A CFD study on optimal venting volume and air flow distribution in a special designed hood system for controlling dust flow *Song Gaoju 1,3, Yang Lei 1,2, and Shen Henggen 1 (1. College of Environmental Science and Engineering, Donghua University, Shanghai 201620, China; 2. Zhongyuan University of Technology, Zhengzhou 450007, China; 3. No.6 Institute of Project Planning & Research of Machinery Industry, Zhengzhou 450007, China) Abstract: A novel hood structure has been designed for the dust control system in the foundry in order to improve the working environment. A composite strategy has been applied for comparative analysis of the optimal venting volume and the airflow distribution between the conventional hood and the novel one in this study. A Computational Fluid Dynamic (CFD) method is used to simulate the airflow fields and dust-polluted air moving paths. The CFD results show that a two-outlet hood, with one outlet located on the left of the hood, is better for improving dustpolluted air than the hood with one outlet only. It can be concluded that the number of the outlets as well as their location on the hood has a significant influence on the air flow pattern in the hood. The optimal venting volume is also a major consideration that is discussed in the study. The venting volume should be designed by considering both the effective level of air flow velocity around the dust source and the energy saving. The optimal airflow distribution may reduce the turbulence in the hood system. Key words: venting volume; air flow distribution; hood; flask shaker; CFD CLC numbers: TG234 +.6 Document code: A Article ID: 1672-6421(2011)03-316-05 Clean industrial plants and modern workplaces are characteristic of a modern foundry. Sand casting is the most common technique used in foundry industry in China, as it is around the world. Foundry is often perceived as a workplace being dirty and environmentally unfriendly. The sand casting process involves the use of a furnace, metal, pattern, and sand mold. The metal is melted in the furnace and then ladled and poured into the cavity of the sand mold, which is formed by the pattern. The sand mold separates along a parting line and the solidified casting can be removed. Once the metal has been poured, the castings may be removed manually or using vibration stations that shake the refractory material away from the castings. Much dust can easily escape into the ambient air from the hood during the shakeout operation [1]. Usually, a high level of dust is generated during the operation, particularly sand dust [2-3]. Hence, the major issues facing the industry are to exhaust and collect the large *Song Gaoju Male, born in 1972, senior engineer, master s degree. Research interests mainly focus on air conditioning systems and their energy saving technologies. Academic research has led to the publication of more than 10 papers in recent years. E-mail: songgaoju@sina.com Received: 2010-12-07; Accepted: 2011-05-24 volumes of floating dust and to maximize cleanliness and keep the foundry environment healthy and safe during the shakeout process [4-6]. However, the problems relating to fugitive dust can never be fully resolved without understanding the air flow characteristics passing through the hood system. This study is aimed at approaching the optimal venting volume and better air flow distribution for the hood of the hood system. For this purpose, a Computational Fluid Dynamic (CFD) method is used for the simulation of the air flow field and dust movement path in the hood in order to reveal what the optimal airflow distribution and the venting volume are. 1 Experimental detail 1.1 Test procedure A camera was used in these experiments to get photographs of the fugitive dust from the hood when the shakeout process operated. This could give the qualitative results of the environmental pollution caused by the shakeout. Meanwhile, some airflow parameters relating to the hood system would be measured and logged, which included its structure and geometric dimensions, and the operational parameters, as the exhaust volume. The actual exhaust volume of the tested hood has been 316
August 2011 measured and calculated with the pitometer and the pitot tube as shown in Table 1. Here, the actual exhaust volume is 75,813 m 3 h -1 and the area of the opening is 39.5 m 2. Hence, the air flow velocity at the opening can be calculated by dividing the exhaust volume by the area of all openings, i.e., the front opening and the top opening, as shown in Fig. 1. According to this computation equation, the calculated average air flow velocity of all openings at the hood without steel cover is 0.53 Research & Development m s -1. When the top steel cover has been fitted, the calculated average velocity of the front opening can be 0.76 m s -1. In addition to the computed air velocity, the study has also actually measured air flow velocities at three sites at the front opening of the conventional hood, as shown in Fig.1. The final air velocity of the front opening would be the average based on the velocity data at S1, S2 and S3. Table 1: Test apparatus and specification Apparatus Model Resolution Accuracy 1 Multi-function thermal anemometer KANOMAX A531 [7] 0.01 m s -1 +/- 2% of reading or +/- 3 fpm 2 Pitometer Testo 512 [8] 0.01 hpa, 0.1 m s -1 reading 0.5% 3 Pitot tube L K = 1 Fig. 1: Schematic of the sites for measuring air velocity in the hood The structure size of the conventional hood could be obtained by its orthographic views as shown in Fig. 2. 1.2 Test apparatus All the apparatus used in the test procedure are listed in Table 1. Here, the multi-function thermal anemometer could log the air velocity values at the openings of the hood when the shakeout operation stopped. The pitometer and the pitot tube were used for logging the actual exhaust volume of the hood. 1.3 New hood model for the hood This study aims to improve the performance of the hood system in controlling dust-polluted air. For the purpose, an additional outlet feature has been designed on the left of the hood for better controlling of the fugitive dust when the shakeout process operates. The newly designed hood structure has the same size as the conventional one, as shown in Fig. 3 and Fig. 4. Fig. 2: Orthographic views of the conventional hood The top opening of the conventional hood was designed for the transportation of the used sand mould. In this way, it is easy for dust to escape. A steel plate was used to cover the top opening of the hood in order to reduce the escaping dust when the shakeout process is in operation. Fig. 3: Schematic drawing of the new hood 317
In this study, the k-ε model of CFD is used. The boundary conditions include the applications of steady laminar incompressible flow, dry air at standard atmospheric pressure and fixed walls. The segregated solver with an implicit formulation is used. The residual values of all variables solved are monitored during the iteration procedure with mass balance set to less than 1.0E-6. 2 Results 2.1 The actually measured operation parameter of the conventional hood Fig. 4: Orthographic views of the new hood 1.4 CFD method The airflow fields of the conventional hood and the newly designed one are analyzed by a CFD method. Three simulation trials on the variable airflow fields for the case with the one-outlet hood have been performed with the exhaust volumes of 75,813 m 3 h -1, 100,000 m 3 h -1 and 150,000 m 3 h -1. In the simulation process, the starting point is to identify a typical structure of a hood as a reference that can be reproduced as a 3-Dimensional CAD SolidWorks engineering drawing package. There are many different sizes and designs available to choose but only one simple design is selected for the purpose of this paper, as shown in Fig. 1. The drawing files are then imported into the CFD software, remodeled into different sections and refined to generate a finite volume meshing. This is a critical step to precisely define the geometrical details of the hood structure. The domain for air flow is also created and the final meshing of all components needs to be accurate. Any errors in the drawings and flow area will need to be corrected. The second step is to import the files into the preprocessing procedure of the CFD software and prepare certain boundary condition data for solving the flow equations. Here, the boundary conditions associated with the flow fields need to be set up. These conditions and parameters may include inlet air flow velocity, outlet air flow velocity, outlet pressure and fluid properties. The next step is to define the nature of the simulated project such as 3-Dimensional, steady and laminar problems. The simulation process is preceeded by the CFD code with automatic data processing and the basic theory of fluid mechanics by balancing the mass continuity and the momentum equations in numerical form, thereafter, to make numerical predictions of the flow variations. This numerical iteration procedure needs to be monitored for convergence and can be recycled if the numerical errors are beyond an unsatisfied preset condition. The final step is the post processing that analyzes the output data and presents them in the form of a velocity vector distribution and contour plots. The study used the multi-function thermal anemometer to measure the actual air flow velocity at the opening. The actual air flow velocity at the opening of the conventional hood without the top cover is about 0.45 m h -1. The velocity for the conventional hood with the top steel cover is about 0.52 m h -1. 2.2 The results of CFD (a) Comparisons between variable outlet designs CFD can simulate the air flow and pressure fields based on the specific boundary conditions. Figure 5 shows the air flow field in the conventional hood, which has a single outlet and with the steel cover on the top opening. The airflow contours in Fig. 5 and Fig. 6 are colored to show variable flow stream paths. Figure 6 shows the air flow field in the new hood, which has two outlets and a steel cover on the top opening. Moreover, the turbulent flow field in the new hood is also simulated, as illustrated in Fig. 7. The contours of Fig.7 are colored to show turbulence intensity. Fig. 5: The airflow field of the conventional hood (a) View 1 318
August 2011 Research & Development The control velocity at the front opening of the conventional hood will increase with increasing venting volume. The results are listed in Table 2. Table 2: The control velocities at the front opening of the hood vs variable venting volumes Venting volume (m 3 h -1 ) 75,813 100,000 150,000 Sites on the front opening of the conventional hood S 1 S 2 S 1 S 2 S 1 S 2 Fig. 6: The airflow field of the new hood (b) View 2 Control velocity (m s -1 ) 3 Discussion 0.33 0.43 0.52 0.64 0.78 0.92 3.1 The airflow velocity for controlling the dust-polluted air Fig. 7: The turbulent flow field of the new hood (b) Different venting volume The three exhaust volumes for simulating airflow fields in the one-outlet hood are 75,813 m 3 h -1, 100,000 m 3 h -1 and 150,000 m 3 h -1. The simulated results of airflow field can be seen in Fig. 5, Fig. 8 and Fig. 9. The computed results and the actually measured results of the air flow velocity at the front opening are different for the case with the one-outlet hood. Without the steel cover, the average velocity derived theoretically, which is 0.53 m h -1, is higher than the actual one, which is 0.45 m h -1. This may be explained by the uneven distribution of static pressure at the front opening and the top opening of the hood, and the static pressure at the front opening may be lower than that at the top opening. When the steel plate cover is on the top opening, the calculated result and the actual result of average air flow velocity on the opening are 0.76 m h -1 and 0.52 m h -1, respectively. This also shows a difference. It can be explained that there may be leakage between the steel cover and the top opening of the hood during the actual operation. 3.2 The outlet number and venting volume (a) Variable outlet designs The actual exhaust volume for the conventional hood is 75,813 m 3 h -1. The steel plate covering the top opening may play a positive function in the control of the fugitive dust. Supposing there is no leakage between the steel cover and the top opening, the calculated average air flow velocity at the front openings is 0.76 m h -1. This value can meet the requirement Fig. 8: The airflow field of the conventional hood with the venting volume of 100,000 m -3 h -1 Fig. 9: The airflow field of the conventional hood with the venting volume of 150,000 m -3 h -1 Fig. 10: The fugitive dust from the leakage between the steel cover and the top opening of the hood 319
of the Chinese standard, which is specified as 0.5 m h -1 to 0.7 m h -1 [9]. However, the leakage between the steel cover and the top opening results in a reduction in the efficiency of the exhaust power, which is shown in Fig. 10. However, the leakage between the steel cover and the top opening is difficult to eliminate completely during actual operation. To deal with this, an additional outlet positioned on the left of the hood may contribute to enhancing the efficiency of the exhaust power. This can be explained by the results shown in Figs. 5 through 7. In Fig. 5, there are many vortexes in the left side of the conventional hood with a single outlet. Figures 6 through 7 show that the new structure with the second outlet can reduce those vortexes. The CFD results explain that the new hood system with two outlets on different sides can contribute to energy saving by exhausting less volume of dust. Hence, the optimal venting volume can be achieved by having two or more outlets installed at optimal locations on the hood. (b) The venting volume The results in Figs. 5, 8 and 9 show that the vortexes in the corners near the front opening of the one-outlet hood can be weakened by increasing the venting volume. The three venting volumes may not all meet the requirement of the specified velocity values for the Chinese standard. The venting volume of 100,000 m m 3 h -1 can be suggested as an optimal venting volume since its control velocity on the front opening of the hood had reached 0.5 m h -1. Table 2 shows that 75,813 m 3 h -1 is too low to control the outflow from the hood and 150,000 m 3 h -1 is unnecessarily high and will waste rather than save energy. 4 Conclusions (1) The optimal exhaust volume can be achieved through comprehensive consideration of the number of the outlets as well as their favorable locations on the hood. A good design of the hood with the above factors may contribute to the better control of velocity around the dust source and at the front opening so that the dust-polluted air cannot easily escape from the hood into the workplace in the foundry. (2) The new hood design with two-outlets may contribute to weakening the vortexes inside the hood and enhance the efficiency of the exhaust power. Especially, the better air flow distribution may enable reduction of the turbulence to achieve less fugitive dust airflow. (3) The optimal venting volume can be found to provide a favorable control velocity at the front opening of the hood. This may contribute to energy saving. (4) The CFD numerical modeling technique applied in this study is proved to be very useful in initiating, further and more comprehensive numerical predictions relating to the dust-polluted air control in the hood at the flask shaker in the foundry. References [1] W H O. H a z a r d P r e v e n t i o n a n d C o n t r o l i n t h e Wo r k Environment: Airborne Dust WHO/SDE/OEH/99.14, Available at https://apps.who.int/environmental_information/dust/ preface.htm, Jan. 2010. [2] Xu Xudong, Zhou Yongtian, Liu Meilan, et al. Survey on the occupational hazard of foundry dust. Chinese Journal of Industrial Medicine, 2002, 15 (1): 37-38. (In Chinese) [3] The Austrian Foreign Ministry. The Foundry Industry - An Overview in Cleaner Production in the Foundry Industry of the People s Republic of China, ASIA INVEST Technical Assistance-ASI/B7-301/02/0535-004 (71307). Available at http://www.centric.at/download2.php?f=103&h=a396cc21936d 013baafede0f3013850d, Jan. 2010. [4] Wang Guanmei and Liang Huiqin. Survey on the Health Status of Dust-Exposed Workers of a Foundry Factory. Ningxia Medical Journal, 2007, 29 (5): 462-463. (In Chinese) [5] Lin Ping and Lǘ Jinbiao. Survey on the Hazard of Dust and its Preventive Measure in Small Casting Factories. Chinese Rural Health Service Administration, 2002, 22 (4): 52-53. (In Chinese) [6] Wang Linchao, Wang Rui, Wei Yangzhou, et al. Study of the Influence of Foundry Dust on the Workers' Lung Function. Chinese Journal of Radioligical Health, 2008, 17 (1): 88-89. (In Chinese) [7] Kanomax USA, Inc. Guide for Multi-Function Thermal Anemometer Model A531. Available at http://www.kanomaxusa.com/anemometer/climomaster/climomaster.html, Jan. 2010. [8] TESTO AG. Guide for Testo 512 differential pressure meter. Available at http://www.testo.com/online/abaxx-?$part=portal.int.productcategorydesk&$event=showfrom-menu&categoryid=1223737, Jan. 2010. [9] Liu Zhuxiong, Zhang Jiaping, Song Gaoju, et al. Dust Control Code for Foundry Guide (2nd edition). Beijing: China Machine Press, 2008. (In Chinese) The study is supported by the Shanghai Leading Academic Discipline Project (B604), the Henan Science and Technology Breakthrough Major Project (102102210440) and the High School Funding Scheme for Key Young Teachers, and the Education Department of Henan Province, 2010. 320