Available online at www.sciencedirect.

com

ScienceDirect
Procedia Technology 26 (2016) 544 – 551

3rd International Conference on System-integrated Intelligence: New Challenges for Product and
Production Engineering, SysInt 2016

Thermofluid-Acoustic Analysis of a Ranque-Hilsch Vortex Tube

Wirachman Wisnoea, *, Khairil Muhaimin Abd Rahmana, Yusman Istihatb, Valliyappan
David Natarajana
a
Faculty of Mechanical Engineering, Universiti Teknologi MARA (UiTM), 40450 Shah Alam, Malaysia
b
Faculty of Engineering, Universiti Selangor (UNISEL), 45600 Bestari Jaya, Malaysia

Abstract

This paper aims to present analysis of sound produced from a Ranque-Hilsch Vortex Tube. A microphone was used to record the
sound produced close to the hot tube. It is observed that, for one configuration of RHVT, the sound produced contains a specific
set of frequencies. When the inlet pressure is varied, these frequencies remain almost the same, however, the magnitude changes.
The RHVT produces different set of frequencies when using different swirl generator. These sets of frequencies-magnitudes
represent signatures for each configuration. Different swirl generator nozzles were tested and presented. Frequency signatures with
their related thermofluid performance are obtained.

©©2016
2016The
TheAuthors. Published
Authors. by by
Published Elsevier Ltd.Ltd.
Elsevier This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the organizing committee of SysInt 2016.
Peer-review under responsibility of the organizing committee of SysInt 2016
Keywords: Ranque-Hilsch Vortex Tube; Acoustic Analysis; Thermofluid Performance; Fast-Fourier Transform; Frequency Signatures

1. Introduction

Ranque-Hilsch Vortex Tube (RHVT) (see figure 1) is a mechanical device that is capable of separating compressed
air into hot and cold streams. Compressed air is supplied at the inlet and flows through the swirl generator that creates
swirl motion of the flow along the tube towards the conical valve. The swirl flow at the outer periphery near the tube
wall exits through the valve with higher temperature than the inlet temperature. The center part of the flow hits the
conical valve and bounces back toward the orifice producing lower temperature than the inlet temperature.

* Corresponding author. Tel.: +60-3-5543 5121; fax: +60-3-5543 5160.

E-mail address: wira_wisnoe@salam.uitm.edu.my

2212-0173 © 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the organizing committee of SysInt 2016
doi:10.1016/j.protcy.2016.08.068
 Wirachman Wisnoe et al. / Procedia Technology 26 (2016) 544 – 551 545

Fig. 1. RHVT and its main components.

Even though the process seems to be simple, the phenomena that occur inside the tube are still difficult to explain.
How the temperature can be separated is still not well known. Many empirical studies have been conducted by
researchers around the world. They studied mainly on the effect of RHVT parameters to the performance of the device.
There are many parameters involved on the performance of the RHVT, such as the shape of swirl generator, the number
of intake nozzles, the nozzle area, the shape and the opening of conical valve, the orifice diameter, etc. Mohammadi
and Farhadi [1] conducted experimental analysis to optimize the number of nozzles and their diameters. They observed
that with high number of nozzle the flow became more turbulent that mixed the hot and cold flows that led to the
decreasing performance. The maximum isentropic efficiency obtained in their experiment was 6.8%. Wisnoe et al. [2],
Ismail et al. [3] performed experimental investigation on the effect of conical valve shape, swirl generator, orifice
diameter and inlet pressure on the performance of the RHVT. They obtained maximum isentropic efficiency of 33.6%
when smallest nozzle and medium conical valve were used. Kandil and Abdelghany [4] investigated the effect of
orifice size and the length to tube diameter ratio on the performance of the tube. They conducted computational fluid
dynamics analysis and concluded that the orifice diameter needed to be lowest to obtain maximum cooling. Aydin and
Baki [5] concluded that higher inlet pressure was needed to obtain greater temperature difference.
Measurement of thermofluid parameters of RHVT consists of measuring temperature, pressure and mass flow rate.
Measurement devices are usually tapped directly at the inlet and outlets of RHVT to obtain the readings. This type of
measurement allows obtaining output readings without observing the process inside the tube. Gao [6] conducted an
experiment on the flow inside the RHVT using hot wire to measure the velocity at the cold end. Introducing hotwire
inside the tube may disturb the swirl flow behavior as the presence of the hotwire creates blockage to the flow. This
disturbance may be neglected when the blockage ratio is small enough. But as the size of RHVT is usually small, the
blockage ratio becomes significant. Aydin and Baki [5] visualized the swirl flow pattern inside the tube. They did not
specify the visualization technique used, but it looks like they used air bubbles in water medium. It is understood that
water is an incompressible fluid. On the other side, the temperature separation in RHVT can occur only if compressible
fluid is used. Hence, no quantitative data can be extracted from the visualization results. Kurosaka [7] attempted to
analyze the RHVT effect through acoustic streaming. He found that the temperature separation was related to the flows
which created acoustic wave known as whistle effect. The whistle effect occurs when the air flows through a blockage
which makes the air to vibrate. This phenomenon can also be explained by the acoustic of flute. When a musician
blows a rapid jet of air across the embouchure hole the air flow creates resonances and vibrate. The vibration is released
in form of sound energy out of the end of the holes. This is explained by Fritz and Wolfe [8] in their paper. Another
study on acoustic signals in an RHVT has been conducted by Istihat and Wisnoe [9]. They presented the wavelet
transform of the acoustic signals recorded at the outlet of cold and hot tubes.
In this paper, the effect of swirl generator nozzle area (or nozzle depth) on the thermofluid performance of an RHVT
and acoustic sounds produced close to the hot tube at different inlet pressure is presented. This is to observe the
frequency signatures of the RHVT for different configuration.
546 Wirachman Wisnoe et al. / Procedia Technology 26 (2016) 544 – 551

Nomenclature

FFT Fast-Fourier Transform

h nozzle depth
݉ሶ௖ mass flow rate of at the cold outlet
݉ሶ௜ mass flow rate of at the inlet
Pa atmospheric pressure
Pi inlet pressure
RHVT Ranque-Hilsch Vortex Tube
SG swirl generator
Tc cold temperature
Ti inlet temperature

Greek letters
'Tc difference of temperatures of the inlet to the cold outlet (or cooling temperature)
J ratio of specific heats (for air, J = 1.4)
Kisen isentropic efficiency

2. Methodology

The schematic diagram of the experimental setup is shown in figure 2(a). It is designed to capture the acoustic
signals produced by the RHVT close to the hot tube and also to measure the thermofluid data such as pressure,
temperature and mass flow rate at the inlet and the two outlets. An anechoic box covers the RHVT to isolate it from
outside noise. A microphone is used at the mid length of the hot tubes to capture the sound produced by the RHVT.
Three inlet pressures were applied in the experiment: 10 psi (| 68947.6 Pa), 15 psi (| 103421 Pa) and 20psi
(| 137895 Pa). Four swirl generator nozzle areas represented by their depth were used: 1 mm, 2, mm, 3 mm and 4 mm
(see figure 2(b)). For this experiment, the position (or the opening) of the conical valve was maintained fix at one
rotation from the closed end. The experiment was conducted in a laboratory at 100.6 kPa - 100.9 kPa of atmospheric
pressure.
Three thermofluid entities were calculated and plotted from the readings: cold mass fraction (eq. 1), cooling
temperature (eq. 2) and isentropic efficiency (eq. 3).

m c
Pc (1)
m i

'Tc Ti  Tc (2)

Ti  Tc
Kisen J 1 ·
(3)
§
¨ §P · J ¸
Ti ¨1  ¨¨ a ¸¸ ¸
¨ © Pi ¹ ¸
© ¹
 Wirachman Wisnoe et al. / Procedia Technology 26 (2016) 544 – 551 547

(a) (b)

Fig. 2. (a) Experimental setup schematic diagram; (b) Swirl generator used in the experiment.

For the acoustic side, the recorded sounds are analyzed using Fast-Fourier Transform (FFT). Main frequencies are
extracted and replaced by bars representing the frequency location and magnitude.

3. Results and Analysis

3.1. Thermofluid Performance

Figure 3 represents the performance of the RHVT for 3 different inlet pressures. Figure 3(a) shows the variation of
cold mass fraction when the swirl generator nozzle depth is varied. It is observed that for this experiment, the cold
mass fraction is almost constant for each of the inlet pressures. For inlet pressure of 15 psi and 20 psi, the cold mass
fraction is around 0.45; and for inlet pressure of 10 psi, the cold mass fraction is around 0.38. The cold mass fraction
can be varied if necessary by adjusting the conical valve.

Cold Mass Fraction Ti - Tc Isentropic Efficiency

0.6 20 30%
0.4 15
20%
(deg C)

(%)

10
0.2 5 10%
0 0 0%
0 1 2 3 4 5 0 1 2 3 4 5 0 1 2 3 4 5
Nozzle depth (mm) Nozzle depth (mm) Nozzle depth (mm)

Pin = 10 psi Pin = 15 psi Pin = 20 psi Pin = 10 psi Pin = 15 psi Pin = 20 psi Pin = 10 psi Pin = 15 psi Pin = 20 psi

(a) (b) (c)

Fig. 3. (a) Cold mass fraction; (b) Cooling temperature; (c) Isentropic efficiency.

Figure 3(b) shows the variation of cooling temperature of the RHVT when different nozzle depth is used. It is
observed that, for all inlet pressures, the cooling temperature, which is the difference of the inlet temperature with the
cold temperature, is getting lower when the depth of the nozzle is increased. The cooling temperature is bigger when
the inlet pressure is higher. It tends towards almost the same value when the nozzle depth is increased. A maximum
difference of temperature of 14.8°C is obtained when using a nozzle depth of 1 mm and 20 psi pressure at inlet. When
bigger nozzle is used, the expansion process when the compressed air passes through the nozzle is not as strong as
when the nozzle is small.
548 Wirachman Wisnoe et al. / Procedia Technology 26 (2016) 544 – 551

The variation of isentropic efficiency with nozzle depth is shown in figure 3(c). The trend of all curves is similar
to the cooling temperature. The efficiency drops when the nozzle depth is increased. Higher pressure at inlet produces
better efficiency. The maximum isentropic efficiency recorded here is 23% when using a nozzle depth of 1 mm and
20 psi pressure at inlet.

3.2. Acoustic Analysis

Figure 4 shows the sound recorded for 80 ms duration for 3 inlet pressures using 4 different nozzle depths. The
signals are presented in time domain (horizontal axis) with amplitudes shown in Pascal (vertical axis). It is observed
that when the inlet pressure is increased the amplitude of the acoustic signals becomes more significant; the sound
produced is louder. When bigger nozzle is used, the sound produced is weakened.

10 psi 15 psi 20 psi

0.1 0.1 0.1

Nozzle depth = 1 mm

0.05 0.05 0.05

Pa

Pa

Pa
0 0 0
0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08
-0.05 -0.05 -0.05

-0.1 -0.1 -0.1

Time (s) Time (s) Time (s)

0.1 0.1 0.1

Nozzle depth = 2 mm

0.05 0.05 0.05

Pa

Pa

Pa
0 0 0
0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08
-0.05 -0.05 -0.05

-0.1 -0.1 -0.1

Time (s) Time (s) Time (s)

0.1 0.1 0.1

Nozzle depth = 3 mm

0.05 0.05 0.05

Pa

Pa

Pa

0 0 0
0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08
-0.05 -0.05 -0.05

-0.1 -0.1 -0.1

Time (s) Time (s) Time (s)

0.1 0.1 0.1

Nozzle depth = 4 mm

0.05 0.05 0.05

Pa

Pa

Pa

0 0 0
0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08 0 0.02 0.04 0.06 0.08
-0.05 -0.05 -0.05

-0.1 -0.1 -0.1

Time (s) Time (s) Time (s)

Fig. 4. Time representation of sounds recorded for different inlet pressure using different nozzle depth.

When Fast-Fourier Transform is applied to the signals in figure 4, the frequency representation of the signals is
obtained as shown in figure 5. Here the magnitude (in Pa) is presented in place of the complex amplitude of the Fourier
transform. Similar to figure 4, it is noticed that higher inlet pressure gives overall stronger magnitude. The magnitude
is weakened when bigger nozzle is used. It is observed that for the same swirl generator (same nozzle depth), the
frequencies representing the peaks remains almost at the same location, only the magnitude changes.
 Wirachman Wisnoe et al. / Procedia Technology 26 (2016) 544 – 551 549

10 psi 15 psi 20 psi

0.004 0.004 0.004

Nozzle depth = 1 mm

0.003 0.003 0.003

Pa
Pa

0.002 0.002

Pa
0.002

0.001 0.001 0.001

0 0 0
0 5000 10000 15000 20000 0 5000 10000 15000 20000 0 5000 10000 15000 20000
Frequency (Hz) Frequency (Hz) Frequency (Hz)

0.004 0.004 0.004

Nozzle depth = 2 mm

0.003 0.003 0.003

Pa
0.002
Pa

Pa
0.002 0.002

0.001 0.001 0.001

0 0 0
0 5000 10000 15000 20000 0 5000 10000 15000 20000 0 5000 10000 15000 20000
Frequency (Hz) Frequency (Hz) Frequency (Hz)

0.004 0.004 0.004

Nozzle depth = 3 mm

0.003 0.003 0.003

Pa

0.002
Pa

Pa
0.002 0.002

0.001 0.001 0.001

0 0 0
0 5000 10000 15000 20000 0 5000 10000 15000 20000 0 5000 10000 15000 20000
Frequency (Hz) Frequency (Hz) Frequency (Hz)

0.004 0.004 0.004

Nozzle depth = 4 mm

0.003 0.003 0.003

Pa

0.002
Pa

Pa
0.002 0.002

0.001 0.001 0.001

0 0 0
0 5000 10000 15000 20000 0 5000 10000 15000 20000 0 5000 10000 15000 20000
Frequency (Hz) Frequency (Hz) Frequency (Hz)

Fig. 5. FFT results for different inlet pressure using different nozzle depth.

To clarify this, the significant peaks in the FFT results are replaced by bars located at the peak location with the
same magnitude, producing a set of frequencies called signature. This is shown in figure 6. It is clearly observed that
increasing the pressure at inlet (within the range of this study) does not displace the frequencies significantly when
the same swirl generator is used. The magnitude is, however, getting stronger. On the other hand, different swirl
generator produces different set of frequencies. When the nozzle depth increases, the strength of frequencies produced
is decreased. Note that, at 10 psi, most of frequencies are still very weak and they start to appear at higher pressure. It
is also noted that only a few frequencies appear when using bigger nozzles for the range of pressure studied. The
frequency representations as shown in figure 6 form signatures of the RHVT at certain configuration of the RHVT.
Each set of frequencies can be correlated with the thermofluid performance of the device as shown on the title of each
graph. Overall, it is observed that, for all nozzle depths, the amplitude of the frequencies becoming more significant
with the increase of efficiency and the increase of cooling temperature.
550 Wirachman Wisnoe et al. / Procedia Technology 26 (2016) 544 – 551

10 psi 15 psi 20 psi

μc=0.367͕ѐdc=6.8ºC ,ɻisen=16.69% μcсϬ͘ϰϰϯ͕ѐdc=10.5ºC,ɻisen=19.54% μcсϬ͘ϰϯϱ͕ѐdĐсϭϰ͘ϴºC,ɻisen=23.05%

Nozzle depth = 1 mm

0.004 0.004 0.004

0.003 0.003 0.003
Pa

0.002

Pa

Pa
0.002 0.002
0.001 0.001 0.001
0 0 0
0 5000 10000 15000 20000 0 5000 10000 15000 20000 0 5000 10000 15000 20000
Frequency (Hz) Frequency (Hz) Frequency (Hz)

μcсϬ͘ϰ͕ѐdĐсϮ͘ϯºC,ɻisen=5.66% μcсϬ͘ϰϰ͕ѐdc=5.3ºC,ɻisen=9.89% μcсϬ͘ϰϱϳ͕ѐdc=7.9ºC,ɻisen=12.32%

Nozzle depth = 2 mm

0.004 0.004 0.004

0.003 0.003 0.003

Pa

Pa
0.002 0.002
Pa

0.002
0.001 0.001 0.001
0 0 0
0 5000 10000 15000 20000 0 5000 10000 15000 20000 0 5000 10000 15000 20000
Frequency (Hz) Frequency (Hz) Frequency (Hz)

μcсϬ͘ϯϴϮ͕ѐdc=2ºC,ɻisen=4.92% μcсϬ͘ϰϳϮ͕ѐdc=3.7ºC,ɻisen=6.89% μcсϬ͘ϰϲϯ͕ѐdc=5.5ºC,ɻisen=8.57%

Nozzle depth = 3 mm

0.004 0.004 0.004

0.003 0.003 0.003
Pa

0.002
Pa

Pa
0.002 0.002
0.001 0.001 0.001
0 0 0
0 5000 10000 15000 20000 0 5000 10000 15000 20000 0 5000 10000 15000 20000
Frequency (Hz) Frequency (Hz) Frequency (Hz)

μcсϬ͘ϯϴϳ͕ѐdc=1.4ºC,ɻisen=3.44% μcсϬ͘ϰϲϱ͕ѐdc=2.1ºC,ɻisen=3.92% μcсϬ͘ϰϱϳ͕ѐdc=2.9ºC,ɻisen=4.53%

Nozzle depth = 4 mm

0.004 0.004 0.004

0.003 0.003 0.003
Pa

0.002
Pa

0.002 Pa 0.002
0.001 0.001 0.001
0 0 0
0 5000 10000 15000 20000 0 5000 10000 15000 20000 0 5000 10000 15000 20000
Frequency (Hz) Frequency (Hz) Frequency (Hz)

Fig. 6. Frequencies signature for different inlet pressure using different nozzle depth.

4. Conclusion

In this study, thermofluid and acoustic analysis has been conducted using different swirl generators which are
represented by their nozzle depth. It is concluded that isentropic efficiency of a Ranque-Hilsch Vortex Tube increases
when the inlet pressure is higher. With high inlet pressures, cooling temperature (cooling capability) is stronger or
cooled temperature produced is lower, which means better cooling. Increasing the nozzle area reduces the isentropic
efficiency and the cooled temperature is getting higher. Overall, the isentropic efficiencies obtained in this study are
considered low (less than 23%).
Increasing the inlet pressure of the RHVT does not displace the frequencies significantly when the same swirl
generator is used within the range of inlet pressures studied. The magnitude of the frequencies, however, changes
significantly when different swirl generator is used. Different swirl generator produces different set of frequencies.
The set of frequencies-magnitudes represent unique signature of the RHVT for that specific configuration. This
signature can be correlated with the thermofluid performance of the device.
It is recommended to widen this thermofluid-acoustic study by varying other parameters of RHVT, such as the
orifice diameter, the opening of conical valve (or the cold mass fraction), etc. Reducing the number of nozzles may
reduce the number of frequencies produced. It is also suggested to increase the inlet pressure, so that more significant
frequencies can be captured. Precaution needs to be considered as higher inlet pressure increases the risk of leakage.
Further and deeper study on how each opening/discontinuity (nozzles, orifice, conical valve, etc.) inside the device
 Wirachman Wisnoe et al. / Procedia Technology 26 (2016) 544 – 551 551

produces acoustic signals is also interesting to be conducted as this may result more accurate acoustic signatures. The
use of several microphones at different location of the tube may help to locate the source of sound.

Acknowledgements

Authors would like to thank the Malaysian Ministry of Higher Education (MOHE), the Research Management
Centre (RMC) and the Faculty of Mechanical Engineering, Universiti Teknologi MARA (UiTM) for providing fund
and support for this research under Fundamental Research Grant Scheme (FRGS), File No.: 600-RMI/FRGS 5/3
(74/2014).

References

[1] Mohammadi, S., Farhadi, F. Experimental analysis of a Ranque-Hilsch vortex tube for optimizing nozzle numbers and diameter. Applied
Thermal Engineering, 2013; 61(2): 500-506
[2] Wisnoe, W., Ismail, N., Remeli, M. F., Zakaria, M. F. Experimental investigation on the effect of conical valve shape and swirl generator to the
performance of Ranque-Hilsch Vortex Tube. IEEE Business Engineering and Industrial Applications Colloquium (BEIAC), 2013: 812-817
[3] Ismail, N., Wisnoe, W., Remeli, M. F. Experimental Investigation on the Effect of Orifice Diameter and Inlet Pressure to the Ranque-Hilsch
Vortex Tube Performance. Applied Mechanics and Materials, 2014; 465-466: 515-519
[4] Kandil, H.A., Abdelghany, S.T. Computational investigation of different effects on the performance of the Ranque–Hilsch vortex tube. Energy,
2015; 84: 207-218
[5] $\GÕQ, O., Baki, M. An experimental study on the design parameters of a counterflow vortex tube. Energy, 2006; 31: 2763–2772
[6] Gao, C. Experimental Study on The Ranque-Hilsch Vortex Tube. Technische Universiteit Eindhoven, 2005
[7] Kurosaka, M. Acoustic streaming in a swirling flow and the Ranque-Hilsch (vortex tube) effect. Journal of Fluid Mechanics, 1982; 124: 139-
172.
[8] Fritz, C., Wolfe, J. How do clarinet players adjust the resonances of their vocal tracts for different playing effects? J. Acoust. Soc. Am., 2005;
118 (5): 3306-3315
[9] Istihat, Y., Wisnoe, W. Wavelet Transform of Acoustic Signal From A Ranque-Hilsch Vortex Tube. Proceedings of The 7th International
Conference on Cooling & Heating Technologies (ICCHT), 2014