 Advances in Technology Innovation, vol. 3, no. 3, 2018, pp. 133 - 140 Flowfield Analysis of a Pneumatic Solenoid Valve Sheam-Chyun Lin 1,* , Yu-Ming Lin 1 , Yu-Song Haung 1 , Cheng-Liang Yao 2 , Bo-Syuan Jian 2 1 Department of Mechanical Engineering, National Taiwan University of Science and Technology, Taipei, Taiwan 2 Metal Industries Research & Development Centre, Kaohsiung, Taiwan , R.O.C. Received 21 July 2017; received in revised from 24 July 2017; accept ed 17 Sept ember 2017 Abstract Pneumatic solenoid valve has been widely used in the vehicle control systems for meeting the rapid -reaction demand triggered by the dynamic conditions encountered during the driv ing course of vehic le. For ensuring the safety of human being, the reliable and effect ive solenoid valve is in great de mand to shorten the reaction time and thus becomes the topic of this research. Th is nume rica l study chooses a comme rcia l 3/ 2-way solenoid valve as the reference valve for analys ing its performance. At first, CFD software Fluent is adopted to simu late the flow field associated with the valve configurat ion. Then, the co mprehensive flo w v isualization is imp le mented to identify the locations of adverse flow patterns. Accordingly, it is found that a high-pressure region exists in the zone between the nozzle e xit and the top of the iron core. Thereafter, the no zzle dia meter and the distance between nozzle and spool are identified as the important design parameters for improving the pressure response characteristics of valve. In conclusion, this work establishes a rigorous and systematic CFD sche me to evaluate the performance of pneu matic solenoid valve. Ke ywor ds: pneumatic solenoid valve, co mpressible nume rica l simulat ion, transient characteristics, pressure-rising process 1. Introduction Pneumatic system has been used extensively in many areas of industrial applicat ions, such as automation control, medica l instruments, and control unit of the vehicle. It is essential to choose an appropriate valve as the interface to electronic controls for performing the required adjustments or actions in accordance to its function design. Among the automobile safety system, an accurate and reliab le pneumatic solenoid valve with a short response time is crit ical in the anti-loc k braking system (known as ABS), which offers imp roved vehicle control and decreases stopping distance. Therefore, the understanding on the flow patterns inside the solenoid valve is in great demand to shorten the reaction time , and thus becomes the goal of this research. Usually, servo valve and on–off valve are two types of electro -pneumatic valves used in controlling the pneumatic actuator. The expensive servo valves with comple x structure are used to achieve the high linear control accuracy. On the other hand, due to the low cost, compact size , and simple structure, the fast-switching on–off valves have received considerable attention in vehicle industry and researchers [1-4]. In 2006, Topçu et al. [5] develops the simple, ine xpensive fast-switching valve for applications of pneumatic position control. Four prototype valves have been built and the basic mode of operation confirmed. In addit ion, the switching characteristics of the on–off valve with 2/2-way function has been investigated both theoretically and experimentally . Simu lated results of the valves dynamics were in agree ment with the experimental results, and thus the validity of the proposed mathematical model was confirmed. * Corresponding author. E-mail address: sclynn@mail.ntust.edu.tw Tel.: +886-2-7376453; Fax: +886-2-27376460 Advances in Technology Innovation, vol. 3, no. 3, 2018, pp. 133 - 140 Copyright © TAETI 134 As expected, the magnetic fie ld has a dominant influence on the response characteristics of a solenoid valve. Thus, an optima l design of the magnetic fie ld of a high -speed response solenoid valve is executed by Tao et al. [6]. They used the finite ele ment method to optimize the solenoid va lve for achiev ing la rger magnetic force and low power v ia the changes on parameters and materials. Later, Wang et al. [7] investigated influences of cross -sectional area of the iron core and ampere turn on the static electro magnetic characteristics through numerical simu lation. They found that the ampere turn has great effect on electro magnetic force for the magnetic saturation phenomenon. Besides , the simu lation method is validated by the experiment. As regards the flow field analysis, several CFD reports [8-12] are focused on analyzing the flow fie ld inside the valve. Peng et al. [8] adopted the comme rcia l CFD software Fluent to establish CFD mode l for simulat ing the inner flo w fie ld of a servo valve when the valve spool is located in certain positions. Also, several improve ments in the core shape of valve are raised and evaluated via the established numerical mode l. Later, Ma and Sun [9] used CFD software Fluent to simulate the static and dynamic flow fields of an e lectro magnetic valve. The function between the mass flow rate and the drop of pressure through the electromagnetic valve was obtained from the results of static flow fie ld simulat ion. Fro m the nume rica l simu lation of the unsteady flow, the valve closed procedure was calculated. The results indicate that the inner flow field nu merical simulation of the valve by Fluent can reflect its working procedure. More recently, in 2016, Liu et al. [10] conducted a research on a solenoid valve used in the hydraulic control system. Based on the conditions occurring in the operation of the hydraulic drive system, the therma l field o f the head is analy zed b y ANSYS. It is illustrated that the solenoid valve has a good performance under high te mperature condition. They presented a method to monitor the performance of the valve while the reactor is working. Fro m the previous papers, it is demonstrated that numerical simu lation can be adopted as a reliable and useful tool in the valve design. Later, Liu et al. [11] p resented a nonlinear dynamic model of a la rge flo w solenoid with the mult i-physics dynamic simulat ion software called Simulat ionX. The dynamic characteristics of this solenoid valve are analyzed and validated by comparing the test and CFD results. In fact, Co mputational Flu id Dynamics (CFD) is increasingly being used as a reliable method for determin ing performance characteristics of other valve. Fa rre ll et al. [12] e xecuted a series of CFD investigation on characterizing the opening and c losing of check valves. They adopted CFX which is a part o f the ANSYS suite of finite e le ment progra ms, to predict and characterize the performances of swing check and lift chec k valves. Also, the good agreements are found via co mparing the available test data of the modeled valves with the numerical results . Fig. 1 Methodology of this numerical simulation over a pneumatic solenoid valve Advances in Technology Innovation, vol. 3, no. 3, 2018, pp. 133 - 140 Copyright © TAETI 135 Therefore, th is computational flu id dynamics (CFD) study chooses a commerc ial 3/2-way solenoid valve, wh ich is used e xtensively in vehicle control system, to e xa mine its dynamic performance. At first, flo w-fie ld simu lation associated with the valve construction is e xecuted by using the commerc ial CFD code Ansys Fluent. Then, the co mprehensive flow visualization is imple mented to identify the locations of adverse flow patterns, which a re critica l for proposing the improving alternatives. Also, the flowchart for this valve research is illustrated in Fig. 1 2. Working Principle of Charging Process and Description of Physical Model 2.1. Work ing principle of charging process Fig. 2 shows the overall valve system, which is co mposed of the solenoid valve, p iston connector, connecting duct, and the outlet vessel. It is necessary to describe the working principle for the pressure -rise process of solenoid valve in brief. For increasing pressure to its setting value for activating the ABS system, a h igh-pressure (10.1 Bar) a ir source is connected to the nozzle inside the top portion of valve (see Fig. 3). Thus, this big pressure difference generates a chocking situation (sonic speed at the nozzle e xit ) and an in flo w with the constant mass flow rate at the beginning of this filling process. However, after the vessel pressure reaches a fixed value (near 52.8 % of the source pressure), the flow rate of this inlet airstream beco mes smaller with a rising vessel pressure. Finally, this process ends when vessel pressure is equivalent to the pressure source . Fig. 2 Overall system of the pneumatic solenoid valve Fig. 3 Open mode of the pneumatic solenoid valve Clearly, this charging process is an unsteady flow undergoing a significant pressure variation; thus the transient simu lation and compressible assumption are needed to realize the comp licate physical phenomena. Ho wever, it is known that the unsteady CFD s imu lation not only needs a high -performance server with huge me mory resource, but also takes a much longer CPU time to obtain the result. Thus, the steady simu lation is carried out on the complete valve system with an opened vessel end in this study for evaluating the flow patterns inside the geometry. Therea fter, a co mprehensive analysis o f the flow pattern inside of the valve system is executed via the simulation outcomes for finding out the modification possibilities. Advances in Technology Innovation, vol. 3, no. 3, 2018, pp. 133 - 140 Copyright © TAETI 136 2.2. Description of physical model The actual valve configuration is quiet co mple x and difficult to establish a nume rical model fo r CFD simu lation. Thus, proper simplifications on the CAD file a re needed to attain an effective nu merica l mode l, wh ich is div ided into several portions with d ifferent grid densities as indicated in Fig. 4(a ). Genera lly, to capture the actual physical phenomenon precisely, the intense grid distribution is placed on regions with an abrupt property variat ion on veloc ity, pressure, or direct ion. Fo r the va lve cons idered here, as illustrated in Fig. 4(b ), these locations include the nozzle, the s mall c learance between nozzle e xit and the armature, e xpansion part in the connector, and junction between the connecting pipe and the vessel. The total grid nu mber of this numerical model is 8.6 million. (a) Overall system of the valve (b) Connecting duct to the vessel Fig. 4 Grid system of the pressure-increasing process for a pneumatic solenoid valve 3. Numerical Scheme This study simulates the comple x flow patterns inside the electromagnetic valve by utilizing the commerc ia l computational fluid dyna mics (CFD) software Fluent [ 13] to solve the fully three-dimensional co mpressible Navier -Stokes equations with the standard k-ε turbulence model. Also, the Se mi-Imp lic it Method for Pressure-Linked Equations (SIM PLE) [14] is imple mented to solve the velocity and pressure coupling calculation for steady cases. Hence, the flo w v isualization inside the valve can be performed and observed carefully to locate the reversed flow p atterns. In this work, several appropriate assumptions and boundary conditions were made to simulate the actual flo w patterns inside a ceiling fan. They are described as : (1) Inlet boundary condition: The inlet boundary condition of valve is set as Pabs=11 bar for serving as an extremely high-pressure input. (2) Outlet boundary condition The outlet boundary condition at the right wall of vessel is set as the atmospheric pressure. (3) Wall boundary condition This numerical model sets the no-slip boundary condition on the solid surfaces of the solenoid valve system. All kinds of flowing flu id proble ms are determined by physical principles, which a re e xp ressed in conservative form for mathe matica l description. They are mass equation (continuity equation) and mo mentum equ ation. Moreover, as the fluid is under the turbulent condition, the additional turbulent equation is needed to incorporate with the governing equations. The continuity and momentum equations in conservation form are expressed as follows: (1) Continuity conservative equation m i i S x u t       )( (1) Here iu is the velocity,  is the density, and m S is the source term. Advances in Technology Innovation, vol. 3, no. 3, 2018, pp. 133 - 140 Copyright © TAETI 137 (2) Momentum conservative equation iiiii Fgpuuu t     )()()(  (2) where p is the static pressure, 𝜏𝑖𝑗 is the stress tensor, and ρ𝑔𝑖 and 𝐹𝑖 are gravitational and external body forces, respectively. Also, the k-ε turbulence model is utilized to solve the Nav ier-Stokes equations. With respect to the incompressible flo w and no source condition under the steady -state, the momentum equation is :       ji jl l ij i j j i ji ji j i uu xx u x u x u xx p uu x u t                                         3 2 (3) Note that Eq. (3) is ca lled the Reynolds -averaged Navie r-Stokes (RANS) equation, where the Reynolds stress -ρui ′ uj ′ should be appropriately modeled by the Boussinesq hypothesis [15] for relat ing to the mean velocity gradients . The advantage of this approach is the relatively lo w co mputational cost associated with the computation of the turbulent viscosity. The k-ε model computes the turbulent viscosity as a function of turbulence kinetic energy k and turbulence dissipation rate ε: (4) (5) (6) where 𝐺𝑘 = 𝑢𝑖 ( ∂𝑢𝑖 ∂𝑥𝑗 + ∂𝑢𝑗 ∂𝑥𝑖 ) ∂𝑢𝑖 ∂𝑥𝑗 is the turbulent kinetic energy generated by the mean veloc ity gradients. 𝐶1𝜀 , 𝐶2𝜀 , 𝐶𝜇 , 𝜎𝐾 , and 𝜎𝜀 are mode l constants with the fo llo wing e mp irica lly derived values: 𝐶1𝜀= 1.44, 𝐶2𝜀= 1.92, 𝐶𝜇 = 0.09, 𝜎𝐾= 1.0 and 𝜎𝜀= 1.3, respectively [16]. 4. Numerical Simulations and Discussions                                     k jk t j i i G x k x k x k t     k C k GG xxxt k j t j i i 2 21                                         2 C t  (a) Overall velocity distribution (b) Region associated with nozzle exit (c) Expansion part of the valve connector (d) Connecting duct to the storage vessel Fig. 5 Velocity distribution for the pressure-rising process inside a pneumatic solenoid valve Advances in Technology Innovation, vol. 3, no. 3, 2018, pp. 133 - 140 Copyright © TAETI 138 This numerical study chooses a comme rcia l 3/2 -way solenoid valve as the reference valve for analy zing its performance. Firstly, CFD software Fluent is adopted to simu late the steady flow field associated with the valve configuration. Later, with the aids of analyzing numerical results, the comprehensive flow visualization is imple mented to identify the locations of adverse flow patterns, which are c ritica l for proposing the improving a lternatives. Hence, the thorough realization on performance features of this valve is attained. Fig. 5(a) shows the overall velocity distribution inside this pneumatic solenoid valve. The high-pressure incoming air stream flows into the valve through the nozzle and undergoes an accelerating and expanding process. Then, this high-speed stream at the nozzle e xit enters the small c learance between the nozzle and the armature. Certa inly, the comp ressed air hits the armature strongly and directly before it flo ws into the inner space of valve. As indicated in Fig. 5(b), there are several significant circu lations occurred in the right portion while a much wea ker c ircu lation e xists in the right part , which is due to an air outlet provided by the connector. Later, owing to the stepwise geometry inside the connector, expansion and circulation are observed in these area-enlarging locations (see Fig. 5c). Finally, the compressed air reaches the 1-liter vessel via the connecting duct with a sma ll c ross section. Certain ly, as de monstrated in Fig. 5(d ), t wo circulat ions generate on both sides of the incoming air flow. In addition, the pressure distribution in the overa ll system of this pneu matic solenoid valve is illustrated in Fig. 6(a). Obviously, the pressure trend decreases along the flow path from the nozzle, the connector, the connecting duct, and the storage vessel as expected. Certain ly, the most dra matic pressure variation occurs inside the nozzle and region near the clearance between nozzle and a rmature as indicated in Fig. 6(b). As a result, the comp rehensive flow visualizat ion yie lds the locations of adverse flow patterns, Also, circulation and reserved flows are observed at region near the nozzle e xit, e xpansion part of valve connector, and junction of the connecting duct to vessel. The above informat ion is c rit ical for proposing the improving a lternatives. Accordingly, the no zzle dia meter and the d istance between nozzle and spool top are identified as t he important design parameters to enhance the pressure response characteristics of valve. (a) Overall system (b) Region near the nozzle exit inside the valve Fig. 6 Pressure distribution for the pressure-rising process inside a pneumatic solenoid valve Advances in Technology Innovation, vol. 3, no. 3, 2018, pp. 133 - 140 Copyright © TAETI 139 5. Conclusions The flo w patterns and response characteristics of a 3/2-way solenoid valve under the charging mode are analy zed in this numerical investigation. With the aids of comprehensive flo w visualizat ion, the locations of adverse flow mechanis ms are realized and identified as the foundation for further modifications on its reaction performance. It fo llo ws that the locations of circulat ion and reserved flo ws are observed at region near the nozzle e xit, e xpansion part of valve connect or, and junction of the connecting duct to the storage vessel. Also, it is found that a high-pressure region e xists in the reg ion between the nozzle e xit and the top of iron core. Accordingly, the nozzle dia meter and the distance between nozzle and spool top are recognized as the important design parameters for improving the pressure response characteristics of solenoid valve. Clearly, the reaction time can be reduced by increasing the nozzle dia meter with an appropriate d istance between nozzle and spool top. Moreover, the charging time for th is valve is estimated successfully via a transient CFD ca lculation in an acceptable deviation fro m test result. In conclusion, this work de monstrates a rigorous and systematic CFD scheme to evaluate the perfo rmance characteristics and to provide important information on key design parameters of the pneumatic solenoid valve Nomenclature Cμ、C1ε、C2ε constants of standard turbulent k-ɛ mode l ρ fluid density Gk turbulent kinetic energy generated by the mean velocity gradients ρa air density t time 𝜏𝑖𝑗 shear stress tensor u absolute velocity tensor σk Prandtl constant of turbulent kinetic equation μ turbulent viscosity σε Prandtl constant of turbulent dissipation equation References [1] R. B. Van Varseveld and G. M. Bone, “Accurate position control of a pneumatic actuator using on/off solenoid valves ,” IEEE/ASME Transactions on Mechatronics , pp. 195-204, September 1997. [2] A. Gentile, N. I. Giannoccaro, and G. Re ina, “ Experimental tests on position control of a pneumatic actuator using on/off solenoid valves,” Industrial Technology, pp. 555-559, December 2002. [3] G. Be lforte, S. Mauro, and G. Mattiazzo, “A method for increasing the dynamic performance of pneumat ic servo systems with digital valves ,” Mechatronics , vol. 14, no. 10, pp. 1105-1120, December 2004. [4] T. A. Parlikar, W. S. Chang, Y. H. Qiu, M. D. Seeman, and D. J. Perreault, “Design and experimental implementation of an electromagnetic engine valve drive,” IEEE/ASME Transactions on Mechatronics, pp. 482-494, October 2005. [5] E. E. Topçu, İ. Yüksel, and Z. Kamış, “Development of electro-pneumatic fast switching valve and investigation of its characteristics,” Mechatronics, vol. 16, no 6, pp. 365-378, July 2006. [6] G. Tao, H. Y. Chen, Y. Y. J, and Z . B. He, “Opt ima l design of the magnetic field of a h igh -speed response solenoid valve,” Journal of Materials Processing Techn ology, vol. 129, no. 1-3, pp. 555-558, October 2002. [7] L. Wang, G. X. Li, C. L. Xu, X. Xi, X. J. Wu, and S. P. Sun, “ Effect o f characteristic para meters on the magnetic properties of solenoid valve for high-pressure common rail diesel engine,” Energy Conversion and Management, vol. 127, pp. 656-666, November 2016. [8] Z. F. Peng, C. G. Sun, R. B. Yuan, and P. Zhang, “The CFD analysis of main valve flo w fie ld and structural optimizat ion for double-nozzle flapper servo valve,” Procedia Engineering, vol. 31, pp. 115-121, 2012. [9] Y. X. Ma and D. C. Sun, “The numerical simulation of the flow field in an electromagnetic valve ,” Machine Tool & Hydraulics, vol. 36, no. 1, pp. 111-113, January. 2008. [10] Q. F. Liu, F. L. Zhao, and H. L. Bo, “Numerical simulation of the head of the direct action solenoid valve under the high temperature condition,” 24th International Conf. on Nuclear Engineering, ASME Press, June 26-30, 2016. [11] Z. Liu, X. Han, and Y. F. Liu, “ Dynamic simu lation of large flo w solenoid valve,” International Mechanica l Engineering Congress and Exposition, ASME Press, November 11-17, 2016. [12] R. Farrell, L. I. Ezekoye, and M. Rain, “Check valve flow and disk lift simulation using cfd ,” 2017 Pressure Vessels and Piping Conf. Paper, ASME Press, July 16-20, 2017. [13] Ansys Fluent User’s Guide-14.5, ANSYS Inc, 2012. [14] B. E. Launder and D. B. Spalding, Lectures in mathematical & models of turbulence, London, England, July 1972. Advances in Technology Innovation, vol. 3, no. 3, 2018, pp. 133 - 140 Copyright © TAETI 140 [15] J. O. Hinze, Turbulence, McGraw-Hill Publishing Co, 1975. [16] S. V. Patankar and D. B. Spalding, “A calculation procedure for heat mass and momentum transfer in three -dimensional parabolic flows,” International Journal of Heat Mass Transfer, vol. 15, no. 10, pp. 1787-1806, October 1972.