applied

sciences
Article
Effect of Venturi Structures on the Cleaning
Performance of a Pulse Jet Baghouse
Xingcheng Liu * and Henggen Shen
School of Environmental Science and Engineering, Donghua University, Shanghai 201620, China
* Correspondence: 1159197@mail.dhu.edu.cn; Tel.: +86-021-6779-2513




Received: 8 August 2019; Accepted: 3 September 2019; Published: 5 September 2019


Featured Application: The experimental data and opinions of this study are expected to provide
guidance for the application of venturi tube in the dust cleaning of the pulse jet baghouse.

Abstract: In order to study the effect of venturi structures on the cleaning performance of a pulse
jet baghouse, several cleaning tests are carried out applying the pulse jet test bench. Experiments
are performed using different filter bags with venturi tubes of varying throat diameters and lengths.
Peak pressure and peak pressure arrival time are selected to evaluate the cleaning efficiency under
different conditions. On this basis, a series of comparative experiments are conducted between a
typical venturi-type and a new venturi tube design on a semi-industrial pulse jet baghouse. The effects
of the new venturi-type on baghouse cleaning performance and dust emission are studied. The results
show that pulsing air pressure has a significant effect on bag-cleaning performance; the larger the
throat diameter and the shorter the venturi length, the higher the peak pressure and the shorter the
peak pressure arrival time. This improves the cleaning efficiency of the filter bags. The new venturi
design changes the distribution of the cleaning pressure in the vicinity of the bag mouth, effectively
extending the cleaning cycle and reducing the energy consumption as well as reducing the dust
emission concentration of the baghouse.

Keywords: baghouse; cleaning performance; venturi; dust emission

1. Introduction
The baghouse is a high-efficiency dust removal device widely used in the treatment of pollutants
present in industrial flue gases. In these types of operations, particulate matter in the dust-laden flue
gas is filtered and trapped by the filter bags, and a dust cake is deposited on the surface of the filter
bag. With an increasing dust cake thickness, the operating resistance of the baghouse continuously
increases [1,2]. Therefore, it is necessary to clean the dust cake on a regular basis, and the pulse jet
cleaning is the method most commonly used for this purpose [3,4]. Pulse jet cleaning usually uses
compressed air, which is injected into the filter bag through the bag mouth with an instant release
mode. At the same time, the compressed air entrains a large amount of secondary airflow near the
bag mouth to enter the filter bag. The dust cake falls off from the surface of the filter bag under the
combined action of primary compressed air and secondary airflow, thus achieving the purpose of the
bag cleaning [5–7].
Effective cleaning of filter bags is of great significance for a stable baghouse operation. Such
processes maintain reasonable filtration resistance and reduce operational energy consumption. Among
them, the installation of venturi tubes in the filter bags improves the efficiency of filter bag cleaning.
Some scholars have studied the effect of venturi on the cleaning performance of the baghouse. Morris [8]
determined that the installation of venturi can reduce the energy consumption of the pulse jet baghouse
during a stable operation by about 30% on average. Grannell et al. [9] considered that the venturi

Appl. Sci. 2019, 9, 3687; doi:10.3390/app9183687 www.mdpi.com/journal/applsci

Appl. Sci. 2019, 9, 3687 2 of 16

tube should be preferred when the dust cake is too viscous to be easily cleaned by the compressed
air. Cursley et al. [10] demonstrated that bag cleaning performance was dissimilar when five different
shapes of venturi tubes were used on a cleaning test. In addition, in the absence of the venturi tube,
a higher pulsing air pressure should be used in order to achieve the same cleaning effect. Lu et al. [11,12]
conducted an experimental study on the cleaning performance of filter bags in the presence and absence
of venturi. The results showed that when the filter bag displayed a high resistance coefficient, a proper
cleaning effect was only achieved in the presence of the venturi tube. Tsai et al. [13] determined that
when the residual resistance of the filtration system was high, the use of venturi produced an elevated
pulse pressure on the inner wall of the filter bag, improving the cleaning effect. Bi et al. [14] found
out that when the same cleaning effect was achieved, the consumption of compressed air required
by the filter bag with the venturi tube was about 40% lower than that without the venturi tube.
Andersen et al. [15] pointed out the importance of optimizing venturi tube design in low-pressure
filters by computational fluid dynamics (CFD) numerical simulation, and considered that the working
principle of the venturi was to restrict backflow, which was detrimental to the pressure rise in the bags.
Previous research has reported that the cleaning performance of filter bags was affected by
the presence of the venturi tubes as well as by their shapes. This, in turn, affected the cleaning
baghouse performance. However, studies on the effect of venturi structural parameters on the cleaning
performance of the filter bag, and of the venturi tube type on dust cleaning performance and the dust
emission concentration of the baghouse, are limited. This paper reports on the effect of different lengths
and throat diameters of venturi tubes on the cleaning performance of long filter bags (8000 mm). Given
that filter bags with typical venturi tubes display poor cleaning efficiency in the bag mouth area, a new
venturi type is designed. This new tube is tested in order to determine the cleaning performance and
dust concentration in the baghouse emissions.

2. Materials and Methods

2.1. Theoretical Background

Former study [16] has shown that under the action of the cleaning airflow, when the dust cake
and the filter bag are moving radially outward and separated due to the action of the filter bag tension,
the separation force per unit area is given by Equation (1), as follows:
 2 1/3
mc 1  Ppk D 
Fd = 1.65 5/6 K 2   , (1)
m tR 

where Fd is the separation force per unit area when the dust cake is separated from the filter bag, N; mc
is the mass of the dust cake present on the filter bag surface per unit area, kg; m is the sum of the filter
material and dust cake masses per unit area, kg; K is the elastic coefficient of the filter bag; Ppk is the
peak pressure of the cleaning airflow on the side wall of the filter bag, Pa; D is the free displacement of
the filter material and dust cake, m; and tR is the peak pressure arrival time, s.
According to Equation (1), the main factors affecting the cleaning effect (separation force Fd ) are
the peak pressure Ppk and the peak pressure arrival time tR when the filter material is selected and the
filtration process is in equilibrium. This means that, the greater the peak pressure and the shorter the
peak pressure arrival time, the greater the separation force of the dust on the filter bag and the better
the cleaning effect [17–20]. In this paper, Ppk and tR (the definitions of peak pressure and peak pressure
arrival time shown in Figure 1) were selected to characterize the separation performance of the filter
bag and the dust cake (the cleaning performance).
Appl. Sci. 2019, 9, x 3 of 16
Appl. Sci. 2019, 9, 3687 3 of 16
Appl. Sci. 2019, 9, x 3 of 16

Figure 1. Schematic diagram of peak pressure and peak pressure arrival time at a measuring point.
Figure
Figure1.
1.Schematic
Schematicdiagram
diagramof
ofpeak
peakpressure
pressureand
and peak
peak pressure
pressure arrival
arrival time
time at
at aa measuring
measuring point.
point.
2.2. Experimental Materials
2.2. Experimental
2.2. ExperimentalMaterials
Materials
Figure 2 shows the two types of venturi tubes that were tested in this research as well as their
Figure
arrangement 2 shows
Figure 2 showsbag
in the thecage.
the two types
two types
Figure of venturi
of venturi
2a–c shows tubesthe
tubes thatbag
that were
were tested
cage
tested in
in in this research
which
this research as well
the typical
as well as their
venturi
as their
was
arrangement
installed, in
and Figure
arrangement the bag
in the 2d–f cage. Figure
corresponds
bag cage. 2a–c
Figure 2a–c shows
to theshowsthe bag
bag the
cagecage
bag in which
adapted
cage inwiththe
which typical
thethenewventuri was installed,
venturi-type
typical venturi was tube.
and
Taking Figure 2d–f
the typical
installed, and Figurecorresponds
venturi to the
2d–fascorresponds bag
reference, the cage adapted
to position
the bag cage with
of the the new
new venturi
adapted venturi-type
with the in the
newbag tube. Taking
cage was tube.
venturi-type the
moved
typical
upward.
Taking Withventuri as reference,
this modification,
the typical the position
the air the
venturi as reference, of
inlet the new
in the of
position venturi
new in
the venturi the
new venturi bag
was in cage
located was
the bag moved
above upward.
the filter
cage was moved bag
With
upward. this modification,
With this the air
modification, inlet in
the the
air new
inlet venturi
in the was
new located
venturi above
was
mouth plane. The inlet and outlet diameters of both types of venturi tubes were of 120 mm and 80 the
located filter bag
above mouth
the plane.
filter bag
The inlet
mouth and outlet
plane. TheThe diameters
inlet of both
and outlet types ofofventuri
diameters tubesof were of 120 mm were
and 80 mm, respectively.
mm, respectively. throat diameter (𝑑 ) and bothlengthtypes
(h) of venturi tubes
the typical venturioftube
120 mm
wereandthe80 test
The throat diameter (d ) and length (h) of the typical venturi tube were the
mm, respectively. The ethroat diameter (𝑑 ) and length (h) of the typical venturi tube were the test test variables.
variables.
variables.

(a)
(a) (b)
(b) (c)
(c)

(d)
(d) (e)
(e) (f)(f)
Figure2. 2.
Figure
Figure 2. The
The The two
twotwo types
types
types ofof
of venturi tubes
venturitubes
venturi and
tubesand their
andtheir arrangement
arrangement in
theirarrangement inthe
in thebag
the bagcage
bag cageand
cage andfilter
and filter bag.
bag.
filter (a)
bag. Bag
(a)(a)
Bag
Bag
cage
cagecage with
withwith the
thethe typical
typical
typical venturi. (b)
venturi.(b)
venturi. Structural
(b)Structural parameters
Structuralparameters
parameters ofof the
of the typical
the typical venturi.
typical venturi. (c)
venturi.(c) Typical
(c)Typical venturi
Typicalventuri tube
venturitube
tube
installed
installed in in
installed aafilter
ainfilter
filter bag.
bag.bag. (d)
(d) Comparison
(d) Comparison
Comparison ofbag
of bag
of bagcages adapted
cages
cages adapted
adapted with the
with
with two
thethe types
two
two ofofventuri
types
types of tubes.
venturi
venturi (e)(e)
tubes.
tubes.
The
(e) newnew
The venturi. (f) The
venturi. (f) newnew
The venturi tubetube
venturi installed in ain
installed filter
a bag.bag.
filter
The new venturi. (f) The new venturi tube installed in a filter bag.
Appl. Sci. 2019, 9, 3687 4 of 16

2.3. Experimental Design

In this study, we first determined the effects of the pulsing air pressure (Ppu ) and the pulsing air
injection time (tP ) on the filter bag cleaning performance. Based on these results, the experimental
conditions for testing the effect of the venturi structure on the cleaning performance of the filter bag
were selected. For this purpose, the influence of the throat diameter and length on the cleaning
performance of filter bag was determined for both the typical and the new venturi tubes. Experimental
results were then compared. Finally, a semi-industrial experiment was carried out in order to compare
the effects of the two venturi tubes used in the present research on the cleaning performance and dust
emission concentration of the baghouse. Experimental design and arrangement are shown in Table 1.

Table 1. Experimental design and arrangement.

The Throat
The Pulsing Air The Pulsing Air The Length of
Diameter of
Pressure. Injection Time. The Type of Venturi.
Test Device Test No. Venturi.
Ppu tP Venturi. h
de
(MPa) (ms) (mm)
(mm)
1 0.3, 0.4, 0.5 100 Typical 40 200
Experimental bench for the 2 0.4 100, 150, 200 Typical 40 200
performance of pulse jet cleaning 3 0.4 100 Typical 40, 50, 60 200
(see Figure 3) 4 0.4 100 Typical 60 160, 180, 200
5 0.4 100 New 60 160
Semi-industrial baghouse 6 0.4 100 Typical 60 160
experimental device (see Figure 4) 7 0.4 100 New 60 160

2.4. Experimental Apparatus

Figure 3 shows the experimental bench approach for determining the performance of the pulse
jet cleaning. This consists of a compressed air supply system, a pulse jet system, a filter bag and bag
cage installed with the venturi tube (Figure 2), and a data acquisition system. The compressed air was
supplied by an air compressor (Model AX2025; PUMA Inc., Memphis, TN, USA); the air tank was
used to store the compressed air and maintain a constant the pulsing airflow pressure, which was
monitored by a pressure gauge (Model Y-60A; Shanghai Automation Instrument Co., Ltd., Shanghai,
China). The pulse controller (Model DMK-3CS; SBFEC. Inc., Shanghai, China) was used to control
the opening and closing time of the electromagnetic pulse valve (Model DMF-Y-102S; SBFEC. Inc.,
Shanghai, China), thus adjusting the blowing time of the compressed pulsing air (the pulsing air
injection time). The distance between the nozzle and the bag mouth was 170 mm, and the diameter of
the nozzle hole was 18 mm. The tested filter bag was made of polyester with a thickness of 1.8 mm,
a diameter of 130 mm and a length of 8000 mm. The bag cage diameter was 125 mm, and the length
7950 mm. The position distribution of the measuring points on the filter bag is shown in Figure 3b.
The measuring points located 1000, 4000, and 8000 mm from the mouth of the bag were named
“top”, “middle”, and “bottom”, respectively. The pressure transmitters (Model 316B2ER; Shanghai
Automation Instrument Co., Ltd., Shanghai, China, measurement accuracy: 0.1Pa) were installed at
the filter bag measuring points for collecting the cleaning pressure data. An oscilloscope (Model DPO
2014; TEKTRONIX, INC., Beaverton, OR, USA, sampling accuracy: 1 ns) and computer were used
to record the duration of the cleaning pressure. In this paper, the effect of the venturi structure on
the cleaning performance of the filter bag was studied by collecting the peak pressure and the peak
pressure arrival time at different points on the test filter bags.
Appl.Sci.
Appl.
Appl. Sci.2019,
Sci. 2019, 9,xx3687
2019,9,
9, 555of
of16
of 16
16

(a)
(a) (b)
(b)
Figure 3.
Figure 3.
Figure Experimental
Experimental
3. Experimental bench
bench
bench approach
approach
approach forfor
for determining
determining
determining the the
the pulse jet cleaning
pulse
pulse jet cleaning performance.
jet cleaning (a)
performance.
performance. (a)
Schematic
(a) Schematic
Schematic diagram
diagram of the
diagram
of the pulse
of pulse
the jet cleaning
pulse
jet cleaning experimental
jet cleaning device
experimental
experimental device and and
device
and (b) the
(b) the distribution
(b) distribution of the
the distribution
of the
of
measuring
measuring points
the measuring
points on the
points
on the filter
onfilter bag.bag.
the filter
bag. P0–P8
P0–P8 are the
P0–P8
are the numbers
arenumbers of the
the numbers
of the measuring
ofmeasuring
the measuringpoints.
points.
points.

Figure4.
Figure
Figure Semi-industrialbaghouse
4. Semi-industrial
Semi-industrial baghouseexperimental
baghouse experimentaldevice.
experimental device.
device.

Based on
Based on the
theresults
resultsrelated
related to the
to theinfluence
the influence of theofventuri tube structure
the venturi
venturi on the filter
tube structure
structure on thebagfilter
the cleaning
bag
Based on the results related to influence of the tube on filter bag
performance, this
cleaning performance, research
performance, this group
this research determined
research groupgroup determined the effects
determined the of two
the effects types
effects of of two of venturi
two types
types of on the
of venturi
venturi oncleaning
on the
the
cleaning
performance
cleaning and dustand
performance emission
dust concentration
emission of the baghouse
concentration of the in a semi-industrial
baghouse in a scenario.scenario.
semi-industrial Figure 4
cleaning performance and dust emission concentration of the baghouse in a semi-industrial scenario.
shows a4 semi-industrial
Figure baghouse experimental
shows aa semi-industrial
semi-industrial baghouse experimental device, which
experimental can simulate
device, which can the working
can simulatecharacteristics
the working
working
Figure 4 shows baghouse device, which simulate the
of a baghouse
characteristics ofunder real
of aa baghouse filtering
baghouse under conditions.
under real The
real filtering semi-industrial
filtering conditions.
conditions. The baghouse was
The semi-industrial
semi-industrialdesigned with a dust
baghouse was
characteristics baghouse was
feeder, a pulse
designed with jet
a cleaning
dust feeder, system,
a pulse a filtering
jet cleaningunit,system,
a powerasystem,
filtering and a test
unit, a system.
power The dust
system, and feeder
a test
designed with a dust feeder, a pulse jet cleaning system, a filtering unit, a power system, and a test
was able
system. Theto control
The dust
dust feederthe
feeder wasdust
was ableconcentration
able toto control
control the in the
the dust dust-laden
dust concentration airflow
concentration in of
in the the baghouse
the dust-laden
dust-laden airflowby adjusting
airflow of of the
the
system.
the amount
baghouse by of dust generated.
adjusting the amount The
of air volume
dust generated. andThe theairfiltration
volume velocity
and the in the baghouse
filtration velocity were
in the
baghouse by adjusting the amount of dust generated. The air volume and the filtration velocity in the
regulated were
baghouse with aregulated
fan and awith flow meter in athe power system. The baghouse
system. operating resistance was
baghouse were regulated with aa fanfan and
and flow
a flow meter
meter in the
in the power
power system. The baghouse
The baghouse operating
operating
monitored with
resistance was a pressure
was monitored
monitored with transducer,
with aa pressure and the
pressure transducer, dust emission
transducer, and was
and the measured
the dust
dust emission in 6
emission was s intervals
was measured using
measured in a dust
in 66 ss
resistance
spectrometer
intervals using (Model
a dust 1.108; Grimm Aerosol
spectrometer (Model Technik
1.108; GmbHAerosol
Grimm & Co., Ainring,
Technik Germany).
GmbH & TheAinring,
Co., filtering
intervals using a dust spectrometer (Model 1.108; Grimm Aerosol Technik GmbH & Co., Ainring,
unit was composed
Germany). The filtering of six
filtering unitpolyester filter bags
was composed
composed with
of six a diameter
six polyester
polyester filterof bags
130 mm withand length ofof1500 mm.
Germany). The unit was of filter bags with aa diameter
diameter 130 mm
of 130 mm
3
The experimental
and length
length ofof 1500 conditions
1500 mm.mm. The were
The experimentalas follows:
experimental conditions the dust
conditions were concentration
were as as follows: was
follows: the maintained
the dust at
dust concentration 100
concentration was g/m
was;
and
the dust for at
maintained
maintained the
at 100testg/m
100 was33;;made
g/m the dust
the of alumina
dust for the
for theparticles waswith
test was
test made
made a weight-mean
of alumina
of diameterwith
alumina particles
particles of 4 µm;
with the filtration
aa weight-mean
weight-mean
velocity inof
diameter
diameter ofthe baghouse
44 μm;
μm; the was set to
the filtration
filtration 2 m/min,
velocity
velocity and
in the
in the operating
the baghouse
baghouse waspressure
was set to
set to 22 drop
m/min,
m/min, of theand
and dust
thecleaning
the operating
operating to
pressure drop
pressure drop of
of the
the dust
dust cleaning
cleaning to
to 1000
1000 Pa;
Pa; and
and particulate
particulate matter
matter in
in the
the dust-laden
dust-laden airflow
airflow was
was
Appl. Sci. 2019, 9, 3687 6 of 16

1000 Pa; and particulate matter in the dust-laden airflow was captured by the filter bag and deposited
on its surface. This increased the operating resistance in the baghouse. When the filtration resistance
measured by the pressure transducer reached the preset value of 1000 Pa, the system automatically
released the compressed pulsing air to clean the filter bag, which continued filtering the dust-laden
airflow. Once the baghouse filtration resistance reached again 1000 Pa, the system started performing
the next cleaning step. The time between any two cleaning intervals of the system was called a cleaning
cycle. When testing the cleaning performance of the baghouse with the two types of venturi tubes,
the experiments were continuously operated for 24 h under the same filtration-cleaning cycle conditions.
After system stabilization, the number of cleaning cycles (N) and the dust emission concentrations Ci
(i = 1,2,3 . . . 600) of the baghouse were continuously recorded for 1 h, and the average filtration cycle
time T and average dust emission concentration C of the baghouse were calculated according to the
Equations (2) and (3), respectively:
3600
T= (2)
N
P600
Ci
C = i=1 (3)
600

3. Results and Discussion

3.1. Effect of Pulsing Air Pressure and Pulsing Air Injection Time on the Cleaning Performance of the Filter Bag
Figure 5 shows the peak pressure distribution of the airflow in the filter bag under different
pulsing air pressure values (0.3, 0.4, and 0.5 MPa) at a pulsing air injection time of 100 ms. According to
the data, the peak pressure (PPk ) on the side wall of the filter bag increased with increasing pulsing air
pressure values (PPu ) and the peak pressure decreased from the top (1000 mm) to the bottom (8000 mm)
in the bag. In addition, the attenuation range of the peak pressure in the upper half of the filter bag was
significantly greater than that observed in the lower half. This occured because when the compressed
airflow was ejected from the nozzle, it would move downward rapidly and induce a large amount
of secondary airflow to enter the filter bag. The compressed airflow (the primary airflow) and the
secondary airflow mixed to form the total cleaning airflow. The cleaning airflow expanded as it moved
downward, thus increasing the air pressure in the space it passed through. Moreover, due to the air
permeability of the filter bag, the cleaning airflow was continuously diffused to the outside of the filter
bag. Furthermore, the cleaning airflow would rub against the filter bag in the downward movement
process, which inevitably consumed a certain amount of energy. Therefore, the peak pressure on the
filter bag decreased with the increase of the distance between the measuring points and the bag mouth.
On the other hand, the greater the pulsing air pressure, the greater the kinetic energy of the cleaning
airflow, the more the amount of the cleaning airflow diffused outside the filter bag, and the more
intense the friction with the filter bag, so the peak pressure differed greatly on the upper part of the
filter bag under different pulsing air pressure values. When the cleaning airflow moved to the lower
half of the filter bag, the friction between the cleaning airflow and the filter bag was weakened due to
the energy attenuation of the cleaning airflow so that the gap of the peak pressure on the lower part of
the filter bag became smaller under the different pulsing air pressure values. Further analysis showed
that, at a pulsing air pressure (Ppu ) of 0.5 Mpa, the peak pressure of the filter bag at the three measuring
points was 53.57% (top), 51.04% (middle), and 113.33% (bottom) higher than that obtained with a
pulsing air pressure of 0.3 MPa, respectively. A similar analysis comparing a pulsing air pressure of
0.4 MPa and 0.3 MPa indicated that the values were of 28.57% (top), 30.21% (middle), and 83.33%
(bottom), correspondingly. However, when the pulsing air pressure was 0.4 MPa, the peak pressure
at the three measuring points was only 16.28% (top), 13.79% (middle), and 14.06% (bottom) lower
than that obtained at a pulsing air pressure of 0.5 MPa. This indicated that as the pulsing air pressure
increased, the rate of increase in the peak pressure of the filter bag decreased.
 Appl. Sci. 2019, 9, 3687 7 of 16
Appl. Sci. 2019, 9, x 7 of 16

Figure
Figure Effectofofthe
5.5.Effect thepulsing
pulsingair
air pressure
pressure on
on peak
peakpressure
pressureofoffilter
filterbag.
bag.

Figure 6 shows that the peak pressure arrival time (t ) decreased with an increase in the pulsing air
Figure 6 shows that the peak pressure arrival time (R 𝑡𝑅 ) decreased with an increase in the pulsing
pressure (PPu ) at the three measuring points. In addition, at the “top” point, the peak pressure arrival
air pressure ( 𝑃𝑃𝑢 ) at the three measuring points. In addition, at the “top” point, the peak pressure
time values obtained under the three different pulsing air pressures were almost the same. Moreover,
arrival time values obtained under the three different pulsing air pressures were almost the same.
with an increasing distance between the measuring points and the bag mouth, the difference between
Moreover, with an increasing distance between the measuring points and the bag mouth, the
the peak pressure arrival time under the three pulsing air pressures increased. The peak pressure
difference between
arrival time the peak point
at the “bottom” pressure
was 262arrival time
ms (0.3 under
MPa), 204the
ms three pulsing
(0.4 MPa), and air
202pressures
ms (0.5 MPa); increased.
thus,
The peak pressure arrival time at the “bottom” point was 262 ms (0.3 MPa),
when the pulsing air pressure was 0.3 MPa, the peak pressure arrival time at the lower part of the 204 ms (0.4 MPa), and
202filter
ms bag
(0.5 was
MPa); thus, when
significantly the pulsing
higher than that airobtained
pressure at was 0.3 MPa,
a pulsing the peakofpressure
air pressure 0.4 and 0.5 arrival
MPa. time
This at
theresult
lowerwas
partconsistent with the regular pattern related to the influence of different pulsing air pressure of
of the filter bag was significantly higher than that obtained at a pulsing air pressure
0.4values
and 0.5
(PPuMPa.
) on theThis
peak result was(Pconsistent
pressure with the regular pattern related to the influence of
Pk ), namely, for long filter bags, when the cleaning airflow passed
different pulsing air pressure values
through the upper half of the filter bag, 𝑃𝑢 ( 𝑃 due to the very
) on peakhighpressure
airflow ( 𝑃velocity,
𝑃𝑘 ), namely,
the timefor of
long
the filter bags,
cleaning
when the cleaning
airflow reaching airflow
the “top” passed
point through
was similar the under
upperdifferent
half of the filter bag,
pulsing due to thewhich
air pressures, very high
meant airflow
the
peak pressure arrival time at the “top” point tended to be the same. However,
velocity, the time of the cleaning airflow reaching the “top” point was similar under different pulsing in the lower half of the
airfilter bag, under
pressures, which different
meant pulsing
the peakairpressure
pressures, the energy
arrival time atattenuation
the “top” pointof the tended
cleaningtoairflow
be the wassame.
different, so that the peak pressure arrival time values were different. In
However, in the lower half of the filter bag, under different pulsing air pressures, the energy addition, when the pulsing air
pressure was 0.4 MPa and 0.5 MPa, the difference between the peak pressure
attenuation of the cleaning airflow was different, so that the peak pressure arrival time values were arrival time at the three
measuring
different. points was
In addition, whensmall.
theAdditionally, it was observed
pulsing air pressure was 0.4 MPa that the
andhigher
0.5 MPa,thethe
pulsing air pressure,
difference between
the greater the consumption of compressed air, and the higher the energy
the peak pressure arrival time at the three measuring points was small. Additionally, it was consumption of the cleaning
observed
thatbag.
theMoreover,
higher thehigh pulsingair
pulsing air pressure,
pressure values causedthe
the greater excessive cleaning of
consumption of the filter bag, destroying
compressed air, and the
the initial layer of dust cake and reducing the filtration efficiency of the filter bag. Therefore, when
higher the energy consumption of the cleaning bag. Moreover, high pulsing air pressure values
selecting the pulsing air pressure, the relationship between the cleaning effect, the energy consumption
caused excessive cleaning of the filter bag, destroying the initial layer of dust cake and reducing the
of the pulse jet cleaning device, and the filtration efficiency of the filter bag should be systematically
filtration efficiency of the filter bag. Therefore, when selecting the pulsing air pressure, the
considered. In this investigation, a pulsing air pressure value of 0.4 MPa was selected to carry out the
relationship between the cleaning effect, the energy consumption of the pulse jet cleaning device, and
follow-up experimental study.
the filtration efficiency of the filter bag should be systematically considered. In this investigation, a
According to Figure 7, at a constant pulsing air pressure (PPu ) value of 0.4 Mpa, the effect of
pulsing air pulsing
different pressureair value of 0.4times
injection MPa(100,
was 150,
selected to carry
and 200 ms) on outthethepeak
follow-up
pressure experimental study.
was not significant.
This figure shows that applying the three injection times previously mentioned, the peak pressure of
the filter bag was similar in the two measuring points labeled as “top” and “middle”, respectively.
However, these values were significantly different at the measuring point labeled as “bottom”. On the
other hand, Figure 8 shows that the peak pressure arrival time (tR ) increased with increasing pulsing
air injection time values (tP ), and the difference between the peak pressure arrival time (tR ) at the
same measuring point increased when the distance between the measuring points and the bag mouth
increased. This indicates that the influence of the pulsing air injection time (tP ) on both the peak
pressure (PPk ) and the peak pressure arrival time (tR ) was not the same. This phenomenon may be
explained by the fact that, at a constant value of the compressed pulsing air pressure, by increasing the
Appl. Sci. 2019, 9, 3687 8 of 16

pulsing air injection time, the amount of compressed air injected into the filter bag also increases; in
addition, the peak pressure arrival time at the “top” measuring point was less than 40 ms, which was
significantly smaller than the injection time of the compressed pulsing air set in the experiment. This
indicates that the cleaning airflow pressure at the “top” measuring point had reached its peak value
before the end of the pulsing air injection. Therefore, the compressed airflow injected between the
peak pressure arrival time and the end of the pulsing airflow injection had no significant influence on
the peak pressure and the peak pressure arrival time at the point labeled as “top”. On the other hand,
with the increase of the injection time, the total flow volume of the primary airflow increased, which
in turn increased the total flow volume of the secondary airflow and ultimately led to the increase
of the total amount and duration of the cleaning airflow. Therefore, at the “bottom” point, the total
amount of the cleaning airflow increased, leading to an increase in the refluxing of the cleaning airflow
(this reflux was generated when the cleaning airflow reached the bottom of the bag), thereby causing
the peak pressure to become larger. In addition, the increase of the refluxing lengthened the reflux
action time of the cleaning air flow, which in turn increased the peak pressure arrival time. According
to the test results, when the pulsing air pressure was constant, increasing the pulsing air injection
time did not significantly increase the peak pressure; on the contrary, the peak pressure arrival time
increased, which was not favorable for the cleaning process. Moreover, it increased the consumption
of compressed air, also increasing energy consumption. In the follow-up test, the pulsing air injection
time was2019,
Appl. Sci. set 9,tox 100 ms. 8 of 16

Appl. Sci. 2019, 9, x Figure 6. Effect of pulsing air pressure

pressure on
on the
the peak
peak pressure
pressure arrival
arrival time.
time. 9 of 16

According to Figure 7, at a constant pulsing air pressure (𝑃𝑃𝑢 ) value of 0.4 Mpa, the effect of
different pulsing air injection times (100, 150, and 200 ms) on the peak pressure was not significant.
This Figure shows that applying the three injection times previously mentioned, the peak pressure
of the filter bag was similar in the two measuring points labeled as “top” and “middle”, respectively.
However, these values were significantly different at the measuring point labeled as “bottom”. On
the other hand, Figure 8 shows that the peak pressure arrival time (𝑡 ) increased with increasing
pulsing air injection time values 𝑡 , and the difference between the peak pressure arrival time (𝑡 )
at the same measuring point increased when the distance between the measuring points and the bag
mouth increased. This indicates that the influence of the pulsing air injection time 𝑡 on both the
peak pressure (𝑃𝑃𝑘 ) and the peak pressure arrival time (𝑡 ) was not the same. This phenomenon may
be explained by the fact that, at a constant value of the compressed pulsing air pressure, by increasing
the pulsing air injection time, the amount of compressed air injected into the filter bag also increases;
in addition, the peak pressure arrival time at the “top” measuring point was less than 40 ms, which
Figure
was significantly 7. Effect
smaller thanofthe
theinjection
pulsing air injection
time of thetime pressure
on peak pressure
compressed ofair
pulsingof filter
filter bag.
setbag.
in the experiment.
This indicates that the cleaning airflow pressure at the “top” measuring point had reached its peak
value before the end of the pulsing air injection. Therefore, the compressed airflow injected between
the peak pressure arrival time and the end of the pulsing airflow injection had no significant influence
on the peak pressure and the peak pressure arrival time at the point labeled as “top”. On the other
hand, with the increase of the injection time, the total flow volume of the primary airflow increased,
which in turn increased the total flow volume of the secondary airflow and ultimately led to the
 Appl. Sci. 2019, 9, 3687 9 of 16
Figure 7. Effect of the pulsing air injection time on peak pressure of filter bag.

Effectofofthe
Figure8.8.Effect
Figure thepulsing
pulsingair
airinjection
injection time
time on
on the peak pressure
pressurearrival
arrivaltime
timeofoffilter
filterbag.
bag.

3.2. Effect of Venturi Length and Throat Diameter on the Cleaning Performance of the Filter Bag
3.2. Effect of Venturi Length and Throat Diameter on the Cleaning Performance of the Filter Bag
Figure 9a illustrates the variation of the peak pressure on the filter bag under different venturi
Figure 9a illustrates the variation of the peak pressure on the filter bag under different venturi
throat diameters (40, 50, and 60 mm) at a pulsing air pressure of 0.4 Mpa, pulsing air injection time of
throat diameters (40, 50, and 60 mm) at a pulsing air pressure of 0.4 Mpa, pulsing air injection time
100 ms, and venturi length of 200 mm. The peak pressure in the point P0 was negative, later increasing
of 100 ms, and venturi length of 200 mm. The peak pressure in the point P0 was negative, later
until reaching a maximum value at the point P1. Subsequently, the peak pressure decreased with
increasing until reaching a maximum value at the point P1. Subsequently, the peak pressure
an increasing distance from the measuring point to the bag mouth. Finally, the peak pressure rose
decreased with an increasing distance from the measuring point to the bag mouth. Finally, the peak
again at point P8. As shown in Figure 9b, when the typical venturi-type tube was installed in the bag
pressure
cage, theroseareaagain at point
near the P8. Aswas
bag mouth shown
in thein“blind
Figureaction
9b, whenzone”the typical
of the venturi-type
cleaning tube was
airflow (primary
installed
airflow andin the bag cage,airflow).
secondary the areaInnear the bag
addition, mouth
when the was in theairflow
cleaning “blindwasaction zone”
ejected fromof the cleaning
the venturi
airflow (primary airflow and secondary airflow). In addition, when the cleaning
tube, the airflow outside the filter bag near P0 would be induced to the inside, creating a negative airflow was ejected
from the venturi
pressure zone in tube,
the areathenear
airflow outside
the bag mouth. theItfilter
can be bag
seennear
thatP0thewould be induced
area near to the
P0 is the one withinside,
the
creating a negative pressure zone in the area near the bag mouth. It can be seen
worst cleaning performance. This inference is consistent with the literature [3,20]. Since, in this point, that the area near P0
isthe
thedust
one with
layerthe worst
cannot becleaning
effectivelyperformance.
cleaned, theThis filterinference
materialisinconsistent with its
this area loses thefiltration
literatureability.
[3,20].
Since,
Thus, it should be considered that installing the typical venturi-type tube implies that the area nearloses
in this point, the dust layer cannot be effectively cleaned, the filter material in this area the
itsfilter
filtration ability.will
bag mouth Thus, it should
exhibit be considered
low filtration that installing
performance. the typical
The maximum venturi-type
value of the peak tube implies
pressure
that the area
occurred at near the filter
P1, which was bag mouthwith
consistent will the
exhibit lowoffiltration
results literatureperformance.
[21,22]. In theThe maximum
process value
of moving
ofdownward
the peak pressure
from theoccurred
bag mouth, at P1,
thewhich wasairflow
cleaning consistent with the
expanded and results of literature
diffused outside the[21,22].
filterIn the
bag.
process
Since the of cleaning
moving airflow
downward couldfromnot be the bag mouth,
infinitely expanded,the cleaning airflow expanded
and the consumption and diffused
and energy loss of
outside the filter
the cleaning airflowbag. Since thethe
accompanied cleaning airflowit was
whole process, could not be that
inevitable infinitely expanded,
there would and the
be a maximum
consumption
pressure on the and energy
side wall ofloss of the
the filter cleaning
bag. After theairflow
maximum accompanied the whole
pressure occurred, process, it
the expansion was
effect
inevitable that there
of the cleaning airflow would be aand
decreased maximum
the energy pressure
attenuationon the side wall
increased. of the
Later, thepeak
filterpressure
bag. After
on thethe
maximum
filter bag pressure
decreased. occurred,
The peak the expansion
pressure at P8effect
was of the than
larger cleaning
that airflow
obtained decreased and was
at P7, which the energy
likely
caused by the refluxing of the cleaning airflow.
According to Figures 9a and 10, with an increase in the venturi throat diameter, the peak pressure
on the filter bag increased and the peak pressure arrival time decreased. This indicates that a better
cleaning effect can be obtained by increasing the throat diameter of the venturi. This may be explained
by the fact that the smaller the venturi throat diameter, the greater the resistance of the clearing airflow
(the primary airflow + the secondary airflow) passing through the venturi, which led to the greater
energy loss of the cleaning airflow, thus making the peak pressure smaller and the peak pressure
arrival time greater.
 According
pressure on the to Figure
filter bag 9a and Figure
increased and 10,
thewith
peakan increasearrival
pressure in thetime
venturi throat diameter,
decreased. the peak
This indicates that
pressure on the filter bag increased and the peak pressure arrival time decreased.
a better cleaning effect can be obtained by increasing the throat diameter of the venturi. This may This indicates thatbe
aexplained
better cleaning effect can be obtained by increasing the throat diameter of the venturi.
by the fact that the smaller the venturi throat diameter, the greater the resistance of the This may be
explained by the(the
clearing airflow factprimary
that theairflow
smaller+ the venturi throat
the secondary diameter,
airflow) passingthethrough
greaterthe
theventuri,
resistance of the
which led
clearing
to the
Appl. airflow
greater
Sci. (the
2019, 9, energy
3687 primary airflow + the secondary airflow) passing through the venturi, which
loss of the cleaning airflow, thus making the peak pressure smaller and the10peak ofled16
to the greater energy loss
pressure arrival time greater. of the cleaning airflow, thus making the peak pressure smaller and the peak
pressure arrival time greater.

(a) (b)
(a) (b)
Figure 9.
Figure 9. Effect
Effect of venturi
venturi throat
throat diameter
diameter onon the
the peak
peak pressure
pressure of
of filter
filterbag.
bag. (a)
(a) Plotted
Plotted data
data indicates
indicates
Figure
the peak
the 9. pressure
peak Effect of venturi
pressure each throat
at each
at diameter
measuring
measuring pointononthe peak
the filterpressure of filter
bags with bag. venturi
different (a) Plotted datadiameters.
throat indicates
the
(b) peak pressure
(b) Schematic
Schematic at each
diagram
diagram measuring
showing
showing thethe point on the
distribution filter
of the
distribution bags
of cleaning
the with different
pressure
cleaning on aventuri
pressure filter
on abag throat
with
filter bagdiameters.
thewith
typical
the
(b) Schematic
venturi-type diagram
tube. tube.
typical venturi-type showing the distribution of the cleaning pressure on a filter bag with the
typical venturi-type tube.

Figure 10. Effect of venturi throat diameter on peak pressure arrival time.
Figure
Figure 10.
10. Effect
Effect of
of venturi
venturi throat
throat diameter on peak pressure arrival time.
Figures 11 and 12 show the influence of different venturi lengths (160, 180, and 200 mm) on the
peakFigures
pressure11and
andthe 12 peak
showpressure
the influence
arrivaloftime,
different venturi lengths
respectively. (160, 180,
Experiments wereand
run200at amm) on the
pulsing air
peak pressure
pressure and
and the
the peak
peak pressure
pressure arrival
arrival time,
time, respectively.
respectively. Experiments
Experiments
pressure of 0.4 MPa, pulsing air injection time of 100 ms, and a venturi throat diameter of 60 mm. were
were runrun
at at
a a pulsing
pulsing air
air pressure
pressure
According ofthe
of to
0.4 0.4
MPa,MPa, pulsing
pulsing
results, airair injection
injection
as venturi time
time
tube length of of 100
100 ms,
decreases, ms,andand
the a aventuri
peak venturi throat
throat
pressure diameter
diameter
increases; of60
of 60 mm.
however, mm.the
According to
to the
the results,
results, asas venturi
venturi tube
tube length
length decreases,
decreases, the the
peak peak pressure
pressure increases;
increases;
peak pressure arrival time decreases. This suggests that it is possible to improve the cleaning effect however,
however, the
the
peak peak pressure
pressure
by reducing the arrival
arrival
length timetime
of decreases.
decreases.
the venturiThis This
tube. suggests
suggests
In that
that
addition, it
is ispossible
itwhen possibletotoimprove
reducing improve
the thelength,
the
venturi cleaning
cleaning effect
effect
a larger
by reducing
reducingthethelength
lengthofofthe theventuri
venturitube. In In
tube. addition, when
addition, when reducing the venturi
reducing length,
the venturi a larger
length, effect
a larger
was observed on the peak pressure in the upper half of the filter bag, compared to that in the second
half. On the contrary, a more significant effect was observed in the lower half of the bag for the peak
pressure arrival time under the same reduced-venturi length condition. This occurred because, for the
cleaning airflow, the venturi tube is a resistance component. The longer the venturi tube, the longer
the distance the cleaning airflow moves in the tube, and the greater the energy loss of the cleaning
airflow. Thus, the peak pressure on the filter bag decreases and the peak pressure arrival time increases.
In addition, according to the test results, when the cleaning pressure in point P4 (4000 mm) reached its
effect was observed on the peak pressure in the upper half of the filter bag, compared to that in the
second half. On the contrary, a more significant effect was observed in the lower half of the bag for
the peak pressure arrival time under the same reduced-venturi length condition. This occurred
because, for the cleaning airflow, the venturi tube is a resistance component. The longer the venturi
tube, the2019,
Appl. Sci. longer the distance the cleaning airflow moves in the tube, and the greater the energy
9, 3687 11 loss
of 16
of the cleaning airflow. Thus, the peak pressure on the filter bag decreases and the peak pressure
arrival time increases. In addition, according to the test results, when the cleaning pressure in point
peak
P4 value,
(4000 mm) thereached
injectionitsofpeak
compressed
value, thepulsing air was
injection not finishedpulsing
of compressed (100 ms).
air This
was indicates
not finishedthat(100
the
increase of the cleaning pressure on the upper part of the filter bag was within
ms). This indicates that the increase of the cleaning pressure on the upper part of the filter bag was the time range of the
continuous
within action
the time of the
range of compressed
the continuousair, action
which of made the velocity of
the compressed air,the cleaning
which made airflow in the upper
the velocity of the
part of the filter bag very large, so that the time for the the cleaning airflow to
cleaning airflow in the upper part of the filter bag very large, so that the time for the the cleaningreach the P0~P4 point
airflow to reach the P0~P4 point under different venturi tube lengths was almost the same. Thus,half
under different venturi tube lengths was almost the same. Thus, the peak pressure for the upper the
of thepressure
peak filter bagforwiththedifferent venturi
upper half tube
of the lengths
filter was reached
bag with differentatventuri
the same time,
tube but thewas
lengths peak pressure
reached at
wassame
the quitetime,
different (especially
but the for P1).was
peak pressure Forquite
the lower half (especially
different of the filter for
bag,P1).
theFor
longer the length
the lower of the
half of the
venturi
filter bag,tube,
thethe greater
longer the the energy
length loss
of the of the tube,
venturi cleaning
the airflow,
greater and the longer
the energy lossthe time
of the for the cleaning
cleaning airflow,
and the longer the time for the cleaning airflow to reach the P5~P8 point, which made thearrival
airflow to reach the P5~P8 point, which made the peak pressure smaller and the peak pressure peak
time greater.
pressure smaller and the peak pressure arrival time greater.

Figure
Figure 11. Effect
Effect of
of venturi
venturi length on the arrival time of peak pressure.

Figure 12. Effect

Effect of
of venturi
venturi length on peak pressure arrival time.

3.3. Effect
3.3. Effect of
of the
the New
New Venturi-Type
Venturi-Type on
on the
the Cleaning
Cleaning Performance
Performance of
of the
the Filter
Filter Bag
Bag
Figures 13a
Figures 13a and
and 14
14 display
display the
the results
results for
for the
the effects
effects of
of two
two types
types of
of venturi
venturi tubes
tubes on
on the
the cleaning
cleaning
performance of
performance of the
the filter
filter bags.
bags. Experiments
Experimentswerewereperformed
performedatata apulsing
pulsingairair pressure
pressure ofof0.40.4 MPa,
MPa, a
a pulsing air injection time of 100 ms, and a venturi throat diameter and length of 60 mm
pulsing air injection time of 100 ms, and a venturi throat diameter and length of 60 mm and 160 mm, and 160 mm,
respectively. Data showed that, when using the new venturi tube and for every measuring point,
the peak pressure was higher and the peak pressure arrival time was shorter compared to the values
obtained when using the typical venturi tube. The filter bag with the new venturi displayed a positive
peak pressure at P0 (200 Pa). It is obvious that this new tube had an effect on the peak pressure in
the area near the filter bag mouth. Since the new venturi was placed at a higher height compared to
 Appl. Sci. 2019, 9, x 12 of 16

respectively. Data showed that, when using the new venturi tube and for every measuring point, the
peak pressure was higher and the peak pressure arrival time was shorter compared to the values
obtained when using the typical venturi tube. The filter bag with the new venturi displayed a positive
Appl. Sci. 2019, 9, 3687 12 of 16
peak pressure at P0 (200 Pa). It is obvious that this new tube had an effect on the peak pressure in the
area near the filter bag mouth. Since the new venturi was placed at a higher height compared to that
for for
that the the
typical venturi
typical (as(as
venturi shown
shown in in
Figure
Figure2),2),when
whena acompressed
compressedairflow
airflow(primay
(primayairflow)
airflow)was
was
injected into the new venturi, two streams of secondary airflow near the bag mouth weregenerated.
injected into the new venturi, two streams of secondary airflow near the bag mouth were generated.
AsAsshown
shownininFigure
Figure13b,
13b,the
thesecondary
secondary airflow
airflow that
that is
is above
above the
the inlet
inlet enters
enters the
the venturi,
venturi,mixes
mixesitself
itself
with the primary airflow, and is later ejected. On the other hand, the secondary airflow thatisisbelow
with the primary airflow, and is later ejected. On the other hand, the secondary airflow that below
theventuri
the venturiinlet
inletenters
enters the
the filter
filter bag
bag through
through the
the side
side of
of the
the venturi,
venturi, and
and later
later mixes
mixesitself
itselfwith
withthe
the
cleaning airflow at the venturi outlet. Therefore, when the new venturi was installed, the area near
cleaning airflow at the venturi outlet. Therefore, when the new venturi was installed, the area near
the bag mouth was impacted from the inside to the outside of the bag by the secondary airflow. This
the bag mouth was impacted from the inside to the outside of the bag by the secondary airflow. This
produced a positive peak pressure in P0. In addition, given that in the new venturi the secondary
produced a positive peak pressure in P0. In addition, given that in the new venturi the secondary
airflow was larger than that in the typical venturi, the peak pressure observed at P1 in the filter bag
airflow was larger than that in the typical venturi, the peak pressure observed at P1 in the filter bag
adapted with the new venturi was significantly higher than that where the typical type venturi was
adapted with the new venturi was significantly higher than that where the typical type venturi was
installed. In this case, the value increased by about 440 Pa. Data also demonstrated that, when the
installed. In this case, the value increased by about 440 Pa. Data also demonstrated that, when the
flow rate of the secondary airflow increased, the new venturi did not extend the action time of the
flow rate of the secondary airflow increased, the new venturi did not extend the action time of the
secondary airflow on the lower half of the filter bag. This effect was different to that obtained when
secondary airflow on the lower half of the filter bag. This effect was different to that obtained when the
the pulsing air injection time was increased. Therefore, the peak pressure arrival time on the filter
pulsing air injection time was increased. Therefore, the peak pressure arrival time on the filter bag with
bag with the new venturi was shorter than that observed when using the typical venturi-type (See
the new venturi was shorter than that observed when using the typical venturi-type (See Figure 14).
Figure 14).

(a) (b)
Figure 13.Effect
Figure13. Effect of
of the
the new venturi-type on peak
peak pressure.
pressure. (a)
(a)Peak
Peakpressure
pressureatateach
eachmeasuring
measuringpoint.
point.
(b)
Appl. Sci. Schematic
(b)2019, 9, x diagram
Schematic diagramshowing thethe
showing cleaning pressure
cleaning distribution
pressure on aon
distribution filter bagbag
a filter withwith
the new venturi
the new tube.
13 of 16
venturi
tube.

Figure
Figure 14.
14. Peak
Peak pressure
pressure arrival
arrival time
time measured
measured in
in different
different points
points of
of the
the filter
filter bag
bag adapted
adapted with
with the
the
new venturi tube.
new venturi tube.

3.4. Effect of the New Venturi-type on the Baghouse Cleaning Performance

Experimental results for the cleaning cycle time and dust emission concentration of baghouses
adapted with the typical or the new venturi tubes are shown in Figure 15. In these experiments, the
pulsing air pressure was set to 0.4 MPa and the pulsing air injection time to 100 ms. According to the
Appl. Sci. 2019, 9, 3687 13 of 16

3.4. Effect of the New Venturi-type on the Baghouse Cleaning Performance

Experimental results for the cleaning cycle time and dust emission concentration of baghouses
adapted with the typical or the new venturi tubes are shown in Figure 15. In these experiments,
the pulsing air pressure was set to 0.4 MPa and the pulsing air injection time to 100 ms. According to
the data, the average cleaning cycle time of the baghouse with the new venturi (321 s) was significantly
longer than that of the baghouse with the typical one (257 s). The test conditions of the two groups
were identical, which indicated that the baghouse adapted with the new venturi displayed better dust
cleaning performance. In this new tube, the cleaning airflow was more efficient in removing the dust
cake deposited on the filter bag. Therefore, after dust cleaning, the baghouse installed with the new
venturi displayed lower resistance, and the cleaning cycle time increased. Thus, compared to the
baghouse with the typical venturi, the one with the new venturi tube showed longer cleaning cycle
time, less compressed air consumption, and lower dust cleaning energy consumption during the same
filtration time. In addition, the average dust emission concentration was lower in the new venturi
baghouse (4.38 mg/m3 ), compared to the typical one (6.82 mg/m3 ). These results may indicate that the
new venturi baghouse displayed a better control effect on dust emission. A potential explanation for
this phenomenon is as follows: the cleaning airflow might have destroyed the dust cake deposited on
the filter bag. In consequence, the filter bag was exposed to the dust-laden airflow. Shortly after the
cleaning and before the dust cake was redeposited again, an instantaneous peak of the dust emission
concentration occurred [23–25]. During the 1-h test period, the baghouse with the typical venturi
showed a shorter cleaning cycle and more bag cleaning times (as shown in Figure 16). Additionaly,
the number of peak values for the dust emission concentrations, the dust leakage quantity, and the dust
emission concentration were higher. Additionally, in the new venturi-type design, the dust cleaning
pressure in the area near the mouth of the filter bag was changed into a positive one. This allowed a
more effective removal of the dust cake deposited near the mouth of the filter bag (Figure 13b). Thus,
compared to the baghouse adapted with the typical venturi-type (as shown in Figure 9b), the former
displayed a larger filtration area. This was one of the reasons for the low dust emission concentration
of the baghouse installed with the new venturi.
Appl. Sci. 2019, 9, x 14 of 16

Figure 15. Average

Figure 15. Averagecleaning
cleaningcycle
cycle time
time andand average
average dust emission
dust emission concentration
concentration of baghouses
of baghouses adapted
adapted with aventuri
with a typical typicaltube
venturi
or atube
newor a newdesign.
venturi venturi design.
 Figure 15. Average cleaning cycle time and average dust emission concentration of baghouses
Appl. Sci. 2019, 9,with
adapted 3687 a typical venturi tube or a new venturi design. 14 of 16

(a) (b)

(c) (d)
Figure
Figure16.16.The variation
The of operating
variation resistance
of operating and dust
resistance and emission concentration
dust emission of baghouses
concentration adapted
of baghouses
with a typical
adapted withventuri tube
a typical or a new
venturi tubeventuri design.
or a new (a) The
venturi variation
design. ofvariation
(a) The operatingofresistance
operatingofresistance
baghouse
with the typical
of baghouse venturi.
with (b) The
the typical variation
venturi. (b)of dust
The emission
variation of concentration
dust emissionofconcentration
baghouse with of the typical
baghouse
venturi.
with the(c)typical
The variation
venturi. of
(c)operating resistance
The variation of baghouse
of operating withofthe
resistance new venturi.
baghouse (d) new
with the The variation
venturi.
of(d)
dust
Theemission
variationconcentration of baghouse
of dust emission with of
concentration thebaghouse
new venturi.
with the new venturi.

4. Conclusions
In this paper, the effect of venturi structures on baghouse dust cleaning performance was reported.
Peak pressure and peak pressure arrival time were tested on the filter bags. On this basis, a new
venturi-type tube was designed, and the influence of the new device on the cleaning cycle time and
dust emission concentration of the baghouse were studied. The specific conclusions are as follows:

1. The pulsing air pressure had a great influence on the cleaning performance of the filter bag:
the greater the pulsing air pressure, the greater the peak pressure and the smaller the peak
pressure arrival time. Increasing the pulsing air injection time had no significant effect on the
peak pressure of the filter bag; however, it increased the peak pressure arrival time, which might
have weakened the dust cleaning efficiency.
2. The smaller the venturi throat diameter, the greater the resistance of the cleaning airflow passing
through the venturi, and the greater the cleaning airflow energy loss, thus reducing the peak
pressure and increasing the peak pressure arrival time. In additon, the shorter the venturi length,
the smaller the resistance and the cleaning airflow energy loss, the larger the peak pressure,
and the shorter the peak pressure arrival time.
Appl. Sci. 2019, 9, 3687 15 of 16

3. Compared with the typical venturi, in the new design, the peak pressure as well as the secondary
volumetric airflow entering the filter bag increased. In addition, the peak pressure distribution
changed, and a positive pressure was achieved near the bag mouth. With this, the area near the
filter bag mouth can be effectively cleaned and the filtering capacity properly restored. This effect
is equivalent to increasing the filtration area in the vicinity of the bag mouth, compared with the
typical venturi.
4. Compared to the typical venturi baghouse, the one with the new design displayed longer average
cleaning cycle time and lower energy consumption. Additionally, the new venturi tube reduced
the average dust emission concentration and improved the dust-collecting effect of the baghouse.

Author Contributions: X.L. had the original idea for the study and performed the research. All authors of this
paper also participated directly, either in the planning, execution, or analysis of this study. H.S. was responsible
for the recruitment and follow-up of study participant. All authors read and approved the final manuscript.
Funding: This research was supported by the National Key Research and Development Program of China
(Grant No. 2018YFC0705300) and the Fundamental Research Funds for the Central Universities (Grant
No. CUSF-DH-D-2017093).
Acknowledgments: The authors would like to thank all the editors and reviewers who provided constructive
feedback and suggestions.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Thomas, D.; Penicot, P.; Contal, P.; Leclerc, D.; Vendel, J. Clogging of fibrous filters by solid aerosol particles
experimental and modelling study. Chem. Eng. Sci. 2001, 56, 3549–3561. [CrossRef]
2. Jeon, K.J.; Jung, Y.-W. A simulation study on the compression behavior of dust cakes. Powder Technol. 2004,
141, 1–11. [CrossRef]
3. Simon, X.; Chazelet, S.; Thomas, D.; Bémer, D.; Régnier, R. Experimental study of pulse-jet cleaning of bag
filters supported by rigid rings. Powder Technol. 2007, 172, 67–81. [CrossRef]
4. Lo, L.M.; Chen, D.-R.; Pui, D.Y.H. Experimental study of pleated fabric cartridges in a pulse-jet cleaned dust
collector. Powder Technol. 2010, 197, 141–149. [CrossRef]
5. Sievert, J.; Löffler, F. Fabric cleaning in pulse-jet filters. Chem. Eng. Process. Process Intensif. 1989, 26, 179–183.
[CrossRef]
6. Saleem, M.; Krammer, G. Effect of filtration velocity and dust concentration on cake formation and filter
operation in a pilot scale jet pulsed bag filter. J. Hazard. Mater. 2007, 144, 677–681. [CrossRef]
7. Simon, X.; Bémer, D.; Chazelet, S.; Thomas, D.; Régnier, R. Consequences of high transitory airflows generated
by segmented pulse-jet cleaning of dust collector filter bags. Powder Technol. 2010, 201, 37–48. [CrossRef]
8. Morris, W.J. Cleaning mechanisms in pulse jet fabric filters. Filtr. Sep. 1984, 21, 50–54.
9. Grannell, S.K.; Seville, J.P.K. Effect of Venturi Inserts on Pulse Cleaning of Rigid Ceramic Filters; Karlsruhe
Institute of Technology (KIT): Karlsruhe, Germany, 1999.
10. Morris, K.; Cursley, C.J.; Allen, R.W.K. The role of venturis in pulse-jet filters. Filtr. Sep. 1991, 28, 33–36.
[CrossRef]
11. Lu, H.C.; Tsai, C.J. A pilot-scale study of the design and operation parameters of a pulse-jet baghouse.
Aerosol Sci. Technol. 1998, 29, 510–524. [CrossRef]
12. Lu, H.C.; Tsai, M.L.; Tsai, C.J. The influence of different types venturi on bag cleaning performance.
J. Aerosol Sci. 1998, 29, 241. [CrossRef]
13. Tsai, C.-J.; Tsai, M.-L. Effect of filtration velocity and filtration pressure drop on the bag-cleaning performance
of a pulse-jet baghouse. Sep. Sci. Technol. 2000, 35, 211–226. [CrossRef]
14. Bi, Y.X.; Zhang, M.X.; Chen, J.D.; Chen, H.Y.; Qian, Y.L.; Zhou, R. Influence of venturis on bag cleaning of a
pulse-jet filtration system. Chin. J. Environ. Eng. 2013, 7, 4481–4485.
15. Andersen, B.O.; Nielsen, N.F.; Walther, J.H. Numerical and experimental study of pulse-jet cleaning in fabric
filters. Powder Technol. 2016, 291, 284–298. [CrossRef]
16. Humphries, W.; Madden, J.J. Fabric filtration for coal-fired boilers:dust dislodgement in pulse jet filters.
Filtr. Sep. 1983, 20, 40–42.
Appl. Sci. 2019, 9, 3687 16 of 16

17. Lu, H.C.; Tsai, C.J. Influence of design and operation parameters in bag-cleaning performance of puls-jet
baghouse. J. Environ. Eng. 1999, 125, 583–591. [CrossRef]
18. Lo, L.M.; Hu, S.-C.; Chen, D.-R.; Pui, D.Y.H. Numerical study of pleated fabric cartridges during pulse-jet
cleaning. Powder Technol. 2010, 198, 75–81. [CrossRef]
19. Yan, C.; Liu, G.; Chen, H. Effect of induced airflow on the surface static pressure of pleated fabric filter
cartridges during pulse jet cleaning. Powder Technol. 2013, 249, 424–430. [CrossRef]
20. Qian, Y.; Bi, Y.; Zhang, M.; Chen, H.; Xu, G. Effect of filtration operation and surface treatment on pulse-jet
cleaning performance of filter bags. Powder Technol. 2015, 277, 82–88. [CrossRef]
21. Li, J.-W.; Shen, H.-G.; Yang, Y.; Qu, X.-Y.; Huang, H. Experimental study and analysis on the sidewall
maximum pressure of the pulse bag filter. J. Saf. Environ. 2016, 16, 268–271.
22. Lü, J.; Yan, C.-P.; Yuan, C.-Y.; Fu, Y.; Li, X.; Chen, H.-Y. Effect of the pulse pressure and the jet distance on the
cleaning performance of the long pulse-jet filter bag. J. Saf. Environ. 2016, 16, 285–288.
23. Yao, Y.P.; Shen, Z.A.; Liu, M.L. Research on dust emissions impact factors of electrostatic fabric hybrid
particulate collector and fabric filter. Environ. Eng. 2016, 34, 70–74.
24. Shim, J.; Joe, Y.-H.; Park, H.-S. Influence of air injection nozzles on filter cleaning performance of pulse-jet
bag filter. Powder Technol. 2017, 322, 250–257. [CrossRef]
25. Liu, X.; Shen, H.; Nie, X. Study on the filtration performance of the baghouse filters for ultra-low emission as
a function of filter pore size and fiber diameter. Int. J. Environ. Res. Public Health 2019, 16, 247. [CrossRef]
[PubMed]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).