Title of Invention	A CYCLONE SEPARATOR-BASED AIR FILTER AND METHODS THEREOF
Abstract	Title of the Invention: A CYCLONE SEPARATOR-BASED AIR FILTER AND METHODS THEREOF A new cyclone separator-based air filter (Fig. 2 and Fig. 5) operating in a dusty environment has been simulated, this design shows reduced pressure drop and increased dust collection efficiency over conventional systems, the flow direction was made tangential so that the pressure drop can be reduced, a radically new concept in air filtration which does not require polyurethane foam-based main filter at all, in other words, some of the so-called pre-filter designs simulated here show a dust collection efficiency of 100% for particles of 4 \|im and higher, this implies that this cyclone separator based air filter can do the job of the pre-filter and main filter configuration thus this radical concept can significantly lower the cost of the air filter unit, and more importantly, the running costs also are reduced to zero, since there is no replaceable part at all.

Title of Invention

A CYCLONE SEPARATOR-BASED AIR FILTER AND METHODS THEREOF

Abstract

Title of the Invention: A CYCLONE SEPARATOR-BASED AIR FILTER AND METHODS THEREOF A new cyclone separator-based air filter (Fig. 2 and Fig. 5) operating in a dusty environment has been simulated, this design shows reduced pressure drop and increased dust collection efficiency over conventional systems, the flow direction was made tangential so that the pressure drop can be reduced, a radically new concept in air filtration which does not require polyurethane foam-based main filter at all, in other words, some of the so-called pre-filter designs simulated here show a dust collection efficiency of 100% for particles of 4 |im and higher, this implies that this cyclone separator based air filter can do the job of the pre-filter and main filter configuration thus this radical concept can significantly lower the cost of the air filter unit, and more importantly, the running costs also are reduced to zero, since there is no replaceable part at all.

Full Text	FIELD OF THE INVENTION The present-invention relates to a cyclone separator- based Air filter for separating particles from air. BACKGROUND OF THE INVENTION Dust consists of tiny solid particles carried by air currents. These particles are formed by disintegration or fracture process, such as grinding, crushing, or impact. Dust is generated by a wide range of manufacturing, domestic, and industrial activities. Construction, agriculture, and mining are among the industries that contribute most to atmospheric dust levels. A wide range of particle sizes is produced during a dust generating process. Particles that are too large to remain airborne settle while others remain in the air indefinitely. The escape of dust particles into the workplace atmosphere is undesirable. Excessive dust emissions can cause both health and industrial problems: • Health hazards - Occupational respiratory diseases - Irritation to eyes, ears, nose, throat and skin • Damage to equipment • Impaired visibility • Unpleasant odors • Problems in community relations PRIOR ART OF THE INVENTION Present air pre-filters are based on a design that causes the entering air to make 180 degree turn before it exits the device. This gives reasonably good dust collection efficiency, however the pressure loss is high. There have been several attempts in the past to reduce the dust collection efficiency, but there has not been any significant reduction possible. However, these air-filters have certain limitations as follows: The pressure drop in present cyclone separator-based air filters is the high across the entire device. This also affects the internal combustion engine’s performance in terms of reduced volumetric efficiency, which in turn affects the power generated and their dust-holding and storage capacities, servicing facilities, and maintenance periods have been sacrificed. In this section a brief review of the relevant literature is presented. Zhou and Soo (1990) worked on gas solid flow and collection of solids in a cyclone separator. A method of improving the performance of cyclone separator with central body was illustrated by them. They conducted Laser Doppler Velocimetry (LDV) measurements to compare and validate computations of flow and pressure drop in a cyclone separator by an analytical solution of constant viscosity and a solution based on the k- e turbulence model. The detailed measurements of gas flow field in cyclone separator by LDV and numerical simulation of this flow field by using k-s turbulence model and SIMPLE (Semi Implicit Method for Pressure Linked Equations) algorithm gave a complete picture of whole flow field. Results explain the significance of solid body core of vortex and the difficulty of k-e method when applied to vortex motion. A theoretical analysis was performed by Soo on the gas flow field, dust dispersion and dust collection in a cyclone, based on the fundamentals of multiphase fluid dynamics. Zhou et al. developed numerical modeling and LDV measurements of gas-solid flows, and also proposed an innovative cyclone separator with a special central body to remarkably reduce the pressure drop. Avci and Karagoz (2003) presented their work on the effects of flow and geometrical parameters on the collection efficiency in cyclone separators. Predictions of the collection efficiency and the cut-off size for a cyclone are difficult problems since the behavior of the mixture inside the cyclone are affected by many parameters, such as flow characteristics, particles and mixture properties and external forces on particles. Therefore, detail knowledge of physical phenomena and appropriate interpretation of experimental results are required in order to be successful in modeling the cyclone flow. A mathematical model has been developed by defining average flow and geometrical parameters, and applying conservation equations to the cyclone flow. Despite the simple approximations to the friction coefficient and the vortex number, the computed results are in good agreement with experimental values, showing that the proposed model is successful for all type of cyclones comparing with other models in literature. The following conclusions can also be drawn from the present study. • Collection efficiency depends on the flow regime which is affected by the flow parameters such as inlet velocity, temperature, viscosity and surface roughness. In big cyclones, flow is generally fully turbulent and the effects of flow parameters decrease with Reynolds number, whereas, in small cyclones or at low velocities, flow is laminar and the effects of flow parameters increase up to a certain point and than start to decrease after that. The characteristics of transitional flow regime are required to be investigated in detail so that more efficient cyclones would be designed. Prediction of the friction coefficient seems to be very important and it plays an important role on the present model. Smooth surfaces were assumed for comparison with experimental values in this model and this may cause some deviation of the computed results from the experimental values. Decreasing in surface resistance which eventually leads to decrease of velocity is an important way to high efficiency due to increasing in natural vortex length and residence time inside cyclone. The diameter of cone apex affects acceleration and consequently collection efficiency and pressure losses. Very small cone apex diameter is not preferable from the point of view of pressure losses and vortex length, whereas very big cone apex diameter is not relevant for collection efficiency. Experimental results show that, keeping other geometrical ratios constant, cyclone height affects collection efficiency positively up to a certain point. These behaviors were also verified by the present model. Although it is possible to say that the optimum cyclone length may be increased if the friction coefficient is decreased, this length also depends on the geometry and flow conditions. Therefore, further studies are required to be performed on this subject. Analysis of experimental and theoretical results has shown that the inlet width has very big influence on the cut-off size in some cases of geometry and operational conditions, whereas it has minor importance otherwise. Therefore, additional studies are required in order to investigate the effects of the inlet width on the cyclone performance. Another interesting parameter is the inlet height, which may be important from the point of view of frictional surfaces. Although the inlet height is only included in the definition of ad, it is easy to take account of this parameter by defining the hydraulic diameter, which may cause some alteration of the results. • Due to variation of velocity, hydraulic diameter and relative roughness in cyclone, it is difficult to compute Reynolds number and to determine dominant flow regime. Therefore, one of the problems is the prediction of the friction coefficient correctly. In this model the friction coefficient was computed with the reference to the inlet conditions for practical usage. Fraser et. al. (1997) presented work on computational and experimental investigations in a cyclone dust separator. They simulated three-dimensional turbulent flow in a model cyclone using PHOENICS code and experimental studies carried out using Laser Doppler Anemometry (LDA) system. The experimental results were used to validate computed velocity distributions based on the standard and a modified k-e model. Simulation The standard k-e model was found to be unsatisfactory for the prediction of flow field inside the cyclone chamber. By considering the strong swirling flow and the streamlined curvature, a k-e model modified to take account of Richardson number, provided better velocity distributions and better agreement with the experimental results. The following form of Richardson number has been used in their study: Ri=K2 / Є2 u/ r2 ∂(ru)/ ∂z where r is the radial distance across the cyclone and z is the distance down the cyclone. The standard k-e turbulence model assumes that the turbulence is isotropic, which is not the case for swirling flows. The Richardson number is based on several terms including the swirl gradient, which allows for the mixing length to increase or decrease as the gradient becomes negative or positive. When this is used to modify the dissipation it makes a first attempt at making the turbulence non-isotropic and hence producing a better simulation. The simulation was carried out assuming incompressible fluid flow on the basis that the maximum velocity vector using the measured swirl and axial components was less than 20 m/s. Experiments The test model was constructed in Perspex and based on the high efficiency Stairmand cyclone design. Air was used as the working fluid and is drawn into the reservoir chamber via a two-stage axial fan at average ambient conditions. The air is supplied to the cyclone chamber through the piping system and from the cyclone chamber the air is exhausted to the atmosphere. The experiment is conducted at a constant flow rate. For the purpose of dust loading and seeding particles for LDA measurement, a specially prepared milk powder in the range of 3-50 μm in diameter is used in the feeder. LDA system is optimized for particle sizes in the range of 5-10 μm and these particles are not separated by the cyclone and so exit via the vortex outlet while the larger particles are separated and drop out through the bottom of the cyclone. Although the flow has a large centrifugal effect it has been estimated that the smaller particles follow the fluid flow with only minor variations. Results The air flow inside the cyclone chamber is non-axisymmetric and with high swirl and streamline curvature. The flow rotates strongly at the axis towards the outlet tube and reduces rapidly at the cyclone wall. In the conical chamber, the swirl velocity reduces rapidly in the core region compared to the cylindrical chamber. The results of the experimental and modified k-e model show that the vortex motion inside the cyclone chamber consists of two parts, the core of the forced vortex at the centre where velocity increases with radius and an outer free vortex where velocity reduces with increasing radius. Maximum velocity occurs at the centre and reduces rapidly with radius. Avci and Karagoz (2000) presented their work on the mathematical model of two-phase flow in tangential cyclone separator using parameters such as effects of cyclone geometry, surface roughness and concentration of particles. The critical diameters, fractional efficiencies and pressure losses are calculated under the assumptions that each phase has the same velocity and acceleration in the spiral motion of flow, a relative velocity occurs in the radial direction and drag coefficient remains constant under certain conditions. The results obtained under these assumptions are compared with their experimental and theoretical values, and a very good agreement is observed with experimental values. Experiments prove that efficiency increase with increase of cyclone height. But it should be a limit for the height due to friction and separation from the surfaces and this should be analyzed experimentally. Molerus and Gluckler (1996) developed a cyclone separator with new design for gentle separation of particles from gas phase (with respect to breakage and attrition of the particle) for usual solid loadings and which is suitable for downer systems. Test experiments were carried out to evaluate quality of performance that gave overall collection efficiency of up to 99.99 % for particles in the range of 10 jam. Any change in direction of a gas-solid flow creates a change in momentum of the particles. This leads to collisions and/or sliding along walls, which causes breakage and attrition. Therefore the new separator should operate with as little change in momentum of the solid stream as possible. A new cyclone separator with a specific inlet device was designed, tested and modified until the performance was satisfactory. The cyclone consists of a frustum of a circular cone and a cover with conventional vortex finder in an inclined arrangement. The gas-solid suspension is fed into the cyclone vertically downward through the cover with an angle of 30 ° to the main axis of the cone. The inlet device is shaped in such a way that it fits with the circular contour of the cone edge and lines up with the wall of the cone. Therefore the feed stream is homogeneously accelerated without any direct collision with the cone wall. Several experiments were carried out to evaluate the quality of performance of the new cyclone. The overall collection efficiencies are satisfactory and independent of the solid loading as long as the solid mass flux is not too high. For solid mass fluxes exceeding a certain value, the separated particles will not flow out of the cyclone fast enough. They will be mixed back into the gas stream to a certain extent which results in a less efficient separation. This effect also occurs with conventional cyclones and can he avoided by correct dimensioning of the solid outlet of the cyclone. However, the collection efficiencies of normally more than 99%, in a range of the solid loading 1 Li et al. (1996) presented results on the study of solid-gas separation mechanism of cyclone with impulse excitation. On the basis of experimental study of the performance of cyclone impulse electrostatic precipitation (CIESP), the solid-gas separation mechanism of CIESP is presented by them. CIESP is a new technique, which combines merits of cyclone precipitation and electrostatic precipitation. The basic idea was to use high voltage impulse electrostatic field in cyclone, when airflow with particles passes, the particles are separated from gas by centrifugal force and impulse electrostatic force. By analyzing the charging and acting forces on particles, the theoretical formula of particles migration velocity and grade efficiency are concluded and verified with experiments. An experimental investigation was performed by Giulio Solero and Aldo Coghe (2002) on a laboratory scale model of a gas cyclone separator inserted in a closed/open loop test facility built up to study the flow field of both the gas and solid phase. The cyclone is made entirely by Plexiglas in order to allow optical access for velocity measurements through Laser Doppler Velocimetry. Three velocity components of the gas phase were measured inside the various regions of the apparatus (tangential inlet chamber, cylindrical body and exhaust duct) and a preliminary analysis of the solid phase behavior has been carried out in some restricted regions of the device, identified as critical for the separation process. Glass particles of mean diameter 37 μm were added to the fluid flow to simulate the solid phase behavior. The gas phase flow pattern has been analyzed inside the main components of the cyclone to provide boundary conditions and validation data for CFD code development, and a better understanding of the main physical processes. Tangential velocity is the predominant component in this chamber. It has its maximum close to the outer wall of the exhaust duct. The circumferential velocity slowly decreases going towards the chamber wall. Radial and axial velocity profiles show that the flow is essentially downward directed (positive axial velocity) and moves towards the center of the annular chamber (negative radial velocity). Turbulence intensity profiles are quite similar for the various velocity components, indicating a certain isotropy of the flow, at least in this region. Examination of the tangential velocity profiles indicates the expected forced/free combination of the Rankine vortex, consisting in an outer potential (free) vortex and a region of solid forced) body rotation at the core. The highest swirl velocity is found at the interface of the combined vortexes and results approximately 1.7 times the inlet mean velocity. The values of tangential velocity in the free vortex zone are higher in the cylinder body than in the annulus. Thus the separating force is higher and this suggests that the annulus should be short. The central region is characterized by an intense rigid vortex off-centered with respect to geometrical axis. The angular velocity of this vortex is higher than that in cylindrical body owing to the conservation of angular momentum and area contraction from the cylindrical body to the exhaust duct. The presence of the central recirculation region is usually undesirable, because it induces higher pressure drops, possible working instability, mechanical vibrations and reduced separation efficiency. The simplest solution resulted in the insertion of a pole of rectangular section introduced inside the exhaust duct, along the internal wall, from the exit till the bottom cone of the cyclone. The presence of the pole highly attenuates the rigid central vortex and the recirculation zone is no more present. The flow field within a cyclone separator can be regarded as a subsonic, two-dimensional gas-particle flow. The analysis of such a flow field is complicated by the fact that not only must the conservation equations account for the mass, momentum and energy of each phase, but additional equations are required to relate the momentum and energy coupling between phases. The aerodynamic drag on a particle due to particle-gas velocity differences gives rise to momentum coupling between phases. A finite difference formulation of the differential equations known as the “tank-and-tube” formulation provides systems of simultaneous algebraic equations for the flow properties which are solved on a digital computer. Crowe and Pratt (1974) have shown how this “tank-and-tube” formulation can be extended to the analysis of gas-particle flows, and it is this scheme which is applied to the problem. They have analyzed an axial-inlet, peripheral-discharge type cyclone separator. Swirl vanes at the inlet produce a rotational motion of the dust cloud and the centrifugal forces accelerate the dust particles toward the wall. The gas streamlines converge and exit through the gas-discharge pipe. The particles move to and subsequently along the separator wall toward the peripheral collector and then are removed. Only a small portion of the total gas flows through the separator is bled off with the dust particles. Of course, some of the smaller dust particles do not reach the wall and are carried out by the primary gas flow. The collection efficiency of the separator is defined as the mass of dust collected divided by the total mass of dust which enters he separator. The gas was assumed to be air at standard conditions. The gas mass bled off through the particle collector was neglected although it could be included if a more detailed solution were desired. It was also assumed that the particles and gas were in velocity equilibrium at entry. The important feature of the gas-particle flow analysis used in this study is the fact that it accounts for the particle-gas momentum coupling. When dust particles are introduced into the flow the radial speed of the gas toward the centerline is reduced. In fact, the reduction in radial speed varies monotonically with increased dust loading. This reduction in radial speed renders the gas flow field less effective in moving the dust particles towards the centerline and allowing their escape through the gas-discharge tube. Thus, the effect of increased dust loading is to increase collection efficiency which qualitatively agrees with and explains the experimental results. As dust loading is increased the collection efficiency of the smaller particles increases accordingly. One notes as before a monatomic increase in efficiency with dust loading. Avci and Karagoz (2001) developed a mathematical model with the definition of new parameters to predict the pressure loss coefficient including friction and local losses, and it is calculated in terms of geometrical and flow parameters, under some simplifying assumptions. They have analyzed a tangential cyclone separator. Having tangential, axial and radial velocity components, fluid mixture mainly accelerates towards to cone apex, while a small parts of the mixture leaves from the main flow to the outlet pipe before attaining to the cone apex. Inlet velocity affects the pressure losses. Increasing the inlet velocity over a value which is necessary to create the vortex inside cyclone, results in increasing of Reynolds number and decrease of friction factor. But, in fully turbulent region, the friction factor depends on surface roughness and is not affected from the velocity. Increase of roughness causes decrease of acceleration and increase of flow losses. The final result is low velocity and low pressure losses, except in case of very small inlet width. The diameter of the outlet pipe in cyclones is an important parameter and has a complex effect on pressure losses and efficiency. It also increases friction surface area and decreases the inlet width. This results in slightly decreasing the pressure losses. The influence of the diameter of cone apex is clearly seen in this model. Because of increase of acceleration and decrease of flow losses, pressure losses increases with decreasing of the cone apex diameter. Experimental studies show that the height of the outlet pipe partially increases pressure losses. This parameter has two effects in the present model. The first one is that increase of outlet pipe height causes decrease of pressure losses due to increase of friction surface. The second one has an inverse effect, which is augmentation of outlet pipe height leads to decrease of flow losses, increases of acceleration and finally increase of pressure losses. The global effect is the slightly increase of pressure loss coefficient due to the second effect. Another geometrical parameter is the cone height. Increase of the cone height decreases the friction surface and forces the flow vortices to diminish their radius. The former one which is included in the model causes increase of pressure losses and the latter one gives raise in the flow velocity and pressure losses. An increase of the resistance generally causes decrease of the pressure losses due to insufficient acceleration and low velocities. Increasing of inlet angle, that is, as approaching axial flow at the inlet causes the number of swirl and acceleration to decrease. As a result, flow losses become high, mean velocity will be low, flow path will be short, pressure losses and efficiency of separation will be low. According to Thorn (1998), cyclone efficiency will increase with particle density, particle size, inlet velocity, cyclone body length, number of air stream revolutions, ratio of body diameter to air outlet diameter and static pressure drop across the cyclone. The efficiency will decrease with an increase in cyclone diameter, inlet width and area, air viscosity and density. To summarize the literature review, most of the work is related to tangential cyclone separators with exit from top of the cyclone. Several techniques such as LDV and LDA experiments have been used to study the flow within a cyclone. Simulations were conducted in once case to develop a new cyclone with central body to reduce swirl and hence pressure drop. Some mathematical models have been developed to see the effect of different geometrical and other parameters on pressure drop and dust collection efficiency. The present work deals with a new design related to axial entry cyclone separators with uniflow direction so that the pressure drop can be reduced. The next chapter presents the details of the problem definition and the objectives of the thesis. OBJECTS OF THE INVENTION The main object of the present-invention is to address the above limitations through a novel solution based on the idea of maintaining the direction of the incoming flow thereby reducing pressure loss from that of the conventional design. Another object of the invention is to optimize this design for minimal pressure loss, maximum dust collection efficiency and minimum overall size (to reduce cost). Yet another object of the invention is increased in dust-holding and storage capacities, servicing facilities, and maintenance periods. Still another object of the present invention is to develop a filter to achieve the aforementioned requirements. Still another object of the present invention is to develop a method of filtering using said pre-filter. Still another object of the present invention is to develop a method of manufacturing said filter. STATEMENT OF THE INVENTION The present invention relates to a cyclone separator-based air filter having tangential airflow from top to bottom with reduced pressure drop, said filter comprises (i) frustum of cone with ratio of upper and lower diameter of the cone ranging between 1.3 to 2.5; (ii) tangential inlet passage or ports attached to the cone for the air to enter tangentially and to swirl intensely; (iii) dust collection chamber located below the cone for dust collection; and (iv) suction pipe placed at center of the chamber below the cone for passage of filtered air, wherein separation gap is provided between the bottom of cone and suction pipe for separation of dust into dust chamber. Also, A method for separating particle from air using cyclone separator-based air filter without any substantial pressure drop, said method comprising steps of: a. passing air tangentially from top to bottom into the cone of the filter through tangential inlet ports, b. allowing the air to swirl intensely through the cone, and c. separating particle from the air by collecting the particle into dust particle collection chamber placed below the cone. And also A method of manufacturing a cyclone separator-based air filter having unidirectional airflow from top to bottom without substantial pressure drop, said method comprises: a. having frustum of cone with ratio of inlet and exit diameter of the cone ranging between 1.3 to 2.0; b. attaching angled inlet blades to the cone for the air to enter tangentially and to swirl intensely; c. placing dust particle collection chamber below the cone for trapping separated particles; and d. placing suction pipe at center of the chamber below the cone for passage of filtered air; wherein separation gap is provided between the bottom of cone and suction pipe for collection of particle into the dust chamber. BRIEF DESCRIPTION OF THE ACCOMPANYING DRAWINGS Figure 1 Photograph and the schematic of a conventional cyclone separator. Figure 2 Photograph of the new design of cyclone separator. Figure 3 & 4 Detailed drawing of the new design proposed for the cyclone separator. Figure 5 Mesh of the tangential cyclone separator geometry used for simulation. Figure 6 Experimental setup of the present invention. Figure 7 Schematic of the Wind tunnel Figure 8 Schematic of the dust feeder. Figure 9 Experimental setup for testing the dust collection efficiency of the conventional cyclone separator. Figure 10 Experimental setup for testing the dust collection efficiency of the new design of cyclone separator Figure 11 A schematic of the three variations of the new design: axial, angled and tangential entry. Figure 12 Pressure distribution (Pa) in the central plane of cyclone. Figure 13 Velocity vectors in the domain of cyclone. Figure 14 Pressure distribution (Pa) in the central plane of cyclone. Figure 15 Velocity vectors (m/s) in the domain of cyclone. Figure 16 Pressure distribution (Pa) in the central plane of cyclone. Figure 17 Velocity vectors (m/s) in the domain of cyclone. Figure 18 Velocity vectors in the conical section of the axial entry cyclone Figure 19 Velocity vectors in the conical section of the angled entry cyclone Figure 20 Velocity vectors in the conical section of the tangential cyclone Figure 21 Velocity vectors in a section near the bottom of the cone for axial entry, angled entry and tangential entry cyclone separators. Figure 22 Particle trajectories for the case corresponding to particle size of 1 μm Figure 23 Particle trajectories for the case corresponding to particle size of 2 μm. Figure 24 Particle trajectories for the case corresponding to particle size of 3 μm. Figure 25 Particle trajectories for the case corresponding to particle size of 4 μm. Figure 26 Variation of dust collection efficiency for tangential cyclone separator. DETAILED DESCRIPTION OF THE INVENTION The present-invention relates to a cyclone separator- based air filter for separating particles from air. The invention relates to a cyclone separator-based air filter having unidirectional airflow from top to bottom without substantial pressure drop, said filter comprises i. frustum of cone with ratio of upper and lower diameter of the cone ranging between 1.3 to 2.5; ii. tangential inlet passage or ports attached to the cone for the air to enter tangentially and to swirl intensely; iii. dust collection chamber located below the cone for dust collection; and iv. suction pipe placed at center of the chamber below the cone for passage of filtered air, wherein separation gap is provided between the bottom of cone and suction pipe for separation of dust into dust chamber. In yet another embodiment of the present invention, numbers of inlet blades are ranging between 6 to 20. In still another embodiment of the present invention, number of inlet ports is about 8. In still another embodiment of the present invention, height of the cone is ranging between 80 to 140 mm. In still another embodiment of the present invention, the separation gap is ranging between 2 to 10 mm. In still another embodiment of the present invention, separation gap is about 6 mm. In still another embodiment of the present invention, exit diameter of the cone is ranging between 45 mm to 120 mm. In still another embodiment of the present invention, lower diameter of the cone is about 70 mm. In still another embodiment of the present invention, upper diameter of the cone is varied from 120 to 150 mm. In still another embodiment of the present invention, upper diameter of the cone is about 134 mm. In still another embodiment of the present invention, the ratio of upper to lower diameters of the cone is about 2. In still another embodiment of the present invention, the filter is an accessory used with IC engines operating in a dusty environment in vehicles selected from a group comprising tractors and earthmovers. In still another embodiment of the present invention, the filter is used in several fields selected from a group comprising food, pharmaceutical, automobile, pollution control, photographic film, and infrastructure. A method for separating particles from air without substantial pressure drop using cyclone separator-based air pre-filter, said method comprising steps of: i. passing air uni-directionally from top to bottom into frustum of cone of the filter through angled inlet blades, ii. allowing the air to swirl intensely through the cone, and iii. separating particles through a separation gap into collection chamber placed below the cone. In still another embodiment of the present invention, frustum of cone with ratio of inlet (upper) and exit (lower) diameter of the cone ranging between 1.3 to 2.0; In still another embodiment of the present invention, number of inlet blades is about 8. In still another embodiment of the present invention, height of the cone is ranging between 60 to 140 mm. In still another embodiment of the present invention, the separation gap is ranging between 2 to 10 mm. In still another embodiment of the present invention, separation gap is about 6mm. In still another embodiment of the present invention, lower diameter of the cone is ranging between 45 mm to 120 mm. In still another embodiment of the present invention, lower diameter of the cone is about 70 mm. In still another embodiment of the present invention, upper diameter of the cone is varied from 120 to 150 mm. In still another embodiment of the present invention, upper diameter of the cone is about 134 mm. In still another embodiment of the present invention, the ratio of inlet to exit diameters of the cone is about 2. In still another embodiment of the present invention, the filter is an accessory used with IC engines operating in a dusty environment in vehicles selected from a group comprising tractors and earthmovers. In still another embodiment of the present invention, dust collection efficiency is 100% for particles of size greater than or equal to 4 μm In still another embodiment of the present invention, average dust collection efficiency for particles of size ranging between 0 to 5 μm is higher than 50%. In still another embodiment of the present invention, efficiency of dust collection increases with reduction in exit diameter of the cone. A method of manufacturing a cyclone separator-based air filter having unidirectional airflow from top to bottom without substantial pressure drop, said method comprises: a. having frustum of cone with ratio of inlet and exit diameter of the cone ranging between 1.3 to 2.0; b. attaching angled inlet blades to the cone for the air to enter tangentially and to swirl intensely; c. placing dust particle collection chamber below the cone for trapping separated particles; and d. placing suction pipe at center of the chamber below the cone for passage of filtered air; wherein separation gap is provided between the bottom of cone and suction pipe for collection of particle into the dust chamber. In case of IC engines operating in a dusty environment, such as in tractors and earth movers, the filter elements get fouled within no time. In such cases, cyclone separators are used prior to the air filter. This will separate out the heavy dust particles. Only the very fine dust enters the filters thus increasing the life of the air filter. The conventional cyclone separator which is available commercially is shown in Figure 1. This cyclone separator separates the heavy dust particles and sends the semi pure air to the main filter. But there are also some draw backs of this design. The energy for generating the swirl is obtained by a certain amount of pressure drop which can affect the performance of the engine. It is also observed from Figure 1 that the flow entering is required to make a 180-degree turn before entering the suction pipe. Also, there is a chance of dust entering directly into the suction pipe. Though the cyclone separator is effective in removing the coarse dust particles, these are some of the shortcomings of the conventional model. Thus, it is required to have a cyclone separator design which minimizes pressure drop, while maintaining the required dust collection efficiency. In order to meet these requirements, a new design is proposed. Figures 2, 3 and 4 show the complete details of the new cyclone separator design. It consists of a frustum of cone attached to a certain number of inlet blades. The cone is located at the top of the separator. It consists of feed openings on the top of the cone. There are 8 passages or ports present. This enables the flow to swirl and the swirl intensifies as the flow passes through the conical section. The suction pipe of the engine is placed at the bottom part of the cone leaving a small gap to allow the particles to escape into the dust collection chamber. The inlets are open to the atmosphere. The suction pipe is connected to the filter. In the conventional model, the cone was kept inversed and the flow was required to make a 180° turn. In the present model, the flow is unidirectional, thus minimizing the pressure loss. This new design also provides sufficient gap for the effective removal of dust particles. This design intuitively seems better than the conventional model. However, there is a need for systematic study of the flow field inside the separator to assess the pressure loss and dust collection efficiency for various particle sizes. Furthermore, there is a need to optimize the system to maximize the dust collection efficiency and minimize the pressure loss. The next chapter presents this analysis and validation of this analysis with experimental data. • CFD Simulation Computational Fluid dynamics (CFD) is concerned with the development and application of numerical methods in solving the equations of fluid motion. Using CFD, we can test a wide variety of alternatives in much less time and at a significant savings in cost as compared with iterations using prototypes. It deals with the simulation of both compressible and incompressible flows. Typical activities include structured and unstructured grid generation, flow algorithm development for compressible and incompressible Navier -Stokes equations based on finite difference, finite volume and finite element methods, turbulence modeling, graphical post-processing, high performance computing and applications. In the present work, simulations of the two-phase (gas-solid) flow in the cyclone separator are performed using FLUENT 6.1, a CFD package. The cyclone separator modeling essentially consists of the study of two-phase swirling flows, with one phase representing the gaseous phase (air) and the other representing the solid phase (dust particles). Swirling flows are among the most common and most complex in the processing systems as cyclone separator. The CFD model helps to simulate these complex tasks. The physical models available in FLUENT allow to accurately predict laminar, transitional and turbulent flows, various modes of heat transfer, chemical reaction, multiphase flows and other complex phenomena. GAMBIT which is available with FLUENT, provides a concise and powerful set of solid modeling-based geometry tools for generating meshes and grids that facilitate CFD simulations. Different CFD problems require different mesh types, and GAMBIT provides all options required in a single package. GAMBIT’S meshing toolkit decomposes geometries for structured hex meshing or perform automated hex meshing with control over clustering. Formulation This section describes the equations governing both the gaseous and solid phases, and their interaction. The Mass Conservation Equation The equation for conservation of mass, or continuity equation, can be written as follows: ∂p/∂t +∆.pv=0 (4.1) This is the general form of the mass conservation equation and is valid for incompressible as well as compressible flows. Momentum Conservation Equations Conservation of momentum in an inertial (non-accelerating) reference frame is described as follows: ∂/∂T(PVP)+∆.P(PVPVP)= -∆P+∆.(Ŧ)+PGP+FP where p is the static pressure, r is the stress tensor (described below), pjgand F are the gravitational body force and external body forces (e.g., that arise from interaction with the dispersed phase), respectively. The stress tensor f is given by Ŧ=μ/ where// is the molecular viscosity, / is the unit tensor, and the second term on the right hand side is the effect of volume dilation. Turbulence Modeling: The RNG k - G model The RNG k - G model was derived using a rigorous statistical technique (called renormalization group theory). It is similar in form to the standard k - G model, but includes the following refinements: • The RNG model has an additional term in its G equation that significantly improves the accuracy for rapidly strained flows. • The effect of swirl on turbulence is included in the RNG model, enhancing accuracy for swirling flows. • While the standard k - G model is a high-Reynolds-number model, the RNG theory provides an analytically-derived differential formula for effective viscosity that accounts for low-Reynolds-number effects. Effective use of this feature does, however, depend on an appropriate treatment of the near-wall region. These features make the RNG k - G model more accurate and reliable for a wider class of flows than the standard k - G model. The RNG-based k - G turbulence model is derived from the instantaneous Navier-Stokes equations, using a mathematical technique called “renormalization group” (RNG) methods. The analytical derivation results in a model with constants different from those in the standard k - G model, and additional terms and functions in the transport equations for k and G. The turbulence kinetic energy k and its rate of dissipation G are obtained from the following transport equations: , d (pkui) = dk a w . a .. „ . .. .„ „ Gk+Gb-ps-YM+Sk (4.4) at ox: ox; -Cuj(Gk+CieGb)-C2Ep^--Re +Se (4.5) , k k dt 3x: dx In these equations, Gk represents the generation of turbulence kinetic energy due to the mean velocity gradients, Gh is the generation of turbulence kinetic energy due to buoyancy, YM represents the contribution of the fluctuating dilatation in compressible turbulence to the overall dissipation rate. The quantities errand «£are the inverse effective Prandtl numbers for k and e, respectively. Sk and SE are user-defined source term. The main difference between the RNG and standard k - 8 models lies in the additional term in the 8 equation given by: (4.6) where 7j = Sk/s, rjo=438, fi = 0.012. The model constants Cl£and C2s in have values derived analytically by the RNG theory. These values, used by default in FLUENT, are : Cu= 1.42, Cl£ =1.68. Equations of Motion for Particle Particle force balance FLUENT predicts the trajectory of a discrete phase particle (or droplet or bubble) by integrating the force balance on the particle, which is written in a Lagrangian reference frame. This force balance equates the particle inertia with the forces acting on the particle, and can be written (for the ^direction in Cartesian coordinates) as: vFx (4.7) p dt Pp where FD(u-u ) is the drag force per unit particle mass and Here, u is the fluid phase velocity, up is the particle velocity, /J is the molecular viscosity of the fluid, p is the fluid density, p is the density of the particle, and dp is the particle diameter. Re is the relative Reynolds number, which is defined as u = pd p \| up- u\|/ u (4.9) In the present modeling, the 3D version of FLUENT is used. In this model, segregated (Implicit solver) solution algorithm is used. The flow being solved is considered to be in steady state, and the time-independent solution is solved as a first-order implicit solution (i.e., first-order accurate dual time-stepping). For air and particle flow, discrete phase model has been used and the turbulent flow is calculated using the RNG k-e turbulence model. Particles of density 3300 kg/m3 have been injected into the inlet for calculating the dust collection efficiency. The efficiency will be calculated by finding the number of particles coming out of the exit pipe divided by the total number of particles injected into inlet. Boundary Conditions Inlet: The inlet is opened to the atmosphere. Therefore the inlet boundary is specified to be at atmospheric pressure. Outlet: The outlet boundary condition is estimated using the mass flow rate. The mass flow rate values have been taken from experiments and given as the outlet boundary condition. The new design of cyclone separator with the complete mesh is shown in the Figure 5 • Model Validation Experimental Setup Any computational modeling effort requires experimental validation. Hence, a prototype of the new design of cyclone separator was fabricated and the pressure drop and dust particle collection efficiency were measured. The objective is to validate one set of simulations; and once validation is achieved, the model can be used to perform various simulations and optimization. Thus, Experiments were conducted for measuring the pressure drop and the dust collection efficiency of conventional and the new cyclone separator. The schematic diagram of the total experimental setup is shown in Figure 6. The various components of the experimental setup are described below. Wind tunnel: Wind tunnel is of suction type as shown in Figure 7. It is equipped with a two-stage blower to provide required suction pressure inside the wind tunnel. It is equipped with an orifice meter which is connected to a digital readout calibrated in mm of water. Flow control valve can be used for varying the flow rate. The cyclone separator is connected to the inlet side of the wind tunnel. Manometer: Manometer is an ordinary U-tube manometer. It is of one meter length. It uses water as the manometric fluid. It is used to find the pressure drop across the cyclone separator. Dust feeder: Dust feeder is used in testing the dust feeding efficiency of the cyclone separator. Figure 8 shows a schematic of the dust feeder. It is provided with timing adjustment to feed the required amount of the dust in specified time. It has an arrangement to introduce the dust into the air flow. Compressed air mixes with the dust and carries it along with air. This air can be fed to cyclone separator to check its efficiency. Experimental Procedure Using the experimental setup shown in Figure 6, experiments were conducted to measure the pressure drop and dust collection efficiency of the conventional and new design of cyclone separators. Pressure drop across the cyclone separator:- The pressure drop was measured at different air mass flow rates through the cyclone separators. Procedure: 1) The cyclone separator was fixed to the experimental setup. 2) The flow rate of air was set using the flow control valve. 3) The flow rate and pressure drop across the cyclone separator were noted. 4) The procedure is then repeated for different flow rates for both conventional and new models. Past collection efficiency:- The dust collection efficiency is important to estimate as the primary aim of a cyclone separator is to separate the dust from the air stream. The tests were conducted for the new and the conventional models for the purpose of comparison. The experimental setups for dust collection efficiency of conventional and new design of separators are shown in Figures 9 and 10. Procedure: 1) The flow control valve was set to a particular flow rate and was maintained constant throughout the experiment. 2) The separator was weighed and was fixed to the experimental setup. Weighed amount (110 gm) of dust was filled into the dust feeder. 3) The feed rate of air into the cyclone separator was set to 237.6 kg/hr. The timer of the dust feeder was set such that 110 gm of dust was fed in 1 hr. 4) After all the dust was fed, the weight of the separator was measured. The difference between the weights gives the dust collected, from which dust collection efficiency of the cyclone separator was calculated. Observations The detailed experimental data is given below. Pressure drop The pressure drop is calculated as follows: Ap - pgh Pa where p - 1000 kg/m for water g =9.81 m/s2 h = manometer reading in m of water Dust collection efficiency The dust collection efficiency for both conventional and new design of cyclone separator was found by weighing the mass of dust collected and mass of dust sent to the separators. _ ,, . nr. . Mass of dust collected in separator Dust collection efficiency = Mass of dust supplied to the separator The experimental values of dust collection efficiency are given in Table 4.2. Table 4.2 Dust collection efficiency of the cyclone separators The dust which was used in experiments consists of many particle sizes and particles of many materials. The details of particle size distribution and the chemical composition of particles are given in Tables 4.3 and 4.4. Table 4.4 Chemical composition of the dust particles PARAMETRIC STUDY The previous chapters concerned the analysis of a new design of the cyclone separator, and also validation of the numerical model with the experimental data. From the results, it was clear that the new design that was based on intuition is clearly superior to the conventional design in terms of both pressure loss and dust collection efficiency. It is however, important to explore variations in this new design and further to optimize it. In this regard, three different variations based on the air entry were studied here. These are: 1. Axial entry cyclone 2. cyclone with angled entry 3. Tangential entry cyclone 1. Axial entry cyclone: This is the new design of the cyclone separator that was described in the previous chapter. The results of this separator will be compared with the other two cyclone separators. The comparison will be done using the results of simulations. 2. Cyclone with angled entry: In this variation of the new design, the air entry is partially axial and partially tangential, with the plane of each entry duct making an angle of 45° with the axis of the separator as shown in Figure 11. 3, Tangential entry cyclone: In this variation of the new design, the air enters completely tangential to the axis of the separator as shown in Figure 11. Performance Comparison of Design Variations Each particular variation in the new design was simulated using FLUENT. The two-phase flow was computed. The boundary conditions being same as in the simulation of the first new design i.e. atmospheric pressure at inlet and the mass flow rate at the outlet. Figure 12 shows the absolute pressure distribution in the central plane of the axial entry cyclone. It can be observed that the pressure decreases as the air flows down the cone. This can be attributed to the fact that the velocities increase as the flow reaches the bottom of the cone and hence pressure decreases. The major pressure drop is near the entrance to the exit pipe because of the change in area of flow. Figure 13 shows the velocity vectors with magnitudes for the design with complete axial entry. It can be observed from the figure that the air entering has some tangential velocity component with the axial velocity. The velocity increases as the air passes down the conical section. The peak velocity occurs at the bottom of the conical section and is about 30 m/s. The velocity magnitudes in the dust collection chamber are very less as the air slows down in the chamber. The velocity is more in exit pipe compared to the inlet because of the reduction in area. Figure 14 shows the pressure distribution in the central plane of the cyclone with angled entry. The pressure is very less near the axis of the conical section. This is because of very high velocities near the axis. Figure 15 shows the velocity vectors of the cyclone with angled entry. The colour bar shows the magnitudes. It can be seen from the velocity vectors that the tangential velocity is very high at the top of the cone section. The velocities increase as the air flows down the conical section and the tangential velocity is maximum at the end of the conical section. The magnitude of the maximum velocity is about 60 m/s. This magnitude is almost double than that in the axial cyclone separator. Thus the pressure drop is much more in this case compared to the axial entry cyclone as can be seen from the Figure 14. Figure 16 shows the pressure contours in the central plane of the design with tangential entry. In this case also, the pressure is very less near the axis of the conical section because of very high tangential velocities. Figure 17 shows the velocity vectors for the cyclone with angled entry. In this design the inlet velocity is complete tangential as can be seen in the figure. The velocities in this model are somewhat lower than in angled entry design and higher than in axial entry cyclone. The maximum velocity is around 48 m/s which occur at the end of the conical section. The pressure drop is also somewhat lower compared with that of angled entry model but is still very high compared with axial entry model. Figure 18 shows the velocity vectors in the conical part of the axial entry cyclone separator. It can be observed from the figure that the velocities are partly axial and partly tangential. Figure 19 shows the velocity vectors in the conical part of the cyclone with angled entry. In this cyclone the velocities are completely tangential The magnitude of velocities are also very high. Figure 20 shows the velocity vectors in the conical part of the tangential entry cyclone separator. In this case also the velocities are very high and are mostly tangential. Figure 21 shows that the velocities near the bottom of the conical section are partly axial and partly tangential for axial entry cyclone and the velocities are almost tangential for the angled entry and tangential entry cyclones. From the simulations, the total pressure drops across the three types of design are found to be as follows: Axial entry cyclone : 576.1 Pa Cyclone with angled entry : 5200Pa Tangential entry cyclone : 2060Pa Since the pressure drop for the second and third design variations is very high compared with the first model, which is axial cyclone separator, these are not considered for further study. Hereafter, a parametric study has been conducted for the axial entry cyclone separator in order to optimize the design for maximum dust collection efficiency and minimum pressure drop. In conventional terms, the air filtering is done in two stages: a pre-filter, i.e., cyclone separator to separate heavier dust particles and a main filter to separate the finer particles. The work presented in this thesis so far concerns the development of a new design for the pre-filter. The assumption is that there is a conventional main filter present downstream which will separate the finer dust particles. Now, in conventional systems, the main filter is supposed to separate the particles of size 5 micron and upwards. Thus, the filter typically removes particles above 30 - 40 um, and the main filter is supposed to trap particles between 5-30 um. Now, some of the new designs simulated in this study have shown dramatic results with respect to dust collection efficiency. These results prompted us to propose a radically new concept in air filtration which does not require polyurethane foam-based main filter at all. In other words, some of the so-called filter designs simulated here show a dust collection efficiency of 100% for particles of 4 um and higher. This implies that this cyclone separator based filter can do the job of the pre-filter and main filter configuration. This radical concept can significantly lower the cost of the air filter unit, and more importantly, the running costs also are reduced to zero, since there is no replaceable part at all (as opposed to the conventional filter where the air filter unit needs to be replaced periodically). However, it is important to also estimate the pressure drop of this new design. First of all, Figure 5 shows the configuration of this design along with the mesh generated. As seen from Figure 5, the entries are all tangential which causes the swirl velocities inside the separator to be very high. This in turn explains the very high dust collection efficiencies of this design for various particle sizes beyond 4 um The particle trajectories for the different particle sizes from 1 nm to 4 um are shown in Figs. 22-25. It can be observed from these figures that the efficiency increases drastically with particle size. As mentioned before, the high efficiency is due to the very high swirl velocities. Figure 26 shows the plot of variation of dust collection efficiency with particle size. It shows that the efficiency is 100% for particle sizes of 4 um and above. Therefore the use of this tangential cyclone separator eliminates the necessity of using a main filter downstream. The overall pressure drop predicted is around 2000 Pa. This result is extremely encouraging since the total pressure drop for a conventional pre-filter and main filter configuration is around 5000 Pa. Thus, this design promises less than half the pressure drop, and significantly lower costs, while maintaining effective separation for particle sizes of 4 micron and above. However, this predicted performance needs to be confirmed with experiments. In any case, based on the earlier model validation, there is a very high probability of computations matching results. Parametric Study A number of geometric variables related to the cyclone separator geometry were initially investigated and four parameters have been selected for detailed study and optimization. These are: 1. Separation gap (between cone and exit pipe) 2. Cone height 3. Ratio of entry diameter to exit diameter of the cone (D/d) 4. Number of vanes at the inlet. Separation gap: The gap between the bottom of frustum of cone and the exit pipe is varied from 2mm to 8mm to evaluate its effect on the performance. The results are displayed in the form of comparison of dust collection efficiency and also pressure drop across the separator. The pressure drop increases as the gap decreases. This can be attributed to the fact that as the gap reduces; more and more air will be going to the collection chamber causing the loss of momentum, thereby subsequent increase in pressure drop. The efficiency is almost same for small particles. As the particle size increases the efficiency increases. The efficiency is higher for the design with higher gap at the separation point. Cone height: The height of the cone is varied to find out the effect of its size on the dust collection efficiency and pressure drop. The efficiency increases with the length of the cone. This is because the efficiency increases with the residence time and the residence time will be more for cyclone of more height. At the same time, the pressure drop decreases with the increase in height of the cone. This is because the convergence angle of the cone decreases as the height of the cone increases fixing the diameters constant. Therefore the velocities will be lower and hence lower pressure drop at higher cone heights. Cone diameter ratio: The exit diameter of the cone is varied by 70, 80, 90 and 100mm. The dust collection efficiency increases with decrease in lower diameter of the cone. The increase in swirl velocity at the separation point because of reduction in the lower diameter of cone is the main reason for the increase in efficiency, since higher swirl velocity causes a larger centrifugal force on the particle. The pressure drop increases with the increase in the lower diameter. This is because, as the diameter increases, the amount of air (at high velocity) entering the collection chamber increases, causing increase in the pressure drop. Number of vanes: The number of vanes at the inlet of the separator is varied from 8 to 16. The dust collection efficiency increases with the increase in number of vanes at inlet. This is because of the better directing capability with more number of vanes, which implies a higher tangential velocity component leading to a higher swirl velocity at separation point. The pressure drop, at the other hand, increases with increase in number of vanes. This is because of more resistance to the flow due to more number of vanes. Based on all the above four parameters, the best possible combination of all the four parameters is selected to give good dust collection efficiency while giving low pressure drop. Simulations were again conducted on the model with this combination of parameters. The values of these parameters for the best combination are as follows: The gap where the separation of particles take place = 6mm The height of the cone = 100mm The lower diameter of the cone = 70mm No. of vanes at the inlet = 8 Purpose of Dust Collection Dust filtration is important in many applications, some of which are described below: • Air-pollution control, as in fly-ash removal from power plant flue gases. • Equipment maintenance reduction, as in filtration of engine intake air or pyrites furnace gas treatment prior to its entry to a contact sulphuric acid plant. • Safety or health hazard elimination, as in collection of siliceous and metallic dusts around grinding and drilling equipment and in some other metallurgical operations and flour dusts from milling or bagging operations. • Product quality improvement, as in air cleaning in production of pharmaceutical products and photographic film. • Recovery of a valuable product, as in collection of dusts from dryers and smelters. Powdered product collection, as in pneumatic conveying, the spray drying of milk, eggs and soap; and manufacturing of high purity zinc oxide and carbon black. CONCLUSION A new cyclone separator-based filter with applications for IC Engines operating in a dusty environment has been simulated. This design shows reduced pressure drop and increased dust collection efficiency over conventional systems. The flow direction was made axial and uni-directional so that the pressure drop can be reduced. In the uni-flow cyclone the air enters axially through number of inclined vanes which provide some tangential velocity thereby increasing the swirl at the separation region and the particles are get separated at the periphery and the air leaves through the central outlet pipe. This enables a good separation of dust particles. Experiments have been done to measure the pressure drop and dust collection efficiency for both the conventional and the new cyclone separator design. A huge reduction in pressure drop was observed with the new cyclone separator compared to the conventional cyclone separator. More significantly, the dust collection efficiency for the new cyclone separator was more than that of the conventional cyclone separator. Simulations were done on the new cyclone separator and the results were compared with the values of experiments. There was a good comparison of pressure drop between experiments and simulations except at very low mass flow rates. This may be because of the uncertainty associated with the measurement of the mass flow rate at low values. The dust collection efficiency from simulations was observed to closely match with the experimental value. A detailed parametric study was performed on the new cyclone separator, having axial entry and uni-directional flow, with respect to its geometrical parameters to optimize the performance in terms of pressure drop across the cyclone separator and the dust collection efficiency. A new optimized design is proposed based on the results of the parametric study. The results of the simulation for the tangential entry cyclone separator indicated very high dust collection efficiencies, i.e., 100% for particle sizes of 4 micron and higher. The air entry for this new design of separator is completely tangential. This gives very high tangential velocities in the conical section of the cyclone separator thus providing very high separation efficiency. The pressure drop is around 2000 Pa. Now, this is much less than the total pressure drop across a conventional pre-filter cum air filters combination, where the pressure loss is around 5000 Pa. This prompted a radically new idea in air filtration for such applications. The idea is to completely eliminate the use of the polyurethane foam-based main filter, and use only a tangential entry cyclone separator. This design anyway satisfies the requirement those particles of size 5 micron and above need to be completely filtered. However, experiments are needed to validate the results of this new design. In conclusion, this design has potential not just in improving performance of the engine (by means of a reduced pressure drop), but also in significantly reducing the costs (both initial and running) of air filters for tractors and earth moving equipment applications.

Full Text

FIELD OF THE INVENTION
The present-invention relates to a cyclone separator- based Air filter for separating
particles from air.
BACKGROUND OF THE INVENTION
Dust consists of tiny solid particles carried by air currents. These particles are formed by disintegration or fracture process, such as grinding, crushing, or impact. Dust is generated by a wide range of manufacturing, domestic, and industrial activities. Construction, agriculture, and mining are among the industries that contribute most to atmospheric dust levels. A wide range of particle sizes is produced during a dust generating process. Particles that are too large to remain airborne settle while others remain in the air indefinitely.
The escape of dust particles into the workplace atmosphere is undesirable. Excessive dust emissions can cause both health and industrial problems:
• Health hazards
- Occupational respiratory diseases
- Irritation to eyes, ears, nose, throat and skin

• Damage to equipment
• Impaired visibility
• Unpleasant odors
• Problems in community relations
PRIOR ART OF THE INVENTION
Present air pre-filters are based on a design that causes the entering air to make 180
degree turn before it exits the device. This gives reasonably good dust collection
efficiency, however the pressure loss is high. There have been several attempts in the
past to reduce the dust collection efficiency, but there has not been any significant
reduction possible.
However, these air-filters have certain limitations as follows:
The pressure drop in present cyclone separator-based air filters is the high across the
entire device. This also affects the internal combustion engine’s performance in terms
of reduced volumetric efficiency, which in turn affects the power generated and their

dust-holding and storage capacities, servicing facilities, and maintenance periods have been sacrificed.
In this section a brief review of the relevant literature is presented. Zhou and Soo (1990) worked on gas solid flow and collection of solids in a cyclone separator. A method of improving the performance of cyclone separator with central body was illustrated by them. They conducted Laser Doppler Velocimetry (LDV) measurements to compare and validate computations of flow and pressure drop in a cyclone separator by an analytical solution of constant viscosity and a solution based on the k- e turbulence model. The detailed measurements of gas flow field in cyclone separator by LDV and numerical simulation of this flow field by using k-s turbulence model and SIMPLE (Semi Implicit Method for Pressure Linked Equations) algorithm gave a complete picture of whole flow field. Results explain the significance of solid body core of vortex and the difficulty of k-e method when applied to vortex motion. A theoretical analysis was performed by Soo on the gas flow field, dust dispersion and dust collection in a cyclone, based on the fundamentals of multiphase fluid dynamics. Zhou et al. developed numerical modeling and LDV measurements of gas-solid flows, and also proposed an innovative cyclone separator with a special central body to remarkably reduce the pressure drop.
Avci and Karagoz (2003) presented their work on the effects of flow and geometrical parameters on the collection efficiency in cyclone separators. Predictions of the collection efficiency and the cut-off size for a cyclone are difficult problems since the behavior of the mixture inside the cyclone are affected by many parameters, such as flow characteristics, particles and mixture properties and external forces on particles. Therefore, detail knowledge of physical phenomena and appropriate interpretation of experimental results are required in order to be successful in modeling the cyclone flow. A mathematical model has been developed by defining average flow and geometrical parameters, and applying conservation equations to the cyclone flow. Despite the simple approximations to the friction coefficient and the vortex number, the computed results are in good agreement with experimental values, showing that the proposed model is successful for all type of cyclones comparing with other models in literature. The following conclusions can also be drawn from the present study.
• Collection efficiency depends on the flow regime which is affected by the flow parameters such as inlet velocity, temperature, viscosity and surface roughness.

In big cyclones, flow is generally fully turbulent and the effects of flow parameters decrease with Reynolds number, whereas, in small cyclones or at low velocities, flow is laminar and the effects of flow parameters increase up to a certain point and than start to decrease after that. The characteristics of transitional flow regime are required to be investigated in detail so that more efficient cyclones would be designed.
Prediction of the friction coefficient seems to be very important and it plays an important role on the present model. Smooth surfaces were assumed for comparison with experimental values in this model and this may cause some deviation of the computed results from the experimental values. Decreasing in surface resistance which eventually leads to decrease of velocity is an important way to high efficiency due to increasing in natural vortex length and residence time inside cyclone.
The diameter of cone apex affects acceleration and consequently collection efficiency and pressure losses. Very small cone apex diameter is not preferable from the point of view of pressure losses and vortex length, whereas very big cone apex diameter is not relevant for collection efficiency. Experimental results show that, keeping other geometrical ratios constant, cyclone height affects collection efficiency positively up to a certain point. These behaviors were also verified by the present model. Although it is possible to say that the optimum cyclone length may be increased if the friction coefficient is decreased, this length also depends on the geometry and flow conditions. Therefore, further studies are required to be performed on this subject.
Analysis of experimental and theoretical results has shown that the inlet width has very big influence on the cut-off size in some cases of geometry and operational conditions, whereas it has minor importance otherwise. Therefore, additional studies are required in order to investigate the effects of the inlet width on the cyclone performance.
Another interesting parameter is the inlet height, which may be important from the point of view of frictional surfaces. Although the inlet height is only included in the definition of ad, it is easy to take account of this parameter by defining the hydraulic diameter, which may cause some alteration of the results.

• Due to variation of velocity, hydraulic diameter and relative roughness in cyclone, it is difficult to compute Reynolds number and to determine dominant flow regime. Therefore, one of the problems is the prediction of the friction coefficient correctly. In this model the friction coefficient was computed with the reference to the inlet conditions for practical usage.
Fraser et. al. (1997) presented work on computational and experimental investigations in a cyclone dust separator. They simulated three-dimensional turbulent flow in a model cyclone using PHOENICS code and experimental studies carried out using Laser Doppler Anemometry (LDA) system. The experimental results were used to validate computed velocity distributions based on the standard and a modified k-e model. Simulation
The standard k-e model was found to be unsatisfactory for the prediction of flow field inside the cyclone chamber. By considering the strong swirling flow and the streamlined curvature, a k-e model modified to take account of Richardson number, provided better velocity distributions and better agreement with the experimental results. The following form of Richardson number has been used in their study:
Ri=K2 / Є2 u/ r2 ∂(ru)/ ∂z
where r is the radial distance across the cyclone and z is the distance down the cyclone. The standard k-e turbulence model assumes that the turbulence is isotropic, which is not the case for swirling flows. The Richardson number is based on several terms including the swirl gradient, which allows for the mixing length to increase or decrease as the gradient becomes negative or positive. When this is used to modify the dissipation it makes a first attempt at making the turbulence non-isotropic and hence producing a better simulation. The simulation was carried out assuming incompressible fluid flow on the basis that the maximum velocity vector using the measured swirl and axial components was less than 20 m/s. Experiments
The test model was constructed in Perspex and based on the high efficiency Stairmand cyclone design. Air was used as the working fluid and is drawn into the reservoir chamber via a two-stage axial fan at average ambient conditions. The air is supplied to the cyclone chamber through the piping system and from the cyclone chamber the air is exhausted to the atmosphere. The experiment is conducted at a constant flow rate. For

the purpose of dust loading and seeding particles for LDA measurement, a specially prepared milk powder in the range of 3-50 μm in diameter is used in the feeder. LDA system is optimized for particle sizes in the range of 5-10 μm and these particles are not separated by the cyclone and so exit via the vortex outlet while the larger particles are separated and drop out through the bottom of the cyclone. Although the flow has a large centrifugal effect it has been estimated that the smaller particles follow the fluid flow with only minor variations. Results
The air flow inside the cyclone chamber is non-axisymmetric and with high swirl and streamline curvature. The flow rotates strongly at the axis towards the outlet tube and reduces rapidly at the cyclone wall. In the conical chamber, the swirl velocity reduces rapidly in the core region compared to the cylindrical chamber. The results of the experimental and modified k-e model show that the vortex motion inside the cyclone chamber consists of two parts, the core of the forced vortex at the centre where velocity increases with radius and an outer free vortex where velocity reduces with increasing radius. Maximum velocity occurs at the centre and reduces rapidly with radius. Avci and Karagoz (2000) presented their work on the mathematical model of two-phase flow in tangential cyclone separator using parameters such as effects of cyclone geometry, surface roughness and concentration of particles. The critical diameters, fractional efficiencies and pressure losses are calculated under the assumptions that each phase has the same velocity and acceleration in the spiral motion of flow, a relative velocity occurs in the radial direction and drag coefficient remains constant under certain conditions. The results obtained under these assumptions are compared with their experimental and theoretical values, and a very good agreement is observed with experimental values. Experiments prove that efficiency increase with increase of cyclone height. But it should be a limit for the height due to friction and separation from the surfaces and this should be analyzed experimentally.
Molerus and Gluckler (1996) developed a cyclone separator with new design for gentle separation of particles from gas phase (with respect to breakage and attrition of the particle) for usual solid loadings and which is suitable for downer systems. Test experiments were carried out to evaluate quality of performance that gave overall collection efficiency of up to 99.99 % for particles in the range of 10 jam. Any change in direction of a gas-solid flow creates a change in momentum of the particles. This

leads to collisions and/or sliding along walls, which causes breakage and attrition. Therefore the new separator should operate with as little change in momentum of the solid stream as possible. A new cyclone separator with a specific inlet device was designed, tested and modified until the performance was satisfactory. The cyclone consists of a frustum of a circular cone and a cover with conventional vortex finder in an inclined arrangement. The gas-solid suspension is fed into the cyclone vertically downward through the cover with an angle of 30 ° to the main axis of the cone. The inlet device is shaped in such a way that it fits with the circular contour of the cone edge and lines up with the wall of the cone. Therefore the feed stream is homogeneously accelerated without any direct collision with the cone wall. Several experiments were carried out to evaluate the quality of performance of the new cyclone. The overall collection efficiencies are satisfactory and independent of the solid loading as long as the solid mass flux is not too high. For solid mass fluxes exceeding a certain value, the separated particles will not flow out of the cyclone fast enough. They will be mixed back into the gas stream to a certain extent which results in a less efficient separation. This effect also occurs with conventional cyclones and can he avoided by correct dimensioning of the solid outlet of the cyclone. However, the collection efficiencies of normally more than 99%, in a range of the solid loading 1 Li et al. (1996) presented results on the study of solid-gas separation mechanism of cyclone with impulse excitation. On the basis of experimental study of the performance of cyclone impulse electrostatic precipitation (CIESP), the solid-gas separation mechanism of CIESP is presented by them. CIESP is a new technique, which combines merits of cyclone precipitation and electrostatic precipitation. The basic idea was to use high voltage impulse electrostatic field in cyclone, when airflow with particles passes, the particles are separated from gas by centrifugal force and impulse electrostatic force. By analyzing the charging and acting forces on particles, the theoretical formula of particles migration velocity and grade efficiency are concluded and verified with experiments.
An experimental investigation was performed by Giulio Solero and Aldo Coghe (2002) on a laboratory scale model of a gas cyclone separator inserted in a closed/open loop test facility built up to study the flow field of both the gas and solid phase. The cyclone is made entirely by Plexiglas in order to allow optical access for velocity

measurements through Laser Doppler Velocimetry. Three velocity components of the gas phase were measured inside the various regions of the apparatus (tangential inlet chamber, cylindrical body and exhaust duct) and a preliminary analysis of the solid phase behavior has been carried out in some restricted regions of the device, identified as critical for the separation process. Glass particles of mean diameter 37 μm were added to the fluid flow to simulate the solid phase behavior.
The gas phase flow pattern has been analyzed inside the main components of the cyclone to provide boundary conditions and validation data for CFD code development, and a better understanding of the main physical processes.
Tangential velocity is the predominant component in this chamber. It has its maximum close to the outer wall of the exhaust duct. The circumferential velocity slowly decreases going towards the chamber wall. Radial and axial velocity profiles show that the flow is essentially downward directed (positive axial velocity) and moves towards the center of the annular chamber (negative radial velocity). Turbulence intensity profiles are quite similar for the various velocity components, indicating a certain isotropy of the flow, at least in this region.
Examination of the tangential velocity profiles indicates the expected forced/free combination of the Rankine vortex, consisting in an outer potential (free) vortex and a region of solid forced) body rotation at the core. The highest swirl velocity is found at the interface of the combined vortexes and results approximately 1.7 times the inlet mean velocity. The values of tangential velocity in the free vortex zone are higher in the cylinder body than in the annulus. Thus the separating force is higher and this suggests that the annulus should be short.
The central region is characterized by an intense rigid vortex off-centered with respect to geometrical axis. The angular velocity of this vortex is higher than that in cylindrical body owing to the conservation of angular momentum and area contraction from the cylindrical body to the exhaust duct. The presence of the central recirculation region is usually undesirable, because it induces higher pressure drops, possible working instability, mechanical vibrations and reduced separation efficiency. The simplest solution resulted in the insertion of a pole of rectangular section introduced inside the exhaust duct, along the internal wall, from the exit till the bottom cone of the cyclone. The presence of the pole highly attenuates the rigid central vortex and the recirculation zone is no more present.

The flow field within a cyclone separator can be regarded as a subsonic, two-dimensional gas-particle flow. The analysis of such a flow field is complicated by the fact that not only must the conservation equations account for the mass, momentum and energy of each phase, but additional equations are required to relate the momentum and energy coupling between phases. The aerodynamic drag on a particle due to particle-gas velocity differences gives rise to momentum coupling between phases. A finite difference formulation of the differential equations known as the “tank-and-tube” formulation provides systems of simultaneous algebraic equations for the flow properties which are solved on a digital computer. Crowe and Pratt (1974) have shown how this “tank-and-tube” formulation can be extended to the analysis of gas-particle flows, and it is this scheme which is applied to the problem. They have analyzed an axial-inlet, peripheral-discharge type cyclone separator. Swirl vanes at the inlet produce a rotational motion of the dust cloud and the centrifugal forces accelerate the dust particles toward the wall. The gas streamlines converge and exit through the gas-discharge pipe. The particles move to and subsequently along the separator wall toward the peripheral collector and then are removed. Only a small portion of the total gas flows through the separator is bled off with the dust particles. Of course, some of the smaller dust particles do not reach the wall and are carried out by the primary gas flow. The collection efficiency of the separator is defined as the mass of dust collected divided by the total mass of dust which enters he separator. The gas was assumed to be air at standard conditions. The gas mass bled off through the particle collector was neglected although it could be included if a more detailed solution were desired. It was also assumed that the particles and gas were in velocity equilibrium at entry. The important feature of the gas-particle flow analysis used in this study is the fact that it accounts for the particle-gas momentum coupling. When dust particles are introduced into the flow the radial speed of the gas toward the centerline is reduced. In fact, the reduction in radial speed varies monotonically with increased dust loading. This reduction in radial speed renders the gas flow field less effective in moving the dust particles towards the centerline and allowing their escape through the gas-discharge tube. Thus, the effect of increased dust loading is to increase collection efficiency which qualitatively agrees with and explains the experimental results. As dust loading is increased the collection efficiency of the smaller particles

increases accordingly. One notes as before a monatomic increase in efficiency with dust loading.
Avci and Karagoz (2001) developed a mathematical model with the definition of new parameters to predict the pressure loss coefficient including friction and local losses, and it is calculated in terms of geometrical and flow parameters, under some simplifying assumptions. They have analyzed a tangential cyclone separator. Having tangential, axial and radial velocity components, fluid mixture mainly accelerates towards to cone apex, while a small parts of the mixture leaves from the main flow to the outlet pipe before attaining to the cone apex.
Inlet velocity affects the pressure losses. Increasing the inlet velocity over a value which is necessary to create the vortex inside cyclone, results in increasing of Reynolds number and decrease of friction factor. But, in fully turbulent region, the friction factor depends on surface roughness and is not affected from the velocity. Increase of roughness causes decrease of acceleration and increase of flow losses. The final result is low velocity and low pressure losses, except in case of very small inlet width. The diameter of the outlet pipe in cyclones is an important parameter and has a complex effect on pressure losses and efficiency. It also increases friction surface area and decreases the inlet width. This results in slightly decreasing the pressure losses. The influence of the diameter of cone apex is clearly seen in this model. Because of increase of acceleration and decrease of flow losses, pressure losses increases with decreasing of the cone apex diameter. Experimental studies show that the height of the outlet pipe partially increases pressure losses. This parameter has two effects in the present model. The first one is that increase of outlet pipe height causes decrease of pressure losses due to increase of friction surface. The second one has an inverse effect, which is augmentation of outlet pipe height leads to decrease of flow losses, increases of acceleration and finally increase of pressure losses. The global effect is the slightly increase of pressure loss coefficient due to the second effect.
Another geometrical parameter is the cone height. Increase of the cone height decreases the friction surface and forces the flow vortices to diminish their radius. The former one which is included in the model causes increase of pressure losses and the latter one gives raise in the flow velocity and pressure losses. An increase of the resistance generally causes decrease of the pressure losses due to insufficient acceleration and low velocities. Increasing of inlet angle, that is, as approaching axial flow at the inlet causes

the number of swirl and acceleration to decrease. As a result, flow losses become high,
mean velocity will be low, flow path will be short, pressure losses and efficiency of
separation will be low.
According to Thorn (1998), cyclone efficiency will increase with particle density,
particle size, inlet velocity, cyclone body length, number of air stream revolutions, ratio
of body diameter to air outlet diameter and static pressure drop across the cyclone. The
efficiency will decrease with an increase in cyclone diameter, inlet width and area, air
viscosity and density.
To summarize the literature review, most of the work is related to tangential cyclone
separators with exit from top of the cyclone. Several techniques such as LDV and LDA
experiments have been used to study the flow within a cyclone. Simulations were
conducted in once case to develop a new cyclone with central body to reduce swirl and
hence pressure drop. Some mathematical models have been developed to see the effect
of different geometrical and other parameters on pressure drop and dust collection
efficiency. The present work deals with a new design related to axial entry cyclone
separators with uniflow direction so that the pressure drop can be reduced. The next
chapter presents the details of the problem definition and the objectives of the thesis.
OBJECTS OF THE INVENTION
The main object of the present-invention is to address the above limitations through a
novel solution based on the idea of maintaining the direction of the incoming flow
thereby reducing pressure loss from that of the conventional design.
Another object of the invention is to optimize this design for minimal pressure loss,
maximum dust collection efficiency and minimum overall size (to reduce cost).
Yet another object of the invention is increased in dust-holding and storage capacities,
servicing facilities, and maintenance periods.
Still another object of the present invention is to develop a filter to achieve the
aforementioned requirements.
Still another object of the present invention is to develop a method of filtering using
said pre-filter.
Still another object of the present invention is to develop a method of manufacturing
said filter.

STATEMENT OF THE INVENTION
The present invention relates to a cyclone separator-based air filter having tangential airflow from top to bottom with reduced pressure drop, said filter comprises
(i) frustum of cone with ratio of upper and lower diameter of the cone
ranging between 1.3 to 2.5; (ii) tangential inlet passage or ports attached to the cone for the air to enter
tangentially and to swirl intensely;
(iii) dust collection chamber located below the cone for dust collection; and (iv) suction pipe placed at center of the chamber below the cone for passage of
filtered air,
wherein separation gap is provided between the bottom of cone and suction pipe for separation of dust into dust chamber. Also,
A method for separating particle from air using cyclone separator-based air filter without any substantial pressure drop, said method comprising steps of:
a. passing air tangentially from top to bottom into the cone of the filter through
tangential inlet ports,
b. allowing the air to swirl intensely through the cone, and
c. separating particle from the air by collecting the particle into dust particle
collection chamber placed below the cone. And also
A method of manufacturing a cyclone separator-based air filter having unidirectional airflow from top to bottom without substantial pressure drop, said method comprises:
a. having frustum of cone with ratio of inlet and exit diameter of the cone ranging
between 1.3 to 2.0;
b. attaching angled inlet blades to the cone for the air to enter tangentially and to
swirl intensely;
c. placing dust particle collection chamber below the cone for trapping separated
particles; and
d. placing suction pipe at center of the chamber below the cone for passage of
filtered air;
wherein separation gap is provided between the bottom of cone and suction pipe for collection of particle into the dust chamber.

BRIEF DESCRIPTION OF THE ACCOMPANYING DRAWINGS
Figure 1 Photograph and the schematic of a conventional cyclone separator.
Figure 2 Photograph of the new design of cyclone separator.
Figure 3 & 4 Detailed drawing of the new design proposed for the cyclone separator.
Figure 5 Mesh of the tangential cyclone separator geometry used for simulation.
Figure 6 Experimental setup of the present invention.
Figure 7 Schematic of the Wind tunnel
Figure 8 Schematic of the dust feeder.
Figure 9 Experimental setup for testing the dust collection efficiency of the
conventional cyclone separator.
Figure 10 Experimental setup for testing the dust collection efficiency of the new
design of cyclone separator
Figure 11 A schematic of the three variations of the new design: axial, angled and
tangential entry.
Figure 12 Pressure distribution (Pa) in the central plane of cyclone.
Figure 13 Velocity vectors in the domain of cyclone.
Figure 14 Pressure distribution (Pa) in the central plane of cyclone.
Figure 15 Velocity vectors (m/s) in the domain of cyclone.
Figure 16 Pressure distribution (Pa) in the central plane of cyclone.
Figure 17 Velocity vectors (m/s) in the domain of cyclone.
Figure 18 Velocity vectors in the conical section of the axial entry cyclone
Figure 19 Velocity vectors in the conical section of the angled entry cyclone
Figure 20 Velocity vectors in the conical section of the tangential cyclone
Figure 21 Velocity vectors in a section near the bottom of the cone for axial entry,
angled entry and tangential entry cyclone separators.
Figure 22 Particle trajectories for the case corresponding to particle size of 1 μm
Figure 23 Particle trajectories for the case corresponding to particle size of 2 μm.
Figure 24 Particle trajectories for the case corresponding to particle size of 3 μm.
Figure 25 Particle trajectories for the case corresponding to particle size of 4 μm.
Figure 26 Variation of dust collection efficiency for tangential cyclone separator.

DETAILED DESCRIPTION OF THE INVENTION
The present-invention relates to a cyclone separator- based air filter for separating particles from air.
The invention relates to a cyclone separator-based air filter having unidirectional airflow from top to bottom without substantial pressure drop, said filter comprises
i. frustum of cone with ratio of upper and lower diameter of the cone ranging between 1.3 to 2.5;
ii. tangential inlet passage or ports attached to the cone for the air to enter tangentially and to swirl intensely;
iii. dust collection chamber located below the cone for dust collection; and
iv. suction pipe placed at center of the chamber below the cone for passage of
filtered air,
wherein separation gap is provided between the bottom of cone and suction pipe for separation of dust into dust chamber.
In yet another embodiment of the present invention, numbers of inlet blades are ranging between 6 to 20.
In still another embodiment of the present invention, number of inlet ports is about 8. In still another embodiment of the present invention, height of the cone is ranging between 80 to 140 mm.
In still another embodiment of the present invention, the separation gap is ranging between 2 to 10 mm.
In still another embodiment of the present invention, separation gap is about 6 mm. In still another embodiment of the present invention, exit diameter of the cone is ranging between 45 mm to 120 mm.
In still another embodiment of the present invention, lower diameter of the cone is about 70 mm.
In still another embodiment of the present invention, upper diameter of the cone is varied from 120 to 150 mm.
In still another embodiment of the present invention, upper diameter of the cone is about 134 mm.
In still another embodiment of the present invention, the ratio of upper to lower diameters of the cone is about 2.

In still another embodiment of the present invention, the filter is an accessory used with
IC engines operating in a dusty environment in vehicles selected from a group
comprising tractors and earthmovers.
In still another embodiment of the present invention, the filter is used in several fields
selected from a group comprising food, pharmaceutical, automobile, pollution control,
photographic film, and infrastructure.
A method for separating particles from air without substantial pressure drop using
cyclone separator-based air pre-filter, said method comprising steps of:
i. passing air uni-directionally from top to bottom into frustum of cone of the filter through angled inlet blades,
ii. allowing the air to swirl intensely through the cone, and
iii. separating particles through a separation gap into collection chamber placed
below the cone.
In still another embodiment of the present invention, frustum of cone with ratio of inlet (upper) and exit (lower) diameter of the cone ranging between 1.3 to 2.0; In still another embodiment of the present invention, number of inlet blades is about 8. In still another embodiment of the present invention, height of the cone is ranging between 60 to 140 mm.
In still another embodiment of the present invention, the separation gap is ranging between 2 to 10 mm.
In still another embodiment of the present invention, separation gap is about 6mm. In still another embodiment of the present invention, lower diameter of the cone is ranging between 45 mm to 120 mm.
In still another embodiment of the present invention, lower diameter of the cone is about 70 mm.
In still another embodiment of the present invention, upper diameter of the cone is varied from 120 to 150 mm.
In still another embodiment of the present invention, upper diameter of the cone is about 134 mm.
In still another embodiment of the present invention, the ratio of inlet to exit diameters of the cone is about 2.

In still another embodiment of the present invention, the filter is an accessory used with
IC engines operating in a dusty environment in vehicles selected from a group
comprising tractors and earthmovers.
In still another embodiment of the present invention, dust collection efficiency is 100%
for particles of size greater than or equal to 4 μm
In still another embodiment of the present invention, average dust collection efficiency
for particles of size ranging between 0 to 5 μm is higher than 50%.
In still another embodiment of the present invention, efficiency of dust collection
increases with reduction in exit diameter of the cone.
A method of manufacturing a cyclone separator-based air filter having unidirectional
airflow from top to bottom without substantial pressure drop, said method comprises:
a. having frustum of cone with ratio of inlet and exit diameter of the cone ranging
between 1.3 to 2.0;
b. attaching angled inlet blades to the cone for the air to enter tangentially and to
swirl intensely;
c. placing dust particle collection chamber below the cone for trapping separated
particles; and
d. placing suction pipe at center of the chamber below the cone for passage of
filtered air;
wherein separation gap is provided between the bottom of cone and suction pipe for collection of particle into the dust chamber.
In case of IC engines operating in a dusty environment, such as in tractors and earth movers, the filter elements get fouled within no time. In such cases, cyclone separators are used prior to the air filter. This will separate out the heavy dust particles. Only the very fine dust enters the filters thus increasing the life of the air filter. The conventional cyclone separator which is available commercially is shown in Figure 1. This cyclone separator separates the heavy dust particles and sends the semi pure air to the main filter. But there are also some draw backs of this design. The energy for generating the swirl is obtained by a certain amount of pressure drop which can affect the performance of the engine. It is also observed from Figure 1 that the flow entering is required to make a 180-degree turn before entering the suction pipe. Also, there is a chance of dust entering directly into the suction pipe.

Though the cyclone separator is effective in removing the coarse dust particles, these are some of the shortcomings of the conventional model. Thus, it is required to have a cyclone separator design which minimizes pressure drop, while maintaining the required dust collection efficiency. In order to meet these requirements, a new design is proposed.
Figures 2, 3 and 4 show the complete details of the new cyclone separator design. It consists of a frustum of cone attached to a certain number of inlet blades. The cone is located at the top of the separator. It consists of feed openings on the top of the cone. There are 8 passages or ports present. This enables the flow to swirl and the swirl intensifies as the flow passes through the conical section. The suction pipe of the engine is placed at the bottom part of the cone leaving a small gap to allow the particles to escape into the dust collection chamber. The inlets are open to the atmosphere. The suction pipe is connected to the filter.
In the conventional model, the cone was kept inversed and the flow was required to make a 180° turn. In the present model, the flow is unidirectional, thus minimizing the pressure loss. This new design also provides sufficient gap for the effective removal of dust particles.
This design intuitively seems better than the conventional model. However, there is a need for systematic study of the flow field inside the separator to assess the pressure loss and dust collection efficiency for various particle sizes. Furthermore, there is a need to optimize the system to maximize the dust collection efficiency and minimize the pressure loss. The next chapter presents this analysis and validation of this analysis with experimental data.
• CFD Simulation
Computational Fluid dynamics (CFD) is concerned with the development and application of numerical methods in solving the equations of fluid motion. Using CFD, we can test a wide variety of alternatives in much less time and at a significant savings in cost as compared with iterations using prototypes. It deals with the simulation of both compressible and incompressible flows. Typical activities include structured and unstructured grid generation, flow algorithm development for compressible and incompressible Navier -Stokes equations based on finite difference, finite volume and finite element methods, turbulence modeling, graphical post-processing, high performance computing and applications.

In the present work, simulations of the two-phase (gas-solid) flow in the cyclone
separator are performed using FLUENT 6.1, a CFD package. The cyclone separator
modeling essentially consists of the study of two-phase swirling flows, with one phase
representing the gaseous phase (air) and the other representing the solid phase (dust
particles). Swirling flows are among the most common and most complex in the
processing systems as cyclone separator. The CFD model helps to simulate these
complex tasks. The physical models available in FLUENT allow to accurately predict
laminar, transitional and turbulent flows, various modes of heat transfer, chemical
reaction, multiphase flows and other complex phenomena.
GAMBIT which is available with FLUENT, provides a concise and powerful set of
solid modeling-based geometry tools for generating meshes and grids that facilitate
CFD simulations. Different CFD problems require different mesh types, and GAMBIT
provides all options required in a single package. GAMBIT’S meshing toolkit
decomposes geometries for structured hex meshing or perform automated hex meshing
with control over clustering.
Formulation
This section describes the equations governing both the gaseous and solid phases, and
their interaction.
The Mass Conservation Equation
The equation for conservation of mass, or continuity equation, can be written as
follows:
∂p/∂t +∆.pv=0 (4.1)
This is the general form of the mass conservation equation and is valid for
incompressible as well as compressible flows.
Momentum Conservation Equations
Conservation of momentum in an inertial (non-accelerating) reference frame is
described as follows:

∂/∂T(PVP)+∆.P(PVPVP)= -∆P+∆.(Ŧ)+PGP+FP
where p is the static pressure, r is the stress tensor (described below), pjgand F are
the gravitational body force and external body forces (e.g., that arise from interaction with the dispersed phase), respectively.
The stress tensor f is given by

Ŧ=μ/

where// is the molecular viscosity, / is the unit tensor, and the second term on the right hand side is the effect of volume dilation.
Turbulence Modeling: The RNG k - G model
The RNG k - G model was derived using a rigorous statistical technique (called renormalization group theory). It is similar in form to the standard k - G model, but includes the following refinements:
• The RNG model has an additional term in its G equation that significantly
improves the accuracy for rapidly strained flows.
• The effect of swirl on turbulence is included in the RNG model, enhancing
accuracy for swirling flows.
• While the standard k - G model is a high-Reynolds-number model, the RNG
theory provides an analytically-derived differential formula for effective
viscosity that accounts for low-Reynolds-number effects. Effective use of this
feature does, however, depend on an appropriate treatment of the near-wall
region.
These features make the RNG k - G model more accurate and reliable for a wider class of flows than the standard k - G model. The RNG-based k - G turbulence model is derived from the instantaneous Navier-Stokes equations, using a mathematical technique called “renormalization group” (RNG) methods. The analytical derivation results in a model with constants different from those in the standard k - G model, and additional terms and functions in the transport equations for k and G.
The turbulence kinetic energy k and its rate of dissipation G are obtained from the following transport equations:

, d
(pkui) =
dk
a w . a .. „ . .. .„ „ Gk+Gb-ps-YM+Sk (4.4)

at ox: ox;

-Cuj(Gk+CieGb)-C2Ep^--Re +Se (4.5)
, k k
dt 3x: dx
In these equations, Gk represents the generation of turbulence kinetic energy due to the mean velocity gradients, Gh is the generation of turbulence kinetic energy due to buoyancy, YM represents the contribution of the fluctuating dilatation in compressible turbulence to the overall dissipation rate. The quantities errand «£are the inverse effective Prandtl numbers for k and e, respectively. Sk and SE are user-defined source
term.
The main difference between the RNG and standard k - 8 models lies in the additional
term in the 8 equation given by:
(4.6)
where 7j = Sk/s, rjo=438, fi = 0.012.
The model constants Cl£and C2s in have values derived analytically by the RNG
theory. These values, used by default in FLUENT, are : Cu= 1.42, Cl£ =1.68.
Equations of Motion for Particle
Particle force balance
FLUENT predicts the trajectory of a discrete phase particle (or droplet or bubble) by integrating the force balance on the particle, which is written in a Lagrangian reference frame. This force balance equates the particle inertia with the forces acting on the particle, and can be written (for the ^direction in Cartesian coordinates) as:
vFx (4.7)
p
dt Pp
where FD(u-u ) is the drag force per unit particle mass and
Here, u is the fluid phase velocity, up is the particle velocity, /J is the molecular viscosity of the fluid, p is the fluid density, p is the density of the particle, and dp is the particle diameter. Re is the relative Reynolds number, which is defined as

u = pd p | up- u|/
u (4.9)
In the present modeling, the 3D version of FLUENT is used. In this model, segregated
(Implicit solver) solution algorithm is used. The flow being solved is considered to be
in steady state, and the time-independent solution is solved as a first-order implicit
solution (i.e., first-order accurate dual time-stepping). For air and particle flow, discrete
phase model has been used and the turbulent flow is calculated using the RNG k-e
turbulence model. Particles of density 3300 kg/m3 have been injected into the inlet for
calculating the dust collection efficiency. The efficiency will be calculated by finding
the number of particles coming out of the exit pipe divided by the total number of
particles injected into inlet.
Boundary Conditions
Inlet:
The inlet is opened to the atmosphere. Therefore the inlet boundary is specified to be at
atmospheric pressure.
Outlet:
The outlet boundary condition is estimated using the mass flow rate. The mass flow
rate values have been taken from experiments and given as the outlet boundary
condition. The new design of cyclone separator with the complete mesh is shown in
the Figure 5
• Model Validation Experimental Setup
Any computational modeling effort requires experimental validation. Hence, a prototype of the new design of cyclone separator was fabricated and the pressure drop and dust particle collection efficiency were measured. The objective is to validate one set of simulations; and once validation is achieved, the model can be used to perform various simulations and optimization.
Thus, Experiments were conducted for measuring the pressure drop and the dust collection efficiency of conventional and the new cyclone separator. The schematic diagram of the total experimental setup is shown in Figure 6.

The various components of the experimental setup are described below.
Wind tunnel:
Wind tunnel is of suction type as shown in Figure 7. It is equipped with a two-stage
blower to provide required suction pressure inside the wind tunnel. It is equipped with
an orifice meter which is connected to a digital readout calibrated in mm of water. Flow
control valve can be used for varying the flow rate. The cyclone separator is connected
to the inlet side of the wind tunnel.
Manometer:
Manometer is an ordinary U-tube manometer. It is of one meter length. It uses water as
the manometric fluid. It is used to find the pressure drop across the cyclone separator.
Dust feeder:
Dust feeder is used in testing the dust feeding efficiency of the cyclone separator.
Figure 8 shows a schematic of the dust feeder. It is provided with timing adjustment to
feed the required amount of the dust in specified time. It has an arrangement to
introduce the dust into the air flow. Compressed air mixes with the dust and carries it
along with air. This air can be fed to cyclone separator to check its efficiency.
Experimental Procedure
Using the experimental setup shown in Figure 6, experiments were conducted to
measure the pressure drop and dust collection efficiency of the conventional and new
design of cyclone separators.
Pressure drop across the cyclone separator:-
The pressure drop was measured at different air mass flow rates through the cyclone
separators.
Procedure:
1) The cyclone separator was fixed to the experimental setup.
2) The flow rate of air was set using the flow control valve.
3) The flow rate and pressure drop across the cyclone separator were noted.
4) The procedure is then repeated for different flow rates for both conventional and
new models.
Past collection efficiency:-
The dust collection efficiency is important to estimate as the primary aim of a cyclone separator is to separate the dust from the air stream. The tests were conducted for the new and the conventional models for the purpose of comparison. The experimental

setups for dust collection efficiency of conventional and new design of separators are
shown in Figures 9 and 10.
Procedure:
1) The flow control valve was set to a particular flow rate and was maintained
constant throughout the experiment.
2) The separator was weighed and was fixed to the experimental setup. Weighed
amount (110 gm) of dust was filled into the dust feeder.
3) The feed rate of air into the cyclone separator was set to 237.6 kg/hr. The
timer of the dust feeder was set such that 110 gm of dust was fed in 1 hr.
4) After all the dust was fed, the weight of the separator was measured. The
difference between the weights gives the dust collected, from which dust
collection efficiency of the cyclone separator was calculated.
Observations
The detailed experimental data is given below. Pressure drop
The pressure drop is calculated as follows: Ap - pgh Pa
where p - 1000 kg/m for water
g =9.81 m/s2
h = manometer reading in m of water Dust collection efficiency
The dust collection efficiency for both conventional and new design of cyclone separator was found by weighing the mass of dust collected and mass of dust sent to the separators.
_ ,, . nr. . Mass of dust collected in separator
Dust collection efficiency =
Mass of dust supplied to the separator The experimental values of dust collection efficiency are given in Table 4.2.

Table 4.2 Dust collection efficiency of the cyclone separators
The dust which was used in experiments consists of many particle sizes and particles of many materials. The details of particle size distribution and the chemical composition of particles are given in Tables 4.3 and 4.4.

Table 4.4 Chemical composition of the dust particles
PARAMETRIC STUDY
The previous chapters concerned the analysis of a new design of the cyclone separator, and also validation of the numerical model with the experimental data. From the results, it was clear that the new design that was based on intuition is clearly superior to the conventional design in terms of both pressure loss and dust collection efficiency. It is however, important to explore variations in this new design and further to optimize it. In this regard, three different variations based on the air entry were studied here. These are:
1. Axial entry cyclone
2. cyclone with angled entry
3. Tangential entry cyclone
1. Axial entry cyclone:
This is the new design of the cyclone separator that was described in the previous chapter. The results of this separator will be compared with the other two cyclone separators. The comparison will be done using the results of simulations.
2. Cyclone with angled entry:
In this variation of the new design, the air entry is partially axial and partially tangential, with the plane of each entry duct making an angle of 45° with the axis of the separator as shown in Figure 11.

3, Tangential entry cyclone:
In this variation of the new design, the air enters completely tangential to the axis of the separator as shown in Figure 11.
Performance Comparison of Design Variations
Each particular variation in the new design was simulated using FLUENT. The two-phase flow was computed. The boundary conditions being same as in the simulation of the first new design i.e. atmospheric pressure at inlet and the mass flow rate at the outlet.
Figure 12 shows the absolute pressure distribution in the central plane of the axial entry cyclone. It can be observed that the pressure decreases as the air flows down the cone. This can be attributed to the fact that the velocities increase as the flow reaches the bottom of the cone and hence pressure decreases. The major pressure drop is near the entrance to the exit pipe because of the change in area of flow. Figure 13 shows the velocity vectors with magnitudes for the design with complete axial entry. It can be observed from the figure that the air entering has some tangential velocity component with the axial velocity. The velocity increases as the air passes down the conical section. The peak velocity occurs at the bottom of the conical section and is about 30 m/s. The velocity magnitudes in the dust collection chamber are very less as the air slows down in the chamber. The velocity is more in exit pipe compared to the inlet because of the reduction in area.
Figure 14 shows the pressure distribution in the central plane of the cyclone with angled entry. The pressure is very less near the axis of the conical section. This is because of very high velocities near the axis. Figure 15 shows the velocity vectors of the cyclone with angled entry. The colour bar shows the magnitudes. It can be seen from the velocity vectors that the tangential velocity is very high at the top of the cone section. The velocities increase as the air flows down the conical section and the tangential velocity is maximum at the end of the conical section. The magnitude of the maximum velocity is about 60 m/s. This magnitude is almost double than that in the axial cyclone separator. Thus the pressure drop is much more in this case compared to the axial entry cyclone as can be seen from the Figure 14.
Figure 16 shows the pressure contours in the central plane of the design with tangential entry. In this case also, the pressure is very less near the axis of the conical section because of very high tangential velocities. Figure 17 shows the velocity vectors for the

cyclone with angled entry. In this design the inlet velocity is complete tangential as can be seen in the figure. The velocities in this model are somewhat lower than in angled entry design and higher than in axial entry cyclone. The maximum velocity is around 48 m/s which occur at the end of the conical section. The pressure drop is also somewhat lower compared with that of angled entry model but is still very high compared with axial entry model.
Figure 18 shows the velocity vectors in the conical part of the axial entry cyclone separator. It can be observed from the figure that the velocities are partly axial and partly tangential. Figure 19 shows the velocity vectors in the conical part of the cyclone with angled entry. In this cyclone the velocities are completely tangential The magnitude of velocities are also very high. Figure 20 shows the velocity vectors in the conical part of the tangential entry cyclone separator. In this case also the velocities are very high and are mostly tangential.
Figure 21 shows that the velocities near the bottom of the conical section are partly axial and partly tangential for axial entry cyclone and the velocities are almost tangential for the angled entry and tangential entry cyclones.
From the simulations, the total pressure drops across the three types of design are found to be as follows:
Axial entry cyclone : 576.1 Pa
Cyclone with angled entry : 5200Pa
Tangential entry cyclone : 2060Pa
Since the pressure drop for the second and third design variations is very high compared with the first model, which is axial cyclone separator, these are not considered for further study. Hereafter, a parametric study has been conducted for the axial entry cyclone separator in order to optimize the design for maximum dust collection efficiency and minimum pressure drop.
In conventional terms, the air filtering is done in two stages: a pre-filter, i.e., cyclone separator to separate heavier dust particles and a main filter to separate the finer particles. The work presented in this thesis so far concerns the development of a new design for the pre-filter. The assumption is that there is a conventional main filter present downstream which will separate the finer dust particles. Now, in conventional systems, the main filter is supposed to separate the particles of size 5 micron and

upwards. Thus, the filter typically removes particles above 30 - 40 um, and the main filter is supposed to trap particles between 5-30 um.
Now, some of the new designs simulated in this study have shown dramatic results with respect to dust collection efficiency. These results prompted us to propose a radically new concept in air filtration which does not require polyurethane foam-based main filter at all. In other words, some of the so-called filter designs simulated here show a dust collection efficiency of 100% for particles of 4 um and higher. This implies that this cyclone separator based filter can do the job of the pre-filter and main filter configuration. This radical concept can significantly lower the cost of the air filter unit, and more importantly, the running costs also are reduced to zero, since there is no replaceable part at all (as opposed to the conventional filter where the air filter unit needs to be replaced periodically).
However, it is important to also estimate the pressure drop of this new design. First of all, Figure 5 shows the configuration of this design along with the mesh generated. As seen from Figure 5, the entries are all tangential which causes the swirl velocities inside the separator to be very high. This in turn explains the very high dust collection efficiencies of this design for various particle sizes beyond 4 um The particle trajectories for the different particle sizes from 1 nm to 4 um are shown in Figs. 22-25. It can be observed from these figures that the efficiency increases drastically with particle size. As mentioned before, the high efficiency is due to the very high swirl velocities. Figure 26 shows the plot of variation of dust collection efficiency with particle size. It shows that the efficiency is 100% for particle sizes of 4 um and above. Therefore the use of this tangential cyclone separator eliminates the necessity of using a main filter downstream. The overall pressure drop predicted is around 2000 Pa. This result is extremely encouraging since the total pressure drop for a conventional pre-filter and main filter configuration is around 5000 Pa. Thus, this design promises less than half the pressure drop, and significantly lower costs, while maintaining effective separation for particle sizes of 4 micron and above. However, this predicted performance needs to be confirmed with experiments. In any case, based on the earlier model validation, there is a very high probability of computations matching results.

Parametric Study
A number of geometric variables related to the cyclone separator geometry were initially investigated and four parameters have been selected for detailed study and optimization. These are:
1. Separation gap (between cone and exit pipe)
2. Cone height
3. Ratio of entry diameter to exit diameter of the cone (D/d)
4. Number of vanes at the inlet.
Separation gap:
The gap between the bottom of frustum of cone and the exit pipe is varied from 2mm to 8mm to evaluate its effect on the performance. The results are displayed in the form of comparison of dust collection efficiency and also pressure drop across the separator. The pressure drop increases as the gap decreases. This can be attributed to the fact that as the gap reduces; more and more air will be going to the collection chamber causing the loss of momentum, thereby subsequent increase in pressure drop. The efficiency is almost same for small particles. As the particle size increases the efficiency increases. The efficiency is higher for the design with higher gap at the separation point.
Cone height:
The height of the cone is varied to find out the effect of its size on the dust collection efficiency and pressure drop. The efficiency increases with the length of the cone. This is because the efficiency increases with the residence time and the residence time will be more for cyclone of more height. At the same time, the pressure drop decreases with the increase in height of the cone. This is because the convergence angle of the cone decreases as the height of the cone increases fixing the diameters constant. Therefore the velocities will be lower and hence lower pressure drop at higher cone heights.
Cone diameter ratio:
The exit diameter of the cone is varied by 70, 80, 90 and 100mm. The dust collection efficiency increases with decrease in lower diameter of the cone. The increase in swirl velocity at the separation point because of reduction in the lower diameter of cone is the main reason for the increase in efficiency, since higher swirl velocity causes a larger centrifugal force on the particle. The pressure drop increases with the increase in the lower diameter. This is because, as the diameter increases, the amount of air (at

high velocity) entering the collection chamber increases, causing increase in the pressure drop.
Number of vanes:
The number of vanes at the inlet of the separator is varied from 8 to 16. The dust collection efficiency increases with the increase in number of vanes at inlet. This is because of the better directing capability with more number of vanes, which implies a higher tangential velocity component leading to a higher swirl velocity at separation point. The pressure drop, at the other hand, increases with increase in number of vanes. This is because of more resistance to the flow due to more number of vanes. Based on all the above four parameters, the best possible combination of all the four parameters is selected to give good dust collection efficiency while giving low pressure drop. Simulations were again conducted on the model with this combination of parameters. The values of these parameters for the best combination are as follows:
The gap where the separation of particles take place = 6mm
The height of the cone = 100mm
The lower diameter of the cone = 70mm
No. of vanes at the inlet = 8
Purpose of Dust Collection
Dust filtration is important in many applications, some of which are described below:
• Air-pollution control, as in fly-ash removal from power plant flue gases.
• Equipment maintenance reduction, as in filtration of engine intake air or
pyrites furnace gas treatment prior to its entry to a contact sulphuric acid
plant.
• Safety or health hazard elimination, as in collection of siliceous and metallic
dusts around grinding and drilling equipment and in some other metallurgical
operations and flour dusts from milling or bagging operations.
• Product quality improvement, as in air cleaning in production of
pharmaceutical products and photographic film.
• Recovery of a valuable product, as in collection of dusts from dryers and
smelters.

Powdered product collection, as in pneumatic conveying, the spray drying of milk, eggs and soap; and manufacturing of high purity zinc oxide and carbon black.
CONCLUSION
A new cyclone separator-based filter with applications for IC Engines operating in a dusty environment has been simulated. This design shows reduced pressure drop and increased dust collection efficiency over conventional systems. The flow direction was made axial and uni-directional so that the pressure drop can be reduced. In the uni-flow cyclone the air enters axially through number of inclined vanes which provide some tangential velocity thereby increasing the swirl at the separation region and the particles are get separated at the periphery and the air leaves through the central outlet pipe. This enables a good separation of dust particles. Experiments have been done to measure the pressure drop and dust collection efficiency for both the conventional and the new cyclone separator design. A huge reduction in pressure drop was observed with the new cyclone separator compared to the conventional cyclone separator. More significantly, the dust collection efficiency for the new cyclone separator was more than that of the conventional cyclone separator. Simulations were done on the new cyclone separator and the results were compared with the values of experiments. There was a good comparison of pressure drop between experiments and simulations except at very low mass flow rates. This may be because of the uncertainty associated with the measurement of the mass flow rate at low values. The dust collection efficiency from simulations was observed to closely match with the experimental value. A detailed parametric study was performed on the new cyclone separator, having axial entry and uni-directional flow, with respect to its geometrical parameters to optimize the performance in terms of pressure drop across the cyclone separator and the dust collection efficiency. A new optimized design is proposed based on the results of the parametric study.
The results of the simulation for the tangential entry cyclone separator indicated very high dust collection efficiencies, i.e., 100% for particle sizes of 4 micron and higher. The air entry for this new design of separator is completely tangential. This gives very high tangential velocities in the conical section of the cyclone separator thus providing very high separation efficiency. The pressure drop is around 2000 Pa. Now, this is much less than the total pressure drop across a conventional pre-filter cum air filters

combination, where the pressure loss is around 5000 Pa. This prompted a radically new idea in air filtration for such applications. The idea is to completely eliminate the use of the polyurethane foam-based main filter, and use only a tangential entry cyclone separator. This design anyway satisfies the requirement those particles of size 5 micron and above need to be completely filtered. However, experiments are needed to validate the results of this new design. In conclusion, this design has potential not just in improving performance of the engine (by means of a reduced pressure drop), but also in significantly reducing the costs (both initial and running) of air filters for tractors and earth moving equipment applications.

Documents:

634-che-2006-abstract.pdf

634-che-2006-abstractimage1.jpg

634-che-2006-abstractimage2.jpg

634-che-2006-claims.pdf

634-che-2006-correspondence-others.pdf

634-che-2006-correspondence-po.pdf

634-che-2006-description(complete).pdf

634-che-2006-drawings.pdf

634-che-2006-form 1.pdf

634-che-2006-form 26.pdf

634-che-2006-form 3.pdf

634-che-2006-form 5.pdf

634-che-2006-form 9.pdf

« Previous Patent

Next Patent »

Patent Number

234381

Indian Patent Application Number

634/CHE/2006

PG Journal Number

29/2009

Publication Date

17-Jul-2009

Grant Date

26-May-2009

Date of Filing

06-Apr-2006

Name of Patentee

INDIAN INSTITUTE OF SCIENCE

Applicant Address

CENTRAL OFFICE (FIRST FLOOR) BANGALORE 560 012

Inventors:

#	Inventor's Name	Inventor's Address
1	R.V. RAVIKRISHNA	DEPARTMENT OF MECHANICAL ENGINEERING, INDIAN INSTITUTE OF SCIENCE BANGALORE 560 012
2	M.V. NARASIMHAN	DEPARTMENT OF MECHANICAL ENGINEERING, INDIAN INSTITUTE OF SCIENCE BANGALORE 560 012
3	K.MADUSUDHAN	DEPARTMENT OF MECHANICAL ENGINEERING, INDIAN INSTITUTE OF SCIENCE BANGALORE 560 012

PCT International Classification Number

BO1D46/00

PCT International Application Number

N/A

PCT International Filing date

PCT Conventions:

#	PCT Application Number	Date of Convention	Priority Country
1			NA