Advanced Powder Technol, Vol. 14. No 6, pp. 657-672(2003) Review po Gas cleaning at high temperatures using rigid ceramic filters JONATHAN SEVILLE TEONG GUAN CHUAH. VUSUMUZISIBANDA and PETER KNight Centre for Formulation Engineering, Department of Chemical Engineering, University of Birmingham, Edgbaston, Birmingham, B15 2TT, Ul Received 31 March 2003: accepted 16 May 2003 Abstract-Rigid ceramic filters have emerged in the last two decades as the most promising technology for particulate removal from process gases at temperatures up to 1000C. Granular nd fibrous forms of media have been developed and both are commonly employed in the form of ylindrical'candles' which are periodically cleaned by application of a reverse gas pulse. Research has focused on this cleaning process, which governs the long-term performance of the filter. The problem is 2-fold: to determine the dust cake'detachment stress, which depends on the dust particle roperties and cake structure, and to understand the propagation of the cleaning pulse which is applied to remove it. The results of research in these areas are summarized and experimental methods for the investigation of filter cleaning briefly described. The implications for design and further development of ceramic filters are discussed Keywords: Ceramic filters; fibrous filters; cake detachment; reverse pulse; fluid mechanics, particle mechanics 1 INTRODUCTION The development of technologies for particulate removal from gases at high temperatures has been rapid over the last two decades, following the pioneering work of the UK/US/German collaborative project at the Grimethorpe Pressurized Fluidized Bed Combustion(PFBC) Facility in the late 1970s and early 1980s. The long-term needs of the power generators may have driven this early development, but the focus has now shifted to the chemical and process industries. Environmental legislation being rapidly implemented in most industrialized countries means that their needs are anything but long-term! Furthermore, their filtration requirements whom correspondence should be addressed. E-mail: J.P.K. Seville@bham ac uk
Advanced Powder Technol., Vol. 14, No. 6, pp. 657– 672 (2003) Ó VSP and Society of Powder Technology, Japan 2003. Also available online - www.vsppub.com Review paper Gas cleaning at high temperatures using rigid ceramic lters JONATHAN SEVILLE¤, TEONG GUAN CHUAH, VUSUMUZI SIBANDA and PETER KNIGHT Centre for Formulation Engineering, Department of Chemical Engineering, University of Birmingham, Edgbaston, Birmingham, B15 2TT, UK Received 31 March 2003; accepted 16 May 2003 Abstract—Rigid ceramic lters have emerged in the last two decades as the most promising technology for particulate removal from process gases at temperatures up to 1000±C. Granular and brous forms of media have been developed and both are commonly employed in the form of cylindrical ‘candles’ which are periodically cleaned by application of a reverse gas pulse. Research has focused on this cleaning process, which governs the long-term performance of the lter. The problem is 2-fold: to determine the dust cake ‘detachment stress’, which depends on the dust particle properties and cake structure, and to understandthe propagation of the cleaning pulse which is applied to remove it. The results of research in these areas are summarized and experimental methods for the investigation of lter cleaning briey described. The implications for design and further development of ceramic lters are discussed. Keywords: Ceramic lters; brous lters; cake detachment; reverse pulse; uid mechanics, particle mechanics. 1. INTRODUCTION The development of technologies for particulate removal from gases at high temperatures has been rapid over the last two decades, following the pioneering work of the UK/US/German collaborative project at the Grimethorpe Pressurized Fluidized Bed Combustion (PFBC) Facility in the late 1970s and early 1980s. The long-term needs of the power generators may have driven this early development, but the focus has now shifted to the chemical and process industries. Environmental legislation being rapidly implemented in most industrialized countries means that their needs are anything but long-term! Furthermore, their ltration requirements ¤To whom correspondence should be addressed. E-mail: J.P.K.Seville@bham.ac.uk
658 J. Seville et al Table 1 Applications for hot gas cleaning and their operational requirements (1l Ap Operating conditions Gas environment Filter device Ie PFBC oxidizing with alkali turbine protection; meet envi ronmental standards integrated gasifica- 600-800 10 reducing with alkali turbine protection; meet envi- tion combined cycle +H?s ronmental standards, protect sulfur capture beds onventional <700 oxidizing meet environmental standards. Chemical proces 300-750 1-3 varied, can be severe enhanced product recovery: metal refining reduced environmental emi calcination/ drying sions, resource recovery, en- catalytic cracking ergy recovery precious metal Incineration up to 100 oxidizing. contain- Reduce environmental er hazardous waste ing reactive chemi- sions: improve incineration cal species process; protect downstream kiln furnace equipment are at least as challenging as the high-pressure, high-temperature filtration problems which stimulated the earlier development of the technology Table 1 [1] gives some examples of applications for hot gas cleaning and their operating requirements. The first group comprises three distinct types of system for electrical power generation from coal, all of which have their own requirements for gas cleaning at high temperatures ne chemical and process industries and in incineration, the need for gas cleaning is being driven increasingly by the requirements of environmental legislation, which has been directed specifically at particulates, acid gases, heavy metal compounds, hydrogen chloride, and organic chlorides such as dioxins and furans. The reasons for interest in cleaning gases hot rather than cold are many and various. They include a desire to remain well above acid dew points, improved thermodynamic efficiency, especially where downstream heat recovery is employed, and an improvement in the versatility of the overall process. The ability to clean process gases hot also allows the simultaneous removal of gaseous components by 'dry scrubbing,, whereby finely divided reactant or adsorbent solids are introduced up-steam of the filter [2]. The solids most commonly employed are calcium carbonate, oxide or hydroxide, to remove SO and HCl, and activated carbon to remove organic chlorides and heavy metals. a further poss advantage concerns the reformation of chlorinated organics, which is thought to be catalyzed by elements of the particulates in the process gases as they are cooled [3]
658 J. Seville et al. Table 1. Applications for hot gas cleaning and their operational requirements [1] Application Operating conditions Gas environment Filter device requirement Tempera- Pressure ture ( ±C) (bar) Power generation PFBC 800 10 oxidizing with alkali turbine protection; meet environmental standards integrated gasication combined cycle 600– 800 10– 30 reducing with alkali C H2S turbine protection; meet environmental standards, protect sulfur capture beds conventional <700 1 oxidizing meet environmentalstandards; low 1p Chemical process metal rening calcination/ drying catalytic cracking precious metal recovery 300– 750 1–3 varied, can be severe enhanced product recovery; reduced environmental emissions; resource recovery; energy recovery Incineration hazardous waste municipal waste kiln furnace up to 1000 1 oxidizing, containing reactive chemical species Reduce environmental emissions; improve incineration process; protect downstream equipment are at least as challenging as the high-pressure, high-temperature ltration problems which stimulated the earlier development of the technology. Table 1 [1] gives some examples of applications for hot gas cleaning and their operating requirements. The rst group comprises three distinct types of system for electrical power generation from coal, all of which have their own requirements for gas cleaning at high temperatures. In the chemical and process industries and in incineration, the need for gas cleaning is being driven increasingly by the requirements of environmental legislation, which has been directed specically at particulates, acid gases, heavy metal compounds, hydrogen chloride, and organic chlorides such as dioxins and furans. The reasons for interest in cleaning gases hot rather than cold are many and various. They include a desire to remain well above acid dew points, improved thermodynamic efciency, especially where downstream heat recovery is employed, and an improvement in the versatility of the overall process. The ability to clean process gases hot also allows the simultaneous removal of gaseous components by ‘dry scrubbing’, whereby nely divided reactant or adsorbent solids are introduced up-steam of the lter [2]. The solids most commonly employed are calcium carbonate, oxide or hydroxide, to remove SOx and HCl, and activated carbon to remove organic chlorides and heavy metals. A further possible advantage concerns the reformation of chlorinated organics, which is thought to be catalyzed by elements of the particulates in the process gases as they are cooled [3];
Gas cleaning at high temperature 659 removal of such particulate catalysts should therefore prevent emissions of these In many applications, it is conventional to carry out the preliminary coarse particle capture in one or two stages of cyclones. However, cyclones have been found to be ineffective for collection of particles much below about 7 um so that it is conventional to apply a so-called tertiary gas cleaning stage, which at near-ambient temperature is frequently a fabric or bag filter. There is still debate about necessity for tertiary gas cleaning and the newest generation of PFBC plants has only two stages of cyclones to protect the turbine Of the available tertiary cleaning methods, rigid ceramic barrier filters are the most promising and the most highly developed [4], although advances continue to be made in the development of high-temperature fabrics for bag filters, and there are likely to be specific application for metal filters [5, 6], granular bed filters [7] and electrostatic precipitators [8]. It is the intention here to review the recent advances understanding of the behavior of rigid ceramic filters. For a wider view of hot cleaning in general, the reader is referred to the Proceedings of the four International Conferences on 'Gas Cleaning at High Temperatures[9-12] 2. ADVANCES IN DESIGN AND UNDERSTANDING For those unfamiliar with the technology, Fig. I represents a typical filter arrange ment. Conventionally, the filter medium is provided in the form of long hollow tubes, or,, closed at one end and hung vertically from a tube plate such that the gas to be filtered passes from the outside inwards, depositing a dust cake on the CLEANED CAS HOT DUSTY LOCH Figure 1. Typical candle filter house(after British Coal)
Gas cleaning at high temperatures 659 removal of such particulate catalysts should therefore prevent emissions of these damaging compounds. In many applications, it is conventional to carry out the preliminary coarse particle capture in one or two stages of cyclones. However, cyclones have been found to be ineffective for collection of particles much below about 7 ¹m so that it is conventional to apply a so-called tertiary gas cleaning stage, which at near-ambient temperature is frequently a fabric or ‘bag’ lter. There is still debate about the necessity for tertiary gas cleaning and the newest generation of PFBC plants has only two stages of cyclones to protect the turbine. Of the available tertiary cleaning methods, rigid ceramic barrier lters are the most promising and the most highly developed [4], although advances continue to be made in the development of high-temperature fabrics for bag lters, and there are likely to be specic application for metal lters [5, 6], granular bed lters [7] and electrostatic precipitators [8]. It is the intention here to review the recent advances in understanding of the behavior of rigid ceramic lters. For a wider view of hot gas cleaning in general, the reader is referred to the Proceedings of the four International Conferences on ‘Gas Cleaning at High Temperatures’ [9 – 12]. 2. ADVANCES IN DESIGN AND UNDERSTANDING For those unfamiliar with the technology, Fig. 1 represents a typical lter arrangement. Conventionally, the lter medium is provided in the form of long hollow tubes, or ‘candles’, closed at one end and hung vertically from a tube plate such that the gas to be ltered passes from the outside inwards, depositing a dust cake on the Figure 1. Typical candle lter house (after British Coal).
660 J. Seville et al 59, 20.0 ky 100um Labor Lang 三84 (b) Figure 2.(a) Granular ceramic medium (Schumacher Dia-Schumalith F40).(b) Fibrous ceramic medium(Cerel Cerafil 2001)
660 J. Seville et al. (a) (b) Figure 2. (a) Granular ceramic medium (Schumacher Dia-Schumalith F40). (b) Fibrous ceramic medium (Cerel Ceral 2001).
Gas cleaning at high temperatures outside of the candle. Each candle is cleaned periodically by applying a short pulse of cleaning gas which causes the gas flow direction to reverse momentarily so that the dust cake is removed and falls into a suitable receiving vessel. This conventional arrangement is very similar to that employed in fabric filter technology There are two generic types of ceramic filtration media(Fig. 2): the high-density granular media typified by sintered or bonded silicon carbide and low-density fibrous media usually formed from bonded fibers consisting largely of alumina. Fibrous media have been developed more recently than the granular types, and because of their high void fraction(typically 90-95%)they are lighter, show less resistance to flow and are less prone to thermal shock. They are also cheaper and their development has opened up many new applications to ceramic filters which were not previously practical using granular media. Having developed a suitable hot gas filtration medium, it was already known from bag filter experience that the long-term pressure drop history was inextricably linked to the cleaning behavior. Both bag filters and rigid ceramic filters are designed to operate as surface filters; they share the characteristic that if the gas velocity is too high, particles will penetrate into the bulk of the medium, resulting in irreversible long-term rise in resistance to flow. The first major advance in understanding however, was the recognition that the cleaning mechanisms for bag filters and rigid ceramic filters are different. To a first approximation, the dust cake detaches when it experiences a tensile stress sufficient to overcome either the strength of the adhesive bond between the cake and the filter medium or the internal cohesion of the cake As soon as the strength of this adhesive or cohesive bond is exceeded(by whatever cleaning mechanism), the cake detaches. In a bag filter, the required cleaning stress is set up primarily by the movement caused by the cleaning pulse or, in the case of a mechanically cleaned filter, th shaking action. Pulse cleaning displaces the fabric outwards. When it becomes aut,it decelerates sharply, normally at many times gravitational acceleration. the cake then experiences a tensile stress which depends on its areal density and on the deceleration. Rigid media such as the ceramics considered here show no splacement on cleaning. The tensile stress is therefore entirely the result of th oressure drop imposed across the cake due to reverse flow of cleaning gas(Fig. 3) Reverse flow of cleaning gas can be an important mechanism for flexible media too, especially if the bag is very long This understanding of the mechanisms of rigid filter cleaning has several impor- nt consequences, one of which is that, other things being equal, thicker cakes require less cleaning gas flow for their detachment. Furthermore, the cleaning effi ciency must depend on the peak pressure difference generated across the cake, and is only indirectly related to the pressure in the cleaning gas reservoir and the pres sure drop across the medium-plus-cake. These and other factors relating to cleaning efficiency are fully discussed in the references listed at the end of this paper In practice, the filter cake does not detach everywhere simultaneously when some critical stress is reached. Neither the cake detachment stress nor the applied stress
Gas cleaning at high temperatures 661 outside of the candle. Each candle is cleaned periodically by applying a short pulse of cleaning gas which causes the gas ow direction to reverse momentarily so that the dust cake is removed and falls into a suitable receiving vessel. This conventional arrangement is very similar to that employed in fabric lter technology. There are two generic types of ceramic ltration media (Fig. 2): the high-density granular media typied by sintered or bonded silicon carbide and low-density brous media usually formed from bonded bers consisting largely of alumina. Fibrous media have been developed more recently than the granular types, and because of their high void fraction (typically 90– 95%) they are lighter, show less resistance to ow and are less prone to thermal shock. They are also cheaper and their development has opened up many new applications to ceramic lters which were not previously practical using granular media. Having developed a suitable hot gas ltration medium, it was already known from bag lter experience that the long-term pressure drop history was inextricably linked to the cleaning behavior. Both bag lters and rigid ceramic lters are designed to operate as surface lters; they share the characteristic that if the gas velocity is too high, particles will penetrate into the bulk of the medium, resulting in irreversible long-term rise in resistance to ow. The rst major advance in understanding, however, was the recognition that the cleaning mechanisms for bag lters and rigid ceramic lters are different. To a rst approximation, the dust cake detaches when it experiences a tensile stress sufcient to overcome either the strength of the adhesive bond between the cake and the lter medium or the internal cohesion of the cake. As soon as the strength of this adhesive or cohesive bond is exceeded (by whatever cleaning mechanism), the cake detaches. In a bag lter, the required cleaning stress is set up primarily by the movement caused by the cleaning pulse or, in the case of a mechanically cleaned lter, the shaking action. Pulse cleaning displaces the fabric outwards. When it becomes taut, it decelerates sharply, normally at many times gravitational acceleration. The cake then experiences a tensile stress which depends on its areal density and on the deceleration. Rigid media such as the ceramics considered here show no displacement on cleaning. The tensile stress is therefore entirely the result of the pressure drop imposed across the cake due to reverse ow of cleaning gas (Fig. 3). Reverse ow of cleaning gas can be an important mechanism for exible media too, especially if the bag is very long. This understanding of the mechanisms of rigid lter cleaning has several important consequences, one of which is that, other things being equal, thicker cakes require less cleaning gas ow for their detachment. Furthermore, the cleaning ef- ciency must depend on the peak pressure difference generated across the cake, and is only indirectly related to the pressure in the cleaning gas reservoir and the pressure drop across the medium-plus-cake. These and other factors relating to cleaning efciency are fully discussed in the references listed at the end of this paper. In practice, the lter cake does not detach everywhere simultaneously when some critical stress is reached. Neither the cake detachment stress nor the applied stress
J. Seville et al Gas flow △Pr Figure 3. Pressure distribution in medium and cake during reverse flow. is entirely uniform across the filter surface and this may be why 'patchy cleaning is observed, i. e cake is almost completely removed in some retained in others. Space does not permit discussion of the consequences of patchy cleaning for filter operation, save to note that when the filtration flow is restored after incomplete cleaning, the local gas velocity in the cleaned areas may initially be many times greater than its average or overall face velocity [13] 3 CURRENT RESEARCH Figure 4 summarizes the research issues which have been of most concern to both academic and industrial workers For the academic, the major difficulty in obtaining a full understanding of filter operation is that both filtration and cake detachment steps depend on both the fluid mechanics of the gas flow and the particle mechanics of cake formation and cake detachment (Table 2). These issues are discussed further below For the industrial user, reassurance that the ceramic filter will provide long-term stable operation is the first essential. Stephen [15]su Does the filter have sufficient mechanical strength and erosion resistance to withstand its operating environment? How can a minimal and stable operating pressure drop be achieved over the filter
662 J. Seville et al. Figure 3. Pressure distribution in medium and cake during reverse ow. is entirely uniform across the lter surface and this may be why ‘patchy cleaning’ is observed, i.e. cake is almost completely removed in some areas and completely retained in others. Space does not permit discussion of the consequences of patchy cleaning for lter operation, save to note that when the ltration ow is restored after incomplete cleaning, the local gas velocity in the cleaned areas may initially be many times greater than its average or overall face velocity [13]. 3. CURRENT RESEARCH Figure 4 summarizes the research issues which have been of most concern to both academic and industrial workers. For the academic, the major difculty in obtaining a full understanding of lter operation is that both ltration and cake detachment steps depend on both the uid mechanics of the gas ow and the particle mechanics of cake formation and cake detachment (Table 2). These issues are discussed further below. For the industrial user, reassurance that the ceramic lter will provide long-term stable operation is the rst essential. Stephen [15] summarizes the issues: ² Does the lter have sufcient mechanical strength and erosion resistance to withstand its operating environment? ² How can a minimal and stable operating pressure drop be achieved over the lter life span?
Gas cleaning at high temperatures cake formation Filtration mechanical prop chemical resistance Cake detachment geometry fracture mechani novel geometries filter surface modification <Combinations of cyclone filter (Combinations of physical chemical separators Figure 4. Research issues in the development of ceramic filters for hot gases [141 Table 2 Academic issues in ceramic filter research [141 Fluid mechanics Particle mechanics Filtration flow field distribution cake formation Cleaning pulse propagation cake detachment What is the effect of the challenging dust, in relation to the operating temperature. on the operating pressure drop? Is the filter medium suitable in relation to the particle size distribution of the challenging dust and system operating conditions? How can the energy requirement for cleaning be minimized? What are the optimum layout and geometry of the filter vessel? Industrially, these issues are being investigated through the use of pilot plants and advanced power generation demonstration plants, e. g. the 71 Mw(th) PFBC demonstration plant at Wakamatsu, Japan and the 15 MW(e)hybrid demonstration plants in Wilsonville, USA. Additionally, controlled pilot-scale investigations have been conducted. However, the majority of industrial knowledge(for both power and non-power generation applications) is gained from plant operating experience rather than a prior fundamental understanding of the principles involved. At present, the most pressing issue in industrial applications concerns the mechanical strength and chemical resistance of the filter medium in relation to its operating environment
Gas cleaning at high temperatures 663 Figure 4. Research issues in the development of ceramic lters for hot gases [14]. Table 2. Academic issues in ceramic lter research [14] Fluid mechanics Particle mechanics Filtration ow eld distribution cake formation Cleaning pulse propagation cake detachment ² What is the effect of the challenging dust, in relation to the operating temperature, on the operating pressure drop? ² Is the lter medium suitable in relation to the particle size distribution of the challenging dust and system operating conditions? ² How can the energy requirement for cleaning be minimized? ² What are the optimum layout and geometry of the lter vessel? Industrially, these issues are being investigated through the use of pilot plants and advanced power generation demonstration plants, e.g. the 71 MW(th) PFBC demonstration plant at Wakamatsu, Japan and the 15 MW(e) ‘hybrid’ demonstration plants in Wilsonville, USA. Additionally, controlled pilot-scale investigations have been conducted. However, the majority of industrial knowledge (for both power and non-power generation applications) is gained from plant operating experience rather than a prior fundamental understanding of the principles involved. At present, the most pressing issue in industrial applications concerns the mechanical strength and chemical resistance of the lter medium in relation to its operating environment.
J. Seville et al For academic investigations, an important issue is the need to reproduce the operating environment of a filter candle so as to give results which are meaningful industrial users. However, reproducing the exact filter operating conditions is difficult and sometimes not the most cost-effective way to conduct the required investigation. For this reason, different techniques have been proposed by which to predict the conditioning behaviour of the filter candle, e.g. experiments performed on small flat filter pieces or 'coupons. Flat coupons are used since they have uniform fow conditions across the diameter (in both filtration and cleaning modes) and require low dust masses for the test work. Coupon filters can be used to predict conditioning behavior [16] in addition to exploring issues surrounding dust cake formation and detachment [17] The following sections outline the development of understanding of filter fluid mechanics and par 3. Fluid mechanics Most ceramic filter elements are long cylinders with one closed end and the flow passing through the wall from the outside inwards. Although this is a practical shape to manufacture, it has certain disadvantages in both the filtration and cleaning parts of the cycle of operation, which are associated with flow maldistribution. In filtration, the internal pressure on the axis falls in the direction from the closed end towards the open end (Fig. 5). Therefore, the pressure difference which exists locally across the wall near the open end can be rather more than the average for the candle as a whole, leading to local high gas velocities [18]. In an extreme case, the ratio between the local filtration velocities at the open and closed ends could be as high as a factor of 2. There is clearly less of a problem if the resistance to flow of the filtration medium is high, but for low-density fibrous media at high face velocities, the effect is enough to cause 'blinding near the neck of the candle. The problem Wall Pressur Drop, dP Figure 5. Flow maldistribution in a candle: filtration mode
664 J. Seville et al. For academic investigations, an important issue is the need to reproduce the operating environment of a lter candle so as to give results which are meaningful to industrial users. However, reproducing the exact lter operating conditions is difcult and sometimes not the most cost-effective way to conduct the required investigation. For this reason, different techniques have been proposed by which to predict the conditioning behaviour of the lter candle, e.g. experiments performed on small at lter pieces or ‘coupons’. Flat coupons are used since they have uniform ow conditions across the diameter (in both ltration and cleaning modes) and require low dust masses for the test work. Coupon lters can be used to predict conditioning behavior [16] in addition to exploring issues surrounding dust cake formation and detachment [17]. The following sections outline the development of understanding of lter uid mechanics and particle mechanics. 3.1. Fluid mechanics Most ceramic lter elements are long cylinders with one closed end and the ow passing through the wall from the outside inwards. Although this is a practical shape to manufacture, it has certain disadvantages in both the ltration and cleaning parts of the cycle of operation, which are associated with ow maldistribution. In ltration, the internal pressure on the axis falls in the direction from the closed end towards the open end (Fig. 5). Therefore, the pressure difference which exists locally across the wall near the open end can be rather more than the average for the candle as a whole, leading to local high gas velocities [18]. In an extreme case, the ratio between the local ltration velocities at the open and closed ends could be as high as a factor of 2. There is clearly less of a problem if the resistance to ow of the ltration medium is high, but for low-density brous media at high face velocities, the effect is enough to cause ‘blinding’ near the neck of the candle. The problem Figure 5. Flow maldistributionin a candle: ltration mode.
Gas cleaning at high ten Figure 6. Typical pressure signals in kPa obtained during reverse pulsing for different positions along a low density filter candle at 4 bar(g)reservoir pressure (221 is self-correcting, in that local high velocity regions will form thicker cakes, but the damage may already have been done In pulse cleaning, the internal pressure is higher near the closed end, because recove ry of the dynamic energy of the pulse gas. Here, the problem is that
Gas cleaning at high temperatures 665 Figure 6. Typical pressure signals in kPa obtained during reverse pulsing for different positions along a low density lter candle at 4 bar (g) reservoir pressure [22]. is self-correcting, in that local high velocity regions will form thicker cakes, but the damage may already have been done. In pulse cleaning, the internal pressure is higher near the closed end, because of recovery of the dynamic energy of the pulse gas. Here, the problem is that
666 J. Seville et al the pressure difference generated across the candle wall may be high enough for cake detachment near the closed end, but not high enough near the open end (Fig. 6). In addition, if the pulse tube is wrongly placed, secondary entrainment may occur through the candle wall near the open end, in the filtration direction, thus ensuring no cake detachment in that region. Properly designed venturis can reduce the requirement of high pressure pulse gas for a particular cleaning duty; n effect, they act as pumps for the pulse gas [19]. However, they can make little difference to the distribution of the cleaning pressure [20]. What is required is a way of modifying this distribution by better candle design. Chuah et al. [21] propose one such modification; Grannell [22] has investigated others. Maldistribution will obviously be more of a problem for candles with more extreme length-to-diameter ratios; Clift et al. [18] give some guidance on this a Pulse cleaning is more complex than filtration from the point of view of fluid chanics, partly because the fow velocities are often larger, but also because th components which give rise to the internal axial pressure drop are in opposition. This internal pressure drop occurs partially because of the momentum changes in the gas and partially because of wall frictional effects, as in conventional pipe flow In filtration, the momentum change In pulse cleaning, however, the two effects is the more significant factor [18, but both contributions act in the same direction act in the opposite sense, so that a minimum can occur in the internal pressure if high reverse flows are reached [15]. It is reassuring to note that all of these effects are predictable using both simple one-dimensional models and more sophisticated computational fluid dynamics(CFD)codes [21] Clearly, if resistance to flow of the medium is not uniform along the candle length, this can also lead to flow maldistribution. Little has been published in this area on granular candles, but low-density candles can show considerable changes in resistance along their length and in opposite senses, according to whether they ar made with the former on the outside or the inside. If the variation is in the right ense,it can help to counteract the maldistribution problems discussed above [1 3. 2. Cake detachmen Particle mechanics as a subject is at a much earlier stage of development than fluid mechanics(see, e.g. [23). Although there are in principle some possible approaches to predicting the failure stress for a compact of identical spherical particles, even this problem has not been resolved satisfactorily [24] and in practice it is necessary to fall back on experiment. The basic validity of the coupon test has been shown by Koch et al. [17] who compared the stresses necessary to remove filter cakes by reverse flow and acceleration in a centrifuge, with comparable results. Figure 7 shows how the results of such a determination might be used in practice On the left-hand side is a set of cake detachment curves; on the right, an imaginary axial distribution of cleaning pressure. If the measured cake detachment stress curve is'a', then most of the cake will be removed by the pulse; if it is 'c', then very little will. This simplistic approach begs lots of questions, however. What level of
666 J. Seville et al. the pressure difference generated across the candle wall may be high enough for cake detachment near the closed end, but not high enough near the open end (Fig. 6). In addition, if the pulse tube is wrongly placed, secondary entrainment may occur through the candle wall near the open end, in the ltration direction, thus ensuring no cake detachment in that region. Properly designed venturis can reduce the requirement of high pressure pulse gas for a particular cleaning duty; in effect, they act as pumps for the pulse gas [19]. However, they can make little difference to the distribution of the cleaning pressure [20]. What is required is a way of modifying this distribution by better candle design. Chuah et al. [21] propose one such modication; Grannell [22] has investigated others. Maldistribution will obviously be more of a problem for candles with more extreme length-to-diameter ratios; Clift et al. [18] give some guidance on this. Pulse cleaning is more complex than ltration from the point of view of uid mechanics, partly because the ow velocities are often larger, but also because the components which give rise to the internal axial pressure drop are in opposition. This internal pressure drop occurs partially because of the momentum changes in the gas and partially because of wall frictional effects, as in conventional pipe ow. In ltration, the momentum change is the more signicant factor [18], but both contributions act in the same direction. In pulse cleaning, however, the two effects act in the opposite sense, so that a minimum can occur in the internal pressure if high reverse ows are reached [15]. It is reassuring to note that all of these effects are predictable using both simple one-dimensional models and more sophisticated computational uid dynamics (CFD) codes [21]. Clearly, if resistance to ow of the medium is not uniform along the candle length, this can also lead to ow maldistribution. Little has been published in this area on granular candles, but low-density candles can show considerable changes in resistance along their length and in opposite senses, according to whether they are made with the former on the outside or the inside. If the variation is in the right sense, it can help to counteract the maldistribution problems discussed above [15]. 3.2. Cake detachment Particle mechanics as a subject is at a much earlier stage of development than uid mechanics (see, e.g. [23]). Although there are in principle some possible approaches to predicting the failure stress for a compact of identical spherical particles, even this problem has not been resolved satisfactorily [24] and in practice it is necessary to fall back on experiment. The basic validity of the coupon test has been shown by Koch et al. [17] who compared the stresses necessary to remove lter cakes by reverse ow and acceleration in a centrifuge, with comparable results. Figure 7 shows how the results of such a determination might be used in practice. On the left-hand side is a set of cake detachment curves; on the right, an imaginary axial distribution of cleaning pressure. If the measured cake detachment stress curve is ‘a’, then most of the cake will be removed by the pulse; if it is ‘c’, then very little will. This simplistic approach begs lots of questions, however. What level of
