Document Detail

Influence of operating parameters on cake formation in pilot scale pulse-jet bag filter.
Jump to Full Text
MedLine Citation:
PMID:  24415802     Owner:  NLM     Status:  Publisher    
Abstract/OtherAbstract:
Bag filters are commonly used for fine particles removal in off-gas purification. There dust laden gas pervades through permeable filter media starting at a lower pressure drop limit leaving dust (called filter cake) on the filter media. The filter cakeformation is influenced by many factors including filtration velocity, dust concentration, pressure drop limits, and filter media resistance. Effect of the stated parameters is investigated experimentally in a pilot scale pulse-jet bag filter test facility where lime stone dust is separated from air at ambient conditions. Results reveal that filtration velocity significantly affects filter pressure drop as well as cake properties; cake density and specific cake resistance. Cake density is slightly affected by dust concentration. Specific resistance of filter cake increases with velocity, slightly affected by dust concentration, changes inversely with the upper pressure drop limit and decreases over a prolonged use (aging). Specific resistance of filter media is independent of upper pressure drop limit and increases linearly over a prolonged use.
Authors:
Mahmood Saleem; Gernot Krammer; Rafi Ullah Khan; M Suleman Tahir
Publication Detail:
Type:  JOURNAL ARTICLE    
Journal Detail:
Title:  Powder technology     Volume:  224     ISSN:  0032-5910     ISO Abbreviation:  Powder Technol     Publication Date:  2012 Jul 
Date Detail:
Created Date:  2014-1-13     Completed Date:  -     Revised Date:  -    
Medline Journal Info:
Nlm Unique ID:  9887892     Medline TA:  Powder Technol     Country:  -    
Other Details:
Languages:  ENG     Pagination:  28-35     Citation Subset:  -    
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Descriptor/Qualifier:

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine

Full Text
Journal Information
Journal ID (nlm-ta): Powder Technol
Journal ID (iso-abbrev): Powder Technol
ISSN: 0032-5910
ISSN: 1873-328X
Publisher: Elsevier Sequoia
Article Information
© 2012 Elsevier B.V.
License:
Received Day: 19 Month: 8 Year: 2011
Revision Received Day: 4 Month: 2 Year: 2012
Accepted Day: 11 Month: 2 Year: 2012
pmc-release publication date: Day: 1 Month: 7 Year: 2012
Print publication date: Month: 7 Year: 2012
Volume: 224 Issue: 1-2
First Page: 28 Last Page: 35
PubMed Id: 24415802
ID: 3886375
Publisher Id: S0032-5910(12)00089-7
DOI: 10.1016/j.powtec.2012.02.016

Influence of operating parameters on cake formation in pilot scale pulse-jet bag filter
Mahmood Saleemab Email: msaleem.icet@pu.edu.pk
Gernot Krammerb
Rafi Ullah Khana
M. Suleman Tahirb
aInstitute of Chemical Engineering and Technology, University of the Punjab, Quaid-i-Azam Campus, 54590- Lahore, Pakistan
bGraz University of Technology, Graz, Austria
Corresponding author. Tel.: + 92 42 99230065; fax: + 92 42 99231159. msaleem.icet@pu.edu.pk

Introduction

Bag filters are frequently employed for off gas purification [1]. Dust laden gas pervades through permeable filter media starting at a lower pressure drop leaving dust (filter cake) on the filter surface. The filter cake is a source of increasing pressure drop (ΔP). Therefore, it has to be removed either at an upper pressure drop limit (ΔPmax) or at a predefined time interval. At the ΔPmax, short duration pulses of high pressure air in reverse direction remove the cake. At the lower pressure drop limit (ΔPmin), the pulses stop and a next filtration cycle starts. A fraction of total filtration area is regenerated normally at the end of a filtration cycle leading to non-uniform distribution of filter cake. Non uniform distribution of cake area load (mass per unit filter area) possesses a number of consequences like non-uniform gas flow distribution, modified cake properties (specific resistance, and porosity), which may affect cake formation and detachment. The ΔP is dependent on properties of filter cake represented in terms of specific resistance, distribution of cake area load, filtration velocity, properties of the dust and gas. Cake properties (density, specific resistance) depend on many factors [1–5].

Usually the overall pressure drop, ΔP is taken as sum of pressure drop across regenerated filter media (ΔPm) and pressure drop across the cake (ΔPc) [Eq. (1)].

[Formula ID: fo0005]
(1) 
ΔP=ΔPm+ΔPc

Since the filtration velocity (superficial velocity) is generally low where Reynolds number is small (Re < 1), Darcy's law can be applied to describe gas flow through the fibrous media and the dust cake to compute individual pressure drops [1].

[Formula ID: fo0010]
(2) 
ΔPm=1BmμLmu
and
[Formula ID: fo0015]
(3) 
ΔPc=1BcμLcu
where μ is gas viscosity, u is filtration velocity or superficial velocity, Lm and Lc are the thickness of filter media and the cake respectively, Bm and Bc are permeability of regenerated filter media and the cake respectively.

Overall pressure drop can be written as:

[Formula ID: fo0020]
(4) 
ΔP=1BmμLmu+1BcμLcu.

However, Bc depends on thickness and mass of filter cake, which can be calculated using mass balance at any time t:

[Formula ID: fo0025]
(5) 
Lcρs(1−ϵc)Af=cutSeAf=wAf
where c is dust concentration, t is filtration time, ρs is density of particles, Se is total separation efficiency, ϵc is porosity of cake, and w is cake area load (cake mass per unit area).

Combining Eqs. (4) and (5) one gets:

[Formula ID: fo0030]
(6) 
ΔP=μBmLmu+μBcwρc(1−ϵc)u
or simply:
[Formula ID: fo0035]
(7) 
ΔP=kmμu+kcμwu
where km=LMBm is specific resistance of filter media (m− 1) and kc=1Bcρs(1−ϵc) is specific resistance of the cake (m/kg).

Eq. (7) is the most commonly used for describing filter pressure drop [1]. It is clear that ΔP rises as filtration velocity increases at all other conditions being fixed only due to proportional increase of filter cake area load or cake height.

Resistance parameters (km and kc) are not constant as is often assumed but complicated functions made up of many other variables. Previous investigations present various correlations for estimating resistance parameters [1,6]. Specific resistance of filter media, km, depends on pulse pressure [7], reservoir pressure [8], decay of pressure pulse in the bag, and velocity [1,9].

Experiments with coal ash and char dust at high temperature with fabric filters made from ceramic fibers at dust concentration from 25,000 to 100,000 ppm (weight basis) and filtration velocity of 12–45 mm/s have shown a linear increase in residual ΔP with increasing velocity, and also an increase with increasing dust concentration [2]. Filtration velocity also affects the cake formation and detachment [10]. It is also observed that the cake porosity decreases and (kc) increases with increasing velocity [11]. Increasing filtration velocity increases the mean particle size of the cake due to an altered segregation behavior of the incoming dust. Also the cake/fabric adhesion force is increased at the higher filtration velocity. Filtration velocity, in addition to ΔP, also affect the cleaning performance of pulse-jet filters in a study with fly ash and limestone dust [12]. Higher velocity increases dust retention which results in significantly increased residual pressure drop [13]. The pressure drop at the same mean cake load is highest when the dust is uniformly distributed. Wider distribution of cake area load lead to a lower pressure drop and vice versa. The profiles of cake height on the filter surface are related to the pressure drop. Non uniform distribution of cake height is found besides higher cake thickness near the bottom of the bag [14].

Aging effect due to dust penetration into the body of the filter media over long time use has been studied by [15] on a large scale filter unit. The filter has been operated at 50 g/m3 dust concentration for 1000 h taking samples of dust at 250 h. A decrease of the fine fraction resulting in an increase in coarse fraction is reported.

This paper investigates the effect of operating parameters like ΔPmax, filtration velocity (u) as well as dust concentration (c) on residual cake height distribution and bag filter operation. The phenomena of aging are also researched experimentally.


Experimental
2.1  Bag filter set-up

The experimental set up [16] is briefly described here. A two screw single component feeder delivers a controlled mass (gravimetric control, variation 1% at steady state) of dust into the dispersion nozzle through a vibrating chute. Compressed, filtered and dried gas (air) meets the dust tangentially in the dispersion nozzle at controllable pressure at 2 bar. Ambient air is sucked in and mixed with dispersed dust (raw gas). The raw gas flows through a 100 mm diameter pipe into the bag filter near the bottom. The filter bags are arranged in three rows, each row containing one bag. The dust collector rests freely on a plate supported on a load cell. Both (the load cell and dust collector) are enclosed in the housing. The load cell is calibrated from 0 to 4000 g at ± 2 g. The gas pervades through the filter bags into the clean gas header and then to the discharge fan. The differential pressure regulator (3 ms time constant) monitors pressure drop across the filter. Cake detachment is accomplished on-line with jet-pulses. A pulse of compressed air enters the bag through an 8 mm diameter nozzle in the gas supply pipe (27 mm in diameter) tapping secondary gas from the clean gas side. The pulses are issued for the rows of bags in a cyclic order either at the upper pressure drop limit or at a pre-defined time interval. The reservoir pressure of the cleaning air can be set up to 4 bar, time of cleaning pulse can be set between 50 and 200 ms, and cleaning pulses interval between 2 and 450 s.

The gas flow rate is measured using an orifice plate. The ΔP across the orifice plate, absolute upstream pressure and temperature are recorded at 1 s resolution. The gas flow is calculated using Labview(R) software according to DIN EN-ISO 5167-1: 1995 (large fan) and ISO 5167:2003 (small fan) depending on the fan in use and recorded. The gas flow is calibrated using Pitot tube measurements. A frequency converter regulates the gas flow rate.

Dust fraction settling in the filter and dust concentration relevant for filtration are calculated since transient dust feed and dust collected in the filter are known. All tests are carried out at ambient conditions.

2.2  Optical stereo set-up

The facility incorporates an optical system (Fig. 1) for in-situ dust cake height measurement on the filter surface [17]. The measurement system is based on stereo vision. The development of the optical technique and its boundary conditions are discussed elsewhere [17].


2.2.1  Materials

The needle felt is made of Polyimide (PI) and Polyphenylensulfide (PPS) polymers [Table 1]. The diameter and length of each bag are 0.012 m and 1.8 m respectively. Available filter area per bag is 0.067824 m2.

Commercial grade non precipitated Limestone (CaCO3) (Commercially known as OMYACARB 5-GU) is used as dust. Some of the dust characteristics are; solid density (ρs) = 2700 kg/m3, median particle diameter (d50, 3) = 5 μm, and bulk density (ρb) = 1200 kg/m3. Air at ambient conditions is used as gas.

2.3  Procedures
2.3.1  A filtration cycle

Filtration process is semi-cyclic by nature. The cycle may be decomposed as filtration or cake formation followed by a jet-pulse cleaning (regeneration). In on-line cleaning, which is normally the case with jet pulse filters, the filtration step shall follow a jet pulse cleaning and a jet pulse cleaning shall follow a filtration. Thus the cycle time (tc) is comprised of a filtration time, (tf) and jet pulse cleaning time, (tcl). Mathematically one may write:

[Formula ID: fo0040]
(8) 
tc=tf+tcl.

The cleaning time is generally short as compared to filtration time and in most cases it may be negligible at all. Thus in many cases filtration time is taken as cycle time. If the cleaning time is higher, it indicates extra costs on cleaning and less overall filtration throughput.

Specific resistance of filter media and filter cake are determined from Eq. (7) taking resistance to flow of gas offered by the filter media and the filter cake in series and applying the Darcy's law [1] for flow of fluids through porous beds. The filtration velocity is time averaged volume flow per unit filtration area. The cloth resistance is calculated straightforwardly by assuming no cake on the bag (w = 0).

[Formula ID: fo0045]
(9) 
km=ΔPmu⋅μ.

The error linked to the value of km can be estimated by:

[Formula ID: fo0050]
(10) 
ekm=|kmu|eu+|kmΔPm|eΔPm

The specific cake resistance is calculated from the time derivative of Eq. (7) where the time derivative of ΔP is replaced by the slope of the regression fit of the pressure drop curve:

[Formula ID: fo0055]
(11) 
kc=AΔ(ΔP)Δtm˙⋅u

The error associated with the kc values is given by:

[Formula ID: fo0060]
(12) 
ekc=|kcΔ(ΔP)Δt|e(Δ(ΔP)Δt)+|kcu|eu+|kcm˙|em˙

2.3.2  Effect of velocity and dust concentration on cake formation

The tests started with conditioned and thoroughly pulsed bags. The reference measurement was made prior to dust injection. Dust feeder was switched on and filter cake was built. At the upper pressure drop level, dust supply was turned off and cake height was measured after the dust had settled and glass window was cleaned. Afterwards the bags were pulsed again. Residual cake heights were measured after the dust had been settled and glass window had cleaned. Dust feed was resumed and filtration continued for a number of cycles. The tests were repeated at different velocity and/or dust concentration (Sr. 1–5 in Table 2).

2.3.3  Effect of upper limit of pressure drop (ΔPmax)

Filter bags were repeatedly pulsed and filter media resistance was measured at steady gas flow (Sr. 6–9 in Table 2). Filter cake was built up. Dust was turned off and final cake height was optically measured. Then bags were pulsed. After filter housing became clear, residual cake height was measured. The bags were then repeatedly cleaned using jet pulses at 4 bar before starting the next test. Cake height distribution was measured in-situ on the regenerated bags.

2.3.4  Effect of aging on specific resistance of filter media and the cake

Because of long time period involved, study of this behavior in laboratory is difficult. However, since the clogging is most affected by the frequency of pulses, a filtration test was conducted where frequency of pulsing was increased by selecting narrow filtration intervals (one cleaning pulse to one row of bags every 120 s). Seven tests were conducted of 6–9 hours duration. Towards the end of every test, all bags were thoroughly pulsed, a cake was built, and all bags were repeatedly pulsed again. The specific resistances of filter media (km) and filter cake (kc) were computed from the slope of the pressure drop, filtration velocity, cake area load, and temperature of the gas using Eqs. (9) and (11) respectively.


Results and discussions

The microscopic image of the surface of the needle felt and histograms of surface roughness are shown in Fig. 2. Obviously the needle felt surface is not smooth on dust side. Black spots are the regions where surface reconstruction by the microscope was not possible. Permeability distribution exists naturally, therefore, non uniform distribution of gas flow is expected at the on-set of filtration cycle.

3.1  Effect of filtration velocity and dust concentration on cake formation

Filtration tests are performed at various combinations of velocity and dust concentrations. The test conditions are listed in Table. 2.

3.1.1  Pressure drop and filtration time

As expected, the slope of the transient ΔP curve (Fig. 3) is higher at higher dust concentration and/or higher filtration velocity resulting in a decrease of filtration cycle time at the same upper ΔP limit. Since cake area load is the product of velocity and dust concentration, both parameters are interlinked. It can be seen in Fig. 3 that ΔP curves are concave irrespective of low or high dust concentration and/or velocity. At the same velocity, but different dust concentration, the ΔP curves evolve in the same way. The slope of the curve, however, becomes higher at higher dust concentration shortly after the beginning of the filtration cycle. Under the premise of perfect filter cleaning and constant mean resistance (bag and cake) values, computed from the linear part of the ΔP curve, the filter ΔP is estimated. The computed cycle time will always be higher than the experimentally observed one because the computations predict ΔP using cake resistance parameter obtained from the linear part of the ΔP curve. Therefore, between the same lower and upper ΔP limits, a longer cycle than the experimentally observed is expected. If the non linear rise is considered, the mean cake resistance parameter would be higher and a shorter filtration cycle would be predicted. At higher velocity (34.1 mm/s) the residual ΔP is already higher and the slope of the transient ΔP curve is steeper naturally [9].

Since the overall ΔP is the sum of ΔP across filter media (at w = 0) and ΔP across the deposited cake, therefore, ΔP due to cake formation (ΔP − ΔPw = 0) is plotted in Fig. 4 versus cake area load. One can see that the two ΔP curves, at the same velocity (20.5 mm/s) but different dust concentrations, are very similar. The ΔP is significantly different at different filtration velocity and approximately same dust concentrations. Thus the ΔP is significantly affected by the filtration velocity but not by the dust concentration. A slightly higher ΔP at 20.5 mm/s and low dust concentration at the beginning is noticeable. The ΔP at 20.5 mm/s and high dust concentration is more fluctuating. All ΔP curves turn to a moderate and linear rise after approximately 25 g/m2 cake area load. It seems that a certain amount of dust deposit is necessary to equilibrate the differences in permeability of the filter medium. A particle–fiber layer forms before a true surface filtration starts on the surface treated needle felts [18].

It is obvious that the filtration cycle time between the same upper and lower limits of ΔP decreases at higher velocity and/or dust concentration because the ΔP depends on velocity and cake area load. Filtration cycle time at a filtration velocity of 20.5 mm/s decreases from 3388 s to 3060 s at low dust concentration (4.53 g/m3) and from 2012 s to 1560 s at higher dust concentration (7.3 g/m3) starting with thoroughly pulsed bags provided the upper ∆P limit remains the same (see Fig. 5). Reduction of filtration time is because of accumulation of residual cake on the bags with increasing filtration cycles. Moreover also the upper ΔP limit is slightly different (between 1233 Pa and 1323 Pa) which also explains the changing filtration time.

The cake resistance is obtained only from the linear section of the transient pressure drop profiles (see e.g. Fig. 3) leaving out the initial concave rise which is discussed in more detail elsewhere [5]. The effect of velocity is depicted in Fig. 6. As expected a shortening of filtration time (Fig. 6) at higher filtration velocity at nearly the same dust concentration is found. The filtration time is again decreasing from the start of filtration to the end. At 20.5 mm/s the filtration time is 3387 s which reduces to 3060 s at the end of 3rd cycle at 4.53 g/m3 dust concentration. At 34.1 mm/s filtration velocity, the filtration cycle time is only 674 s and decreases to 545 s in 6 cycles at 4.81 g/m3 dust concentration. A small difference in dust concentrations (4.53 and 4.81) is due to a different degree of dust settling in the filter.

3.1.2  Dust settling

Provisions for continuous measurement of dust streams entering and leaving (assuming 100% separation) allows the computation of transient settling of dust during filtration cycles. Dust settling is higher at lower velocity as compared to that at higher velocity. Transient cumulative mass of collected dust versus cumulative dust injected from the start of a filtration cycle is displayed in Fig. 7 at different velocities. One can observe that dust settling curves follow a similar trend up to 40 mm/s. However, the dust settling is slightly reduced at 45 mm/s. Rate of dust settling increases with filtration time non-linearly. A probable reason of such behavior is the agglomeration of dust particles inside the filter housing. The rate of dust settling gradually increases as the size and mass of agglomerates increases.

Fraction of dust which settles without forming cake per cycle is plotted in Fig. 8. Fraction of dust settling at low velocity and lower dust concentration is higher and vice versa. At 20.5 mm/s filtration velocity it is around 12% at 4.53 g/m3 and around 5% at 7.3 g/m3. At 34.1 mm/s it is between 1 and 4%. Longer cycles (owing to low velocity or dust concentration) have shown higher dust settling which is most probably due to agglomeration of dust particles. It has been observed that agglomerates grow on dead spots especially at the top plate inside the filter housing. They grow with time and start detaching as reaching a critical mass. They settle by gravity and partially reach the dust collector. Their length as long as one centimeter was observed which depends on how much time they have to grow and also the flow conditions inside the housing. They are almost removed on pulse-jet cleaning. These phenomena are observed with naked eye as well as captured by camera but has not been investigated in detail. It can be seen in Fig. 7 that dust settling increases with time during a cycle. Evidence of agglomeration is found in the performed tests (see Fig. 9).

3.1.3  Cake height and density

In-situ measured mean cake height is decreasing from first to the last filtration cycle at respective constant dust concentration in Fig. 5. The mean cake density (Fig. 10) is calculated using corrected dust concentrations, gas volume flow, filtration cycle time, mean cake height, and filter area. The density (kg/m3) is higher (785 compared to 735) at low dust concentration at the end of the first cycle. It increases from first to the last cycle in both cases. In fact the dust reaching the bags during a filtration cycle is gradually declining as reflected by decreasing filtration time with the number of cycles at constant filtration velocity and dust concentration. Obviously there must be a trend in the cake height measurements as is the case otherwise, therefore, data point 2 in data set corresponding to 20.5 mm/s and 7.32 g/m3 is erroneous and termed as outlier. The cake resistance is also higher at lower dust concentration and proportionately increases over cycles. At low dust concentration and constant filtration velocity, the filtration cycles are naturally longer. Probably more compact cakes are formed which lead to higher cake density and cake resistance. However, the effect of dust concentration on cake height and density is not pronounced.

3.1.4  Specific cake resistance

Specific cake resistance and mean cake density (Fig. 11) are higher at higher filtration velocity. Fig. 12 shows the correlation between the average cake resistance and the filtration velocity where this trend is confirmed. Moreover this graph also contains the results at different concentrations indicating that a higher dust concentration might have a reduction of the cake resistance as a consequence. However, this difference is not significant when looking at the error bars. Both, the specific cake resistance and the mean cake density, increase from start of filtration to the end again. At higher velocity, the increase in cake resistance seems to be leveled off after a few cycles. However the cake density is increasing. The reason may lie in the error in cake height measurement where the mean cake height is too low (0.15 mm) at very low dust loads (200 g/m2) and therefore percentage of error is higher.

3.2  Effect of ΔPmax on cake formation

Transient ΔP, filtration velocity (u) and dust concentration (c) are displayed in Fig. 13 for four filtration cycles each with higher ΔPmax. The ΔP curves show a concave rise gradually turning to linear. In the fourth test, sudden jumps in the ΔP curve are possible because of cake compaction at 1600 Pa, 2000 Pa, and 2200 Pa. A discontinuity in ΔP curve in the third test is due to disturbance in the gas flow. The concave rise at the beginning of cake formation is characteristic of PPS filter media and has been observed in all tests presented here and otherwise. It takes a short time to level off the permeability distribution after which the cake growth and ΔP rise becomes linear. Despite the cake compaction, slope of ΔP curve is not changed apparently.

During the experiments, it is observed that on regeneration at 1200 Pa, the residual patches are more in number, larger in size and thicker as compared to those at 800 Pa. A further improved cake detachment is observed at 1600 Pa and 2400 Pa, i.e. thicker cakes. Thus thicker cakes support cake detachment provided the detachment force can overcome the cohesive stress of the filter cake.

Mean specific resistance of pulse-jet cleaned filter medium and filter cake is displayed in Fig. 14. Specific resistance of filter media remains nearly constant for thoroughly pulsed filter media at constant flow. However, specific resistance of filter cake is decreasing with increasing ΔPmax at all other conditions being the same. The decrease from 800 Pa to 1200 Pa is nearly 12% but afterwards it is less. A small but gradual reduction is associated with modified cake formation at higher ΔPmax. As one observes cake compaction at higher ΔPmax, a higher cake resistance is expected too. However the mean kc value is smaller for thicker cake formed at otherwise similar conditions. One may think of a relatively porous cake formation as the cake height keeps on growing. But this is in contradiction to the observation where cake compaction is observed. A thicker cake due to higher pressure drop shall be compressed and therefore compact. On the other hand this decrease of kc might be considered due to reduced % error of cake area load estimation. An explanation can be given that at 800 Pa the cycle is relatively short and less cake area load on the bags. Percent error for small cake area load is high. However, percent error in cake area load estimation at higher cake loads (longer filtration cycles) is lower leading to proportional decrease in the value of kc. The fairly close values computed at 1200, 1600 and 2400 Pa are in favor of this argument.

3.3  Effect of aging on specific resistance of filter medium and filter cake

The calculated specific resistance of filter medium and the filter cake over nearly 1500 cycles is displayed in Fig. 15. It can be seen that the specific resistance of filter medium is increasing linearly and the specific resistance of filer cake is decreasing with the filtration cycles. The decrease is, however, in the range of scatter in the estimated values of kc.


Conclusions

The effect of operating parameters like ΔPmax, filtration velocity, dust concentration on bag filter operation is investigated in a pilot scale test facility. The phenomena of aging are also researched experimentally with lime stone dust and air at ambient conditions.

The filtration velocity has a pronounced effect on ΔP as well as on cake properties, cake density and specific cake resistance. Cake density and specific resistance increase with increasing filtration velocity at constant dust concentration. Cake density is also affected by dust concentration. A denser cake evolves at low dust concentration. Besides the obviously shorter filtration times at a higher filtration velocity but constant dust concentration, a denser cake evolves and a higher cake resistance parameter is determined at higher gas velocities. However, the nature of this experimental evidence cannot be explained yet.

Measurements with the settling dust under gravity reveal that the rate of dust settling is a function of filtration time which is perhaps linked to agglomeration in the filter housing.

Data analyses reveal that ΔPmax will influence the cake formation at higher levels. At higher value cake compaction is observed. Longer cycles, owing to higher ΔPmax, promote agglomeration which causes enhanced dust settling. Dust settling decreases at higher velocity and vice versa. Specific resistance of the filter media is independent of ΔPmax. Specific resistance of filter cake is higher at lower pressure drop and vice versa.

Specific resistance of filter media is increasing linearly while the specific resistance of filer cake is decreasing on aging.


References
1. Löffler F.,Dietrich H.,Flatt W.. Dust Collection with bag and Envelop FiltersYear: 1988Fried Vieweg and SonsBraunschweig, Germany
2. Ergudenler A.,Tang W.,Clive B.M.H.,Lim C.J.,Grace J.R.,Gennrich T.J.. Performance of high-temperature fabric filters under gasification and combustion conditionsSeparation and Purification Technology111Year: 1997116
3. Saleem M.,Krammer G.,Rüther M.,Bischof H.. Optical measurement of cake thickness distribution and cake detachment on patchily cleaned commercial bag filters (Proceedings of the International Conference and Exhibition for Filtration and separation Technology).Vol. IIYear: 2005FILTECH Exhibitions GermanyWiesbaden, Germany5158
4. Kavouras A.,Krammer G.. Deriving cake detachment versus cake area load in a jet pulsed filter by a mechanistic modelPowder Technology133Year: 2003134146
5. Koch M.,Krammer G.. The permeability distribution (pd) method for filter media characterizationAerosol Science and Technology426Year: 2008433444 arXiv:. http://www.tandfonline.com/doi/pdf/10.1080/02786820802172053 doi:10.1080/02786820802172053. URL http://www.tandfonline.com/doi/abs/10.1080/02786820802172053
6. H. Risnes, High temperature filtration in biomass combustion and gasification processes, Ph.D. thesis, NTNU Trondheim (2002).
7. Strangert S.. Predicting performance of bag filtersFiltration and Separation January-FebruaryYear: 19784248
8. Dennis R.,Klemm H.A.. Modeling concepts for pulse jet filtrationJournal of the air pollution control association30Year: 19803848
9. Saleem M.,Krammer G.. Effect of filtration velocity and dust concentration on cake formation and filter operation in a pilot scale jet pulsed bag filterJournal of Hazardous Materials1443Year: 200767768117412496
10. Cheng Y.H.,Tsai C.J.. Factors influencing pressure drop through a dust cake during filtrationAerosol Science and Technology29Year: 1998315328
11. Silva C.R.N.,Negrini V.S.,Aguiar M.L.,Coury J.R.. Influence of gas velocity on cake formation and detachmentPowder Technology1012Year: 1999165172
12. Tsai C.J.,Tsai M.L.,Lu H.C.. Effect of filtration velocity and filtration pressure drop on the bag cleaning performance of a pulse-jet baghouseSeparation Science and Technology352Year: 2000211226
13. Ellenbecker M.,Leith D.. The effect of dust retention on pressure drop in a high velocity pulse-jet fabric filterPowder Technology252Year: 1980147154
14. Klingel R.,Löffler F.. Dust collection and cleaning efficiency of a pulse jet fabric filterFiltration and SeparationYear: 1983205208
15. Ehrenberg C.,Leibinger H.,Höflinger W.. The change of the particle size distribution during a test on a large scale aging test equipment for surface dust filtersProceedings of Filtech EuropaYear: 2003DüsseldorfGermany
16. Saleem M.,Krammer G.. Optical in-situ measurement of filter cake height during bag filter plant operationPowder Technology173Year: 200793106
17. Rüther M.,Saleem M.,Bischof H.,Krammer G.. In-situ measurement of dust deposition on bag filters using stereo vision and non-rigid registrationAssembly Automation253Year: 2005196203
18. Höflinger M.G.,Koschutnig W.W.. Cleaning behaviour of textile filter mediaIn Proceedings of European Conference on Filtration and SeparationYear: 2002GotenburgGermany

Acknowledgments

The authors acknowledge the project funding by Austrian Science Foundation (FWF) under project P 16313-No. 7, financial support of Higher Education Commission, Islamabad, Pakistan, supply of needle felts by M/S Inspec Fibers (Lenzing, Austria), and M/S Alicona Imaging (Graz, Austria) for microscopic analysis of needle felts.


Article Categories:
  • Article

Keywords: Keywords Off-gas purification, Pulse-jet bag filter, Operating parameters, Cake properties, Needle felt.

Previous Document:  Experimental study of cake formation on heat treated and membrane coated needle felts in a pilot sca...
Next Document:  The effect of operating conditions on resistance parameters of filter media and limestone dust cake ...