-------
Table 8. VOLUME FLOW RATES DURING THE TEST
Date Test
June
June
June
June
June
June
June
June
23
23
24
24
25
25
26
26
Average flow rate
1
2
3
4
5
6
7
8
5.04
Flow rate
m3/sec
5
5
5
5
5
5
4
4
±0.21
.17
.08
.27
.27
.04
.04
.73
.73
m3/sec
(dry standard) *
(ft3/sec)
(1.
(1.
(1.
(1.
(1.
(1.
(1.
(1.
(1.78
53
79
86
86
78
78
67
67
±0
x 102)
x 102)
x 102)
x 102)
x 102)
x 102)
x 102)
x 102)
.61 x 102 ft3/sec)
kDry standard P = 760 Torr, T = 21°C.
62
-------
Tests with a gas velocity of 1.8 m/sec scheduled for the last
day of test were cancelled due to an electrical short in the
precipitator.
Inlet Grain Loading
An average grain loading of 7.62 g/Nm3 (3.09 gr/dscf) was ob-
tained in the inlet ducting for the eight tests for which
rapping data are compared. The measurements for each test
are tabulated in Table 9. A comparison with Brink impactor
measurements made within one meter of the inlet of the pre-
cipitator indicated that the particulate loading had decreased
to 6.44 g/Nm3 (2.61 gr/dscf) at the inlet to the precipitator.
The correlation of the inlet feed rates determined from mass
train measurements and the hopper feed rates is shown in
Figure 19. These data indicate that 33% of dust leaving the
feed hoppers did not reach the precipitator. This was not
surprising considering the amount of ducting the gas passed
through before reaching the precipitator (See Figure 7).
PLATE ACCELERATIONS
To remove a precipitated dust layer from a surface by rapping,
enough acceleration-has to be given to the surface to create
a force between the dust layer and the surface sufficient to
cause the two to separate. If the acceleration is too small,
the dust layer is not removed and may even be compacted. If
the acceleration is too large, energy can be imparted to the
dust layer, reentraining it in the gas stream.
The peak X, Y, and Z axis plate accelerations measured during
the test with 1 to 2 mm thick dust layers collected on the
plates were averaged for each accelerometer location. These
accelerations are tabulated in Tables 10 and 11 for pressure
63
-------
Table 9. INLET PARTICULATE LOAD
AND PARTICULATE FEED RATES
Date
June
June
June
June
June
June
June
June
June
June
Avg.
19
20
23
23
24
24
25
25
26
26
Test
Test
0-1
0-2
1
2
3
4
5
6
7
8
Hopper
Grain Load Feed rate Feed rate
g/Nm3 (gr/dscf) kg/hr (Ib/hr) kg/hr (Ib/hr)
9
6
5
8
4
7
7
6
8
8
l-8b 7
±0
Avg. collection
(load cells)
Feed
Brink
.57
.78
.66
.06
.73
.79
.13
.29
.37
.10
.62
.76
rate
110
rate at entrance
impactor - 108
(3.
(2.
(2.
(3.
(1.
(3.
(2.
(2.
(3.
(3.
(3.
±0.
880)
751)
296)a
268)
920)a
160)
892)
549)
394)
286)
092
316)
142
106
98
137
137
120
106
132
127
127
±12
(313)
(234)
(215)
(302)
(303)
(265)
(233)
(291)
(280)
(279
±26)
158
180
151
163
156
163
160
176
163
(348)
(397)
(333)
(360)
(344)
(360)
(353)
(389)
(360)
on precipitator plates ,
kg/hr (243 Ib/hr)
to precipitator as determined
kg/hr (239 Ib/hr)
by
Hole in filter
Omitted tests 1 and 3 from average
CNm3 - Normal cubic meter T = 21°C, P = 760 Torr
64
-------
180
160
m
<
oc
Q
UJ
UJ
LL
Q
Ol
CC
140
120
100
100
70
V
8O
120 140
HOPPER FEED RATE, kg/hr
160
180
Figure 19. Correlation of measured feed rate and hopper feed rate.
65
-------
Table 10. Composite Average of Plate Acclerations During
Rapping for AP = 6.2 x 105 Pa, Dust Covered Plate
Verticala
location
0.91 m
3.05 m
5.18 m
Acceleration
Axis
X
Y
Z
X
Y
Z
X
Y
Z
Plate #1
22
19 ±1.5
—
13
—
11
—
11
—
Plate #2
—
24 ±6
20 ±4
15 ±2
20 ±9
11
12
14
9
, G
Plate #3
26
19
12
11
18
11
—
—
—
±3
±4
— No available data
From the top of the plate
66
-------
Table 11. Composite Average of Plate Accelerations During
Rapping for AP = 4.8 x 10s Pa, Dust Covered Plate
Vertical
location
0.91 m
3.05 m
5.18 m
Axis
X
Y
Z
X
Y
Z
X
Y
Z
Acceleration,
Plate #1 Plate #2
— —
— 19
— 18
11 14.3
— 11
9 11
— —
— —
— —
G
Plate #3
—
14
13
11
15
11
—
—
—
— No available data
aFrom the top of the plate
67
-------
drops across the rapper cylinders of 4.8 x 1Q5 Pa and
6.2 x 105 Pa (70 psi and 90 psi). A problem with the charge
amplifiers occurred during the test, thus some acceleration
data is missing. The X axis was parallel to the direction
of gas flow. The Y axis was in the vertical plane of the
plates, and the Z axis was perpendicular to the plates. The
method of rapping the plates and the accelerometer descrip-
tions and locations were discussed in an earlier section.
The measurements of the accelerations of plate 5-2 from both
the upstream and downstream rappers were averaged. The
variations in rapping intensity as a function of the differ-
ential pressure across the rappers with no dust load on the
plates are shown in Figures 20 and 21 for the Y and Z axis
of the accelerometers on plate 4-1. The accelerations obtained
at all nine locations without a collected dust layer (P =
6.2 x 105 Pa) are given in Table 12. These accelerations
are significantly higher than the accelerations obtained
with the dust layers on the plates.
The range of accelerations that were available with dust
layers on the plates of the precipitator was small. The
change in rapping intensity from 80% to 100% of maximum pro-
duced only a small change in plate accelerations (see
Tables 10 and 11). A modification of the air reserve for
the pneumatic rappers after completion of the test appears
to have significantly increased the rapping accelerations
(see Table 13). The effects of changing plate accelerations
could be studied during any subsequent test with the FluiDyne
pilot precipitator.
Typical time traces of plate accelerations with a dust layer
on the plates are shown in Figure 22. Time traces of plate
68
-------
50
C3
m
o
o
40
30
20
10
NO DUST ON PLATES
DUST COVERED PLATE
100 200 300 400 500
DIFFERENTIAL PRESSURE ON RAPPER, kPa
600
700
Figure 20. Plate 5-1 accelerations as a functional differential
pressure on rapper, upstream rapper.
69
-------
30
CO
z
O
cc
LJ
UJ
o
20
IO
Y AXIS UPPER
100 200 300 400 500
DIFFERENTIAL PRESSURE ON RAPPER, kPa
600
700
Figure 21.
Plate 5-1 accelerations as a functional differential
pressure on rapper, upstream rapper, no dust on plates.
70
-------
Table 12. Composite Average of Plate Accelerations
During Rapping for AP = 6.2 x 10s Pa, Clean Plates
Vertical a
location Axis
0.91 m X
Y
Z
3.05 m X
Y
Z
5.18 m X
Y
Z
Acceleration,
Plate
49
24
29.
34.
26
37.
28
20.
13.
#1
5
5
5
5
5
Plate
—
38
34
36.
24.
26.
42.
64.
17.
#2
5
5
0
5
5
0
G
Plate
45
53
27
29.
23.
33
32
31
13
#3
5
0
— No available data
From the top of the plate
71
-------
Table 13. Composite Average of Plate
Accelerations During Rapping with Modified
Rappers for AP = 6.2 x 10^ Pa, Light Dust Layer
Vertical
location Axis
0.91 m X
Y
Z
3.05 m X
Y
Z
5.18 m X
Y
Z
Acceleration,
Plate #1
73
—
51
41
80
—
75
50
33
Plate #2
IV-
70/107
41/47
40/68
33/52
-/-
69/87
51/113
35/64
G
Plate #3
151
99
35
105
33
—
37
34
34
— No available data
From the top of the plate
72
-------
(/}
b
o
cc
LU
_J
LU
8
i-
_i
Q.
25
25
25
25
25
25
•VtaA.
X AXIS
Y AXIS
Z AXIS
0.03 0.06 0.09 0.12 0.15 0,18 0.21
TIME, sec
Figure 22.
Typical traces of plate accelerations for a dust
covered plate [plate 5-2, 3m from top of plate,
AP = 6.2 x 105 Pa (90 psi)].
73
-------
accelerations without dust layers are given in Figure 23.
The plate accelerations of plate 5-1 were small when the down-
stream rappers were activated and the acceleration of plate
5-3 were small when the upstream rappers were activated.
The accelerations of plate 4-3 suspended from load cells
were compared with the accelerations of plate 5-3. The
results of the comparison of peak accelerations for the lower
accelerometers is as follows:
Plate 4-3 Plate 5-3
Y axis 20G 25G
X axis 19G 31G
Z axis 13G 12G
This data indicates that the load cells dampened the accelera-
tions , but that in the important Z axis the accelerations
were nearly the same.
74
-------
CO
o
a
CC
UJ
UJ
o
UJ
50
25
0
25
25
0
25
25.
0
25
50
25
0
25
50.
25
0
25
1 I I T
AXIS
y|ftA4MMlM
AXIS
AXIS
Y2AXIS
0.03 0.06 0.09 0.12 0.15 0.18 0.21
TIME, sec
Figure 23.
Typical traces of plate accelerations for a clean
plate [plate 5-2; Xi, Yi, Zi, 0.9 m from top of plate;
X2, Y2 3m from top of plate; AP = 6.2 x 10s Pa
(90 psi)].
75
-------
SECTION V
RESULTS AND DISCUSSION
The measurements discussed in this section were aimed at
quantification of rapping reentrainment in terms of the per-
centage and the particle size distribution of the reentrained
particulate and at determining the conditions for removal of
dry dust by rapping. The results of the measurements indicate
that efficiency of dust removal increases as a function of mass
per unit area collected on the plates between raps while the
percent of material reentrained decreases as a function of mass
per unit area. These results are in agreement with those pre-
viously obtained by Plato11 and by Sanayev and Reshidov.15
Measurements of particle size distribution showed that the mass
median diameters of the particles emitted during the raps in-
creased with increased time between raps or equivalently with
mass per unit area collected on the plates between raps.
MASS EMISSIONS
The results of the mass emission measurements provide data for
determining the effect of changes in rapping variables, for
determining the vertical stratification of the dust at the
outlet, and for determining the percentage of total emissions
due to reentrainment for various test conditions. Independent
measurements of the mass emissions of the dust that passed
through the electrified region without being collected and
the mass emissions due to rapping reentrainment were made using
the apparatus and procedures described in an earlier section.
Separate measurements were made of emissions from the upper
68% and the lower 32% of the precipitator.
76
-------
Separate sets of mass train data were obtained for the fol-
lowing rapping conditions:
Rapping Intensity Rapping Interval
Test 1 100% of max 150 minutes
Test 2 100% of max 120 minutes
Test 3 80% of max 12 minutes
Test 4 80% of max 32 minutes
Test 5 100% of max 12 minutes
Test 6 100% of max 32 minutes
Test 7 80% of max 52 minutes
Test 8 100% of max 52 minutes
Since the difference observed in plate acceleration (for the
two different rapping intensities) was small (see previous
section), data from tests having the same rapping intervals
were combined to generate an average set of data. Data for
the two tests with rapping intervals of 120 and 150 minutes
were also combined and averaged. The resulting set of mass
train data showed that hourly emission rates for the emis-
sion due to rapping reentrainment decreased as the time in-
terval between raps increased. Emissions due to rapping were
reduced by collecting thicker dust layers on the plates be-
fore they were rapped.
Precipitator efficiency eventually decreases if the collected
dust layer becomes too thick. However, the between rap data
showed that the emission rates and particulate loadings
between raps remained constant during these tests as the time
interval between raps was increased, for time intervals less
than 155 minutes. The above effects are illustrated in Fig-
ures 24 through 29. The effective particulate loading for
the rap contribution given in Figures 27 and 29 were calculated
by dividing the rate of emission due to rapping by the
77
-------
00
5 <
LU cc
10 20
40 50
60 70 80 90
TIME BETWEEN RAPS, minutes
100 110 120 130 140 150
Figure 24. Mass emission rates for lower 32% of the precipitator
cross sectional area as a function of the time interval
between raps.
-------
-J
vo
5 <
\u a: 2
TOTAL UPPER 68%
BETWEEN RAPS UPPER
RAP UPPER
10 20 30 40 50 60 70 80 90
TIME BETWEEN RAPS, minutes
100
110
120 130
140
150
Figure 25. Mass emission rates for upper 68% of the precipitator
cross sectional area as a function of the time interval
between raps.
-------
O)
UJ
<
CC
O
13
12
11
10
9
8
7
6
5
4
3
2
1
20
COMBINED BETWEEN RAPS
40
60
80
100
120
140
TIME BETWEEN RAPS, minutes
Figure 26. Combined upper and lower mass emission rates as a
function of the time interval between raps.
80
-------
CO
E
Z
~oi
Q"
O
_J
UJ
5
o
en
u_
§
fe
40 60 80 100 120
TIME INTERVAL BETWEEN RAPS, minutes
140
Figure 27. Particulate loading - lower 32% of the precipitator as
a function of time interval between raps.
81
-------
E
2
o>
Q"
O
LLI
I-
D
O
IT
2.4
2.2
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
- ' |
Figure 28.
1.0
0.90
0.80
0.70
0.60 u_
0.50 ^
0.40
0.30
0.20
0.10
40 60 80 100 120
TIME INTERVAL BETWEEN RAPS, minutes
140
Particulate loading - upper 68% of the precipitator as
a function of time interval between raps.
82
-------
00
n
a>
d
O
K
tr
2.4
2.2
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
~ ' | ' | ' I ' I ' | ' I ' I '
— —
I — —
— • STANDARD EFFICIENCY
TEST —
—
__
—
—
— —
a
~~ ^^v^ —
~~ S~---— _ COMBINED TOTAL _^_
"V^ ° COMBINED BETWEEN RAP
^^^^^ ^""^
— **"^^_ COMBINERAP, UPPER AND LO\NER — A«—
1 .U
0.90
0.80
0.70
0:60
0.50
0.40
0.30
0:20
0.10
u_
O
a
O)
20 40 60 80 100 120
TIME INTERVAL BETWEEN RAPS, minutes
140
Figure 29. Combined particulate load as a function of time interval
between raps.
-------
appropriate volume flow rates. The total emission rates of
the combined rap and between rap periods, for the upper and
lower portions of the precipitates were obtained by adding
the rap and between rap emission rates. The total upper and
lower emission rates were added to give total emission rates.
The total effective particulate loading was calculated from
the total emission rates.
Good agreement was obtained between the total effective out-
let grain loading determined by the above procedure and an
outlet grain loading measured when the precipitator was
operated in a normal manner. During the normal test, the
precipitator was operated with a continuous dust feed. The
upstream rapper was activated every 30 minutes and the down-
stream rapper every 60 minutes. During this test, mass
emissions were determined using standard in-stack filter
techniques. The procedures used during the intensive test
program gave an effective total grain loading of 0.58 g/Nm3
(0.235 gr/dscf) for the 32 minute interval test and 0.49 g/Nm3
(0.19 gr/dscf) for the 52 minute interval test. The more nearly
normal procedure gave a grain loading of 0.53 g/Nm3 (0.215 gr/
dscf) for the 30-60 minute rap interval test.
Based on the data given in Figure 29 and the average inlet
grain loading, the overall precipitator efficiencies were de-
determined to be:
12 minute interval tests 88.6%
32 minute interval tests 92.4%
52 minute interval tests 93.9%
120-150 minute interval tests 93.1%
-------
For the normal test (30-60 minute rap interval) the measured
efficiency was 93.0%. The average efficiency of collection
without including the rapping loss was 94.8%. These values
show that significant changes in emission were obtained by
changing rapping intervals. However, it should be noted that
the pilot scale precipitator is basically only one section of
a full scale unit. The effect of rapping reentrainment for
larger, more efficient units is discussed later in this section.
The data given in Figures 24 through 29 show that the emissions
between raps and during the raps in the lower 32% of the
precipitator were significantly higher than the emissions
in the upper 68% of the precipitator. For the tests with
32-minutes rapping intervals, the emission due to rapping in
the lower 32% of the precipitator accounted for 82% of the
total emissions due to rapping and the emissions between raps
in the lower 32% of the precipitator accounted for 60% of the
between rap emissions. The between-rap vertical gradient is
probably due to gravitational settling, sneakage through the
hoppers, and reentrainment in the hoppers. The vertical
stratification of the rapping puffs was visually observed and
was photographically recorded. The major sources of vertical
stratification during the rapping puff appeared to be
gravitational settling and hopper "boil-up". The dust re-
moved from the plates appeared to drop into the hoppers and
then to boil up over the hopper baffles and be carried out of
the precipitator. The vertical stratification in particulate
concentration was clearly evident with the real time system
and is discussed in more detail later in the report.
As mentioned in the introduction, an improvement in overall
efficiency with increasing time between raps had been pre-
viously observed during a Southern Research Institute study
85
-------
of rapping reentrainment at the Bull Run Steam Plant in the
TVA system,17 by Plato,11 and by Sanayev and Reshidov.15 In
Figure 30 the percentage of particulate removed from the
plates that escaped the precipitator is shown as a function
of mass per unit area of dust collected on the precipitator
plates between raps. Data obtained from the Southern Research
Institute-FluiDyne experiment and from the laboratory studies
of Sanayev and Reshidov are shown. The Southern Research
Institute percentages were calculated in two ways. First,
the amount of emissions per rap was divided by the average
mass collected between raps. The average mass collected was
determined for the entire length of the precipitator from
load cell measurements which are discussed later in this sec-
tion. Second, the percentages were calculated in terms of
mass collected only in the first 0.9 meters of the precipita-
tor neglecting the smaller quantity of dust collected in the
last 1.8 meters. According to this data, during the Southern
Research Institute-FluiDyne experiments only 2.7% to 7.2%
of the dust collected on the plates was reentrained and
emitted from the precipitator. The data of Sanayev and
Reshidov show that these percentages depend not only on the
dust layer mass per unit area, but also on flow velocity.
The data in Figure 30 indicate that the most efficient re-
moval of the dust from the plates and transfer to the hoppers
for the ash used in these experiments was obtained with a
dust surface density of 0.5 kg/m2 or greater. Sanayev and
Reshidov15 explained the dependence of reentrainment on dust
surface density as follows. They claimed that thin dust layers
are more intensely disaggregated during rapping than thicker
layers. The thin layers produce fine aggregates that are easily
reentrained, while thicker layers lead to an increase in
size and number of large aggregates reaching the hopper.
86
-------
20
cr>
(3
z ?
0 fc
o <
UJ rr
LU I—
-I LU
8 °
3*
Q t
<-> T
S5 f
Q. (/)
LU
15
10
SOUTHERN RESEARCH INSTITUTE DATA
BASED ON MASS FIRST. .91m
SOUTHERN RESEARCH INSTITUTE DATA
BASED ON TOTAL MASS ON PLATES
SANAYEV & RESHIDOV5 GAS VELOCITY 2m/sec
SANAYEV & RESHIDOV15 GAS VELOCITY 3m/sec
V
1.0
1.5
DUST SURFACE DENSITY, kg/m2
Figure 30.
Percent of dust collected on precipitator plates that
is emitted due to rapping.
-------
These effects were also observed by Plato.11 Plato found
that with dust surface densities greater than 1 kg/m2, dust
was removed in cakes, and for values less than this it formed
clouds. Impactor measurements made during the Southern
Research Institute-FluiDyne experiments showed that the
thicker layers produced larger aggregates than the thin
layers. We expect the percentage of reentrained dust to
decrease faster when large agglomerates are produced than
when small agglomerates or discrete particles are entrained
due to rapping. The large agglomerates will be recollected
faster by gravitational settling and by electrostatic pre-
cipitation than the small ones.
It should be noted, as mentioned by Sanayev and Reshidov, that
the data shown in Figure 30 characterize the optimum con-
ditions for the rapping process and do not consider the
precipitation process. At FluiDyne the effect of the col-
lected dust on the precipitation process appeared to be
minimal. As is discussed later, this is not the case for higher
resistivity dusts.
RESULTS OF IMPACTOR MEASUREMENTS
Precipitator performance, in addition to being a function
of many other variables, is a function of the particle size
distribution of the dust being collected in the precipitator.
Reentrainment when the plates are rapped significantly changes
the particle size distribution of the dust in the precipitator.
To obtain data on this change, we measured the particle size
distribution of the particulate in the gas stream at the inlet,
at the outlet between raps and at the outlet during raps.
The measurements were made with impactors. The resultant data
are used to indicate changes in the particle size distribution
of the suspended particulate when the plates are rapped, to
calculate fractional collection efficiencies with and without
88
-------
rapping reentrainment and to provide input data for a com-
puter model of electrostatic precipitation of dust including
rapping reentrainment.
The particle size distribution of the particulate at the
inlet of the precipitator is shown in Figures 31 and 32. A
procedure by Davies22 was used to correct the distribution for
an isokinetic error. The impactor samples at the inlet were
collected with a sample velocity of 1.5 m/sec while the flue
velocity at the sampling point was 1.2 m/sec. The corrected
distribution is shown by the dashed curve in Figure 31. The
measurements shown in this graph were made within a half meter
of the entrance screen of the precipitator.
The average cumulative percent distribution of the inlet
particulate distribution is given in Figure 32. This data
represents a composite of data obtained from separate loca-
tions at the precipitator entrance. This data indicates that
the inlet distribution was log normal with a mass median
diameter of 13 ym and with a geometric standard deviation of
3.3.
Outlet Between Rap Particle Size Distribution
The average outlet particle size distributions between raps
for all of runs is shown in Figure 33. The distributions for
both the lower and the upper impactor are shown. In Figure 31,
the weighted average of these two distributions is given. The
distributions were weighted according to the area of the pre-
cipitator sampled by the lower and upper mass trains. The im-
pactors were situated near the middle of each of these areas.
The lower impactor showed a higher loading in the 1 to 8 ym
range. This was probably due to gravitational settling and
sneakage through the hopper area. The finding of the same
89
-------
1000
01
O)
o
13
E
100
CORRECT FOR AN ISOKINETIC ERROR —I
32% LOWER
68% UPPER
1 10
PARTICLE DIAMETER, um
100
Figure 31. Inlet and weighted outlet particle size distributions,
dm/d log D.
90
-------
100
DC
HI
HI
5
O
LU
_l
O
a:
10
0.5
I Tl I T
I I ' I ' I ' I
j I . .I I . I , I i I
0.1 1 10 20 40 60 80 90
PERCENT LESS THAN INDICATED SIZE, by mass
Figure 32. Inlet cumulative percent particle size distribution.
91
-------
104
5
8
O
^i
a"
S
TJ
10"
10'
0.1
• INLET
• OUTLET LOWER
A OUTLET UPPER
1.0 10
PARTICLE DIAMETER, urn
100
Figure 33. Inlet and outlet (lower 32% and upper 68% of the
precipitator) particle size distribution, dm/d log D.
92
-------
concentrations for particles greater than 10 ym was unexpected
and may be an artifact of the measurement methods resulting
from the lack of a complete exit plane traverse. The outlet
emission rates between raps are given in Figure 34 for various
size intervals.
Outlet Particle Size Distribution During Raps
The outlet emissions measurements with the impactors agreed
with the results obtained with the mass trains and the real
time system. All three systems measured a decrease in emis-
sions as the time interval between raps was increased. The
results of the impactor measurements are shown in Figure 35.
The particulate loadings during the rapping puffs determined
from the impactor data are shown in Figure 36. These data
have considerable scatter. The real time system also showed
large variations in rapping puff concentrations. The particu-
late loading during the rap was as much as 10 to 200 times
the particulate loading between raps. The particulate load-
ing during the rapping puffs increased as the dust layer
thickness increased. Data given in Figure 35 and 36 also
illustrates the vertical stratification of the rapping puffs.
The particle size distribution of the rapping puffs measured
with the lower impactor are shown in Figures 37 and 38. The
data in Figure 38 indicate that as the time interval between
raps was increased the mass median diameter of the particles
increased. These data show that a smaller percentage of fine
particles and a larger percentage of large particle agglomerates
are produced with thick dust layers than thin dust layers. An
inspection of the upper impactor stages for impactors sampling
rapping puffs indicated that a substantial number of particles
smaller than the lower cut off for the stages were present.
Particles on these stages appeared to be agglomerates made
up of fine particles.
93
-------
10'
g
I
§
W
100
oc
z
o
HI
10-1
10-2
UPPER 68% OF THE
PRECIPITATOR
LOWER 32% OF THE
PRECIPITATOR
SB
SI
se
Stage
S1
S2
S3
S4
S5
S6
S7
S8
S5
S4
S3
S2
S1
STAGE
Size Interval
8.5
5.3
3.7
2.4
1.1
0.67
0.50
12.2 urn
12.2pm
8.5pm
5.3pm
3.7 pm
2.4pm
1.1 pm
0.67 pm
Figure 34.
Outlet emission rates between raps for various size
intervals.
94
-------
U1
01
5
DC
20
LOWER IMP ACTOR
UPPER IMPACTOR
40 60 80 100
TIME INTERVAL BETWEEN RAPS, minutes
120
140
160
Figure 35. Outlet emission rates measured with the impactors as
a function of the time interval between raps.
-------
CTi
n
E
1C
o
z
IT
Q
Q
uu
O
<
DC
LLI
OJ
ID
O
tr
<
0.
96
90
84
78
72
66
60
54
48
42
36
30
24
18
15
6
20
40 60 80 100 120
TIME INTERVAL BETWEEN RAPS, minutes
140
160
Figure 36. Average particulate load during a rap as a
function of the time interval between raps.
-------
105
10*
o>
1
103
AVG.
• TESTS
* TESTS 4 & 6
A TESTS7&8
• TESTS 4, 5,6, 7, & 8
102
0.1
1.0 10
PARTICLE DIAMETER, urn
100
Figure 37. Particle size distribution for rapping puffs,
dm/d log D's (average volume of gas sampled during
the rapping puffs was determined from the sampling
rate and the average persistence time of the rapping
puffs) tests 5, 4 & 6, 7 & 8 had 12, 32, and 52
minutes, respectively, between rap intervals.
97
-------
100
10
oc
LU
5
LU
o
OL
1.0
0.1
I ' ' I
I I ' I ' I
BARS REPRESENT SCATTER
FOR TWO TEST EACH FOR
THE 32MIN AND52MIN
RAP INTERVAL TEST
I
I I.I.I.
0.01 0.1 1 10 20 40 60
PERCENT LESS THAN INDICATED SIZE, by mass
80
Figure 38.
Cumulative percent distribution for rapping puffs, lower
32% of the precipitator cross sectional area, rapping
intervals of 12, 32 and 52 minutes.
98
-------
The emission rates for various particle size intervals (each
irapactor stage) are shown in Figures 39 and 40 for the upper
and the lower impactors. The lower area emission rates were
calculated on the basis of a sampled cross sectional area
of 5.6 m2 (60 ft2) and the upper area emissions were calculated
on the basis of a sampled cross sectional area of 8.9 m2 (96 ft2)
By summing the emission rate measured by the upper and lower
impactor in a particle size band, the total emission rate in
each size band can be obtained. A comparison of the data
shown in Figures 39 and 40 shows that the emissions in the lower
portion of the precipitator dominated the rapping emissions.
As already mentioned previously, the rapping emissions in the
lower portion of the precipitator appeared to be due to gravi-
tational settling and hopper boil-up. The particle size
distributions measured in the upper portion of the precipitator
have large fluctuations. The fluctuations in the data are
possibly due to real variations in the quantity and distri-
bution of the reentrained particulate.
The percent contributions of rapping puffs to total emissions
for each particle size are shown in Figure 41. This data
shows that rapping puffs contributed up to 60% of the emis-
sion of particles with diameters greater than 1 ym. The
data also indicate that the percent contribution for the parti-
cle sizes less than 3 um decreases as the time interval be-
tween raps is increased. These percentages are equivalent to
the percentages defined by equation 5 in the introduction.
Their application for computing rapping emissions is discussed
later in the report.
99
-------
LU
I-
<
CC
Z
g
55
10'1
i LU
. o
IS
r
< LU
CC 0.
10'3
12MINTEST
— X 32MIN TEST
52 WIN TEST
S8
S7
S6
S5
S4
S3
S2
S1
STAGE
Stage Size Interval
S1
S2
S3
S4
S5
S6
S7
S8
8.5
5.3
3.7
2.4
1.1
0.67
0.50
12.2 urn
12.2 urn
8.5 wm
5.3 urn
3.7 «m
2.4 jjm
1.1 urn
0.67 «m
Figure 39.
Mass emission rates for rapping reentrainment, lower
32% of precipitator exit for various particle size
intervals•
100
-------
1x10°
HI
cc
2
01
-------
Ill
N
53
LU
DC
<
o.
cr
o
LL
a.
O
z
X
o
CC
CO
c/J
LU
70
60
50
40
30
20
10
TIME BETWEEN
RAPS
12min
30 min
50 min
6 8 10
PARTICLE DIAMETER, urn
12
14
16
Figure 41. Percent emissions due to rapping reentrainment as a
function of particle size.
102
-------
RESULTS OF REAL TIME MEASUREMENTS OF RELATIVE CONCENTRATIONS
IN FIVE SEPARATE PARTICLE SIZE INTERVALS
Dust removal from the precipitator plates by rapping produces
temporal and spatial variations in the concentration and parti-
cle size distribution of the dust leaving the precipitator.
The rapping puffs can account for a significant portion of the
emissions for some particle sizes. Two modified Royco Model
225 particle counters with five-channel analog ratemeters were
used to monitor rapping emissions in time and space in the
following five diameter intervals: 1.5 to 3 ym, 3 to 6 ym,
6 to 12 ym, 12 to 24 ym, and greater than 24 ym.
Relative emissions between raps for each size interval were
determined by multiplying average count rates between raps
by the time interval between raps. The relative emissions from
rapping puffs for each size interval were calculated by inte-
grating the ratemeter output over the time interval for which
the rapping puff persisted. The dust feed was turned off
when the plates were rapped; thus rapping puff emissions
were observed independently of the emissions not due to
rapping.
Although a size-selective dilution system was used, concentra-
tions in the 1.5 to 3 ym and 3 to 6 ym diameter intervals were
too high for the Royco system during some raps for accurate
analysis, while concentrations in the 12 to 24 ym and greater
than 24 ym intervals were so low that a statistical analysis
based on a Poisson distribution of counts had to be used to
determine total relative emissions in the large particle size
intervals.
Vertical concentration gradients between raps were explored
with the real time system. Significant increases in concen-
tration toward the bottom of the precipitator were observed.
Between raps, the vertical distribution of particles in the
1.5 to 3 ym size interval was found to be relatively constant
103
-------
while the vertical distribution in the other particle diameter
ranges decreased with increasing elevation. An example of
the vertical stratification is shown in Figure 42. These data
indicate that there was some change in the particle size dis-
tribution of the suspended particulate with increased elevation,
The vertical stratification of the particulate is attributed
to several causes. In the lower portion of the precipitator
the increased emissions are due to hopper sneakage and to
gravitational settling of the particulate, especially for
the larger particle sizes. Sneakage through the corona wire
and plate support structure appears to explain the increase in
emissions observed at the top of the precipitator.
The relative total emissions between raps in the five particle
size intervals studied with the real time system were compared
for time intervals of 12, 32, and 52 minutes between raps.
The data indicated that the between-rap emission rates did
not change with the increase from 12 minutes to 32 minutes
between raps. However, there were increased emission rates
at the sampling point just above the lower baffle for the 52
minute interval test. Emissions at this point are probably
dominated by sneakage and may be signficant only for this
particular precipitator-hopper geometry. The increase may
have resulted from voluntary rapping or "sloughing". Relative
emissions with respect to the 12-minute interval between rap
tests are tabulated in Table 14. In the absence of changes in
collection efficiency with increases in dust layer thicknesses
on the plates, the ratios in Table 14 would be expected to be
the same as the ratios of the time intervals between raps.
The vertical stratification of the rapping puffs that was ob-
served with the impactors and mass sampling trains is
clearly shown by the real time particle size data given in
104
-------
z
o
I
i
Q
LU
Z
3
o
o
ta
in
_l
O
IT
d
z
10'
106
10*
103
102
10
i i i r
PARTICLE DIAMETER BAND
• 1.5-3/jm
• 3 -6pm
^ 6- 12 urn
A >-24/jm
I I I
B 20
A
F
F
L
E
40 60 80 100 330 350 370 390 410
DISTANCE FROM BOTTOM BAFFLE, cm
Figure 42.
Spatial distribution of between rap emissions for
various particle sizes.
105
-------
Table 14. RATIO OF RELATIVE TOTAL BETWEEN RAP
EMISSIONS IN FIVE PARTICLE SIZE INTERVALS FOR
RAPPING INTERVALS OF 12, 32, AND 52 MINUTES
Particle Diameter Interval
(microns)
Time Between
Raps
12 min
32 min
52 min
12 min
32 min
52 min
Time
Ratio
1
2.7
4.3
1
2.7
4.3
1.5-3
3-6
6-12
12-24
>24
Emission Ratio Lower Sampling Points
1
2.7
6.0
(5.9)
1
2.9
10.0
(12.7)
1
1.1
3.7
(8.0)
1
2.5
6.0
(14.3)
1
.8
2.9
(5.3)*
Emission Ratio Upper Sampling Points
1
2.7
5.5
1
2.5
7.8
1
2.5
4.4
1
3.1
2.8
1
4.6
2.3
*Includes lowest sampling point
106
-------
Figures 43 and 44. The data in Figure 44 shows that stratifi-
cation occurs for both the upstream and the downstream raps.
The emissions at 23 cm and 42 cm above the lower baffle were
nearly the same. For higher elevations above the baffle the
dust concentration in the puffs decreased with increasing height.
The puffs in the lower portion of the precipitator were ob-
served to last approximately 28 seconds and 24 seconds respec-
tively for the upstream and downstream raps. The rapping puffs
observed in the upper portion of the precipitator lasted about
9 seconds for the upstream rap and 7 seconds for the downstream
rap. Typical traces of upstream and downstream rapping puffs
for both the upper and lower sampling locations are shown in
Figures 45 through 48. On the basis of the flow velocity
(^ .9 m/sec) through the precipitator, it was estimated that
the rapping puffs should last only 2 seconds for both the up-
stream and downstream rap. This estimate was based on a rap
of a 1.8 meter length of the precipitator and the assumption
that the reentrained particulate moves with the velocity of
the gas through the precipitator. The raps observed with the
real time system persisted for 3 to 19 times the 2 second
estimate. In the lower half of the precipitator two separate
puffs were observed. The first puff appeared to last for
approximately 2 seconds, and the second puff approximately 25
seconds for the 1.5-3.0 um diameter particles. The persistance for
large particles was somewhat less. Examples of rapping puffs
in the lower portion of the precipitator are displayed in
Figure 45 for both upstream and downstream raps. The two
separate rapping puffs from both the upstream and downstream
raps were recorded on film. This data indicates that some of
the dust removed from the plates when they are rapped is
reentrained directly into the gas stream and is carried out
of the precipitator at the gas velocity. The remainder of the
dust drops into the hoppers. Of the dust that drops into the
107
-------
IV
105
LL
LL
Q.
O
? 104
Q.
CC
z
03
HI
I
0
i io3
Q.
d
z
102
ml
I I I I I "|
DOWNSTREAM
0 UPSTREAM I PARTICLE D.AMETER
RAP BAND
2 Q O*1.5- 3jjm
O O • 3- 6tim
O •O^6-12um
^ ^ _ D O •l2-24jum
« D
^ 0 0
£ «
B 5^
— - 0^0 •—
• B ^^
o o
* , 8 §
^— ^ • *— -«
^
CJ ^^
6
A A 0
— A A 0 —
A &
A i
I I I I I 7, I
20 40 60 80 100
POSITION ABOVE BAFFLE, cm
406
Figure 43.
Spatial distribution of particles in upstream and
downstream rapping puffs for 12 minute rap interval
test
108
-------
Q
UJ
Z
< I-
1- UJ
Z W
o*1
UJ <
o«
WOE
<|
o 5
II
oc z
u. <
z-
? IE
1- £0 Q.
10b
10B
103
102
U
I I I I I "I
I I I
PARTICLE DIAMETER
• 1.5- 3.0 jam
• 3.0- 6.0 ijm
^ 6.0-12.0 jam
• 12.0-24.0 jam
A > 24 jam
-• •
8 20
A
F
F
L
E
-U
\ \ \
40 60 80 100 330 350 370 390 410
DISTANCE FROM BOTTOM BAFFLE, cm
Figure 44. Spatial distribution of particles in combined upstream
and downstream rapping puff.
109
-------
-H
o
c
o)
I-
z
D
O
o
50
40
30
20
10
500
400
« 300
z
O 200
o
100
> 24 urn DIAMETER PARTICLES
I I I I
I. I I I I
Ll
12 - 24 urn DIAMETER PARTICLES
5x10J
I 2
O
o
1
6 - 12 urn DIAMETER PARTICLES
5x103
o
c. 3
_ UPSTREAM
z
u
5x103|
3 - 6 urn DIAMETER PARTICLES
DOWNSTREAM
1.5 - 3 wm DIAMETER PARTICLES
I I I Lf I I IVI I
I I JI KL_
TIME. 12sec/div
Figure 45. Rapping puffs lower portion of the precipitator, 32
minutes between raps. First set of puffs, upstream
rap. Second set of puffs, downstream rap.
110
-------
24 «m DIAMETER PARTICLES
o
-?
6
c
(/>
1-
Z
O
u
ou
40
30
20
10
I I I I I I 1 I 1 I I I I I
__ __
^_ ^^
—
— • L —
k il A .
12 - 24 «m DIAMETER PARTICLES
8
i
o
o
40
30
20
10
—
—
_
—
7 i
I -
-
6.0 - 12.0 «m DIAMETER PARTICLES
40
30
20
10
—
—
,
J
I-
~~
-
-
I
3.0 - 6.0 urn DIAMETER PARTICLES
auu
•8 400
o
c
OT 300
g 200
O
100
__ __
UPSTREAM
k,
DOWNSTREAM
, n-
5x103
i
o
o
1.5 - 3.0 urn DIAMETER PARTICLES
I I I
TIME, 12sec/div
Figure 46. Rapping puffs upper portion of the precipitator,
32 minutes between raps. First set of puffs,
upstream rap. Second set of puffs, downstream rap.
Ill
-------
> 24 urn DIAMETER PARTICLES
z
o
u
ill
O
I-
i
D
O
U
o
tc
a.
50
40
30
20
10
-I I
—
—
—
__
A . 1 ,
j
1 1 1 1 1 1 1-
... 1
—
—
12-24 Mm DIAMETER PARTICLES
JL
g 5x103
i 4
I 3
O
u 2
6 - 12 Wm DIAMETER PARTICLES
IT
a.
g 5x1 03
o
c. 4
§ 3
O
2 ^
_i
u
p 1
tc
a.
J - o Mm UlAivie i tn r«n i li^uca
UPSTREAM
~ i
-A
M
— 1 1 DOWNSTREAM —
1 L
- IV L -
z
o
o
ui
5x103
4
3
2
1
1.5 - 3 urn DIAMETER PARTICLES
•Lr
TIME, 12sec/div
Figure 47. Rapping puffs, lower portion of the precipitator.
12 minutes between raps. First set of puffs, upstream
rap. Second set of puffs, downstream rap.
112
-------
24 Mm DIAMETER PARTICLES
-2
§°
en"
1-
z
3
O
o
UJ
o
H
oc
<
0.
1
o
C
w"
1-
z
3
O
o
UJ
_i
PARTIC
i
.
o
y>
r~
3
0
O
Ul
_j
y
DC
<
Q.
U
-H
O
c
to"
H-
3
O
O
UJ
o
1-
oc
<
a.
u
-2
d
c
P
z
3
O
O
UJ
^ml
O
1—
It
<
a.
3U
40
30
20
10
| |
i i i i i i i i
i^ ._
•
1
1
1
1
1
- -
L
•
• ^^««
n.
fl i i
A II 1 I
12 - 24 urn DIAMETER PARTICLES
50
40
30
20
10
—
^~
—
—
—
^M
—
il.r
6 - 12 urn DIAMETER PARTICLES
50
40
30
20
10
—
"
— —
.^
•>ii . . .•!.«
3 - 6 urn DIAMETER PARTICLES
400
300
200
100
•~- ~^
-
1
1
Ln^ !• _
1
— 1 —
11
, 1.5 - 3 urn DIAMETER PARTICLES
4
1
_ ^_
^^ ^^
•» — ^
y»
-urH-u-uuw-k.
TIME, 12sec/div
Figure 48. Rapping puffs, upper portion of the precipitator.
12 minutes between raps. First set of puffs,
upstream rap. Second set of puffs, downstream rap
113
-------
hoppers a portion rebounds, breaks up, and is carried out of
the precipitator. At FluiDyne the time separation of the
maxima in concentration for the two puffs were approximately
7 seconds for the upstream rap and 10 seconds for the down-
stream rap. It was also noted that the puff from the upstream
rap for the lower portion of the precipitator was first ob-
served approximately 3 seconds after the rapper was activated
and the downstream puff was observed approximately 1 second
after the downstream rapper was activated. The puffs in the
upper portion of the precipitator lasted for approximately
4 seconds. Multiple puffs were also observed in the upper
sampling area with each puff persisting approximately 2 seconds.
There are several possible mechanisms for the secondary puffs
in the upper region of the precipitator. The most likely
mechanism is a secondary breakaway of the dust initially dis-
lodged by the rap but not immediately removed from the plate.
Some of the upper secondary puffs were observed to occur as
much as 17 seconds after the primary puff. These delayed puffs
may have been due to hopper boil-up and turbulent transport
of the hopper boil-up. Examples of rapping puffs in the upper
portion of the precipitator are shown in Figures 46 and 48.
The data displayed in Figure 43 and the rapping puffs shown in
Figures 45 and 47 show that in the lower portion of the pre-
cipitator the upstream rap always produced the largest contri-
bution to the rapping emissions in the lower half of the pre-
cipitator. This phenomenon occurs because precipitators col-
lect dust in an exponential fashion. The first plate of the
FluiDyne precipitator collected the thicker dust layer. Since
the dust surface density was higher than in other parts of the
precipitator and the particle size larger, the dust was more
easily removed from the first plate. Thus, on rapping, a
larger quantity of dust is removed from the first two plates and
dropped into the hoppers than from the third plate. Some of this
dust rebounds and passes through the hoppers without being
114
-------
recollected. On the downstream plate, the dust surface density
was much smaller thus did not contribute as much to rapping
emissions. In the upper portion of the precipitator, the rela-
tive size of the rapping puffs from the upstream and downstream
rap did not depend on the amount of hopper boil-up. When the
dust surface density in the downstream section was too small
for dust removal, the upstream rap contributed the larger
portion of the emissions in the upper portion of the precipita-
tor, as indicated by the rapping puffs shown in Figure 48,
which were obtained with a 12 minute interval between raps.
However, when the dust surface density on the downstream plates
was high enough that dust was removed, the downstream rap pro-
duced the largest quantity of dust in the upper portion, as
shown by the rapping puffs displayed in Figure 46 which were
recorded with a 32 minute interval between raps. Although a
larger quantity of material was initially reentrained in the
inlet of the precipitator, some of this particulate was re-
precipitated. Reprecipitation also occurs for material re-
entrained in the outlet section, but at a much lower efficiency.
It is for this reason that outlet field raps in full scale pre-
cipitators play a major part in rapping reentrainment, while
reentrainment from the inlet section is usually not a severe
factor in performance unless hopper boilup can sneak through
the hoppers. The data from FluiDyne indicates that the effect
of inlet raps on outlet emissions will depend on what happens
to the hopper boilup. Hopper boilup in the outlet section of
a full scale precipitator can play a major part in rapping
reentrainment according to these experiments.
The total percent contribution of the raps to total emission
in each size band determined with the Royco system for dif-
ferent time intervals between raps are given in Figure 49.
The percentages obtained with the real-time system are
115
-------
100
HI
N
O
-------
comparable to the percentages calculated from the impactor
data. The real-time system data show a decrease in the net
total emission rates as the time interval between raps is
increased, in agreement with the impactor and mass train data.
The decrease was most significant for particles with optical
diameters less than 6 ym. As mentioned previously, this is
most likely due to agglomeration of the particles with the
thicker dust layers. The data given in Figures 41 and 49
shows that the percent increase in the concentration of the
large particles during rapping is significantly larger than
the percentage increase for the small particles. This occurs
because the rapping puffs emissions have a larger percentage
of large particles than the between rap emissions. For a
high efficiency precipitator which removes nearly 100% of
the large particles entering the precipitator, rapping can
contribute up to 100% of the particle emissions for particles
with large diameters. Total percent emissions were not de-
termined from the real time system, since comparison could
only be made for data from the same size interval without
correcting the data for size dependent probe losses, size
selective dilution and anisokinetic sampling. Correction
factors for probe losses, size selective dilution, and aniso-
kinetic sampling were not considered accurate enough to
justify the effort, especially since the percent of total
emissions due to rapping were available from mass train and
impactor data. However, one comparison between the real
time data and the impactor data for a between rap case was
attempted.
Expected counting rates for various particle sizes were gen-
erated from the impactor data. The expected counting rates
included the correction factors for probe losses, size selec-
tive dilution, and anisokinetic sampling in the real time
system. The results of the comparison are given in Table 15.
The sampling flow rate used with the real time system was ap-
proximately 3 times greater than that required for isokinetic
117
-------
Table 15. COMPARISON OF REAL-TIME SYSTEM
DATA AND IMPACTION DATA OBTAINED BETWEEN RAPS
Predicted
Count Rate
Particle from Measured
Diameter Impactor Data Count Rate
ym number/sec number/sec
1.5-3 4240 4500
3-6 590 475
6-12 12 4
12-24 2.3 0.4
24-48 0.83 0.2
118
-------
sampling. This high flow rate was used to solve a condensa-
tion problem in the diluter. The real time test results ob-
tained during the test with 120 minute and 150 minute intervals
between raps were confused by condensation forming in the
diluter. Increasing the sample flow velocity raised the tem-
perature in the diluter enough that the condensation did not
occur during the other tests.
The real time particle size system also proved useful for
studying variations in between-rap outlet concentrations as a
function of precipitator currents and voltages.
CONTINUOUS MONITORS, KONITEST AND TRANSMISSOMETER
The dust concentrations at the outlet were continuously
monitored with a Lear Siegler RM41p transmissometer and with
a Konitest meter. These instruments provided fast real time
response for determining changes in precipitator operation
during the test. The effects of sparking and changes in
current density were immediately observed with these instru-
ments.
The amplitude of the Konitest current tended to correlate with
the amount of dust removed from the plates, as shown in Figure
50. Unfortunately, neither the transmissometer nor the Konitest
were monitored closely enough during the test and the majority
of the rapping puffs produced signals that were off-scale on
the recording instruments. Thus, these instruments were not
used to make quantitative measurements of rapping emissions.
An additional reason for not using the transmissometer data
is that the transmissometer output is a function of the
particle size distribution of the particulate in the view
volume and this changed when rapping puffs occurred.
119
-------
0.25
0.20
Z
LU
CC
tr
D
O
111
j-
o
0.15
0.10
0.05
234
MASS OF DUST REMOVAL, kg
Figure 50.
Correlation of mass of dust removed from plate 4-1
and amplitude of Konitest meter current output for
the upstream rapping puff for test 5 (12 minute rap
interval).
120
-------
Between-rap data obtained with the Konitest during tests 5
and 6 indicated an increase in the between-rap loading during
the tests. However this was not observed during the previous
tests 3 and 4 with the Konitest nor during any of these tests
with the transmissometer.
DUST COLLECTION RATES ON THE PLATES AND DUST REMOVAL FROM THE
PLATES
In this section, the removal of the collected particulate from
the plates is discussed. The data provide a background for
modelling the process of removal of the collected particulate
from the precipitator plates by rapping and for determining
optimum rapping conditions.
In the FluiDyne test facility the suspended particulate is
collected on large plates by electrostatic precipitation and
removed from the plates by rapping them with a pneumatic
rapper. The design of the rapping system was described in an
earlier section. In place of the pneumatic system, mechanical
drop hammers, magnetic drop hammers, or vibrators might have
been used with similar results.
For proper rapping, these systems must supply sufficient
force to remove dust from the plates and at the same time not
directly reentrain the dust into the gas stream. Ideally, the
collected layer should be freed just enough to slide down the
precipitator plates and into the hoppers. Even under this
ideal condition dust can still be reentrained when it falls
into the hoppers.
The rapping system at FluiDyne was successful in removing
dust from the plates of the first meter (3 ft) of the preci-
pitator but unsuccessful in removing the collected particulate
efficiently from the last 1.8m(6 ft) of the precipitator. The
mass of particulate collected on plates 4-1, 4-2, and 4-3 as
121
-------
a function time for tests with rap intervals of 12 and 32
minutes are shown in Figures 51 through 54.
From this load cell data and similar load cell data for the
various test conditions, we calculated the collection rates/
removal rates, and percentage removal rates as functions of
the time between raps.
The data given in Figure 51 shows that no dust was removed
from plate 4-1 until more than 15 kg of particulate had col-
lected on the plate. The data also show that the residual
layer remaining after a rap increased throughout the 12-minute
interval rap test. The data in Figures 52 and 54 showed that
very little dust was removed from the outlet plate when it
was rapped.
It was also observed that during the night about 10 kg of dust
fell off plate 4-1. According to the load cell data, dust fell
off the plates every time the precipitator was turned off
after several hours of testing.
The collection rates determined from load cell data are dis-
played in Figure 55. The collection rates show the expected
theoretical exponential dependence on precipitator length and
a collection rate that is independent of the time interval be-
tween raps. A decrease in collection rate with increasing
time between raps was expected. This was not observed. From
data, it was estimated that 76% of the particulate was collected
in the first meter of the unit.
Particulate removal from the plates was observed to vary as a
function of the time interval between raps and as a function of
location in the precipitator. The dependence of particulate
122
-------
30
25
20
O
Q
O
UJ
O
O
OT
15
10
O DUST FEED ON
D UPSTREAM RAP
O DOWNSTREAM RAP
X MASS ON PLATE AT
END OF PREVIOUS TEST
>v14 HRS BETWEEN TEST
PRECIPITATOR OFF
60
120
180
240
TIME, minutes
Figure 51.
Load cell data, mass collected on plate 4-1, 12 minute
interval between raps.
123
-------
45
40
°?
v
Q
Z
UJ
H
Q.
Z
O
O
UJ
O
UJ
O
O
CO
CO
<
35
15
O DUST FEED ON
O UPSTREAM RAP
O DOWNSTREAM RAP
X MASS ON PLATES AT
END OF PREVIOUS TEST
-v14 MRS BETWEEN TEST,
PRECIPITATOR OFF
60
120
180
240
TIME, minutes
Figure 52.
Load cell data mass collected on plates 4-2 and 4-3,
12 minute intervals between raps.
124
-------
35
UJ
a.
Z
O
a
UJ
o
LU
o
o
oo
CO
30
25
20
15
10
60
PLATE NO. 4-1
O DUST FEED ON
a UPSTREAM RAP
O DOWNSTREAM RAP
I
O PRECIP. OFF
120 180
TIME, minutes
240
Figure 53. Load cell data, mass collected on plate 4-1, 32 minute
intervals between raps.
125
-------
O)
n
<
IN
V)
Ul
O
Q
UJ
u
UJ
O
U
45
40
35
20
15
10
PRECIPITATOR OFF
O DUST FEED ON
D UPSTREAM RAP
O DOWNSTREAM RAP
,
60
120 180
TIME, minutes
240
Figure 54. Load cell data, mass collected on plates 4-2 and 4-3,
32 minute intervals between raps.
126
-------
u.tu
0.35
c 0.30
OJ
£ 0.25
DC
Z
° 0.20
O
LLJ
_l
_J
8 0.15
to
-o
0.10
0.05
1 | I | I | 1 | 1 | 1 | 1 | 1 |
& PLATE NO. 4-1
O PLATE NO. 4-2
D PLATE NO. 4-3
— —
A A
— A —
— —
— . —
— ' —
00 ° _
~~ D 0 0
n n
I ,D I , I , I , I , V , I , I
20 40 60 80 100 120 140 160
TIME INTERVAL BETWEEN RAPS, minutes
Figure 55. Collection rate of particulate as a function of time
interval between raps.
-------
removal on these parameters can be explained by the effect
that dust surface density (mass per unit ares) plays in removal
of the dust from the plates and the effects of consolidation
time on the cohesion characteristics of the collected dust
layer. In Table 16, the removal efficiencies, the percent of
freshly collected dust that is removed by a rap, and the per-
cent of total mass on the plate that is removed by a rap are
given for time intervals of 12, 32, 52, and 120 minutes be-
tween raps. The removal efficiencies versus time between raps
are displayed in Figure 56. These data were computed from
load cell data similar to that in Figures 51 through 54. The
data show that efficient removal (>75%) of the dust collected
between raps occurred only when the mass per unit area collected
between raps exceeded .8 kg/m2. It should be noted that the
percentage dust removed from the plates varied from rap to rap
even for the same conditions. There are also indications that
no removal occurred unless the mass per unit area on the plates
was greater than 2.5 kg/m2. The dust that is not removed with
the first rap after it is collected appears to develop higher
tensile and shear strengths than the freshly collected layer.
The load cell data for plates 4-1, 4-2 and 4-3 indicated that
dust was not removed even when the mass per unit area of the
dust collected on these plates finally exceeded the levels for
which freshly collected dust was removed from plates. Raps with
insufficient intensity to remove the dust layer may have com-
pacted the dust layer making it even harder to remove.
In addition to the consolidation that appears to have occurred
during these experiments, it is suspected that the collected
layer is harder to remove from the second and third plates
due to changes in the physical properties of the layer.
128
-------
Table 16. MASS PER UNIT AREA ON PLATES WHEN RAPPED,
MASS PER UNIT AREA COLLECTED BETWEEN RAPS AND
MASS PER UNIT AREA REMOVED BY RAP
NJ
Rap
Interval
Test minutes Plate
3 12 4-1
4-2
4-3
4 32 4-1
4-2
4-3
7 52 4-1
4-2
4-3
2 120 4-1
4-2
4-3
Mass per Unit
Area of Dust
on Plate When
Rapped
kg/m2
2.54
3. 30
0.78
2.78
3.98
1.08
3.03
2.60
1.65
4.54
*
— *
Mass per Unit
Area of Dust
Collected Be-
tween Raps
kg/m2
0.40
0.08
0.05
0.82
0. 13
0.09
1. 35
0.23
0.12
2.91
1.07
0.25
Mass per Unit
Area Removed
By Raps
kg/m2
0.27
0.00
0.00
0.77
0.02
0.03
1.07
0.01
0.00
3.16
0.77
0.04
Efficiency of
Removal of
Particulate
Collected
Between Raps
%
68
00
00
94
15
33
79
4
0
109
72
16
*Load cell zero shifted.
-------
OJ
o
Ol
O.
100
90
80
70
60
50
40
30
20
10
.26
20
40
% OF PARTICULATE
COLLECTED BETWEEN RAPS
O % OF TOTAL PARTICULATE
COLLECTED ON THE PLATES
60
80
100
120
.78
TIME INTERVAL BETWEEN RAPS, minutes
1.3 1.8 2.3 2.9
MASS/AREA GAINED BETWEEN RAPS, kg/m2
140
3.4
160
Figure 56. Dust removal efficiency for dust on plate 4-1 versus
the time interval between raps.
-------
The cohesive strength of the collected dust layer theoretically
increases along the length of the precipitator, since the parti-
culate is fractionated into particle size distributions with
decreasing mass median diameters from the inlet to the outlet.
The feed dust at FluiDyne had a mass median diameter of 16 ym
and the dust collected in the last 20 cm length of the precipi-
tator had a mass median diameter of ^3 ym.
The simple theory of dust removal discussed in the introduction
(see equations 2 and 3) indicates that, for a given accelera-
tion perpendicular to the collection plates, the collected
layer will be removed if the mass per unit area exceeds some
minimum value. The average acceleration perpendicular to the
plates at FluiDyne was approximately l.lxlO1* cm/sec2 (11 G) .
An estimated value for the tensile strength of the collected
dust layer is ^2x103 dynes/cm2 based on the tensile strength
tabulated in Table 1.
The application of this tensile strength to other precipitated
fly ash layers is highly uncertain, since cohesion and adhesion
of dust layers depend on many parameters, including the physi-
cal properties of the dust, the electrical conditions at the
collection plate, and the chemical composition of the gas from
which the dust is precipitated. If we ignore the above objections
to the use of this tensile strength, a minimum mass per unit
area of 1.8 kg/m2 was calculated for dust removal (from the
ratio of the above tensile strength and plate acceleration).
This minimum dust surface density agrees nicely with the data
from the FluiDyne experiment, which indicates that 100% re-
moval of the freshly collected layer occurred at a mass per unit
area between 1 and 3 kg/m2.
131
-------
These load cell data show that if plate accelerations and
time intervals between raps are not properly chosen to match
the dust layer tensile strength, 100% removal of the dust
from the precipitator plates will not occur. The data indi-
cate that larger percentages of dust were removed from the
plates as the time interval between the raps was increased.
The data also indicate that, if the residual dust layers
collected during the test were to be removed, plate accelera-
tions needed to be increased. The amount of increase needed
was not determined, but estimates could possibly be obtained
by measuring dust removal with increased plate accelerations or
by measuring the effects of consolidation time, particle size,
and vibrational compaction on the tensile strength of the
precipitated dust layer.
The FluiDyne experiments show the buildup of a residual dust
layer. Precipitator performance is often observed to degrade
after a week or two weeks of operation after starting with clean
plates. This degradation is usually ascribed to the buildup
of a steady-state layer of collected particulate on the pre-
cipitator plates. One solution to the residual dust layer may
be the use of a set of auxiliary rappers that would rap the
plates with enough force to remove the layers on a daily or
weekly basis. An increase in plate acceleration to continually
prevent any buildup would in many cases produce undesirable
reentrainment characteristics.
The data indicate that for efficient dust removal from the
plates, the plates should not be rapped at all until there is
sufficient dust buildup to give a reasonable chance for
efficient removal with the available plate acceleration.
132
-------
The upper limit on the time interval between raps is set by
the effect the dust layer has on precipitator operation.
Obviously, if the dust layers become too thick they interfere
with gas flow through the precipitator and inhibit performance.
The collected dust layers also affect the electrical operation
of the precipitator. For low resistivity dust the electrical
effects are small but for high resistivity dust they can be
significant. The resistivity of the dust layer affects
precipitator operation in two ways: (1) by setting the current
density at which the precipitator operates, if the resistivity
is high enough that the current density is limited by back
corona to a current density below the maximum clean plate cur-
rent density, and (2) by adding a resistivity element to the
circuit. The first effect is relatively independent of the
thickness of the dust layer and depends only on the dust
resistivity- The second effect depends on the thickness of
the dust layer, dust resistivity, and current density. To
prevent an excessive voltage drop in the dust layer, the
thickness of the layer must be kept to a reasonable value.
Another electrical effect of collecting too thick a layer be-
fore rapping is a reduction of collection efficiency during
rapping due to corona quenching produced by the space charge
that develops with a large particulate load in the gas stream.
The impactor data indicate that particulate loads range from
4.9 to 49 g/Nm3 (2 to 20 gr/dscf) during the rapping puffs.
However, the data from these experiments also indicate that
the collected mass per unit area would need to exceed ^4 kg/m2
for the particulate load produced by rapping to have a large
effect on overall emissions.
This data on dust removal from precipitator plates does not supply
all the information desired for modelling dust removal in any
full scale precipitator. The effects of variations in dust
properties and in plate acceleration were not determined. The
133
-------
FluiDyne data showed substantial variations in the removal
efficiency as a function of location in the precipitator and
as a function of the time interval between raps. It is sus-
pected that the variations will be even larger in full scale
precipitators. These variations present problems for the
modelling of rapping reentrainment. The data show that only
under ideal conditions can it be assumed that all the dust
collected at a particular location on a plate is subject to
reentrainment every time the plates are rapped.
COLLECTION RATES, FRACTIONAL EFFICIENCIES, AND CONTRIBUTION
OF RAPPING REENTRAINMENT TO TOTAL EMISSIONS
During this study of rapping reentrainment, the performance
of the precipitator excluding rapping reentrainment losses
was determined by measuring the fractional efficiency between
raps and by measuirng the collection rate of the particulate
on the precipitator plates between raps. The collection rate
data and the fractional efficiency data obtained between
raps were compared with predictions made with the Southern
Research Institute computer model of electrostatic precipita-
tion. 1 The comparison of the collection rates measured with
the load cells and the collection rates calculated with the
computer model is shown in Figure 57. Reasonable agreement
between measured and calculated rates was obtained. However,
the computer model calculation was low for the last meter of
the precipitator length. This discrepancy could be due either
to a load cell error or to an unmodelled collection effect.
The measured fractional efficiency between raps and the
measured fractional efficiency including rapping losses
were compared with theoretical fractional efficiencies
(Figure 58). The theoretical fractional efficiencies were
calculated using the Southern Research Institute computer
model, the design parameters for the FluiDyne pilot
134
-------
0.30
c
E
UJ
oc
2
g
i-
o
111
o
o
u.
O
'E
IT
O
u.
tr
g
a
o
o
0.25
0.20
0.15
O.IO
0.5
—— THEORY, SRI Model1
EXPERIMENTAL
O
i
O.5 1.0 1.5 2.0
LOCATION INTHEPRECIPITATOR, m
2.5
Figure 57. Collection rate versus horizontal location in the
precipitator.
135
-------
u>
0.01
0.1
cc
UJ
z
UJ
o_
10
20
40
60
0.1
1.0
10
99.99
99.9
99
98
95
90
o
o
u.
80 tu
60
40
100
PARTICLE DIAMETER, >un
Figure 58,
Fractional collection efficiency versus particle
diameter, solid line theory, circles experimental
between rap, diamonds experimental, include rapping
reentrainment.
-------
precipitator, and the measured electrical conditions, volume
flow rate, and inlet particle size distribution for the
FluiDyne test. The fractional efficiencies are compared
for the case in which nonidealities are excluded from the
computer calculations. The shapes of the computed and
measured fractional efficiency curves did not agree either
when rapping losses were included or when they were excluded.
For the no rap condition, the computer model underpredicted
the fractional collection efficiencies for particle sizes
between .7 ym and 4.5 ym and overpredicted the fractional
collection efficiencies for particle sizes greater than 4.5 ym.
The discrepancy for the large particles has been previously
reported.23 It was speculated that the discrepancy was due
to rapping reentrainment. The measured fractional efficiency
curve without rapping reentrainment losses indicates that
rapping reentrainment is not the only cause for the discrepancy.
A comparison of the computed and measured fractional effici-
ency curves indicates a shift in the particle size distribution
of the particulate inside the precipitator that is not accounted
for by the different migration velocities calculated for each
particle size. A portion of the discrepancy is undoubtedly
due to reentrainment between raps. The reentrainment between
raps of agglomerates from hoppers, collection plates, chamber
walls, corona wires, plate support structures, etc., was not
included in the computer calculations. Sneakage of particulate
through hoppers, and corona wire and plate support areas which
permits large particles to escape collection can also cause the
discrepancy. Coagulation of the particulate in the gas stream
may be another possibility. Strong electric fields in electro-
static precipitators produce coagulation and increase the
efficiency of precipitation.21* This effect is not included in
the present computer model and possibly could produce the
effects observed, explaining the load cell measurements and
the fractional efficiency curve without rapping losses.
137
-------
The fractional efficiency curve including rapping given in
Figure 58 is for the tests with 12 minute intervals between
raps. The largest rapping emissions were observed during these
tests. The fractional efficiency curve including rapping was
computed from the measured inlet feed rates in each size band
and the calculated total outlet emission rates in each size
band. The total outlet emissions were obtained by summing out-
let emissions for each size band for the raps and for the
between rap intervals.
As expected, the inclusion of the rapping losses increased the
discrepancy between the computed fractional efficiencies and
the measured fractional efficiencies for the large particles.
Contribution of Rapping Reentrainment to Total Emissions
In the FluiDyne tests, rapping losses were observed to account
for a significant portion of the total emissions. The per-
centage of total emissions resulting from rapping reentrain-
ment ranged from 18% to 53%, depending on the amount of dust
on the plates when they were rapped.
A plot showing the percentage contribution of the rapping puffs
to total emissions for all size bands is given in Figure 59.
This data indicates that minimum rapping losses would have
been obtained with a time interval between raps somewhere
between 52 and 150 minutes.
Since the FluiDyne test was conducted under conditions
representative of the inlet sections of many precipitators
attached to coal-fired boilers, the data indicate that im-
provements in efficiency can be obtained by increasing the
time interval between raps. Many field units operate with
time intervals between raps for inlet sections on the order
of 2 to 6 minutes; the FluiDyne data indicate that these times
138
-------
U)
LU
Q
l/J
UJ 5
II
u_ ill
O UJ
50
40
30
20
10
20
I I
I I
40 60 80 100 120
TIME INTERVAL BETWEEN RAPS, min
140
160
Figure 59. Percentage of total emission due to rapping reentrainment.
-------
should be increased by a factor of 6 to 15. The exact factor
depends on many parameters/ including the effect on between-
rap operation.
EFFECT OF RAPPING REENTRAINMENT ON EFFICIENCY, PENETRATION,
AND SIZING OF ELECTROSTATIC PRECIPITATORS
Generally, if rapping reentrainment were entirely prevented
the percentage increase in efficiency would be small, but the
percentage decrease in penetration would be large. Examples
using the percentages obtained during the FluiDyne experiments
are given in Tables 17 and 18. These examples also show that
the percentage increase in efficiency decreases with increas-
ing efficiency of the precipitator. Due to the exponential
dependence of precipitator efficiency on particle migration
velocity, plate collection area, and gas volume flow rate, the
effect of rapping reentrainment on the size of a precipitator
required for a given efficiency is substantial. Estimates of
the effect of rapping reentrainment on precipitator size are
also tabulated in Tables 17 and 18 for various conditions.
Two approaches were used to generate the tabulated values.
One approach was to assume that the percentage penetration due
to rapping was independent of the size of the unit. This
assumption, although not strictly true, may fairly represent
real situations. According to the data obtained during the
experiments at FluiDyne, the rapping puff concentration has a
significant vertical gradient and the particle size distri-
butation of the rapping puff consists of large particles.
Thus, in a multi-section precipitator the rapping puffs from
upstream sections may be recollected in the lower portion of
a succeeding section and later transferred from this lower
position into the hoppers without making any significant con-
tibution to outlet emissions. In this case, outlet fields
would contribute the largest portion of the rapping emission
140
-------
Table 17. EFFECT OF RAPPING REENTRAINMENT ON EFFICIENCY FOR RAPPING LOSSES
ACCOUNTING FOR 20%, 30%, 40% AND 50% OF EMISSIONS FOR VARIOUS OVERALL EFFICIENCIES
(first approach)
SCA* required
Assumed Penetration
percent of Due to
total pene- rapping
tration due reentrain-
to rapping ment,
reentrainment %
3.75
2.25
50 0.75
0.25
0.05
3.00
1.80
40 0.60
0.20
0.04
2.25
1.35
30 0.45
0.15
0.03
1.50
0.90
20 0.30
0.40
0.08
Without
rapping
reentrain-
ment. Total
3.75
2.25
0.75
0.25
0.05
4.50
2.70
0.90
0.60
0.06
5.25
3.15
1.05
0.35
0.07
6.00
4.60
1.20
0.10
0.02
7.50
4.50
1.50
0.50
0.10
7.50
4.50
1.50
0.50
0.10
7.50
4.50
1.50
0.50
0.10
7.50
4.50
1.50
0.50
0.10
Assumed
Overall
Efficiency
n
92.50
95.50
98.50
99.50
99.90
92.50
95.50
98.50
99.50
99.90
92.50
95.50
98.50
99.50
99.90
92.50
95.50
98.50
99.50
99.90
Efficiency
without
rapping re-
entrainment ,
riK, %
96.25
97.75
99.25
99.75
99.95
95.50
97.30
99.10
99.70
99.94
94.75
96.85
98.95
99.65
99.93
94.00
96.40
98.80
99.90
99.92
For n without
Increase in rapping re-
r| without entrainment,
rapping re- (ft2/
entrainment, m2/ 1000 ft3/
% m3/sec sec)
3.90
2.31
0.76
0.25
0.05
3.24
1.88
0.61
0.20
0.04
2.43
1.41
0.46
0.15
0.03
1.62
0.94
0.30
0.10
0.02
26.2
41.8
68.9
26.2
41.8
68.9
26.2
41.8
68.9
26.2
41.8
68.9
(133)
(212)
(350)
(133)
(212)
(350)
(133)
(212)
(350)
- (133)
(212)
(350)
For n,R or for n Additional SCA
with rapping required because
reentrainment, of reentrainment Addi-
(ft2/ (ft2/ tional
' m2/ 1000 ft3/ m2/ 1000 ft3/ SCA,
m3/sec sec) m3/sec sec) %
16.2
21.3
35.5
53.2
81.7
18.7
32.9
50.2
78.8
30.5
47.3
75.8
29.2
68.9
73.3
( 82)
(108)
(180)
(270)
(415)
( 95)
(167)
(255)
(400)
(155)
(240)
(385)
(148)
(350)
(372)
9.3
11.4
12.8
6.7
8.5
9.8
•••
4.5
5.5
6.9
3.0
3.5
4.3
(47)
(58)
(65)
(34)
(43)
(50)
(23)
(28)
(35)
(15)
(18)
(22)
353
27.4
18.6
25.6
20.3
14.3
16.5
13.2
10.0
11.3
8.5
6.3
•Specific Collection Area (SCA) based on data given by Gooch, et al.
(Figure 26 computed performance curves at
nA/cm2).
-------
Table 18. EFFECT OF RAPPING REENTRAINMENT ON PERFORMANCE FOR VARIOUS REENTRAINMENT
PERCENTAGES AS A FUNCTION OF NUMBER OF SECTIONS AND A CONSTANT COLLECTION EFFICIENCY PER SECTION.
Assumed
percent of
collected
material
per sec- Number
tion re- of
entrained Sections
1
5 9 2
3
4
1
2 3 2
3
4
1
0.97 2
3
4
SCA* Needed
Penetration
Due to
rapping
reentrain-
ment ,
%
6.2
1.02
0.136
0.0193
2.23
0.28
0.027
0.0023
0.90
0.10
0.009
0.0007
Without
rapping
reentrain-
ment.
%
5.2
0.27
0.014
0.0007
5.20
0.27
0.014
0.0007
5.20
0.27
0.014
0.0007
Total
%
11.4
1.29
0.150
0.020
7.43
0.55
0.041
0.0030
6.1
0.37
0.023
0.0014
Percent of
penetration
due to rap-
ping reen-
trainment.
%
52
79
91
97
30
51
66
77
15
27
39
50
Efficiency
n.
%
88.6
98.7
99.85
99.98
92.57
99.45
99.959
99.997
93.9
99.63
99.97
99.9986
Assumed
efficiency
HR without
rapping re-
entrainment.
%
94.8
99.73
99.986
99.9993
94.80
99.73
99.986
99.9993
94.8
99.73
99.986
99.9993
Increase
Without rapping
in n with reentrainment to
no rapping obtain n
reentrain-
ment.
%
6.54
1.03
0.14
0.02
2.35
0.28
0.03
0.002
0.95
0.10
O.O16
0.001
m2/
m3/sec
28.0
62.1
98.5
40.4
83.7
108.5
46.3
91.0
108.3
(ft2/
1000 ft3/
sec)
(142)
(315)
(500)
(205)
(425)
(550)
(235)
(462)
(550)
With
rapping
reentrainment
m2/
m3/sec
51.6
105.4
108.3
51.6
105.4
108.3
51.6
105.4
108.3
(ft2/
1000 ft3/
sec)
(262)
(535)
(550)
(262)
(535)
(550)
(262)
(535)
(550)
Additional SCA
mV
ra3/sec
23.6
43.3
11.2
21.7
5.3
14.4
(ft2/
1000 ft3/
sec)
(120)
(220)
( 57)
(110)
( 27)
( 73)
Addi-
tional
SCA,
%
84
69
28
25
11
16
*Specific Collection Area (SCA) based on data given by Gooch, et al.
(Figure 26 computed performance curves at 20 nA/cm).
-------
and the percent emission due to rapping should be independent
of the size of the unit. Estimates based on this approach are
tabulated in table 17.
1 8
The second procedure was based on the development by Francis,
which was discussed in the introduction. This approach assumes
that the fraction of collected dust per section reentrained
and emitted is the same for every section, that the between-
rap sectional efficiency is constant from one section to
another, and that the recollection efficiency for the rapping
puffs is the same as the between rap sectional efficiency.
According to Francis1 procedure, the penetration of any down-
stream section is equal to the penetration of the first
section (Pi) raised to the power of the number of preceding
sections plus one. Thus, the penetration of the N section (PN)
is given by:
PN = (Pi)N (8)
Using this equation, the efficiencies of a precipitator having
1, 2, 3, or 4 sections with a between-rap sectional efficiency
of 94.8% were calculated for percent penetrations of the first
section due to rapping reentrainments of 52%, 30%, and 15%.
These percentages correspond to 5.9%, 2.3%, and 0.97% of the
collected material per section being reentrained and emitted
from a section. These values were chosen to correspond with
those obtained during the FluiDyne experiments. The results
of the calculations based on this procedure are tabulated in
Table 18. According to this second procedure for the con-
ditions corresponding to the worst rapping reentrainment case
at FluiDyne, the contribution of rapping reentrainment to
total emissions would be greater than 90% for a three or four
section precipitator with 94.8% sectional efficiency. For
143
-------
a case in which reentrainment accounts for only 15% of the
emissions for the first section, a situation slightly better
than the best obtained during the FluiDyne experiments, rapping
reentrainment according to the Francis procedure would account
for 39% of emissions for a 3-section precipitator and 50% of
emissions for 4-section precipitator, if the precipitator had
a sectional efficiency of 94.8%. These calculations indicate
that, as the size of the precipitator increases, the percentage
contribution of rapping reentrainment to total emissions in-
creases. As previously discussed, this may not happen. Further
investigations with a multi-section precipitator are needed.
To obtain an indication of the effect of rapping reentrainment
on the sizing of a precipitator, estimates of the theoretical
specific collection area needed to obtain a given collection
efficiency for a particular set of conditions were obtained
from the model developed by Gooch, et al.1 For the example
case, a typical particle size distribution for an ash from a
coal-fired boiler plant was selected and a precipitator operat-
ing current density of 20 nA/cm2 was assumed (see the report
of Gooch, et al,1 for further details).
The theoretical specific collection area (SCA) required to
obtain a desired collection efficiency with rapping reentrain-
ment was compared with a theoretical SCA that would give the
same efficiency if there were no rapping emissions. This gave an
estimate of the increase in SCA needed to counterbalance the
rapping reentrainment emissions. The theoretical SCA needed
to obtain the desired efficiency with rapping reentrainment
was determined by first computing the efficiency that would be
obtained if there were no rapping reentrainment, by assuming
that either the percentage of emissions due to rapping re-
entrainment or the fraction of collected dust reentrained and
emitted from a section due to rapping were known. Once this
144
-------
efficiency was calculated and assuming that there were no
other nonidealities beside rapping reentrainment, the required
theoretical SCA was obtained from data given by Gooch, et al.
Then, again using the data given by Gooch, et al, the SCA was
determined that would give the desired efficiency if there were
no nonidealities such as rapping reentrainment. For the parti-
cular cases considered, the increase in required SCA due to
the rapping reentrainment ranged from 6% to 84% (see Tables 17
and 18). Thus, although the percentage decrease in efficiency
due to rapping reentrainment is relatively small, the effect
on the size required for a given efficiency can be significant.
The data obtained from the FluiDyne experiments indicates that
even moderate changes in rapping parameters can significantly
affect the size of a precipitator for a given efficiency.
EFFECT OF AGGLOMERATION AND FRACTIONAL REENTRAINMENT PERCENT-
AGES OF COLLECTED PARTICIPATE FOR VARIOUS PARTICLE SIZES
Modern electrostatic precipitator technology is concerned not
only with the overall collection efficiency but also with the
fractional collection efficiency,that is, the collection ef-
ficiency for various particle size ranges. The fraction of
collected material that is emitted due to rapping reentrain-
ment for various particle sizes is in the over simplified case
independent of particle size.
For a homogeneous mixture of discrete particles of various
sizes if one quarters out from the mixture a sample, theoreti-
cally the percentage of particles removed from the mixture
would be the same for all sizes. Rapping reentrainment does
not produce this simplified case. The particle size distribu-
tion of the discrete particles reentrained is not the same as
the particle size distribution of the individual particles
collected on the plates. Many of the discrete particles that
are reentrained are agglomerates of the individual particles
145
-------
Appendix I
Log of Precipitator Data, June 23, 1975
Test 1
Power Supply
Time
1040
1050
1100
1110
1120
1130
1140
1150
1200
1210
1220
1230
1240
1250
1300
1310
Water,
gal/min
4.5
4.5
4.5
4.5
4.6
4.6
4.6
4.6
4.6
4.6
4.6
4.6
4.6
4.6
4.6
Dust off
Avg . duct
Temp . @
inlet, °F
283
283
278
278
278
278
278
283
283
283
283
283
288
288
293
and on
Secondary
voltage
kV
36
38
39
39
39
39
39
39
39
40
38
38
39
39
39
Secondary
current
mA
43
43
43
43
42
42
42
42
42
42
24
24
32
32
32
152
-------
Appendix I
Log of Precipitator Data, June 23, 1975
Test 2
Power Supply
Time
1320
1330
1440
1450
1500
1510
1520
1530
1540
1600
1610
1620
1630
1640
1650
1700
1710
Water,
gal/min
4.6
4.6
4.6
4.6
4.6
4.6
4.1
5.0
5.5
5.5
5.5
5.5
5.5
5.5
5.5
5.5
5.5
Avg. duct
Temp . @
inlet, °F
293
293
303
308
308
308
308
293
278
273
273
273
273
263
263
263
263
Secondary
voltage
kV
38
37
37
37
37
37
37
40
40
40
40
38
38
39
39
42
42
Secondary
current
mA
23
24
24
24
24
24
24
34
34
34
34
16
16
17
17
32
32
153
-------
Appendix I
Log of Precipitator Data, June. 24, 1975
Test 3
Power Supply
Time
1110
1120
1130
1140
1150
1200
1210
1220
1230
1240
1250
1300
1310
1320
1330
Water ,
gal/min
6.0
6.0
6.0
6.0
6.0
5.8
Dust off
Dust on
5.8
Dust Off
5.8
Dust off
5.8
Dust off
5.8
Avg. duct
Temp . @
inlet, °F
250
250
250
258
258
258
263
and on
263
and on
263
and on
266
Secondary
voltage
kV
38
40
41
41
41
41
42
42
42
42
Secondary
current
mA
40
40
40
40
40
40
40
40
40
40
154
-------
Appendix I
Log of Precipitator Data, June 24, 1975
Test 4
Power Supply
Time
1500
1510
1520
1530
1540
1550
1600
1610
1620
1630
1640
1650
1700
Water,
gal/min
5.8
5.8
5.8
5.8
Dust off
5.8
5.8
5.8
Dust off
5.8
5.8
6.0
Dust off
Avg. duct
Temp . °
inlet, °F
258
263
263
263
and on
263
266
266
and on
266
266
260
Secondary
voltage
kV
42
44
43
43
43
43
43
43
43
44
Secondary
current
mA
31
41
40
41
40
40
39
40
40
40
155
-------
Appendix I
Log of Precipitator Data, June 25, 1975
Test 5
Power Supply
Time
0950
1000
1010
1020
1030
1040
1050
1100
1110
1120
1130
1140
1150
Water,
gal/min
5.7
5.8
5.8
5.8
5.8
5.5
5.5
5.7
5.7
5.7
5.7
5.7
5.7
Avg . duct
Temp . @
inlet, °F
256
256
263
263
263
268
268
268
258
258
258
258
262
Secondary
voltage
kV
40
40
40
40
40
40
41
41
42
42
42
40
42
Secondary
current
mA
40
40
40
40
40
40
40
40
40
40
40
40
40
156
-------
Appendix I
Log of Precipitator Data, June 25, 1975
Test 6
Power Supply
Time
1200
1210
1220
1350
1400
1410
1420
1430
1440
1450
1500
1510
1520
1530
1540
1550
1600
1610
1620
1630
1640
1650
1700
Water,
gal/min
5.7
5.7
5.7
5.7
5.8
5.9
5.9
6.0
6.0
6.0
6.0
6.0
6.0
5.9
5.9
5.9
5.9
5.9
5.9
5.9
5.9
5.9
5.9
Avg. duct
Temp . @
inlet, °F
262
262
262
274
264
264
264
264
264
264
264
264
264
265
266
266
266
266
267
268
268
269
269
Secondary
voltage
kV
42
40
40
43
43
43
43
44
44
42
42
43
44
44
44
44
44
44
44
44
44
44
44
Secondary
current
mA
40
40
40
40
40
40
40
40
40
36
39
37
40
38
40
40
40
40
38
38
38
38
38
157
-------
Appendix I
Log of Precipitator Data, June 26, 1975
Test 7
Power Supply
Time
0950
1000
1010
1020
1030
1040
1050
1100
1110
1120
1130
1140
1150
1200
1210
Water,
gal/min
5.9
5.9
5.9
5.9
5.9
5.9
5.9
5.9
5.9
6.0
6.0
6.0
6.0
6.0
6.0
Avg. duct
Temp . @
inlet, °F
257
257
257
258
258
258
264
264
267
267
267
261
269
269
269
Secondary
voltage
kV
39
40
40
40
41
41
41
41
42
42
42
43
41
Secondary
current
mA
40
44
44
41
41
41
41
41
40
40
40
41
41
158
-------
Appendix I
Log of Precipitator Data, June 26, 1975
Test 8
Power Supply
Time
1220
1230
1240
1250
1300
1310
1420
1430
1440
1450
1500
1510
1520
1530
1540
1550
1600
1610
1620
1630
1640
1650
1700
1710
1720
1730
Water,
gal/min
6.0
6.0
6.0
6.0
6.0
6.0
6.0
6.0
6.0
6.0
6.0
6.1
6.1
6.1
6.1
6.1
6.1
6.1
6.1
6.1
6.1
6.1
6.1
6.1
6.1
Avg. duct
Temp . @
inlet, °F
269
269
269
269
269
269
264
265
265
265
265
267
267
267
264
264
264
264
264
264
264
264
264
264
264
Secondary
voltage
kV
43
43
43
43
43
43
43
43
43
43
43
43
44
44
44
44
44
44
45
45
45
43
44
45
45
Secondary
current
mA
40
40
41
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
Test stopped
159
-------
Precipitator Ash Loading Data
Tests 1 and 2
Time
Test 1 1050
1108
1126
1139
1151
1155
1210
1230
1247
1300
1310
Test 2 1333
1440
1500
1525
1547
1608
1625
1645
1700
1715
Actual
Feed
Rate
(Ibs) (Ibs/hr) Comments
930 348
850
760
700
900
875
800
675
560
500
440 Dust off (13:04-15:25) -
44Q reload hopper
1300
1300
1300
1200 407
1075
950
825
725
640 Dust off
160
-------
Precipitator Ash Loading Data
Tests 3 and 4
Time
Test 3 1116
1133
1144
1155
1203
1216
1232
1250
1308
1327
Test 4 1500
1514
1527
1546
1600
1615
1627
1649
1658
(Ibs)
1300
1300
1280
1225
1160
1140
1075
1000
925
875
1175
1090
1010
950
875
760
740
600
Actual
Feed
Rate
(Ibs/hr)
332
338
335
369
342
380
358
Dust
Dust
Dust
Dust
Dust
Dust
Dust
Dust
Dust
Dust
on
off
off
off
off
off
off
off
off
off
Comments
(12:05-12
(12:25-12
(12:42-12
(13:01-13
(13:18-13
(13:37-14
(15:37-15
(16:17-16
:10)
:30)
:48)
:07)
:25)
:56)
:44)
:24)
161
-------
Precipitator Ash Loading Data
Test 5
Time
Test 5 0951
1005
1014
1025
1039
1043
1051
1104
1113
1122
1126
1134
1143
1148
1155
1207
(Ibs)
1225
1130
1080
1035
970
925
925
860
860
775
750
750
700
680
680
620
Actual
Feed
Rate
(Ibs/hr)
358
352
340
326
327
400
Comments
Dust off (10:25-10:31)
Dust off (10:43-10:51)
Dust off (11:04-11:13)
Dust off (11:26-11:34)
Dust off (11:48-11:55)
Fill hopper
Dust off
162
-------
Precipitator Ash Loading Data
Test 6
Time
Test 6 1356
1407
1414
1423
1432
1442
1446
1455
1504
1527
1533
1542
1549
1559
1607
1620
1626
1634
1642
1648
1655
1700
(Ibs)
1200
1150
1100
1050
990
930
920
920
870
720
720
670
630
560
520
520
480
430
380
350
310
270
Actual
Feed
Rate
(Ibs/hr) Comments
362 Fill hopper
Dust off (12:07-13:56)
Dust off (14:46-14:55)
357
Dust off (15:27-15:33)
358
Dust off (16:07-16:20)
360
Dust off
163
-------
Precipitator Ash Loading Data
Test 7
Time
Test 7 1009
1027
1048
1657
1127
1151
1200
1208
1224
1238
1246
1257
1308
(Ibs)
1100
980
870
810
660
520
470
420
360
280
230
170
100
Actual
Feed
Rate
(Ibs/hr)
341
350 Dust
Dust
356
Comments
off (11:11-11:15)
off (12:08-12:15)
164
-------
Precipitator Ash Loading Data
Test 8
Time
Test 8 1426
1447
1457
1506
1515
1523
1535
1544
1552
1613
1617
1624
1634
1643
1651
1707
1716
1724
1730
Actual
Feed
Rate
(Ibs) (Ibs/hr) Comments
1250 382 Fill hopper
Dust off (13:08-14:26)
1120
1060
1000
950
900
820
760
700
590 420 Dust off (15:52-16:00
560
520
450
400
340
280 350 Dust off (16:51-16:59)
220
160
Dust off
165
-------
TABLE OF CONVERSION FACTORS
MASS
1 grain = 0.0648 grams
1 gram = 15.4324 grains
AREA
1 ft2 = 0.0929 m2
1m2 =10.76 ft2
VOLUME
1 ft3 = 0.0283 m3
1 m3 = 35.31 ft3
PRESSURE
1 kPa = 0.1451 Psi
1 Pa =1 N/m2
1 Psi = 6.894 kPa (kilopascal)
ACCELERATION
1 G = 9.8 m/sec2
VELOCITY
1 m/sec = 0.3048 ft/sec
1 ft/sec = 3.28 m/sec
166
-------
TECHNICAL REPORT DATA
(Please read Inunctions on the reverse before completing)
1. REPORT NO.
EPA-600/2-76-140
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
Rapping Reentrainment in a Nearly Full-Scale
Pilot Electrostatic Precipitator
5. REPORT DATE
May 1976
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Herbert W. Spencer, III
8. PERFORMING ORGANIZATION REPORT NO.
SORI-EAS-76-061
3489-V
9. PERFORMING OROANIEATION NAME AND ADDRESS
Southern Research Institute
2000 Ninth Avenue South
Birmingham, Alabama 35205
10. PROGRAM ELEMENT NO.
1AB012; ROAP 21ADL-027
11. CONTRACT/GRANT NO.
68-02-1875
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final; 3/75-1/76
14. SPONSORING AGENCY CODE
EPA-ORD
15. SUPPLEMENTARY NOTES ffiRL-RTP Project Officer for this report is Leslie E. Sparks,
Mail Drop 61, Ext 2925.
16. ABSTRACT
The report gives results of an experimental investigation of rapping
reentrainment in a nearly full-scale pilot electrostatic precipitator. The study
included a fundamental examination of the mechanics of removal of dry dust by
rapping and the quantification of rapping reentrainment in terms of the percentage
and the particle size distribution of the reentrained dust. During the study, the
contribution of rapping reentrainment to total emissions ranged from 53 to 18%,
depending on rapping conditions. These percentages corresponded to 5. 4 and 2. 7%
of the dust collected on the plates being emitted from the precipitator during plate
A major portion of the reentrained material was observed to result from
hopper 'boil-up.' A decrease in rapping emissions was obtained by increasing the
time interval between raps. The mass median diameters of the size distributions
of the particles emitted during the raps were observed to increase with increased
time between raps.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Air Pollution
Electrostatic Precipitators
Dust
Air Pollution Control
Stationary Sources
Rapping
Reentrainment
Particulate
13B
11G
8. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
178
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
167
-------