EPA-600/2-76-154b
July 1976
A.P.S.
ELECTRO-TUBE
EVALUATION
by
Seymour Calvert, Christian Chris tensen, and Charles Lake
Air Pollution Technology, Inc.
4901 Morena Boulevard, Suite 402
San Diego, California 92117
Contract No. 68-02-1496
ROAPNo. 21ADL-004
Program Element No. 1AB012
EPA Project Officer: Dale L. Harmon
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
iBRARY
-------
ABSTRACT
Fine particle collection efficiency as a function of
particle size has been computed from data taken on an Air
Pollution Systems, Inc. (A.P.S.) Electro-Tube.
The Electro-Tube was operated at gas flow rates of
16.9, 18.9, and 22.9 Am3/min at ambient conditions. Titanium
dioxide was generated as the test aerosol having an approxi-
mate mass mean aerodynamic diameter of 1.2 ymA with a geo-
metric standard deviation of 2.2. Test results indicated
that the overall collection efficiency was 96.91 for the
high gas flow rates, 98.2% for the medium gas flow rates,
and 99.3% for the low gas flow rates.
111
-------
TABLE OF CONTENTS
Page
Abstract iii
List of Figures v
List of Tables vii
List of Abbreviations and Symbols viii
Sections
Conclusions 1
Recommendations 2
Introduction 3
Manufacturer's Description of Device 4
Source and Control System 6
Test Method 9
Conditions for Runs 12
Operating Conditions 14
Cascade Impactor Particle Data 15
Diffusion Battery Data 18
Particle Penetration 20
Computation Method 20
References 25
Appendix 26
IV
-------
LIST OF FIGURES
No. Page
1 Diagram of A.P.S. Electro-Tube 7
2 Schematic Diagram of Particle Generator. ... 8
3 Modified E.P.A. Sampling Train with
In-Stack Cascade Impactor 10
4 Penetration versus Aerodynamic Particle
Diameter for Low Gas Flow Runs 15, 16, 19. . . 21
5 Penetration versus Aerodynamic Particle
Diameter for Medium Gas Flow Runs 7, 10,
12, 13, 14 22
6 Penetration versus Aerodynamic Particle
Diameter for High Gas Flow Runs 3, 4,
5, 18 23
A-l Inlet and Outlet Size Distribution
for Run 3 27
A-2 Inlet and Outlet Size Distribution
for Run 4 27
A-3 Inlet and Outlet Size Distribution
for Run 5 28
A-4 Inlet and Outlet Size Distribution
for Run 7 28
A-5 Inlet and Outlet Size Distribution
for Run 10 29
A-6 Inlet and Outlet Size Distribution
for Run 12 .29
A-7 Inlet and Outlet Size Distribution
for Run 13 30
A-8 Inlet and Outlet Size Distribution
for Run 14 , 30
A-9 Inlet and Outlet Size Distribution
for Run 15 .31
v
-------
LIST OF FIGURES (Continued)
No.
A-10 Inlet and Outlet Size Distribution
for Run 16 31
A-ll Inlet and Outlet Size Distribution
for Run 18 32
A-12 Inlet and Outlet Size Distribution
for Run 19 32
A-13 Diffusion Battery Size Distribution
for Inlet and Outlet Run 17 33
A-14 Diffusion Battery Size Distribution
for Inlet and Outlet Run 18 33
A-15 Diffusion Battery Size Distribution
for Inlet and Outlet Run 19 34
VI
-------
LIST OF TABLES
No. Page
1 Summary of Test Runs 13
2 Operating Conditions 14
3 Size Distribution, Mass Loading, and
Overall Penetration Data 16
4 Number Basis Size Distribution Data for
Diffusion Battery Tests 19 •
A-
A-
A-
A-
A-
A-
A-
A-
A-
A-
A-
A-
A-
1
2
3
4
5
6
7
8
9
10
11
12
13
Total Filter Particle
Inlet and
for Run 3 .
Inlet and
for Run 4 .
Inlet and
for Run 5 .
Inlet and
for Run 7.
Inlet and
for Run 10
Inlet and
for Run 12
Inlet and
for Run 13
Inlet and
for Run 14
Inlet and
for Run 15
Inlet and
for Run 16
Inlet and
for Run 18
Inlet and
for Run 19
Outlet
Outlet
Outlet
Outlet
Outlet
Outlet
Outlet
Outlet
Outlet
Outlet
Outlet
Outlet
Sampl
Sampl
Sampl
Sampl
Sampl
Loading Tests. . . .
e
e
e
e
e
Sample
Sampl
Sampl
Sampl
Sampl
e
e
e
e
Sample
Sampl
e
Particle
Particle
Particle
Particle
Particle
Particle
Particle
Particle
Particle
Particle
Particle
Particle
Data
Data
Data
Data
Data
Data
Data
Data
Data
Data
Data
Data
• • 35
• • 36
• • 36
• • 36
• • 36
• • 37
• • 37
• - 37
• • 37
• • 38
• • 38
- • 38
• • 38
VI1
-------
LIST OF ABBREVIATIONS AND SYMBOLS
Latin
d - particle diameter, ym
d - aerodynamic particle diameter, ymA
d - aerodynamic cut diameter, pmA
d - geometric mean particle diameter, ym or ymA
r &
d - diameter of particle collected with 50% efficiency,
PSO ymA
Greek
ymA - aerodynamic diameter, ym(g/cm3) 2
p - particle density, g/cm3
a - geometric standard deviation of particle size
^ distribution
Vlll
-------
CONCLUSIONS
This evaluation was one of a series of such evaluations
being conducted by the Industrial Environmental Research La-
boratory of the Environmental Protection Agency to identify
and test novel devices which are capable of high efficiency
collection of fine particulates. The test methods used were
not the usual compliance-type methods but were, rather, state-
of-the-art techniques for measuring efficiency as a function
of particle size using total filters, cascade impactors, and
diffusion battery tests upstream and downstream from the
A.P.S. Electro-Tube control device.
Experimental tests of the A.P.S. Electro-Tube were done
with an aerosol of Ti02 with a mass mean aerodynamic diameter
of 1.2 ymA and a standard deviation of 2.2. Overall collec-
tion efficiencies ranged from 96.9% for high gas flow rates
(22.9 Am3/min) to 99.31 for low gas flow rates (16.9 Am3/min),
Experimentally determined penetration for 0.5 ymA aero-
dynamic diameter particles was 8.6% for high gas flow, 3.7%
for medium gas flow, and 1.1% for low gas flow. For 1.0 ymA
aerodynamic diameter particles the penetrations were 4.3% for
high gas flow, 2.27% for medium gas flow, and 0.68% for low
gas flow. For 2.0 ymA aerodynamic diameter particles the
penetrations were 0.22% for high gas flow, 0.39% for medium
gas flow, and 0.38% for low gas flow.
The Electro-Tube performance was found to be similar to
that which can be achieved in small wet electrostatic pre-
cipitators.
The A.P.S. Electro-Tube was tested using impactors with
no charge neutralizers preceding them. The statically
charged particles could have affected the performance of
the impactors both on the outlet and the inlet sample points.
-------
RECOMMENDATIONS
For operations where wet electrostatic precipitators
are practical, the Electro-Tube should be considered as
an alternative which may represent a savings in power
consumption at equivalent collection efficiency.
-------
INTRODUCTION
Air Pollution Technology, Inc. (A.P.T.), in accordance
with EPA Contract No. 68-02-1496, "Experimental Tests of
Novel Fine Particulate Collection Devices," conducted a
performance evaluation on Air Pollution Systems, Inc.
Electro-Tube system. The objective of the performance
test was to determine fine particle penetration as a
function of aerodynamic particle size and control device
parameters.
Simultaneous inlet-outlet particle sampling measure-
ments were taken for high, manufacturer's maximum, and
two lower gas flow rates during the testing period of
July 9 to 16, 1975. The maximum design flow used by
A.P.S. for a twelve-inch tube was 22.9 Am3/min.
In-stack cascade impactors, total filters, condensa-
tion nuclei counters and a portable diffusion battery
were used to obtain mass loadings and size distribution
data.
The results of this series of tests on the Air Pol-
lution Systems Electro-Tube are presented in the text.
-------
MANUFACTURER'S DESCRIPTION OF DEVICE
Air Pollution Systems has developed a two-stage elec-
trostatic device called an "ELECTRO-TUBE" which combines
a wetted wall electrostatic precipitator with hi-intensity
particle precharging. The saturation charge on the par-
ticle is increased substantially from the normal 4-5 kV/cm
field in a conventional precipitator to 12 kV/cm by first
passing the gases through the Hi-intensity Ionizer.
The increased electrostatic charge allows a more
effective migration of the fine particles in the collec-
ting electric field of the precipitator. The collecting
precipitator, a wetted wall pipe type, has a passive high-
voltage electrode which emits no corona current and
operates at an average applied field of 5 to 10 kV/cm.
Tests have shown a high degree of stability in the elec-
tric fields up to a gas flow velocity of 6.1 m/sec (20 FPS)
Despite the shorter residence time in the charging field,
there is no apparent deterioration of particle charging
efficiency.
A single power supply provides hi-voltage to both
ionizer charger and collector sections with a total power
consumption being less than 8.5 W/(m3/min). Pressure
drop through the ELECTRO-TUBE is less than 0.8 cm W.C.
Since only a small amount of stack heat is transferred to
the wetted walls, the ELECTRO-TUBE wet collection process
does not quench the gas stream. Water utilization rate
ranges between 134 and 268 liters/1000 m3/roirt depending
on inlet dust loading and degree of prequenching desired.
The ELECTRO-TUBE is designed so that entrained water does
not affect electric field stability. The unit is self de-
misting as any entrained water droplets are collected as
fine particulate.
-------
Installed cost for an ELECTRO-TUBE should be less
than a conventional wet pipe ESP since higher flow volumes
can be treated in the same size unit at higher collection
efficiencies. Operating costs should be somewhat lower
than a wet pipe ESP in that less electrical power is used
with recycling expense and pressure drop being about the
same.
-------
SOURCE AND CONTROL SYSTEM
The pilot scale Electro-Tube of Air Pollution Systems
(A.P.S.) is basically a tube electrostatic precipitator
with a central rod electrode and wetted wall collector.
Figure 1 is a schematic diagram of the pilot system. The
inlet particles are charged in a high energy field (12 kV/cm)
by a high intensity ionizer at the base of the electrode.
The charged particles then migrate to the wetted wall in
the body of the device in a field of 5-10 kV/cm. A.P.S.
indicates that initial saturation charge on the particles
is higher than the usual 4-5 kV/cm for a conventional ESP
(electrostatic precipitator) and facilitates increased
migration in the collecting electric field.
Under testing conditions the Electro-Tube was operated
at gas flow rates of 16.9, 18.9 and 22.9 Am3/min (596, 668,
and 808 ACFM, respectively) at 20°C. Liquid flow rates
could range from 1.7 to 6.4 £/min according to the manu-
facturer's specifications, but the liquid flow was held
constant at 3.8 £/min during the entire test period.
The test aerosol for this study was titanium diox-
ide (Ti02) which had a single particle density of 4.1 g/cm3.
The mass median aerodynamic diameter of the dispersed aero-
sol was about 1.2 ymA, geometric standard deviation 2.2,
and agglomerate density about 3.0 g/cm3.
Figure 2 is a schematic diagram of the particle gene-
rator used during the test which consists of a feed auger,
intermediate blower, deagglomeration orifice chamber, and
main blower. A cross-shaped baffle was placed at the system
aerosol inlet to insure adequate mixing of the aerosol with
the incoming ambient air.
-------
FROM HIGH
VOLTAGE SOURCE
OPTIONAL
SECONDARY
AND TERTIARY
IONIZATION
ZONES
AEROSOL
INLET
AEROSOL OUTLET
TO BLOWER
WATER
INLET
ELECTRODE
•BODY
HIGH INTENSITY
IONIZER SECTION
I]
-TANK
u
OUTLET
WATER DRAIN
Figure 1, Diagram of A.P.S. Electro-Tube
7
-------
l-l Pi
P-1
(D
c
rt
E
0)
xi
o
CO
CN)
0)
S-.
3
00
-------
TEST METHOD
The performance characteristic of the A.P.S. Electro-
Tube was determined by measuring the particle size dis-
tribution and mass loadings of the inlet and outlet gas
streams simultaneously.
All tests were performed using modified EPA type
sampling trains with in-stack University of Washington
Mark III (U.W.) cascade impactors. Figure 3 is a schematic.
diagram of the sample train. Greased aluminum substrates
were used in the impactors to prevent particle bounce and
minimize wall losses.
The Air Pollution Technology portable screen diffu-
sion battery (A.P.T. - S.D.B.) was used for particle
measurements from 0.1 ym to 0.01 pm (actual).
The A.P.T. - S.D.B. uses Brownian diffusion to accom-
plish the size fractionation of particles smaller than
approximately 0.1 ym. Because smaller particles diffuse
more readily than larger ones, successively larger parti-
cles are captured as they proceed through the battery.
Using a condensation nuclei counter (CNC) to deter-
mine the total number of particles at various points in
the battery, one can obtain data which will correspond to
a unique size distribution. The size distribution compu-
tation is based on calibration of the S.D.B. in the A.P.T.
laboratory.
Computation of size distributions from the data on
penetration is accomplished through the use of measure-
ments of density, pressure, temperature, flow rate, and
moisture, in addition to the CNC calibration factors.
During an impactor run, several inlet and outlet
fine particle size measurements were taken alternately
with the portable diffusion battery. Since the system
-------
S-4
o
-(-I
o
nJ
P.
0)
T3
rt
u
in
rt
u
o
rt
t/>
£
-P
•H
•H
05
60
P!
•H
(D
•H
M-i
•H
to
-------
remained fairly constant during each run, alternate inlet
and outlet S.D.B. measurements were considered to approxi-
mate simultaneous sampling.
In-stack filter samples were also taken to obtain
total particulate loadings and overall collection effi-
ciencies of the system.
Impactor blank runs were performed periodically to
insure that, the greased aluminum substrates did not react
with the stack gases. A blank impactor run consisted of
an impactor preceded by two glass fiber filters run at
identical sample conditions as the actual sampling runs.
Total filter loadings were also obtained during the blank
impactor run to furnish simultaneously inlet and outlet
mass concentrations and overall collection efficiency data.
Gas flow rates for all tests were determined by means
of a calibrated standard-type pitot tube for velocity head
measurement along with in-stack taps for continuous wet
and dry bulb temperature measurements. Velocity traverses
of the inlet and outlet were performed according to EPA
standards and an average velocity point was selected for
one point sampling. Sample f]ow rates were measured with
the usual EPA train instruments so as to obtain isokinetic
sampling.
The inlet sample port was located twelve duct dia-
meters downstream from the dust feeder and five diameters
upstream of the nearest disturbance. The outlet port was
located eleven duct diameters downstream from the nearest
disturbance and six diameters upstream of the blower inlet.
Velocity traverses of both the inlet and the outlet gas
streams revealed fully developed flow profiles.
11
-------
CONDITIONS FOR RUNS
A total of nineteen simultaneous inlet and outlet
sample runs was performed on the Electro-Tube, Three sets
of gas flow rates were tested during the evaluation period.
Diffusion battery data were taken during the last three
days of testing.
Table 1, "Summary of Test Runs," is a summary of the
characteristics of each test, gas flow rates and sample
devices used (impactor, filter, diffusion battery). Runs
1-6, 8-9, and 18 were taken during high gas flow conditions
of 22.9 Am3/min. Runs 7, 10-14, and 17 were taken during
medium gas flow conditions of 18.9 Am3/min. Low gas flow
conditions, 16.9 Am3/min, were tested during runs 15, 16,
and 19. To insure a representative dust sample, a mixing
baffle was inserted at the aerosol inlet prior to run 9
and left in place for the remainder of the test period.
A power failure in the pilot plant building resulted
in a loss of run 11. The cascade impactor data for run 17
were eliminated due to failure of a sample train vacuum
pump while the run was under way.
12
-------
Table 1. SUMMARY OF TEST RUNS
Date
7/9/75
7/9/75
7/9/75
7/9/75
7/10/75
7/10/75
7/10/75
7/10/75
7/11/75
7/11/75
7/11/75
7/11/75
7/14/75
7/14/75
7/14/75
7/15/75
7/15/75
7/15/75
7/16/75
Run #
1
2
3
4
5
6
7
8
9
10
11(1)
12
13
14
15
16
17(2)
18
19
Run Type
Total Filter
Total Filter
Tmpactor
Impactor
Impactor
Total Filter
Impactor
Total Filter
Total Filter
Impactor
Impactor
Flow, Am3/min
22.9
22.9
22.9
22.9
22.9
22.9
18.9
22.9
22.9 Mixing baffle at
aerosol inlet
18 .9 Mixing baffle at
aerosol inlet
18.9 Mixing baffle at
aerosol inlet
Impactor J 18.9 Mixing baffle at
Impactor
Impactor
Impactor
Impactor
aerosol inlet
18.9 Mixing baffle at
aerosol inlet
18.9 Mixing baffle at
aerosol inlet
16.9 Mixing baffle at
aerosol inlet
lt». 9 Mixing baffle at
aerosol inlet
Impactor, DB 18.9 Mixing baffle at
aerosol inlet
Impactor, DB 22.9 Mixing baffle at
! aerosol inlet
Impactor, DB 16.9 Mixing baffle at
aerosol inlet
(1) Power failure in pilot plant building - no data obtained,
(2) Vacuum train sample pump failure - no data obtained.
13
-------
OPERATING CONDITIONS
The operating conditions for the A.P.S. Electro-Tube
are tabulated in Table 2 shown below:
Table 2, OPERATING CONDITIONS
Condition
Inlet
Outlet
Temperature
Velocity in Electro-Tube
Am3/min 8 22.5°C
ACFM g 22.5°C
DNmVmin @ 0°C
DSCFM @ 21°C (70°F)
Vol % H20 Vapor
Static Pressure at
Sample Ports
18r27°C
4.3-5.8 m/sec
(14.1-19.1 FPS)
16.9-22.9
596.0-808.2
15.4-20.9
585.0-794.0
1.3-2.4
-0.5 to
-1.5 cm W.C.
18-27°C
4.4-6.0 m/sec
(14.5-19.9 FPS)
17.4-23.8
615.0-841.0
15.9-21.7
605.0-826.0
1.3-2.4
-2.5 to
-4.6 cm W.C.
Pressure drop across Electro-Tube: 0.53 (low flow) to
0.64 (high flow) cm W.C.
Ionizer wall water wash rate: 3.8 £/min (1 GPM)
14
-------
CASCADE IMPACTOR PARTICLE DATA
Data sets for three gas flow rates were obtained
for the A.P.S. Electro-Tube. As the gas flow was fully
developed one-point sample locations were used for all
data points.
Table 3 is a summary of the size distribution of
inlet and outlet mass loading, and overall penetration
data for the various runs made. Table A-l in the appen-
dix summarizes the total filter loading tests while
Tables A-2 to A-13 summarize inlet and outlet particle
data for the impactor runs. Figures A-l to A-12 are
plots of inlet and outlet size distributions for the
impactor runs. Figures A-13 to A-15 are plots of the
inlet and outlet distributions for the diffusion battery
data.
In this report, the symbol "d " refers to aero-
dynamic diameter, which is equal to the particle diameter
(d ) in microns (ym) times the square root of the particle
density (p ) in grams per cubic centimeter (g/cm3) times
the square root of the Cunningham slip correction factor
(C1). The symbol "vimA" represents the units of aero-
dynamic size.
dpa
The symbol "d " refers to the aerodynamic cut dia-
meter or the particle diameter in which the impactor stage
collection efficiency is 50%. The "d " may also refer
pc
to the aerodynamic cut diameter of the control device in
which, at the given particle diameter, the penetration
is equal to 0.50. The symbol "d " refers to the mass
r o
mean aerodynamic particle diameter for a given size dis-
tribution.
15
-------
Table 3. SIZE DISTRIBUTION, MASS LOADING, AND OVERALL
PENETRATION DATA*
Run
No.
1
2
3
4
5
6
7
8
9
10
12
13
14
15
16
18
19
INLET
d ,ymA
Pg'
Filt
a
g
.er
Filter
1.02
1.15
1.47
2.1
2.1
2.4
Filter
1.55
2.4
Filter
Filter
1.30
0.87
1. 27
1.11
1.05
1.20
1.25
1.20
2.2
2.0
2.1
2.2
2.0
2.5
2.2
1.9
mg/DNm3
177
242
185
248
571
775
1080
606
662
364
167
389
248
375
738
274
240
OUTLET
d , ymA
Pg'
Fill
a
g
ber
Filter
0.69
0.71
0.70
1.4
1.5
1.5
Filter
0.68
Fir
1.5
ter
Filter
0.82
0.69
0.75
0.74
0.68
0. 79
0.66
1.00
1.7
1.6
1.9
1.4
1.8
1.8
1.5
2.2
mg/DNm3
8.7
11.0
8.5
8.7
13.0
14.5
13.7
19.0
9.1
5.0
4.1
7. 0
4.2
2.5
5.8
7.8
1.2
Overall
Penetration
(%)
4.9
4.6
4.6
3.5
2.3
1.9
1.3
3.1
1.4
1.4
2. 5
1.8
1.7
0.7
0.8
2.9
0.5
*d is the geometric mass mean aerodynamic particle
P^ diameter in this table.
16
-------
Isokinetic or near isokinetic sampling was undertaken
for all the test runs; however, isokinetic conditions are
not that crucial for sampling fine particles. For example,
the error caused by sampling 4 ymA particles at a velocity
501 higher or lower than the gas stream velocity would
only be about 2 or 3% of the concentration.
Single point sampling is also generally sufficient
when measuring fine particle size and concentration. The
fine particles will be distributed well in the gas stream,
except in cases where streams with different particle con-
centrations have not had sufficient time to mix. To
illustrate that one point sampling is sufficient for fine
particles, we may note that Stokes stopping distance of a
3 ymA particle with an initial velocity of 15 m/sec
(50 ft/sec) is about 0.04 cm (0.016 inches) and for a
1 ymA diameter particle is one ninth of that. Since the
stopping distance is the maximum that a particle can be
displaced from a gas streamline by going around a right
angle bend, it becomes apparent that fine particle distri-
bution in the gas stream will be negligibly affected by
flow direction changes.
-------
DIFFUSION BATTERY DATA
Diffusion battery data were taken during the last
three days of testing. The runs were made simultaneously
with cascade impactors and numbered accordingly. Runs 14
and 16 were preliminary tests to determine count and flow
adjustment techniques; therefore, they were not included
in the data set.
Performances during runs 17, 18, and 19 were deter-
mined by statistical conversion of inlet-outlet size
distribution to cumulative number concentration and
calculating the particle penetrations in the same manner
as the cascade impactor analysis. Calculations were
performed by a computerized FORTRAN program.
Table 4 of the text contains the size distribution
summary of the diffusion battery tests. Note that the
sizes given are actual rather than aerodynamic as with
cascade impactor analysis. This is necessary when re-
ducing data from the diffusion battery as the particles
are evaluated in the actual size regime. Conversion to
aerodynamic diameter for comparison with impactor analysis
requires knowledge of the density.
The titanium dioxide powder used had a density of
4.1 g/cm3 as measured by a pycnometer. This value agreed
closely with published data for Ti02. The solid unitary •
particles would have this value for their density. However,
the batch of Ti02 that was used consisted of many agglo-
merated as well as single particles. Most of'the single
particles and the detectible units of the agglomerates had
diameters in the range of 0.1 to 0.3 pm as measured from
scanning electron micrographs. Therefore, particles larger
than about 0.3 ym were mainly agglomerates and, because of
their irregular shape and possible voids, had a density
less than 4.1 g/cm3. The density of the larger particles
was estimated to be about 3 g/cm3.
18
-------
Table 4. NUMBER BASIS SIZE DISTRIBUTION DATA FOR
DIFFUSION BATTERY TESTS*
Run
No.
17
18
19
INLET
dpg'ym
0.37
0.14
0.34
a
g
7.8
7.2
7.9
OUTLET
d ,um
Pg
0,12
0.035
0.19
°g
7.6
4.3
7.5
*d is the geometric number (count) mean
Pg particle diameter in this table.
Size distribution plots of diffusion battery runs
have been included in Figures A-13 through A-15. The
penetrations calculated from the diffusion battery data
were converted to penetrations corresponding to aerody-
namic diameter using both the 4.1 g/cm3 and 3 g/cm3 den-
sities, and are shown in Figures 4, 5, and 6 of the next
section. The lower density results in 5 to 15 percent
lower particle penetrations for the three runs shown.
19
-------
PARTICLE PENETRATION
The overall penetration summary is presented in
Table 3. Total mass loadings were taken by cascade im-
pactors or total filter samples. Overall penetrations
tended to decrease with decreasing volumetric gas flow
rates.
Average overall penetrations for the Electro-Tube
were 3.1% for high flow, 1.8% for medium flow, and 0.7%
for low flow. The average aerosol "d " was 1.2 ymA with
a geometric standard deviation of 2.2.
Particle penetration versus particle size were plotted
for the data obtained from the Electro-Tube and appear in
Figures 4-6. The data have been plotted together accor-
ding to the gas flow rate of the device with diffusion
battery penetration analysis included on each figure.
Particle density assumptions account for discrepancies
between impactor and diffusion battery penetrations.
COMPUTATION METHOD
The penetrations described above were calculated by
means of a machine computation program. This method
utilizes inlet and outlet particle size data which are
fitted to log-normal distributions. Next the size distri-
bution parameters and the mass concentrations are used to
compute penetration as a function of particle size. Dif-
fusion battery data yield penetration related to physical
size while cascade impactor data are in terms of aero-
dynamic size. In order to put the results on the same
basis, it is necessary to know the particle density so
that one can convert physical size to aerodynamic size
(or vice versa) .
Particle size distribution data are conveniently
represented on logarithmic probability graph paper. Often
20
-------
0.1
DIFFUSION
BATTERY
0.001
2.0 3.0
0.05
AEROYNAMIC PARHCLE DIAMETER, dpa,
flow,
21
-------
0.3
o
I—I
u
O
t—t
H
H
W
2
W
PL,
CJ
t— I
H
Pi
0.1
0.05
0.01
0.005
0.001
DIFFUSION
BATTERY
RUN 17
0.05 0.1
0.5
1.0
2.0 3.0
AERODYNAMIC PARTICLE DIAMETER, d , ymA
pa
Figure 5. Penetration versus aerodynamic
particle diameter for medium gas
flow, Runs 7, 10, 12, 13, 14
22
-------
1.0
0.5
o
i—i
H
U
P-,
0.1
0.05 --
0.01
0.005 _
0.001
DIFFUSION;
BATTERY ;
RUN 18
0.05
0.1
0.5 1.0 2.0 3.0
AERODYNAMIC PARTICLE DIAMETER, d . ymA
pa'
Figure 6. Penetration versus aerodynamic
particle diameter for high gas
flow, Runs 3, 4, 5, 18
23
-------
the data fit a straight line on log-probability paper, and
thus can be mathematically represented by a log-normal size
distribution.
Caution must be taken when using a mathematical ex-
pression for the penetration. First, the fit of the data
to a log-normal distribution must be close. Second, there
is uncertainty introduced by extrapolating data beyond the
size range measured. The penetration equation should not
be used for particle sizes smaller or larger than those
measured.
Previous calculations of penetrations have been done
by manually and visually determining the ratio of the
slopes of the outlet and inlet cumulative mass versus par-
ticle size curves at different particle sizes. Penetra-
tion versus particle size curves were drawn for the A.P.S.
Electro-Tube by the visual method and by the machine
method described above. The deviations in penetrations
between the two methods were generally less than ±0.01
or 1 percent penetration.
Within limits of the particle sizes measured, the
accuracy of the log-normal penetration equation depends
only on how well the data fit log-normal distributions.
Since the particle size data have many other inaccuracies
because of the difficulties of measurement, the log-normal
penetration is accurate enough, considering the ease with
which it can be used. It is also most advantageous to
eliminate the subjective errors possible with the visual
method.
24
-------
REFERENCES
Calvert, S., J. Goldshmid, D. Leith, and D. Mehta.
"Scrubber Handbook," A.P.T., Inc. Riverside, California.
EPA Contract No. CPA-70-95. August 1972. PB-213-016.
Calvert, S., "Engineering Design of Fine Particle Scrub-
bers," J. of A.P.C.A., 24_, No. 10, p. 929. October 1974.
Gooch, J., J. McDonald, and S. Oglesby. "A Mathematical
Model of Electrostatic Precipitators." Southern Research
Institute, Birmingham, Alabama. EPA Contract No. 68-02-
0265. April 1975.
Oglesby, S. and G. Nichols. "A Manual of Electrostatic
Precipitator Technology." Southern Research Institute,
Birmingham, Alabama. EPA Contract No. CPA 22-69-73.
August 1970.
25
-------
APPENDIX
26
-------
APPENDIX
26
-------
or:
la
Q
"
3
O
•a
Ci
«
•p -*
-------
E-i
Z
w
u
1/5
CO
T3
0>
•O
C
03
3
00
H
2
u
cri
la
28
-------
w
KJ
in
O
-
OS
HJ
OH
3
,0
^H C
C 3
3
BO
'B p 'HH13NVKI 313II«Vd
3
XI
rH C
C 3
I-H ai
3
bo
B
' p 'aaiawvia
29
-------
Q
2
a
2
PJ
U
oi
tq
rt
^j-
4-> rH
-------
O
•z.
u
C4
w
p '
C 3
1-1 n;
<
CD
31
-------
«
Nl
to
u
ci
a
4-1
<
0>
}-i
3
DIWVNAQOHHV
at
ia
a
z
2
HJ
-------
,0
•H
^ 00
c/3
K
tq
UJ
u
CK
m
a,
OS
>H 3
CD o
•M
C +J
o a>
•H i— *
w C
3 -H
•H O
Q m
3
H
fc
p 'HHiawvia
l-i I-
•P t-H
BJ
Q
z
uirt ' p '
1- 3
CD O
rt is
n n)
(3 -i-1
O d>
•H <—<
to C
3 -H
m
<
CD
3
M
-------
w
?
h—I
Q
20 30 50 70 90 95
NUMBER PERCENT UNDERSIZE
Figure A-15. Diffusion battery size distribution
for inlet and outlet, Run 19
34
------- |