Chapter 1
Introduction
Ranque-Hilsch vortex tubes, often referred to as vortex tubes, are fascinating devices that
have the ability to divide a compressed gas stream into hot and cold streams without the need
for any additional energy input or moving elements beyond the compression process itself.
These devices, which were discovered in 1933 by Georges J. Ranque and further developed
by Rudolf Hilsch in 1947, use the laws of thermodynamics and fluid dynamics to separate
temperatures.
A vortex tube typically consists of a cylindrical chamber with two outputs for the divided hot
and cold gas streams and a tangential input for compressed gas. A fast-moving vortex is
produced when compressed air is delivered through the tangential intake. Centrifugal forces
cause the outer layers of the gas to heat up as it spirals around the tube's periphery towards
the hot outlet, while a cooler inner core develops and escapes through the cold outlet.
Without the need for refrigerants or moving components, this phenomenon—known as the
Ranque-Hilsch effect—allows for effective thermal separation.
The dependability, compact size, and simplicity of vortex tubes make them valuable. They
are used in many different sectors, such as the cooling of electronic parts, cutting and
welding, and controlling temperature in gas storage and medicinal processes. The exact
mechanics behind the thermal separation of the vortex tube are still up for debate in science,
despite a great deal of research being done in this intriguing area.
The aim of the current study is to change the working fluids and observe the change in the
cooling effect of the vortex tube. For this purpose, the design of the vortex tube is kept the
same for all the working fluids. Two designs are considered for this study: the nominal design
and the grooved geometry design. It has been observed that the grooved design gives a better
cooling effect as compared to the nominal design for all the different working fluids.
The geometry was also varied in the case of the grooved design, as in the length of the tube
was changed while keeping the diameter fixed. As the L/D ratio was increased it was
observed that the temperature at the cold outlet kept decreasing. But beyond L/D = 15 there
was not a substantial difference for our model. For the current study we took a L/D ratio of
12 as the model gave the most coherent results for this ratio.
Chapter 2
Literature Review
An inventive tool that may thermally separate gases without the requirement for moving parts
is the Ranque-Hilsch Vortex Tube (RHVT). While experimenting with a vortex-based pump
for vacuuming iron fittings in 1933, Georges J. Ranque made the initial discovery of this
extraordinary potential. Rudolf Hilsch improved the gadget later, and his thorough research,
which was published in 1947, greatly increased both its design and use. The RHVT's capacity
to separate gas mixtures was demonstrated conclusively by Linderstrom-Lang in 1967.
A cylindrical tube with two outputs for the expulsion of hot and cold gas streams and a single
entrance for compressed air is used by the RHVT to operate. A vortex generator is used to
introduce gas into the tube by using nozzles angled tangentially to produce a whirling flow.
Part of this flow produces a cooler, swirling core that escapes through the cold outlet as it
moves around the tube's edge towards the hot outlet. The Ranque-Hilsch effect, a thermal
separation effect, demonstrates how the apparatus can produce significant temperature
variations without the use of mechanical labour or refrigerants.
The scientific community is very interested in the RHVT because of its ease of use, small
size, and effectiveness in producing heat separation (Zhang and Guo). Its many industrial
uses include rapid cooling and temperature maintenance in medical procedures and storage,
the separation of oxygen from air in the aeronautical industry, welding, thermal sealing, and
part machining, as well as preventing the freezing of compressed gases in storage units.
There is still no widely accepted explanation explaining the observed thermal separation,
despite extensive research and the publication of hundreds of papers targeted at
understanding the working principles of the vortex tube phenomena (Subudhi and Sen [2]). In
the beginning, Georges J. Ranque [3] postulated that adiabatic compression and expansion
was the cause of the thermal separation. Later, Ahlborn and Gordon compared thermal
separation to a traditional thermodynamic refrigeration cycle and proposed that secondary
circulation close to the inlet may be the origin of thermal separation.
Xue et al. [6] conducted additional experimental and numerical studies and found that the
temperature rise was brought on by partial flow stagnation in the multi-circulation zone close
to the hot exit, whereas the temperature reduction was caused by a drop in pressure near the
tube's intake. According to Polihronov and Straatman [7], adiabatic expansion of the rotating
inflow—which converts internal and rotational energy into the rotational energy of the
system by viscous shearing—is the cause of cooling at the center of rotation in a rotating
heat-wave venturi tube (RHVT). Furthermore, Liew et al. [8] proposed that temperature
separation is accomplished by the working fluid expanding and compressing radially
adiabatically in the presence of turbulent eddies that function as discrete heat pumps.
Using a two-dimensional axisymmetric model with a k-ϵ turbulence model with Keyes
correlation, Frohlingsdorf and Unger [9] were the first to publish a numerical analysis on the
vortex tube using the CFX code system. They compared their findings to experiments carried
out by Bruun [10]. Many studies (Kazantseva et al. [11], Behera et al. [12], Farouk and
Farouk [13], Secchiaroli et al. [14], Rafiee and Sadeghiazad [15], Guo et al. [16]) have
studied the temperature separation mechanism and flow field inside the vortex tube using
turbulence models and different CFD programs in the last few years. The performance of an
industrially distributed vortex tube (EXAIR 708 slpm) was initially analysed using CFD by
Skye et al. [17], who found good agreement with experimental data. In order to compare CFD
results with various turbulence models, Wang and Suen evaluated the energy separation
effect's performance with air and different refrigerants. Guo et al. [16] simulated the flow
using the LES turbulence model; post-processed findings provided a thorough picture of the
large-scale unsteadiness and processing vortices inside the tube. The numerous numerical
studies that have been conducted and published recently validate CFD as a trustworthy
technique for analysing, predicting, and researching the performance and flow field of vortex
tube designs.
The majority of experimental and computational studies on vortex tube design optimization
that are currently available in the literature concentrate on the various portions of the vortex
tube's size and dimensions. A small number of research have examined the impact of the
geometric shape of the hot tube. In order to reduce the tube length, Takahama et al. [19]
carried out an experimental investigation on the divergent vortex tube design, which
produced a 10% improvement in performance over the straight cylindrical tube design.
Chang et al. [20] then examined variables impacting a diverging vortex tube's performance.
Afterwards, to find out how the conical valve angle affected the diverging vortex tube's
performance, Devade et al. ran a CFD study.
Pouraria and Zangooee [22] investigated the impact of the tube's angle of divergence using a
similar approach. In their experimental investigation of the impact of curvature on the
temperature separation of the tube, Valipour and Niazi [23] presented a design idea. In order
to verify the findings of the experiments and investigate the impact of changing the tube's
angle of curvature, Bovand et al. [24] ran a CFD simulation. An RHVT with a convergent-
divergent tube design demonstrated improved temperature separation performance, according
to Zangana and Barwari [25]. By adding fins of different forms to the cold tube, Bazgir et al.
[26] studied the performance of vortex tubes and found that fins improved the temperature
distribution profile, with parallelogram fins offering the greatest temperature separation.
As seen above, a large number of research have been carried out, but none of them have
thoroughly investigated the flow mechanics of a vortex tube design with enhanced
performance. By examining the improvement in performance of the innovative design, this
numerical study aims to characterize the important factors that impact the device's
performance, assess the RHVT, and offer a fresh design that enhances its performance. This
paper makes an effort to evaluate the fundamental flow physics entailed in the RHVT.
These many theories highlight the RHVT's intricacy and the difficulties in pinpointing a
single explanatory mechanism for its thermal separation effect. The RHVT is a topic of
ongoing scientific study and innovation due to the continual interest in and research into these
phenomena, which highlights the RHVT's potential for different industrial applications.
Despite the numerous studies that have been conducted, as shown above, there aren’t any
substantial studies that have conducted a thorough investigation into the flow physics of a
vortex tube design with improved performance. The purpose of this numerical study is to
propose a novel design that improves the performance of the RHVT, to analyse the RHVT,
and describe the key parameters that influence the performance of this device by
investigating the increase in the performance of the novel design.
This study also aims to provide a thorough understanding on the cooling effect of the
different types of working fluids. Not only are the different fluids compared but we also
compare the results obtained from a nominal geometry vortex tube with that of a modified
design vortex tube which aims to improve the performance.
2.1 Literature Gaps
A numerical comparison study of the different types of working fluids with modified
vortex tube geometry is not conducted before.
The dependence of the thermal separation on the length of the tube for the ribbed
vortex tube is not done.
The 2 phase numerical study for CO2 is also not well established.
Transient and density based models have not been used in most of the numerical
studies done.
2.2 Objectives
The objective of the current study is to do a comparison study of different working fluids
with respect to both nominal and ribbed geometry vortex tubes. The difference of the cooling
effect achieved with respect to the change in L/D ratio is also done.
The shift in the dividing streamline in each case is also observed and compared with the
result obtained from the ribbed geometry and also the other working fluids.
Chapter 3
Mathematical Modelling
3.1 Computational Domain (Nominal RHVT)
Figures 3.1 and 3.2 represent the geometry obtained from the work of Skye et al. [18]. A 106
mm length and 11 mm diameter cylindrical tube is part of the design of the vortex tube under
analysis. With an exterior diameter that matches the main tube and an inner diameter of 10
mm, the hot exit is shaped like an annular aperture. The diameter of the chilly outlet is 6.2
mm.
The gas swirling motion is produced by the vortex generator, which has dimensions of 12.8
mm in diameter and 0.97 mm in breadth. Six nozzles with rectangular cross-sections, each
measuring 0.97 mm in width and 1.4 mm in length, are included in this design. These nozzles
are positioned such that the gas enters the cylindrical tube tangentially, forming the vortex
that is required for thermal separation.
Sr. No. Parameter Description Dimension (mm)
1. L Length of Vortex Tube 106
2. D Diameter of Vortex Tube 11
3. Dvg Diameter of Vortex Tube Generator 12.8
4. Dho Hot Outlet Inner Diameter 10
5. Dco Cold Outlet Diameter 6.2
6. Wn Inlet Nozzle Width 0.97
7. Hn Inlet Nozzle Height 1.4
Fig. 3.1: (a) Nominal VT Computational Domain Schematic (b) 3D Computational
Domain
3.2 Computational Domain (Ribbed RHVT)
Figures 3.3 and 3.4 shows the ribbed geometry model obtained from the work of Rohan et al.
[]. The model geometry in itself is similar to that of the nominal vortex tube but there are
regions of sudden expansions along the length of the tube. The ribs are in the shape of a right
angle with height of 2 mm and width of 10 mm equally spaced along the length of the hot
exit of the tube.
Sr. No. Parameter Description Dimension (mm)
1. L Length of Vortex Tube 121
2. D Diameter of Vortex Tube 55
3. DVG Diameter of Vortex Generator 12.8
4. DHO Hot Outlet Diameter 10
5. DCO Cold Outlet Diameter 6.2
6. WN Width of Nozzle 0.97
7. HN Height of Nozzle 1.4
8. M Rib Spacing 10
9. WRIB Rib Width 2
10. HRIB Rib Height 17.67
Fig. 3.2: (a) Ribbed VT Computational Domain Schematic (b) 3D Computational
Domain
3.3 Governing Equations
The 3D compressible Navier-Stokes equations were solved numerically using the finite
volume method to conduct the CFD study. The mass, momentum, and energy conservation
concepts are the three main conservation principles included in these equations.
δρ
+∇ . ( ρu ) =0
δt
δ( ρu)
+∇ . ( ρuu )=∇ . ( τ ) −∇ p
δt
δ( ρH ) δρ k
+ ∇ . ( ρuH )= + ∇( ∇ . H −∇ ( τ . u ) )
δt δt Cp
where ρ is the density of the fluid, τ is the viscous stress tensor, u is the fluid
velocity, H is the total enthalpy, k is the thermal conductivity of the fluid,
CP is the specific heat of the fluid and p is the static pressure. In this
study we are considering ideal gas hence the equation of state was
adopted. This was the case for all the different working fluids.
P= ρRT
where R is the ideal gas constant and T is the absolute temperature of the gas.
The standard k-ԑ model was utilised to simulate the turbulence effects and the near wall
boundary layer conditions of the flow. The k-ԑ equations are provided below [28]:
δ( ρk) δ( ρk ui ) δ
δt
+
δ xi
=
δxj [( ) ]
μ δk
μ+ t
σk δ x j
+Gk +Gb− ρε−Y m + Sk
δ( ρε) δ(ρε ui ) δ μ δε ε ε2
+ = [(μ+ t ) ]+C 1 ε (Gk +C 3 ε Gb )−C 2 ε ρ + Sε
δt δ xi δxj σε δ x j k k
where where Gk is the turbulent kinetic energy generation from velocity gradients, Gb is the
turbulent kinetic energy generation from buoyancy, Y m is the contribution of the fluctuating
dilation in compressible turbulence to the overall dissipation rate, σ k and σ ε are the turbulent
Prandtl numbers for k and ϵ, Sk and Sε are user-defined source terms, μt is the turbulent
viscosity and C μ =0.09, C 1 ε = 1.44, C 2 ε = 1.92, σ k = 1.0, σ ε = 1.3.
The turbulent viscosity term in the above equations can be calculated from k and ԑ with the
help of the below equation:
2
k
μt =ρ C μ
ϵ
3.4 Computational Steps
The ANSYS® Fluent package [28] was utilized for domain modeling, simulation, and flow
post-processing. Using SIMPLEC pressure-velocity coupling, a three-dimensional steady-
state simulation was carried out. PRESTO! technique was used for pressure discretization,
whereas second-order upwind scheme was used for density, momentum, energy, turbulent
kinetic energy, and turbulent dissipation rate spatial discretization. The Full Multi-grid
approach was used for the flow initialization. It took 3000 iterations of the solution to obtain
convergence.
3.5 Boundary Conditions
A mass flow inlet boundary condition was applied at the nozzle inlets with a mass
flow rate of 0.0831 Kg/s and a temperature of 294.2K.
The nozzle inlet was also given an initial pressure of 3.3 MPa.
Both the cold and hot outlets were given 0 Pa gauge pressure and 294.2K initial
temperature.
The hot outlet was also given a targeted mass flow rate in order to simulate the
presence of a valve at that position.
These boundary conditions were kept constant for all the different working fluids.
3.6 Grid System
ANSYS® Mechanical was used to create the geometry's computational grid. Three different
cell types were used in the hybrid meshing technique: prismatic inflating layer meshing at the
walls, tetrahedral meshing at the periphery, and hexahedral meshing at the domain center.
The grid and the regions of grid refinement employed in this investigation are shown in
Figures 3.3 and 3.4. 513,884 mesh elements and 150,390 mesh nodes make up the grid. Of
these, 10,535 nodes are found in 7,704 hexahedral elements, while 145,015 nodes are found
in 506,180 tetrahedral elements and prismatic inflation layer cells.
Using BOIs, grid refinement was carried out in three locations, as seen in Figure 3.4 (a). The
core, annulus, and inlet areas' body sizes were established at 0.5 mm, 0.5 mm, and 0.4 mm,
correspondingly. The six layers that made up the inflating layer applied to the mesh had a
growth rate of 1.5 and a total thickness of 0.2 mm. To capture the flow characteristics in the
high shear vortex breakdown flow area, refinement in the annular zone was used. In order to
keep the minimum y+ value above 30, an average cell size of 0.18 mm was utilized in the
vicinity of the walls. The resultant mesh had cell sizes ranging from 0.055 mm to 1.5 mm,
with a maximum skewness of 0.59. The Maximum and minimum aspect ratios were 342.1
and 1.004 respectively with an average of 5.395.
Fig. 3.3: Mesh Grid System (Nominal VT)
Fig. 3.4: (a) Grid Refinement Regions (b) Cross-section of Mesh of Nominal VT
The flow was calculated with a total of eight distinct grid sizes in order to examine how the
dependency on the mesh size. For this investigation, the metric was the Hot Temperature
Difference (HTD) measured at 0.8 Cold Mass Fraction (CMF). The variability of HTD with
cell number is seen in Figure 6. The variance in HTD was less than 1 percent beyond 513,884
cells. Therefore, all subsequent analyses in this work were carried out using this grid size and
layout in order to strike a compromise between high accuracy and reduced computing effort.