IOP Conference Series: Earth and Environmental Science Experimental study on the cavitation of vortex diode based on CFD To cite this article: L Jiao et al 2012 IOP Conf. Ser.: Earth Environ. Sci. 15 062058 View the article online for updates and enhancements. Related content - Cavitation simulation and NPSH prediction of a double suction centrifugal pump P Li, Y F Huang and J Li - Numerical study of cavitation flows inside a tubular pumping station X L Tang, W Huang, F J Wang et al. - Study on flow instability and countermeasure in a draft tube with swirling flow T Nakashima, R Matsuzaka, K Miyagawa et al. This content was downloaded from IP address 37.44.202.46 on 15/02/2018 at 06:29
Experimental study on the cavitation of vortex diode based on CFD L Jiao,P P Zhang,C N Chen, J L Yin and L Q Wang Institute of chemical machinery and process equipment, Zhejiang University, Hangzhou 310027, China Email:hj_wlq4@zju.edu.cn Abstract. To investigate effects of the boundary conditions on vortex diode performance, CFD simulations and experiment were carried out on vortex diodes with different inlet velocity and outlet pressure. The results show that the cavitation flow is made up of the vapor caused by the low pressure in the chamber and the non-condensing gas transported to the center by the pressure drop of the swirling flow according to Dicipline of Henrry. The size of the cavitation flow increases as the Reynolds Number increases, and decreases as the outlet pressure which makes initial velocity of cavitation higher.the performance of vortex diode is deeply influenced by cavitation flow due to the forced vortex. The above-mentioned methods and results are of great guiding significance to predict the performance of vortex diode. 1. Introduction Vortex diode is used in the nuclear industry and other high-risk liquid transportation systems because of no rotating parts and leakage. Its principle is similar to the diode in circuit, when the fluid enters in the axial port and out of the tangential port, the vortex diode gives low resistance. On the other hand, in the reverse flow condition, great resistance can be achieved because of confined swirling flow due to tangential velocity. The scientists have fined that the pressure in center of swirling flow may be lower than vaporization pressure of liquid if the tangential velocity is high enough, which can lead to cavitation flow in vortex diode. The published literatures have showed some optimization techniques for performance and structure of vortex diode, focusing on the fluid control in low flow rate. However, the information available in the literatures does not reveal several important phenomena to the cavitation of vortex diode. Although T Wada referred to the cavitation flow in the reverse flow of vortex diode, did not research it deeply [5-6]. Kulkarni and Ranad have made some preliminary study about the structure of axial port, tangential port and chamber by simulations and experiments [7-8]. Certain amount of advancements in the study on CFD simulations of fluid machinery cavitation have come recent years. Cavitation flow has great influence in vortex diode performance. In this paper, a deep study has been carried out about the cavitation mechanism and flow resistance characteristic of vortex diode through experiment and CFD. Published under licence by Ltd 1
CFD simulations were carried out on vortex diode. We hope to research the detail of flow field and its internal mechanism. The CFD can also be used to study the optimization of structure which is difficult for experiment. The modals with different inlet velocity and outlet pressure are simulated to research the methods of optimization design. Besides the CFD simulations, we test the development process of cavitation flow in vortex diode through the experiment equipment made up of glass and PIV system with color-coded test functions. Study the effects on cavitation of the inlet Reynolds number and outlet pressure. Based on above experiment results, we research on the source of cavitation flow and its impact on overall performance of the vortex diode 2. Simulation Method and Experiment Equipment 2.1. Simulation Method 2.1.1. Cavitation Model. It is reasonable to use method of simulation proposed by Singhal [13] and SST κ-ω turbulence model. CFD simulations were carried out on vortex diode for cavitation with the phase -2 transport equations and hypothesis of gas-liquid mixed-phase. There are some non-condensable gas in water in the actual project,which can not condensate to liquid under high pressure(unlike vapor), so the fluid within the vortex diode is made up of water, vapor and a small amount of non-condensable gas. 2.1.2. Boundary Condition. The whole geometry is meshed using hexahedral elements (Figure 1). Unsteady condition is selected according the working condition. The boundary conditions are specified as velocity-inlet and pressure-outlet. PRESTO is available for discretization of pressure, speed is coupled with SIMPLE algorithm. The second order upwind scheme is selected in momentum, turbulent kinetic energy and dissipation rate, QUICK scheme for phase -2 transport equations. Figure 1. Mesh Figure 2. Experiment system 2.2. Experiment Equipment Figure 2 shows the experiment system[15-16]. Vortex diode is made up of tangential port, axial port (plexiglass) and chamber (cover as plexiglass).the pump provides pressure to water in tank, valve and regulationis fixed in the pump outlet, which can set the flow rate of vortex diode, and then the water runs into the vortex diode. We use the camera to capture the internal flow, the gas column will be amplified by refraction of gas, water and plexiglass at first, and then amplified by camera, finally amplified by the computer. The first amplification can be removed by giving the ratio factor. The second amplification can be removed through comparing with the standard rule. The last one could be removed by the computer. The error is within 5% caused by the inlet flow pulsation. 3. Results and Discussions 2
3.1. Effect on Cavitation Flow of Outlet Pressure In order to study the effect on cavitation flow of back pressure of vortex diode outlet and investigate the mechanism of cavitation, simulations were carried out on vortex diodes with different outlet pressure. We define Gas column as the part where the gas volume fraction is more than 90%, diameter of gas column as Dv,length of gas column as Lv,the diameter of port as D(D = 19mm) and the length of axial port as L(L = 222mm). Simulations with inlet velocity as 12m/s and different outlet pressure (0kpa,10kpa, 50kpa,100kpa and 150kpa) were carried out. The results show that Dv and Lv decreases as the outlet pressure increases (Figure 3). Figure 4 shows the gas phase volume fraction contour with different back pressures. The gas column runs throughout the axial port at first, then dwindles with the outlet pressure increases, especially the column beyond the outlet, which will gradually disappear. We also made the experiment for the vortex diode cavitation with different outlet pressure (17.7kpa, 23.5kpa,35kpa and 45kpa), and focused on the structure of gas column. The results are shown in Figure 6 and 8 which are agreed with the CFD simulations. The velocity at the same flow rate is stable. The static pressure is important factor for cavitation flow, which should drop lower enough to keep the cavitation flow under high outlet pressure. The pressure drop between the wall and center of chamber should increases, so it requires higher tangential velocity and higher inlet pressure. In this work, the static pressure gets higher and results in less cavitation flow because of higher outlet pressure when the inlet boundary remain unchanged. The law provides some effective means for controlling the size of the gas column. 3.2. Effect on Cavitation Flow of Re In order to study the effect on cavitation flow of inlet pressure of vortex diode, simulations were carried out on vortex diodes with fixed outlet pressure and different inlet pressure. The Reynolds number Re for inlet here is: ρmd0vm Re = µ (1) m Where D0 is diameter of tangential port, vm is the velocity in the place of D0, μm is dynamic viscosity of mixed-phase.,l v 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 L v 0.0-1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 P o /10 4 Pa,L v 1.00 0.95 0.90 0.85 0.80 0.75 0.70 0.65 0.60 L 0.55 v 0.50 0.45 0.40 0.35 0.30 1.5 2.0 2.5 3.0 3.5 4.0 4.5 P o /10 3 Pa Figure 3. Plot of Lv and Dv vs.po in CFD Figure 4. Plot of Lv and Dv vs.po in experiment 3
(a) Po=0kpa (b) Po=10kpa (c) Po=50kpa (d) Po=100kpa (e) Po=150kpa Figure 5. Contour of phase-2 volume fraction for different Po (a) Po=0kpa (b) Po=10kpa (c)po=3.5kpa (d)po=4.5kpa Figure 6. Gas column for different Po in experiment The size of the gas column gradually increases as Re increases from the simulations (Figure 7). When Re is 310,200, the gas is just like a core, but when Re is 448,800, it has run throughout the whole axial port (Figure 9). When the Re increases, in the axial port,the tangential velocity increases, and the static pressure decreases; in the region of cavitation, the cavitation phenomenon intensifies, and the size of gas column increases. Whatever, gas-core doesn t come out in the chamber, but in the axial port where the conical tube connected to the straight tube. The contraction of the axial port leads to smaller radius of swirling flow, higher tangential velocity and lower static pressure, which means the joint of conical and straight tube has the lowest pressure. 4
We also made the experiment for the vortex diode cavitation with different inlet velocity, and focused on the structure of gas column. The results are shown in Figure 8 and 10 which are agreed with the CFD simulations. Figure 10 shows the process of cavitation flow as the inlet velocity gradually increasing. when the inlet velocity is low, the gas column is small spiral cone and gradually disappears in the outlet as shown in Figure 10 (a); when the inlet velocity large, the diameter of gas column increases and extended toward the outlet of axial port, as shown in Figure 10 (b) and Figure 10 (c); when the inlet velocity is large enough, the gas column has been throughout the axial port and chamber, shown in Figure 10 (d). The results are consistent with the simulations. 1.1 1.0 1.0 0.9 0.9 0.8,L v 0.7 0.6 0.5 0.4 0.3 0.2 0.1 L v,l v 0.8 0.7 0.6 0.5 L v 0.0-0.1 10 15 20 25 30 Re/10 4 0.4 6 8 10 12 14 16 v i /m/s Figure 7. Plot of Lv and Dv/s.Re in CFD Figure 8. Plot of Lv and Dv/s. Re in experiment (a) Re=310200 (b) Re=334400 (c) Re=360800 (d) Re=402600 (e) Re=448800 Figure 9. Contour of phase-2 volume fraction with Re 5
(a) v i=6m/s (b) v i=9m/s (c) v i=12m/s (d) v i =15m/s Figure 10. Gas column for different Po in experiment 3.3. Effect on Vortex Diode Performance of Cavitation Flow In order to study the effect on vortex diode performance of cavitation, we investigate the resistance coefficient Eur of the vortex diode while different sizes of the gas column, which is defined as: Pi Po Eur = (2) 2 0.5ρv where Pi is inlet pressure, Po is outlet pressure. Figure 11 shows the plot of Eur vs. size of gas column with different Re and outlet pressure. It can be seen from the figures that Eur increases as the size increases both with different Re and outlet pressures. Whatever, Eur with Po is larger than with Re what mean Re gets greater influence in Eur. Flow in chamber of vortex diode is made up of free vortex outside and forced vortex in the center [2] ; The forced vortex leads to large turbulent kinetic energy and turbulent dissipation rate. The performance of vortex diode depends on the forced vortex. According to Figures 3 and 7, when the Re increases, the gas column mainly develops to the direction of chamber where the strong forced vortex exists, the gas column reduces the space of flow, but occupies the space of forced vortex, reduce the energy loss and make the vortex diode performance weaker. When the outlet pressure decreases, the gas column mainly develops to the direction of the outlet, almost have no effect on the forced vortex. Thus, Eur increases as the size increases, which are mainly results of flow rate. The column (especially which beside the chamber) makes vortex diode perform weak. Eur 26 24 22 20 18 16 14 12 10 8 Eur for Re Eur for P o 6 0.0 0.1 0.2 0.3 0.4 0.5 0.6 Figure11. Eur vs Dv for different Re and Po 6
4. Conclusions (1) The size of the cavitation flow increases with the Reynolds number increases and decreases with the outlet pressure increases, which is affected by static pressure (2) The outlet pressure will enlarge initial inlet velocity necessary for cavitation, then affect the Initial condition of cavitation. (3) The cavitation flow makes great impact on the performance of vortex diode due to the space of forced vortex occupied, which depends on the structure of cavitation flow. References [1] Tippetts J R, Priestman G H and Thompson D 1981 Transactions of the ASME 103 342-351 [2] Sun Q J Jiao L and Wang L Q 2008 Journal of Engineering Thermophysics 29(12) 2046-2048 [3] Syred N 2008 A review of vortex devices and power fluidics Proc. of the 3rd Int. Conf. on Heat and Mass Transfer and Hydrodynamics in Swirling Flow (Moscow, Russia, 2008) [4] Jacobes B 1972 The steady-state and transient performance of some large-scale vortex diodes Fifth Cranfield Fluidics Conf. (Uppsala, Sweden, 1972) 17-32 [5] Yoshitomi H, Wada T and Koizumi T et al. 1989 Japanese Mechanical Society Essays 3434-3439 [6] Wada T, Takagi M and Shimizu A 1986 Fluid Control and Measurement 421-426 [7] Kulkarni A A, Ranade V V and Rajeev R et al.2009 Chemical Engineering Science 64 1285-1292 [8] Kulkarni A A and Ranade V V 2008 AIChE J 24 (5) 1139-1152 [9] Kubota A, Kato H and Yamaguti H A 1992 J Fluid Mech. 240 59-96 [10] Delgosha C O, Reboud J L and Delannoy Y 2003 Int. J Numerical Meth Fluids 42 527-548 [11] Senocak I and Shyy W 2004 Int. J Numerical Meth Fluids 44 975-995 [12] Yin J L,Liu J T and Jiao L et al. 2010 Journal of Engineering Thermophysics 31(5) 769-772 [13] Singhal A K, Athavale M M and Li Huiying et al. 2002 J Fluids Eng, Trans ASME 124 617-24 [14] Wang P, Bai X S and Wessman M 2004 Phys Fluids 16 (9) 3306-3324 [15] Escudier M P, Bornstein J and Zehnder N 1980 J Fluid Mech. 98(1) 49-63 [16] Smagorinsky J 1963 The basic experiment Month Weather Rev 91(3) 99-164 [17] Kayo I and Michael G 1998 Dyn Atmos Oceans 27 333-350 [18] Priestman G H and Tippetts J R 1984 Chem. Eng Res Design 62 67-80 7