.2010.06.024
2506 O. Aydın et al. / Applied Thermal Engineering 30 (2010) 2505e2511

Nomenclature yc Cold mass fraction

cp Specific heat at constant pressure (kJ kg1 K1) Greek letters

COP Coefficient of performance G a rate related to specific heat ratio, k-1/k
d Inner diameter of helical swirl flow generator (mm), din inner diameter of the inlet nozzle
Diameter of cold exit dhot inner diameter of the hot end nozzle
D Outer diameter of helical swirl flow generator, Inner his isentropic efficiency
diameter of tube (mm) f angle of the control valve
h Helical length (mm)
k Specific heat ratio, cp/cv Subscripts
L Vortex tube length (mm) a environment
m_ Mass flow rate (kg s 1) c cold stream
P Pressure (bar) CM cooling machine
Q_ Thermal power (kW) h hot stream
R Gas constant (kJ kg1 K1) HP heat pump
T Temperature  (C) i inlet
DT Temperature difference  (C) t total
Wm Mechanical energy (kW)

Fig. 1. The flow pattern in a counter-flow vortex tube.

2 2
P P 8

4
5 6
1 3
7
1. Compressor 12 7
13
2. Pressure gauge
3. Air tank
4. Valve 14 9
5. Dehumidifier
6. Pressure regulator P 10
7. Thermocouple 11 7
8. Rotameter
9. Conical valve
10. Tube 8
11. Digital manometer
12. Inlet nozzle
13. Hot end nozzle (hot exit)
14. Cold exit

Fig. 2. Schematic diagram of the experimental setup.

study, the aim of this study is focused on investigating the effect of

the helical length for varying geometrical parameters and working Table 1
conditions. Characteristics and uncertainties of the measurement instruments.

Instrument Range Uncertainty

2. Experimental study
Digital manometer HHP-2082 0e2 [bar] 0.15%
Digital thermometer (OM-2041) 200 to 1370 [ C] 0.1% (full scale)
The schematic diagram of the experimental setup used in the Volumetric flow meter CZ-32458-65 80e560 [LPM] 3% (full scale)
experiments is shown in Fig. 2. The experimental setup consists of Humidity meter HT-3006HA 10e95 [%R.H.] 3% (<70% R.H.)
a compressor, an air tank, a dehumidifier, a test section, valves and Barometer Testo 435-1 600e1150 [mbar] 3%
(Part no: 0632 1535)
joining components.
 O. Aydın et al. / Applied Thermal Engineering 30 (2010) 2505e2511 2507

Fig. 3. An exploded view of the vortex tube.

Fig. 4. Schematic diagram of the helical vortex generator (a) control valve (b).

The compressed working fluid supplied from the compressor (1) there is a digital manometer (11). The pressure of cold air leaving
passes through the air tank (3) and the dehumidifier (5). After the from the vortex tube is measured using this manometer. Volu-
dehumidifier, it passes through a pressure regulator (6) where its metric flow rate values are then calculated. The characteristics and
pressure is adjusted to desired level by reading from pressure gauge uncertainties of the measurement instruments are given in Table 1.
(2). Before the working fluid enters the test section (from inlet The experiments are performed in a tube (10) with inner diam-
nozzle, 12) tangentially, its temperature is measured with ther- eter of 10 mm made of plexiglass. Outer surface of the tube is
mocouples (7). By moving the conical valve (9), mass flow rate of covered with an insulating material to prevent heat loss to the
the working fluid in the cold (14) and hot exits (13) can be adjusted. environment. The inlet nozzle, the outlet nozzle and the helical swirl
Outside of the hot and cold exits, temperature of the working fluid flow generator are made of brass. The inlet, cold end and hot end
is measured by thermocouples which are located 1 cm beyond chambers are made of delrin. Various values of the lengthediameter
exits. Volumetric flow rate of fluid is gauged by rotameters (8) ratio (L/D) and the length of the helical swirl flow generators are
which are located outside of the both exits. Also, out of the cold exit, tested to investigate their interactive effects comparatively. Then
optimization is sought over these results. The effect of these
geometrical parameters is investigated in three different values of
70
the inlet pressure: 3, 4 and 5 bar (absolute). In the experiments, inner
Pi = 3bar h = 10mm
diameters of the inlet nozzle (din) and hot end nozzle (dhot) are 5 mm
60 Pi = 4bar L/D = 30
and kept constant. Diameter of the cold exit equals inner diameter of
Pi = 5bar φ = 30o helical swirl flow generator (d) and is 4 mm. An exploded view of
Temperature difference (oC)

50
vortex tube tested are shown in Fig. 3.
The parameters mentioned above are examined for varying the
40
gas inlet pressure and the cold mass fraction.
30
The cold mass fraction is defined as follows:

_c
m
20 yc ¼ (1)
_t
m
ΔTc
10
ΔTh where m _ c represents the mass flow rate of the cold stream released,
while m_ t represents the inlet or the total mass flow rate. For every
0 different configuration, the temperature values just before the inlet
0.0 0.2 0.4 0.6 0.8 1.0
nozzle and just after the outlet nozzles are measured for any value
Cold mass fraction, yc
of the cold mass fraction tested.
Fig. 5. Temperature differences DTh and DTc as functions of cold mass fraction, yc, for
The performance of the vortex tube is marked by cooling effect
different inlet pressures at h ¼ 10 mm, L/D ¼ 30. (DTc) and heating effect (DTh) [4], which are defined as follows:
2508 O. Aydın et al. / Applied Thermal Engineering 30 (2010) 2505e2511

Fig. 6. Temperature differences DTh and DTc as functions of cold mass fraction, yc, for different values of L/D in different pressures for h ¼ 10 mm (a) and h ¼ 30 mm (b).

of the helical vortex generator and the control valve are given in
DTc ¼ Ti  Tc (2) Fig. 4 a and b, respectively.
To calculate the cooling efficiency of the vortex tube, the prin-
DTh ¼ Th  Ti (3) ciple of adiabatic expansion of ideal gas is used. As the air flows into
the vortex tube, the expansion in isentropic process occurs [3]. The
Summing Eqs. (2) and (3), the total temperature difference
isentropic efficiency can be written as follows:
(DT ¼ DTc þ DTh) is obtained as in the following:

DT ¼ Th  Tc (4) Ti  Tc
his ¼   (5)

k1
Each of the helical swirl flow generators is designed in different Ti 1  ppai k
length but in the same helical pitch (one pitch). Schematic diagram
 O. Aydın et al. / Applied Thermal Engineering 30 (2010) 2505e2511 2509

Fig. 7. Temperature differences DTh and DTc as functions of cold mass fraction, yc, for different values of helical vortex generator length, h, at different pressures for L/D ¼ 30 (a) and
L/D ¼ 10 (b).

where, Pi, Pa and k are the inlet air pressure, the environment where, Q_ c is the thermal power in point of cooling side and can be
pressure and the specific heat ratio, respectively. defined as follows:
The vortex tube can be considered as both a cooling machine _
and a heat pump. The efficiency of a cooling machine can be Q_ c ¼ m
_ c cp ðTi  TÞc (7)
expressed in terms of coefficient of performance (COP) [23] and it is Wm is the required mechanical energy to supply cooling or heating.
defined as follows: Since compressed air has been used in this experiment, the input
power can be found as follows [11].
Q_ c  
COPCM ¼ (6) _m ¼ m P
_ m W _ i RTi ln i (8)
W Pc
2510 O. Aydın et al. / Applied Thermal Engineering 30 (2010) 2505e2511

60 Using Eq. (7) and Eq. (8) in Eq. (6) and rearranging the expres-
o
h = 10mm = 30 sion, Eq. (9) can be obtained:
h = 15mm L/D = 30
50 h = 20mm
Pi = 3bar 1 yc ðTi  Tc Þ
h = 25mm COPCM ¼   (9)
h = 30mm G T ln Pi
i Pc
40

where, k is the specific heats ratio and G ¼ k  1=k.

(%)

30 Similarly, as a heat pump coefficient of performance can be

is

expressed as follows:

20 1 ð1  yc ÞðTh  Ti Þ
COPHP ¼   (10)
G Ti ln PPci
10

3. Results and discussion

0
0.0 0.2 0.4 0.6 0.8 1.0
Pressure and temperature of the input air, temperature of the
Cold mass fraction, yc
exhausting air flow at both the hot and cold ends have been
Fig. 8. Isentropic efficiency versus cold mass fraction for different helical vortex measured. The cold mass fraction and the temperature differences
generator lengths. have been calculated according to values measured.
The experimental temperature differences for the hot stream
and cold stream as a function of cold mass fraction, yc, at different
inlet pressures for a set of geometrical parameters are shown in
Fig. 5. Temperature differences are getting increased when inlet
pressure is increased. But the rate of increase slows down. This
situation can be explained by the chocking of the flow. When the
inlet pressure is increased, flow velocity outside of the inlet nozzle
increases. After the chocking takes place, velocity and mass flow
rate outside of the inlet nozzle don’t increase anymore, even
though inlet pressure increases.
Fig. 6a and b show the temperature difference for the hot stream
and cold stream as a function of the cold mass fraction, yc, at various
values of the inlet pressure and L/D at h ¼ 10 mm and h ¼ 30 mm,
respectively. The effect of L/D is influenced by the length of the
helical swirl flow generator, h and the inlet pressure, pi. For
instance, for h ¼ 10 mm, L/D ¼ 10 presents a worse performance
compared to the L/D ¼ 20, 30 and 40 while its performance
becomes comparable for h ¼ 30 mm.
Fig. 7a and b show the temperature difference as a function of yc
for various values of the inlet pressure and the helical swirl flow
generator lengths, h, at L/D ¼ 30 and L/D ¼ 10. From Fig. 7a and b, it
can be seen that helical length has an obvious impact on energy
separation. But, like the results obtained from Fig. 6, its effect
changes according to the tube length. From Fig. 7a, it is seen that
temperature differences are getting higher when helical length is
getting shortened for high L/D ratios. For example, for L/D ¼ 30, at 4
and 5 bar, h ¼ 10 mm is the best length while h ¼ 30 mm is the worst
one. On the other hand, Fig. 7b shows that at a low L/D, shorter helical
lengths, like h ¼ 10 and 15 mm, present a sudden breakdown in
efficiency. For this situation, h ¼ 10 mm becomes the worst value.
These results are related to the energy of the working fluid.
There is probably a critical energy for the vortex tube. Higher
kinetic energy means higher velocity at the same time. Using the
helical swirl flow generator increases efficiency because of reducing
sudden change of flow direction. Also it helps to reduce the
generation of eddies which cause energy loss. It supplies a smooth
transition from an inlet straight pipe to a circular pipe. Long helical
length increases friction loses and, in follows, decreases tempera-
ture difference. That’s why there is a great and obvious difference
between short and long helical lengths. Short ones are better for
h ¼ 10 mm. But, near hot end region, flow is desired to have a low
peripheral velocity. In short tube lengths, if flow has much energy,
turbulence effects will increase and this will cause hot and cold
Fig. 9. Coefficients of performance versus cold mass fraction with different helical flow to mix in the tube. Also flow has high peripheral velocity near
vortex generator lengths for COPCM (a) and COPHP (b). far end of the tube. As seen from the graphs, this situation is much
 O. Aydın et al. / Applied Thermal Engineering 30 (2010) 2505e2511 2511

clearer for the hot end side (see Fig. 6. Even though flow has high [2] O. Aydin, M. Baki, An experimental study on the design parameters of
a counter-flow vortex tube, Energy 31 (2006) 2763e2772.
energy, if a longer tube is used, flow can complete its motion, effects
[3] S. Eiamsa, P. Promvonge, Review of RanqueeHilsch effects in vortex tubes,
of turbulence decrease and flow will have lower velocity at the hot Renewable & Sustainable Energy Review 12 (2008) 1822e1842.
end because of increasing wall friction (owing to long tube). At [4] Y.T. Wu, Y. Ding, Y.B. Ji, C.F. Ma, M.C. Ge, Modification and experimental
a low value of the helical length, if the inlet pressure increases, the research on vortex tube, International Journal of Refrigeration 30 (2007)
1042e1049.
critical L/D will increase. Then, the best configuration is obtained at [5] M.G. Ranque, Experiences sur la delente avec production simultanees dun
h ¼ 10 mm and L/D ¼ 30 at Pi ¼ 5 bar (abs.) for both the ends. echappement dair chaud et dun echappement dair froid, Journal de Physique
According to the experimental results, it can be said that if the et de Radium 7 (1933) 112e115.
[6] R. Hilsch, The use of the expansion of gases in a centrifugal fields as cooling
pressure was increased, for example to 7,8. bar etc., L/D ¼ 40 could process, Review of Scientific Instruments 18 (1947) 108e113.
be the best ratio at h ¼ 10 mm. [7] C.D. Fulton, Ranque’s tube, Journal of American Society of Refrigerating
Fig. 8 shows the isentropic efficiency versus cold mass fraction at Engineers 58 (1950) 473e479.
[8] M. Kurosaka, Acoustic streaming in swirling flow and the Rangue-Hilsch
different values of the helical vortex generator length. As seen from (Vortex-Tube) effect, Journal of Fluid Mechanics 124 (1982) 139e172.
the figure, especially up to yc z 0.5, isentropic efficiency increases [9] K. Stephan, S. Lin, M. Durst, F. Huang, D. Seher, An investigation of energy
with a decrease in the helical length. Between yc z 0.5 O 0.9 h ¼ 10 separation in a vortex tube, International Journal of Heat Mass Transfer 26
(1983) 341e348.
and 15 mm show the same trend. But for the whole yc values, there is [10] K. Dincer, S. Baskaya, B.Z. Uysal, I. Ucgul, Experimental investigation of the
a great difference between h ¼ 10 and 30 mm. performance of a RanqueeHilcsh vortex tube with regard to a plug located at
Fig. 9a and b shows the coefficients of performance versus cold the hot outlet, International Journal of Refrigeration 32 (2009) 87e94.
[11] Y. Xue, M. Arjomandi, The effect of vortex angle on the efficiency of the
mass fraction at different helical vortex generator lengths for vortex
RanqueeHilcsh vortex tube, Experimental Thermal and Fluid Science 33
tubes as a cooling machine and a heat pump, respectively. Coeffi- (2008) 54e57.
cient of performance increases with decreasing helical length. [12] U. Behera, P.J. Paul, S. Kasthurirengan, R. Karunanithi, S.N. Ram, K. Dinesh,
COPCM reaches its maximum value at nearly yc ¼ 0.8 and COPHP S. Jacob, CFD analysis and experimental investigations towards optimizing the
parameters of RanqueeHilsch vortex tube, International Journal of Heat Mass
reaches its maximum value at nearly yc ¼ 0.7. Transfer 48 (2005) 1961e1973.
[13] K. Dincer, S. Tasdemir, S. Baskaya, B.Z. Uysal, Modeling of the effects length to
4. Concluding remarks diameter ratio and nozzle number on the performance of counter flow Ran-
queeHilsch vortex tubes using artificial neural networks, Applied Thermal
Engineering 28 (2008) 2380e2390.
A series of experiments are conducted to investigate some [14] A.M. Pinar, O. Uluer, V. Kirmaci, Optimization of counter flow RanqueeHilsch
design features of the counter-flow vortex tubes at different pres- vortex tube performance using Taguchi method, International Journal of
sures. An innovative and new design of vortex tube was proposed, Refrigeration 32 (2009) 1487e1494.
[15] U. Behera, P.J. Paul, K. Dinesh, S. Jacob, Numerical investigations on flow
in which the swirl flow was enters into the vortex tube through behavior and energy separation in RanqueeHilsch vortex tube, International
a helical swirl flow generator. The results show that this new design Journal of Heat Mass Transfer 51 (2008) 6077e6089.
has an obvious and superior effect on temperature separation. The [16] H.M. Skye, G.F. Nellis, S.A. Klein, Comparison of CFD analysis to empirical data
in a commercial vortex tube, International Journal of Refrigeration 29 (2006)
effect of the helical swirl length was tested for various values of L/D
71e80.
and the inlet pressure. From the results it is seen that at a high L/D, [17] S. Eiamsa-ard, P. Promvonge, Numerical investigation of the thermal separa-
short helical generators are better. The best configuration is tion in a RanqueeHilsch vortex tube, International Journal of Heat Mass
obtained when h ¼ 10 mm, L/D ¼ 30 and Pi ¼ 5 bar (abs.). In this Transfer 50 (2007) 821e832.
[18] T. Farouk, B. Farouk, Large eddy simulations of the flow field and temperature
configuration, maximum temperature differences are: DT ¼ 83.1  C, separation in the RanqueeHilsch vortex tube, International Journal of Heat
DTc ¼ 45.9  C, DTh ¼ 61.5  C (Tc ¼ 36.3  C and Th ¼ 71.5  C). Mass Transfer 50 (2007) 4274e4735.
[19] S. Eiamsa-ard, P. Promvonge, Numerical simulation of flow field and
Acknowledgements temperature separation in a vortex tube, International Communications Heat
Mass Transfer 35 (2008) 937e947.
[20] N.F. Aljuwayhel, G.F. Nellis, S.A. Klein, Parametric and internal study of the
The first author of this article is also indebted to the Turkish vortex tube using a CFD model, International Journal of Refrigeration 28
Academy of Sciences (TUBA) for the financial support provided under (2005) 442e450.
[21] A. Secchiaroli, R. Ricci, S. Montelpare, V. D’Alessandro, Numerical simulation
the Programme to Reward Success Young Scientists (TUBA-GEBIT). of turbulent flow in a RanqueeHilsch vortex tube, International Journal of
Heat Mass Transfer 52 (2009) 5496e5511.
References [22] B. Markal, O. Aydin, M. Avci, An experimental study on the effect of the valve
angle of counter-flow RanqueeHilsch vortex tubes on thermal energy sepa-
[1] S.U. Nimbalkar, M.R. Muller, An experimental investigation of theoptimum ration, Experimental Thermal and Fluid Science 34 (2010) 966e971.
geometry for the cold end orifice of a vortex tube, Applied Thermal Engi- [23] Y.A. Cengel, M.A. Boles, Thermodynamics: An Engineering Approach, second
neering 29 (2008) 509e514. ed. McGraw-Hill, Literatür, Istanbul, 1996.