Available online at www.sciencedirect.com
ScienceDirect Energy Procedia 75 (2015) 3098 – 3104
The 7th International Conference on Applied Energy – ICAE2015
Validation and Visualization of Decaying Vortex Flow in an Annulus Baiman Chena*, Kelvin Hob , Frank G.F. Qina, Runhua Jianga, Yousif A. Abakrb, Andrew Chanb a
Key Laboratory of Distributed Energy Systems of Guangdong Province, Department of Energy and Chemical Engineering, Dongguan University of Technology, Dongguan 523808, China b The University of Nottingham, Faculty of Engineering, Jalan Broga, 43500, Semenyih, Selangor Darul Ehsan, Malaysia
Abstract Swirling jets produced using inlet swirlers are often used to increase the heat or mass diffusion rates in heat exchangers or reactors, separation of smoke and dust suspended in the air in ventilation ducts etc. The aim of this work is to validate a CFD model through streak-line comparison and to define the fully laminar and turbulent regime for decaying vortex flow produced using a set of axial swirlers in an annular pipe. Flow visualization was conducted using fluorescent dye to visualize on eddy formation along the pipe. Decaying vortex flow was produced using static axial swirlers fitted at the inlet of the annular pipe. In this study, three different swirl angles were investigated: 30°, 45° and 60°. In addition, the Reynolds number (Re) was varied from 10 to 1800. Results shows that the flow regimes for decaying vortex flow in annuli can be defined with reference to axial annular pipe flow. © 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license © 2015 The Authors. Published by Elsevier Ltd. (http://creativecommons.org/licenses/by-nc-nd/4.0/). Selection and/or peer-review under responsibility of ICAE Peer-review under responsibility of Applied Energy Innovation Institute Keywords: decaying vortex flow; fluorescent dye seeding; axial swirl vane; annuli; validation; CFD
1. Introduction The uses of swirling jet/flow are vast, commonly in heat exchangers [1-3], regenerators, rotors, power cables, combustors, cyclones etc. It is known that the additional of a tangential component to an axial flow enhances the thermal performance [4, 5]. Swirling jets can be used as mixers, separators and flow distributors. According to [6], swirling flows can be produced through active/passive methods. Rotating cylinder is a well-researched active method (attributed to the work of Taylor [7, 8]). Other passive methods are, tangential inlets [9], guided vanes [10, 11], and radial or axial swirl generators [12-14].
* Corresponding author. Tel.: +86-769-2286-2038; fax: +86-769-2286-2038. E-mail address:
[email protected].
1876-6102 © 2015 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 Applied Energy Innovation Institute doi:10.1016/j.egypro.2015.07.640
Baiman Chen et al. / Energy Procedia 75 (2015) 3098 – 3104
Nomenclature A
clearance area between two pipes
²
density of water
μ
dynamic viscosity of water
șv
vane angle
Dh
hydraulic diameter
Tight clearances and complex geometrical configurations is common in engineering designs, alongside cost and energy constraints, limits the use of active swirlers and guided vanes, thus, the use of inlet swirlers are warranted. Although the use of static inlet swirlers poses as a viable solution for the said engineering constraints, but, its swirl momentum diffuses downstream. Nozaki et. al. [15] investigated compound swirl jet with a thick rim through flow visualization using smoke seeding method. The authors found out that compound swirl jets are very useful in diffusion control in ventilation ducts, and this can be done through selecting an adequate flow rate ratio. Chang et. al. [16-18] performed an experimental study on turbulent swirling flow in a concentric pipe using smoke and dye seeding techniques. In the work of [19] the authors produced swirling jets using a rotating nozzle for a moderate Reynolds number of (Re=1000). The aim of the study was to investigate on the dominant role of the Kelvin-Helmholtz instability and the centrifugal instability that exists in the axial shear layer. Using the apparatus [20], flow visualization was carried out using the dye seeding method in a water tunnel for three different swirl angles. The governing parameters for this study are the Reynolds number (Re) and the vane angle (șv). The results were used to validate a CFD model that will be used for future work. 2. Experimental Apparatus and Methods The apparatus in figure 1 was designed and assembled. The apparatus was designed for flow visualization through dye injection. Plexiglas is used because the cost and weight is greatly reduced and it is more ductile compared to glass. The working fluid will be circulated between reservoir 1 and 2 through the use of a water pump. Before the fluid enters the test section, the flow is first straightened using stacked straws with a diameter of 5 mm and a length of 50 mm before a swirling decaying flow was produced using an axial swirl vane. Detailed information and dimensions of the setup and vanes can be found in [20].
Fig 1: Experimental apparatus
3099
3100
Baiman Chen et al. / Energy Procedia 75 (2015) 3098 – 3104
In order to validate the apparatus, flow visualization was performed on a normal pipe flow and results showed that the transition from laminar to turbulent regime occurred in the vicinity of a Reynolds number of 2950 [20]. The definition of the Reynolds number is given by the following; UVDh ª kg / m 3 m 3 / s m º Re « » (1) AP ¬« m 2 kg / ms ¼» ȡ is the density of water and Dh is the hydraulic diameter of the concentric pipe, A is the clearance area between the two pipes and μ is the dynamic viscosity of water. The resultant velocity for swirling decaying flow consists of both axial and circumferential component vaxial and vcircumferential. The average axial velocity along the pipe will have a constant value for a constant volume flow rate. The circumferential velocity will decrease along the pipe and eventually reach a value of NIL for a long pipe. The vane angle governs the maximum swirl angle of the flow and it can be defined as;
Tv
tan(T )
(2)
3. Numerical Validation In this section, a comparison will be made between streak lines captured through flow visualization and those produced through numerical simulation. The aim is to have a validated simulation model to increase the confidence of future predictions. 3.1. Governing equations The following are the generalized continuity, momentum and energy equations for fluid flow and heat transfer. With the applied material properties and boundary conditions, swirling decaying flow in a concentric annular duct is solved numerically using ANSYS CFX codes. (U f U )
(3)
0
The simplified time-independent non-viscous Navier-Stokes momentum equation with the two equation k-İ model for eddy prediction is given in eq 4, ( U ) U f U P eff 2U P (P eff U ) T 0 (4) P eff
where,
Pt
P Pt CP U f
ke 2
(5)
H
The turbulent kinetic energy equation for ke is given in eq 6 and turbulent kinetic energy dissipation equation for İ is given in eq 7, w( U f k e ) wt w( U f H ) wt
ª§ P ( U f k eU ) «¨¨ P t V k ¬«©
ª§ P ( U f HU ) «¨¨ P t VH ¬«©
º · ¸k e » U f H ¸ ¹ ¼»
U H2 · º ¸H » CH f ¸ ke ¹ ¼»
0
0
(6) (7)
The constants used in computing the turbulent kinetic energy and the turbulent kinetic dissipation rate are given by Launder and Spalding [21].
Baiman Chen et al. / Energy Procedia 75 (2015) 3098 – 3104
3.2. Boundary conditions, initial conditions and computation The governing equations in the previous section are subjected to the following boundary conditions. At both walls, no slip and impermeable (u=v=0). The flow in the annular duct is assumed steady and uniform throughout the length of the duct. A numerical solution is obtained using the CFD codes provided by ANSYS CFX using the SIMPLEC algorithm. The Barth and Jesperson (high resolution scheme) and a series of structured and unstructured mesh discussed above is used to discretize the governing equations. The transient term is discretized using the second order backward Euler scheme. 3.3. Validation The decaying nature of the flow is made visible using fluorescent dye. This section provides a comparison between experimental streak-lines and modeled streak-lines (simulation validation). Figure 2 compares the experimental and modeled streak-lines for the 30º vane.
Fig 2: Experimental streak-line and predicted stream line for the 30º vane for volume flow rate of 15 ml/s(Re=260).
The experimental streak-line for a volume flow rate of 15±1 ml/s (Re=260) and the modeled streakline with a volume flow rate of 14 ml/s (Re=243) is shown in figure 2. The model is found to have overpredicted the swirl momentum for the 2nd swirl. A comparison between both experimental and modeled streak-line for the 45º vane is shown in figure 3, comparing the experimental streak-line for a volume flow rate of 15±1 ml/s (Re=260) to the modeled streak-line with a volume flow rate of 14 ml/s (Re=243).
Fig 3: Experimental streak-line and predicted stream line for line for the 45º vane for volume flow rate of 15 ml/s (Re=260).
Fig 4: Experimental streak-line and predicted stream line for line for the 60º vane for a volume flow rate of 60 ml/s (Re=1041).
A comparison between the experimental streak-line for a volume flow rate of 60±1 ml/s (Re=1041) to the modelled streak-line with a volume flow rate of 60 ml/s for the 60º vane is shown in figure 4 the prediction is observed to have an exact match with the experimental streak-line. The minor deviation between experimental and modeled streak-lines can be attributed to the following reasons: Deviation in losses through the vanes and pipe between physical experiments and modeling; the difference between water and dye density; the uncertainly during volume flow rate measurements. Taking into account the reasons stated above, the experimental streak-lines were found to be accurate in the laminar regime only as observed by other researchers [22]. The existence of eddies after (Reres>1500) will render the flow turbulent, thus, deviating experimental and predicted streak-lines.
3101
3102
Baiman Chen et al. / Energy Procedia 75 (2015) 3098 – 3104
4. Flow Visualization Figure 5 shows a series of flow visualization images for the 30º vane (S=0.5) for three different flow regimes (laminar, transition and turbulent). Figure 5(a)-(d) depicts the streak-lines for Re (Reres) equals to (350(700), 430(860), 520(1040) and 690(1380)). Streak-lines were uniform and no traces of eddies were found in the early laminar regime, except for Re=690, whereby, traces of eddies were found after the second swirl. The angle for each swirl approaches 30Û as Re increases and at the same time the distance between consecutive swirls also decreases. Figure 5(e)-(g) depicts the streak-lines for Re (Reres) equals to (1040(2080), 1130(2260), and 1220(2440)). This is the transition regime and eddies starts to develop with and increasing intensity as Re increases. Along with the previous observations (increase in swirl angle and decrease in distance between consecutive swirls), the mass diffusivity was observed to increase greatly with increasing Re (Evident by the rapid mixing of water and dye). Figure 5(h) shows the streak line for Re (Reres) equals to 1740(3480). This is in the early turbulent regime and eddies were clearly observed and three dimensional (evident by the ripples along the flow path and shredding of dye into multiple streak-lines along the test section). Due to the high mass diffusivity, much more dye was needed to trace the flow. Even so, the dye diffuses rapidly after seeding and only clouds of yellow ripples was observed at the second half of the section rather than streak-lines.
(a)
(b)
(c)
(e)
(f)
(g)
(h) (d) Fig 5: Flow visualization images for 30º vanes; Re= (a) 350, (b) 430 (c) 520, (d) 690, (e) 1040, (f) 1130, (g) 1220, and (h) 1740.
The use of static vanes as a mean to increase the resultant velocity along a concentric pipe presents a way to enhance the mass and heat transfer rate without the expense of active power sources. Such vanes are easily fitted at the inlet and able to be fitted at tight spaces without affecting the duct. 5. Conclusion Flow visualization for swirling decaying flow produced using an axial swirl vane in a concentric pipe has been conducted using the dye seeding method using fluorescent dye. The following conclusions based on the experimental observations, the designed vanes are able to produce uniform swirling flow along the duct; the use of axial swirl vanes to inflate the resultant Reynolds number is shown and presents itself as a viable means for numerous industrial applications; the validated CFD model and the experimental results will be used as a precursor for subsequent heat transfer and flow studies.
Baiman Chen et al. / Energy Procedia 75 (2015) 3098 – 3104
6. Copyright Authors keep full copyright over papers published in Energy Procedia Acknowledgements This work was supported by the China National Natural Science Foundation Project (No. 51406036) and the Science and Technology Development Project of Dongguan Society (2013108101008). References [1] Goldstein, R.J., et al., Heat transfer--A review of 2005 literature. International Journal of Heat and Mass Transfer, 2010. 53(2122): p. 4397-4447. [2] Goldstein, R.J., et al., Heat transfer--A review of 2004 literature. International Journal of Heat and Mass Transfer, 2010. 53(2122): p. 4343-4396. [3] Goldstein, R.J., et al., Heat transfer--A review of 2003 literature. International Journal of Heat and Mass Transfer, 2006. 49(3-4): p. 451-534. [4] Newton, I., Newton's Principia: The mathematical principles of natural philosophy. 1846, New York: Daniel Ader. [5] Joule, J.P., On the Surface-Condensation of Steam. Philosophical Transactions of Royal Society of London, 1861. 151: p. 133160. [6] Bergles, A.E., Heat transfer enhancement-The encouragement and accommodation of high heat fluxes. Journal of Heat Transfer(Trans ASME), 1997. 119: p. 8-19. [7] Taylor, G.I., Experiments with Rotating Fluids. Proc. R. Soc. Lond. A, 1921. 100: p. 114-121. [8] Taylor, G.I., Stability of a Viscous Liquid Contained between Two Rotating Cylinders. Phil. Trans. R. Soc. Lond. A, 1923. 223: p. 289-343. [9] de Farias Neto, S.R., P. Legentilhomme, and J. Legrand, Finite element simulation of mass transfer in laminar swirling decaying flow induced by means of a tangential inlet in an annulus. Computer Methods in Applied Mechanics and Engineering, 2001. 190(35-36): p. 4713-4731. [10] Saha, S.K. and A. Dutta, Thermohydraulic Study of Laminar Swirl Flow Through a Circular Tube Fitted With Twisted Tapes. Journal of Heat Transfer, 2001. 123(3): p. 417-427. [11] Saha, S.K., A. Dutta, and S.K. Dhal, Friction and heat transfer characteristics of laminar swirl flow through a circular tube fitted with regularly spaced twisted-tape elements. International Journal of Heat and Mass Transfer, 2001. 44(22): p. 4211-4223. [12] Zhang, J., et al., Simulation of swirling turbulent flows and heat transfer in an annular duct Numerical Heat Transfer, Part A: Applications: An International Journal of Computation and Methodology, 2003. 44(6): p. 591 - 609. [13] Zhang, J., et al., Simulation of annular swirling turbulent flows with a new algebraic reynolds stress model. Numerical Heat Transfer, Part B: Fundamentals: An International Journal of Computation and Methodology, 1997. 31(2): p. 235 - 249. [14] Ahmadvand, M., A. Najafi, and S. Shahidinejad, An experimental study and CFD analysis towards heat transfer and fluid flow characteristics of decaying swirl pipe flow generated by axial vanes. Meccanica, 2010. 45(1): p. 111-129. [15] Nozaki, T., et al., Flow visualization of a compound swirl jet with a thick rim. Journal of visualization, 2003. 6(2): p. 135-142. [16] Chang, T. and H. Lee, An experimental study on swirling flow in a 90 degree circular tube by using particle image velocimetry. Journal of visualization, 2003. 6(4): p. 343-352. [17] Chang, T.-H., Experimental stuudy on turbulent swirling flow in a cylindrical annuli by using the PIV technique. International Journal of Automotive Technology, 2004. 5(1): p. 17-22. [18] Chang, T.-H. and K.-S. Lee, An experimental study on swirling flow in a cylindrical annuli using the PIV technique. Journal of visualization, 2010. 13(4): p. 293-301. [19] Liang, H.Z. and T. Maxworthy, An experimental investigation of swirling jets. Journal of fluid mechanics, 2005. 525: p. 115159. [20] Ho, K., Y.A. Abakr, and A. Chan, An Experimental Set-Up For Investigating Swirling Decaying Flow in an Annular Pipe. International Communications in Heat and Mass Transfer, 2011. 38(9). [21] Launder, B.E. and D.B. Spalding, The numerical computation of turbulence flows. Computer Methods in Applied Mechanics and Engineering, 1974. 3(2): p. 269-289. [22] Lim, T.T. and A.J. Smits, Flow Visualization: Techniques & Example. 2000: Imperial College Press.
3103
3104
Baiman Chen et al. / Energy Procedia 75 (2015) 3098 – 3104
Biography Baiman Chen received his Ph.D. in mechanical engineering from the University of Nottingham in 2013. He continued his research on fluid dynamics at the Dongguan University of Technology. Dr. Baiman has research interest on thermoacoustics, heat and mass transfer applications, sustainable technology and computational fluid dynamics.
