- Email: [email protected]
Experimental research on the dynamics of airflow parameters in a six-channel cyclone-separator
Powder Technology 283 (2015) 328–333
Contents lists available at ScienceDirect
Powder Technology journal homepage: www.elsevier.com/locate/powtec
Experimental research on the dynamics of airflow parameters in a six-channel cyclone-separator Pranas Baltrėnas, Aleksandras Chlebnikovas ⁎ Vilnius Gediminas Technical University, Saulėtekio ave. 11, 10223 Vilnius, Lithuania
a r t i c l e
i n f o
Article history: Received 18 February 2015 Received in revised form 9 May 2015 Accepted 1 June 2015 Available online 9 June 2015 Keywords: Cyclone-separator Particulate matter Air flow Dynamics Efficiency
a b s t r a c t The dynamics of airflow parameters were researched in a six-channel cyclone (separator with a tangential flow). The dependencies of dynamic pressures in cyclone's channels and of cleaning efficiency on an airflow inlet velocity were analyzed. Standard and upgraded semi-rings were used in the cyclone's internal structure. Changes in cleaning efficiency at different inlet concentrations of glass and technical salt particles were analyzed. Research results were obtained at inflow velocities of 10.9 m/s to 21.9 m/s. The highest cleaning efficiency was 97.3%. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Cyclone-separators are unrivaled leaders among devices used to remove particulate matters (PMs) from a dirty airflow. In terms of their price, simplicity and operation, cyclone devices have been unparalleled. Cyclones will stay competitive in the market for a long time due to their unique structure, absence of any moving parts and filtering surfaces that require regular service, comparatively low aerodynamic resistance and high efficiency [2,4,14,25]. Air cleaning devices of the cyclone type are most often used in industries which are characterized by high volumes of polluted air emissions. Most devices are suitable for the separation of PM of large-scale dispersion in multi-phase flows. The work of the common hollow filter is based on the widespread principle of separation of PM with the appearance of centrifugal forces that are influenced by the turbulent airflow inside the body of the equipment. The effectiveness of the mentioned cyclone-separator is from 75 to 85%; the air is cleaned from the PMs that are more than 20 μm in diameter. The structure of the upgraded six-channel cyclone is designed for the separation of dispersed fine PM, up to 10 μm in size [1,9,10,24]. Although cyclone-separators operate under the influence of centrifugal and gravitational forces, their cleaning efficiency depends on the characteristics of inlet airflow [11]. When airflow parameters – velocity and pressure – change, aerodynamic forces in the cyclone's system also change. The aim of this research is to evaluate changes ⁎ Corresponding author. Tel.: +370 63693865. E-mail addresses: [email protected] (P. Baltrėnas), [email protected] (A. Chlebnikovas).
http://dx.doi.org/10.1016/j.powtec.2015.06.005 0032-5910/© 2015 Elsevier B.V. All rights reserved.
in airflow pressure and cleaning efficiency at different airflow inlet velocities in a six-channel cyclone. The cyclone's basic structure has remained unchanged for more than a hundred years. There are two main types of cyclone structures: a direct cyclone and a reverse-flow cyclone [21,22]. However, the most frequently used are reverse-flow cyclones, which are also divided into axial flow and rotational flow cyclones with a tangential gas inlet [3,4,10,15]. Patterson and Munz [18] studied the effect of PM concentration (of up to 235.2 g/m), fluid temperature (from 300 K to 2000 K) and gas inlet velocity (from 3 m/s to 42 m/s) on cyclone's collection efficiency and determined that an increase in PM concentration had improved the efficiency of collection, in particular at a high temperature [8,23]. Turbulent mixing of the flow may be associated by a free flow of gas kinetic theory. By this case the velocity of air molecules distributed by the normal law. According to this law, in the distribution of velocities in the gaps between the curved half-rings is defined by dependence [10]: σ¼
2ρaϑ2ϕ δ3r
2
¼
2ρak
ð1Þ
δ2 r2 2
where: σ — standard deviation of velocity variation, σ ¼ 6kρaϑ ;ϑ — δ3 r2 velocity of airflow, m/s, ρ — air density, kg/m3, a — distance between the aspects of half-rings, m, δ — diameter ofair molecules, m, r — 1ffi ϑ2 exp − channel of cyclone radius, also f ðν Þ ¼ pffiffiffiffi 2 : 2σ 2πσ Cyclone-separators are used as drying machines, reactors and catalysts [5,12,19]. Although recent patented cyclones differ by
P. Baltrėnas, A. Chlebnikovas / Powder Technology 283 (2015) 328–333
their originality, they still retain the main distinguishing feature of these devices — a turbulent flow [7,20]. Highly efficient cyclones and filters are used to reduce dustiness in industry [17]. Cyclone's two most important working parameters, collection efficiency and pressure drop, largely depend on the concentration of dispersed PM [6,23]. The work of channel cyclone-separator is based on centrifugal forces and additionally occurring filtration process. Due to the interaction between inlet flow from the (peripheral) channel coming next and the flow following the direction towards the axis of the cyclone along the channel (transit), additional filtration takes place. Air flow is filtered through the peripheral flow — a curtain appearing behind the separation zone of curved half-rings (quarter-rings), which raises cleaning efficiency up to 15%. The aim of this work is to determine six-channel cyclone's airflow parameters and analyze air cleaning efficiency at different dispersities of PM. 2. Materials and methods Experiments were carried out on an upgraded six-channel cyclone located at the Environmental Technologies Laboratory of Vilnius Gediminas Technical University (VGTU). A schematic diagram of the cyclone's basic design is presented in Fig. 1. The cyclone's air cleaning efficiency was determined by introducing into it ground PM of glass and technical salt of up to 50 μm and 20 μm. The standard dynamic pressure (1/2ρV2) of the airflow was measured in the respective channels of the cyclone's structure using a Pitot (Prandtl) dynamic tube connected to a multifunctional measurement instrument Testo-400. Measurements were made at the following characteristic points (Fig. 1) of the separation chamber inside the structure: at the beginning, in the middle and at the end of the channel, and also at the intermediate points (between the beginning and the middle, and between the middle and the end of the channel). The dynamic tube was inserted at the points directly through special holes made in the separation chamber cover plate. The pressure setting for the reliability of results was performed over
Fig. 1. Cyclone cross-section with semi-ring at position I: I–VI — cyclone channels; R1–R5 — semi-rings; 1–45 — dynamic pressure measurement points; a — segment circular spacings.
329
all cross-sections of each cyclone-separator channel, by making a 9-point network (on the periphery boundary layer, in the middle of each channel cross-section, and on the inner wall of the channel). Dynamic pressures were measured by using standard curvilinear semi-rings and upgraded semi-rings with windows. All pressures were measured by changing the airflow yield supplied by the channel ventilator. The supplied airflow was changed with a lever of the ventilator's control unit. The aim of this regulation was to achieve the optimum pressure for the cyclone's highest cleaning efficiency. Fig. 1 shows the cyclone's internal structure and the location of semirings. The positions of semi-rings (only R1, R3, R5) are regulated as follows: the rings are shifted by 10 mm to the left in position II and by 10 mm to the right in position III, but other semi-rings remain in the same places. Regulation was made only in the x-axis direction. In order to determine the best application of the air cleaning device concerned in a particular industry, PMs of different density, bulkiness and dispersity were selected. The physical density of glass (SiO2) PM was 2670 kg/m3 and that of technical salt (NaCl) was 2210 kg/m3. For determining the cyclone's efficiency a high-pressure compressor producing an air pressure of up to 6 bars was used during experiments. The compressor was connected to a feeding nozzle which sucks in and delivers dust PM directly to the cyclone's air inlet, 200 mm in diameter, thus forming a dispersed two-phase flow. An airflow supply ventilator (RUCK RS200L, power 190 W) was installed in front of the six-channel cyclone and connected to an air duct, 2 m long. Pressure changes in the system were registered with a differential pressure meter (PCE Instruments DSM-1). A possible error for stream unevenness that might occur due to a non-uniform distribution of PM within the entire stream of the inlet air was eliminated. Glass vessels for specimens, scales for weighing specimens (error ± 0.1 g), and a second meter Sekonda for registering air inlet time (error ± 0.2 s) were used during tests. Experimental research on the cleaning efficiency of the six-channel cyclone was carried out according to the Methodology “LAND 28-98/ M-08 Determination of the concentration of dust (PM). Weighted method.” [13]. Before being tested, the specimens of glass and technical salt PM were dried up to a constant weight in a laboratory electric stove at a temperature of 100 °C. PMs were ground with the grinder Retsch RM200. Only PMs of up to 20 and up to 50 μm fraction were used for the experiments. Studies using these PM sizes allow us to investigate the multi-cyclone cleaning efficiency of fine PM and compared with conventional centrifugal cyclone efficiency. The remaining portion was sieved once again. During experiments the ambient air temperature varied from 20.1 °C to 22.2 °C, while relative humidity reached 52%. A two-phase airflow tangentially enters the first channel through the inlet (1) in the separation chamber. The channel is delimited by a peripheral wall and the first curvilinear semi-ring. The airflow is thus distributed within channels of different curvature and is filtered through spacings between semi-rings. PMs are precipitated on the bottom of the six-channel cyclone by a turbulent flow created by centrifugal forces. On entering segment circular spacings (9) they fall down and accumulate in the cyclone's hopper (10). After passing through all six channels clean air leaves the system through the airflow outlet (8). Dusty air is filtered in the active zone of channel spacings. Filtration also takes place during the interaction of PM when they coagulate and separate from the airflow (Fig. 1, Fig. 2). Experiments were carried out by using the standard (without a window) and upgraded (with a window) structure of curvilinear semi-rings. In the case of upgraded structure a polluted airflow can be returned to the previous channels. Thus, this structural solution has extended the time of filtration of a dirty airflow in cyclone's channels. The experimental research on dynamic pressures analyzes cases involving the first three external semi-rings with cut windows of 10 mm curved width. Tests on air cleaning efficiency were carried out by using upgraded semi-rings: only the first (R1) upgraded semiring; the first two (R1 and R2) upgraded semi-rings, and the first three (R1, R2 and R3) upgraded semi-rings (Fig. 2).
330
P. Baltrėnas, A. Chlebnikovas / Powder Technology 283 (2015) 328–333
Fig. 2. Structure of six-channel cyclone (a), standard (b) and upgraded (c) semi-rings: 1 — airflow inlet; 2 — diffuser; 3–7 — semi-rings; 8 — airflow outlet; 9 — cyclone bottom; 10 — conical hopper; 11 — standard semi-ring; 12 — upgraded semi-ring; 13 — window.
For the experiments at issue scientific sources most often present works carried out at concentrations varying between 1.0 g/m3 and 25 g/m3. The concentrations of PM in dirty air ranged from 500 mg/m3 to 10 g/m3. Multichannel cyclone-separators may be applied as air cleaning device to contaminate PM near workplaces. Also, as well as industrial facilities, device to clean polluted air (gas) flow in industrial plants. For this purpose, experimental research was carried out used a rather wide concentration range of contaminated air. The concentration of PM mass in the air is calculated according to the standard LAND 28/M-08 (1998): ðm2 −m1 Þ ; C¼ V0
ð2Þ
where: C — dust concentration, mg/m3; m1 — mass of the filter without PM, mg; m2 — mass of the filter with PM, mg; V0 — volume of the air sucked through the filter and recalculated under normal conditions, m3. As experiments used dust of different nature, i.e., glass and technical salt, after testing specimens of type I the entire system – hoses and cyclone structure – was cleaned by blowing a high-speed flow through it.
3. Results and discussion Average dynamic pressures inside cyclone's channels were registered at the maximum (≈22 m/s), medium (≈14 m/s) and minimum (≈ 11 m/s) inflow velocities by determining three positions of semirings. The results of experimental research on dynamic pressures in the optimum case – position I – are presented in Figs. 3 and 4. 3.1. Dynamic pressure with standard semi-rings Dynamic pressure was measured to determine the dependence of aerodynamic characteristics on the inflow velocity in the cyclone. Standard dynamic pressures were measured during these experiments. The measurement results show the dependence of two-phase flow distribution in cyclone on turbulence flows of different levels. All selected velocities created a high-level flow of turbulence. Turbulence of fluid flow is measured by the Reynolds number, which is composed of the inlet velocity and the height of channel. At the maximum velocity (≈ 22 m/s) — Re ≈ 4.35 × 105, at the medium (≈ 14 m/s) — Re ≈ 2.8 × 105 and at the minimum (≈ 11 m/s) — Re ≈ 2.2 × 105. The results of dynamic pressure measurements inform
Fig. 3. Effect of inflow velocity on the average dynamic pressure in the cyclone channels with standard semi-rings in position I.
P. Baltrėnas, A. Chlebnikovas / Powder Technology 283 (2015) 328–333
331
Fig. 4. Effect of inflow velocity on the average dynamic pressure in the cyclone channels with upgraded semi-rings in position I.
us about aerodynamic processes and changes in a turbulence level in the cyclone. The highest values of dynamic pressure were obtained at the maximum inlet velocity of 21.9 m/s. When standard semi-rings are in position I, the value is equal to 173.4 Pa. Meanwhile values obtained with standard semi-rings in positions II and III are 171 Pa and 174.6 Pa, respectively. It can, therefore, be stated that the highest velocity (axial) of the sixth channel is obtained when standard semi-rings are in position III. According to previous tests, cleaning efficiency decreases in positions II and III. This might have resulted from inappropriate transit/peripheral flow ratio. The optimum ratio is achieved in position I. As the obtained results show, when an airflow passes from one cyclone channel to another, the average dynamic pressure changes, on average, by 1.15 to 1.45 times as a result of a decrease in the channel's cross-section and a simultaneous increase in dynamic pressure and velocity. When air flows from the second channel to the third one a difference in pressures is bigger than mentioned above — 1.88 times at the minimum airflow in position I. In positions II and III this pressure difference is smaller equaling, on average, 1.3 times. The smallest measured average dynamic pressure, 12.4 Pa, was recorded in the first channel at the lowest inflow velocity, 10.9 m/s, and in position I of standard semi-rings. At the medium and maximum inflow velocities, average dynamic pressures in all the three positions in question were 73.5 Pa and 110.4 Pa respectively (Fig. 3). The analyzed results show that when the arrangement of semi-rings and also spaces among them are changed, the size of the cross-section used for the filtration of dirty air changes, thus increasing he dynamic pressure of air. As Fig. 3 shows, dynamic pressures are distributed quite evenly and increase in a linear manner from the cyclone inlet (entry to the first channel) towards the cyclone's center line (the cyclone's channel 6) and air outlet. Dynamic pressure is influenced by the change of inlet velocity and the cross-sections of channels. 3.2. Dynamic pressure with upgraded semi-rings The maximum value of dynamic pressure, 180.3 Pa, was recorded in semi-ring position III in the cyclone's channel 6 when three external semi-rings with windows cut in them were used. Meanwhile values obtained using upgraded semi-rings in positions I and II stand at179.5 Pa and 177 Pa, respectively. It can, therefore, be stated that the highest velocity (axial) of the sixth channel is achieved when upgraded semi-rings are in position III. When an airflow passes from the second channel to the third a difference in pressures is bigger reaching 1.83 times at the minimum airflow in position III. In positions I and II
this pressure difference is smaller and equals, on average, 1.6 times (Fig. 4). The tendency of pressure distribution remains similar to that of standard semi-rings. Optimum position I was selected for tests on cleaning efficiency. Using upgraded semi-rings, the minimum dynamic pressure of 17 Pa was recorded in the first channel at an inflow velocity of 10.9 m/s (Fig. 4). At the medium and maximum inflow velocities, average dynamic pressures in all the three positions in question were 79 Pa and 116.6 Pa respectively. 3.3. Cleaning efficiency with standard semi-rings Results obtained using the PM of glass and of technical salt are comparable as the PM specimens were prepared according to the same methodology and were used in experimental research in the same manner. The highest cleaning efficiency, 97.3%, was achieved at 10 g/m3 concentration of PM, 50 μm in size. The lowest efficiency, 94.3%, was registered at 500 mg/m3 concentration of the same fraction. When PM dispersity changes to 20 μm the cyclone's air cleaning efficiency falls, on average, by 1.5 times. The maximum cleaning efficiency, 74.1%, achieved at standard internal geometry was also recorded at the maximum inlet concentration of 10 g/m3. The minimum efficiency, 48.5%, was established at an initial concentration of 500 mg/m3 (see Fig. 5). Tests were carried out using curvilinear semi-rings with windows. Due to this upgrading a polluted airflow returned to the previous channel and additional part of PM could precipitate and enter the hopper via segment circular spacings (see Fig. 2). The highest efficiency of removing technical salt PM (up to 50 μm) from an airflow is 94.9% at the highest inlet concentration of 10 g/m3. An analysis of the degree of catching dispersed fine PM of up to 20 μm shows that the efficiency of catching glass PM is higher than the efficiency of catching technical salt PM. The average difference in cleaning efficiency of the PM of these types accounts for nearly 3%. The maximum air cleaning efficiency using standard semi-rings reaches 72.3%, which is by 2.5% smaller than in the case of glass PM. The tendency of dependencies of air cleaning efficiency on concentration remains the same — air cleaning efficiency increases when concentration increases. After completing experimental research on the effect of PM concentration on gas flow and on separation of PM using cyclones of a standard structure, researchers Mothes and Loffler [16] determined that the centrifugal force inside the cyclone had decreased at a high concentration of PM. As studies on efficiency carried out by Mothes and Loffler [16] show, cyclone's performance improved at a high concentration due to the agglomeration of PM [10].
332
P. Baltrėnas, A. Chlebnikovas / Powder Technology 283 (2015) 328–333
Fig. 5. Effect of type and size of PM on the air cleaning efficiency of the six-channel cyclone with standard semi-rings.
3.4. Cleaning efficiency with upgraded semi-rings Air cleaning efficiency was determined using upgraded semirings, i.e., the first (R1), the first two (R1 and R2) and the first three (R1, R2 and R3) upgraded semi-rings. The remaining semi-rings in each case were standard. To analyze the degree of cleaning of the six-channel cyclone with upgraded semi-rings (internal geometry) dispersed fine PM, up to 20 μm in size, were used. The maximum efficiency in removing glass PM from an airflow was 78.4%. It was obtained using the first three external upgraded semi-rings at the highest inlet concentration of 10 g/m3 . It should be noted that when the number of upgraded semi-rings (with windows in the first and second semi-rings, and in the first one only) is reduced and the inlet concentration of glass PM is decreased, cleaning efficiency also decreases. With upgraded semi-rings, the lowest efficiency, 47.7%, was recorded at an inlet concentration of 500 mg/m3 of technical salt, and using only the first (external) upgraded semiring (Fig. 6).
Cleaning efficiencies recorded for upgraded curvilinear semi-rings insignificantly differ from the analyzed efficiencies while using standard semi-rings for glass PM, but are smaller. It can be noted that the introduced upgraded semi-rings enhance air cleaning efficiency and when their number increases, i.e., all the three upgraded semi-rings are used, the maximum cleaning percentage reaches 76.9%, but efficiency is by 1.5% lower than in the case of the glass PM at issue. When spaces between upgraded semi-rings and the peripheral wall decrease, active dynamic forces increase, which leads to the change of other parameters. Changed dynamic pressures affect the efficiency of dust PM catch when PMs are taken to the hopper through segment circular openings in the bottom. Air cleaning efficiency increases with the inlet concentration of PM increasing under the influence of increasing centrifugal forces and filtration effects. Air cleaning efficiency in the presence of higher centrifugal forces (at an airflow velocity above 22 m/s) was not analyzed. However, this dependence is likely to change according to asymptote [10]. The particularity of the six-channel cyclone-separator lies in the fact that a
Fig. 6. Effect of type and size of PM on the air cleaning efficiency of the six-channel cyclone with upgraded semi-rings.
P. Baltrėnas, A. Chlebnikovas / Powder Technology 283 (2015) 328–333
dynamic gas-dust layer forms in the spaces between curvilinear semirings when airflow passes through channels. When a dirty airflow is filtered through this layer, part of PM is caught and at the same time cleaning efficiency is enhanced [16]. One of the parameters of the cyclone-separator, which defines the energy loss during cleaning process, is the aerodynamic resistance. As known the aerodynamic resistance of traditional cyclone such as CN-15, which is similar to tested multi-channel cyclone by capacity is equal to 700 Pa approximately. Maximal aerodynamic resistance of multi-channel cyclone equal to 450 Pa, using upgraded semi-rings at an airflow velocity is 22 m/s. Under flow distribution rates, resistance is decreasing because of changes in the flow paths of turbulent flows, which makes less energy losses. An analysis of the removal of the PM concerned from an airflow leads to the conclusion that the air cleaning device in question is not so efficient in removing the PM of technical salt from a polluted flow, which might be determined by the specific characteristics of these PMs, such as density, form, and degree of agglomeration processes. The centrifugal force effect increases at higher speeds. The PMs are mostly taken to the peripheral wall, precipitate on the bottom of cyclone, and after entering segment circular spacings fall and accumulate in the cyclone's hopper. 4. Conclusions Air cleaning efficiency in a six-channel cyclone-separator improves when the inflow velocity in cyclone increases from 11 m/s to 22 m/s. Glass and technical salt particulate matter is effectively removed from an airflow at the optimal inflow velocity in cyclone of 22 m/s using upgraded semi-rings. The greatest efficiency, 78.4%, is reached when dispersed fine glass PMs of up to 20 μm are used. Air cleaning efficiency of 74.1% is determined using standard semi-rings when glass PMs of up to 20 μm are introduced at 10 g/m3 concentration and at 22 m/s inflow velocity. The upgraded semi-rings have improved cleaning efficiency by about 4 to 5%. The new structural solution with upgraded semi-rings extended the time of filtration of a dirty fluid flow in cyclone's channels, because a polluted airflow can be returned to the previous channels. The maximum dynamic pressure of 179.5 Pa was registered while using upgraded semi-rings at the highest inflow velocity of 22 m/s in the sixth channel of cyclone in position I. When the position of curvilinear semi-rings changes from position I to positions II and III, pressure decreases due to a change in turbulent flow trajectories causing additional energy growth. A comparison of dynamic pressures in a sixchannel cyclone-separator has shown that they are significantly (approximately by 10–15%) bigger when upgraded semi-rings are used. After evaluating the results of pilot research on multichannel cyclone-separator, it can be concluded that this device is suitable for removing gaseous effluents from small dispersion particulate matter, up to 20 μm in size, into the atmosphere.
333
References [1] S. Altmeyer, V. Mathieu, S. Jullemier, P. Contal, N. Midoux, S. Rode, J.P. Leclers, Comparision of different models of cyclone prediction performance for various operating conditions using a general software, Chem. Eng. 43 (2004) 511–522. [2] A. Avci, I. Karagoz, Effect of flow and geometrical parameters on the collection efficiency in cyclone separators, J. Aerosol Sci. 34 (2003) 937–955. [3] P. Baltrėnas, J. Morkūnienė, P. Vaitiekūnas, Mathematical simulation of solid particle dispersion in the air of Vilnius city, J. Environ. Eng. Landsc. Manag. 16 (1) (2008) 15–22. [4] S. Bernardo, M. Mori, A.P. Peres, R.P. Dionisio, 3-D computational fluid dynamics for gas and gas–particle flows in a cyclone with different inlet section angles, Powder Technol. 162 (2006) 190–200. [5] J. Bradulienė, S. Vasarevičius, Experimental investigation into reduction of gravel road dustiness using “SAFECOTE” solution and its mixture with CaCl2, J. Environ. Eng. Landsc. Manag. 18 (2) (2010) 145–153. [6] L. Curtis, W. Rea, P. Smith-Willis, E. Fenyves, Y. Pan, Adverse health effects of outdoor air pollutants, Environ. Int. 32 (2006) 815–830. [7] J. Dirgo, D. Leith, Performance of theoretically optimized cyclones, Filtration and Separation 22 (1985) 119–125. [8] W. Gujun, S. Guogang, X. Xiaohu, S. Mingxian, Solids concentration simulation of different size particles in a cyclone separator, Powder Technol. 183 (2008) 94–104. [9] L.Y. Hu, L.X. Zhou, J. Zhang, M.X. Shi, Studies on strongly swirling flows in the full space of volute cyclone separator, Chem. Eng. J. 51 (3) (2005) 740–749. [10] I. Jakštonienė, D. Serebryanskyy, P. Vaitiekūnas, Experimental research on the work of centrifugal filter when eliminating solid particles from clinker cooling system, The 8th International conference “Environmental engineering”: selected papers, May 19–20, 2011 Vilnius, Lithuania, Vol. 1., Vilniu, Technika 134–138 (2011). [11] L.C. Kenny, R.A. Gussman, Characterization and modeling of a family of cyclone aerosol preseparators, J. Aerosol Sci. 26 (1995) S777–S778. [12] S. Konoverskytė, P. Vaitiekūnas, Analysis of solid particle transfer in the C-30 cyclone, Aplinkos apsaugos inžinerija 10 (2010) 19–24. [13] LAND 28–98/M-08, Determination of the concentration of dust (solid particles). Weighted method, Lithuanian Ministry of Environment Protection Order Nr. 69, Vilnius, Lithuania1998. [14] C.E. Lapple, Processes use many collector types, Chem. Eng. 58 (1951) 144–151. [15] M.E. Moore, A.R. McFarland, Performance modeling of single-inlet aerosol sampling cyclones, Environ. Sci. Technol. 27 (1993) 1842–1848. [16] H. Mothes, F. Loffler, Prediction of particle removal in cyclone separators, Int. Chem. Eng. 28 (1988) 51–55. [17] I. Myers, R.L. Maynard, Polluted air — outdoors and indoors, Occup. Med. 55 (2005) 432–438. [18] P.A. Patterson, R.J. Munz, Cyclone efficiencies at very high temperatures, Can. J. Chem. Eng. 67 (1989) 51–55. [19] A. Pushnov, M. Berengarten, Ecological aspects of industrial cooling towers exploitation and it's influence to environment, J. Environ. Eng. Landsc. Manag. 19 (2) (2011) 158–166. [20] A. Raoufi, M. Shams, H. Kanani, CFD analysis of flow field in square cyclones, Powder Technol. 183 (2008) 1–9. [21] C.J. Stairmand, Design and performance of cyclone separator, Trans. Inst. Chem. Eng. 29 (1951) 356–373. [22] N. Strauss, Industrial Gas Cleaning, 15Pergamon Press, 1966. 56–63. [23] P. Vaitiekūnas, I. Jakštonienė, Analysis of numerical modeling of turbulence in a conical reverse-flow cyclone, J. Environ. Eng. Landsc. Manag. 18 (4) (2010) 321–328. [24] B. Zhao, Y. Su, J. Zhang, Simulation of gas flow pattern and separation efficiency in cyclone with conventional single and spiral double inlet configuration. Trans. Inst. Chem. Engineering. PA, Chem. Eng. Res. Des. 84 (2006) 1158–1165. [25] L.X. Zhoue, S.L. Soo, Gas solid flow and collection of solids in a cyclone separator, Powder Technol. 63 (1) (1990) 45–53.
Contents lists available at ScienceDirect
Powder Technology journal homepage: www.elsevier.com/locate/powtec
Experimental research on the dynamics of airflow parameters in a six-channel cyclone-separator Pranas Baltrėnas, Aleksandras Chlebnikovas ⁎ Vilnius Gediminas Technical University, Saulėtekio ave. 11, 10223 Vilnius, Lithuania
a r t i c l e
i n f o
Article history: Received 18 February 2015 Received in revised form 9 May 2015 Accepted 1 June 2015 Available online 9 June 2015 Keywords: Cyclone-separator Particulate matter Air flow Dynamics Efficiency
a b s t r a c t The dynamics of airflow parameters were researched in a six-channel cyclone (separator with a tangential flow). The dependencies of dynamic pressures in cyclone's channels and of cleaning efficiency on an airflow inlet velocity were analyzed. Standard and upgraded semi-rings were used in the cyclone's internal structure. Changes in cleaning efficiency at different inlet concentrations of glass and technical salt particles were analyzed. Research results were obtained at inflow velocities of 10.9 m/s to 21.9 m/s. The highest cleaning efficiency was 97.3%. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Cyclone-separators are unrivaled leaders among devices used to remove particulate matters (PMs) from a dirty airflow. In terms of their price, simplicity and operation, cyclone devices have been unparalleled. Cyclones will stay competitive in the market for a long time due to their unique structure, absence of any moving parts and filtering surfaces that require regular service, comparatively low aerodynamic resistance and high efficiency [2,4,14,25]. Air cleaning devices of the cyclone type are most often used in industries which are characterized by high volumes of polluted air emissions. Most devices are suitable for the separation of PM of large-scale dispersion in multi-phase flows. The work of the common hollow filter is based on the widespread principle of separation of PM with the appearance of centrifugal forces that are influenced by the turbulent airflow inside the body of the equipment. The effectiveness of the mentioned cyclone-separator is from 75 to 85%; the air is cleaned from the PMs that are more than 20 μm in diameter. The structure of the upgraded six-channel cyclone is designed for the separation of dispersed fine PM, up to 10 μm in size [1,9,10,24]. Although cyclone-separators operate under the influence of centrifugal and gravitational forces, their cleaning efficiency depends on the characteristics of inlet airflow [11]. When airflow parameters – velocity and pressure – change, aerodynamic forces in the cyclone's system also change. The aim of this research is to evaluate changes ⁎ Corresponding author. Tel.: +370 63693865. E-mail addresses: [email protected] (P. Baltrėnas), [email protected] (A. Chlebnikovas).
http://dx.doi.org/10.1016/j.powtec.2015.06.005 0032-5910/© 2015 Elsevier B.V. All rights reserved.
in airflow pressure and cleaning efficiency at different airflow inlet velocities in a six-channel cyclone. The cyclone's basic structure has remained unchanged for more than a hundred years. There are two main types of cyclone structures: a direct cyclone and a reverse-flow cyclone [21,22]. However, the most frequently used are reverse-flow cyclones, which are also divided into axial flow and rotational flow cyclones with a tangential gas inlet [3,4,10,15]. Patterson and Munz [18] studied the effect of PM concentration (of up to 235.2 g/m), fluid temperature (from 300 K to 2000 K) and gas inlet velocity (from 3 m/s to 42 m/s) on cyclone's collection efficiency and determined that an increase in PM concentration had improved the efficiency of collection, in particular at a high temperature [8,23]. Turbulent mixing of the flow may be associated by a free flow of gas kinetic theory. By this case the velocity of air molecules distributed by the normal law. According to this law, in the distribution of velocities in the gaps between the curved half-rings is defined by dependence [10]: σ¼
2ρaϑ2ϕ δ3r
2
¼
2ρak
ð1Þ
δ2 r2 2
where: σ — standard deviation of velocity variation, σ ¼ 6kρaϑ ;ϑ — δ3 r2 velocity of airflow, m/s, ρ — air density, kg/m3, a — distance between the aspects of half-rings, m, δ — diameter ofair molecules, m, r — 1ffi ϑ2 exp − channel of cyclone radius, also f ðν Þ ¼ pffiffiffiffi 2 : 2σ 2πσ Cyclone-separators are used as drying machines, reactors and catalysts [5,12,19]. Although recent patented cyclones differ by
P. Baltrėnas, A. Chlebnikovas / Powder Technology 283 (2015) 328–333
their originality, they still retain the main distinguishing feature of these devices — a turbulent flow [7,20]. Highly efficient cyclones and filters are used to reduce dustiness in industry [17]. Cyclone's two most important working parameters, collection efficiency and pressure drop, largely depend on the concentration of dispersed PM [6,23]. The work of channel cyclone-separator is based on centrifugal forces and additionally occurring filtration process. Due to the interaction between inlet flow from the (peripheral) channel coming next and the flow following the direction towards the axis of the cyclone along the channel (transit), additional filtration takes place. Air flow is filtered through the peripheral flow — a curtain appearing behind the separation zone of curved half-rings (quarter-rings), which raises cleaning efficiency up to 15%. The aim of this work is to determine six-channel cyclone's airflow parameters and analyze air cleaning efficiency at different dispersities of PM. 2. Materials and methods Experiments were carried out on an upgraded six-channel cyclone located at the Environmental Technologies Laboratory of Vilnius Gediminas Technical University (VGTU). A schematic diagram of the cyclone's basic design is presented in Fig. 1. The cyclone's air cleaning efficiency was determined by introducing into it ground PM of glass and technical salt of up to 50 μm and 20 μm. The standard dynamic pressure (1/2ρV2) of the airflow was measured in the respective channels of the cyclone's structure using a Pitot (Prandtl) dynamic tube connected to a multifunctional measurement instrument Testo-400. Measurements were made at the following characteristic points (Fig. 1) of the separation chamber inside the structure: at the beginning, in the middle and at the end of the channel, and also at the intermediate points (between the beginning and the middle, and between the middle and the end of the channel). The dynamic tube was inserted at the points directly through special holes made in the separation chamber cover plate. The pressure setting for the reliability of results was performed over
Fig. 1. Cyclone cross-section with semi-ring at position I: I–VI — cyclone channels; R1–R5 — semi-rings; 1–45 — dynamic pressure measurement points; a — segment circular spacings.
329
all cross-sections of each cyclone-separator channel, by making a 9-point network (on the periphery boundary layer, in the middle of each channel cross-section, and on the inner wall of the channel). Dynamic pressures were measured by using standard curvilinear semi-rings and upgraded semi-rings with windows. All pressures were measured by changing the airflow yield supplied by the channel ventilator. The supplied airflow was changed with a lever of the ventilator's control unit. The aim of this regulation was to achieve the optimum pressure for the cyclone's highest cleaning efficiency. Fig. 1 shows the cyclone's internal structure and the location of semirings. The positions of semi-rings (only R1, R3, R5) are regulated as follows: the rings are shifted by 10 mm to the left in position II and by 10 mm to the right in position III, but other semi-rings remain in the same places. Regulation was made only in the x-axis direction. In order to determine the best application of the air cleaning device concerned in a particular industry, PMs of different density, bulkiness and dispersity were selected. The physical density of glass (SiO2) PM was 2670 kg/m3 and that of technical salt (NaCl) was 2210 kg/m3. For determining the cyclone's efficiency a high-pressure compressor producing an air pressure of up to 6 bars was used during experiments. The compressor was connected to a feeding nozzle which sucks in and delivers dust PM directly to the cyclone's air inlet, 200 mm in diameter, thus forming a dispersed two-phase flow. An airflow supply ventilator (RUCK RS200L, power 190 W) was installed in front of the six-channel cyclone and connected to an air duct, 2 m long. Pressure changes in the system were registered with a differential pressure meter (PCE Instruments DSM-1). A possible error for stream unevenness that might occur due to a non-uniform distribution of PM within the entire stream of the inlet air was eliminated. Glass vessels for specimens, scales for weighing specimens (error ± 0.1 g), and a second meter Sekonda for registering air inlet time (error ± 0.2 s) were used during tests. Experimental research on the cleaning efficiency of the six-channel cyclone was carried out according to the Methodology “LAND 28-98/ M-08 Determination of the concentration of dust (PM). Weighted method.” [13]. Before being tested, the specimens of glass and technical salt PM were dried up to a constant weight in a laboratory electric stove at a temperature of 100 °C. PMs were ground with the grinder Retsch RM200. Only PMs of up to 20 and up to 50 μm fraction were used for the experiments. Studies using these PM sizes allow us to investigate the multi-cyclone cleaning efficiency of fine PM and compared with conventional centrifugal cyclone efficiency. The remaining portion was sieved once again. During experiments the ambient air temperature varied from 20.1 °C to 22.2 °C, while relative humidity reached 52%. A two-phase airflow tangentially enters the first channel through the inlet (1) in the separation chamber. The channel is delimited by a peripheral wall and the first curvilinear semi-ring. The airflow is thus distributed within channels of different curvature and is filtered through spacings between semi-rings. PMs are precipitated on the bottom of the six-channel cyclone by a turbulent flow created by centrifugal forces. On entering segment circular spacings (9) they fall down and accumulate in the cyclone's hopper (10). After passing through all six channels clean air leaves the system through the airflow outlet (8). Dusty air is filtered in the active zone of channel spacings. Filtration also takes place during the interaction of PM when they coagulate and separate from the airflow (Fig. 1, Fig. 2). Experiments were carried out by using the standard (without a window) and upgraded (with a window) structure of curvilinear semi-rings. In the case of upgraded structure a polluted airflow can be returned to the previous channels. Thus, this structural solution has extended the time of filtration of a dirty airflow in cyclone's channels. The experimental research on dynamic pressures analyzes cases involving the first three external semi-rings with cut windows of 10 mm curved width. Tests on air cleaning efficiency were carried out by using upgraded semi-rings: only the first (R1) upgraded semiring; the first two (R1 and R2) upgraded semi-rings, and the first three (R1, R2 and R3) upgraded semi-rings (Fig. 2).
330
P. Baltrėnas, A. Chlebnikovas / Powder Technology 283 (2015) 328–333
Fig. 2. Structure of six-channel cyclone (a), standard (b) and upgraded (c) semi-rings: 1 — airflow inlet; 2 — diffuser; 3–7 — semi-rings; 8 — airflow outlet; 9 — cyclone bottom; 10 — conical hopper; 11 — standard semi-ring; 12 — upgraded semi-ring; 13 — window.
For the experiments at issue scientific sources most often present works carried out at concentrations varying between 1.0 g/m3 and 25 g/m3. The concentrations of PM in dirty air ranged from 500 mg/m3 to 10 g/m3. Multichannel cyclone-separators may be applied as air cleaning device to contaminate PM near workplaces. Also, as well as industrial facilities, device to clean polluted air (gas) flow in industrial plants. For this purpose, experimental research was carried out used a rather wide concentration range of contaminated air. The concentration of PM mass in the air is calculated according to the standard LAND 28/M-08 (1998): ðm2 −m1 Þ ; C¼ V0
ð2Þ
where: C — dust concentration, mg/m3; m1 — mass of the filter without PM, mg; m2 — mass of the filter with PM, mg; V0 — volume of the air sucked through the filter and recalculated under normal conditions, m3. As experiments used dust of different nature, i.e., glass and technical salt, after testing specimens of type I the entire system – hoses and cyclone structure – was cleaned by blowing a high-speed flow through it.
3. Results and discussion Average dynamic pressures inside cyclone's channels were registered at the maximum (≈22 m/s), medium (≈14 m/s) and minimum (≈ 11 m/s) inflow velocities by determining three positions of semirings. The results of experimental research on dynamic pressures in the optimum case – position I – are presented in Figs. 3 and 4. 3.1. Dynamic pressure with standard semi-rings Dynamic pressure was measured to determine the dependence of aerodynamic characteristics on the inflow velocity in the cyclone. Standard dynamic pressures were measured during these experiments. The measurement results show the dependence of two-phase flow distribution in cyclone on turbulence flows of different levels. All selected velocities created a high-level flow of turbulence. Turbulence of fluid flow is measured by the Reynolds number, which is composed of the inlet velocity and the height of channel. At the maximum velocity (≈ 22 m/s) — Re ≈ 4.35 × 105, at the medium (≈ 14 m/s) — Re ≈ 2.8 × 105 and at the minimum (≈ 11 m/s) — Re ≈ 2.2 × 105. The results of dynamic pressure measurements inform
Fig. 3. Effect of inflow velocity on the average dynamic pressure in the cyclone channels with standard semi-rings in position I.
P. Baltrėnas, A. Chlebnikovas / Powder Technology 283 (2015) 328–333
331
Fig. 4. Effect of inflow velocity on the average dynamic pressure in the cyclone channels with upgraded semi-rings in position I.
us about aerodynamic processes and changes in a turbulence level in the cyclone. The highest values of dynamic pressure were obtained at the maximum inlet velocity of 21.9 m/s. When standard semi-rings are in position I, the value is equal to 173.4 Pa. Meanwhile values obtained with standard semi-rings in positions II and III are 171 Pa and 174.6 Pa, respectively. It can, therefore, be stated that the highest velocity (axial) of the sixth channel is obtained when standard semi-rings are in position III. According to previous tests, cleaning efficiency decreases in positions II and III. This might have resulted from inappropriate transit/peripheral flow ratio. The optimum ratio is achieved in position I. As the obtained results show, when an airflow passes from one cyclone channel to another, the average dynamic pressure changes, on average, by 1.15 to 1.45 times as a result of a decrease in the channel's cross-section and a simultaneous increase in dynamic pressure and velocity. When air flows from the second channel to the third one a difference in pressures is bigger than mentioned above — 1.88 times at the minimum airflow in position I. In positions II and III this pressure difference is smaller equaling, on average, 1.3 times. The smallest measured average dynamic pressure, 12.4 Pa, was recorded in the first channel at the lowest inflow velocity, 10.9 m/s, and in position I of standard semi-rings. At the medium and maximum inflow velocities, average dynamic pressures in all the three positions in question were 73.5 Pa and 110.4 Pa respectively (Fig. 3). The analyzed results show that when the arrangement of semi-rings and also spaces among them are changed, the size of the cross-section used for the filtration of dirty air changes, thus increasing he dynamic pressure of air. As Fig. 3 shows, dynamic pressures are distributed quite evenly and increase in a linear manner from the cyclone inlet (entry to the first channel) towards the cyclone's center line (the cyclone's channel 6) and air outlet. Dynamic pressure is influenced by the change of inlet velocity and the cross-sections of channels. 3.2. Dynamic pressure with upgraded semi-rings The maximum value of dynamic pressure, 180.3 Pa, was recorded in semi-ring position III in the cyclone's channel 6 when three external semi-rings with windows cut in them were used. Meanwhile values obtained using upgraded semi-rings in positions I and II stand at179.5 Pa and 177 Pa, respectively. It can, therefore, be stated that the highest velocity (axial) of the sixth channel is achieved when upgraded semi-rings are in position III. When an airflow passes from the second channel to the third a difference in pressures is bigger reaching 1.83 times at the minimum airflow in position III. In positions I and II
this pressure difference is smaller and equals, on average, 1.6 times (Fig. 4). The tendency of pressure distribution remains similar to that of standard semi-rings. Optimum position I was selected for tests on cleaning efficiency. Using upgraded semi-rings, the minimum dynamic pressure of 17 Pa was recorded in the first channel at an inflow velocity of 10.9 m/s (Fig. 4). At the medium and maximum inflow velocities, average dynamic pressures in all the three positions in question were 79 Pa and 116.6 Pa respectively. 3.3. Cleaning efficiency with standard semi-rings Results obtained using the PM of glass and of technical salt are comparable as the PM specimens were prepared according to the same methodology and were used in experimental research in the same manner. The highest cleaning efficiency, 97.3%, was achieved at 10 g/m3 concentration of PM, 50 μm in size. The lowest efficiency, 94.3%, was registered at 500 mg/m3 concentration of the same fraction. When PM dispersity changes to 20 μm the cyclone's air cleaning efficiency falls, on average, by 1.5 times. The maximum cleaning efficiency, 74.1%, achieved at standard internal geometry was also recorded at the maximum inlet concentration of 10 g/m3. The minimum efficiency, 48.5%, was established at an initial concentration of 500 mg/m3 (see Fig. 5). Tests were carried out using curvilinear semi-rings with windows. Due to this upgrading a polluted airflow returned to the previous channel and additional part of PM could precipitate and enter the hopper via segment circular spacings (see Fig. 2). The highest efficiency of removing technical salt PM (up to 50 μm) from an airflow is 94.9% at the highest inlet concentration of 10 g/m3. An analysis of the degree of catching dispersed fine PM of up to 20 μm shows that the efficiency of catching glass PM is higher than the efficiency of catching technical salt PM. The average difference in cleaning efficiency of the PM of these types accounts for nearly 3%. The maximum air cleaning efficiency using standard semi-rings reaches 72.3%, which is by 2.5% smaller than in the case of glass PM. The tendency of dependencies of air cleaning efficiency on concentration remains the same — air cleaning efficiency increases when concentration increases. After completing experimental research on the effect of PM concentration on gas flow and on separation of PM using cyclones of a standard structure, researchers Mothes and Loffler [16] determined that the centrifugal force inside the cyclone had decreased at a high concentration of PM. As studies on efficiency carried out by Mothes and Loffler [16] show, cyclone's performance improved at a high concentration due to the agglomeration of PM [10].
332
P. Baltrėnas, A. Chlebnikovas / Powder Technology 283 (2015) 328–333
Fig. 5. Effect of type and size of PM on the air cleaning efficiency of the six-channel cyclone with standard semi-rings.
3.4. Cleaning efficiency with upgraded semi-rings Air cleaning efficiency was determined using upgraded semirings, i.e., the first (R1), the first two (R1 and R2) and the first three (R1, R2 and R3) upgraded semi-rings. The remaining semi-rings in each case were standard. To analyze the degree of cleaning of the six-channel cyclone with upgraded semi-rings (internal geometry) dispersed fine PM, up to 20 μm in size, were used. The maximum efficiency in removing glass PM from an airflow was 78.4%. It was obtained using the first three external upgraded semi-rings at the highest inlet concentration of 10 g/m3 . It should be noted that when the number of upgraded semi-rings (with windows in the first and second semi-rings, and in the first one only) is reduced and the inlet concentration of glass PM is decreased, cleaning efficiency also decreases. With upgraded semi-rings, the lowest efficiency, 47.7%, was recorded at an inlet concentration of 500 mg/m3 of technical salt, and using only the first (external) upgraded semiring (Fig. 6).
Cleaning efficiencies recorded for upgraded curvilinear semi-rings insignificantly differ from the analyzed efficiencies while using standard semi-rings for glass PM, but are smaller. It can be noted that the introduced upgraded semi-rings enhance air cleaning efficiency and when their number increases, i.e., all the three upgraded semi-rings are used, the maximum cleaning percentage reaches 76.9%, but efficiency is by 1.5% lower than in the case of the glass PM at issue. When spaces between upgraded semi-rings and the peripheral wall decrease, active dynamic forces increase, which leads to the change of other parameters. Changed dynamic pressures affect the efficiency of dust PM catch when PMs are taken to the hopper through segment circular openings in the bottom. Air cleaning efficiency increases with the inlet concentration of PM increasing under the influence of increasing centrifugal forces and filtration effects. Air cleaning efficiency in the presence of higher centrifugal forces (at an airflow velocity above 22 m/s) was not analyzed. However, this dependence is likely to change according to asymptote [10]. The particularity of the six-channel cyclone-separator lies in the fact that a
Fig. 6. Effect of type and size of PM on the air cleaning efficiency of the six-channel cyclone with upgraded semi-rings.
P. Baltrėnas, A. Chlebnikovas / Powder Technology 283 (2015) 328–333
dynamic gas-dust layer forms in the spaces between curvilinear semirings when airflow passes through channels. When a dirty airflow is filtered through this layer, part of PM is caught and at the same time cleaning efficiency is enhanced [16]. One of the parameters of the cyclone-separator, which defines the energy loss during cleaning process, is the aerodynamic resistance. As known the aerodynamic resistance of traditional cyclone such as CN-15, which is similar to tested multi-channel cyclone by capacity is equal to 700 Pa approximately. Maximal aerodynamic resistance of multi-channel cyclone equal to 450 Pa, using upgraded semi-rings at an airflow velocity is 22 m/s. Under flow distribution rates, resistance is decreasing because of changes in the flow paths of turbulent flows, which makes less energy losses. An analysis of the removal of the PM concerned from an airflow leads to the conclusion that the air cleaning device in question is not so efficient in removing the PM of technical salt from a polluted flow, which might be determined by the specific characteristics of these PMs, such as density, form, and degree of agglomeration processes. The centrifugal force effect increases at higher speeds. The PMs are mostly taken to the peripheral wall, precipitate on the bottom of cyclone, and after entering segment circular spacings fall and accumulate in the cyclone's hopper. 4. Conclusions Air cleaning efficiency in a six-channel cyclone-separator improves when the inflow velocity in cyclone increases from 11 m/s to 22 m/s. Glass and technical salt particulate matter is effectively removed from an airflow at the optimal inflow velocity in cyclone of 22 m/s using upgraded semi-rings. The greatest efficiency, 78.4%, is reached when dispersed fine glass PMs of up to 20 μm are used. Air cleaning efficiency of 74.1% is determined using standard semi-rings when glass PMs of up to 20 μm are introduced at 10 g/m3 concentration and at 22 m/s inflow velocity. The upgraded semi-rings have improved cleaning efficiency by about 4 to 5%. The new structural solution with upgraded semi-rings extended the time of filtration of a dirty fluid flow in cyclone's channels, because a polluted airflow can be returned to the previous channels. The maximum dynamic pressure of 179.5 Pa was registered while using upgraded semi-rings at the highest inflow velocity of 22 m/s in the sixth channel of cyclone in position I. When the position of curvilinear semi-rings changes from position I to positions II and III, pressure decreases due to a change in turbulent flow trajectories causing additional energy growth. A comparison of dynamic pressures in a sixchannel cyclone-separator has shown that they are significantly (approximately by 10–15%) bigger when upgraded semi-rings are used. After evaluating the results of pilot research on multichannel cyclone-separator, it can be concluded that this device is suitable for removing gaseous effluents from small dispersion particulate matter, up to 20 μm in size, into the atmosphere.
333
References [1] S. Altmeyer, V. Mathieu, S. Jullemier, P. Contal, N. Midoux, S. Rode, J.P. Leclers, Comparision of different models of cyclone prediction performance for various operating conditions using a general software, Chem. Eng. 43 (2004) 511–522. [2] A. Avci, I. Karagoz, Effect of flow and geometrical parameters on the collection efficiency in cyclone separators, J. Aerosol Sci. 34 (2003) 937–955. [3] P. Baltrėnas, J. Morkūnienė, P. Vaitiekūnas, Mathematical simulation of solid particle dispersion in the air of Vilnius city, J. Environ. Eng. Landsc. Manag. 16 (1) (2008) 15–22. [4] S. Bernardo, M. Mori, A.P. Peres, R.P. Dionisio, 3-D computational fluid dynamics for gas and gas–particle flows in a cyclone with different inlet section angles, Powder Technol. 162 (2006) 190–200. [5] J. Bradulienė, S. Vasarevičius, Experimental investigation into reduction of gravel road dustiness using “SAFECOTE” solution and its mixture with CaCl2, J. Environ. Eng. Landsc. Manag. 18 (2) (2010) 145–153. [6] L. Curtis, W. Rea, P. Smith-Willis, E. Fenyves, Y. Pan, Adverse health effects of outdoor air pollutants, Environ. Int. 32 (2006) 815–830. [7] J. Dirgo, D. Leith, Performance of theoretically optimized cyclones, Filtration and Separation 22 (1985) 119–125. [8] W. Gujun, S. Guogang, X. Xiaohu, S. Mingxian, Solids concentration simulation of different size particles in a cyclone separator, Powder Technol. 183 (2008) 94–104. [9] L.Y. Hu, L.X. Zhou, J. Zhang, M.X. Shi, Studies on strongly swirling flows in the full space of volute cyclone separator, Chem. Eng. J. 51 (3) (2005) 740–749. [10] I. Jakštonienė, D. Serebryanskyy, P. Vaitiekūnas, Experimental research on the work of centrifugal filter when eliminating solid particles from clinker cooling system, The 8th International conference “Environmental engineering”: selected papers, May 19–20, 2011 Vilnius, Lithuania, Vol. 1., Vilniu, Technika 134–138 (2011). [11] L.C. Kenny, R.A. Gussman, Characterization and modeling of a family of cyclone aerosol preseparators, J. Aerosol Sci. 26 (1995) S777–S778. [12] S. Konoverskytė, P. Vaitiekūnas, Analysis of solid particle transfer in the C-30 cyclone, Aplinkos apsaugos inžinerija 10 (2010) 19–24. [13] LAND 28–98/M-08, Determination of the concentration of dust (solid particles). Weighted method, Lithuanian Ministry of Environment Protection Order Nr. 69, Vilnius, Lithuania1998. [14] C.E. Lapple, Processes use many collector types, Chem. Eng. 58 (1951) 144–151. [15] M.E. Moore, A.R. McFarland, Performance modeling of single-inlet aerosol sampling cyclones, Environ. Sci. Technol. 27 (1993) 1842–1848. [16] H. Mothes, F. Loffler, Prediction of particle removal in cyclone separators, Int. Chem. Eng. 28 (1988) 51–55. [17] I. Myers, R.L. Maynard, Polluted air — outdoors and indoors, Occup. Med. 55 (2005) 432–438. [18] P.A. Patterson, R.J. Munz, Cyclone efficiencies at very high temperatures, Can. J. Chem. Eng. 67 (1989) 51–55. [19] A. Pushnov, M. Berengarten, Ecological aspects of industrial cooling towers exploitation and it's influence to environment, J. Environ. Eng. Landsc. Manag. 19 (2) (2011) 158–166. [20] A. Raoufi, M. Shams, H. Kanani, CFD analysis of flow field in square cyclones, Powder Technol. 183 (2008) 1–9. [21] C.J. Stairmand, Design and performance of cyclone separator, Trans. Inst. Chem. Eng. 29 (1951) 356–373. [22] N. Strauss, Industrial Gas Cleaning, 15Pergamon Press, 1966. 56–63. [23] P. Vaitiekūnas, I. Jakštonienė, Analysis of numerical modeling of turbulence in a conical reverse-flow cyclone, J. Environ. Eng. Landsc. Manag. 18 (4) (2010) 321–328. [24] B. Zhao, Y. Su, J. Zhang, Simulation of gas flow pattern and separation efficiency in cyclone with conventional single and spiral double inlet configuration. Trans. Inst. Chem. Engineering. PA, Chem. Eng. Res. Des. 84 (2006) 1158–1165. [25] L.X. Zhoue, S.L. Soo, Gas solid flow and collection of solids in a cyclone separator, Powder Technol. 63 (1) (1990) 45–53.