svEPA
United States Industrial Environmental Research
Environmental Protection Laboratory
Agency Research Triangle Park NC 27711
EPA-600/7-78-129
July 1978
Evaluation
of Electrostatic
Precipitator During
SRC Combustion
Tests
Interagency
Energy/Environment
R&D Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of, control technologies for energy
systems; and integrated assessments of a wide range of energy-related environ-
mental issues.
REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication. Approval does not signify that the contents necessarily reflect the
views and policies of the Government, nor does mention of trade names or commercial
products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/7-78-129
July 1978
Evaluation of Electrostatic Precipitator
During SBC Combustion Tests
by
Grady B. Nichols and William J. Barrett
Southern Research Institute
2000 Ninth Avenue, South:
Birmingham, Alabama 35205
Contract No. 68-02-2610
Task No. 2
Program Element No. EHE623A
EPA Project Officer. William J. Rhodes
Industrial Environmental Research Laboratory
Office, of Energy, Minerals, and Industry
Research Triangle Park, NG 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
ABSTRACT
This report deals with the evaluation of an electrostatic
precipitator and associated environmental factors during the
burning of solvent refined coal in a boiler at Plant Mitchell
of the Georgia Power Company. The effort was part of an overall
study of the use of solvent refined coal in a full-scale
electric power plant. The results of a performance evaluation
of the electrostatic precipitator are reported and interpreted.
Samples of stack emissions were collected with a Source
Assessment Sampling System (SASS train) for chemical analysis,
the results of which are to be reported by ano.ther contractor.
11
-------
CONTENTS
Abstract ii
Figures iv
Tables v
1. Introduction 1
2 . Conclusions and Recommendations 2
3. Performance of the Electrostatic Precipitator System.. 4
Introduction 4
Plant Description 5
Test Program 5
Electrostatic Precipitators 7
Test Results 10
Conclusion and Recommendation 15
4. Electrical Resistivity of the Fly Ash 17
Background 17
Resistivity Measurements at Plant Mitchell 24
Electric Power Set Voltage-Current Curves 27
5. Collection of Samples for Chemical Analysis.. 41
Procedures 41
Samples Collected during Phase II 42
Samples Collected during Phase III 44
Comments on Operation of the SASS Train 44
References 48
111
-------
FIGURES
Number Page
1. Electrostatic precipitator flowsheet—Georgia Power
Company, Plant Mitchell 6
2. ESP No. 1—Georgia Power Company, Plant Mitchell 8
3. ESP No. 3—Georgia Power Company, Plant Mitchell 11
4. Voltage versus current for a precipitator with 23-cm
plate spacing and 0. 25-cm corona wire 19
5. Point-to-plane resistivity probes equipped for thick-
ness measurement 21
6. Typical voltage-current relationships for point-to-
plane resistivity probe 22
7. Voltage-current curves. Plant Mitchell full load coal
Phase I, 1/25/77, Test 1 25
8. Voltage-current curves. Plant Mitchell full load coal
Phase I, 1/25/77, Test 2 26
9. Electrical equivalent circuit of a precipitator elec-
trode system with a dust layer 29
10. Voltage-current relationship in an ideal capacitor
resistor parallel combination 30
11. Voltage divider network for measuring precipitator
secondary voltages 32
12. Sample V-I curve data sheet 33
13. Voltage-current curves. ESP No. 1 full load coal
5/24/77—In Al and Out Al. ESP No. 1 full load
SRC 6/22/77—In A4 and Out A4 36
IV
-------
FIGURES (Concluded)
Number Page
14. Voltage-current curves. ESP No. 3 full load coal
6/8/77—In A3 and Out A3. ESP No. 3 full load SRC
6/22/77—In A5 and Out A5 . 37
15. Voltage waveforms applied to the precipitators at
Plant Mitchell 39
TABLES
Number Page
1. Electrostatic Precipitator Performance Data., 12
2 . Resistivity Data—Phase II Tests 27
3. Average Resistivities—Plant Mitchell Phase II .-. . 27
4. Data from Runs with Regular Coal (Phase II) 43
5. Data from Runs with Solvent Refined Coal (Phase III)... 45
v
-------
SECTION 1
INTRODUCTION
This report deals with the evaluation of an electrostatic
precipitator (ESP) and associated environmental factors during
the burning of solvent refined coal (SRC) in a boiler at Plant
Mitchell of the Georgia Power Company. The tests described in
this report were conducted during May and June 1977.
This effort was a part of an overall study of the use of
solvent refined coal in a full-scale electric power plant. The
study was sponsored by the U. S. Energy Research and Development
Administration and the U. S. Environmental Protection Agency with
the assistance of several contractors and subcontractors. The
part of the overall study covered in this report was performed
under Work Assignment No. 2 of Environmental Protection Agency
Contract No. 68-02-2610 with Southern Research Institute. Under
a subcontract with Southern Research Institute, York Research
Corporation obtained plant operating data during the burning of
unrefined coal (Phase I and Phase II) and later during the burn-
ing of solvent refined coal (Phase III). These data were ana-
lyzed and interpreted by Rust Engineering Company under another
subcontract, with the assistance of Southern Research Institute
personnel.
Institute personnel also collected various samples during
Phase II and Phase III with a Source Assessment Sampling System
(SASS train). These samples were delivered to Hittman Associates,
who had an assignment to perform analyses for organic and inor-
ganic emissions under another Environmental Protection Agency
contract.
This report consists of three principal parts. The first
part deals with the performance of the electrostatic precipitator
system and was prepared by Rust Engineering Company. The second,
closely associated part is concerned with the electrical resistiv-
ity of the fly ash and was prepared by Southern Research Institute,
The third part describes the collection of samples for chemical
analysis, and includes some comments on the operation of the SASS
train.
-------
SECTION 2
CONCLUSIONS AND RECOMMENDATIONS
The data on electrostatic precipitator performance were
inconsistent and inconclusive, suggesting operational abnormali-
ties that were not revealed by the test data. In particular, the
performance of the precipitator was abnormal during baseline
firing with unrefined coal. The data were not adequate to define
the design parameters for an electrostatic precipitator operating
at an SRC-fired boiler. The high carbon content, low electrical
resistivity, and low bulk density of the SRC particulate emis-
sions, in conjunction with operational factors such as gas veloc-
ities and plate electrode areas, appeared to be associated with
the low observed collection efficiency. It is recommended that
additional test programs be performed to define more accurately
the behavior of a precipitator collecting particulate emissions
from the burning of solvent refined coal.
Even though there are potential problems with the test data
developed in this program, one could expect an electrostatic pre-
cipitator to behave at least as well as ESP No. 3 under normal
conditions.* One would expect a new precipitator, designed with
a specific collection electrode area of about 1.3 m2/m3/niin
(400 ft2/1000 ft3/min) and a gas velocity of about 0.6 m/sec
(2 ft/sec) to attain a collection efficiency of about 95% with an
outlet mass loading of about 0.05 g/m3 (0.02 gr/scf) as given by
tests 32 through 34 of Table 1.
The question of the need for additional tests is based on
the need to better define the required rapping and gas flow char-
acteristics for an electrostatic precipitator installed in a new
solvent refined coal boiler. This additional test could perhaps
be conducted in a pilot-scale operation rather than a full size
boiler because of the limited amount of solvent refined coal
available for additional tests.
* See Figure 1 on page 6 and the text on page 10 for descriptions
of the emission control system and ESP No. 3.
-------
No major difficulties were encountered in the operation of
the SASS train. However, the small particle size and possibly
the carbonaceous nature of the SRC particulate emissions resulted
in rapid clogging of the filters, necessitating frequent filter
changes. This minor problem might be alleviated by the use of
larger filters, with redesign of the oven to provide more space
for the cyclone-filter assembly.
-------
SECTION 3*
PERFORMANCE OF THE ELECTROSTATIC
PRECIPITATOR SYSTEM
INTRODUCTION
The Rust Engineering Company participated in a program
designed for the evaluation of operating parameters in a full-
scale commercial steam generating unit when fired with a fuel con-
sisting of solvent refined coal.
The program was directed and managed by Southern Company
Services, Inc., the engineering and research subsidiary of the
Southern Company, Atlanta, Georgia, a public utility holding com-
pany whose operating subsidiaries include Georgia Power Company,
Alabama Power Company, Gulf Power Company, Mississippi Power Com-
pany, and Southern Electric Generating Company. The test burn was
sponsored by the US Energy Research and Development Administration,
The ESP evaluation was sponsored by the US Environmental Protec-
tion Agency. Other participating companies included:
The Babcock and Wilcox Company
Southern Research Institute
York Research Corporation
The Pittsburgh & Midway Coal Mining Company
Hittman Associates, Inc.
Battelle Northwest
TRW Systems and Energy
Bechtel Corporation
Wheelabrator-Frye, Inc.
The test program was performed in various stages between
January and June 1977. This report pertains to the part of the
program concerned with the evaluation of the performance of the
electrostatic precipitation system, installed for the removal of
solid particulate matter from the exhaust gases emanating from
the test boiler.
* Section 3 was prepared by John E. Paul, Senior Staff Engineer,
Rust Engineering Company.
-------
PLANT DESCRIPTION
The test program was conducted at the Mitchell Power Station
of the Georgia Power Company, an operating subsidiary of the
Southern Company, Atlanta, Georgia. The plant is located on State
Highway 3, south of Albany, Georgia.
The test unit is identified as Boiler No. 1, a 22.5-MW rated
B & W boiler, fired with pulverized fuel. The boiler is equipped
with three B & W coal mills, front-wall burner arrangement, and
features in its exhaust gas path a Ljungstrom-type air preheater,
a primary electrostatic precipitator (Research Cottrell, Inc.)
identified as ESP No. 1, an induced draft (ID) fan, and a secon-
dary electrostatic precipitator (American Standard), identified
as ESP No. 3. ESP No. 3 serves to clean the combined exhaust
gases emanating from No. 1 and No. 2 Boilers (see Figure 1).
TEST PROGRAM
The test program was divided into three phases:
Phase I: Baseline data acquisition.
Test boiler operation with conventional
burners on normal coal.
Phase II: Baseline data acquisition.
Test boiler operation with modified
burners on normal coal.
Phase III: Test data acquisition.
Test boiler operation with modified
burners and readjusted fuel mills
on solvent refined coal.
Each phase included boiler operation at rated full load,
two-thirds load, and one-third of full load. Data acquisition
included:
1. Pulverizer (mill) performance.
2. Boiler performance.
3. Fuel handling equipment.
4. Combustion performance.
5. Particulate mass loading in exhaust gases (ASME and
EPA Method 5 tests).
a. Upstream of ESP No. 1.
-------
ESP *l
AID PRE-HEATER
I
1
1
1
I/I ] h
1
1
1
1
1
r\
1C
'3 BOILER
HV TRANSFORMER RECTIFIERS
"2 BOILER
W
h
h
ESP '2
Figure 1. Electrostatic precipitator flowsheet — Georgia Power
Company, Plant Mitchell.
-------
b. Downstream of ESP No. 1.
c. Downstream of ESP No. 3.
6. Particulate sizing, same locations as 5.
7. Exhaust gas analysis for Oa , CO, and COa.
8. S02 concentration.
9. NOX concentration.
10. Hydrocarbon concentration.
11. Chemical analysis of fuel, bottom ash/ and fly ash.
12. Electrical resistivity of fly ash.
13. Precipitation current and V/I relationship.
ELECTROSTATIC PRECIPITATORS
Description of ESP No. 1
ESP No. 1 is of Research Cottrell, Inc. (RC), design and
was put into operation in 1947. It is of the horizontal gas flow
type, generally as shown in Figure 2.
The precipitator consists of one chamber having two indepen-
dent electrical fields in series, which were originally energized
by high voltage (HV) transformers with mechanical, synchronous
motor driven rectifiers. High voltage cables are used for power
transmission.
Several years ago the mechanical rectifiers were replaced
with new solid state rectifiers housed integrally with the trans-
formers and provided with modern saturable core reactor voltage
control equipment.
One HV energizing unit supplies the required precipitation
power to both inlet and outlet fields via a double half-wave
bridge rectifier arrangement. Voltage control upon input signal
based on spark rate limitation, dc current, or voltage limitation
thus affects both fields.
The precipitator is equipped with collecting electrodes made
of expanded metal sheets tied together by anvil bars at the front
and rear ends and suspended from structural members located in
the roof structure. Discharge electrodes are single round wires
of 0.25 cm (0.1 in.) diameter, suspended from a framework located
between fields, and supported from bushing type ceramic insulators
-------
EXPANDING METAL ELECTRODE
COLLECTING PLATES
00
0.2 m
(8-1/2")"
6.9 m
(22'-6">
27 SPACES AT 0.2 m (8J&")
COLLECTING PLATES
- 0.2 m
(8-1/2")
Figure 2. ESP No. 1 - Georgia Power Company,
Plant Mitchell.
-------
in the roof-mounted insulator compartment. These wires are held
in plumb suspension between collecting electrodes by weights fas-
tened to their lower extremities.
Collecting electrode (CE) plates are cleaned by side-wall
mounted, solenoid-operated impact rappers, acting horizontally
and perpendicular to the plates and the gas stream, thus promoting
thrust for reentrainment of collected material. No equipment is
provided for discharge electrode (DE) cleaning.
The precipitator is equipped with perforated gas velocity
distribution plates located in the inlet and outlet plenum.
Field measurements and Research Cottrell supplied design
information are as follows:
Design gas volume 4800 m3/min (170,000 ACFM)
Design gas temperature 175°C (350°F)
Design efficiency 90%
CE area (effective) 963 m2 (10,368 ft2)
DE length 2960 m (9,720 ft)
Design specific collecting 12 m2/m3/sec
area (SCA) (61 ft2/1000 ACFM)
Free face opening 25.8 m2 (277.6 ft2)
Design treatment velocity 3.1 m/sec (10.2 ft/sec)
CE height 5.5 m (18 ft)
Number of CE plates 28
CE length (two plates in 1.8 m (6 ft)/field
each of two fields)
Total effective treatment 3.66 m (12 ft)
length
Design treatment time 1.3 sec
CE spacing 20 cm (8 in.)
The precipitator was inspected during the week of January 17,
1977. The following comments apply.
The internal components of the precipitator were found to be
in excellent condition. Specifically, there were no broken or
-------
missing discharge electrode wires, and no deterioration or
corrosion of collecting electrodes. No excessive accumulation of
ash was observed on collecting electrode surfaces and wires showed
only isolated spots of ash buildup in the form of hardened
enlargements up to 6 cm (2.5 in.) diameter. All insulators were
found to be in good workable condition. High voltage cables are
regularly topped with insulating liquid.
The perforated plates in the inlet were clean; there was
some accumulation of ash on the outlet side.
The ideal alignment between discharge wires and collecting
electrode surfaces would be 9.8 cm (3-7/8 in.); the worst align-
ment measured was 11.7 cm (4-5/8 in.) one side to 8.7 cm
(3-7/16 in.) on the opposite side.
Prior to commencement of testing, the precipitator was
thoroughly cleaned by plant personnel. This included inlet and
outlet plenums and ductwork.
Description of ESP No. 3
ESP No. 3 is of American Standard design and was placed in
service in 1962. It was originally installed to treat the gases
exhausting from No. 3 Boiler, a 125-MW unit not a subject of this
test series. The ducting to and from this ESP was later modified
to provide for its operation on the combined gases from No. 1 and
No. 2 Boilers.
The precipitator is of the horizontal gas flow type, gener-
ally as shown in Figure 3. The precipitator consists of two cham-
bers, having two independent electrical fields in series, and it
is energized by two HV transformer-rectifier power units with
double half-wave bridge configuration and saturable core reactor
voltage control equipment. The two discharge electrode bus
systems in each chamber are energized each by one half-leg of the
open electrically negative rectifier bridges, rated at 1,200 mA
pulsating dc output. The total effective collecting electrode
surface area has been measured at 4630 m2 (49,882 ft2).
A physical inspection of the precipitator revealed no dis-
crepancies in discharge electrode or collecting electrode systems
which would affect the normal performance of the equipment. Sev-
eral high voltage support bushings showed hairline cracks. These
bushings were exchanged for new bushings prior to commencement of
testing.
TEST RESULTS
A summary of the relevant test data is presented in Table 1.
10
-------
/ir\
14.2 m (46'-4V,") OUTSIDE
ROOF EL. 88 m_J
<287'-2")
CONTROL
X
CHAMBER 1
CHAMBER 2
S.R. = SATURABLE CORE REACTOR
DOUBLE HALF WAVE POWER SUPPLY
PENTHOUSE
BLOWER
GAS FLOW
TRANSFORMERS
ACCESS
DOOR
PLAN AT A-A
Figure 3. ESP No. 3 — Georgia Power Company,
Plant Mitchell.
-------
TABLE 1. ELECTROSTATIC PRECIPITATOR PERFORMANCE DATA
N>
Phase I Data
Parameter
Date
Boiler load, MW
Gas flow, ESP outlet.
m'/min
Gas temp, °C
H20, %
Particulate concn,
inlet, g/m3 (dry)
Particulate concn,
outlet, g/m3 (dry)
Efficiency, %
Av power input, W/m2
SCA, mVlOOO mVmin
W, calculated, cm/sec
Inlet dust, %
1 um
5 um
10 um
Fractional efficiency, <
Smaller than 1 um
Smaller than 5 um
Smaller than 10 um
Boiler operating data
C-burning efficiency,
02, %
Fuel flow, kg/min
Air flow, kg/min
C in ash, %
i
3/22
22.
3656
161
4.
9.
2.
71.
-
263
7.
1.
20
25
i
15.
32.
48.
% 97.
4.
186
1700
14.
5
0
21
64
3
912
2
0
0
0
94
8
3
2
3/23
22.5
3392
152
6.2
8.86
1.78
79.9
-
283
9.420
-
-
-
-
-
-
97.16
4.9
181
1710
15.5
3
3/24
15.0
2493
136
6.0
8.85
0.45
94.9
-
386
12.884
1.5
10
22
12.0
61.0
80.0
97.40
5.7
145
1330
17.1
(ESP No.
4
3/24
7.5
1660
126
6.9
6.75
0.06
99.1
-
580
13.599
1.2
14
30
78.0
93.0
95.0
96.48
11.2
67
937
21.6
1)
Test
5
No.
3/26
7.
1595
130
5.
6.
0.
99.
-
604
12.
-
' -
-
-
-
-
97.
9.
67
960
17.
5
4
78
07
0
763
25
7
8
6
3/27
7.5
1791
128
5.4
6.27
0.08
98.7
-
538
13.437
-
-
-
-
-
-
96.52
11.1
69
1000
19.1
7
3/28
22.5
3497
159
6.0
10.66
2.11
80.2
-
275
9.800
-
-
-
-
-
-
97.58
4.3
179
1750
11.8
8
3/29
15.0
2491
148
8.4
9.00
0.23
97.4
-
387
15.775
-
-
-
-
-
-
97.20
6.1
119
1350
15.2
9
3/30
15.0
2565
148
6.8
10.73
0.41
96.2
-
376
14.532
-
-
•
-
-
-
97.70
6.1
122
1350
11.1
(Continued)
-------
TABLE 1 (continued).
U>
Parameter
ESP No.
Date
Boiler load, MW
Gas flow, ESP outlet.
m'/min
Gas temp, °C
HiO, %
Particulate concn,
inlet, g/m1 (dry)
Particulate concn,
outlet, g/m1 (dry)
Efficiency, %
Av power input, W/ra1
SCA, m*/1000 m'/min
W, calculated, cm/sec
Inlet dust, %
1 urn
5 urn
10 wm
Fractional efficiency.
Smaller than 1 urn
Smaller than 5 urn
Smaller than 10 urn
Boiler operating data
C-burning efficiency.
0,, «
Fuel flow, kg/min
Air flow, kg/min
C in ash, %
10
1
5/24
21.0
3687
152
6.7
9.14
1.66
81.8
•
261
10.884
2.0
24
45
%
49.4
64.2
68.6
98.34
4.5
169
1910
9.5
11
1
5/25
14.0
3000
147
6.2
9.96
0.52
94.8
-
321
15.367
3.5
18
40
74.8
75.0
82.6
96.36
6.5
116
1490
16.2
Phase II
13
1
5/27
21.0
3755
150
5.9
5.70
1.29
77.4
-
257
9.641
-
'-
_
-
-
-
97.77
4.2
160
1900
18.7
Data (ESP No. 1 and ESP No.
14
1
5/28
21.0
3775
152
7.4
4.54
1.32
70.9
-
255
8'. 022
-
-
_
-
-
_
97.94
4.1
151
1900
21.5
15
1
5/29
14.0
3201
144
6.1
3.02
0.40
86.8
-
300
11.221
-
-
_
-
-
-
-
6.0
106
1590
22.8
Test No.'
16
1
5/30
14.0
3151
143
6.1
6.58
0.82
87.5
_
306
11.339
_
_
_
-
_
_
-
6.3
115
1600
22.7
, 3)
»
17
1
5/31
7.5
1806
146
6.2
4.94
0.11
97.8
-
533
11.826
3.0
30
50
85.5
91.5
93.1
-
11.6
71
1130
16.9
18
1
«/l
7.5
1934
138
5.6
6.05
0.28
95.4
_
498
10.268
-
-
_
-
-
-
-
11.6
73
1150
16.8
19
3
6/5
21.0
3320
139
7.0
4.02
0.09
97.8
5.87
1394
4.519
2.2
20
40
-
-
_
97.93
4.6
152
1930
30.0
20
3
6/5
21.0
3499
146
6.6
3.74
0.03
99.2
6.12
1325
5.896
-
-
_
-
-
_
98.32
4.6
150
1900
25.1
21
3
6/6
21.0
3420
137
6.9
4.40
0.06
98.6
5.76
1355
S.251
-
-
-
-
-
_
97.62
4.7
152
1940
29.1
* Test No. 12 on Hay 26 aborted because the fuel feeder tripped.
(Continued)
-------
TABLE 1 (continued).
Phase III Data (ESP No. 1
and ESP No. 3)
Test No.*
Parameter
ESP No.
Date
Boiler load, MM
Gas flow, ESP outlet,
m'/min
Gas temp, "C
H20, *
Particulate concn,
inlet, g/m» (dry)
Particulate concn,
outlet, g/ra! (dry)
Efficiency, »
Av power input, W/m2
SCA, raVlOOO m'/min
W, calculated, cm/sec
Inlet dust, %
1 pm
5 ura
10 ym
Fractional efficiency, %
Smaller than 1 ym
Smaller than 5 ym
Smaller than 10 ym
Boiler operating data
C-burning efficiency, %
Oj, %
Fuel flow, kg/min
Air flow, kg/min
C in ash, »
* Test No. 30 on June 21
t The plate rapping cycle
I The plate rapping cycle
§ The plate rapping cycle
22
1
6/13
21.0
3609
161
6.8
1.07
0.87
18.4
-
267
23
1
6/14
14.0
3076
144
6.2
1.28
1.00
21.9
-
313
1.267 1.320
4.0
20
30
20.9
-
-
98.51
5.4
132
1880
75.7
5.0
18
22
44.1
-
-
97.03
7.2
91
1690
88.5
24
1
6/15
7.5
1947
151
5.4
1.08
0.49
56.6
-
495
2.817
4.2
15
22
41.5
17.8
-
96.89
11.7
54
1120
89.7
25
1
6/16
21.0
3621
166
6.7
1.00
0.83
17.0
-
266
1.159
-
-
-
~-
-
-
98.37
5.6
134
1840
82.4
26
1
6/17
21.0
3629
160
7.3
1.00
0.80
20.0
-
265
-
-
-
-
-
-
-
98.28
5.8
133
1860
84.8
27
1
6/18
7.5
2007
147
6.0
1.08
0.62
42.3
-
480
1.912
-
-
-
-
-
-
-
10.9
56
1160
88.9
28
1
6/19
7.5
2005
147
5.9
1.09
0.59
45.9
-
481
2.116
-
-
-
_
-
-
-
11.2
56
1130
89.7
was aborted because of a chanae in load demand.
was 5 min (continuously sequential) .
was 15
was 25
min (manually intermittent) .
min (manually intermittent) .
29 31
1 3
6/20 6/22
14.0 21.0
3229 3282
147 153
7.6 7.4
1.66 0.99
1.23 0.13t
25.9 86.9
5.19
298 1410
1.688 2.404
6.0
26
40
74.5
83.8
85.7
98.28
7.4 6.7
95 136
1780 1880
86.0 80.5
32
3
6/22
21.0
3271
157
8.2
0.84
0.03f
96.4
6.07
1417
3.762
-
-
-
_
-
-
98.66
6.7
134
1890
77.0
Conversion factors
To convert
ra'/min
g/m'
mVlOOO m'/min
W/m2
kg/min
to
ACFM
gr/SCF
ftViooo
W/ft2
Ib/hr
33
3
6/23
21.0
3328
148
6.6
0.82
0.04$
95.1
7.35
1391
3.587
-
-
-
.
-
-
98.65
5.7
133
1960
75.0
34
3
6/23
21.0
3471
152
9.1
0.73
0.03§
95.9
7.37
1335
4.002
-
-
-
_
-
-
98.80
5.7
133
1950
66.3
for Table 1
multiply by
ACFM
35.314
0.4370
0.3048
0.0929
132.3
-------
The low efficiencies obtained by ESP No. 1, the primary
precipitator, during SRC firing reflects the difficulty of precip-
itating material of extremely low electrical resistivity. This
was expected.
It was not expected, however, that ESP No. 3, the secondary
precipitator, would operate at calculated migration velocities of
only 2.5 to 4 cm/sec during SRC firing. Its subnormal perfor-
mance at migration velocities of 4.5 to 5.2 cm/sec during full-
load base coal firing is highly suspect of some operational
abnormality, as yet not identified. A possible cause could be
bad gas velocity distribution within the precipitator inter-
electrode space.
Neither precipitation power input nor measured particle
size distribution provide any explanation for the apparent per-
formance discrepancy. It is possible that the size distribution
measurements are subject to error not identifiable at this time.
The specific problems associated with interpreting the data
from ESP No. 3 involve the uncertainties in particle size distri-
bution provided during the short test period allocated to this
collector. As stated previously, the exhaust gases from Boilers
1 and 2 are combined prior to introduction into this control
device. In order to conduct measurements on~ the effluent from
Boiler No. 1, Boiler No. 2 must be shut down. The shutdown of
Boiler No. 2 resulted in a very low gas velocity through the col-
lector—approximately 0.6 m/sec (1.9 ft/sec).
It is to be noted that the gas flow control turning vanes
were designed for gas velocities of about 1.5 to 1.8 m/sec (5 to
6 ft/sec). Thus, it is not unreasonable to expect the gas veloc-
ity distribution to be more nonuniform when the gas comes from
only one boiler.
A second factor of concern is the behavior of the control
device when collecting the very high carbon content particulates.
It is to be expected that the proper rapping conditions—acceler-
ation and interval—for the solvent refined coal residue will be
significantly different than for fly ash. The optimization of
these factors was not attempted in this test period.
CONCLUSION AND RECOMMENDATION
The data available from this test program do not appear to
be adequate to arrive at a definitive conclusion as to the sizing
requirements for an electrostatic precipitation system operating
at an SRC-fired boiler. If anything, the interpretation of the
data would result in a very conservative design approach, pointing
to an electrode surface area requirement of 80 to 120 m2 (400 to
15
-------
600 ft2) of collecting electrode area per cubic meter per second
(1000 ACFM) for an outlet loading of approximately 0.05 g/m3
(0.02 gr/SCF).
Several items were mentioned above that could degrade the
performance of the electrostatic precipitator. Therefore, one
cannot state conclusively what the optimum design for a precipi-
tator for the new SRC facility would be. One can, however, state
that if the conditions used in this test were duplicated, then
the new control device would be expected to operate at least as
well as the performance observed in ESP No. 3 while collecting
particulate from Boiler No. 1.
A word of caution is in order. The particulate that results
from the combustion of solvent refined coal is a very low density
material that is high in carbon content. This particulate closely
resembles that from the combustion of fuel oil. This highly con-
ductive material is especially susceptible to reentrainment.
Therefore, the principles of design for electrostatic precipita-
tors for oil-fired units should be applied to any new SRC plant.
Specifically, special attention should be paid to attaining a
near uniform gas velocity distribution and the face velocity
through the electrostatic precipitator should be maintained at
about 0.6 m/sec (2 ft/sec).
It is recommended to pursue additional test programs to
define more accurately the behavior of an electrostatic precipita-
tor collecting SRC dust.
The additional tests recommended for a further definition
of the particulate and electrostatic precipitator parameters
should center on determining the particle size distribution and
gas velocity distribution. It is recognized that another full-
scale SRC test is not practical at this time. Since the particle
size distribution measurements are so critical to the design of a
new control device, these data would be extremely useful. If an
acceptable small-scale combustor could be located that would
duplicate the time-temperature profile of the furnace as well as
the particle size distribution, a pilot-scale evaluation could
provide useful data. However, care must be exercised in the selec-
tion of the pilot-scale device. One possible candidate is the
Babcock and Wilcox minicombustor.
A test that could provide useful data at Plant Mitchell is
to determine the gas velocity distribution with ESP No. 3. This
measurement would require a 2-day shutdown of Boilers 1 and 2 and
hot wire anemometer measurements within the interelectrode space.
We suggest that these measurements be considered.
16
-------
SECTION 4*
ELECTRICAL RESISTIVITY OF THE FLY ASH
BACKGROUND
The electrical resistivity of particulate matter present in
the effluent gas stream is one of the primary factors that deter-
mine the operating characteristics of an electrostatic precipita-
tor. The following paragraphs excerpted from a report by Nichols1
are reproduced here to provide an understanding of the importance
of the resistivity determinations and the methods by which in
situ resistivity was measured.
Significance of Particulate Resistivity to Electrostatic Precipi-
tator Operation
In a conventional single-stage, dry-electrode electrostatic
precipitator, the total corona current flows through the collected
dust layer to reach the grounded collection electrode. This flow
of current establishes an electric field (E) in the dust layer
which is proportional to the corona current density (j) and the
particulate resistivity (p) as given by
E = jp. (1)
The electric field in the dust layer yields a voltage drop (AV)
across the dust layer which is proportional to the dust layer
thickness (t) and is given by
. AV = Et. (2)
* This section was prepared by Sherman M. Banks, Electrical Engi-
neer, and Dr. Jack R. McDonald, Head, Theoretical and Basic
Studies Section, Southern Research Institute.
1. Nichols, Grady B. "Techniques for Measuring Fly Ash Resis-
tivity", EPA Report No. 650/2-74-079, US Environmental Protec-
tion Agency, pp. 2-4.
17
-------
If the resistivity of the dust layer is increased while the cur-
rent density is held constant, the electric field in the layer
increases proportionately (Equation 1). If the electric field in
the dust layer exceeds the field strength for corona initiation
(electrical breakdown), an electron avalanche will occur in the
dust layer similar to that which occurs adjacent to the corona
wire. This electrical breakdown acts as a limit on the allowable
electrical conditions in the precipitator as is discussed below.
The manner in which this breakdown limits the precipitator
performance is dependent upon the value of the resistivity of the
dust and the thickness of the layer. If the resistivity is in
the moderately high range (^1012 ohm-cm) breakdown of the dust
layer will occur at a voltage too low to propagate a spark across
the interelectrode region. This gives rise to a condition of
reverse ionization or back corona. Figure 4 illustrates these
two conditions. The figure shows the current density as a func-
tion of applied voltage for an electrostatic precipitator with a
0.28-cm (0.109-in.) diameter corona wire, a 23-cm (9-in.) plate
spacing, and a dust layer thickness of 1 cm (0.5 in.). If the
dust layer resistivity is in the moderately high range, e_.g_. ,
2 x 10ll ohm-cm, electrical breakdown in the dust layer will occur
at an applied voltage greater than that required for sparking
between clean electrodes, and sparking will occur at a reduced
current density. For a very high resistivity dust with the condi-
tions shown, electrical breakdown will occur in the dust layer at
a much lower current density and applied voltage. Under these
conditions, the dust layer will be continuously broken down elec-
trically and will interject into the interelectrode space ions of
an opposite electrical polarity from those produced by the corona.
The precipitator electrical operating conditions are thus limited
by a high resistivity dust layer and the precipitator is con-
strained to operate at lower currents and voltages than one col-
lecting a lower resistivity dust. The magnitude of the reduction
in electrical operating conditions is a direct function of the
dust resistivity.
In view of the importance of the resistivity of the dust
layer as a primary factor in limiting the performance of a pre-
cipitator, it becomes necessary to determine the resistivity of
the material to be collected in order to estimate the conditions
to be expected in a precipitator.2
2. White, H. J. "Resistivity Problems in Electrostatic Precipi-
tation". J. Air Pollut. Control Assoc., "2A_ (4), April 1974,
p. 316.
18
-------
60
2 50
o
HI
cc
O
o- 40
GO
LLJ
tr
<
O
z
2 30
GO
Z
UJ
Q
t-
cc
D
O
<
o
cc
O
SPARK
p = 2x 1011 OHM-CM
./p = 2 x 1012 OHM-CM
20
30 40 50
APPLIED VOLTAGE, kilovolts
60
Figure 4. Voltage versus current for a precipitator with 23-cm plate spacing
and 0.25-cm corona wire. Solid curve is for a clean electrode. The
two dashed curves represent conditions for a 0.5-cm layer of dust
with the resistivities indicated.
19
-------
Point-to-Plane Probe
The point-to-plane probe for measuring resistivity has been
in use since the early 1940's in this country.2 Two models of
this device are shown in Figure 5. The probe is inserted directly
into the dust-laden gas stream and allowed to come to thermal
equilibrium. The particulate sample is deposited electrically
onto the measurement cell through the electrostatic action of the
corona point and plate electrode. A high voltage is impressed
across the point and plane electrode system such that a corona is
formed in the vicinity of the point. The dust particles are
charged by the ions and perhaps by free electrons from this corona
in a manner analogous to that occurring in a precipitator.
The dust layer is formed through the interaction of the
charged particulate with the electrostatic field adjacent to the
collection plate. Thus, this device is intended to simulate the
behavior of a full-scale electrostatic precipitator and to provide
a realistic value for the resistivity of the dust that would be
comparable to that in the actual device.
In the point-to-plane technique, two methods of making mea-
surements on the same sample may be used. The first is the "V-l"
method. In this method, a voltage-current curve is obtained
before the electrostatic deposition of the dust, while the col-
lecting disc is clean. A second voltage-current curve is obtained
after the dust layer has been collected. After the layer has been
collected and the clean and dirty voltage-current curves obtained,
the second method of making a measurement may be used. In the
second method, a disc the same size as the collecting disc is
lowered on the collected sample. Increasing voltages are then
applied to the dust layer and the current obtained is recorded
until the dust layer breaks down electrically and sparkover occurs.
The geometry of the dust sample, together with the applied voltage
and current, provides sufficient information for determination of
the dust resistivity.
In the point-to-plane method, the voltage drop across the
dust layer is determined by the shift in the voltage-versus-
current characteristics along the voltage axis as shown in
Figure 6. The situation shown is for resistivity values ranging
from 103 to 1011 ohm-cm.
If the parallel disc method is used, dust resistance is
determined from the voltage measured just prior to sparkover. In
both methods the resistivity is calculated as the ratio of the
electric field to the current density.
The practice of measuring the resistivity with increasing
voltage has evolved because the dust layer behaves as a nonlinear
resistor. As the applied voltage is increased, the current
increases greater than that attributable to the increase in
20
-------
PICOAMMETER
CONNECTION
HIGH VOLTAGE
CONNECTION
DIAL INDICATOR
PICOAMMETER
CONNECTION
MOVABLE
SHAFT
STATIONARY
POINT
GROUNDED
RING
(a)
(b)
Figure 5. Point-to-plane resistivity probes equipped for
thickness measurement.
21
-------
3.0
2.5
(M
5 2.0
o
a.
a
V-
2
UJ
X
I 1.0
0.5-
SPARK
NO DUST
DEPOSIT ON
PLATE
VOLTAGE DROP ACROSS
DUST LAYER (Vd) FOR
DUST THICKNESS
(xd) = 0.001 METER
8 12
VOLTAGE, kV
16
20
Figure 6. Typical voltage-current relationships for point-to-plane
resistivity probe.
22
-------
voltage. Therefore, as described in the A.S.M.E. Power Test Code
No. 28 procedure, the value just prior to sparkover is reported
as the resistivity.
There is considerable justification for using the value of
resistivity prior to electrical breakdown as the resistivity,
since it is precisely at electrical breakdown that the resistivity
causes problems within the precipitator. The electrical breakdown
in the dust layer in the operating precipitator either initiates
electrical sparkover or reverse ionization (back corona) when the
resistivity is the factor limiting precipitator behavior. If
neither of these events occurs, the dust layer merely represents
an additional voltage drop to the precipitator power supply.
Even though there are many similarities between the opera-
tion of the point-to-plane device and a full-scale precipitator,
several problems also exist. The first problem encountered is
the determination of the thickness of the dust layer. Some
devices use a thickness measurement system built into the probe.
In other devices, the instrument is withdrawn from the duct and
the thickness of- the layer is estimated visually by inspecting
the dust layer. However, the dust layer is almost always dis-
turbed by the air flow through the sampling port and extreme care
is required to preserve the layer intact.
The advantages of utilizing the point-to-plane probe for in
situ measurements are:
1. The particulate collection mechanism is the same as
that in an electrostatic precipitator.
2. The dust-gas and dust-electrode interfaces are the
same as those in an electrostatic precipitator.
3. The measurement electric field and current densi-
ties are comparable to those in the precipitator.
4. Flue gas conditions are preserved.
5. The values obtained for the resistivity are, in gen-
eral, consistent with the electrical behavior
observed in the precipitator.
6. Measurements can often be made by two different
methods.
The disadvantages are:
1. The measurement of the dust layer thickness can be
difficult.
2. High voltages are required for collection.
23
-------
3. Considerable time is required for each test.
4. Experienced personnel are required for testing.
5. A number of measurements are required for gaining
confidence in the measured value (there is consider-
able scatter in the data).
6. Particle size of the collected dust is not repre-
sentative.
7. Sample size is small.
8. Carbon in the ash can hamper resistivity measure-
ments .
RESISTIVITY MEASUREMENTS AT PLANT MITCHELL
Attempts were made to measure the in situ electrical resis-
tivity of the fly ash after the air preheater during all three
phases of the test program. Unfortunately, these attempts met
with varying degrees of success.
The fly ash during the Phase I testing had a relatively high
carbon content which caused considerable error in the measurement
of the resistivity of the remainder of the particulate with the
disc-to-disc, direct contact method of measurement. However, some
data may be gleaned from the V-I method of measurement, as the
accompanying Figures 7 and 8 illustrate. The resistivity referred
to in the figures is the average of the resistivities computed at
each data point indicated. These resistivities were
2.5 x 10ll ohm-cm from Test 1 and 2.1 x 10ll ohm-cm from Test 2.
These were computed at significantly different field strengths
and densities. The reason for the differences in these V-I curves
has not been determined.
Considerably more data were collected during the Phase II
testing. Again, the percentage of combustibles ran quite high;
however, direct contact data could be obtained. These data are
summarized in Table 2.
Some tests, such as 1, 6, and 7, may have been biased by
the presence of the large percentage of carbon in the fly ash.
Tests 1 through 5, conducted on the 24th, were collected while
the boiler was operating at full load. The lower resistivities
the next day were determined under two-thirds load conditions.
From these data, the averages were computed and are presented in
Table 3.
24
-------
SRI RESISTIVITY PROBE
1.50
1.25
IIrl I I I I
O CLEAN-PLATE
D DIRTY PLATE
RESISTIVITY = 2.5 x 1011
1.00
a
I
ui
oc
oc
o
0.75
0.50
0.25
456
II I |_ I
8 9
VOLTAGE, kV
I I
10 1.1 12 13
Figure 7. Voltage-current curves. Plant Mitchell full load
coal Phase I, 1/25/77, Test 1.
25
-------
SRI RESISTIVITY PROBE
5.00(—
4.50
i.
4.00
RESISTIVITY = 2.
3.50
3.00
u
E
2 2.50
S 2.00
1.50
1.00
0.50
0.00
I I I I I
O CLEAN PLATE
D DIRTY PLATE
/
!.1x1011 /
//
/ D
/ /
^7
II I I
8 10 12 14 16 18 20 22
VOLTAGE, kV
Figure 8. Voltage-current curves. Plant Mitchell full load
coal Phase /, 7/25/77, Test 2.
26
-------
TABLE 2. RESISTIVITY DATA—PHASE II TESTS
Test
No.
1
2
3
4
5
6
7
8
9
10
Date
5/24
5/24
5/24
5/24
5/24
5/25
5/25
5/25
5/25
5/25
Temperature/
°C (°F)
143
138
155
138
154
131
145
132
132
137
(290)
(280)
(311)
(280)
(309)
(268)
(293)
(270)
(270)
(278)
Resistivity,
V-I method,
ohm- cm
9
2
1
1
2
4
1
1
6
4
.0 x
.8 x
.3 x
.7 x
.4 x
.0 x
.9 x
.3 x
.5 x
.4 x
10
10
10
10
10
10
10
10
10
10
1 0
i i
i i
i i
i i
1 0
1 0
1 1
1 0
1 0
Resistivity,
direct contact,
ohm- cm
3.
2.
1.
1.
1.
•1.
1.
6.
7.
4.
5
4
0
9
3
5
6
6
5
8
x
X
X
X
X
X
X
X
X
X
107
10 J1
1011
1011
1011
106
10"
1010
109
109
TABLE 3. AVERAGE RESISTIVITIES—PLANT MITCHELL PHASE II
Condition V-I
Full load, 1.8 x 10 ' '
Tests 1 through 5
Full load, 2.1 x 10 J 1
without Test No. 1
Two-thirds load, 6.1 x 10 10
Tests 6 through 10
Two-thirds load, 8.0 x 10 10
without Tests 6 and 7
Direct
1.3
1.6
1.6
2.6
contact
x 101 1
x 101 J
x 1010
x 1010
Phase III testing was conducted by burning the SRC. The
extremely high carbon content of the ash made any sort of deter-
mination of resistivity with the in situ point-plane probe vir-
tually impossible.
ELECTRIC POWER SET VOLTAGE-CURRENT CURVES
Data were collected during Phases II and III by recording
the electrical power set's (TR's) secondary voltage versus second-
ary current relationships. The method of collecting these data
and background information is outlined below. Actual curves and
waveforms are reported in subsequent paragraphs.
27
-------
Background
The electrical equivalent circuit of a precipitator is
shown in Figure 9,3 where
V = voltage applied in volts
I = total conventional current flow in amperes
Cp = equivalent precipitator capacitance in farads
RG = effective resistance of the interelectrode gap
in ohms
CD = effective capacitance of the dust layer in farads
RD = effective resistance of the dust layer in ohms.
The voltage normally applied to a precipitator is either half-
wave or full-wave rectified 60 Hz ac. Neglecting for a moment
the effects of CD and RD, the capacitor, Cp, charges on the
increasing portion of the voltage waveform and discharges on the
decreasing portion. The current from the discharging capacitance
flows through the resistance RG tending to maintain the peak volt-
age applied. There is an exponential decay of this voltage depen-
dent on the time constant of the &Gcp circuit. This time constant
is given by: 4
T = RC seconds
where: T = time in seconds for the waveform to decrease
to approximately 37% of its peak value after
the voltage is removed
RG = equivalent resistance of the interelectrode
region in ohms
Cp = equivalent capacitance of the electrode
system in farads.
The current, I, will flow in the return leg of the circuit only
during the charging of the capacitor. During the remainder of
the cycle, the current supplied to RG is the discharge current
from Cp. These relationships are shown in Figure 10. In this
example T is assumed to be greater than 8 msec or one-half cycle
of the line voltage.
3. Oglesby, S., Jr., and Grady B. Nichols. A Manual of Electro-
static Precipitation Technology Part 1—Fundamentals. NTIS
PB 196 380, The National Air Pollution Control Administration,
Cincinnati, Ohio, 1970, p. 251.
4. Oglesby, S., Jr., and Grady B. Nichols. A Manual of Electro-
static Precipitation Technology Part 1—Fundamentals. NTIS
PB 196 380, The National Air Pollution Control Administration,
Cincinnati, Ohio, 1970, p. 254.
28
-------
CP:
RETURN O-
Figure 9. Electrical equivalent circuit of a prec/'pitatbr electrode
system with a dust layer. '
29
-------
VOLTAGE ACROSS RG
APPLIED VOLTAGE
CURRENT I
TIME
Figure 10. Voltage-current relationship in an ideal capacitor
resistor parallel combination.
30
-------
Normally the effective impedance presented by the parallel
combination of CD and RD is negligible compared to the Impedance
of RQ. Thus, the time domain response of the precipitator is
determined by the combination of Cp and RQ. However, this is not
true when the dust layer is in a breakdown condition and possibly
exhibiting back corona. The breakdown may effectively short out
the dust layer and a portion of RG thereby reducing the time con-
stant, T, and increasing the current I. This change in time con-
stant may be monitored on an oscilloscopic presentation of the
voltage waveform and used to support evidence that breakdown of
the dust layer is occurring.
The voltages and currents in a precipitator are most often
measured by the installed power set instrumentation as root-mean-
square (RMS) or effective values. The capacitances and resis-
tances vary slowly with time so that the equivalent circuit of a
precipitator in normal operation can be approximated as a pure
resistance across the terminals of a dc source. The voltage-
current relationship is simply V = RI where R is the effective
value of the resistance in ohms, V is in RMS volts, and I is in
RMS amperes. An actual precipitator departs from the ideal in
that R is a nonlinear function of the current. The graphical
presentation of precipitator voltage versus secondary current is
not the straight line as generated with an ohmic resistance but
generally curved, and referred to as a V-I curve.
V-I Curve Measurement Techniques
The measurement of V-I curves at Plant Mitchell entailed
the direct measurement of power set secondary voltages and cur-
rents. Today, most precipitator manufacturers install secondary
kV and current (milliamp) meters on each TR set. These readings
may be used directly when taking V-I data. However, in the
instances when there is no secondary kV meter, as occurred with
ESP No. 1, or when greater accuracy is required, calibration with
known voltage dividers and an accurate voltmeter may be necessary,
The secondary voltage meter calibration involves inserting
a known resistive voltage divider in parallel across the high-
voltage bus section of the precipitator and taking comparative
readings with the dc kV meter installed on the power set con-
troller. The installation of this device is shown schematically
at point No. 1 in Figure 11.
Figure 12 is a facsimile of a data sheet used to collect
data from which voltage current relationships may be plotted. In
the general heading, information is recorded which will identify
the test, the power supply (TR Set), the plate area fed by the
power set, and the determined calibration factors for the voltage
and current. Data are taken as the manual set control is gradu-
ally increased until some current flow is detected. This is
recorded as the corona starting voltage. Subsequent points are
31
-------
TO VOLTMETER FOR
SECONDARY VOLTAGE
PRECIPITATOR CONTROL
PANEL PRIMARY VOLTAGE
AND CURRENT
~T
CONTROL
"
I
TRA
I
'
r
I
NSFORMER
TO VOLTMETER FOR
I 2 I
SECONDARY CURRENT *•
J
~\
1 R3 S '
1 3< I
I 1
_i
^
vv
t
o
1 RECTIFIERS
*S
{*
T
E.S.P.
1
S
A
T
S.A. = SURGE ARRESTO
Figure 11. Voltage divider network for measuring precipitator
secondary voltages.
32
-------
DATE/TIME
POWER SET
VOLTAGE-CURRENT CURVE DATA SHEET
T/R SET NO. COLLECTING AREA
to
U)
PRIMARY
VOLTS
PRIMARY
AMPS
DCKV
T/R SET
METER
DCMA
T/R SET
METER
VC
T/R SET
SPARK
RATE
)LTAGE
DCMA (
DCMA
CORR.
DIV. MUL1
XJRRECTI
DC VOLTS
VOLTAGE
DIV.
ON
DCKV
CORR.
MA/
ft2
IMA/
cm ^
TERMINAL POINT
DETERMINED BY:
(CIRCLE ONE)
1. SPARKING
2. SEC. CURRENT LIMIT
3. SEC. VOLTAGE LIMIT
4. OTHER
COMMENTS:
Figure 12. Sample V-l curve data sheet.
-------
taken by increasing the control for some increment of current and
recording the meter readings at that point. Readings are taken
until some limiting factor is reached. This factor is recorded
on the right-hand side of the data sheet and is usually excessive
sparking or a current or voltage limitation of the power set.
The columns usually completed for each point include the
primary voltage and current, the secondary kV meter on the TR set,
the secondary current from the TR set, the spark rate, and the
dc voltage from the voltage divider. At a later time the DCMA
correction factor is applied to the TR set meter reading and the
corrected DCMA column is completed.* The DCKV CORR column is
completed by multiplying the dc Voltage DIV. column by the voltage
divider multiplier. The last two columns are completed by divid-
ing the DCMA CORR. by the appropriate collecting area in square
feet or square centimeters and applying a multiplicative factor
of 10~3. A plot is then made on linear graph paper of the DCKV
CORR. versus yA/ft2 or nA/cm2, depending on the experimental
requirement.
Electrical Energization
The effluent gas from Boiler No. 1 passes through ESP No. 1
and combines with the gas from ESP No. 2 to pass through a third
precipitator, for the purposes of the report, called ESP No. 3.
The two electric sets for ESP Nos. 1 and 2 are housed in a single
building on the ground between the two precipitators. The high
voltage is switchable so that TR Set No. 1 may feed each inlet
field in dual half-wave operation or it may feed the inlet and
outlet field of ESP No. 1 also in a dual half-wave configuration.
Testing was conducted utilizing the second aforementioned option,
ESP No. 3 had two TR's, one each for the inlet and outlet fields
in a full-wave configuration. The controls were located some
60 m (200 ft) away from the penthouse above the precipitator
which contained the transformer-rectifier sets.
Voltage dividers were installed on the high voltage busses
of ESP No. 1 for each V-I curve due to the nonexistence of kV
meters. The kV meters and voltage dividers were installed during
Phase III for calibration purposes in accordance with the preced-
ing paragraphs. Photographs of the voltage waveforms were taken
with an oscilloscope camera monitoring the low level signal from
the voltage dividers.
* On a dual half-wave installation where the voltage is measured
on one independent HV bus but the current is the sum of both
sections, the secondary current must also be multiplied by the
ratio of the plate area of the section under test to the total
plate area in order to approximate the secondary current in
that power supply leg.
34
-------
Figure 13 is a plot of the V-I curves from ESP No. 1, inlet
and outlet fields, for both full-load coal and SRC. As the
curves on ESP No. 1 were taken, the inlet and outlet fields had
essentially the same voltages applied. There were no independent
voltage controls on the inlet and outlet fields. Normally all
fields upstream of the field under test are at their operating
points while a V-I curve is being taken. Thus, there may be some
ambiguity in the lower portion of the curve for the outlet field.
Figure 14 shows the curves from ESP No. 3. These curves do not
show the corona initiation point due to a design limitation on
the mechanical voltage adjustment.
The large shift directly to the right in the curves from
ESP No. 1 and the relatively smaller shift for the curves from
ESP No. 3 may be explained by the following arguments if the
resistivity of the dust were in the typical ranges found in the
fly ash from coal fired boilers. (1) Space charge current sup-
pression would be likely if the particle size distributions show
a large increase in the number of smaller particles, or (2) some
buildup on the discharge electrodes from the time the precipitator
was cleaned and coal burned, and the burning of the SRC. However,
these arguments alone could not explain the large shift in voltage
which has occurred in ESP No. 1. A third, most likely possibility
results from the fact that the fly ash from the SRC had an
extremely low resistivity due to the excessively high carbon con-
tent.
Bickelhaupt5 has shown that in the range of temperature at
this installation (about 150°C), surface conduction is the primary
conduction component in the dust layer. The mechanism of this
conduction consists of the migration of alkali metal ions toward
the negative electrode. The resistivity of the dust determines
the ease of this migration.
Gooch and Piulle6 indicate that the electrical force tend-
ing to hold a dust layer to the surface of the collection elec-
trode may, in fact, be negative, thus actually repelling the dust
layer. This can occur with very low resistivity dusts as in the
case of the fly ash from the SRC.
5. Bickelhaupt, R. E. "Surface Resistivity and the Chemical
Composition of Fly Ash". APCA Journal, Vol. 25, No. 2.
February 1975, pp. 148-152
6. Gooch, J. P., and W. Piulle. "Studies of Particle Reentrain-
ment Resulting from Electrode Rapping". June 1977 Denver
Proceedings.
35
-------
•1
V)
a
E
(0
O
u
E
K
U'
cc
cc
O
40.0
37.5
'"
35.0
32.5
30.0
27.5
25.0
22.5
20.0
1,7.5
15.0
12.5
10.0
7.5
5:0
2.5
0.0
1
^™
- • IN A1
• OUT A1
- D IN A4 gg f
O OUT A4
—
• •
• •
—
• •
~ • • DO
••CD
• • CD
- •• OD
•• DO
- »
i i i b a ' ' ' ' ' ' ' '
4 18 22 26 30 34 38
VOLTAGE, kV
Figure 13. Voltage-current curves.
ESP No. 1 full load coal 5/24/77 - In A1 and Out A.I.
ESP No. 1 full load SRC 6/22/77 -, In A4 and.Out A4.
36
-------
40
35
(N
a
E
ra
O
UJ
oc
cc
D
O
30
20
15
10
• IN A3
• OUT A3
O IN A5
O OUT A5
O
•
I
30
32
34
VOLTAGE, kV
36
38
Figure 14. Voltage-current curves.
ESP No. 3 full load coal 6/8/77 - In A3 and Out A3.
ESP No. 3 full load SRC 6/22/77 - In A5 and Out AS.
37
-------
From the above discussion it may be theorized that positive
charge could build up on the plates and collected dust and could
be rapidly transferred to the dust particle on impact. Then this
particle is repelled and reentrained in the gas stream. The pres-
ence of these positively charged particles greatly reduce the
efficiency of the unit and causes a decrease in current for a
given voltage. Thus, the position of the V-I curves in Figures 13
and 14 may be explained by large numbers of positively charged
particles being reentrained into the gas stream.
Facts in support of this include:
1. Measured efficiency of ESP No. 1 while burning SRC
was 20% average.
2. Measured carbon content of the fly ash from SRC was
more than 80%.
3. ESP No. 1's collection electrodes were expanded
metal plates, decreasing the area to which a parti-
cle might adhere.
4. The gas velocity in ESP No. 1 was almost four times
the velocity in ESP No. 3 tending to increase the
reentrainment.
5. The SCA of ESP No. 3 was five times that of ESP No. 1,
which would tend to reduce the amount of current
suppression.
6. The measured bulk density of the fly ash during the
SRC burn was 0.2 g/cm3, compared with usually mea-
sured bulk densities of the order of 2.0 g/cm3, which
also would tend to increase the tendency toward
reentrainment.
7. The ash appeared to be nonadhesive since only those
large particles trapped in the crevices of the
point-plane probe were found when the power was
removed from the probe and the probe withdrawn from
the flue.
It is hoped the above discussion explains (1) the reason the
curves are shifted to the right when burning SRC, and (2) the
reason the curves on ESP No. 1 are shifted more than ESP No. 3.
Voltage Waveform Photographs
Figure 15a, b, and c are photographs of the oscilloscope
presentations of the high-voltage waveforms from the precipitators
under full-load conditions. Figure 15a and b are from ESP No. 1
while firing coal and SRC, respectively. Figure 15c is from
38
-------
40
5 MS/CM
MITCHELL
-40
>
H
tr
oc
o
-40
10 MS/CM
MITCHELL
5 MS/CM
MITCHELL
Figure 15. Voltage waveforms applied to the precipitators
at Plant Mitchell.
§ Jt
8 c/>
--5-
III
-------
ESP No. 3 during SRC operation. The top trace of each photograph
is the inlet field; the bottom trace, the outlet. The zero refer-
ence for the top trace on all the photographs is the upper hori-
zontal line of the graticule marked with a zero in the right-hand
margin. The zero near the center of the photograph in the right-
hand margin is the zero reference for the lower trace. The
horizontal time per division is given in the lower margin.
In observing the photographs it should be noted that these
are negative voltages and no polarity inversion has been intro-
duced. Therefore, the top of the waveform, as viewed, is the
minimum magnitude of the negative voltage. This minimum normally
coincides with the corona starting voltage. However, in cases
where the resistivity of the dust layer is high enough to generate
a back corona, the minimum may approach zero more closely; that is,
it may fall below the corona initiation voltage.7
There is a noticeable change in the top of the voltage wave-
form in Figure 15a which corresponds to a corona starting voltage
in the range of 10 to 15 kV, as compared to Figure 15b where the
corona starting voltage is now in the range of 22 to 28 kV. This
was also seen in the V-I curves in Figure 13 as a shift to the
right and was discussed at that time. Figure 15c shows the full
wave rectified waveforms applied to ESP No. 3 during the Phase III
tests. Visual inspection of the three waveforms show no transi-
tion of the waveforms below the corona start, and a relatively
short delay time, both supportive of the assumption that the
resistivity of the dust is not so high as to be detrimental to
the electrical operation.
From the resistivity, V-I curves, and voltage waveform data
presented, it can be concluded that during the Phase II burn,
resistivity of the dust did not appear to limit the precipitator
performance. However, in the Phase III burn the extremely low
dust resistivity had a disastrous effect on the operation of the
precipitators, as exemplified by the reduced efficiency of ESP
No. 1 from 76 to 20%.
7. Nichols, Grady B. "Techniques for Measuring Fly Ash Resis-
tivity", EPA Report No. 650/2-74-079, US Environmental Pro-
tection Agency, p. 326.
40
-------
SECTION 5*
COLLECTION OF SAMPLES FOR CHEMICAL ANALYSIS
PROCEDURES
This section describes the sampling program conducted with
the Source Assessment Sampling System during the burning of both
regular coal (Phase II) and solvent refined coal (Phase III).
Samples were collected primarily at the inlet and outlet of ESP
No. 1 under full- (21 MW), medium- (14 MW), and low- (7 MW) load
conditions. Two additional samples were obtained at the outlet
of ESP No. 3, with Unit No. 2 offline and Unit No. 1 at 21 MW,
during burns with each type of coal.
Two types of gasket and O-ring material—Viton and Teflon—
were used during the Phase II test program. Two sets of runs
were made, under supposedly identical plant operating conditions,
to assess the physical and chemical integrity of Viton against
that of Teflon.
All samples were processed and combined according to the
protocol given in IERL-RTP Procedures Manual: Level I Environ-
mental Assessment, June 1976, or the April 24, 1977, amendment to
that manual. The samples were then shipped to Hittman Associates,
Inc., along with coal samples and SASS train data sheets. The
results of the analyses of these samples will be found in a sepa-
rate report prepared by Hittman Associates, Inc., under Contract
No. 68-02-2162.
The SASS train supplied by the Environmental Protection
Agency was cleaned and rechecked in our laboratory and found to
perform satisfactorily. A spare filter holder, Teflon and Viton
gaskets and 0-rings, and impinger bottles were obtained from the
manufacturer, the Aerotherm Division of Acurex, Inc. A later
part of this section is devoted to observations on the performance
of the SASS train during the field tests.
* This section was prepared by Walter R. Dickson, Senior Chemist,
and William J. Barrett, Head, Analytical and Physical Chemistry
Division, Southern Research Institute.
41
-------
Approximately 3.5 kg of XAD-2 resin was cleaned according
to a modification of the Level I procedure recommended by Dr.
Phillip L. Levins of Arthur D. Little, Inc. The modification con-
sisted of extracting the resin for 24 hr with methylene chloride
instead of the usual diethyl ether and pentane. No problems were
experienced during the lengthy cleanup procedure. The cleaned
resin contained 8 yg/150 g of extractable organic material, as
determined by gas chromatography with flame ionization detection,
corresponding to a concentration of 0.2 yg/m3 for a normal SASS
train run of 5 hr at 0.11 m3/min (4 ftVniin) .
Samples were collected at three locations during both
phases of the test program: at the inlet and outlet of ESP No. 1,
and at the outlet of ESP No. 3. Since the flue gas from ESP
No. 2 was combined with that from ESP No. 1 prior to entering ESP
No. 3, Unit No. 2 was taken offline for the ESP No. 3 tests.
During Phase II it was necessary to delay the start of the
SASS train runs at ESP No. 1 for about 2 hr until a port was
cleared by other test crews. Additional sampling ports were
installed during the interval between Phases II and III to elimi-
nate this problem. At each sampling location the SASS probe was
positioned in the duct in a region representative of the average
velocity across the duct at that particular port. No attempt was
made to traverse the duct during any of the runs.
SAMPLES COLLECTED DURING PHASE II
Table 4 presents a compilation of data from the SASS train
runs during the Phase II test program. In most runs considerable
difficulty was experienced in maintaining an 0.11 m3/min
(4 ft3/min) sampling rate at either the inlet or outlet of ESP
No. 1 due to clogging of the particulate filter. Because of
restrictions imposed by plant operations on the time available
for sample collection and because approximately 20 to 30 min was
required to replace filter holders and reheat the oven to 200°C
(400°F), it was impossible to change filters as often as was
desired. Since the grain loading at the outlet of ESP No. 3 was
very low, no significant pressure drop was encountered, even
though nearly 80% of the particulate appeared to be less than
1 ym in size.
A black stain was observed in the upper region of the
organic module on all runs. This stain was apparently insoluble
in the methylene chloride/methanol rinse solution and could only
be partially removed by vigorous scrubbing with a nylon brush.
The aqueous condensate collected by the organic module was always
light green in color. Immediately following the termination of
each run, the condensate—usually 700 to 1000 ml—was.extracted
three times with about 100 ml of methylene chloride. The color
remained in the aqueous phase in all .samples.
42
-------
TABLE 4. DATA FROM RUNS WITH REGULAR COAL (PHASE II)
OJ
Date
5/25
5/26
5/27
5/28
5/29
5/30
5/31
6/2
6/5
6/6
Sampling
location
Outlet ESP No. 1
Outlet ESP No. 1
Outlet ESP No. 1
Inlet ESP No. 1
Inlet ESP No. 1
Outlet ESP No. 1
Outlet ESP No. 1
Inlet ESP No. 1
Outlet ESP No. 3
Outlet ESP No. 3
Gasket
material
Teflon
Teflon
Teflon
Teflon
Viton
Viton
Viton
Viton
Viton
Teflon
Load
condi-
tion,
MW
14
7
21
21
14
14
7
7
21
21
Volume
sampled.
m3 (dry)
29.6
*
30.5
29. Ot
30.0
30.0.
27. 6f
26. 2\
35.15
18. 6#
Average
sampling
rate,
mVmin (dry)
0
0
0
0
0
0
0
0
0
.0974
.1050
.0994
.1025
.1039
.1062
.1048
.1110
.1113
Total
particulate
collected,
g
14.67
32.61
132.81
90.50
28.51
8.08
154.39
0.076
0.052
Weight % of
particulate
Cyclone
10 pm
37.3
44.1
48.4
42.5
51.5
38.6
49.9
4.5
4.7
3 ym
29.0
24.0
37.3
39.0
25.6
33.7
34.4
3.3
0.2
1 ym
28.1
24.0
11.7
15.8
16.4
24.6
12.3
14.6
15.5
Filters
5.5
7.9
2.6
2.7
6.6
3.0
3.4
77.7
80.4
Particulate
concn ,
g/m3 (dry)
0.494
1.071
4.576
3.020
0.954
0.295
5.880
§
0.002
* The fire went out in the boiler during the early stages of this run. The SASS test was aborted.
t The fire again went out in the boiler. The unit was restarted on fuel oil without notifying us in time, thus this
sample was contaminated.
T These tests were terminated early because the plant needed to go on full load because of a power shortage.
§ The probe nozzle was not facing in the direction of the gas flow during 75% of the run.
# This test was terminated early when Unit No. 2 was brought on line because of a power shortage.
-------
The appearance of the impinger solutions varied somewhat
from run to run. The peroxide impinger was clear while the first
and second silver nitrate/ammonium persulfate impingers were
either clear or amber in color with fine black particles gener-
ally present in the first impinger and a small amount of a white
agglomerated material in the second impinger.
Since the SASS train data represent only a single point sam-
ple at each location, a critical evaluation of precipitator per-
formance, particle-size distribution, and mass loading would be
presumptuous. The results do suggest, however, that ESP No. 1
was relatively inefficient, but that ESP No. 3 performed well.
The particulate not collected by ESP No. 3 was predominantly less
than 1 ym in size while that particulate passing through ESP No. 1
was considerably larger.
SAMPLES COLLECTED DURING PHASE III
The results obtained from the SASS train runs during the
burning of solvent refined coal are given in Table 5. Severe
pressure drops were encountered after about 10 min of sampling at
both the inlet and outlet of ESP No. 1, but not at ESP No. 3.
Obviously, filter changes could not be made this often and still
complete a test run in a reasonable length of time. Therefore,
filters were changed only every 35 to 40 min despite the fact
that the 0.11 m3/min (4 ft3/min) sampling rate could not be main-
tained over this sampling period.
The data suggest that a proportionately greater fraction of
small particles were produced during the combustion of the solvent
refined coal than from regular coal. Two factors could, however,
invalidate this conclusion: a lower average sampling velocity
through the cyclones, and an ash that appeared to have a bulk
density considerably less than the ash from regular coal.
The black stains found on the upper section of the organic
module during the burning of regular coal were observed only on
those runs at the outlet of ESP No. 3 for the Phase III series.
On the second of these runs a narrow ring of green crystals was
apparent on the wall of the inner cooler where the hot gas
impinged on the cold surface.
No significant difference in the appearance of the impinger
solutions was obvious between Phase II and Phase III samples.
COMMENTS ON OPERATION OF THE SASS TRAIN
Both Teflon and Viton gaskets were used during the Phase II
test program. Teflon gaskets have a tendency to leak, especially
after having been used before, and considerable difficulty can
44
-------
TABLE 5. DATA FROM RUNS WITH SOLVENT REFINED COAL (PHASE III)*
U1
Date
6/13
6/14
6/15
6/16
6/17
6/18
6/19
6/20
6/21
6/22
6/23
6/24
Sampling
location
Outlet ESP No. 1
Outlet ESP No. 1
Outlet ESP No. 1
Outlet ESP No. 1
Inlet ESP No. 1
Inlet ESP No. 1
Outlet ESP No. 1
Outlet ESP No. 1
Inlet ESP No. 1
Outlet ESP No. 3
Outlet ESP No. 3
Outlet ESP No. 1
Load
condi- Volume
tion, sampling,
MW m3 (dry)
21
14
7
21
21
7
7
14
14
21
21
23.5
28.77
28.60
28.26t
28.46
30.04?
29.87
30.16
§
*
32.22
32.85
30.30
Average
sampling
rate,
ra3/min (dry)
0
0
0
0
0
0
0
0
0
0
.0960
.0932
.1019
.0943
.0946
.0951
.0980
.1073
.1096
.0977
Total
particulate
collected,
. g
24.02
30.64
15.68
23.93
41.75
33.71
15.62
0.186
0.193
18.03
Weight % of
particulate Particulate
Cyclone
10 vim
54.3
57.2
44.5
53.7
62.2
47.7
44.7
3.9
13.7
41.6
3 um
19.7
19.0
17.2
12.4
12.1
14.2
20.4
1.8
3.6
26.9
1 um
1.6
1.0
3.0
1.3
0.8
1.0
1.7
2.7
4.1
1.7
Filters
24.4
22.8
35.4
32.7
24.9
37.2
33.3
91.6
78.6
40.8
concn,
g/m3 (dry)
0.830
1.046
0.551
0.837
1.382
1.123
0.501
0.007
0.007
0.593
* The gasket material used in all tests was Teflon.
t This test was terminated after a near loss of fire in the boiler.
T Power was lost during the run, resulting in impinger solutions backing up into the organic module. All
samples except the particulates were discarded.
§ Power was insufficient to run the SASS train. This run was aborted.
ft The plant lost a coal mill during the early part of this run and all testing was cancelled.
-------
be expected in attaining a satisfactory leak rate check. One
solution to this problem was to wrap all outer cyclone, filter,
and organic module flanges with Teflon tape before applying the
clamps. By following this procedure an acceptable leak rate test
was obtained with the same gaskets on 13 sampling runs.
Viton gaskets provided an excellent seal for the five runs
in which they were used, but the cyclones became increasingly
more difficult to dismantle with each test. Finally, the flanges
had to be pried apart. The Viton appeared to be partially bonded
to the metal and was physically distorted. The same problem did
not occur with the seals in the organic module, which were, of
course, not subjected to the 200°C (400°F) oven temperature.
Although the flange clamps were loosened appreciably from the
tension normally applied to the Teflon gaskets, there is a possi-
bility that the Viton gaskets may still have been overly com-
pressed. However, it seems likely that the bonding problem would
have been encountered on the initial runs if that were the case.
Since only one set of the Viton gaskets was available, additional
testing of Viton could not be conducted during this field test.
Teflon gaskets, therefore, were used exclusively for the Phase III
test program.
Only minor problems were experienced with the SASS train
during this field test; however, some suggestions for improvements
in the system seem warranted.
• At locations where a significant concentration of fine
particulate is present a thimble preceding the abso-
lute filter would reduce the number of filter changes
required to maintain the desired sampling rate. A
somewhat larger oven would facilitate aligning the
cyclones, tightening connections, and changing filters.
An auxiliary heater in the oven would also decrease
the downtime following a filter change.
• The present fiberglass impinger case should be
enlarged and insulated.
• The new impinger bottles, with the VMP trademark,, pro-
vide such a tight fit with the Teflon heads that it
was difficult to dismantle them without breaking the
impinger bottles.
• The glass pour spout for the condensate bottle is
extremely fragile. A Teflon Swagelok bulkhead fitting
through a cap on an amber bottle would be less subject
to breakage.
46
-------
• The braided stainless steel transfer line from the
oven to the organic module should be insulated to
prevent burns to test personnel. A longer line
would provide greater flexibility in setting up the
system at various types of sampling locations.
47
-------
REFERENCES
1. Nichols, Grady B. "Techniques for Measuring Fly Ash Resis-
tivity", EPA Report No. 650/2-74-079, US Environmental Protec-
tion Agency, pp. 2-4.
2. White, H. J. "Resistivity Problems in .Electrostatic Precipi-
tation". J. Air Pollut. Control Assoc., 2_4 (4), April 1974,
p. 316. i
3. Oglesby?, S., Jr., and Grady B. Nichols. A Manual of Electro-
static Precipitation Technology Part 1—Fundamentals. NTIS
PB 196 380J The National Air Pollution Control Administration,
Cincinnati', Ohio, 1970, p. 251.
j i
4.. Oglesby, S>, Jr., and Grady B. Nichols. A Manual of Electro-
stati-c Precipitation Technology Part 1—Fundamentals. NTIS
PB 196 380', The National Air Pollution Control Administration,
Cincinnati,/ Ohio, 1970, p. 254.
5. Bickelhaupt, R. E. "Surface Resistivity and the Chemical Com-
position of Fly Ash". APCA Journal, Vol. 25, No. 2. February
1975, pp. 148-152.
6. Gooch, J. P., and W. Piulle. "Studies of Particle Reentrain-
ment Resulting from Electrode Rapping". June 1977 Denver
Proceedings.
7. Nichols, Grady B. "Techniques for Measuring Fly Ash Resis-
tivity", EPA Report No. 650/2-74-079, US Environmental Protec-
tion Agency, p. 326.
48
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
REPORT NO.
EPA-600/7-78-129
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
Evaluation of Electrostatic Precipitator During
SRC Combustion Tests
5. REPORT DATE
July 1978
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Grady B. Nichols and William J. Barrett
B. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Southern Research Institute
2000 Ninth Avenue, South
Birmingham, Alabama 35205
10. PROGRAM ELEMENT NO.
EHE623A
11. CONTRACT/GRANT NO.
68-02-2610, Task 2
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 ANDPERIOD COVERED
13. TYPE OF REPORT ANDPEF
Task Final; 4-8/77
14. SPONSORING AGENCY CODE
EPA/600/13
is. SUPPLEMENTARY NOTES IERL-RTP project officer is William J. Rhodes, Mail Drop 61,
919/541-2851.
is. ABSTRACT The report deals with the evaluation of an electrostatic precipitator (ESP)
and associated environmental factors during the burning of solvent refined coal (SRC)
in a boiler at Plant Mitchell of the Georgia Power Company. The effort was part of
an overall study of the use of SRC in a full-scale electric power plant. Results of a
performance evaluation of the ESP are reported and interpreted. Samples of stack
emissions were collected with a Source Assessment Sampling System (SASS) train
for chemical analysis: results of the analysis are to be reported later.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Air Pollution
Electrostatic
Precipitators
Combustion
Coal
Solvent Extraction
Tests
Sampling
Electric Power
Plants
Flue Gases
Dust
Air Pollution Control
Stationary Sources
Solvent Refined Coal
Particulate
Source Assessment
Sampling System
SASS Train
13B
131
2 IB
21D
07D
14B
10B
11G
13. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (This Report}
Unclassified
21. NO. OF PAGES
54
20. SECURITY CLASS (Tillspage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
49
------- |